Monitoring in

Neurocritical Care
 Monitoring in
Neurocritical Care

Peter D. le Roux, MD, FACS

Associate Professor of Neurosurgery
Perelman School of Medicine at the University of
Pennsylvania;
Department of Neurosurgery
Pennsylvania Hospital
Philadelphia, Pennsylvania

Joshua M. Levine, MD
Assistant Professor
Departments of Neurology, Neurosurgery, and
Anesthesiology and Critical Care
Perelman School of Medicine at the University of
Pennsylvania;
Co-Director, Neurointensive Care Unit
Hospital of the University of Pennsylvania
Philadelphia, Pennsylvania

W. Andrew Kofke, MD, MBA, FCCM

Professor
Departments of Anesthesiology and Critical Care and
Neurosurgery
Perelman School of Medicine at the University of
Pennsylvania;
Director, Neuroanesthesia
Department of Anesthesiology and Critical Care
University of Pennsylvania Health System;
Co-Director, Neurointensive Care Unit
Hospital of the University of Pennsylvania
Philadelphia, Pennsylvania
1600 John F. Kennedy Blvd.
Ste 1800
Philadelphia, PA 19103-2899

MONITORING IN NEUROCRITICAL CARE ISBN: 978-1-4377-0167-8

Copyright © 2013 by Saunders, an imprint of Elsevier Inc.
Chapter 22: “Renal, Electrolyte, and Acid Base Assessment”: Guy M. Dugan retains copyright to his original
contribution.
Chapter 23: “Gastrointestinal and Hepatic Disorders” by Christiana E. Hall and Aashish R. Patel:
Christiana E. Hall retains copyright to her original portion of the contribution only.

No part of this publication may be reproduced or transmitted in any form or by any means, electronic or
mechanical, including photocopying, recording, or any information storage and retrieval system, without
permission in writing from the Publisher. Details on how to seek permission, further information about the
Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance
Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions.

This book and the individual contributions contained in it are protected under copyright by the Publisher
(other than as may be noted herein).

Notices

Knowledge and best practice in this field are constantly changing. As new research and experience broaden
our understanding, changes in research methods, professional practices, or medical treatment may become
necessary.
Practitioners and researchers must always rely on their own experience and knowledge in evaluating
and using any information, methods, compounds, or experiments described herein. In using such
information or methods, they should be mindful of their own safety and the safety of others, including
parties for whom they have a professional responsibility.
With respect to any drug or pharmaceutical products identified, readers are advised to check the most
current information provided (i) on procedures featured or (ii) by the manufacturer of each product to be
administered, to verify the recommended dose or formula, the method and duration of administration,
and contraindications. It is the responsibility of practitioners, relying on their own experience and
knowledge of their patients, to make diagnoses, to determine dosages and the best treatment for each
individual patient, and to take all appropriate safety precautions.
To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume
any liability for any injury and/or damage to persons or property as a matter of products liability,
negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas
contained in the material herein.

Library of Congress Cataloging-in-Publication Data

  Monitoring in neurocritical care / [edited by] Peter D. Le Roux, Joshua M. Levine, W. Andrew Kofke.
    p. ; cm.
   Includes bibliographical references and index.
   ISBN 978-1-4377-0167-8 (hardcover : alk. paper)
   I. Le Roux, Peter D.  II. Levine, Joshua M.  III. Kofke, W. Andrew.
   [DNLM:1.  Monitoring, Intraoperative–methods.  2.  Neurosurgical Procedures.  3.  Central Nervous
System Diseases–surgery.  4.  Intensive Care–methods.  5.  Intensive Care Units–organization &
administration. WL 368]
   616.02’8–dc23
2012036807

Content Strategist: Charlotta Kryhl

Senior Content Development Specialist: Janice Gaillard
Content Development Specialist: Angela Rufino
Publishing Services Manager: Anne Altepeter
Senior Project Manager: Cheryl A. Abbott
Project Manager: Louise King
Design Direction: Louis Forgione Working together to grow
libraries in developing countries
Printed in China www.elsevier.com | www.bookaid.org | www.sabre.org

Last digit is the print number:  9  8  7  6  5  4  3  2  1 

To the ICU nurses who take care of all our patients
Preface
Neurocritical care has evolved rapidly in the past 10 years.
It is a specialty that focuses on the critical care manage-
ment of patients with catastrophic neurologic diseases,
including primary neurologic pathologies such as traumatic
brain injury (TBI), ischemic stroke, intracerebral hemorrhage
(ICH), subarachnoid hemorrhage (SAH), brain tumors,
infection (e.g., HIV, TB, meningitis), spinal cord injury, and
acute ascending neuropathies. In addition, neurologic dys-
function occurs in many diverse systemic disorders, including
hypoxia (e.g., post cardiac arrest, near drowning), liver dys-
function, electrolyte abnormalities, high altitude sickness,
eclampsia, lead intoxication, and malignant hypertension.
There are many reasons why brain injury or damage occurs
in patients with neurologic disorders. In particular, multiple
experimental and clinical studies have demonstrated a close
relationship between variables such as hypoxia, increased
intracranial pressure, arterial hypotension, hyperglycemia,
fever, and seizures with neurologic outcome, and accumulat-
ing evidence suggests that brain damage evolves over time.
Minimizing the burden of this delayed, or “secondary,” brain
injury, has become the focus of modern neurocritical care.
However, despite much research, trials in neuroprotection
have largely failed, in part because of their association with
prognostic heterogeneity, multiple mechanisms of cellular
damage, and a paucity of early mechanistic endpoints. This
has led to a realization that strategies of care, or “bundles,”
rather than single agents, and approaches that are tailored to
individual patient physiology and pathophysiology are neces-
sary to improve outcome.
In daily practice neurointensivists focus on the recognition
of subtle changes in the neurologic condition, interactions
between the brain and systemic derangements, and brain
physiology. The challenge for intensivists today is to identify
individuals who are at risk of developing disease or secondary
injury, determine disease severity, and distinguish responders
from nonresponders to therapy (i.e., individualized and tar-
geted medicine). Monitoring is one tool that may answer these
challenges, and it has become central to the management of
secondary brain injury and to individualized care. In recent
years (interestingly in concert with the evolution of neuro-
critical care as a distinct specialty), technology developments
have resulted in several new monitoring techniques that
provide the neurointensivist with information about brain
and cellular function. Techniques to better monitor function
of the heart, lung, liver, kidney, and blood also have evolved.
In addition, when the various techniques are combined (“mul-
timodal monitoring”), a more accurate overall picture of brain
function is produced. This approach, along with new com-
puter systems that integrate data at the bedside, and the
emerging field of bioinformatics may change the way patients
with brain injury are managed in the future.
In the last decade, there have been many advances in
neurocritical care monitoring technology and a better
understanding of what information the technology provides.
These advances have been chronicled in numerous contribu-
tions to the scientific literature, the sheer volume of which
makes it difficult for healthcare providers and device engi-
neers to keep up to date with the knowledge necessary to
provide the best patient care. In addition, although several
textbooks on critical care or head injury briefly discuss moni-
toring in a chapter or two, there is a paucity of information
that summarizes all aspects of neuromonitoring and no text-
book that is dedicated to monitoring in neurocritical care.
This book, Monitoring in Neurocritical Care, represents a
comprehensive review of neuromonitoring. We have designed
this textbook to provide the reader with a practical but in-
depth reference that describes the scientific basis and rationale
for use of a particular monitor, the information it provides,
and how this information can be used to manage the neuro-
critical care patient in an integrated fashion. We have been
fortunate to have chapters written by authors from around the
world. The contributors are clinicians, engineers, information
technology experts, and researchers who have extensive expe-
rience in the field, and each has provided an excellent and
timely review. We hope that the reader will gain a comprehen-
sive understanding about neuromonitoring and neurocritical
care and an insight into existent controversies and potential
future management. Its contents will be relevant to neurolo-
gists, neurosurgeons, neuroanesthesiologists, neurointensiv-
ists, and neuroscience nurses, and will serve as a useful resource
to intensivists working in medical and surgical ICUs. For those
who are interested in clinical or laboratory research on brain
injury in its broad sense, this book will provide many ideas
and references and will be a stepping-stone to further progress
in understanding a complex problem. The text also will serve
as a reference and guide for many engineers, bioengineers, and
computer experts who work on medical device and bioinfor-
matics development.
We have divided the book into seven sections. Section I,
Background, provides information about cerebral metabo-
lism, the principles of neurocritical care, informatics, quality
assessment, the role of ICU design and nursing, specific con-
siderations in children, the effects of anesthetic agents on
monitors, and a discussion on the relationship between bio-
ethics and monitoring. Section II, Clinical and Laboratory
Assessment, reviews clinical evaluation, sedation, pain, delir-
ium, outcomes such as neuropsychological and brain death,
extracerebral organ systems, laboratory analysis, and the role
of biomarkers. Section III, Electrophysiology, is devoted to
evoked potentials and electroencephalography. Section IV,
Radiology, discusses the use and integration of various tech-
niques, including computed tomography, xenon-CT, MRI,
PET, and SPECT in neurocritical care. Section V, Cerebral
Blood Flow, is a review of techniques such as neurosonology,
laser Doppler flowmetry, thermal diffusion flowmetry, jugular
bulb oximetry, and near infrared spectroscopy. Section VI,
vii
viii Preface
Intracranial Monitoring, provides an in-depth review of
invasive techniques, including intracranial pressure, brain
oxygen, cerebral microdialysis, and brain temperature. The
final section, Computers, Engineering, and the Future, pro-
vides a description of device development, engineering, sim-
ulation, telemedicine, robotics, information processing, data
acquisition and storage, medical informatics and multimo-
dality monitoring, noninvasive brain monitoring, and a dis-
cussion of potential future developments. It is important for
the reader to realize that the “ideal brain monitor” does not
yet exist and no single monitor will by itself affect outcome.
Instead, it is the information provided by a monitor and how
we as healthcare providers interpret and apply the informa-
tion that has the potential to improve outcome and lead to
new insights into disease processes.
We would like to express our appreciation and acknowledge
the efforts of all contributors to this volume. We also thank
the editorial, design, and production staff at Elsevier Science,
in particular Janice Gaillard, Charlotta Kryhl, Julie Goolsby,
Kate Crowley, Angela Rufino, Louis Forgione, Cheryl Abbott,
Louise King, and Anne Altepeter, who have been very helpful
in producing this volume, and Yolanda Caban who provided
excellent administrative assistance. Finally, we thank Barbara
Williams who provided outstanding editorial assistance and
made this book possible.
Peter D. le Roux, MD, FACS
Joshua M. Levine, MD
W. Andrew Kofke, MD, MBA, FCCM
Contributors
Pippa G. Al-Rawi, BSc Rosette C. Biester, PhD Maurizio Cereda, MD
Research Associate Polytrauma Neuropsychologist Assistant Professor
Neurosurgery Unit Department of Behavioral Health Department of Anesthesiology and Critical Care
Department of Clinical Neurosciences Philadelphia Veterans Affairs Medical Center; Perelman School of Medicine at the University
University of Cambridge/Addenbrooke’s Auxiliary Health Care Provider of Pennsylvania
Hospital Department of Physical Medicine and Philadelphia, Pennsylvania
Cambridge, United Kingdom Rehabilitation
University of Pennsylvania Health System Randall M. Chesnut, MD, FCCM, FACS
Pamela J. Amelung, MD Philadelphia, Pennsylvania Integra Endowed Professor of Neurotrauma
Clinical Associate Professor Department of Neurological Surgery
Department of Medicine Peter M. Black, MD, PhD Department of Orthopaedic Surgery
University of Maryland School of Medicine; Center for Advanced Brain and Spine Surgery Adjunct Professor, School of Global Health
Physician Liaison Natick, Massachusetts; Harborview Medical Center, University of
Philips VISICU Professor of Neurosurgery Washington
Baltimore, Maryland Department of Surgery Seattle, Washington
Harvard Medical School
Michal Arkuszewski, MD, PhD Boston, Massachusetts Jan Claassen, MD
Department of Neurology, Central University Attending Neurointensivist
Hospital Thomas P. Bleck, MD, FCCM Department of Neurology
Medical University of Silesia Professor Division of Neurocritical Care
Katowice, Poland; Departments of Neurological Sciences, Columbia University College of Physicians and
Department of Radiology, Neuroradiology Neurosurgery, Anesthesiology, and Medicine Surgeons
Division Rush Medical College; New York, New York
Perelman School of Medicine at the University Associate Chief Medical Officer for Critical Care
of Pennsylvania Rush University Medical Center Wendy A. Cohen, MD
Philadelphia, Pennsylvania Chicago, Illinois Professor
Departments of Radiology and Neurological
Syed T. Arshad, MD Jens Bracht, Dipl.-Phys. Surgery
Neuro-Intensivist Technical Director R&D, PD University of Washington School of Medicine
Department of Neurosurgery GMS mbH Seattle, Washington
Sacramento Medical Center/Kaiser Permanente Kiel, Germany
North Valley E. Sander Connolly, Jr., MD
Sacramento, California M. Ross Bullock, MD, PhD Bennett M. Stein Professor and Vice Chairman
Professor Department of Neurological Surgery
Ramani Balu, MD, PhD Department of Neurological Surgery Columbia University College of Physicians and
Fellow Director, Clinical Neurotrauma Surgeons;
Department of Neurology Department of Neurological Surgery Director, Cerebrovascular Research Laboratory
Division of Neurocritical Care University of Miami Miller School of Medicine Surgical Director, Neuro-Intensive Care Unit
Perelman School of Medicine at the University Miami, Florida Department of Neurological Surgery
of Pennsylvania Columbia University Medical Center/New
Philadelphia, Pennsylvania Andrew P. Carlson, MD York-Presbyterian
Department of Neurological Surgery New York, New York
Sarice L. Bassin, MD University of New Mexico School of Medicine
Assistant Professor Albuquerque, New Mexico Marek Czosnyka, PhD
Department of Neurology Reader in Brain Physics
Northwestern University Feinberg School of Emmanuel Carrera, MD Department of Clinical Neurosciences,
Medicine; Department of Neurology Neurosurgery Unit
Program Director, Neurocritical Care Fellowship Centre Hospitalier Universitaire Vaudois University of Cambridge/Addenbrooke’s
Division of Neurocritical Care (CHUV) Hospital
Northwestern Memorial Hospital Lausanne University Hospital Cambridge, United Kingdom
Chicago, Illinois Lausanne, Switzerland
John A. Detre, MD
David M. Benglis, Jr., MD Professor
Atlanta Brain and Spine Care Departments of Neurology and Radiology
Atlanta, Georgia Perelman School of Medicine at the University
of Pennsylvania
Philadelphia, Pennsylvania

ix
x Contributors

Martin E. Doerfler, MD Thomas Geeraerts, MD, PhD K.T. Henrik Huttunen, MD

Associate Professor Anesthesiology and Critical Care Department Clinical Assistant Professor of Anesthesiology
Departments of Medicine and Science Education University Hospital of Toulouse Department of Anesthesiology, Pharmacology,
Hofstra North Shore-LIJ School of Medicine at University Paul Sabatier and Therapeutics
Hofstra University Toulouse, France University of British Columbia Faculty of
Hempstead, New York; Medicine;
Vice President, Evidence Based Clinical Practice Vicente H. Gracias, MD Attending Anesthesiologist, Vancouver Acute
and Clinical Integration Professor Department of Anesthesia
North Shore-LIJ Health System Department of Surgery Division of Neuroanesthesia
Lake Success, New York Chief, Trauma/Surgical Critical Care Vancouver General Hospital
UMDNJ-Robert Wood Johnson Medical School; Vancouver, British Columbia, Canada
Guy M. Dugan, MD, FCCP Director, Level I Trauma Center
Director Robert Wood Johnson University Hospital Peter J. Kirkpatrick, MSc, FRCS(SN),
Department of Critical Care and Neurocritical New Brunswick, New Jersey FMedSci
Care Fellow of Medical Academy of Sciences
Alexian Brothers Medical Center David M. Greer, MD, MA, FCCM, FAHA Society of British Neurosurgeons (SBNS)
Elk Grove Village, Illinois Dr. Harry M. Zimmerman and Dr. Nicholas and Consultant Neurosurgeon
Viola Spinelli Professor and Vice Chairman Honorary University Lecturer
Richard P. Dutton, MD, MBA Director, Neurosciences Intensive Care Unit Department of Neurosurgery
Clinical Associate Neurology Residency Program Director Addenbrooke’s Hospital
Department of Anesthesia and Critical Care Director of Medical Studies Cambridge University Hospitals
University of Chicago; Department of Neurology Cambridge, United Kingdom
Executive Director Yale School of Medicine
American Quality Institute (AQI) New Haven, Connecticut Michel Kliot, MD
American Society of Anesthesiologists Professor of Clinical Neurosurgery
Park Ridge, Illinois Christiana E. Hall, MD, MS Department of Neurological Surgery
Associate Professor of Neurology and University of California, San Francisco School of
E. Wesley Ely, MD, MPH, FACP, FCCM Neurotherapeutics and Neurological Surgery Medicine;
Professor of Medicine and Critical Care Division of Neurocritical Care Director, Center for Management and Surgery of
Department of Medicine University of Texas Southwestern Peripheral Nerve Disorders
Center for Health Services Research Dallas, Texas San Francisco, California
Vanderbilt University School of Medicine;
Associate Director of Aging Research, VA J. Claude Hemphill III, MD, MAS W. Andrew Kofke, MD, MBA, FCCM
GRECC Professor of Clinical Neurology Professor
Veterans Affairs Tennessee Valley Healthcare Department of Neurology Departments of Anesthesiology and Critical
System University of California, San Francisco School of Care and Neurosurgery
Nashville, Tennessee Medicine; Perelman School of Medicine at the University
Director, Neurocritical Care of Pennsylvania;
Ronald G. Emerson, MD Department of Neurology Director, Neuroanesthesia
Adjunct Professor of Clinical Neurology San Francisco General Hospital Department of Anesthesiology and Critical Care
Department of Neurology San Francisco, California University of Pennsylvania Health System;
Columbia University Medical Center; Co-Director, Neurointensive Care Unit
Attending Neurologist Jiri Horak, MD Hospital of the University of Pennsylvania
Department of Neurology Assistant Professor of Clinical Anesthesiology Philadelphia, Pennsylvania
Hospital for Special Surgery and Critical Care
New York, New York Department of Anesthesiology and Critical Care Jaroslaw Krejza, MD, PhD
Perelman School of Medicine at the University Research Associate Professor
Per Enblad, MD, PhD of Pennsylvania Department of Radiology
Professor of Neurosurgery Philadelphia, Pennsylvania Perelman School of Medicine at the University
Department of Neuroscience of Pennsylvania
Section of Neurosurgery Peter Horn, MD Philadelphia, Pennsylvania;
Uppsala University Associate Professor of Neurosurgery Al-Imam Muhammad Ibn Saud Islamic
Uppsala University Hospital Department of Neurosurgery University
Uppsala, Sweden Charité Universitätsmedizin Berlin Riyadh, Saudi Arabia
Berlin, Germany
Anthony A. Figaji, MD, FCS, PhD Monisha A. Kumar, MD
Professor David A. Horowitz, MD Assistant Professor
Division of Neurosurgery Assistant Professor of Clinical Medicine Departments of Neurology, Neurosurgery, and
University of Cape Town; Perelman School of Medicine at the University Anesthesiology and Critical Care
Head of Pediatric Neurosurgery of Pennsylvania; Perelman School of Medicine at the University
Red Cross Children’s Hospital Associate Chief Medical Officer of Pennsylvania;
Cape Town, Western Cape, South Africa University of Pennsylvania Health System Director, Neurocritical Care Fellowship Program
Philadelphia, Pennsylvania Hospital of the University of Pennsylvania
Damien Galanaud, MD, PhD Philadelphia, Pennsylvania
CNRS UPR 640 LENA
Université Pierre et Marie Curie (Paris VI);
Neuroradiology
Pitié Salêtrière Hospital
Paris, France
 Contributors xi

Arthur M. Lam, MD, FRCPC David K. Menon, MD, PhD, FRCP, FRCA, DaiWai M. Olson, PhD, RN
Medical Director of Neuroanesthesia and FFICM, FMedSci Assistant Professor
Neurocritical Care Professor and Head Department of Medicine
Swedish Neuroscience Institute Division of Anaesthesia Division of Neurology
Swedish Medical Center; University of Cambridge/Addenbrooke’s Duke University School of Medicine
Clinical Professor of Anesthesiology and Pain Hospital; Durham, North Carolina
Medicine Honorary Consultant
University of Washington Perioperative Care Pratik P. Pandharipande, MD, MSCI
Member, Physician Anesthesia Services Addenbrooke’s Hospital; Associate Professor
Seattle, Washington Co-Chair, Acute Brain Injury Program Department of Anesthesiology
Wolfson Brain Imaging Centre Division of Critical Care and Perioperative
Peter D. le Roux, MD, FACS University of Cambridge Medicine
Associate Professor of Neurosurgery Cambridge, United Kingdom Vanderbilt University School of Medicine
Perelman School of Medicine at the University Nashville, Tennessee
of Pennsylvania; Asako Miyakoshi, MD
Department of Neurosurgery Assistant Professor Jose L. Pascual, MD, PhD, FRCS(C), FACS
Pennsylvania Hospital Department of Radiology Assistant Professor
Philadelphia, Pennsylvania Division of Neuroradiology Department of Surgery
University of Washington School of Medicine Division of Traumatology, Surgical Critical Care,
Joshua M. Levine, MD Seattle, Washington and Emergency Surgery
Assistant Professor Perelman School of Medicine at the University
Departments of Neurology, Neurosurgery, and Richard S. Moberg, MSE of Pennsylvania;
Anesthesiology and Critical Care President Attending Surgeon
Perelman School of Medicine at the University Moberg Research, Inc. Department of Surgery
of Pennsylvania; Ambler, Pennsylvania Hospital of the University of Pennsylvania
Co-Director, Neurointensive Care Unit Philadelphia, Philadelphia
Hospital of the University of Pennsylvania Pierre D. Mourad, PhD
Philadelphia, Pennsylvania Associate Professor Aashish R. Patel, DO
Department of Neurological Surgery Fellow in Neurocritical Care
Geoffrey T. Manley, MD, PhD Principal Physicist Department of Neurological Surgery
Professor and Vice Chairman Applied Physics Laboratory University of Texas Southwestern
Co-Director, Brain and Spinal Injury Center University of Washington Dallas, Texas
Department of Neurological Surgery Seattle, Washington
University of California, San Francisco School of Frederik A. Pennings, MD, PhD
Medicine; Barnett R. Nathan, MD Assistant Professor
Chief of Neurosurgery Associate Professor Department of Neurosurgery
San Francisco General Hospital Departments of Neurology and Internal University of Massachusetts Medical School
San Francisco, California Medicine Worcester, Massachusetts
University of Virginia
Basil F. Matta, MB, MA, BCh, FRCA Charlottesville, Virginia Ian Piper, PhD
Associate Lecturer Clinical Scientist
Department of Medicine Patrick J. Neligan, MD Clinical Physics
Division of Anaesthesia Senior Clinical Lecturer of Anaesthesia and University of Glasgow;
University of Cambridge/Addenbrooke’s Intensive Care Brain-IT Group Coordinator, Intensive Care
Hospital; Galway University Hospitals Monitoring
Divisional Director and Associate Medical National University of Ireland, Galway Department of Clinical Physics
Director Galway, Ireland Southern General Hospital Trust
Emergency and Perioperative Care Glasgow, United Kingdom
Cambridge University Foundation Trust Anoma Nellore, MD
Hospitals Fellow Amit Prakash, MBBS, MD, FRCA, EDIC
Cambridge, United Kingdom Department of Medicine Consultant, Department of Anaesthesia and
Division of Infectious Disease Intensive Care
Jonathan McEwen, MD Massachusetts General Hospital Addenbrooke’s Hospital
Clinical Assistant Professor of Anesthesiology Harvard Medical School Cambridge University Hospital Foundation
Department of Anesthesiology, Pharmacology, Boston, Massachusetts Trust
and Therapeutics Cambridge, United Kingdom
University of British Columbia Faculty of Mauro Oddo, MD
Medicine; Staff Physician J. Javier Provencio, MD, FCCM
Attending Anesthesiologist, Vancouver Acute Department of Intensive Care Medicine Associate Professor
Department of Anesthesia, Division of CHUV-University Hospital Departments of Neurology, Neurological
Neuroanesthesia Faculty of Biology and Medicine University of Surgery, and Neurosciences
Vancouver General Hospital Lausanne Cleveland Clinic Lerner College of Medicine of
Vancouver, British Columbia, Canada Lausanne University Hospital Case Western Reserve University;
Lausanne, Switzerland Director, Neurocritical Care Fellowship Program
Cleveland Clinic
Cleveland, Ohio
xii Contributors

Louis Puybasset, MD, PhD J. Michael Schmidt, PhD, MSc Martin Smith, MBBS, FRCA, FFICM
Département d’Anesthésie-Réanimation Assistant Professor of Clinical Neurophysiology Consultant and Honorary Professor in
Université Pierre et Marie Curie (Paris VI); in Neurology Neurocritical Care
Neurosurgical Intensive Care Unit Department of Neurology The National Hospital for Neurology and
Pitié Salêtrière Hospital Columbia University College of Physicians and Neurosurgery
Paris, France Surgeons; University College London Hospitals
Director London, United Kingdom
Rohan Ramakrishna, MD Neuro-ICU Neuromonitoring and Informatics
Resident Columbia University Medical Center Marco D. Sorani, PhD
Department of Neurological Surgery New York, New York Adjunct Assistant Professor
University of Washington Medical Center Department of Neurological Surgery
Seattle, Washington Sarah E. Schmitt, MD University of California, San Francisco
Assistant Professor of Clinical Neurology San Francisco, California
Mahbub Rashid, PhD Department of Neurology
Professor Perelman School of Medicine at the University Alejandro M. Spiotta, MD
University of Kansas School of Architecture, of Pennsylvania; Resident
Design and Planning Director, Electroencephalography Laboratory Department of Neurosurgery
Lawrence, Kansas Department of Neurology Cleveland Clinic
Hospital of the University of Pennsylvania Cleveland, Ohio
Gerald P. Roston, PhD, PE Philadelphia, Pennsylvania
Technology Consultant and Managing Partner John J. Stern, MD
Pair of Docs Consulting, LLC Patricia D. Scripko, MD Clinical Professor
Saline, Michigan Resident Department of Medicine
Department of Neurology Perelman School of Medicine at the University
Stuart Russell, PhD Massachusetts General Hospital of Pennsylvania;
Professor of Computer Science Brigham and Women’s Hospital Chief, Division of Infectious Diseases
Michael H. Smith and Lotfi A. Zadeh Chair in Boston, Massachusetts Department of Medicine
Engineering Pennsylvania Hospital
Computer Science Division †John M. Sewell, BSEE Philadelphia, Pennsylvania
University of California, Berkeley Chief Engineer
Berkeley, California; Active Signal Technologies, Inc. Nino Stocchetti, MD
Adjunct Professor Linthicum, Maryland Professor of Anesthesia and Intensive Care
Department of Neurological Surgery Terapia Intensiva Neuroscienze
University of California, San Francisco Robert G. Siman, PhD Fondazione IRCCS Cà Granda
San Francisco, California Research Professor University of Milan
Department of Neurosurgery Milan, Italy
Owen B. Samuels, MD Perelman School of Medicine at the University
Associate Professor of Neurosurgery and of Pennsylvania Jose I. Suarez, MD
Neurology Philadelphia, Pennsylvania Professor
Emory University School of Medicine; Departments of Neurology and Neurosurgery
Director, Division of Neuroscience Critical Care Carrie A. Sims, MD, FACS Baylor College of Medicine;
Emory Healthcare Assistant Professor Director
Atlanta, Georgia Department of Surgery Vascular Neurology and Neurocritical Care
Division of Traumatology, Surgical Critical Care, Baylor College of Medicine
Matthew R. Sanborn, MD and Emergency Surgery Houston, Texas
Resident Perelman School of Medicine at the University
Department of Neurosurgery of Pennsylvania Farzana Tariq, MD
Perelman School of Medicine at the University Philadelphia, Pennsylvania Cerebrovascular and Skull Base Fellow
of Pennsylvania Department of Neurological Surgery
Philadelphia, Pennsylvania Richard O. Sinnott, PhD University of Washington
Director, eResearch Seattle, Washington
Bernhard Schmidt, PhD University of Melbourne
Department of Neurology Melbourne, Australia Kyla P. Terhune, MD
Chemnitz Medical Centre Assistant Professor of Surgery and
Chemnitz, Germany Alan Siu Anesthesiology
Resident Division of General Surgery
Eric Albert Schmidt, MD, PhD Department of Neurological Surgery Vanderbilt University School of Medicine
Department of Neurosurgery The George Washington University Medical Nashville, Tennessee
Hôpital Purpan Center
Toulouse, France Washington, District of Columbia Brett Trimble, BSME
Director, Advanced Technology
Integra LifeSciences Corporation
†Deceased San Diego, California
 Contributors xiii

David K. Vawdrey, PhD Brandon von Tobel, MD, MBE Elisa R. Zanier, MD
Assistant Professor of Clinical Biomedical Vice President of Finance and Operations Department of Neuroscience
Informatics ImaCor, Inc. Instituto Mario Negri
Department of Biomedical Informatics New York, New York Milan, Italy
Columbia University College of Physicians and
Surgeons Howard Yonas, MD Craig Zimring, PhD
New York, New York Professor and Chairman Professor of Architecture and Psychology
Department of Neurosurgery Colleges of Architecture and Psychology
Paul M. Vespa, MD University of New Mexico School of Medicine Georgia Institute of Technology
Professor of Neurosurgery and Neurology Albuquerque, New Mexico Atlanta, Georgia
Department of Neurosurgery
David Geffen School of Medicine at UCLA; Brad E. Zacharia, MD
Director Resident
Neurocritical Care Department of Neurological Surgery
Ronald Reagan UCLA Medical Center Columbia University Medical Center/New
Los Angeles, California York-Presbyterian
New York, New York
Abbreviations
NOTE: Abbreviations may have more than one meaning, depending on their context.

3-H  hypertension, hemodilution, and hypervolemia ASTM  American Society for Testing and Materials
AACN  American Association of Critical Care Nurses AT  antithrombin
AAN  American Academy of Neurology ATC  automatic tube compensation
AARP  American Association of Retired Persons ATN  acute tubular necrosis
ABA  American Bar Association ATP  adenosine triphosphate
ABM  acute bacterial meningitis ATS  American Thoracic Society
ABP  arterial blood pressure AU  arbitrary units
ACA  anterior cerebral artery AV  audiovisual
ACE  angiotensin converting enzyme AVDO2  arteriovenous difference in oxygen
ACEI/ARBs  angiotensin-converting enzyme inhibitor/ AVM  arteriovenous malformation
angiotensin-receptor blocker BA  basilar artery
ACGME  Accreditation Council for Graduate Medical Education BAEP  brainstem auditory evoked potential
AChE  acetylcholinesterase BAM  brain acoustic monitor
ACCP  American College of Chest Physicians BANN  Bayesian Artificial Neural Network
ACoA  anterior communicating artery BBB  blood-brain barrier
ACS  abdominal compartment syndrome BFV  blood flow velocity
ACT  activated clotting time BG  blood glucose
ACTH  adrenocorticotrophic hormone BHI  breath holding index
ACV  assist control ventilation BIS  bispectral index
AD  axial diffusivity BMI  body mass index
ADC  apparent diffusion coefficient BMR  basal metabolic rate
ADH  antidiuretic hormone BOLD  blood oxygen level dependent
ADL  activities of daily living BOOST  Brain Oxygen and Outcome Study in Traumatic Brain Injury
ADNI  Alzheimer’s Disease Neuroimaging Database BP  blood pressure
ADP  adenosine diphosphate BPI  bactericidal permeability-increasing protein
ADQI  Acute Dialysis Quality Initiative BrainIT  brain monitoring with information technology
ADR  alpha/delta ratio BSI  bloodstream infection
ADT  admission/discharge/transfer BSM  bedside monitors
AED  antiepileptic drug BT  brain temperature
aEEG  amplitude-integrated electroencephalography BTO  balloon test occlusion
AEP  auditory evoked potentials BUN  blood urea nitrogen
AF  atrial fibrillation CA  cerebral autoregulation (Chapters 30, 46)
AG  anion gap CA  cardiac arrest (Chapter 25)
AHA/ASA  American Heart Association/American Stroke Association CA-BSI  catheter-associated bloodstream infection
AI  artificial intelligence CAD  coronary artery disease
AIS  acute ischemic stroke CAM-ICU  Confusion Assessment Method for the ICU
AKI  acute kidney injury CAP  College of American Pathologists
AKIN  Acute Kidney Injury Network CAR  cerebral arterial resistance
ALF  acute liver failure CAS  carotid angioplasty and stenting
ALFSG  Acute Liver Failure Study Group CASL  continuous arterial spin labeling
ALI  acute lung injury CBF  cerebral blood flow
AMA  American Medical Association CBFV  cerebral blood flow velocity
AMID  active implantable medical device CBV  cerebral blood volume
ANH  artificial nutrition and hydration CCAT  Computerized Cognitive Assessment Tool
ANN  artificial neural network CCO  continuous cardiac output
APACHE  Acute Physiology and Chronic Health Evaluation CCT  central conduction time
aPL  antiphospholipid antibodies Ccw  compliance of the chest wall
APN  advanced practice nurse CDC  Centers for Disease Control and Prevention
APP  abdominal perfusion pressure CDSA  color density spectral array
aPTT  activated partial thromboplastin time CEA  carotid endarterectomy
ARAS  ascending reticular activating system cEEG  continuous electroencephalography
ARC  absolute reticulocyte count CES  cholesterl emboli syndrome
ARDS  acute respiratory distress syndrome CEUs  continuing education units
ARi (or ARI)  autoregulation index CFM  cerebral function monitoring
ASA  American Society of Anesthesiology C-FMZ  C-flumazenil
aSAH  aneurysmal subarachnoid hemorrhage Cho  choline
ASIA  American Spinal Injury Association CHr  reticulocyte hemoglobin content
ASL  arterial spin labeling CI  coagulation index

xv
xvi Abbreviations

CINMA  critical illness neuromuscular abnormalities DHHS  Department of Health and Human Services
CIPM  critical illness polyneuromyopathy DI  diabetes insipidus
CIPNM  critical illness polyneuropathy and myopathy DIC  disseminated intravascular coagulation
CIRCI  critical illness related corticosteroid insufficiency DIND  delayed ischemic neurologic deficit
CK  creatine kinase DIT  drug-induced thrombocytopenia
Cl  compliance of the lung DITP  drug-induced immune thrombocytopenia
CLIA 88  Clinical Laboratory Improvements Amendments of 1988 dIVC  inferior vena cava diameter
CLAB  central line-associated bacteraemia DLCO  diffusing capacity of the lung for carbon monoxide
CLABSI  central line-associated bloodstream infection DMN  default mode network
CMAP  compound muscle action potentials DO2  oxygen delivery
CMO  comfort measures only DoD  Department of Defense
CMRO2  cerebral metabolic rate of oxygen DoE  Department of Energy
CMRGluc  cerebral metabolic rate of glucose DRG  diagnostic related group
CMS  Centers for Medicare and Medicaid Services DRS  Disability Rating Scale
CMV  cytomegalovirus DS  Down syndrome
CNS  central nervous system DSA  digital subtraction angiography
CO  cardiac output DSM  Diagnostic and Statistical Manual of Mental Disorders
CO2  carbon dioxide DSP  digital signal processing
COGIF  Consensus on Grading Intracranial Flow DTI  diffusion tensor imaging
COI  cerebral oxygenation index DUS  duplex ultrasonography
COM  communication DV  data validation
COMBI  Center for Outcome Measurement in Brain Injury DVT  deep vein thrombosis
COMPACCS  Committee on Manpower for Pulmonary and Critical DWI  diffusion-weighted imaging
Care Societies EAA  excitatory amino acids
COPD  chronic obstructive pulmonary disease EBM  evidence-based medicine
COx  cerebral oximetry index EBNP  evidence-based nursing practice
CPAP  continuous positive airway pressure EBP  evidence-based practice
CPOE  computerized order entry EC-IC  extracranial to intracranial
CPP  cerebral perfusion pressure ECA  external carotid artery
CPR  cardiopulmonary resuscitation ECCO  Essentials of Critical Care Orientation
CPSE  complex partial status epilepticus ECF  extracellular fluid
CPT  current procedural terminology ECoG  electrocorticogram
Cr  creatine ED  emergency department
CRH  corticotropin-releasing hormone EDC  extended differential count
CRM  crew/crisis resource management EDH  epidural hematoma
CRMP  collapsin response mediator protein EDM  esophageal Doppler monitor
CRP  C-reactive protein EDTA  ethylene diamine tetra acetate
CRRT  continuous renal replacement therapy EEG  electroencephalography; electroencephalogram
CRS  Coma Recovery Scale EF  ejection fraction
CRS-R  Coma Recovery Scale–Revised E-GOS  Extended Glasgow Outcome Scale
Crs  compliance of respiratory system EHR  electronic health record
CSA  cross-sectional area EIT  electrical impedance tomography
CSD  cortical spreading depression EKG  electrocardiogram
CSE  convulsive status epilepticus ELISA  enzyme-linked immunological sample assay
CSF  cerebrospinal fluid EMG  electromyogram
CSW  cerebral salt wasting EMI  electromagnetic interference
CT  computed tomography EMR  electronic medical record
CTA  computed tomography angiography EMS  emergency medical services
CTP  computed tomography perfusion (Chapters 13, 26) EN  enteral nutrition
CTP  Child-Turcotte-Pugh (Chapter 23) eNAA  extracellular N-acetyl aspartate
CTT  central conduction time EOG  electrooculogram
CTV  cerebral venous thrombosis EP  evoked potential
CVC  central venous catheter EPIC  extended prevalence of infection in intensive care
CVP  central venous pressure EPL  estimated percent lysis
CVR  cerebrovascular resistance EPO  erythropoietin
CVT  cerebral venous thrombosis EPOR  erythropoietin receptor
CVVH  continuous veno-venous hemofiltration ESA  erythropoiesis-stimulating agents
CXR  chest x-ray ESICM  European Society of Intensive Care Medicine
D  diameter of conduit ESO  European Stroke Organization
DAI  diffuse axonal injury ESRD  end-stage renal disease
dARi  dynamic autoregulation index ET  endotracheal tube
DBN  dynamic Bayesian network etCO2  end-tidal carbon dioxide
DBP  diastolic blood pressure ETF  Emerging Technology Fund
DBS  deep brain stimulator EU  European Union
DC  decompressive craniectomy EVD  external ventricular drain
DCI  delayed cerebral ischemia EVLWI  extravascular lung water index
DCS  diffuse correlation spectroscopy FA  fractional anisotropy
DHCA  deep hypothermic circulatory arrest FC  Foley catheter
 Abbreviations xvii

FDA  Food and Drug Administration HSE  Herpes simplex encephalitis

FD&C  Food, Drug, and Cosmetic Act HSV  Herpes simplex virus
FDG  fluorodeoxyglucose HUS  hemolytic uremic syndrome
Fe  iron HV  hyperventilation
FENa  fractional excretion of sodium IAP  intra-abdominal pressure
FET  field effect transistor IBW  ideal body weight
FFF  family, friends, and fools ICA  internal carotid artery
FFP  fresh frozen plasma ICAMs  intracellular adhesion molecules
FFT  fast-Fourier transformation ICE  integrated clinical environment
FIM  Functional Independence Measure ICG  indocyanine green
FiO2  inspiratory oxygen fraction ICH  intracerebral hemorrhage
FLAIR  fluid-attenuated inversion recovery ICP  intracranial pressure
fMRI  functional magnetic resonance imaging ICDSC  Intensive Care Delirium Screening Checklist
FNHTR  febrile non-hemolytic transfusion reactions ICU  intensive care unit
FNN  Foundations of Neuroscience Nursing IDE  investigational device exemption
FOIA  Freedom of Information Act IDSA  Infectious Diseases Society of America
FOUR  Full Outline of UnResponsiveness IEEE  Institute of Electrical and Electronic Engineers
FRBC  fragmented RBC iEEG  intermittent electroencephalography
FTc  flow time correction IFNα  interferon-α
F/V  Flotrac-Vigileo IH  intracranial hypertension
FV  flow velocity IHD  ischemic heart disease
FVl  flow velocity, left IIT  intensive insulin therapy
FVL  Factor V Leiden IJV  internal jugular vein
FVr  flow velocity, right IL  interleukin
GAAP  generally accepted accounting practices IMZ  iomazenil
GABA  gamma-aminobutyric acid IND  investigational new drug
GAP  growth associated protein INR  international normalized ratio
GB  gigabyte IOM  Institute of Medicine
GCS  Glasgow Coma Scale IOP  intraocular pressure
GDP  gross domestic product IP  intellectual property
GFAP  glial fibrillary acidic protein IPC  intermittent pneumatic compression
GFR  glomerular filtration rate IPF  immature platelet fraction
GH  growth hormone IPS  intensive care unit physician staffing
GHBP  growth-hormone binding protein IRF  immature reticulocyte fraction
GHRP  GH-releasing peptide ISF  International Sepsis Forum
GI  gastrointestinal ISHEN  International Society for Hepatic Encephalopathy and
GMDI  Glioma Molecular Diagnostic Initiative Nitrogen Metabolism
GMP  good manufacturing practices ISO  International Organization for Standardization
GN  glomerulonephritis ISS  Injury Severity Score
GNP  gross national product IT  information technology
GOS  Glasgow Outcome Scale ITAA  Information Technology Association of America
G-PEDs  generalized periodic epileptiform discharges ITBVI  intrathoracic blood volume index
1
H-MRS  proton magnetic resonance spectroscopy IV  intravenous
HAC  hospital-acquired condition IVC  inferior vena cava
HAI  hospital-acquired infection IVIg  intravenous immunoglobulin
HAP  hospital-acquired pneumonia KIM-1  kidney injury molecular 1
HbO2  oxyhemoglobin LC  liquid chromatography
HCAP  heathcare-associated pneumonia LCD  liquid crystal display
Hct  hematocrit LDF  laser Doppler flowmetry
HE  hepatic encephalopathy LDH  lactate dehydrogenase
HELLP  hemolytic anemia, elevated liver enzymes, low platelets LDUH  low-dose unfractionated heparin
syndrome LGIB  lower gastrointestinal bleeding
Hgb  hemoglobin LGR  lactate : glucose ratio
HHb  deoxyhemoglobin LIP  lower inflection point
HHCFA  Health Care Financing Organization LMWH  low molecular weight heparin
HHNK  hyperglycemic hyperosmolar nonketotic lp(a)  lipoprotein (a)
HIE  hypoxic-ischemic encephalopathy LOC  loss of consciousness
HIPAA  Health Insurance Portability and Accountability Act LOH  Loop of Henle
HIT  heparin-induced thrombocytopenia LOI  lactate oxygen index
HiTT  high dose thrombin time LOS  length of stay
HIV  human immunodeficiency virus LP  lumbar puncture
HLA  human leukocyte antigens LPR  lactate : pyruvate ratio
HMG CoA  3-hydroxy-3-methylglutaryl coenzyme A LUS  lung ultrasound
HMS  Haemostasis Management System LV  left ventricle
HPA  hypothalamic-pituitary axis LVEDA  left ventricular end-diastolic area
HPS  human patient stimulator LVEDV  left ventricular end-diastolic volume
HR  heart rate MA  maximum amplitude
HRS  hepatorenal syndrome MAC  mid arm circumference (Chapter 14)
xviii Abbreviations
MAC  minimum alveolar concentration (Chapter 9)
MAP  mean arterial pressure
MAP2  microtubule-associate protein type 2
MB  megabyte
MBP  myelin basic protein
MBs  microbubbles
MCA  middle cerebral artery
MCHC  mean corpuscular hemoglobin concentration
MCI  mild cognitive impairment
MCMC  Monte Carlo Markov Chain
MCS  minimally conscious state
MCV  mean corpuscular volume
MCVr  reticulocyte volume
MD  mean diffusion (Chapter 28)
MD  microdialysis (Chapters 36, 48)
MDCT  multidetector computed tomography
MDPnP  Medical Device Plug and Play
MDS  myelodysplastic syndrome
MEE  measured energy expenditure
MEMS  micro electro-mechanical system
MEP  motor evoked potential
MES  microembolic signal
MeSH  medical subject heading
MFI  microcirculatory flow index
mGy  milli-Gray
MI  Maastricht Index (Chapter 14)
MI  myocardial infarction (Chapter 15)
MIB  Medical Information Bus
MICU  medical intensive care unit
MIPS  maximum intensity projections
MMM  multi modality monitor
MMP  matrix metalloproteinases
MMPF  multimodal pressure-flow
MMSE  Mini-Mental Status Examination
MOANS  Mayo’s Older Americans Normative Studies
MOCA  Montreal Cognitive Assessment
MODS  Multiple Organ Dysfunction Syndrome
MPAI  Mayo Portland Adaptability Inventory
MPM  Mortality Probability Model
MPV  mean platelet volume
MR  magnetic resonance
MRA  magnetic resonance angiography
MRI  magnetic resonance imaging
MRP  magnetic resonance perfusion
MRS  magnetic resonance spectroscopy
mRS  modified Rankin Scale
MRSA  methicillin resistant Staphylococcus aureus
MS  mass spectrometry
mSv  milli-Sieverts
MTT  mean transit time
Mx  autoregulation
NAA  N-acetyl aspartate
NADH  nicotinamide adenine dinucleotide
NAG  N-acetyl-glucosaminidase
NBH  normobaric hyperoxia
nCPPl  noninvasive cerebral perfusion pressure, left
nCCPr  noninvasive cerebral perfusion pressure, right
NCCU  neurocritical care unit
NCS  nonconvulsive seizures
NCSE  nonconvulsive status epilepticus
NEMS  nano electro-mechanical system
NFH  neurofilament heavy chain
NFL  neurofilament light chain
NFM  neurofilament middle chain
NGAL  neutrophil gelatinase-associated lipocalin
NHSN  National Healthcare Safety Network
NIBP  noninvasive blood pressure
NICE  National Institute for Clinical Excellence
NiCO  noninvasive cardiac output
NICU  neurointensive care unit
NIH  National Institutes of Health
NIHSS  National Institutes of Health Stroke Scale
NINDS  National Institute of Neurological Disease and Stroke
NIRS  near infrared spectroscopy
NMDA  N-methyl-D-aspartic
NNIS  National Nosocomial Infections Surveillance
NO  nitric oxide
NOS  nitric oxide synthase
NOx  nitric oxide metabolite
NPH  normal pressure hydrocephalus
NPV  negative predictive value
NRBC  nucleated red blood cell
NRCPR  National Registry of Cardiopulmonary Resuscitation
NRI  Nutritional Risk Index
NSAID  nonsteroidal anti-inflammatory drug
NSE  neuron-specific enolase
NSF  National Science Foundation (Chapter 38)
NSF  nephrogenic systemic fibrosis (Chapter 22)
NTIS  nonthyroidal illness syndrome
NTP  Network Time Protocol
OAST  Optimizing Analysis of Stroke Trials
OEF  oxygen extraction fraction
OGR  oxygen-glucose ratio
ONS  optic nerve sheath
OPS  orthogonal polarizing spectral
OR  operating room
P4P  pay-for-performance
PA  pulmonary artery
PAC  pulmonary artery catheter
PaCO2  arterial carbon dioxide tension
PACS  picture archiving and communication system
PAD  pulmonary artery diastolic
PAGE  polyacrylamide gel electrophoresis
PAI  plasminogen activator inhibitor
Palv  alveolar pressure
PaO2  arterial oxygen tension
PAOP  pulmonary artery occlusion pressure
PAP  pulmonary pressures
PAR  pressure autoregulation
PAS  pulmonary artery systolic
PASL  pulsed arterial spin labeling
PAV  percent alpha variability
PAV+  proportional assist ventilation
Paw  airway pressures
PbtO2  brain tissue oxygen tension
PCA  posterior cerebral artery
pCAM-ICU  Pediatric Confusion Assessment Method for Intensive
Care Unit
PCI  percutaneous coronary interventions
PCM  pulse contour method
PCR  polymerase chain reaction
PCT  proximal convoluted tubule
PCWP  pulmonary capillary wedge pressure
PDSA  Plan-Do-Study-Act
PE  pulmonary embolism
PED  periodic epileptiform discharges
PEEP  positive end-expiratory pressure
PEEPi  intrinsic positive end-expiratory pressure
Pes  esophageal pressure
PET  positron emission tomography
PF4  platelet factor 4
PI  pulsatility index
PICARD  Program to Improve Care in Acute Renal Disease
PICC  peripherally inserted central catheter
PID  peri-infarct depolarizations
PICU  pediatric intensive care unit

PILOT  Pediatric Intensity Level of Therapy
PIM  Pediatric Index of Mortality
Pip  peak inspiratory pressure
Pl  transpulmonary pressure
PLED  periodic lateralized epileptiform discharge
Pleth-HR  heart rate from oxygen saturation monitor
PMA  premarket approval
PN  parenteral nutrition
PO2  partial pressure of oxygen
POC  point of care
PoCoA  posterior communicating artery
POCT  point of care testing
Pplat  plateau pressure
PPV  pulse pressure variation
PRAM  pressure recording analytical method
PRBC  packed red blood cell
PRESS  posterior reversible leukoencephalopathy syndrome
PRx  cerebrovascular pressure reactivity index
PSV  pressure support ventilation
PT  prothrombin time
PT/INR  prothrombin time/international normative ratio
PTS  Pediatric Trauma Score
PTSD  post-traumatic stress disorder
PTT  partial thromboplastin time
PVD  patient-ventilator dyssynchrony
PVI  pressure-volume index
PvO2  oxygen partial pressure in venous blood
PVS  persistent vegetative state
PWI  perfusion-weighted imaging
QFD  Quality Functional Deployment
QI  quality initiatives
qMRA  quantitative magnetic resonance angiography
QODD  quality of dying and death
QOL  quality of life
QUASAR  quantitative STAR labeling of arterial regions
RA  right atrium
RAAS  renin-angiotensin-II-aldosterone system
RAI  relative adrenal insufficiency
RAIDs  redundant arrays of independent disks
RALS  remote automated laboratory system
RAP  cerebrospinal compensatory reserve
RASS  Richmond Agitation-Sedation Scale
RBC  red blood cell
RBCT  red blood cell transfusion
RC  reticulocyte count
rCBF  regional cerebral blood flow
RCM  radiocontrast media
RCT  randomized clinical trial
RDW  red cell distribution width
REE  resting energy expenditure
REG  rheoencephalography
RF  radio frequency
rhuEPO  recombinant human erythropoietin
Ri  inspiratory resistance
RN  registered nurse
ROI  regions of interest
ROSC  return of spontaneous circulation
ROTEM  rotational thromboelastometry
RPT  robotic telepresence
RQ  respiratory quotient
RR  respiratory rate
RRT  renal replacement therapy
RS  Ramsay Sedation Scale
RSE  refractory status epilepticus
rSO2  regional cerebral oxygen saturation
rSO2C  regional cerebral tissue oxygenation
rSO2R  regional renal tissue oxygenation
rSO2S  regional splanchnic tissue oxygenation
Abbreviations xix
rTEG  rapid thromboelastography
rtPA  recombinant tissue plasminogen activator
RTS  Revised Trauma Score
RTV silicone  room temperature vulcanization silicone
RV  right ventricle
SAH  subarachnoid hemorrhage
SaO2  arterial oxygen saturation
SAPS  Simplified Acute Physiology Score
SAS  Riker Sedation-Agitation Scale
SATs  spontaneous awakening trial
SBDP  spectrin breakdown degradation product
SBI  secondary brain insults
SBIR  Small Business Innovative Research
SBP  systolic blood pressure
SBP 120  spectrin breakdown products
SBTs  spontaneous breathing trials
SC  serum creatinine
SCAT  Standardized Concussion Assessment Tool
SCCM  Society of Critical Care Medicine
SCI  spinal cord injury
SCM  sternocleidomastoid muscle
SCUF  slow continuous ultrafiltration
ScVO2  oxygen saturation in superior vena cava (Chapters 19, 40)
ScvO2  mixed venous oxygen saturation (Chapter 23)
SDE  subdural empyema
SDF  sidestream dark field
SDH  subdural hematoma
SE  status epilepticus
SF  short form
SIADH  syndrome of inappropriate antidiuretic hormone
sICH  spontaneous intracerebral hemorrhage
SIMV  Synchronized Intermittent Mechanical Ventilation
SIP  Sickness Impact Profile
SIRS  systemic inflammatory response syndrome
SjvO2  jugular venous oxygen saturation
SLE  systemic lupus erythematosus
SLT  scanning laser tomography
SMBG  self-monitoring blood glucose
SMR  standardized mortality ratio
SNAP  superlattice nano wire pattern transfer
SOFA  Sequential Organ Failure Assessment
SPECT  single photon emission computed tomography
SpO2  oxygen saturation
SR  suppression ratio
SSC  Surviving Sepsis Campaign
SSEP  somatosensory evoked potential
STC  shock trauma center
StcO2  transcranial oxygen saturation
StO2  tissue oxygen saturation
STTR  Small Business Technology Transfer
SV  stroke volume
SVC  superior vena cava
SVM  support vector machine
SVO2  venous oxygen saturation
SVV  stroke volume variability
SWI  susceptibility-weighted imaging
T  body temperature (Chapters 45, 48)
T  translational (Chapter 6)
T3  triiodothyronine
T4  thyroxine
TBI  traumatic brain injury
Tc99m-ECD  technetium-99m-ethylcysteinate dimer
Tc99m-HMPAO  technetium-99m-hexamethylpropyleneamine oxide
TCCS  transcranial color-coded duplex sonography
TCD  transcranial Doppler
TcE  transcranial electrical (stimulation)
TcEMEP  transcranial electrical motor evoked potential
TcMMEP  transcranial magnetic motor evoked potential
xx Abbreviations
TDF  thermal diffusion flowmetry
TDM  time division multiplex
TEE  transesophageal echocardiography
TEG  thromboelastography
TF  tissue factor
TH  therapeutic hypothermia
THb  total hemoglobin
THR  transient hyperemic response
THx  total hemoglobin reactivity
TI  thermal index
TIBC  total iron binding capacity
TIBI  thrombolysis in brain ischemia
TIL  Therapeutic Intensity Level
TLC  total lymphocyte count
TNF  tumor necrosis factor
TOI  tissue oxygen index
TOR  brain tissue oxygen response
tPA  tissue plasminogen activator
TPL  transplant
TR  tricuspid regurgitation
TRALI  transfusion-related acute lung injury
TRC  tanned red cell
TRH  thyrotropin-releasing hormone
TRICC  transfusion requirements in critical care
TRISS  Trauma Injury Severity Score
TS  test standard
TSF  triceps skinfold thickness
TSH  thyroid stimulating hormone
T-tau  total tau
TTE  transthoracic echocardiography
TTP  thrombotic thrombocytopenic purpura
TV  tidal volume
UCH  ubiquitin C-terminal hydrolase
UFH  unfractionated heparin
UGIB  upper gastrointestinal bleeding
UIP  upper inflection point
UMLS  Unified Medical Language System
UNa  urinary sodium concentration
UO  urine output
US  ultrasound
UTI  urinary tract infection
UUN  urine urea nitrogen
VA  vertebral artery
VALI  ventilator-associated lung injury
VAP  ventilator-associated pneumonia
VASO  vascular space occupancy
VC  venture capital (Chapter 38)
VC  vital capacity (Chapter 20)
VCAM-1  vascular cell adhesion molecule-1
VE  expired minute volume
vEEG  video electroencephalography
VEGF  vascular endothelial cell growth factor
VEP  visual evoked potentials
V’I  inspiratory flow
VILI  ventilator-induced lung injury
VKA  vitamin K antagonist
VMRr  vasomotor reactivity
VO  virtual organization
VO2  oxygen consumption
VPR  volume-pressure response
VRE  vancomycin-resistant enterococcus
VS  vegetative state
VSP  vasospasm
vT  tidal volume
VTE  venous thromboembolism
VTI  velocity time integral
vWD  von Willebrand’s disease
vWF  von Willebrand multimers
WBC  white blood cell
WDM  wavelength division multiplex
WFNS  World Federation of Neurological Societies
WHO  World Health Organization
WNV  West Nile virus
WNVE  West Nile virus encephalitis
Xe-CT  xenon-enhanced computed tomography
 I
Chapter
1  
Principles of Cerebral
Metabolism and Blood Flow
Brad E. Zacharia and E. Sander Connolly, Jr.
Introduction
The brain’s survival and function depend on its ability to
maintain a constant supply of oxygen and energy-rich sub-
strate. To accomplish this feat the complex architecture of the
brain that accounts for approximately 2% of total body mass
demands approximately 20% of the total cardiac output, and
consumes roughly one quarter of resting total body oxygen
consumption.1,2
Cerebral Hemodynamics
An understanding of the complex cerebral circulation first
requires an understanding of basic physical principles about
flow. Ohm’s law predicts that flow (Q) is proportional to the
pressure gradient between inflow and outflow (ΔP) divided by
the resistance to flow (R).
Q = ∆P / R
Analogously in the brain, cerebral perfusion pressure (CPP),
the difference between arterial inflow and venous outflow
pressure, represents the driving pressure for cerebral blood
flow (CBF). CPP is the difference between mean arterial pres-
sure (MAP) and the pressure in the thin-walled veins.1 Venous
pressure changes with changes in intracranial pressure (ICP)
and is typically 2 to 5 mm Hg higher than the central venous
pressure. CPP, therefore, can be described as:
CPP = MAP − ICP
Poiseuille’s law demonstrates that in addition to CPP (ΔP),
blood viscosity (η) and vessel radius (r) are key determinants
of CBF (Q). Vessel length (L) is generally not measured in
physiologic systems.
Q = (πr 4 ∆P)/ 8 η L)
Viscosity, the internal friction of blood flow, is frequently
overlooked because direct measurement is difficult. Impor-
tantly, however, viscosity is felt to vary directly with changes
in hematocrit and any other process that alters blood’s cellular
composition. Blood viscosity also varies inversely with vessel
diameter. This is a consequence of the increased velocity gra-
dient of laminar flow as vessel size decreases, a parameter
known as the shear rate.3 Thus for a given blood velocity, shear
rates are greater in smaller vessels and apparent viscosity is
2 © Copyright 2013 Elsevier Inc. All rights reserved.
consequently lower in the microcirculation. This effect is
known as the Fahraeus-Lindquist effect.1,4
The most powerful factor in Poiseuille’s law that can influ-
ence CBF is vessel radius. For example, the maximum
constriction that can be obtained by hyperventilation is
approximately 20% from baseline. This, however, leads to a
decrease in CBF of approximately 60%.5 From a practical
standpoint, all of this diameter regulation takes place in the
microcirculation.
With knowledge about how much blood flows to the brain
and how much oxygen the brain extracts from this blood
(arteriovenous difference in oxygen, AVDO2), one can calcu-
late cerebral metabolic rate of oxygen (CMRO2) consumption.
CMRO2 = CBF × AVDO2
Physiology of Cerebral Blood Flow
and Cerebral Blood Volume
Normal CBF is approximately 50 mL/100 g brain tissue/min.6
Flow is normally greater in gray matter than white matter.
Under normal conditions there are critical thresholds for CBF
in the brain to maintain tissue health and cellular integrity.
When CBF is reduced to 25 mL/100 g/min there is electroen-
cephalographic slowing, and at 20 mL/100 g/min loss of con-
sciousness occurs. When CBF is less than 18 mL/100 g/min,
cellular homeostasis becomes jeopardized and neurons convert
to anaerobic metabolism.2,7,8 At a CBF of 10 mL/100 g/min,
membrane integrity is compromised and irreversible brain
damage is inevitable (Table 1.1). Tissue infarction, however, is
related not only to CBF, but to time as well.7
Many different factors are postulated to maintain adequate
CBF, but it is believed that local metabolic factors are of
primary importance. Under normal circumstances, in areas of
increased cerebral activity, vasoactive substances are released,
which alters vascular tone and local perfusion. The compen-
satory increase in perfusion then creates a local washout
effect, which leads to a reduction in perfusion. Key local
metabolites include, but are not limited to, carbon dioxide
(CO2), potassium, adenosine, nitric oxide, histamine, and
prostaglandins.2
Cerebral blood volume (CBV) is determined by CBF and
capacitance vessel diameter. CBV thus increases with vasodila-
tion and decreases with vasoconstriction. The relationship
 Section I—Background 3

Table 1.1  Cerebral Blood Flow Thresholds

Cerebral Blood Flow (mL/100 g/min) Threshold Consequences
40-60 — Normal
20-30 Neurologic function Start of neurologic symptoms
Altered mental status
16-20 Electrical failure Isoelectric electroencephalogram
Loss of evoked potentials
10-12 Ionic pump failure Na+ and K+ pump failure
Cytotoxic edema
<10 Metabolic failure Complete metabolic failure with gross
disturbance of cellular energy homeostasis

From Doberstein C, Martin NA. Cerebral blood flow in clinical neurosurgery. In Youmans R, editor. Neurological surgery: a comprehensive reference guide to the
diagnosis and management of neurosurgical problems. Philadelphia: WB Saunders; 1996. p. 521, Table 21-4.

between CBF and CBV, however, is complex and under both Vasodilatory Vasoconstrictory
cascade cascade
normal and pathologic situations these variables may be
inversely related. Furthermore, blood volume is not equally Autoregulation
distributed throughout the brain, with greater volume per Passive Zone of normal breakthrough
collapse autoregulation zone
unit weight in gray matter than white matter. The central
volume principle relates the volume that intravascular blood Cerebral blood flow
occupies within the brain (CBV in mL) and the volume of VASCULAR CALIBER
75 60
(mL/100 g/min)
blood, which moves through the brain per unit time (CBF in

mL/min).9
CBF = CBV /τ
This equation implies that a change in vascular diameter will
directly affect CBV, but not necessarily CBF if mean transit
time (τ) is simultaneously altered. Normally, increases in CBV
are handled by the vasculature in two ways: (1) restricted
inflow via constriction of the major feeding arteries and (2)
increased venous outflow. On the other hand, dog studies have
shown that within the autoregulatory range, increased CBV
and mean transit time maintain CBF constancy.10 In healthy
volunteers with intact autoregulation, transcranial Doppler
studies suggest that a sudden steep decrease of arterial blood
pressure results in an increase of intracranial blood volume,11
that is to say, vasodilatation may protect the brain from a
decrease in CBF during a decrease in CPP. In pathologic states
the relationship between CBV and CBF is more complex and
may depend on the region affected, CO2 reactivity, autoregula-
tion, and blood-brain barrier permeability.12,13
Cerebral Autoregulation
The normal cerebral vasculature is able to regulate its own
blood flow through metabolic or pressure regulation. In meta-
bolic autoregulation, CBF changes proportionally with
changes in CMRO2 demand. In pressure autoregulation14 arte-
rial diameter increases or decreases to actively control CBF
and maintain constant flow over a range of perfusion pres-
sures; that is to say, CBF remains constant despite altered CPP.
When CPP exceeds the limits of the autoregulatory plateau,
cerebral resistance vessels respond passively to further changes
in pressure. Thus once the limits of autoregulation are
exceeded, CBF changes passively with increases or reductions
in perfusion pressure (Fig. 1.1). In an attempt to compensate
for reduced CBF, the oxygen extraction fraction (OEF) from
ICP (mmHg)
50 40
25 20
25 50 75 100 125 150 175
Cerebral perfusion pressure (mm Hg)
Fig. 1.1  Relationship between extremes of cerebral perfusion pressure
(CPP) and intracranial pressure (ICP) in states of normal (purple line) and
reduced intracranial compliance (blue line). In the vasodilatory cascade
zone, CPP insufficiency and intact pressure autoregulation lead to reflex
cerebral vasodilation and increased ICP; the treatment is to raise CPP. In
the autoregulation breakthrough zone, pressure and volume overload,
which overwhelms the brains capacity to autoregulate, leads to increased
cerebral blood volume and ICP; the treatment is to lower CPP. (From Rose
JA, Mayer SA. Optimizing blood pressure in neurological emergencies. Neurocrit
Care 2004;1:289.)
the blood is increased in an attempt to maintain oxygen deliv-
ery (DO2) for metabolism and to prevent ischemia. When OEF
is increased (oligemia), the brain may not be able to compen-
sate for added reductions in DO2 and, if not corrected, infarc-
tion is likely. Clinical symptoms develop once the reduction
in CPP exceeds the brain’s ability to extract enough oxygen to
satisfy its metabolic demands. Breakthrough of the upper
limits of autoregulation also is accompanied by segmental
dilation of arterial vessels and breakdown of the blood-brain
barrier (BBB) and vascular endothelium, which ultimately
contributes to cerebral edema.
The mechanisms responsible for CBF autoregulation are
not yet clearly understood. Several mechanisms are proposed:
(1) The myogenic hypothesis states that smooth muscle in
resistance arteries responds directly to alterations in CPP;15
(2) The metabolic hypothesis states that reduced CBF
4 Section I—Background
stimulates the release of vasoactive substances from the brain
that then stimulate the dilatation of cerebral vessels; (3) Auto-
regulation also may be modified by the sympathetic nervous
system, which shifts both the upper and lower limits of auto-
regulation to higher perfusion pressures.1
Carbon Dioxide Reactivity
The cerebral vasculature is exquisitely sensitive to changes in
CO2, a phenomenon known as CO2 reactivity (Fig. 1.2). With
hyperventilation (and resultant hypocapnea) cerebral vaso-
constriction ensues and CBF decreases, whereas with hyper-
capnea the reverse occurs.16 The effect can be profound with
an increase or decrease in CBF of approximately 4% for a
corresponding change in PaCO2 of 1 mm Hg.17 Autoregula-
tion is fundamentally different from CO2 reactivity. In both
metabolic and pressure autoregulation, vessel diameter
changes are compensatory responses to maintain a constant
AVDO2. In CO2 reactivity the diameter changes are primary
and not related to metabolic needs such that CBF and AVDO2
follow passively.2
Cerebral Oxygen Tension
There are many factors that influence brain oxygen tension
(PbtO2). Under physiologic conditions, PbtO2 has a close rela-
tionship with arterial oxygen tension (PaO2). In acute brain
injury PbtO2 appears to depend on the interaction between
plasma oxygen tension and CBF: PbtO2 = CBF × AVTO2.18
Because oxygen is consumed in the tissue, PbtO2 can be con-
sidered as a continuum that can vary from approximately
90 mm Hg close to the arteriolar side of capillaries to approxi-
mately 35 mm Hg in the more distal regions.19 Reductions of
partial pressure of arterial PaO2 can alter CBF to compensate
for reduced oxygen delivery. This may depend also in part on
hemoglobin concentration among other factors but once PaO2
is less than 65 mm Hg the ability to perform complex tasks is
impaired in humans. Loss of consciousness occurs when PaO2
is equal to or less than 30 mm Hg.20 Observational clinical
studies and microdialysis studies suggest that cerebral infarc-
tion may occur in a time-dependent fashion when PbtO2 is less
than 10 to 15 mm Hg.21-28
125 PaCO2
100
Cerebral blood flow
(mL/100 g/min)
Mean arterial
75 blood pressure
50 PaO2
25
25 50 75 100 125 150 175
Pressure (mm Hg)
Fig. 1.2  Comparison of the response of cerebral blood flow to mean
arterial blood pressure, arterial oxygen tension (PaO2), and arterial carbon
dioxide tension (PaCO2). ICP, Intracranial pressure. (From Lee KR, Hoff JT.
Intracranial pressure. In: Youmans JR, editor. Neurological surgery: a comprehensive
reference guide to the diagnosis and management of neurosurgical problems.
Philadelphia: WB Saunders; 1996. p. 510, Fig. 20-14.)
Intracranial Pressure and
Cerebral Hemodynamics
The Monro-Kellie hypothesis states that the brain, its vascular
network, and cerebrospinal fluid (CSF) are contained within
a rigid bony skull and membranous dura mater with a fixed
volume and so has a limited potential for expansion. ICP
reflects the volume of the three compartments. Increases in
any one component are compensated for by removal of an
equivalent amount of another; if not ICP will increase. The
adult intracranial space is composed of approximately 87%
brain, 9% CSF, and 4% blood.29 These compartments allow
fluid and chemical shifts under normal as well as pathologic
conditions (Fig. 1.3). Normally, no net change occurs in brain
water or sodium content because the influx into the brain
from the blood vessels is matched closely by extracellular bulk
flow into the CSF.30
An increase in ICP can affect neurologic function through
its effects on cerebral blood flow.31-33 However, there also are
circumstances in which increased flow (hyperemia) can cause
increased ICP. The pressures in arteries of the subarachnoid
space are estimated to be about the same as the systemic arte-
rial pressures.1 Normally cerebral venous pressure is approxi-
mately the same as ICP because the pressure is transmitted
through the subarachnoid space to compliant veins. However,
when the ICP is pathologically elevated, it can become dissoci-
ated from CVP. CBV is a consequence of arterial inflow,
venous drainage, and cerebrovascular tone. Cerebral vasodila-
tion increases cerebral blood volume, whereas vasoconstric-
tion decreases blood volume. Manipulation of these factors
can be taken advantage of to treat elevated ICP, but to be used
Elastance
Low High
C
High
Pressure
intracranial pressure
A B
Increasing
Low
Increasing volume
Fig. 1.3  Elastance curve showing the region of low intracranial pressure
and low elastance (A), low intracranial pressure and high elastance (B),
and high intracranial pressure and high elastance (C). Region A describes
a state of normal intracranial pressure and a “safe” amount of volume
buffering capacity. Region B includes patients with normal intracranial
pressure with a dangerously low amount of volume reserve. Region C is
clearly abnormal and readily diagnosed by the level of intracranial
pressure. (From Lee KR, Hoff JT. Intracranial pressure. In: Youmans JR, editor.
Neurological surgery: a comprehensive reference guide to the diagnosis and
management of neurosurgical problems. Philadelphia: WB Saunders; 1996. p. 503,
Fig. 20-8.)

correctly requires knowledge of autoregulation, and the effect
may depend in part on the autoregulatory state.34
CBV can vary between 13 and 51 mL in the microcircula-
tion.2 The pressure volume index (PVI) can be used to
understand what these volume differences mean to ICP (see
also Chapter 34). The PVI is defined as the fluid volume
change (in mL) in the intracranial space that would produce
a 10-fold increase in ICP [PVI = V/loge(Po/Pm)].35 Normal
PVI is 20 to 25 mL and reflects intracranial elastance.36 Because
maximal constriction can reduce blood volume by almost
40 mL, this equation helps explain how going from normo-
capnea to hypocapnea can reduce ICP.2,37 This may not,
however, always be desired because the effect of vasoconstric-
tion on blood vessel diameter is raised to the fourth power
(Poiseuille’s equation) and so can diminish CBF.7
Cerebral Metabolism
The brain only comprises 2% to 3% of the total body weight,
yet uses up to 20% of energy generated. Approximately 50%
of the energy produced by the brain is for synaptic activity,
25% is used for restoring ionic gradients across the cell mem-
brane, and the remaining energy is spent on biosynthesis.2 The
majority of energy is consumed by neurons. Although glial
cells account for almost half of the brain volume, they have a
much lower metabolic rate and account for less than 10% of
total cerebral energy consumption.38 The brain is metaboli-
cally unique in its use of fuel relative to other organs because
it lacks the ability to store fuel and thus requires a constant
supply of energy metabolites.39 Brain energy metabolism in
humans with acute brain injury is complex.40-43 Although
glucose is the major substrate, a large body of evidence sug-
gests that endogenous lactate, produced by aerobic glycolysis,
also can be an important substrate for neurons.44 Further-
more, there is a glucose lactate shuttle between astrocytes and
neurons.
Cerebral Energy Production
High-energy phosphates, such as adenosine triphosphate
(ATP), are the main energy substrates in the brain. Under
normal physiologic conditions, a constant supply of these sub-
strates is made possible through aerobic metabolism of
glucose. In the cell, the complete oxidation of 1 mole of
glucose is associated with the formation of 38 moles of ATP,
according to the following equation:
Glucose + 6 O2 + 38 ADP + 38 Pi → 6 CO2 + 44 H2O + 38 ATP
The above equation is a summary of many individual reac-
tions. The first of these reactions, glycolysis, occurs in the
cytoplasm. Glucose is initially phosphorylated to glucose-6-
phosphate and then through a series of reactions is converted
into pyruvate. Once pyruvate has been synthesized it can
either be reversibly converted to lactate or in the presence of
oxygen, enter the citric acid cycle. The citric acid or Krebs cycle
represents a cyclic series of reactions that occur in the mito-
chondria and achieve the complete oxidation of pyruvate to
carbon dioxide and water. The Krebs cycle is the common
pathway for the oxidation of fuel molecules and also serves as
a source of building blocks for biosynthesis. The electron
transport chain is the final step in the production of ATP from
the NADH and FADH2 formed during glycolysis, fatty acid
Section I—Background 5
oxidation, and the citric acid cycle. The electrons are donated
to molecular oxygen, resulting in a large amount of free
energy, which is used to generate ATP.45
Coupling of Cerebral Metabolism
and Blood Flow
There is a relationship between cerebral metabolism and CBF;
that is to say, blood flow is determined by regional energy
demands. Strong evidence of tight coupling between local
metabolism and flow has been obtained with in vivo radioac-
tive deoxyglucose methods.46,47 Many local metabolites have
been implicated in this coupling. Among them, adenosine
appears to provide a consistent link between flow and metabo-
lism.48 This substance is a potent cerebral vasodilator and a
product of dephosphorylation of adenosine monophosphate
that accumulates during increased neuronal activity. This
“metabolic theory” of coupling has been challenged,49 and it
is proposed that there is an intrinsic neural system capable of
influencing CBF independent of changes in metabolism. Evi-
dence for this hypothesis comes from examples of dispropor-
tionate increases in flow compared to metabolism during
manipulation of cerebellar and brainstem centers.49,50 It seems
reasonable that many neurotransmitters can overcome meta-
bolic regulation and exert a primary effect on blood flow, or
can directly influence metabolism and regulate blood flow in
kind.49-51
Cerebral Hemodynamic
and Metabolic Response
to Neurologic Injury
Although a detailed discussion of the hemodynamic and met-
abolic responses of the brain in various pathologic conditions
is beyond the scope of this introduction, a basic understand-
ing will provide a contextual foundation upon which to
understand the indications for, and potential benefits of, mon-
itoring in neurocritical care.
Primary and Secondary Injury
The pathophysiology of brain injury can be thought of in
terms of primary and secondary events. Primary brain injury
is the physical brain injury sustained at the moment of impact
in traumatic brain injury (TBI) or at the time of an insult, for
example, aneurysm rupture or reduced flow with ischemic
stroke, whereas secondary injury are events that occur at any
later stage. Primary injuries generally are believed to be imme-
diate and irreversible, but some mechanisms of cell death
from primary injury may occur over hours, suggesting the
possibility of reversibility.52,53 Secondary insults consist of a
wide range of ischemic, metabolic, and immunologic insults
often initiated by the primary event that can occur at a lower
threshold in a susceptible brain and lead to further brain
damage. The etiology of theses secondary events is diverse and
can include systemic and intracranial phenomena such as
hypotension, hypoxemia, intracranial hypertension, edema, or
seizures.53-56 In addition, secondary events may be iatrogenic
and cumulative. Secondary injuries of some form occur in
nearly all patients treated in a neurointensive care unit. It is
6 Section I—Background
the goal of monitoring in neurocritical care to detect poten-
tially harmful pathophysiologic events before they cause irre-
versible secondary brain damage, and so allow effective
treatment and improved patient outcomes.53,57-59
Ischemia
Ischemia is defined as a decrease in blood flow below the level
necessary to sustain normal cell structure and function. Isch-
emia can be global, as in cardiac arrest or severe oligemia as
may be seen with intracranial hypertension, or focal as in
occlusion of an intracranial vessel by embolism or thrombus.
Focal ischemia does not necessarily lead to irreversible isch-
emic damage in the entire area of reduced perfusion because
of collateral blood flow. The central pathophysiology of cere-
bral ischemia is energy failure; that is to say, when metabolic
demands are not satisfied, there is a rapid decrease in high-
energy intermediates, and anaerobic glycolysis is stimulated.
The loss of an adequate energy supply impairs membrane-
bound ionic pumps, causing a loss of ionic homeostasis and
a dramatic rise in the intracellular concentration of calcium,
an event of major pathophysiologic importance through its
activation of several deleterious enzyme systems and intracel-
lular signaling pathways.1,60
Conceptually there are two distinct areas of hemodynamic
and metabolic functioning in focal ischemia: (1) a central
densely ischemic core destined to die, and (2) a collateral
area composed of viable cells at risk of ischemic damage,
known as the penumbra.60,61 Blood flow within the penum-
bra lies in a range that is between the threshold for energy
failure and complete infarction. The ultimate survival of
penumbral cells depends on restoration of adequate blood
flow, which leads to the concept of a flow-dependent thera-
peutic time window.1
Ischemia commonly impairs the normal regulatory
responses of the cerebral circulation and can lead to a state of
vasomotor paralysis, which is manifest by a completely passive
blood flow response to changes in perfusion pressure. A major
factor thought to impair autoregulatory mechanisms in isch-
emia is reduced tissue pH secondary to lactate accumulation.
Positron emission tomography studies measure cerebral
hemodynamics and have provided valuable information about
the events that take place during ischemia. The first alteration
is a decrease in the ratio of CBF to CBV secondary to mainte-
nance of blood flow through autoregulatory vasodilation.
When this mechanism is exhausted, oxygen extraction increases
in hypoperfused areas. If this increased oxygen extraction
cannot maintain an adequate oxygen supply, then CMRO2
declines and the threshold for an infarct will be reached.1,62 In
part the goal of monitoring is to help define the penumbra or
detect pathologic changes while they still are reversible.
Traumatic Brain Injury
TBI commonly is associated with disturbed cerebral hemo-
dynamics, which is particularly detrimental given the vulner-
ability of injured brain to secondary insults.1,63,64 Secondary
cerebral ischemia is very common after severe TBI, and can
be associated with an unfavorable outcome.65,66 However, not
all secondary injury after TBI is ischemic in nature. Follow-
ing trauma, autoregulatory mechanisms can fail and lead
to a detrimental uncoupling of CBF and CMRO2. Cerebral
autoregulation can be impaired even in mild TBI, and in the
setting of normal ICP and MAP values.67 It is unknown why
autoregulatory mechanisms are so vulnerable to injury, but
disturbed autoregulation may have a variable time course
that various monitoring techniques may help detect.68-70
Although pressure autoregulation commonly is disturbed
following TBI, CO2 vasoreactivity may be preserved. Dis-
turbed CO2 reactivity, however, is usually seen in patients
with poor neurologic status.1,68,71 Defective autoregulation is
often associated with poor outcome, and in some studies
focally altered autoregulation may have a greater impact on
poor outcome than globally altered autoregulation.71,72
Xenon blood flow studies demonstrate that cerebral isch-
emia occurs early after TBI and often is more severe in patients
with intracranial hematomas and those with diffuse edema.73
It has been proposed that focal ischemia may be due to
increased ICP that compresses small parenchymal arteries or
results from increased diffusion distance for oxygen from the
microvasculature secondary to tissue swelling and cytotoxic
edema.32 In these situations the brain may require higher
tissue oxygen tensions to maintain sufficient tissue oxygen-
ation.19 An early reduction in CBF often is followed by a
period of hyperemia; that is to say, blood flow is in excess of
cellular needs.69,74,75 Hyperemia then can be a significant con-
tributor to increased ICP elevations following TBI in some
patients, and can contribute to or aggravate diffuse cerebral
edema (i.e., there can be two CBF patterns associated with
ICP: oligemic or hyperemic).1
Aneurysmal Subarachnoid Hemorrhage
Immediately following aneurysm rupture, global reductions
in CBF and even circulatory arrest can be observed, often
proportional to the clinical severity of the SAH.76,77 Patients in
poor neurologic condition frequently develop intracranial
hypertension and vasospasm. The relationship between clini-
cal grade and CBF generally remains even in the absence of
increased ICP.1 It is possible that the initial flow reduction is
related to direct metabolic depression from blood or its
by-products.78,79 Associated with a decrease in CBF, many
patients have a reduction in CMRO2 and CBV that can reduce
cerebral perfusion.37,78,80-83 There often is a progressive decrease
in CBF during the first 14 days after hemorrhage, after which
it usually increases.84 Cerebral infarction is one of the princi-
pal causes of death and disability in patients who survive the
initial hemorrhage and have their aneurysm secured.84,85 Arte-
rial vasospasm is considered the most common cause of isch-
emia; particularly delayed cerebral ischemia following SAH
and its severity is associated with the development of ischemic
deficits. However, there are many other causes for delayed
infarction including intracranial hypertension, cerebral
edema, microthrombosis, and systemic hypotension among
others. Both pressure and CO2 autoregulation are impaired
after SAH, particularly in the acute period. Disturbances in
autoregulation are worse with poor clinical grade and the
occurrence of vasospasm.
Conclusion
A thorough understanding of the basic principles that govern
cerebral blood flow and metabolism is essential for clinicians
who care for patients with critical neurologic disease. It is only

when equipped with this knowledge that we can fully take
advantage of neuromonitoring to attenuate secondary pro-
cesses leading to worse brain damage. It further stands to
reason that information learned from multimodality neuro-
monitoring may alter and enhance our understanding of the
fundamental mechanisms of cerebral hemodynamic and met-
abolic regulation.
References
1. Dobserstein C, Martin NA. Cerebral blood flow in clinical neurosurgery. In:
Winn RH, editor. Youmans neurological surgery. 5th ed. Philadelphia: WB
Saunders; 2004. pp. 519–69.
2. Verweij BH, Amelink GJ, Muizelaar JP. Current concepts of cerebral oxygen
transport and energy metabolism after severe traumatic brain injury. Prog
Brain Res 2007;161:111–24.
3. Chien S. Shear dependence of effective cell volume as a determinant of
blood viscosity. Science 1970;168:977–9.
4. Albrecht KH, Gaehtgens P, Pries A, et al. The Fahraeus effect in narrow
capillaries (i.d. 3.3 to 11.0 micron). Microvasc Res 1979;18:33–47.
5. Kontos HA, Raper AJ, Patterson JL. Analysis of vasoactivity of local pH,
PaCO2 and bicarbonate on pial vessels. Stroke 1977;8:358–60.
6. Sokoloff L. Metabolism of the central nervous system in vivo. In: Field J,
Magoun HW, Hall VE, editors. Handbook of physiology. Washington, DC:
American Physiological Society; 1960. pp. 1843–64.
7. Jones TH, Morawetz RB, Crowell RM, et al. Thresholds of focal cerebral
ischemia in awake monkeys. J Neurosurg 1981;54:773–82.
8. Schroder ML, Muizelaar JP, Kuta AJ, et al. Thresholds for cerebral ischemia
after severe head injury: relationship with late CT findings and outcome.
J Neurotrauma 1996;13:17–23.
9. Celsis P, Chan M, Marc-Vergnes JP, et al. Measurement of cerebral
circulation time in man. Eur J Nucl Med 1985;10:426–31.
10. Ferrari M, Wilson DA, Hanley DF, et al. Effects of graded hypotension on
cerebral blood flow, blood volume, and mean transit time in dogs. Am J
Physiol 1992;262(6 Pt 2):H1908–14.
11. Rosengarten B, Rüskes D, Mendes I, et al. A sudden arterial blood pressure
decrease is compensated by an increase in intracranial blood volume.
J Neurol 2002;249(5):538–41.
12. Singer OC, de Rochemont Rdu M, Foerch C, et al. Relation between relative
cerebral blood flow, relative cerebral blood volume, and mean transit time in
patients with acute ischemic stroke determined by perfusion-weighted MRI.
J Cereb Blood Flow Metab 2003;23(5):605–11.
13. Nordström CH. Physiological and biochemical principles underlying
volume-targeted therapy—the “Lund concept.” Neurocrit Care
2005;2(1):83–95.
14. Muizelaar JP, Lutz HA, 3rd, Becker DP. Effect of mannitol on ICP and CBF
and correlation with pressure autoregulation in severely head-injured
patients. J Neurosurg 1984;61:700–6.
15. Mitagvaria NP. Regulation of local cerebral blood flow. Adv Exp Med Biol
1984;180:861–79.
16. Muizelaar JP, Marmarou A, Ward JD, et al. Adverse effects of prolonged
hyperventilation in patients with severe head injury: a randomized clinical
trial. J Neurosurg 1991;75:731–9.
17. Rangel-Castilla L, Gasco J, Nauta HJ, et al. Cerebral pressure autoregulation
in traumatic brain injury. Neurosurg Focus 2008;25:E7.
18. Rosenthal G, Hemphill JC III, Sorani M, et al. Brain tissue oxygen tension is
more indicative of oxygen diffusion than oxygen delivery and metabolism in
patients with traumatic brain injury. Crit Care Med 2008;36:1917–24
19. Zauner A, Daugherty WP, Bullock MR, et al. Brain oxygenation and energy
metabolism: part I-biological function and pathophysiology. Neurosurgery
2002;51:289–301; discussion 2.
20. Siesjo B. Brain energy metabolism. Chichester, UK: Wiley; 1978.
21. Maloney-Wilensky E, Le Roux P. The physiology behind direct brain oxygen
monitors and practical aspects of their use. Childs Nerv Syst
2010;26(4):419–30.
22. Scheufler K-M, Lehnert A, Rohrborn H-J, et al. Individual values of brain
tissue oxygen pressure, microvascular oxygen saturation, cytochrome redox
level and energy metabolites in detecting critically reduced cerebral energy
state during acute changes in global cerebral perfusion. J Neurosurg
Anesthesiol 2004;16:210–19
23. Valadka AB, Goodman JC, Gopinath SP, et al. Comparison of brain tissue
oxygen tension to microdialysis-based measures of cerebral ischemia in
fatally head-injured humans. J Neurotrauma 1998;15(7):509–19.
Section I—Background 7
24. Valadka AB, Hlatky R, Furuya Y, et al. Brain tissue PO2: correlation with
cerebral blood flow. Acta Neurochir Suppl 2002;81:299–301.
25. Maas AI, Fleckenstein W, de Jong DA, et al. Monitoring cerebral oxygenation:
experimental studies and preliminary clinical results of continuous
monitoring of cerebrospinal fluid and brain tissue oxygen tension. Acta
Neurochir Suppl (Wien) 1993;59:50–7.
26. Hlatky R, Valadka AB, Goodman JC, et al. Patterns of energy substrates
during ischemia measured in the brain by microdialysis. J Neurotrauma
2004;21(7):894–906.
27. Hlatky R, Valadka AB, Goodman JC, et al. Evolution of brain tissue injury
after evacuation of acute traumatic subdural hematomas. Neurosurgery
2007;61(1 Suppl):249–54; discussion 254–5.
28. Doppenberg EM, Zauner A, Watson JC, et al. Determination of the ischemic
threshold for brain oxygen tension. Acta Neurochir Suppl 1998;71:166–9.
29. Plum F, Posner JB. The diagnosis of stupor and coma. Philadelphia: FA
Davis; 1980.
30. Cserr HF, Knopf PM. Cervical lymphatics, the blood-brain barrier and the
immunoreactivity of the brain: a new view. Immunol Today 1992;13:507–12.
31. Lee KR, Hoff JT. Intracranial pressure. In: Youmans JR, editor. Neurological
surgery: a comprehensive reference guide to the diagnosis and management
of neurosurgical problems. 4th ed. Philadelphia: WB Saunders; 1996.
32. Lindenberg R. Compression of the brain arteries as pathogenetic factor for
tissue necroses and their areas of predilection. J Neuropathol Exper Neurol
1955;14:223–43.
33. Miller J. The pathophysiology of raised intracranial pressure. In: Greenfield
JG, Adams JH, Corsellis JAN, et al, editors. Greenfield’s neuropathology.
London: Edward Arnold; 1984. pp. 53–84.
34. Huseby JS, Pavlin EG, Butler J. Effect of positive end-expiratory pressure on
intracranial pressure in dogs. J Appl Physiol 1978;44(1):25–7.
35. Marmarou A, Shulman K, Rosende RM. A nonlinear analysis of the
cerebrospinal fluid system and intracranial pressure dynamics. J Neurosurg
1978;48:332–44.
36. Shapiro K, Marmarou A, Shulman K. Characterization of clinical CSF
dynamics and neural axis compliance using the pressure-volume index. I.
The normal pressure-volume index. Ann Neurol 1980;7:508–14.
37. Bouma GJ, Muizelaar JP, Bandoh K, et al. Blood pressure and intracranial
pressure-volume dynamics in severe head injury: relationship with cerebral
blood flow. J Neurosurg 1992;77:15–9.
38. Siesjo BK. Cerebral circulation and metabolism. J Neurosurg
1984;60:883–908.
39. Guyton A. Textbook of medical physiology. Philadelphia: WB Saunders;
1986.
40. Vespa P, Bergsneider M, Hattori N, et al. Metabolic crisis without brain
ischemia is common after traumatic brain injury: a combined microdialysis
and positron emission tomography study. J Cereb Blood Flow Metab
2005;25(6):763–74.
41. Pellerin L. How astrocytes feed hungry neurons. Mol Neurobiol
2005;32(1):59–72.
42. Schurr A, Miller JJ, Payne RS, et al. An increase in lactate output by brain
tissue serves to meet the energy needs of glutamate-activated neurons.
J Neurosci 1999;19(1):34–9.
43. Pellerin L, Magistretti PJ. How to balance the brain energy budget while
spending glucose differently. J Physiol 2003;546(Pt 2):325.
44. Gallagher CN, Carpenter KL, Grice P, et al. The human brain utilizes lactate
via the tricarboxylic acid cycle: a 13C-labelled microdialysis and high-
resolution nuclear magnetic resonance study. Brain 2009;132(Pt 10):2839–
49. Epub 2009 Aug 20.
45. Pasqualin A. Cerebral metabolism. In: Youmans JR, editor. Neurological
surgery: a comprehensive guide to the diagnosis and management of
neurosurgical problems. 4th ed. Philadelphia: WB Saunders; 1996.
46. Sokoloff L. Relationships among local functional activity, energy metabolism,
and blood flow in the central nervous system. Fed Proc 1981;40:2311–6.
47. Sokoloff L. Localization of functional activity in the central nervous system
by measurement of glucose utilization with radioactive deoxyglucose.
J Cereb Blood Flow Metab 1981;1:7–36.
48. Wahl M, Kuschinsky W. The dilatatory action of adenosine on pial arteries
of cats and its inhibition by theophylline. Pflugers Arch 1976;362:55–9.
49. Edvinsson L, MacKenzie ET, McCulloch J. Cerebral blood flow and
metabolism. New York: Raven Press; 1993.
50. Bonvento G, Lacombe P, Seylaz J. Effects of electrical stimulation of the
dorsal raphe nucleus on local cerebral blood flow in the rat. J Cereb Blood
Flow Metab 1989;9:251–5.
A complete list of references for this chapter can be found online at
www.expertconsult.com.

References
1. Dobserstein C, Martin NA. Cerebral blood flow in clinical neurosurgery. In:
Winn RH, editor. Youmans neurological surgery. 5th ed. Philadelphia: WB
Saunders; 2004. pp. 519–69.
2. Verweij BH, Amelink GJ, Muizelaar JP. Current concepts of cerebral oxygen
transport and energy metabolism after severe traumatic brain injury. Prog
Brain Res 2007;161:111–24.
3. Chien S. Shear dependence of effective cell volume as a determinant of
blood viscosity. Science 1970;168:977–9.
4. Albrecht KH, Gaehtgens P, Pries A, et al. The Fahraeus effect in narrow
capillaries (i.d. 3.3 to 11.0 micron). Microvasc Res 1979;18:33–47.
5. Kontos HA, Raper AJ, Patterson JL. Analysis of vasoactivity of local pH,
PaCO2 and bicarbonate on pial vessels. Stroke 1977;8:358–60.
6. Sokoloff L. Metabolism of the central nervous system in vivo. In: Field J,
Magoun HW, Hall VE, editors. Handbook of physiology. Washington, DC:
American Physiological Society; 1960. pp. 1843–64.
7. Jones TH, Morawetz RB, Crowell RM, et al. Thresholds of focal cerebral
ischemia in awake monkeys. J Neurosurg 1981;54:773–82.
8. Schroder ML, Muizelaar JP, Kuta AJ, et al. Thresholds for cerebral ischemia
after severe head injury: relationship with late CT findings and outcome.
J Neurotrauma 1996;13:17–23.
9. Celsis P, Chan M, Marc-Vergnes JP, et al. Measurement of cerebral
circulation time in man. Eur J Nucl Med 1985;10:426–31.
10. Ferrari M, Wilson DA, Hanley DF, et al. Effects of graded hypotension on
cerebral blood flow, blood volume, and mean transit time in dogs. Am J
Physiol 1992;262(6 Pt 2):H1908–14.
11. Rosengarten B, Rüskes D, Mendes I, et al. A sudden arterial blood pressure
decrease is compensated by an increase in intracranial blood volume.
J Neurol 2002;249(5):538–41.
12. Singer OC, de Rochemont Rdu M, Foerch C, et al. Relation between relative
cerebral blood flow, relative cerebral blood volume, and mean transit time in
patients with acute ischemic stroke determined by perfusion-weighted MRI.
J Cereb Blood Flow Metab 2003;23(5):605–11.
13. Nordström CH. Physiological and biochemical principles underlying
volume-targeted therapy—the “Lund concept.” Neurocrit Care 2005;2(1):
83–95.
14. Muizelaar JP, Lutz HA, 3rd, Becker DP. Effect of mannitol on ICP and CBF
and correlation with pressure autoregulation in severely head-injured
patients. J Neurosurg 1984;61:700–6.
15. Mitagvaria NP. Regulation of local cerebral blood flow. Adv Exp Med Biol
1984;180:861–79.
16. Muizelaar JP, Marmarou A, Ward JD, et al. Adverse effects of prolonged
hyperventilation in patients with severe head injury: a randomized clinical
trial. J Neurosurg 1991;75:731–9.
17. Rangel-Castilla L, Gasco J, Nauta HJ, et al. Cerebral pressure autoregulation
in traumatic brain injury. Neurosurg Focus 2008;25:E7.
18. Rosenthal G, Hemphill JC III, Sorani M, et al. Brain tissue oxygen
tension is more indicative of oxygen diffusion than oxygen delivery and
metabolism in patients with traumatic brain injury. Crit Care Med 2008;36:
1917–24.
19. Zauner A, Daugherty WP, Bullock MR, et al. Brain oxygenation and energy
metabolism: part I-biological function and pathophysiology. Neurosurgery
2002;51:289–301; discussion 2.
20. Siesjo B. Brain energy metabolism. Chichester, UK: Wiley; 1978.
21. Maloney-Wilensky E, Le Roux P. The physiology behind direct brain oxygen
monitors and practical aspects of their use. Childs Nerv Syst
2010;26(4):419–30.
22. Scheufler K-M, Lehnert A, Rohrborn H-J, et al. Individual values of brain
tissue oxygen pressure, microvascular oxygen saturation, cytochrome redox
level and energy metabolites in detecting critically reduced cerebral energy
state during acute changes in global cerebral perfusion. J Neurosurg
Anesthesiol 2004;16:210–19
23. Valadka AB, Goodman JC, Gopinath SP, et al. Comparison of brain tissue
oxygen tension to microdialysis-based measures of cerebral ischemia in
fatally head-injured humans. J Neurotrauma 1998;15(7):509–19.
24. Valadka AB, Hlatky R, Furuya Y, et al. Brain tissue PO2: correlation with
cerebral blood flow. Acta Neurochir Suppl 2002;81:299–301.
25. Maas AI, Fleckenstein W, de Jong DA, et al. Monitoring cerebral oxygenation:
experimental studies and preliminary clinical results of continuous
monitoring of cerebrospinal fluid and brain tissue oxygen tension. Acta
Neurochir Suppl (Wien) 1993;59:50–7.
26. Hlatky R, Valadka AB, Goodman JC, et al. Patterns of energy substrates
during ischemia measured in the brain by microdialysis. J Neurotrauma
2004;21(7):894–906.
Section I—Background 7.e1
27. Hlatky R, Valadka AB, Goodman JC, et al. Evolution of brain tissue injury
after evacuation of acute traumatic subdural hematomas. Neurosurgery
2007;61(1 Suppl):249–54; discussion 254–5.
28. Doppenberg EM, Zauner A, Watson JC, et al. Determination of the ischemic
threshold for brain oxygen tension. Acta Neurochir Suppl 1998;71:166–9.
29. Plum F, Posner JB. The diagnosis of stupor and coma. Philadelphia: FA
Davis; 1980.
30. Cserr HF, Knopf PM. Cervical lymphatics, the blood-brain barrier and the
immunoreactivity of the brain: a new view. Immunol Today 1992;13:507–12.
31. Lee KR, Hoff JT. Intracranial pressure. In: Youmans JR, editor. Neurological
surgery: a comprehensive reference guide to the diagnosis and management
of neurosurgical problems. 4th ed. Philadelphia: WB Saunders; 1996.
32. Lindenberg R. Compression of the brain arteries as pathogenetic factor for
tissue necroses and their areas of predilection. J Neuropathol Exper Neurol
1955;14:223–43.
33. Miller J. The pathophysiology of raised intracranial pressure. In: Greenfield
JG, Adams JH, Corsellis JAN, et al, editors. Greenfield’s neuropathology.
London: Edward Arnold; 1984. pp. 53–84.
34. Huseby JS, Pavlin EG, Butler J. Effect of positive end-expiratory pressure on
intracranial pressure in dogs. J Appl Physiol 1978;44(1):25–7.
35. Marmarou A, Shulman K, Rosende RM. A nonlinear analysis of the
cerebrospinal fluid system and intracranial pressure dynamics. J Neurosurg
1978;48:332–44.
36. Shapiro K, Marmarou A, Shulman K. Characterization of clinical CSF
dynamics and neural axis compliance using the pressure-volume index. I.
The normal pressure-volume index. Ann Neurol 1980;7:508–14.
37. Bouma GJ, Muizelaar JP, Bandoh K, et al. Blood pressure and intracranial
pressure-volume dynamics in severe head injury: relationship with cerebral
blood flow. J Neurosurg 1992;77:15–9.
38. Siesjo BK. Cerebral circulation and metabolism. J Neurosurg
1984;60:883–908.
39. Guyton A. Textbook of medical physiology. Philadelphia: WB Saunders; 1986.
40. Vespa P, Bergsneider M, Hattori N, et al. Metabolic crisis without brain
ischemia is common after traumatic brain injury: a combined microdialysis
and positron emission tomography study. J Cereb Blood Flow Metab
2005;25(6):763–74.
41. Pellerin L. How astrocytes feed hungry neurons. Mol Neurobiol
2005;32(1):59–72.
42. Schurr A, Miller JJ, Payne RS, et al. An increase in lactate output by brain
tissue serves to meet the energy needs of glutamate-activated neurons.
J Neurosci 1999;19(1):34–9.
43. Pellerin L, Magistretti PJ. How to balance the brain energy budget while
spending glucose differently. J Physiol 2003;546(Pt 2):325.
44. Gallagher CN, Carpenter KL, Grice P, et al. The human brain utilizes lactate
via the tricarboxylic acid cycle: a 13C-labelled microdialysis and high-
resolution nuclear magnetic resonance study. Brain 2009;132(Pt 10):2839–
49. Epub 2009 Aug 20.
45. Pasqualin A. Cerebral metabolism. In: Youmans JR, editor. Neurological
surgery: a comprehensive guide to the diagnosis and management of
neurosurgical problems. 4th ed. Philadelphia: WB Saunders; 1996.
46. Sokoloff L. Relationships among local functional activity, energy metabolism,
and blood flow in the central nervous system. Fed Proc 1981;40:2311–6.
47. Sokoloff L. Localization of functional activity in the central nervous system
by measurement of glucose utilization with radioactive deoxyglucose.
J Cereb Blood Flow Metab 1981;1:7–36.
48. Wahl M, Kuschinsky W. The dilatatory action of adenosine on pial arteries
of cats and its inhibition by theophylline. Pflugers Arch 1976;362:55–9.
49. Edvinsson L, MacKenzie ET, McCulloch J. Cerebral blood flow and
metabolism. New York: Raven Press; 1993.
50. Bonvento G, Lacombe P, Seylaz J. Effects of electrical stimulation of the
dorsal raphe nucleus on local cerebral blood flow in the rat. J Cereb Blood
Flow Metab 1989;9:251–5.
51. Tuor UI, Edvinsson L, McCulloch J. Catecholamines and the relationship
between cerebral blood flow and glucose use. Am J Physiol
1986;251:H824–33.
52. Reilly PL. Brain injury: the pathophysiology of the first hours. “Talk and die
revisited.” J Clin Neurosci 2001;8:398–403.
53. Tisdall MM, Smith M. Multimodal monitoring in traumatic brain injury:
current status and future directions. Br J Anaesth 2007;99:61–7.
54. Chesnut RM. Secondary brain insults after head injury: clinical perspectives.
New Horiz 1995;3:366–75.
55. Choi DW. Ionic dependence of glutamate neurotoxicity. J Neurosci
1987;7:369–79.
56. Unterberg AW, Stover J, Kress B, et al. Edema and brain trauma.
Neuroscience 2004;129:1021–9.
7.e2 Section I—Background
57. Fakhry SM, Trask AL, Waller MA, et al. Management of brain-injured
patients by an evidence-based medicine protocol improves outcomes and
decreases hospital charges. J Trauma 2004;56:492–9; discussion 9–500.
58. Jones PA, Andrews PJ, Midgley S, et al. Measuring the burden of secondary
insults in head-injured patients during intensive care. J Neurosurg
Anesthesiol 1994;6:4–14.
59. Patel HC, Menon DK, Tebbs S, et al. Specialist neurocritical care and
outcome from head injury. Intensive Care Med 2002;28:547–53.
60. Siesjo BK. Pathophysiology and treatment of focal cerebral ischemia. Part I:
Pathophysiology. J Neurosurg 1992;77:169–84.
61. Hakim AM. The cerebral ischemic penumbra. Can J Neurol Sci 1987;14:557–9.
62. Baron JC, Rougemont D, Soussaline F, et al. Local interrelationships of
cerebral oxygen consumption and glucose utilization in normal subjects and
in ischemic stroke patients: a positron tomography study. J Cereb Blood
Flow Metab 1984;4:140–9.
63. Hlatky R, Valadka AB, Robertson CS. Intracranial pressure response to
induced hypertension: role of dynamic pressure autoregulation.
Neurosurgery 2005;57:919–23.
64. Miller JD, Sweet RC, Narayan R, et al. Early insults to the injured brain.
JAMA 1978;240:439–42.
65. Bouma GJ, Muizelaar JP, Choi SC, et al. Cerebral circulation and metabolism
after severe traumatic brain injury: the elusive role of ischemia. J Neurosurg
1991;75:685–93.
66. Graham DI, Adams JH. Ischaemic brain damage in fatal head injuries.
Lancet 1971;1:265–6.
67. Junger EC, Newell DW, Grant GA, et al. Cerebral autoregulation following
minor head injury. J Neurosurg 1997;86:425–32.
68. Enevoldsen EM, Cold G, Jensen FT, et al. Dynamic changes in regional CBF,
intraventricular pressure, CSF pH and lactate levels during the acute phase
of head injury. J Neurosurg 1976;44:191–214.
69. Fieschi C, Battistini N, Beduschi A, et al. Regional cerebral blood flow and
intraventricular pressure in acute head injuries. J Neurol Neurosurg
Psychiatry 1974;37:1378–88.
70. Bouma GJ, Muizelaar JP. Cerebral blood flow, cerebral blood volume, and
cerebrovascular reactivity after severe head injury. J Neurotrauma 1992;9
Suppl 1:S333–48.
71. Obrist WD, Langfitt TW, Jaggi JL, et al. Cerebral blood flow and metabolism
in comatose patients with acute head injury. Relationship to intracranial
hypertension. J Neurosurg 1984;61:241–53.
72. Schmidt EA, Piechnik SK, Smielewski P, et al. Symmetry of cerebral
hemodynamic indices derived from bilateral transcranial Doppler.
J Neuroimaging 2003;13:248–54.
73. Bouma GJ, Muizelaar JP, Stringer WA, et al. Ultra-early evaluation of
regional cerebral blood flow in severely head-injured patients using
xenon-enhanced computerized tomography. J Neurosurg 1992;77:360–8.
74. Gobiet W, Grote W, Bock WJ. The relation between intracranial pressure,
mean arterial pressure and cerebral blood flow in patients with severe head
injury. Acta Neurochir (Wien) 1975;32:13–24.
75. Thomas DJ, du Boulay GH, Marshall J, et al. Cerebral blood-flow in
polycythaemia. Lancet 1977;2:161–3.
76. Eng CC, Lam AM, Byrd S, et al. The diagnosis and management of a
perianesthetic cerebral aneurysmal rupture aided with transcranial Doppler
ultrasonography. Anesthesiology 1993;78:191–4.
77. Kobayashi KI, Koike T. Cerebral blood flow and metabolism in patients with
ruptured aneurysms. Acta Neurol Scand Suppl 1979;60:292–3.
78. Fein JM. Cerebral energy metabolism after subarachnoid hemorrhage. Stroke
1975;6:1–8.
79. Gelmers HJ, Beks JW, Journee HL. Regional cerebral blood flow in
patients with subarachnoid haemorrhage. Acta Neurochir (Wien) 1979;47:
245–51.
80. Fein J, Boulos R. Local cerebral blood flow in experimental middle cerebral
artery vasospasm. J Neurosurg 1973;39:337–47.
81. Ishii R. Regional cerebral blood flow in patients with ruptured intracranial
aneurysms. J Neurosurg 1979;50:587–94.
82. Nornes H, Knutzen HB, Wikeby P. Cerebral arterial blood flow and
aneurysm surgery. Part 2: Induced hypotension and autoregulatory capacity.
J Neurosurg 1977;47:819–27.
83. Meyer CH, Lowe D, Meyer M, et al. Progressive change in cerebral blood
flow during the first three weeks after subarachnoid hemorrhage.
Neurosurgery 1983;12:58–76.
84. Sundt TM, Jr, Whisnant JP. Subarachnoid hemorrhage from intracranial
aneurysms. Surgical management and natural history of disease. N Engl J
Med 1978;299:116–22.
85. Adams HP, Jr, Kassell NF, Torner JC, et al. Early management of aneurysmal
subarachnoid hemorrhage. A report of the Cooperative Aneurysm Study.
J Neurosurg 1981;54:141–5.
 I
Chapter
2  
Why Monitor and Principles
of Neurocritical Care
Syed T. Arshad and Jose I. Suarez
Introduction
The collective burden of critical neurologic illnesses is over-
whelming. Stroke occurs in about 800,000 individuals each
year in the United States, with case fatality rates from acute
ischemic stroke (AIS) being close to 15% to 20%. Although
primary intracerebral hemorrhage (ICH) and aneurysmal
subarachnoid hemorrhage (SAH) are less common than isch-
emic stroke, their case fatality rates are nearly two to three
times higher than AIS.1 Traumatic brain injury (TBI), accord-
ing to the World Health Organization, will surpass many dis-
eases as the major cause of death and disability worldwide by
the year 2020.2 The number of patients with hypoxic-ischemic
encephalopathy following resuscitation is between 100,000
and 200,000 per year in the United States.3 The mortality in
this particular cohort of critically ill neurologic patients has
been decreasing steadily, in part because of advances in scien-
tific understanding of the clinical disorder but perhaps more
so because of the impact of neurologically oriented critical
care. In addition, patients with AIS have better outcomes
when admitted to a dedicated stroke/neurocritical care unit
rather than a general medical intensive care unit (ICU). In
addition, studies have shown that admission of ICH patients
to specialized neurocritical care units is associated with
improved clinical outcome. The institution of such dedicated
neurocritical care units has improved resource utilization,
introduced more efficient patient care protocols, established
neuroprotective measures, reduced the impact of comorbid
illness, and allowed for prevention and management of post-
stroke complications.4
Principles of Neurocritical Care
A Brief History of Neurocritical Care
Historians date the beginning of neurocritical care back to the
16th century, at the time resuscitation practices and attempts
at artificial ventilation first appeared. Modern neurocritical
care probably started with the polio epidemics and evolved
thereafter with the introduction of the iron lung in the early
to mid-20th century. At that time neurologists were the
primary treating physicians for these patients and probably
laid the groundwork for the first large-scale use of mechanical
ventilation. In 1923, a neurosurgeon, Dr. W.E. Dandy, created
a three-bed unit for postoperative neurosurgical patients.
8 © Copyright 2013 Elsevier Inc. All rights reserved.
World War II further expanded the field of intensive care with
the designation of “shock wards.”
The advent of polio vaccines and few important develop-
ments in neurosurgical techniques led to a somewhat a nihil-
istic approach to critically ill neurologic and neurosurgical
patients in the 1950s. By 1960, health care witnessed a rapid
spread of resuscitation and surgical techniques, including the
creation of recovery rooms. Toward the late 1960s almost 95%
of all acute care hospitals in the United States had some sort
of critical care unit. Advances in neuroanesthesia and the
organization of critical care protocols allowed for a more com-
prehensive care model for the critically ill. In the 1960s, tech-
niques to measure cerebral blood flow (CBF) and intracranial
pressure (ICP) were introduced and evolved. During the
1970s, neurologists David Jackson and Alan Ropper, and an
anesthesiologist, Sean Kennedy, spearheaded the concept of a
“neuro-ICU” in North America. Later pioneers included Dr.
Daniel Hanley and Dr. Thomas Bleck. In 2003, the Neurocriti-
cal Care Society was formed with Dr. Bleck as its first presi-
dent.5,6 The scope of neurocritical care has further evolved
with the advent of more sophisticated monitoring techniques
and informatics and computer systems and guidelines for
program requirements for neurocritical care training and a
core curriculum and competencies for training.7,8
Scope of Practice of the
Modern Neurointensivist
In day-to-day clinical practice, neurointensivists focus on
subtle changes in the neurologic exam and physiologic or
pathophysiologic interactions between the brain and other
organ systems. The main argument in favor of staffing
neuro-ICUs with full-time neurointensivists stems from the
notion that these individuals are specially trained to recog-
nize the specific interactions between the intracranial physi-
ology and systemic derangements of the neurocritically ill
patient.9 The members of a neurocritical care team are more
likely to be aware of secondary physiologic insults to the
brain that can include fever, hyperglycemia, anemia, hypo-
natremia, and delirium. A neurocritical care team typically
will care for patients with AIS, ICH, SAH, intracranial neo-
plasms, TBI and spinal cord injury, status epi­lepticus, neuro-
muscular respiratory failure, postoperative neurosurgical
care, hypoxic-ischemic brain injury after cardiac arrest,

postprocedure neurovascular care, and acutely ill medical
patients with neurologic injury requiring critical care, for
example, hepatic encephalopathy. Neurointensivists have
intimate knowledge of acute circulatory, respiratory, and
metabolic disturbances and general skills required for
advanced cardiac life support, cardioversion, and intubation
and ventilator management, and for the insertion of invasive
hemodynamic monitoring. In addition, given their knowl-
edge with nervous system function, neurointensivists have
significant experience in end-of-life questions, prognosis in
severe brain injury, and defining brain death. Consequently
neurointensivists lead the way at the interface for organ pro-
curement and donation.
Impact of Specialized
Neurocritical Care Team
Several observational studies suggest that the evolution
of neurocritical care units (NCCUs) has led to improved out-
comes and optimal utilization of resources for patients with
severe acute brain injury (Table 2.1).9-17 Mortality after TBI
has improved because of prehospital advanced life support
and specialized neurointensivist-led teams.10 In patients with
AIS and who are critically ill, admission to neurointensivist-
led NCCU is asso­ciated with improved condition at discharge
and shortened hospital length of stay.4,11 ICH traditionally has
carried the highest mortality rate and been the most disabling
form of stroke. Recent data suggest that those patients admit-
ted to the NCCU have improved outcomes, shorter lengths of
stay, and lower total cost of care compared with a national
benchmark.12,13 In addition, studies show that mortality
among patients who undergo surgery for ruptured intracra-
nial aneurysms is significantly reduced in high-volume centers
that provide access to specialized multidisciplinary care.14 In
addition, aggressive treatment through specialized care is
associated with improved outcome and proven to be cost
effective in poor-grade subarachnoid hemorrhage patients.15,16
Hypoxic-ischemic brain injury after cardiac arrest is one of
the most devastating conditions; historically only 15% of
patients successfully resuscitated in the field have survived to
hospital discharge. Studies have shown that induced therapeu-
tic hypothermia after cardiac arrest is associated with improved
survival and better functional outcomes.17,18 In many institu-
tions neurointensivists have become the ideal practitioners to
care for these critically ill patients. Studies also have shown
that availability of neurointensivists is associated with
improved documentation in the medical records and an
increased rate of organ and tissue donation after brain death
declarations.19,20
Physiologic Parameters in Neurocritical
Care Patients
The adult brain (1200-1400 g) comprises 2% to 3% of total
body weight and yet receives 15% to 20% of cardiac output.
However, it has a high cerebral metabolic rate of oxygen
(CMRO2) and uses glucose predominantly as a substrate for
its energy needs. Normal CBF in humans averages 50 mL/100  stimulate multidisciplinary efforts of scientists from diverse
g/min brain tissue per minute. Irreversible neuronal damage
occurs when CBF is less than 10 mL/100 g/min. The brain
has no significant storage capacity; hence cerebral metabo-
lism, CBF, and oxygen extraction are tightly coupled. This
Section I—Background 9
relationship is expressed by the Fick’s equation: CMRO2 = CBF
× AVDO2, in which AVDO2 represents arteriovenous differ-
ence of oxygen. Under normal conditions the brain maintains
a constant AVDO2 by responding to changes in metabolism,
cerebral perfusion pressure (CPP), and blood viscosity with
changes in vessel caliber, a phenomenon referred to as
autoregulation.21-23 Ischemia occurs if oxygen delivery is below
the metabolic demand of the tissue (despite the increased
AVDO2). Hyperemia, or “luxury perfusion,” occurs if oxygen
delivery is greater than the metabolic demand. These various
changes may not always manifest clinical features, and moni-
toring techniques can help detect and define these parameters
before irreversible injury occurs.
CPP, defined as mean arterial pressure (MAP) minus ICP,
is well maintained in healthy subjects. In pathologic states
there can be initial compensatory changes to maintain CPP;
for example, small increases in intracranial volume such as an
intracranial mass lesion can be accommodated for by translo-
cation of cerebrospinal fluid (CSF) into distensible spinal sub-
arachnoid space with little effect on ICP. This is expressed as
the Monro-Kellie doctrine. Exhaustion of this compensatory
mechanism can result in large increases in ICP, and so decrease
CPP or CBF. Uncorrected, this results in a vicious cycle of
further increases in ICP and decreases in CBF. Various moni-
tors can help define and detect this delicate balance.
A large body of evidence suggests that knowledge of normal
CBF physiology and alterations caused by specific diseases is
paramount in the treatment of critically ill patients with acute
brain injury, although not all cellular dysfunction results from
altered blood flow. CBF is equal to CPP divided by the cere-
brovascular resistance (CVR) (CBF = CPP/CVR). Measure-
ment of CBF in the critically ill can be technically difficult, so
CPP (i.e., ICP and MAP) is monitored as a surrogate for the
adequacy of CBF. Exactly what optimal CPP is remains
debated,24 but adequate values for CPP generally are consid-
ered to be between 55 and 100 mm Hg. This is a wide range,
and these values, however, may depend on the specific patient
and pathology and so emphasize a need for monitoring. In
addition, a normal CPP represents a normal CBF only if CVR
is normal. If CVR is high, a normal CPP still may be accom-
panied by ischemia. Alternatively, if CVR is low, a normal CPP
may be associated with hyperemia and increased ICP. Thus it
is important to maintain CPP in acute neurologic disorders,
preferably in a range that is targeted to the patient especially
in situations in which autoregulation is impaired. Although
subtle changes in serial neurologic examinations may suggest
altered CBF, the sensitivity of these clinical findings can be
enhanced by careful attention to other bedside indicators,
such as CPP.
Neurocritical Care Delivery
and Patient Outcome
The Senate and House of Representatives of the United States
designated the decade beginning January 1, 1990, as the
“Decade of the Brain.” The goal of this proclamation was to
areas to better understand “the structure of the brain and how
it affects our development, health and behavior.” An immedi-
ate consequence of this has been a significant increase in the
number of new tools available to monitor the central nervous
Table 2.1  Summary of Studies That Investigate the Impact of Specialized Neurocritical Care on Outcomes of Critically Ill
Neurologic and Neurosurgical Patients.
Author Objective
Diringer To determine whether mortality rate after
et al, 2001 intracerebral hemorrhage (ICH) is lower
in patients admitted to a neurologic or
neurosurgical (neuro) intensive care unit
(ICU) compared with those admitted to
general ICUs
Patel et al, To document the effect of neurocritical
2002 care, delivered by specialist staff and
based on protocol-driven therapy aimed
at intracranial pressure (ICP) and
cerebral perfusion pressure (CPP)
targets, on outcome in acute head injury
Wilby et al, To prospectively assess outcome and cost
2003 for poor-grade subarachnoid
hemorrhage patients presenting to a
regional neurosurgical center
Berman et al, To examine the impact of hospital
2003 characteristics on outcome after the
treatment of ruptured and unruptured
cerebral aneurysms
Suarez et al, Length of stay and mortality in
2004 neurocritically ill patients: impact of a
specialized neurocritical care team
Helms et al, Measuring the effect of uncoupling and
2004 removal of the treating physician from
organ and tissue donation requests on
consent rates for donation in the
neurocritical care unit
Varelas et al, To evaluate the impact of a newly
2004 appointed neurointensivist on
neurosciences intensive care unit (NICU)
patient outcomes and quality of care
variables
Varelas et al, Examined the hypothesis that a newly
2005 appointed neurointensivist may alter
documentation practices in a university
hospital setting
Varelas et al, Impact of a neurointensivist on outcomes
2006 in patients with head trauma treated in
an NICU
Bershad Impact of a specialized neurointensive care
et al, 2008 team on outcomes of critically ill acute
ischemic stroke patients
Varelas et al, The impact of a neurointensivist on
2008 patients with stroke admitted to an
NICU
10
Design Outcome Measure Conclusion
Outcomes, Hospital mortality, hospital For patients with acute ICH, admission to a neuro- vs.
cross-sectional length of stay (LOS), general ICU is associated with reduced mortality rate. The
with concurrent readmission rates, presence of a full-time intensivist was associated with a
control, long-term mortality lower mortality rate
retrospective
Cohort with Hospital mortality Specialist neurocritical care with protocol-driven therapy is
historical control, associated with a significant improvement in outcome for
retrospective all patients with severe head injury. Such management
may also benefit patients requiring no surgical therapy,
some of whom may need complex therapeutic
Section I—Background
interventions
Outcomes, Hospital mortality, hospital Poor-grade aneurysmal subarachnoid hemorrhage is
cross-sectional length of stay, cost associated with a high mortality, but a significant subset
analysis of patients can achieve favorable outcomes.
Outcomes, Hospital mortality, hospital Hospital procedural volume and the propensity of a
cross-sectional length of stay hospital to use endovascular therapy are both
independently associated with better outcome
Cohort with Hospital mortality, hospital Introduction of a neurocritical care team, including a
historical control, length of stay, readmission full-time neurointensivist who coordinated care, was
retrospective rates, long-term mortality associated with significantly reduced in-hospital mortality
and length of stay without changes in readmission rates
or long-term mortality.
Cohort with End-of-life care and organ Neurointensivist-led policy change resulted in an increase in
historical control, donation consent rates for organ donation.
retrospective
Cohort with Hospital mortality, hospital The institution of a neurointensivist-led team model was
historical control, length of stay, readmission associated with an independent positive impact on
retrospective rates, long-term mortality patient outcomes, including a lower ICU mortality, LOS,
and discharge to a skilled nursing facility and a higher
discharge home.
Cohort with Length of stay, readmission A major change was implemented in the NICU regarding
historical control, rates documentation after a neurointensivist was appointed.
retrospective
Cohort with Hospital mortality, hospital Neurointensivist-led team model had an independent,
historical control, length of stay, readmission positive impact on patient outcomes, including a lower
retrospective rates, long-term mortality NICU-associated mortality rate and hospital LOS,
improved disposition, and better chart documentation
Cohort with Hospital mortality, hospital Institution of a dedicated neurocritical care team was
historical control, length of stay, readmission associated with a reduction in resource utilization and
retrospective rates, long-term mortality improved patient outcomes at hospital discharge.
Cohort with Hospital mortality, hospital The direct patient care offered and the organizational
historical control, length of stay, readmission changes implemented by a neurointensivist shortened
retrospective rates, long-term mortality the NCCU stay and hospital LOS and improved the
disposition of patients with strokes admitted to an NCCU.

system.25 A multidisciplinary team comprised of neurologists,
neurointensivists, neurosurgeons, and neuroradiologists is
frequently seen in the NCCU of the academic medical center
of today. The efficiency and core proficiency of any NCCU
further depends on a central core of nurses with neurocritical
care training.
There is ongoing debate about whether general ICUs
require a dedicated, full-time team. Several studies have shown
that admission to closed units is associated with a decrease in
morbidity and mortality rates, average ventilation time,
reduced length of stay, complications, resource utilization, and
total hospital charges. For patients with ICH, TBI, and severe
AIS, the introduction of a multidisciplinary NCCU team has
been associated with decreased mortality, shortened the hos-
pital length of stay, and lowered total cost of care, and has led
to better disposition at discharge, that is to say, increased rate
of home discharges and a concomitant decrease in nursing
home discharges.16,26 Differences in care, including the use of
more invasive intracranial monitoring, tracheotomy, and less
use of intravenous (IV) sedation, may account for some of the
observed better outcome in NCCUs.27 It remains to be seen
whether true “closed” or “open” NCCUs provide better out-
comes. However, it does appear that patients with neurologic
disorders that require ICU admission are better in dedicated
NCCUs than general ICUs.12 Although few studies have
directly addressed the value of using information from a
monitor in the NCCU, a recent meta-analysis of the TBI lit-
erature that included more than 100,000 patients suggests
there is a value to using an ICP monitor and aggressive treat-
ment in severe TBI.28 Similarly clinical studies suggest that use
of a brain oxygen monitor and ICP monitor and management
based on that may help improve outcome,29 whereas microdi-
alysis studies suggest that changes in the brain can be observed
before altered ICP occurs.30
Why Monitor Patients in the NCCU:
Historical Perspective
Between the 1960s and 1980s monitoring in the ICU was
restricted to clinical examination, heart and respiratory rate,
blood pressure, oxygen saturation, body temperature, and
central venous pressure. NCCUs at that time focused almost
exclusively on the care of postoperative neurosurgical patients,
and neuromonitoring was restricted primarily to serial neu-
rologic examination and, in some units, ICP monitoring. This
era can be thought of as the age of clinical neuromonitoring.
Between 1980 and 2000, use of ICP monitoring became more
widespread and the age of physiologic neuromonitoring was
ushered in. The idea was to detect and treat increases in ICP
before they led to obvious and often irreversible clinical dete-
rioration. The current decade has focused on monitoring and
management based on brain-derived physiologic information
to detect secondary injury in its earliest phase. Monitoring
techniques in the ICU can be divided into “whole-brain moni-
toring,” which includes ICP monitoring devices, jugular bulb
catheters, electroencephalography (EEG), evoked potentials,
and “regional brain monitoring,” which includes transcranial
Doppler (TCD) ultrasonography, xenon-133 clearance, laser
Doppler flowmetry, thermal diffusion flowmetry, microdialy-
sis catheters, and brain-tissue oxygen probes. These various
monitors can be supplemented with radiologic studies. The
Section I—Background 11
future of neurocritical care will see a further development of
multimodality monitoring in large part because it now is real-
ized that no one monitor can provide all the necessary infor-
mation about what is happening in the brain. In addition, the
evolution of computing power and informatics means that the
information from various monitors can in the future be used
for neurophysiologic decision support. Advanced monitoring
techniques that can provide real-time information about the
relative health or distress of the brain and specifically cellular
health, such as brain tissue oxygen tension, signal-processed
continuous EEG, and neurochemical analysis using microdi-
alysis likely will be used to develop strategies to maintain an
optimal physiologic environment for the comatose injured
brain. This may be particularly important because most
studies of single agents or neuroprotection, although success-
ful in the laboratory, have not proven beneficial in clinical
practice. This suggests that strategies that are based on under-
standing of physiology and pathophysiology (i.e., that are
monitoring based) may be useful given the complex and het-
erogeneous consequences of any injury to the brain.
Primary and Secondary Brain Injury
It is now well established that any acute insult to the brain,
whether it be trauma, ischemia, hemorrhage, edema, or
infection among others, is associated with primary and sec-
ondary brain injury. The same is true for spinal cord injury.
In large part the prevention and management of secondary
injury through early detection with a variety of monitors has
become the focus of modern NCCUs. Recognition of primary
and secondary brain injury allows the NCCU team to orga-
nize the proper resources to influence clinical outcome in the
critically ill patient. Secondary injury can be complex and
represent an interaction between systemic and intracranial
factors. Therefore knowledge of cellular mechanisms of
injury through monitoring becomes important to allow early
intervention. There are many biochemical and pathophysio-
logic processes that are activated. For example, when CPP
and as a consequence, CBF are reduced, failure of aerobic
metabolism and thus the electron transport system may
result. This in turn leads to disruption of the mitochondrial
membrane potential and depletion of adenosine triphos-
phate (ATP). During the process oxygen radicals are released
and the ensuing anaerobic metabolism lowers the pH and
can induce cellular edema; this prevents repolarization
further and exacerbates ATP depletion. There is also release
of excitatory amino acids, and CSF concentrations of gluta-
mate, glycine, and aspartate can increase significantly.21 A
change in the immune response triggers the production of
excitatory cytokines (interleukin [IL]-6 and tumor necrosis
factor) that contribute to dysfunction of the blood-brain
barrier. These various intracranial processes generally are
detected by indirect measures (e.g., ICP, glucose metabolism
through microdialysis, EEG, brain oxygen), but newer tech-
niques are evolving to allow more direct assessments. In
addition it appears that processes considered to be “symp-
toms,” for example, seizures, are actually secondary insults.
Many secondary injuries also result from systemic factors
such as hypoxia, hypotension, hyperglycemia, hyperthermia,
shivering anemia, coagulation disorders, and organ
dysfunction.31-34 In addition secondary injury may result
from treatment; for example, hyperventilation may reduce
12 Section I—Background
ICP but simultaneously decrease CBF. The ability to monitor
for systemic dysfunction and the effects of treatment hence
become important in caring for a patient in the NCCU.
Rationale for Systemic Monitoring
Patient monitoring is an integral part of the management of
severely brain-injured patients in the NCCU and can be clas-
sified broadly into systemic monitoring and brain-specific
techniques. Brain-specific techniques may be further classified
into radiographic techniques or continuous methods. Second-
ary brain injury can result from systemic insults. Furthermore,
efforts to treat the intracranial pathophysiology, although suc-
cessful in reversing abnormalities in the brain, can aggravate
function in other organ systems and so not improve outcome.35
Observational data suggest that efforts to correct systemic
insults can help improve outcome in the NCCU.36 Systemic
monitoring includes invasive arterial blood pressure record-
ing, core body temperature, heart rate, systemic arterial oxygen
saturation (SaO2), and central venous pressure. Invasive blood
pressure monitoring may be indicated when excessive blood
pressure elevations may lead to ICH or hematoma expansion
such as in the post-thrombolysis patient or with primary ICH;
in those with loss of cerebral autoregulation, for example,
poor-grade SAH; or in patients with autonomic instability
such as severe Guillain-Barré syndrome. In addition, invasive
blood pressure monitoring can help guide therapy in those
patients with neurologic injury with septic shock or those
with critical vascular stenosis and fluctuations of blood
pressure.
Monitoring central venous pressure (CVP) and in special
circumstances the use of pulmonary artery catheterization
that can provide an estimate of intravascular volume status in
conditions such as pulmonary edema, cardiac failure, cerebral
vasospasm, severe preeclampsia, and sepsis are useful. Heart
rate, telemetry, respiratory rate, and oxygen saturation help
alert the neurocritical care team to early signs of sepsis, pul-
monary embolism, myocardial infarction, ventricular tachy-
cardia, and pneumonia. Temperature monitoring in the
NCCU is essential because hyperthermia is common and
known to aggravate outcome in a variety of conditions admit-
ted to the NCCU, whereas normothermia, perhaps through
suppression of the immune response, can increase the likeli-
hood of favorable outcome.37,38 Regular laboratory analysis
of blood samples and chest x-rays supplement continuous
monitoring.
Rationale for Neurologic Monitoring
The “ideal” neurologic monitor does not exist. Instead what
has evolved in recent years in NCCUs is the concept of mul-
timodality monitoring, that is, the use of more than one com-
plementary method to monitor a single organ, when no one
single method can provide complete information. This is par-
ticularly true for the brain, although more relevant than the
monitor is what is done with the information. Current neu-
romonitoring techniques involve a range of tools that have
evolved from the study of cerebral physiology and from
advances in the understanding of the pathophysiology of
acute brain injury. The following text presents a brief overview
of several available techniques. A more detailed discussion on
each technique is presented in other chapters in this book.
Broadly these monitors provide either single measurements at
a specific point in time (radiographic techniques) or continu-
ous information.
An important advance in the management of critically ill
patients has been the emergence of functional imaging and
sophisticated magnetic resonance imaging (MRI) techniques.
Consequently radiology has undergone a paradigm shift from
simply providing an anatomic image to providing informa-
tion about tissue metabolism. These techniques remain limited
by fixed time windows despite the development of faster scan-
ners and higher-resolution detectors.
Positron emission tomography (PET) may be considered
the current gold-standard method to image brain metabolism.
Although largely a research tool, it has been shown to be
specific for changes in CBF, cerebral blood volume (CBV),
and oxygen and glucose requirements (rCMRO2/rCMRGlu)
including regional and relative changes. PET also can be used
to identify some traumatic brain injuries not associated with
overt anatomic abnormalities. For example, diffuse axonal
injury (DAI) following TBI is associated with diffuse cortical
hypometabolism and a decrease in CMRO2 in the visual
cortex.39 Newer MRI techniques are largely replacing some of
PET’s role. In addition PET is expensive and available in only
a few major medical centers.
Single-photon emission computed tomography (SPECT) is
available in most medical centers and shares similar basic
principles to PET scanning but has less spatial resolution and
provides qualitative rather than quantitative information.
SPECT can be used to measure CBV and mean transit time
(MTT). CBF is then calculated from the equation: CBF =
CBV/MTT. Abnormalities detected using SPECT within 24
hours after TBI even with “negative” initial head computed
tomography (CT) scans are associated with patient outcome.40
CT perfusion is a newer imaging modality that provides
similar qualitative information as SPECT and is able to
measure rCBF, rCBV and rMTT. In addition CT perfusion can
be used to detect of the state of cerebral autoregulation. Even
though a single examination in time, this information can
have important clinical value because in patients with normal
autoregulation, alterations of mean arterial pressure are
unlikely to affect CPP, whereas in those with defective auto-
regulation, elevation of the MAP is more likely to result in
elevated ICP rather than CPP.41
ICP monitoring is the central monitoring technique in
much of NCCU care. Increased ICP (>20 mm Hg) is an
important secondary insult and associated with mortality
after TBI. Patients with an admission Glasgow Coma Scale
(GCS) score between 3 and 8 and an abnormal CT scan are
most likely to develop intracranial hypertension, and it is rec-
ommended that these patients receive an ICP monitor. Knowl-
edge about ICP is necessary to calculate CPP that is a
“surrogate” of CBF where direct CBF monitoring is not used.
The use of an ICP monitor is described in detail in the Brain
Trauma Foundation’s Guidelines for severe TBI. Despite
nearly 50 years of use, there is only class II evidence literature
that supports a role for ICP monitors,42-45 although meta-
analysis of the literature suggests use of an ICP is associated
with better outcome after TBI.28
Histopathologic evidence for ischemia is a common finding
in autopsy studies in TBI and SAH, and it is likely that inad-
equate CBF contributes to the occurrence of post-traumatic

secondary brain insults and increases the probability of a poor
outcome. In addition CBF threshold values for infarction and
the penumbra (which remains potentially salvageable) have
been defined. The tissue with a CBF between these two thresh-
olds will proceed to infarction if CBF is not restored within a
limited time window. Therefore CBF monitoring offers a
rational approach to detect and prevent secondary insults.46-48
Bedside CBF monitors can be characterized as quantitative or
qualitative and use either direct or indirect methods. Jugular
oximetry and TCD provide nonquantitative or “adequacy of
CBF” data and are the methods used most often in the NCCU.
Jugular venous oxygen saturation (SjvO2) provides informa-
tion about the adequacy of global CBF in relation to metabolic
demands. Reduction in SjvO2 to less than 50% after TBI is
associated with poor outcome, and there is some evidence to
suggest that therapies based on this information may help
improve outcomes. However, SjvO2 is limited by its lack of
sensitivity to regional changes.49,50 TCD was introduced to
clinical practice in 1981 and is based on the Doppler principle.
It is noninvasive and can be used repeatedly but its interpreta-
tion is operator dependent. TCD measures CBF velocity and
not CBF. However, reasonable correlations have been reported
between TCD and xenon CT or PET CBF measurements.
Invasive techniques to measure CBF continually include
such techniques as laser Doppler flowmetry (LDF) and
thermal diffusion flowmetry (TDF). Both are available as
intraparenchymal probes placed through a small craniotomy
or held in place by a skull bolt and offer the advantage of
assessing tissue perfusion in the microcirculation rather than
the major vessels. LDF measures erythrocyte flux and relative
changes in CBF. In TDF the catheter contains a distal thermis-
tor and a second, more proximal located temperature probe.
The thermistor is heated to few degrees above tissue tempera-
ture, and the temperature probe samples temperature con-
stantly. The temperature difference is thus a reflection of heat
transfer and may be translated to a measure of CBF. Initial
data suggest that TDF provides a sensitive real-time assess-
ment of intraparenchymal CBF that agrees with the xenon CT
CBF measurements.51 Monitors that provide a measure of the
adequacy of CBF and insights into metabolism include direct
brain oxygen tension measurement and cerebral microdialy-
sis. Brain oxygen monitoring has demonstrated that many
of the secondary processes that complicate primary brain
injury—hypotension, hypoxia, intracranial hypertension—all
can decrease brain oxygen tension.52,53 The result of oxygen
deprivation is increased anaerobic metabolism, which in turn
can lead to cell death and progressive inflammation.54 Cere-
bral microdialysis can be used to detect such metabolic
changes before irreversible injury and may complement cere-
bral oximetry or ICP measurements. Severe hypoxia or isch-
emia is typically associated with marked increases in the
lactate/pyruvate ratio that correlates with the PET-measured
oxygen extraction fraction. An increase in the lactate-
to-pyruvate ratio greater than what is considered normal
(20 : 25) is associated with a poor outcome in severe TBI. Both
brain oxygen monitors and cerebral microdialysis can be used
in patients with epilepsy, stroke, SAH, and TBI, where their
use often may detect pathology that is not detected by more
conventional means.55-61
Seizures are a source of secondary insult to the injured
brain. Recent data from continuous electroencephalography
(cEEG) studies demonstrate that seizures, mostly not clinically
Section I—Background 13
manifest, occur in more than 20% of patients in the NCCU
with a variety of pathologies including TBI, SAH, or ICH,
among others. In addition EEG can be used to detect ischemia
that may be particularly useful in patients with SAH who
develop vasospasm. Other applications of cEEG include
prognostication of outcome, for example, the use of alpha
variability in cEEG recordings may help predict outcome
after TBI.62-64
Conclusion
Neuromonitoring has evolved and become an integral part
of the care of patients with neurocritical care diseases. The
primary justification for monitoring is that detection of early
neurologic worsening can prevent irreversible brain damage.
Neuromonitoring may help to individualize patient care
decisions, guide patient management, and monitor the ther-
apeutic response of interventions and so avoid any potential
adverse effects. In addition, monitoring may allow physicians
and nurses to understand the pathophysiology of complex
disorders, to design and implement management protocols
based on real-time data, and to improve neurologic outcome
and quality of life in survivors of severe brain injury. Tradi-
tional monitoring in the NCCU has been largely reactive, but
with the introduction of more advanced neuromonitoring
the information provided by newer monitoring techniques
now allows trends to be defined and for the proactive treat-
ment of patients. The application of clinical informatics
and the use of multimodal monitoring are evolving and
have become standard in some NCCUs; this is centered
on patient-specific physiologic targets. Further evolution of
monitoring and its integration is likely in the coming
years such that decision support will become commonplace
in the NCCU.
References
1. Rosamond W, Flegal K, Furie K, et al. Heart disease and stroke statistics—
2008 Update: a report from the American Heart Association Statistics
Committee and Stroke Statistics Subcommittee. Circulation 2008;117:
e25–146.
2. Hyder AA, Wunderlich CA, Puvanachandra P, et al. The impact of
traumatic brain injuries: a global perspective. NeuroRehabilitation 2007;22:
341–53.
3. Wijdicks EFM, Hijdra A, Young GB, et al. Practice parameter: prediction of
outcome in comatose survivors after cardiopulmonary resuscitation (an
evidence-based review): report of the Quality Standards Subcommittee of
the American Academy of Neurology. Neurology 2006;67:203–10.
4. Bershad EM, Feen ES, Hernandez OH, et al. Impact of a specialized
neurointensive care team on outcomes of critically ill acute ischemic stroke
patients. Neurocrit Care 2008;9:287–92.
5. Suárez JI. Critical care neurology and neurosurgery. Totowa, NJ: Humana
Press; 2004.
6. Ropper AH, Gress DR, Diringer MN, et al. Neurological and neurosurgical
intensive care. 4th ed. Philadelphia: Lippincott Williams & Wilkins; 2003.
7. Mayer SA, Coplin WM, Chang C, et al. Neurocritical Care Society; American
Academy of Neurology Section on Critical Care and Emergency Neurology;
Society of Neurosurgical Anesthesia and Critical Care. Program
requirements for fellowship training in neurological intensive care: United
Council for Neurologic Subspecialties guidelines. Neurocrit Care
2006;5(2):166–71.
8. Mayer SA, Coplin WM, Chang C, et al. Neurocritical Care Society; American
Academy of Neurology Section on Critical Care and Emergency Neurology;
Society of Neurosurgical Anesthesia and Critical care. Core curriculum and
competencies for advanced training in neurological intensive care: United
Council for Neurologic Subspecialties guidelines. Neurocrit Care 2006;5(2):
159–65.
14 Section I—Background
9. Suarez JI. Outcome in neurocritical care: advances in monitoring and
treatment and effect of a specialized neurocritical care team. Crit Care Med
2006;34:S232–8.
10. Varelas PN, Eastwood D, Yun HJ, et al. Impact of a neurointensivist on
outcomes in patients with head trauma treated in a neurosciences intensive
care unit. J Neurosurg 2006;104:713–19.
11. Varelas PN, Schultz L, Conti M, et al. The impact of a neuro-intensivist on
patients with stroke admitted to a neurosciences intensive care unit.
Neurocrit Care 2008;9:293–9.
12. Diringer MN, Edwards DF. Admission to a neurologic/neurosurgical
intensive care unit is associated with reduced mortality rate after
intracerebral hemorrhage. Crit Care Med 2001;29:635–40.
13. Mirski MA, Chang CW, Cowan R. Impact of a neuroscience intensive care
unit on neurosurgical patient outcomes and cost of care: evidence-based
support for an intensivist-directed specialty ICU model of care. J Neurosurg
Anesthesiol 2001;13:83–92.
14. Berman MF, Solomon RA, Mayer SA, et al. Impact of hospital-related factors
on outcome after treatment of cerebral aneurysms. Stroke 2003;34:2200–7.
15. Wilby MJ, Sharp M, Whitfield PC, et al. Cost-effective outcome for treating
poor-grade subarachnoid hemorrhage. Stroke 2003;34:2508–11.
16. Suarez JI, Zaidat OO, Suri MF, et al. Length of stay and mortality in
neurocritically ill patients: impact of a specialized neurocritical care team.
Crit Care Med 2004;32:2311–17.
17. Bernard SA, Gray TW, Buist MD, et al. Treatment of comatose survivors of
out-of-hospital cardiac arrest with induced hypothermia. N Engl J Med
2002;346:557–63.
18. Hypothermia After Cardiac Arrest Study Group. Mild therapeutic
hypothermia to improve the neurologic outcome after cardiac arrest. N Engl
J Med 2002;346:549–56.
19. Varelas PN, Spanaki MV, Hacein-Bey L. Documentation in medical records
improves after a neurointensivist’s appointment. Neurocrit Care
2005;3:234–6.
20. Helms AK, Torbey MT, Hacein-Bey L, et al. Standardized protocols increase
organ and tissue donation rates in the neurocritical care unit. Neurology
2004;63:1955–7.
21. Rosenthal G, Hemphill 3rd JC, Sorani M, et al. Brain tissue oxygen tension is
more indicative of oxygen diffusion than oxygen delivery and metabolism in
patients with traumatic brain injury. Crit Care Med 2008;36:1917–24.
22. Obrist WD, Langfitt TW, Jaggi JL, et al. Cerebral blood flow and metabolism
in comatose patients with acute head injury: relationship to intracranial
hypertension. J Neurosurg 1984;61:241–53.
23. Lassen N. Cerebral blood flow and oxygen consumption in man. Physiol Rev
1959;39:183–238.
24. Rosner M, Rosner S, Johnson A. Cerebral perfusion pressure: management
protocol and clinical results. J Neurosurg 1995;83:949–62.
25. Sahuquillo J. Does multimodality monitoring make a difference in
neurocritical care? Eur J Anaesthesiol Suppl 2008;42:83–6.
26. Varelas PN, Conti MM, Spanaki MV, et al. The impact of a neurointensivist-
led team on a semiclosed neurosciences intensive care unit. Crit Care Med
2004;32:2191–8.
27. Kurtz P, Fitts V, Sumer Z, et al. How does care differ for neurological patients
admitted to a neurocritical care unit versus a general ICU? Neurocrit Care
2011;Apr 26. Epub ahead of print.
28. Stein SC, Georgoff P, Meghan S, et al. Relationship of aggressive monitoring
and treatment to improved outcomes in severe traumatic brain injury.
J Neurosurg 2010;112(5):1105–12.
29. Spiotta AM, Stiefel MF, Gracias VH, et al. Brain tissue oxygen-directed
management and outcome in patients with severe traumatic brain injury.
J Neurosurg 2010;113:571–80.
30. Adamides AA, Rosenfeldt FL, Winter CD, et al. Brain tissue lactate elevations
predict episodes of intracranial hypertension in patients with traumatic
brain injury. J Am Coll Surg 2009;209:531–9.
31. Chesnut RM, Marshall LF, Klauber MR, et al. The role of secondary brain
injury in determining outcome from severe head injury. J Trauma 1993;34:
216–22.
32. Oddo M, Schmidt JM, Carrera C, et al. Impact of tight glycemic control on
cerebral glucose metabolism after severe brain injury: a microdialysis study.
Crit Care Med 2008;36:3233–8.
33. Oddo M, Milby A, Chen I, et al. Hemoglobin concentration and cerebral
metabolism in patients with aneurysmal subarachnoid hemorrhage: a
microdialysis study. Stroke 2009;40(4):1275–81.
34. Oddo M, Frangos S, Maloney-Wilensky E, et al. Effect of shivering on brain
tissue oxygenation during induced normothermia in patients with severe
brain injury. Neurocrit Care 2010;12(1):10–6.
35. Robertson CS, Valadka AB, Hannay HJ, et al. Prevention of secondary
ischemic insults after severe head injury. Crit Care Med 1999;27:2086–95.
36. Elf K, Nilsson P, Enblad P. Outcome after traumatic brain injury improved
by an organized secondary insult program and standardized neurointensive
care. Crit Care Med 2002;30:2129–34.
37. Markgraf C, Clifton G, Moody M. Treatment window for hypothermia in
brain injury. J Neurosurg 2001;95:979–83.
38. Clifton GL, Choi SC, Miller ER, et al. Intercenter variance in clinical trials of
head trauma—experience of the National Acute Brain Injury Study:
hypothermia. J Neurosurg 2001;95:751–5.
39. Newberg AB, Alavi A. Neuroimaging in patients with head injury. Semin
Nucl Med 2003;33:136–47.
40. Jacobs A, Put E, Ingels M, et al. Prospective evaluation of technetium-99m-
HMPAO SPECT in mild and moderate traumatic brain injury. J Nucl Med
1994;35:942–47.
41. Wintermark M, Chiolero R, Van Melle G, et al. Cerebral vascular
autoregulation assessed by perfusion-CT in severe head trauma patients.
J Neuroradiol 2006;33:27–37.
42. The Brain Trauma Foundation. The American Association of Neurological
Surgeons. The Joint Section on Neurotrauma and Critical Care. Indications
for intracranial pressure monitoring. J Neurotrauma 2000;17:479–91.
43. Becker DP, Miller JD, Ward JD, et al. The outcome from severe head injury
with early diagnosis and intensive management. J Neurosurg 1977;47:
491–502.
44. Ghajar JB, Hairiri RJ, Paterson RH, et al. Improved outcome from traumatic
coma using only ventricular CSF drainage for ICP control. Adv Neurosurg
1993;21:173–7.
45. Narayan RK, Kishore PR, Becker DP, et al. Intracranial pressure: to monitor
or not to monitor? A review of our experience with severe head injury.
J Neurosurg 1982;56:650–9.
46. Bouma GJ, Muizelaar JP, Choi SC, et al. Cerebral circulation and metabolism
after severe traumatic brain injury: the elusive role of ischemia. J Neurosurg
1991;75:685–93.
47. Robertson CS, Contant CF, Gokaslan ZL, et al. Cerebral blood flow,
arteriovenous oxygen difference, and outcome in head injured patients.
J Neurol Neurosurg Psychiatry 1992;55:594–603.
48. Symon L, Held K, Dorsch N. A study of regional autoregulation in the
cerebral circulation to increased perfusion pressure in normocapnia and
hypercapnia. Stroke 1973;4:139–47.
49. Murr R, Schurer L. Correlation of jugular venous oxygen saturation to
spontaneous fluctuations of cerebral perfusion pressure in patients with
severe head injury. Neurol Res 1995;17:329–33.
50. Robertson CS, Gopinath SP, Goodman JC, et al. SjvO2 monitoring in
head-injured patients. J Neurotrauma 1995;12:891–6.
A complete list of references for this chapter can be found online at
www.expertconsult.com.

References
1. Rosamond W, Flegal K, Furie K, et al. Heart disease and stroke statistics—
2008 Update: a report from the American Heart Association Statistics
Committee and Stroke Statistics Subcommittee. Circulation 2008;117:
e25–146.
2. Hyder AA, Wunderlich CA, Puvanachandra P, et al. The impact of traumatic
brain injuries: a global perspective. NeuroRehabilitation 2007;22:341–53.
3. Wijdicks EFM, Hijdra A, Young GB, et al. Practice parameter: prediction of
outcome in comatose survivors after cardiopulmonary resuscitation (an
evidence-based review): report of the Quality Standards Subcommittee of
the American Academy of Neurology. Neurology 2006;67:203–10.
4. Bershad EM, Feen ES, Hernandez OH, et al. Impact of a specialized
neurointensive care team on outcomes of critically ill acute ischemic stroke
patients. Neurocrit Care 2008;9:287–92.
5. Suárez JI. Critical care neurology and neurosurgery. Totowa, NJ: Humana
Press; 2004.
6. Ropper AH, Gress DR, Diringer MN, et al. Neurological and neurosurgical
intensive care. 4th ed. Philadelphia: Lippincott Williams & Wilkins; 2003.
7. Mayer SA, Coplin WM, Chang C, et al. Neurocritical Care Society; American
Academy of Neuorology Section on Critical Care and Emergency Neurology;
Society of Neurosurgical Anesthesia and Critical Care. Program
requirements for fellowship training in neurological intensive care: United
Council for Neurologic Subspecialties guidelines. Neurocrit Care
2006;5(2):166–71.
8. Mayer SA, Coplin WM, Chang C, et al. Neurocritical Care Society; American
Academy of Neurology Section on Critical Care and Emergency Neurology;
Society of Neurosurgical Anesthesia and Critical care. Core curriculum and
competencies for advanced training in neurological intensive care: United
Council for Neurologic Subspecialties guidelines. Neurocrit Care 2006;5(2):
159–65.
9. Suarez JI. Outcome in neurocritical care: advances in monitoring and
treatment and effect of a specialized neurocritical care team. Crit Care Med
2006;34:S232–8.
10. Varelas PN, Eastwood D, Yun HJ, et al. Impact of a neurointensivist on
outcomes in patients with head trauma treated in a neurosciences intensive
care unit. J Neurosurg 2006;104:713–19.
11. Varelas PN, Schultz L, Conti M, et al. The impact of a neuro-intensivist on
patients with stroke admitted to a neurosciences intensive care unit.
Neurocrit Care 2008;9:293–9.
12. Diringer MN, Edwards DF. Admission to a neurologic/neurosurgical
intensive care unit is associated with reduced mortality rate after
intracerebral hemorrhage. Crit Care Med 2001;29:635–40.
13. Mirski MA, Chang CW, Cowan R. Impact of a neuroscience intensive care
unit on neurosurgical patient outcomes and cost of care: evidence-based
support for an intensivist-directed specialty ICU model of care. J Neurosurg
Anesthesiol 2001;13:83–92.
14. Berman MF, Solomon RA, Mayer SA, et al. Impact of hospital-related
factors on outcome after treatment of cerebral aneurysms. Stroke 2003;34:
2200–7.
15. Wilby MJ, Sharp M, Whitfield PC, et al. Cost-effective outcome for treating
poor-grade subarachnoid hemorrhage. Stroke 2003;34:2508–11.
16. Suarez JI, Zaidat OO, Suri MF, et al. Length of stay and mortality in
neurocritically ill patients: impact of a specialized neurocritical care team.
Crit Care Med 2004;32:2311–17.
17. Bernard SA, Gray TW, Buist MD, et al. Treatment of comatose survivors of
out-of-hospital cardiac arrest with induced hypothermia. N Engl J Med
2002;346:557–63.
18. Hypothermia After Cardiac Arrest Study Group. Mild therapeutic
hypothermia to improve the neurologic outcome after cardiac arrest. N Engl
J Med 2002;346:549–56.
19. Varelas PN, Spanaki MV, Hacein-Bey L. Documentation in medical records
improves after a neurointensivist’s appointment. Neurocrit Care
2005;3:234–6.
20. Helms AK, Torbey MT, Hacein-Bey L, et al. Standardized protocols increase
organ and tissue donation rates in the neurocritical care unit. Neurology
2004;63:1955–7.
21. Rosenthal G, Hemphill 3rd JC, Sorani M, et al. Brain tissue oxygen tension is
more indicative of oxygen diffusion than oxygen delivery and metabolism in
patients with traumatic brain injury. Crit Care Med 2008;36:1917–24.
22. Obrist WD, Langfitt TW, Jaggi JL, et al. Cerebral blood flow and metabolism
in comatose patients with acute head injury: relationship to intracranial
hypertension. J Neurosurg 1984;61:241–53.
23. Lassen, N. Cerebral blood flow and oxygen consumption in man. Physiol
Rev 1959;39:183–238.
Section I—Background 14.e1
24. Rosner M, Rosner S, Johnson A. Cerebral perfusion pressure: management
protocol and clinical results. J Neurosurg 1995;83:949–62.
25. Sahuquillo J. Does multimodality monitoring make a difference in
neurocritical care? Eur J Anaesthesiol Suppl 2008;42:83–6.
26. Varelas PN, Conti MM, Spanaki MV, et al. The impact of a neurointensivist-
led team on a semiclosed neurosciences intensive care unit. Crit Care Med
2004;32:2191–8.
27. Kurtz P, Fitts V, Sumer Z, et al. How does care differ for neurological patients
admitted to a neurocritical care unit versus a general ICU? Neurocrit Care
2011;Apr 26. [Epub ahead of print]
28. Stein SC, Georgoff P, Meghan S, et al. Relationship of aggressive monitoring
and treatment to improved outcomes in severe traumatic brain injury.
J Neurosurg 2010;112(5):1105–12.
29. Spiotta AM, Stiefel MF, Gracias VH, et al. Brain tissue oxygen-directed
management and outcome in patients with severe traumatic brain injury.
J Neurosurg 2010;113:571–80.
30. Adamides AA, Rosenfeldt FL, Winter CD, et al. Brain tissue lactate elevations
predict episodes of intracranial hypertension in patients with traumatic
brain injury. J Am Coll Surg 2009;209:531–9.
31. Chesnut RM, Marshall LF, Klauber MR, et al. The role of secondary brain
injury in determining outcome from severe head injury. J Trauma
1993;34:216–22.
32. Oddo M, Schmidt JM, Carrera C, et al. Impact of tight glycemic control on
cerebral glucose metabolism after severe brain injury: a microdialysis study.
Crit Care Med 2008;36:3233–8.
33. Oddo M, Milby A, Chen I, et al. Hemoglobin concentration and cerebral
metabolism in patients with aneurysmal subarachnoid hemorrhage: a
microdialysis study. Stroke 2009;40(4):1275–81.
34. Oddo M, Frangos S, Maloney-Wilensky E, et al. Effect of shivering on brain
tissue oxygenation during induced normothermia in patients with severe
brain injury. Neurocrit Care 2010;12(1):10–6.
35. Robertson CS, Valadka AB, Hannay HJ, et al. Prevention of secondary
ischemic insults after severe head injury. Crit Care Med 1999;27:2086–95.
36. Elf K, Nilsson P, Enblad P. Outcome after traumatic brain injury improved
by an organized secondary insult program and standardized neurointensive
care. Crit Care Med 2002;30:2129–34.
37. Markgraf C, Clifton G, Moody M. Treatment window for hypothermia in
brain injury. J Neurosurg 2001;95:979–83.
38. Clifton GL, Choi SC, Miller ER, et al. Intercenter variance in clinical trials of
head trauma—experience of the National Acute Brain Injury Study:
hypothermia. J Neurosurg 2001;95:751–5.
39. Newberg AB, Alavi A. Neuroimaging in patients with head injury. Semin
Nucl Med 2003;33:136–47.
40. Jacobs A, Put E, Ingels M, et al. Prospective evaluation of technetium-99m-
HMPAO SPECT in mild and moderate traumatic brain injury. J Nucl Med
1994;35:942–47.
41. Wintermark M, Chiolero R, Van Melle G, et al. Cerebral vascular
autoregulation assessed by perfusion-CT in severe head trauma patients.
J Neuroradiol 2006;33:27–37.
42. The Brain Trauma Foundation. The American Association of Neurological
Surgeons. The Joint Section on Neurotrauma and Critical Care. Indications
for intracranial pressure monitoring. J Neurotrauma 2000;17:479–91.
43. Becker DP, Miller JD, Ward JD, et al. The outcome from severe head injury
with early diagnosis and intensive management. J Neurosurg
1977;47:491–502.
44. Ghajar JB, Hairiri RJ, Paterson RH, et al. Improved outcome from traumatic
coma using only ventricular CSF drainage for ICP control. Adv Neurosurg
1993;21:173–7.
45. Narayan RK, Kishore PR, Becker DP, et al. Intracranial pressure: to monitor
or not to monitor? A review of our experience with severe head injury.
J Neurosurg 1982;56:650–9.
46. Bouma GJ, Muizelaar JP, Choi SC, et al. Cerebral circulation and metabolism
after severe traumatic brain injury: the elusive role of ischemia. J Neurosurg
1991;75:685–93.
47. Robertson CS, Contant CF, Gokaslan ZL, et al. Cerebral blood flow,
arteriovenous oxygen difference, and outcome in head injured patients.
J Neurol Neurosurg Psychiatry 1992;55:594–603.
48. Symon L, Held K, Dorsch N. A study of regional autoregulation in the
cerebral circulation to increased perfusion pressure in normocapnia and
hypercapnia. Stroke 1973;4:139–47.
49. Murr R, Schurer L. Correlation of jugular venous oxygen saturation to
spontaneous fluctuations of cerebral perfusion pressure in patients with
severe head injury. Neurol Res 1995;17:329–33.
50. Robertson CS, Gopinath SP, Goodman JC, et al. SjvO2 monitoring in
head-injured patients. J Neurotrauma 1995;12:891–6.
14.e2 Section I—Background
51. Vajkoczy P, Roth H, Horn P, et al. Continuous monitoring of regional
cerebral blood flow: experimental and clinical validation of a novel thermal
diffusion microprobe. J Neurosurg 2000;93:265–74.
52. Gopinath SP, Valadka AB, Uzura M, et al. Comparison of jugular venous
oxygen saturation and brain tissue Po2 as monitors of cerebral ischemia after
head injury. Crit Care Med 1999;27:2337–45.
53. Kiening KL, Härtl R, Unterberg AW, et al. Brain tissue pO2-monitoring in
comatose patients: implications for therapy. Neurol Res 1997;19:233–40.
54. Siesjö BK. Pathophysiology and treatment of focal cerebral ischemia. Part II:
mechanisms of damage and treatment. J Neurosurg 1992;77:337–54.
55. Clausen T, Alves OL, Reinert M, et al. Association between elevated brain
tissue glycerol levels and poor outcome following severe traumatic brain
injury. J Neurosurg 2005;103:233–8.
56. Meixensberger J, Kunze E, Barcsay E, et al. Clinical cerebral microdialysis:
brain metabolism and brain tissue oxygenation after acute brain injury.
Neurol Res 2001;23:801–6.
57. Hutchinson PJ, Gupta AK, Fryer TF, et al. Correlation between cerebral
blood flow, substrate delivery, and metabolism in head injury: a combined
microdialysis and triple oxygen positron emission tomography study. J Cereb
Blood Flow Metab 2002;22:735–45.
58. Vespa P, Martin NA, Nenov V, et al. Delayed increase in extracellular glycerol
with post-traumatic electrographic epileptic activity: support for the theory
that seizures induce secondary injury. Acta Neurochir 2002;81(Suppl.):355–7.
59. Ståhl N, Mellergård P, Hallström A, et al. Intracerebral microdialysis and
bedside biochemical analysis in patients with fatal traumatic brain lesions.
Acta Anaesthesiol Scand 2001;45:977–85.
60. Reinstrup P, Ståhl N, Mellergård P, et al. Intracerebral microdialysis in
clinical practice: baseline values for chemical markers during wakefulness,
anesthesia, and neurosurgery. Neurosurgery 2000;47:701–9.
61. Vespa PM, McArthur D, O’Phelan K, et al. Persistently low extracellular
glucose correlates with poor outcome 6 months after human traumatic brain
injury despite a lack of increased lactate: a microdialysis study. J Cereb Blood
Flow Metab 2003;23:865–77.
62. Scheuer ML. Continuous EEG monitoring in the intensive care unit.
Epilepsia 2002;43(Suppl 3):114–27.
63. Vespa PM, Boscardin WJ, Hovda DA, et al. Early and persistent impaired
percent alpha variability on continuous electroencephalography monitoring
as predictive of poor outcome after traumatic brain injury. J Neurosurg
2002;97:84–92.
64. Vespa PM, Nuwer MR, Nenov V, et al. Increased incidence and impact of
nonconvulsive and convulsive seizures after traumatic brain injury as
detected by continuous electroencephalographic monitoring. J Neurosurg
1999;91:750–60.

Designing the Neurocritical
Care Unit for Better Patient Care
Mahbub Rashid, Craig Zimring, and Owen B. Samuels
Introduction
There is growing evidence that the physical design of clinical
spaces affects the quality, efficiency, and experience of care in
intensive care units (ICUs) and other settings.1-4 These effects
are complex. In some cases the design directly leads to out-
comes that might have clinical significance, for example,
whether improved air filtering reduces airborne pathogens. In
other cases the environment contributes to noise or lighting
that can increase or decrease stress for patient, families, or
staff. In yet other cases, the environment contributes to staff
or patient behavior that may affect care. For instance, if staff
see and encounter each other informally over the course of
their day, they may better coordinate care,5 or if they see hand
hygiene sinks or rubs, they may increase their compliance with
handwashing.3
This chapter examines the effects of the ICU design and
particularly the design of the neurocritical care unit (NCCU)
on patient care. Addressed are the important areas within the
ICU, key design considerations, main design guidelines, and
emerging evidence on how ICU design influences patient care.
The chapter ends with a case study in which the authors were
involved, from the perspective of the medical director.
Evidence-Based Intensive Care
Unit Design
Research in ICU environmental design is an emerging field of
study, and therefore high-quality research articles are still
limited in number. In addition, many confounding variables
can influence the outcome of ICU design research. Therefore
along with the findings of ICU design research, the authors
also present expert opinions and findings of research studies
in other health care settings that may be relevant to ICU
design. This section reviews the primary functional and pro-
cedural issues associated with ICU design and addresses the
key areas of the ICU.
Unit Layout
Unit layout defines the size and location of functional areas
and their relationships within a unit. Major determinants of
© Copyright 2013 Elsevier Inc. All rights reserved.
Chapter
3  
I
the layout include: (1) direct monitoring of patients by clinical
staff, (2) an outside window for each patient room, and (3)
reduction of cross-infection, traffic volume, and noise level,
among other factors.
Monitoring
In NCCUs, patients need constant visual monitoring because
their conditions may change quickly and unpredictably, and
these changes need to be recognized early. NCCU patients also
require frequent neurologic assessments or bedside proce-
dures that may involve multiple members of a care team. A
physical layout that supports effective face-to-face interaction
in a patient’s room and within the unit is thus required. In a
teaching hospital the rooms or corridors also need to support
the large number of participants who make patient rounds.
Windows
U.S. law requires that ICU patients must have direct access to
natural light. This same requirement also applies in several
other countries (e.g., India, Saudi Arabia, and the United
Arab Emirates). The number and arrangement of patient
rooms consequently depend on the amount of peri­meter wall
available in the unit. Therefore a review of best practice
examples shows that designers often select compact shapes
with high area-to-perimeter ratios to accommodate the
maximum number of patient rooms for any given area. They
also put most support areas, including nurse work areas, in
the core that do not require any outside window to reduce the
walking distance between clinical support areas and patient
rooms.6
Limiting Infection, Traffic, and Noise
Cross-infection, traffic volume, and noise level often may help
shape unit configurations, and each of these factors is of
greater significance in larger hospital units. Studies suggest
that patients in larger units have greater risk of hospital-
acquired infection.7,8 Larger units also have more traffic and
noise sources. These noise sources commonly include noises
of other patients (e.g., snoring, crying), monitor alarms, tele-
phone rings and conversations, conversations among staff,
15
16 Section I—Background
staff entering or leaving, staff wandering, sudden voices, foot-
steps, falling objects, noises of respirators, doors closing, and
visitors talking, among others.9 Breaking a larger unit into
smaller units, pods, or clusters may reduce infection and noise.
However, pods can break down the visual and social cohesive-
ness of a unit, and multiple pods may make movement of
supplies difficult because they create more service stops. Thus
the appropriate configuration for a large ICU remains a matter
of striking the right balance among various contradictory
factors.
Unit Size
According to the Society of Critical Care Medicine’s Guide-
lines for Intensive Care Unit Design (henceforth, the Guide-
lines), 8 to 12 beds per unit are considered best from a
functional perspective.10-13 In one survey, a group of ICU
experts also suggested that the ideal number of patient beds
in an ICU should be 9 or 10, and no less than 6.14 An NCCU
with fewer than 6 beds may be inefficient to operate and
manage. In contrast, in a very large unit without proper unit
design and a sufficiently large nursing staff, patient monitor-
ing and care may become difficult.
The total gross area of a unit, another indicator for unit
size, is somewhat related to the number of beds in the unit—
the more beds the greater the ICU gross area per bed.6 The
amount of circulation spaces, another determinant of the
gross area of a unit, is an important indicator of the square
footage efficiency of an ICU. Circulation spaces within an
ICU consist of internal hallways, corridors, or aisles used by
all ICU users—patients, ICU staff, and visitors. Like any
other facility, an ICU may be inefficient with too much or too
little circulation space. Circulation spaces are sociologically
important because they determine the interconnectedness of
people and functions within a facility. A narrow corridor
within an ICU can impede transfer of knowledge as much as
it can impede the flow of goods and people. Properly designed
circulation spaces in ICUs may hold a great potential to
enhance the transfer of tacit knowledge through face-to-face
interactions.15
In the United States, the National Fire Protection Associa-
tion’s Life Safety Code also affects ICU size and design. The
Life Safety Code limits the size of any suite to 5000 square feet
if it does not have intervening smoke partitions and fire-rated
doors.16 The area per bed of a unit ranges from 650 to 1200
square feet or more depending on function on the unit, so
most units require smoke partitions or “hold open” smoke
doors. Hospitals and designers need to ensure that when any
one part of an ICU larger than 5000 square feet becomes
unavailable in the event of fire or smoke breakout, the other
parts of the unit have the required components for patient and
staff safety.
Unit Location
Important departmental relationships of the unit should be
carefully considered during the site selection process. Conve-
nient physical movement across departments may help reduce
many safety risks associated with patient transfer. In general,
patients are transferred from one place to another as often as
three to six times during their hospital stay to receive the care
that matches their level of acuity.17-19 This rate may be higher
for NCCU patients. Delays, communication discontinuities,
loss of information, and changes in computers and systems
during patient transfer can contribute to increased medical
errors and loss of staff time and productivity.18,20 Patients can
get hurt, and most nurses’ back injuries occur during patient
transfer (see the following text). Thus design interventions
must include factors that help reduce the time and effort
involved in patient transfer.
Patient Room Design
A patient’s room is the basic working unit of an ICU, and it
affects patient care including safety, privacy, and comfort. For
example, nurses can save multiple trips to nursing stations or
storage rooms if a patient’s room includes a space for chart-
ing or storage space for supplies. A longer distance between
the bed and the toilet can increase physical stress for a nurse
who needs to help the patient to the toilet. If there is a well-
defined area within the room for families, they can be present
for patient comfort and help caregivers by providing infor-
mation and assistance with care. Important design consider-
ations of patient rooms in the NCCU, similar to other ICUs,
include to: (1) create well-defined functional zones to elimi-
nate conflicts, (2) provide enough space that is appropriate
for patient care, (3) provide necessary life support systems,
(4) balance visibility and privacy, (5) provide toilet facilities,
(6) reduce hospital-acquired infections and psychosis among
patients, and (7) reduce patient and staff injury.
Patient Room Layout
The patient room layout defines the size and location of func-
tions and their relationships within the room. It affects how
functions are performed and how patients and caregivers
interface in the room. Jastremski and Harvey suggest that an
ideal room should have three zones: a patient zone, a family
zone, and a caregiver zone.21 Hamilton and Shepley22 defined
four different zones within a patient’s room: (1) patient, (2)
hygiene, (3) staff, and (4) family zones. The patient zone
includes the bed, bedside, and overbed tables, and the imme-
diate area occupied by clinicians when they provide care. The
hygiene zone includes the patient’s toilet, sink, and activities
associated with hygiene. The staff zone includes the area just
inside or outside the entry to the patient’s room to support
nursing and caregiver functions; this may include a writing
surface, provisions for hand hygiene, patient information,
medication, and supplies. The family zone may include seating
or provisions for overnight stay, storage space, separate light-
ing, Internet access, and a writing surface, among others.
Defining patient room in terms of what is in a patient’s view
and what is not can be important. Evidence suggests that
patients lying on an ICU bed are stressed when they see
medical equipment, accessories, and monitors.23 Therefore it
is reasonable to keep these devices away from the patient’s
view, and instead put family space within the patient’s view.
Architectural treatment of areas within a patient’s view also
can be important because it appears that “positive distrac-
tions” such as nature can reduce stress and pain3 and that
natural light can reduce analgesic use.24 Soothing color and
light, natural materials, and paintings of nature may be used

in the area within the patient’s view to make his or her experi-
ence more comfortable.
Patient Room Size
The need for larger patient rooms is increasing in ICUs. Advo-
cates of infection prevention recommend that each patient
room should have dedicated patient care equipment to reduce
cross-infection with more resistant microbial strains. Others
also recommend that each room should have a dedicated
family space, with amenities to improve family integration
with patient care. As medical breakthroughs and advance-
ments occur, more technology is brought into patient rooms,
and this requires more space. The increasing multidisciplinary
nature of patient care in the NCCU also requires patient
rooms to accommodate larger medical teams. Additional
space also may be needed in patient rooms to support research
and the increasing number of procedural interventions that
now are performed in ICUs, such as bronchoscopy, echocar-
diography, and placement of external ventricular drains.
The Guidelines stipulate, “Ward-type ICUs should allow at
least 225 square feet of clear floor area per bed. ICUs with
individual patient modules should allow at least 250 square
feet per room [assuming one patient per room]. …”12 In a
survey of the best-practice ICUs in the United States built
between 1993 and 2003, the average size of a patient room was
250 square feet.6 However, since 2000, many best-practice
ICUs have also used more than 250 square feet for patient
rooms,25 suggesting that hospitals are aware of trends such as
the demand for family spaces in patient rooms.
Privacy and Visibility in a Patient Room
Privacy is an important factor associated with individual sat-
isfaction in many environments including hospitals and even
in moments of extreme crisis, privacy may be required to
preserve individual dignity. Patient and family surveys suggest
hospitals with more private rooms tend to have higher patient
satisfaction rates (www.pressganey.com). There are several
advantages to private rooms. First, they provide dedicated
space for individualized care without disturbing other patients
and can help reduce noise, improve patient sleep quality, and
support staff-patient communication.21,26 Second, hospital-
acquired infection rates in ICUs with private rooms are less
because of improved airflow, better ventilation, and more
accessible handwashing facilities.1,27-30 Finally, the private
room design allows for a patient care environment with more
control over optimal environmental conditions.31 However, in
private rooms patient visibility is often at a stake and patients
are afraid of being left alone. From a clinical viewpoint it is
easier to make a case for open patient rooms in ICUs; this
permits easy visual monitoring of the patient by the clinical
staff that can affect patient safety. ICU staff often prefer an
open ICU to see everything that happens, to readily seek help
from others in a crisis, and to act immediately in groups
without the constraints of walls of a private patient room.32
In private patient rooms, where glass often is used for patient
visibility, measures to ensure visual and acoustic privacy of
patients and their families when needed also are necessary.
Hospitals often prefer breakaway glass doors because they can
be closed for privacy, noise reduction, and infection control
Section I—Background 17
and still maintain maximum visibility of patients and moni-
tors. Breakaway glass doors also allow maximum clearance to
move patients in and out of the room. In an emergency, break-
away glass doors allow the room to become more open than
the other doors. However, with breakaway doors closed it may
be difficult to hear patient alarms, for example, ventilator
alarms.
Life Support Systems in a Patient Room
Easy access to the patient’s head and the ability to move a
patient bed around during procedures are important in the
NCCU. Traditional head wall systems that have been used
since the 1970s to provide the life-support systems do not
allow easy access to the patient’s head, nor do they allow clini-
cians to reorient the bed when needed. Headwall systems
include power outlets and outlets for medical gases and
vacuum on one or both sides of a patient bed. Some installa-
tions also include wall-mounted monitors and equipment for
a patient’s vitals.
Power columns, whether rotating or static, are now used in
many ICUs to provide life support systems. Equipped with
medical utilities, power outlets, and monitors, these power
columns allow easy access to the patient’s head and may allow
clinicians to reorient the bed. Two, instead of one, power
columns—one on each side of the bed—can be installed to
help increase the symmetry of functions around the patient
bed. However, it can be difficult to work around power
columns during a procedure. Other innovations—the ceiling-
mounted boom, ceiling columns, or the Draeger Ponta
beam—help provide easy access and sufficient flexibility for
proper patient care in NCCUs and in particular provide
access to the patient’s entire body, especially to the head.
These ceiling-mounted systems, however, add cost and often
require additional structural support. These systems occa-
sionally can conflict with a patient lift system in an NCCU
patient room.
Toilets in a Patient Room
ICU patient rooms are not required to have toilets, but there
are many good reasons to have a toilet in the room for both
the patient and family. Toilets in patient rooms can be used to
dump and clean bedpans, which can be a major source of
aerosol contaminants and infectious organisms,33 and con-
tained systems often are used to reduce aerosols. The location,
use, and design of a toilet or of other devices to eliminate
waste have a significant effect on ICU design. A review of
current design practice shows that toilets are placed in many
different locations in relation to a patient room. Inboard
toilets are on the corridor side of patient rooms; outboard are
on the window side. A third option is to place the toilets of
two adjacent patient rooms next to each other in a wet zone.
Each location has advantages and disadvantages.6
Hospital-Acquired Infections
In patient room design, air quality, single-bed patient rooms,
lighting conditions, noise level, and handwashing sink are
important because they can help reduce infections among
patients. Most studies show that (1) private patient rooms can
18 Section I—Background
reduce cross-infection among ICU patients; (2) single-bed
patient rooms with high-quality high-efficiency particulate
air (HEPA) filters and with negative- or positive-pressure
ventilation are more effective in preventing airborne patho-
gens; and (3) multibed rooms are more difficult to decon-
taminate and have more surfaces that act as a reservoir for
pathogens.2,30,34 The 2006 American Institute of Architects
Guidelines for Design and Construction of Healthcare Facili-
ties therefore has adopted the single-bed room as the stan-
dard for all new construction in the United States.35
Infrequent handwashing by health care staff is associated
with hospital-acquired infections.36 Several design factors
may discourage handwashing, including: (1) poor sink loca-
tion, (2) poor visibility, (3) uncomfortable sink height, and
a (4) lack of redundancy and wide spatial separation of
resources that are used sequentially in handwashing.36-39
There are conflicting results on how physical design influ-
ences handwashing compliance.40-43 There is, however, a con-
sensus that a multi-strategy intervention that includes staff
education, easy visual and physical access to sinks, standard
sink locations in all patient rooms, comfortable sink heights,
and alcohol-based dispensers can help increase handwashing
compliance.36,37
A review of current design practice shows that handwashing
sinks are placed at many different locations in a patient room,
including: (1) locations directly outside the room entrance-
way, (2) immediately after the entranceway either on the foot-
wall or on the head wall, (3) somewhere in the middle of the
footwall, and (4) at the far end of the room walls. Designers
and hospitals must consider the advantages and disadvantages
of each of these locations.6 (For a discussion on this topic, see
Rashid.6)
Psychosis Among Intensive Care
Unit Patients
ICU patients who are confined to bed may suffer from delir-
ium, also known as “ICU psychosis.”44 There is growing evi-
dence from non-ICU settings that both natural light and
outdoor views can help patients maintain sensory orientation
and circadian rhythm.45-49 This in turn can reduce the likeli-
hood of delirium. Consequently, the Guidelines recommend:
“Windows are an important aspect of sensory orientation,
and as many rooms as possible should have windows to rein-
force day/night orientation. … If windows cannot be pro-
vided in each room, an alternate option is to allow a remote
view of an outside window or skylight.”12 Artificial lighting
conditions that mimic the variations in natural light also can
help patients maintain sensory orientation and circadian
rhythm. Looking at a ceiling or a wall painted in solid colors
for a long period also may contribute to patient delirium.
Warm colors and paintings of nature may improve the envi-
ronment. Sometimes a calendar that shows the date or a
digital clock that shows date and time may help patients
adjust their internal physiologic rhythm. However, the monot-
onous clicking sound of an analog clock can cause distress in
ICU patients.
ICU psychosis also may be associated with a high level of
noise.50,51 Noise level in the ICU ranges from 50 to 75 dB, with
peaks up to 85 dB.52 This is much higher than World Health
Organization recommendations for noise levels in hospital
patient rooms and units.53 Common sources of noise in
hospitals include telephones, alarms, trolleys, ice machines,
paging systems, nurse shift change, staff caring for other
patients, doors, staff conversations, and patients crying out or
coughing.54,55 Hospital noise can be improved if proper design
and management measures are in place.
Reducing Patient Falls and Fall Severity
The physical environment can be a root cause for patient
falls.56 Among specific interior design elements, flooring can
contribute to the incidence of falls and the severity of injury
from a fall.57 Patients may suffer more injuries when they fall
on vinyl floors than carpeted floors.58 Subfloors also may
influence the injury from falls; the risk of fracture is less for
wooden than concrete subfloors.59 Several design factors can
help reduce the incidence of falls including decentralized
observation units next to patient rooms rather than a central-
ized nursing station,17 patient lifts and other transfer devices,
and innovative acuity adaptable patient rooms.
Other Patient and Staff Injuries
Manual lifting of ICU patients poses a high risk of injury for
patients, such as dislodgement of invasive tubes and lines,
shoulder dislocation, fracture of fragile bones, or being
dropped.60 Skin tears and abrasions also may occur when
patients are pulled up or across beds, and manual patient
handling can contribute to pain in critically ill patients. Pain
experienced by these patients during turning or repositioning
sometimes is greater than that experienced during tracheal
suctioning, tube advancement, and wound dressing changes.60
Staff also are at risk for work-related injuries when moving
patients. The use of ceiling lifts can help limit risks associated
with manual moving to both patient and staff. If possible, lifts
should extend into the toilet room.3 If patient volumes and
staffing patterns allow, ICU rooms that can flex in acuity from
super-acute to step-down also may reduce patient transfers
and discontinuity in care.
Staff Work Areas
and Support Spaces
ICUs can be stressful workplaces that can endanger the physi-
cal and mental health of the clinical staff. In the United States
there is a shortage of critical care nursing staff, and the current
staff is older and has a high turnover rate. Staff turnover rate
in many institutions is greater in ICUs than other wards
because of the stressful working conditions. ICU design,
however, can help relieve staff stress by reducing unnecessary
physical labor, by providing amenities, and by providing posi-
tive distractions for staff physical and mental recovery.
Staff Work and Support Area Location
and Layout
There are three basic nursing unit configurations: (1) central-
ized nursing station, (2) nursing substation, and (3) nursing
observation unit. In older hospital units a centralized nursing
station is generally the main component of the staff work area.
Along with the patients’ records room, the central monitoring
station, and staff workstations for patient charting and medical

recording, it serves as the hub of all unit functions. This type
of nursing station usually has a centralized support and service
area next to it that accommodates medical and supply storage,
pharmacy, conference rooms, administrative offices, and other
ancillary functions. In older units a centralized nursing station
is where clinical management, staff interaction, mentoring,
and socialization occur. However, centralized nursing stations
may contribute to errors and inefficiency because of noise,
crowding, and considerable walking distance from patient
rooms.
In hospital units built in the 1990s and 2000s, the central-
ized nursing station often is replaced with several decentral-
ized observation units with direct observation of one or two
rooms and located either just outside or inside the patient
room. Typical functions include a work surface for patient
charting, a computer to record and access patient informa-
tion, and telecommunication services. There may or may not
be storage spaces for medication and supplies, handwashing
facilities, and image retrieval systems. It is suggested that the
decentralized observation units increase efficiency, but
perhaps at the cost of staff “social life.” Team workstations,
sometimes distributed throughout a unit, provide spaces for
interdisciplinary teamwork, mentoring, clinical management,
and social functions that may not be feasible in the totally
decentralized observations units.
Hospitals may use different combinations of the basic
nursing unit configurations in ICUs.6 The relation of the
support or service areas to nursing units differs from unit to
unit. In an exploratory study Zborowsky et al.61 investigated
how nursing station design (i.e., centralized and decentralized
nursing station layouts) affected nurses’ use of space, patient
visibility, noise levels, and perceptions of the work environ-
ment. They concluded that the “hybrid” nursing design model
in which decentralized nursing stations are coupled with
centralized meeting rooms for consultation between staff
members may strike a balance between the increase in com-
puter duties and the ongoing need for communication and
consultation that addresses the conflicting demands of tech-
nology and direct patient care. In another recent study, Hen-
drich et al. examined how nurses adopt distinct movement
strategies based on features of unit topology and nurse assign-
ments and observed that the spatial qualities of nurse assign-
ments and unit layout affect nurse strategies for moving
through units and affect how frequently nurses enter patient
rooms and the nurse station.62 Therefore how various design
features of nursing unit and support space configuration
affect staff stress and effectiveness, staff communication, and
task performance meets the needs of a particular ICU need to
be considered.
Staff Stress and Effectiveness
The location of supplies and equipment and of electronic
charting impact the time spent in patient care by nurses.
Nurses spend a lot of time walking, which includes the time
to locate and gather supplies and equipment and to find other
staff members.63,64 Studies suggest that bringing staff and sup-
plies physically and visually closer to the patient reduces the
time nurses spend walking about the unit.65,66 Decentralized
nurses’ stations and supplies’ servers next to patient rooms
allows nurses to spend more time in direct patient care
activities.25-27
Section I—Background 19
Nurses may spend between 15% and 25% of their time in
charting.67-69 Computers for charting may be found at the
nursing observation units next to the patient room, in the
patient room, on mobile carts, and/or at the central nursing
stations away from the patient room. When charting is not
done at or near the bedside nurses may spend more time away
from the bedside to prepare and store charts at other locations
or make more mistakes in charting because of memory lapses
between the time of information collection at the bedside and
putting it on the chart.
Several factors contribute to staff stress in the ICU. Among
them is noise that can be associated with emotional exhaus-
tion and burnout among ICU nurses.70,71 Excessively high
noise levels in healthcare settings and frequent interruptions
also can interfere with work.72 For example, in a recent time-
and-motion study that included 40 doctors for more than 210
hours, Westbrook et al. observed that interruptions led doctors
to spend less time on the tasks they were working on and, in
nearly a fifth of cases, to give up on the task.73
Patients in NCCUs often have numerous monitoring and
recording devices with alarms that may also contribute to staff
stress and fatigue. Nurses may become insensitive to alarms,
because every alarm may not need immediate attention. There-
fore they can miss that important alarm which requires imme-
diate attention. For example, an investigation by the Boston
Globe suggested that 216 deaths from 2005 to 2010 nationwide
were associated with monitor alarm problems.74 Device-related
incidents may be underreported and it is possible that the
number of adverse events associated with alarm problems is
higher. According to the Globe’s investigation, the deaths listed
in the U. S. Food and Drug Administration (FDA) data were
linked to problems with alarms on patient monitors that track
heart functions, breathing and other vital signs. The problem
typically was not a broken device. Instead, most cases occurred
because medical staff did not notice an alarm or react with
urgency. This includes not hearing the alarm, ignoring an
alarm, mis­programming complicated monitors or forgetting
to turn them on.
The interface between technology and patients also can
challenge NCCU staff and contribute to stress. Often, patient
monitor and access lines are built up with multiple ports and,
even with proper training and instructions it may difficult to
sort out where to put this or that connector. ICU nurses also
need to ensure that medications are compatible and so need
to know how to control the stopcock. It is expected that tech-
nology innovations will help overcome some of the these
interface related problems, e.g., each port can be made unique
to help eliminate mixing up of lines, devices can be created to
help identify incompatible drugs or integrated printers in
every patient room for prescriptions may help eliminate
medical errors. Together, several simple fail-safe devices may
remove some of the cognitive load that contributes to stress
among nurses.75 For example, Kobayashi et al. used simple
human factors engineering principles to develop an experi-
mental chart binder system with alternating color-based
chart groupings, simple and prominent identifiers, and
embedded visual cues.76 This was associated with a significant
reduction in chart binder location problems, so contributing
to safe patient care delivery. Similarly Blomkvist et al.,77 who
examined the effects of changing the acoustic conditions
(sound-absorbing versus sound-reflecting ceiling tiles) in a
coronary ICU, observed that there were positive staff
20 Section I—Background
outcomes, including improved speech intelligibility and
reduced perceived work demands, pressure and strain during
the periods of improved acoustic conditions.
Staff Communication
Staff communication is a major variable in health care and
ICU safety and several studies demonstrate that a high degree
of involvement and interaction among caregivers can influ-
ence patient outcomes78 and length of stay.79 For example,
Baggs et al.80 examined interdisciplinary collaboration and
patient outcomes in a medical ICU and observed that patients
had a 5% chance of death or readmission where nurses
believed they had worked successfully with medical residents.
The risk was tripled when the residents made decisions about
patient care without adequate nurse consultation. The value
of better communication is strongest in the very sick, complex
patients.81 Research in “Magnet” hospitals also indicates that
healthy collaborative relationships among caregivers are pos-
sible and appear linked to optimal patient outcome.82 A
growing body of literature in several environmental design
research areas including offices, laboratories, housing com-
plexes, and acute care health facilities show that the physical
design of an environment may affect communication, interac-
tion, and/or or collaboration (For a review of the literature,
see Ulrich et al.3; Rashid83; Elsbach and Pratt84; Rashid and
Zimring.85) When designing staff areas in ICUs, the following
should be considered: (1) Physical design can affect the quality
and the quantity of interaction.86 (2) Location of people and
activity and physical distance among workers may be linked
to their informal communication.87 (3) Proximity of work-
spaces may predispose to the development of an informal
group among compatible people as an outgrowth of the infor-
mal communication associated with proximity.87 (4) Spatial
arrangement including the location of functions, walls, parti-
tions, furnishings, and other barriers may affect cohesiveness
and interaction among groups.88 (5) Visibility and accessibility
play a powerful role in the way individuals perceive and use
workplaces and communicate within.89 (6) Time spent in
walking by staff may be related to time spent in patient care
activities including nurse-physician interactions.17
Lighting Conditions and Task Performance
There are few studies on the effects of lighting conditions on
task performance among ICU staff. Studies in other work set-
tings show that staff may make more mistakes in inappropri-
ate lighting conditions, and performance on visual tasks gets
better as light levels increase.90 For example, medication dis-
pensing errors among hospital workers are more frequent
when daylight hours are fewer91; these errors can be reduced
at an illumination level greater than the baseline level of 45
foot candles.92 Studies in offices also indicate the importance
of appropriate lighting levels for complex tasks that require
excellent vision.93
Spaces for Families
Many patients in the NCCU cannot communicate for them-
selves. Therefore family members have an important role as
surrogate decision makers.94-96 Family members can also help
patients (1) perform daily functions, (2) understand concerns
about health, (3) foster a link to the environment, (4) reinforce
self-esteem, and (5) enhance positive relationships by offering
love and comfort.97 In addition, families can help busy nurses
and physicians.98,99
Location of Family Space
The location of family waiting spaces in ICUs has both prac-
tical and symbolic importance. Having family members
nearby often helps to shift an ICU’s culture and make fami-
lies a greater part of decision making (though sometimes this
also requires clinicians to create structured ways of dealing
with stressed family members). Symbolically the presence of
family spaces located outside the unit may suggest that fami-
lies are not integrated with patient care, whereas those pro-
vided within the patient room may indicate that families are
integrated with patient care. Depending on their accessibility
and comfort, family spaces provided within the unit but not
in the patient room can suggest the family role is in a state of
flux in the unit. These impressions can, at times, be wrong.
Some units may still not allow families to play an active role
in patient care even with families present in the patient
room. Hospitals should consider providing family spaces at
several of these locations for different functions and ameni-
ties.6 (For a discussion on the possible sociologic implica-
tions of different locations of family spaces in ICUs, see
Rashid, 2006.6)
Family Waiting Area Layout
Family waiting areas may be broken down into zones with
varying degrees of privacy and control similar to a home. The
number of ICU patient rooms, the availability of family space
in the patient room, the average length of patient stay, and the
types of amenities are some of the issues to consider to deter-
mine the size of a common waiting space. Waiting areas should
be divided into sections to provide more intimate and quieter
resting spaces and relatively busy and noisy activity spaces.
Solid partitions, dividers, glass walls, or planters can separate
each section according to the need of these spaces. Each
section should contain comfortable chairs or sofas for the
family, and resting and activity spaces should include private
spaces or booths for telephone conversations. Activity spaces
should include: (1) computers with Internet access that allow
families to access these computers to stay up-to-date with
patient status and the outside world and (2) study carrels with
health care information for families. There should be a media
room to separate the television from the rest of the waiting
area, so families who wish to have quiet and solitude are not
disturbed. Families with children should be given spaces sepa-
rate from a “quiet zone” for adults only. A play area for chil-
dren within the direct visual reach of the adult family members
should be considered in the area designated for families with
children. Some hospitals provide family sleep rooms with
private bathrooms, kitchenettes, and laundry in their waiting
areas for family members who must stay at the hospital for an
extended period.
Family Space in Patient Rooms
When possible, each patient room should include a well-
defined family area to allow families to be present for shift
change report, teaching sessions, care-planning discussions,
and daily medical rounds. A family space within a patient’s

room provides families a more comfortable environment for
activities, and being able to sleep in the patient’s room pro-
vides social support and reassurance for the patient and family
members.
Furnishings and Finishes
Furnishings in family areas should include comfortable seating
for the family and the option of a wall-mounted fold-down
bed or foldout chair-bed. Flexible furniture arrangements that
allow families to change furniture layout to meet their needs
are preferable, because seating arrangements that cannot be
changed or chairs that cannot be moved may cause frustration
among families. Unnecessary sources of visual stimulation
should be minimized and wall furnishings should not be of
bold patterns or colors that can be misperceived as threatening
objects (e.g., bugs, animals) by patients or their families. Wall
coverings and colors should be soothing and relaxing. In
general, the attractiveness of the physical environment in
waiting areas has been shown to be significantly associated
with higher perceived quality of care, less anxiety, and higher
reported positive interaction with staff.100
Access to Patients and Caregivers
Family spaces should provide easy visual or physical access to
patient rooms so family members can see the patient. Families
also should have easy access to caregivers when needed and
know when caregivers are available in the unit. For patient
safety it is better for families to enter the unit through a sepa-
rate entry other than the one used by service and clinical staff.
Designers and hospitals need to ensure that such a system of
entrances does not make interfaces among families and care-
givers difficult. If the ICU design restricts family-caregiver
interfaces, families may gather at places where they are likely
to find caregivers.
Staff-Family Communication
In ICUs, staff-family communication can provide emotional,
informational, and tangible supports to family members, and
can facilitate family members’ involvement in patient care.
Hospitals should consider the following to help improve
family-staff interactions and communication. (1) Central
nursing stations and glass partitions around the staff area can
limit family access to staff. (2) Decentralized nursing stations
may provide more opportunities for a nurse to spend time in
patient rooms. These stations also can be used during medical
rounds by medical teams to retrieve patients’ records. (3)
Private patient rooms and well-designed consultation rooms
may provide opportunities for confidential discussions. (4)
Within hallways, alcoves can provide private spaces for confi-
dential discussions. (5) Seating that is arranged side-by-side
along family space walls can discourage social interaction. (6)
Private and peaceful spaces can help improve communication.
(7) In dim lighting conditions and in rooms with softer floor
materials, people may interact longer.101-107
Noise Reduction in Family Space
Carpeting in corridors next to family waiting spaces can
reduce the sound of footsteps, rolling carts, staff member
conversations, and other common noises in ICUs, and
Section I—Background 21
high-performance sound-absorbing materials can be used to
reduce reverberation time, sound propagation, and noise
intensity levels. Storage areas, staff lounges, and utility rooms
also can be located away from patient rooms and family spaces
to reduce noise. Internal corridors between storage and utility
rooms can help clinical and support staff members perform
necessary tasks without disturbing patients or families.
Music in Family Space
Sometimes music can be used to mask distressing environ-
mental noise that cannot otherwise be eliminated. Sounds of
nature accompanied by soft music also can be used in family
waiting areas to calm anxious families or visitors. However,
not all music can produce a desired calming effect. Music
often evokes emotions and feelings that are rooted in an indi-
vidual’s past experiences and personal preferences.108 Thus it
is essential to respect music preferences of family members
and provide them choices.
Artwork in Family Space
Appropriate artwork can help reduce stress among patient
families.109 The choice of artwork needs to be sensitive to
culture, religion, the specific geographic area, and the interior
design scheme. Hospitals should consider developing systems
that allow artwork to be changed by the patient family as easily
as changing television channels. Because art varies enormously
in subject matter and style, not all artwork is suitable for high-
stress health care spaces.
Lighting in Family Space
Appropriate lighting can influence mood or create a relaxed
ambience. Therefore attention should be given to make
natural light available in all family spaces in the unit. When
the family space is within the patient room, it is necessary to
make sure that the intensity of natural light is comfortable
to patients. The use of slightly tinted or reflective glass can
reduce glare and heat production from sunlight. Vertical
blinds and other window treatments can be used to adjust
light intensity as desired by the patient. For artificial light in
family spaces, it is important to consider multiple lighting
options that can be controlled by the family when appropri-
ate. Where natural light is not an option (e.g., older ICUs),
full-spectrum fluorescent lighting can be used for compara-
ble benefits. A dynamic lighting solution that allows the
color and temperature levels to be changed according to the
time of day may be suitable for a patient room or family
space.
Odors and Aromas
Pleasing aromas can help reduce blood pressure, slow the rate
of respiration, lower pain perception levels, improve the
immune system, and help increase a sense of well-being
among family members who are under severe mental and
physical stress. In contrast, odors (“negative smells”) may
stimulate anxiety, fear, and stress.110 Hospitals, particularly
ICUs, are well known for their unpleasant odors or chemical
smells. If possible, strong-smelling cleaning agents should be
avoided near family areas. Aroma should be used with caution
in ICU family areas.
22 Section I—Background

high-quality care. Consequently a larger NCCU was proposed,

Nature, Spirituality, and Religion
Nature can have a positive effect on physical and emotional
well-being; hence it is preferable to design family areas with
windows to the outside. If possible these spaces need to be
close to hospital gardens with plants, water, and other natural
objects. When family spaces do not have a view of or access to
nature, nature may be brought into the unit (e.g., potted
plants, sound of nature, and nature-related artwork). Hospi-
tals also should consider outdoor labyrinths in gardens as a
focus for spirituality.111-113
A working knowledge of common cultural and religious
needs of patients and families is required to design family
spaces. Sometimes a focus group with local spiritual and
ethnic leaders before ICU design or redesign may help identify
common design concerns of these groups. It may help families
to have a community room for tai chi, yoga, and other spiri-
tual modalities and a chapel or a sanctuary close to the ICU
where religious services can take place. Sometimes these places
are the only quiet refuge for families from the chaos of the
hospital. Hospitals should also provide religious books, inspi-
rational texts, and texts on grief and coping written in mul-
tiple languages in the chapel or the community room, and
consider having space for common religious items used by
people of different faiths (e.g., rosaries, crucifixes, and holy
water for Catholic users; clean clothing, prayer rugs, and com-
passes for Muslim users; Sabbath kits for Jewish users) in the
chapel or the community room. When building a new unit, a
tile marker could be inserted in all family spaces and patient
rooms to signify the direction of prayer.
initially to appeal to the bottom line: if the unit had more beds
and could attract more patients, it could generate more
revenue and, most important, fulfill staff ’s mission as the hos-
pital of last resort for many of these patients with complex
brain injury. In 2007 Emory Medical Center was in the early
phase of planning a replacement hospital, and the administra-
tion did not want to spend a lot of money on a new ICU that
would be demolished in 5 years when the new hospital opened.
Early on, the new NCCU was described as the “throwaway”
NCCU, and the hospital agreed to meet the state and federal
requirements so that it could quickly open to meet the imme-
diate ICU bed shortages.
The initial design was for a 24-bed ICU with a “track”
around it; visitors would enter the patient rooms from the
back so as not to disrupt the central area used by the doctors
and nurses. The rooms measured 200 square feet, as required
by the state of Georgia, with no dedicated space for family
members. This design essentially duplicated the current ICUs
at that time. In these units the typical patient room was so
crowded with specialized equipment that it was difficult to get
to the patient without tripping over cords or knocking out
invasive lines Figure 3.1 shows a comparison of the ICU
before and after construction. It took time to respond to an
emergency and to maintain sterility was near impossible.
During patient care rounds there was little room for the ICU
team’s numerous members in crowded rooms and hallways.
In the central areas, nurses were crowded around desks with
charts spread all over tables; this increased the potential for
mistakes in documentation, record keeping, and medication
administration. Families were limited to dark, common spaces

Case Study: Emory University

Hospital Neuroscience Critical
Care Unit
In 2005 an architecture firm that worked with the Emory
Hospital in Atlanta, Georgia, presented a design to replace the
NCCU. The unit’s medical director (senior author Owen
Samuels, MD) was concerned that the intended design might
not address some of the issues described earlier that could
affect patient outcomes. Dr. Samuels approached Craig
Zimring of Georgia Institute of Technology School of Archi-
tecture about published evidence on the effects of ICU design.
Dr. Zimring agreed to have his graduate students, under the
guidance of Dr. Rashid, search the literature and develop
design concepts based on research findings. This exercise ulti-
mately led to an interdisciplinary design charrette with the
architectural firm and a revised NCCU design that included
the design implications from the literature. This section
reviews how the NCCU was designed primarily from the per-
spective of the medical director.

Evidence for a Better Way

Neurocritical care admissions at Emory increased from 587
patients in 1999 to more than 1400 patients in 2007. This
rapid growth meant that in 2007 the existing NCCU was out
of beds and patients with severe neurologic disorders were
cared for in three geographically separated small units (two
seven-bed units, and one nine-bed unit). This approach was Fig. 3.1  Top panel, Crowded initial conditions. Bottom panel, Completed
inefficient and compromised patient safety and delivery of project.
 Section I—Background 23

in the outside hall away from patients and were restricted to

visiting during morning rounds. Discussions between doctor Table 3.1  Project Goals and Metrics
and families, including those about prognosis, brain death, for ICUD Design
organ donation, and end-of-life concerns took place either in Outcome
the cluttered patient rooms or in public hallways with no Design Drivers Design Response Measures
privacy. The new proposed space promised little more than Support families Family zone in Greater
some new converter chairs. patient room satisfaction on
There was concern that the proposed design was not ideal. Kid’s room Press-Ganey and
To convince the administration to pursue a completely new Lockers and Emory ICU
concept, designers focused on key people: the chief nursing showers surveys
Family quiet room Fewer complaints
officer for neurosciences and the chief executive officer of and litigation
Emory Healthcare. The design team told them that current
Support more Medical gas booms Fewer patient
ICUs were uncomfortable for families, the space was inher- procedures at Larger patient transfer
ently dangerous with a large potential for avoidable medical the bedside zone complications
mistakes that was largely unrecognized, and staff burnout was Improved and lower costs
common. Doctors and nurses make clinical decisions based in ergonomics Fewer errors
part on evidence from the literature. Shouldn’t such evidence Shorter stays
More time spent
also inform how hospitals and ICUs are designed? Although by the ICU staff
research in ICU design is an emerging field, there is a large in the ICU area
body of scientific evidence on how the physical environment Reduce infection Numerous rubs Improved
affects patient outcome, staff effectiveness, family well-being, and handwashing
and costs. The team therefore proposed a new design founded handwashing compliance
on an evidence-based approach for patient and family- stations Lower MSRA and
centered care supported by a healing environment but flexible hospital-
acquired
enough that it could continue to change in the future. Devel- infection rates
opers were confident that an improved design could reduce
Reduce medical Improved ceiling Fewer medical and
staff stress and enhance performance. As an academic institu- errors and tiles medication
tion, the team also wanted to study the effect of a new type increase patient Carpet where errors
of ICU. safety appropriate Less litigation
Charting niches Reduced
Zoned caregiver self-extubation
Emphasis About Family Involvement area Decreased falls
and injuries
Many factors other than technologic advancement can con- related to
tribute to patient outcome. When the ICU was designed, the patients leaving
team had several goals or “design drivers” (Table 3.1) in mind, beds
with accompanying measurable outcome variables to be ICU, Intensive care unit; MRSA, methicillin-resistant Staphylococcus aureus.
tracked. A primary driver for the new ICU was family support.
It was proposed to eliminate signs that restricted family visita-
tion, and the team was tempted to replace the sign “Physician
Rounds in Progress” with “Physician Rounds in Progress, American Association of Critical Care Nurses, and the Ameri-
Family Presence Encouraged.” The plan included a family zone can Institute of Architects jointly give this award. Next the
located within the patient’s room, as the “family studio,” a team partnered with the division of health care design at
children’s area located immediately outside the ICU, family Georgia Institute of Technology’s College of Architecture, led
lockers and showers, clothing washers and dryers, and a quiet by an environmental psychologist who specializes in the use
room for families and close friends. Outcome measures to be of architectural design as a healing tool. Several charrette ses-
collected included patient and family satisfaction scores, sions were held, which included key stakeholders from Emory,
hospital-acquired infections, nursing retention rates, ICU the architects (HKS, Inc.), and Georgia Tech.
length of stay, organ donation success, and number of litiga- A mock-up of the proposed ICU that included nurses’
tion filings. This family-centered approach was to be balanced station, patient rooms, wall-to-ceiling booms, and family areas
with the ability to turn the patient ICU room into a “mini- was created. Several procedures, including resuscitation, intu-
operating room” to reduce patient transfers and support more bation, intracranial monitor implantation, nursing shift-
procedures at the bedside. Other important design drivers change, and interactions between families and staff were
were reduction of medical errors; increased patient safety, and enacted through role-playing and recorded by videographers
staff satisfaction. Each goal had measurable outcomes to be for later analysis. A mock-up of the family studio allowed
tracked. analysis of functionality and family flow. Family members of
patients who had recently been discharged from or were still
in the ICU were involved with the unit’s design throughout
A Dynamic Design Process the process. As developers learned from such experiences, the
To determine factors such as patient room size and configura- design of the ICU was altered even as construction was under
tion and the design of family spaces, the design team analyzed way. It was originally planned to distribute the nurses’ stations
best practices of the prior 10 years’ winners of the ICU Design throughout the unit, but later it was decided to keep a com-
Citation Award. The Society of Critical Care Medicine, the munal area as well, because the team realized that nurses and
24 Section I—Background
8.5m
(28’0”)
5.3m
(17’3”)
Fig. 3.2  Illustration of two adjacent ICU patient
rooms (shaded blue). The nurse’s station outside
the patient rooms and the adjoining family
areas (shaded green) also are visible. Room
dimensions are provided.
Fig. 3.3  Patient room, view from above.
doctors at times need to be in the same area to better support
one another.
Proposal Becomes Reality
The new NCCU opened February 2007. The rooms are
between 345 and 450 square feet (Fig. 3.2) whereas the old
rooms were 120 to 200 square feet. Pneumatic booms lift
equipment off the floor. This limits clutter and allows rapid
360-degree access to the patient’s head in emergency situa-
tions. The beds and doors are configured so that patients who
are awake have a direct line of sight to the nurses’ station or a
bay window. Each room is a suite that consists of the patient
room and a family area separated by a curved wall with large
glass-block windows that lets in natural light (Fig. 3.3). The
2.6m
(8’7”)
1.5m
(4’9”)
1.5m
(4’9”)
1.6m
(5’4”)
2.9m
(9’5”)
4.5m (15’) 9.1m (30’)
family studio has a table, chairs, comfortable sleeping arrange-
ments, flat-screen television, wireless Internet access, music,
and a white-noise system to blunt surrounding noises. The
new unit allows staff to do things that could not be done
before including hold a private conversation with a family
member when visiting a patient. Family members can leave
the room for some respite and still be just a stone’s throw away
from their loved one.
The following case that bridged the transition between
the old and new NCCU illustrates some of the advantages of
the new family-centered unit. David was a 31-year-old com-
puter programmer, the father of a 3-year-old girl, and about
to be married. He was admitted to the NCCU with a grade
IV subarachnoid hemorrhage and was in the old ICU for 4
days, and then moved to the new NCCU when it opened.
David later developed neurogenic pulmonary edema, severe
pneumonia, acute respiratory distress syndrome, and heart
failure and required induced coma for intracranial hyperten-
sion. He subsequently had an aneurysm rebleed and pro-
gressed to brain death and his family decided to donate his
organs.
His family, parents, and his fiancée kept a rotating vigil 24
hours a day. They always felt they were in the way in the old
ICU, whereas they felt welcome in the new facility. The family
often stood at David’s bedside as the team explained the
purpose of the complex monitors and instruments and did
not have to leave the bedside for discussions about brain death
or organ donation. The mother said, “This was our home for
a month, and it got so that the nurses could tell when we
needed a hug.”114 David’s father said, “No one ever misled us
or told us anything but the truth … and most importantly, we
were there for everything.”115
Hospital staff did everything they could for David, and
nothing could change his ultimate outcome. But the way

someone dies is important. The circumstances of how David
was treated probably helped allow the family to donate his
organs and better come to terms with his death. They later
generously donated their time to help the NCCU develop the
family-centered approach by participating in many discus-
sions about their experiences.
Family-Centered Units
Pose Challenges
The concept of care around the patient and their family makes
sense. However, there is resistance to it, even from the most
dedicated health care workers. The team at Emory spent about
a year and a half preparing for family-centered care; starting
the process was the real challenge. In a unit that is designed for
both patients and their families, how does one care for patients
and their families simultaneously? There are several challenges:
 Team rounding: Doctors and nurses may be apprehensive
about inviting families to rounds. Training doctors and
nurses with families present is a real paradigm shift and
raises many controversial issues.
 Nurses in view: Imagine a nurse operating six or seven
intravenous pumps and trying to figure out medications
while having a family member—or three or four
members—continuously present.
 Urgent or frightening treatment: How does staff deal with
resuscitation? What if the family is right by the bedside?
Should they be asked to leave? What kind of support do
they need?
The authors do not have all the answers to such problems but
are currently studying them to figure out best practices.
Outcomes
Emory’s NCCU won the 2008 ICU Design Citation Award
from the Society of Critical Care Medicine, the American
Association of Critical Care Nurses, and the American Insti-
tute of Architects Academy on Architecture for Health. The
authors have started to look at outcomes associated with the
NCCU. Patient and staff satisfaction have increased, according
to self-assessments. Most recent Press-Ganey Family Satisfac-
tion scores indicate greater than 85% overall satisfaction.
Other variables of quality care benchmarks include significant
decrease in pneumonia rates and central venous line blood-
stream infections, increased compliance with critical care
pathways, increased nursing retention, and a trend toward
decreased medical legal litigation.
Conclusion
This chapter has briefly discussed environmental design crite-
ria for the ICU, the patient room, staff and support areas, and
family areas of ICUs to help improve patient, staff, and family
outcomes. These processes can influence patient outcomes
and have an effect on how staff monitor individual patients
and overall outcomes and quality in the ICU. Although some
of these issues will likely persist for the foreseeable future,
others are being addressed by technical and process changes.
For example, as manufacturers of medical equipment and
supplies continue to merge and consolidate and develop
Section I—Background 25
standardized protocols for integration of medical equipment
and devices, many interface problems related to technology
that we see in ICUs today may be eliminated.
References
1. Bartley J, Streifel AJ. Design of the environment of care for safety of patients
and personnel: does form follow function or vice versa in the intensive care
unit? Crit Care Med 2010;38(Suppl. 8):388–98.
2. Ulrich RS, Zimring C, Joseph A, et al. The role of the physical environment
in the hospital of the 21st century: a once-in-a-lifetime opportunity.
Concord, CA: The Center for Health Design; 2004.
3. Ulrich RS, Zimring C, Zhu X, et al. A review of the research literature on
evidence-based healthcare design. HERD 2008;1(3):61–125.
4. Zimring C, Ulrich RS, Joseph A, et al. The environment’s impact on safety.
In: Marberry SO, editor. Improving healthcare with better building design.
Chicago: Health Administration Press; 2006. p. 63–81.
5. Cai H, Do EY-L, Zimring CM. Extended linkography and distance graph in
design evaluation: an empirical study of the dual effects of inspiration
sources in creative design. Design Studies 2010;31(2):146–68.
6. Rashid M. A decade of adult intensive care unit design: a study of the
physical design features of the best-practice examples. Crit Care Nurs Quart
2006;29(4):282–311.
7. Kohlenberg A, Schwab F, Behnke M, et al. Screening and control of
methicillin-resistant Staphylococcus aureus in 186 intensive care units:
different situations and individual solutions. Critical Care 2011;15:R285.
8. Vincent JL, Bihari DJ, Suter PM, et al. The prevalence of nosocomial
infection in intensive care units in Europe: results of the European
Prevalence of Infection in Intensive Care (EPIC) Study. EPIC International
Advisory Committee. JAMA 1995;274(8):639–44.
9. Akansel N, Kaymakçi S. Effects of intensive care unit noise on patients: a
study on coronary artery bypass graft surgery patients. J Clin Nurs
2008;17(12):1581−90.
10. Piergeorge AR, Ceserano FL, Casanova DM. Designing the critical care unit:
a multidisciplinary approach. Crit Care Med 1983;11(7):541–45.
11. Society of Critical Care Medicine (SCCM). NIH consensus development
conference on critical care medicine. Crit Care Med 1983;11:466–9.
12. Society of Critical Care Medicine (SCCM). Guidelines for intensive care unit
design. Crit Care Med 1995;23(3):582–8.
13. Stoddard JC. Design, staffing and equipment requirements for an intensive
care unit. Int Anesthesiol Clin 1981;19(2):77–95.
14. Hamilton DK, editor. ICU 2010: ICU design for the future, a critical care
design symposium. Houston: Center for Innovation in Health Facilities;
2000.
15. Carthey J. Reinterpreting the hospital corridor: “wasted space” or essential
for quality multidisciplinary clinical care? HERD 2008;2(1):17–29.
16. National Fire Protection Association. Section 19.2.5.7.2.3. Quincy, MA; 2012.
17. Hendrich A. Case study: the impact of acuity adaptable rooms on future
designs, bottlenecks and hospital capacity. Paper presented at the Impact
Conference on Optimizing the Physical Space for Improved Outcomes,
Satisfaction and the Bottom Line, Atlanta, 2003.
18. Hendrich A, Fay J, Sorrells A. Effects of acuity-adaptable rooms on flow of
patients and delivery of care. Am J Crit Care 2004;13(1):35–45.
19. Hendrich A, Lee N. Intra-unit patient transports: time, motion, and cost
impact on hospital efficiency. Nurs Econ 2005;23(4):157–64.
20. Cook RI, Render M, Woods DD. Gaps in the continuity of care and progress
on patient safety. BMJ 2000;320(7237):791–4.
21. Jastremski CA, Harvey M. Making changes to improve the intensive care unit
experience for patients and their families. New Horiz 1998;6(1):99–109.
22. Hamilton DK, Shepley MM. Design for critical care: an evidence based
approach. Oxford: Elsevier Ltd; 2010.
23. Tanja-Dijkstra K. The impact of bedside technology on patients’ well being.
Health Environments Research & Design Journal 2011;5(1):43−51.
24. Walch JM, Rabin BS, Day R, et al. The effect of sunlight on post-operative
analgesic medication usage: a prospective study of spinal surgery patients.
Psychosom Med 2005;67(1):156–63.
25. Cadenhead C, Anderson D. Critical care unit design, the winners and future
trends: an investigative study. World Health Design Journal 2009;2:72–7.
26. Chaudhury H, Mahmood A, Valente M. Advantages and disadvantages of
single-versus multiple-occupancy rooms in acute care environments: a
review and analysis of the literature. Enviro Behav 2005;37:760–86.
27. Levin PD, Golovanevski M, Moses AE, et al. Improved ICU design reduces
acquisition of antibiotic resistant bacteria: a quasi-experimental
observational study. Critical Care 2011;15:R21.
26 Section I—Background
28. Maki DG. Nosocomial infection in the intensive care unit. In: Parillo JE,
Bone RC, editors. Critical care medicine-principles of diagnosis and
management. St. Louis: Mosby; 1995. p. 893–954.
29. Shirani KZ, McManus AT, Vaughan GM, et al. Effects of environment on
infection in burn patients. Arch Surg 1986;121(1):31–6.
30. Teltsch DY, Hanley J, Loo V, et al. Infection acquisition following intensive
care unit room privatization. Arch Intern Med 2011;171(1):32–8.
31. Van Enk RA, Steinberg F. Comparison of private room with multiple-bed
ward neonatal intensive care unit environments. Health Environments
Research & Design Journal 2011;5(1):52–63.
32. France DJ, Throop P, Walczyk B, et al. Does patient-centered design
guarantee patient safety? Using human factors engineering to find a balance
between provider and patient needs. J Patient Saf 2005;1(3):145–53.
33. Burrington M, Architects WHR. An alternative method for human waste
disposal in critical care. In Hamilton DK, editor. ICU 2010: ICU design for
the future, a critical care design symposium. Houston: Center for Innovation
in Health Facilities; 2000.
34. Mullin B, Rouget C, Clément C, et al. Association of private isolation rooms
with ventilator-associated Acinetobacter baumanii pneumonia in a surgical
intensive-care unit. Infection Control and Hospital Epidemiology
1997;18(7):499–503.
35. Facilities Guidelines Institute. Guidelines for design and construction of health
care facilities. Washington, DC: The American Institute of Architects; 2006.
36. Joseph A. The impact of the environment on infections in healthcare
facilities. Concord, CA: The Center for Health Design; 2006.
37. Boyce, J. Antiseptic technology: access, affordability, and acceptance.
Emerging Infectious Diseases 2001;7(2):231–3.
38. Lankford MG, Zembower TR, Trick WE, et al. Influence of role models and
hospital design on hand hygiene of healthcare workers. Emergency
Infectious Diseases 2003;9(2):217–23.
39. Larson E, Albrecht S, O’Keefe M. Hand hygiene behavior in a pediatric
emergency department and a pediatric intensive care unit: comparison of
use of 2 dispenser systems. American Journal of Critical Care
2005;14(4):304–11.
40. Kaplan LM, McGuckin M. Increasing handwashing compliance with more
accessible sinks. Infection Control 1986;7(8):408–10.
41. Vernon MO, Trick WE, Welbel SF, et al. Adherence with hand hygiene: does
number of sinks matter? Infection control and hospital epidemiology
2003;24(3):224–5.
42. Vietri NJ, Dooley DP, Davis CE, et al. The effect of moving to a new hospital
facility on the prevalence of methicillin-resistant Staphylococcus aureus.
American Journal of Infection Control 2004;32(5):262–7.
43. Trick W, Vernon M, Welbe LS, et al. Multicenter intervention program to
increase adherence to hand hygiene recommendations and glove use and to
reduce the incidence of antimicrobial resistance. Infection Control and
Hospital Epidemiology 2007;28(1):42–9.
44. Dyson M. Intensive care unit psychosis, the therapeutic nurse-patient
relationship and the influence of the intensive care setting: analyses of
interrelating factors. Journal of Clinical Nursing 1999;8:284–90.
45. Beauchemin KM, Hays P. Sunny hospital rooms expedite recovery from
severe and refractory depressions. Journal of Affective 1996;40(1-2):
49–51.
46. Benedetti F, Colombo C, Barbini B, et al. Morning sunlight reduces length of
hospitalization in bipolar depression. Journal of Affective Disorders
2001;62(3):221–3.
47. Lewy AJ, Bauer VK, Cutler NL, et al. Morning vs. evening light treatment of
patients with winter depression. Archives of General Psychiatry
1998;55(10):890–6.
48. Ulrich RS. View through a window may influence recovery from surgery.
Science 1984;224(4647):420–1.
49. Ulrich RS. Effects of interior design on wellness: theory and recent scientific
research. Journal of Health Care Design 1991;3:97–109.
50. Gelling L. Causes of ICU psychosis: the environmental factors. Nursing in
Critical Care 1999;4:22–6.
A complete list of references for this chapter can be found online at
www.expertconsult.com.

References
1. Bartley J, Streifel AJ. Design of the environment of care for safety of
patients and personnel: does form follow function or vice versa in the
intensive care unit? Crit Care Med 2010;38(Suppl. 8):388–98.
2. Ulrich RS, Zimring C, Joseph A, et al. The role of the physical environment
in the hospital of the 21st century: a once-in-a-lifetime opportunity.
Concord, CA: The Center for Health Design; 2004.
3. Ulrich RS, Zimring C, Zhu X, et al. A review of the research literature on
evidence-based healthcare design. HERD 2008;1(3):61–125.
4. Zimring C, Ulrich RS, Joseph A, et al. The environment’s impact on safety.
In: Marberry SO, editor. Improving healthcare with better building design.
Chicago: Health Administration Press; 2006. p. 63–81.
5. Cai H, Do EY-L, Zimring CM. Extended linkography and distance graph in
design evaluation: an empirical study of the dual effects of inspiration
sources in creative design. Design Studies 2010;31(2):146–68.
6. Rashid M. A decade of adult intensive care unit design: a study of the
physical design features of the best-practice examples. Crit Care Nurs Quart
2006;29(4):282–311.
7. Kohlenberg A, Schwab F, Behnke M, et al. Screening and control of
methicillin-resistant Staphylococcus aureus in 186 intensive care units:
different situations and individual solutions. Critical Care 2011;15:R285.
8. Vincent JL, Bihari DJ, Suter PM, et al. The prevalence of nosocomial
infection in intensive care units in Europe: results of the European
Prevalence of Infection in Intensive Care (EPIC) Study. EPIC International
Advisory Committee. JAMA 1995;274(8):639–44.
9. Akansel N, Kaymakçi S. Effects of intensive care unit noise on patients: a
study on coronary artery bypass graft surgery patients. J Clin Nurs
2008;17(12):1581−90.
10. Piergeorge AR, Ceserano FL, Casanova DM. Designing the critical care unit:
a multidisciplinary approach. Crit Care Med 1983;11(7):541–45.
11. Society of Critical Care Medicine (SCCM). NIH consensus development
conference on critical care medicine. Crit Care Med 1983;11:466–9.
12. Society of Critical Care Medicine (SCCM). Guidelines for intensive care
unit design. Crit Care Med 1995;23(3):582–8.
13. Stoddard JC. Design, staffing and equipment requirements for an intensive
care unit. Int Anesthesiol Clin 1981;19(2):77–95.
14. Hamilton DK, editor. ICU 2010: ICU design for the future, a critical care
design symposium. Houston: Center for Innovation in Health Facilities;
2000.
15. Carthey J. Reinterpreting the hospital corridor: “wasted space” or essential
for quality multidisciplinary clinical care? HERD 2008;2(1):17–29.
16. National Fire Protection Association. Section 19.2.5.7.2.3. Quincy, MA;
2012.
17. Hendrich A. Case study: the impact of acuity adaptable rooms on future
designs, bottlenecks and hospital capacity. Paper presented at the Impact
Conference on Optimizing the Physical Space for Improved Outcomes,
Satisfaction and the Bottom Line, Atlanta, 2003.
18. Hendrich A, Fay J, Sorrells A. Effects of acuity-adaptable rooms on flow of
patients and delivery of care. Am J Crit Care 2004;13(1):35–45.
19. Hendrich A, Lee N. Intra-unit patient transports: time, motion, and cost
impact on hospital efficiency. Nurs Econ 2005;23(4):157–64.
20. Cook RI, Render M, Woods DD. Gaps in the continuity of care and
progress on patient safety. BMJ 2000;320(7237):791–4.
21. Jastremski CA, Harvey M. Making changes to improve the intensive care
unit experience for patients and their families. New Horiz 1998;6(1):
99–109.
22. Hamilton DK, Shepley MM. Design for critical care: an evidence based
approach. Oxford: Elsevier Ltd; 2010.
23. Tanja-Dijkstra K. The impact of bedside technology on patients’ well being.
Health Environments Research & Design Journal 2011;5(1):43−51.
24. Walch JM, Rabin BS, Day R, et al. The effect of sunlight on post-operative
analgesic medication usage: a prospective study of spinal surgery patients.
Psychosom Med 2005;67(1):156–63.
25. Cadenhead C, Anderson D. Critical care unit design, the winners and future
trends: an investigative study. World Health Design Journal 2009;2:72–7.
26. Chaudhury H, Mahmood A, Valente M. Advantages and disadvantages of
single-versus multiple-occupancy rooms in acute care environments: a
review and analysis of the literature. Enviro Behav 2005;37:760–86.
27. Levin PD, Golovanevski M, Moses AE, et al. Improved ICU design reduces
acquisition of antibiotic resistant bacteria: a quasi-experimental
observational study. Critical Care 2011;15:R21.
28. Maki DG. Nosocomial infection in the intensive care unit. In: Parillo JE,
Bone RC, editors. Critical care medicine-principles of diagnosis and
management. St. Louis: Mosby; 1995. p. 893–954.
Section I—Background 26.e1
29. Shirani KZ, McManus AT, Vaughan GM, et al. Effects of environment on
infection in burn patients. Arch Surg 1986;121(1):31–6.
30. Teltsch DY, Hanley J, Loo V, et al. Infection acquisition following intensive
care unit room privatization. Arch Intern Med 2011;171(1):32–8.
31. Van Enk RA, Steinberg F. Comparison of private room with multiple-bed
ward neonatal intensive care unit environments. Health Environments
Research & Design Journal 2011;5(1):52–63.
32. France DJ, Throop P, Walczyk B, et al. Does patient-centered design
guarantee patient safety?: Using human factors engineering to find a
balance between provider and patient needs. J Patient Saf 2005;1(3):
145–53.
33. Burrington M, Architects WHR. An alternative method for human waste
disposal in critical care. In Hamilton DK, editor. ICU 2010: ICU design for
the future, a critical care design symposium. Houston: Center for
Innovation in Health Facilities; 2000.
34. Mullin B, Rouget C, Clément C, et al. Association of private isolation
rooms with ventilator-associated Acinetobacter baumanii pneumonia in a
surgical intensive-care unit. Infection Control and Hospital Epidemiology
1997;18(7):499–503.
35. Facilities Guidelines Institute. Guidelines for design and construction of
health care facilities. Washington, DC: The American Institute of Architects;
2006.
36. Joseph A. The impact of the environment on infections in healthcare
facilities. Concord, CA: The Center for Health Design; 2006.
37. Boyce, J. Antiseptic technology: access, affordability, and acceptance.
Emerging Infectious Diseases 2001;7(2):231–3.
38. Lankford MG, Zembower TR, Trick WE, et al. Influence of role models and
hospital design on hand hygiene of healthcare workers. Emergency
Infectious Diseases 2003;9(2):217–23.
39. Larson E, Albrecht S, O’Keefe M. Hand hygiene behavior in a pediatric
emergency department and a pediatric intensive care unit: comparison of
use of 2 dispenser systems. American Journal of Critical Care
2005;14(4):304–11.
40. Kaplan LM, McGuckin M. Increasing handwashing compliance with more
accessible sinks. Infection Control 1986;7(8):408–10.
41. Vernon MO, Trick WE, Welbel SF, et al. Adherence with hand hygiene: does
number of sinks matter? Infection control and hospital epidemiology
2003;24(3):224–5.
42. Vietri NJ, Dooley DP, Davis CE, et al. The effect of moving to a new
hospital facility on the prevalence of methicillin-resistant Staphylococcus
aureus. American Journal of Infection Control 2004;32(5):262–7.
43. Trick W, Vernon M, Welbe LS, et al. Multicenter intervention program to
increase adherence to hand hygiene recommendations and glove use and to
reduce the incidence of antimicrobial resistance. Infection Control and
Hospital Epidemiology 2007;28(1):42–9.
44. Dyson M. Intensive care unit psychosis, the therapeutic nurse-patient
relationship and the influence of the intensive care setting: analyses of
interrelating factors. Journal of Clinical Nursing 1999; 8:284–90.
45. Beauchemin KM, Hays P. Sunny hospital rooms expedite recovery from
severe and refractory depressions. Journal of Affective 1996;40(1-2):49–51.
46. Benedetti F, Colombo C, Barbini B, et al. Morning sunlight reduces length
of hospitalization in bipolar depression. Journal of Affective Disorders
2001;62(3):221–3.
47. Lewy AJ, Bauer VK, Cutler NL, et al. Morning vs. evening light treatment of
patients with winter depression. Archives of General Psychiatry
1998;55(10):890–6.
48. Ulrich RS. View through a window may influence recovery from surgery.
Science 1984;224(4647):420–1.
49. Ulrich RS. Effects of interior design on wellness: theory and recent
scientific research. Journal of Health Care Design 1991;3:97–109.
50. Gelling L. Causes of ICU psychosis: the environmental factors. Nursing in
Critical Care 1999;4:22–6.
51. Hughes J. Hallucinations following cardiac surgery in a pediatric intensive
care unit. Intensive and Critical Care Nursing 1994;10:209–11.
52. Gabor J, Cooper A, Crombach S, et al. Contribution of the intensive care
unit environment to sleep disruption in mechanically ventilated patients
and healthy subjects. American Journal of Respiratory and Critical Care
Medicine 2003;167:708–15.
53. Busch-Vishniac I, West J, Barnhill C, et al. Noise levels in Johns Hopkins
Hospital. Journal of the Acoustical Society of America 2005;118:
3629–1645.
54. Cropp A, Woods L, Raney D. Name that tone: the proliferation of alarms in
the intensive care unit. Chest 1994;105:1217–20.
55. Ulrich RS, Lawson B, Martinez M. Exploring the patient environment: an
NHS estates workshop. London: The Stationery Office; 2003.
26.e2 Section I—Background
56. Joint Commission on Accreditation of Healthcare Organizations. Root
causes of patient falls. Sentinel event statistics [Graph] 2006 [cited 2007
7/22/07]; Available from: http://www.jointcommission.org/NR/rdonlyres/
FA5A080F-C259-47CC-AAC8-BAC3F5C37D84/0/se_rc_patient_falls.jpg
57. Drahota A, Gal D, Windsor J. Flooring as an intervention to reduce injuries
from falls in healthcare settings. Quality in Ageing and Older Adults
2007;8(1):3–9.
58. Healey F. Does flooring type affect risk of injury in older in-patients?
Nursing Times 1994;90(27):40–1.
59. Simpson AH, Lamb S, Roberts PJ, et al. Does the type of flooring affect the
risk of hip fracture? Age and Ageing 2004;33(3):242–6.
60. Nelson AL, Matz M, Chen F, et al. Development and evaluation of a
multifaceted ergonomics program to prevent injuries associated with
patient handling tasks. Journal of International Nursing Studies 2006;43:
717–33.
61. Zborowsky T, Bunker-Hellmich L, Morelli A, et al. Centralized vs.
decentralized nursing stations: effects on nurses’ functional use of space
and work environment. Health Environments Research & Design Journal
2010;3(4):19−42.
62. Hendrich A, Chow MP, Bafna S, et al. Unit-related factors that affect
nursing time with patients: spatial analysis of the time and motion study.
Health Environments Research & Design Journal 2009;2(2):5–20.
63. Burgio L, Engel B, Hawkins A, et al. A descriptive analysis of nursing staff
behaviors in a teaching nursing home: differences among NAs, LPNs and
RNs. The Gerontologist 1990;30(1):107–12.
64. Tucker A, Spear S. Operational failures and interruptions in hospital
nursing. Health Services Research 2006;41(3):643–62.
65. Shepley MM. Predesign and postoccupancy analysis of staff behavior in a
neonatal intensive care unit. Children’s Health Care 2002;31(3):237–53.
66. Shepley MM, Davies K. Nursing unit configuration and its relationship to
noise and nurse walking behavior: an AIDS/HIV unit case study. AIA
Academy Journal 2003. Available from: http://www.aia.org/aah/journal/
0401/article0404.asp.
67. Deane D, McElroy MJ, Alden S. Documentation: meeting requirements
while maximizing productivity. Nursing Economics 1986;4(4):174–7.
68. Pierpont GL, Thilgen D. Effect of computerized charting on nursing
activity in intensive care. Critical Care Medicine 1995;23(6):1067–73.
69. Quist BD. Work sampling nursing units. Nursing Management 1992;23(9):
50–1.
70. Morrison W, Haas E, Shaffner D, et al. Noise stress and annoyance in a
pediatric intensive care unit. Critical Care Medicine 2003;31(1):35–7.
71. Topf M, Dillon E. Noise-induced stress as a predictor of burnout in critical
care nurses. Heart and Lung 1988;17:247–50.
72. Bayo MV, Garcia AM, Garcia A. Noise levels in an urban hospital and
workers’ subjective responses. Archives of Environmental Health 1995;
50(3):247–51.
73. Westbrook JI, Coiera E, Dunsmuir WT, et al. The impact of interruptions
on clinical task completion. Qual Saf Health Care 2010;19(4):284−9.
74. Boston Globe. “Patient alarms often unheard, unheeded.” Feb. 11, 2011.
Available at: http://www.boston.com/lifestyle/health/articles/2011/02/13/
patient_alarms_often_unheard_unheeded/ (accessed Feb. 29, 2012).
75. Rashid M. Technology and the future of ICU design. Critical Care Nursing
Quarterly 2011;34(4):332–60.
76. Kobayashi L, Boss RM , Gibbs FJ, et al. Color-coding and human factors
engineering to improve patient safety characteristics of paper-based
emergency department clinical documentation. Health Environments
Research & Design Journal 2011;4(4):79−88.
77. Blomkvist V, Eriksen C, Theorell T, et al. Acoustics and psychosocial
environment in intensive coronary care. Occupational and Environmental
Medicine 2005;62(3):1–8.
78. Knaus WA, Draper EA, Wagner DP, et al. An evaluation of outcome from
intensive care in major medical centers. Ann Intern Med 1986;104:410–8.
79. Shortell SM, Zimmerman JE, Rousseau DM, et al. The performance of
intensive care units: does good management make a difference? Medical
Care 1994;32(5):508–25.
80. Baggs JG, Ryan SA, Phelps CE, et al. The association between
interdisciplinary collaboration and patient outcomes in a medical/intensive
care unit. Heart and Lung 1992;21(1):18–24.
81. Baggs JG, Schmitt M, Mushlin A, et al. Association between nurse-physician
collaboration and patient outcomes in three intensive care units. Critical
Care Medicine 1999;27(9):1991–8.
82. Kramer M, Schmalenberg C. Securing “good” nurse physician relationship.
Nursing Management 2003;34(7):34–8.
83. Rashid M. Hospital design and face-to-face interaction among clinicians: a
theoretical model. Health Environments Research & Design Journal
2009;2(4):62−84.
84. Elsbach KD, Pratt MG. The physical environment in organizations. The
Academy of Management Annals 2007;1(1):181−224.
85. Rashid M, Zimring C. On psychosocial constructs and office settings: a
review of the empirical literature. In Chaudhury H. editor. Proceedings of
the Environmental Design Research Association (EDRA) 36 Conference.
Vancouver, Canada: EDRA; 2005. p. 107–22.
86. Sundstrom E, Herbert RK, Brown DW. Privacy and communication in an
open-plan office: a case study. Environment and Behavior 1982;1:379–92.
87. Allen T. Managing the flow of technology. Cambridge, MA: MIT Press;
1977.
88. Bobele HK, Buchanan PJ. Building a more productive environment.
Management World 1979;8(1):8–10.
89. Hillier B, Penn A. Visible colleges: structure and randomness in the place of
discovery. Science in Context 1991;4(1):23–49.
90. Boyce P, Hunter C, Howlett O. The benefits of daylight through windows.
Troy, New York: Rensselaer Polytechnic Institute; 2003.
91. Roseman C, Booker JM. A seasonal pattern of hospital medication errors in
Alaska. Psychiatry Research 1995;57(3):251–7.
92. Buchanan TL, Barker KN, Gibson JT, et al. Illumination and errors in
dispensing. American Journal of Hospital Pharmacy 1991;48(10):2137–45.
93. Rashid M, Zimring C. A review of the empirical literature on the
relationships between indoor environment and stress in healthcare and
office settings: problems and prospects of sharing evidence. Environment
and Behavior 2008;40(3):151–90.
94. Bozett FW, Gibbons R. The nursing management of families in the critical
care setting. Critical Care Update 1983;10(2):22–7.
95. Byers, JF. Family discussion groups, a way to meet the needs of the ICU
patient’s family. Critical Care Nurse 1983;3(1):16–20.
96. Rashid M. Environmental design for patient families in intensive care units.
Journal of Healthcare Engineering 2010;1(3):367–97.
97. Simpson T. The family as a source of support for the critically ill adult.
AACN Clinical Issues in Critical Nursing 1991;2(2):229–35.
98. Hodovanic BH, Reardon D, Reese W, et al. Family crisis intervention program
in the medical intensive care unit. Heart and Lung 1984;13(3):243–9.
99. Molter NC. Needs of relatives of critically ill patients: a descriptive study.
Heart and Lung 1979;8(2):332–9.
100. Pati D, Nanda U. Influence of positive distractions on children in two clinic
waiting areas. Health Environments Research & Design Journal
2011;4(3):124–40.
101. Gurascio-Howard L, Malloch K. Centralized and decentralized nurse
station design: an examination of caregiver communication, work activities,
and technology. Health Environments Research and Design Journal
2007;1(1):44–57.
102. Melin L, Gotestam KG. The effects of rearranging ward routines on
communication and eating behaviors of psychogeriatric patients. Journal of
Applied Behavior Analysis. 198;14(1):47–51.
103. Peterson R, Knapp T, Rosen J. The effects of furniture arrangement on the
behavior of geriatric patients. Behavior Therapy 1977;8:464–7.
104. Holahan C. Seating patterns and patient behavior in an experimental
dayroom. Journal of Abnormal Psychology 1972;80(2):115–24.
105. Sommer R, Ross H. Social interaction on a geriatrics ward. The
International Journal of Social Psychiatry 1958;4(2):128–33.
106. Astedt-Kurki P, Paavilainen E, Tammentie T, et al. Interaction between adult
patients’ family members and nursing staff on a hospital ward.
Scandinavian Journal of Caring Sciences 2001;15(2):142–50.
107. Miwa Y, Hanyu K. The effects of interior design on communication and
impressions of a counselor in a counseling room. Environment and
Behavior 2006;38(4):484–502.
108. Daykin N, Bunt L. Music as a Resource for Health and Well-Being. In:
Söderback Ingrid, editor. International Handbook of Occupational Therapy
Interventions. New York: Springer; 2009. p. 453–46.
109. Ulrich RS, Gilpin L. Healing arts. In: Frampton SB, gilpin L, Charmel P,
editors. Putting patients first: Designing and practicing patient-centered
care. San Francisco: Jossey-Bass; 2003. p. 117–46.
110. Cooke B, Ernst E. Aromatherapy: A systematic review. British Journal of
General Practice 2000;50:493–6.
111. Marcus CC, Barnes M, editors. Healing gardens: Therapeutic benefits and
design recommendations. New York: Wiley; 1999.
112. O’Brien ME. Spirituality in Nursing. Sudbury, CT: Jones and Bartlett; 2003.
113. Young C. Spiritually healing environments. In: Spirituality, Health, and
Healing. Jones & Bartlett Publishers; 2006. p. 147–68.
114. Landro L. ICUs’ new message: welcome, families. Wall St. J. 2007. (Retrieved
from http://oruine.wsj.com/article/SB118419577614963880.html/).
115. Mantone J. Building family-friendly lCUs. Wall St. J. 2007. <http://
blogs.wsj.com/health/2007/07/12/building-familyfriendly-icus/>. (accessed
30.10.2008).

Informatics Infrastructure
for the Neurocritical
Care Unit
J. Michael Schmidt, David K. Vawdrey, and Richard S. Moberg
Introduction
S.P. is a 60-year-old female with a history of hypertension who
presented comatose with a left thalamic intracerebral hemor-
rhage associated with intraventricular hemorrhage, midline
shift, and localized mass effect. Emergently, an external ven-
tricular drain was placed into the left frontal horn, and on day
5 an additional drain was placed into the left temporal horn
for persistent left temporal dilation. Invasive multimodality
monitoring including intracranial pressure (ICP), and brain
tissue oxygen tension (PbtO2), and cerebral microdialysis was
placed into the right frontal lobe. She received repeated doses
of mannitol and hypertonic saline for persistent elevated ICP.
Managing patients with intracranial hypertension, like S.P.,
is a mainstay in neurocritical care and involves a structured
treatment protocol1 to continually assess the effect of each
intervention. Tracking and quantifying physiologic measures
such as cerebral perfusion pressure (CPP) also is crucial and
needs to be put into the context of sedation level, osmothera-
pies, ventilator settings, and temperature modulation among
many other interventions. In a digital world clinical staff
should be able to systematically, continually, and rapidly eval-
uate the effect of various treatments for increased ICP. For
instance, when the impact of mannitol and hypertonic saline
administration in S.P. is evaluated, the neurocritical care unit
(NCCU) staff should have immediate access to basic variables
such as osmolality and ICP at the bedside (Fig. 4.1). As fun-
damental as this is, a plot that contains these core elements is
difficult to create in real time and even retrospectively at most
institutions. But imagine being able to look at the patient’s
display screen in the NCCU and to be able to answer the fol-
lowing questions easily:
1. When mannitol is administered, by how much (mm Hg)
is ICP lowered; how quickly is ICP lowered, and how long
until ICP again increases to greater than 20 mm Hg?
2. Are repeated administrations of mannitol equally effective
or does repeated administration fail to lower ICP as much,
as quickly, or as long?
3. Does adding hypertonic saline provide additional ICP
reduction?
© Copyright 2013 Elsevier Inc. All rights reserved.
Chapter
4  
I
4. Do osmolar data suggest that additional osmotic therapy
would be dangerous to the patient?
Today clinicians generally are required to answer these ques-
tions in their heads, approximating when interventions were
made by thumbing through the hourly recorded data, estimat-
ing each therapy’s impact on ICP, and manually checking
laboratory data to estimate the osmolar gap. Ideally other
aspects of ICP management protocols also should be evalu-
ated, for example, end-tidal carbon dioxide (CO2) and tem-
perature. Further, analysis tools at the bedside should exist to
evaluate the patient’s physiologic status, such as cerebral auto-
regulation, which would indicate how ICP would respond to
changes in blood pressure. This then may reduce the need to
discuss how other invasive neuromonitoring data (e.g., cere-
bral blood flow, PbtO2, cerebral metabolism) may facilitate ICP
management. Any attempt to understand these dynamic rela-
tionships among ICP, interventions, and other physiology is
complicated, and yet this is only one aspect of patient care.
The promise of clinical information systems is to improve
patient care, reduce medical errors, reduce documentation
time, and improve efficiency.2,3 Paper-based hospital medical
records are generally inefficient (e.g., illegible handwriting,
manual data entry) and have been shown to contribute to
medical errors.4,5 Billions of dollars are being devoted to the
development and implementation of electronic health records
(EHRs)6 and significant progress has been made to bring these
data together for documentation purposes.4 Intensive care
units (ICUs) have been migrating from paper-based to com-
puterized charting systems for more than a decade.7
In the ICU, however, problems that face clinicians and
nurses extend beyond the need for more efficient documenta-
tion. During rounds of critically ill patients each morning, a
physician may be confronted with more than 200 variables,8,9
and some clinical information systems acquire and store phys-
iologic variables and device parameters online at least every
minute.10 People, however, are not able to judge the degree of
relatedness between more than two variables.11 The ability to
store high-dimensional data far exceeds the intellectual capac-
ity to understand it unassisted12; this greatly contributes to
conditions of constant “information overload” that can lead
27
28 Section I—Background
ICP
60
40
ICP (mm Hg)
20
0 200
–20
Fig. 4.1  Example of minimum integrated data display
to support intracranial pressure management. ICP,
Intracranial pressure.
to preventable medical errors.11,13 A clinical informatics infra-
structure in the NCCU that only supports documentation is
simply not sufficient. Ideally NCCU staff should be able to use
clinical decision support systems to help understand what a
patient’s data is trying to tell them.14
Currently there is a paucity of clinical decision support
systems in use in the ICU,12 but this is understandable because
the data and infrastructure to support their development and
use are generally not in place. When patient data is collected
in digital form, it usually resides on isolated systems without
a way to bring this data together in meaningful ways.4 The
objective of this chapter is to provide a blueprint for work
with hospital administration, information technology, bio-
medical engineering, industry vendors, and clinical staff to
create an informatics infrastructure that provides decision
support. Many hospitals already have data-collection systems
in place for electronic documentation, laboratory results, and
physician-order systems but need strategies to integrate these
data with high-resolution bedside monitor data to support
clinical decision support systems. First, clinicians need to be
able to articulate a clear message as to why patient data must
be stored at a high resolution and be accessible in real time to
support clinical decision making.
The Justification: Why Clinical Data
Is Needed
Data are a commodity that has many purposes but in health
care enable hospitals to provide medical care and meet associ-
ated legal obligations. Patient data must be maintained securely
over a long period of time and done so in a way that maintains
patient privacy. Storing data digitally can be more efficient,
help reduce medical errors, improve medical device manage-
ment, and reduce health care costs.3,14,15 It is important for
clinicians and nursing staff to voice how patient data can be
used to improve patient care. In the absence of that, other
more mundane administrative concerns usually take prece-
dence. In the NCCU the clinical need for real-time access to
patient data and clinical decision support systems falls into
three categories. First, NCCU staff need tools to evaluate the
effectiveness of treatment(s) meant to modify physiologic
Osmolal-Serium 3% Saline Mannitol 20%
500
340 2500
400
2000
320
75 mL/hr
300
1500
300
1000
280
100 500
260
0 0
00:00 00:00
1/9/2007 1/10/2007 1/11/2007
endpoints or improve clinical conditions such as cerebral vaso-
spasm or sepsis. Second, patient data should provide under­
standing of the patient’s physiologic status (e.g., dysautonomia,
cerebral autoregulation failure), and how such physiology
impacts other metrics of brain health (see Chapter 40). Third,
complex relationships within patient data should be automati-
cally processed to enable better understanding of prognosis
and earlier diagnosis of secondary events before clinical symp-
toms occur. This automatic processing of data further sup-
ports clinical decision making, and can enable computerized
implementation of clinical protocols16 (see Chapter 45) or
decision making on prognosis or outcome prediction if appro-
priate models are used.17,18
Clinical Computing Systems
The Blueprint
The goal is to create the informatics infrastructure necessary
to enable health care providers to leverage data from database
systems that contain physiologic, laboratory, pharmacy, and
other clinical data for the continued development and imple-
mentation of better clinical support tools to provide patients
the best care possible.3,4 There are many ways to accomplish
these goals, but it is essential to work closely with the institu-
tion’s information technology (IT) department to devise the
right solution for the ICU and institution because it is often
difficult to tell what is needed before the project starts. Regard-
less of the specifics of the solution, each will contain the same
basic elements: (1) collection and storage of different kinds of
patient data to a database system; (2) creation of a data ware-
house whereby patient data are integrated into a single data-
base with copies of all databases to support real-time analytics;
and (3) use of software to perform various methods of analysis
that together create a clinical decision support system (Fig.
4.2). Chapters 40 and 45 describe some NCCU specific solu-
tions that have been implemented.
Planning
Informatics is a rapidly changing in field, and therefore a
department should not be locked in any one system at an
 Section I—Background 29

Patient room

Bedside devices
Monitor network

Patient monitor

Patient monitor mainframe

Hospital network

Serial-to-TCP/IP
A B Raid storage drive
Bedside data server

D C Data warehouse

Bedside Lab EHR EEG

data

Clinical decision support Decision support

analytics

Other data collection and

Knowledge advancement Clinical research storage systems

Fig. 4.2  Theoretical data collection infrastructure for the neurocritical care unit. A, Medical devices that monitor patient physiology can plug
into the bedside patient monitor or be transmitted over the network using a serial-to-TCP/IP converter. B, A data server dedicated to collecting and
storing physiologic data can receive data from the patient monitor mainframe or directly from a serial-to-TCP/IP converter that is installed on the
server as a COM port. C, An IT-managed data warehouse receives all forms of patient data and makes it available for multiple purposes. D, Clinical
decision support systems use the data warehouse to facilitate real-time clinical decision making in the intensive care unit. The data warehouse also
supports retrospective analysis of a patient for quality assurance, clinical research, and other aspects of knowledge advancement. COM, Communication;
EEG, electroencephalogram; EHR, electronic health record; IT, information technology; TCP/IP, transmission control protocol/Internet protocol.

infrastructure level. Data should be stored in an open non-
proprietary format such as Structured Query Language (SQL)
(Open Database Connectivity [ODBC] compliant) database
that will allow any system to access the data. Many ICU direc-
tors strive to include an informatics infrastructure when
building a new ICU. This is a great time to do this because
there will be a capital budget, and essential items such as
installing power and internet connectivity in patient rooms
will be cheapest during the construction phase. An ICU con-
struction project also will mean that health care providers will
have the attention of hospital administrators and information
technology personnel. It also is vital to create an operating
budget to maintain the infrastructure over time and to
upgrade equipment19 and to decide whether remote data
viewing is important to put such a system in place. There are
commercial products, such as Citrix MetaFrame, that securely
enable this functionality and are gaining favor in ICU and
emergency department (ED) settings.20 Decisions also should
be made about whether to integrate the NCCU data with
other ICUs, both in the same institution and other medical
centers. Creation of such integrative databases may help facili-
tate syndrome surveillance, decision support, comparative
effectiveness research, and outcome research.21,22 Data collec-
tion systems and the EHR, however, do not replace verbal
communication for efficient interdisciplinary exchange of
patient information in the NCCU but can be used to supple-
ment this.23
Collecting Bedside Device Data
Overview
Current EHR systems generally are adequate to store clinical
documentation and laboratory, pathology, and radiology
30 Section I—Background
information. However, most EHRs are not designed to capture
and store high-frequency physiologic measurements obtained
from patient monitoring equipment. In the NCCU, invasive
and noninvasive monitoring techniques provide continuous
measurement of physiologic parameters, and this high-
frequency data can be vital to understand the course of
critical illness and optimize treatment. Data documented in
an EHR by the NCCU staff, typically on a hourly basis, are
not sufficient to fully represent this physiologic data stream,
and therefore it is desirable to establish a thoughtful strategy
to collect and use data from bedside medical devices such
as ventilators, patient monitors, infusion pumps, and other
neurospecific monitors. Bedside medical device data are
generally the most difficult data to obtain, whereas continu-
ous electroencephalographic and associated patient video
data create the greatest demands on network and storage
capacity. The informatics infrastructure needed to collect
and store continuous electroencephalographic data in the
ICU environment is described in detail elsewhere.24
Collecting Bedside Patient Monitor Data
Via a Network Portal
Many bedside medical devices are plugged into the standard
patient monitor, which serves to display all the digital and
waveform data on a single bedside display in real time.
Most clinical information systems from patient monitor
vendors store these data up to 72 hours for clinical review
purposes before they are deleted forever. Some may transfer
hourly values into an EHR for nurse verification. Networked
patient monitors facilitate patient data display in areas other
than the patient’s room, such as a central station or the
nursing pod area, by allowing other systems to receive that
data from a central server. Networked systems represent one
potential strategy to collect all data on every bedside patient
monitor from a single networked portal at a high temporal
resolution.
Some medical device manufacturers, including Philips
Healthcare (Andover, MA) and GE Healthcare (Chalfont St.
Giles, UK), have developed their own proprietary solutions
for automated data acquisition through this single portal.
Customers rarely use these solutions because they tend to be
expensive, do not provide adequate data frequency, and
sometimes do not integrate easily with other clinical infor-
mation systems. These systems do continue to improve,
however, and it is sensible to investigate data collection and
storage solutions from the institution’s patient monitor
vendor. Third-party products such as DataCaptor (Capsule
Technologie, Paris, France) are designed to convert the non-
standard output from bedside devices into HL7 format,
which can then be sent to EHR systems such as Cerner Mil-
lennium (Cerner Corporation, Kansas City, MO) and Eclip-
sys Sunrise Clinical Manager (Eclipsys Corporation, Atlanta,
GA). Open-source efforts to collect bedside device data for
both research and clinical care purposes is described else-
where.25,26 In the Columbia University Medical Center neuro-
logical ICU, Bedmaster XA (Excel Medical Electronics,
Jupiter, FL) is used to collect all the bedside patient monitor
data including parameter data at least every 5 seconds and all
visible waveform data at a resolution of 240 Hz. This system
works for both the General Electric and Philips patient
monitors.
Collecting Data from Individual Bedside
Patient Monitors or Devices
It may not always be possible to collect bedside device data
through a network portal, and many medical devices do not
plug into the patient monitor, or if a device does plug into the
monitor, not all the data are transferred. Data then need to be
collected from the individual medical device or sometimes
even each individual patient monitor. Most bedside devices
(including the patient monitor) are equipped with RS-232
(digital) and analog (waveform) data output ports for auto-
matic data output capabilities. Specifications for data com-
munication are generally available in the device’s technical
manual or can be obtained from the device manufacturer.
Medical devices output data usually as a comma-delimited
text string (e.g., 21, 15.6, 78.24, 12, 15) every few seconds. This
text string must be converted into data that fill specific fields
in a database, which requires knowledge about what each
number in the text string represents and in what field in the
database it should be stored. This translation is referred to as
a device interface (a small bit of software that automatically
converts data from the device into sensible data for a particu-
lar database). Provided it is known what data output is sent
from the machine, what each number means, and where it
goes in the database, actually writing a device interface is not
complicated.
One limitation of RS-232 for bedside device communica-
tion is that there are many different ways to send data over the
serial interface. Several different pin-out and connector styles
exist, and there are multiple options for baud rate, parity, stop
bits, handshaking, flow control, and signaling.27 For example,
many bedside devices output the data string at a set time
frequency without prompting, whereas with other devices,
such as the patient bedside monitor, the software interface
must send a coded request to receive data. Therefore the
device interface must be written to accommodate the specifics
of each device or patient monitor configuration. Some newer
modes to output data from devices include Universal Serial
Bus (USB), 802.3 (Ethernet), IEEE 11073, or even wireless
(e.g., 802.11 b/g, Bluetooth) data communication capabilities.
Because the life span for some medical devices is 10 to 15
years, it is unlikely that RS-232 ports will disappear in the
near future.
At Columbia University it was decided to use Bedmaster XA
to collect bedside data in the NCCU, because the company
also wrote device interfaces for devices that did not connect
to the bedside monitor, or did so incompletely. For example,
it is possible to collect all 22 parameters outputted from the
Arctic Sun cooling device (Medivance, Louisville, CO). This
allows caregivers to automatically track the device’s water tem-
perature, which decreases when the machine works harder to
maintain body temperature (i.e., fever spike), and other
parameters such as ICP and PbtO2. There also is a device inter-
face for the Camino ICP monitor (Integra Neuroscience,
Paramus, NJ), even though this monitor can plug into the
patient bedside monitor. Many brain injury patients may have
an external ventricular drain and a parenchymal ICP monitor.
It is important to distinguish each ICP from the other.
However, there are limited methods to maintain the same data
label for each device regardless of where it was plugged in on
the patient monitor. Instead the Camino is now to both the
patient monitor and the secondary system, where plugged in

the data label can be controlled. It is important to understand
how the patient monitor handles different device scenarios
when considering how to collect patient data.
Standards in Bedside Device Collection
Technical standards developed for medical device communi-
cation are available. These standards are supposed to simplify
the problem of getting device data into multiple data systems,
but often are not adopted by device manufacturers. Efforts to
establish a medical device connectivity standard began in the
1980s with the Medical Information Bus (MIB). The MIB was
developed by representatives from medical device manufac-
turers, health care institutions, and academic departments
who advocated “plug-and-play” data sharing among bedside
devices irrespective of device vendor.28,29 In 1996 the MIB
specification was approved by the Institute of Electrical and
Electronic Engineers (IEEE) Standards Board under the name
“IEEE 1073 Standard for Medical Device Communica-
tions.”30-32 The IEEE 1073 standard continued to evolve, and
in 2004 the International Organization for Standardization
(ISO) formally ratified IEEE 1073 as “ISO 11073 Point-of-
Care Medical Device Communication Standard. Despite this,
the ISO/IEEE 11073 standard still is not widely used by the
medical device industry. Consequently, most hospitals have
limited or no integration of bedside device data with their
EHR systems and other clinical information systems.
Regulatory concerns present another barrier to medical
device connectivity at least in the United States. The U.S. Food
and Drug Administration (FDA) requires a manufacturer of
a data collection device to test the connections to each device
from which it receives data rather than just test to ensure
proper implementation of a standard (such as ISO/IEEE
11073). This presents an unnecessary burden on the manufac-
turer, similar to a computer manufacturer’s having to test the
connection to every printer rather than just test to the USB
standard. The FDA realizes the benefits of medical device con-
nectivity and the regulatory issues are being addressed as part
of the mission of the Medical Device Plug and Play (MDPnP)
program.33
Transferring Data Over the Network
The simplest configuration to transfer data from a medical
device to a computer is to plug a RS-232 cable purchased from
the local computer store into the RS-232 output of the medical
device and then plug the other end into the nine-pin com-
munication (COM) port of a regular personal computer. A
serial-to-TCP/IP converter enables one to send data from
multiple bedside devices over the hospital or local network to
a server located in another room or building. Placed in the
patient’s room and “installed” on the server, it creates multiple
COM ports that allow many devices to plug into it and trans-
fer its data over the network using standard TCP/IP protocols.
These converters are available in many configurations includ-
ing virtual and wireless.
Special care is needed when wireless communication is used
to acquire data from bedside medical devices. First, if mobile
monitoring solutions rely on battery power, power consump-
tion issues must be considered. Second, wireless communica-
tion devices are a cause of radio frequency (RF) electromagnetic
interference (EMI) that may cause undesirable effects to
Section I—Background 31
medical equipment. It is recommended that testing be per-
formed for EMI between mobile wireless communication and
medical devices.34 Finally, with many devices and multiple
wireless technologies competing to use the same frequency
spectrum to transmit, crowding can occur that may decrease
throughput and affect the reliability of data exchange. Device
manufacturers that use RF communication should provide an
analysis of data communication requirements, including esti-
mated bandwidth utilization at periods of peak and normal
usage.
Data overload is one of the greatest informatics chal-
lenges in the NCCU for both medical informaticians and
intensive care clinicians.35-39 For example, at Columbia Uni-
versity NCCU bedside monitors store approximately 200
megabytes (MB) per day per bed, whereas continuous elec-
troencephalograms (EEGs) can generate approximately 1
gigabyte (GB) per day for EEG alone, and 20  GB per day
when there is video recording. When multiple systems in
multiple patient rooms record simultaneously, network per-
formance can slow down or data loss is possible. Modern
Ethernet networks that use 1  GB per second or greater
connections between switches and routers should be used
throughout the network to help avoid this problem. It also
is important to check for potential bottleneck areas where
older systems that use 10 or 100 MB per second connec-
tions may exist (e.g., when networks span new and old
buildings).24
Linking Patients to Their Data
Medical device connectivity is complex, and this extends
beyond simply the physical connectors and communication
protocols. Perhaps more important is to establish patient
context and ensure that device(s) data are reliably associated
with the correct patient. A wired infrastructure simplifies
patient association because each device can be connected
unambiguously to a particular location (i.e., room or bed
number). Use of existing electronic admission/discharge/
transfer (ADT) systems to map a specific patient to a location
can link device data to a patient. However, mobile devices
(e.g., infusion pumps) and wireless data communication
introduce additional challenges for patient identification.
Some devices allow a patient’s medical record number to be
entered into the device interface, or provide bar code scanning
capabilities to accomplish the same task.
It is essential to consider how data will be linked to a patient
before device purchase. There may be an advantage to auto-
matic data documentation directly to an electronic medical
record (EMR) from devices such as ventilators and infusion
pumps; several studies show this is more accurate and saves
nursing time.33,40,41 In some ICUs ventilators are always con-
nected to the bedside monitor and their data automatically
transmitted to an EMR for verification. Infusion pumps
should have the capability to send streaming data, but not all
are able to do this; some that do, send anonymous drug and
dose data to a server to help biomedical departments with asset
management and device maintenance. Transmission of anon-
ymous data bypasses the Health Insurance Portability and
Accountability Act (HIPAA) and privacy concerns. This allevi-
ates potential headaches for biomedical engineering and infor-
matics departments; however, this does not help clinicians
manage patients. These various issues need to be addressed
32 Section I—Background
before making purchase decisions. No matter the mechanism
used to link device data to patients, local testing and periodic
validation should be performed to ensure accuracy.
Time Synchronization
It is critical to synchronize data together from multiple
devices for accurate bedside device data collection. Each
device may maintain its own internal clock, and additional
clocks may exist for the data acquisition system and the hos-
pital clinical information system. When data that are used for
time-sensitive analyses (e.g., when does an infused medica-
tion affects vital signs, when do alarms trigger, and what is
the duration?) are recorded, it is important to know exactly
the order and delay between events. The Network Time Proto-
col (NTP) is the most commonly used Internet time protocol
and can be used if utilities for time synchronization are oth-
erwise unavailable.
Other Data Collection Considerations
It is important to anticipate how the data may be used to
prepare an effective strategy for physiologic data acquisition
in the NCCU. Is the purpose of data collection to facilitate
clinical care, research, or both? If the data are to be used to
provide patient care and there is an existing EHR system in
use, it is important to decide whether bedside device data will
be automatically transferred to the EHR, and if so, how this
will be accomplished. Undoubtedly there will be workflow
changes for care providers associated with automatic data col-
lection (e.g., documentation and data review processes). To
help anticipate some of the sociotechnical challenges that may
arise, both technologic and associated with personnel or
workflow, the following questions should be addressed:
 How often will data be transferred to the existing EHR
system?

Will manual verification be required before the data are
included in the patient’s record?

For each data element that is currently documented in the
patient’s record, does that element exist in the bedside
device’s output stream?

Will filters be applied to the data to reduce artifact
(i.e., “noise” in the signal) or select representative measure-
ments?

Will the same type of measurement be obtained simulta-
neously from separate devices (e.g., respiratory rate col-
lected from a patient monitor and a mechanical ventilator)?
How will these measurements be dealt with?

How will downtime procedures be created and enforced
(for the EHR system as well as the data collection system)?
Some clinical information systems may not be capable of han-
dling second-by-second or even minute-by-minute recording
of physiologic variables. If data are to be stored in a secondary
or research database instead of or in addition to an EHR
system, several other questions should be considered:
 How will data be stored (e.g., flat files, relational database,
XML database)?
 How will data backups be performed?
 How frequently can or should data be acquired and stored
(e.g., 500 Hz, 1 sec, 5 sec, 1 min, 15 min, 1 hr)?
Data Warehouse Configurations
In this chapter the focus has been on how to get high-frequency
data (i.e., parameter data every 5 seconds) from the patient
monitor and other bedside medical devices to a “server.”
Servers are simply a class of computers with large computa-
tional and storage capacities that manage, store, and retrieve
data for other computers or devices.19 A data warehouse is a
collection of decision support technologies to enable better
and faster decisions42 that is composed of multiple servers and
networked data storage to support clinical decision making
and research.3,43
A data warehouse is where laboratory, imaging, interven-
tion, physiologic, and all other patient databases reside. Tra-
ditional data warehouses are set up to bring several different
kinds of data (e.g., laboratory and physiologic) together into
a unified database to be used by clinical support software
tools. Software programs translate data output from the device
into an intelligent form to allow the data to be added into a
database at medical device interfaces. Similarly to transfer data
from one database to another database also requires a software
interface. Database interfaces rely on standards such as HL7
to streamline the process. An interface engine is software that
manages all the standardized “messages” being sent back and
forth between databases.44 An alternative and more flexible
strategy is to create a real-time replica of each database paired
with software that supports real-time and ad hoc queries from
decision support software.43 The exact strategy to be used in
the NCCU will require discussion with the institutional IT
department and vendors.
There are several important considerations to link data-
bases. First, performance bottlenecks are not specific to the
network and can occur at the server level. There needs to be
adequate storage capacity for primary data collection of
patient monitors and medical devices, a backup of these data
in case of hardware failure, and replicas or integrated data-
bases that will support clinical support applications to prevent
bottlenecks. Second, the file server and associated disk systems
need to be sufficiently robust to avoid performance slow-
downs during data collection, data review, and other mainte-
nance functions. Third, file servers used in critical care should
be capable of operating without interruption, despite common,
predictable component failures. Strategies such as redundant
power supplies and redundant arrays of independent disks
(RAIDs) rather than storing data on single hard drives may
prevent this. Though there are many variations of a RAID
drive, in its basic form data are divided and written across
multiple devices simultaneously and redundantly. This can
increase performance and provide fault tolerance.19,24
Clinical Decision Support
Research still is needed to determine what data should be col-
lected at the bedside, and how information is best summarized
and presented to health care providers in the NCCU. Innova-
tive graphic displays show particular promise but have yet to
be adopted on a large scale.37,45-49 Display of physiologic trends
also is important in critical care, perhaps more so than simple
numeric displays or threshold alarms. Adoption of automatic
data acquisition from bedside devices will finally enable the
research and patient care environment that Peters and Stacy
anticipated almost a half-century ago:

“The ability to combine analogue and digital computation
… moves the field of analysis of physiologic data to that
point where mathematical postulates can be set up and
tested in the real situation. The enhancement of this skill
and propensity will permit us to take physiology out of the
data-collecting-for-data’s-sake stage to the more imagina-
tive methods now employed in physical sciences. This step
will mean as much to the advancement of knowledge in
physiology as it has meant in the physical sciences.”50
Chapters 40 and 45 address these issues in further detail.
However, how the data will be used needs to be considered to
collect it in a manner that enables proper use. Currently there
is little regulatory guidance on clinical decision support
systems despite reports that have surfaced about their poten-
tial for harm.51 Two independent organizations are address-
ing this52,53 and will hopefully provide some certification
guidelines. Thus the regulatory issues ahead are certain to
become more stringent, but the actual outcome remains
unpredictable.
Conclusion
To create the infrastructure necessary for a successful neuro-
monitoring and clinical informatics program in the NCCU is
feasible, and there are enough commercial resources available
to help with this process. The key to success is to recognize
that there are actually four problems to consider when the
infrastructure is developed: data collection, data visualization,
data analysis, and decision support. The first and most impor-
tant step is data collection, because without the data nothing
else is possible. It is critical to clearly articulate to hospital
administrators and information technologists what data the
NCCU needs, how it will be used, and what positive impact it
should have on patient care, for example, better outcomes or
reduced length of stay. Each hospital constituent group has
different needs for the same data, and unless the NCCU needs
are understood, the required data may turn out to be unavail-
able or inaccessible. Therefore NCCU staff should work with
the information technology group to make sure that network
and storage concerns are addressed up front and that there is
an equipment maintenance plan. The data should be stored
in a manner such that it can be automatically exported in a
standard format to any number of viewing or analysis systems.
In addition, communication between different health care
providers (e.g., nurses, pharmacists, physicians) is necessary
to ensure a synchronized and efficient system.54 These steps
will ensure that the infrastructure will withstand the test of
time and subsequently enable leverage of whatever technolo-
gies emerge in the years to come.
References
1. Mayer S, Chong J. Critical care management of increased intracranial
pressure. J Intensive Care Med 2002;17(2):55–67.
2. Payne T. Introduction and overview of clinical computing systems within a
medical center. In: Payne T, editor. Practical guide to clinical computing
systems: design, operations, and infrastructure. Burlington, MA; Academic
Press; 2008.
3. Martich G, Waldmann C, Imhoff M. Clinical informatics in critical care.
J Intensive Care Med 2004;19(3):154–163.
4. Varon J, Marik PE. Clinical information systems and the electronic
medical record in the intensive care unit. Curr Opin Crit Care 2002;8(6):
616–24.
Section I—Background 33
5. Mador R, Shaw N. The impact of a critical care information system (CCIS)
on time spent charting and in direct patient care by staff in the ICU: a
review of the literature. Int J Med Inform 2009;78(7):435–45.
6. Gill JM. EMRs for improving quality of care: promise and pitfalls.
[comment] Fam Med 2009;41(7):513–15.
7. Donati A, Gabbanelli V, Pantanetti S, et al. The impact of a clinical
information system in an intensive care unit. J Clin Monit Comput 2008;
22(1):31–6.
8. Morris G, Gardner R. Computer applications. In: Hall, J, Schmidt G,
Wood L, editors. Principles of critical care. McGraw-Hill: New York; 1992.
p. 500–14.
9. Gather U, Imhoff M, Fried R. Graphical models for multivariate time series
from intensive care monitoring. Stat Med 2002;21(18):2685–2701.
10. Imhoff M, Fried R, Gather U, et al. Dimension reduction for physiological
variables using graphical modeling. AMIA Annu Symp Proc 2003:313–17.
11. Imhoff M, Fried R, Gather U. Detecting relationships between physiological
variables using graphical modeling. Proc AMIA Symp 2002:340–4.
12. De Turck F, Decruyenaere J, Thysebaert P, et al. Design of a flexible platform
for execution of medical decision support agents in the intensive care unit.
Comput Biol Med 2007;37(1):97–112.
13. Jennings D, Amabile T, Ross L. Informal assessments: data-based versus
theory-based judgments. In: Kahnemann D, Slovic P, Tversky A, editors.
Judgments under uncertainty: heuristics and biases. Cambridge University
Press: New York; 1982. p. 211–30.
14. Adhikari N, Lapinsky S. Medical informatics in the intensive care unit:
overview of technology assessment. J Crit Care 2003;18(1):41–7.
15. Clemmer TP. Computers in the ICU: where we started and where we are
now. J Crit Care 2004;19(4):201–7.
16. Campion TR Jr, Waitman LR, May AK, et al. Social, organizational, and
contextual characteristics of clinical decision support systems for intensive
insulin therapy: a literature review and case study. Int J Med Inform 2010;
79(1):31–43.
17. Peelen L, de Keizer NF, Jonge E, et al. Using hierarchical dynamic Bayesian
networks to investigate dynamics of organ failure in patients in the intensive
care unit. J Biomed Inform 2010;43(2):273–86.
18. Ji SY, Smith R, Huynh T, et al. A comparative analysis of multi-level
computer-assisted decision making systems for traumatic injuries. BMC Med
Inform Decis Mak 2009;9:2.
19. Chou D, Sengupta S. Infrastructure and security. In: Practical guide to
clinical computing systems: design, operations, and infrastructure. Payne T,
editor. Burlington, MA: Academic Press; 2008.
20. Payne T. Architecture of clinical computer systems. In Payne T, editor.
Practical guide to clinical computing systems: design, operations, and
infrastructure. Burlington, MA: Academic Press; 2008.
21. Herasevich V, Pickering BW, Dong Y, et al. Informatics infrastructure for
syndrome surveillance, decision support, reporting, and modeling of critical
illness. Mayo Clin Proc 2010;85(3):247–54.
22. Platt R, Takvorian SU, Septimus E, et al. Cluster randomized trials in
comparative effectiveness research: randomizing hospitals to test methods
for prevention of healthcare-associated infections. Med Care 2010;48
(6 Suppl):S52–7.
23. Collins SA, Bakken S, Vawdrey DK, et al. Model development for EHR
interdisciplinary information exchange of ICU common goals. Int J Med
Inform 2010; Oct 23. Epub ahead of print.
24. Kull L, Emerson R. Continuous EEG monitoring in the intensive care unit:
technical and staffing considerations. J Clin Neurophysiol 2005;22(2):
107–18.
25. Kroth PJ, Belsito A, Overhage JM, et al. Bedside vital signs capture for the
non-ICU setting—an open source, PC-based solution. Proc AMIA Symp
2001:344–8.
26. Vawdrey DK, Hum RS. OpenMDC: an open-source framework for medical
device communication. In: AMIA Annu Symp Proc. Washington, DC; 2008.
27. TAL tech: Instrumental software solutions. [cited 1/2/2010]; Available from:
www.taltech.com/resources/intro-sc.html.
28. Gardner RM, Hawley WL, East TD, et al. Real time data acquisition:
recommendations for the Medical Information Bus (MIB). Int J Clin Monit
Comput 1991;8(4):251–8.
29. Shabot MM. Standardized acquisition of bedside data: the IEEE P1073
medical information bus. Int J Clin Monit Comput 1989;6(4):197–204.
30. Garnsworthy J. Standardizing medical device communications: the Medical
Information Bus. Med Device Technol 1998;9(3):18–21.
31. Kennelly RJ. Improving acute care through use of medical device data. Int J
Med Inf 1998;48(1-3):145–9.
32. Kennelly RJ, Gardner RM. Perspectives on development of IEEE 1073: the
Medical Information Bus (MIB) standard. Int J Clin Monit Comput
1997;14(3):143–9.
34 Section I—Background
33. Dalto JD, Johnson KV, Gardner RM, et al. Medical information bus usage for
automated IV pump data acquisition: evaluation of usage patterns. Int J Clin
Monit Comput 1997;14(3):151–4.
34. Bruns B, Dimantha S. Evaluating EMI in a multi-hospital facility. Biomed
Instrum Technol 2006;Suppl:40–2.
35. Alberdi E, Gilhooly K, Hunter J, et al. Computerisation and decision making
in neonatal intensive care: a cognitive engineering investigation. J Clin Monit
Comput 2000;16(2):85–94.
36. Calvelo D, Chambrin MC, Pomorski D, et al. Towards symbolization using
data-driven extraction of local trends for ICU monitoring. Artif Intel Med
2000;19(3):203–23.
37. Clemmer TP, Gardner RM. Data gathering, analysis, and display in critical
care medicine. Respir Care 1985;30(7):586–601.
38. Cole WG, Stewart JG. Human performance evaluation of a metaphor
graphic display for respiratory data. Methods Inf Med 1994;33(4):390–6.
39. Sims AJ, Pay, DA, Watson BG. An architecture for the automatic acquisition
of vital signs by clinical information systems. IEEE Trans Inf Technol
Biomed 2000;4(1):74–5.
40. Sapo M, Wu S, Asgari S, et al. A comparison of vital signs charted by nurses
with automated acquired values using waveform quality indices. J Clin
Monit Comput 2009;23(5):263–71.
41. Vawdrey D, Gardner RM, Evans RS, et al. Assessing data quality in
manual entry of ventilator settings. J Am Med Informat Assoc 2007;14(3):
295–303.
42. Chaudhuri S, Dayal U. An overview of data warehousing and OLAP
technology. ACM Sigmod Record 1997;26(1):65–74.
43. Chelico J, Wajngurt D. Architectural design of a data warehouse to support
operational and analytical queries across disparate clinical databases. AMIA
Annu Symp Proc 2007;11:901.
44. Payne T, Hoath J. Creating and supporting interfaces. In Payne T, editor.
Practical guide to clinical computing systems: design, operations, and
infrastructure. Burlington, MA: Academic Press; 2008.
45. Cole WG, Stewart JG. Metaphor graphics to support integrated decision
making with respiratory data. Int J Clin Monit Comput 1993;10(2):91–100.
46. Horn W, Popow C, Unterasinger L. Support for fast comprehension of ICU
data: visualization using metaphor graphics. Methods Inf Med 2001;40(5):
421–4.
47. Syroid ND, Agutter J, Drews FA, et al. Development and evaluation of a
graphical anesthesia drug display. Anesthesiology 2002;96(3):565–75.
48. Powsner SM, Tufte ER. Graphical summary of patient status. Lancet 1994;
344(8919):386–9.
49. Drews FA, Westenskow DR. The right picture is worth a thousand numbers:
data displays in anesthesia. Hum Factors 2006;48(1):59–71.
50. Peters RM, Stacy RW. Automatized clinical measurement of respiratory
parameters. Surgery 1964;56:44–52.
A complete list of references for this chapter can be found online at
www.expertconsult.com.

References
1. Mayer S, Chong J. Critical care management of increased intracranial
pressure. J Intensive Care Med 2002;17(2):55.
2. Payne T. Introduction and overview of clinical computing systems within a
medical center. In: Payne T, editor. Practical guide to clinical computing
systems: design, operations, and infrastructure. Academic Press: Burlington,
MA;2008.
3. Martich G, Waldmann C, Imhoff M. Clinical informatics in critical care.
J Intensive Care Med 2004;19(3):154.
4. Varon J, Marik PE. Clinical information systems and the electronic medical
record in the intensive care unit. Curr Opin Crit Care 2002;8(6):616–24.
5. Mador R, Shaw N. The impact of a critical care information system (CCIS)
on time spent charting and in direct patient care by staff in the ICU: a
review of the literature. Int J Med Inform 2009;78(7):435–45.
6. Gill JM. EMRs for improving quality of care: promise and pitfalls.
[comment] Fam Med 2009;41(7):513–15.
7. Donati A, Gabbanelli V, Pantanetti S, et al. The impact of a clinical
information system in an intensive care unit. J Clin Monit Comput
2008;22(1):31–6.
8. Morris G, Gardner R. Computer applications. In: Hall, J, Schmidt G, Wood
L, editors. Principles of critical care. McGraw-Hill: New York; 1992.
p. 500–14.
9. Gather U, Imhoff M, Fried R. Graphical models for multivariate time series
from intensive care monitoring. Stat Med 2002;21(18):2685–2701.
10. Imhoff M, Fried R, Gather U, et al. Dimension reduction for physiological
variables using graphical modeling. AMIA Annu Symp Proc 2003:313–17.
11. Imhoff M, Fried R, Gather U. Detecting relationships between physiological
variables using graphical modeling. Proc AMIA Symp 2002:340–4.
12. De Turck F, Decruyenaere J, Thysebaert P, et al. Design of a flexible platform
for execution of medical decision support agents in the intensive care unit.
Comput Biol Med 2007;37(1):97–112.
13. Jennings D, Amabile T, Ross L. Informal assessments: data-based versus
theory-based judgments. In: Kahnemann D, Slovic P, Tversky A, editors.
Judgments under uncertainty: heuristics and biases. Cambridge University
Press: New York; 1982. p. 211–30.
14. Adhikari N, Lapinsky S. Medical informatics in the intensive care unit:
overview of technology assessment. J Crit Care 2003;18(1):41–7.
15. Clemmer TP. Computers in the ICU: where we started and where we are
now. J Crit Care 2004;19(4):201–7.
16. Campion Jr TR, Waitman LR, May AK, et al. Social, organizational, and
contextual characteristics of clinical decision support systems for intensive
insulin therapy: a literature review and case study. Int J Med Inform
2010;79(1):31–43.
17. Peelen L, de Keizer NF, Jonge E, et al. Using hierarchical dynamic Bayesian
networks to investigate dynamics of organ failure in patients in the intensive
care unit. J Biomed Inform 2010;43(2):273–86.
18. Ji SY, Smith R, Huynh T, et al. A comparative analysis of multi-level
computer-assisted decision making systems for traumatic injuries. BMC Med
Inform Decis Mak 2009;9:S6.
19. Chou D, Sengupta S. Infrastructure and security. In: Practical guide to
clinical computing systems: design, operations, and infrastructure. Payne T,
editor. Burlington, MA: Academic Press; 2008.
20. Payne T. Architecture of clinical computer systems. In: Payne T, editor.
Practical guide to clinical computing systems: design, operations, and
infrastructure. Burlington, MA: Academic Press; 2008.
21. Herasevich V, Pickering BW, Dong Y, et al. Informatics infrastructure for
syndrome surveillance, decision support, reporting, and modeling of critical
illness. Mayo Clin Proc 2010;85(3):247–54.
22. Platt R, Takvorian SU, Septimus E, et al. Cluster randomized trials in
comparative effectiveness research: randomizing hospitals to test methods
for prevention of healthcare-associated infections. Med Care 2010;48
(6 Suppl):S52–7.
23. Collins SA, Bakken S, Vawdrey DK, et al. Model development for EHR
interdisciplinary information exchange of ICU common goals. Int J Med
Inform 2010; Oct 23. [Epub ahead of print]
24. Kull L, Emerson R. Continuous EEG monitoring in the intensive care unit:
technical and staffing considerations. J Clin Neurophysiol
2005;22(2):107–18.
25. Kroth PJ, Belsito A, Overhage JM, et al. Bedside vital signs capture for the
non-ICU setting—an open source, PC-based solution. Proc AMIA Symp
2001:344–8.
Section—Background 34.e1
26. Vawdrey DK, Hum RS. OpenMDC: an open-source framework for medical
device communication. In: AMIA Annu Symp Proc. Washington, DC; 2008.
27. TAL tech: Instrumental software solutions. [cited 1/2/2010]; Available from:
www.taltech.com/resources/intro-sc.html.
28. Gardner RM, Hawley WL, East TD, et al. Real time data acquisition:
recommendations for the Medical Information Bus (MIB). Int J Clin Monit
Comput 1991;8(4):251–8.
29. Shabot MM. Standardized acquisition of bedside data: the IEEE P1073
medical information bus. Int J Clin Monit Comput 1989;6(4):197–204.
30. Garnsworthy J. Standardizing medical device communications: the Medical
Information Bus. Med Device Technol 1998;9(3):18–21.
31. Kennelly RJ. Improving acute care through use of medical device data. Int J
Med Inf 1998;48(1-3):145–9.
32. Kennelly RJ, Gardner RM. Perspectives on development of IEEE 1073: the
Medical Information Bus (MIB) standard. Int J Clin Monit Comput
1997;14(3):143–9.
33. Dalto JD, Johnson KV, Gardner RM, et al. Medical information bus usage for
automated IV pump data acquisition: evaluation of usage patterns. Int J Clin
Monit Comput 1997;14(3):151–4.
34. Bruns B, Dimantha S. Evaluating EMI in a multi-hospital facility. Biomed
Instrum Technol 2006;Suppl:40–2.
35. Alberdi E, Gilhooly K, Hunter J, et al. Computerisation and decision making
in neonatal intensive care: a cognitive engineering investigation. J Clin Monit
Comput 2000;16(2):85–94.
36. Calvelo D, Chambrin MC, Pomorski D, et al. Towards symbolization using
data-driven extraction of local trends for ICU monitoring. Artif Intel Med
2000;19(3):203–23.
37. Clemmer TP, Gardner RM. Data gathering, analysis, and display in critical
care medicine. Respir Care 1985;30(7):586–601.
38. Cole WG, Stewart JG. Human performance evaluation of a metaphor
graphic display for respiratory data. Methods Inf Med 1994;33(4):390–6.
39. Sims AJ, Pay, DA, Watson BG. An architecture for the automatic acquisition
of vital signs by clinical information systems. IEEE Trans Inf Technol
Biomed 2000;4(1):74–5.
40. Sapo M, Wu S, Asgari S, et al. A comparison of vital signs charted by nurses
with automated acquired values using waveform quality indices. J Clin
Monit Comput 2009;23(5):263–71.
41. Vawdrey D, Gardner RM, Evans RS, et al. Assessing data quality in manual
entry of ventilator settings. J Am Med Informat Assoc 2007;14(3):295–303.
42. Chaudhuri S, Dayal U. An overview of data warehousing and OLAP
technology. ACM Sigmod Record 1997;26(1):65–74.
43. Chelico J, Wajngurt D. Architectural design of a data warehouse to support
operational and analytical queries across disparate clinical databases. AMIA
Annu Symp Proc 2007;11:901.
44. Payne T, Hoath J. Creating and supporting interfaces. In: Payne T, editor.
Practical guide to clinical computing systems: design, operations, and
infrastructure. Burlington, MA; Academic Press; 2008.
45. Cole WG, Stewart JG. Metaphor graphics to support integrated decision
making with respiratory data. Int J Clin Monit Comput 1993;10(2):91–100.
46. Horn W, Popow C, Unterasinger L. Support for fast comprehension of ICU
data: visualization using metaphor graphics. Methods Inf Med
2001;40(5):421–4.
47. Syroid ND, Agutter J, Drews FA, et al. Development and evaluation of a
graphical anesthesia drug display. Anesthesiology 2002;96(3):565–75.
48. Powsner SM, Tufte ER. Graphical summary of patient status. Lancet
1994;344(8919):386–9.
49. Drews FA, Westenskow DR. The right picture is worth a thousand numbers:
data displays in anesthesia. Hum Factors 2006;48(1):59–71.
50. Peters RM, Stacy RW. Automatized clinical measurement of respiratory
parameters. Surgery 1964;56:44–52.
51. Tsai TL, Fridsma DB, Gatti G. Computer decision support as a source of
interpretation error: the case of electrocardiograms. J Am Med Inform Assoc
2003;10:478–83.
52. Healthcare Information Technology Standards Panel (HITSP), New York:
American National Standards Institute.
53. Certification Commisssion for Health Information Technology (CCHIT),
Chicago, IL.
54. Pingenot AA, Shanteau J, Sengstacke LT. Description of inpatient medication
management using cognitive work analysis. Comput Inform Nurs
2009;27(6):379–92.

Nursing and Education
DaiWai M. Olson
Introduction
Nurses and physicians come from two distinctly different pro-
grams. In the neurocritical care unit (NCCU) they work side
by side but may not have been provided with the opportunity
to understand each other’s roles. The purpose of this chapter
is to provide a common ground of understanding for nurses
and physicians to help improve multimodal monitoring
systems by exploring the role of nursing in patient monitoring
and how nursing education influences that role.
Historical Perspective of Nursing
and Monitoring
Almost a century ago George Santayana1 wrote, “Those who
cannot remember the past are condemned to repeat it.” The
history of nursing parallels the development of patient moni-
toring and equally affords an insight to predict some of the
future directions in patient monitoring. Today nurses view the
terms multitasking and multimodal monitoring not as new
terms, but as evolving paradigms.2 In this chapter the histori-
cal influence of medicine on the current role of the nurse
serves as a mechanism to understand how nursing can
optimally contribute to emerging trends in neurologic
monitoring.
Ancient Past
In ancient times nursing and medicine were intertwined, and
there were healers who brewed medicinal teas and tended to
injuries with a combination of mysticism, herbalism, and
remedies handed down through oral history.3 After each cup
of “medicine” and after each broken bone was set, the healers
monitored their patients and observed their successes and
failures that they shared with other healers. When human
societies settled down and towns were established, medical
care was regimented and evolved into a discipline.
As the base of medical knowledge expanded, humankind
began to document the effect of various treatments. Oral
history or direct observation no longer could adequately
transfer medical knowledge that now was shared across time
and distance in written form. Although physical evidence
exists to support splinting of bones, herbal remedies, and even
trepanation, much of known early medical history is attrib-
uted to Egyptian papyrus writings.3 Several key writings
provide clues to the role of nursing in emerging societies.
Except for midwifery, nursing was predominantly a male
© Copyright 2013 Elsevier Inc. All rights reserved.
Chapter
5  
I
profession. Wreszinski4 translated the Ebbers papyrus such
that future English versions report “wet nurse” as the figure
who provides milk to a child. Historians find no female ver-
sions of a nurse’s role as an attendant or assistant to the physi-
cian. The nosocomii were men who were educated by
physicians and taught how to run a hospital; the role of the
nosocomii (from which we derive nosocomial) is not fully
described but is thought to be the first ancestor of modern
nursing.
More modern times provide a few clues to the emerging
role of nursing. The physician’s role evolved over time and was
symbolized through Asclepius, the god of medicine and
healing. The rod of Asclepius often is shown with a coiled
snake. Of the five daughters of Asclepius, it is Hygeia, goddess
of health, from whom many believe the modern nurse descends
(Fig. 5.1). Drawings and sculptures of Hygeia often depict her
holding a snake, cup, or both. Further symbolism of the rod
of Asclepius is the caduceus. This symbol, which has come to
represent medicine, shows twin snakes (or a two-headed
snake) coiled about the rod of Asclepius. The caduceus was
adopted in the early 1900s to represent medicine and often is
confused with the twin snakes and winged staff of Hermes (a
symbol of commerce). Controversy exists whether the snake
represents the twin battles of life and death, or more directly
the practice of removing worms from open wounds by
winding them about a small staff.
Immediate Past
During the middle ages the role of God in health care expanded
and in times of plague, men and women devoted to God began
to house and care for the sick. Most historical scholars believe
that the nun’s habit of the 15th and 16th centuries is the fore-
runner to the nurses’ cap of the mid-20th century. Nursing
care was a separated workforce, both by sex and time. Male
nurses associated with religious brotherhoods and female
nurses associated with sisterhoods often were separated by
distances that were not easily crossed.5
The end of the Middle Ages marked the onset of modern
history and a further evolution in the healing arts. In particu-
lar global commerce, increasing life span, growing population
centers, and war created new opportunities and challenges for
the medical team. During this time it was realized that a physi-
cian alone could not attend to the sick and wounded. Instead,
nurses around the world took on this role. Florence Nightin-
gale provides one of the earliest and most resilient personifica-
tions of the nursing profession and the role of nursing care in
patient outcome. She had been educated in mathematics and
35
36 Section I—Background
Fig. 5.1  Asclepius and Hygeia.
through careful observation was able to demonstrate a signifi-
cant change in outcomes by modifying the postoperative care
and environment of soldiers during the Crimean War. By war’s
end she had penned texts on care and conduct in both military
and civilian hospitals and helped to establish a center for
nursing training. After the war she wrote Notes on Nursing,
which provided details on how nurses should care for and
monitor patients in their wards.6 In this book her words
provide a glimpse of the future:
“What you want are facts, not opinion—for what can have any
opinion of any value as to whether the patient is better or worse,
excepting the constant medical attendant, or the really observing
nurse?” F. Nightingale (1860)
This insight forecasts two needs in the establishment of fact
and a foundation of monitoring in neurocritical care. First,
the phrase “constant medical attendant” implies a need to
obtain data without gaps across time. Second is the use of the
word “observing” as a method to obtain facts about the patient
rather than provide opinions.
Recent History
In the last century nursing and medicine have witnessed an
astounding evolution in the ability to acquire and store data.
Although nothing has replaced the utility of direct observa-
tion, new tools have been developed to enhance observational
skills and provide for continuous assessment. Feeling for the
threadiness of a pulse was replaced with blood pressure aus-
cultation, then noninvasive blood pressure cuffs and subse-
quently with continuous intra-arterial blood pressure
monitoring and now an effort to use noninvasive blood flow
monitoring.
Just as the tools to find facts have evolved, so have the
methods to communicate and record these facts. Oral
histories and verbal reports of facts were replaced with short
notes. Notes become cumbersome and were placed in charts,
which became disorganized and were replaced with tab-
separated sections. Physicians wrote orders for nurses to
follow (being careful to press down hard enough to ensure
that their notes filtered through multiple layers of carbon-
copy paper. Electronic medical records (EMRs) now dominate
the field and provide the opportunity to think of data acquisi-
tion beyond traditional boundaries.
Changes in how we observe and record data have paral-
leled the changes in who records and observes data. After the
U.S. Civil War, nursing became a primarily female occupa-
tion. The “Angels of the Battlefield” became famous leaders.
Dorothea Dix became the first female superintendent of the
Union Army, Mary Todd Lincoln advocated for nursing
before and after her husband’s assassination, and Clara
Barton became the founder of the American Red Cross. With
each new war the face of nursing changed again. By the end
of World War II, nearly 60,000 women joined the military
nurse corps. With the advantage of rapid transport, medical
evacuation, and mobile army surgical hospital (MASH) tech-
nology, men became trained as corpsmen in subsequent
wars. Many of those corpsmen turned to nursing as a postwar
profession.7
Education of nurses has evolved quickly to meet the growing
demand for what Nightingale predicted: “a really observing
nurse.” In 1873 Linda Richards became the first nurse to grad-
uate as a Nurse with a Diploma in the United States. Thirty
years later, North Carolina became the first state to require
nursing licensure. By 1956 Columbia University became the
first school to award a master’s degree in nursing, and in 1965,
the University of Colorado established the first nurse practi-
tioner position. Although nurse educators and nurse scientists
began entering academia with doctoral degrees in affiliated
areas (e.g., biology, education, psychology), the first PhD pro-
grams specifically in nursing were established in 1954.7
Currently nurses have a wide range of programs and career
paths that include certification and specialization that range
from the licensed practical nurse (LPN) to the doctorally pre-
pared nurse. Table 5.1 provides a list of titles and descriptions
for professional nurses with advanced education to give a
general understanding of the primary role designated to each
discipline.
Nursing Theory—Practice
and Process
The discipline of nursing is a unique blend of art and science
that evolved over many years even before nursing education
became compulsory.8,9 The various views on epistemology in
nursing are summarized by Walker and Avant.10 Empirical
knowing is guided by practice and observation. In the empiri-
cal realm, knowledge is created from an extension of the
senses; it arises from experiment.11 The need for nursing
knowledge as a separate discipline was a major catalyst in the
evolutionary process of nursing.12 As nurses became educated
they contributed to the fundamental knowledge base of
nursing; nursing theory emerged as the driving force to test
and explain nursing practice.13
The advantages to this approach are well described.14,15
Nursing science provides the best evidence for nursing care of

Table 5.1.  A Short Description of Common Nursing Titles
Certified Registered Doctorate in
Registered Nurse Clinical Nurse Specialist Nurse Practitioner Nurse Anesthetist Nursing Practice Doctor of Philosophy
(RN) (CNS)* (NP)* (CRNA) (DNP) (PhD)
Career focus Diploma, associate Expert in clinical nursing, Midlevel provider of Midlevel provider of Leadership roles as Nursing research
or baccalaureate master’s degree direct care, master’s anesthesia care, an advanced-
degree in nursing degree master’s degree practice nurse
Degree Entry to practice as A CNS is an advanced- An advanced-practice Advanced-practice Nursing leaders in To prepare nurse
objectives a nursing practice nurse familiar with nurse who diagnoses nursing. CRNAs interdisciplinary scientists to develop
professional the theory and research and manages common provide anesthetics health care new knowledge for
related to a nursing acute and stable chronic to patients in every teams. Using the the science and
specialty area, such as health problems. NPs practice setting, and tools and skills practice of nursing.
critical care. Traditionally, a perform comprehensive for every type of specific to Graduates will lead
CNS influences patient physical exams, order surgery or procedure. translating interdisciplinary
outcomes through direct and interpret diagnostic They are the sole evidence gained research teams,
clinical practice, tests, obtain specialty anesthesia providers through nursing design, and conduct
collaboration, education, consults, and perform in two thirds of all research into research studies, and
research, consultation and and prescribe rural hospitals, and practice, they disseminate
system leadership. CNSs are therapeutic measures, the main provider of improve systems knowledge for nursing
uniquely prepared, by including most classes anesthesia to of care, and and related disciplines,
experience and education, of medications. NPs, expectant mothers measure particularly addressing
to manage the complexities individually accountable and to men and outcomes of trajectories of chronic
of negotiating complex for their practice and women serving in patient groups, illness and care
health care delivery decisions, collaborate the U.S. Armed populations, and systems.
systems. closely with physicians. Forces. communities.
Curriculum Basic nursing care Improve outcomes for Direct care provider with Direct care provider Translation of Independent researcher
focus patients by influencing prescriptive authority with prescriptive evidence to
systems level, team, and authority practice at the
patient change systems level
Core Nursing assessment, Applied statistics, research Applied statistics. Neurobiology, spinal Evidence-based Philosophy of science
courses† nutrition, design and methodology. Graduate-level and epidural practice, applied and theory
pharmacology, Graduate-level pharmacology, anesthesia, and statistics data development,
human growth pharmacology, assessment, assessment, and advanced airway driven health care advanced research
and development, and pathophysiology. CNS pathophysiology, and management. improvement, design, statistics and
psychology, students are required to courses in an elected Graduate-level financial data analysis
maternal/child, take specialty focus courses specialty (e.g., pharmacology. management and longitudinal and
microbiology, and advanced health management of CRNAs are required budget planning, qualitative research
communication, assessment. critically ill patients, for to achieve a number effective methods. PhD
anatomy and the acute care NP, of clinical internship leadership, health students are required
physiology, and versus child health in hours (varies by systems to take courses related
mental health family care, for the state). transformation to their intended area
nursing family NP) of research.
Prescriptive No No‡ Yes Yes No‡ No
authority
Point of High school BSN BSN BSN BSN or MSN BSN or MSN
entry diploma
Program 2-4 years 2-3 years 2-3 years 2-3 years 2-3 years 4-7 years
length§
Section I—Background

*There are a variety of subspecialties (i.e., family practice, acute care).

†
Course requirements vary by state and university. The core requirements are provided here only as a sample of general requirements.
‡
May have prescriptive authority.
37

§
Full-time student.
38 Section I—Background
patients.16 Caring, although not solely in the domain of
nursing, requires a theoretical basis, just as curing, not solely
in the domain of medicine, requires a theoretical basis.17,18
Theory provides a foundation of knowledge about a problem
or situation; what is known, what are the components, media-
tors, moderators, and co-variables. Importantly, theory pro-
vides the platform from which new hypotheses may be
generated and tested. Nursing knowledge is not limited to
academic research or rigid theories. Theories that are not sup-
ported by scientific testing give way to those that hold up to
such scrutiny.
Nurses at all levels are responsible for generating and testing
theories, and for participating in research. Most recently, the
push for evidence-based practice (EBP) has emerged as a
dominant paradigm in nursing care. Understanding the role
of nursing theory in nursing education and bedside care pro-
vides insight for members of other disciplines that interact
with nurses. One of the most widely explored theories in criti-
cal care setting is Benner’s Novice to Expert Theory.19
The Novice to Expert Theory
Benner’s Novice to Expert theory of nursing care is often cited
in the critical care setting19 and describes five stages of profes-
sional development: (1) novice, (2) advanced beginner, (3)
competent, (4) proficient, and (5) expert. At first appearance
the NCCU is a chaotic environment filled with sights, sounds,
and even smells that rarely are encountered in daily life. In the
midst of this chaos, nurses are focused on the task of patient
care. When nurses first come to the intensive care unit (ICU)
they enter at the novice level.19 When novices, nurses often
may be unsuccessful in recognizing and interpreting key data.
The focus is on task completion, and novices are governed by
rules (medical orders) such as “record B/P q1h, notify medical
doctor if SBP<180 mm Hg.” The novice has yet to gain the
experience required to understand how the context of one
situation varies from another situation.
Through a carefully orchestrated orientation most novices
are prepared to move to the stage of advanced beginner within
1 year.19 Advanced beginners have developed the ability to
quickly and accurately interpret much of the data, most of the
time. The competent nurse is one who has been working in
the ICU for 3 to 5 years and has begun to link incoming data
to long-term outcomes. This stage is embodied by the nurse
who consciously plans a strategy to provide care but may lack
the skills of more advanced professionals to quickly respond
to change.
The nurse who can understand the holism of patient care
and quickly recognize how to acquire and interpret needed
data is considered proficient. Proficiency is not garnered by
the number of years a nurse has worked or by his or her level
of education. Proficiency is demonstrated in actions that
move the patient forward along a continuum of care. The
proficient nurse relies on years of experience to provide care
guided by maxims with subtle nuances for patient-specific
situations.
Critical thinking skills have been directly linked to nurses
advancing from novice to expert.19,20 At the expert level nurses
have accomplished performance aspects that have allowed
them to progress to higher levels of professionalism. The
expert nurse is one who has become an active participant in
the care of ICU patients, relies on experience as a foundation
for action, and is able to recognize the entirety of the situa-
tion.19 To accomplish these performance aspects, the nurse
must possess critical thinking skills and have acquired new
schema.21 The progress from novice nurse to expert nurse
typically occurs over many years. As nurses gain experience
they develop enhanced critical thinking skills, which leads to
the development of new schema and they become more adept
at recognizing and interpreting information.18,19
A clinical example provides additional meaning to these
five stages. It is 2 pm and the nurse prepares to turn her
patient in bed. The novice states, “It is 2 pm; we turn patients
at 2 pm,” but does not recognize the implications of the
patient’s heart rate. The advanced beginner recognizes the
patient’s heart rate as she turns the patient, and then (having
seen results in the past) administers analgesia. The competent
nurse notes the heart rate before she turns her patient; she
plans to administer analgesia and wait for a more normal
heart rate before turning the patient. The proficient nurse
accepts the holism of the situation noting that the patient was
recently given a diuretic; she acquires additional data (blood
pressure, oxygen saturation) and follows the maxim that
“fluid is the answer” before seeking a new medical order.
Finally, the expert nurse is one who can step out of maxims,
rules, and guidelines. Intuitively, the expert talks soothingly
to the patient, touching gently, planning her approach to turn
the patient with assistance only after allowing family to visit
and turning down the radio that inadvertently had been left
playing.
Care Models
Nursing theory leads to the development of care models. An
information-inclusive model of nursing knowledge is pre-
ferred to a model in which knowledge is limited by its point
of origin. Jacobs-Kramer and Chinn22 modify Carper’s23
model of nursing knowledge only to the extent of suggesting
that the model become an active template to synthesize groups
of knowledge from multiple disciplines. Nursing models that
embrace both empirical and nonempirical approaches to grow
and develop the knowledge base provide the greatest potential
to advance the contributions of nursing toward improving
patient outcomes. The American Association of Critical-Care
Nurses (AACN) embraces the synergy model for care of criti-
cally ill patients.24 Additionally, many hospitals and universi-
ties have adopted various evidence-based nursing practice
(EBNP) models.25
Evidence-Based Practice
A sharp distinction exists between EBP and practice sup-
ported by evidence. As the term implies, EBP is that which is
rooted in research.26 The implementation of EBP requires that
the practitioner begin with a question and then evaluate all
the available evidence that surrounds the clinical question;
practice is thus driven from the results of this process.27 Prac-
tice supported by evidence is a significantly less rigorous
process in which the practitioner has a practice pattern or
paradigm and then seeks evidence to support that paradigm.27
Many health care systems have adopted the principles of EBP
and created more individualized models. The Iowa model is
one such example that has been successfully adopted by dif-
ferent health care systems.28,29

Synergy Model
The synergy model focuses on the needs of the patients but
describes a dyadic relationship between nurses’ competencies
and patients’ characteristics.30 The model identifies eight
patient characteristics and eight dimensions of nursing prac-
tice. Levels for the characteristics and dimensions are described
in detail elsewhere.31 The nursing competencies are clinical
judgment, advocacy and moral agency, caring practices, col-
laboration, systems thinking, response to diversity, facilitator
of learning, and clinical inquiry. The patient characteristics are
resiliency, vulnerability, stability, complexity, resource avail-
ability, participation in care, participation in decision making,
and predictability.31 The health care environment is a vital
element in the synergy model and can either promote or
hinder outcomes. For optimal synergy the environment needs
to be one in which the nurse-patient dyad is supported.
In the neurocritical care setting, nursing provides an
optimum contribution only through thoughtful collabora-
tions. Nursing is a vital component of any care delivery model,
and it is primarily tasked with bedside decision making within
the context of primarily physician-derived parameters.32
Nurses benefit from an understanding of the history of the
medical model, and physicians benefit from an understanding
of the history and development of nursing theory and care
delivery models.
Monitoring Tools
A wide variety of monitoring tools exist and are discussed
throughout this textbook. Because the purpose of this chapter
is to provide information about the role of the nurse in mul-
timodal monitoring, noninvasive and invasive monitoring is
only briefly discussed. More complete explanation of the
many modes of monitoring, their benefits and limitations, are
provided in other chapters.
Basic nursing education provides information about the
use and interpretation of only a few of the monitoring tools
used in critical care. For critical care nurses, the term monitor
is often used interchangeably in both the verb and noun
forms. Hence, a nurse may monitor the patient with a monitor!
Most nurses adhere to hospital-specific or unit-specific pro-
tocols and policies. Within the United States these protocols
are developed based on each state’s board of nursing scope of
practice. Although most of the differences noted are small and
seemingly insignificant, when nurses change position or relo-
cate to a new NCCU, there is a need to review all of the policies
and procedures for that new position. Likewise, the physician
who relocates will benefit from understanding the nuances
associated with monitoring in any given unit.
Noninvasive Monitoring
There are many types of monitors described throughout this
book. Noninvasive monitors in general are those that do not
require that you break the skin or invade the body to obtain
data. Most nursing units allow that nurses with a registered
nurse (RN) degree or higher (see Table 5.1) may place nonin-
vasive monitors without a physician’s order.
The most basic and primary mode of noninvasive monitor-
ing has been and will remain the physical examination. Tools
have been developed to assist with this examination (e.g.,
Section I—Background 39
Glasgow Coma Scale, National Institutes of Health [NIH]
stroke score, etc.) with varying degrees of success. In the
NCCU such tools are commonplace. Incumbent on the cadre
of seasoned nurses and physicians is the responsibility to
explore and understand the interrater reliability of each tool
both across populations and within their individual units.
During their primary training most nurses receive only a
few days of clinical experience in an ICU setting. Although
most RNs likely will have been exposed to telemetry monitors,
noninvasive blood pressure monitors, and pulse oximetry, it
is very likely that an RN will graduate without ever being
inside an NCCU. Most hospitals provide a significant orienta-
tion period during which an RN becomes familiar with the
care paradigms and equipment in their new unit. The RN who
initiates and maintains noninvasive neuromonitoring most
likely will be exposed to that particular monitor for the first
time during his or her unit-based orientation.
Invasive Monitoring
Beginning with the moment a decision is made to initiate
invasive monitoring, the nurse and physician should collabo-
rate to provide optimum success. The major portion of inva-
sive monitoring falls within the category of physician-directed,
nurse-driven care. Invasive monitoring takes many forms in
the NCCU. Perhaps the most common example is intra-
arterial blood pressure monitoring. In this example the nurse
and physician must work together to initiate monitoring.
Although most state boards of nursing permit RNs to insert
intra-arterial catheters, it is most common that either an
advanced practice nurse (APN) or physician perform this
activity. While the physician or APN inserts the catheter, the
RN must prepare the pressure line and tubing (Fig. 5.2) Often
the nurse becomes the eyes and ears for the medical team
during bedside procedures to monitor the patient and provide
updates on the patient’s vital signs (Fig. 5.3).
A single monitoring device may be used for multimodal
monitoring. For example, an intra-arterial pressure line will
provide a blood pressure reading, access to obtain blood to
sample biomarkers and blood gases, and the ability to provide
additional blood flow measurements such as cardiac output
and stroke volume. However, with a single monitor actions
or parameters can be mutually exclusive. For example, if
Fig. 5.2  Nurse monitoring vital signs while the physician performs a
procedure.
40 Section I—Background
Fig. 5.3  The nurse may become the eyes and ears for the medical team
during a bedside procedure.
an RN is directed to drain cerebrospinal fluid from an intra-
ventricular catheter, the ability to monitor intracranial pres-
sure (ICP) temporarily ceases. When this is the case, nurses
often are tasked to coordinate the timing during each compo-
nent of monitoring to calibrate monitors to maintain accu-
racy, record data from monitors, interpret the meaning of the
data, and if necessary take action on those findings.32
Nurses employed as RNs do not have prescriptive authority,
and it may be the case that the most appropriate course of
action is to inform the physician or APN of their findings.
When this is the case, the RN must decide if and when to leave
the bedside, a course of action that may leave the patient unat-
tended and unmonitored. Systems designed to reduce the
period of time when patients are unattended are highly desir-
able; delegating to technology is one such system.33 The sim-
plest example of this is setting alarm parameters. More
complex methods are being developed and are discussed later
in this text (see Chapters 4, 40, 43, 45, 48). These technologic
developments and the ability to delegate the role of data acqui-
sition to computers represent an evolution in the practice
paradigm.34-36 Thus technology has allowed ICU nurses the
ability to acquire more data points and to acquire data from
more sources, but data are only useful if they are interpreted
(correctly).
Nursing and Monitoring
In the NCCU, nurses monitor the monitors. Sandelowski11
writes of the link between nursing and technology that “as the
primary machine tenders in health care, nurses often acquire
an understanding of how to apply, operate, and interpret the
products of devices that becomes an integral part of the tacit
know-how of clinical practice.” In this statement she identifies
several vital components: (1) that the nurses are indeed the
primary machine tenders, (2) that there is a knowledge basis
required for nursing care, and (3) that nurses have acquired a
unique ability to interpret meaning from monitors.
Nurses can, and have, walked into a patient’s room and
quickly gathered data on heart rate, systolic and diastolic
blood pressure, oxygen saturation, intracranial and cerebral
perfusion pressure, ventilatory status, and temperature all
while noting an increase in the patient’s level of agitation,
100
HR ICP
80
60
40
20
0
3:00 3:15 3:30 3:45 4:00
3:00 PM 3:15 PM 3:30 PM 3:45 PM 4:00 PM
HR 64 66 68 69 74
ICP 12 13 11 13 12
Fig. 5.4  Trends. The figure shows 1 hour of data. The graph (top)
indicates trends noted from heart rate (HR) and intracranial pressure (ICP)
sampled once per minute. The electronic medical record (EMR) data
(bottom) indicate values that were sampled and documented at
15-minute intervals.
darker than normal urine, low volume of intravenous (IV)
fluid remaining in the primary infusion, soft rattle with each
respiratory cycle, and the conspicuously close proximity of the
patient’s hand to his or her endotracheal tube. Nurses have
indeed developed a unique skill set for multimodal monitor-
ing. Nowhere is this more evident than in the NCCU. Just
as there are significant accomplishments of nursing, it should
be noted that since nurses are neither trained in, nor licensed
as, diagnosticians, they are obligated to communicate their
findings.37
Nurses have the luxury and obligation to make continual
observations but have the tools to report only interval obser-
vations. When taking care of critically ill neuroscience patients,
the nurse often spends long periods of time in one room with
one patient. The nurse therefore may observe continual
changes in a patient’s vital signs or neurologic status. However,
current documentation systems allow for the recording of
only finite amounts of data at discrete intervals. Because the
continual observations are reported at intervals, there is a risk
of ghosting or aliasing. Ghosting is the emergence of false
frequencies and is related to sampling rate.38 In an example
showing 1 hour of data (Fig. 5.4) the graph at top is based on
HR and ICP sampled at 1-minute intervals and indicates that
the patient’s heart rate steadily decreases with periods of wide
fluctuation that appear to correspond to an increasing ICP. Yet
in this same example, the HR and ICP were sampled and
documented in the patient record at 15-minute intervals and
indicate an increasing HR and stable ICP.
The purpose of the medical record (electronic or paper) is
to provide an avenue of communication among health care
providers39—literally, to tell the patient’s story. Each event in
the patient’s time in the ICU has meaning. “The primary role
of documentation should be to clearly describe and commu-
nicate what is going on with the patient.”39 Only by capturing
and recording data from the many monitors can nurses
provide physicians with the information they need. The
duality of this obligation is noteworthy. Just as nurses are
obligated to collect and present a comprehensive and com-
plete dataset, physicians are obligated to ensure that they in
 Section I—Background 41

turn evaluate each data element and if not, should not order
that data to be collected. Box 5.1  Considerations During Intrahospital
Transport
Monitoring During Transport  Pretransport communication
 Is this transport necessary?
The mantra of the nurse who monitors a patient during intra-
 Is the receiving area ready?
hospital transport is: “Plan for the worst—hope for the best—
 Transport personnel
be ready for anything.” Intrahospital transport, defined as any
 How many?
time when a patient is moved from the primary area of care
 Qualifications
to another location within the same facility, requires that the
 Do the transport personnel have coverage for other
patient receive the same level of care throughout the intrahos-
patients they may be caring for in the unit?
pital transport as would have been received had he or she
 Monitoring equipment
remained in the primary care area.40,41 The medical record
 The patient should receive the same minimum standard of
should indicate that the patient was monitored throughout
care during transport as was prescribed in the critical care
the transport and indicate the patient’s response during each
unit.
phase of transport. With most current systems of data acquisi-
 Battery life
tion, a loss of some patient data is inevitable, and in particular
 Physical location of the monitor during transport
this often may occur during patient transport.
 Time and distance
The very nature of critical care drives the need for addi-
 How long will the patient and nurse be involved in the
tional investigational and treatment options that exist only
intrahospital transport?
beyond the confines of the patient’s room. Nearly half of all
 What physical barriers exist between the critical care unit
patients admitted to critical care experience at least one intra-
and the transport destination?
hospital transport, and this most often occurs early during the
 Is the destination on a different floor?
critical care stay.42 It is well established that any decision to
 Is the destination readily accessible?
transport a patient must be made by balancing the potential
 Does the patient need to pass through high-access
risks and benefits (Box 5.1).41,43
areas?
For each episode of intrahospital transport there are three
 Resources
phases during which the patient is exposed to different risk
 Phase I and phase III
factors. Phase I begins with the decision that transport will
 Must be able to competently deliver the standard of
occur and ends when the patient reaches the intrahospital
care
destination. Phase II is a maintenance phase that begins when
 Emergency equipment
the patient arrives at a new destination and persists through-
 Airway and cardiac resuscitation equipment
out the patient’s stay at the new location. Phase III encom-
 Emergency medications
passes that period of time during which the patient is
 Phase II
transported back to the ICU and concludes only when the
 Must be equal to or greater than the standard of care
patient is stable or has returned to pretransport status.44 Spe-
in the critical care unit
cialization and education along with highly prepared staff are
 Documentation
associated with lower rates of adverse events during intrahos-
 Ability to acquire and record data
pital transport.45
 Ability to store and retrieve data

Phase I—From ICU to Destination

During phase I of the transport, most commonly it is the staff
nurse or a combination of staff nurse and respiratory therapist
who monitor the patient.45 The physician must clearly balance
the need to travel against the risk of problems during travel.
Having made the decision that intrahospital transport is indi-
cated, there is a joint responsibility to determine monitoring
needs. The frequency of intrahospital transport may be a
factor in the purchase of new monitoring devices. Battery life,
portability, data storage and transfer, and even the weight of
a monitor will factor into the decision of which monitors are
used during transport.
Loss of data during phase I of the transport may be associ-
ated with a combination of variables. Data may be lost if the
signal quality is such that the data are uninterpretable; such is
the case with artifact because patient movement is a common
cause of artifact.46 Transport inevitably creates a variety of
movement. Excessive artifact may result in data not being
displayed or in displayed data that do not reflect the true
patient condition. Few studies have explored data loss during
physical transport.40,41
Phase II—Maintenance at the
Intrahospital Destination
The degree of multimodal monitoring that takes place during
phase II of intrahospital transport often is determined by
temporal parameters. Expert level and experienced nursing
staff know to expect that a noncontrasted head computed
tomography (CT) scan typically is performed in only a few
minutes. The decision to re-level transducers, reconnect mon-
itors, or troubleshoot artifact is balanced against the risk of
keeping the patient flat for a prolonged period of time.47
Several studies have begun to explore solutions to this conun-
drum.40,42,43 One such example is the development of a scoring
system to grade the level of acuity and risk associated with the
event.45 Higher-acuity patients require additional monitoring,
resources, and staffing. Alternatively portable CT scanners in
the ICU can help reduce the need for some intrahospital
transport.
42 Section I—Background
Phase III—Returning to the ICU
Phase III begins when the patient is transported back to the
point of origin (the NCCU). The goals and concerns during
this portion of phase III are the same as those during phase I.
On return to the NCCU the staff must simultaneously tend
the monitoring systems while assessing the patient. The nurse
should perform a comprehensive patient assessment that
includes ensuring that all access lines, drains, and life-support
devices are secure, intact, and properly functioning. In addi-
tion, the nurse should evaluate the effects of any medications
or interventions performed during the intrahospital transport
and should report any patient condition changes.
The nurse must ensure that all monitoring devices func-
tion properly. Devices that were taken with the patient during
transport may need to be recalibrated or re-leveled. Alarm
parameters should be reviewed to ensure that any parameters
that were reset during transport are now set to the desired
level, and any device that requires electricity again is plugged
into an outlet. Many NCCUs now use an EMR. Monitors
may need to be synchronized to central hubs such that
there is a transfer of data to the EMR. The final phase of an
intrahospital transport is marked by a combination of two
conditions: (1) the patient has been returned to the neuro-
critical care unit, and (2) the patient has returned to a stable
condition.44
Collaboration
Collaboration is a key element in monitoring the patient with
a neurologic injury. Nurses who take care of patients must
collaborate on a variety of levels. The nurse-to-nurse collabo-
ration takes place both within and between nursing shifts.
Nurses collaborate with midlevel practitioners (APNs and
physician assistants [PAs]) as well as with medical residents
and attending physicians to modify the patient’s plan of care.
Nurses also may collaborate with ancillary department per-
sonnel (e.g., pharmacy, respiratory care, physical therapy) to
coordinate the timing of care delivery. Finally, the nurse will
collaborate with the patient and family to determine the goals
of care and provide education.
Nurse-to-Nurse Collaboration
The nurse-to-nurse collaboration takes place on two fronts
and in particular during the nursing handover. The nursing
handover (also called shift report, or handoff) occurs when one
nurse ends her or his shift and hands over treatment of the
patient to the next nurse. During this time, the two nurses
communicate about the events and observations made during
the previous shift.48 Handover typically lasts from 15 to 30
minutes but currently is not a standardized process.49 The
oncoming nurse and the care nurse whose shift is ending work
jointly to review the patient’s medical history and events of
the day and develop a plan of care for the next shift.48,49
The second form of nurse-to-nurse collaboration occurs
during and throughout the nursing shift. Throughout a
nursing shift, the expert and proficient nurse provide informal
education to the novice, advanced beginner, and competent
nurses.19 It is not possible for any one nurse to be in more than
one place at one time, and often nurses are assigned to provide
care to multiple patients. Although most bedside monitors
have the capacity to send and display some information from
any one given patient to the room of another patient, few if
any monitors provide a full spectrum of readily available data
to be displayed outside of a single patient room. A nurse who
cares for two patients must collaborate with her peers to
provide continual monitoring. This action may be a simple
request to listen for alarms or as complicated as the shift hand­
over. If one nurse needs help she asks other nurses to come to
the bedside—they collaborate when one nurse makes a new
observation.
Nurse and Midlevel Practitioner
Technology has provided increased levels of information to
obtain and monitor critically ill patients. To meet this need,
the number of eyes and ears needed to run an ICU also has
increased. Many NCCUs now employ a variety of midlevel
practitioners (see Table 5.1). The APNs or PAs typically have
a minimum of a master’s degree. During the past decade mid-
level practitioners have begun to take more active roles in the
daily functions of how an ICU runs. Because most midlevel
practitioners have at least a limited prescriptive authority, they
may be the first person the staff nurse (RN) approaches when
seeking a new order or seeking to clarify an existing order.
Nurse and Physician
Without active collaboration between the nurse who provides
direct care and the physician who directs care, the system can
fall apart. Physicians must understand not only their own
respective roles, but also must have an understanding of the
role, abilities, and limitations of their counterpart (the nurse).
Active collaboration on the part of the nurse means providing
the physician with the products of monitoring. This exchange
may be in a more formal or ritualized pattern; for example,
during morning rounds the nurse may report the patient’s
condition to the ICU team (Fig. 5.5). Other collaborations
may occur in less formal exchanges, when the nurse calls or
speaks with the physician or physician designee. Whether in
oral or written form, the physician requires data from which
to make vital decisions. Equally, the physician must work
with the nursing staff to develop and continually refine the
plan of care.
Fig. 5.5  The neurocritical care team discussing patient care during
rounds.

Continuous Education
Neurocritical care is a relatively new field. New treatment
options or strategies, medications, and interventions continu-
ally are being developed. New methods and tools available to
monitor patients also are being developed and introduced
at an amazing pace. To maintain competence and provide
the highest level of care, nurses (and physicians) must engage
in a program that encourages and provides continuing
education.
Nursing Education
A nurse officially becomes a nurse upon graduation from an
accredited school of nursing. However, only after passing the
licensure examination is a nurse registered with the state
board of nursing and conferred the status of RN. RN status
indicates that a nurse has met the minimum licensing stan-
dards and may begin professional practice. During nursing
school, few nurses have adequate exposure to the NCCU to
be considered competent enough to provide care indepen-
dently. For this reason, the orientation period (discussed
earlier) often is the first post-degree education that RNs
receive.
Nationally, several programs have been developed that are
geared to provide specific aspects of neuroscience knowledge
and critical care knowledge to the novice and advanced begin-
ner nurse. The Foundations of Neuroscience Nursing (FNN)
program was developed by the American Association of Neu-
roscience Nurses.50 The Essentials of Critical Care Orientation
(ECCO) was created by the AACN orientation program.51
The FNN program is an electronically distributed educa-
tional package. The course is composed of 21 modules that
range between 30 and 150 minutes in length and covers a
broad spectrum of topics related to the care of patients with
neurologic and neurosurgical injuries. It is offered to indi-
viduals who wish to increase their knowledge and to institu-
tions that wish to provide the content to one or more
employees. An option to negotiate a multisite contract also
exists. The target audience for the FNN course is nurses with
less than 2 years of experience who provide care to neurosci-
ence patients. The program is designed to be facilitated by an
instructor (experienced nurse or nurse educator) from the
institution that purchased the product.50
The most recent version of the ECCO program is a series
of more than 40 lessons in 10 modules. ECCO is specifically
targeted as an orientation package to nurses who enter a criti-
cal care unit and covers each of the major human biologic
systems and the most common diseases found in the critical
care setting. Completion of this Internet-based course pro-
vides the RN with 69 contact hours.51
In most states the board of nursing requires that nurses
demonstrate that they have maintained their nursing compe-
tency by obtaining continuing education. Nurses obtain and
maintain contact hours through a variety of venues. Local,
regional, and national conferences offer a chance for nurses to
obtain continuing education units (CEUs). Although these
annual events are popular, time, travel, and money limit the
ability of many to attend. Hospital-based inservices often
provide a unique opportunity for low-cost education.
Hospital- and unit-based inservices are a nearly universal
experience in the nursing lexicon. Nursing inservices offer the
Section I—Background 43
benefit of low production cost and ease of attendance. A hos-
pital employee (e.g., nurse educator, APN, experienced nurse,
or hospital administrator), a physician faculty member or
invited guest, or an industry representative may present the
inservice. The obvious advantages of inservicing should be
balanced against the risk of bias. It is possible that a presenta-
tion could provide material that is out-of-date or be designed
to provide only a single view of a topic. Repeated inservicing
by a limited number of individuals may result in the same
material being delivered again and again and so prevent the
influx of new ideas.
Patient and Family Education
Another aspect of education about new monitoring is the
need to educate patients and families. The concept of open-
visiting, although not universally embraced, has become wide-
spread. Gone are the days when family members visited
patients twice each day in 10-minute intervals. Today loved
ones are omnipresent and considered part of the care team
(see Chapter 3).52 As members of the care team, patients and
family members must be educated to understand the moni-
tors. No two patients and no two families are alike, and edu-
cational needs must be tailored to the individual. Promoting
patient and family education requires is a collaborative effort
between physicians and nurses.53
Conclusion
As a distinct discipline, neurocritical care is continually
expanding. New procedures and treatment options have pro-
vided patients with neurologic injury and illness the oppor-
tunity to lead full, productive lives long after their hospital
discharge. A vast array of monitoring options already exist
that can be used to observe for changes in patient condi-
tions, and newer devices are in development. To be optimally
effective, these current and future monitoring devices will
require highly educated staff that specialize in the care and
treatment of neurocritically ill patients. The role of multi-
modal monitoring in neurocritical care is expanding; the
role of the nurse in critical care is equally expanding to meet
this need.
References
1. Santayana G. The life of reason. New York: C. Scribner’s Sons; 1922.
2. Olson DM. Multimodal neurological monitoring. In: Kaplow R, Hardin SR,
editors. Critical care nursing: synergy for optimal outcomes. Sudbury, Mass:
Jones & Bartlett; 2007. p. 359–74.
3. Davidson GR. Medicine through the ages. London, New York: Methuen; Roy
Publishers; 1968.
4. Wreszinski W. Königliche Museen zu Berlin. Der grosse medizinische
Papyrus des Berliner Museums (Pap. Berl. 3038) in Facsimile und Umschrift
mit Übersetzung, Kommentar und Glossar. Leipzig: J.C. Hinrichs; 1909.
5. Mackintosh C. A historical study of men in nursing. J Adv Nurs 1997;26(2):
232–6.
6. Nightingale F. Florence Nightingale’s notes on nursing. Skretkowicz V, editor.
Reviewed with additions ed. London: Scutari Press; 1992.
7. Shuttleworth A, O’Dowd A. 100 years of nursing. Nurs Times 2005;101(19):
17–28, 30–18, 40–18.
8. Barrett EA. What is nursing science? Nurs Sci Q 2002;15(1):51–60.
9. Parse RR. Nursing: the discipline and the profession. Nurs Sci Q 1999;12(4):
275–6.
10. Walker LO, Avant KC. Strategies for theory construction in nursing. 3rd ed.
Norwalk, Conn: Appleton & Lange; 1995.
44 Section I—Background
11. Sandelowski M. Knowing and forgetting: the challenge of technology for a
reflexive practice science of nursing. In: Thorne SE, Hayes VE, editors.
Nursing praxis knowledge and action. Thousand Oaks, Calif: Sage
Publications; 1997. p. 69–85.
12. Donaldson SK, Crowley DM. The discipline of nursing. Nurs Outlook 1978;
26(2):113–20.
13. Verran JA. The value of theory-driven (rather than problem-driven)
research. Semin Nurse Manag 1997;5(4):169–72.
14. Chinn PL, Jacobs MK. Theory and nursing: a systematic approach. St. Louis:
CV Mosby; 1983.
15. Sidani S, Braden CJ. Evaluating nursing interventions: a theory-driven
approach. Thousand Oaks, CA: Sage Publications; 1998.
16. Parse RR. Paradigms: a reprise. Nurs Sci Q 2000;13(4):275–6.
17. Watson J, Smith MC. Caring science and the science of unitary human
beings: a trans-theoretical discourse for nursing knowledge development.
J Adv Nurs 2002;37(5):452–61.
18. Beeby JP. Intensive care nurses’ experiences of caring. Part 1: consideration
of the concept of caring. Intensive Crit Care Nurs 2000;16(2):76–83.
19. Benner P. From novice to expert: excellence and power in clinical nursing
practice. Menlo Park, CA: Addison-Wesley, Nursing Division; 1984.
20. McKane CL, Schumacher L. Professional advancement model for critical care
orientation. J Nurs Staff Dev 1997;13(2):88–91.
21. Rashotte J, Thomas M. Incorporating educational theory into critical care
orientation. J Contin Educ Nurs 2002;33(3):131–7.
22. Jacobs-Kramer MK, Chinn PL. Perspectives on knowing: a model of nursing
knowledge. In: Nicoll LH, editor. Perspectives on nursing theory. 3rd ed.
Philadelphia: Lippincott; 1997. p. 323–30.
23. Carper B. Fundamental patterns of knowing in nursing. ANS Adv Nurs Sci
1978;1(1):13–23.
24. Hardin SR, Kaplow R. Synergy for clinical excellence: the AACN synergy
model for patient care. Sudbury, MA: Jones & Bartlett; 2005.
25. Keele R. Nursing research and evidence-based practice: ten steps to success.
Sudbury, MA: Jones & Bartlett Learning; 2011.
26. Youngblut JM, Brooten D. Evidence-based nursing practice: why is it
important? AACN Clin Issues 2001;12(4):468–76.
27. Stillwell SB, Fineout-Overholt E, Melnyk BM, et al. Evidence-based practice,
step by step: asking the clinical question: a key step in evidence-based
practice. Am J Nurs 2010;110(3):58–61.
28. Titler MG, Kleiber C, Steelman VJ, et al. The Iowa model of evidence-based
practice to promote quality care. Crit Care Nurs Clin North Am 2001;13(4):
497–509.
29. Stetler CB, Ritchie JA, Rycroft-Malone J, et al. Institutionalizing evidence-
based practice: an organizational case study using a model of strategic
change. Implement Sci 2009;4:78.
30. Curley MA. Patient-nurse synergy: optimizing patients’ outcomes. Am J Crit
Care 1998;7(1):64–72.
31. Kaplow R, Hardin SR. Critical care nursing: synergy for optimal outcomes.
Sudbury, MA: Jones & Bartlett; 2007.
32. Olson DM, Graffagnino C. Consciousness, coma, and caring for the
brain-injured patient. AACN Clin Issues 2005;16(4):441–55.
33. Crocker C, Timmons S. The role of technology in critical care nursing. J Adv
Nurs 2009;65(1):52–61.
34. Cuthbertson BH, Boroujerdi M, Prescott G. The use of combined
physiological parameters in the early recognition of the deteriorating acute
medical patient. J R Coll Physicians Edinb 2010;40(1):19–25.
35. Stylianides N, Dikaiakos MD, Gjermundrod H, et al. Intensive care window:
real-time monitoring and analysis in the intensive care environment. IEEE
Trans Inf Technol Biomed 2011;15(1):26–32.
36. Burgess LP, Herdman TH, Berg BW, et al. Alarm limit settings for early
warning systems to identify at-risk patients. J Adv Nurs 2009;65(9):1844–52.
37. Lacey SR, Cox KS. Nursing: key to quality improvement. Pediatr Clin North
Am 2009;56(4):975–85.
38. Webster JG, Clark JW. Medical instrumentation: application and design. 4th
ed. Hoboken, NJ: John Wiley & Sons; 2010.
39. Schiff GD, Bates DW. Can electronic clinical documentation help prevent
diagnostic errors? N Engl J Med 2010;362(12):1066–9.
40. Day D. Keeping patients safe during intrahospital transport. Crit Care Nurse
2010;30(4):18–32; quiz 33.
41. Warren J, Fromm Jr RE, Orr RA, et al. Guidelines for the inter- and
intrahospital transport of critically ill patients. Crit Care Med 2004;32(1):
256–62.
42. Voigt LP, Pastores SM, Raoof ND, et al. Review of a large clinical series:
intrahospital transport of critically ill patients: outcomes, timing, and
patterns. J Intensive Care Med 2009;24(2):108–15.
43. Jarden RJ, Quirke S. Improving safety and documentation in intrahospital
transport: development of an intrahospital transport tool for critically ill
patients. Intensive Crit Care Nurs 2010;26(2):101–7.
44. Noa Hernandez JE, Gonzalez EC, Romero JM, et al. [Intrahospital
transportation of the seriously ill patient. The need for an action guideline.].
Enferm Intensiva 2011;22(2):74–7.
45. Kue R, Brown P, Ness C, et al. Adverse clinical events during intrahospital
transport by a specialized team: a preliminary report. Am J Crit Care 2011;
20(2):153–61; quiz 162.
46. Jevon P. An introduction to electrocardiogram monitoring. Nurs Crit Care
2010;15(1):34–8.
47. Bekar A, Ipekoglu Z, Tureyen K, et al. Secondary insults during intrahospital
transport of neurosurgical intensive care patients. Neurosurg Rev 1998;
21(2-3):98–101.
48. Chaboyer W, McMurray A, Johnson J, et al. Bedside handover: quality
improvement strategy to “transform care at the bedside.” J Nurs Care Qual
2009;24(2):136–42.
49. Manser T, Foster S. Effective handover communication: an overview of
research and improvement efforts. Best Pract Res Clin Anaesthesiol 2011;
25(2):181–91.
50. Foundations of Neuroscience Nursing [computer program]: Helmick K,
Fitzsimmons B, Forsyth L, editors. Glenview, Ill: American Association of
Neuroscience Nurses; 2006.
A complete list of references for this chapter can be found online at
www.expertconsult.com.

References
1. Santayana G. The life of reason. New York: C. Scribner’s Sons; 1922.
2. Olson DM. Multimodal neurological monitoring. In: Kaplow R, Hardin SR,
editors. Critical care nursing: synergy for optimal outcomes. Sudbury, Mass:
Jones & Bartlett; 2007. p. 359–74.
3. Davidson GR. Medicine through the ages. London, New York: Methuen; Roy
Publishers; 1968.
4. Wreszinski W. Königliche Museen zu Berlin. Der grosse medizinische
Papyrus des Berliner Museums (Pap. Berl. 3038) in Facsimile und Umschrift
mit Übersetzung, Kommentar und Glossar. Leipzig: J.C. Hinrichs; 1909.
5. Mackintosh C. A historical study of men in nursing. J Adv Nurs 1997;26(2):
232–6.
6. Nightingale F. Florence Nightingale’s notes on nursing. Skretkowicz V, editor.
Reviewed with additions ed. London: Scutari Press; 1992.
7. Shuttleworth A, O’Dowd A. 100 years of nursing. Nurs Times 2005;101(19):
17–28, 30–18, 40–18.
8. Barrett EA. What is nursing science? Nurs Sci Q 2002;15(1):51–60.
9. Parse RR. Nursing: the discipline and the profession. Nurs Sci Q 1999;12(4):
275–6.
10. Walker LO, Avant KC. Strategies for theory construction in nursing. 3rd ed.
Norwalk, Conn: Appleton & Lange; 1995.
11. Sandelowski M. Knowing and forgetting: the challenge of technology for a
reflexive practice science of nursing. In: Thorne SE, Hayes VE, editors.
Nursing praxis knowledge and action. Thousand Oaks, CA: Sage
Publications; 1997. p. 69–85.
12. Donaldson SK, Crowley DM. The discipline of nursing. Nurs Outlook 1978;
26(2):113–20.
13. Verran JA. The value of theory-driven (rather than problem-driven)
research. Semin Nurse Manag 1997;5(4):169–72.
14. Chinn PL, Jacobs MK. Theory and nursing: a systematic approach. St. Louis:
C.V. Mosby; 1983.
15. Sidani S, Braden CJ. Evaluating nursing interventions: a theory-driven
approach. Thousand Oaks, CA: Sage Publications; 1998.
16. Parse RR. Paradigms: a reprise. Nurs Sci Q 2000;13(4):275–6.
17. Watson J, Smith MC. Caring science and the science of unitary human
beings: a trans-theoretical discourse for nursing knowledge development.
J Adv Nurs 2002;37(5):452–61.
18. Beeby JP. Intensive care nurses’ experiences of caring. Part 1: consideration
of the concept of caring. Intensive Crit Care Nurs 2000;16(2):76–83.
19. Benner P. From novice to expert: excellence and power in clinical nursing
practice. Menlo Park, CA: Addison-Wesley, Nursing Division; 1984.
20. McKane CL, Schumacher L. Professional advancement model for critical care
orientation. J Nurs Staff Dev 1997;13(2):88–91.
21. Rashotte J, Thomas M. Incorporating educational theory into critical care
orientation. J Contin Educ Nurs 2002;33(3):131–7.
22. Jacobs-Kramer MK, Chinn PL. Perspectives on knowing: a model of nursing
knowledge. In: Nicoll LH, editor. Perspectives on nursing theory. 3rd ed.
Philadelphia: Lippincott; 1997. p. 323–30.
23. Carper B. Fundamental patterns of knowing in nursing. ANS Adv Nurs Sci
1978;1(1):13–23.
24. Hardin SR, Kaplow R. Synergy for clinical excellence: the AACN synergy
model for patient care. Sudbury, MA: Jones & Bartlett; 2005.
25. Keele R. Nursing research and evidence-based practice: ten steps to success.
Sudbury, MA: Jones & Bartlett Learning; 2011.
26. Youngblut JM, Brooten D. Evidence-based nursing practice: why is it
important? AACN Clin Issues 2001;12(4):468–76.
27. Stillwell SB, Fineout-Overholt E, Melnyk BM, et al. Evidence-based practice,
step by step: asking the clinical question: a key step in evidence-based
practice. Am J Nurs 2010;110(3):58–61.
28. Titler MG, Kleiber C, Steelman VJ, et al. The Iowa model of evidence-based
practice to promote quality care. Crit Care Nurs Clin North Am 2001;13(4):
497–509.
29. Stetler CB, Ritchie JA, Rycroft-Malone J, et al. Institutionalizing evidence-
based practice: an organizational case study using a model of strategic
change. Implement Sci 2009;4:78.
Section I—Background 44.e1
30. Curley MA. Patient-nurse synergy: optimizing patients’ outcomes. Am J Crit
Care 1998;7(1):64–72.
31. Kaplow R, Hardin SR. Critical care nursing: synergy for optimal outcomes.
Sudbury, MA: Jones & Bartlett; 2007.
32. Olson DM, Graffagnino C. Consciousness, coma, and caring for the
brain-injured patient. AACN Clin Issues 2005;16(4):441–55.
33. Crocker C, Timmons S. The role of technology in critical care nursing. J Adv
Nurs 2009;65(1):52–61.
34. Cuthbertson BH, Boroujerdi M, Prescott G. The use of combined
physiological parameters in the early recognition of the deteriorating acute
medical patient. J R Coll Physicians Edinb 2010;40(1):19–25.
35. Stylianides N, Dikaiakos MD, Gjermundrod H, et al. Intensive care window:
real-time monitoring and analysis in the intensive care environment. IEEE
Trans Inf Technol Biomed. 2011;15(1):26–32.
36. Burgess LP, Herdman TH, Berg BW, et al. Alarm limit settings for early
warning systems to identify at-risk patients. J Adv Nurs 2009;65(9):1844–52.
37. Lacey SR, Cox KS. Nursing: key to quality improvement. Pediatr Clin North
Am 2009;56(4):975–85.
38. Webster JG, Clark JW. Medical instrumentation: application and design. 4th
ed. Hoboken, NJ: John Wiley & Sons; 2010.
39. Schiff GD, Bates DW. Can electronic clinical documentation help prevent
diagnostic errors? N Engl J Med 2010;362(12):1066–9.
40. Day D. Keeping patients safe during intrahospital transport. Crit Care Nurse
2010;30(4):18–32; quiz 33.
41. Warren J, Fromm Jr RE, Orr RA, et al. Guidelines for the inter- and
intrahospital transport of critically ill patients. Crit Care Med 2004;32(1):
256–62.
42. Voigt LP, Pastores SM, Raoof ND, et al. Review of a large clinical series:
intrahospital transport of critically ill patients: outcomes, timing, and
patterns. J Intensive Care Med 2009;24(2):108–15.
43. Jarden RJ, Quirke S. Improving safety and documentation in intrahospital
transport: development of an intrahospital transport tool for critically ill
patients. Intensive Crit Care Nurs 2010;26(2):101–7.
44. Noa Hernandez JE, Gonzalez EC, Romero JM, et al. [Intrahospital
transportation of the seriously ill patient. The need for an action guideline.].
Enferm Intensiva 2011;22(2):74–7.
45. Kue R, Brown P, Ness C, et al. Adverse clinical events during intrahospital
transport by a specialized team: a preliminary report. Am J Crit Care 2011;
20(2):153–61; quiz 162.
46. Jevon P. An introduction to electrocardiogram monitoring. Nurs Crit Care
2010;15(1):34–8.
47. Bekar A, Ipekoglu Z, Tureyen K, et al. Secondary insults during intrahospital
transport of neurosurgical intensive care patients. Neurosurg Rev
1998;21(2-3):98–101.
48. Chaboyer W, McMurray A, Johnson J, et al. Bedside handover: quality
improvement strategy to “transform care at the bedside.” J Nurs Care Qual
2009;24(2):136–42.
49. Manser T, Foster S. Effective handover communication: an overview of
research and improvement efforts. Best Pract Res Clin Anaesthesiol 2011;
25(2):181–91.
50. Foundations of Neuroscience Nursing [computer program]: Helmick K,
Fitzsimmons B, Forsyth L, editors. Glenview, Ill: American Association of
Neuroscience Nurses; 2006.
51. Kaddoura MA. Effect of the essentials of critical care orientation (ECCO)
program on the development of nurses’ critical thinking skills. J Contin
Educ Nurs 2010;41(9):424–32.
52. Livesay S, Gilliam A, Mokracek M, et al. Nurses’ perceptions of open visiting
hours in neuroscience intensive care unit. J Nurs Care Qual 2005;20(2):
182–9.
53. Farahani MA, Sahragard R, Carroll JK, et al. Communication barriers to
patient education in cardiac inpatient care: a qualitative study of multiple
perspectives. Int J Nurs Pract 2011;17(3):322–8.

Quality Assessment in the
Neurocritical Care Unit
Anoma Nellore, Peter D. le Roux, and David A. Horowitz
Introduction
Intensive care is one of the most expensive forms of care
delivered by a hospital system. In 1993, intensive care repre-
sented 25% to 30% of acute hospital costs and constituted 1%
of the gross national product (GNP).1 It has become increas-
ingly important in an era of public and private accountability
to measure and benchmark quality of intensive care unit
(ICU) care; however, this is a complicated task. Quality
measures of health care are divided into three broad areas:
(1) process (appropriate delivery of health care), (2) outcome
(measured endpoints of care), and (3) structure (adequate
resources to provide health care).2 Process measures include
appropriate uses of stress ulcer and deep vein thrombosis
(DVT) prophylaxis, appropriate transfusion thresholds, daily
interruption of sedation, appropriate glycemic control, and
use of aspirin and beta-blockers in acute myocardial infarc-
tions among other benchmarks. Outcome measures include
such parameters as mortality, incidence of catheter-related
bloodstream infections, or rates of ventilator-associated pneu-
monia (VAP). Typical structural quality indicators include the
presence of an ICU medical director, daily rounds by an inten-
sivist, appropriate ICU nurse-to-patient ratio, and multidisci-
plinary rounds involving a pharmacist.
Use of each type of quality metric has its advantages and
disadvantages. For example, although structural quality indi-
cators are easy to define and measure, they may not translate
to immediate improvement in care. Process quality indicators
may require tailoring to specific disease processes and may
become cumbersome. Although the public (and health care
providers) are most interested in outcome metrics, data that
pertain to outcome measures may take a long time to obtain
and there is little definition of appropriate standards. In addi-
tion, outcome measures often require risk factor stratification
for valid comparisons and although much work has been done
in general critical care, there has been less research on bench-
marking specific to neurocritical care.
History of and Impetus for
Qualitative Initiatives
The Institute of Medicine (IOM) defines quality as “the degree
to which health services for individuals and populations
© Copyright 2013 Elsevier Inc. All rights reserved.
Chapter
6  
I
increase the likelihood of desired health outcomes and are
consistent with current professional knowledge.” Patient safety
is integral to quality because it is difficult to provide successful
care when safety is compromised; that is, it is difficult to
examine quality and safety in isolation. The IOM defines
safety as “the absence of clinical error, either by commission
[unintentionally doing the wrong thing] or omission [unin-
tentionally not doing the right thing].” This is similar to the
World Health Organization (WHO) definition.
The concept of quality in health care is central to clinical
practice and is as old as medicine. Indeed the phrase primum
non nocere (first do no harm) dates back to Hippocrates and is
one of the principal tenets of medical ethics. In the early 20th
century, Codman advocated public reporting of outcomes by
both physicians and hospitals.3 Modern quality initiatives
(QIs) started in other industries that realized that unexplained
variation leads to poor quality, and processes that decrease
variation coupled with continuous evaluation can improve
quality and therefore profit. For example, in the 1950s Japa-
nese car manufacturers applied Deming’s theory that improv-
ing quality will reduce expenses while increasing productivity
and market share; this led to an international demand for
Japanese automobiles.4 For example, Toyota became a model
of operational excellence, in part through kaizen, a Japanese
philosophy of continuous improvement. However, in 2009,
there was a recall of more than 8 million vehicles and a decline
in stock value thought by some to be a subtle abandonment
of these principles. In the 1980s Donabedian introduced the
structure, process, and outcome paradigm to assess quality in
health care but also cautioned that before assessment can
begin it must be determined how quality is to be defined,
something that is still ongoing.5 During the past decade there
has been increased focus on quality and safety of medical care
with the recognition of the high cost of health care, potential
for harm, and a perceived need for transparency about hospi-
tals to the public.2,6-8 The report “To Err Is Human” by the
IOM in the United States in 1999 led to an increased focus on
safety and quality of care. In 2009 the European Society of
Intensive Care Medicine (ESICM) initiated a task force to
improve the safety and quality of care provided to ICU
patients, leading to a publication of prospectively defined
indicators to improve quality of critical care.9 QIs are increas-
ingly a focus of hospitals because pay-for-performance
(P4P) programs are under development by organizations that
45
46 Section I—Background
purchase health care such as the Leapfrog Group and the
Centers for Medicare & Medicaid Services (CMS). To measure
quality a combination of process and outcome indicators may
be best,10 but measuring ICU performance can be a complex
and time-consuming task11 because it depends on multiple
domains such as patient- and family-centered outcomes, the
work environment, personnel satisfaction, and economic per-
formance among others.
Process and Outcome
Quality Indicators
Donabedian proposed that structure, process, and outcome be
reviewed to improve quality of health care.5 Processes of care
usually are the easiest to target initially, modify them if neces-
sary, and then measure outcomes associated with the process.
Outcomes such as mortality are easy to measure. However,
many variables and processes influence mortality, and so
attributing variation in mortality to a single intervention or
process change is not always feasible. There thus can be some
overlap between outcome and process in QIs. For example,
process can include use of available technology and staff edu-
cation and training, whereas outcome may include resource
use and appropriate use of diagnostic or therapeutic proce-
dures. The following text reviews process and outcome indica-
tors together.
Risk Adjustment
Quality and benchmarking initiatives require accurate strati-
fied risk adjustment. The main prognostic models used to
assess overall illness severity of adults admitted to the ICU
are the Acute Physiology and Chronic Health Evaluation
(APACHE), Simplified Acute Physiology Score (SAPS), and
Mortality Probability Model (MPM). The SAPs and MPM
have been updated to their third versions and APACHE to its
fourth version (see also Chapter 10).
Mortality—Outcome
Short-term (e.g., 30 days) mortality often is used to measure
success of management strategies in general ICU care.
However, mortality as an outcome quality indicator has limi-
tations. Crude mortality cannot be used to evaluate ICU care
because it does not adjust for underlying patient mix or
disease state. The standardized mortality ratio (SMR) com-
pares the observed to the expected mortality rate. The expected
mortality rate may be derived from a number of physiologic
parameters including APACHE scoring, SAPS, and the MPM.
These models, however, all have limitations in external valida-
tion series when applied to data sets outside their original
population.12-14 For example, Sirio et al. observed consistent
overestimation of the SMR (actual/predicted mortality) devel-
oped from a national sample that was applied to a community-
based sample of hospital ICUs in a metropolitan area.6 The
data further suggested that the methods used to describe stan-
dard outcome measures may require recalibration when
applied to new samples and to reflect practice changes over
time. Glance et al. compared the SMRs based on three differ-
ent scoring systems (APACHE II, SAPS II, and MPM II) to see
if each system could identify the same low-performance ICUs.7
There was fair to moderate agreement in identifying ICU
quality outliers among the three scoring systems. However, the
majority of ICUs in the study were classified as above average
and so the authors questioned the utility of SMRs if high-
performance ICUs cannot be reliably highlighted as role
model institutions and questioned the ability of the various
models to be used for benchmarking. More recent study from
Europe using validation sets each of more than 50,000 patients
and from multiple ICUs suggest that the original APACHE IV
shows good discrimination and accuracy but poor calibra-
tion,15 whereas a risk profile management using the new SAPS
II can be used to evaluate individual ICUs according to the
specific risk for patients to die compared with a reference
sample over the whole spectrum of hospital mortality.16
There are other limitations to the hospitalized SMR because
it was developed for all patients admitted to a hospital with a
primary diagnosis that is associated with 80% of all in-hospital
mortality based on national outcomes.17 SMR, therefore, may
better reflect hospital rather than ICU mortality. Consistent
with this Brinkman et al., in a cohort study that included
66,564 ICU patients from 55 Dutch ICUs, found that a model
based on clinical data (SAPS II) outperformed SMR.18 Hence
when comparing ICU performance it may be better to use a
clinical (ICU) rather than an administrative or hospital model,
because there is great variation across ICUs in disease severity
and hospitals (or ICUs) that admit more severely ill patients
are disadvantaged when compared with institutions that
admit less severely ill patients. Instead the SMR could be used
to indicate when performance may be poor rather than as a
quality indicator. Furthermore, although short-term mortality
may be sufficient to differentiate outcomes in general ICUs, it
is not an adequate measure to reflect recovery after many
neurologic conditions that require ICU admission or to assess
quality in a neurocritical care unit (NCCU) because long-term
functional recovery is more important in these patients.
Process Quality Indicators
Process indicators usually refer to how something is done
(or not). These indicators describe the interactions among
patients, their families, and health care providers and the
complex tasks performed by the ICU staff to achieve a speci-
fied outcome. There often may be a gap between process
quality indicator establishment and use in ICU care. For
example, Pronovost et al. developed and tested ICU perfor-
mance measures in a study that involved 13 ICUs, including 3
medical, 3 surgical, and 7 combined ICUs at nonteaching hos-
pitals.8 Three types of performance monitors—infection
control forms, daily rounding forms, and team leader forms—
were used to gather data. There was significant variation in
compliance with quality measures among ICUs and within
individual ICUs. For example, elevated head-of-bed position
to prevent VAP was observed in 67% (range 42%-99%) of the
ICUs. Other quality measures—appropriate blood transfusion
goals (9%-66%) appropriate peptic ulcer (71%-98%), or DVT
prophylaxis (48%-98%)—also had a wide range. Based on
compliance with the process quality indicators the authors
suggest that 1604 excess ICU and 198 excess hospital days
per year were incurred. Detailed data derived from patients
contained in the eICU Research Institute (eRI) data repository
who were discharged from the hospital during 200819 further
describe adherence with critical care best practice including

stress ulcer (90.8%) and venous thrombosis prophylaxis
(86.8%), beta-blocker use (68%-79%), low tidal volume ven-
tilation (28% for <6 mL/kg), and glycemic control (56% in
110-180 mg/dL) in current clinical practice in 271 ICUs in 188
institutions across diverse regions of the United States. The
following text examines individual process QIs and suggests
strategies toward their implementation.8
End-of-Life Care
Quality end-of-life care, also sometimes referred to as pallia-
tive care, is an important aspect of ICU treatment because one
in five U.S. deaths occurs in the ICU. Many of these deaths are
in patients in whom treatment goals are changed to comfort
measures. Interviews with family members after a patient dies
using a 31-item Quality of Dying and Death (QODD) ques-
tionnaire demonstrate that higher scores (i.e., quality) are
associated with death at home or in a location of the patient’s
choice, lower symptom burden, satisfaction with symptom
management, communication about treatment preferences,
and satisfaction with communication from health care pro-
viders.20 Other investigators have found that among ICU
patients high QODD scores are associated with documenta-
tion of a living will, withdrawal of tube feeding, a family
conference to discuss the patient’s end-of-life care, family at
the bedside at the time of death, and absence of cardiopulmo-
nary resuscitation (CPR).21 In particular, symptom control,
preparation for death, perceived preservation of dignity and
respect, and enough time to visit family members are associ-
ated with high QODD scores.22
Quality indicators for palliative care have been developed.
However, most of these quality indicators cover one specific
setting or target group, for example, cancer or patients who
are at home.23 Ghafoor et al have recently described a set of
indicators for palliative care and for support of relatives
before or after the patient’s death that appears applicable to
all settings where palliative care is provided to adult patients.
In addition, there are policy, standard orders, and quality-
assurance monitoring described for palliative sedation
therapy,24 but there are few validated outcome measures to
use in improving end-of-life care in the ICU. There are two
main models to integrate ICU and palliative care: (1) “con-
sultative,” which increases the involvement of palliative care
consultants for ICU patients and their families, particularly
those patients at greatest risk for poor outcome; and (2)
“integrative,” which embeds palliative care into daily practice
for all patients and families in the ICU. These models,
however, are not mutually exclusive.25
Ventilator-Associated Pneumonia
VAP is a common event in the ICU (8%-28% of mechanically
ventilated patients) and can increase ICU length of stay and
adversely affect outcome. For example, Yang et al. found that
traumatic brain injury (TBI) and stroke patients who develop
VAP have greater hospital expenses, longer length of stay in
the hospital and in the ICU, and a greater number of readmis-
sions than those without VAP.26 Consequently the incidence
and efforts to prevent VAP often are used as a benchmark of
quality of ICU care. However, the complexity and subjectivity
of VAP surveillance can limit its value to assess and compare
care for ICU patients, and so the use of VAP as a benchmark
Section I—Background 47
of ICU “quality” is debated for several reasons. First, VAP can
be difficult to diagnose because of underlying cardiopulmo-
nary disorders (e.g., pulmonary contusion, acute respiratory
distress syndrome, atelectasis) and the nonspecific radio-
graphic and clinical signs associated with VAP. One third of
patients diagnosed with VAP have no evidence of VAP at
autopsy and one fourth of mechanically ventilated patients
who are not diagnosed with VAP actually have a VAP at
autopsy. Second, clinical criteria described by the American
Thoracic Society (ATS) lack specificity because other condi-
tions may mimic VAP. For example, purulent secretions often
are present in intubated patients because of poor mucociliary
clearance. Furthermore colonization of oral secretions can
confuse respiratory culture data, and surveillance for VAP
requires dedicated and skilled staff.27,28 Third, reporting of
VAP is variable. For example, Eggimann et al. examined dif-
ferent denominators (1000 patient-days, patient-days at risk,
ventilator-days, and ventilator-days at risk) used to report VAP
and found that the method of reporting VAP could underes-
timate its incidence, particularly if reported as ventilator-days
or ventilator-days at risk.29 External reporting pressures also
may encourage stricter interpretation of subjective signs and
other surveillance initiatives that then may artificially lower
VAP rates.
In 2005 the ATS and the Infectious Diseases Society of
America (ATS/IDSA) published guidelines to manage hospital-
acquired pneumonia (HAP), VAP, and health care–associated
pneumonia (HCAP). There is consistent evidence that strate-
gies that affect the primary pathophysiologic mechanisms of
VAP reduce the incidence of the disease. Preventive measures
are best implemented when grouped into “bundles” to improve
overall efficacy and sustain compliance.30-35 Several different
protocols (or bundles) to reduce VAP are described. For
example, Lansford et  al. described use of head-of-bed eleva-
tion, twice-daily chlorhexidine oral cleans, once-daily ventila-
tor wean, and conversion from nasogastric to orogastric tube
in a trauma ICU; the VAP incidence decreased from 6.9
episodes per 1000 patient-days to 2.8 episodes per 1000
patient-days.36 Cocanour et  al. described a ventilator bundle
(head-of-bed elevation, stress gastritis prophylaxis, DVT pro-
phylaxis, and planned sedation weans).37 They found that the
VAP incidence was not altered with use of the ventilator
bundle alone. Instead a benefit was observed when compli-
ance with the ventilator bundle was audited daily with weekly
caregiver feedback. Other techniques that can be included
into VAP bundles include proper hand hygiene, high nurse-
to-patient ratio, avoidance of unnecessary transfer of ven­
tilated patients and manipulation of ventilator circuits,
incorporation of sedation control and weaning protocols into
patient care, use of noninvasive mechanical ventilation, main-
tenance of adequate pressure of endotracheal cuffs, selective
digestive tract decontamination, and use of oral rather than
nasal intubation, among others. These techniques need to be
supplemented with appropriately educated and trained staff
and a daily checklist. Initiation of a VAP bundle generally is
associated with a reduced VAP incidence (on average from
10 cases/1000 ventilator-days to 3 or 4 cases/1000 ventilator
days) and with cost savings, but it remains unclear if this
reduces overall ICU mortality.38-40
A reduction in the number of days of mechanical ven­
tilation is an important goal to reduce VAP. This may be
accomplished through use of protocol-driven weaning
48 Section I—Background
parameters by a multidisciplinary team, including nursing
staff and respiratory therapists.41-44 For example, Kollef et al.
in a randomized trial of different weaning methods in medical
and surgical ICUs observed that weaning protocols were asso-
ciated with significant reduction in duration of mechanical
ventilation. The least success with a weaning protocol occurred
in the ICU without an ICU director and critical care fellows.41
Use of a multidisciplinary weaning protocol also can help
reduce ventilation time in patients ventilated greater than 7
days.44 Inclusion of parameters such as assessment of sedation
status using a standardized form and standard clinical criteria
for weaning eligibility including hemodynamic status, oxy-
genation requirements, and minute ventilation can be used to
help clear patients for T-piece or pressure support weans.42
Today many hospitals report VAP rates at or close to zero,
so health care–regulating bodies propose that HAP should not
be reimbursed and potentially should be a “never event.”
However, VAP rates reported by hospital administrative
sources appear to be less accurate than physician-reported
rates and underestimate VAP incidence, that is, a never event
may not be possible.45 It also is difficult to unravel the relative
contribution of care improvements rather than surveillance
effects on the low VAP rates. Furthermore, whether a VAP rate
of zero is obtainable in the ICU is unclear because many
patients who are admitted already intubated have evidence of
early pneumonia or bacterial growth within 48 hours of
arrival. This suggests that early infection or colonization
occurred before admission.46,47 In the future, surveillance defi-
nitions that use more objective criteria may better mirror and
inform clinical practice.48 This includes objective alternatives
such as average duration of mechanical ventilation, length of
stay, mortality, and antibiotic prescribing to estimate disease
burden and guide quality improvement.
Pressure Ulcer
The National Quality Forum has identified a pressure ulcer as
a hospital-acquired condition (HAC) that is high cost and may
be prevented by using evidence-based guidelines. The CMS
intends to no longer reimburse acute care facilities for the cost
associated with facility-acquired (FA) ulcers. ICU patients are
at risk for pressure ulcer development because of comorbid
diseases, immobility, and use of sedation and analgesia that
reduce patient perception of high pressure points. In the Inter-
national Pressure Ulcer Prevalence Survey (IPUPS), an obser-
vational, cross-sectional cohort study conducted in the United
States during 2008 and 2009, FA pressure ulcer rates were
highest in ICUs (~10%), although the incidence varied accord-
ing to ICU type. In 2009, 3.3% of ICU patients developed
severe pressure ulcers (stage III, stage IV).49 There are several
risk assessment scales for pressure ulcers, although often they
are not specific to ICU populations.50 The APACHE score also
correlates with pressure ulcer risk. Development of hospital-
wide protocols including early skin assessments and transfers
to pressure-reducing mattresses51 can help reduce the inci-
dence of pressure ulcers or increase the length of time it takes
for an ulcer to develop.
Bloodstream Infections
Hospital acquired infections (HAIs) and in particular
bloodstream infections (BSIs), are common and adversely
affect patient outcome. The U.S. Centers for Disease
Control and Prevention (CDC) National Healthcare Safety
Network (NHSN; formerly the National Nosocomial Infec-
tions Sur­veillance [NNIS] group) collects national data on
catheter-associated bloodstream infections (CA-BSIs). How
the condition is defined may affect the incidence. For
example, CA-BSIs and catheter-related bloodstream infec-
tions (CR-BSIs) differ in the degree of proof needed to
show that the catheter “causes” the infection, so the CR-BSI
rate is significantly lower than the CA-BSI rate.52 In October
of 2008, Medicare enacted new regulations that reduced
reimbursement for HAIs and CA-BSIs and in 2009 the IDSA
published updated guidelines for the diagnosis and manage-
ment of all intravascular catheter-related infections.
Several factors such as central venous catheters (CVCs),
total parenteral nutrition (TPN) days, and hemodialysis cath-
eters are associated with nosocomial BSIs.53 Consequently a
variety of protocols and techniques now are in use to reduce
the incidence of infections associated with central lines,
that is, central line–associated bacteremia (CLAB) or central
line–associated bloodstream infections (CLABSIs). First, is
the use of chlorhexidine or silver sulfadiazine or antibiotic-
impregnated CVCs54-56 that at the very least appear to reduce
the risk of colonization and perhaps BSIs. These catheters can
be expensive and there is a concern that use of antibiotic-
impregnated catheters may be associated with an increased
risk of antimicrobial resistance. In addition, adherence to
standardization of the processes of care associated with CVC
placement rather than use of coated CVCs (i.e., placement,
dressing, and maintenance/removal protocols) appears to be
a more important factor in reduction of CR-BSIs.57 Second is
use of chlorhexidine-soaked dressings. For example, the Bio-
patch dressing surrounds the hub of the CVC and is saturated
with chlorhexidine. Data from the manufacturer suggest that
its use is associated with a 60% decrease in BSIs and 44%
decrease in local site infections (see www.jnjgateway.com).
Third, a reduction in the duration of central venous access and
peripherally inserted central catheter (PICC) rather than CVC
use may help reduce the incidence of CR-BSI particularly in
patients who are in the ICU a long time.58 Fourth, resident
oversight and credentialing policy for CVC placement can
help reduce CLABSIs.59 Finally and perhaps most important
is a bundle approach with physician and patient specific steps
and use of a checklist. Compliance with these protocols
(bundles) has improved the safety of hospitalized patients,
and better compliance appears to be associated with a reduc-
tion in infection and CLABSIs.60 Bundle adherence and
CLABSI rates may depend on the institution; where compli-
ance is low the sites often lack a functional team, forcing
functions, and feedback systems.61
Glycemic Control (See Also Chapter 14)
Both hyperglycemia and hypoglycemia can adversely affect
ICU outcome.62,63 Furthermore increased insulin administra-
tion and poor glycemic control in nondiabetic patients
may aggravate outcome, hence insulin and glycemic control
protocols have developed. The most successful protocols
include bedside glucose monitoring, nursing-driven proto-
cols, and computerized decision-making algorithms.64-66
Although there are many clinical and fiscal benefits to
glucose control67-71 there are barriers to “tight” glucose

control in the ICU that include lack of a defined target
glucose range, health care provider fear of hypoglycemia,
and changes to subcutaneous insulin, par­ticularly when
patients’ nutritional and clinical status is in flux.67-72
Hyperglycemia is common among ICU patients, and 90%
of critically ill patients develop blood glucose concentrations
greater than 110 mg/dL (6.1 mmol/L). Van den Bergh’s land-
mark study published in 2001 supported the use of intensive
insulin therapy (IIT) to target normoglycemia (a blood
glucose concentration of 80-110 mg/dL [4.4-6.1 mmol/L]) in
the critically ill and postoperative patient.70 However, other
investigators have not been able to replicate these findings, and
a meta-analysis that included 29 randomized trials with 8432
patients showed that tight glucose control is not associated
with reduced hospital mortality but is associated with an
increased risk of hypoglycemia in critically ill adult patients.73
A paradigm shift in glucose control in the ICU occurred after
the publication in 2009 of the normoglycemia in intensive
care evaluation survival using glucose algorithm regulation
(NICE-SUGAR).74 This trial demonstrated increased mortal-
ity and incidence of hypoglycemia in patients managed with
IIT. Current recommendations are more moderate and target
a blood glucose (BG) concentration between 144 mg/dL and
180 mg/dL (8-10 mmol/L). These “moderate” insulin proto-
cols are in use in many ICUs and appear to avoid hyper­
glycemia and low glucose variability while rarely inducing
hypoglycemia.75,76 These various studies were conducted
largely in non-NCCU patients, and laboratory studies that
indicate the deleterious effects of severe hyperglycemia in the
context of anaerobic conditions in the brain and clinical
microdialysis studies in severe brain injury that show intersti-
tial hypoglycopenia in NCCU patients with tight glucose
control77 underscore the still unsettled question on optimal
glucose management in NCCU patients.
Most hospitals have established glycemic control programs.
Based on reports from the 2009 Remote Automated Labora-
tory System (RALS) that provides trends in glycemic control
between 2006 and 2009 from 576 U.S. hospitals (including
175 million BG results, 25% from the ICU) for many institu-
tions, glycemic control has improved: the mean range of BG
results in ICU patients in 2009 was 121.1 to 217 mg/dL.78
Earlier RALS reports show that hospital hyperglycemia
(>180 mg/dL) prevalence was 46.0% in the ICU and 31.7%
outside the ICU. Hospital hypoglycemia (<70 mg/dL) preva-
lence was 10.1% in the ICU and 3.5% for non-ICU readings.79
What is unclear is whether quality indicators for diabetes care
are associated with patient outcomes. For example, Sidoren-
kov et al. conducted a systematic literature review of 24 studies
(but not limited to ICU patients) to test this relationship.80
There was insufficient evidence that quality indicators pre-
dicted better patient outcomes. Consequently insulin infusion
protocols (IIPs) to achieve desired BG targets must be indi-
vidualized and validated to the ICU and institution where they
are used. The implementation of IIPs is not a simple process
but requires a carefully planned, inclusive, and continual team
effort. Continual assessment of protocol errors, adverse events,
staff satisfaction, and outcomes is vital to success, the goal of
which is effective glucose control while avoiding hyperglyce-
mia, hypoglycemia, and glucose variability, all of which are
detrimental.81,82 Accurate and efficient glucose monitoring
devices, including automated point-of-care (POC) BG data
management software therefore are essential. Each of these
Section I—Background 49
devices is subject to federal and state regulations, and accredi-
tation standards developed by the College of American
Pathologists and The Joint Commission. Future research will
need to address whether computerized decision support
systems and newer technologies that allow accurate and con-
tinuous or near-continuous BG measurements improves the
quality of glycemic control.83
Structural Quality Indicators
Structure in the ICU refers to the type and size of the ICU,
ICU design, availability of technology, and staffing among
others (see Chapter 3 for a review on ICU design). Interven-
tions that influence structure usually take longer to implement
and are more expensive than changes to processes of care. The
following text discusses organization and management in
critical care by reviewing ICU staffing, transport, handovers,
and telemedicine.
Physician Staffing Models
Models of physician staffing in the ICU include low intensity,
or open, and high intensity, or closed. In low-intensity systems
there is either no or only partial intensivist direction. A
patient’s primary attending physician controls the patient’s
admission to and discharge from the ICU. Intensivists may
serve as part of an acute resuscitation team or be consulted
for a particular management question. In a closed unit, a
patient’s attending physician yields control of patient care to
a critical care team. The team model that characterizes the
closed unit empowers medical and nursing administrators to
effect standard protocols of care. In the United States, the
model of ICU care historically has been a low-intensity
system.84,85 For example, in 2000 Angus et al. published a
survey of 393 ICUs; 23% were managed by a full-time inten-
sivist; an intensivist consultant was available in 14%; and 29%
had no access to an intensivist at all.84 High-intensity units
usually are associated with teaching hospitals, hospitals with
large patient volume, surgical ICUs, or trauma ICUs and are
more prevalent.85
Several studies describe a benefit to closed rather than open
systems.86-98 For example, Pronovost et al. reviewed 27 ran-
domized or observational trials that described the effect of
physician staffing on mortality and length of stay in medical,
surgical, mixed medical and surgical, or pediatric ICUs. In
nearly every study, decreased mortality was associated with a
closed ICU and in half decreased ICU length of stay was asso-
ciated with a closed ICU system.1 The benefits of a high-
intensity model may be greatest when patients are admitted at
night or during the weekend,88 patients admitted more than
48 hours, or among elderly patients with more comorbid dis-
eases.92 The benefits of closed systems on patient outcome,
ventilator-days, length of stay, morbidity, cost, and discharge
disposition among other measures also are seen in various
subspecialty ICUs including trauma, surgical, and neurocriti-
cal care89-92,96,97 in patients with acute lung injury or who
undergo specific high-risk procedures such as abdominal
aortic surgery or esophageal resection.93-95 For example,
Varelas et al. compared patient outcome and disposition using
historical controls in a university hospital NCCU that changed
from an open ICU to one with a dedicated neurointensivist.89
Hospital and ICU length of stay were reduced and fewer
50 Section I—Background
patients were discharged to nursing homes when the NCCU
was led by a neurointensivist. Implementing a high-intensity
staffing model including continuous (24 hour) onsite pres-
ence of a critical care specialist also is associated with improved
processes of care and staff satisfaction.98
Evidence-Based Protocols
The mechanisms that underlie improved outcomes in high-
intensity ICUs are unclear but may be associated with
increased use of evidence-based quality indicators and
increased use of evidence-based guidelines or protocols.99
These protocols include DVT prophylaxis, stress ulcer pro-
phylaxis, sedation interruption, spontaneous breathing trials,
and glycemic control, among others. Even just having an
available intensivist can contribute to evidence-based proto-
col enforcement.99 Other processes that may be associated
with better outcomes and are improved in a closed system
include weaning by respiratory therapist protocol, daily
rounds with pharmacists, nurse-to-patient ratio of greater
than 1 : 2, and ICU mortality and morbidity reviews.100 A
mandatory bedside checklist can be a simple, cost-effective
method to prevent errors of omission and improve imple-
mentation of ICU best practices.101
Introduction of evidence-based protocols requires an audit
of current practice, expert-led education sessions, and dis-
semination of algorithms (because adoption of protocols
with low baseline adherence changes clinical practice little).102
Evidence-based guidelines create a base to allow change in
health care practitioner behavior. However, they do not in
and of themselves effect change or outcome unless their use
is accompanied by conversion of a core of key goals in the
guideline recommendations to quality indicators, and with
regular feedback on performance, for example, weekly.103,104
In addition, the protocols must be updated to incorporate
new guidelines and evidence if these new guidelines and evi-
dence meet rigorous critical appraisal. Most appraisal systems
consider randomized controlled trials (RCTs) the gold stan-
dard given their ability to limit bias. However, many RCTs in
critical care have failed,105,106 or a number of trials have had
results that are initially positive and then subsequent trials
are less positive.107 This has led to uncertainty about the sci-
entific approach to ICU research and that guideline develop-
ment may need reevaluation. In part this may be explained by
the heterogeneous nature of ICU patients, who all have dif-
fering physiologic requirements over time. Consequently,
ICU clinicians routinely adjust treatment dose and care based
on multiple factors that can be difficult to simulate in a clini-
cal trial.108 Procedure or disease-specific registries, observa-
tional cohorts, or studies that aim to understand variation
in performance and delivery of care and how to reduce
individual and organizational underperformance may help
answer some of the limitations observed with RCTs in critical
care.105,107 The past decade also has seen the bringing together
of positive studies and interventions into bundles of
care (goal-directed therapy, protocolized care). While these
bundles appear to be associated with improved outcomes,
there needs to be a balance between individualized versus
protocolized care.109
There are several concerns with protocols in ICU care. First,
the use of clinical protocols may negatively affect medical
education by eliminating clinical decision making from
trainees. However, a study by Prasad et al. suggests this is not
so.110 Second, the ICU environment is complex and highly
dynamic and often requires that clinicians adjust and deviate
from standard protocol when confronted with nonstandard
circumstances. These deviations may be errors or innovations.
Observations in acute trauma care indicate that protocol devi-
ations by more experienced physicians are a combination of
errors and innovations, whereas the deviations by less experi-
enced practitioners are mostly errors.111 Finally, bundles of
care often are implemented under pressure from hospital
administrators or regulatory bodies with little regard for the
context in which the evidence was generated and without
understanding the difference between efficacy and effective-
ness;112 that is, the interventions could be implemented to a
larger group of ICU patients that differ from those in whom
the evidence was gathered. The risk then is that the bundle
tools or protocols can be used for quality control, performance
evaluation, and legal and regulatory purposes out of context
and without robust validation. This will put patients and pro-
fessionals at risk of being harmed by the protocols or bundles
of care.
Intensive Care Unit Physician
Staffing Plan
In the United States there still is a relative paucity of ICUs
that are closed and have a board-certified intensivist available
at all times.84-95 These staffing issues are changing in part
because of the Leapfrog initiative or addressed through the
development of the eICU and telemedicine (see Chapters 42
through 44). The Leapfrog organization recommends all
ICUs have a board-certified intensivist available onsite during
daytime hours and be immediately reachable through a pager
during nighttime hours with a critical care physician or “phy-
sician extender” available within 5 minutes of the ICU at
night. These rules formulate the Leapfrog’s 2002 ICU Phy­
sician (IPS) Standard (www.leapfroggroup.org).113 There
are, however, several obstacles to the Leapfrog standard,
including (1) primary physician resistance to yield control
over patient care, (2) perceived loss of continuity of patient
care, (3) shortage of critical care physicians to staff ICUs,
(4) limited fiscal incentive for critical care–certified physi-
cians, and (5) poor fiscal incentive for hospitals with small
ICUs to hire full-time ICU physicians. These factors are
greater in smaller nonacademic hospitals where there may
not even be an ICU director.100,114 To overcome some of these
obstacles, many ICU directors recommend financial incentive
from either a hospital, the government, or a third party to
ensure compliance with the Leapfrog initiative.114 Consistent
with this a survey of ICU directors as well as chief medical
officers suggest that adoption of the IPS standard was facili-
tated by intensivist salary support.115
What is the cost of achieving the Leapfrog IPS standard?
Hypothetical financial models have been applied to 6-, 12-,
and 18-bed ICUs. The variables in the model include annual
admissions, intensivist salary, revenue from intensivist billing,
and salary of critical care physician extenders among others.
Savings, because of improved patient outcomes and shorter
ICU length of stays, range between $510,000 and $3.3
million.116 The greatest savings occur in larger ICUs and are
associated with ICU beds, length of stay, and cost of ICU bed

per day. However, under worst-case scenario conditions, costs
may range from $890,000 to $1.3 million. Small ICUs may be
disproportionately more vulnerable to worst-case conditions
and experience net loss more frequently under the IPS
standard.114
The final obstacle is having enough staff. The Committee
on Manpower for Pulmonary and Critical Care Societies
(COMPACCS), commissioned by the American College of
Chest Physicians, the ATS, and the Society of Critical Care
Medicine, assessed the future supply and demand for intensiv-
ists until 2030. COMPACCS found that the supply of intensiv-
ists likely will be static until 2030, whereas the demand for
critical care services, particularly by older patients, will increase
as the baby boomer generation ages.117,118 Other critical care
workforce analyses estimate a 35% shortage of intensivists
by 2020, particularly in surgical critical care in the United
States.119 The likely intensivist workforce shortage calls for
creative staffing solutions. Among these are (1) use of “novel”
critical care personnel including critical care–certified physi-
cian assistants and nurse practitioners and physicians from
emergency medicine and anesthesiology who receive ICU
training; (2) regionalization; and (3) telemedicine. In many
countries around the world anesthesiologists provide critical
care but in the United States in 2006, less than 4% of board-
certified anesthesiologists were also certified in critical care.
There are two major barriers to dual certification: a paucity of
programs that offer such certification, and surgeons and anes-
thesiologists who practice critical care tend to take a pay cut.117
An alternative may be increased use of acute care nurse prac-
titioners (ACNPs). In a “semiclosed” surgical intensive care
unit (SICU) the presence of ACNPs increases compliance with
clinical practice guidelines.120
In 2004 the American Association of Critical-Care Nurses,
the American College of Chest Physicians, the ATS, and the
Society of Critical Care Medicine, convened the Critical Care
Workforce Partnership and published a white paper titled
“FOCCUS—Framing Options for Critical Care in the United
States.” The conference highlighted that although trauma
patients are classified by need into level of trauma service,
there is no tiered critical care delivery system. Such a regional-
ized and standardized triage system may appropriately direct
intensivist workforce demand to higher-level, higher-acuity
centers.117
Clinical Handovers
Staffing requirements and transport of ICU patients for diag-
nostic or therapeutic procedures mean clinical handover is
a necessary process in ICU care. There is a well-described
association between intrahospital transport of critically ill
patients and complications including exacerbation of pulmo-
nary function and intracranial physiology.121-124 The rate of
adverse events during patient transport, both intra- and
interhospital, which by definition include transitions of care,
can be reduced use of a specialized transport team and struc-
tured process.125,126 In addition, deficiencies in the handover
process can introduce error and discontinuities in care and
compromise patient care, whereas distractions during hando-
ver can adversely affect information transfer.127 By contrast, a
formal, structured handover process, including during transi-
tions of care can reduce medical errors and improve team-
work.128 However, many hospitals lack safe and effective
Section I—Background 51
clinical handover practices (i.e., mechanisms to transfer
information, responsibility, and authority) and many new
doctors feel unprepared for the handover process.129 Studies
in ICU care also show that medical members of the
ICU team provide data that can differ from the medical
record when asked to recall clinical information discussed
at handover.130
Efforts to improve handover safety and effectiveness are
ongoing. Multidisciplinary teams administer critical care, and
it is conceivable that behavioral methods to enhance team
performance may affect the quality of care. This can be
achieved in development of safe handover practices that
requires (1) awareness of handover problems and opportuni-
ties; (2) identification of solutions; (3) adaptation of local best
practices, for example, structured handovers that include both
written and verbal components to improve information
exchange; (4) multidisciplinary education about the clinical
handover process to encourage teamwork and a common lan-
guage among different services in the hospital; and (5) insti-
tutionalizing practice changes.131 Three factors—information
transfer, shared understanding, and working atmosphere—
predict handover quality.132 However, handover is not only a
skills-based task; there also are complex interactions between
individual and systems factors. A full understanding of team
principles including leadership, psychological safety, transac-
tive memory, and accountability all are important to the han-
dover process (and to critical care) and can be used to improve
patient care.133 These same team principles also can be applied
to rapid response systems, in which there are few benchmark-
ing tools to allow comparison for quality assurance, and car-
diopulmonary resuscitation (CPR).134,135
Telemedicine
Telemedicine is described in Chapters 42 and 44 but is briefly
discussed here to address physician staffing. Since the concept
was introduced in 1977, several publications have described
its use. In particular, the presence of an off-site intensivist who
monitors patient care remotely using cameras and data trans-
mission equipment and then communicates his or her recom-
mendations to the onsite personnel either during formal
rounds or during informal communication with the onsite
attending staff is associated with decreased ICU and hospital
mortality and decreased ICU length of stay and costs.136 These
same benefits may be seen when a single intensivist team
monitors multiple ICU sites. Together these early observa-
tional data suggest that telemedicine could be used to stem the
future demand for intensivists.137
In 2003 VISICU, Inc was founded to provide ICU telemedi-
cine. These sites are typically multihospital systems, and
several are associated with large academic institutions. In
smaller networks (fewer than 70 beds), a remote team consists
of one critical care nurse and one critical care physician. In
larger networks (more than 120 ICU beds), two critical care
physicians and three critical care nurses remotely monitor
patient care. The hours of remote monitoring vary by site and
need (e.g., some sites only use VISICU at nights and week-
ends). There are several challenges to implement telemedicine
in the ICU, including (1) cost, because ICU telemedicine does
not command separate billing and so health systems assume
the costs of salary and infrastructure; (2) restructuring of
rounds to multidisciplinary rounds when the remote team can
52 Section I—Background
inform the onsite team; (3) staff training to use the remote
system; (4) maintenance costs for software or hardware down-
time, which may have real clinical implications if an ICU is
entirely dependent on remote monitoring; (5) privacy of
patient information and confidentiality; and (6) different
management policies and protocols between the “remote”
intensivist and onsite personnel.138
Despite these challenges ICU telemedicine may help address
the shortage of board-certified critical care physicians to staff
ICUs. In addition, new standards for supervision and duty
hours for residency programs released by the Accreditation
Council for Graduate Medical Education (ACGME) were
introduced in July 2011. In ICUs that depend on residents for
portions of patient care, telemedicine, among other options
such as increased use of nurse practitioners and physician
assistants or partnerships with hospitalists, may offer a solu-
tion to the change in workflow. However, further research is
needed to examine how the new duty hours influence patient
safety, continuity of care, resident learning, and ICU staff-
ing.139 The University Health System Consortium ICU Tele-
medicine Task Force, an independent organization with no
financial interest in VISICU, has issued recommendations for
institutions that are interested in a remote ICU program.
Among these recommendations are warnings that a telemedi-
cine program does not obviate the need for onsite full or
part-time intensivists and that such a program does not serve
hospitals with small ICU capacity.140 These cautions serve as
important reminders that the challenge of appropriate physi-
cian staffing to maintain quality in ICU care still need to be
resolved.
Benchmarking and Quality
Improvement in the Future
The Practice of Neurocritical Care
To improve quality (of therapy), it is important to know, eval-
uate, and make transparent daily ICU processes. Much of the
literature and research associated with quality assurance and
benchmarking is, in general, medical or surgical critical care
and there is little that is specific to neurocritical care. Can
these results be extrapolated to the NCCU or are there differ-
ences that suggest NCCU specific research is needed? Large
databases that describe the range of critical care practice
matched to acuity-adjusted outcomes have only recently
become available. For example, Lilly et al. (2011) provided
benchmark data from 243,553 adult admissions, including
7692 patients with neurologic disorders from 271 ICUs all
supported with ICU telemedicine monitoring.19 There were
important patient demographic (age) and clinical differences
(APACHE IV) between NCCUs and other ICUs. The average
length of stay in the ICU was just over 3 days. However, neu-
roscience and trauma patients had longer stays, and NCCU
patients more frequently required rehabilitation or chronic
care than other ICU types. Unadjusted mortality rates were
less for nontrauma surgical or mixed ICUs than medical or
neuroscience ICUs (hospital 10.1%, ICU 6.4%). The APACHE
IV model overpredicted hospital and ICU mortality for all
ICU types. These data suggest that there may be enough dif-
ferences between NCCUs and other ICUs that specific NCCU
safety, and quality assurance may better inform select practices
in neurocritical care.
Which Indicators Can Drive
Safety and Quality in the
Intensive Care Unit?
Quality assurance in health care continues to mature and
benchmarks are used more frequently to compare perfor-
mance between institutions. A critical assessment of the diver-
sity in critical care practice and organization and their
association with outcomes can help inform the development
of evidence-based guidelines. This is important given the
increasing use of critical care services and the significant cost
of ICU care. However, the robustness of this comparison
depends on which quality indicators or benchmarks are used.
Other important considerations in benchmark use include
definition, time frame, case-mix adjustment, and the context
in which the original benchmark was described and then is
used. In 2009 the ESICM formed a Safety and Quality Task
Force with the intent of improving the safety and quality of
ICU care. This task force developed a directive for change
signed by 57 national and international critical care organiza-
tions in the Declaration of Vienna141 and next identified a set
of indicators that could be used to measure the quality of
ICU care in a prospective fashion. To do this the Safety and
Quality Task Force used 18 nominated experts from 9 coun-
tries and a modified Delphi process. A final set of 9 indicators,
each of which had a greater than 90% agreement rate, from
111 potential indicators, were agreed upon (Table 6.1).9
Similar efforts have been made in stroke research to ensure
common standards to define quality indicators, collect infor-
mation, and promote valid comparisons of acute stroke care
between institutions.142,143
Does Benchmarking
Influence Outcome?
An ICU benchmark may be defined as a quantitative, stan-
dardized measurement to compare the performance of differ-
ent ICUs. Industries other than health care have used
Table 6.1  Prospectively Defined Indicators
to Improve the Safety and Quality of Care
for Critically Ill Patients*
Intensive care unit (ICU) fulfils national requirements to
provide intensive care
24-hour availability of a consultant level intensivist
Adverse event reporting system
Presence of routine multidisciplinary clinical ward rounds
Standardized handover procedure for discharging patients
Reporting and analysis of standardized mortality ratio (SMR)
ICU readmission rate within 48 hours of ICU discharge
The rate of central venous catheter–related bloodstream
infection
The rate of unplanned endotracheal extubations
Modified from Rhodes A, Moreno RP, Azoulay E, et al. Prospectively defined
indicators to improve the safety and quality of care for critically ill patients:
a report from the Task Force on Safety and Quality of the European Society
of Intensive Care Medicine (ESICM). Intensive Care Med 2012; Jan 26. Epub
ahead of print.
*Final set of prospectively defined indicators agreed upon by the Task Force
on Safety and Quality of the European Society of Intensive Care Medicine.

benchmarks for many years in large part to compare their
performance against competitors and so gain greater market
share or reduce costs. The primary goal is to increase efficiency
and reduce variability that in turn should increase quality.
Benchmarking has become more frequent in health care,
driven in large part by a need to deliver care at lower costs
(i.e., increased efficiency) and to standardize care (i.e., vari-
ability reduction). When the standards, including clinical
guidelines, order sets, and care bundles are evidence based it
is postulated that the reduced variability will improve the
quality of care and so enhance outcome. However, it is difficult
to demonstrate that benchmarking is associated with improved
outcomes.144-147 For example, Hernandez et al. linked Medi-
care claims with the Get With the Guidelines-Heart Failure
registry of the American Heart Association for heart failure
patients greater than 65 years old and admitted to hospital.146
The agreement between different methods to rank hospital-
based quality of care and 30-day mortality or readmission was
poor. Similarly Sciacovelli et al., who collected data on quality
indicators over a 3-year period from a clinical laboratory,
observed that processes under laboratory control but not pro-
cesses that required cooperation between the patient care pro-
viders and the laboratory improved (i.e., more effective
laboratory function did not necessarily mean better patient
care).147 On the other hand, Woodhouse et al. in a benchmark-
ing program in eight ICUs in the Netherlands observed a 3%
decline in the SMR each year over an 8-year period.144 There
were three main components to their benchmarking: (1)
quality (e.g., SMR), (2) availability (e.g., was a bed available
for a patient who required ICU care), and (3) efficiency (e.g.,
length of stay and ventilator duration). They expected bench-
marking would lead to a convergence of “best practices”
between hospitals. However, they found that over time all the
ICUs improved in such a way that the gap between the best
and worst performing ICUs remained.
It has become increasingly common to release information
about the performance of hospitals, health care providers, and
health care organizations into the public domain. Ideally
public reporting should be focused on the program rather
than individual health care provider because modern medi-
cine including critical care or neurosurgery requires a
team-based approach. However, it is unclear whether public
reporting affects the performance of health care professionals,
improves quality, or alters behavior of consumers. A Cochrane
analysis148 found a small effect for coronary bypass surgery
and low-complication outliers for lumbar diskectomy. These
effects disappeared within 2 months after public release.
However, overall the analysis concluded that public release of
performance data does not change consumer behavior or
improve care. Similarly, Ryan and Blustein, who examined the
Massachusetts Medicaid’s (MassHealth) hospital-based pay-
for-performance program, did not find improved quality
despite substantial financial incentives.149 Furthermore, rather
than outcome, it appears that improvements in process and
structure are most likely to increase a patient’s overall evalu-
ation of the quality of care.150 Physicians also need to be
mindful of unintended negative consequences associated with
performance measurement reporting. For example, many
payers and regulatory bodies measure hospital quality of care
by the rate of HAP. Public reporting of these rates may then
drive increased use of antibiotics and so lead to bacterial
resistance.151
Section I—Background 53
How to Ensure a Successful
Quality Assurance or
Benchmarking Initiative
Several prerequisites are necessary to ensure a successful
quality assurance or benchmarking initiative: (1) a willingness
to compare ICUs, (2) intent to act on the results with quality
improvement programs, (3) accurate clinical documentation
with physician buy-in,152,153 (4) mature data collection tools,
and (5) a synergistic effort between the hospital administra-
tors and the medical staff with assembly of a multidisciplinary
team.154 Within each ICU or department the foundation
needed to measure and enhance quality is good team spirit,
communication, and cooperation.155 Program implementa-
tion requires (1) choosing the right tool to measure perfor-
mance and outcome and identification of clinical processes
that need to be changed; (2) standard measurement defini-
tions, a statement of desired goals, and a clear timeline; (3)
teaching sessions and daily goal sheets or checklists; (4) con-
sistent measurement and feedback practices of both process
compliance and outcome data156; and (5) use of human factor
analysis that can provide insights into how interactions
between organizations, tasks, and the individual worker influ-
ence behavior and affect systems reliability because unreliable
delivery of best practice is a contributor to medical error. Data
that are collected particularly for benchmarking should be (1)
incorporated into the daily routine with minimal added work;
(2) clinically relevant; (3) comparable across various ICUs
with risk adjustment; (4) valid (i.e., a measurement change
results in a change in the outcome of interest); (5) robust; (6)
standardized; and (7) anonymous.144 The collected data can
then be used to guide quality improvement. This may be best
achieved through small tests of change in an incremental
process using repeat evaluation and refinement such as a Plan-
Do-Study-Act (PDSA) cycle. Pronovost et al. have suggested a
five-phase translational (T) framework to develop programs
to reduce preventable harm or enhance quality157 (Table 6.2).
It should be remembered that sustained high performance is
not a permanent state that an institution or ICU reaches but
rather a dynamic balance and one that is constantly moving
forward.158
Table 6.2  Five-Phase Translational (T)
Framework to Develop Programs to Reduce
Preventable Harm or Enhance Quality
Phase T0: discover opportunities and approaches to prevent
adverse health care events
Phase T1: use T0 discoveries to develop and test interventions
on a small scale
Phase T2: broaden and strengthen the evidence base for
promising interventions to develop evidence-based
guidelines
Phase T3: translate guidelines into clinical practice
Phase T4: implement and evaluate T3 work on a national and
international scale
Modified from Pronovost PJ, Cardo DM, Goeschel CA, et al. A research
framework for reducing preventable patient harm. Clin Infect Dis
2011;52(4):507–13.
54 Section I—Background
Conclusion
Measurement of hospitals’ performance has become more
commonplace in part to inform patient choices and payer
reimbursement decisions including pay-for-performance ini-
tiatives and potential nonpayments for readmissions and
adverse events. Consequently, many institutions have replaced
revenue growth models of external market and service expan-
sion with cost management and enhancement of institutional
capabilities to ensure fiscal stability. The significant cost of
critical care services has prompted the medical community
and regulatory agencies to examine the range of organization
and practice in critical care and how this is associated with
outcomes. This provides opportunities to improve delivery of
care and enhance quality of care particularly when coupled
with evidence-based guidelines. Much of the focus has been
on preventable complications (e.g., medical errors, health
care–associated infections, and venous thromboembolism)
that are common in hospitalized and ICU patients in general.
A large body of scientific evidence exists from clinical trials to
provide evidence-based guidelines to manage many of these
conditions. However, whether these recommendations apply
to all patients in the NCCU or how best to enhance outcome
through quality assurance initiatives in the NCCU still requires
many questions to be answered.
References
1. Pronovost PJ, Angus DC, Dorman TM, et al. Physician staffing patterns and
clinical outcomes in critically ill patients: a systematic review. JAMA 2002;
288(17):2151–62.
2. Pronovost PJ, Miller MR, Dorman TM, et al. Developing and implementing
measures of quality of care in the intensive care unit. Curr Opin Crit Care
2001;7(4):297–303.
3. Mallon WJ. Earnest Amory Codman: the end result of a life in medicine.
Philadelphia: Saunders; 2000.
4. Deming WE. Out of the crisis. Cambridge, MA: MIT Press; 1986.
5. Donabedian A. The quality of care: how can it be assessed? JAMA 1988;
260(12):1743–8.
6. Sirio CA, Shepardson LB, Rotondi AJ, et al. Community-wide assessment of
intensive care outcomes using a physiologically based prognostic measure:
implications for critical care delivery from Cleveland health quality choice.
Chest 1999;115(3):793–801.
7. Glance LG, Osler TM, Dick AP. Rating the quality of intensive care units: is
it a function of the intensive care unit scoring system? Crit Care Med 2002;
30(9):1976–82.
8. Pronovost PJ, Berenholtz SM, Ngo K, et al. Developing and pilot testing
quality indicators in the intensive care unit. J Crit Care 2003;18(3):
145–55.
9. Rhodes A, Moreno RP, Azoulay E, et al. Prospectively defined indicators to
improve the safety and quality of care for critically ill patients: a report from
the Task Force on Safety and Quality of the European Society of Intensive
Care Medicine (ESICM). Intensive Care Med 2012; Jan 26. Epub ahead of
print.
10. Lilford R, Pronovost PJ. Using hospital mortality rates to judge hospital
performance: a bad idea that just won’t go away. BMJ 2010;340:955–7.
11. Huang DT, Clermont G, Kong L, et al. Intensive care unit safety culture and
outcomes: a US multicenter study. Int J Qual Health Care 2010;22(3):
151–61.
12. Sakr Y, Marques J, Mortsch S, et al. Is the SAPS II score valid in surgical
intensive care unit patients? J Eval Clin Pract 2010;Sep 22. Epub ahead of
print.
13. Poole D, Rossi C, Anghileri A, et al. External validation of the Simplified
Acute Physiology Score (SAPS) 3 in a cohort of 28,357 patients from 147
Italian intensive care units. Intensive Care Med 2009;35(11):1916–24.
14. Niskanen M, Reinikainen M, Pettilä V. Case-mix-adjusted length of stay and
mortality in 23 Finnish ICUs. Intensive Care Med 2009;35(6):1060–7.
15. Brinkman S, Bakhshi-Raiez F, Abu-Hanna A, et al. External validation of
Acute Physiology and Chronic Health Evaluation IV in Dutch intensive care
units and comparison with Acute Physiology and Chronic Health Evaluation
II and Simplified Acute Physiology Score II. J Crit Care 2011;26(1):
105.e11–18.
16. Moreno RP, Bauer P, Metnitz PG. Characterizing performance profiles of
ICUs. Curr Opin Crit Care 2010;16(5):477–81.
17. Jarman B, Aylin P: Death rates in England and Wales and the United States:
variation with age, sex, and race. BMJ 2004;329:1367.
18. Brinkman S, Abu-Hanna A, van der Veen A, et al. A comparison of the
performance of a model based on administrative data and a model based on
clinical data: effect of severity of illness on standardized mortality ratios of
intensive care units. Crit Care Med 2012;40(2):373–8.
19. Lilly CM, Zuckerman IH, Badawi O, et al. Benchmark data from more than
240,000 adults that reflect the current practice of critical care in the United
States. Chest 2011;140(5):1232–42.
20. Curtis JR, Patrick DL, Engelberg RA, et al. A measure of the quality of dying
and death: initial validation using after-death interviews with family
members. J Pain Symptom Manage 2002;24(1):17–31.
21. Glavan BJ, Engelberg RA, Downey L, et al. Using the medical record to
evaluate the quality of end-of-life care in the intensive care unit. Crit Care
Med 2008;36(4):1138–46.
22. Mularski RA, Heine CE, Osborne ML, et al. Quality of dying in the ICU:
ratings by family members. Chest 2005;128(1):280–7.
23. Claessen SJ, Francke AL, Belarbi HE, et al. A new set of quality indicators for
palliative care: process and results of the development trajectory. J Pain
Symptom Manage. 2011;42(2):169–82.
24. Ghafoor VL, Silus LS. Developing policy, standard orders, and quality-
assurance monitoring for palliative sedation therapy. Am J Health Syst
Pharm 2011;68(6):523–7.
25. Nelson JE, Bassett R, Boss RD, et al. Improve Palliative Care in the Intensive
Care Unit Project. Models for structuring a clinical initiative to enhance
palliative care in the intensive care unit: a report from the IPAL-ICU project
(Improving Palliative Care in the ICU). Crit Care Med 2010;38(9):1765–72.
26. Yang CC, Shih NC, Chang WC, et al. Long-term medical utilization
following ventilator-associated pneumonia in acute stroke and traumatic
brain injury patients: a case-control study. BMC Health Serv Res 2011;
11:289.
27. Klompas MM, Platt RM. Ventilator-associated pneumonia: the wrong quality
measure for benchmarking. Ann Intern Med 2007;147(11):803–5.
28. Uckay I, Ahmed QA, Sax H, et al. Ventilator-associated pneumonia as a
quality indicator for patient safety? Clin Infect Dis 2008;46(4):557–63.
29. Eggimann P, Hugonnet S, Sax H, et al. Ventilator-associated pneumonia:
caveats for benchmarking. Intensive Care Med 2003;29(11):2086–9.
30. Mangino JE, Peyrani P, Ford KD, et al. Development and implementation of
a performance improvement project in adult intensive care units: overview
of the Improving Medicine Through Pathway Assessment of Critical Therapy
in Hospital-Acquired Pneumonia (IMPACT-HAP) study. Crit Care 2011;
15(1):R38.
31. Ramirez P, Bassi GL, Torres A. Measures to prevent nosocomial infections
during mechanical ventilation. Curr Opin Crit Care 2012;18(1):86–92.
32. Rello J, Lode H, Cornaglia G, et al. VAP care bundle contributors:
a European care bundle for prevention of ventilator-associated pneumonia.
Intensive Care Med 2010;36(5):773–80.
33. Roberts N, Moule P. Chlorhexidine and tooth-brushing as prevention
strategies in reducing ventilator-associated pneumonia rates. Nurs Crit Care
2011;16(6):295–302.
34. Labeau SO, Van de Vyver K, Brusselaers N, et al. Prevention of ventilator-
associated pneumonia with oral antiseptics: a systematic review and
meta-analysis. Lancet Infect Dis 2011;11(11):845–54.
35. Li Bassi G, Torres A. Ventilator-associated pneumonia: role of positioning.
Curr Opin Crit Care 2011;17(1):57–63.
36. Lansford T, Moncure M, Carlton E, et al. Efficacy of a pneumonia prevention
protocol in the reduction of ventilator-associated pneumonia in trauma
patients. Surg Infect 2007;8(5):505–10.
37. Cocanour CS, Peninger M, Li T, et al. Decreasing ventilator-associated
pneumonia in a trauma ICU. J Trauma 2006;61(1):122–9.
38. Bird D, Zambuto A, O’Donnell C, et al. Adherence to ventilator-associated
pneumonia bundle and incidence of ventilator-associated pneumonia in the
surgical intensive care unit. Arch Surg 2010;145(5):465–70.
39. Stone Jr ME, Snetman D, O’ Neill A, et al. Daily multidisciplinary rounds to
implement the ventilator bundle decreases ventilator-associated pneumonia
in trauma patients: but does it affect outcome? Surg Infect 2011;12(5):373–8.
40. Sundar KM, Nielsen D, Sperry P. Comparison of ventilator-associated
pneumonia (VAP) rates between different ICUs: implications of a zero VAP
rate. J Crit Care 2011;Jul 5. Epub ahead of print.
41. Kollef MH, Shapiro SD, Silver P, et al. A randomized, controlled trial of
protocol-directed versus physician-directed weaning from mechanical
ventilation. Crit Care Med 1997;25(4):567–74.

42. Dries DJ, McGonigal MD, Malian MS, et al. Protocol-driven ventilator
weaning reduces use of mechanical ventilation, rate of early reintubation,
and ventilator-associated pneumonia. J Trauma Inj Infect Crit Care 2004;
56(5):943–52.
43. Henneman EA, Dracup KA, Ganz T, et al. Effect of a collaborative weaning
plan on patient outcome in the critical care setting. Crit Care Med 2001;
29(2):297–303.
44. Henneman EA, Dracup KA, Ganz T, et al. Using a collaborative weaning
plan to decrease duration of mechanical ventilation and length of stay in the
intensive care unit for patients receiving long-term ventilation. Am J Crit
Care 2002;11(2):132–40.
45. Thomas BW, Maxwell RA, Dart BW, et al. Errors in administrative-reported
ventilator-associated pneumonia rates: are never events really so? Am Surg
2011;77(8):998–1002.
46. Wahl WL, Zalewski C, Hemmila MR. Pneumonia in the surgical intensive
care unit: is every one preventable? Surgery 2011;150(4):665–72.
Section I—Background 55
47. Bouadma L, Deslandes E, Lolom I, et al. Long-term impact of a multifaceted
prevention program on ventilator-associated pneumonia in a medical
intensive care unit. Clin Infect Dis 2010;51(10):1115–22.
48. Klompas M. Is a ventilator-associated pneumonia rate of zero really possible?
Curr Opin Infect Dis 2012;Jan 13. Epub ahead of print.
49. VanGilder C, Amlung S, Harrison P, et al. Results of the 2008-2009
International Pressure Ulcer Prevalence Survey and a 3-year, acute care,
unit-specific analysis. Ostomy Wound Manage 2009;55(11):39–45.
50. Paul JA, Keller JW, van Ramshorst B, et al. Pressure ulcers in intensive care
patients: a review of risks and prevention. Intensive Care Med
2002;28:1379–88.
A complete list of references for this chapter can be found online at
www.expertconsult.com.

References
1. Pronovost PJ, Angus DC, Dorman TM, et al. Physician staffing patterns and
clinical outcomes in critically ill patients: a systematic review. JAMA 2002;
288(17): 2151–62.
2. Pronovost PJ, Miller MR, Dorman TM, et al. Developing and implementing
measures of quality of care in the intensive care unit. Curr Opin Crit Care
2001;7(4):297–303.
3. Mallon WJ. Earnest Amory Codman: the end result of a life in medicine.
Philadelphia: Saunders; 2000.
4. Deming WE. Out of the crisis. Cambridge, MA: MIT Press; 1986.
5. Donabedian A. The quality of care: how can it be assessed? JAMA 1988;
260(12):1743–8.
6. Sirio CA, Shepardson LB, Rotondi AJ, et al. Community-wide assessment of
intensive care outcomes using a physiologically based prognostic measure:
implications for critical care delivery from Cleveland health quality choice.
Chest 1999;115(3):793–801.
7. Glance LG, Osler TM, Dick AP. Rating the quality of intensive care units: is
it a function of the intensive care unit scoring system? Crit Care Med 2002;
30(9):1976–82.
8. Pronovost PJ, Berenholtz SM, Ngo K, et al. Developing and pilot testing
quality indicators in the intensive care unit. J Crit Care 2003;18(3):145–55.
9. Rhodes A, Moreno RP, Azoulay E, et al. Prospectively defined indicators
to improve the safety and quality of care for critically ill patients: a report
from the Task Force on Safety and Quality of the European Society of
Intensive Care Medicine (ESICM). Intensive Care Med 2012; Jan 26. Epub
ahead of print.
10. Lilford R, Pronovost PJ. Using hospital mortality rates to judge hospital
performance: a bad idea that just won’t go away. BMJ 2010;340:955–7.
11. Huang DT, Clermont G, Kong L, et al. Intensive care unit safety culture
and outcomes: a US multicenter study. Int J Qual Health Care 2010;22(3):
151–61.
12. Sakr Y, Marques J, Mortsch S, et al. Is the SAPS II score valid in surgical
intensive care unit patients? J Eval Clin Pract 2010;Sep 22. Epub ahead of
print.
13. Poole D, Rossi C, Anghileri A, et al. External validation of the Simplified
Acute Physiology Score (SAPS) 3 in a cohort of 28,357 patients from 147
Italian intensive care units. Intensive Care Med 2009;35(11):1916–24.
14. Niskanen M, Reinikainen M, Pettilä V. Case-mix-adjusted length of stay
and mortality in 23 Finnish ICUs. Intensive Care Med 2009;35(6):1060–7.
15. Brinkman S, Bakhshi-Raiez F, Abu-Hanna A, et al. External validation of
Acute Physiology and Chronic Health Evaluation IV in Dutch intensive care
units and comparison with Acute Physiology and Chronic Health
Evaluation II and Simplified Acute Physiology Score II. J Crit Care 2011;
26(1):105.e11–18.
16. Moreno RP, Bauer P, Metnitz PG. Characterizing performance profiles of
ICUs. Curr Opin Crit Care 2010;16(5):477–81.
17. Jarman B, Aylin P: Death rates in England and Wales and the United States:
variation with age, sex, and race. BMJ 2004;329:1367.
18. Brinkman S, Abu-Hanna A, van der Veen A, et al. A comparison of the
performance of a model based on administrative data and a model based
on clinical data: effect of severity of illness on standardized mortality ratios
of intensive care units. Crit Care Med 2012;40(2):373–8.
19. Lilly CM, Zuckerman IH, Badawi O, et al. Benchmark data from more than
240,000 adults that reflect the current practice of critical care in the United
States. Chest 2011;140(5):1232–42.
20. Curtis JR, Patrick DL, Engelberg RA, et al. A measure of the quality of
dying and death: initial validation using after-death interviews with family
members. J Pain Symptom Manage 2002;24(1):17–31.
21. Glavan BJ, Engelberg RA, Downey L, et al. Using the medical record to
evaluate the quality of end-of-life care in the intensive care unit. Crit Care
Med 2008;36(4):1138–46.
22. Mularski RA, Heine CE, Osborne ML, et al. Quality of dying in the ICU:
ratings by family members. Chest 2005;128(1):280–7.
23. Claessen SJ, Francke AL, Belarbi HE, et al. A new set of quality indicators
for palliative care: process and results of the development trajectory. J Pain
Symptom Manage. 2011;42(2):169–82.
24. Ghafoor VL, Silus LS. Developing policy, standard orders, and quality-
assurance monitoring for palliative sedation therapy. Am J Health Syst
Pharm 2011;68(6):523–7.
25. Nelson JE, Bassett R, Boss RD, et al. Improve Palliative Care in the Intensive
Care Unit Project. Models for structuring a clinical initiative to enhance
palliative care in the intensive care unit: a report from the IPAL-ICU
project (Improving Palliative Care in the ICU). Crit Care Med 2010;38(9):
1765–72.
26. Yang CC, Shih NC, Chang WC, et al. Long-term medical utilization
following ventilator-associated pneumonia in acute stroke and traumatic
Section I—Background 55.e1
brain injury patients: a case-control study. BMC Health Serv Res
2011;11:289.
27. Klompas MM, Platt RM. Ventilator-associated pneumonia: the wrong
quality measure for benchmarking. Ann Intern Med 2007;147(11):803–5.
28. Uckay I, Ahmed QA, Sax H, et al. Ventilator-associated pneumonia as a
quality indicator for patient safety? Clin Infect Dis 2008;46(4):557–63.
29. Eggimann P, Hugonnet S, Sax H, et al. Ventilator-associated pneumonia:
caveats for benchmarking. Intensive Care Med 2003;29(11):2086–9.
30. Mangino JE, Peyrani P, Ford KD, et al. Development and implementation
of a performance improvement project in adult intensive care units:
overview of the Improving Medicine Through Pathway Assessment of
Critical Therapy in Hospital-Acquired Pneumonia (IMPACT-HAP) study.
Crit Care 2011;15(1):R38.
31. Ramirez P, Bassi GL, Torres A. Measures to prevent nosocomial infections
during mechanical ventilation. Curr Opin Crit Care 2012;18(1):86–92.
32. Rello J, Lode H, Cornaglia G, et al. VAP care bundle contributors:
a European care bundle for prevention of ventilator-associated pneumonia.
Intensive Care Med 2010;36(5):773–80.
33. Roberts N, Moule P. Chlorhexidine and tooth-brushing as prevention
strategies in reducing ventilator-associated pneumonia rates. Nurs Crit Care
2011;16(6):295–302.
34. Labeau SO, Van de Vyver K, Brusselaers N, et al. Prevention of ventilator-
associated pneumonia with oral antiseptics: a systematic review and
meta-analysis. Lancet Infect Dis 2011;11(11):845–54.
35. Li Bassi G, Torres A. Ventilator-associated pneumonia: role of positioning.
Curr Opin Crit Care 2011;17(1):57–63.
36. Lansford T, Moncure M, Carlton E, et al. Efficacy of a pneumonia
prevention protocol in the reduction of ventilator-associated pneumonia in
trauma patients. Surg Infect 2007;8(5):505–10.
37. Cocanour CS, Peninger M, Li T, et al. Decreasing ventilator-associated
pneumonia in a trauma ICU. J Trauma 2006;61(1):122–9.
38. Bird D, Zambuto A, O’Donnell C, et al. Adherence to ventilator-associated
pneumonia bundle and incidence of ventilator-associated pneumonia in
the surgical intensive care unit. Arch Surg 2010;145(5):465–70.
39. Stone Jr ME, Snetman D, O’ Neill A, et al. Daily multidisciplinary rounds to
implement the ventilator bundle decreases ventilator-associated pneumonia
in trauma patients: but does it affect outcome? Surg Infect 2011;12(5):
373–8.
40. Sundar KM, Nielsen D, Sperry P. Comparison of ventilator-associated
pneumonia (VAP) rates between different ICUs: implications of a zero VAP
rate. J Crit Care 2011;Jul 5. Epub ahead of print.
41. Kollef MH, Shapiro SD, Silver P, et al. A randomized, controlled trial of
protocol-directed versus physician-directed weaning from mechanical
ventilation. Crit Care Med 1997;25(4):567-74.
42. Dries DJ, McGonigal MD, Malian MS, et al. Protocol-driven ventilator
weaning reduces use of mechanical ventilation, rate of early reintubation,
and ventilator-associated pneumonia. J Trauma Inj Infect Crit Care
2004;56(5):943–52.
43. Henneman EA, Dracup KA, Ganz T, et al. Effect of a collaborative weaning
plan on patient outcome in the critical care setting. Crit Care Med
2001;29(2):297–303.
44. Henneman EA, Dracup KA, Ganz T, et al. Using a collaborative weaning
plan to decrease duration of mechanical ventilation and length of stay in
the intensive care unit for patients receiving long-term ventilation. Am J
Crit Care 2002;11(2):132–40.
45. Thomas BW, Maxwell RA, Dart BW, et al. Errors in administrative-reported
ventilator-associated pneumonia rates: are never events really so? Am Surg
2011;77(8):998–1002.
46. Wahl WL, Zalewski C, Hemmila MR. Pneumonia in the surgical intensive
care unit: is every one preventable? Surgery 2011;150(4):665–72.
47. Bouadma L, Deslandes E, Lolom I, et al. Long-term impact of a
multifaceted prevention program on ventilator-associated pneumonia in a
medical intensive care unit. Clin Infect Dis 2010;51(10):1115–22.
48. Klompas M. Is a ventilator-associated pneumonia rate of zero really
possible? Curr Opin Infect Dis 2012;Jan 13. Epub ahead of print.
49. VanGilder C, Amlung S, Harrison P, et al. Results of the 2008-2009
International Pressure Ulcer Prevalence Survey and a 3-year, acute care,
unit-specific analysis. Ostomy Wound Manage 2009;55(11):39–45.
50. Paul JA, Keller JW, van Ramshorst B, et al. Pressure ulcers in intensive care
patients: a review of risks and prevention. Intensive Care Med 2002;28:
1379–88.
51. de Laat EH, Pickkers P, Schoonhoven L, et al. Guideline implementation
results in a decrease of pressure ulcer incidence in critically ill patients. Crit
Care Med 2007;35(3):815–20.
52. Sihler KC, Chenoweth C, Zalewski C, et al. Catheter-related vs. catheter-
associated blood stream infections in the intensive care unit: incidence,
microbiology, and implications. Surg Infect 2010;11(6):529–34.
55.e2 Section I—Background
53. Taylor GD, McKenzie M, Kirkland T. The impact of health care
restructuring on nosocomically acquired blood stream infectious. Can J
Infect Dis 2000;11(1):29–32.
54. Douglas JE, Schuerer JEZ, Thomas J, et al. Effect of chlorhexidine/silver
sulfadiazine-impregnated central venous catheters in an intensive care unit
with a low blood stream infection rate after implementation of an
educational program: a before–after trial. Surg Infect 2007;8(4):445–54.
55. Rupp ME, Lisco SJ, Lipsett PA, et al. Effect of a second-generation venous
catheter impregnated with chlorhexidine and silver sulfadiazine on central
catheter-related infections: a randomized, controlled trial. Ann Intern Med
2005;143(8):570–80.
56. Darouiche RO, Raad II, Heard SO, et al. A comparison of two
antimicrobial-impregnated central venous catheters. N Engl J Med 1999;
340(1):1–8.
57. Cherry-Bukowiec JR, Denchev K, Dickinson S, et al. Prevention of
catheter-related blood stream infection: back to basics? Surg Infect 2011;
12(1):27–32. Epub 2010, Dec 20.
58. Gunst M, Matsushima K, Vanek S, et al. Peripherally inserted central
catheters may lower the incidence of catheter-related blood stream
infections in patients in surgical intensive care units. Surg Infect
2011;12(4):279–82. Epub 2010, Jul 14.
59. Cherry RA, West CE, Hamilton MC, et al. Reduction of central venous
catheter associated blood stream infections following implementation of a
resident oversight and credentialing policy. Patient Saf Surg 2011;5(1):15.
60. Seddon ME, Hocking CJ, Mead P, et al. Aiming for zero: decreasing central
line associated bacteraemia in the intensive care unit. N Z Med J 2011;
124(1339):9–21.
61. Render ML, Hasselbeck R, Freyberg RW, et al. Reduction of central line
infections in Veterans Administration intensive care units: an observational
cohort using a central infrastructure to support learning and improvement.
BMJ Qual Saf 2011;20(8):725–32. Epub 2011, Apr 2.
62. Finney SJ, Zekveld C, Elia A, et al. Glucose control and mortality in
critically ill patients. JAMA 2003;290(15):2041–7.
63. Gale SC, Sicoutris CP, Reilly PM, et al. Poor glycemic control is associated
with increased mortality in critically ill trauma patients. Am Surg 2007;
73(5):454–60.
64. Shulman R, Finney SJ, O’Sullivan C, et al. Tight glycaemic control: a
prospective observational study of a computerised decision-supported
intensive insulin therapy protocol. Crit Care 2007;11(4):R75.
65. Meynaar IA, Dawson L, Tangkau PL, et al. Introduction and evaluation of
a computerised insulin protocol. Intensive Care Med 2007;33(4):591–6.
66. Quinn JA, Berghoff JL, Colombo CS, et al. A practical approach to
hyperglycemia management in the intensive care unit: evaluation of an
intensive insulin infusion protocol. Pharmacotherapy 2006;26(10):1410–20.
67. Krinsley JS, Jones RL. Cost analysis of intensive glycemic control in
critically ill adult patients. Chest 2006;129(3):644–50.
68. Schmeltz LR, DeSantis AJ, Thiyagarajan V, et al. Reduction of surgical
mortality and morbidity in diabetic patients undergoing cardiac surgery
with a combined intravenous and subcutaneous insulin glucose
management strategy. Diabetes Care 2007;30(4):823–8.
69. Van den Berghe G, Wilmer A, Hermans G, et al. Intensive insulin therapy in
the medical ICU. N Engl J Med 2006;354(5):449–61.
70. Krinsley JS. Effect of an intensive glucose management protocol on the
mortality of critically ill adult patients. Mayo Clin Proc 2004;79(8):
992–1000.
71. van der Voort PHJ, Feenstra RA, Bakker AJ, et al. Intravenous glucose
intake independently related to intensive care unit and hospital mortality:
an argument for glucose toxicity in critically ill patients. Clin Endocrinol
(Oxf) 2006;64(2):141-5.
72. Anger KE. Barriers to glucose control in the intensive care unit.
Pharmacotherapy 2006;26(2):214–28.
73. Wiener RS, Wiener DC, Larson RJ. Benefits and risks of tight glucose
control in critically ill adults: a meta-analysis. JAMA 2008;300(8):933–44.
74. NICE-SUGAR Study Investigators, Finfer S, Chittock DR, et al. Intensive
versus conventional glucose control in critically ill patients. N Engl J Med
2009;360(13):1283–97. Epub 2009, Mar 24.
75. Egi M, Finfer S, Bellomo R. Glycemic control in the ICU. Chest 2011;
140(1):212–20.
76. Kutcher ME, Pepper MB, Morabito D, et al. Finding the sweet spot:
identification of optimal glucose levels in critically injured patients.
J Trauma 2011;71(5):1108–14.
77. Oddo M, Schmidt JM, Carrera C, et al. Impact of tight glycemic control on
cerebral glucose metabolism after severe brain injury: a microdialysis study.
Crit Care Med 2008;36:3233–8.
78. Anderson M, Zito D, Kongable G. Benchmarking glucose results through
automation: the 2009 remote automated laboratory system report.
J Diabetes Sci Technol 2010;4(6):1507–13.
79. Cook CB, Kongable GL, Potter DJ, et al. Inpatient glucose control: a
glycemic survey of 126 U.S. hospitals. J Hosp Med 2009;4(9):E7–E14.
80. Sidorenkov G, Haaijer-Ruskamp FM, de Zeeuw D, et al. Review: relation
between quality-of-care indicators for diabetes and patient outcomes: a
systematic literature review. Med Care Res Rev 2011;68(3):263–89. Epub
2011, May 2.
81. Duning T, van den Heuvel I, Dickmann A, et al. Hypoglycemia aggravates
critical illness-induced neurocognitive dysfunction. Diabetes Care
2010;33(3):639–44. Epub 2009, Dec 23.
82. Hermanides J, Vriesendorp TM, Bosman RJ, et al. Glucose variability is
associated with intensive care unit mortality. Crit Care Med 2010;38(3):
838–42.
83. Meyfroidt G, Wouters P, De Becker W, et al. Impact of a computer-
generated alert system on the quality of tight glycemic control. Intensive
Care Med 2011;37(7):1151–7. Epub 2011, Mar 3.
84. Angus DC, Kelley MA, Schmitz RJ, et al. Current and projected workforce
requirements for care of the critically ill and patients with pulmonary
disease: can we meet the requirements of an aging population? JAMA 2000;
284(21):2762–70.
85. Angus DC, Shorr AF, White A, et al. Critical care delivery in the United
States: distribution of services and compliance with leapfrog
recommendations. Crit Care Med 2006;34(4):1016–24
86. Carlson RW, Weiland DE, Srivathsan K. Does a full-time, 24-hour
intensivist improve care and efficiency? Crit Care Clin 1996;12(3):525–51.
87. Burchardi H, Moerer O. Twenty-four hour presence of physicians in the
ICU. Crit Care 2001;5(3):131–7.
88. Arabi YM, Alshimemeri AM, Taher SM. Weekend and weeknight
admissions have the same outcome of weekday admissions to an intensive
care unit with onsite intensivist coverage. Crit Care Med 2006;34(3):605–11.
89. Varelas P, Varelas PN, Schultz L, et al. The impact of a neuro-intensivist on
patients with stroke admitted to a neurosciences intensive care unit.
Neurocrit Care 2008;9(3):293–9.
90. Ghorra SMD, Reinert SE, Cioffi W, et al. Analysis of the effect of conversion
from open to closed surgical intensive care unit. Ann Surg 1999;229(2):
163–71.
91. Hanson III CW, Deutschman CS, Anderson III HL, et al. Effects of an
organized critical care service on outcomes and resource utilization: a
cohort study. Crit Care Med 1999;27(2):270–4.
92. Nathens AB, Rivara FP, MacKenzie EJ, et al. The impact of an intensivist-
model ICU on trauma-related mortality. Ann Surg 2006;244(4):545–54.
93. Pronovost PJ, Jenckes MW, Dorman TM, et al. Organizational
characteristics of intensive care units related to outcomes of abdominal
aortic surgery. JAMA 1999;281(14):1310–17.
94. Dimick JB, Pronovost PJ, Heitmiller RF, et al. Intensive care unit physician
staffing is associated with decreased length of stay, hospital cost, and
complications after esophageal resection. Crit Care Med 2001;29(4):
753–8.
95. Treggiari MM, Martin DP, Yanez ND, et al. Effect of intensive care unit
organizational model and structure on outcomes in patients with acute
lung injury. Am J Respir Crit Care Med 2007;176(7):685–90.
96. Chittawatanarat K, Pamorsinlapathum T. The impact of closed ICU model
on mortality in general surgical intensive care unit. J Med Assoc Thai
2009;92(12):1627–34.
97. Treggiari MM, Martin DP, Yanez ND, et al. Effect of intensive care unit
organizational model and structure on outcomes in patients with acute
lung injury. Am J Respir Crit Care Med 2007;176(7):685–90. Epub 2007,
Jun 7.
98. Gajic O, Afessa B, Hanson AC, et al. Effect of 24-hour mandatory versus
on-demand critical care specialist presence on quality of care and family
and provider satisfaction in the intensive care unit of a teaching hospital.
Crit Care Med. 2008;36(1):36–44.
99. Kahn JM, Brake H, Steinberg KP. Intensivist physician staffing and the
process of care in academic medical centres. Qual Saf Health Care
2007;16(5):329–33.
100. Nathens AB, Maier RV, Jurkovich GJ, et al. The delivery of critical care
services in us trauma centers: is the standard being met? J Trauma
2006;60(4):773–84.
101. Byrnes MC, Schuerer DJ, Schallom ME, et al. Implementation of a
mandatory checklist of protocols and objectives improves compliance with
a wide range of evidence-based intensive care unit practices. Crit Care Med
2009;37(10):2775–81.
102. Scales DC, Dainty K, Hales B, et al A multifaceted intervention for quality
improvement in a network of intensive care units: a cluster randomized
trial. JAMA 2011;305(4):363–72. Epub 2011, Jan 19.
103. Schramm GE, Kashyap R, Mullon JJ, et al. Septic shock: a multidisciplinary
response team and weekly feedback to clinicians improve the process of
care and mortality. Crit Care Med 2011;39(2):252–8.

104. Daniels R. Surviving the first hours in sepsis: getting the basics right (an
intensivist’s perspective). J Antimicrob Chemother 2011;66(Suppl 2):
ii11–23.
105. Vincent JL. We should abandon randomized controlled trials in the
intensive care unit. Crit Care Med 2010;38(10 Suppl):S534–8.
106. Ospina-Tascon GA, Buchele GL, Vincent JL: Multicenter, randomized,
controlled trials evaluating mortality in intensive care: doomed to fail? Crit
Care Med 2008;36:1311–22.
107. Walker S, Brett S. Oiling the wheels of intensive care to reduce “machine
friction”: the best way to improve outcomes? Crit Care Med 2010;38(10
Suppl):S642–8.
108. Tobin MJ. Counterpoint: evidence-based medicine lacks a sound scientific
base. Chest 2008;133(5):1071–4; discussion 1074–7.
109. Wendon J. Critical care “normality”: individualized versus protocolized
care. Crit Care Med 2010;38(10 Suppl):S590–9.
110. Prasad M, Holmboe ES, Lipner RS, et al. Clinical protocols and trainee
knowledge about mechanical ventilation. JAMA 2011;306(9):935–41.
111. Kahol K, Vankipuram M, Patel VL, et al. Deviations from protocol in a
complex trauma environment: errors or innovations? J Biomed Inform
2011;44(3):425–31. Epub 2011, Apr 13.
112. Moreno R, Rhodes A. Evidence should not be viewed in isolation. Crit Care
Med 2010;38(10 Suppl):S528–33.
113. Gasperino J. The Leapfrog initiative for intensive care unit physician
staffing and its impact on intensive care unit performance: a narrative
review. Health Policy 2011;102(2-3):223–8. Epub 2011, Mar 24.
114. Kahn JM, Matthews FA, Angus DC, et al. Barriers to implementing the
Leapfrog group recommendations for intensivist physician staffing: a survey
of intensive care unit directors. J Crit Care 2007;22(2): 97–103.
115. Pronovost PJ, Thompson DA, Holzmueller CG, et al. Impact of the
Leapfrog group’s intensive care unit physician staffing standard. J Crit Care
2007;22(2):89–96.e24.
116. Pronovost PJ, Needham DM, Waters HR, et al. Intensive care unit physician
staffing: Financial modeling of the Leapfrog standard. Crit Care Med 2004;
32(6):1247–53.
117. Kozar RA, Shackford SR, Cocanour CS. Challenges to the care of the
critically ill: novel staffing paradigms. J Trauma 2008;64(2):366–73.
118. Pronovost PJ, Waters HP, Dorman TM. Impact of critical care physician
workforce for intensive care unit physician staffing. Curr Opin Crit Care
2001;7(6):456–9.
119. Napolitano LM, Fulda GJ, Davis KA, et al. Challenging issues in surgical
critical care, trauma, and acute care surgery: a report from the critical care
committee of the American Association for the Surgery of Trauma.
J Trauma 2010;69(6):1619–33.
120. Gracias VH, Sicoutris CP, Stawicki SP, et al. Critical care nurse practitioners
improve compliance with clinical practice guidelines in “semiclosed”
surgical intensive care unit. J Nurs Care Qual 2008;23(4):338–44.
121. Lovell MA, Mudaliar MY, Klineberg PL. Intrahospital transport of critically
ill patients: complications and difficulties. Anaesth Intensive Care 2001;
29(4):400–5.
122. Warren J, Fromm RE Jr, Orr RA, et al. Guidelines for the inter- and
intrahospital transport of critically ill patients. Crit Care Med
2004;32(1):256–62.
123. Bercault N, Wolf M, Runge I, et al. Intrahospital transport of critically ill
ventilated patients: a risk factor for ventilator-associated pneumonia—a
matched cohort study. Crit Care Med 2005;33(11):2471–8.
124. Swanson E, Mascitelli J, Stiefel M, et al. The effect of patient transport on
brain oxygen in comatose patients. Neurosurgery 2010;66:925–32.
125. Kue R, Brown P, Ness C, et al. Adverse clinical events during intrahospital
transport by a specialized team: a preliminary report. Am J Crit Care
2011;20(2):153–61; quiz 162.
126. Wiegersma JS, Droogh JM, Zijlstra JG, et al. Quality of interhospital
transport of the critically ill: impact of a mobile intensive care unit
with a specialized retrieval team. Crit Care 2011;15(1):R75. Epub 2011,
Feb 28.
127. Lyons MN, Standley TD, Gupta AK. Quality improvement of doctors’
shift-change handover in neuro-critical care. Qual Saf Health Care
2010;19(6):e62. Epub 2010, Apr 27.
128. Joy BF, Elliott E, Hardy C, et al. Standardized multidisciplinary protocol
improves handover of cardiac surgery patients to the intensive care unit.
Pediatr Crit Care Med 2011;12(3):304–8.
129. Cleland JA, Ross S, Miller SC, et al. “There is a chain of Chinese
whispers …”: empirical data support the call to formally teach
handover to prequalification doctors. Qual Saf Health Care 2009;18(4):
267–71.
130. Pickering BW, Hurley K, Marsh B. Identification of patient information
corruption in the intensive care unit: using a scoring tool to direct quality
improvements in handover. Crit Care Med 2009;37(11):2905–12.
Section I—Background 55.e3
131. Clarke CM, Persaud DD. Leading clinical handover improvement: a change
strategy to implement best practices in the acute care setting. J Patient Saf
2011;7(1):11–18.
132. Manser T, Foster S, Gisin S, et al. Assessing the quality of patient handoffs
at care transitions. Qual Saf Health Care 2010;19(6):e44.
133. Manthous CA, Hollingshead AB. Team science and critical care. Am J
Respir Crit Care Med 2011;184(1):17–25. Epub 2011, Mar 18.
134. Oglesby KJ, Durham L, Welch J, et al. ‘Score to door time’, a benchmarking
tool for rapid response systems: a pilot multi-centre service evaluation. Crit
Care 2011;15(4):R180.
135. Hunziker S, Johansson AC, Tschan F, et al. Teamwork and leadership in
cardiopulmonary resuscitation. J Am Coll Cardiol 2011;57(24):2381–8.
136. Rosenfeld BA, Dorman TM, Breslow MJ, et al. Intensive care unit
telemedicine: alternate paradigm for providing continuous intensivist care.
Crit Care Med 2000;28(12):3925–31.
137. Breslow MJ, Rosenfeld BA, Doerfler M, et al. Effect of a multiple-site
intensive care unit telemedicine program on clinical and economic
outcomes: an alternative paradigm for intensivist staffing. Crit Care Med
2004;32(1):31–8.
138. Breslow MJ. Remote ICU care programs: current status. J Crit Care
2007;22(1):66–76.
139. Pastores SM, O’Connor MF, Kleinpell RM, et al. The Accreditation Council
for Graduate Medical Education resident duty hour new standards: history,
changes, and impact on staffing of intensive care units. Crit Care Med
2011;39(11):2540–9.
140. Cummings J, Krsek C, Vermoch K, et al. Intensive care unit telemedicine:
review and consensus recommendations. Am J Med Qual 2007;22(4):
239–50.
141. Moreno RP, Rhodes A, Donchin Y. Patient safety in intensive care medicine:
the Declaration of Vienna. Intensive Care Med 2009;35:1667–72.
142. Whiteman R, Gould L, Oczkowski W, et al. Using a quality improvement
process to create measurable improvement in care delivery for acute stroke.
Healthc Quart 2011;14(3):75–9.
143. Wiedmann S, Norrving B, Nowe T, et al. Variations in quality indicators of
acute stroke care in 6 European countries: the European Implementation
Score (EIS) collaboration. Stroke 2012;43(2):458–63. Epub 2011, Dec 8.
144. Woodhouse D, Berg M, van der Putten J, et al. Will benchmarking ICUs
improve outcome? Curr Opin Crit Care 2009;15(5):450–5. Review.
145. Kardooni S, Haut ER, Chang DC, et al. Hazards of benchmarking
complications with the National Trauma Data Bank: numerators in search
of denominators. J Trauma. 2008;64(2):273–7.
146. Hernandez AF, Fonarow GC, Liang L, et al. The need for multiple measures
of hospital quality: results from the Get with the Guidelines-Heart Failure
Registry of the American Heart Association. Circulation 2011;124(6):
712–19. Epub 2011, Jul 25.
147. Sciacovelli L, Sonntag O, Padoan A, et al. Monitoring quality indicators in
laboratory medicine does not automatically result in quality improvement.
Clin Chem Lab Med 2011;Dec 13. Epub ahead of print.
148. Ketelaar NA, Faber MJ, Flottorp S, et al. Public release of performance data
in changing the behaviour of healthcare consumers, professionals or
organisations. Cochrane Database Syst Rev 2011;11:CD004538.
149. Ryan AM, Blustein J. The effect of the MassHealth hospital pay-for-
performance program on quality. Health Serv Res 2011;46(3):712–28.
doi: 10.1111/j.1475–6773.2010.01224.x. Epub 2011, Jan 6.
150. Rademakers J, Delnoij D, de Boer D. Structure, process or outcome: which
contributes most to patients’ overall assessment of healthcare quality? BMJ
Qual Saf 2011;20(4):326–31. Epub 2011, Feb 21.
151. Metersky ML. Should management of pneumonia be an indicator of
quality of care? Clin Chest Med 2011;32(3):575–89.
152. Breuer S, Arquilla V. Clinical documentation improvement: focus on
quality. Healthc Financ Manage 2011;65(8):84–6, 88, 90.
153. Zegers M, de Bruijne MC, Spreeuwenberg P, et al. Quality of patient record
keeping: an indicator of the quality of care? BMJ Qual Saf 2011;20(4):
314–18. Epub 2011, Feb 8.
154. Goeschel CA, Wachter RM, Pronovost PJ. Responsibility for quality
improvement and patient safety: hospital board and medical staff
leadership challenges. Chest 2010;138(1):171–8.
155. Gisvold SE, Fasting S. How do we know that we are doing a good job—can
we measure the quality of our work? Best Pract Res Clin Anaesthesiol
2011;25(2):109–22.
156. Pinto A, Burnett S, Benn J, et al. Improving reliability of clinical care
practices for ventilated patients in the context of a patient safety
improvement initiative. J Eval Clin Pract 2011;17(1):180–7.
157. Pronovost PJ, Cardo DM, Goeschel CA, et al. A research framework for
reducing preventable patient harm. Clin Infect Dis 2011;52(4):507–13.
158. Wolf JA. Sustaining high performance: dynamic balancing in an otherwise
unbalanced system. Adv Health Care Manage 2011;10:367–84.
 I
Chapter
7  
Brain Monitoring Issues
in Pediatrics
Anthony A. Figaji
Introduction
Neuromonitoring in the intensive care unit (ICU) is under-
taken for several acute neurologic conditions in children.
Traumatic brain injury (TBI) constitutes the most common
indication for neuromonitoring because injury is the leading
cause of death in children over the age of 1 year,1,2 and those
who sustain TBI have the highest chance of dying or being
permanently disabled.3,4 Early treatment to prevent second-
ary insults may avoid as much as 30% of pediatric TBI
deaths.5 Therefore, monitoring for these potential secondary
insults is the foundation of management for children.
Importantly, experimental studies demonstrate that neuro-
monitoring findings, including intracranial pressure (ICP),
brain oxygen, and microdialysis correlate with neuropathy
of animal models of pediatric TBI.6 The aim of this chapter
is not to provide a comprehensive summary of monitoring
options, but rather to highlight some differences between
adults and children with respect to neuromonitoring. In
addition, the reader is reminded that neuromonitoring may
be valuable in children and adolescents with nontraumatic
coma (e.g., meningitis, hepatic encephalopathy, stroke, near
drowning, and hypoxic-ischemic injury).7-10
Children and Adults Are Different
Potential secondary insults in children are similar to those in
adults; however, children have different physical dimensions
and physiologic responses. They are rapidly growing and
developing, and normal ranges and thresholds for treatment
not only are different from adults, but also vary between age
groups. For these reasons, and because of the smaller research
base, the best evidence-based recommendations for specific
aspects of clinical management in pediatric TBI are still mostly
set at the level of options.11
Glasgow Coma Scale
The Glasgow Coma Scale (GCS) is the standard method of
assessing and monitoring the level of consciousness in pedi-
atric TBI.12,13 In children the utility of the GCS is limited in
younger patients; therefore, modifications of the GCS such as
the Pediatric Coma Scale14 and the Children’s Coma Scale15 are
56
used. However, the GCS as a monitoring tool is limited by
prehospital and ICU sedation16 and because children with
higher GCS may deteriorate suddenly.17 Methods of assessing
initial clinical severity, such as the GCS and acute illness
scores, do not predict which patients are at risk for all types
of secondary injury.18
Prehospital Hypotension
and Hypoxia
Polytrauma is common in children with severe TBI,19,20 and
the addition of extracranial injury increases the risk of hypo-
tension and hypoxia.21 Hypoxia and hypotension occur in
31% to 68% of children who sustain severe TBI and are associ-
ated with poor outcome, especially hypotension.12,21 Although
the diagnosis of hypoxia in children is standard, and consis-
tent with that in adults, the definition of hypotension varies.
Systolic blood pressure (SBP) is commonly used to define
hypotension, either as SBP less than 90 mm Hg or less than
the fifth percentile for age,22,23 but these approaches have limi-
tations. First, mean arterial pressure (MAP) is more appropri-
ate for determining likely cerebral perfusion pressure (CPP).
Second, height influences the normal range of blood pressure
(BP) in children, but is rarely considered. Height is related to
BP independently of age, so there are significant differences
across height-for-age percentile blood pressure ranges.24,25
Cerebral Blood Flow
There are important differences in cerebral blood flow (CBF)
values between adults and children in health and disease.
Specifically, the role of hyperemia as the predominant cause
of diffuse swelling observed after TBI in children has been
debated.26-28 When CBF values in children are compared with
mean values for age, the incidence of true hyperemia appears
to be much lower than previously believed.28 CBF values tend
to increase from the lowest values at birth to a peak at age
3 to 5 years and then decrease to adult levels.28,29 Therefore,
when interpreting high CBF values in children with TBI (high
and low), age-specific ranges must be considered. Low CBF is
common early after injury and is associated with worse
outcome; however, data on practical CBF monitoring for
© Copyright 2013 Elsevier Inc. All rights reserved.

children in critical care are limited.30,31 Ideally, absolute CBF
values should be interpreted with metabolic demand and
flow-metabolism coupling.
Intracranial Pressure: Indications,
Methods, and Thresholds
Indications for intracranial pressure (ICP) monitoring in chil-
dren vary among different institutions.3,32 Current recommen-
dations support ICP monitoring for children with severe TBI,
regardless of whether fontanels and sutures are open.11 Still,
in the absence of randomized controlled trials, some question
whether ICP monitoring is beneficial in pediatric TBI33; the
reasons for this are complex.34 In the United Kingdom only
about 60% of children presenting with severe TBI undergo
ICP monitoring, with large among-center variation,32 and
many patients receive ICP-targeted therapy without an ICP
monitor in situ. But adherence to published guidelines appears
to be increasing.35 Indications for ICP monitoring in children
are based on less evidence than in adults because of the
paucity of studies.11 Because computed tomography (CT)
signs cannot reliably exclude intracranial hypertension,36 it is
recommended to monitor all patients with severe TBI who are
not likely to be extubated within 12 hours.
Standard methods for ICP monitoring in adults are also
used in children; intraparenchymal devices are most com-
monly used and complications are rare.37 A ventriculostomy
can be useful in some patients38 but is less commonly used,32
most likely because of difficulties in cannulating small ven-
tricles in the typical child with diffuse swelling. For practical
purposes the bolt system for intracranial catheters in pediat-
rics is usually suitable even in younger children, but the very
young may require a formal burr hole and subgaleal tunneling
if the skull is particularly thin.
In children the definition of elevated ICP is less clear
because of changing physiologic norms with age, the differ-
ent pressure-volume relationship of the brain in children,38
and uncertainty about whether the autoregulatory curve in
children is similar to that of adults.39 Lower thresholds than
those recommended for treatment in adults are often used
for younger children, but there are no reliable data to support
this. The treatment threshold of 20 mm Hg is recommended
at the level of an option for children.11 If the anatomic char-
acteristics of the brain that determine the point at which a
pediatric brain undergoes herniation are similar to that of an
adult brain, and the autoregulatory curve is similar,39 then
targeting a similar threshold appears sensible, so long as an
adequate CPP is maintained. This ICP threshold for treat-
ment is supported by the findings in children of an inverse
relationship between CBF and ICP greater than 20 mm Hg27
and significant changes in the pressure-volume index when
ICP is greater than 20 mm Hg.38 However, the relationship
between ICP and CBF or oxygenation of the brain is com-
plex. Several pathophysiologic mechanisms may underlie
an increase in ICP in individual children, including classic
cerebral edema, seizures, impaired pressure autoregulation,
hyperemia (uncoupled flow-metabolism coupling), increased
CO2, and so on. Therefore, the precise relationship between
ICP and perfusion of the brain depends not only on the
effect of ICP on perfusion, but also on the reason for the
increased ICP.40 In the author’s experience, ICP affects oxy-
genation of the brain at different levels, depending on the age
Section I—Background 57
of the patient, the absolute ICP value, and the underlying
cause of the increased ICP. Although it is important to treat
elevated ICP, additional data are required to guide sound
clinical decisions. Therefore, a multimodal approach has par-
ticular value.
Cerebral Perfusion Pressure
Low CPP is known to be associated with poor outcome;
however, this may be a surrogate marker for high ICP. However,
some argue that absolute ICP values are of lesser importance
if CPP is adequate. Therefore, as in adults, the merits of ICP-
targeted treatment versus CPP-targeted treatment are also
debated in children.41 Higher CPP targets in adult TBI have
been recommended42 but may increase the risk of acute respi-
ratory distress syndrome,43 and if autoregulation is impaired
it may increase ICP44 and tissue edema.45,46 The Lund approach
using brain volume regulation and microdialysis monitoring
to accept lower CPP levels also has been described in chil-
dren47 but is not without controversy and risks.48
It has not been established that active modulation of CPP,
beyond avoiding hypotension, benefits pediatric TBI patients,
so there is little agreement on what constitutes an appropriate
CPP target for age. Chambers et al.49 recommended CPP
thresholds of 48, 54, and 58 for age-groups 2 through 6,
7 through 10, and 11 through 15 years, respectively, and
45 mm Hg overall.50 Prabhakaran et al.41 examined ICP- and
CPP-targeted approaches but reached no definitive conclu-
sion. Vavilala et al.51 found that all children who had an admis-
sion SBP less than 90 mm Hg had a poor outcome, but that
the 75th percentile of SBP for age was a better predictor for
outcome. Jones et al.52 and Hackbarth et al.20 recommended
CPP greater than 50 to 60 mm Hg, but Downard et al.53 found
that CPP greater than 50 mm Hg conferred no survival benefit
and argued that benefits of CPP elevation may simply be a
proxy for the avoidance of hypotension.
None of the these studies considered height for age in the
evaluation of MAP or CPP. Furthermore, all recommenda-
tions for specific targets may not be appropriate given
that impaired cerebral autoregulation is common in children
with severe TBI.39 Nevertheless, current recommendations
for pediatric TBI support the avoidance of CPP less than
40 mm Hg in all children and targeting CPP greater than
50 mm Hg as an option. These have no clear recommenda-
tions for an age-related approach.11 It is likely that a single CPP
value is not appropriate for all patients given differences in age
and metabolic needs, the tissue pressure effect of increased
ICP (rather than just the CPP-lowering effect thereof), the
status of pressure autoregulation, and the variable coupling
between metabolism and CBF.40,54,55
Brain Tissue Oxygen
Several studies have examined the relationship between brain
tissue oxygen tension (PbtO2) and outcome in adults,56 but
few in children.57-61 Narotam et al.57 (n = 16) found that no
children with normal initial PbtO2 died. Stiefel et al.58 (n = 6)
found PbtO2 was significantly lower if ICP was greater than
20 mm Hg and CPP was less than 40 mm Hg. Figaji et al.61
(n = 52) found that low PbtO2 was an independent predictor
of poor outcome. Poor outcome was linked to the depth and
duration of brain hypoxia, especially when PbtO2 was less than
58 Section I—Background
60
50
40
Intracranial pressure
30
20
10
0
35
30
25
20 Brain tissue oxygen
15
10
0 Transcranial Doppler
Fig. 7.1  Graph of continuous monitoring of intracranial pressure (ICP)
and brain tissue oxygen tension (PbtO2) in a 7-year-old child, recorded
over a 3-hour period, showing dynamic changes in ICP and its effect on
PbtO2.
10 mm Hg. In addition, episodes of critical brain hypoxia
(PbtO2 <10 mm Hg) occurred in almost one third of patients
despite recommended treatment targets being met for ICP,
CPP, arterial oxygenation, and hemoglobin concentration.59
These episodes appear to be not only outcome predictors
but also a manifestation of ongoing or secondary injury.62
In children, low PbtO2 is associated with poor outcome, is a
manifestation of secondary injury, and responds to global
changes in blood pressure, ICP (Fig. 7.1), and systemic
oxygenation.18,54,55,61,63-66
Near-Infrared Spectroscopy
Theoretically, near-infrared spectroscopy (NIRS) may be
more promising in young children and neonates67 in whom
transillumination is easier. Although many report the useful-
ness of NIRS in adults and children when there is no
primary brain injury, as in carotid endarterectomy and pedi-
atric cardiac anesthesia,68,69 studies of its utility in the pedi-
atric TBI population are limited.66 Although some suggest
promising results,70 technical factors that interfere with
signal reliability, especially in head trauma, may limit the
utility of the method with current technology. Reduced cor-
relation between NIRS values and other markers of oxygen-
ation, such as jugular venous oxygen saturation (SjvO2) and
PbtO2, has been reported,71-73 and normal NIRS values have
been found with complete ischemia.74 However, the develop-
ment of newer technologies may reduce these technical limi-
tations. NIRS may be used as a noninvasive tool for
measuring physiologic phenomena, such as autoregulation
and the effect of ICP.75,76
Microdialysis
Very few cerebral microdialysis studies have been performed
in children,77-79 but these have highlighted possible patho-
physiologic differences between adult and pediatric TBI. In
general, the principles are expected to be largely similar in
children, as would be the limitations. The expense, infrastruc-
ture required, intermittent sampling technique, and lack of an
“early warning system” have limited the more widespread use
of microdialysis in pediatric ICUs.
Jugular Bulb Venous
Oxygen Saturation
There are fewer reports of SjvO2 monitoring in children com-
pared with adults.80,81 The technique in children has been
described.82 Perez et al.81 found that poor outcome in pediatric
TBI was associated with episodes of SjvO2 less than 55% and
pathologic arteriovenous difference in lactate. The technical
limitations of SjvO2 have not been specifically described in
children, but are probably similar to those in adults.83,84
Transcranial Doppler (TCD) monitoring has been used as a
noninvasive monitoring tool in pediatric TBI, hydrocephalus,
and meningitis.85-87 Insonation of the basal vessels is usually
easier in children because of the thinner temporal bone. The
optimal depth of insonation of the middle cerebral artery
(MCA), is shorter in children than in adults and depends on
age and other physical factors such as scalp swelling after
trauma. Normal values for TCD-derived flow velocity (FV) in
the MCA (FVMCA) depend on age. FVMCA is approximately
24 cm/second in newborns, increases with age to peak at
97 cm per second at 6 to 9 years old, and then decreases to adult
values of about 50 cm per second.39 The reported age for peak
mean FVMCA in children is slightly higher than the age reported
for peak CBF.28,29 There also appear to be slight gender differ-
ences in anterior and posterior circulation FV in children.88
The TCD-derived pulsatility index (PI) may be useful for
the noninvasive estimation of ICP and CPP in adults.89,90
Although this is not as well defined in children, the PI may be
associated with outcome after pediatric TBI91,92 and may be
useful as a guide in treating hydrocephalus.93 The Lindegaard
index may be helpful in monitoring vasospasm in adult TBI.94
Although vasospasm appears to be uncommon after pediatric
severe TBI,87,95 monitoring for vasospasm may be useful for
subarachnoid hemorrhage or meningitis.85
Pressure Autoregulation
Pressure autoregulation (PAR) may be impaired in about 40%
of children with severe TBI.39,96 Although this has implications
for clinical management, PAR is not commonly measured in
children with TBI. If PAR is impaired, maintenance of ade-
quate CBF may require increased CPP, but higher CPP may
also increase cerebral blood volume and ICP.44 Impaired PAR
is associated with poor outcome in adults and children.97,98
Testing of PAR has been well described99 and bedside methods
have been validated.100,101 TCD-based methods are the most
common bedside techniques used in children. The strength of
PAR in children may vary between the two hemispheres and
may change over time.102,103 Continuous measurement of the
strength of PAR over time104 has also been described in chil-
dren.105 The status of PAR has profound consequences for the
effect of BP changes on ICP and perfusion of the brain in
pediatric TBI.54 Ideally, the status of PAR should be known
when making decisions about what BP target is optimal in
individual patients. Important issues that still need to be clari-
fied in children include whether the autoregulation curve is
similar to that of adults39 and whether information from
testing PAR leads to therapeutic decisions that benefit the
patient.

Computed Tomography Scanning
Head CT is used for diagnosis, to estimate prognosis, to
monitor progress, and as a guide to treatment. Unlike adult
TBI, diffuse injury is more common than focal injury in chil-
dren with severe TBI106 and mass lesions requiring surgical
removal are less common. Diffuse brain swelling has been
reported to be the most common CT finding in children with
severe TBI107 and increases the likelihood of poor outcome.108
Subarachnoid hemorrhage on head CT has also been associ-
ated with poor outcome in pediatric TBI.106 In the presence of
brain swelling, often the basal cisterns on head CT are effaced
or obliterated; however, in children increased ICP can occur
in the presence of completely normal-appearing cisterns.36
There is increasing concern about the effects of head CT on
the developing brain. Careful thought should be given to chil-
dren who require multiple CT scans, including those of other
body parts and studies such as CT-perfusion or angiography,
and alternative examinations sought, if feasible, to reduce the
total radiation dose and the potential for cumulative deleteri-
ous effects.109-115,115a For example, susceptibility-weighted
imaging (SWI), a relatively novel MRI sequence may be useful
in select patients to detect hemorrhagic lesions.116 However,
its clinical usefulness in children is still to be elucidated.
Serum Biomarkers
Various serum biomarkers have been examined in adult and
pediatric TBI for prognosticating and predicting the need
for imaging. Berger et al.117 found that biomarkers (neuron-
specific enolase, myelin-basic protein and S100-B) were pre-
dictive of outcome. Specifically, the prediction of good
outcome (97%) was better than the prediction of poor
outcome (75%), and peak concentrations correlated better
with outcome than initial concentrations. However, the rela-
tive strengths of association between biomarkers and outcome
for different childhood age groups still need further clarifica-
tion,117,118 and it is unclear whether biomarker levels will influ-
ence acute management or be used primarily to predict
outcome.119
Glucose, Hemoglobin, and
Temperature Monitoring
Global trends have favored more conservative thresholds for
blood transfusion in critically ill children120; however, acute
anemia may have particular risks in TBI patients.121 Studies
examining the effect of blood transfusion on brain oxygen-
ation in adults and children65,122 demonstrate increased oxy-
genation with transfusion; however, this tends to be temporary.
Transfusion does not universally increase oxygenation, and
the prediction of the effect based on pretransfusion character-
istics is incomplete. Therefore, optimal hemoglobin concen-
trations, or transfusion triggers, in critically ill children with
neurologic pathology remain debated. It may be that the
optimal level depends on the underlying severity of brain
injury and risk of ischemia.
Glucose control is similarly controversial. Hyperglycemia is
associated with worse outcome in children with TBI,123-125 and
strict control of glucose in pediatric trauma patients has
been recommended.126 However, whether hyperglycemia is a
Section I—Background 59
modifiable factor that causes secondary injury or is merely a
marker of injury severity remains uncertain. Furthermore,
strict control of glucose may significantly reduce cerebral
glucose, increase markers of tissue injury, and worsen
outcome.127,128 No studies have examined the impact on
outcome of strict glycemic control in children with severe TBI.
Core temperature is routinely monitored in pediatric criti-
cal care, and hyperthermia should be avoided. Induced hypo-
thermia remains controversial for pediatric TBI,129,130 but is
better established as a therapy for other conditions, such as
perinatal asphyxia.131
Other
Although not strictly a monitoring modality, imaging of the
brain has evolved substantially and has become a powerful
tool for research and outcome prediction in TBI. Tools
such as diffusion-weighted imaging, susceptibility-weighted
imaging, magnetic resonance spectroscopy, and diffusion
tensor imaging have contributed to the appreciation of edema,
microhemorrhages, ischemia, and white matter injury in adult
and pediatric TBI.132 Continuous electroencephalographic
monitoring can be used to diagnose convulsive and noncon-
vulsive epileptiform activity and monitor the depth of barbi-
turate sedation. Seizures may be seen in as many as one third
of children in the critical care environment, with the great
majority of these being nonconvulsive.133 Cortical spreading
depolarization events happen frequently in adult TBI, and are
associated with metabolic disturbances and poor outcome.134
The role of these events in pediatric neurocritical care has not
yet been determined.
Conclusion
There are fewer reports that examine neuromonitoring tools
in critically ill children than in adults, so the evidence base for
making clinical decisions is small. This is compounded by the
increased complexity of changing physiology with age in chil-
dren, and so the choice of monitoring tools and appropriate
targets is arguably more challenging than in adults. It is
important to remember that the rules governing monitoring
issues in adults may not apply to children, that the nature of
the pathology may be different, and that age-related physical
and physiologic issues must be taken into account when treat-
ing children. Given the increased complexity of brain injury
pathophsyiology in children it is arguable that a multimodal
approach has even greater value than in adults.
References
1. Segui-Gomez M, MacKenzie EJ. Measuring the public health impact of
injuries. Epidemiol Rev 2003;25:3–19.
2. Arias E, Anderson RN, Kung HC, et al. Deaths: final data for 2001. Natl Vital
Stat Rep 2003;52(3):1–116.
3. Cantais E, Paut O, Giorgi R, et al. Evaluating the prognosis of multiple,
severely traumatized children in the intensive care unit. Intensive Care Med
2001;27:1511–17.
4. Mayer T, Walker ML, Johnson DG, et al. Causes of morbidity and mortality
in severe pediatric trauma. JAMA 1981;245:719–21.
5. Sharples PM, Storey A, Aynsley-Green A, et al. Avoidable factors contributing
to death of children with head injury. BMJ 1990;300:87–91.
6. Friess SH, Ralston J, Eucker SA, et al. Neurocritical care monitoring
correlates with neuropathology in a swine model of pediatric traumatic
brain injury. Neurosurgery 2011;69(5):1139–47; discussion 1147.
60 Section I—Background
7. Seshia SS, Bingham WT, Kirkham FJ, et al. Nontraumatic coma in children
and adolescents: diagnosis and management. Neurol Clin.
2011;29(4):1007–43.
8. Odetola FO, Clark SJ, Lamarand KE, et al. Intracranial pressure monitoring
in childhood meningitis with coma: A national survey of neurosurgeons in
the United States. Pediatr Crit Care Med 2011;12(6):e350–6.
9. Kamat P, Kunde S, Vos M, et al. Invasive intracranial pressure monitoring is
a useful adjunct in the management of severe hepatic encephalopathy
associated with pediatric acute liver failure. Pediatr Crit Care Med
2012;13(1):e33–8.
10. Allen BB, Hoffman CE, Traube CS, et al. Continuous brain tissue
oxygenation monitoring in the management of pediatric stroke. Neurocrit
Care. 2011;15(3):529–36.
11. Kochanek PM, Carney N, Adelson PD, et al. Guidelines for the acute medical
management of severe traumatic brain injury in infants, children, and
adolescents—second edition. Pediatr Crit Care Med 2012;13(Suppl. 1):
S1–82.
12. Ducrocq SC, Meyer PG, Orliaguet GA, et al. Epidemiology and early
predictive factors of mortality and outcome in children with traumatic
severe brain injury: experience of a French pediatric trauma center. Pediatr
Crit Care Med 2006;7:461–7.
13. Salorio CF, Slomine BS, Guerguerian AM, et al. Intensive care unit variables
and outcome after pediatric traumatic brain injury: a retrospective study of
survivors. Pediatr Crit Care Med 2008;9:47–53.
14. Simpson D, Reilly P. Pediatric coma scale. Lancet 1982;2:450.
15. Hahn YS, McLone DG. Risk factors in the outcome of children with minor
head injury. Pediatr Neurosurg 1993;19:135–42.
16. Lu J, Murray GD, Steyerberg EW, et al. Effects of Glasgow outcome scale
misclassification on traumatic brain injury clinical trials. J Neurotrauma
2008;25:641–51.
17. Lobato RD, Rivas JJ, Gomez PA, et al. Head-injured patients who talk and
deteriorate into coma. Analysis of 211 cases studied with computerized
tomography. J Neurosurg 1991;75:256–61.
18. Figaji AA, Zwane E, Fieggen AG, et al. Acute clinical grading in pediatric
severe traumatic brain injury and its association with subsequent
intracranial pressure, cerebral perfusion pressure, and brain oxygenation.
Neurosurg Focus 2008;25:E4.
19. Feickert HJ, Drommer S, Heyer R. Severe head injury in children: impact of
risk factors on outcome. J Trauma 1999;47:33–8.
20. Hackbarth RM, Rzeszutko KM, Sturm G, et al. Survival and functional
outcome in pediatric traumatic brain injury: a retrospective review and
analysis of predictive factors. Crit Care Med 2002;30:1630–5.
21. Tepas 3rd JJ, DiScala C, Ramenofsky ML, et al. Mortality and head injury:
the pediatric perspective. J Pediatr Surg 1990;25:92–5.
22. Coates BM, Vavilala MS, Mack CD, et al. Influence of definition and location
of hypotension on outcome following severe pediatric traumatic brain
injury. Crit Care Med 2005;33:2645–50.
23. Adelson PD, Bratton SL, Carney NA, et al. Guidelines for the acute medical
management of severe traumatic brain injury in infants, children, and
adolescents. Pediatr Crit Care Med 2003;4:S12–8.
24. Haque IU, Zaritsky AL. Analysis of the evidence for the lower limit of
systolic and mean arterial pressure in children. Pediatr Crit Care Med
2007;8:138–44.
25. Rosner B, Prineas RJ, Loggie JM, et al. Blood pressure nomograms for
children and adolescents, by height, sex, and age, in the United States.
J Pediatr 1993;123:871–86.
26. Bruce DA, Raphaely RC, Goldberg AI, et al. Pathophysiology, treatment
and outcome following severe head injury in children. Childs Brain
1979;5:174–91.
27. Sharples PM, Stuart AG, Matthews DS, et al. Cerebral blood flow and
metabolism in children with severe head injury. Part 1: relation to age,
Glasgow coma score, outcome, intracranial pressure, and time after injury.
J Neurol Neurosurg Psychiatry 1995;58:145–52.
28. Zwienenberg M, Muizelaar JP. Severe pediatric head injury: the role of
hyperemia revisited. J Neurotrauma 1999;16:937–43.
29. Chiron C, Raynaud C, Maziere B, et al. Changes in regional cerebral blood
flow during brain maturation in children and adolescents. J Nucl Med 1992;
33:696–703.
30. Figaji AA. Practical aspects of bedside cerebral hemodynamics monitoring in
pediatric TBI. Childs Nerv Syst 2010;26:431–9.
31. Adelson PD, Srinivas R, Chang Y, et al. Cerebrovascular response in
children following severe traumatic brain injury. Childs Nerv Syst
2011;27:1465–76.
32. Morris KP, Forsyth RJ, Parslow RC, et al. Intracranial pressure complicating
severe traumatic brain injury in children: monitoring and management.
Intensive Care Med 2006;32:1606–12.
33. Plotz FB, Kneyber M, van Heerde M, et al. Traumatic pediatric brain injury
and intracranial pressure monitoring: does it really improve outcome?
Intensive Care Med 2007;33:1675.
34. Figaji AA, Adelson PD. Does ICP monitoring in children with severe head
injuries make a difference? Am Surg 2009;75:441–2.
35. Dean NP, Boslaugh S, Adelson PD, et al. Physician agreement with
evidence-based recommendations for the treatment of severe traumatic
brain injury in children. J Neurosurg 2007;107:387–91.
36. Kouvarellis AJ, Rohlwink UK, Sood V, et al. The relationship between basal
cisterns on CT and time-linked intracranial pressure in paediatric head
injury. Childs Nerv Syst 2011;27:1139–44.
37. Jensen RL, Hahn YS, Ciro E. Risk factors of intracranial pressure monitoring
in children with fiberoptic devices: a critical review. Surg Neurol 1997;47:
16–22.
38. Shapiro K, Marmarou A. Clinical applications of the pressure-volume index
in treatment of pediatric head injuries. J Neurosurg 1982;56:819–25.
39. Udomphorn Y, Armstead WM, Vavilala MS. Cerebral blood flow and
autoregulation after pediatric traumatic brain injury. Pediatr Neurol 2008;38:
225–34.
40. Rohlwink UK, Zwane E, Fieggen AG, et al. The relationship between
intracranial pressure and brain oxygenation in children with severe
traumatic brain injury. Neurosurgery 2012;70:1220–30; discussion 1231.
41. Prabhakaran P, Reddy AT, Oakes WJ, et al. A pilot trial comparing cerebral
perfusion pressure-targeted therapy to intracranial pressure-targeted
therapy in children with severe traumatic brain injury. J Neurosurg 2004;
100:454–9.
42. Rosner MJ, Rosner SD, Johnson AH. Cerebral perfusion pressure:
management protocol and clinical results. J Neurosurg 1995;83:949–62.
43. Robertson CS, Valadka AB, Hannay HJ, et al. Prevention of secondary
ischemic insults after severe head injury. Crit Care Med 1999;27:2086–95.
44. Bouma GJ, Muizelaar JP, Bandoh K, et al. Blood pressure and intracranial
pressure-volume dynamics in severe head injury: relationship with cerebral
blood flow. J Neurosurg 1992;77:15–9.
45. Durward QJ, Del Maestro RF, Amacher AL, et al. The influence of systemic
arterial pressure and intracranial pressure on the development of cerebral
vasogenic edema. J Neurosurg 1983;59:803–9.
46. Nordstrom CH. Assessment of the optimal cerebral perfusion pressure in
head-injured patients. Anesth Analg 2005;101(1):299–300.
47. Wahlstrom MR, Olivecrona M, Koskinen LO, et al. Severe traumatic brain
injury in pediatric patients: treatment and outcome using an intracranial
pressure targeted therapy—the Lund concept. Intensive Care Med 2005;31:
832–9.
48. Andrews PJ, Citerio G. Lund therapy—pathophysiology-based therapy or
contrived over-interpretation of limited data? Intensive Care Med 2006;32:
1461–3.
49. Chambers IR, Jones PA, Lo TY, et al. Critical thresholds of intracranial
pressure and cerebral perfusion pressure related to age in paediatric head
injury. J Neurol Neurosurg Psychiatry 2006;77:234–40.
50. Chambers IR, Treadwell L, Mendelow AD. Determination of threshold levels
of cerebral perfusion pressure and intracranial pressure in severe head injury
by using receiver-operating characteristic curves: an observational study in
291 patients. J Neurosurg 2001;94:412–16.
A complete list of references for this chapter can be found online at
www.expertconsult.com.

References
1. Segui-Gomez M, MacKenzie EJ. Measuring the public health impact of
injuries. Epidemiol Rev 2003;25:3–19.
2. Arias E, Anderson RN, Kung HC, et al. Deaths: final data for 2001. Natl
Vital Stat Rep 2003;52(3):1–116.
3. Cantais E, Paut O, Giorgi R, et al. Evaluating the prognosis of multiple,
severely traumatized children in the intensive care unit. Intensive Care Med
2001;27:1511–17.
4. Mayer T, Walker ML, Johnson DG, et al. Causes of morbidity and mortality
in severe pediatric trauma. JAMA 1981;245:719–21.
5. Sharples PM, Storey A, Aynsley-Green A, et al. Avoidable factors
contributing to death of children with head injury. BMJ 1990;300:87–91.
6. Friess SH, Ralston J, Eucker SA, et al. Neurocritical care monitoring
correlates with neuropathology in a swine model of pediatric traumatic
brain injury. Neurosurgery 2011;69(5):1139–47; discussion 1147.
7. Seshia SS, Bingham WT, Kirkham FJ, et al. Nontraumatic coma in children
and adolescents: diagnosis and management. Neurol Clin.
2011;29(4):1007–43.
8. Odetola FO, Clark SJ, Lamarand KE, et al. Intracranial pressure monitoring
in childhood meningitis with coma: A national survey of neurosurgeons in
the United States. Pediatr Crit Care Med 2011;12(6):e350–6.
9. Kamat P, Kunde S, Vos M, et al. Invasive intracranial pressure monitoring is
a useful adjunct in the management of severe hepatic encephalopathy
associated with pediatric acute liver failure. Pediatr Crit Care Med
2012;13(1):e33–8.
10. Allen BB, Hoffman CE, Traube CS, et al. Continuous brain tissue
oxygenation monitoring in the management of pediatric stroke. Neurocrit
Care. 2011;15(3):529–36.
11. Kochanek PM, Carney N, Adelson PD, et al. Guidelines for the acute
medical management of severe traumatic brain injury in infants, children,
and adolescents—second edition. Pediatr Crit Care Med 2012;13(Suppl. 1):
S1–82.
12. Ducrocq SC, Meyer PG, Orliaguet GA, et al. Epidemiology and early
predictive factors of mortality and outcome in children with traumatic
severe brain injury: experience of a French pediatric trauma center. Pediatr
Crit Care Med 2006;7:461–7.
13. Salorio CF, Slomine BS, Guerguerian AM, et al. Intensive care unit variables
and outcome after pediatric traumatic brain injury: a retrospective study of
survivors. Pediatr Crit Care Med 2008;9:47–53.
14. Simpson D, Reilly P. Pediatric coma scale. Lancet 1982;2:450.
15. Hahn YS, McLone DG. Risk factors in the outcome of children with minor
head injury, Pediatr Neurosurg 1993;19:135–42.
16. Lu J, Murray GD, Steyerberg EW, et al. Effects of Glasgow outcome scale
misclassification on traumatic brain injury clinical trials. J Neurotrauma
2008;25:641–51.
17. Lobato RD, Rivas JJ, Gomez PA, et al. Head-injured patients who talk and
deteriorate into coma. Analysis of 211 cases studied with computerized
tomography. J Neurosurg 1991;75:256–61.
18. Figaji AA, Zwane E, Fieggen AG, et al. Acute clinical grading in pediatric
severe traumatic brain injury and its association with subsequent
intracranial pressure, cerebral perfusion pressure, and brain oxygenation.
Neurosurg Focus 2008;25:E4.
19. Feickert HJ, Drommer S, Heyer R. Severe head injury in children: impact of
risk factors on outcome. J Trauma 1999;47:33–8.
20. Hackbarth RM, Rzeszutko KM, Sturm G, et al. Survival and functional
outcome in pediatric traumatic brain injury: a retrospective review and
analysis of predictive factors. Crit Care Med 2002;30:1630–5.
21. Tepas 3rd JJ, DiScala C, Ramenofsky ML, et al. Mortality and head injury:
the pediatric perspective. J Pediatr Surg 1990;25:92–5.
22. Coates BM, Vavilala MS, Mack CD, et al. Influence of definition and
location of hypotension on outcome following severe pediatric traumatic
brain injury. Crit Care Med 2005;33:2645–50.
23. Adelson PD, Bratton SL, Carney NA, et al. Guidelines for the acute medical
management of severe traumatic brain injury in infants, children, and
adolescents. Pediatr Crit Care Med 2003;4:S12–8.
24. Haque IU, Zaritsky AL. Analysis of the evidence for the lower limit of
systolic and mean arterial pressure in children. Pediatr Crit Care Med
2007;8:138–44.
25. Rosner B, Prineas RJ, Loggie JM, et al. Blood pressure nomograms for
children and adolescents, by height, sex, and age, in the United States.
J Pediatr 1993;123:871–86.
26. Bruce DA, Raphaely RC, Goldberg AI, et al. Pathophysiology, treatment
and outcome following severe head injury in children. Childs Brain
1979;5:174–91.
Section I—Background 60.e1
27. Sharples PM, Stuart AG, Matthews DS, et al. Cerebral blood flow and
metabolism in children with severe head injury. Part 1: relation to age,
Glasgow coma score, outcome, intracranial pressure, and time after injury.
J Neurol Neurosurg Psychiatry 1995;58:145–52.
28. Zwienenberg M, Muizelaar JP. Severe pediatric head injury: the role of
hyperemia revisited. J Neurotrauma 1999;16:937–43.
29. Chiron C, Raynaud C, Maziere B, et al. Changes in regional cerebral blood
flow during brain maturation in children and adolescents. J Nucl Med
1992;33:696–703.
30. Figaji AA. Practical aspects of bedside cerebral hemodynamics monitoring
in pediatric TBI. Childs Nerv Syst 2010;26:431–9.
31. Adelson PD, Srinivas R, Chang Y, et al. Cerebrovascular response in
children following severe traumatic brain injury. Childs Nerv Syst
2011;27:1465–76.
32. Morris KP, Forsyth RJ, Parslow RC, et al. Intracranial pressure complicating
severe traumatic brain injury in children: monitoring and management.
Intensive Care Med 2006;32:1606–12.
33. Plotz FB, Kneyber M, van Heerde M, et al. Traumatic pediatric brain injury
and intracranial pressure monitoring: does it really improve outcome?
Intensive Care Med 2007;33:1675.
34. Figaji AA, Adelson PD. Does ICP monitoring in children with severe head
injuries make a difference? Am Surg 2009;75:441–2.
35. Dean NP, Boslaugh S, Adelson PD, et al. Physician agreement with
evidence-based recommendations for the treatment of severe traumatic
brain injury in children. J Neurosurg 2007;107:387–91.
36. Kouvarellis AJ, Rohlwink UK, Sood V, et al. The relationship between basal
cisterns on CT and time-linked intracranial pressure in paediatric head
injury. Childs Nerv Syst 2011;27:1139–44.
37. Jensen RL, Hahn YS, Ciro E. Risk factors of intracranial pressure
monitoring in children with fiberoptic devices: a critical review. Surg
Neurol 1997;47:16–22.
38. Shapiro K, Marmarou A. Clinical applications of the pressure-volume index
in treatment of pediatric head injuries. J Neurosurg 1982;56:819–25.
39. Udomphorn Y, Armstead WM, Vavilala MS. Cerebral blood flow and
autoregulation after pediatric traumatic brain injury. Pediatr Neurol
2008;38:225–34.
40. Rohlwink UK, Zwane E, Fieggen AG, et al. The relationship between
intracranial pressure and brain oxygenation in children with severe
traumatic brain injury. Neurosurgery 2012;70:1220–30; discussion
1231.
41. Prabhakaran P, Reddy AT, Oakes WJ, et al. A pilot trial comparing cerebral
perfusion pressure-targeted therapy to intracranial pressure-targeted
therapy in children with severe traumatic brain injury. J Neurosurg
2004;100:454–9.
42. Rosner MJ, Rosner SD, Johnson AH. Cerebral perfusion pressure:
management protocol and clinical results. J Neurosurg 1995;83:949–62.
43. Robertson CS, Valadka AB, Hannay HJ, et al. Prevention of secondary
ischemic insults after severe head injury. Crit Care Med 1999;27:2086–95.
44. Bouma GJ, Muizelaar JP, Bandoh K, et al. Blood pressure and intracranial
pressure-volume dynamics in severe head injury: relationship with cerebral
blood flow. J Neurosurg 1992;77:15–9.
45. Durward QJ, Del Maestro RF, Amacher AL, et al. The influence of systemic
arterial pressure and intracranial pressure on the development of cerebral
vasogenic edema. J Neurosurg 1983;59:803–9.
46. Nordstrom CH. Assessment of the optimal cerebral perfusion pressure in
head-injured patients. Anesth Analg 2005;101:(1):299–300.
47. Wahlstrom MR, Olivecrona M, Koskinen LO, et al. Severe traumatic brain
injury in pediatric patients: treatment and outcome using an intracranial
pressure targeted therapy—the Lund concept. Intensive Care Med
2005;31:832–9.
48. Andrews PJ, Citerio G. Lund therapy—pathophysiology-based therapy
or contrived over-interpretation of limited data? Intensive Care Med
2006;32:1461–3.
49. Chambers IR, Jones PA, Lo TY, et al. Critical thresholds of intracranial
pressure and cerebral perfusion pressure related to age in paediatric head
injury. J Neurol Neurosurg Psychiatry 2006;77:234–40.
50. Chambers IR, Treadwell L, Mendelow AD. Determination of threshold
levels of cerebral perfusion pressure and intracranial pressure in severe
head injury by using receiver-operating characteristic curves: an
observational study in 291 patients. J Neurosurg 2001;94:412–16.
51. Vavilala MS, Bowen A, Lam AM, et al. Blood pressure and outcome after
severe pediatric traumatic brain injury. J Trauma 2003;55:1039–44.
52. Jones PA, Andrews PJ, Easton VJ, et al. Traumatic brain injury in
childhood: intensive care time series data and outcome. Br J Neurosurg
2003;17:29–39.
60.e2 Section I—Background
53. Downard C, Hulka F, Mullins RJ, et al. Relationship of cerebral perfusion
pressure and survival in pediatric brain-injured patients. J Trauma
2000;49(4):654–8; discussion 658–9.
54. Figaji AA, Zwane E, Fieggen AG, et al. Pressure autoregulation, intracranial
pressure, and brain tissue oxygenation in children with severe traumatic
brain injury. J Neurosurg Pediatr 2009;4:420–8.
55. Figaji AA, Zwane E, Thompson C, et al. Brain tissue oxygen tension
monitoring in pediatric severe traumatic brain injury: part 2: relationship
with clinical, physiological, and treatment factors. Childs Nerv Syst
2009;25:1335–43.
56. Maloney-Wilensky E, Gracias V, Itkin A, et al. Brain tissue oxygen and
outcome after severe traumatic brain injury: a systematic review. Crit Care
Med 2009;37:2057–63.
57. Narotam PK, Burjonrappa SC, Raynor SC, et al. Cerebral oxygenation in
major pediatric trauma: its relevance to trauma severity and outcome.
J Pediatr Surg 2006;41:505–13.
58. Stiefel MF, Udoetuk JD, Storm PB, et al. Brain tissue oxygen monitoring
in pediatric patients with severe traumatic brain injury. J Neurosurg
2006;105:281–6.
59. Figaji AA, Fieggen AG, Argent AC, et al. Does adherence to treatment
targets in children with severe traumatic brain injury avoid brain hypoxia?
A brain tissue oxygenation study. Neurosurgery 2008;63(1):83–91;
discussion 91–2.
60. Figaji AA, Zwane E, Thompson C, et al. Brain tissue oxygen tension
monitoring in pediatric severe traumatic brain injury: part 2: relationship
with clinical, physiological, and treatment factors. Childs Nerv Syst
2009;25(10):1335–43.
61. Figaji AA, Zwane E, Thompson C, et al. Brain tissue oxygen tension
monitoring in pediatric severe traumatic brain injury: part 1: relationship
with outcome. Childs Nerv Syst 2009;25:1325–33.
62. Figaji AA, Zwane E, Fieggen AG, et al. Acute clinical grading in pediatric
severe traumatic brain injury and its association with subsequent
intracranial pressure, cerebral perfusion pressure, and brain oxygenation.
Neurosurg Focus 2008;25:E4.
63. Figaji AA, Zwane E, Graham Fieggen A, et al. The effect of increased
inspired fraction of oxygen on brain tissue oxygen tension in children with
severe traumatic brain injury. Neurocrit Care 2010;12:430–7.
64. Figaji AA, Kent SJ. Brain tissue oxygenation in children diagnosed with
brain death. Neurocrit Care 2010;12:56–61.
65. Figaji AA, Zwane E, Kogels M, et al. The effect of blood transfusion on
brain oxygenation in children with severe traumatic brain injury. Pediatr
Crit Care Med 2010;11:325–31.
66. Rohlwink UK, Figaji AA. Methods of monitoring brain oxygenation. Childs
Nerv Syst 2010;26:453–64.
67. Wardle SP, Yoxall CW, Weindling AM. Determinants of cerebral fractional
oxygen extraction using near infrared spectroscopy in preterm neonates.
J Cereb Blood Flow Metab 2000;20:272–9.
68. Abdul-Khaliq H, Troitzsch D, Berger F, et al. Regional transcranial oximetry
with near infrared spectroscopy (NIRS) in comparison with measuring
oxygen saturation in the jugular bulb in infants and children for
monitoring cerebral oxygenation. Biomed Tech (Berl) 2000;45:328–32.
69. Shimizu N, Gilder F, Bissonnette B, et al. Brain tissue oxygenation index
measured by near infrared spatially resolved spectroscopy agreed with
jugular bulb oxygen saturation in normal pediatric brain: a pilot study.
Childs Nerv Syst 2005;21:181–4.
70. Adelson PD, Nemoto E, Colak A, et al. The use of near infrared
spectroscopy (NIRS) in children after traumatic brain injury: a preliminary
report. Acta Neurochir Suppl 1998;71:250–4.
71. Buchner K, Meixensberger J, Dings J, et al. Near-infrared spectroscopy—not
useful to monitor cerebral oxygenation after severe brain injury. Zentralbl
Neurochir 2000;61:69–73.
72. Holzschuh M, Woertgen C, Metz C, et al. Dynamic changes of cerebral
oxygenation measured by brain tissue oxygen pressure and near infrared
spectroscopy. Neurol Res 1997;19:246–8.
73. Nagdyman N, Fleck T, Schubert S, et al. Comparison between cerebral
tissue oxygenation index measured by near-infrared spectroscopy and
venous jugular bulb saturation in children. Intensive Care Med 2005;31:
846–50.
74. Schwarz G, Litscher G, Kleinert R, et al. Cerebral oximetry in dead subjects.
J Neurosurg Anesthesiol 1996;8:189–93.
75. Weerakkody RA, Czosnyka M, Zweifel C, et al. Near infrared spectroscopy
as possible non-invasive monitor of slow vasogenic ICP waves. Acta
Neurochir Suppl 2012;114:181–5.
76. Zweifel C, Castellani G, Czosnyka M, et al. Noninvasive monitoring of
cerebrovascular reactivity with near infrared spectroscopy in head-injured
patients. J Neurotrauma 2010;27:1951–8.
77. Tolias C, Richards D, Bowery N, et al. Investigation of extracellular amino
acid release in children with severe head injury using microdialysis. A pilot
study. Acta Neurochir Suppl 2002;81:377–9.
78. Tolias CM, Richards DA, Bowery NG, et al. Extracellular glutamate in the
brains of children with severe head injuries: a pilot microdialysis study.
Childs Nerv Syst 2002;18:368–74.
79. Richards DA, Tolias CM, Sgouros S, et al. Extracellular glutamine to
glutamate ratio may predict outcome in the injured brain: a clinical
microdialysis study in children. Pharmacol Res 2003;48:101–9.
80. Nakamura T, Nagao S, Kawai N, et al. Significance of multimodal cerebral
monitoring under moderate therapeutic hypothermia for severe head
injury. Acta Neurochir Suppl 1998;71:85–7.
81. Perez A, Minces PG, Schnitzler EJ, et al. Jugular venous oxygen
saturation or arteriovenous difference of lactate content and outcome in
children with severe traumatic brain injury. Pediatr Crit Care Med 2003;4:
33–8.
82. Gayle MO, Frewen TC, Armstrong RF, et al. Jugular venous bulb
catheterization in infants and children. Crit Care Med 1989;17:385–8.
83. Dearden NM, Midgley S. Technical considerations in continuous jugular
venous oxygen saturation measurement. Acta Neurochir Suppl (Wien)
1993;59:91–7.
84. Kiening KL, Unterberg AW, Bardt TF, et al. Monitoring of cerebral
oxygenation in patients with severe head injuries: brain tissue PO2 versus
jugular vein oxygen saturation. J Neurosurg 1996;85:751–7.
85. Kilic T, Elmaci I, Ozek MM, et al. Utility of transcranial Doppler
ultrasonography in the diagnosis and follow-up of tuberculous
meningitis-related vasculopathy. Childs Nerv Syst 2002;18:142–6.
86. Nadvi SS, Du Trevou MD, Van Dellen JR, et al. The use of transcranial
Doppler ultrasonography as a method of assessing intracranial pressure in
hydrocephalic children. Br J Neurosurg 1994;8:573–7.
87. Mandera M, Larysz D, Wojtacha M. Changes in cerebral hemodynamics
assessed by transcranial Doppler ultrasonography in children after head
injury. Childs Nerv Syst 2002;18:124–8.
88. Vavilala MS, Kincaid MS, Muangman SL, et al. Gender differences in
cerebral blood flow velocity and autoregulation between the anterior and
posterior circulations in healthy children. Pediatr Res 2005;58:574–8.
89. Chan KH, Miller JD, Dearden NM, et al. The effect of changes in cerebral
perfusion pressure upon middle cerebral artery blood flow velocity and
jugular bulb venous oxygen saturation after severe brain injury. J Neurosurg
1992;77:55–61.
90. Bellner J, Romner B, Reinstrup P, et al. Transcranial Doppler sonography
pulsatility index (PI) reflects intracranial pressure (ICP). Surg Neurol
2004;62(1):45–51; discussion 51.
91. Meyer PG, Ducrocq S, Rackelbom T, et al. Surgical evacuation of acute
subdural hematoma improves cerebral hemodynamics in children: a
transcranial Doppler evaluation. Childs Nerv Syst 2005;21:133–7.
92. Trabold F, Meyer PG, Blanot S, et al. The prognostic value of transcranial
Doppler studies in children with moderate and severe head injury. Intensive
Care Med 2004;30:108–12.
93. Vajda Z, Buki A, Veto F, et al. Transcranial Doppler-determined pulsatility
index in the evaluation of endoscopic third ventriculostomy (preliminary
data). Acta Neurochir (Wien) 1999;141:247–50.
94. Lindegaard KF, Nornes H, Bakke SJ, et al. Cerebral vasospasm diagnosis by
means of angiography and blood velocity measurements. Acta Neurochir
(Wien) 1989;100:12–24.
95. Figaji AA, Zwane E, Fieggen AG, et al. Transcranial Doppler pulsatility
index is not a reliable indicator of intracranial pressure in children with
severe traumatic brain injury. Surg Neurol 2009;72(4):389–94.
96. Vavilala MS, Lee LA, Boddu K, et al. Cerebral autoregulation in pediatric
traumatic brain injury. Pediatr Crit Care Med 2004;5:257–63.
97. Vavilala MS, Muangman S, Tontisirin N, et al. Impaired cerebral
autoregulation and 6-month outcome in children with severe traumatic
brain injury: preliminary findings. Dev Neurosci 2006;28:348–53.
98. Panerai RB, Kerins V, Fan L, et al. Association between dynamic cerebral
autoregulation and mortality in severe head injury. Br J Neurosurg 2004;18:
471–9.
99. Panerai RB. Assessment of cerebral pressure autoregulation in humans—a
review of measurement methods. Physiol Meas 1998;19:305–38.
100. Newell DW, Aaslid R, Lam A, et al. Comparison of flow and velocity during
dynamic autoregulation testing in humans. Stroke 1994;25:793–7.
101. Steiner LA, Coles JP, Johnston AJ, et al. Assessment of cerebrovascular
autoregulation in head-injured patients: a validation study. Stroke 2003;34:
2404–9.
102. Vavilala MS, Tontisirin N, Udomphorn Y, et al. Hemispheric differences in
cerebral autoregulation in children with moderate and severe traumatic
brain injury. Neurocrit Care 2008;9(1):45–54.

103. Tontisirin N, Armstead W, Waitayawinyu P, et al. Change in cerebral
autoregulation as a function of time in children after severe traumatic
brain injury: a case series. Childs Nerv Syst 2007;23:1163–9.
104. Czosnyka M, Smielewski P, Piechnik S, et al. Continuous assessment of
cerebral autoregulation—clinical verification of the method in head injured
patients. Acta Neurochir Suppl 2000;76:483–4.
105. Brady KM, Shaffner DH, Lee JK, et al. Continuous monitoring of
cerebrovascular pressure reactivity after traumatic brain injury in children.
Pediatrics 2009;124:e1205–12.
106. Pillai S, Praharaj SS, Mohanty A, et al. Prognostic factors in children with
severe diffuse brain injuries: a study of 74 patients. Pediatr Neurosurg
2001;34:98–103.
107. Levin HS, Aldrich EF, Saydjari C, et al. Severe head injury in children:
experience of the Traumatic Coma Data Bank. Neurosurgery 1992;31:435,
443.
108. Aldrich EF, Eisenberg HM, Saydjari C, et al. Diffuse brain swelling in
severely head-injured children. A report from the NIH Traumatic Coma
Data Bank. J Neurosurg 1992;76:450–4.
109. Dougeni E, Faulkner K, Panayiotakis G. A review of patient dose and
optimisation methods in adult and paediatric CT scanning. Eur J Radiol
2012;81(4):e665–83.
110. Guillerman RP. Newer CT applications and their alternatives: what is
appropriate in children? Pediatr Radiol 2011;41(Suppl 2):534–48.
111. Mahesh M. Advances in CT technology and application to pediatric
imaging. Pediatr Radiol 2011;41(Suppl 2):493–7.
112. Mueller DL, Hatab M, Al-Senan R, Cohn SM, et al. Pediatric radiation
exposure during the initial evaluation for blunt trauma. J Trauma
2011;70(3):724–31.
113. Kuppermann N, Holmes JF, Dayan PS, et al, Pediatric Emergency Care
Applied Research Network (PECARN). Identification of children at very
low risk of clinically important brain injuries after head trauma: a
prospective cohort study. Lancet 2009;374(9696):1160–70.
114. Brenner DJ, Hall EJ. Computed tomography—an increasing source of
radiation exposure. N Engl J Med 2007;357(22):2277–84.
115. Scaife ER, Rollins MD. Managing radiation risk in the evaluation of the
pediatric trauma patient. Semin Pediatr Surg 2010;19(4):252–6.
115a.  Nievelstein RA, van Dam IM, van der Molen AJ. Multidetector CT in
children: current concepts and dose reduction strategies. Pediatr Radiol
2010;40(8):1324–44.
116. Beauchamp MH, Ditchfield M, Babl FE, et al. Detecting traumatic brain
lesions in children: CT versus MRI versus susceptibility weighted imaging
(SWI). J Neurotrauma 2011;28(6):915–27.
117. Berger RP, Beers SR, Richichi R, et al. Serum biomarker concentrations and
outcome after pediatric traumatic brain injury. J Neurotrauma 2007;24:
1793–801.
Section I—Background 60.e3
118. Shore PM, Berger RP, Varma S, et al. Cerebrospinal fluid biomarkers versus
glasgow coma scale and glasgow outcome scale in pediatric traumatic brain
injury: the role of young age and inflicted injury. J Neurotrauma 2007;24:
75–86.
119. Sandler SJ, Figaji AA, Adelson PD. Clinical applications of biomarkers in
pediatric traumatic brain injury. Childs Nerv Syst 2010;26:205–13.
120. Lacroix J, Hebert PC, Hutchison JS, et al. Transfusion strategies for patients
in pediatric intensive care units. N Engl J Med 2007;356:1609–19.
121. Hare GM, Mazer CD, Hutchison JS, et al. Severe hemodilutional anemia
increases cerebral tissue injury following acute neurotrauma. J Appl Physiol
2007;103:1021–9.
122. Smith MJ, Stiefel MF, Magge S, et al. Packed red blood cell transfusion
increases local cerebral oxygenation. Crit Care Med 2005;33:1104–8.
123. Michaud LJ, Rivara FP, Longstreth Jr WT, et al. Elevated initial blood
glucose levels and poor outcome following severe brain injuries in children.
J Trauma 1991;31:1356–62.
124. Cochran A, Scaife ER, Hansen KW, et al. Hyperglycemia and outcomes
from pediatric traumatic brain injury. J Trauma 2003;55:1035–8.
125. Smith RL, Lin JC, Adelson PD, et al. Relationship between hyperglycemia
and outcome in children with severe traumatic brain injury. Pediatr Crit
Care Med 2012;13:85–91.
126. Tuggle DW, Kuhn MA, Jones SK, et al. Hyperglycemia and infections in
pediatric trauma patients. Am Surg 2008;74:195–8.
127. Strong AJ, Boutelle MG, Vespa PM, et al. Treatment of critical care patients
with substantial acute ischemic or traumatic brain injury. Crit Care Med
2005;33(9):2147–9.
128. Vespa P, Boonyaputthikul R, McArthur DL, et al. Intensive insulin therapy
reduces microdialysis glucose values without altering glucose utilization or
improving the lactate/pyruvate ratio after traumatic brain injury. Crit Care
Med 2006;34:850–6.
129. Hutchison JS, Ward RE, Lacroix J, et al. Hypothermia therapy after
traumatic brain injury in children. N Engl J Med 2008;358:2447–56.
130. Kochanek PM, Bell MJ, Adelson P. Hypothermia therapy after traumatic
brain injury in children. N Engl J Med 2008;359:1179.
131. Jacobs S, Hunt R, Tarnow-Mordi W, et al. Cooling for newborns with
hypoxic ischaemic encephalopathy. Cochrane Database Syst Rev 2007;(4):
CD003311.
132. Friess SH, Kilbaugh TJ, Huh JW. Advanced neuromonitoring and imaging
in pediatric traumatic brain injury. Crit Care Res Pract 2012;361310.
133. McCoy B, Sharma R, Ochi A, et al. Predictors of nonconvulsive seizures
among critically ill children. Epilepsia 2011;52(11):1973–8.
134. Hartings JA, Bullock MR, Okonkwo DO, et al. Spreading depolarisations
and outcome after traumatic brain injury: a prospective observational
study. Lancet Neurol 2011;10(12):1058–64.

Bioethics and the Family
Patricia D. Scripko and David M. Greer
“Medical technology has effectively created a twilight zone
of suspended animation where death commences while life,
in some form, continues.”
—Justice William Brennan, 1990, regarding the case of
Nancy Cruzan v. The State of Missouri
Introduction
Medicine’s purpose is to improve and lengthen life. When a
person dies, therefore, medicine, in a sense, has failed. These
simple truths lead to the far more complex and contentious
question of when that failure has occurred. Philosophers,
doctors, and lawyers have debated the question of how to
define death for centuries. The answer has impli­cations for
political and medical issues such as organ donation.
In recent times, the issue has been complicated further
by steady progress in medical technology. Some of this
technology directly affects clinical actions to prevent and
diagnose death by improving the assessment of it, inhibiting
its progress, or altering the understanding of death alto-
gether. Technologies that aid the ability to reverse certain
pathologic processes and thereby help avoid raising the
white flag to death, also pose the related question: Is use
of these technologies in a particular patient futile? Futility
is a highly debated concept that concerns irreversible states
with depreciated quality or an inevitable progression to the
end of life. The question of futility is both unavoidable,
in that it affects decisions in critical care, and ethically
complicated, because it raises questions of patient auton-
omy in end-of-life decision making and proper resource
allocation.
Neuromonitoring encompasses a specific set of technologic
tools that may have an effect on current clinical protocols for
patients with critical neurologic illnesses. It is worth consider-
ing how the potential of these tools may be realized, and what
ethical implications they may have in the intensive care
unit (ICU). This chapter attempts to integrate medical
and legal concepts into a discussion on the ethics of neuro-
monitoring in the ICU, specifically focusing on its role in cases
of perceived medical or neurologic futility. We begin by
defining human death, for to understand what is futile,
one must understand what defines irreversible versus revers-
ible neurologically. Table 8.1 lists definitions of states of
© Copyright 2013 Elsevier Inc. All rights reserved.
Chapter
8  
I
consciousness and of death that are useful in this discussion.
The text continues with the utility of neuromonitoring, spe-
cifically in preventing futile care and its limitations. Finally,
ethical dilemmas in which families are engaged in conversa-
tions about the concepts of neuromonitoring, death, and futil-
ity are discussed.
Brain Death: A Definition
of Human Death
History
Although the terminology is misleading, brain death in the
United States and most countries worldwide legally is equated
to death of the person, not simply the death of the brain in an
otherwise living person. The ambiguous nomenclature is an
artifact of historical changes to clinically identify and define
death (Table 8.2). Before the 19th century, cessation of
breathing, an easily observable function, equated human
death. With the invention of the stethoscope in the 1800s,
cessation of an audible heartbeat became the more consid-
ered definition. Once mechanical ventilation and resuscita-
tion became routine in critical care ICUs in the 1950s and
1960s1,2 and so allowed for continued ventilation and circula-
tory support in patients in the absence of neurologic activ-
ity,* brain death arose as an alternative definition for death in
patients after catastrophe and with absent neurologic func-
tion.1,3 The Harvard Commission clarified brain death crite-
ria in 1968;4 although altered slightly on multiple occasions
since (see reference 1 for review),5-10 the criteria typically
include coma, or total unawareness of external and internal
stimuli; cranial areflexia; no spontaneous breathing (apnea);
and a flat electroencephalogram (EEG), all observed in the
absence of alternative explanations for the state, including
intoxication, metabolic disorders, or hypothermia.11 Legal
definitions depend on medical criteria, with vague terminol-
ogy that allows room for these criteria to change and still
coincide with the law. For example, as written by the
American Bar Association (ABA) in 1975, “For all legal pur-
poses, a human body with irreversible cessation of total brain
function … shall be considered dead.”12 Specifics on what
*Albeit in a disintegrated fashion, according to the 1981 President’s Commission Report.
61
62 Section I—Background

Table 8.1  Relevant Terms to Discuss Human Life and Death

Term Definition References
Locked-in syndrome Caused by a specific lesion in the pons. This syndrome consists of anarthria and 41
quadriplegia with preservation of cognition and conscious awareness. Patients typically
can respond only by blinking or with vertical eye movements. This syndrome must be
distinguished from states of impaired consciousness (i.e., MCS, PVS, coma).
Minimally conscious Identified by partial consciousness; sleep-wake cycles are present and the patient is able 62
state (MCS) to localize both noxious stimuli and location of sound origin. Communication consists
of intelligible verbalization and/or contingent vocalization.
Persistent vegetative For at least 1 month no conscious awareness or communicative ability. Sleep-wake cycles, 110
state (PVS) spontaneous, nonpurposeful movement, and stimulated arousal are present.
Permanent For at least 1 year after traumatic brain injury, or 3 months following ischemic injury (a), (a) 51
vegetative state the patient has no conscious awareness or communicative ability. Sleep-wake cycles (b) 110
spontaneous, nonpurposeful movements and stimulated arousal are present (b).
Coma No conscious awareness, no communicative ability, and no sleep-wake cycles, with an 41
inability to be aroused, even by noxious stimuli
Brain death Equated with human death, but complicated by differing specifics in its definition, which 111
depends on certain clinical diagnostic criteria
Irreversible Two connotations: a strong one, which means that the function can never, not even in 16
the future, be restored, and a weak one that means given the present tools and
situation, function cannot be restored.
Permanent “Lasting or intended to last or remain unchanged indefinitely” 15
“Lasting or continuing without interruption”
Human A being with specific traits identified as necessary and sufficient to declare one
“human.” These traits span from human deoxyribonucleic acid (DNA) to conscious
awareness, and are controversial.
Person Often associated with traits specific to an individual human, but, in this paper, is equated
with human.
Higher brain “Capacity to think, to perceive and to respond” (a) (a) 13
Sentience: “able to perceive or feel” (b) (b) 15
Lower brain The capacity to “integrate bodily functions” 13, 112
Whole brain Both higher and lower brain function 14, 32

constitutes “irreversible,” “total,” and “function” were not pro-
vided; these are discussed in the following text.
A Closer Look at the Definition: “Total”
and “Irreversible”
Brain death carries a legal definition in accordance with whole
brain theorists, or those persons who believe that the irrevers-
ible cessation of both higher and lower brain function are
individually necessary and together sufficient to declare
someone a deceased human. Higher and lower brain functions
are defined, respectively, as the “capacity to think, to perceive
and to respond” (cortical function) and the capacity to “inte-
grate bodily functions” (brainstem function).13 Several authors
support “whole brain” death criteria and claim its conservative
nature is necessary to protect against misdiagnosis of a person
as brain dead who actually is transitioning to an improved
neurologic state or has the potential for such a change. To
fulfill these criteria, destruction of neuronal tissue must be
so widespread that there is no doubt the person’s state is
irreversible.14
Irreversible is itself an inconclusive term, with both strong
and weak interpretations,15 as noted by Bernat.14 According to
Cole, under the stronger definition, irreversible means that
something will not be recovered now and cannot ever be
recovered later. The weaker definition refers to presently unre-
coverable objects or states that may or may not be recoverable
in the future (with better technology).16 Aulisio et al.17 sug-
gested an alternative weak interpretation of the word: an
inability to restore without intervention. These weaker defini-
tions are in line with the concept of the contextual depen-
dency of death that becomes very relevant in the neurocritical
care environment. For example, Ropper recognized that death
depends on technologic opportunities to preserve human life
that are available at a given time.18 This means that so long as
there is no machine at the present to substitute for, or restore
higher and lower brain function, then brain death is human
death. A historical example of Ropper’s view is evident in the
impact of mechanical ventilation and resuscitation that elimi-
nated cardiac death as brain death (see Table 8.2). Further
expanding the contextual dependency of irreversible, under
many state statutes, such as the often-cited Capron and Kass
model,19 physicians are bestowed the power to define and
identify “irreversible” and “relevant” (brain) functions.†20
These statutes call for diagnostic decisions to be based on best
†
The Uniform Determination of Death Act also reads that “determination of death must
be made in accordance with accepted medical standards.”
 Section I—Background 63

Table 8.2  Important Events in Defining Human Life and Death By Chronologic Order
Year Event References
Pre-1800s Death assessed by cessation of breathing, a readily observable event 3
1800s Stethoscope is invented, leading physicians to identify death by the now-observable cessation of 3
beating of the heart (“cardiac death”)
1950s Mechanical ventilation and resuscitation are staples in the intensive care unit (ICU), leading physicians 1
to shift to a neurologic base for death criteria (“brain death”)
1968 Harvard criteria are written and define brain death as (1) deep coma/no withdrawal to painful 4
stimuli; (2) cranial and spinal arreflexia; (3) apnea when the ventilator is disconnected for 3
minutes; (4) a flat electroencephalogram (EEG); (5) exclusion of hypothermia and drugs as a cause
for the findings; and (6) repeat examination in 24 hours.
1969 Luis Kutner conceived advance directives as a practical concept. He proposed legislature in 1968. 96
1971 Minnesota criteria are published. Criteria were similar to the Harvard criteria, except that a requisite 5
flat EEG was omitted, the repeat examination was to occur at 12 hours, and an “irrefutable
intracranial lesion” was identified in the patient.
1973 The case of Roe v. Wade ruled in favor of abortion before the fetus is “viable,” at which point the 108
fetus is granted legal rights as a person, a living human. This had a great impact on how society
viewed human life.
1973 The case of Edelin: A chief obstetrics resident performed a late second-term abortion at the request 57
of a teenage girl and her mother. A 2-year trial culminated in Edelin’s conviction of criminal
negligence (but notably, not murder or manslaughter) for not attempting to save a human life (the
human fetus). His publicized conviction was given a few weeks before the Quinlan case began.
1975 The case of Karen Quinlan: A 21-year-old in New Jersey went into a coma, possibly induced by 57
barbiturates. Family asked to withdraw mechanical ventilation, but the Catholic hospital, St Clare’s,
would not. The New Jersey Supreme Court voted in favor of the parents’ wish, and 10 years later,
Karen died. The American Medical Association (AMA) and the Catholic church at this time did not
recognize a difference between actively killing and withdrawing life support. This case is
monumental about decisions to dissociate life maintained by machines with life maintained by the
person alone.
1975 Since there were several state statutes that defined brain death, the American Bar Association (ABA) 12
announced in a position statement that: “for all legal purposes, a human body with irreversible
cessation of total brain function…shall be considered dead.”
1976 Barry Keene got the first advance directive legislation passed and California became the first state to 109
have such legislation (see “Natural Death Act”).
1976 United Kingdom code eliminated the repeat examination as necessary for brain death, and required a 6, 7
specific CO2 level during an apnea trial rather than time to determine that the apnea test did not
elicit medullary-driven respirations in the patient.
1977 U.S. Collaborative Study eliminated the apnea test and reintroduced a flat EEG and the repeat 8
examination, now at 30 to 60 minutes.
1980 Uniform Determination of Death Act is approved for the United States. It holds that irreversible 20
cessation of all functioning of the brain, of whole brain death, is human death, as is irreversible
cessation of circulatory and respiratory function.
1981 U.S. President’s Commission (a) supported the Uniform Determination of Death Act (b) and (a) 9
reinstituted the apnea test and a repeat examination. Also only required cerebral blood flow (b) 20
assessment if necessary to determine brain death
1983 The case of Elizabeth Bouvier: A 25-year-old paraplegic who also had little function in her arms, Ms. 57
Bouvier sought suicide by self-starvation and became a media-popularized figure of a daunting
struggle with life and wish for a right to die. She was force-fed by a court order, which she fought.
At the end of the fight, the court sympathized with Ms. Bouvier and made the notable statement:
“It is incongruous, if not monstrous for medical practitioners to assert their right to preserve a life
that someone else must…endure. No criminal or civil liability attaches to honoring a competent,
informed patient’s refusal for medical service.”
1990 AMA supports withdrawal of care for patients in Persistent Vegetative State (PVS). 58
1988 American Academy of Neurology (AAN) position on withdrawal of care in PVS cases, specifically noted 30
that physicians, unless an advance directive requests otherwise, have no obligation to administer
medical treatment, including artificial nutrition and hydration, if they cannot improve the patient’s
neurologic state.
1989 The case of Carrie Coons: An 86-year-old in New York regained consciousness after being in a 35, Ch 1
supposed state of irreversible unconsciousness (PVS). She “awoke” just before a court-ordered
removal of her feeding tube took place. This laid doubt on the accuracy of the PVS diagnosis.

Continued
64 Section I—Background

Table 8.2  Important Events in Defining Human Life and Death By Chronologic Order—cont’d
Year Event References
1989 The case of Nancy Cruzan: The first “right to death” case heard by the U.S. Supreme Court. A patient 95
in PVS, Cruzan’s family asked for her feeding tube to be removed but were voted down by the
Missouri Supreme Court based on the role of the “parens patria” of the state or, the guardian of
the incompetent, which must err on the side of life if it must err. The U.S. Supreme Court decided
that there was no evidence that Ms. Cruzan would have wanted to endure what the state of
Missouri was defining for her as her “life.”
1991 The state of New Jersey enacted a law that provided citizens with a religious exemption from brain 44
death determination.
1995 AAN criteria published that eliminated a flat EEG as necessary to declare brain death, and added 10
strict testing details for the apnea test, such as delivery of 100% oxygen to avoid harm from the
test itself.
1990-2005 The case of Terri Schiavo: After suffering from a cardiac arrest, Ms. Schiavo was comatose, 49
transitioning to PVS later. Her husband, based on his understanding of Ms Schiavo’s wishes, fought
against her parents to have her feeding tube removed. The court voted in favor of removing the
tube. The case reignited questions over what is sufficient to declare a being a living human, and
how proxies should be identified for the neurologically incompetent.

hypotheses, and recognize that absolute facts are not a reality.
Overall, the subjectivity of irreversible to time, technology,
and physicians’ best hypotheses leaves room for doubt, debate,
and fear of being incorrect.
A Closer Look at the Definition: “Function”
In the ABA’s statement that “for all legal purposes, a human
body with irreversible cessation of total brain function…shall
be considered dead,”12 it is brain function rather than the
presence and viability of the brain or its cellular components
that is cited as the necessary criterion for human life. This is
evident in medical criteria, as well. What is not obvious in
this statement, but is in others and in medical brain-death
criteria, is that this loss of brain function refers to a loss of
clinical function, not cellular function. Certain statutes, such
as that modeled by Capron and Kass, speak of “relevant” brain
function,19 making this point clear. Viable, functional cells no
doubt persist in cases of brain death, but the integration of
these cells’ functions that confers the “relevant” brain activity
sufficient to declare the patient a living human, does not.21
Bedside assessments can distinguish cellular from clinical
function. In contrast to the use of in vitro analytical tools and
EEG to determine the viability and activity of cells, clinical
function is assessed by such things as pupil reactivity, other
brainstem reflexes, medullary response in the apnea test, and
withdrawal from painful stimuli. Distinctions between clinical
function and neuronal function were made by the President’s
Commission in 1981,9 because it was recognized that clinical
functions are observable demonstrations of coordinated brain
activity that may irreversibly be lost even when particular areas
of the brain remain viable. This report called for all U.S. juris-
dictions to accept the Uniform Determination of Death Act,
which equated irreversible cessation of all clinical function of
the brain with brain death.20 Consistent with this the American
Academy of Neurology (AAN) criteria in 1995 eliminated a
flat EEG, a measure of cellular function, as necessary to declare
brain death.10 In one review, the Quality Standards Subcom-
mittee of the AAN found no published reports of recovery in
neurologic function in adults who fulfilled the diagnosis of
brain death outlined by the 1995 AAN criteria.11
The distinctions made between cellular and clinical func-
tion in brain death criteria point to an important fact some
still dispute: being alive is different from being a living human.
To elaborate, persons may die, but their bodies and the discon-
nected parts of their brain may still live, metabolize, grow, and
even carry out previously initiated human reproduction, as in
the case of brain-dead mothers.22,23 Bartlett and Youngner
note that “life may go on after a person has died … [but death
of a person] is a question of when we no longer [are] … a
living human being… the death of the organism [may] outlive
the death of a person.”24 This, along with insufficient evalua-
tion of neurologic status25 has led to some confusion made
obvious by Alan Shewmon’s declarations of “evidence” (i.e.,
organ survival) of human life (rather than just life) persisting
when brain-death criteria are “met.”26,27
Through this chapter, book, and in day-to-day clinical prac-
tice, the reader is asked to keep in mind the difference between
cellular and integrated, clinical function. Sometimes neuro-
monitoring provides output about one, the other, or both, and
thus must be appropriately weighted when considering its
value in providing insight about the prognosis of maintaining
meaningful human life.
Loosening Brain Death Criteria to Include
Other Disorders of Consciousness: Issues in
Diagnostics and Prognostics
Some clinical conditions with no or significantly impaired
higher brain function, or consciousness, do not fulfill the
Harvard and similar criteria for brain death, but often are
argued as equivalent to human death or to irreversibly dying
(see reference 28 for review).28 One such condition is the per-
manent or persistent vegetative state (PVS), in which higher
cortical functions (including consciousness) are absent and
cannot be restored.‡29 Brainstem function remains intact. In
1988, the AAN,30 echoed by the American Medical Association
‡
Posner suggests that no restoration is possible in the case of permanent (>1 year) vegeta-
tive state, but restoration may occur when the vegetative state is persistent (>1 month
<1 year). Some regard the use of “permanent” and “persistent” as confusing, and suggest
these adjectives be replaced by a physician’s statement on the prognosis of the vegetative
state.

(AMA) in 1990,31 announced approval of withdrawing care
from patients who met criteria for PVS, but did not specifi-
cally endorse equating PVS to brain death. Hesitation to
extend the clinical definition of brain death to include PVS
arises from one of two origins: (1) a belief that consciousness
is not necessary for human life to persist, declaring brainstem
function, with its role in physiologic homeostasis, as necessary
(and possibly sufficient) for human life; and (2) a fear that
such an extension would increase the number of false-positive
brain-dead diagnoses. The latter is the focus here because the
concept of consciousness being necessary for human life is
discussed in detail elsewhere.14,24,32
A fear of false positives is understandable given media-
popularized stories of “miracles” in the ICU; a patient
declared brain dead awakens to life. This leaves the patient’s
loved ones distrustful of medical diagnostic criteria for life,
and elicits inquiries from readers across the nation as to
whether their own deceased loved ones were truly, irreversibly
dead. Unfortunately, no media story begins with “this is rarely
the case, and is likely due to an error in diagnosis rather than
a miracle, but … .” People recall the cases of individuals like
Carrie Coons, who in 1989 regained consciousness before a
court-ordered act to remove her feeding tube took place.33
Similar stories are present in academic literature. In 2006,
Voss et al. wrote of a 39-year old male shown to be cortically
dead, but not dead by whole brain or brainstem criteria, who
transitioned to the minimally conscious state (MCS).34 This
may suggest that rather than improved accuracy to diagnose
cortical death, conservative criteria that guarantee the pres-
ence of widespread, irreversible destruction, such as the
current whole brain death criteria,14 are necessary because
patients in MCS may recover.35 In addition, studies using
advanced magnetic resonance imaging (MRI) techniques
including functional MRI (fMRI) have reframed questions
about consciousness and coma and the capacity of recovery
in severe brain injury.36-38
Conservative criteria may only mask the true problem;
diagnostic accuracy needs improvement. This problem was
addressed by Andrews et al.39 who reported a misdiagnosis of
40% of their 40 patients who were diagnosed as being in PVS
(they were actually in MCS). Patient blindness or severe motor
disability or insufficient patient observation, confusion with
terminology used, and evaluation by inexperienced physicians
were the most common reasons for the misdiagnosis. These
issues become relevant in the neurocritical care unit (NCCU)
where consciousness and neurologic function often may be
altered by therapies designed to correct an abnormality
detected by a monitor. Inexperience with the neuromonitor
can lead to potentially serious errors if something such as the
absence of intracranial blood flow is confused with simply
decreased blood flow.40 In addition, it is unlikely that informa-
tion provided by a monitor presently used in the NCCU, even
if that information is consistent with current clinical criteria
for brain death, will replace that criteria because monitors for
the most part assess one aspect of brain function. The infor-
mation from a monitor may be useful to decide when a
“brain-death” examination is necessary.
If death is considered to be an event, not a process,32
neuromonitoring may lend insight to the intricacies of dying
and the process that precedes the event of death.§ This could
§
Note that some view death itself as a process, rather than an event.
Section I—Background 65
increase prognostic accuracy. It also may improve diagnostic
accuracy and serve as a confirmatory test, which is deemed
necessary by some neurologists.14 Specifically, neuromonitor-
ing could identify events, such as cessation of cerebral blood
flow, that are indicative of severe brain dysfunction or hernia-
tion and not compatible with life41 and other means of
irreversible neuronal destruction. Furthermore, because neu-
romonitoring can improve diagnostic and prognostic abili-
ties, it may simplify critical care ethics by relieving the
physician of the burden of doubt. As prognostic and diagnos-
tic tools are improved, however, caution is needed not to
veer too far from conservative criteria that add extra protec-
tion against false positives of brain death. For the moment
the concept of imminent brain death is based on clinical cri-
teria including: (1) mechanically ventilated deeply comatose
patient, admitted to an ICU, with (2) irreversible catastrophic
brain damage of known origin, and (3) a Glasgow Coma
Scale score of 3 and the progressive absence of at least three
of six brainstem reflexes or a Full Outline of UnResponsive-
ness (FOUR) score of Eyes (E), Motor (M), Brainstem (B),
Respiration (R).42 This definition may be useful in identifying
potential organ donors.
Futility
The end of human life has numerous conceptual and clinical
definitions, but these definitions do not clarify the irreversible
points that lead to this inevitable end. The theoretical concept
of futility seeks to identify these points, but there is no con-
sensus on a practical delineation of states or measures that
strictly define futility. Differences in these practical definitions
often depend on the religious beliefs and cultural practices
that assign particular restrictions, definitions, and priorities to
life, death, and medical intervention.43 Governing bodies in
the United States tend to be sensitive to these differences. In
1991 the state of New Jersey enacted a law that provided citi-
zens with a religious exemption from brain death determina-
tion.44 In these cases only cardiopulmonary arrest is an
acceptable criterion for death. Similarly, New York hospitals
must have a protocol that allows for a “reasonable accommo-
dation [to] religious or moral objections” to neurologically
determined death, although, unlike in New Jersey, patients are
not ubiquitously given power to veto brain-death determina-
tions.45 Although such culturally sensitive policies protect citi-
zens’ First Amendment right to free exercise of religion
(applicable to state law through the due process clause of the
14th constitutional amendment),46,47 they limit physicians’
ability to avoid futile measures.
In medicine, futility is most simply understood as any
measure that will not sufficiently restore quality or quantity
of life-years. ||48 The AAN set forth three key considerations
for physicians when treatment may be considered futile (spe-
cifically in the setting of PVS). They are as follows:
1. A patient’s right to self-determination is central to the
medical, ethical, and legal principles relevant to medical
treatment decisions.
||
It differs from rationing in that in determining whether a treatment is futile, the physi-
cian asks whether it will achieve a beneficial end, rather than if it is worth using a treat-
ment knowing its likely result weighed against its financial and resource consumption
(i.e., hospital beds) costs.
66 Section I—Background
2. A physician also must attempt to promote the patient’s
well-being, either by relieving suffering or addressing or
reversing a pathologic process. Where medical treatment
fails to promote a patient’s well-being, there is no longer
an ethical obligation to provide it.
3. Treatments that provide no benefit to the patient or the
family may be discontinued. Medical treatment, including
the medical provision of artificial nutrition and hydration
(ANH), provides no benefit to patients in a persistent veg-
etative state, once the diagnosis has been established to a
high degree of medical certainty.30
The second consideration is most important; physicians who
cannot reduce suffering or stop a pathologic process are freed
of an ethical “obligation” to treat the underlying pathologic
process (symptomatic or comfort-directed treatment, however,
is still appropriate)30 because such treatment would be futile.
Physicians carry the burden to determine whether feeding or
other means of support actually prolong death and not life
(i.e., are futile).49 The only exception to this “rule” occurs if a
patient’s advance directive, or known previously communi-
cated opinion, indicates a wish for continued treatment in the
setting of inevitable death (futile treatment). In this case, the
patient’s wish must be upheld to protect his or her autonomy,
and here the importance of the physician-patient relationship
and physician honesty must be emphasized.50 The AMA
echoed these positions a year after the AAN voiced them.31
Because brain death defines an irreversible state given the
technology available, no current medical intervention can
“improve” a brain-dead person’s well-being. Thus all medical
measures used in a brain-dead patient, including neuromoni-
toring, are futile. Similarly, in PVS patients, it is assumed that
the patient will not regain consciousness, and so treatment is
futile.¶39,51 The war against futility often drives policy and posi-
tion statements. This has been true for some time but contin-
ues to stir controversy. An early example of academia that
convened to combat futility was when the Harvard Commis-
sion put forth their set of brain dead criteria.4 Other examples
include the AAN and AMA statements that support with-
drawal of care when futile,30,31 all of which arose from intents
to avoid futile measures in PVS patients.
Several recent legislative bills, antagonistic to efforts to
minimize futility, were drafted and introduced because of the
Terri Schiavo case.49,52 These bills aim to reverse the policy that
favors withdrawal of artificial nutrition and hydration and
other treatments when a physician believes medical measures
are futile, and no advance directive dictates otherwise.53 This
would transfer decisive power from physicians and families
who have insight to the patient’s wishes to a detached govern-
ing body, and would set the default to administering poten-
tially futile treatments. The AAN does not support this, as
evidenced by its 1988 position statement,30 and its public
opposition in 2005 to “all state and federal legislation that
would presume to prescribe the patient’s preferences for arti-
ficial hydration and nutrition in situations where the patient
lacks an advance directive or living will.”31 The AMA adopted
this latter position to complement their policies that limit
¶
It is possible that the rare reports of patients diagnosed as in PVS, who later recovered
consciousness outside of the probable timeframes for recovery of 3 months (nontrau-
matic) and 12 months (traumatic) originally may have been in MCS, and simply were
misdiagnosed.
legal jurisdiction in end-of-life decision making, and aim to
avoid futile treatment.54 Although the legal system does not
entirely coincide with these position statements and leaves
loopholes for vetoing physicians’ and families’ decisions to
withdraw care if clear advance directives or unified proxy
support for withdrawal are not available, historically, courts
have voted in favor of such withdrawal, so long as clear oppos-
ing directives also are not available. This is consistent with
Americans’ constitutional First Amendment rights, which
grant a coveted luxury of society: autonomy, even when
autonomous wishes are known only from loved ones and sur-
rogates (see reference 55 for discussion).55 The Cruzan case
was an important national precedent in establishing this right
and is summarized in Table 8.2.
Futility can be considered in relation to autonomous indi-
viduals, but it also has important implications for societal
good. What is futile for one may be valuable to others. For
example, when life-sustaining ICU measures are considered
moral practices, some futile interventions may offer ritualistic
benefit to patients, families, and ICU staff and so help facili-
tate the process of dying.56 An individual’s outcome may not
be improved by monitoring in the NCCU, but monitoring
may elucidate pathologic processes that underlie inevitable
death or a highly morbid state, and so could lead to improved
treatment regimens and drug discovery that will benefit
society. A comparison may be drawn to clinical drug trials in
oncology when death of the patient is inevitable, or, more
closely, to body donation for medical education. Although
such use of neuromonitoring may be ethically sound so long
as the proper intents and consent support it, caution must be
exhibited to not take an entirely utilitarian position. Such a
position is antagonistic to John Stuart Mill’s belief that the
goal of moral actions is to create freedom and autonomy, not
to further the good of any particular division of society.57 It
also is antagonistic to both the AMA’s policy (E-8.081), which
states that “any quality of life considerations should be mea-
sured as the worth to the individual whose course of treatment
is in question, and not as a measure of social worth,”58 and
most importantly, to Americans’ First Amendment, constitu-
tional rights.46
Futile measures carry financial, emotional, and ethical costs
that neuromonitoring may either exacerbate by being futile
itself or relieve by helping to identify futile treatments. It must
be stressed that physicians are not obligated to take futile
measures, but that they may be pushing the ethical envelopes
if they do so. For example, hospital or ICU beds used by
patients in whom measures are futile are not available to other
patients who require care. Thus rationing of care in the ICU
is necessary to maintain an ethical practice.59 Even in cases in
which society may benefit from a futile action (e.g., neuro-
monitoring in a PVS patient), both lack of patient consent for
continued care and a lack of financial capacity ethically disfa-
vor futile treatment. Neuromonitoring in this case includes
continuous physiologic data but may also include serial “bio-
markers” collected over several days or sophisticated imaging
techniques (reviewed in Chapters 18 and 28). A balance is
necessary because this information can help decide when
management is futile or help “predict” which patient may have
a “good” outcome or a “poor” outcome, the definition of
which may depend on whether a caregiver or family member
is asked. For example, a family member may regard a good
outcome as recovery of functional communication even

though the individual may remain physically dependant.
Functional communication recovery can be assessed through
the Coma Recovery Scale-Revised (CRS-R).60,61
A feasible way to decrease the financial burden of patients
with disorders of consciousness requires improved diagnostic
and prognostic assessments.62 This particularly benefits deci-
sion making for those patients who meet criteria for the MCS
due to any etiology, and patients with a traumatic brain injury
(TBI) that results in “stunned” neuronal tissue.63 Unlike brain
dead and PVS cases, MCS and TBI patients may have a pos-
sible, but ill-defined chance of recovery28,51,62 that calls into
question assessments of futility. Perhaps with improved prog-
nostic tools, the probability of such transitions with particular
interventions will be easier to assess. Neuromonitoring may
harbor such prognostic potential that would not only classify
monitoring as valuable in MCS and TBI, rather than as a futile
measure in these cases, but also as a tool to avoid other futile
measures. These tools also may include sophisticated and
advanced neuroimaging techniques or biomarkers.64-67 This
would lighten the economic and thus ethical loads carried by
potentially futile care.
Contrary to the beliefs of some, actions to reduce futility
should not be viewed as mere attempts to increase organ
donation. Neuromonitoring should have the same aim to
combat futility, but it should be recognized that even with this
as its aim, some people may perceive it as a tool used to
achieve earlier and increased organ donation from patients.
Physicians must explain with sincerity to patients, families,
and the media that this is not its purpose. The recent birth of
this technology and society’s unfamiliarity with it may further
contribute to the family’s suspicion and hesitation to condone
its use.
The Utility of Monitoring
Ethically Weighted by Prognostic
Ambiguity and Futility
The early evolution of neurocritical care as a distinct subspe-
cialty has occurred in parallel with the ability to monitor and
image physiologic function of the human brain. What role
does monitoring in the NCCU play in futility decisions? This
section discusses the practicality to avoid futile measures,
while promoting individual patient’s wishes. The text first
addresses the utility and limits of monitoring in the NCCU to
provide a sense of its potential (although this is discussed in
greater depth in the chapters about each monitor), then con-
tinues with a conversation about shared decision making with
the patient’s family in cases that may involve neuromonitor-
ing. At the outset it is important to realize that use of a
monitor depends primarily on the accuracy of interpretation
of the data provided and whether effective therapies or deci-
sions based on that information exist and are validated.
Monitoring in critical care may be of most use in the fol-
lowing situations, suggested by Masdeu as indicative of a need
for ancillary medical technology and confirmatory tests: (1)
when clinical diagnosis is uncertain, (2) when metabolic
derangements are present, (3) when the brainstem is selec-
tively damaged, (4) when evaluation of brainstem function is
hindered (i.e., by facial trauma), (5) in young children, and
(5) as a confirmatory test for brain death.68 Studies are under
way to assess the prognostic value of neuromonitoring. In the
Section I—Background 67
NCCU, monitoring also may have the potential to provide
information about a condition that is still reversible and
before it becomes irreversible.
Monitoring should, in theory, yield information on the
patient’s status that will direct surgical, medical, and personal
management to maximize benefit tailored to the patient’s
individual wants. This should improve outcomes as viewed by
the patient and physician. The data, however, on how moni-
toring in the NCCU affects outcome are mixed. Most of these
studies have addressed the role of intracranial pressure (ICP)
monitors in TBI; some studies suggest outcome is not
improved by monitoring.69-72 However, a meta-analysis and
literature review that included more than 100,000 patients
suggests that adherence to guidelines for severe TBI and use
of an ICP monitor is associated with better outcome after
severe TBI.73,74 Consistent with this Whitmore et al.75 using a
decision-analytical model observed that “aggressive” care,
including use of invasive monitoring, for TBI is associated
with better outcomes and is more cost effective than “routine”
care and even for older patients. In addition, potential bene-
fits to use of a monitor may depend on the use of more than
one monitor (i.e., better targeting of therapy).76 For example,
in severe TBI information from a brain oxygen monitor
appears to provide valuable information about outcome
beyond that provided by an ICP monitor, both in adults and
children,77-79 and when the two monitors are used together,
therapy based on that information may provide further
improvements in outcome.80 The role of monitoring in other
conditions treated in the NCCU is less well studied than for
TBI, in which again the combination of monitors may be
more useful than a single monitor alone. For example,
Dohmen et al. and Heiss et al.81,82 observed that although use
of microdialysis, ICP, and tissue oxygen pressure could classify
the pathophysiology of a malignant middle cerebral artery
infarct, this monitoring did not predict it in time to perform
a hemicraniectomy, whereas positron emission tomography
(PET) imaging did.
These mixed results associated with how monitoring affects
NCCU care may reflect several issues, including: (1) neuro-
monitoring only assesses a small portion of brain tissue82 and
thus may suggest a generalization that misrepresents the
patient’s condition (or it provides global information and
misses the regional data that is more important); (2) the risks
of an invasive device or how information provided by the
device is used may outweigh the benefits; (3) user dependency
or the learning curve may further skew the risk-to-benefit
ratio;83 (4) inexperience with what to do with the information
obtained from neuromonitoring may lead to errors;40
(5) more information (that may mean more treatment) does
not always mean better outcome; or (6) the information pro-
vided by a monitor may indicate a treatment that is useful but
the treatment then has unexpected adverse effects.84,85
To date no level I (i.e., a randomized controlled trial) evi-
dence demonstrates that a monitor (or monitors) is helpful in
the NCCU. This may be important because randomized
studies in general critical care showed that pulmonary artery
catheter use did not always contribute to better patient
outcomes.86-88 However, it can be argued that randomized
trials may not be the gold standard answer to clinical ques-
tions in critical care,89,90 because much of what is done during
day-to-day practice is based on understanding pathophysiol-
ogy and how treatment may influence that pathophysiology.
68 Section I—Background
In many conditions managed in the NCCU and for many
therapies, this knowledge about pathophysiology is only
beginning to be elucidated in part through the use of neuro-
monitoring. Hence an argument could be made for neuro-
monitoring both for the patient being treated and for patients
likely to be treated in the future. However, any technologic
advance can have unintended consequences and so create a
pseudodisease that requires “treatment.”91 How these conse-
quences are viewed legally differs from how they are seen in
health care: in medicine results, conclusions, and recommen-
dations about care are subject to continuing review—they can
change over time. By contrast, the courts make conclusions
only once, at trial.92
Because the prognostic and diagnostic value of some aspects
of neuromonitoring still remains to be clearly defined, and
because neuromonitoring may be a futile measure in some
patients, the utility of neuromonitoring must be empirically
studied, and families must be well informed of its potential
limits and benefits. Physicians are more apt to individually
make decisions with little to no invited input from families
and patients when the ambiguity of the utility of a medical
treatment or tool is great.93 This is yet another motivator to
clarify the value and use of neuromonitoring: to allow a bal-
anced conversation to occur between families and physicians
to decide on treatments for loved ones.94
When and How to Use
Neuromonitoring in the Context
of Patients’ and Their Families’
Needs and Wants
Because neuromonitoring may provide more insight about a
patient’s prognosis, the family and treatment team may make
a more informed, patient-centric decision. A few issues can
complicate this promising thought. Neuromonitoring itself
may seem to represent a shift from personal, patient-oriented
care to generalized, machine-controlled care. Since ventilators
were introduced to the ICU in the 1950s and 1960s, talk of
machines, not humans, carrying out life has permeated ethical
literature. The impact of machines on the definitions, assess-
ment, and perceptions of life and death is obviously great (see
Table 8.2).
The recognition of the technologic dependency of life and
death is evident in legal documents and cases, including the
statement by a Florida Court of Appeals judge of the Cruzan
case,95 who noted that “it may be determined by reason of the
advanced scientific and medical technologies of this day that
Life has, through causes beyond human control, reached the
unconscious and vegetative state where all that remains is the
forced function of the body’s vital functions.” Many ethically
weighted court cases (see reference 57, case of Karen Quinlan,
as state precedent; reference 95, case of Nancy Cruzan as
national precedent; and reference 55, for summary) that fol-
lowed publication of the Harvard Criteria4 for brain death
often centered on what should be done with these life-
supporting (or death-prolonging, so as not to suggest bias)
machines. The case of Karen Quinlan,57 a young woman in a
PVS whose parents desired to withdraw machine support
(see Table 8.2), was specifically referenced as evidence of dep-
ersonalized, machine-controlled medicine.57 Perceptions of
machine-dependent medicine and life permeated the dialogue
16 years later in the Cruzan case, and continue to maintain a
presence in discussions.95
How may providers ensure that neuromonitoring devices
are viewed as simply ancillary tools for physicians rather
than as hardwired controllers of life? The first step is to
define the potential value of these tools. Second, clarification
about the purpose and value of these tools as they pertain to
the individual patient must be provided to proxies. Finally,
quality assessment of the ethical conduct displayed when
neuromonitoring-related decisions are made, as is suggested
for all practices in medicine,55 may motivate ethical behavior
and boost confidence in the ethical use of these technolo-
gies. These efforts toward individualizing patient care must
be conducted with the overarching goal to avoid all mea-
sures the patient would deem futile, and pursue all measures
the patient would deem worthwhile.
In the NCCU, a patient’s wishes most often are unknown.
These preferences need to be considered when neuromoni-
toring is used for personalized end-of-life care. These prefer-
ences are difficult to determine, in part because this is a new
set of technologies that yields a new set of data and more
generally because few persons have advance directives even
though they were conceptually conceived in 1969.55,96 Even
when advance directives are available, their specifications can
be cryptic and sometimes they are disregarded. State-to-state
variability in legal regulation of directives can create addi-
tional confusion for patients and physicians. Efforts to publi-
cize and standardize the collection and use of advance
directives49 are therefore worthwhile. A federal statute, the
Patient Self-Determination Act that mandates that covered
medical institutions actively distribute, collect, and honor
information on their protocol for handling advance direc-
tives, has not been sufficient. This suggests there is a need to
fix the protocol, not just publicize it.55 The Uniform Health-
Care Decisions Act of 1993 represents an effort worth atten-
tion.97 This act lends clarity, not just motivation, to the
proper collection use of advance directives consent, advance
directives, and appointment of durable power of attorneys in
a manner that likely will best promote the patient’s presumed
wishes. Both the AARP (formerly the American Association
of Retired Persons) and the ABA endorsed this act, but
only a fraction of states have adopted it (see discussion in
reference 49).49
As our legal system progresses to meet patients’ needs, spe-
cifically those prominent in disorders of consciousness, direct
reference to decisions that involve neuromonitoring and the
data it provides should perhaps be included. This is worth
careful thought, because the system to identify suitable
proxies for these patients is often insufficient and how sur-
rogates view futility or decide about care may have implica-
tions for health care providers. For example, one study
indicates that proxies did no better at fulfilling the neuro­
logically compromised patient’s wishes than chance alone,98
whereas other studies indicate that up to two thirds of
surrogates express doubts about the accuracy of futility
predictions.99
Many legal and ethical obstacles complicate the process to
choose a sufficient proxy for consent of three areas that
may concern neuromonitoring: treatment, research, or with-
drawal of care. First, potential proxies must be deemed suit-
able to act as the legal representative of the impaired person,
and be willing to do so. In a study by Mason et al., it was

found that a large number of impaired persons did not have
a suitable proxy, and among those who did, the majority of
the potential proxies were not willing to fill this role.100 To
claim a proxy as “suitable” a person must determine the
motives behind the proxy’s decision, and ensure their choice
is based on their best estimation of what the patient would
want. This assumes that the proxy understands the concep-
tual basis for the treatment, comfort measures only (CMO)
status, experimental therapies, or other routes being pursued.
Second, it must be confirmed that the proxy will not acquire
a secondary gain by prolonging or hastening an inevitable
death (e.g., financial, property, capitalizing on family dis-
agreements). In the case of multiple proxies, “disagreements
among proxies with equal priorities” may arise101 and pose
yet another obstacle in this process. Finally, the legal specifics
of proxy consent are similar to brain death criteria32,55,56
in that they are under the jurisdiction of the state, and
there is much heterogeneity between and confusion within
states.55,102 Similarly there is significant variability in institu-
tional review board surrogate consent practices and what
risk is acceptable in research on incapacitated adults.103
Indeed, there are limits#55 to studies that investigate the
success of proxy appointments, but what is consistent with
most of them is that the decision-making process for the
neurologically incompetent patient is inadequate. Still,
current protocols for decision making in futile cases (see ref-
erence 49 for review) offer a usable framework for decisions
about neuromonitoring.
There likely will be times when the treatment team believes
that neuromonitoring is not indicated and views monitoring
as futile, but the family insists on its use (i.e., there is dis-
agreement with the perception of futility). In these circum-
stances, neuromonitoring may be viewed in the same light as
artificial nutrition and hydration, antibiotics, and other
medical treatments. Only so long as information provided by
monitoring can help prolong or improve life, in the physi-
cian’s best opinion, is the physician obligated to use it, unless
a patient’s advance directive or informally expressed wishes
request otherwise.49 The origin of the family’s request also
must be acknowledged and responded to in the following
cases:
1. If the family seems to harbor false hope for a recovery, or
if the family seems to be in denial, the physician should
clearly reiterate the prognosis and diagnosis.
2. If the family is simply unwilling to “give up,” the physician
must assure them that they have done all they can, and
that further efforts are not in the best interest of the
patient. The physician also should comfort the family by
telling them that by withdrawing monitors, the family is
not giving up on their loved one, but rather, accepts that
the monitors simply do not provide data to benefit the
patient.
3. If the family appears to have a desire for knowledge about
the etiology of the patient’s state, physicians should help
them find closure in some other knowledge-based way. An
example of this is to share the most complete information
available that is relevant to the patient’s condition, offer
one’s contact information for further questions, and then
direct families to support and educational groups that have
#
Limits include variable methodology in data collection and assessment.
Section I—Background 69
published resources that the family can use to expand their
knowledge.
An inability of physicians to clearly and fully address futility
to families means families may insist on increased end-of-life
care, prolong emotional suffering by the family, and render a
poor reputation to the hospital as one that carelessly spends
patients’ money without an impact on outcomes. This is par-
ticularly concerning because Consumer Reports now lists such
figures as average cost and intensity of end-of-life care for the
public to compare hospitals.104 In end-of-life care, mediation
between families and the physician usually is successful in
resolving conflict if it exists, although it still is to be fully
elucidated on how to optimally use an ethics consult in neuro-
critical care.105,106 This can be important because there are
differences in how the public and health care professionals
view futility. For example, Jacobs et al.107 in a survey of the
general public and of trauma professionals found that more
of the public (57.4%) than the professionals (19.5%) believe
that divine intervention may save a person when treatment
was thought to be futile.
A final communicative issue that is more specific to neuro-
monitoring than other medical modalities used near the end
of life concerns families and legal authorities who may misin-
terpret data from neuromonitoring. This misinterpretation
may result in confusion, anger, and poor substituted judgment
by the proxy. Physicians should be clear and complete in
explaining the output of the monitors to prevent this confu-
sion. If misinterpretation-driven frustration still arises, the
physician should again make a personal attempt to clarify the
situation and, if need be, call on a hospital ethicist if his or her
attempt fails.
Conclusion
In any conversations about neuromonitoring, the relation of
these tools to ethical issues relevant to critical care must be
addressed. The most pertinent is futility. This chapter defined
futility, describing irreversible and reversible points that may
progress to the ultimate irreversible end—human death, with
a definition that is still evolving. It also addressed the ethical
context of futile measures in medicine, bound by economical
limits, and ethically weighted by how these limits restrict
medical care for other patients. It recognized that neuromon-
itoring, like other technologies in critical care, is ethically
complicated and entwined with issues of brain death and
futility given its impact on the quality and length of human
life. The potential for neuromonitoring to lighten ethical
burdens also was discussed, specifically, that brain death,
along with states of less severe neurologic dysfunction (i.e.,
PVS and MCS) may be more accurately predicted, assessed,
and diagnosed with neuromonitoring. This could help
prevent costly futile measures, or premature withdrawal of
care. Finally, it was stressed that the full impact of neuro-
monitoring has yet to be realized. It is particularly important
that as this realization comes to fruition, as the prognostic
value of neuromonitoring improves, caution be expressed.
Overconfidence in neuromonitoring’s ability to predict a
poor outcome may lead to hasty decisions that surrender
human life before its end is inevitable. This possibility is
expressed by the court-appointed guardian for Karen
Quinlan, Coburn, who said, “One human being, by conduct
70 Section I—Background
or lack of conduct, is going to cause the death of another
human being.”57 Overconfidence also could lead to misguided
rationing, and physicians may suggest against care they incor-
rectly deem futile based on data from a monitor. This may be
less likely when multimodality monitoring is used. On the
other hand, inexperience with monitoring increases these
risks,‡‡39 and a lack of clear patient directives about end-of-
life care complicates decision making.
Finally, physicians must understand patient and family
values, concerns, and views of monitoring in the NCCU. Par-
ticular attention must be paid to ensure that patients are not
viewed as hostages to machines, but rather that these tools
enhance patient-centric care and decision making.
References
1. Daroff R. The historical evolution of brain death from former definitions of
death: the Harvard Criteria to the present. In: Marcelo Sanchez Sorondo HE,
editor. The signs of death: the proceedings of the working group. Scripta
varia 110. Rome: Pontificia Academia Sciantiarvm; 2006. p. 217–21.
2. Hacke W. Brain death: an artifact created by critical care medicine or the
death of the brain has always been the death of the individuum. In: Marcelo
Sanchez Sorondo HE, editor. The signs of death: the proceedings of the
working group. Scripta varia 110. Rome: Pontificia Academia Sciantiarvm;
2006. p. 84–91.
3. Jennett B. Foreword. In: Wijdicks EFM, editor. Brain death. Philadelphia:
Lippincott Williams & Wilkins; 2001. p. IX–X.
4. Ad Hoc Committee of the Harvard Medical School. A definition of
irreversible coma. JAMA 1968;205:337–40.
5. Mohandas A, Chou SN. Brain death—a clinical and pathologic study.
J Neurosurg 1971;35:211–18.
6. Conference of Medical Royal Colleges and Their Facilities in the United
Kingdom. Diagnosis of brain death. BMJ 1976:1187–8.
7. Conference of Medical Royal Colleges and Their Facilities in the United
Kingdom. Diagnosis of brain death. Lancet 1976:1069–70.
8. U.S. Collaborative Study. An appraisal of the criteria of cerebral death. JAMA
1977;237:982–6.
9. U.S. President’s Commission. Guidelines for the determination of death.
JAMA 1981;246:2184–6.
10. American Academy of Neurology. Practice parameters for determining brain
death in adults. Neurology 1995;45:1012–14.
11. Wijdicks EF, Varelas PN, Gronseth GS, et al. Evidence-based guideline
update: determining brain death in adults: report of the Quality Standards
Subcommittee of the American Academy of Neurology. Am Acad Neurol
2010;74(23):1911–18.
12. House of Delegates redefines death, urges redefinition of rape, and undoes
the Houston amendments. Am Bar Assoc J 1975;61:463–4.
13. Veith FJ, Fein JM, Tendler MD. Brain death: a status report of medical and
ethical considerations. JAMA 1977;238:1651–5.
14. Bernat J. How do physicians prove irreversibility in the determination of
death? In: Marcelo Sanchez Sorondo HE, editor. The signs of death: the
proceedings of the working group. Scripta varia 110. Rome: Pontificia
Academia Sciantiarvm; 2006. p. 159–76.
15. Oxford English Dictionary. 2nd ed. Oxford: Oxford University Press. 2006.
16. Cole D. Statutory definitions of death and the management of terminally ill
patients who may become organ donors after death. Kennedy Inst Ethics J
1993;3(2):145–55.
17. Aulisio MP, Devita M, Luebke D. Taking values seriously: ethical challenges
in organ donation and transplantation for critical care professionals. Crit
Care Med 2007;35(2 Suppl):S95–101.
18. Ropper A. The apnea test and rationale for brain death as death. In: Marcelo
Sanchez Sorondo HE, editor. The signs of death: the proceedings of the
working group. Scripta varia 110. Rome: Pontificia Academia Sciantiarvm;
2006. p. 237–49.
19. Capron AM, Kass LR. A statutory definition of the standards for determining
human death: an appraisal and a proposal. Univ Penn Law Rev
1972;121:87–118.
20. Uniform Determination of Death Act, Uniform Laws Annot 12A:589
www.law.upenn.edu/bll/archives/ulc/fnact99/1980s/udda80.htm.
‡‡
Inexperience was one of the top five reasons for the misdiagnosis of PVS in a set of
MCS patients.
21. Bernat J. Philosophical and ethical aspects of brain death. In: Wijdicks EFM,
editor. Brain death. Philadelphia: Lippincott Williams & Wilkins; 2001. p.
171–88.
22. Brain dead mother is taken off life support. Available at: www.cbsnews.com/
stories/2005/06/17/health/main702568.shtml and www.washingtonpost.com/
wp-dyn/content/article/2005/08/03/AR2005080300224.html (accessed
September 2009).
23. Posner J. Alleged awakenings from prolonged coma and brain death and
delivery of live babies from brain-dead mothers do not negotiate brain
death. In: Marcelo Sanchez Sorondo HE, editor. The signs of death: the
proceedings of the working group. Scripta varia 110. Rome: Pontificia
Academia Sciantiarvm; 2006.
24. Bartlett E, Youngner S. Human death and the destruction of the neocortex.
In: Zaner RN, editor. Death: beyond whole-brain criteria. Dordrecht, NL:
Kluwer Academic Publishers; 1988. p. 199–215.
25. Wang MY, Wallace P, Gruen JB. Brain death documentation: analysis and
issues. Neurosurgery 2002;51:731–5.
26. Shewmon DA. Chronic ‘brain death’ meta-analysis and conceptual
consequences. Neurology 1998;51:1538–45.
27. Repertinger S, et al. Long survival following bacterial meningitis-associated
brain destruction. J Child Neurol 2006;21(7):591–5.
28. Owen A. Disorders of consciousness. Ann N Y Acad Sci 2008;1124:
225–38.
29. Group discussion. In: Marcelo Sanchez Sorondo HE, editor. The signs of
death: the proceedings of the working group. Scripta varia 110. Rome:
Pontificia Academia Sciantiarvm; 2007. p. 21–3.
30. Certain aspects of the care and management of the persistent vegetative state
patient. AAN position statement adopted on April 21, 1988. Available at
www.aan.com/about/ethics/109556.pdf.
31. Persistent vegetative state and the decision to withdraw or withhold life
support. Council on Scientific Affairs and Council on Ethical and Judicial
Affairs. JAMA 1990;263(3):426–30.
32. Bernat J, Culver CM, Gert B. On the definition and criterion of death. Ann
Intern Med 1981;94:389–94.
33. Steinbock B. Recovery from the persistent vegetative state? The case of Carrie
Coons. Hastings Cent Rep 1989;19:14–5.
34. Voss H, Uluc AM, Dyke JP. Possible axonal regrowth in late recovery from
the minimally conscious state. J Clin Invest 2006;116(7):2005–11.
35. Lammi MH, Smith VH, Tate RL, et al. The minimally conscious state and
recovery potential: a follow-up study 2 to 5 years after traumatic brain
injury. Arch Phys Med Rehabil 2005;86(4):746–54.
36. Laureys S, Schiff ND. Coma and consciousness: paradigms (re)framed by
neuroimaging. Neuroimage 2011;Dec 27. Epub ahead of print.
37. Fins JJ, Schiff ND, Foley KM. Late recovery from the minimally conscious
state: ethical and policy implications. Neurology 2007;68(4):304–7.
38. Giacino JT, Schnakers C, Rodriguez-Moreno D, et al. Behavioral assessment
in patients with disorders of consciousness: gold standard or fool’s gold?
Prog Brain Res 2009;177:33–48.
39. Andrews K. Recovery of patients after four months or more in the persistent
vegetative state. BMJ 1993;306(6892):1597–600.
40. Coimbra CG. Implications of ischemic penumbra for the diagnosis of brain
death. Braz J Med Biol Res 1999;32:1479–87.
41. Plum F, Posner JB. The diagnosis of stupor and coma. 3rd ed. Philadelphia:
FA Davis Co; 1980. p. 88–101.
42. de Groot YJ, Jansen NE, Bakker J, et al. Imminent brain death: point of
departure for potential heart-beating organ donor recognition. Intensive
Care Med 2010;36(9):1488–94. Epub 2010, Mar 16.
43. Gallagher C, Wijdicks EFM. Religious and cultural aspects of brain death. In:
Wijdicks EFM, editor. Brain death. Philadelphia: Lippincott Williams &
Wilkins; 2001. p. 135–50.
44. Olick RS. Brain death, religious freedom, and public policy: New Jersey’s
landmark legislative initiative. Kennedy Inst Ethics J 1991:275–88.
45. NY comp codes, rules & regs, Title 10, section 400.16(d), (e)(3) (1992)
46. U.S. Const. Amend 1
47. U.S. Const. Amend 14
48. Truog R, Burns J. Futility: a concept in evolution. Chest 2007;132(6):
1987–93.
49. Bacon D, Williams MA, Gordon J. Position statement on laws and
regulations concerning life-sustaining treatment, including artificial
nutrition and hydration, for patients lacking decision-making capacity.
Neurology 2007;68:1097–100.
50. Bitsori M, Georgopoulos D, Galanakis E. The question of futility and Roger
C. Bone. 2009;12(4):477–81. Epub 2009, Mar 4.
A complete list of references for this chapter can be found online at
www.expertconsult.com.

References
1. Daroff R. The historical evolution of brain death from former definitions
of death: the Harvard Criteria to the present. In: Marcelo Sanchez Sorondo
HE, editor. The signs of death: the proceedings of the working group.
Scripta varia 110. Rome: Pontificia Academia Sciantiarvm; 2006. p. 217–21.
2. Hacke W. Brain death: an artifact created by critical care medicine or the
death of the brain has always been the death of the individuum. In:
Marcelo Sanchez Sorondo HE, editor. The signs of death: the proceedings
of the working group. Scripta varia 110. Rome: Pontificia Academia
Sciantiarvm; 2006. p. 84–91.
3. Jennett B. Foreword. In: Wijdicks EFM, editor. Brain death. Philadelphia:
Lippincott Williams & Wilkins; 2001. p. IX–X.
4. Ad Hoc Committee of the Harvard Medical School. A definition of
irreversible coma. JAMA 1968;205:337–40.
5. Mohandas A, Chou SN. Brain death—a clinical and pathologic study.
J Neurosurg 1971;35:211–18.
6. Conference of Medical Royal Colleges and Their Facilities in the United
Kingdom. Diagnosis of brain death. BMJ 1976:1187–8.
7. Conference of Medical Royal Colleges and Their Facilities in the United
Kingdom. Diagnosis of brain death. Lancet 1976:1069–70.
8. U.S. Collaborative Study. An appraisal of the criteria of cerebral death.
JAMA 1977;237:982–6.
9. U.S. President’s Commission. Guidelines for the determination of death.
JAMA 1981;246:2184–6.
10. American Academy of Neurology. Practice parameters for determining
brain death in adults. Neurology 1995;45:1012–14.
11. Wijdicks EF, Varelas PN, Gronseth GS, et al. Evidence-based guideline
update: determining brain death in adults: report of the Quality Standards
Subcommittee of the American Academy of Neurology. Am Acad Neurol
2010;74(23):1911–18.
12. House of Delegates redefines death, urges redefinition of rape, and undoes
the Houston amendments. Am Bar Assoc J 1975;61:463–4.
13. Veith FJ, Fein JM, Tendler MD. Brain death: a status report of medical and
ethical considerations. JAMA 1977;238:1651–5.
14. Bernat J. How do physicians prove irreversibility in the determination of
death? In: Marcelo Sanchez Sorondo HE, editor. The signs of death: the
proceedings of the working group. Scripta varia 110. Rome: Pontificia
Academia Sciantiarvm; 2006. p. 159–76.
15. Oxford English Dictionary. 2nd ed. Oxford: Oxford University Press. 2006.
16. Cole D. Statutory definitions of death and the management of terminally ill
patients who may become organ donors after death. Kennedy Inst Ethics J
1993;3(2):145–55.
17. Aulisio MP, Devita M, Luebke D. Taking values seriously: ethical challenges
in organ donation and transplantation for critical care professionals. Crit
Care Med 2007;35(2 Suppl):S95–101.
18. Ropper A. The apnea test and rationale for brain death as death. In:
Marcelo Sanchez Sorondo HE, editor. The signs of death: the proceedings
of the working group. Scripta varia 110. Rome: Pontificia Academia
Sciantiarvm; 2006. p. 237–49.
19. Capron AM, Kass LR. A statutory definition of the standards for
determining human death: an appraisal and a proposal. Univ Penn Law Rev
1972;121:87–118.
20. Uniform Determination of Death Act, Uniform Laws Annot 12A:589
www.law.upenn.edu/bll/archives/ulc/fnact99/1980s/udda80.htm.
21. Bernat J. Philosophical and ethical aspects of brain death. In: Wijdicks
EFM, editor. Brain death. Philadelphia: Lippincott Williams & Wilkins;
2001. p. 171–88.
22. Brain dead mother is taken off life support. Available at:
www.washingtonpost.com/wp-dyn/content/article/2005/08/03/
AR2005080300224.html (accessed September 2009).
23. Posner J. Alleged awakenings from prolonged coma and brain death and
delivery of live babies from brain-dead mothers do not negotiate brain
death. In: Marcelo Sanchez Sorondo HE, editor. The signs of death: the
proceedings of the working group. Scripta varia 110. Rome: Pontificia
Academia Sciantiarvm; 2006.
24. Bartlett E, Youngner S. Human death and the destruction of the neocortex.
In: Zaner RN, editor. Death: beyond whole-brain criteria. Dordrecht, NL:
Kluwer Academic Publishers; 1988. p. 199–215.
25. Wang MY, Wallace P, Gruen JB. Brain death documentation: analysis and
issues. Neurosurgery 2002;51:731–5.
26. Shewmon DA. Chronic ‘brain death’ meta-analysis and conceptual
consequences. Neurology 1998;51:1538–45.
27. Repertinger S, et al. Long survival following bacterial meningitis-associated
brain destruction. J Child Neurol 2006;21(7):591–5.
28. Owen A. Disorders of consciousness. Ann N Y Acad Sci 2008;1124:225–38.
Section I—Background 70.e1
29. Group discussion. In: Marcelo Sanchez Sorondo HE, editor. The signs of
death: the proceedings of the working group. Scripta varia 110. Rome:
Pontificia Academia Sciantiarvm; 2007. p. 21–3.
30. Certain aspects of the care and management of the persistent vegetative
state patient. AAN position statement adopted on April 21, 1988. Available
at www.aan.com/about/ethics/109556.pdf.
31. Persistent vegetative state and the decision to withdraw or withhold life
support. Council on Scientific Affairs and Council on Ethical and Judicial
Affairs. JAMA 1990;263(3):426–30.
32. Bernat J, Culver CM, Gert B. On the definition and criterion of death. Ann
Intern Med 1981;94:389–94.
33. Steinbock B. Recovery from the persistent vegetative state? The case of
Carrie Coons. Hastings Cent Rep 1989;19:14–5.
34. Voss H, Uluc AM, Dyke JP. Possible axonal regrowth in late recovery from
the minimally conscious state. J Clin Invest 2006;116(7):2005–11.
35. Lammi MH, Smith VH, Tate RL, et al. The minimally conscious state and
recovery potential: a follow-up study 2 to 5 years after traumatic brain
injury. Arch Phys Med Rehabil 2005;86(4):746–54.
36. Laureys S, Schiff ND. Coma and consciousness: paradigms (re)framed by
neuroimaging. Neuroimage 2011;Dec 27. Epub ahead of print.
37. Fins JJ, Schiff ND, Foley KM. Late recovery from the minimally conscious
state: ethical and policy implications. Neurology 2007;68(4):304–7.
38. Giacino JT, Schnakers C, Rodriguez-Moreno D, et al. Behavioral assessment
in patients with disorders of consciousness: gold standard or fool’s gold?
Prog Brain Res 2009;177:33–48.
39. Andrews K. Recovery of patients after four months or more in the
persistent vegetative state. BMJ 1993;306(6892):1597–600.
40. Coimbra CG. Implications of ischemic penumbra for the diagnosis of brain
death. Braz J Med Biol Res 1999;32:1479–87.
41. Plum F, Posner JB. The diagnosis of stupor and coma. 3rd ed. Philadelphia:
FA Davis Co; 1980. p. 88–101.
42. de Groot YJ, Jansen NE, Bakker J, et al. Imminent brain death: point of
departure for potential heart-beating organ donor recognition. Intensive
Care Med 2010;36(9):1488–94. Epub 2010, Mar 16.
43. Gallagher C, Wijdicks EFM. Religious and cultural aspects of brain death.
In: Wijdicks EFM, editor. Brain death. Philadelphia: Lippincott Williams &
Wilkins; 2001. p. 135–50.
44. Olick RS. Brain death, religious freedom, and public policy: New Jersey’s
landmark legislative initiative. Kennedy Inst Ethics J 1991:275–88.
45. NY comp codes, rules & regs, Title 10, section 400.16(d), (e)(3) (1992)
46. U.S. Const. Amend 1
47. U.S. Const. Amend 14
48. Truog R, Burns J. Futility: a concept in evolution. Chest 2007;132(6):
1987–93.
49. Bacon D, Williams MA, Gordon J. Position statement on laws and
regulations concerning life-sustaining treatment, including artificial
nutrition and hydration, for patients lacking decision-making capacity.
Neurology 2007;68:1097–100.
50. Bitsori M, Georgopoulos D, Galanakis E. The question of futility and Roger
C. Bone. 2009;12(4):477–81. Epub 2009, Mar 4.
51. Medical aspects of the persistent vegetative state. The Multi-Society Task
Force on PVS. N Engl J Med 1994;330(22):1572–9.
52. Terri Schiavo case. In: Phelps S, editor. Gale encyclopedia of everyday
law. Gale Group, 2003. eNotes.com. Available at http://www.enotes.com/
topic/Terri_Schiavo_case. Accessed July 26, 2012 (accessed September
2009).
53. MergerWatch. Summary of state bills restricting end-of-life choices
introduced in 2005 and 2006. March 28. Available online at
www.mergerwatch.org/pdfs/fs_end_of_life_legislation.pdf, 2006 (accessed
September 2009).
54. Resolution 209 (A-05), as introduced at the 2005 Annual Meeting of the
American Medical Association. Available online at www.ama-assn.org/
meetings/public/annual05/209a05.doc.
55. Wolf SM, et al. Sources of concern about the Patient Self-Determination
Act, 325. New Engl J Med 1991;1666–71.
56. Mohammed S, Peter E. Rituals, death and the moral practice of medical
futility. Nurs Ethics 2009;16(3):292–302.
57. Pence GE. Classical cases in medical ethics: accounts of the cases that have
shaped medical ethics with philosophical, legal, and historical backgrounds.
New York: McGraw-Hill, 1990. Ch 1-2.
58. YPS delegate’s report, AMA Annual Meeting, June 17–22, 2005, Chicago, IL.
Available at www.ama-assn.org/ama1/pub/upload/mm/17/
a05delegatesreport.pdf.
59. Truog RD, Brock DW, Cook DJ, et al. For the Task Force on Values, Ethics,
and Rationing in Critical Care (VERICC). Rationing in the intensive care
unit. Crit Care Med 2006;34(4):958–63.
70.e2 Section I—Background
60. Giacino JT, Kalmar K, Whyte J. The JFK Coma Recovery Scale-Revised:
measurement characteristics and diagnostic utility. Arch Phys Med Rehabil
2004;85:2020–9.
61. Maas AI, Harrison-Felix CL, Menon D, et al. Common data elements for
traumatic brain injury: recommendations from the interagency working
group on demographics and clinical assessment. Arch Phys Med Rehabil
2010;91:1641–9.
62. Giacino JT. The minimally conscious state: definition and diagnostic
criteria. Neurology 2002;58:349–53.
63. Hennerici W. Surviving areas of brain tissue in brain death: is the whole
more than the sum of its parts? In: Marcelo Sanchez Sorondo HE, editor.
The signs of death: the proceedings of the working group. Scripta varia 110.
Rome: Pontificia Academia Sciantiarvm; 2006. p. 101–14.
64. Lescot T, Galanaud D, Puybasset L. Exploring altered consciousness states
by magnetic resonance imaging in brain injury. Ann N Y Acad Sci
2009;1157:71–80.
65. Galanaud D, Puybasset L. Cardiac arrest: has the time of MRI come? Crit
Care 2010;14(2):135. Epub 2010, Mar 26.
66. Tshibanda L, Vanhaudenhuyse A, Galanaud D, et al. Magnetic resonance
spectroscopy and diffusion tensor imaging in coma survivors: promises and
pitfalls. Prog Brain Res 2009;177:215–29.
67. Siman R, Toraskar N, Dang A, et al. A panel of neuron-enriched proteins as
markers for traumatic brain injury in humans. J Neurotrauma
2009;26(11):1867–77.
68. Masdeu J. Neuroimaging: a window into total brain destruction and the
unresponsive states. In: Marcelo Sanchez Sorondo HE, editor. The signs of
death: the proceedings of the working group. Scripta varia 110. Rome:
Pontificia Academia Sciantiarvm; 2006. p. 222–34.
69. Idris Z, Ghani RI, Musa KI, et al. Prognostic study of using different
monitoring modalities in treating severe traumatic brain injury. Asian J
Surg 2007;30(3):200–8.
70. Steiner LA, Andrews PJ. Monitoring the injured brain: ICP and CBF. Br J
Anaesth 2006;97(1):26–38.
71. Cremer OL, van Dijk GW, van Wensen E, et al. Effect of intracranial
pressure monitoring and targeted intensive care on functional outcome
after severe head injury. Crit Care Med 2005;33(10):2207–13.
72. Shafi S, Diaz-Arrastia R, Madden C, et al. Intracranial pressure monitoring
in brain-injured patients is associated with worsening of survival. J Trauma
2008;64(2):335–40.
73. Stein SC, Georgoff P, Meghan S, et al. Relationship of aggressive monitoring
and treatment to improved outcomes in severe traumatic brain injury.
J Neurosurg 2010;112(5):1105–12.
74. Stein SC, Georgoff P, Meghan S, et al. 150 years of treating severe traumatic
brain injury: a systematic review of progress in mortality. J Neurotrauma
2010;27(7):1343–53.
75. Whitmore RG, Thawani JP, Grady MS, et al. Is aggressive treatment of
traumatic brain injury cost-effective? J Neurosurg 2012;Mar 6. Epub ahead
of print.
76. Wartenberg KE, Schmidt JM, Mayer SA. Multimodality monitoring in
neurocritical care. Crit Care Clin 2007;23(3):507–38.
77. Oddo M, Levine JM, Mackenzie L, et al. Brain hypoxia is associated with
short-term outcome after severe traumatic brain injury independently of
intracranial hypertension and low cerebral perfusion pressure.
Neurosurgery 2011;69(5):1037–45; discussion 1045.
78. Figaji AA, Zwane E, Thompson C, et al. Brain tissue oxygen tension
monitoring in pediatric severe traumatic brain injury. Part 1: relationship
with outcome. Childs Nerv Syst 2009;25(10):1325–33.
79. Maloney-Wilensky E, Gracias V, Itkin A, et al. Brain tissue oxygen and
outcome after severe traumatic brain injury: a systematic review. Crit Care
Med 2009;37(6):2057–63.
80. Nangunoori R, Maloney-Wilensky E, Stiefel MD, et al. Brain tissue oxygen
based therapy and outcome after severe traumatic brain injury: a systematic
literature review. Neurocrit Care 2011;Aug 16. Epub ahead of print.
81. Dohmen C, Bosche B, Graf R, et al. Prediction of malignant course in MCA
infarction by PET and microdialysis. Stroke 2003;34(9):2152–8.
82. Heiss WD, Dohmen C, Sobesky J, et al. Identification of malignant brain
edema after hemispheric stroke by PET-imaging and microdialysis. Acta
Neurochir 2003;86:237–40.
83. Sahuquillo J. Does multimodality monitoring make a difference in
neurocritical care? Eur J Anaesthesiol 2008;42:83–6.
84. Robertson CS, Valadka AB, Hannay HJ, et al. Prevention of secondary
ischemic insults after severe head injury. Crit Care Med 1999;27(10):2086–95.
85. Contant CF, Valadka AB, Gopinath SP, et al. Adult respiratory distress
syndrome: a complication of induced hypertension after severe head injury.
J Neurosurg 2001;95(4):560–8.
86. Harvey S, Harrison DA, Singer M, et al. Assessment of the clinical
effectiveness of pulmonary artery catheters in management of patients in
intensive care (PAC-Man): a randomised controlled trial. Lancet
2005;366(9484):472–7.
87. Richard C, Warszawski J, Anguel N, et al. Early use of the pulmonary artery
catheter and outcomes in patients with shock and acute respiratory distress
syndrome: a randomized controlled trial. JAMA 2003;290(20):2713–20.
88. Sandham JD, Hull RD, Brant RF, et al. A randomized, controlled trial of the
use of pulmonary-artery catheters in high-risk surgical patients. N Engl J
Med 2003;348(1):5–14.
89. Ospina-Tascon GA, Buchele GL, Vincent JL. Multicenter, randomized,
controlled trials evaluating mortality in intensive care: doomed to fail? Crit
Care Med 2008;36:1311–22.
90. Vincent JL. We should abandon randomized controlled trials in the
intensive care unit. Crit Care Med 2010;38(Suppl):S534–8.
91. Freeman JM. Beware: the misuse of technology and the law of unintended
consequences. Neurotherapeutics 2007;4(3):549–54.
92. Dickens BM, Cook RJ. The legal effects of fetal monitoring guidelines. Int J
Gynaecol Obstet 2010;108(2):170–3. Epub 2009, Nov 22.
93. Parascandola M, Hawkins J, Danis M. Patient autonomy and the challenge
of clinical uncertainty. Kennedy Inst Ethics J 2002;12(3):245–64.
94. Bernat J. Ethical aspects of determining and communicating prognosis in
critical care. Neurocritical Care 2004;1(1):L107–17.
95. Cruzan, by her parents and co-guardians, Cruzan et ux. v. Director,
Missouri Department of Health, et al. No. 88-1503 Supreme Court of the
United States497 U.S. 261; 110 S. Ct. 2841; 111 L. Ed. 2d 224; 1990 U.S.
LEXIS 3301; 58 U.S.L.W. 4916
96. Brown BA. The history of advance directives: a literature review. J Gerontol
Nurs 2003;29(9):4–14.
97. Franck M, Fry TP, Hill MK, et al. Uniform Health Care Decisions Act.
Available at: www.law.upenn.edu/bll/archives/ulc/fnact99/1990s/
uhcda93.htm (accessed July 26, 2012).
98. Seckler AB, et al. Substituted judgment: how accurate are proxy predictions?
Ann Intern Med 1991;115(2):92–8.
99. Zier LS, Burack JH, Micco G, et al. Surrogate decision makers’ responses to
physicians’ predictions of medical futility. Chest 2009;136(1):110–7.
100. Mason S, et al. Brief report on the experience of using proxy consent for
incapacitated adults. J Med Ethics 2006;32:61.
101. Alzheimer’s Association. Research consent for cognitively impaired adults:
recommendations for IRB and investigators. Alzheimer Dis Assoc Disord
2004;18:171.
102. Kim SY, et al. Proxy and surrogate consent in geriatric neuropsychiatric
research: update and recommendations. Am J Psychiatry 2004;162:777.
103. Gong MN, Winkel G, Rhodes R, et al. Surrogate consent for research
involving adults with impaired decision making: survey of Institutional
Review Board practices. Crit Care Med 2010;38(11):2146–54.
104. Best Hospitals. Available at: www.consumerreports.org/health/doctors-and-
hospitals/hospital-selector.htm (accessed July 26, 2012).
105. Boissy AR, Ford PJ, Edgell RC, et al. Ethics consultations in stroke and
neurological disease: a 7-year retrospective review. Neurocrit Care
2008;9(3):394–9.
106. Luce JM. A history of resolving conflicts over end-of-life care in intensive
care units in the United States. Crit Care Med 2010;38(8):1623–9.
107. Jacobs LM, Burns K, Bennett Jacobs B. Trauma death: views of the public
and trauma professionals on death and dying from injuries. Arch Surg
2008;143(8):730–5.
108. Roe et al. v. Wade, district attorney of Dallas County no. 70-18 Supreme
Court of the United States 410 u.s. 113; 93 S. Ct. 705; 35 L. Ed. 2d 147;
1973 U.S. LEXIS 159
109. California Health & Safety Code 7185 to 7194.5 (West Supp. 1994)
110. Jennett B, Plum F. Foreword in: Persistent vegetative state after brain
damage: a syndrome in search of a name. Lancet 1972;1:734–7, pp ix-x.
111. Jennett B. Foreward. In Wijdicks EFM, editor. Brain death. Philadelphia:
Lippincott Williams & Wilkins; 2001, pp. ix–x.
112. Pallis C. On the brainstem criterion of death. In: Youngner S, Arnold RM,
Shapiro R, editors. The definition of death: contemporary controversies.
Baltimore: Johns Hopkins University Press; 1999. p. 93–100.

Monitors During Anesthesia:
Effects of Anesthetic Agents
on Monitors
Jonathan McEwen, K.T. Henrik Huttunen, and Arthur M. Lam
Introduction
Need for Anesthesia and Monitoring
During Surgery
Monitoring is an essential component of patient care in the
operating room, and standard monitoring mandated by the
American Society of Anesthesiology includes blood pressure,
electrocardiogram, pulse oximetry, end-tidal CO2 (Et CO2),
and temperature. However, for the neurosurgical patient,
either the surgical procedure or the patient’s pathology may
necessitate specific monitoring of the central nervous system.
The interpretation of these monitored variables is potentially
subjected to the following influences: (1) pathophysiologic
changes (related to the patient’s disease or surgical manipula-
tion), (2) physiologic influences (related to changes that are
within “physiologic range”), and (3) anesthetic influences
(changes that are anesthesia related). Pathophysiologic inter-
pretation of these monitors is covered elsewhere in this text-
book, and “physiologic changes” are relatively minor. The
focus of this chapter is on the potential interactions and influ-
ence of anesthetic agents on these monitors.
Common Anesthetic Techniques
Although most neurosurgical procedures are performed under
general anesthesia, some procedures such as functional neu-
rosurgery require either a cooperative patient or one with
intrinsic cortical or subcortical electroencephalographic activ-
ity unaltered by any central nervous system depressants. In
general the anesthetic techniques can be classified into three
categories:
1. General anesthesia. The patient is rendered unconscious
in the conventional manner with the use of hypnotic/
potential inhaled or intravenous agents/narcotics with or
without neuromuscular blocking agents.
2. Regional anesthesia. A field block or regional block with
local anesthetic with and without use of sedative is used.
This has limited application in neurosurgical procedures,
© Copyright 2013 Elsevier Inc. All rights reserved.
Chapter
9  
I
and is primarily used for carotid endarterectomy or func-
tional neurosurgery where sedatives usually are omitted.
3. Monitored anesthesia care. Only light sedation is used in
these cases to supplement local anesthetic at the surgical
site. This technique may apply to epilepsy surgery or tumor
surgery that encroaches on eloquent areas where patient
cooperation is required. It also is used commonly in neu-
roradiologic procedures, some endovascular procedures,
or peripheral nerve surgery.
Anesthetic Agents
Anesthetic agents include hypnotics, analgesics, neuromuscu-
lar blockers, and other adjuncts. Their effects on interpreta-
tion of monitors are mediated by (1) direct effect—the specific
intrinsic influence of each anesthetic agent on the physiologic
activity of the brain (i.e., blood flow, metabolism, or func-
tion), or (2) indirect effect—changes in monitored variables
secondary to decreased perfusion of the brain secondary to
the systemic/cerebral effects of the anesthetic. The influence
is unique to the anesthetic agent and to the monitored vari-
able, and is discussed under the monitored modalities later.
The common anesthetic agents in current practice include the
following:
Hypnotics
Inhaled agents: desflurane, sevoflurane, isoflurane,
halothane, and nitrous oxide (N2O)
Intravenous agents: propofol, thiopental, benzodiaze-
pines, ketamine, and etomidate
Analgesics
Morphine and other synthetic opiates, nonsteroidal
anti-inflammatory drugs, gabapentin, and pregabalin.
Neuromuscular blockers
Short-acting agents, such as succinylcholine; inter­
mediate-duration agents, such as rocuronium and
vecuronium; and long-acting agents, such as pan-
curonium
Other adjuncts
Neuroleptics and α-2 agonists, such as dexmedetomi-
dine
71
72 Section I—Background
Monitors and the Influence
of Anesthetic Agents
Cerebral Blood Flow and Surrogates
Continuous direct measurement of cerebral blood flow (CBF)
continues to be an elusive goal, although laser Doppler1 and
thermodilution flow meters2 are available. They can be diffi-
cult to use and are not widely used during surgical procedures.
Anesthetic agents do not directly influence the function of
these monitors, but can alter CBF (see Transcranial Doppler
Ultrasonography).
Jugular Venous Oximetry
Jugular venous oxygen saturation reflects the balance between
CBF and cerebral oxygen metabolism. Thus a higher jugular
venous oxygen saturation (>70%) suggests flow in excess of
metabolic need, and conversely, a low saturation (<50%) indi-
cates increased extraction and is suggestive of ischemia. Anes-
thetic agents can influence this balance between flow and
metabolism. Inhaled agents tend to result in higher jugular
venous saturation, whereas intravenous agents such as propo-
fol result in lower saturation.3 More important than the anes-
thetic influence is the cerebrovascular response to change in
carbon dioxide (CO2) tension; thus hyperventilation can
result in low jugular venous saturation, and hypoventilation
may have the opposite effect. It is generally considered prudent
to maintain jugular venous saturation greater than 50%.
Although not a widely adopted practice, the use of jugular
venous oximetry is considered a useful intraoperative monitor
for craniotomy at some centers where hyperventilation
is used.4
Intracranial Pressure and
Cerebral Perfusion Pressure Monitor
Monitoring of intracranial pressure (ICP) provides useful
information about the status of the intracranial compartment,
and just as importantly the net perfusion pressure to the brain
(cerebral perfusion pressure = mean arterial pressure − ICP).
Anesthetic agents do not per se interfere with the measure-
ment of ICP, but they can alter ICP by causing vasodilation or
vasoconstriction. In general, inhaled agents cause either no
change or a dose-related increase in CBF, whereas intravenous
agents such as propofol cause vasoconstriction and so a reduc-
tion in ICP (ketamine is an exception). The Prx index, that is,
the moving correlation coefficient between mean ICP and
mean arterial blood pressure, can provide information about
autoregulation.5 However, the effects of anesthetics on Prx
remain unclear.
Transcranial Doppler Ultrasonography
Although not widely used in the operating room because of
technical and labor-intensive demands, transcranial Doppler
(TCD), when properly applied, can be a useful continuous
monitor of relative changes in CBF.6
TCD is a noninvasive technique that assesses blood flow
velocity (BFV) and direction in the cerebral arteries by means
of a pulsed ultrasonic beam. Aaslid et al first described the
TCD technique in 1982.7 Today there are multiple clinical
applications for TCD that are described in detail in Chapter
30. This chapter addresses the use of TCD as an intraoperative
monitor and effects of anesthetic agents on TCD.
TCD has the ability to noninvasively assess local CBF veloci-
ties in the major intracranial arteries. This determination of
cerebral blood velocities allows for an indirect assessment of
CBF, and TCD can quantitatively measure relative changes in
CBF. This property makes TCD an appealing intraoperative
monitor, in that maintenance of adequate cerebral perfusion
is a goal in any operative procedure.
Bilateral insonation of the middle cerebral arteries (MCAs)
through the temporal bone (temporal window) is the most
commonly used intraoperative application of TCD. This
allows for comparison of blood flow velocities between
hemispheres, emboli monitoring, assessment of cerebral auto-
regulation, and CO2 reactivity. However, TCD has several
limitations that prevent routine application of it as an intra-
operative monitor:
1. In neurosurgical patients there is limited access to acoustic
windows required for accurate flow velocity measurements
because of the surgical incision sites and the sterile surgical
field during a craniotomy. This limitation, however, may
not apply during a carotid endarterectomy.
2. Patient positioning complicates placement of the ultra-
sound transducers in stable positions. For reliable, con-
tinuous assessment of cerebral BFV, TCD probes require a
constant insonation angle (CBF velocity is proportional to
the cosine of the angle of insonation). This requires the
use of a fixation device, such as the LAM-Rack, to stabilize
the ultrasound transducers, because handholding tech-
niques or intermittent measurements are unreliable.
3. About 20% of patients have inadequate acoustic windows
because of temporal bone hyperostosis.
Despite its limitations, TCD can be a valuable intraoperative
monitor. It is particularly useful during carotid endarterec-
tomy,8-11 in which it can be more accurate than near infrared
spectroscopy (NIRS), stump pressure and somatosensory
evoked potentials (SSEP) in patients at risk for intraoperative
cerebral hypoperfusion or emboli.12 TCD also allows for
assessment of postoperative embolic phenomenon13-19 and
hyperperfusion.20-25 Both of these conditions are associated
with postoperative neurologic insult and stroke.26-27
Anesthetic Effects on Blood Flow Velocity
Unlike other neurophysiologic monitoring techniques, TCD
is relatively resistant to interference from anesthetic agents.
However, changes observed in CBF under anesthesia may
reflect physiologic changes of the patient associated with anes-
thetic influence. Consequently, TCD has been used to study
neurophysiologic effects of anesthetic agents.28 Table 9.1 sum-
marizes these anesthetic physiologic effects measured by TCD.
Inhalational Anesthetics
CBF-metabolism coupling is generally preserved during inha-
lation anesthesia, and has been extensively studied using TCD,
because changes in cerebral blood volume reflect correspond-
ing CBF changes. Although flow-metabolism coupling is pre-
served with volatile anesthetics29; the net effect of a volatile
 Section I—Background 73

Table 9.1  Anesthetic Effects on Cerebral Hemodynamics Measured by Transcranial Doppler

Anesthetic Agent CBFV Autoregulation CO2 Reactivity References
N2O ↑ Impaired Preserved 38-41
Isoflurane (0.5-1.5 MAC)  Preserved at 1 MAC,  >1MAC Preserved at 1 MAC 26, 28, 35, 37
Desflurane (0.5-1.5 MAC)  Preserved at 1 MAC, Impaired Preserved at 1 MAC 32-34
at 1.5 MAC
Sevoflurane (0.5-2 MAC)  Preserved at 1.5 MAC Preserved at 2 MAC 27-31
Midazolam (0.15 mg/kg bolus) ↔ 45
Thiopental (5 mg/kg bolus) ↓ 45
Propofol (2 mg/kg bolus) ↓ 45
Propofol (25-200 mcg/kg/min)  Preserved Preserved 42, 43, 44, 100
Etomidate (0.3 mg/kg bolus) ↓ Preserved 51-53
Ketamine (1.5 mg/kg bolus) ↑/↔ Preserved 45, 47, 48
Dexmedetomidine (various doses)  Impaired Impaired 49, 50
Fentanyl (various doses) ↔ Preserved 45, 55
Remifentanil (various doses) ↔ Preserved Preserved 42, 56
Sufentanil (0.1-0.2 vs 6 mcg/kg) ↔/ 54, 55

CBFV, Cerebral blood flow velocity; ↑, increased; ↓, decreased; , dose dependent increase; , dose dependent decrease; ↔, minimal effect; MAC, minimum alveolar
concentration.

anesthetic on CBF depends on the balance between the agent’s
direct vasodilatory effect and flow-metabolism coupling-
mediated vasoconstriction.30
Sevoflurane is a commonly used volatile anesthetic during
neurosurgical procedures because it has relatively low blood-
gas solubility that allows rapid titration and emergence, and
has favorable cerebral hemodynamic effects compared with
other volatile anesthetics. Sevoflurane, similar to other volatile
anesthetics, is a central nervous system depressant that can
decrease the cerebral metabolism. This depressant effect
causes TCD-measured cerebral BFV to decrease secondary to
flow-metabolism coupling.31 As concentrations of sevoflurane
are increased, TCD flow velocities tend to increase, but to a
lesser extent than other volatile agents32; this reflects the influ-
ence of the intrinsic cerebrovasodilatory effect of sevoflurane33
and can result in luxury perfusion states with concentrations
greater than 1.5 minimum alveolar concentration (MAC). It
is unusual that clinical concentrations are required above
this level. Furthermore, at 1.5 to 2 MAC, cerebral autoregula-
tion mechanisms34 and cerebral reactivity to CO2 are well
preserved.35
Desflurane has the lowest blood-gas solubility coefficient of
the halogenated ether anesthetics, making it desirable for use
in neuroanesthesia because it facilitates a rapid emergence.
However, cerebral physiologic effects of desflurane are not as
favorable as for sevoflurane.36 Desflurane causes a dose-
dependent increase in cerebral BFVs; this takes effect at much
lower MAC values than for sevoflurane. To reduce the luxury
perfusion, increased cerebral blood volume and potential
increase in ICP that can be caused by desflurane, concentra-
tions should be limited to 1 MAC. Cerebral autoregulation
and CO2 reactivity are preserved at this clinically relevant
concentration.37,38
Isoflurane has a long history of use as a neuroanesthetic
agent. Similar to other volatile agents, isoflurane has intrinsic
dose-dependent cerebral vasodilatory properties.32 When
unopposed by flow-metabolism coupling-mediated vasocon-
striction, isoflurane, in concentrations of 0.5 to 1.5 MAC,
causes significant increases to cerebral BFVs.30 The effect of
isoflurane on cerebral autoregulation is dose dependent, and
this homeostatic mechanism is preserved up to concentrations
of 1 MAC,39 beyond which it is impaired. However, this
depression of cerebral autoregulation can be restored with
hyperventilation.40 CO2 reactivity is maintained at 1 MAC.41
N2O is most commonly used as an adjuvant to other anes-
thetic agents, because hyperbaric conditions are required to
reach anesthetic concentrations with N2O alone. N2O is a
potent cerebral vasodilator, and results in increased cerebral
BFVs.42 N2O also increases cerebral metabolism and can
increase CBF through vasodilation.43 Furthermore, cerebral
autoregulation is impaired with N2O.44 Together these factors
make N2O an undesirable agent in patients with limited intra-
cranial compliance. However, cerebrovasomotor reactivity to
CO2 is preserved.45
Intravenous Anesthetics
Propofol is commonly used as an anesthetic induction
agent and to maintain anesthesia. It is easily and rapidly titrat-
able and well tolerated for use during evoked potential moni-
toring, and its cerebrovascular effects are favorable in the
neurosurgical patient. Propofol causes a dose-dependent
decrease in cerebral BFVs,46 supporting its preservation of
flow-metabolism coupling. Propofol also has direct cerebral
vasoconstricting properties that contribute to the observed
BFV decreases. Cerebral autoregulation and CO2 reactivity are
well preserved.47,48
Thiopental is a barbiturate anesthetic, which like propofol,
can be used for both induction and maintenance of anesthe-
sia. However, thiopental’s pharmacodynamic profile can be
74 Section I—Background
undesirable in the neurosurgical patient in that it delays emer-
gence. Nevertheless, thiopental is frequently used as a neuro-
protective agent during aneurysm surgery (e.g., during
temporary arterial occlusion), because it decreases cerebral
metabolism with the induction of burst suppression. Induc-
tion doses of 5 mg/kg decrease cerebral BFV.49
Ketamine is an intravenous anesthetic agent with dissocia-
tive anesthetic properties. It has a controversial history in the
neurosurgical patient in that older evidence showed ketamine
could increase CBF velocities, cerebral metabolism, and ICP.50
Ketamine increases cerebral BFV during induction,49 but as an
adjuvant to an anesthetic under controlled ventilation, ket-
amine does not increase cerebral BFV.51 When used with pro-
pofol as part of a total intravenous technique, ketamine
preserves cerebral autoregulation.52
Dexmedetomidine is a selective α-2 agonist used as a seda-
tive in intensive care units and in the operating room for
procedures such as awake craniotomies and carotid endarter-
ectomies. It has sedative properties with minimal respiratory
depression that makes it appealing for these applications. Dex-
medetomidine decreases cerebral BFVs in a dose-dependent
manner,53,54 which appears to result from decreased cerebral
metabolic rate with preserved flow-metabolism coupling.54
However, at sedative doses, dexmedetomidine may cause a
slight decrease in both dynamic cerebral autoregulation53 and
CO2 reactivity.54
Etomidate is an imidazole derivative intravenous anesthetic
agent commonly used as an induction agent, particularly in
trauma patients or those with limited cardiac reserve, because
of its hemodynamic stability. Induction doses of etomidate
lead to decreased middle cerebral artery BFVs.55,56 CO2 reactiv-
ity is preserved during etomidate infusion57; however, infu-
sions are not recommended because there may be adrenal
suppression and propylene glycol toxicity.
Midazolam is a commonly used intravenous benzodiaze-
pine. It is used frequently as a preoperative anxiolytic or as an
infusion in intensive care units as a sedative. There is conflict-
ing evidence on the effects of midazolam on cerebral BFVs. In
one study midazolam at 0.15 mg/kg with fentanyl did not
affect MCA BFVs.49 In other studies, however, at a much lower
dose of 20 and 40 µg/kg, cerebral BFVs decreased about 20%
but returned to baseline in 5 minutes.58
Opioids are used frequently during neurosurgery for mul-
tiple reasons: to blunt sympathetic responses to stimulation,
to treat pain, and as adjuvants to help reduce the doses of
other anesthetic drugs. In clinically relevant concentrations
with controlled ventilation, opioids have limited effects on
cerebral physiology. Fentanyl and remifentanil do not signifi-
cantly change cerebral BFVs. CO2 reactivity and cerebral auto-
regulation also remain intact.46,49,59,60 Low doses of sufentanil
(0.1 to 0.2 µg/kg) do not affect cerebral BFVs; however,
increasing doses can decrease cerebral BFV in a dose-
dependent fashion.58,59
Cerebral Function
Electroencephalography (Unprocessed
and Computer Processed)
Electroencephalogram (EEG) activity can be monitored intra-
operatively for a variety of reasons, such as to detect cerebral
ischemia or to localize seizure activity. However, all EEG
correlates or derivatives are influenced by anesthetic agents.
Both intravenous anesthetics and inhalation anesthetics cause
dose-related depression that ultimately may result in electrical
silence. Low-dose inhalation agents can cause initial stimula-
tion with an increase in EEG activity in the alpha band. With
progressive increase in depth, there is an increase in delta
wave, and disappearance of beta and alpha bands. With further
increase in anesthetic depth, burst-suppression progressing to
electrical silence will occur. The dose-response curve is differ-
ent among the inhaled anesthetics. Isoflurane will cause burst-
suppression at 1.5 MAC and electrical silence at 2 MAC,
whereas sevoflurane causes burst-suppression at 2 MAC and
electrical silence at 2.5 MAC.61,62 There also is an age-anesthetic
effect interaction, with lower anesthetic doses needed to
achieve burst suppression in older patients.63 N2O paradoxi-
cally can cause an increase in EEG activity when added to an
inhalation anesthetic. Seizure activity also has been reported
with most anesthetic agents.64-66 Even though some agents
have a higher propensity to cause epileptogenic activity, the
clinical significance in unclear. For example, inhalation anes-
thetics including sevoflurane and enflurane can cause epilep-
togenic spikes in healthy patients without a history of
epilepsy.67 Seizure activity can interfere with monitoring of
computer-processed EEG such as the Bispectral Index. Intra-
venous agents in general do not cause stimulation and slow
waves develop early. Both thiopental and propofol at increas-
ing doses will result in burst suppression pattern followed by
electrical silence.68 This is used often for brain protection
during periods of potential ischemia.
Because anesthetic agents can have a profound influence on
EEG, anesthetic depth must be maintained at a relatively steady
state to reliably monitor cerebral ischemia, to allow meaning-
ful interpretation of changes. These agents and benzodiaze-
pines must be avoided during functional surgery when
mapping of electrical activity or epileptic foci is important. For
both deep brain stimulator implantation and resection of epi-
leptic foci in patients who are otherwise awake, dexmedetomi-
dine as the sole sedative agent has been used with some success,
without interfering with electrophysiologic recording.69,70
Evoked Potentials
Multimodality evoked potentials increasingly are used intra-
operatively to monitor and help preserve integrity of the
central nervous system structures placed at risk by the surgical
procedure. Anesthetic agents have unique and profound influ-
ence on evoked potentials, and these effects must be under-
stood to allow meaningful recording and interpretation of
these signals. The modalities include SSEP, motor evoked
potentials (MEP) and cranial nerves with motor components
(V, VII, IX, X, XII).
Somatosensory Evoked Potential Monitoring
An important clinical decision when using SSEP monitoring
is the choice between inhalational and intravenous anesthetic.
Even though SSEPs are not degraded to the same extent as
MEPs by inhalational anesthetics, they exhibit a dose-related
increase in latency and decrease in amplitude under volatile
anesthesia. This is particularly true for cortical SSEPs that
involve multiple synapses. Brainstem and spinal components
are relatively resistant to anesthetic effects. When volatile
 Section I—Background 75

Table 9.2  Effects of Anesthetic Agents on Brainstem Auditory Evoked Potentials and
Somatosensory Evoked Potentials
BAEP SSEP
Latency Amplitude Latency Amplitude
Thiopental
Low dose ↔ ↔ ↑ ↓
High dose
↑ ↓ ↑ 

Propofol ↑ ↔ ↑ ↔ or ↓
Dexmedetomidine ? ? ↔ ↔
Midazolam ? ? ↔ ↓
Etomidate ↔ ↔ ↑ 

Ketamine ↔ ↔ ↔ 

Fentanyl ↔ ↔ ↑ ↓
Remifentanil ↔ ↔ ↔ ↔
Halogenated volatile agents
0.5 Mac
↑ ↔ ↔ or ↑ ↔ or ↓
>1 Mac
 

Nitrous oxide ↔ ↔ ↔ 

The relative influence of anesthetics on brainstem auditory evoked potentials (BAEP) and somatosensory evoked potentials (SSEP) is depicted. Halogenated volatile
agents referenced include desflurane, sevoflurane, and isoflurane.
↔, No change; ↑, clinically insignificant increase; ↓, clinically insignificant decrease; , clinically significant increase; , clinically significant decrease; ?, no
information.

agents are used, the dose should be kept below 1 MAC to
maintain SSEP signal quality. This may be facilitated by the
coadministration of potent opiates, which do not affect signal
quality. The addition of N2O typically is avoided, because this
further degrades the signal.71 In addition, the anesthetic con-
centration should be kept constant during critical periods to
reduce anesthetic influence on the interpretation of changes.
The use of intravenous anesthetics, such as propofol, thiopen-
tal, or midazolam infusion, often results in better signal quality
than with use of volatile agents. Ketamine and etomidate, in
particular, have been reported to increase the amplitude of
SSEPs, although the effect that this has on sensitivity and
utility remains a matter of speculation. Regardless of the anes-
thetic regimen, it is important to avoid bolus doses or rapid
changes in anesthetic depth during key portions of an inter-
vention, such as vessel occlusion or manipulation of the spine,
to avoid confounding influences on signal integrity at a critical
time when the monitoring modality is most useful.
Other factors, such as blood pressure, temperature, and
patient position, may influence evoked potential monitoring.
Nerve injury associated with patient position, such as crush
injury at pressure points, or traction injury must be consid-
ered. A decrease in blood pressure may also alter SSEPs and
even mild hypothermia can increase latency and decrease the
amplitude of evoked potentials because thermoregulatory
behaviors and physiologic responses are blunted in the anes-
thetized patient.
Brainstem Auditory Evoked
Potential Monitoring
Brainstem auditory evoked potentials (BAEPs) are resistant to
anesthetic effects and can be recorded with virtually any
anesthetic technique. Inhalation agents can increase the
latency but do not abolish BAEPs.72 The effects of anesthetic
agents on BAEP and SSEP are summarized in Table 9.2.
Motor Evoked Potentials
MEPs are electrical potentials monitored along the motor
pathway after stimulation of the motor cortex or spinal cord.
MEPs are elicited by direct or transcranial stimulation with
either electrical (TcEMEP) or magnetic (TcMMEP) tech-
niques. Responses can be recorded over the spinal cord from
electrodes placed in the epidural space (spinal D waves or
epidural MEP), along peripheral motor nerves (neurogenic
MEP), or as compound muscle action potentials (CMAPs)
over muscle (myogenic MEP). Myogenic MEPs are large
amplitude biphasic potentials in response to motor unit depo-
larization, and so, unlike SSEPs, signal averaging is not
required. This allows for near instant motor pathway feedback.
TcMMEPs are exquisitely sensitive to anesthetic agents and so
are unsuitable for intraoperative monitoring.
The entire motor pathway from cerebral motor cortex
through brainstem, corticospinal tracks, anterior horn,
peripheral nerve, and neuromuscular junction to skeletal
muscle can be assessed using transcranial stimulation with
myogenic MEP monitoring. Motor pathway injury can result
from ischemia, or disruption, retraction, compression, or dis-
traction of neural tissue. Each of these insults can cause func-
tional alterations that may be neurophysiologically evident.73
MEPs therefore are an appealing intraoperative monitor and
are applicable to any surgical procedure which places the
motor pathway at risk for injury: craniotomy, spinal surgery,
or thoracoabdominal aortic aneurysm repair. Intraoperatively,
MEPs are usually monitored with SSEPs.
76 Section I—Background
Isoflurane 0%
Fig. 9.1  Graphs that illustrate augmentation of
motor evoked potentials (MEP) responses when
using multipulse stimuli with increasing isoflurane
doses. (Ubags LH, Kalkman CJ, Been HD. Influence of
isoflurane on myogenic motor evoked potentials to
single and multiple transcranial stimuli during nitrous
oxide/opioid anesthesia. Neurosurgery. 1998;43(1):
90–4.)
MEPs are useful to monitor for ischemia, particularly of
the anterior horn cells (alpha motor neurons), which are
thought to be most sensitive to ischemic insults.73 Many pub-
lications describe the use of MEP for intraoperative monitor-
ing, although there are no clearly defined thresholds for
latency or amplitude changes that correlate with risk of
motor pathway injury. In general, latency is robust to anes-
thetic or surgical insults; therefore amplitude changes are
assessed more commonly. The threshold considered concern-
ing for injury is the complete loss of signal rather than a fixed
percentage decrease in amplitude (e.g., 50%). Other authors
consider the increase in stimulation energy required to repro-
duce baseline signals as the marker for injury.74 The safety of
MEPs is well known.75 However, induction of seizures is a
potential risk with TcEMEPs.
Anesthetic Effects
MEPs are sensitive to anesthetics; therefore a tailored approach
is required to select the most appropriate anesthetic regimen.76
Transcranial magnetic stimulation–generated MEPs are more
sensitive to anesthetics than transcranial electrical (TcE) stim-
ulation, and usually are abolished with induction, regardless
of the agents used. TcE stimulation therefore is the technique
used exclusively during intraoperative monitoring. Transcra-
nial magnetic stimulation currently has a very limited role in
the operating room. Early work with TcEMEPs showed that
single-pulse signals were extremely sensitive to volatile agents
even at subclinical doses.77,78 Subsequently, multipulse stimu-
lation techniques have been shown to improve both low-
amplitude baseline signals and MEP signals depressed by
anesthesia (Fig. 9.1). Table 9.3 summarizes how different
anesthetic agents affect multipulse TcEMEPs, which is the
current standard during intraoperative monitoring. “Anes-
thetic fade,” a time-dependent decrease in MEP signal ampli-
tude proportionate to the length of surgery regardless of the
anesthetic regimen can occur.79 Increasing the stimulation
1 stimulus 3 stimuli 5 stimuli
0.2%
0.4%
0.5 mV
0.6%
0 100 Time (msec) 0 100 Time (msec) 0 100 Time (msec)
energy may attenuate anesthetic fade, but increases the risk for
potential false-positive signal changes.
Volatile anesthetics are used commonly to maintain anesthe-
sia. All currently used halogenated anesthetics cause signifi-
cant dose-dependent depression of myogenic MEP signals.80-85
This depression is believed to result from volatile anesthetic-
induced inhibition of pyramidal activation of spinal motor
neurons.86 At full anesthetic concentrations of 1 MAC, multi-
pulse TcEMEPs with isoflurane are only recordable in 8% of
subjects.82 TcEMEP in response to paired TcE stimulation
during sevoflurane anesthesia at 1 MAC is unrecordable,84
whereas with a four-pulse TcE stimulation this result improves
to more than 80%.85 However, this is still inadequate for clini-
cal use. With desflurane or sevoflurane at 0.5 MAC with either
N2O or a narcotic infusion, MEPs are recordable in 100% of
patients,81,83 and with isoflurane MEP monitoring is possible
in 90%.78 Total intravenous anesthetic regimens therefore
seem preferable and only one study to date has shown ade-
quate clinical recording conditions with a volatile anesthetic
compared to total IV technique.81
N2O is used frequently as an anesthetic adjuvant. When
50% N2O is used with intravenous infusion of narcotic with
or without propofol, multipulse TcEMEP amplitudes are
reduced by about half, but can be consistently recorded.87,88
With concentrations ranging from 20% to 60%, N2O is associ-
ated with a dose-dependent decrease in MEP amplitude, yet
signals remain recordable in 100% of subjects.89 Given these
findings, N2O is an acceptable anesthetic adjuvant during
MEP monitoring when administered with other agents that
have limited depressive effects.
Propofol is currently the most popular anesthetic agent to
maintain anesthesia during MEP monitoring in that it is easily
titratable and has a favorable pharmacokinetic profile. Titrat-
ing propofol infusion rate intraoperatively to achieve clinically
relevant concentrations does not affect MEP responses in
a clinically significant fashion.90 However, dose-dependent
depression of myogenic MEPs can be demonstrated at higher
 Section I—Background 77

Table 9.3  Anesthetic Effects on Transcranial Electrical Stimulation on Motor Evoked

Potential Responses
Change in Amplitude % of Subjects with
Anesthetic Agent TcE Pulses from Baseline Response References
Isoflurane (0.17 MAC) 5 ↓↓ 100% 81
Isoflurane (0.33 MAC) 5 ↓↓ 100% 81
Isoflurane (0.5 MAC) 5 ↓↓↓ 90% 81
Isoflurane (0.75 MAC) 4 NA 61% 82
Isoflurane (1 MAC) 5 NA 8% 83
Desflurane (0.5 MAC) 5 NA 100% 84
Sevoflurane (0.25 %) 2 ↓↓↓ 85% 85
Sevoflurane (0.5 MAC) 2/4 ø/NA 55%/100% 85, 86
Sevoflurane (0.75 MAC) 2/4 ø/NA 10%/90%–100% 85, 86
Sevoflurane (1 MAC) 2/4 ø/NA 0%/80%–90% 85, 86
N2O (50%) 2 ↔ 100% 88, 89
N2O (20%, 40%, 60%) 6  100% 90
Midazolam (0.05 mg/kg bolus) 1 ↓↓↓ 100% 92
Thiopental  NA 101
Propofol (67 µg/kg/min) 5 NA 100% 95
Propofol (>100 µg/kg/min) 4 97% 82
Etomidate (0.1 mg/kg bolus) 3 ↓ 100% 94
Etomidate (5-10 µg/kg/min) 1 100% 96
Ketamine (0.5 mg/kg bolus) 3 ↔ 100% 94
Ketamine (1 mg/kg/hr) 5 NA 100% 96
Dexmedetomidine (1 µg/kg bolus) ↔ 100% 97
Dexmedetomidine (0.15 µg/kg/hr) 7 ↔ 100% 98
Dexmedetomidine (~0.3 µg/kg/hr) 7 ↔ 100% 98
Fentanyl (3 µg/kg bolus) 1 ↔ 100% 92
Remifentanil (9 ng/mL)  100% 101
Sufentanil (0.5 µg/kg/hr) 3 100% 81
Neuromascular blockade  /ø
T1 45%–55% of baseline 6 ↔ 100% 102
T1 5%–15% of baseline 6 ↓↓↓ 100% 102

↔, Minimal/no change; ↓, decrease 5% to 30% ; ↓↓, decrease 30% to -60%; ↓↓↓, decrease greater than 60%; ø, absent or less than 10%; , dose-dependent
decrease; MAC, minimum alveolar concentration; blank indicates no reliable data.

doses.76 Bolus administration of propofol can lead to MEP
signal loss. However, since propofol has rapid redistribution
and metabolism, signals return within minutes91 (Fig. 9.2).
Therefore during intraoperative MEP monitoring propofol is
preferable to many of the other intravenous agents because of
its desirable neurophysiologic effects and pharmacokinetic
profile.
Midazolam boluses of 50 µg/kg cause significant depression
of single-pulse TcEMEP signals; the signals, however, are still
recordable in all patients.91 Infusions of 0.1 mg/kg/hr of mid-
azolam do not depress TcMMEP signals.76 Studies have not
been done in humans with multipulse TcE, but one would
expect signals to be better than those described for TcM or
single pulse TcE. Other benzodiazepines have similar effects
as midazolam.
Ketamine can be used in boluses (0.5 mg/kg) or as continu-
ous infusions (1 mg/kg/hr) without altering MEP responses.92-94
Excellent recording conditions are achieved in all cases.
However, ketamine has dissociative anesthetic side effects
and potentially undesirable or controversial neurophysiologic
effects. These factors need to be balanced with the desire for
reliable intraoperative MEP signals.
Etomidate has appealing hemodynamic stability for anes-
thesia induction. It also can be used as a continuous infusion
to maintain sedation. However, it may contribute to adrenal
insufficiency and propylene glycol toxicity. Nevertheless,
etomidate allows for excellent recording conditions for MEPs,
with only a short duration of suppression following bolus
administration and stable recording conditions when given as
an infusion.91,93,95
78 Section I—Background
Propofol
+
2 mV
–
Control
2 min
5 min
10 min
Fig. 9.2  The effect of intravenous boluses of
anesthetic agents on motor evoked potentials 30 min
(MEPs). (Kalkman CJ, Drummond JC, Ribberink AA,
et al. Effects of propofol, etomidate, midazolam, and
fentanyl on motor evoked responses to transcranial
0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100
electrical or magnetic stimulation in humans. Anesthe-
siology. 1992;76(4):502–9.)
Dexmedetomidine is a relatively new sedative and anesthetic
agent. The effects of dexmedetomidine on MEP amplitudes
have been studied in patients undergoing spine surgery.96-98
Earlier studies showed that a standard infusion rate of dexme-
detomidine has little effect on MEP amplitude and signals are
recordable in 100% of subjects.96,97 However, a more recent
study suggests that dexmedetomidine may cause a clinically
significant dose-related decrease in MEP amplitude98 and loss
of signals has been described in pediatric spine surgery.99
Dexmedetomidine has propofol-sparing effects; therefore a
low dose infusion may be a useful adjunct during propofol
anesthesia when MEPs are monitored, but the effects of chang-
ing infusion rate/plasma concentration must be taken into
account.
Opioids are administered commonly with other anesthetic
agents. Opioids frequently are used when MEP monitoring is
required because they have minimal effects. A single bolus of
fentanyl at 3 µg/kg does not decrease MEP amplitude.91 Suf-
entanil at an infusion rate of 0.5 µg/kg/hr allows reliable
recordings in 100% of patients.80 Other investigations have
shown that opioids decrease myogenic MEPs in a dose-
dependent manner; however, at target-controlled plasma con-
centrations remifentanil has the least effect.100
Neuromuscular blockade generally is used to facilitate tra-
cheal intubation or provide an immobile patient optimal sur-
gical exposure. However, boluses of muscle relaxants
inevitably abolish MEPs, in that they block signal transduc-
tion at the neuromuscular junction. Careful infusions of neu-
romuscular blockers to achieve 20% to 50% maintenance of
single twitch height have been shown to be compatible with
reliable MEP recordings when used with a minimally sup-
pressant anesthetic and multipulse TcE. However, deeper
blockade leads to unsatisfactory signals.101 To improve MEP
amplitude, a posttetanic stimulation can be used.102,103
Caution is necessary in patients with preoperative neurologic
deficits because they appear to be more sensitive to the
depressant effects of neuromuscular blocking drugs.95 It gen-
erally is acceptable and reasonable to use a single dose of a
short-acting neuromuscular blocking agent at the time of
induction because its effects should wear off by the time
intraoperative MEP recording begins. The exception is when
a baseline recording is desired. However, maintenance of
Midazolam Etomidate Fentanyl
Time (msec)
neuromuscular blockade or rebolusing can limit interpreta-
tion of MEPs and should be avoided.
Cerebral Metabolism
Two modalities can be used intraoperatively to estimate cere-
bral oxygen metabolism: jugular venous oximetry (see also
Chapter 32) and transcranial NIRS (see also Chapter 33). Both
modalities provide information on the balance between flow
and metabolism rather than on cerebral oxygen metabolism
itself. Jugular venous oximetry is discussed in the section on
CBF in this chapter.
Near Infrared Spectroscopy
NIRS is a noninvasive technique that can measure cerebral
oxygen saturation, cerebral blood volume, and even mitochon-
drial redox states using differential absorption of near infrared
light (700 to 1000 nm).104 It is an evolving technology that is
starting to be adopted for intraoperative neurophysiologic
monitoring, and so its role and utility is only beginning to be
elucidated. NIRS is used most commonly in cardiac surgery
during cardiopulmonary bypass and during carotid endarter-
ectomy. In carotid endarterectomy it is used as an ischemia
monitor during carotid cross clamping and appears to be
effective and comparable to other commonly used monitors.
However, relative and absolute thresholds for changes in
regional cerebral oxygen saturation remain to be defined.105
NIRS also has been used during neuroendovascular proce-
dures for cerebral aneurysm embolization106 and preopera-
tively to assess patients with skull base tumors and giant
aneurysms to determine cerebral oxygen reserve during vessel
test occlusion.107 NIRS has several theoretical advantages that
make it an appealing intraoperative monitor: (1) it is noninva-
sive, (2) it allows for continuous assessment, (3) NIRS machines
are portable, and (4) spectroscopy is unaffected by anesthetics.
However, there also are several disadvantages: (1) it can only be
applied to the forehead, (2) the signal is contaminated by
extracranial absorption and scattering, and (3) the signals can
be affected by ambient light. NIRS is a promising technology,
but further research is required to demonstrate its clinical effi-
cacy and justify its routine use during anesthesia.108
 Section I—Background 79

Anesthetic Effects Volatile Anesthetics

Measurement of oxyhemoglobin and deoxyhemoglobin by The observed NIRS changes seen with the halogenated inhala-
spectroscopy is unaffected by anesthetics. However, anesthetics tional agents support findings associated with the neurophysi-
can affect cerebral metabolism and CBF. Therefore indirect ologic effects of these agents. Desflurane, sevoflurane, and
changes can be observed in NIRS monitoring that result from isoflurane have all been examined at clinically relevant concen-
the pharmacodynamic and consequent physiologic effects of trations and at 1 MAC do not affect rSO2112-114 despite decreases
each anesthetic agent. Because NIRS is a relatively new and in blood pressure and cardiac output. These results demon-
evolving monitoring modality, there is limited evidence about strate that cerebral autoregulation is maintained at these con-
observed NIRS changes that occur with different anesthetic centrations. Both desflurane and sevoflurane can result in a
agents. The known effects are summarized in Table 9.4. Intra- dose-dependent increase in rSO2113 suggesting either a shift in
operative use of intravenous dyes such as indigo carmine109 the flow-metabolism coupling or cerebrovasodilation at higher
and indocyanine green110,111 can cause low regional cerebral concentrations. Sevoflurane, at inhalational induction concen-
oxygen saturation (rSO2), similar to that observed with pulse trations of 8%, has been shown to result in cerebral hyperemia,
oximetry. although these results are confounded by the use of N2O.115
The effects of N2O on NIRS have not been specifically studied;
however, N2O has been used as an anesthetic adjuvant in many
Table 9.4  Observed Changes in Regional of the studies published on NIRS and may be a confounding
Cerebral Oxygen Saturation Under Various factor.
Anesthetics
Anesthetic Agent Change in rSO2 References Intravenous Anesthetics
Isoflurane (~IMAC) ↔ 114 Because of the ability to rapidly and fairly predictably achieve
Isoflurane (~IMAC) ↔ 113 a state of unconsciousness, bolus administration of intrave-
nous anesthetic commonly facilitates induction of anesthesia.
Desflurane (>IMAC)  113
However, limited information is available on the effect of dif-
Sevoflurane (~IMAC) ↔ 112 ferent intravenous agents on rSO2. There is some evidence that
Sevoflurane (>IMAC) 113,115 sedative doses of midazolam (0.05 mg/kg), do not change

rSO2 .113 One study examined the induction effects of etomi-
Midazolam (0.05 mg/kg ↔ 114 date, propofol, and thiopental on cerebral oxygenation and
bolus)
found that propofol and thiopental slightly increased rSO2
Thiopental (6 mg/kg bolus) ↑ 116 while etomidate caused a small decrease in cerebral oxygen-
Propofol (3 mg/kg bolus) ↑ 116 ation116 (Fig. 9.3). These differences were explained by varying
Etomidate (0.3 mg/kg ↓ 116 flow-metabolism coupling and cerebral autoregulation prop-
bolus) erties of the agents. Surprisingly, induction with etomidate
resulted in decreased cerebral oxygenation. This observation
↔, Equivocal/no change; ↑, increase;  ,dose-dependent increase;
↓, decrease; MAC, minimum alveolar concentration.
supports animal work that shows etomidate decreases CBF
despite maintenance of mean blood pressure before its

20

Thiopental

10
∆[HbO2] (µMolar)

Propofol

0
Fig. 9.3  Line graphs that illustrate relative
changes in oxyhemoglobin (HbO2) measured
using near infrared spectroscopy (NIRS) after
induction of anesthesia with three intravenous
Etomidate anesthetic agents. (Lovell AT, Owen-Reece H, Elwell
CE, et al. Continuous measurement of cerebral
–10
oxygenation by near infrared spectroscopy during
–2 –1 0 1 2 3
induction of anesthesia. Anesth Analg 1999;88(3):
Time (min) 554–8.)
80 Section I—Background
sedative effects can decrease metabolism. How maintenance
or steady-state effects of intravenous anesthetic agents affect
NIRS is not clear.
Extracranial Monitors
Extracranial monitors commonly used in the operating room
include the ECG, noninvasive and invasive blood pressure,
central venous pressure, pulse oximetry, and EtCO2. Anes-
thetic agents do not have any direct influence on the function
of these monitors. However, the various anesthetic agents may
affect the physiologic variables recorded by these monitors by
virtue of their effects on the cardiovascular and respiratory
system. The systemic effects of anesthetic agents on these vari-
ables are reviewed in standard anesthesia texts (e.g., Miller’s
Anesthesia, 7th edition, 2010, Elsevier, or Clinical Anesthesia,
6th edition, 2009, Lippincott Williams & Wilkins).
Conclusion
Advanced neuromonitoring frequently is used during neuro-
surgical procedures to guide surgical decisions and help
improve patient safety. The successful deployment of these
techniques requires a team approach that involves the neuro-
surgeon, the anesthesiologist, and the neurophysiologist. The
rational use of these monitoring modalities and how the
information is interpreted depends on use of an optimal anes-
thetic regimen. This is only possible with a full understanding
of how anesthetic agents influence these various monitors.
This chapter provides the basic information and the frame-
work to understand how anesthetic agents affect monitors
both directly or indirectly because of physiologic effects. This
knowledge is valuable in both the operating room and neuro-
critical care unit.
References
1. Fukui K, Morioka T, Hashiguchi K, et al. Relationship between regional
cerebral blood flow and electrocorticographic activities under sevoflurane
and isoflurane anesthesia. J Clin Neurophysiol 2010;27(2):110–5.
2. Mielck F, Brauer A, Radke O, et al. Changes of jugular venous blood
temperature associated with measurements of cerebral blood flow using the
transcerebral double-indicator dilution technique. Eur J Anaesthesiol
2004:289–95.
3. Yoshitani K, Kawaguchi M, Iwata M, et al. Comparison of changes in jugular
venous bulb oxygen saturation and cerebral oxygen saturation during
variations of haemoglobin concentration under propofol and sevoflurane
anaesthesia. Br J Anaesth 2005;94(3):341–6.
4. Matta BF, Lam AM, Mayberg TS, et al. A critique of the intraoperative use of
jugular venous bulb catheters during neurosurgical procedures. Anesth Analg
1994;79(4):745–50.
5. Czosnyka M, Smielewski P, Kirkpatrick P, et al. Continuous assessment of
the cerebral vasomotor reactivity in head injury. Neurosurgery 1997;41:11–7.
6. Lam AM, Newell DW. Intraoperative use of transcranial Doppler
ultrasonography. Neurosurg Clin N Am 1996;7:709–22.
7. Kincaid MS. Transcranial Doppler ultrasonography: a diagnostic tool of
increasing utility. Curr Opin Anaesthesiol 2008;21(5):552–9.
8. Schatlo B, Pluta RM. Clinical applications of transcranial Doppler
sonography. Rev Rec Clin Trials 2007;2(1):49–57.
9. Ogasawara K, Suga Y, Sasaki M, et al. Intraoperative microemboli and low
middle cerebral artery blood flow velocity are additive in predicting
development of cerebral ischemic events after carotid endarterectomy. Stroke
2008;39:3088–91.
10. Ackerstaff RG, Moons KG, van de Vlasakker CJ, et al. Association of
intraoperative transcranial Doppler monitoring variables with stroke from
carotid endarterectomy. Stroke 2000;31(8):1817–23.
11. Halsey JH. Risks and benefits of shunting in carotid endarterectomy. The
International Transcranial Doppler Collaborators. Stroke 1992;23(11):
1583–7.
12. Moritz S, Kasprzak P, Arlt M, et al. Accuracy of cerebral monitoring in
detecting cerebral ischemia during carotid endarterectomy: a comparison of
transcranial Doppler sonography, near-infrared spectroscopy, stump
pressure, and somatosensory evoked potentials. Anesthesiology
2007;107(4):563–9.
13. Ackerstaff RGA. Cerebral circulation monitoring in carotid
endarterectomy and carotid artery stenting. Front Neurol Neurosci 2006;
21229–38.
14. Gossetti B, Gattuso R, Irace L, et al. Embolism to the brain during carotid
stenting and surgery. Acta Chir Belg 2007;107(2):151–4.
15. Telman G, Kouperberg E, Eran A, et al. Microemboli in MCA ipsilateral to
occluded common carotid artery: an observation and short review of the
literature. Neurol Res 2008;30(7):684–6.
16. Imray CHE, Pattinson KTS. Potential role for TCD-directed antiplatelet
agents in symptomatic carotid artery dissection. Stroke 2006;37(3):767.
17. Zuromskis T, Wetterholm R, Lindqvist JF, et al. Prevalence of micro-emboli
in symptomatic high grade carotid artery disease: a transcranial Doppler
study. Eur J Vasc Endovasc Surg 2008;35(5):534–40.
18. Alexandrov AV, Babikian VL, Adams RJ, et al. The evolving role of
transcranial Doppler in stroke prevention and treatment. J Stroke
Cerebrovasc Dis 1998;7(2):101–4.
19. Roy J, Akhtar N, Watson T, et al. Transcranial Doppler microembolic signal
monitoring is useful in diagnosis and treatment of carotid artery dissection:
two case reports. J Neuroimaging 2007;17(4):350–2.
20. Sfyroeras GS, Karkos CD, Arsos G, et al. Cerebral hyperperfusion after
carotid stenting: a transcranial doppler and spect study. Vasc Endovascular
Surg 2009;43:150–6.
21. Adhiyaman V, Alexander S. Cerebral hyperperfusion syndrome following
carotid endarterectomy. Q J Med 2007;100(4):239–44.
22. Ogasawara K, Sakai N, Kuroiwa T, et al. Intracranial hemorrhage associated
with cerebral hyperperfusion syndrome following carotid endarterectomy
and carotid artery stenting: retrospective review of 4494 patients. J
Neurosurg 2007;107(6):1130–6.
23. Jansen C, Sprengers AM, Moll FL, et al. Prediction of intracerebral
haemorrhage after carotid endarterectomy by clinical criteria and
intraoperative transcranial Doppler monitoring. Eur J Vasc Surg 1994;8(3):
303–8.
24. Jansen C, Sprengers AM, Moll FL, et al. Prediction of intracerebral
haemorrhage after carotid endarterectomy by clinical criteria and
intraoperative transcranial Doppler monitoring: results of 233 operations.
Eur J Vasc Surg 1994;8(2):220–5.
25. Magee TR, Davies AH, Horrocks M. Transcranial Doppler evaluation of
cerebral hyperperfusion syndrome after carotid endarterectomy. Eur J Vasc
Surg 1994;8(1):104–6.
26. Riles TS, Imparato AM, Jacobowitz GR, et al. The cause of perioperative
stroke after carotid endarterectomy. J Vasc Surg 1994;19(2):206–16.
27. Spencer MP. Transcranial Doppler monitoring and causes of stroke from
carotid endarterectomy. Stroke 1997;28(4):685–91.
28. Fodale V, Schifilliti D, Conti A, et al. Transcranial Doppler and anesthetics.
Acta Anaesthesiol Scand 2007;51(7):839–47.
29. Lam AM, Matta BF, Mayberg TS, et al. Change in cerebral blood flow
velocity with onset of EEG silence during inhalation anesthesia in humans:
evidence of flow-metabolism coupling? J Cereb Blood Flow Metab
1995;15(4):714–7.
30. Matta BF, Mayberg TS, Lam AM. Direct cerebrovasodilatory effects of
halothane, isoflurane, and desflurane during propofol-induced isoelectric
electroencephalogram in humans. Anesthesiology 1995;83(5):980–5.
31. Molnár C, Settakis G, Sárkány P, et al. Effect of sevoflurane on cerebral
blood flow and cerebrovascular resistance at surgical level of anaesthesia: a
transcranial Doppler study. Eur J Anaesthesiol 2006;24(02):179–84.
32. Matta BF, Heath KJ, Tipping K, et al. Direct cerebral vasodilatory effects of
sevoflurane and isoflurane. Anesthesiology 1999;91(3):677–80.
33. Bundgaard H, von Oettingen G, Larsen KM, et al. Effects of sevoflurane on
intracranial pressure, cerebral blood flow and cerebral metabolism:
a dose-response study in patients subjected to craniotomy for cerebral
tumours. Acta Anaesthesiol Scand 1998;42(6):621–7.
34. Wong GT, Luginbuehl I, Karsli C, et al. The effect of sevoflurane on cerebral
autoregulation in young children as assessed by the transient hyperemic
response. Anesth Analg 2006;102(4):1051–5.
35. Nishiyama T, Matsukawa T, Yokoyama T, et al. Cerebrovascular carbon
dioxide reactivity during general anesthesia: a comparison between
sevoflurane and isoflurane. Anesth Analg 1999;89(6):1437–41.

36. Bedforth NM, Hardman JG, Nathanson MH. Cerebral hemodynamic
response to the introduction of desflurane: a comparison with sevoflurane.
Anesth Analg 2000;91(1):152–5.
37. Bedforth NM, Girling KJ, Skinner HJ, et al. Effects of desflurane on cerebral
autoregulation. Br J Anaesth 2001;87(2):193–7.
38. Brenet O, Granry JC, Poirier N, et al. The effect of desflurane on cerebral
blood flow velocity and cerebrovascular reactivity to CO2 in children.
Ann Fr Anesth Reanim 1998;17(3):227–33.
39. Summors AC, Gupta AK, Matta BF. Dynamic cerebral autoregulation during
sevoflurane anesthesia: a comparison with isoflurane. Anesth Analg
1999;88(2):341–5.
40. McCulloch TJ, Boesel TW, Lam AM. The effect of hypocapnia on the
autoregulation of cerebral blood flow during administration of isoflurane.
Anesth Analg 2005;100(5):1463–7.
41. Leon JE, Bissonnette B. Cerebrovascular responses to carbon dioxide in
children anaesthetized with halothane and isoflurane. Can J Anaesth
1991;38(7):817–25.
42. Lam AM, Mayberg TS, Eng CC, et al. Nitrous oxide-isoflurane anesthesia
causes more cerebral vasodilation than an equipotent dose of isoflurane in
humans. Anesth Analg 1994;78(3):462–8.
43. Matta BF, Lam AM. Nitrous oxide increases cerebral blood flow velocity
during pharmacologically induced EEG silence in humans. J Neurosurg
Anesthesiol 1995;7(2):89–93.
44. Iacopino DG, Conti A, Battaglia C, et al. Transcranial Doppler ultrasound
study of the effects of nitrous oxide on cerebral autoregulation during
neurosurgical anesthesia: a randomized controlled trial. J Neurosurg
2003;99(1):58–64.
Section I—Background 81
45. Aono M, Sato J, Nishino T. Nitrous oxide increases normocapnic cerebral
blood flow velocity but does not affect the dynamic cerebrovascular response
to step changes in end-tidal P(CO2) in humans. Anesth Analg
1999;89(3):684–9.
46. Conti A, Iacopino DG, Fodale V, et al. Cerebral haemodynamic changes
during propofol-remifentanil or sevoflurane anaesthesia: transcranial
Doppler study under bispectral index monitoring. Br J Anaesth
2006;97(3):333–9.
47. Strebel S, Lam AM, Matta B, et al. Dynamic and static cerebral
autoregulation during isoflurane, desflurane, and propofol anesthesia.
Anesthesiology 1995;83(1):66–76.
48. Matta BF, Lam AM, Strebel S, et al. Cerebral pressure autoregulation and
carbon dioxide reactivity during propofol-induced EEG suppression. Br J
Anaesth 1995;74(2):159–63.
49. Thiel A, Zickmann B, Roth H, et al. Effects of intravenous anesthetic agents
on middle cerebral artery blood flow velocity during induction of general
anesthesia. J Clin Monit 1995;11(2):92–8.
50. Kochs E, Werner C, Hoffman WE, et al. Concurrent increases in brain
electrical activity and intracranial blood flow velocity during low-dose
ketamine anaesthesia. Can J Anaesth 1991;38(7):826–30.
A complete list of references for this chapter can be found online at
www.expertconsult.com.

References
1. Fukui K, Morioka T, Hashiguchi K, et al. Relationship between regional
cerebral blood flow and electrocorticographic activities under sevoflurane
and isoflurane anesthesia. J Clin Neurophysiol 2010;27(2):110–5.
2. Mielck F, Brauer A, Radke O, et al. Changes of jugular venous blood
temperature associated with measurements of cerebral blood flow using the
transcerebral double-indicator dilution technique. Eur J Anaesthesiol
2004:289–95.
3. Yoshitani K, Kawaguchi M, Iwata M, et al. Comparison of changes in
jugular venous bulb oxygen saturation and cerebral oxygen saturation
during variations of haemoglobin concentration under propofol and
sevoflurane anaesthesia. Br J Anaesth 2005;94(3):341–6.
4. Matta BF, Lam AM, Mayberg TS, et al. A critique of the intraoperative use
of jugular venous bulb catheters during neurosurgical procedures. Anesth
Analg 1994;79(4):745–50.
5. Czosnyka M, Smielewski P, Kirkpatrick P, et al. Continuous assessment of
the cerebral vasomotor reactivity in head injury. Neurosurgery 1997;
41:11–7.
6. Lam AM, Newell DW. Intraoperative use of transcranial Doppler
ultrasonography. Neurosurg Clin N Am 1996;7:709–22.
7. Kincaid MS. Transcranial Doppler ultrasonography: a diagnostic tool of
increasing utility. Curr Opin Anaesthesiol 2008;21(5):552–9.
8. Schatlo B, Pluta RM. Clinical applications of transcranial Doppler
sonography. Rev Rec Clin Trials 2007;2(1):49–57.
9. Ogasawara K, Suga Y, Sasaki M, et al. Intraoperative microemboli and low
middle cerebral artery blood flow velocity are additive in predicting
development of cerebral ischemic events after carotid endarterectomy.
Stroke 2008;39:3088–91.
10. Ackerstaff RG, Moons KG, van de Vlasakker CJ, et al. Association of
intraoperative transcranial Doppler monitoring variables with stroke from
carotid endarterectomy. Stroke 2000;31(8):1817–23.
11. Halsey JH. Risks and benefits of shunting in carotid endarterectomy. The
International Transcranial Doppler Collaborators. Stroke
1992;23(11):1583–7.
12. Moritz S, Kasprzak P, Arlt M, et al. Accuracy of cerebral monitoring in
detecting cerebral ischemia during carotid endarterectomy: a comparison
of transcranial Doppler sonography, near-infrared spectroscopy, stump
pressure, and somatosensory evoked potentials. Anesthesiology 2007;
107(4):563–9.
13. Ackerstaff RGA. Cerebral circulation monitoring in carotid endarterectomy
and carotid artery stenting. Front Neurol Neurosci 2006;21229–38.
14. Gossetti B, Gattuso R, Irace L, et al. Embolism to the brain during carotid
stenting and surgery. Acta Chir Belg 2007;107(2):151–4.
15. Telman G, Kouperberg E, Eran A, et al. Microemboli in MCA ipsilateral to
occluded common carotid artery: an observation and short review of the
literature. Neurol Res 2008;30(7):684–6.
16. Imray CHE, Pattinson KTS. Potential role for TCD-directed antiplatelet
agents in symptomatic carotid artery dissection. Stroke 2006;37(3):767.
17. Zuromskis T, Wetterholm R, Lindqvist JF, et al. Prevalence of micro-emboli
in symptomatic high grade carotid artery disease: a transcranial Doppler
study. Eur J Vasc Endovasc Surg 2008;35(5):534–40.
18. Alexandrov AV, Babikian VL, Adams RJ, et al. The evolving role of
transcranial Doppler in stroke prevention and treatment. J Stroke
Cerebrovasc Dis 1998;7(2):101–4.
19. Roy J, Akhtar N, Watson T, et al. Transcranial Doppler microembolic signal
monitoring is useful in diagnosis and treatment of carotid artery dissection:
two case reports. J Neuroimaging 2007;17(4):350–2.
20. Sfyroeras GS, Karkos CD, Arsos G, et al. Cerebral hyperperfusion after
carotid stenting: a transcranial doppler and spect study. Vasc Endovascular
Surg 2009;43:150–6.
21. Adhiyaman V, Alexander S. Cerebral hyperperfusion syndrome following
carotid endarterectomy. Q J Med 2007;100(4):239–44.
22. Ogasawara K, Sakai N, Kuroiwa T, et al. Intracranial hemorrhage associated
with cerebral hyperperfusion syndrome following carotid endarterectomy
and carotid artery stenting: retrospective review of 4494 patients. J
Neurosurg 2007;107(6):1130–6.
23. Jansen C, Sprengers AM, Moll FL, et al. Prediction of intracerebral
haemorrhage after carotid endarterectomy by clinical criteria and
intraoperative transcranial Doppler monitoring. Eur J Vasc Surg
1994;8(3):303–8.
24. Jansen C, Sprengers AM, Moll FL, et al. Prediction of intracerebral
haemorrhage after carotid endarterectomy by clinical criteria and
intraoperative transcranial Doppler monitoring: results of 233 operations.
Eur J Vasc Surg 1994;8(2):220–5.
Section I—Background 81.e1
25. Magee TR, Davies AH, Horrocks M. Transcranial Doppler evaluation of
cerebral hyperperfusion syndrome after carotid endarterectomy. Eur J Vasc
Surg 1994;8(1):104–6.
26. Riles TS, Imparato AM, Jacobowitz GR, et al. The cause of perioperative
stroke after carotid endarterectomy. J Vasc Surg 1994;19(2):206–16.
27. Spencer MP. Transcranial Doppler monitoring and causes of stroke from
carotid endarterectomy. Stroke 1997;28(4):685–91.
28. Fodale V, Schifilliti D, Conti A, et al. Transcranial Doppler and anesthetics.
Acta Anaesthesiol Scand 2007;51(7):839–47.
29. Lam AM, Matta BF, Mayberg TS, et al. Change in cerebral blood flow
velocity with onset of EEG silence during inhalation anesthesia in humans:
evidence of flow-metabolism coupling? J Cereb Blood Flow Metab
1995;15(4):714–7.
30. Matta BF, Mayberg TS, Lam AM. Direct cerebrovasodilatory effects of
halothane, isoflurane, and desflurane during propofol-induced isoelectric
electroencephalogram in humans. Anesthesiology 1995;83(5):980–5.
31. Molnár C, Settakis G, Sárkány P, et al. Effect of sevoflurane on cerebral
blood flow and cerebrovascular resistance at surgical level of anaesthesia: a
transcranial Doppler study. Eur J Anaesthesiol 2006;24(02):179–84.
32. Matta BF, Heath KJ, Tipping K, et al. Direct cerebral vasodilatory effects of
sevoflurane and isoflurane. Anesthesiology 1999;91(3):677–80.
33. Bundgaard H, von Oettingen G, Larsen KM, et al. Effects of sevoflurane on
intracranial pressure, cerebral blood flow and cerebral metabolism:
a dose-response study in patients subjected to craniotomy for cerebral
tumours. Acta Anaesthesiol Scand 1998;42(6):621–7.
34. Wong GT, Luginbuehl I, Karsli C, et al. The effect of sevoflurane on
cerebral autoregulation in young children as assessed by the transient
hyperemic response. Anesth Analg 2006;102(4):1051–5.
35. Nishiyama T, Matsukawa T, Yokoyama T, et al. Cerebrovascular carbon
dioxide reactivity during general anesthesia: a comparison between
sevoflurane and isoflurane. Anesth Analg 1999;89(6):1437–41.
36. Bedforth NM, Hardman JG, Nathanson MH. Cerebral hemodynamic
response to the introduction of desflurane: a comparison with sevoflurane.
Anesth Analg 2000;91(1):152–5.
37. Bedforth NM, Girling KJ, Skinner HJ, et al. Effects of desflurane on cerebral
autoregulation. Br J Anaesth 2001;87(2):193–7.
38. Brenet O, Granry JC, Poirier N, et al. The effect of desflurane on cerebral
blood flow velocity and cerebrovascular reactivity to CO2 in children. Ann
Fr Anesth Reanim 1998;17(3):227–33.
39. Summors AC, Gupta AK, Matta BF. Dynamic cerebral autoregulation
during sevoflurane anesthesia: a comparison with isoflurane. Anesth Analg
1999;88(2):341–5.
40. McCulloch TJ, Boesel TW, Lam AM. The effect of hypocapnia on the
autoregulation of cerebral blood flow during administration of isoflurane.
Anesth Analg 2005;100(5):1463–7.
41. Leon JE, Bissonnette B. Cerebrovascular responses to carbon dioxide in
children anaesthetized with halothane and isoflurane. Can J Anaesth
1991;38(7):817–25.
42. Lam AM, Mayberg TS, Eng CC, et al. Nitrous oxide-isoflurane anesthesia
causes more cerebral vasodilation than an equipotent dose of isoflurane in
humans. Anesth Analg 1994;78(3):462–8.
43. Matta BF, Lam AM. Nitrous oxide increases cerebral blood flow velocity
during pharmacologically induced EEG silence in humans. J Neurosurg
Anesthesiol 1995;7(2):89–93.
44. Iacopino DG, Conti A, Battaglia C, et al. Transcranial Doppler ultrasound
study of the effects of nitrous oxide on cerebral autoregulation during
neurosurgical anesthesia: a randomized controlled trial. J Neurosurg
2003;99(1):58–64.
45. Aono M, Sato J, Nishino T. Nitrous oxide increases normocapnic cerebral
blood flow velocity but does not affect the dynamic cerebrovascular
response to step changes in end-tidal P(CO2) in humans. Anesth Analg
1999;89(3):684–9.
46. Conti A, Iacopino DG, Fodale V, et al. Cerebral haemodynamic changes
during propofol-remifentanil or sevoflurane anaesthesia: transcranial
Doppler study under bispectral index monitoring. Br J Anaesth
2006;97(3):333–9.
47. Strebel S, Lam AM, Matta B, et al. Dynamic and static cerebral
autoregulation during isoflurane, desflurane, and propofol anesthesia.
Anesthesiology 1995;83(1):66–76.
48. Matta BF, Lam AM, Strebel S, et al. Cerebral pressure autoregulation and
carbon dioxide reactivity during propofol-induced EEG suppression. Br J
Anaesth 1995;74(2):159–63.
49. Thiel A, Zickmann B, Roth H, et al. Effects of intravenous anesthetic agents
on middle cerebral artery blood flow velocity during induction of general
anesthesia. J Clin Monit 1995;11(2):92–8.
81.e2 Section I—Background
50. Kochs E, Werner C, Hoffman WE, et al. Concurrent increases in brain
electrical activity and intracranial blood flow velocity during low-dose
ketamine anaesthesia. Can J Anaesth 1991;38(7):826–30.
51. Mayberg TS, Lam AM, Matta BF, et al. Ketamine does not increase cerebral
blood flow velocity or intracranial pressure during isoflurane/nitrous oxide
anesthesia in patients undergoing craniotomy. Anesth Analg
1995;81(1):84–9.
52. Engelhard K, Werner C, Möllenberg O, et al. S(+)-ketamine/propofol
maintain dynamic cerebrovascular autoregulation in humans. Can J
Anaesth 2001;48(10):1034–9.
53. Ogawa Y, Iwasaki K, Aoki K, et al. Dexmedetomidine weakens dynamic
cerebral autoregulation as assessed by transfer function analysis and the
thigh cuff method. Anesthesiology 2008;109(4):642–50.
54. Drummond JC, Dao AV, Roth DM, et al. Effect of dexmedetomidine on
cerebral blood flow velocity, cerebral metabolic rate, and carbon dioxide
response in normal humans. Anesthesiology 2008;108(2):225–32.
55. Kofke WA, Dong ML, Bloom M, et al. Transcranial Doppler
ultrasonography with induction of anesthesia for neurosurgery.
J Neurosurg Anesthesiol 1994;6(2):89–97.
56. Thiel A, Zickmann B, Roth H, et al. Effects of intravenous anesthetic agents
on middle cerebral artery blood flow velocity during induction of general
anesthesia. J Clin Monit 1995;11(2):92–8.
57. Cold GE, Eskesen V, Eriksen H, et al. CBF and CMRO2 during continuous
etomidate infusion supplemented with N2O and fentanyl in patients with
supratentorial cerebral tumour: a dose-response study. Acta Anaesthesiol
Scand 1985;29(5):490–4.
58. Cheng MA, Hoffman WE, Baughman VL, et al. The effects of midazolam
and sufentanil sedation on middle cerebral artery blood flow velocity in
awake patients. J Neurosurg Anesthesiol 1993;5(4):232–6.
59. Hänel F, Werner C, von Knobelsdorff G, et al. The effects of fentanyl and
sufentanil on cerebral hemodynamics. J Neurosurg Anesthesiol
1997;9(3):223–7.
60. Fodale V, Schifilliti D, Praticò C, et al. Remifentanil and the brain. Acta
Anaesthesiol Scand 2008;52(3):319–26.
61. Frost EA. Inhalation anaesthetic agents in neurosurgery. Br J Anaesth
1984;56(Suppl 1):47S-56S.
62. Kennedy RR, Minto C, Seethepalli A. Effect-site half-time for burst
suppression is longer than for hypnosis during anaesthesia with
sevoflurane. Br J Anaesth 2008;100(1):72–7.
63. Schwartz AE, Tuttle RH, Poppers PJ. Electroencephalographic burst
suppression in elderly and young patients anesthetized with isoflurane.
Anesth Analg 1989;68(1):9–12.
64. Voss LJ, Seleigh JW, Barnard JPM, et al. The howling cortex: seizures and
general anesthetic drugs. Anesth Analg 2008;107:1689–703.
65. Kofke WA. Anesthetic management of the patient with epilepsy or prior
seizures. Curr Opin Anaesthesiol 2010;23:391–9.
66. Mäkelä JP, Iivanainen M, Pieninkeroinen IP, et al. Seizures associated with
propofol anesthesia. Epilepsia 1993;34(5):832–5.
67. Jääskeläinen SK, Kaisti K, Suni L, et al. Sevoflurane is epileptogenic in
healthy subjects at surgical levels of anesthesia. Neurology 2003;
61(8):1073–8.
68. Mustola ST, Baer GA, Toivonen JK, et al. Electroencephalographic burst
suppression versus loss of reflexes anesthesia with propofol or thiopental:
differences of variance in the catecholamine and cardiovascular response to
tracheal intubation. Anesth Analg 2003;97(4):1040–5.
69. Souter MJ, Rozet I, Ojemann JG, et al. Dexmedetomidine sedation during
awake craniotomy for seizure resection: effects on electrocorticography.
J Neurosurg Anesthesiol 2007;19(1):38–44.
70. Rozet I, Muangman S, Vavilala MS, et al. Clinical experience with
dexmedetomidine for implantation of deep brain stimulators in Parkinson’s
disease. Anesth Analg 2006;103:1244–8.
71. Lam AM, Sharar SR, Mayberg TS, et al. Isoflurane compared with nitrous
oxide anaesthesia for intraoperative monitoring of somatosensory-evoked
potentials. Can J Anaesth 1994;41(4):295–300.
72. Manninen PH, Lam AM, Nicholas JF. The effects of isoflurane and
isoflurane-nitrous oxide anesthesia on brainstem auditory evoked
potentials in humans. Anesth Analg 1985;64(1):43–7.
73. de Haan P, Kalkman CJ. Spinal cord monitoring: somatosensory- and
motor-evoked potentials. Anesthesiol Clin N Am 2001;19(4):923–45.
74. Calancie B, Harris W, Brindle GF, et al. Threshold-level repetitive
transcranial electrical stimulation for intraoperative monitoring of central
motor conduction. J Neurosurg 2001;95(2 Suppl):161–8.
75. MacDonald DB. Safety of intraoperative transcranial electrical stimulation
motor evoked potential monitoring. J Clin Neurophysiol 2002;19(5):
416–29.
76. Lotto ML, Banoub M, Schubert A. Effects of anesthetic agents and
physiologic changes on intraoperative motor evoked potentials. J Neurosurg
Anesthesiol 2004;16(1):32–42.
77. Zentner J, Albrecht T, Heuser D. Influence of halothane, enflurane, and
isoflurane on motor evoked potentials. Neurosurgery 1992;31(2):298–305.
78. Haghighi SS, Green KD, Oro JJ, et al. Depressive effect of isoflurane
anesthesia on motor evoked potentials. Neurosurgery 1990;26(6):993–7.
79. Lyon R, Feiner J, Lieberman JA. Progressive suppression of motor evoked
potentials during general anesthesia: the phenomenon of “anesthetic fade.”
J Neurosurg Anesthesiol 2005;17(1):13–19.
80. Ubags LH, Kalkman CJ, Been HD. Influence of isoflurane on myogenic
motor evoked potentials to single and multiple transcranial stimuli during
nitrous oxide/opioid anesthesia. Neurosurgery 1998;43(1):90–4.
81. Pelosi L, Stevenson M, Hobbs GJ, et al. Intraoperative motor evoked
potentials to transcranial electrical stimulation during two anaesthetic
regimens. Clin Neurophysiol 2001;112(6):1076–87.
82. Pechstein U, Nadstawek J, Zentner J, et al. Isoflurane plus nitrous oxide
versus propofol for recording of motor evoked potentials after high
frequency repetitive electrical stimulation. Electroencephalogr Clin
Neurophysiol 1998;108(2):175–81.
83. Lo Y, Dan Y, Tan YE, et al. Intraoperative motor-evoked potential
monitoring in scoliosis surgery: comparison of desflurane/nitrous oxide
with propofol total intravenous anesthetic regimens. J Neurosurg
Anesthesiol 2006;18(3):211–4.
84. Kawaguchi M, Inoue S, Kakimoto M, et al. The effect of sevoflurane on
myogenic motor-evoked potentials induced by single and paired
transcranial electrical stimulation of the motor cortex during nitrous oxide/
ketamine/fentanyl anesthesia. J Neurosurg Anesthesiol 1998;10(3):131–6.
85. Reinacher PC, Priebe H, Blumrich W, et al. The effects of stimulation
pattern and sevoflurane concentration on intraoperative motor-evoked
potentials. Anesth Analg 2006;102(3):888–95.
86. Rampil IJ, King BS. Volatile anesthetics depress spinal motor neurons.
Anesthesiology 1996;85(1):129–34.
87. Ubags LH, Kalkman CJ, Been HD, et al. Differential effects of nitrous oxide
and propofol on myogenic transcranial motor evoked responses during
sufentanil anaesthesia. Br J Anaesth 1997;79(5):590–4.
88. van Dongen EP, ter Beek HT, Schepens MA, et al. The influence of nitrous
oxide to supplement fentanyl/low-dose propofol anesthesia on transcranial
myogenic motor-evoked potentials during thoracic aortic surgery. J
Cardiothorac Vasc Anesth 1999;13(1):30–4.
89. van Dongen EP, ter Beek HT, Schepens MA, et al. Effect of nitrous oxide on
myogenic motor potentials evoked by a six pulse train of transcranial
electrical stimuli: a possible monitor for aortic surgery. Br J Anaesth
1999;82(3):323–8.
90. Scheufler KM, Reinacher PC, Blumrich W, et al. The modifying effects of
stimulation pattern and propofol plasma concentration on motor-evoked
potentials. Anesth Analg 2005;100(2):440–7.
91. Kalkman CJ, Drummond JC, Ribberink AA, et al. Effects of propofol,
etomidate, midazolam, and fentanyl on motor evoked responses to
transcranial electrical or magnetic stimulation in humans. Anesthesiology
1992;76(4):502–9.
92. Inoue S, Kawaguchi M, Kakimoto M, et al. Amplitudes and intrapatient
variability of myogenic motor evoked potentials to transcranial electrical
stimulation during ketamine/N2O- and propofol/N2O-based anesthesia.
J Neurosurg Anesthesiol 2002;14(3):213–7.
93. Ubags LH, Kalkman CJ, Been HD, et al. The use of ketamine or etomidate
to supplement sufentanil/N2O anesthesia does not disrupt monitoring of
myogenic transcranial motor evoked responses. J Neurosurg Anesthesiol
1997;9(3):228–33.
94. Zaarour C, Engelhardt T, Strantzas S, et al. Effect of low-dose ketamine on
voltage requirement for transcranial electrical motor evoked potentials in
children. Spine 2007;32(22):E627–30.
95. Lang EW, Beutler AS, Chesnut RM, et al. Myogenic motor-evoked potential
monitoring using partial neuromuscular blockade in surgery of the spine.
Spine 1996;21(14):1676–86.
96. Tobias JD, Goble TJ, Bates G, et al. Effects of dexmedetomidine on
intraoperative motor and somatosensory evoked potential monitoring
during spinal surgery in adolescents. Paediatr Anaesth 2008;18:1082–8.
97. Bala E, Sessler DI, Nair DR, et al. Motor and somatosensory evoked
potentials are well maintained in patients given dexmedetomidine during
spine surgery. Anesthesiology 2008;109(3):417–25.
98. Mahmoud M, Sadhasivam S, Salisbury S, et al. Susceptibility of transcranial
electric motor-evoked potentials to varying targeted blood levels of
dexmedetomidine during spine surgery. Anesthesiology 2010;
112(6):1364–73.

99. Mahmoud M, Sadhasivam S, Sestokas AK, et al. Loss of transcranial electric
motor evoked potentials during pediatric spine surgery with
dexmedetomidine. Anesthesiology 2007;106(2):393–6.
100. Scheufler K, Zentner J. Total intravenous anesthesia for intraoperative
monitoring of the motor pathways: an integral view combining clinical and
experimental data. J Neurosurg 2002;96(3):571–9.
101. van Dongen EP, ter Beek HT, Schepens MA, et al.Within-patient variability
of myogenic motor-evoked potentials to multipulse transcranial electrical
stimulation during two levels of partial neuromuscular blockade in aortic
surgery. Anesth Analg 1999;88(1):22–7.
102. Hayashi H, Kawaguchi M, Yamamoto Y, et al. The application of tetanic
stimulation of the unilateral tibial nerve before transcranial stimulation can
augment the amplitudes of myogenic motor-evoked potentials from the
muscles in the bilateral upper and lower limbs. Anesth Analg 2008;
107(1):215–20.
103. Yamamoto Y, Kawaguchi M, Hayashi H, et al. The effects of the
neuromuscular blockade levels on amplitudes of posttetanic motor-evoked
potentials and movement in response to transcranial stimulation in
patients receiving propofol and fentanyl anesthesia. Anesth Analg
2008;106(3):930–4.
104. Smith M. Perioperative uses of transcranial perfusion monitoring.
Anesthesiol Clin 2007;25(3):557–77.
105. Duffy CM, Manninen PH, Chan A, et al. Comparison of cerebral oximeter
and evoked potential monitoring in carotid endarterectomy. Can J Anaesth
1997;44(10):1077–81.
106. Bhatia R, Hampton T, Malde S, et al. The application of near-infrared
oximetry to cerebral monitoring during aneurysm embolization: a
comparison with intraprocedural angiography. J Neurosurg Anesthesiol
2007;19(2):97–104.
Section I—Background 81.e3
107. Dujovny M, Slavin KV, Hernandez G, et al. Use of cerebral oximetry to
monitor brain oxygenation reserves for skull base surgery. Skull Base Surg
1994;4(3):117–21.
108. Kakihana Y, Matsunaga A, Yasuda T, et al. Brain oxymetry in the operating
room: current status and future directions with particular regard to
cytochrome oxidase. J Biomed Opt 2008;13(3):033001.
109. McDonagh DL, McDaniel MR, Monk TG. The effect of intravenous indigo
carmine on near-infrared cerebral oximetry. Anesth Analg 2007;105(3):
704–6.
110. Patel J, Marks K, Roberts I, et al. Measurement of cerebral blood flow in
newborn infants using near infrared spectroscopy with indocyanine green.
Pediatr Res 1998;43(1):34–9.
111. McCormick PW, Stewart M, Lewis G, et al. Intracerebral penetration of
infrared light. Technical note. J Neurosurg 1992;76(2):315–8.
112. Nishikawa K, Kanemaru Y, Hagiwara R, et al. The influence of sevoflurane
on the bispectral index, regional cerebral oxygen saturation, and propofol
concentration during propofol/N2O anesthesia. J Clin Monit Comput
2006;20(6):415–20.
113. Fassoulaki A, Kaliontzi H, Petropoulos G, et al. The effect of desflurane and
sevoflurane on cerebral oximetry under steady-state conditions. Anesth
Analg 2006;102(6):1830–5.
114. Kanemaru Y, Nishikawa K, Goto F. Bispectral index and regional cerebral
oxygen saturation during propofol/N2O anesthesia. Can J Anaesth.
2006;53(4):363–9.
115. Iwasaki K, Nomoto Y, Ishiwata M, et al. Vital capacity induction with 8%
sevoflurane and N2O causes cerebral hyperemia. J Anesth 2003;17(1):3–7.
116. Lovell AT, Owen-Reece H, Elwell CE, et al. Continuous measurement of
cerebral oxygenation by near infrared spectroscopy during induction of
anesthesia. Anesth Analg 1999;88(3):554–8.
II
Chapter
10  
Clinical Assessment in the
Neurocritical Care Unit
Ramani Balu, John A. Detre, and Joshua M. Levine
Introduction
Many patients in the neurocritical care unit (NCCU) are in
coma, have altered consciousness, or have a disease process
that may lead to altered consciousness. Coma and other dis-
orders of consciousness are common clinical problems that
indicate a severe disturbance of brain function. Monitoring
and management of patients with disturbed consciousness
depend in large part on the underlying etiology. The princi-
pal means of monitoring the comatose patient are clinical
neurologic examinations, the use of rating scales, neuroimag-
ing, and electrophysiologic tests. In addition, a variety of
invasive and noninvasive monitors of brain function is avail-
able and discussed in other chapters of this book. Goals of
monitoring include establishing an etiologic diagnosis and
excluding coma mimics, determining the location and sever-
ity of injury, assessing response to therapy, and determining
prognosis. This chapter provides an overview of the anatomy
of consciousness, the general clinical approach to the coma
patient, and neurologic assessment, with a focus on the clini-
cal examination, quantitative coma scales, and electrophysio-
logic and imaging studies that may provide insight into the
prognosis of coma. In addition, other critical care scores that
provide insight into extracerebral pathophysiology are briefly
discussed.
Anatomy of Brain Circuits
Involved in Arousal, Alertness,
and Conscious Behavior
Level of arousal (alertness) depends on the coordinated activ-
ity of multiple brainstem nuclei that send projections to
diencephalic targets in the thalamus and hypothalamus and
diffusely to the cerebral cortex. Areas of the thalamus that
receive information from arousal centers in the brainstem in
turn send widespread projections to both cerebral hemi-
spheres. Thus isolated brainstem lesions, bilateral thalamic
damage, or diffuse bilateral cerebral injury can cause global
impairments of consciousness and coma. An important corol-
lary to this general rule is that unilateral lesions should not
cause global reductions in consciousness unless they are large
enough to produce significant mass effect on the contralateral
hemisphere or there is a preexisting contralateral lesion.
84
Because brainstem arousal centers are close to other impor-
tant nuclei and tracts that control eye movements and breath-
ing, the location of lesions causing global impairments in
consciousness often can be readily determined by careful
physical examination.
In the early 20th century Baron Constantin von Economo
first proposed the concept that specific brainstem nuclei
control arousal and alertness after his detailed clinicopatho-
logic study of patients with encephalitis lethargica.1 The vast
majority of these patients had profound somnolence along
with parkinsonism and focal oculomotor abnormalities, and
spent only a few hours awake each day. Postmortem analysis
of these patients demonstrated damage to the paramedian
reticular formation near the junction between the midbrain
and the diencephalon. A small subset of patients who suffered
from debilitating insomnia in contrast had damage to the
anterior hypothalamus. Based on these results, von Economo
proposed that a specific arousal-promoting center exists in the
midbrain, and a separate sleep-promoting center is in the
hypothalamus.
Studies by Morruzi and Magoun in the 1940s2 showed that
selective lesions of the midbrain paramedian reticular forma-
tion in cats caused electroencephalographic slowing and
behavioral unresponsiveness similar to sleep. Lesions in the
midbrain that contained ascending sensory pathways caused
defects in somatosensory and auditory-evoked responses but
had no effect on the normal desynchronized electroencepha-
lographic pattern seen in awake animals. These investigators
coined the term ascending reticular activating system (ARAS)
to describe the critical function of the paramedian reticular
formation in alertness. Subsequent lesion studies in humans
have confirmed the importance of the upper brainstem to
promote and maintain behavioral arousal. Specifically, injury
to either the paramedian midbrain or bilateral dorsolateral
pontine tegmentum causes coma.3 Lesions of other parts of
the brainstem outside of this ascending arousal system,
although often debilitating, do not cause global impairments
of consciousness.3,4 Multiple brainstem nuclei that form dis-
tinct cholinergic and monoaminergic projection systems are
now known to constitute the ascending arousal system.4,5
Although these nuclei are not all located within the parame-
dian midbrain reticular formation, their output fibers in
general course through this structure on their way to targets
in the diencephalon and cerebral cortex.
© Copyright 2013 Elsevier Inc. All rights reserved.
 Section II—Clinical and Laboratory Assessment 85
Cholinergic Projection Systems
The main source of arousal promoting inputs to the thalamus
is from the pedunculopontine and laterodorsal tegmental
nuclei in the pons.5-8 These areas send cholinergic projections
through the midbrain reticular formation mainly to the intra-
laminar thalamic nuclei and the thalamic reticular nucleus,
although they also project to the thalamic relay nuclei. The
intralaminar nuclei are excited by their ascending cholinergic
input and send widespread diffuse excitatory projections to
the cortex. The thalamic reticular nucleus, by contrast, sends
local inhibitory projections to thalamocortical relay nuclei,
and its principal neurons are inhibited by acetylcholine. Thus
activity in the pedunculopontine and laterodorsal tegmental
nuclei—which is greatest during awakening and decreases
during sleep—both activate diffuse excitatory projections
from the intralaminar nuclei and disinhibit excitatory thala-
mocortical relay neurons by inhibiting the reticular nucleus.
A second source of cholinergic inputs arises from the basal
forebrain. These neurons receive inputs from brainstem
monoaminergic projection systems (discussed following) and
neighboring excitatory brainstem nuclei and send widespread
projections throughout the cerebral cortex.5,9 Interestingly,
single axons from the basal forebrain have patchy, spatially
defined projection patterns to defined subsets of cortical
neurons.10 Thus regulated activity of different parts of the
basal forebrain can have spatially restricted effects on different
cortical areas.
Monoaminergic Projection Systems
Multiple monoaminergic projection systems also provide tel-
encephalic input from the brainstem and diencephalon that
influence alertness and conscious behavior. Activity of these
monoaminergic projections has complex effects on cortical
activity and often decreases background activity but increases
the responsiveness of a cortical neuron to its best stimulus,
thereby increasing signal to noise ratio.11-15 These systems
therefore have critical functions in regulating attention,
memory, and other higher order cognitive processes in
addition to providing a more general arousal-promoting
stimulus.
Dopaminergic neurons are found in the midbrain in both
the substantia nigra and in the ventral tegmental area.5 Projec-
tions from the substantia nigra terminate in the basal ganglia
and are crucial for motor control and modulation. The ventral
tegmental projections course through the midbrain reticular
formation and provide a mesolimbic projection to the nucleus
accumbens, basal forebrain, cingulate cortex, hippocampus,
and amygdala and a mesocortical projection to the prefrontal
cortex.5 Mesolimbic dopaminergic pathways are involved in
reward-mediated behavior,16,17 and overactivity of these pro-
jections is thought to be central in the pathogenesis of thought
disorders such as schizophrenia.18 Mesocortical projections
likely are involved in attention and working memory.19
Noradrenergic projections to the cerebral hemispheres arise
mainly from the locus ceruleus, located in the upper pons near
the fourth ventricle. Locus ceruleus activity increases with
awakening, and this area likely is important in behavioral state
switching (i.e., from sleep to wakefulness).7,8 These noradren-
ergic systems, like serotonergic systems (see following text),
also are pharmaceutical targets in mood disorders.20,21
Multiple midline brainstem areas throughout the midbrain,
pons, and medulla that comprise the raphe nuclei provide
serotonergic input to the brain.5 Of these, the rostral raphe
nuclei in the midbrain and pons project diffusely throughout
the cortex, thalamus, and basal ganglia. Serotonin effects on
these areas are complex and can be either excitatory or inhibi-
tory. These pathways also are thought to play a major role in
psychiatric diseases such as depression, anxiety, and obsessive-
compulsive disorder.20,21
A widespread histaminergic pathway arises from the tuber-
omamillary nucleus in the posterior hypothalamus.5 Hista-
minergic neurons are important in sleep and arousal
regulation, and activity of tuberomamillary nucleus neurons
is greatest during waking and rapid eye movement (REM)
sleep.4,7 Histamine receptor blockers that can cross the blood-
brain barrier, such as diphenhydramine, can exert a powerful
sleep-promoting effect.4
Interactions Between Arousal-Promoting
Brain Centers
The anatomy of the pathways described previously and
their aggregate importance in arousal have been recognized
for decades. However, the specific ways in which these path-
ways interact to promote conscious behavior, attention, or
memory and regulate the transitions between multiple behav-
ioral states is only beginning to be elucidated. Other inputs,
including the galanin and gamma aminobutyric acid–secret-
ing (GABAergic) inputs from the ventrolateral preoptic
(VLPO) area of the hypothalamus,7,8 orexin-containing
neurons from the lateral hypothalamus, and descending corti-
cal inputs provide modulatory control over the ARAS.8,22,23
Nevertheless, the concept of an ascending brainstem arousal
system whose activity promotes alertness and conscious
behavior remains invaluable clinically when evaluating
patients with disorders of impaired consciousness.
Coma Etiology
Causes of altered mental status and coma are protean, and
can be divided into primary brain disorders and systemic
derangements that secondarily affect brain function (Table
10.1). Primary brain disorders can be caused by structural
abnormalities (e.g., cerebral infarction, tumors, subdural
hematomas, intraparenchymal hemorrhages, and abscesses
among others) that either directly distort the circuitry of the
ascending arousal system or globally increase intracranial
pressure (ICP), or by diffuse nonstructural disturbances such
as seizures. Often both structural and nonstructural abnor-
malities may exist together in the same patient, for example,
a patient with a cerebral abscess who develops seizures. Sys-
temic disturbances cause encephalopathy through diffuse
bilateral cerebral dysfunction and can include metabolic
derangements, exposure to toxins, systemic infections, or
diffuse encephalitis caused by primary central nervous system
(CNS) infections, autoimmune processes, or paraneoplastic
syndromes.
The most common causes of coma are traumatic brain
injury (TBI), hypoxic-ischemic encephalopathy (HIE), drug
overdose, ischemic and hemorrhagic strokes, CNS infections,
and brain herniation from space-occupying lesions. A detailed
discussion of these conditions is beyond the scope of this
86 Section II—Clinical and Laboratory Assessment

Table 10.1  A Partial List of the Etiologies

Differential Diagnosis (Mimics) of
of Coma and Altered Mental Status Coma and the Vegetative State
PRIMARY BRAIN DISORDERS
Patients may become unresponsive from conditions that
mimic coma or a vegetative state. Care must be taken not to
Structural Lesions
consider these patients as comatose because they have no
1. Traumatic brain injury impairments of conscious behavior and often can communi-
a. Diffuse axonal injury cate with clinical staff and other individuals if an appropriate
b. Contusions
c. Subdural hematomas communication system is devised.24 In the locked-in syn-
d. Epidural hematomas drome, destruction of the ventral pons leaves the patient
2. Cerebrovascular disorders quadriplegic and mute.4 Patients often are aware of their sur-
a. Ischemic strokes roundings and may communicate only through vertical eye
b. Spontaneous intracerebral hemorrhage
c. Subarachnoid hemorrhage
movements and blinking, which are spared. With more rostral
d. Hypoxic-ischemic encephalopathy pontine lesions, vertical eye movements and blinking are lost.
e. Cerebral venous sinus thrombosis In this state that can be likened to receiving a neuromuscular
3. Malignant disease blocking agent without a sedative, the patient has no means
a. Brain tumors of communication. Severe Guillain-Barré syndrome, botu-
4. Infectious diseases
a. Brain abscesses lism, and critical-illness neuropathy may similarly result in
5. Demyelinating disease complete de-efferentation. Catatonia is a manifestation of
a. Acute disseminated encephalomyelitis severe psychiatric illness in which patients open their eyes, do
b. Central pontine myelinolysis not speak or follow commands, and may exhibit waxy flexibil-
6. Hydrocephalus
ity.4,25 The remainder of the neurologic exam and the electro-
Nonstructural Disorders encephalogram (EEG) are normal. Akinetic mutism that
1. Infectious diseases results from bilateral medial frontal lobe injury is a profound
a. Bacterial meningoencephalitis form of abulia (lack of motivation) in which patients are
b. Viral encephalitis unable to speak or to move, but open their eyes, occasionally
2. Malignant disease
a. Carcinomatous or lymphomatous meningitis
track visual stimuli, and sometimes respond when prompted.26,27
3. Generalized seizures, status epilepticus
4. Basilar migraines
SYSTEMIC DISORDERS Neurologic Examination of the
Toxic Encephalopathies Comatose Patient
1. Medication overdose The initial neurologic examination of a patient with altered
a. Opioids, benzodiazepines, barbiturates, tricyclics, etc. consciousness allows the lesion to be localized: this helps to
2. Illicit drug exposure
a. Opioids, alcohols, amphetamines, etc. narrow the list of etiologic possibilities. For example, integrity
3. Environmental toxin exposure of brainstem function and absence of focal signs suggests a
a. Carbon monoxide toxic or metabolic disorder. In contrast, asymmetric findings
b. Heavy metals and brainstem dysfunction are more consistent with a struc-
c. Pesticides
tural etiology. The exam also is used to exclude conditions that
Metabolic Encephalopathies may mimic coma. Serial examinations of the comatose patient
1. Hypoglycemia, hyperglycemia over time are essential and provide information about treat-
2. Hyponatremia, hypernatremia ment efficacy, progression of the primary process, and prog-
3. Hypercalcemia nosis. The clinical examination of the comatose patient is
4. Hepatic encephalopathy
5. Uremia
discussed first, followed by some of the standardized scales
6. Vitamin deficiencies (thiamine, niacin) that have been developed to evaluate coma (Fig. 10.1).
7. Hypothermia, severe hyperthermia The coma examination is focused on four elements:
8. Hypothyroidism, hyperthyroidism (1) determination of the patient’s level of arousal (wake­
9. Urea cycle disorders fulness), (2) eye examination, (3) motor responses and pres-
Infections ence of abnormal reflexes, and (4) observation of breathing
1. Urinary tract infections patterns.
2. Pneumonia
3. Sepsis
Level of Consciousness
Adapted from Stevens RD, Bhardwaj A. Approach to the comatose patient.
Crit Care Med 2006;34(1):31–41. Patients who exhibit spontaneous eyes opening, verbalization
attempts, moaning, tossing, reaching, leg crossing, yawning,
coughing, or swallowing have a higher level of consciousness
chapter; however, a few warrant special mention because rapid than those who do not. The examiner should next assess
diagnosis and urgent treatment are essential to limit brain the patient’s response to a series of stimuli that escalate in
injury. These include seizures, infections, acute hydrocephalus, intensity. The patient’s name should be called loudly. If there
herniation, ischemic and hemorrhagic strokes, subarachnoid is no response, the examiner may stimulate the patient by
hemorrhage (SAH), cerebral venous sinus thrombosis, hyper- gently shaking him or her. If this produces no response, the
tensive encephalopathy, and TBI.4 examiner should use a noxious stimulus, such as pressure
 Section II—Clinical and Laboratory Assessment 87
Initial resuscitation
Airway
Breathing
Circulation
Cervical spine stabilization
Laboratory evaluation
Serum glucose, electrolytes, blood
gas, osmolality, liver and thyroid
function tests, ammonia, complete
blood count, toxicology screen
Coma examination
Initial/empiric treatment
Hyperventilation, mannitol 0.75-1.25 g/kg if evidence of
intercranial hypertension, herniation
Thiamine (100 mg IV) then glucose (50 cc of 50% solution)
Naloxone (0.4-2 mg IV q 3 minutes)
Flumazenil (0.2 mg/min to maximum dose of 1 mg IV) if
benzodiazepine overdose suspected
Gastric lavage with activated charcoal if other
drug intoxication suspected
Antibiotic administration if meningoencephalitis suspected
Induced hypothermia if comatose after VT/VF
Cardiac arrest per ILCOR guidelines
Head CT if etiology unknown
or structural lesion suspected
Detailed history
and physical
examination
Lumbar puncture, EEG,
MRI as indicated
Fig. 10.1  Algorithm for initial approach to the comatose patient.
CT, Computed tomography; EEG, electroencephalogram; MRI, magnetic
resonance imaging; ILCOR, International Liaison Committee on
Resuscitation; IV, intravenous; VT/VF, Ventricular Tachycardia/Ventricular
Fibrillation.
to the supraorbital ridge, nailbeds, or sternum, or nasal tickle
with a cotton wisp. Responses such as grimacing, eye opening,
grunting, or verbalization should be documented. Motor
responses provide information not only about sensation and
limb strength but also level of consciousness. The examiner
should note whether stimuli produce “purposeful,” nonste-
reotyped limb movements, for example, reaching toward the
site of stimulation (“localization”). This implies a degree of
intact cortical function. Stereotyped limb movements gener-
ally are mediated by brain and spinal reflexes and do not
require cortical input. Examples include extension and inter-
nal rotation of the limbs (decerebrate posturing), upper
extremity flexion (decorticate posturing), and flexion at the
ankle, knee, and hip (“triple-flexion”). Several scales designed
to quantitate level of consciousness are available to help
reduce inter- and intraobserver variability and to facilitate
accurate follow-up. These scales (Glasgow Coma Scale [GCS],
Full Outline of UnResponsiveness [FOUR] score, Reaction
Level Scale 85 [RLS85], and Innsbruck Coma Score are dis-
cussed later in this chapter.
The Eye Examination
In coma, the neuro-ophthalmologic examination should focus
on: (1) the pupils, (2) resting eye position and eye movements,
(3) retinal appearance, and (4) the corneal reflex.
Pupillary Examination
Pupillary examination is perhaps the most important part of
the coma examination, because it can help localize and dif-
ferentiate between structural lesions and diffuse metabolic
encephalopathies that cause bilateral cerebral hemispheric
dysfunction. Pupillary constriction is mediated by cholinergic
parasympathetic efferents that arise from the Edinger-
Westphal nucleus in the midbrain and travel within the ocu-
lomotor nerve. Pupillary dilation, in contrast, is mediated by
sympathetic efferents originating in the hypothalamus that
travel through the brainstem and cervical spinal cord and then
synapse onto postganglionic neurons in the superior cervical
ganglion. These noradrenergic neurons then course along the
carotid artery into the cavernous sinus and through the supe-
rior orbital fissure before reaching the iris.
Pupillary size, shape, and reactivity to light should be
assessed. In general, abnormalities of the pupillary light reflex
suggest a structural abnormality. However, certain drugs also
may affect the pupillary light reflex. These agents can cause
pupillary abnormalities that are mistakenly attributed to
structural brain lesions (Table 10.2). Metabolic causes of
coma typically do not affect the pupils.
Normal pupils are round, have equal diameters, and briskly
constrict when illuminated. When unequal pupils (anisoco-
ria) are observed, it is important to establish whether it is the
larger or the smaller pupil that is abnormal. This is accom-
plished by examining the eyes both in the light and in the dark.
When the lights are extinguished, an abnormally small pupil
will fail to dilate fully and the degree of anisocoria will increase.
In contrast, when the abnormal pupil is the larger one, the
degree of anisocoria will be maximal under full illumination
when the larger pupil fails to constrict fully.
In the NCCU, the most important causes of a unilaterally
dilated pupil are compressive lesions of the oculomotor nerve
complex (e.g., uncal herniation or a posterior communicating
artery aneurysm). A complete third nerve palsy results in ipsi-
lateral mydriasis, inferolateral deviation of the eye, and ipsilat-
eral ptosis. As a rule a third nerve palsy with pupil involvement
means a surgical lesion (i.e., compressive), whereas when the
pupil is spared the cause is medical (e.g., diabetes, meningo-
vascular syphilis). Patients with myasthenia gravis may present
with ptosis and ophthalmoparesis secondary to neuromuscu-
lar weakness that is often unilateral. Therefore, this condition
also should be considered in patients who present with what
appears to be a pupil-sparing third nerve palsy.
The most important cause of a unilateral small pupil is
Horner syndrome, which is caused by damage to sympathetic
efferents to the eye and consists of miosis and mild ipsilateral
ptosis. Horner syndrome can be seen from multiple areas of
brain damage, including (1) hypothalamic injury, (2) brain-
stem damage to descending inputs that synapse onto cervical
88 Section II—Clinical and Laboratory Assessment

Table 10.2  Drugs That Affect Pupillary Responses

Drug Class Examples Effect on Pupils Comments
Muscarinic Atropine, scopolamine, Dilate pupils by blocking action Ipratropium and tiotropium can cause
antagonists ipratropium, tiotropium, of acetylcholine inadvertent pupillary dilation if
tropicamide accidentally gets into eye during
respiratory treatment. Tropicamide
eyedrops commonly used to dilate pupils
for ophthalmologic evaluation.
Beta agonists Epinephrine, isoproterenol, Dilate pupils by activating Albuterol, salmeterol, and formoterol can
salmeterol, formoterol, beta-adrenergic receptors cause inadvertent pupillary dilation if
albuterol accidentally gets into eye during
respiratory treatment.
Cholinesterase Physostigmine, edrophonium, Constrict pupils by increasing Rivastigmine and donepezil used for
inhibitors pyridostigmine, rivastigmine, acetylcholine concentration treatment of dementia. Pyridostigmine
donepezil, organophosphates at synaptic cleft used for treatment of myasthenia gravis.
Muscarinic Pilocarpine Constrict pupils by activating Pilocarpine eyedrops are used to
agonists muscarinic acetylcholine distinguish pharmacologic pupil dilation
receptors from oculomotor nerve injury.
Serotonin-2A Lysergic acid diethylamide Dilate pupils through unclear Commonly used recreational hallucinogen.
(5HT-2A) (LSD) mechanism
receptor
agonists
Monoamine Cocaine, hydroxyamphetamine Dilate pupils by blocking Hydroxyamphetamine eyedrops can help
reuptake reuptake of norepinephrine distinguish if constricted pupil is from
inhibitors at synaptic cleft preganglionic or postganglionic injury.
Norepinephrine Hydroxyamphetamine Dilate pupils by increasing See above.
release norepinephrine concentration
potentiators at synaptic cleft
Opiate receptor Morphine, fentanyl, Constrict pupils Commonly used analgesics in hospitalized
agonists hydromorphone and critically ill patients

sympathetic preganglionic neurons, (3) cervical spine injury,
(4) injury to sympathetic paravertebral ganglia (e.g., with an
apical lung mass), or (5) damage to sympathetic postgangli-
onic fibers because they course along the internal carotid
artery (e.g., carotid artery dissection). Depending on the
lesion location, ipsilateral facial anhidrosis and other focal
neurologic signs also may be present. The presence of Horner
syndrome unequivocally places the lesion ipsilateral to the
pupillary abnormality. Bilaterally fixed and dilated pupils are
seen in the terminal stages of brain death but also with anti-
cholinergic medications, such as atropine. Hyperadrenergic
states (e.g., pain, anxiety, cocaine intoxication) produce bilat-
erally large and reactive pupils. Reactive pinpoint (<1 mm)
pupils are observed with opiate and barbiturate intoxication,
and after extensive pontine injury. Patients with long-standing
diabetes mellitus may have small pupils secondary to
hyperglycemia-induced damage of the sympathetic fibers that
mediate pupillary dilation.
Resting Eye Position and Eye Movements
Eye position and spontaneous movements including horizon-
tal or vertical misalignment, spontaneous roving, or rhythmic
and repetitive vertical movements should be looked for and
documented. The frontal lobe cortex (frontal eye fields) medi-
ates conjugate deviation of the eyes toward the contralateral
side. Lateral deviation of both eyes therefore indicates a
destructive lesion in the ipsilateral frontal lobe or an excitatory
focus (seizure) in the contralateral hemisphere. A destructive
unilateral pontine lesion, and rarely, a thalamic lesion, causes
conjugate deviation to the contralateral side. Downward devi-
ation (or lack of upward gaze) of the eyes is caused by dys-
function of the dorsal midbrain and may be seen with
hydrocephalus, tumors, strokes, and pineal region abnormali-
ties. Dysconjugate gaze frequently is seen in sedated patients
and usually represents unmasking of a latent eso- or exopho-
ria. Roving, or slow to-and-fro eye movements, implies
functional integrity of the brainstem.4 Ocular bobbing—fast
conjugate downward gaze followed by a slow upward correc-
tion to midposition—implies extensive pontine injury.28
Ocular dipping—slow conjugate downward gaze followed by
fast upward gaze—also localizes to the pons.29 With a skew
deviation, a vertical misalignment of the eyes that is not
caused by an isolated cranial nerve palsy, the lesion usually is
in the midbrain on the side of the higher eye, or in the ponto-
medullary junction on side of the lower eye.30,31 In comatose
patients there is a high incidence of nonconvulsive seizures,
and jerking movements of the eyes may be the only evidence
of seizure activity.
If spontaneous eye movements are absent, then an oculo-
cephalic response (“doll’s eyes”) should be sought by turning
the head horizontally and vertically. This maneuver should not
be performed on trauma patients with known or suspected
cervical spine instability. Normally the eyes move opposite to
the direction of head turning. Testing the oculocephalic
response may uncover a vertical gaze paresis, a skew deviation,
or a sixth nerve palsy that was not otherwise obvious.
If an oculocephalic response cannot be elicited, then an
oculovestibular (“cold-caloric”) response is sought. First, the
tympanic membrane should be visualized to ensure that is
 Section II—Clinical and Laboratory Assessment 89
intact and unobstructed. The head of the bed should be set at
30 degrees to align the patient’s horizontal semicircular canals
parallel to the floor. Then, using an angiocatheter or a but-
terfly catheter without the needle, 30 to 60 cc of ice-cold water
are instilled into the external auditory canal against the tym-
panic membrane. This inhibits the ipislateral vestibular system
and normally causes the eyes first to move slowly toward the
ipsilateral ear and then to jerk quickly toward the contralateral
ear. The initial slow response is mediated by the unopposed
contralateral vestibular system in the brainstem, and the
frontal eye fields mediate the subsequent corrective nystag-
mus. With bilateral cortical dysfunction and an intact brain-
stem, slow tonic deviation of the eyes toward the ipislateral
ear is observed and is not followed by contralateral nystagmus.
In early metabolic coma, the oculocephalic and oculovestibu-
lar responses are preserved. Absent response indicates diffuse
brainstem dysfunction and is seen in primary brainstem
injury, late transtentorial herniation, barbiturate intoxication,
and brain death.
Retinal Examination
A funduscopic examination should be considered to look for
signs of intracranial hypertension. However, successful fun-
duscopy often requires that the pupil be dilated; if doing this
pharmacologically may confuse overall patient assessment,
retinal examination may be deferred. Papilledema is swelling
of the optic nerve head from increased ICP. It is almost always
bilateral and may be accompanied by retinal hemorrhages,
exudates, and cotton wool spots, and ultimately by enlarge-
ment of the optic cup. Papilledema develops over hours to
days. Its absence therefore does not imply normal ICP, espe-
cially in the acute setting. Spontaneous pulsatility of the
retinal veins implies normal ICP. Terson syndrome is vitreous,
subhyaloid, or retinal hemorrhage associated with SAH. Pap-
illedema itself can be associated with visual loss,32 and in
general when seen, even in a patient with normal conscious-
ness, should prompt immediate investigation as to its etiology
so treatment can be started in a timely manner.
The Corneal Reflex
The cornea of each eye is gently touched with a drop of
saline or a cotton wisp while observing for eyelid closure
to test the corneal reflex. Failure of unilateral eyelid closure
suggests facial nerve dysfunction on that side. Failure of
bilateral eyelid closure with stimulation of one cornea, but
not the other, implies trigeminal nerve dysfunction on the
stimulated side. Failure of bilateral eyelid closure on stimula-
tion of either cornea usually implies pontine dysfunction.
Motor Responses and Abnormal Reflexes
The symmetry of motor responses and reflexes, and the pres-
ence of abnormal movement often permits discrimination
between structural and systemic etiologies of altered mental
status. First, the patient should be observed for any abnormal
or spontaneous movements. Asterixis implies a metabolic dis-
turbance such as uremia or hepatic encephalopathy. Twitching
or jerking of the face or limbs, even if subtle, raises the suspi-
cion for seizures. Asymmetry of resting limb position may be
a sign of weakness. For example, a paretic leg may lie
externally rotated. Next, the patient is stimulated and the
examiner should observe for asymmetry in the patient’s face
(grimace) and limb motor responses. A less vigorous response
on one side of the body indicates a contralateral structural
lesion that involves the motor pathways above the level of the
caudal medulla. In general, weakness associated with a pyra-
midal (upper motor neuron) lesion is found in the extensor
muscle groups of the arms and the flexor muscle groups of
the legs. Hence a patient with a hemiparesis may lie with the
arm flexed and the leg extended. Tone, on the other hand, is
increased in the muscle groups that remain strong. Testing
handgrip strength is not reliable because this function often
is affected late in the development of spastic hemiparesis. In
addition a handgrip may represent an abnormal reflex associ-
ated with frontal lobe disease. Another important reflex is the
plantar response. An up-going toe is abnormal and is called a
Babinski reflex (i.e., there is no such thing as a negative Babin-
ski). This reflex can be difficult to elicit and is best obtained
by stroking the lateral border of the sole and then across the
ball of the foot. The toes must not be touched. It is the first
movement of the big toe that is important. For example, an
individual with very sensitive feet may first flex the big toe and
then extend all the toes and withdraw the leg—this is a normal
response. An up-going toe is associated with an upper motor
neuron lesion. The presence of an abnormal plantar reflex can
be cross-checked with the abdominal reflex obtained by gently
stroking the abdominal wall and looking for abdominal wall
muscle contraction. Absence of an abdominal reflex implies
an upper motor neuron lesion above T9. The reflex can be
difficult to assess in a patient who is obese or has had multiple
abdominal surgeries.
Paraparesis and quadraparesis raise the possibility of spinal
cord injury, especially in the setting of trauma. In these
patients, in whom spinal cord injury or compression is a con-
sideration, how they respond to sensory stimuli needs to be
factored into what motor responses occur. Subtle findings
such as sweat patterns may help indicate a sensory level in a
comatose patient with a spinal cord abnormality. By looking
at plantar and abdominal reflexes, the level of injury can be
differentiated further. Priapism (reflex erection) results from
loss of sympathetic tone with injury to the spinal cord. The
bulbocavernosus reflex is contraction of the anal sphincter
when the penile shaft is pinched or a Foley catheter is pulled.
Its absence is associated with a spinal cord injury. The Ameri-
can Spinal Injury Association (ASIA) Muscle Grading Scale
can be used to classify patients with spinal cord injury (Table
10.3). This is different from the ASIA Impairment Scale that
defines the severity of the injury (Table 10.4).
Breathing Patterns
Several different breathing patterns that depend on the type
of injury and location of the pathology may be observed in
coma. However, in clinical practice breathing patterns often
are obscured by the use of sedatives, paralytics, and mechani-
cal ventilation. Apneustic respirations are characterized by a
prolonged end-inspiratory pause. This pattern may be seen
after focal injury to the dorsal lower half of the pons (e.g.,
stroke), but also may be observed with meningitis, hypoxia,
and hypoglycemia. Sustained hyperventilation can be seen in
metabolic encephalopathies such as hepatic coma, sepsis, and
diabetic ketoacidosis, but also can be seen as a consequence of
90 Section II—Clinical and Laboratory Assessment
Table 10.3  American Spinal Injury
Association Muscle Grading for Spinal
Cord Injury
Grade Clinical Description
0 Total paralysis
1 Palpable or visible contraction
2 Active movement, full range of motion, gravity
eliminated
3 Active movement, full range of motion, against
gravity
4 Active movement, full range of motion, against
gravity and provides some resistance
5 Active movement, full range of motion, against
gravity and provides normal resistance
5* Muscle able to exert, in examiner’s judgment,
sufficient resistance to be considered normal if
identifiable inhibiting factors were not present
Table 10.4  American Spinal Injury
Association Impairment Scale
A: Complete: no motor or sensory function is preserved in the
sacral segments S4-S5.
B: Incomplete: sensory but not motor function is preserved
below the neurologic level and includes the sacral
segments S4-S5.
C: Incomplete: motor function is preserved below the
neurologic level, and more than half of key muscles below
the neurologic level have a muscle grade less than 3.
D: Incomplete: motor function is preserved below the
neurologic level, and at least half of key muscles below the
neurologic level have a muscle grade of 3 or more.
E: Normal: motor and sensory function are normal.
focal injury to the upper pons—usually either from secondary
neurogenic pulmonary edema caused by intrinsic brainstem
injury33 or because of a tumor that creates a local cerebrospi-
nal fluid (CSF) acidosis, which stimulates brain chemorecep-
tors to trigger hyperventilation.34 Cluster breathing consists of
several rapid, shallow breaths followed by a prolonged pause,
and localizes to the upper medulla. Ataxic respirations, or
Biot’s breathing, is a chaotic pattern in which the length and
depth of the inspiratory and expiratory phases are irregularly
irregular. It may occur after injury to the respiratory centers
in the lower medulla. Apnea may be seen in a variety of neu-
rologic and non-neurologic disorders and by itself is of little
localizing value. Kussmaul respirations are rapid, deep breaths
that usually signal metabolic acidosis, but also may be observed
with pontomesencephalic lesions. Cheyne-Stokes respiration
refers to alternating spells of apnea and crescendo-decrescendo
hyperpnea. It has limited value in lesion localization and is
seen with diffuse cerebral injury, hypoxia, hypocapnea, and
congestive heart failure. Agonal gasps reflect bilateral lower
medullary injury and occur in the terminal stages of brain
injury.
Standardized Coma Rating Scales
Coma scales are important because it can be difficult to oth-
erwise define levels of consciousness. States of consciousness
fall along a spectrum, with coma at one end and normal con-
sciousness at the other. Patients who are in coma do not
respond to external stimuli in a “purposeful” manner but may
demonstrate reflexive behavior. Their eyes are closed and
sleep-wake cycles are absent. Coma is usually prolonged and
lasts for at least hours to days, but rarely is permanent. Instead
coma progresses to death or to a higher level of consciousness
including the vegetative or minimally conscious state (see
Chapters 8 and 13). Deterioration of normal consciousness is
often designated by terms such as “confusion” or “delirium,”
“stupor,” and “coma.” These labels are imprecise and have been
defined inconsistently from study to study, and even occasion-
ally within a given study. Several attempts have been made to
codify criteria for coma, vegetative state, minimally conscious
state, or delirium.4 These attempts, although laudable, have
not yet led to universal adoption of standard terminology. For
practical purposes for the intensive care unit (ICU) physician,
it is advisable to simply describe the clinical exam and use a
validated coma scale.
Several quantitative coma scales have been developed to
measure the degree of coma and extent of neurologic injury
in patients. Standardized scales provide an objective measure
of coma that facilitates communication between physicians
and ancillary medical staff, a means of comparing patients
with varying coma etiologies, and a simple mechanism to
track clinical changes over time. Scales describing sedation
and agitation (e.g., Ramsay Sedation Scale),35 Richmond Agi-
tation Sedation Scale,36 Vancouver Interaction and Calmness
Scale,37 Riker Sedation-Agitation Scale,38 and the American
Association of Critical-Care Nurses’ Sedation Assessment
Scale for Critically Ill Patients39 also are used in critical
care and are reviewed in Chapter 11. (See also www.mc.
vanderbilt.edu/icudelirium/docs/CAM_ICU_training.pdf for
training in delirium assessment.)
Glasgow Coma Scale
The GCS (Table 10.5) is the most widely used standardized
scale to assess the degree of coma and neurologic injury in
patients and has been incorporated into several models that
predict outcome in a variety of neurologic insults. It was ini-
tially developed for triage in patients with head trauma40,41 but
it has since been shown to correlate with survival and neuro-
logic outcome in TBI,42-46 nontraumatic coma,47 intracerebral
hemorrhage,48,49 SAH,50 dementia,51 and meningitis.52-54 To
calculate the GCS a patient is given a numeric score in three
categories (eye response, motor response, and verbal response)
and then these numbers are added together to get the full score
(see Table 10.5). The score can range from 3 to 15; a GCS of 13
or higher signifies mild brain injury, 9 to 12 signifies moderate
brain injury and 8 or less signifies severe brain injury and is
frequently used to define coma.4 The GCS is simple to calcu-
late and provides a rapid estimate of the degree of neurologic
injury. It has been widely adopted by emergency medical per-
sonnel to evaluate patients both in the field and in the emer-
gency department (ED) to triage and guide initial therapy. The
GCS has been incorporated into other validated scoring
systems that assess the severity of critical illness such as the
Acute Physiology and Chronic Healthy Evaluation (APACHE
II) score55 and the Simplified Acute Physiology Score (SAPS II)
score56 or of SAH (e.g., the World Federation of Neurosurgical
Societies score; see following text) and independently predicts
 Section II—Clinical and Laboratory Assessment 91

The GCS is also problematic in young children because the

Table 10.5  Glasgow Coma Scale verbal component is not appropriate and they will not be able
Motor Response to appropriately follow motor commands.68,69 Several coma
scores have been developed specifically for children in an
Follows commands 6
attempt to compensate for differences in verbal and motor
Localizes pain 5 capabilities.70 The Pediatric GCS modifies the verbal and
Withdraws to pain 4 motor scores such that the highest verbal score is “alert,
Flexion 3
babbles, coos words or sentences—normal for age” and the
highest motor score is “normal spontaneous” movement.71
Extension 2 Another score developed to assess level of consciousness in
None 1 infants with TBI has low interrater reliability and is not in
VERBAL RESPONSE widespread use.72
Oriented 5
Confused speech 4 Reaction Level Scale 85
Inappropriate words 3 Used almost exclusively in Sweden, the RLS8573 is a linear
Incomprehensible 2 score that is based on a patient’s level of consciousness from
1 (unconscious and unresponsive) to 8 (complete conscious-
None 1
ness). Rather than having subscores in different categories that
EYE OPENING are then added together, the RLS85 assigns a single number to
Spontaneous 4 the patient’s level of consciousness. This removes the potential
To command 3
ambiguity of having multiple ways to arrive at the same score
through different permutations. It has been studied and vali-
To pain 2 dated for a variety of neurologic conditions including isch-
None 1 emic and hemorrhagic stroke, TBI, and drug intoxication.74-77
Adapted from Teasdale G, Jennett B. Assessment of coma and impaired
Multiple studies have compared the RLS85 with the GCS and
consciousness: a practical scale. Lancet 1974;2:81–4. have found good correspondence between the two scoring
systems.74,76-78 In addition, one study used the RLS85 to calcu-
late APACHE II scores rather than the GCS and found good
agreement between GCS- and RLS85-based APACHE II
mortality in the general critical care patient.57,58 In addition, scores.77 However, the RLS85 has not been widely adopted and
the GCS is used to classify patients with acute brain injury and so there are few studies to validate interrater reliability.76,77
select them for various clinical trials.43-46 Websites such as In addition, there is some inherent ambiguity in the score
those based on the Corticosteroid Randomisation after Sig- because an RLS85 score of either 2 or 3 can signify “drowsy
nificant Head Injury (CRASH) study (www.crash.lshtm.ac.uk/ with response to strong stimulation.”
Risk%20calculator/index.html) or the International Mission
for Prognosis and Analysis of Clinical Trials in TBI (IMPACT)
project (www.tbi-impact.org/) use the GCS to help provide Innsbruck Coma Score
online prognostic predictor models after brain injury. This score79 is based on eight different categories: (1) eye
Despite its widespread use and proven utility, the GCS has opening, (2) reaction to acoustic stimuli, (3) reaction to pain,
potential limitations. For example, deafness or blindness may (4) body posture, (5) pupil size, (6) pupil response to light,
affect the GCS. In intubated patients, the verbal score cannot (7) position and movements of the eyeballs, and (8) oral
be calculated and usually is removed from the final calcula- automatisms. This score has not gained popularity outside of
tion. The GCS may be difficult to interpret when applied to the hospital where it was initially developed, and interrater
patients with swollen eyelids or eyes that cannot be evaluated reliability is unclear. In addition, the evaluation of “oral
accurately, locked-in patients, aphasic patients, or intoxicated automatisms” may be ambiguous in a comatose patient with
patients who have slurred speech and fluctuating compliance. occasional reflexive movements.80
In these circumstances the GCS motor score often is used
alone and seems to carry adequate predictor information.59,60
In addition, a given score in the midrange (6-12) may be Full Outline of Unresponsiveness Score
assigned to patients with different degrees of impaired con- The FOUR score was developed in an attempt to rectify the
sciousness through different combinations of scores in each difficulties that may occur when assessing nonverbal (aphasic
of the three categories. Early studies suggested that there was or intubated) patients and the inability to specifically test
excellent reliability between different raters who used the brainstem reflexes using the GCS (Fig. 10.2). The score has
GCS to assess the same patient.40,61 However, more recent four categories: eyes, motor, brainstem, and respiration. Each
studies suggest that interrater reliability may be lower than category is given a score from 0 to 4. The four categories and
expected.62-65 There is some evidence that the score is less consistent rating scales for each category are reinforced by the
accurate when used by untrained or less experienced medical score acronym (FOUR), making the scoring system potentially
staff65, 66 emphasizing the need for proper training for this or easier to remember than the GCS, which uses a different score
any scale or classification system that is used. Structured inter- for each category. Because of the structure of the FOUR score,
views are described to enhance the reliability of the GCS locked-in patients (no motor response but can blink on
and Glasgow Outcome Scale (GOS).67 command) and patients in a persistent vegetative state can be
92 Section II—Clinical and Laboratory Assessment

Fig. 10.2  The FOUR score. Instructions for the assessment of the individual categories of the FOUR (Full Outline of UnResponsiveness) score. A, For
eye response (E), grade the best possible response after at least three trials in an attempt to elicit the best level of alertness. A score of E4 indicates
at least three voluntary excursions. If eyelids are closed, the examiner should open them and examine tracking of a finger or object. Tracking with
the opening of one eyelid will suffice in cases of eyelid edema or facial trauma. If tracking is absent horizontally, examine vertical tracking. Alternatively,
two blinks on command should be documented. This will recognize a locked-in syndrome (patient is fully aware). A score of E3 indicates the absence
of voluntary tracking with open eyes. A score of E2 indicates eyelids opening to a loud voice. A score of E1 indicates eyelids open to pain stimulus.
A score of E0 indicates no eyelid opening to pain. B, For motor response (M), grade the best possible response of the arms. A score of M4 indicates
that the patient demonstrated at least one of three hand positions (thumbs-up, fist, or peace sign) with either hand. A score of M3 (localization)
indicates that the patient touched the examiner’s hand after a painful stimulus compressing the temporomandibular joint or supraorbital nerve. A
score of M2 indicates any flexion movement of the upper limbs. A score of M1 indicates extensor response to pain. A score of M0 indicates no motor
response to pain, or myoclonus status epilepticus. C, For brainstem reflexes (B), grade the best possible response. Examine pupillary and corneal
reflexes. Preferably, corneal reflexes are tested by instilling two or three drops sterile saline onto the cornea from a distance of 4 to 6 inches (this
minimizes corneal trauma from repeated examinations). Sterile cotton swabs can also be used. The cough reflex to tracheal suctioning is tested only
when both of these reflexes are absent. A score of B4 indicates pupil and corneal reflexes are present. A score of B3 indicates one pupil wide and
fixed. A score of B2 indicates either pupil or cornea reflexes are absent. A score of B1 indicates both pupil and cornea reflexes are absent. A score
of B0 indicates pupil, cornea, and cough reflex (using tracheal suctioning) are absent. D, For respiration (R), determine spontaneous breathing pattern
in a nonintubated patient and grade simply as regular (R4), or irregular (R2), Cheyne-Stokes (R3) breathing. In mechanically ventilated patients, assess
the pressure waveform of spontaneous respiratory pattern or the patient triggering of the ventilator (R1). The ventilator monitor displaying respiratory
patterns can be used to identify the patient-generated breaths on the ventilator. No adjustments are made to the ventilator while the patient is
graded, but grading is done preferably with PaCO2 within normal limits. A standard apnea (oxygen-diffusion) test may be needed when a patient
breathes at ventilator rate (R0). (From Wijdicks EF, Bamlet WR, Maramattom BV, et al. Validation of a new coma scale: the FOUR score. Ann Neurol 2005;58:
585–93.)
 Section II—Clinical and Laboratory Assessment 93
unambiguously identified. The motor category allows “no
motor response” or “myoclonic status epilepticus” both to be
given a score of 0. This highlights the extremely poor long-
term prognosis of patients with neurologic injury who develop
myoclonic seizures. In addition, the motor component allows
coarse testing of language comprehension and praxis because
the best motor score involves the patient making a hand
gesture such as a fist or peace sign. Finally, a combined score
of 0 (no eye opening, absent motor responses, no brainstem
reflexes, no breathing over the set ventilator rate) is consistent
with brain death.
The FOUR score was initially validated in a prospective
single center study of 120 patients at the Mayo Clinic.81 These
patients were either directly admitted to an NCCU or were
seen by a consultant neurologist in a medical or surgical ICU
for “unresponsiveness.” Patients had a variety of diagnoses
including stroke, TBI, craniotomy for brain tumor, SAH,
seizure or status epilepticus, and primary CNS infection. This
study found that interrater reliability was similar to the GCS.
Subsequent studies validated this scale for use in both the ED
and medical intensive care unit setting.82-84 The FOUR score’s
simplicity and its ability to detect locked-in patients, uncal
herniation and brain death make it an attractive alternative to
the GCS. Whether it will gain widespread acceptance remains
to be seen.
AVPU and ACDU Score
These two scales were created to provide medical personnel
with a simple bedside tool to assess the severity of coma that
does not rely on detailed physical examination. In the AVPU
scale the patient is given a score based on whether he or she
is alert and oriented (A), responding to voice (V), respond-
ing to pain (P), or unresponsive (U). The similar ACDU
asks whether the patient is alert and oriented (A), confused
(C), drowsy (D), or unresponsive (U). These scales have a
similar accuracy to the GCS, and their simplicity makes
them attractive candidates for rapid assessment in emer-
gency settings.85
Clinical Scales Used in Stroke
Many clinical scales have been developed to evaluate
patients with ischemic stroke, intraparenchymal hemorrhage,
and subarachnoid hemorrhage. This section outlines some
of the most widely used stroke assessment scales. (The reader
also is referred to: www.strokecenter.org/professionals/stroke-
diagnosis/stroke-assessment-scales-overview, which details
a large number of clinical stroke scales and outcome
measures.)
Ischemic Stroke and
Intraparenchymal Hemorrhage
The most widely used neurologic score in stroke is the National
Institutes of Health Stroke Scale (NIHSS).86 It is based on
three previously used scales: the Toronto Stroke Scale, the
Oxbury Initial Severity Scale, and the Cincinnati Stroke
Scale.87 The NIHSS is a 15-item test and has scores that range
between 0 and 42, with a higher number indicating more
neurologic impairment. The NIHSS is complex and does
require certification (www.ninds.nih.gov/doctors/stroke_scale
_training.htm), but with proper training interrater reliability
can be high.88,89 A potential limitation with the NIHSS is that
it tends to give disproportionate weighting to symptoms
attributable to anterior circulation and cortical dysfunction
(e.g., focal weakness, aphasia, neglect). Thus patients with
cerebellar or brainstem infarcts may be markedly debilitated
but have low to moderate NIHSS scores. Despite this, the
NIHSS has proven invaluable to assess the clinical course
of hospitalized patients, to guide treatment decisions, and
to allow comparison in research studies. A full NIHSS
can be viewed at the National Stroke Association website
(www.stroke.org). Other scales used for ischemic stroke
include the Scandinavian Neurological Stroke Scale,90 the
Unified Stroke Scale,91 the European Stroke Scale,92 and the
Canadian Neurological Scale.93
Several scales are used to grade intracerebral hematomas.94
The original is the most widely validated score and provides
an estimate of 30-day mortality after intracerebral hematoma.
Its relationship to long-term functional outcome is less clear.
It is the sum of individual points scored as follows: GCS 3 to
4 (2 points), 5 to 12 (1), 13 to 15 (0); age equal to or more
than 80 years yes (1), no (0); infratentorial origin yes (1), no
(0); intracerebral hematoma volume equal or greater than 30
cm3 (1), less than 30 cm3 (0); and intraventricular hemorrhage
yes (1), no (0). Patients with a score of 0 should survive; those
with a score of 5 likely will die.95
Subarachnoid Hemorrhage
Several assessment scales have been developed to assess clini-
cal severity and provide prognostic information in SAH
patients. Of these, the Hunt and Hess scale and the World
Federation of Neurological Societies (WFNS) scale are two of
the most widely used instruments.
The Hunt and Hess scale was initially proposed as an assess-
ment tool to help neurosurgeons decide when to operate on
patients with aneurysmal SAH.96, 97 Patients receive a score
from I to V, with higher scores signifying increasing SAH
severity (Table 10.6). Grades IV and V are considered poor-
grade SAH patients. If systemic disease is present the patient’s
score should be increased by one level. The Hunt and Hess
scale is simple to administer and widely recognized; however,
many of the terms used to define grades are vague (e.g.,
Table 10.6  Hunt and Hess Scale Used to
Classify Severity of Nontraumatic
Subarachnoid Hemorrhage
Grade* Clinical Description
I Asymptomatic, mild headache, slight nuchal rigidity
II Moderate to severe headache, nuchal rigidity, no
neurologic deficit other than cranial nerve palsy
III Drowsiness or confusion, mild focal neurologic
deficit
IV Stupor, moderate to severe hemiparesis
V Coma, decerebrate posturing
*A modified version adds “grade 0”: unruptured aneurysm, and “grade Ia”:
no acute meningeal reaction but a fixed neurologic deficit.
94 Section II—Clinical and Laboratory Assessment
Table 10.7  World Federation of
Neurosurgical Societies Subarachnoid
Hemorrhage Grading Scale
Motor Deficit (Also
Grade GCS Includes Speech Deficit)
I 15 −
II 14-13 − while he or she is in the ICU. The Sequential Organ Failure
III 14-13 + Assessment (SOFA)106,107 was initially designed to quantify
IV 12-7 ± the degree of sepsis-related organ dysfunction in critically
V 6-3 ±
From Teasdale GM, Drake CG, Hunt W, et al. A universal subarachnoid
hemorrhage scale: report of a committee of the World Federation of
Neurosurgical Societies. J Neurol Neurosurg Psychiatry 1988;51(11):1457.
GCS, Glasgow Coma Scale.
drowsy, stupor) and interrater reliability can therefore be low,
in particular of grade III patients. Despite this, the Hunt and
Hess scale score and association with clinical outcome are well
described.98,99,100
The WFNS scale was proposed in 1988 as an alternative to
the Hunt and Hess scale to improve interrater reliability by
avoiding the inherent ambiguity in terms such as drowsy,
stupor, and deep coma.101 Like the Hunt and Hess scale, the
WFNS scale is graded from 0 to 5, with higher numbers rep-
resenting more severe SAH (Table 10.7). Patients are assigned
a score based on their GCS and the presence of focal neuro-
logic signs. The WFNS score is correlated with clinical
outcome.98, 99
Clinical Scores That Reflect Overall
Disease Severity
Several important scales have been developed to calculate
overall disease severity, guide treatment decisions, and predict
mortality and outcome in critically ill patients. The APACHE
II score is a well-validated score that is widely used in both
clinical practice and research literature to predict mortality
and outcome in adult patients.55 It is applied to patients
admitted to ICUs within the first 24 hours of admission. The
score is produced using physiologic measurements, laboratory
values, and the GCS, with higher numbers indicating more
severe disease. APACHE II score calculation is used to deter-
mine eligibility for certain treatments in the ICU. Importantly,
the score is by definition an admission score and is not recal-
culated sequentially. There are later versions in use (APACHE
III, IV) and several online calculators (e.g., www.mdcalc.com/
apache-ii-score-for-icu-mortality). The SAPS II is an admis-
sion disease severity score similar to the APACHE II that also
strongly predicts mortality56 and has been found useful in
SAH patient assessment.102 An admission score specific to
trauma is the Injury Severity Score (ISS).103 The ISS divides
the body into six sections: head and neck, face, chest, abdomen,
extremity, and external. Each section is assigned an Abbrevi-
ated Injury Score from 1 to 6 (1 being minor and 6 being an
unsurvivable injury). The highest scores in three categories are
taken and squared and then summed (see www.trauma.org/
archive/scores/iss.html). Other trauma scores include the
Trauma Injury Severity Score (TRISS)104 or the Revised
Trauma Score (RTS).105 The RTS is a physiologic scoring
system, with high interrater reliability and demonstrated
accuracy to predict death. It consists of GCS, systolic blood
pressure (BP), and respiratory rate that is scored when the
patient is admitted (see www.trauma.org).
Several scores have been developed to follow a patient
ill patients. It has since been used to track the severity of
illness in nonseptic, critically ill patients. The score assigns
1 to 4 points to six different organ systems (respiratory,
nervous, cardiovascular, liver, coagulation, and renal) based
on the severity of organ dysfunction as measured by labora-
tory values (for respiratory, liver, coagulation, and renal),
vital signs and pressor requirements (for cardiovascular)
and the GCS (for nervous system) (see www.mdcalc.com/
sequential-organ-failure-assessment-sofa-score). High scores
indicate more severe illness. The score has been used to
sequentially track the severity of illness in patients during
the course of their ICU stay. The SOFA score has also been
used to predict mortality, although it was not developed for
this purpose.108 Its use also has been validated in TBI,109 in
which it may have superior discriminative ability and stron-
ger association with outcome compared with other scoring
systems of extracerebral organ dysfunction such as the
multiple organ dysfunction score110 or the multiple-system
organ-failure score.111
The clinical assessment scores described here have not been
validated for children younger than age 15 and should not be
used in the pediatric population because of the different
normal physiologic parameters between children and adults.
There are, however, several pediatric specific scales that can be
used to predict disease severity. For example, the Pediatric
Index of Mortality (PIM 2)112 assesses disease severity in criti-
cally ill children similar to the APACHE II and SAPS II scores
used in adults. The PIM 2 uses the following pediatric inten-
sive care unit (PICU) admission variables to predict outcome:
systolic blood pressure (BP), pupillary reaction (fixed or reac-
tive), ratio of arterial partial pressure of oxygen to fraction of
inspired oxygen (PaO2/FiO2), base excess, elective admission
(yes/no [Y/N]), mechanical ventilation (Y/N), recovery from
surgery (Y/N), cardiac bypass (Y/N), and high risk/low risk
diagnosis. For children with traumatic injury, the Pediatric
Trauma Score (PTS)113 has been used to assess the severity of
injury. The PTS includes six variables: weight, systolic BP,
mental status, airway maintenance, skeletal injury, and open
wounds.
In addition to the clinical assessment scales discussed, a
variety of quantitative rating scales calculate the severity of
illness in a variety of conditions (e.g., liver failure, pulmonary
embolism, systemic inflammatory response syndrome, tran-
sient ischemic attack) that are beyond the scope of this chapter.
The reader is referred to www.mdcalc.com for a comprehen-
sive list of these different assessment instruments. Scores also
exist to predict development of ICU complications (e.g., pres-
sure sores)114, 115 (see www.bradenscale.com). Finally there are
scores that help quantify the intensity of care, for example, the
Therapeutic Intervention Scoring System.116 Of particular rel-
evance to the NCCU is the Therapeutic Intensity Level (TIL).117
 Section II—Clinical and Laboratory Assessment 95
The TIL was developed to measure the effort required to
control ICP in different clinical situations. The original score
does not include some therapies now in use (e.g., decompres-
sive craniectomy, hypothermia, or hypertonic saline), and the
maximum score originally was assigned for barbiturate use,
which now is uncommon. It has, however, been modified for
modern TBI care and is used in current phase III clinical trials
(www.em.emory.edu/protect/toolbox.cfm). The TIL provides
information that provides insight into compliance of the
intracranial compartment; that is, a patient with an ICP of 20
and a high TIL is far sicker than a patient with the same ICP
and low TIL.118 The Pediatric Intensity Level of Therapy
(PILOT)119 aimed to correct the deficiencies of the TIL score.
It assigns points for hypertonic therapy and it does not assign
maximal points for barbiturate use.
Laboratory Investigations to Help
Evaluate Coma
Metabolic and toxic encephalopathies account for a significant
proportion of altered mental status. Commonly encountered
metabolic disturbances include severe hyper- and hyponatre-
mia, hypercalcemia, elevated blood urea nitrogen (BUN),
hyperammonemia, hypo- and hyperglycemia, hypercarbia,
hypoxemia, and severe hyper- and hypothyroidism. A toxicol-
ogy screen detects exposure to common drugs and toxins
(Table 10.8). Plasma osmolality should be measured so that
the plasma osmolal gap can be determined. The osmolal gap
is the difference between the calculated osmolarity (2[Na] +
[BUN]/2.8 + [glucose]/18) and the measured osmolality. The
osmolal gap is elevated with intoxication from alcohols, such
as methanol, and ethylene glycol. A complete blood count
(white blood cell count) and differential should be obtained
to help assess for an infectious cause of encephalopathy.
Neuroimaging Studies and
Coma Evaluation
Noncontrast cranial compued tomography (CT) is indicated
in all new cases of unexplained coma and is the test of first
choice. CT rapidly identifies intra- and extra-axial cerebral
Table 10.8  Agents Commonly Tested for on
Routine Toxicology Screens
Cocaine
Amphetamines
Benzodiazepines
Barbiturates
Ethanol
Tetrahydrocannabinol (active ingredient in marijuana)
Phencyclidine (i.e., PCP)
Opiates*
Acetaminophen
Salicylates
*Routine opiate panels generally only test for morphine and its derivatives
(morphine, codeine, heroin). Testing for other synthetic opioids (e.g.,
hydrocodone, hydromorphone, methadone, fentanyl, oxycodone,
buprenorphine) requires extended opioid testing.
hemorrhages, brain herniation, cerebral edema, and hydro-
cephalus. Magnetic resonance imaging (MRI) is indicated in
patients whose coma remains unexplained after CT. MRI
has a higher sensitivity than CT for acute ischemic stroke,
intracerebral hemorrhage, inflammatory conditions, brain
abscesses, brain tumors, cerebral edema, cerebral venous sinus
thrombosis, and diffuse axonal injury. MRI also provides
much better visualization of brainstem structures than CT.
(The role of CT, MRI, and other imaging studies in the NCCU
is discussed in Chapters 26 through 29.) MRI is generally less
widely available than CT, is more time consuming, and
requires nonferromagnetic equipment, making it less feasible
for many critically ill patients. The risks of transporting
patients to the CT or MRI scanner, and the time it takes to
complete the studies must be weighed against the benefit of
the information they may yield. The development of portable
CT scanning equipment now allows bedside head CT scans to
be obtained in critically ill patients, and many new ICUs now
incorporate advanced imaging such as MRI and positron
emission tomography (PET) scanning in nearby locations.
Prognosis of the Comatose Patient
Coma, by definition, is self-limited. Survivors may recover to
a persistent vegetative state, to complete neurologic recovery,
or to an intermediate state of neurologic disability. The
Glasgow Outcome Scale (GOS) is an established metric of
recovery120 (Table 10.9). Although it fails to capture the full
breadth of possible neurologic outcomes, it is widely used by
clinical investigators for traumatic and nontraumatic coma.
How to determine the prognosis of comatose patients poses
a significant clinical and sometime ethical challenge. Studies
that have addressed the prognosis of coma are plagued by
numerous problems that limit their clinical applicability. First,
coma has been defined inconsistently across studies. Second,
virtually all studies involved patients in whom care was inten-
tionally limited, leading to a “self-fulfilling prophecy” bias.
Third, most studies were performed before major advances in
Table 10.9  The Glasgow Outcome Scale
1 Death
2 Persistent Patient exhibits no obvious cortical
vegetative state function
3 Severe disability (Conscious but disabled.) Patient
depends on others for daily
support due to mental or physical
disability or both.
4 Moderate (Disabled but independent.) Patient
disability is independent as far as daily life is
concerned. The disabilities found
include varying degrees of
dysphagia, hemparesis, or ataxia,
as well as intellectual and memory
deficits and personality changes.
5 Good recovery Resumption of normal activities even
though there may be minor
neurologic or psychological deficits
Adapted from Jennett B, Bond M. Assessment of outcome after severe brain
damage. Lancet 1975;1:480–4.
96 Section II—Clinical and Laboratory Assessment
intensive care (e.g., induced hypothermia) that have been
shown to improve outcome in certain patient populations.
Last, studies have not yet systematically addressed the impact
of clinical confounders such as metabolic disturbances, seda-
tive and paralytic mediations, and shock. Recent efforts by the
IMPACT investigators and others have attempted to address
prognosis after TBI,43-46,121-123 and tools to help calculate prog-
nosis are available online (e.g., www.tbi-impact.org).
The prognosis of coma is based in part on its etiology. For
example, coma after a high-velocity penetrating head injury
usually has a worse prognosis than a closed TBI, whereas a
patient in coma with an extradural hematoma is far more
likely to do well than one with a subdural hematoma. In
general nontraumatic coma has a worse outcome than trau-
matic coma, and among patients with nontraumatic coma,
nonstructural etiologies are associated with a better prognosis
than structural etiologies. However, to “predict” outcome
requires a careful consideration also of clinical signs and ancil-
lary tests including electrophysiologic, neuroimaging, and
biochemical studies.
Levy and colleagues124 performed a landmark prospective
study of outcome in 500 patients with nontraumatic coma,
primarily due to cardiac arrest, and found that prognosis
depended in large part on the duration of coma, neuro-
ophthalmologic signs, and motor function. The Levy criteria
have since become the benchmark to determine prognosis after
severe, nontraumatic brain injury, although newer advances
in critical care, especially therapeutic hypothermia,125,126 have
altered the accuracy of these prognostic criteria. Multiple
studies have examined the utility of additional clinical signs
and ancillary tests for prognosis in cardiac arrest survivors.
These data are summarized in a practice parameter issued by
the Quality Standards Subcommittee of the American Academy
of Neurology.127 Clinical signs that most accurately portend
poor prognosis in these patients include (1) myoclonic status
epilepticus within the first 24 hours after primary circulatory
arrest, (2) absence of pupillary responses within days 1 to 3
after cardiopulmonary resuscitation (CPR), (3) absent corneal
reflexes within days 1 to 3 after CPR, and (4) absent or extensor
motor responses after day 3. Although certain electroencepha-
lographic patterns, such as burst suppression and generalized
epileptiform discharges are associated with poor prognosis,
EEG lacks sufficient prognostic accuracy and is not recom-
mended as a tool to predict outcome.128, 129 Somatosenory
evoked potentials (SSEPs) appear to be more useful for
prognostication130-133 because SSEPs are less susceptible than
EEG to the confounding effects of drugs and metabolic
derangement. Bilateral absence of the N20 component of the
SSEP (see Chapter 24) with median nerve stimulation after
day 1 is associated with poor prognosis. Numerous serum and
CSF biomarkers have been studied for their prognostic value.
To date, the only marker with sufficient prognostic accuracy
to be recommended is neuron-specific enolase (NSE). Serum
NSE levels of greater than 33 mcg/L on days 1 through 3 after
cardiac arrest predict poor outcome127, 134; however, this test is
not yet widely available and there is evidence that the false-
positive rate may be unacceptably high in patients treated with
therapeutic hypothermia.135 Ongoing studies address the prog-
nostic role of cranial CT, MRI, and magnetic resonance spec-
troscopy after coma both from traumatic and nontraumatic
causes.136-143 These are discussed further in Chapters 26
and 28.
The Vegetative and Minimally
Conscious State
Coma survivors who fail to recover consciousness tradition-
ally have been diagnosed as being in “vegetative” or “mini-
mally conscious” states. Patients with vegetative states regain
some arousal, but without apparent true awareness. If there is
no further improvement after 4 weeks, the patient is stated to
be in a persistent vegetative state.144-146 Vegetative patients
may demonstrate sleep-wake cycles, may open their eyes and
track objects, and may be alert and swallow with feeding.
Patients in minimally conscious states have greater evidence
of awareness and interaction including verbal or gestural
responses, reaching for objects, and emotional responses.144
However, the anatomic and physiologic differences between
vegetative and minimally conscious states are poorly under-
stood and remain difficult to define clinically. This results in
both ethical and legal challenges to managing patients in
these conditions.
Specialized rating scales that assess cognitive function in
minimally conscious and vegetative patients have been devel-
oped to help aid outcome prediction and monitor the effec-
tiveness of treatments in long-term rehabilitation centers.
The JFK Coma Recovery Scale (CRS) was initially described
in 1991147 and is comprised of hierarchically arranged
items associated with brainstem and subcortical processes.
The initial CRS was revised148 to aid in the differentiation of
minimally conscious and vegetative patients. The CRS-R com-
prises six subscales that address auditory, visual, motor, oro-
motor, communication, and arousal functions. Patients are
scored based on the presence or absence of different behav-
ioral responses to a variety of stimuli that range from cogni-
tively complex behaviors (e.g., object recognition, functional
object use, intelligible verbalization) to reflexive behaviors
(e.g., startle responses, posturing, oral reflexive movements)
to no response. Validity analyses support use of the scale as an
index of neurobehavioral function and have shown that the
CRS-R is capable of discriminating patients in minimally con-
scious state from those in vegetative state, which is of critical
importance to establish prognosis and formulate treatment
intervention.149
Although clinical assessment of the level of consciousness
has traditionally relied on behavioral responses, functional
imaging methods have begun to be used to directly interrogate
brain function.150 In general, cerebral blood flow and metabo-
lism reflects the level of arousal, with locked-in or minimally
conscious patients showing relatively normal function, vegeta-
tive patients showing reduced function, and brain dead
patients showing no function, though certain networks of
brain regions are clearly more closely tied to the level of con-
sciousness than others. Mental imagery tasks are known to
produce regional brain activation approximating actual task
performance, and a functional MRI study carried out in a
patient thought to be in a persistent vegetative state 5 months
after severe TBI demonstrated task-specific activation patterns
during motor imagery similar to those observed in control
subjects, indicating that the patient was able to understand the
task instruction and modulate brain activity in response to
exogenous cues despite the absence of convincing behavioral
responses.151 A follow-up study152 observed that 5 out of 54
vegetative or minimally conscious patients were able to will-
fully modulate brain activity and produce activation patterns
 Section II—Clinical and Laboratory Assessment 97
similar to those seen in healthy control subjects on functional
MRI (fMRI), and in one case two types of mental imagery
were used to verify “yes” or “no” answers to several questions,
demonstrating the feasibility of using brain responses to
imagery as a means of communication. Although fMRI is not
a practical means for communication, other modalities such
as transcranial optical spectroscopy and EEG could be used to
transduce brain responses at the bedside or to drive brain
computer interfaces in patients who are otherwise locked-in.
Another fMRI approach to assess brain function is to
examine the integrity of correlated activity within distributed
networks that occurs independently of any voluntary task.
Activity in one such network, termed the “default mode
network,”153 is closely tied to a subject’s level of attentiveness
and shows graded alterations in patients with disorders of
consciousness, with locked-in patients showing normal con-
nectivity, and minimally conscious and unconscious patients
showing reduced connectivity.154 These and related studies
emerging in the literature suggest that functional brain
imaging may ultimately supplement clinical examination in
assessing the integrity of conscious awareness, though signifi-
cant technical, practical, and ethical challenges remain in the
development and validation of their use.
Conclusion
Altered mental status and coma occur commonly in the ICU.
Physicians should develop an efficient diagnostic approach
that facilitates rapid treatment to minimize additional brain
injury. The clinical exam, laboratory tests, and neuroimaging
studies are powerful tools that help to determine whether a
structural or systemic abnormality exists, thus narrowing the
differential diagnosis. Standardized clinical scales allow the
measurement of disease severity, help to track clinical prog-
ress, aid in communication between caregivers, and can give
valuable prognostic information. Currently there are few
widely available means of reliably determining prognosis in
comatose patients. Genetic and metabolic markers and
advanced neuroimaging and neurophysiologic methods may
ultimately prove helpful.
References
1. Von Economo C. Sleep as a problem of localization. J Nerv Ment Dis 1930;
71:249–59.
2. Morruzi G, Magoun HW. Brain stem reticular formation and activation of
the EEG. J Neuropsychiatry Clin Neurosci 1949;7:251–67.
3. Parvizi J, Damasio AR. Neuroanatomical correlates of brainstem coma. Brain
2003;126:1524–36.
4. Posner JB, Saper CB, Schiff ND, et al. Plum and Posner’s diagnosis of stupor
and coma. 4th ed. New York: Oxford University Press; 2007.
5. Blumenfeld H. Neuroanatomy through clinical cases. Sunderland, MA:
Sinauer Associates; 2010.
6. Hallanger AE, Levey AI, Lee HJ, et al. The origins of cholinergic and other
subcortical afferents to the thalamus in the rat. J Comp Neurol 1987; 262:
105–24.
7. Saper CB, Scammell TE, Lu J. Hypothalamic regulation of sleep and
circadian rhythms. Nature 2005;437(7063):1257–63.
8. Saper CB, Fuller PM, Pedersen NP, et al. Sleep state switching. Neuron 2010;
68(6):1023–42.
9. Saper CB. Organization of cerebral cortical afferent systems in the rat. II.
Magnocellular basal nucleus. J Comp Neurol 1984;222:313–42.
10. Saper CB. Diffuse cortical projection systems: anatomical organization and
role in cortical function. In: Plum F, editor. Handbook of physiology: the
nervous system. V. Bethesda, MD: American Physiological Society; 1987.
pp. 169–210.
11. Bassant MH, Ennouri K, Lamour Y. Effects of iontophoretically applied
monoamines on somatosensory cortical neurons of unanesthetized rats.
Neuroscience 1990;39:431–9.
12. Kolta A, Reader TA. Modulatory effects of catecholamines on neurons of the
rat visual cortex: single cell iontophoretic studies. Can J Physiol Pharmacol
1989;67:615–23.
13. McCormick DA. Cholinergic and noradrenergic modulation of
thalamocortical processing. Trends Neurosci 1989;12(6):215–21.
14. McCormick DA, Bal T. Sleep and arousal: thalamocortical mechanisms.
Annu Rev Neurosci 1997;20:185–215.
15. Sato H, Fox K, Daw NW. Effect of electrical stimulation of locus coeruleus
on the activity of neurons in the cat visual cortex. J Neurophysiol 1989;62:
946–58.
16. Bromberg-Martin ES, Matsumoto M, Hikosaka O. Dopamine in
motivational control: rewarding, aversive, and alerting. Neuron 2010;68:
815–34.
17. Schultz W. Predictive reward signal of dopamine neurons. J Neurophysiol
1998;80(1):1–27.
18. Sawa A, Snyder SH. Schizophrenia: diverse approaches to a complex disease.
Science 2002;296(5568):692–5.
19. Marié RM, Defer GL. Working memory and dopamine: clinical and
experimental clues. Curr Opin Neurol 2003;S29–35.
20. Haenisch B, Bönisch H. Depression and antidepressants: insights from
knockout of dopamine, serotonin, or noradrenaline re-uptake transporters.
Pharmacol Ther 2011;129(3):352–68.
21. Ravindran LN, Stein MB. The pharmacologic treatment of anxiety disorders:
a review of progress. J Clin Psychiatry 2010;71(7):839–54.
22. Adamantidis AR, Zhang F, Aravanis AM, et al. Neural substrates of
awakening probed with optogenetic control of hypocretin neurons. Nature
2007;450(7168):720–4.
23. Willie JT, Chemelli RM, Sinton CM, et al. To eat or to sleep? Orexin in the
regulation of feeding and wakefulness. Annu Rev Neurosci 2001;24:429–58.
24. Schjolberg A, Sunnerhagen KS. Unlocking the locked-in: a need for team
approach for survivors of locked-in syndrome. Acta Neurol Scand 2011.
Epub ahead of print.
25. Gelenberg AJ. The catatonic syndrome. Lancet 1976;1:1339–41.
26. Cairns H. Head injuries in motor-cyclists: the importance of the crash
helmet. Br Med J 1941;2(4213):465–71.
27. Kumral E, Bayulkem G, Evyapan D, et al. Spectrum of anterior cerebral
artery territory infarction: clinical and MRI findings. Eur J Neurol 2002;9(6):
615–24.
28. Fisher CM. Ocular bobbing. Arch Neurol 1964;11:543–6.
29. Rosenberg ML. Spontaneous vertical eye movements in coma. Ann Neurol
1986;20(5):635–7.
30. Brandt TH, Dieterich M. Different types of skew deviation. J Neurol
Neurosurg Psychiatry 1991;54(6):549–50.
31. Keane JR. Ocular skew deviation: analysis of 100 cases. Arch Neurol 1975;
32(3):185–90.
32. Schirmer CM, Hedges TR. Mechanisms of visual loss in papilledema.
Neurosurg Focus 2007;23(5):E5.
33. Simon RP. Neurogenic pulmonary edema. Neurol Clin 1993;11(2):309–23.
34. Tarulli AW, Lim C, Bui JD, et al. Central neurogenic hyperventilation: a case
report and discussion of pathophysiology. Arch Neurol 2005;62(10):
1632–4.
35. Ramsay MAE, Savege TM, Simpson BRJ, et al. Controlled sedation with
alphaxalone-alphadolone. BMJ 1974;2:656–9.
36. Sessler CN, Gosnell MS, Grap MJ, et al. The Richmond agitation-sedation
scale: validity and reliability in adult intensive care unit patients. Am J Respir
Crit Care Med 2002;166:1338–44.
37. De Lemos J, Tweeddale M, Chittock D. Measuring quality of sedation in
adult mechanically ventilated critically ill patients: the Vancouver interaction
and calmness scale. J Clin Epidemiol 2000;53:908–19.
38. Fraser GL, Riker R. Monitoring sedation, agitation, analgesia, and delirium
in critically ill adult patients. Crit Care Clin 2001;17:1–21.
39. De Jong MMJ, Burns SB, Campbell ML, et al. Development of the American
Association of Critical-Care Nurses’ Sedation Assessment Scale for Critically
Ill Patients Am J Crit Care 2005;14:531–44.
40. Teasdale G, Jennett B. Assessment of coma and impaired consciousness:
a practical scale. Lancet 1974;2:81–4.
41. Teasdale G, Jennett B. Assessment and prognosis of coma after head injury.
Acta Neurochir 1976;34:45–55.
42. Gabbe BJ, Cameron PA, Finch CF. The status of the Glasgow coma scale.
Emerg Med (Fremantle) 2003;15(4):353–60. Review.
43. Marmarou A, Lu J, Butcher I, et al. Prognostic value of the Glasgow coma
scale and pupil reactivity in traumatic brain injury assessed pre-hospital and
on enrollment: an IMPACT analysis. J Neurotrauma 2007;24(2):270–80.
98 Section II—Clinical and Laboratory Assessment
44. Murray GD, Butcher I, McHugh GS, et al. Multivariable prognostic analysis
in traumatic brain injury: results from the IMPACT study. J Neurotrauma
2007;24(2):329–37.
45. Steyerberg EW, Mushkudiani N, Perel P, et al. Predicting outcome after
traumatic brain injury: development and international validation of
prognostic scores based on admission characteristics. PLoS Med 2008;5(8):
e165.
46. Lingsma HF, Roozenbeek B, Steyerberg EW, et al. Early prognosis in
traumatic brain injury: from prophecies to predictions. Lancet Neurol
2010;9(5):543–54.
47. Bates D. The prognosis of medical coma. J Neurol Neurosurg Psychiatry
2001;71(Suppl):i20–23.
48. Tuhrim S, Dambrosia JM, Price TR, et al. Prediction of intracerebral
hemorrhage survival. Ann Neurol 1988;24(2):258–63.
49. Tuhrim S, Dambrosia JM, Price TR, et al. Intracerebral hemorrhage: external
validation and extension of a model for prediction of 30-day survival. Ann
Neurol 1991;29(6):658-63.
50. Hirai S, Ono J, Yamaura A. Evaluation of consciousness by the Glasgow
Coma Scale in aneurysmal subarachnoid hemorrhage: appraisal using
logistic regression analysis. Cerebrovasc Dis 1997;99:S42–43.
A complete list of references for this chapter can be found online at
www.expertconsult.com.
 Section II—Clinical and Laboratory Assessment 98.e1
References
1. Von Economo C. Sleep as a problem of localization. J Nerv Ment Dis
1930;71:249–59.
2. Morruzi G, Magoun HW. Brain stem reticular formation and activation of
the EEG. J Neuropsychiatry Clin Neurosci 1949;7:251–67.
3. Parvizi J, Damasio AR. Neuroanatomical correlates of brainstem coma.
Brain 2003;126:1524–36.
4. Posner JB, Saper CB, Schiff ND, et al. Plum and Posner’s diagnosis of
stupor and coma. 4th ed. New York: Oxford University Press; 2007.
5. Blumenfeld H. Neuroanatomy through clinical cases. Sunderland, MA:
Sinauer Associates; 2010.
6. Hallanger AE, Levey AI, Lee HJ, et al. The origins of cholinergic and other
subcortical afferents to the thalamus in the rat. J Comp Neurol 1987;
262:105–24.
7. Saper CB, Scammell TE, Lu J. Hypothalamic regulation of sleep and
circadian rhythms. Nature 2005;437(7063):1257–63.
8. Saper CB, Fuller PM, Pedersen NP, et al. Sleep state switching. Neuron
2010;68(6):1023–42.
9. Saper CB. Organization of cerebral cortical afferent systems in the rat. II.
Magnocellular basal nucleus. J Comp Neurol 1984;222:313–42.
10. Saper CB. Diffuse cortical projection systems: anatomical organization and
role in cortical function. In: Plum F, editor. Handbook of physiology: the
nervous system. V. Bethesda, MD: American Physiological Society; 1987.
pp. 169–210.
11. Bassant MH, Ennouri K, Lamour Y. Effects of iontophoretically applied
monoamines on somatosensory cortical neurons of unanesthetized rats.
Neuroscience 1990;39:431–9.
12. Kolta A, Reader TA. Modulatory effects of catecholamines on neurons of
the rat visual cortex: single cell iontophoretic studies. Can J Physiol
Pharmacol 1989;67:615–23.
13. McCormick DA. Cholinergic and noradrenergic modulation of
thalamocortical processing. Trends Neurosci 1989;12(6):215–21.
14. McCormick DA, Bal T. Sleep and arousal: thalamocortical mechanisms.
Annu Rev Neurosci 1997;20:185–215.
15. Sato H, Fox K, Daw NW. Effect of electrical stimulation of locus coeruleus
on the activity of neurons in the cat visual cortex. J Neurophysiol 1989;62:
946–58.
16. Bromberg-Martin ES, Matsumoto M, Hikosaka O. Dopamine in
motivational control: rewarding, aversive, and alerting. Neuron 2010;68:
815–34.
17. Schultz W. Predictive reward signal of dopamine neurons. J Neurophysiol
1998;80(1):1–27.
18. Sawa A, Snyder SH. Schizophrenia: diverse approaches to a complex disease.
Science 2002;296(5568):692–5.
19. Marié RM, Defer GL. Working memory and dopamine: clinical and
experimental clues. Curr Opin Neurol 2003;S29–35.
20. Haenisch B, Bönisch H. Depression and antidepressants: insights from
knockout of dopamine, serotonin, or noradrenaline re-uptake transporters.
Pharmacol Ther 2011;129(3):352–68.
21. Ravindran LN, Stein MB. The pharmacologic treatment of anxiety
disorders: a review of progress. J Clin Psychiatry 2010;71(7):839–54.
22. Adamantidis AR, Zhang F, Aravanis AM, et al. Neural substrates of
awakening probed with optogenetic control of hypocretin neurons. Nature
2007;450(7168):720–4.
23. Willie JT, Chemelli RM, Sinton CM, et al. To eat or to sleep? Orexin in the
regulation of feeding and wakefulness. Annu Rev Neurosci 2001;24:429–58.
24. Schjolberg A, Sunnerhagen KS. Unlocking the locked-in: a need for team
approach for survivors of locked-in syndrome. Acta Neurol Scand 2011.
Epub ahead of print.
25. Gelenberg AJ. The catatonic syndrome. Lancet 1976;1:1339–41.
26. Cairns H. Head injuries in motor-cyclists: the importance of the crash
helmet. Br Med J 1941;2(4213):465–71.
27. Kumral E, Bayulkem G, Evyapan D, et al. Spectrum of anterior cerebral
artery territory infarction: clinical and MRI findings. Eur J Neurol
2002;9(6):615–24.
28. Fisher CM. Ocular bobbing. Arch Neurol 1964;11:543–6.
29. Rosenberg ML. Spontaneous vertical eye movements in coma. Ann Neurol
1986;20(5):635–7.
30. Brandt TH, Dieterich M. Different types of skew deviation. J Neurol
Neurosurg Psychiatry 1991;54(6):549–50.
31. Keane JR. Ocular skew deviation. Analysis of 100 cases. Arch Neurol
1975;32(3):185–90.
32. Schirmer CM, Hedges TR. Mechanisms of visual loss in papilledema.
Neurosurg Focus 2007;23(5):E5.
33. Simon RP. Neurogenic pulmonary edema. Neurol Clin 1993;11(2):309–23.
34. Tarulli AW, Lim C, Bui JD, et al. Central neurogenic hyperventilation: a case
report and discussion of pathophysiology. Arch Neurol 2005;62(10):1632–4.
35. Ramsay MAE, Savege TM, Simpson BRJ, et al. Controlled sedation with
alphaxalone-alphadolone. BMJ 1974;2:656–9.
36. Sessler CN, Gosnell MS, Grap MJ, et al. The Richmond agitation-sedation
scale: validity and reliability in adult intensive care unit patients. Am J
Respir Crit Care Med 2002;166:1338–44.
37. De Lemos J, Tweeddale M, Chittock D. Measuring quality of sedation in
adult mechanically ventilated critically ill patients: the Vancouver
interaction and calmness scale. J Clin Epidemiol 2000;53:908–19.
38. Fraser GL, Riker R. Monitoring sedation, agitation, analgesia, and delirium
in critically ill adult patients. Crit Care Clin 2001;17:1–21.
39. De Jong MMJ, Burns SB, Campbell ML, et al. Development of the
American Association of Critical-Care Nurses’ Sedation Assessment Scale
for Critically Ill Patients. Am J Crit Care 2005;14:531–44.
40. Teasdale G, Jennett B. Assessment of coma and impaired consciousness:
a practical scale. Lancet 1974;2:81–4.
41. Teasdale G, Jennett B. Assessment and prognosis of coma after head injury.
Acta Neurochir 1976;34:45–55.
42. Gabbe BJ, Cameron PA, Finch CF. The status of the Glasgow Coma Scale.
Emerg Med (Fremantle) 2003;15(4):353–60. Review.
43. Marmarou A, Lu J, Butcher I, et al. Prognostic value of the Glasgow Coma
Scale and pupil reactivity in traumatic brain injury assessed pre-hospital
and on enrollment: an IMPACT analysis. J Neurotrauma 2007;24(2):270–80.
44. Murray GD, Butcher I, McHugh GS, et al. Multivariable prognostic analysis
in traumatic brain injury: results from the IMPACT study. J Neurotrauma
2007;24(2):329–37.
45. Steyerberg EW, Mushkudiani N, Perel P, et al. Predicting outcome after
traumatic brain injury: development and international validation of
prognostic scores based on admission characteristics. PLoS Med 2008;5(8):
e165.
46. Lingsma HF, Roozenbeek B, Steyerberg EW, et al. Early prognosis in
traumatic brain injury: from prophecies to predictions. Lancet Neurol
2010;9(5):543–54.
47. Bates D. The prognosis of medical coma. J Neurol Neurosurg Psychiatry
2001;71(Suppl):i20–23.
48. Tuhrim S, Dambrosia JM, Price TR, et al. Prediction of intracerebral
hemorrhage survival. Ann Neurol 1988;24(2):258–63.
49. Tuhrim S, Dambrosia JM, Price TR, et al. Intracerebral hemorrhage:
external validation and extension of a model for prediction of 30-day
survival. Ann Neurol 1991;29(6):658-63.
50. Hirai S, Ono J, Yamaura A. Evaluation of consciousness by the Glasgow
Coma Scale in aneurysmal subarachnoid hemorrhage: appraisal using
logistic regression analysis. Cerebrovasc Dis 1997;99:S42–43.
51. Benesch CG, McDaniel KD, Cox C, et al. End-stage Alzheimer’s disease:
Glasgow Coma Scale and the neurologic examination. Arch Neurol 1993;
50:1309–15.
52. Chou CH, Lin GM, Ku CH, et al. Comparison of the APACHE II, GCS and
MRC scores in predicting outcomes in patients with tuberculous
meningitis. Int J Tuberc Lung Dis 2010;14:86–92.
53. Ishihara M, Kamei S, Taira N, et al. Hospital-based study of the prognostic
factors in adult patients with acute community-acquired bacterial
meningitis in Tokyo. Japan. Intern Med 2009;48(5):295–300.
54. Schut ES, Lucas MJ, Brouwer MC, et al. Cerebral infarction in adults with
bacterial meningitis. Neurocrit Care 2011;Oct. Epub ahead of print.
55. Knaus WA, Draper EA, Wagner DP, et al. APACHE II: a severity of disease
classification system. Crit Care Med 1985;13:818–29.
56. Le Gall JR, Lemeshow S, Saulnier F. A new simplified acute physiology score
(SAPS II) based on a European/North American multicenter study. JAMA
1993;270:2957–63.
57. Niskanen M, Kari A, Nikki P, et al. Acute physiology and chronic health
evaluation (APACHE II) and Glasgow Coma Scores as predictors of
outcome from intensive care after cardiac arrest. Crit Care Med 1991;
19(12):1465–73.
58. Bastos PG, Sun X, Wagner DP, et al. Glasgow Coma Scale score in the
evaluation of outcome in the intensive care unit: findings from the Acute
Physiology and Chronic Health Evaluation III study. Crit Care Med
1993;21(10):1459–65.
59. Buechler CM, Blostein PA, Koestner A, et al. Variation among trauma
centers’ calculation of Glasgow Coma Scale score: results of a national
survey. J Trauma 1998;45:429–32.
60. Healey C, Osler TM, Rogers FB, et al. Improving the Glasgow Coma Scale
score: motor score alone is a better predictor. J Trauma 2003;54(4):671–8.
61. Teasdale G, Knill-Jones R, van der Sande J. Observer variability in assessing
impaired consciousness and coma. J Neurol Neurosurg Psychiatry
1978;41:603–10.
98.e2 Section II—Clinical and Laboratory Assessment
62. Gill MR, Reiley DG, Green SM. Interrater reliability of Glasgow Coma Scale
scores in the emergency department. Ann Emerg Med 2004;43:215–23.
63. Gill M, Martens K, Lynch EL, et al. Interrater reliability of 3 simplified
neurologic scales applied to adults presenting to the emergency department
with altered levels of consciousness. Ann Emerg Med 2007;49(4):403–7.
64. Holdgate A, Ching N, Angonese L. Variability in agreement between
physicians and nurses when measuring the Glasgow Coma Scale in the
emergency department limits its clinical usefulness. Emerg Med Australas
2006;18(4):379–84.
65. Rowley G, Fielding K. Reliability and accuracy of the Glasgow Coma Scale
with experienced and inexperienced users. Lancet 1991;337:535–8.
66. Heim C, Schoettker P, Gilliard N, et al. Knowledge of Glasgow Coma Scale
by air-rescue physicians. Scand J Trauma Resusc Emerg Med 2009; 17:39.
67. Wilson JT, Pettigrew LE, Teasdale GM. Structured interviews for the
Glasgow Outcome Scale and the extended Glasgow Outcome Scale:
guidelines for their use. J Neurotrauma 1998;15:573–85.
68. Raimondi AJ, Hirschauer J. Head injury in the infant and toddler: coma
coring and outcome scale. Childs Brain 1984;12:1018–20.
69. Simpson DA, Cockington RA, Hanieh A, et al. Head injuries in infants and
young children: the value of the paediatric Coma Scale. Review of literature
and report on a study. Childs Nerv Syst 1991;7:183–90.
70. Kirkham FJ, Newton CR, Whitehouse W. Pediatric Coma Scales. Dev Med
Child Neurol. 2008;50(4):267–74.
71. James HE, Anas NG, Perkin RM, et al. The Glasgow Coma Scale. In: James
HE, Anas NG, Perkin RM, editors. Brain insults in infants and children.
Orlando, FL: Grune and Stratton; 1985. pp. 179–82.
72. Morray JP, Tyler DC, Jones TK, et al. Coma Scale for use in brain-injured
children. Crit Care Med 1984;12:1018–20.
73. Starmark JE, Stålhammar D, Holmgren E. The Reaction Level Scale
(RLS85): manual and guidelines. Acta Neurochir 1988a;91:12–20.
74. Johnstone AJ, Lohlun JC, Miller JD, et al. A comparison of the Glasgow
Coma Scale and the Swedish Reaction Level Scale. Brain Inj 1993;7:501–6.
75. Stålhammar D, Starmark JE, Holmgren E, et al. Assessment of
responsiveness in acute cerebral disorders: a multicenter study on the
reaction level scale (RLS 85). Acta Neurochir 1988;90:73–80.
76. Tesseris J, Pantazidis N, Routsi C, et al. A comparative study of the Reaction
Level Scale (RLS85) with Glasgow Coma Scale (GCS) and Edinburgh-2
Coma Scale (modified) (E2CS(M)). Acta Neurochir 1991;10:65–76.
77. Walther SM, Jonasson U, Gill H. Comparison of the Glasgow Coma Scale
and the reaction level scale for assessment of cerebral responsiveness in the
critically ill. Intensive Care Med 2003;29:933–8.
78. Starmark JE, Stålhammar D, Holmgren E, et al. A comparison of the
Glasgow Coma Scale and the Reaction Level Scale (RLS85). J Neurosurg
1988;69(5);699–706.
79. Benzer A, Mitterschiffthaler G, Marosi M, et al. Prediction of non-survival
after trauma: Innsbruck Coma Scale. Lancet 1991;338(8773):977–8.
80. Diringer MN, Edwards DF. Does modification of the Innsbruck and the
Glasgow Coma Scales improve their ability to predict functional outcome?
Arch Neurol 1997;54(5):606–11.
81. Wijdicks EF, Bamlet WR, Maramattom BV, et al. Validation of a new Coma
Scale: the FOUR score. Ann Neurol 2005;58:585–93.
82. Iyer VN, Mandrekar JN, Danielson RD, et al. Validity of the FOUR score
Coma Scale in the medical intensive care unit. Mayo Clin Proc 2009;84:
694–701.
83. Stead LG, Wijdicks EF, Bhagra A, et al. Validation of a new Coma Scale, the
FOUR score, in the emergency department. Neurocrit Care 2009;10:50–4.
84. Wolf CA, Wijdicks EF, Bamlet WR, et al. Further validation of the FOUR
score Coma Scale by intensive care nurses. Mayo Clin Proc 2007;82:435–8.
85. McNarry AF, Goldhill DR. Simple bedside assessment of level of
consciousness: comparison of two simple assessment scales with the
Glasgow Coma Scale. Anesthesia 2004;59:34–7.
86. Brott T, Adams HP Jr, Olinger CP, et al. Measurements of acute cerebral
infarction: a clinical examination scale. Stroke 1989;20(7):864–70.
87. Lyden PD, Lau GT. A critical appraisal of stroke evaluation and rating
scales. Stroke 1991;22(11):1345–52.
88. Lyden P, Brott T, Tilley B, et al. Improved reliability of the NIH stroke scale
using video training. Stroke 1994;25:2220–5.
89. Lyden P, Raman R, Liu L, et al. NIHSS training and certification using a
new digital video disk is reliable. Stroke 2005;36(11):2446–9.
90. Scandinavian Stroke Study Group. Multicenter trial of hemodilution in
ischemic stroke: background and study protocol. Stroke 1985;16:885–90.
91. Orgogozo JM, Asplundh K, Boysen G. A unified form for neurological
scoring of hemispheric stroke with motor impairment. Stroke 1992;23;
1678–9.
92. Hantsen L, De Weerdt W, DeKeyser J, et al. The European Stroke Scale.
Stroke 1994;25:2215–19.
93. Cote R, Battista RN, Wolfson C, et al. The Canadian Neurological Scale:
validation and reliability assessment. Neurology 1989;39:638–43.
94. Hwang BY, Appelboom G, Kellner CP, et al. Clinical grading scales in
intracerebral hemorrhage. Neurocrit Care 2010;13(1):141–51.
95. Hemphill JC 3rd, Bonovich DC, Besmertis L, et al. The ICH score: a simple,
reliable grading scale for intracerebral hemorrhage. Stroke 2001;32(4):
891–7.
96. Hunt WE, Hess RM. Surgical risk as related to time of intervention in the
repair of intracranial aneurysms. J Neurosurg 1968;28:14–20.
97. Hunt WE, Kosnik EJ. Timing and perioperative care in intracranial
aneurysm surgery. Clin Neurosurg 1974;21:9–89.
98. Chiang VL, Claus EB, Awad IA. Toward a more rational prediction of
outcome in patients with high-grade subarachnoid hemorrhage.
Neurosurgery 2000;46:28–35.
99. Oshiro EM, Walter KA, Piantadosi S, et al. A new subarachnoid hemorrhage
grading system based on the Glasgow Coma Scale: a comparison with the
Hunt and Hess and World Federation of Neurological Surgeons scales in a
clinical series. Neurosurgery 1997;41(1):140–7.
100. Salary M, Quigley MR, Wilberger JE Jr. Relation among aneurysm size,
amount of subarachnoid blood and clinical outcome. J Neurosurg 2007;
107(1):13–7.
101. Teasdale GM, Drake CG, Hunt W, et al. A universal subarachnoid
hemorrhage scale: report of a committee of the World Federation of
Neurosurgical Societies. J Neurol Neurosurg Psychiatry 1988;51(11):1457.
102. Claassen J, Vu A, Kreiter KT, et al. Effect of acute physiologic derangements
on outcome after subarachnoid hemorrhage. Crit Care Med 2004;32(3):
832–8.
103. Baker SP, O’Neill B, Haddon W Jr, et al. The Injury Severity Score: a
method for describing patients with multiple injuries and evaluating
emergency care. J Trauma 1974;14(3):187–96.
104. Boyd CR, Tolson MA, Copes WS: Evaluating trauma care: the TRISS
method. J Trauma 1987;27:370–8.
105. Champion HR, Sacco WH, Copes WS, et al. A revision of the trauma score.
J Trauma 1989;29:623–9.
106. Vincent JL, Moreno R, Takala J, et al. The SOFA (Sepsis-Related Organ
Failure assessment) score to describe organ dysfunction/failure. Working
Group on sepsis-related problems of the European Society of Intensive
Care Medicine. Intensive Care Med 1996;22(7):707–10.
107. Vincent JL, de Mendonca A, Cantraine F, et al. Use of the SOFA score to
assess the incidence of organ dysfunction/failure in intensive care units:
results of a multicenter, prospective study. Working group on sepsis-related
problems of the European Society of Intensive Care Medicine. Crit Care
Med 1998;26(11):1793–800.
108. Vincent J, Ferreira F, Moreno R. Scoring systems for assessing organ
dysfunction and survival. Crit Care Clin 2000;16:353–66.
109. Zygun D, Berthiaume L, Laupland K, et al. SOFA is superior to MOD score
for the determination of non-neurologic organ dysfunction in patients with
severe traumatic brain injury: a cohort study. Crit Care 2006;10(4):R115.
110. Marshall JC, Cook DJ, Christou NV, et al. Multiple Organ Dysfunction
Score: a reliable descriptor of a complex clinical outcome. Crit Care Med
1995;23(10):1638–52. Review.
111. Ciesla DJ, Moore EE, Johnson JL, et al. The role of the lung in postinjury
multiple organ failure. Surgery 2005;138(4):749–57, discussion 757–8.
112. Slater A, Shann F, Pearson G. Paediatric Index of Mortality (PIM) study
group: PIM2: a revised version of the Paediatric Index of Mortality.
Intensive Care Med 2003;29:278–85.
113. Tepas JJ,3rd, Ramenofsky ML, Mollitt DL, et al. The pediatric trauma score
as a predictor of injury severity: an objective assessment. J Trauma 1988;
28:425–9.
114. Bergstrom N, Braden BJ, Laguzza A, et al. The Braden scale for predicting
pressure sore risk. Nurs Res 1987;36(4):205–10.
115. National Pressure Ulcer Advisory Panel and European Pressure Ulcer
Advisory Panel. Prevention and treatment of pressure ulcers. Clinical
practice guideline. Washington DC: NPUAP; 2009.
116. Miranda DR, de Rijk A, Schaufeli W. Simplified therapeutic intervention
scoring system: the TISS-28 items—results from a multicenter study. Crit
Care Med 1996;24:64–73.
117. Maset AL, Marmarou A, Ward JD, et al. Pressure-volume index in head
injury. J Neurosurg 1987;67(6):832–40.
118. Weiner GM, Lacey MR, Mackenzie L, et al. Decompressive craniectomy for
elevated intracranial pressure and its effect on the cumulative ischemic
burden and therapeutic intensity levels after severe traumatic brain injury.
Neurosurgery 2010;66:1111–9.
119. Shore PM, Hand LL, Roy L, et al. Reliability and validity of the pediatric
intensity level of therapy (PILOT) scale: a measure of the use of
intracranial pressure-directed therapies. Crit Care Med 2006;34(7):1981–7.
 Section II—Clinical and Laboratory Assessment 98.e3
120. Jennett B, Bond M. Assessment of outcome after severe brain damage.
Lancet 1975;1:480–4.
121. Husson EC, Ribbers GM, Willemse-van Son AH, et al. Prognosis of
six-month functioning after moderate to severe traumatic brain injury: a
systematic review of prospective cohort studies. J Rehabil Med 2010;42:
425–36.
122. Lu J, Murray GD, Steyerberg EW, et al. Effects of Glasgow Outcome Scale
misclassification on traumatic brain injury clinical trials. J Neurotrauma
2008;25(6):641–51.
123. Menon DK, Zahed C. Prediction of outcome in severe traumatic brain
injury. Curr Opin Crit Care 2009;15:437–41.
124. Levy DE, Caronna JJ, Singer BH, et al. Predicting outcome from hypoxic-
ischemic coma. JAMA 1985;253:1420–6.
125. Bernard SA, Gray TW, Buist MD, et al. Treatment of comatose survivors of
out-of-hospital cardiac arrest with induced hypothermia. N Engl J Med
2002;346:557–63.
126. Hypothermia after cardiac arrest (HACA) study group. Mild therapeutic
hypothermia to improve the neurologic outcome after cardiac arrest.
N Engl J Med 2002;346:549–56.
127. Wijdicks EF, Hijdra A, Young GB, et al. Practice parameter: prediction
of outcome in comatose survivors after cardiopulmonary resuscitation
(an evidence-based review). Neurology 2006;67:203–10.
128. Rossetti AO, Oddo , Liaudet L, et al. Predictors of awakening from
postanoxic status epilepticus after therapeutic hypothermia. Neurology
2009;72:744–9.
129. Young GB. The EEG in coma. J Clin Neurophysiol 2000;17:473–85.
130. Chen R, Bolton CF, Young B. Prediction of outcome in patients with
postanoxic coma: a clinical and electrophysiological study. Crit Care Med
1996;24:672–8.
131. Gendo A, Kramer L, Häfner M, et al. Time-dependency of sensory evoked
potentials in comatose cardiac arrest survivors. Intensive Care Med 2001;
27(8):1305–11.
132. Young GB, Doig G, Ragazzoni A. Anoxic-ischemic encephalopathy: clinical
and electrophysiological associations with outcome. Neurocrit Care
2005;2:159–64.
133. Young GB. Clinical practice: neurologic prognosis after cardiac arrest.
N Engl J Med 2009;361(6):605–11.
134. Zandbergen EG, Hijdra A, Koelman JH, et al. Prediction of poor outcome
within the first 3 days of postanoxic coma. Neurology 2006;66(1):1133.
135. Fugate JE, Wijdicks EF, Mandrekar J, et al. Predictors of neurologic
outcome in hypothermia after cardiac arrest. Ann Neurol 2010;68(6):
907–14.
136. Galanaud D, Puybasset L. Cardiac arrest: has the time of MRI come? Crit
Care 2010;14(2):135–6.
137. Newcombe V, Chatfield D, Outtrim J, et al. Mapping traumatic axonal
injury using diffusion tensor imaging: correlations with functional
outcome. PLoS One 2011;6(5):e19214.
138. Perlbarg V, Puybasset L, Tollard E, et al. Relation between brain lesion
location and clinical outcome in patients with severe traumatic brain
injury: a diffusion tensor imaging study using voxel-based approaches.
Hum Brain Mapp 2009;30(12):3924–33.
139. Sidaros A, Engberg AW, Sidaros K, et al. Diffusion tensor imaging during
recovery from severe traumatic brain injury and relation to clinical
outcome: a longitudinal study. Brain 2008;131:559–72.
140. Tollard E, Galanaud D, Perlbarg V, et al. Experience of diffusion tensor
imaging and 1H spectroscopy for outcome prediction in severe traumatic
brain injury: preliminary results. Crit Care Med 2009;37(4):1448–55.
141. Tshibanda L, Vanhaudenhuyse A, Galanaud D, et al. Magnetic resonance
spectroscopy and diffusion tensor imaging in coma survivors: promises and
pitfalls. Prog Brain Res 2009;177:215–29.
142. Wu O, Sorensen AG, Benner T, et al. Comatose patients with cardiac arrest:
predicting clinical outcome with diffusion-weighted MR imaging.
Radiology 2009;252:173–81.
143. Wu O, Batista LM, Lima FO, et al. Predicting clinical outcome in comatose
cardiac arrest patients using early noncontrast computed tomography.
Stroke 2011;42:985–92.
144. Giacino JT, Ashwal S, Childs N, et al. The minimally conscious state:
definition and diagnostic criteria. Neurology 2002;58(3):349–53.
145. Giacino JT. The vegetative and minimally conscious states: consensus-based
criteria for establishing diagnosis and prognosis. NeuroRehabilitation 2004;
19:293–8.
146. Working Party of the Royal College of Physicians. The vegetative state:
guidance on diagnosis and management. Clin Med 2003;3(3):249–54.
147. Giacino JT, Kezmarsky MA, DeLuca J, et al. Monitoring rate of recovery to
predict outcome in minimally responsive patients. Arch Phys Med Rehabil
1991;72:897–901.
148. Giacino JT, Kalmar K, Whyte J. The JFK coma recovery scale-revised:
measurement characteristics and diagnostic utility. Arch Phys Med Rehabil
2004;85(12):2020–9.
149. Schnakers C, Majerus S, Giacino J, et al. A French validation study of the
Coma Recovery Scale-Revised (CRS-R). Brain Injury 2008;22(10):786–92.
150. Laureys S, Owen AM, Schiff ND. Brain function in coma, vegetative state,
and related disorders. Lancet Neurol 2004;3(9):537–46.
151. Owen AM, Coleman MR, Boly M, et al. Detecting awareness in the
vegetative state. Science 2006;313(5792):1402.
152. Monti MM, Vanhaudenhuyse A, Coleman MR, et al. Willful modulation of
brain activity in disorders of consciousness. N Engl J Med 2010;362(7):
579–89.
153. Raichle ME, MacLeod AM, Snyder AZ, et al. A default mode of brain
function. Proc Nat Acad Sci U S A 2001;98(2):676–82.
154. Vanhaudenhuyse A, Noirhomme Q, Tshibanda LJ, et al. Default network
connectivity reflects the level of consciousness in non-communicative
brain-damaged patients. Brain 2010;133:161–71.

Pain, Sedation, and Delirium
in Critical Illness
Kyla P. Terhune, E. Wesley Ely, and Pratik P. Pandharipande
Introduction
Patient comfort in an intensive care unit (ICU), from both a
mental and physical health perspective, is of utmost impor-
tance for patient safety and recovery. Proper management of
analgesia and sedation is especially important in patients
admitted to the neurocritical care unit (NCCU), because these
patients have significant comorbidities and often management
goals (e.g., sedation vs. need to evaluate consciousness or neu-
rologic function) are opposing. In addition, health care pro-
viders have to be cognizant of the anxiety experienced by
family members and understand how to alleviate it. Family
anxiety in the ICU is beyond the scope of this chapter and is
discussed elsewhere.1 This chapter provides an overview of
analgesia and sedation in the critically ill and how to monitor
its proper use. The reader also is referred to recent reviews and
guidelines on analgesia, sedation, and delirium in the ICU.2-12
Most research has been on patients in medical or surgical
ICUs with far less specific information available for patients
in the NCCU. The following topics are discussed:
1. Indications for the use of analgesia and sedation in the
ICU
2. Potential complications of analgesic and sedative medica-
tions, with a focus on the role of these medications in
delirium of critical illness
3. Strategies to optimize the delivery of analgesia and seda-
tion
4. Selection of analgesia and sedation for the critically ill
5. Management of delirium
Indications for Analgesia
and Sedation
Comfort and safety are the two main indications for analgesia
and sedation in the ICU, and both analgesia and sedation may
be administered for prolonged periods, particularly when a
patient is mechanically ventilated.2 In addition, in the NCCU
analgesia and sedation are used to treat intracranial pressure
(ICP) and can help improve brain oxygen.13
Pain or discomfort is treated by analgesics. The ICU patient
often has multiple sources of discomfort that arise from
© Copyright 2013 Elsevier Inc. All rights reserved.
Chapter
11  
II
primary disease, invasive monitoring, endotracheal intuba-
tion, phlebotomy, and procedures performed while in the
ICU. Inadequately treated discomfort and pain may cause an
increased stress response manifested by tachycardia, increased
oxygen consumption, hypercoagulability, immunosuppres-
sion, hypermetabolism, increased endogenous catecholamine
activity, and, in patients with acute brain injury, an increase
in ICP.2,14,15 Insufficient pain relief also can lead to sleep defi-
ciency, disorientation, and anxiety.2,14 Oxygen consumption
can be decreased by about 15% from baseline when adequate
analgesia is provided, thus reversing some of these unwanted
effects.14 In addition, a blood pressure increase that can con-
tribute to intracranial bleeding after a craniotomy can be
attenuated by effective pain control rather than antihyperten-
sives in some patients.
Critically ill patients often experience anxiety that can be
alleviated by judicious sedation. Apart from anxiolysis, seda-
tives also can produce amnesia, an effect that should be sought
only when neuromuscular blocking agents are adminis-
tered.2,15 There are also potentially unwanted mental health
effects from how sedation is used in the ICU. Unpleasant
memories of an ICU stay have been associated with post-
traumatic stress disorder (PTSD) symptoms in survivors.16,17
However, some studies suggest that the occurrence of PTSD
may be more specifically related to whether the memories
retained are delusional, and that factual memories, even if
unpleasant, are less likely to cause PTSD.18,19 In addition there
are some emerging concerns that some drugs themselves may
produce cognitive dysfunction post ICU.
The use of analgesia and sedation in the ICU can be
a double-edged sword. Both are inherently necessary but
both can have unwanted consequences. The administration
of analgesia and sedation therefore should account for
patient heterogeneity, including severity and prognosis, to
develop a tailored, optimized approach for each individual.2
There are many approaches to the same patient, and these
often are influenced by unexpected factors such as resources,
technology, and culture.2,20-23 Given the potential deleterious
effects of mismanagement, the Society of Critical Care
Medicine (SCCM)2 guidelines provide recommendations
for appropriate use of sedative and analgesic medications
and outline specific monitoring to be used when these
medications are administered to ICU patients. Adherence
99
100 Section II—Clinical and Laboratory Assessment
to these guidelines can help ensure that the medications
are used properly and side effects are minimized.
Complications of
Sedative Administration
Drug pharmacokinetics and pharmacodynamics can be
unpredictable in ICU patients because of their disease, drug
interactions, organ dysfunction, irregular absorption, and
variable protein binding.15 Each of these parameters can
contribute to complications, including need for mechanical
ventilation, infection, delirium, and chronic illness among
survivors.24 Structured multidisciplinary approaches, includ-
ing protocols and algorithms can help reduce these adverse
effects associated with sedation and analgesia in the ICU.
Sedation and analgesia algorithms should include identifica-
tion of goals and specific targets, use of valid and reliable tools
to assess analgesia, agitation, and sedation, and incorporation
of logical medication selection.25
Respiratory Depression
Overuse of opiates and sedatives in the conscious patient can
cause respiratory depression.15 It therefore is important to
individualize doses of opiates and sedatives to achieve levels
of analgesia and anxiolysis that are appropriate without
needing to establish an artificial airway.
Prolonged Mechanical Ventilation
Deep sedation, especially when administered through con-
tinuous infusions without daily interruption, can be associ-
ated with prolongation of mechanical ventilation and ICU
length of stay.26 The inability to perform adequate daily
patient assessments and neurologic examinations then may
result in increased need to evaluate the neurologic status
with imaging or electrophysiologic studies.27 This contributes
to increased costs of care and potential complications associ-
ated with transport of a critically ill patient.26,27 On the other
hand, inadequate sedation and analgesia in an intubated
patient can contribute to coughing or compromise ventila-
tion (i.e., patient optimization arises from conflicting thera-
peutic priorities).
Critical Illness Polyneuromyopathy
Patients who survive critical illness often have long-lasting
physical, neuropsychiatric, and quality of life impairments.28,29
Prolonged bed rest and stay in the ICU can be associated with
persistent functional debilitation and reduced quality of life,
even after ICU discharge.30-33 This functional debilitation
often manifests as a polyneuropathy or an associated myopa-
thy. Although the two conditions are physiologically distinct
from each another, they frequently coexist and are recognized
as critical illness polyneuropathy and myopathy (CIPNM),
critical illness neuromuscular abnormalities (CINMA), or
critical illness polyneuromyopathy (CIPM).34-36
CIPM manifests itself physically as a generalized, sym­
metrical muscle weakness that spares facial muscles. Although
some suggest that physical examination and history may be
sufficient for the diagnosis,37 nerve conduction studies,
electromyography, and possibly even muscle biopsy increase
the sensitivity to identify this entity.36,38 Additional potential
risk factors include steroid use, age, hyperglycemia, and the
presence of systemic inflammatory response syndrome
(SIRS).34,39 None are definitive, but the association with pro-
longed immobilization makes it reasonable that reduction of
sedation may improve the ability to mobilize critically ill
patients early, thus minimizing the occurrence or the severity
of CIPM.
Neuro-Specific Complications
Analgesics and sedatives often are administered in continuous
drips and have variable accumulation in critically ill patients;
this can predispose patients to experience a withdrawal syn-
drome on discontinuation.2,15,40,41 In addition, many psycho-
logical conditions have been studied in relation to prolonged
analgesia and sedation, particularly depression and PTSD,
with symptoms that appear directly related to the number of
days receiving such medications.42 There is also some sugges-
tion that not being able to recall specifics of their ICU stay
may have negative effects on patients’ long-term cognitive
abilities compared with those who have specific recall.43 ICU
survivors also may recall pain, anxiety, fear, and inability to
sleep during their ICU stay when questioned later.44 These are
important aspects for the intensivist to keep in mind because
they are effects that are unlikely to be manifested in the ICU
but have significant impact on the patient as an individual.
Delirium
Delirium is defined by the Diagnostic and Statistical Manual
of Mental Disorders (DSM IV)45 as a disturbance of con-
sciousness with inattention, accompanied by a change in
cognition or perceptual disturbance that develops over a
short period (hours to days) and fluctuates over time. Delir-
ium occurs in between 20% and 80% of mechanically venti-
lated medical, trauma, and surgical ICU patients.46-48
However, the majority of the cases are not recognized49-51;
this creates a delay in diagnosis, or the symptoms are attrib-
uted to another cause.
One approach to classify delirium is to group patients
according to their level of alertness (hyperactive, hypoactive,
or mixed).52 Peterson et al. found the distribution of the delir-
ium subtypes among ICU patients to be 1.6% hyperactive,
43.5% hypoactive, and 54.1% a mixed type.53 Hyperactive
delirium is assigned to patients with characteristics of agi­
tation, restlessness, attempts to remove catheters or tubes,
violent behaviors, and emotional lability and has an overall
better prognosis compared with hypoactive delirium, which is
manifested by flat affect, withdrawal, apathy, lethargy, and
decreased responsiveness.52,54-57 Given its subtle appearance in
a population that may be subject to polypharmacy, it is not
surprising that hypoactive delirium remains unrecognized in
66% to 74% of patients in a variety of settings.51,58-60 Use of a
validated monitoring system can help guard against the failure
to recognize this type of delirium.
Recognition of delirium in ICU patients is important
because its development is associated with longer time on
mechanical ventilation, ICU length of stay, higher costs,46,61,62
increased likelihood of discharge to a nursing home,45,63,64 an
increased in-hospital mortality,46,61,65 and long-term cognitive
 Section II—Clinical and Laboratory Assessment 101
impairment in survivors.66 Indeed, patients who develop delir-
ium during their ICU stay have a two- to threefold increase
risk of death after adjustment for covariates such as age,
presence of sepsis or acute respiratory distress syndrome
(ARDS) and Acute Physiology and Chronic Health Evaluation
(APACHE) II scores.46,61,65 In addition, a delay in initiating
treatment for delirium and a longer duration of delirium are
strong predictors of increased mortality in the ICU.67-69
The delirium research field continues to expand, and mul-
tiple potential mechanisms of delirium have been proposed
including imbalances in the synthesis, release, and inactivation
of neurotransmitters55,57; inflammatory abnormalities induced
by endotoxins and cytokines70-73; impaired oxidative metabo-
lism74; a relative cholinergic deficiency secondary to this
impaired oxidative metabolism75; and variations in levels of
neurotransmitter precursor amino acids.76,77
Who is at risk for developing delirium in the ICU popula-
tion? Baseline characteristics including genetic factors, demen-
tia, chronic illness, advanced age, and depression are associated
with an increased risk of developing delirium.78-81 Although
many of these risk factors are not modifiable, there are iatro-
genic factors to which clinicians can pay close attention to help
reduce the risk of delirium. These factors include hypoxia,
metabolic and electrolyte imbalances, substance withdrawal
syndromes, infection, dehydration, hyperthermia, vascular
disorders, and sleep deprivation. In addition neurologic
insults such as seizures, head trauma, and intracranial space–
occupying lesions have been suggested to have a role in
delirium.78-80
Benzodiazepines and opiates are implicated in the develop-
ment of delirium in non-ICU and in ICU patients (including
the postsurgical population).48,82-84 This often puts the clini-
cian in a paradoxical situation because these medications may
be necessary and are established as standard practice in clinical
practice guidelines by the SCCM2 to provide analgesia and
anxiolysis to patients who undergo procedures or are on
mechanical ventilation. Hence the need for sedation and anal-
gesia must be balanced against its associated complications.
Although benzodiazepines have been shown to be risk
factors for delirium, the data that implicate fentanyl and
morphine are less convincing.48,82,83,85 For example, Morrison
et al.85 in a prospective cohort study showed that inadequate
analgesia may predispose patients to delirium, with patients
who receive lower doses of parenteral morphine after hip frac-
ture to be at higher risk for developing delirium. This may also
depend in part on other medications (e.g., some anesthe­
siologists often comment that premedication with scopol-
amine makes patients who arouse in pain more likely to suffer
delirium).
Delirium Scales
Screening tools and guidelines allow the clinician to approach
the patient at risk for delirium in a systematic fashion. Two
scales (Fig. 11.1 and Table 11.1), the Confusion Assessment
Method for the ICU (CAM-ICU)47 and the Intensive Care
Delirium Screening Checklist (ICDSC)86 have been validated
to diagnose delirium in mechanically ventilated patients.
Delirium monitoring was included in the 2002 SCCM practice
guidelines for the sustained use of sedatives and analgesics in
the critically ill adult.2 The management guidelines for pain,
agitation, and sedation were under revision in 2012. A
Feature 1: Acute onset of mental status
changes or a fluctuating course
AND
Feature 2: Inattention
AND
Feature 3: Disorganized Feature 4: Altered level
thinking
OR
of consciousness
= DELIRIUM
Fig. 11.1  The confusion assessment method for the intensive
care unit (CAM-ICU). This delirium instrument can be used with a
sedation scale as a two-step approach to assess consciousness and
diagnose delirium. Patients are considered to have delirium if they
have Richmond Agitation-Sedation Scale (RASS) scores of –3 and
greater (responsive to verbal stimulus) and are CAM-ICU positive by
having features 1 and 2, and either 3 or 4 positive. (Adapted with
permission from Dr. E.W. Ely, www.icudelirium.org.)
separate diagnostic instrument to diagnose pediatric delirium
in critically ill children, the Pediatric Confusion Assessment
Method for Intensive Care Unit (pCAM-ICU), has been vali-
dated.87 Importantly, unlike bedside pain assessment tools that
have been validated in critically ill patients with neurologic
disorders, delirium and sedation tools have had less rigorous
research in NCCU patients.10
Sedation Scales
The diagnosis of delirium first requires the assessment of
arousal level, followed by an assessment for delirium. Exam-
ples of sedation scales include the Ramsay Sedation (RS)
scale,88 the Riker Sedation-Agitation Scale (SAS),89 and the
Richmond Agitation-Sedation Scale (RASS).90,91 Although any
of these can be used to assess agitation, this chapter focuses
on the RASS (Table 11.2) because it has excellent interrater
reliability and can be used to detect changes over consecutive
days.92 The RASS90,91 is a 10-point scale that ranges from +4 to
–5. A score of 0 signifies a calm and alert patient. Positive
RASS scores denote levels of aggressive behavior, and negative
RASS scores denote less responsiveness, and differentiate
between response to verbal (–1 to –3) and physical stimuli (–4
and –5). Patients with a RASS score of –3 and higher (lighter
levels of sedation) can further be assessed for delirium.
Confusion Assessment Method
for Intensive Care Unit
The CAM-ICU is a scale that assesses delirium, that can be
done in approximately 60 to 90 seconds.93 It has high specific-
ity and interrater reliability but may not always help detect
early delirium.94,95 In addition, the CAM-ICU allows for de-
lirium assessment in the face of sedative medication. This then
may be a potential limitation in the NCCU, where it is im­
portant to determine if there is neurologic deterioration as
well. The CAM-ICU assesses four features: acute change or
fluctuation in mental status (feature 1), inattention (feature
102 Section II—Clinical and Laboratory Assessment

Table 11.1  The Intensive Care Delirium Table 11.2  Richmond Agitation-Sedation
Screening Checklist (ICDSC) Scale (RASS)
Patient Evaluation Score Term Description
Altered level of +4 Combative Overtly combative, violent,
consciousness (A-E)* immediate danger to staff
Inattention Difficulty in following a +3 Very agitated Pulls or removes tube(s) or
conversation or instructions. Easily catheter(s); aggressive
distracted by external stimuli.
+2 Agitated Frequent nonpurposeful
Difficulty in shifting focuses. Any
movement, fights ventilator
of these scores 1 point.
+1 Restless Anxious but movements not
Disorientation Any obvious mistake in time, place,
aggressive or vigorous
or person scores 1 point.
0 Alert and calm
Hallucinations- The unequivocal clinical
delusion-psychosis manifestation of hallucination or −1 Drowsy* Not fully alert, but has sustained
of behavior probably due to awakening (eye opening/eye
hallucination or delusion. Gross contact) to voice (≥10 seconds)
impairment in reality testing. Any
of these scores 1 point. −2 Light Briefly awakens with eye
sedation* contact to voice (<10 seconds)
Psychomotor Hyperactivity requiring the use of
agitation or additional sedative drugs or −3 Moderate Movement or eye opening to
retardation restraints in order to control sedation† voice (but no eye contact)
potential danger to oneself or −4 Deep No response to voice, but
others. Hypoactivity or clinically sedation† movement or eye opening to
noticeable psychomotor slowing. physical stimulation
Inappropriate speech Inappropriate, disorganized, or −5 Unarousable No response to voice or physical
or mood incoherent speech. Inappropriate stimulation
display of emotion related to
events or situation. Any of these From Sessler CN, Gosnell MS, Grap MJ, et al. The Richmond Agitation-
scores 1 point. Sedation Scale: validity and reliability in adult intensive care unit patients.
Am J Respir Crit Care Med 2002;166:1338–44; Ely EW, Truman B, Shintani A,
Sleep-wake cycle Sleeping less than 4 hours or et al. Monitoring sedation status over time in ICU patients: reliability and
disturbance waking frequently at night (do validity of the Richmond Agitation-Sedation Scale (RASS). JAMA 2003;289:
not consider wakefulness initiated 2983–91.
by medical staff or loud *Verbal stimulation.
environment). Sleeping during †
Physical stimulation.
most of the day. Any of these
scores 1 point.
Symptom fluctuation Fluctuation of the manifestation of
any item or symptom over 24 Intensive Care Delirium Screening Checklist
hours scores 1 point.
The ICDSC has eight diagnostic features (see Table 11.1).86 To
Total score (0-8) Score of 4 or more is consistent be diagnosed with delirium, a patient should have four or
with diagnosis of delirium
more features from the checklist during the evaluation
Adapted from Bergeron N, Dubois MJ, Dumont M, et al. intensive care period.86 Patients who demonstrate some features from the
delirium screening checklist: evaluation of a new screening tool. Intensive
Care Med 2001;27:859–64.
ICDSC but do not meet all the criteria for delirium are con-
*Level of consciousness: sidered to have subsyndromal delirium.96 Work by Ouimet
A: No response, score: None. et al.96 has shown that subsyndromal delirium is part of the
B: Response to intense and repeated stimulation (loud voice and pain), score: spectrum of acute brain dysfunction between normal and
None. full-feature delirium and is associated with worse outcomes
C: Response to mild or moderate stimulation, score: 1.
D: Normal wakefulness, score: 0. than in patients with normal cognition, though better than in
E: Exaggerated response to normal stimulation, score: 1. those with delirium.

2), disorganized thinking (feature 3), or an altered level
of consciousness (feature 4) (see Fig. 11.1). The diagnosis
of delirium using the combination of the RASS and the
CAM-ICU requires:
1. RASS score of –3 or higher and
2. Feature 1 of CAM-ICU (acute change or fluctuation in
mental status) and
3. Feature 2 of CAM-ICU (inattention) and
4. One of the following:
 Feature 3 (disorganized thinking)
 Feature 4 (altered level of consciousness)
Strategies to Optimize the Delivery
of Analgesia and Sedation
Systematic ICU management protocols including rating scales
are important for analgesia, sedation, and delirium, and their
use likely improves outcome including in ventilated ICU
patients.25,97,98 Studies that evaluate the efficacy of sedation
protocols and target-based sedation all show benefits such as
shorter time on mechanical ventilation and less time in the
ICU and hospitals.99,100 However, subjective sedation scales
that depend on patient behavior and clinical examination may
be insensitive during deep sedation.101 This is evident from a
 Section II—Clinical and Laboratory Assessment 103
study that showed that almost 40% of patients admitted to a
medical ICU had episodes of burst suppression on their elec-
troencephalogram (EEG), despite being managed by the use
of targeted sedation with sedation scales. In addition, burst
suppression was associated with an increased in-hospital and
6-month mortality after adjusting for severity of illness and
presence of sepsis.102 The bispectral analysis of the EEG with
a Bispectral Index (BIS) monitor can be measured easily in the
ICU setting. A frontal temporal sensor is attached to the
patient; this derives a signal from several EEG components
that can be visualized but also numerically graded on a scale
from 100 (completely awake) to 0 (deep sedation).103 The
visualization of potential burst suppression allows the clini-
cian to recognize the presence of a drug-induced coma,104,105
and the graded scale allows the clinician to otherwise titrate
sedative medications to their desired effect. This is particularly
helpful in the case of the paralyzed patient, in whom the depth
of sedation and amnesia is difficult to determine. However,
even in the nonparalyzed patient, the BIS monitor may provide
further information to better titrate sedation.
The BIS and the SAS or RASS have been compared in
mechanically ventilated patients and appear comparable (i.e.,
BIS can differentiate inadequate from adequate sedation).106-108
Similarly Riker et al.109 studied 39 ICU patients sedated after
cardiac surgery and observed significant agreement between
BIS and SAS. In addition, the authors described significant
changes in the BIS values as patients became more arousable.
In trauma patients the BIS wave form appears to coincide with
propofol on/off during daily spontaneous awakening trials
(SATs).106 However, the role of the BIS to guide the titration
of sedation in ICU remains unclear, and in one study110 use of
a BIS monitor to support clinical sedation management deci-
sions in ventilated patients did not reduce the amount of
sedation used, the length of mechanical ventilation (MV), or
the length of ICU stay. In addition, the interpretation of BIS
findings in traumatic brain injury (TBI) patients can be sub-
jective.106 Further studies on the applicability of neurologic
monitoring beyond sedation scales are under way. However,
in the interim, sedation protocols and scales need to be imple-
mented to prevent unnecessary administration of sedatives
and analgesics to critically ill patients.
One method used in ICUs to assess sedation is the “daily
wake-up.”111 The role of daily interruptions of continuous
sedative infusions in critically ill patients has been evaluated
in several studies. Patients managed with sedation cessation
had a significant reduction in the duration of MV, shorter ICU
and hospital length of stay, and fewer neuroradiologic proce-
dures and tracheostomies. Importantly there was no increase
in long-term mental health sequelae associated with the daily
“wake-ups.”27,112 Girard et al. in a multicenter randomized
clinical trial, the Awakening Breathing Controlled trial, exam-
ined the efficacy and safety of a linked sedation and ventilator
weaning protocol for mechanically ventilated ICU patients.113
Patients were managed by a “targeted-sedation” strategy and
control group patients received daily spontaneous breathing
trials (SBTs). In the treatment group patients had a mandatory
SAT as step A (total cessation of sedation long enough to wake
to verbal stimulus or tolerate for 4 hours), followed by the SBT
as step B. The “wake-up and breathe” intervention was associ-
ated with a 3-day reduction of MV, 1 day less in coma, 4 days
less in the ICU and hospital, and a 14% absolute reduction in
the risk of death within 1 year (Fig. 11.2). Whether the daily
100
80
Patients alive (%)
SAT + SBT (n = 167)
60
40 Usual care + SBT (n = 168)
20
0
0 60 120 180 240 300 360
Days
Fig. 11.2  Mortality benefit in the Awakening Breathing Controlled
trial. This “wake up and breathe” intervention resulted in 14% absolute
reduction in the risk of death within 1 year (58% in control reduced to
44% in treatment group). SAT, Spontaneous awakening trial; SBT,
spontaneous breathing trial. (From Girard TD, Kress JP, Fuchs BD, et al. Efficacy
and safety of a paired sedation and ventilator weaning protocol for mechanically
ventilated patients in intensive care [Awakening and Breathing Controlled trial]: a
randomised controlled trial. Lancet 2008;371:126–34.)
wake-up is beneficial in the NCCU or the ideal method to
detect neurologic deterioration in a sedated patient is still to
be fully elucidated.
Selection of Analgesia and
Sedation for Critically Ill Patients
The selection of an opioid traditionally has depended on the
likely duration of analgesic infusion and the pharmacology of
the specific opioid.2 There is little evidence that one sedative
agent is better than another for control of ICP or cerebral
perfusion pressure in adults with severe brain injury.114 Mor-
phine and hydromorphone because of their longer duration
of action are used typically for cases of prolonged infusions.2
Hydromorphone has several advantages over morphine
including a lack of histamine release and lack of a clinically
active metabolite. This may result in less vasodilation and
hypotension and less prolonged sedation in patients with
renal insufficiency. Fentanyl’s rapid onset and short duration
of action make it better suited for shorter periods of infusion.
Its ease of titration and lack of histamine release also make it
better suited to the hemodynamically unstable patient.2
Remifentanil is one of the newest µ agonists with a rapid
onset and offset of action. Its lack of accumulation makes it a
useful drug for continuous infusion in the ICU, especially the
NCCU. Dahaba et al. conducted a randomized double blind
study on ventilated patients receiving remifentanil or mor-
phine. Those in the remifentanil group had more optimal
sedation and a shorter duration of mechanical ventilation.115
Muellejans et al. compared remifentanil with fentanyl in ICU
patients; the efficacy of sedation was similar but more propo-
fol was required in the fentanyl group. The time to extubation
was similar. However, the percentage of patients who experi-
enced pain after extubation was greater in the remifentanil
group, suggesting the need for proactive pain management
104 Section II—Clinical and Laboratory Assessment
when weaning remifentanil.40 This is important in patients
with neurologic compromise because increased pain and the
potential for withdrawal may contribute to hypertension and
tachycardia that may be detrimental to the patient. Other
studies suggest that fentanyl and remifentanil provide similar
amounts of analgesia.116
The safety and efficacy of analgesia-based sedation with
remifentanil have been compared with conventional sedation
with hypnotic-based regimens for patients with brain injury
who require prolonged sedation for MV. Neurologic assess-
ment times and time to extubation were shorter for patients
who received remifentanil than those who received propofol
or midazolam supplemented with morphine or fentanyl.117
Breen et al., in another randomized controlled trial that com-
pared remifentanil-based sedation with a midazolam-based
regimen, observed that the duration of mechanical ventilation
and duration of weaning were shorter in patients who received
remifentanil. There also was a trend toward shortened ICU
stay.118
There are several comparative trials between hypnotic sed-
ative regimens. Barr et al. used a pharmacologic model to
compare lorazepam and midazolam infusions for ICU seda-
tion. Lorazepam was found to have twice the sedative potency
and four times the amnestic potency of midazolam. The
emergence times for light and deep sedation were longer for
lorazepam than midazolam.119 In a prospective randomized
controlled study in trauma patients in which infusions of
lorazepam, midazolam, and propofol were compared, McCol-
lam et al. observed that oversedation occurred most fre-
quently with lorazepam, and the greatest number of dosage
adjustments was required by the lorazepam group. Cost was
greatest, and undersedation occurred most often with propo-
fol. The results suggest that midazolam is the most titratable
drug with the least amount of oversedation or undersedation
and that lorazepam was the most cost-effective agent for
sedation.120
Propofol has been compared to individual benzodiazepines
in several studies, including in randomized controlled trials
(RCTs). Carson et al.121 conducted a randomized trial that
compared intermittent lorazepam boluses to propofol infu-
sion with daily interruption of sedatives in both groups.
Patients in the propofol group had fewer mechanical ventila-
tion days, with a trend toward greater number of ventilator-
free survival days.121 In an economic evaluation of propofol
and lorazepam, overall propofol was less costly per patient
than lorazepam despite the considerably lower pharmacy unit
cost of lorazepam. The lower costs likely were associated with
the greater number of ventilator-free days in the propofol
group.122 Compared with midazolam infusion, patients sedated
by propofol infusion have faster and more reliable wake-up
times,123 have equal efficacy of achieving sedation goals,124
spend a larger percentage of time at their target sedation goal,
and have more rapid extubation, although this does not always
mean earlier ICU discharge.125 However, propofol appears to
require more frequent dose adjustments, has a lower nurse-
rated quality of sedation, and may cause more cardiovascular
depression and less amnesia than midazolam, with higher
costs.124 Average time to sedation and change in oxygen
consumption is similar in patients who receive propofol or
midazolam.123 Walder et al. performed a systematic review of
trials that compared sedation with propofol or midazolam in
2001. The duration of adequate sedation was greater with
propofol, independent of the length of sedation, and weaning
times were shorter with propofol, but this was only significant
in patients sedated for less than 36 hours.126 Together these
various data suggest that propofol is associated with better
outcomes when compared with benzodiazepines and appears
to be particularly useful in neurologic patients in whom eva­
luation of the neurologic status is important. In addition
when compared with morphine, use of propofol may reduce
the need for other interventions to control ICP in patients
with TBI.114
Dexmedetomidine, an alpha-2 agonist, provides an alterna-
tive to gamma-aminobutyric acid (GABA)–agonist sedative
medications. There are several potential advantages to its use:
“arousable sedation,” suppression of delirium, and preserved
ventilatory drive.127,128 In 2007, Pandharipande et al. compared
dexmedetomidine to lorazepam in mechanically ventilated
ICU patients in an RCT, and found that sedation with dexme-
detomidine resulted in more days alive without delirium or
coma, a lower prevalence of coma, greater achievement of
target sedation, and an important trend toward reduction in
mortality (27% in lorazepam and 17% in dexmedetomidine
group).129 Studies in ICU patients comparing dexmedetomi-
dine to midazolam130,131 have shown superior outcomes with
dexmedetomidine, including time on mechanical ventilation
and lower probability of delirium.130 Propofol and dexme-
detomidine use is associated with similar outcomes in criti-
cally ill patients.131 After coronary artery bypass surgery,
sedation with dexmedetomidine has similar times to weaning
and extubation as propofol but there is a significant reduction
in narcotic, β-blocker, antiemetic, nonsteroidal anti-
inflammatory drug, epinephrine, and diuretic use in patients
who receive dexmedetomidine.132 Observational data suggest
patients with neurologic disorders in the ICU may require
higher doses of dexmedetomidine to achieve desired sedation
levels.133 It does not seem that higher doses are associated with
more adverse side effects.134
Sedation paradigms have changed in the ICU; there is less
benzodiazepine exposure and unnecessary deep sedation.
However, with use has also come identification of disorders
such as propofol infusion syndrome,135 propylene glycol toxic-
ity from lorazepam (see Chapter 22), and lipid accumulation
with very prolonged emergence with high doses. In addition,
risks can be associated with off-label use of attractive sedatives
(e.g., unexpected deaths were observed when etomidate was
tried off label) and there have been some propofol deaths
related to infusion syndrome. Postmarketing discoveries of
new risks associated with off-label use of dexmedetomidine
and remifentanil may still arise before it can be concluded that
the change in sedation paradigm is an independent factor
associated with improved patient outcomes especially in the
NCCU.
Management of Delirium
Delirium is a symptom of an underlying disorder and it is
common in the ICU; its prevalence may be as high as 80%.
Development of delirium is associated with increased morbid-
ity and excess mortality.12,69,136 Ely et al. have suggested a prag-
matic approach to this: screening, prevention, and restoration
of brain function,137 or a bundle known as the ABCDE (for
awakening and breathing coordination, delirium monitoring,
and exercise/early mobility) approach to ventilated patients.9
 Section II—Clinical and Laboratory Assessment 105
It is important to recognize and proactively treat reversible
causes—hypoxia, hypercarbia, hypoglycemia, sleep depriva-
tion metabolic derangements and shock among others—that
might lead to delirium.138,139 High baseline inflammatory bio-
markers such as procalcitonin and C-reactive protein (CRP)
also may predict the likelihood of delirium in ventilated
patients.140 Intracranial hemorrhage and in particular intrace-
rebral hemorrhage may increase the risk of delirium in NCCU
patients.12
Just as the diagnosis and likely causes of delirium are
multifaceted, so too is prevention. The most important, and
certainly least expensive, interventions are noninvasive inter-
ventions that can be used by every member of the health
care team. For example, Inouye et al. trained volunteers and
nurses to focus on several key goals including provision of
regular stimulating activities, a nonpharmacologic sleep pro-
tocol, early mobilization, appropriate and early removal of
catheters and restraints, optimization of sensory input using
the patient’s own seeing and hearing aids, and attention to
hydration to prevent delirium in 852 hospitalized patients
greater than 70 years old. There was a 40% reduction in de-
lirium, but the interventions did not appear to show benefit
when results were again assessed at 6 months.141 However,
nonpharmacologic approaches are unlikely to be successful
once delirium is established.142 Other studies have shown a
decrease in the duration and severity of delirium but with
little effect on overall incidence,143,144 have shown benefit
only to subgroups of patients,145 or have not shown any
benefit at all.146 Although these studies were not specific to
the ICU setting, they provide some insight into how to alter
the course of delirium. Recently, Wang et al. have suggested
that low-dose intravenous (IV) prophylactic haloperidol may
prevent delirium in elderly patients admitted to the ICU
after noncardiac surgery.147
Because sedatives and analgesics play a role in the genesis
of delirium, there may be benefit to daily interruption of seda-
tives and analgesics, and awakening and breathing trials in a
protocol-driven fashion to decrease psychoactive medication
use.26,27,100,113,148 Although these strategies have been used in the
ICU, the studies have either not evaluated for delirium or
when delirium was evaluated,113 did not show any differences
in delirium rates or duration. However, in a subgroup of septic
patients, the wake-up-and-breathe strategy did reduce delir-
ium duration. Similarly, the Maximizing Efficacy of Targeted
Sedation and Reducing Neurological Dysfunction (MENDS)
trial showed that sedation with dexmedetomidine resulted in
more days alive without delirium or coma and a lower preva-
lence of coma.129
In delirious patients, particularly those with hyperactive
delirium, it may still be necessary to use medication to main-
tain a healthy sleep-wake cycle. This creates a paradox because
it is often difficult to discern whether the medications are the
root of the delirium. These medications also have the potential
to have psychoactive effects of their own, thus potentially pro-
longing delirium. The best approach is to use medications
sparingly and in the smallest dose possible to achieve the
desired effect. This seems self-explanatory, but it is often dis-
regarded with the use of “as needed” medications, which are
not regulated or regularly reassessed.
There is no specific U.S. Food and Drug Administration
(FDA)–approved medication for the treatment of delirium in
the ICU. Benzodiazepines are best avoided in the management
of delirium. However, proper diagnosis is critical because ben-
zodiazepines remain the drugs of choice for alcohol with-
drawal and seizures. Use of benzodiazepines for delirium may
lead to oversedation, exacerbation of delirium, and respiratory
suppression that could lead to hypoxia and hypercarbia and
thus start a vicious cycle. Elderly patients are particularly
susceptible to the effects of benzodiazepines. Haloperidol, a
butyrophenone “typical” antipsychotic, is a good choice in the
absence of risk factors such as prolonged QT syndrome
and is still the most widely used neuroleptic agent for delir-
ium.149 A variety of administration routes have been used—
intermittent intravenous IV, continuous IV infusion, oral or
enteral or intramuscular (IM) haloperidol.150 It does not sup-
press respiratory drive and works as a dopamine receptor
antagonist by blocking the D2 receptor. This treats positive
symptomatology (hallucinations, unstructured thoughts pat-
terns, etc.) and produces a variable sedative effect. The SCCM
guidelines recommend haloperidol to treat delirium, although
the class of evidence (level C data) supporting it is not par-
ticularly strong.2 There are few data from non-ICU studies
that support the use of haloperidol for delirium. For example,
Kalisvaart et al. in an RCT observed a reduction in the severity
of delirium but not the incidence of delirium in elderly
patients who had hip surgery.151 Milbrandt et al.152 also showed
that mechanically ventilated patients who received haloperi-
dol had reduced mortality after adjusting for covariates,
though patients were not evaluated for delirium.
There are newer “atypical” antipsychotic agents (e.g., ris-
peridone, ziprasidone, quetiapine, and olanzapine) that affect
neurotransmitters other than dopamine and may provide
benefit in delirium.153-157 Several studies have addressed their
use. Skrobik et al.156 compared olanzapine to haloperidol in
critically ill patients, and found that both agents had a
similar effect on the severity of delirium, though olanzapine
had fewer side effects. In cardiac surgery patients, Prakanrat-
tana et al.158 showed that a single dose of respiridone reduced
delirum. One case study of a patient with the CYP2D6 geno-
type showed that a switch from risperidone to quetiapine
cleared persistent delirium and extrapyramidal symptoms in
a timely fashion (i.e., genetic predisposition may also affect
responses to treatment with different medications).159
Further studies are needed and based on pilot studies are
feasible.160
Antipsychotics in general have several adverse effects,
including hypotension, dystonias, extrapyramidal effects,
malignant hyperthermia, and anticholinergic effects. However,
the most worrisome and life-threatening adverse effect is the
torsades de pointes. Antipsychotics therefore should not be
given to patients with prolonged QT intervals, and periodic
QT interval measurements should be obtained in those on
such medications. In addition, a recent meta-analysis of out-
patients who received antipsychotic medications for delirium
in the setting of dementia, shows that prolonged antipsychotic
treatment is associated with increased mortality and stroke
among elderly patients.161,162
Conclusion
Accumulating evidence shows that effective use of sedation
and analgesia protocols used in a target-based manner can
benefit patients admitted to the ICU. However, the most effec-
tive protocols to use in the NCCU still are being elucidated.
106 Section II—Clinical and Laboratory Assessment
In addition, analgesia and sedation can be a double-edged
sword because of inherent risks and because the medications
used may limit the ability to accurately assess consciousness
or neurologic function. Therefore patients who receive anal-
gesia and sedation in the ICU need to be carefully monitored
through validated scales such as the CAM-ICU, the ICDSC,
the RS scale, the RASS, and the Riker SAS, among others, to
ensure an optimal approach targeted to each individual.
References
1. Sauls JL, Warise LF. Interventions for anxiety in the critically ill: a guide for
nurses and families. Nurs Clin North Am 2010;45(4):555–67, vi.
2. Jacobi J, Fraser GL, Coursin DB, et al. Clinical practice guidelines for the
sustained use of sedatives and analgesics in the critically ill adult. Crit Care
Med 2002;30:119–41.
3. Erstad BL, Puntillo K, Gilbert HC, et al. Pain management principles in the
critically ill. Chest 2009;135(4):1075–86.
4. Chamorro C, Borrallo JM, Romera MA, et al. Anesthesia and analgesia
protocol during therapeutic hypothermia after cardiac arrest: a systematic
review. Anesth Analg 2010;110(5):1328–35.
5. Martin J, Heymann A, Bäsell K, et al. Evidence and consensus-based German
guidelines for the management of analgesia, sedation and delirium in
intensive care: short version. Ger Med Sci 2010;8:Doc 02.
6. Honiden S, Siegel MD. Analytic reviews: managing the agitated patient in
the ICU: sedation, analgesia, and neuromuscular blockade. J Intensive Care
Med 2010;25(4):187–204.
7. Wu CL, Raja SN. Treatment of acute postoperative pain. Lancet 2011;
377(9784):2215–25.
8. Patel SB, Kress JP. Sedation and analgesia in the mechanically ventilated
patient. Am J Respir Crit Care Med 2012;185(5):486–97. Epub 2011
Oct 20.
9. Morandi A, Brummel NE, Ely EW. Sedation, delirium and mechanical
ventilation: the ‘ABCDE’ approach. Curr Opin Crit Care 2011;17(1):43–9.
10. Teitelbaum JS, Ayoub O, Skrobik Y. A critical appraisal of sedation, analgesia
and delirium in neurocritical care. Can J Neurol Sci 2011;38(6):815–25.
11. Power I. An update on analgesics. Br J Anaesth 2011;107(1):19–24. Epub
2011, May 30.
12. Frontera JA. Delirium and sedation in the ICU. Neurocrit Care
2011;14(3):463–74.
13. Pascual JL, Georgoff P, Maloney-Wilensky E, et al. Reduced brain tissue
oxygen in traumatic brain injury: are most commonly used interventions
successful? J Trauma 2011;70(3):535–46.
14. Kress JP, Pohlman AS, Hall JB. Sedation and analgesia in the intensive care
unit. Am J Respir Crit Care Med 2002;166:1024–8.
15. Gehlbach BK, Kress JP. Sedation in the intensive care unit. Curr Opin Crit
Care 2002;8:290–8.
16. Weinert CR, Sprenkle M. Post-ICU consequences of patient wakefulness
and sedative exposure during mechanical ventilation. Intensive Care Med
2008;34(1):82–90. Epub 2007 Aug 17.
17. Rotondi AJ, Chelluri L, Sirio C, et al. Patients’ recollections of stressful
experiences while receiving prolonged mechanical ventilation in an intensive
care unit. Crit Care Med 2002;30:746–52.
18. Jones C, Griffiths RD, Humprhis G. Disturbed memory and amnesia related
to intensive care. Memory 2000;8:79–94.
19. Jones C, Griffiths RD, Humphris G, et al. Memory, delusions, and the
development of acute posttraumatic stress disorder-related symptoms after
intensive care. Crit Care Med 2001;29:573–80.
20. Weinert CR, Chlan L, Gross C. Sedating critically ill patients: factors
affecting nurses’ delivery of sedative therapy. Am J Crit Care 2001;10:156–65.
21. Weinert CR, Calvin AD. Epidemiology of sedation and sedation adequacy
for mechanically ventilated patients in a medical and surgical intensive care
unit. Crit Care Med 2007;35:393–401.
22. Payen JF, Chanques G, Mantz J, et al. Current practices in sedation and
analgesia for mechanically ventilated critically ill patients: a prospective
multicenter patient-based study. Anesthesiology 2007;106:687–95.
23. Mehta S, Burry L, Fischer S. Canadian survey of the use of sedatives,
analgesics, and neuromuscular blocking agents in critically ill patients. Crit
Care Med 2006;34:374–80.
24. Nseir S, Makris D, Mathieu D, et al. Intensive care unit-acquired infection as
a side effect of sedation. Crit Care 2010;14(2):R30. Epub 2010, Mar 15.
25. Sessler CN, Pedram S. Protocolized and target-based sedation and analgesia
in the ICU. Anesthesiol Clin 2011;29(4):625–50.
26. Kollef MH, Levy NT, Ahrens TS, et al. The use of continuous IV sedation is
associated with prolongation of mechanical ventilation. Chest 1998;114:
541–8.
27. Kress JP, Pohlman AS, O’Connor MF, et al. Daily interruption of sedative
infusions in critically ill patients undergoing mechanical ventilation. N Engl
J Med 2000;342:1471–7.
28. Jackson JC, Mitchell N, Hopkins RO. Cognitive functioning, mental health,
and quality of life in ICU survivors: an overview. Crit Care Clin 2009;25(3):
615–28, x.
29. Desai SV, Law TJ, Needham DM. Long-term complications of critical care.
Crit Care Med 2011;39(2):371–9.
30. Dowdy DW, Eid MP, Sedrakyan A, et al. Quality of life in adult survivors of
critical illness: a systematic review of the literature. Intensive Care Med 2005;
31:611–20.
31. Needham DM, Dowdy DW, Mendez-Tellez P, et al. Studying outcomes of
intensive care unit survivors: measuring exposures and outcomes. Intensive
Care Med 2005;31(9):1153–60.
32. Herridge MS, Cheung AM, Tansey CM, et al. One-year outcomes in survivors
of the acute respiratory distress syndrome. N Engl J Med 2003;348:683–93.
33. Herridge MS, Batt J, Hopkins RO. The pathophysiology of long-term
neuromuscular and cognitive outcomes following critical illness.
Crit Care Clin 2008;24:79–99, x.
34. de Letter MCJ, Schmitz PIM, Visser LH, et al. Risk factors for the
development of polyneuropathy and myopathy in critically ill patients.
Crit Care Med 2001;29:2281–6.
35. Stevens RD, Dowdy DW, Michaels RK, et al. Neuromuscular dysfunction
acquired in critical illness: a systematic review. Intensive Care Med 2007;33:
1876–91.
36. Bednarik J, Lukas Z, Vondracek P. Critical illness polyneuromyopathy: the
electrophysiological components of a complex entity. Intensive Care Med
2003;29:1505–14.
37. Morris C, Trinder JT. Electrophysiology adds little to clinical signs in critical
illness polyneuropathy and myopathy. Crit Care Med 2002;30:2612.
38. Berek K, Margreiter J, Willeit J, et al. Polyneuropathies in critically ill
patients: a prospective evaluation. Intensive Care Med 1996;22:849–55.
39. Weber-Carstens S, Deja M, Koch S, et al. Risk factors in critical illness
myopathy during the early course of critical illness: a prospective
observational study. Crit Care 2010;14(3):R119. Epub 2010, Jun 18.
40. Muellejans B, Lopez A, Cross MH, et al. Remifentanil versus fentanyl for
analgesia based sedation to provide patient comfort in the intensive care
unit: a randomized, double-blind controlled trial [ISRCTN43755713].
Crit Care 2004;8:R1–R11.
41. Vinik HR, Kissin I. Rapid development of tolerance to analgesia during
remifentanil infusion in humans. Anesth Analg 1998;86:1307–11.
42. Nelson BJ, Weinert CR, Bury CL, et al. Intensive care unit drug use and
subsequent quality of life in acute lung injury patients. Crit Care Med 2000;
28:3626–30.
43. Larson MJ, Weaver LK, Hopkins RO. Cognitive sequelae in acute respiratory
distress syndrome patients with and without recall of the intensive care unit.
J Int Neuropsychol Soc 2007;13:595–605.
44. Ethier C, Burry L, Martinez-Motta C, et al. Recall of intensive care unit stay
in patients managed with a sedation protocol or a sedation protocol with
daily sedative interruption: a pilot study. J Crit Care 2011;26(2):127–32.
45. American Psychiatric Association. Diagnostic and statistical manual of
mental disorders. Fourth edition, text revision. Washington, DC:
American Psychiatric Association; 2000.
46. Ouimet S, Kavanagh BP, Gottfried SB, et al. Incidence, risk factors and
consequences of ICU delirium. Intensive Care Med 2007;33:66–73.
47. Ely EW, Inouye SK, Bernard GR, et al. Delirium in mechanically ventilated
patients: validity and reliability of the confusion assessment method for the
intensive care unit (CAM-ICU). JAMA 2001;286:2703–10.
48. Pandharipande P, Cotton B, Shintani A, et al. Prevalence and risk factors for
development of delirium in surgical and trauma intensive care unit patients.
J Trauma 2008;65:34–41.
49. Ely EW, Siegel MD, Inouye SK. Delirium in the intensive care unit: an
under-recognized syndrome of organ dysfunction. Semin Respir Crit Care
Med 2001;22:115–26.
50. Francis J, Martin D, Kapoor WN. A prospective study of delirium in
hospitalized elderly. JAMA 1990;263:1097–101.
A complete list of references for this chapter can be found online at
www.expertconsult.com.
 Section II—Clinical and Laboratory Assessment 106.e1
References
1. Sauls JL, Warise LF. Interventions for anxiety in the critically ill: a guide for
nurses and families. Nurs Clin North Am 2010;45(4):555–67, vi.
2. Jacobi J, Fraser GL, Coursin DB, et al. Clinical practice guidelines for the
sustained use of sedatives and analgesics in the critically ill adult. Crit Care
Med 2002;30:119–41.
3. Erstad BL, Puntillo K, Gilbert HC, et al. Pain management principles in the
critically ill. Chest 2009;135(4):1075–86.
4. Chamorro C, Borrallo JM, Romera MA, et al. Anesthesia and analgesia
protocol during therapeutic hypothermia after cardiac arrest: a systematic
review. Anesth Analg 2010;110(5):1328–35.
5. Martin J, Heymann A, Bäsell K, et al. Evidence and consensus-based
German guidelines for the management of analgesia, sedation and delirium
in intensive care: short version. Ger Med Sci 2010;8:Doc 02.
6. Honiden S, Siegel MD. Analytic reviews: managing the agitated patient in
the ICU: sedation, analgesia, and neuromuscular blockade. J Intensive Care
Med 2010;25(4):187–204.
7. Wu CL, Raja SN. Treatment of acute postoperative pain. Lancet 2011;
377(9784):2215–25.
8. Patel SB, Kress JP. Sedation and analgesia in the mechanically ventilated
patient. Am J Respir Crit Care Med 2012;185(5):486–97. Epub 2011 Oct 20.
9. Morandi A, Brummel NE, Ely EW. Sedation, delirium and mechanical
ventilation: the ‘ABCDE’ approach. Curr Opin Crit Care 2011;17(1):43–9.
10. Teitelbaum JS, Ayoub O, Skrobik Y. A critical appraisal of sedation,
analgesia and delirium in neurocritical care. Can J Neurol Sci 2011;38(6):
815–25.
11. Power I. An update on analgesics. Br J Anaesth 2011;107(1):19–24. Epub
2011, May 30.
12. Frontera JA. Delirium and sedation in the ICU. Neurocrit Care 2011;14(3):
463–74.
13. Pascual JL, Georgoff P, Maloney-Wilensky E, et al. Reduced brain tissue
oxygen in traumatic brain injury: are most commonly used interventions
successful? J Trauma 2011;70(3):535–46.
14. Kress JP, Pohlman AS, Hall JB. Sedation and analgesia in the intensive care
unit. Am J Respir Crit Care Med 2002;166:1024–8.
15. Gehlbach BK, Kress JP. Sedation in the intensive care unit. Curr Opin Crit
Care 2002;8:290–8.
16. Weinert CR, Sprenkle M. Post-ICU consequences of patient wakefulness
and sedative exposure during mechanical ventilation. Intensive Care Med
2008;34(1):82–90. Epub 2007 Aug 17.
17. Rotondi AJ, Chelluri L, Sirio C, et al. Patients’ recollections of stressful
experiences while receiving prolonged mechanical ventilation in an
intensive care unit. Crit Care Med 2002;30:746–52.
18. Jones C, Griffiths RD, Humprhis G. Disturbed memory and amnesia
related to intensive care. Memory 2000;8:79–94.
19. Jones C, Griffiths RD, Humphris G, et al. Memory, delusions, and the
development of acute posttraumatic stress disorder-related symptoms after
intensive care. Crit Care Med 2001;29:573–80.
20. Weinert CR, Chlan L, Gross C. Sedating critically ill patients: factors
affecting nurses’ delivery of sedative therapy. Am J Crit Care 2001;10:156–65.
21. Weinert CR, Calvin AD. Epidemiology of sedation and sedation adequacy
for mechanically ventilated patients in a medical and surgical intensive care
unit. Crit Care Med 2007;35:393–401.
22. Payen JF, Chanques G, Mantz J, et al. Current practices in sedation and
analgesia for mechanically ventilated critically ill patients: a prospective
multicenter patient-based study. Anesthesiology 2007;106:687–95.
23. Mehta S, Burry L, Fischer S. Canadian survey of the use of sedatives,
analgesics, and neuromuscular blocking agents in critically ill patients. Crit
Care Med 2006;34:374–80.
24. Nseir S, Makris D, Mathieu D, et al. Intensive care unit-acquired infection
as a side effect of sedation. Crit Care 2010;14(2):R30. Epub 2010, Mar 15.
25. Sessler CN, Pedram S. Protocolized and target-based sedation and analgesia
in the ICU. Anesthesiol Clin 2011;29(4):625–50.
26. Kollef MH, Levy NT, Ahrens TS, et al. The use of continuous IV sedation is
associated with prolongation of mechanical ventilation. Chest 1998;114:
541–8.
27. Kress JP, Pohlman AS, O’Connor MF, et al. Daily interruption of sedative
infusions in critically ill patients undergoing mechanical ventilation. N Engl
J Med 2000;342:1471–7.
28. Jackson JC, Mitchell N, Hopkins RO. Cognitive functioning, mental health,
and quality of life in ICU survivors: an overview. Crit Care Clin 2009;25(3):
615–28, x.
29. Desai SV, Law TJ, Needham DM. Long-term complications of critical care.
Crit Care Med 2011;39(2):371–9.
30. Dowdy DW, Eid MP, Sedrakyan A, et al. Quality of life in adult survivors of
critical illness: a systematic review of the literature. Intensive Care Med
2005;31:611–20.
31. Needham DM, Dowdy DW, Mendez-Tellez P, et al. Studying outcomes of
intensive care unit survivors: measuring exposures and outcomes. Intensive
Care Med 2005;31(9):1153–60.
32. Herridge MS, Cheung AM, Tansey CM, et al. One-year outcomes in
survivors of the acute respiratory distress syndrome. N Engl J Med 2003;
348:683–93.
33. Herridge MS, Batt J, Hopkins RO. The pathophysiology of long-term
neuromuscular and cognitive outcomes following critical illness.
Crit Care Clin 2008;24:79–99, x.
34. de Letter MCJ, Schmitz PIM, Visser LH, et al. Risk factors for the
development of polyneuropathy and myopathy in critically ill patients.
Crit Care Med 2001;29:2281–6.
35. Stevens RD, Dowdy DW, Michaels RK, et al. Neuromuscular dysfunction
acquired in critical illness: a systematic review. Intensive Care Med 2007;
33:1876–91.
36. Bednarik J, Lukas Z, Vondracek P. Critical illness polyneuromyopathy: the
electrophysiological components of a complex entity. Intensive Care Med
2003;29:1505–14.
37. Morris C, Trinder JT. Electrophysiology adds little to clinical signs in
critical illness polyneuropathy and myopathy. Crit Care Med 2002;30:2612.
38. Berek K, Margreiter J, Willeit J, et al. Polyneuropathies in critically ill
patients: a prospective evaluation. Intensive Care Med 1996;22:849–55.
39. Weber-Carstens S, Deja M, Koch S, et al. Risk factors in critical illness
myopathy during the early course of critical illness: a prospective
observational study. Crit Care 2010;14(3):R119. Epub 2010, Jun 18.
40. Muellejans B, Lopez A, Cross MH, et al. Remifentanil versus fentanyl for
analgesia based sedation to provide patient comfort in the intensive care
unit: a randomized, double-blind controlled trial [ISRCTN43755713].
Crit Care 2004;8:R1–11.
41. Vinik HR, Kissin I. Rapid development of tolerance to analgesia during
remifentanil infusion in humans. Anesth Analg 1998;86:1307–11.
42. Nelson BJ, Weinert CR, Bury CL, et al. Intensive care unit drug use and
subsequent quality of life in acute lung injury patients. Crit Care Med 2000;
28:3626–30.
43. Larson MJ, Weaver LK, Hopkins RO. Cognitive sequelae in acute respiratory
distress syndrome patients with and without recall of the intensive care
unit. J Int Neuropsychol Soc 2007;13:595–605.
44. Ethier C, Burry L, Martinez-Motta C, et al. Recall of intensive care unit stay
in patients managed with a sedation protocol or a sedation protocol with
daily sedative interruption: a pilot study. J Crit Care 2011;26(2):127–32.
45. American Psychiatric Association. Diagnostic and statistical manual of
mental disorders. Fourth edition, text revision. Washington, DC: American
Psychiatric Association; 2000.
46. Ouimet S, Kavanagh BP, Gottfried SB, et al. Incidence, risk factors and
consequences of ICU delirium. Intensive Care Med 2007;33:66–73.
47. Ely EW, Inouye SK, Bernard GR, et al. Delirium in mechanically ventilated
patients: validity and reliability of the confusion assessment method for the
intensive care unit (CAM-ICU). JAMA 2001;286:2703–10.
48. Pandharipande P, Cotton B, Shintani A, et al. Prevalence and risk factors for
development of delirium in surgical and trauma intensive care unit
patients. J Trauma 2008;65:34–41.
49. Ely EW, Siegel MD, Inouye SK. Delirium in the intensive care unit: an
under-recognized syndrome of organ dysfunction. Semin Respir Crit Care
Med 2001;22:115–26.
50. Francis J, Martin D, Kapoor WN. A prospective study of delirium in
hospitalized elderly. JAMA 1990;263:1097–101.
51. Inouye SK. The dilemma of delirium: clinical and research controversies
regarding diagnosis and evaluation of delirium in hospitalized elderly
medical patients. Am J Med 1994;97:278–88.
52. Milisen K, Foreman MD, Godderis J, et al. Delirium in the hospitalized
elderly: nursing assessment and management. Nurs Clin North Am 1998;
33:417–36.
53. Peterson JF, Pun BT, Dittus RS, et al. Delirium and its motoric subtypes: a
study of 614 critically ill patients. J Am Geriatr Soc 2006;54:479–84.
54. O’Keeffe ST, Lavan JN. Clinical significance of delirium subtypes in older
people. Age Ageing 1999;28:115–9.
55. Meagher DJ, Trzepacz PT. Motoric subtypes of delirium. Semin Clin
Neuropsychiatry 2000;5:75–85.
56. Meagher DJ, Hanlon DO, Mahony EO, et al. Relationship between
symptoms and motoric subtype of delirium. J Neuropsychiatry Clin
Neurosci 2000;12:51–6.
57. Justic M. Does “ICU psychosis” really exist? Crit Care Nurse 2000;20:28–37.
106.e2 Section II—Clinical and Laboratory Assessment
58. Marcantonio ER, Goldman L, Mangione CM, et al. A clinical prediction
rule for delirium after elective noncardiac surgery. JAMA 1994;271:
134–9.
59. Sanders AB. Missed delirium in older emergency department patients:
a quality-of-care problem. Ann Emerg Med 2002;39:338–41.
60. Hustey FM, Meldon SW. The prevalence and documentation of impaired
mental status in elderly emergency department patients. Ann Emerg Med
2002;39:248–53.
61. Lin SM, Liu CY, Wang CH, et al. The impact of delirium on the survival of
mechanically ventilated patients. Crit Care Med 2004;32:2254–9.
62. Milbrandt EB, Deppen S, Harrison PL, et al. Costs associated with delirium
in mechanically ventilated patients. Crit Care Med 2004;32:955–62.
63. Inouye SK, Schlesinger MJ, Lydon TJ. Delirium: a symptom of how hospital
care is failing older persons and a window to improve quality of hospital
care. Am J Med 1999;106:565–73.
64. Levkoff SE, Evans DA, Liptzin B, et al. Delirium: the occurrence and
persistence of symptoms among elderly hospitalized patients. Arch Intern
Med 1992;152:334–40.
65. Ely EW, Shintani A, Truman B, et al. Delirium as a predictor of mortality in
mechanically ventilated patients in the intensive care unit. JAMA
2004;291:1753–62.
66. Girard TD, Jackson JC, Pandharipande PP, et al. Delirium as a predictor of
long-term cognitive impairment in survivors of critical illness. Crit Care
Med 2010;38(7):1513–20.
67. Pisani MA, Kong SY, Kasl SV, et al. Days of delirium are associated with
1-year mortality in an older intensive care unit population. Am J Respir
Crit Care Med 2009;180(11):1092–7. Epub 2009, Sep 10.
68. Heymann A, Radtke F, Schiemann A, et al. Delayed treatment of delirium
increases mortality rate in intensive care unit patients. J Int Med Res 2010;
38(5):1584–95.
69. Shehabi Y, Riker RR, Bokesch PM, et al. Delirium duration and mortality in
lightly sedated, mechanically ventilated intensive care patients. Crit Care
Med 2010;38(12):2311–8.
70. Fulkerson W, MacIntyre N, Stamler J, et al. Pathogenesis and treatment of
the adult respiratory distress syndrome. Arch Intern Med 1996;156:29–38.
71. Bellingan G. The pulmonary physician in critical care, 6: the pathogenesis
of ALI/ARDS. Thorax 2002;57:540–6.
72. Arvin B, Neville LF, Barone FC, et al. Brain injury and inflammation:
a putative role of TNF alpha. Ann N Y Acad Sci 1995;765:62–71.
73. Fink MP, Evans TW. Mechanisms of organ dysfunction in critical illness:
report from a Round Table Conference held in Brussels. Intensive Care Med
2002;28:369–75.
74. Lipowski ZJ. Delirium: acute confusional states, Rev. edition. New York:
Oxford University Press; 1990.
75. Blass JP, Gibson GE, Duffy TE, et al. Cholinergic dysfunction: a common
denominator in metabolic encephalopathies. In: Pepeu G, Landinsky H,
editors. Cholinergic Mechanisms. New York, NY: Plenum Press; 1981.
p. 921–8.
76. Van Der Mast RC. Pathophysiology of delirium. J Geriatr Psychiatry Neurol
1998;11:138–45.
77. Van Der Mast RC, van den Broek WW, Fekkes D, et al. Is delirium after
cardiac surgery related to plasma amino acids and physical condition?
J Neuropsychiatry Clin Neurosci 2000;12:57–63.
78. Webb JM, Carlton EF, Geeham DM. Delirium in the intensive care unit:
are we helping the patient? Crit Care Nurs Quat 2000;22:47–60.
79. Francis J. Drug-induced delirium: diagnosis and treatment. CNS Drugs
1996;5:103–14.
80. Inouye SK, Charpentier PA. Precipitating factors for delirium in
hospitalized elderly persons: predictive model and interrelationship with
baseline vulnerability. JAMA 1996;275:852–7.
81. McNicoll L, Pisani MA, Zhang Y, et al. Delirium in the intensive care unit:
occurrence and clinical course in older patients. J Am Geriatr Soc
2003;51:591–8.
82. Marcantonio ER, Juarez G, Goldman L, et al. The relationship of postoperative
delirium with psychoactive medications. JAMA 1994;272:1518–22.
83. Dubois MJ, Bergeron N, Dumont M, et al. Delirium in an intensive care
unit: a study of risk factors. Intensive Care Med 2001;27:1297–304.
84. Pandharipande P, Shintani A, Peterson J, et al. Lorazepam is an independent
risk factor for transitioning to delirium in intensive care unit patients.
Anesthesiology 2006;104:21–6.
85. Morrison RS, Magaziner J, Gilbert M, et al. Relationship between pain and
opioid analgesics on the development of delirium following hip fracture.
J Gerontol A Biol Sci Med Sci 2003;58:76–81.
86. Bergeron N, Dubois MJ, Dumont M, et al. intensive care delirium screening
checklist: evaluation of a new screening tool. Intensive Care Med 2001;27:
859–64.
87. Smith HA, Boyd J, Fuchs DC, et al. Diagnosing delirium in critically ill
children: validity and reliability of the Pediatric Confusion Assessment
Method for the Intensive Care Unit. Crit Care Med 2011;39(1):150–7.
88. Ramsay MA, Savege TM, Simpson BR, et al. Controlled sedation with
alphaxalone-alphadolone. BMJ 1974;2:656–9.
89. Riker RR, Picard JT, Fraser GL. Prospective evaluation of the Sedation-
Agitation Scale for adult critically ill patients. Crit Care Med 1999;27:
1325–9.
90. Ely EW, Truman B, Shintani A, et al. Monitoring sedation status over time
in ICU patients: reliability and validity of the Richmond Agitation-Sedation
Scale (RASS). JAMA 2003;289:2983–91.
91. Sessler CN, Gosnell MS, Grap MJ, et al. The Richmond Agitation-Sedation
Scale: validity and reliability in adult intensive care unit patients. Am J
Respir Crit Care Med 2002;166:1338–44.
92. Ely EW, Truman B, Shintani A, et al. Monitoring sedation status over time
in ICU patients: reliability and validity of the Richmond Agitation-Sedation
Scale (RASS). JAMA 2003;289(22):2983–91.
93. Pun BT, Gordon SM, Peterson JF, et al. Large-scale implementation of
sedation and delirium monitoring in the intensive care unit: a report from
two medical centers. Crit Care Med 2005;33:1199–205.
94. Guenther U, Popp J, Koecher L, et al. Validity and reliability of the
CAM-ICU Flowsheet to diagnose delirium in surgical ICU patients. J Crit
Care 2010;25(1):144–51. Epub 2009, Oct 13.
95. van Eijk MM, van den Boogaard M, van Marum RJ, et al. Routine
use of the confusion assessment method for the intensive care unit: a
multicenter study. Am J Respir Crit Care Med 2011;184(3):340–4. Epub
2011, May 11.
96. Ouimet S, Riker R, Bergeon N, et al. Subsyndromal delirium in the ICU:
evidence for a disease spectrum. Intensive Care Med 2007;33:1007–13.
97. Payen JF, Bosson JL, Chanques G, et al. Pain assessment is associated with
decreased duration of mechanical ventilation in the intensive care unit:
a post hoc analysis of the DOLOREA study. Anesthesiology 2009;111(6):
1308–16.
98. Skrobik Y, Ahern S, Leblanc M, et al. Protocolized intensive care unit
management of analgesia, sedation, and delirium improves analgesia and
subsyndromal delirium rates. Anesth Analg 2010;111(2):451–63. Epub
2010, Apr 7.
99. Brattebo G, Hofoss D, Flaatten H, et al. Effect of a scoring system and
protocol for sedation on duration of patients’ need for ventilator support in
a surgical intensive care unit. BMJ 2002;324:1386–9.
100. Brook AD, Ahrens TS, Schaiff R, et al. Effect of a nursing-implemented
sedation protocol on the duration of mechanical ventilation. Crit Care Med
1999;27:2609–15.
101. De Deyne C, Struys M, Decruyenaere J, et al. Use of continuous bispectral
EEG monitoring to assess depth of sedation in ICU patients. Intensive Care
Med 1998;24:1294–8.
102. Watson PL, Shintani A, Tyson R, et al. Presence of EEG burst suppression
in critically ill patients is associated with increased mortality. Crit Care Med
2008;36(12):3171–7.
103. Sigl JC, Chamoun NG. An introduction to bispectral analysis for the
electroencephalogram. J Clin Monit 1994;10:392–404.
104. Wolter S, Friedel C, Bohler K, et al. Presence of 14Hz spindle oscillations in
the human EEG during deep anesthesia. Clin Neurophysiol 2006;117:157–68.
105. Leistritz L, Jager H, Schelenz C, et al. New approaches for the detection and
analysis of electroencephalographic burst-suppression patterns in patients
under sedation. J Clin Monit Comput 1999;15:357–67.
106. Ogilvie MP, Pereira BM, Ryan ML, et al. Bispectral index to monitor
propofol sedation in trauma patients. J Trauma 2011;71(5):1415–21.
107. Karamchandani K, Rewari V, Trikha A, et al. Bispectral index correlates well
with Richmond agitation sedation scale in mechanically ventilated critically
ill patients. J Anesth 2010;24(3):394–8. Epub 2010, Mar 12.
108. Simmons LE, Riker R, Prato M, et al. Assessing sedation during intensive
care unit mechanical ventilation with the bispectral index and the
sedation-agitation scale. Crit Care Med 1999;27:1499–504.
109. Riker RR, Fraser GL, Simmons LE, et al. Validating the sedation-agitation
scale with the bispectral index and visual analog scale in adult ICU patients
after cardiac surgery. Intensive Care Med 2001;27:853–8.
110. Weatherburn C, Endacott R, Tynan P, et al. The impact of bispectral index
monitoring on sedation administration in mechanically ventilated patients.
Anaesth Intensive Care 2007;35:204–8.
111. Augustes R, Ho KM. Meta-analysis of randomised controlled trials on daily
sedation interruption for critically ill adult patients. Anaesth Intensive Care
2011;39(3):401–9.
112. Kress JP, Gehlbach B, Lacy M, et al. The long-term psychological effects of
daily sedative interruption on critically ill patients. Am J Respir Crit Care
Med 2003;168:1457–61.
 Section II—Clinical and Laboratory Assessment 106.e3
113. Girard TD, Kress JP, Fuchs BD, et al. Efficacy and safety of a paired sedation
and ventilator weaning protocol for mechanically ventilated patients in
intensive care (Awakening and Breathing Controlled trial): a randomised
controlled trial. Lancet 2008;371:126–34.
114. Roberts DJ, Hall RI, Kramer AH, et al. Sedation for critically ill adults with
severe traumatic brain injury: a systematic review of randomized controlled
trials. Crit Care Med 2011;39(12):2743–51.
115. Dahaba AA, Grabner T, Rehak PH, et al. Remifentanil versus morphine
analgesia and sedation for mechanically ventilated critically ill patients:
a randomized double blind study. Anesthesiology 2004;101:640–6.
116. Spies C, Macguill M, Heymann A, et al. A prospective, randomized,
double-blind, multicenter study comparing remifentanil with fentanyl in
mechanically ventilated patients. Intensive Care Med 2011;37(3):469–76.
Epub 2010, Dec 17.
117. Karabinis A, Mandragos K, Stergiopoulos S, et al. Safety and efficacy of
analgesia-based sedation with remifentanil versus standard hypnotic-based
regimens in intensive care unit patients with brain injuries: a randomised,
controlled trial [ISRCTN50308308]. Crit Care 2004;8:R268–80.
118. Breen D, Karabinis A, Malbrain M, et al. Decreased duration of mechanical
ventilation when comparing analgesia-based sedation using remifentanil
with standard hypnotic-based sedation for up to 10 days in intensive care
unit patients: a randomised trial [ISRCTN47583497]. Crit Care 2005;9:
R200–10.
119. Barr J, Zomorodi K, Bertaccini EJ, et al. A double-blind, randomized
comparison of IV lorazepam versus midazolam for sedation of ICU
patients via a pharmacologic model. Anesthesiology 2001;95:286–98.
120. McCollam JS, O’Neil MG, Norcross ED, et al. Continuous infusions of
lorazepam, midazolam, and propofol for sedation of the critically ill
surgery trauma patient: a prospective, randomized comparison. Crit Care
Med 1999;27:2454–8.
121. Carson SS, Kress JP, Rodgers JE, et al. A randomized trial of intermittent
lorazepam versus propofol with daily interruption in mechanically
ventilated patients. Crit Care Med 2006;34:1326–32.
122. Barrientos-Vega R, Mar Sanchez-Soria M, Morales-Garcia C, et al.
Prolonged sedation of critically ill patients with midazolam or propofol:
impact on weaning and costs. Crit Care Med 1997;25:33–40.
123. Kress JP, O’Connor MF, Pohlman AS, et al. Sedation of critically ill patients
during mechanical ventilation: a comparison of propofol and midazolam.
Am J Respir Crit Care Med 1996;153:1012–8.
124. Weinbroum AA, Halpern P, Rudick V, et al. Midazolam versus propofol for
long-term sedation in the ICU: a randomized prospective comparison.
Intensive Care Med 1997;23:1258–63.
125. Hall RI, Sandham D, Cardinal P, et al. Propofol vs midazolam for ICU
sedation: a Canadian multicenter randomized trial. Chest 2001;119:1151–9.
126. Walder B, Elia N, Henzi I, et al. A lack of evidence of superiority of
propofol versus midazolam for sedation in mechanically ventilated critically
ill patients: a qualitative and quantitative systematic review. Anesth Analg
2001;92:975–83.
127. Pichot C, Ghignone M, Quintin L. Dexmedetomidine and clonidine: from
second-to-first-line sedative agents in the critical care setting? J Intensive
Care Med 2011;Apr 27. Epub ahead of print.
128. Mantz J, Josserand J, Hamada S. Dexmedetomidine: new insights.
Eur J Anaesthesiol 2011;28(1):3–6.
129. Pandharipande PP, Pun BT, Herr DL, et al. Effect of sedation with
dexmedetomidine vs lorazepam on acute brain dysfunction in mechanically
ventilated patients: the MENDS randomized controlled trial. JAMA 2007;
298:2644–53.
130. Riker RR, Shehabi Y, Bokesch PM, et al. Dexmedetomidine vs midazolam
for sedation of critically ill patients: a randomized trial. JAMA
2009;301(5):489−99.
131. Jakob SM, Ruokonen E, Grounds RM, et al. Dexmedetomidine vs
midazolam or propofol for sedation during prolonged mechanical
ventilation: two randomized controlled trials. JAMA 2012;307(11):1151−60.
132. Herr DL, Sum-Ping ST, England M. ICU sedation after coronary artery
bypass graft surgery: dexmedetomidine-based versus propofol-based
sedation regimens. J Cardiothorac Vasc Anesth 2003;17:576–84.
133. Grof TM, Bledsoe KA. Evaluating the use of dexmedetomidine in
neurocritical care patients. Neurocrit Care 2010;12(3):356–61.
134. Jones GM, Murphy CV, Gerlach AT, et al. High-dose dexmedetomidine for
sedation in the intensive care unit: an evaluation of clinical efficacy and
safety. Ann Pharmacother 2011;45(6):740–7. Epub 2011, Jun 10.
135. Diedrich DA, Brown DR. Analytic reviews: propofol infusion syndrome in
the ICU. J Intensive Care Med 2011;26(2):59–72.
136. Salluh JI, Soares M, Teles JM, et al. Delirium epidemiology in critical care
(DECCA): an international study. Crit Care 2010;14(6):R210. Epub 2010,
Nov 23.
137. Vasilevskis EE, Pandharipande PP, Girard TD, et al. A screening, prevention,
and restoration model for saving the injured brain in intensive care unit
survivors. Crit Care Med. 2010;38(10 Suppl):S683–91.
138. Weinhouse GL, Schwab RJ, Watson PL, et al. Bench-to-bedside review:
delirium in ICU patients: importance of sleep deprivation. Crit Care
2009;13(6):234. Epub 2009, Dec 7.
139. Van Rompaey B, Elseviers MM, Schuurmans MJ, et al. Risk factors for
delirium in intensive care patients: a prospective cohort study. Crit Care
2009;13(3):R77. Epub 2009, May 20.
140. McGrane S, Girard TD, Thompson JL, et al. Procalcitonin and C-reactive
protein levels at admission as predictors of duration of acute brain
dysfunction in critically ill patients. Crit Care 2011;15(2):R78. Epub 2011,
Mar 2.
141. Inouye SK, Bogardus ST Jr, Charpentier PA, et al. A multicomponent
intervention to prevent delirium in hospitalized older patients. N Engl J
Med 1999;340:669–76.
142. Skrobik Y. Delirium prevention and treatment. Anesthesiol Clin
2011;29(4):721–7.
143. Milisen K, Foreman MD, Abraham IL, et al. A nurse-led interdisciplinary
intervention program for delirium in elderly hip-fracture patients. JAGS
2001;49:523–32.
144. Lundstrom M, Edlund A, Karlsson S, et al. A multifactorial intervention
program reduces the duration of delirium, length of hospitalization, and
mortality in delirious patients. J Am Geriatr Soc 2005;53:622–8.
145. Marcantonio ER, Flacker JM, Wright RJ, et al. Reducing delirium after hip
fracture: a randomized trial. J Am Geriatr Soc 2001;49:516–22.
146. Cole MG, McCusker J, Bellavance F, et al. Systematic detection and
multidisciplinary care of delirium in older medical inpatients: a
randomized trial. CMAJ 2002;167:753–9.
147. Wang W, Li HL, Wang DX, et al. Haloperidol prophylaxis decreases delirium
incidence in elderly patients after noncardiac surgery: a randomized
controlled trial. Crit Care Med 2011;Nov 3. Epub ahead of print.
148. Hooper MH, Girard TD. Sedation and weaning from mechanical
ventilation: linking spontaneous awakening trials and spontaneous
breathing trials to improve patient outcomes. Anesthesiol Clin
2011;29(4):651–61.
149. Ely EW, Stephens RK, Jackson JC, et al. Current opinions regarding the
importance, diagnosis, and management of delirium in the intensive care
unit: a survey of 912 healthcare professionals. Crit Care Med
2004;32:106–12.
150. Wang EH, Mabasa VH, Loh GW, et al. haloperidol dosing strategies in the
treatment of delirium in the critically ill. Neurocrit Care 2011;Oct 26. Epub
ahead of print.
151. Kalisvaart KJ, de Jonghe JF, Bogaards MJ, et al. Haloperidol prophylaxis for
elderly hip-surgery patients at risk for delirium: a randomized placebo-
controlled study. J Am Geriatr Soc 2005;53:1658–66.
152. Milbrandt EB, Kersten A, Kong L, et al. Haloperidol use is associated with
lower hospital mortality in mechanically ventilated patients. Crit Care Med
2005;33:226–9.
153. Tune LE, Damlouji NF, Holland A, et al. Association of postoperative
delirium with raised serum levels of anticholinergic drugs. Lancet
1981;2:651–3.
154. Tune LE, Egeli S. Acetylcholine and delirium. Dement Geristr Cogn Disord
2001;10:342–4.
155. Foreman M, Milisen K, Marcantonia EM. Prevention and treatment
strategies for delirium. Primary Psychiatry 2004;11:52–8.
156. Skrobik YK, Bergeron N, Dumont M, et al. Olanzapine vs haloperidol:
treating delirium in a critical care setting. Intensive Care Med
2004;30:444–9.
157. Alao AO, Soderberg M, Pohl EL, et al. Aripiprazole in the treatment of
delirium. Int J Psychiatry Med 2005;35:429–33.
158. Prakanrattana U, Prapaitrakool S. Efficacy of risperidone for prevention of
postoperative delirium in cardiac surgery. Anaesth.Intensive Care
2007;35:714–9.
159. Kato D, Kawanishi C, Kishida I, et al. Delirium resolving upon switching
from risperidone to quetiapine: implication of the CYP2D6 genotype.
Psychosomatics 2005;46:374–5.
160. Girard TD, Pandharipande PP, Carson SS, et al. Feasibility, efficacy, and
safety of antipsychotics for intensive care unit delirium: the MIND
randomized, placebo-controlled trial. Crit Care Med 2010;38(2):428–37.
161. Wang PS, Schneeweiss S, Avorn J, et al. Risk of death in elderly users of
conventional vs. atypical antipsychotic medications. N Engl J Med
2005;353:2335–41.
162. Schneider LS, Dagerman KS, Insel P. Risk of death with atypical
antipsychotic drug treatment for dementia: meta-analysis of randomized
placebo-controlled trials. JAMA 2005;294:1934–43.

Outcome Scales and
Neuropsychological Outcome
Rosette C. Biester
Introduction
The incidence of mortality and illness-related complications
has long been documented in the medical literature. More
recently, the measurement of outcomes has gained increasing
attention because advanced medical procedures, medications,
and therapies have led to better survival rates. One of the main
reasons to measure status at the time of injury or illness or
soon after is that long-term outcomes can be predicted from
these initial assessments. In addition, the efficacy of treatment
must be determined through objective measures rather than
simply using clinical judgment or behavioral observation. The
medical community has recognized the need to measure out-
comes in several areas, including medical and physical prog-
ress and functional or rehabilitation recovery. However,
inadequate or incomplete assessment occurs all too frequently
and does not provide enough data to reasonably predict long-
term outcome.
The term outcomes no longer refers to simply one or two
functional areas. To establish a complete picture of an indi-
vidual’s status or level of functioning, several outcomes
should be considered, and if possible, measured at different
time points. In this chapter functional outcomes, including
quality of life measures, and neuropsychologic or neurocog-
nitive outcomes, are reviewed and discussed. These measures
are of particular importance to the management of patients
with acute neurologic abnormalities as patients with no phys-
ical deficit can be profoundly disabled because of cognitive or
neurobehavioral deficits. It is now well documented that crit-
ical illness or a stay in an intensive care unit (ICU) can be
associated with long-term neurocognitive function.1 Addi-
tional outcomes, including mood and behavioral function-
ing, also are important in determining a patient’s complete
status, but they are not reviewed in this chapter. This chapter
does not present an exhaustive list of measures, but hopefully
will provide the reader with a representative sampling of
measures that could be useful in critical care and post–critical
care settings. The reader is referred to recent reviews on these
topics, recommendations from the traumatic brain injury
© Copyright 2013 Elsevier Inc. All rights reserved.
Chapter
12  
II
clinical trials network, and the Common Data Elements
Workgroups.2-10
Functional Outcome Scales
Functional outcome scales have been used at the time of
discharge from critical care units but more often in post–
critical care settings to supplement physical and medical
measurements of status, progress, and stability.11-17 These
scales include general measures of disability, more detailed
measures of motor skills and other physical abilities, and com-
prehensive assessments of physical, cognitive, and emotional
functioning.18-20 Outcome scales are typically used by raters to
determine levels of functioning through observation of behav-
ior or alternatively, caregiver report of patterns of behavior.
Factors that should be considered in deciding which scales to
use are:
1. Level of patient arousal and alertness: Patients who are
either in coma, have fluctuating consciousness, or are in a
minimally conscious state will not exhibit many behaviors
that can be rated on these measures. Instead, a coma-
emergent scale should be chosen as a more meaningful
alternative to standard outcome measures. A consistent
level of arousal and alertness is typically needed for mean-
ingful assessment on the measures that are reviewed in the
following text.
2. Environmental factors (e.g., limited available observation
time): Observation periods may be limited due to the
fragile medical status of the patient. This may dictate the
specific scale chosen for outcome ratings, or the frequency
with which ratings can be made.
3. Training of raters: Raters who have extensive training in
observation of critical care patients are much better suited
to using the more in-depth scales with finer gradations of
functioning. In contrast, raters with less training may be
limited to the more popular, easier-to-administer scales.
Two types of functional outcome scales are reviewed:
(1) brief functional outcome measures appropriate for ratings
107
108 Section II—Clinical and Laboratory Assessment
of critical care clients and (2) comprehensive tests of func-
tional outcome.
Brief Tests of Functional Outcome
The measures reviewed here are essentially global ratings com-
pleted by observers. Typically observed patients are either
critically ill or emerging from critical states. These patients are
usually incapable of either completing ratings themselves, or
participating in cognitive testing.
Glasgow Outcome Scale and Extended
Glasgow Outcome Scale
The Glasgow Outcome Scale (GOS) is one of the earliest
outcome measures developed. It has been used frequently in
outcome studies after acute neurologic disorders such as
traumatic brain injury (TBI) for which it was originally
described, as well as stroke and subarachnoid hemorrhage
(SAH).18,21 It has five levels of outcome, and is the basis for
other scales that have followed, including the Extended
Glasgow Outcome Scale (E-GOS) and the Disability Rating
Scale (DRS).22 These two scales are now more widely used
than the GOS, which is regarded as too broad in its catego-
ries and therefore not sensitive to changes in patient status.
In addition the GOS is not an adequate assessment measure
after mild brain injury such as concussion, for which there
are many specific concussion assessment tools: Axon Sports
Computerized Cognitive Assessment Tool (CCAT) or
CogState-Sport,23 Standardized Concussion Assessment Tool
2 (SCAT2),24 and Immediate Post-Concussion Assessment
and Cognitive Testing (ImPACT),25 among others. Research
in mild TBI and sports-related concussion suggests that no
one measure is ideal for obtaining a comprehensive picture
of the client’s status. Instead, a combination of measures or
specific neurocognitive tests assessing multiple domains is
more effective.
The E-GOS added three additional disability and outcome
categories to the original five, resulting in these eight catego-
ries: Dead, Vegetative State, Lower Severe Disability, Upper
Severe Disability, Lower Moderate Disability, Upper Moder-
ate Disability, Lower Good Recovery, and Upper Good
Recovery. Global outcome, rather than specific ability-related
outcomes, is the main outcome assessed by the E-GOS. A
well-designed study of 399 patients with intracerebral hem-
orrhage utilized the E-GOS together with the modified
Rankin Scale described in this chapter to determine global
outcome at 90 days.26 Although not as extensively cited as
the DRS, several studies have used the E-GOS to determine
functional outcome, including investigations several years
after the neurologic event. One study explored functional
outcome 10 years after TBI.27 Poorer outcome on the E-GOS
was associated with significantly longer posttraumatic
amnesia; more impairment on cognitive measures of infor-
mation processing speed, attention, memory, and executive
function; and higher anxiety levels. The E-GOS is used also
to measure functional outcome after SAH.28,29 Rating reli-
ability has been improved from the original GOS with the
addition of a structured interview. The E-GOS is popular in
neurosurgical-based outcome studies, but in rehabilitation
and neuropsychology studies it has not had the popularity
afforded the DRS.
Disability Rating Scale
DRS is a widely used measure consisting of eight items that
fall within four categories. Scores range from 0 (no disabil-
ity) to 29 (extreme vegetative state). It is brief but more
comprehensive than other short scales because it contains
items that measure all three categories listed by the World
Health Organization: Impairment, Disability, and Handi-
cap.30 The first three items were incorporated from the
Glasgow Coma Scale (GCS)31 with slight modifications: Eye
Opening, Communication Ability, and Motor Response.
These items are measures of impairment, and are very
important behaviors to observe and rate in critically ill or
emerging patients whose behavioral repertoire is severely
restricted. Measures of Feed­ing, Toileting, and Grooming
reflect disability, and the category of Employability is a
measure of handicap. An early study indicated that it was
more sensitive than the GCS in measuring clinical changes in
severe TBI patients.32 Similarly, it was found to be more sen-
sitive to detect improvement than the GOS for an inpatient
rehabilitation sample.33
The DRS is used commonly because it has been demon-
strated to be reliable and valid,34,35 is relatively brief and easy
to score, and can either be self-administered (though not a
common mode of administration) or scored through inter-
view with the client, family member, or other caretaker. One
limitation of this scale is that it does not usually detect subtle
changes that may occur in higher functioning individuals,
especially mild TBI survivors.
Modified Rankin Scale
The original Rankin Scale was developed in 1957 by J. Rankin.36
It generally is used to measure the degree of handicap in stroke
patients. This six-level version has been supplanted by the
modified Rankin Scale (mRS), which was published in 1988.37
It has an additional grade ranging from 0 (no symptoms) to
6 (dead). The grades in between all provide a degree of detail
that sets it apart from other less descriptive measures, and
hence increases its utility even though it is unidimensional.
For example, level three is: “Moderate disability. Requires
some help, but able to walk unassisted.” A study using neu-
rologists, nurses, and physiotherapists as raters found reliabil-
ity to be satisfactory.38 To improve interobserver reliability,
another study used a multimedia training process,39 another
feature that sets it apart from other equally brief scales. The
mRS (or dichotomized mRS) has been used extensively in
critical care studies to measure medical status and outcomes
at various time points after serious neurologic events includ-
ing intracerebral hemorrhage and acute ischemic stroke.40-48
Barthel Index
The Barthel Activities of Daily Living (ADL) Index is one of
the earliest measures that focuses specifically on activities of
daily living as well as mobility. It was originally introduced in
1965, and although the initial version is still widely used, a
modified Barthel Index with a 100-point assessment of inde-
pendence in 10 activities of daily living is even more compre-
hensive and allows for finer discriminations between ratings.49
The variables measured are feeding, bathing, grooming, dress-
ing, bowel, bladder, toilet use, transfers (bed to chair
and back), mobility, and stairs. The usefulness of the scale has
 Section II—Clinical and Laboratory Assessment 109
been demonstrated in several diagnostic groups, including
stroke,50,51 TBI,52,53 and even spinal cord injury.54 Obviously,
a scale such as the Barthel Index gives a more specific and
complete picture of the individual than global measures of
disability or handicap. Yet ratings of the 10 variables in this
index can only be made when the individual progresses to a
more advanced stage of recovery.
Ordinal Analysis and Sliding Dichotomy
In most phase III clinical trials ordinal outcome measures are
collapsed into a binary scale. For example, in TBI trials the
5-point GOS at 6 months often is used as the primary outcome
measure. For analysis it is dichotomized to unfavorable (Dead,
Vegetative, or Severe Disability) or favorable outcome (Mod-
erate Disability or Good Recovery). In stroke trials the mRS,
an ordinal scale (six categories), or Barthel Index also is col-
lapsed into binary scales. This collapse of an ordinal scale has
two potential problems: (1) loss of information55 and (2) pre-
cedence is given to a specific transition in the scale. For
example, for the GOS, this is the difference between moderate
and severe disability. The failure of many phase III TBI trials
over the past several decades has raised questions about con-
ventional statistical approaches.56 To address this use of ordinal
rather than dichotomous outcome, analysis is recommended.
Simulation studies suggest that this can increase statistical
power57 because all patients can contribute to the detection of
a treatment effect. The analysis is performed through a sliding
dichotomy in which the point of dichotomy is customized to
each patient’s baseline prognosis.58 Analysis of the Corticoste-
roid Randomization After Significant Head Injury (CRASH)
trial data demonstrates that ordinal analysis of the GOS can
improve statistical power to detect a treatment effect with
equal sample size.59 The Optimising Analysis of Stroke Trials
(OAST) collaboration suggests a benefit of ordinal analysis in
stroke research.60
Comprehensive Tests of Functional
Outcome, Quality of Life, or Health Status
Measures designed to assess health status after the acute stages
of critical care are typically more comprehensive in evaluating
areas of functioning. These measures can be administered as
early as several days after discharge from intensive care, and
indefinitely afterward. These questionnaires are completed
either by the patient or an observer or caretaker familiar with
the person’s functional status. These measures are different
from the brief tests covered in the previous section in which
an observer must rate the patient’s functions rather than the
patient participating actively in answering test items.
Short Form-36 and Short Form-12
The Short Form-36 Health Survey (SF-36) consists of 36 items
that measure health status, outcomes, and quality of life. It
measures 8 health concepts and yields 2 summary measures,
making it a relatively comprehensive instrument often selected
by clinicians and researchers alike to assess status and change.
The 8 health domains cover physical and emotional or psy-
chological status: (1) physical functioning, (2) role limitations
due to physical health problems, (3) bodily pain, (4) general
health, (5) vitality (energy or fatigue), (6) social functioning,
(7) role limitations due to emotional problems and (8) mental
health (psychological distress and psychological well-being).
Each of these domains is important to assess for patients
who have had neurointensive illnesses, injuries, or interven-
tions, because they are all basic areas of functioning that can
be negatively affected in seriously ill patients. The SF-36 is
widely cited in the health and medical literature for many dif-
ferent diagnostic groups including severe TBI, spinal cord
injury, or even elective surgery.61-63 It has even been discussed
as a quality-of-life measure.64 The SF-36 has been very useful
in determining progress over time with repeat assessments65
and in assessing functional outcome many years post injury.66,67
The Short Form-12 (SF-12) is a much shorter version of
the SF-36 that still measures the eight domains but contains
only one or two items from each of these. The SF-12 is gaining
increasing popularity and more recently has been used in place
of its longer counterpart.19,68,69 Several studies have compared
the SF-36 with the SF-12, and although some information
inevitably will be missing with any shortened form including
the SF-12, there appears to be enough evidence suggesting
that this shorter version can be a good substitute, especially
when assessment time is limited or a number of other mea-
sures are also employed in the study. In comparing the SF-12
with the SF-36 in a trauma population, one conclusion was
that the Physical Component Score in both was moderately
responsive to change and was equivalent using either the short
or longer measure.70 Other studies comparing these two mea-
sures in non-neurologic patients have also shown the useful-
ness of the SF-12 as an alternative to the longer version.71,72
Sickness Impact Profile
This health outcome measure is considerably longer than the
more widely used SF-36 but is more comprehensive and
detailed in assessing many important components of everyday
functioning. The 12 categories of the Sickness Impact Profile
(SIP) are (1) Sleep and Rest, (2) Emotional Behavior, (3) Body
Care and Movement, (4) Home Management, (5) Mobility,
(6) Social Interaction, (7) Ambulation, (8) Alertness Behavior,
(9) Communication, (10) Work, (11) Recreation and Pas-
times, and (12) Eating. It is relatively lengthy, and may actually
be the longest well-known outcomes measure at 136 items
that can be self- or interviewer administered. Scoring is done
by categories, broader dimensions, or the total SIP score.
In deciding whether to use the shorter SF-36 or the longer
SIP, the clinician or researcher must weigh the importance of
assessment time limitation versus the need for a very thorough
assessment of everyday functioning. For example, in a study
of the cognitive, mood, and quality-of-life impairments in
acute respiratory distress syndrome (ARDS) survivors, the
authors concluded that because ARDS patients had not
been well characterized in terms of the broad range of deficits
that persist, the most comprehensive quality-of-life measure
should be selected.14 Similarly, Frontera et al. who studied the
effect of delayed cerebral ischemia on quality of life 3 months
after SAH, used the entire SIP rather than a shorter measure.13
Others have used both the SIP, which some describe as a
measure of functional status, and the SF-36, considered to be
a quality of life measure, even though the SIP has been
described as a quality-of-life measure also.12 In another study
of long-term health status, the specific categories of Work,
Ambulation, Home Management, Recreation and Pastimes,
110 Section II—Clinical and Laboratory Assessment
and Alertness Behavior were most problematic at 12- to
18-month follow-up in survivors of major trauma.11 It is pos-
sible, however, to extract only those subscales of the SIP that
are needed for a particular clinical or research purpose. For
example, in a study of stroke survivors assessed at 6 months
post stroke, only 6 of the 12 categories were utilized.16 The
original SIP has sometimes been excluded from outcomes
research because of its length; hence, a shorter alternative
was developed. The SIP 68 was found to correlate well with
the SIP.73
Mayo Portland Adaptability Inventory-4
The Mayo Portland Adaptability Inventory-4 (MPAI-4) is the
newest version of the original Portland Adaptability Inventory
first published in 1987.74,75 This fifth version of the scale rates
the most common sequelae of acquired brain injury (ABI),
including physical, cognitive, emotional, behavioral, and social
problems.76 The areas of communication and employment
were further refined for this version. The MPAI-4 is a carefully
constructed inventory that provides clinical evaluations of
ABI survivors during the postacute stage following injury.
Patients who are suitable for evaluation on this scale range
from mild to severe brain injury. The MPAI-4 may be com-
pleted by persons with ABI, their significant others, and/or
medical or rehabilitation professionals. In one study, 134 indi-
viduals with ABI were rated by the persons with ABI, family
and significant others, and rehabilitation staff.77 Satisfactory
internal consistency for the MPAI-4 was found, regardless of
the source of ratings.
The MPAI-4 has three subscales: (1) Ability Index, (2) Ad-
justment Index, and (3) Participation Index. It has 29 clinical
items, as well as six additional items covering preinjury and
postinjury information about alcohol use, drug use, psychotic
symptoms, law violations, other conditions causing physical
impairment, and other conditions causing cognitive impair-
ment. Scoring consists of a total score and each of the three
index scores. Outcomes are described by five functional levels:
Good Outcome, Mild Limitations, Mild to Moderate Limita-
tions, Moderate to Severe Limitations, and Severe Limitations.
The manual, forms, and other language versions (including
French and German translations) can be easily downloaded
from the Center for Outcome Measurement in Brain Injury
(COMBI) website (http://www.tbims.org/combi), making it
an attractive and comprehensive functional battery.
Functional Independence Measure
The Functional Independence Measure (FIM) is used in inpa-
tient rehabilitation settings and even in residential brain injury
programs to assess disability rather than impairment.78 It is a
comprehensive measure of essential daily functions, and is
widely used across rehabilitation programs. The FIM is
reviewed in this chapter because it is frequently used with
post–critical care patients who have progressed from the ICU
to an acute rehabilitation setting and whose recovery is
advanced enough to warrant this level of detailed assessment.
This instrument measures the person’s performance along
several behavioral dimensions, from total dependence to inde-
pendence. The scale provides classification of individuals by
their ability to carry out an activity either with some level of
assistance or supervision, or independently. Each dimension is
rated on a 7-point scale: 7 is complete independence, 6 is modi-
fied independence, 5 equals supervision or setup, 4 equals
minimal contact assistance (client expends 75% or more of the
effort), 3 is moderate assistance (client expends between 50%
and 75% of the effort), 2 is maximal assistance (client expends
between 25% and 50% of the effort), and 1 equals total assis-
tance (client expends <25% of the effort).79 The FIM is typi-
cally completed by a group of rehabilitation specialists with
expertise in particular functional areas, often in weekly patient
care conferences. Eighteen areas are assessed: Eating; Groom-
ing; Bathing; Dressing—Upper Body; Dressing—Lower Body;
Toileting; Bladder Management—Level of Assistance; Bowel
Management—Level of Assistance; Transfers: Bed, Chair,
Wheelchair; Transfers: Toilet; Transfers: Tub or Shower; Loco-
motion: Walk/Wheelchair; Locomotion: Stairs; Comprehen-
sion; Expression; Social Interaction; Prob­lem Solving; and
Memory.
Neuropsychological Measures
Neuropsychological functioning typically refers to a broad
range of cognitive abilities. However, a comprehensive defini-
tion entails not only cognitive, but also motor, perceptual, and
emotional or psychological functions. After a neurologic
event, some or all of these domains may be negatively affected.
Patients who have been in ICUs are typically assessed after
hospital discharge, rather than in the ICU, because their
medical status is usually too compromised to proceed with
cognitive testing until after discharge.68,80 Tests that measure
these functions (abilities) vary in length and scope. They can
be classified as:
1. Neuropsychological screens
2. Neuropsychological batteries
3. Individual tests measuring specific cognitive domains
In selecting tests and deciding on how extensive testing should
be, the evaluator needs to be aware of guidelines in the field
and cognitive issues specific to the diagnostic group of the
individual being tested. This chapter reviews only neuropsy-
chological screens because they are more appropriate for use
in acute care patient assessment. Moreover, an entire chapter
on neuropsychological assessment alone would be needed to
do full justice to comprehensive batteries and to specific
domain-specific measures. Recent reviews, however, are avail-
able that discuss this.
Neuropsychological Screens
Neuropsychological screens are objective measures adminis-
tered directly to testees to determine cognitive status, and to
assess whether changes in functioning have occurred after
neurologic injury or illness. Unlike functional outcome scales
in which observers typically rate patient levels, neuropsycho-
logical tests are administered in face-to-face sessions to testees
who actively participate in answering specific test questions.
Several well-known screens are reviewed here, with limitations
listed for each measure.
Mini-Mental Status Examination
The Mini-Mental Status Examination (MMSE) is one of the
earliest screens used to measure cognitive functioning.81,82 It
 Section II—Clinical and Laboratory Assessment 111
is a brief but relatively comprehensive test of cognitive abili-
ties, and is directly administered to the patient. As such, the
patient must have a reasonable degree of orientation and lan-
guage ability to participate. Otherwise, a rating scale such as
one of those reviewed earlier in this chapter, must be used to
determine functional level. It is widely used to detect dementia
and even delirium, although there is caution in the literature
about using the MMSE alone to yield a diagnosis for either of
these conditions.83
The cognitive skills measured are (1) time and place orien-
tation, (2) registration (i.e., immediate auditory recall),
(3) attention and calculation, (4) recall, and (5) language
(receptive and expressive). A total of 30 points is the maximum
score, and cutoff levels of 23 and higher have been proposed
to differentiate dementia or cognitive impairment from
normal functioning. It has been the screen of choice among
health care professionals in medical settings, especially in
assessing hospitalized patients with known or suspected neu-
rologic disorders. It is among the most widely used tests of
adult mental status. A comprehensive review regarding the
psychometric properties of the MMSE was done.84 Subjects
ranging from cognitively intact community dwellers to indi-
viduals with dementia were included in the studies reviewed.
Satisfactory reliability and construct validity were found, and
the test was shown to be sensitive for moderate to severe cog-
nitive impairment but had lower sensitivity for mild degrees
of impairment. Content analyses revealed the test to be highly
verbal, and not all items were equally sensitive to cognitive
impairment. MMSE scores were affected by age, education,
and one’s cultural background, but not gender. Age and edu-
cational norms were included in one study, with suggestions
for using different cutoff levels.85
Limitations: (1) The MMSE covers several but not all neu-
ropsychological domains. Some areas that are missing include
visual perception, visual attention, and several subcompo-
nents of executive functioning, including verbal fluency and
mental flexibility. (2) The MMSE can easily miss neurologi-
cally based impairments in higher-functioning individuals.
Therefore false-negatives may result. (3) Only one test of
executive functioning or motor ability is included, namely, the
design construction. Verbal and other nonverbal executive
abilities are often disrupted with neurologic damage, and
assessment of these areas is advisable, even with a relatively
brief screen. (4) There is an overemphasis on language: Other
areas are disrupted as much as or even more than language
after a neurologic event, and a battery that measures other
areas more extensively would have had more utility. (5) Poor
sensitivity has been found for the detection of mild cognitive
impairment (MCI),84 which is a transition stage between
normal aging and the associated cognitive decreases that may
occur, and the more serious cognitive difficulties as a result of
Alzheimer’s disease.
Montreal Cognitive Assessment
An alternative to the MMSE is a more recently developed
cognitive test called the Montreal Cognitive Assessment
(MOCA), which takes about 10 minutes to administer and has
a total of 30 points.86 The cutoff for the MOCA is typically 26
out of 30 points, i.e., scores below 26 are considered impaired.
It is gaining popularity, and has been used as a screen for more
extensive neuropsychological testing.87 The MOCA has more
emphasis on attention and executive functions than the
MMSE, and contains a more difficult memory task, requiring
the recitation and recall of five instead of three words. In
comparing this measure to the MMSE, the MOCA has been
found to be significantly more sensitive. In the original study
highlighting the development of the MOCA as an alternative
to the MMSE, the MOCA had a sensitivity of 100% in detect-
ing mild Alzheimer’s disease, whereas the MMSE had a 78%
sensitivity.86 The difference in detecting Mild Cognitive
Impairment was more dramatic, with a sensitivity of 90% for
the MOCA but 18% for the MMSE. This is expected, because
one of the main reasons for the development of the MOCA
was to create a measure that could detect MCI much more
effectively than the MMSE.
Limitations: (1) The MOCA measures a fairly comprehen-
sive set of cognitive domains, especially for a brief screen, but
does not sample visual perception, an important function that
can be disrupted with TBI, brain tumor, stroke, or other neu-
rologic illness or injury. (2) Although both auditory and visual
attention skills are evaluated, only auditory memory is
assessed. It would be a more complete test if visual memory
also were measured, especially because other visual functions
have been included (visual attention, visual executive func-
tion, and visual motor construction).
Cognistat
The Cognistat is the short term for the Neurobehavioral Cog-
nitive Status Examination, a screening test that is more exten-
sive in scope than the MMSE or MOCA, and is used by
neuropsychologists as preliminary assessment of known or
suspected neurologic disorders. In fact, it is the number one
choice for cognitive screening among U.S. and Canadian neu-
ropsychologists.88 It is often used as a starting point, so that
additional neuropsychological measures can be selected to
further assess specific cognitive functions. It assesses a repre-
sentative sampling but not complete array of cognitive func-
tions, including level of consciousness, orientation, language,
memory, constructional ability, calculations, and reasoning.89
Language is tested through four subtests: Speech Sample, in
which the test taker verbally describes a detailed picture with
people and actions; Comprehension; Repetition; and Naming.
Reasoning has two subtests: Similarities and Judgment. The
Cognistat takes longer than the MMSE and MOCA screens,
typically 10 to 20 minutes to administer. It has a unique feature
that actually helps to quickly determine normal functioning;
Each test has a screen item that is more difficult than the test
items that follow, which are called the metric section. If an
individual successfully answers the difficult screen item, the
metric items under that domain are skipped. If only screen
items are administered, the test actually is fast and efficient,
and may take only 5 minutes to complete. Another useful
feature is that the scores are plotted on the profile sheet, so
that average and impaired scores are clearly illustrated. It has
been translated and is available in nine languages, including
Spanish, Japanese, and Hebrew.
Limitations: (1) Standardization data in the manual are
limited, because only 60 subjects are used, divided into two
age groups of 30 each (ages 20 to 30 and 40 to 66). For a
screen so extensively used and translated, a larger normative
group would have been expected. (2) As with other screens, a
select but not comprehensive battery of cognitive domains is
112 Section II—Clinical and Laboratory Assessment
provided. (3) There is a somewhat uneven emphasis on
certain but not other cognitive areas, particularly too many
tests of language and no tests of executive functions, an area
often affected by neurologic insult. (4) The utility of a sen-
tence repetition test in a general neuropsychological screen is
questionable; for example, aphasic patients would be better
tested using a language battery, such as the Boston Diagnostic
Aphasia Examination.90
Dementia Rating Scale-2
The original Dementia Rating Scale was developed in 1988,
and further refined into a second edition in 2001. The same
36 tasks and 32 stimulus cards have been retained for the
Dementia Rating Scale-2, together with the 5 subscales that
measure specific cognitive abilities vulnerable to deterioration
in aging dementia patients: Attention (8 items), Initiation/
Perseveration (11), Construction (6), Conceptualization (6),
and Memory (5).91 A total score and the 5 subscale scores are
generated, which assists the test evaluator to decide whether
additional testing is necessary. For example, a university pro-
fessor who scores considerably above the cutoff for his age
group but who has significant problems with Memory and
Conceptualization probably should be referred for more com-
prehensive testing. The upgraded aspects of the Dementia
Rating Scale-2 are expanded normative data, greater ease of
administration and scoring based on user-friendly test forms,
and additional validity information in the manual’s literature
review. The norms for the Dementia Rating Scale-2 are based
on a sample of healthy, nondementing adults, ages 56 to 105
from the Mayo Clinic’s impressive and comprehensive Mayo’s
Older Americans Normative Studies (MOANS) database,92
which also has been used for other testing instruments specific
to elderly populations.93 The authors of the Dementia Rating
Scale-2 caution the user to rely more on the subscale scores
than the total score to determine meaningful cognitive decline,
because an individual may score above the total score cutoff
for his or her age range but demonstrate significant cognitive
impairments in two or more subscale areas, for example. With
the original Dementia Rating Scale, there was a single cut-off
score of 123 out of a total of 144, and the common practice
was to rate an individual as having a probable dementia dis-
order if he/she scored below this total. However, researchers
in geriatric assessment have questioned this practice. One
study argued that different cutoff scores should be used for
different age groups.94 The Mayo Clinic research group found
that 16% of adults older than 85 years of age in the MOANS
normative data scored below the cutoff of 123 even though
they were not considered a dementia sample.92
The Dementia Rating Scale-2 also can be used with geriatric
patients who have suspected neuropsychological impairments
that are not necessarily dementia, including head injury, or are
community dwellers with cognitive problems who have not
yet been diagnosed with dementia.95 Like the Cognistat, test
items can be eliminated based on successful performance of
previous items, making it a time-efficient test to administer,
especially for the geriatric population who may have limited
tolerance for testing. For example, a client may have no dif-
ficulty in Initiation/Perseveration, but problems with Atten-
tion and Memory, requiring all of these latter two subscale
items to be administered but only a limited number of items
for the former.
The MOANS database has one limitation: The database
consists of predominantly Caucasian adults with an education
level higher than the average (M = 13.1 years).92 This implies
that scores for non-Caucasians and for persons with less than
8 years of education must be interpreted with caution, espe-
cially if they fall in the dementia range.
Conclusion
Outcome in critical care often has been assessed by mortality.
However, the efficacy of an ICU treatment can no longer be
judged simply by survival. After patients are discharged from
a neurocritical care unit, functional outcome is important to
determine success of treatment and rate of recovery. Func-
tional outcome includes many measures such as (1) physical
abilities, (2) quality of life, (3) activities of daily living, (4)
return to work, (5) cognitive function, (6) emotional or psy-
chological status, and (7) social integration to name a few.
Many different methods assess functional outcome, some of
which are described in this chapter. These may be assessed by
direct interview by a rater, by telephone, completion of a
questionnaire (self-report) using paper and pencil or online
computerized systems, by caregiver reports, or through
observation. What is clear is that no single ideal way to assess
outcome and variables such as age, sex, education, ethnicity,
response bias, test learning, and even the test itself can affect
outcome. Outcome measures therefore should be carefully
selected based on the disease process and expectations.
Neuropsychological testing is a necessary and very effective
method to assess cognition and functioning following the
initial use of outcome scales.
References
1. Hopkins RO, Jackson JC. Long-term neurocognitive function after critical
illness. Chest 2006;130(3):869–78.
2. Wilde EA, Whiteneck GG, Bogner J, et al. Recommendations for the use of
common outcome measures in traumatic brain injury research. Arch Phys
Med Rehabil 2010;91(11):1650–60.e17.
3. Bagiella E, Novack TA, Ansel B, et al. Measuring outcome in traumatic
brain injury treatment trials: recommendations from the traumatic
brain injury clinical trials network. J Head Trauma Rehabil 2010;25(5):
375–82.
4. McCauley SR, Wilde EA, Anderson VA, et al. Recommendations for the use
of common outcome measures in pediatric traumatic brain injury research.
J Neurotrauma 2011; Aug 24. Epub ahead of print.
5. Fields JA, Ferman TJ, Boeve BF, et al. Neuropsychological assessment of
patients with dementing illness. Nat Rev Neurol 2011; Nov 1. Epub ahead of
print.
6. Lewis MW, Babbage DR, Leathem JM. Assessing executive performance
during cognitive rehabilitation. Neuropsychol Rehabil 2011;21(2):145–63.
Epub 2011 Jan 10.
7. American Congress of Rehabilitation Medicine, Brain Injury-
Interdisciplinary Special Interest Group, Disorders of Consciousness Task
Force, et al. Assessment scales for disorders of consciousness: evidence-based
recommendations for clinical practice and research. Arch Phys Med Rehabil
2010;91(12):1795–813.
8. Podell K, Gifford K, Bougakov D, et al. Neuropsychological assessment in
traumatic brain injury. Psychiatr Clin North Am 2010;33(4):855–76.
9. Bilder RM. Neuropsychology 3.0: evidence-based science and practice. J Int
Neuropsychol Soc 2011;17(1):7–13. Epub 2010 Nov 19.
10. Gottesman RF, Hillis AE. Predictors and assessment of cognitive dysfunction
resulting from ischaemic stroke. Lancet Neurol 2010;9(9):895–905.
11. Holtslag HR, Post MW, Lindeman E, et al. Long-term functional health
status of severely injured patients. Injury 2007;38(3):280–9.
12. Van der Schaaf M, Beelen A, Dongelmans DA, et al. Functional status after
intensive care: a challenge for rehabilitation professionals to improve
outcome. J Rehabil Med 2009;41(5):360–6.
 Section II—Clinical and Laboratory Assessment 113
13. Frontera JA, Fernandez A, Schmidt JM, et al. Defining vasospasm after
subarachnoid hemorrhage: what is the most clinically relevant definition?
Stroke 2009;40(6):1963–8.
14. Mikkelsen ME, Shull WH, Biester RC, et al. Cognitive, mood and quality of
life impairments in a select population of ARDS survivors. Respirology 2009;
14(1):76–82.
15. Van Baalen B, Odding E, van Woensel MP, et al. Reliability and sensitivity to
change of measurement instruments used in a traumatic brain injury
population. Clin Rehabil 2006;20(8):686–700.
16. de Haan R, Limburg M, Bossuyt P, et al. The clinical meaning of Rankin
“handicap” grades after stroke. Stroke 1995;26:2027–30.
17. Lipsett PA, Swoboda SM, Dickerson J, et al. Survival and functional outcome
after prolonged intensive care unit stay. Ann Surg 2000;231(2):262–8.
18. Boake C, High WM. Functional outcome from traumatic brain injury. Am J
Phys Med Rehabil 1996;75(2):105–13.
19. von Elm E, Osterwalder JJ, Graber C, et al. Severe traumatic brain injury in
Switzerland—feasibility and first results of a cohort study. Swiss Med Wkly
2008;138(23-24):327–34.
20. Lippert-Gruner M, Maegele M, Haverkamp H, et al. Health-related quality
of life during the first year after severe brain trauma with and without
polytrauma. Brain Inj 2007;21(5):451–5.
21. Broessner G, Helbok R, Lackner P, et al. Survival and long-term functional
outcome in 1155 consecutive neurocritical care patients. Crit Care Med
2007;35(9):2025–30.
22. Sander A. The Extended Glasgow Outcome Scale. The Center for Outcome
Measurement in Brain Injury, 2002, http://www.tbims.org/combi/gose
(accessed September 24, 2012).
23. Collie A, Maruff P, Makdissi M, et al. CogSport: reliability and correlation
with conventional cognitive tests used in postconcussion medical
evaluations. Clin J Sport Med 2003;13:28–32.
24. McCrory P, Meeuwisse W, Johnston K, et al. Consensus statement on
concussion in sport: the 3rd International Conference on Concussion in
Sport, Zurich, November 2008. Br J Sports Med 2009;43(Suppl 1):i76–i90.
25. Iverson GL, Lovell MR, Collins MW. Validity of ImPACT for measuring
processing speed following sports-related concussion. J Clin Exp
Neuropsychol 2005;27(6):683–9.
26. Mayer SA, Brun NC, Begtrup K, et al. Recombinant activated factor VII for
acute intracerebral hemorrhage. N Engl J Med 2005;352:777–85.
27. Ponsford J, Draper K, Schonberger M. Functional outcome 10 years after
traumatic brain injury: its relationship with demographic, injury severity,
and cognitive and emotional status. J Int Neuropsychol Soc
2008;14(2):233–42.
28. Kirkness CJ, Burr RL, Mitchell PH. Intracranial and blood pressure
variability and long-term outcome after aneurysmal sub-arachnoid
hemorrhage. Am J Crit Care 2009;8(3):241–51.
29. Poon WS, Ng SC, Wong GK, et al. Chronic hydrocephalus that requires
shunting in aneurysmal subarachnoid haemorrhage: its impact on clinical
outcome. Acta Neurochir Suppl 2008;102:129–30.
30. World Health Organization (WHO). International Classification of
Impairments, Disabilities, and Handicaps: a manual of classification relating
to the consequences of disease. Geneva, 1980.
31. Teasdale G, Jeanette B. Assessment of coma and impaired consciousness:
practical scale. Lancet 1974;2:81–4.
32. Rappaport M, Hall KM, Hopkins K, et al. Disability rating scale for severe
head trauma: coma to community. Arch Phys Med Rehabil 1982;63:118–23.
33. Hall K, Cope N, Rappaport M. Glasgow outcome scale and disability rating
scale: comparative usefulness in following recovery in traumatic head injury.
Arch Phys Med Rehabil 1985;66:35–7.
34. Gouvier WD, Blanton PD, LaPorte KK, et al. Reliability and validity of the
disability rating scale and the levels of cognitive functioning scale in
monitoring recovery from severe head injury. Arch Phys Med Rehabil
1987;68(2):94–7.
35. Neese LE, Caroselli JS, Klaas R, et al. Neuropsychological assessment and the
disability rating scale (DRS): a concurrent validity study. Brain Inj
2000;14(8):719–24.
36. Rankin J. Cerebral vascular accidents in patients over the age of 60:
prognosis. Scott Med J 1957;2(5):200–15.
37. Van Swieten J, Koudstaal P, Visser M, et al. Interobserver agreement for the
assessment of handicap in stroke patients. Stroke 1988;19:604–7.
38. Shinohara Y, Minematsu K, Amano T, et al. Modified Rankin scale with
expanded guidance scheme and interview questionnaire: interrater
agreement and reproducibility of assessment. Cerebrovasc Dis 2006;21:
271–8.
39. Quinn TJ, Lees KR, Hardemark HG, et al. Initial experience of a digital
training resource for modified Rankin scale assessment in clinical trials.
Stroke 2007;38:2257–61.
40. Gresham GE, Phillips TF, Labi ML. ADL status in stroke: relative merit s of
these standard indexes. Arch Phys Med Rehabil 1980;61(8):355–8.
41. Koch S, Romano JG, Forteza AM, et al. Rapid blood pressure reduction in
acute intracerebral hemorrhage: feasibility and safety. Neurocrit Care
2008;8:316–21.
42. Naval NS, Abdelhak TA, Zeballos P, et al. Prior statin use reduces mortality
in intracerebral hemorrhage. Neurocrit Care 2008;8:6–12.
43. Kwiatkowski TG, Libman RB, Frankel M, et al. Effects of tissue plasminogen
activator for acute ischemic stroke at one year. N Engl J Med
1999;40:1781–7.
44. Albers GW, Bates VE, Clark WM, et al. Intravenous tissue-type plasminogen
activator for treatment of acute stroke. JAMA 2000;283(9):1145–50.
45. Stead LG, Gilmore RM, Bellolio MF, et al. Hyperglycemia as an independent
predictor of worse outcome in non-diabetic patients presenting with acute
ischemic stroke. Neurocrit Care 2009;10(2):181–6.
46. Smith WS, Tsao JW, Billings ME, et al. Prognostic significance of
angiographically confirmed large vessel intracranial occlusion in patients
presenting with acute brain ischemia. Neurocrit Care 2006;4(1):14–7.
47. Murthy JM, Chowdary GV, Murthy TV, et al. Decompressive craniectomy
with clot evacuation in large hemispheric hypertensive intracerebral
hemorrhage. Neurocrit Care 2005;2(3):258–62.
48. Schmidt JM, Rincon F, Fernandez A, et al. Cerebral infarction associated
with acute subarachnoid hemorrhage. Neurocrit Care 2007;7(1):10–7.
49. Shah S, Vanclay F, Cooper B. Improving the sensitivity of the Barthel index
for stroke rehabilitation. J Clin Epidemiol 1989;42(8):703–9.
50. Loewen SC, Anderson BA. Predictors of stroke outcome using objective
measurement scales. Stroke 1990;21:78–81.
A complete list of references for this chapter can be found online at
www.expertconsult.com.
 Section II—Clinical and Laboratory Assessment 113.e1
References
1. Hopkins RO, Jackson JC. Long-term neurocognitive function after critical
illness. Chest 2006;130(3):869–78.
2. Wilde EA, Whiteneck GG, Bogner J, et al. Recommendations for the use of
common outcome measures in traumatic brain injury research. Arch Phys
Med Rehabil 2010;91(11):1650–60.e17.
3. Bagiella E, Novack TA, Ansel B, et al. Measuring outcome in traumatic brain
injury treatment trials: recommendations from the traumatic brain injury
clinical trials network. J Head Trauma Rehabil 2010;25(5):375–82.
4. McCauley SR, Wilde EA, Anderson VA, et al. Recommendations for the use
of common outcome measures in pediatric traumatic brain injury research.
J Neurotrauma 2011; Aug 24. Epub ahead of print.
5. Fields JA, Ferman TJ, Boeve BF, et al. Neuropsychological assessment of
patients with dementing illness. Nat Rev Neurol 2011; Nov 1. Epub ahead of
print.
6. Lewis MW, Babbage DR, Leathem JM. Assessing executive performance
during cognitive rehabilitation. Neuropsychol Rehabil 2011;21(2):145–63.
Epub 2011 Jan 10.
7. American Congress of Rehabilitation Medicine, Brain Injury-
Interdisciplinary Special Interest Group, Disorders of Consciousness Task
Force, et al. Assessment scales for disorders of consciousness: evidence-based
recommendations for clinical practice and research. Arch Phys Med Rehabil
2010;91(12):1795–813.
8. Podell K, Gifford K, Bougakov D, et al. Neuropsychological assessment in
traumatic brain injury. Psychiatr Clin North Am 2010;33(4):855–76.
9. Bilder RM. Neuropsychology 3.0: evidence-based science and practice. J Int
Neuropsychol Soc 2011;17(1):7–13. Epub 2010 Nov 19.
10. Gottesman RF, Hillis AE. Predictors and assessment of cognitive dysfunction
resulting from ischaemic stroke. Lancet Neurol 2010;9(9):895–905.
11. Holtslag HR, Post MW, Lindeman E, et al. Long-term functional health
status of severely injured patients. Injury 2007;38(3):280–9.
12. Van der Schaaf M, Beelen A, Dongelmans DA, et al. Functional status after
intensive care: a challenge for rehabilitation professionals to improve
outcome. J Rehabil Med 2009;41(5):360–6.
13. Frontera JA, Fernandez A, Schmidt JM, et al. Defining vasospasm after
subarachnoid hemorrhage: what is the most clinically relevant definition?
Stroke 2009;40(6):1963–8.
14. Mikkelsen ME, Shull WH, Biester RC, et al. Cognitive, mood and quality of
life impairments in a select population of ARDS survivors. Respirology
2009;14(1):76–82.
15. Van Baalen B, Odding E, van Woensel MP, et al. Reliability and sensitivity to
change of measurement instruments used in a traumatic brain injury
population. Clin Rehabil 2006;20(8):686–700.
16. de Haan R, Limburg M, Bossuyt P, et al. The clinical meaning of Rankin
“handicap” grades after stroke. Stroke 1995;26:2027–30.
17. Lipsett PA, Swoboda SM, Dickerson J, et al. Survival and functional outcome
after prolonged intensive care unit stay. Ann Surg 2000;231(2):262–8.
18. Boake C, High WM. Functional outcome from traumatic brain injury. Am J
Phys Med Rehabil 1996;75(2):105–13.
19. von Elm E, Osterwalder JJ, Graber C, et al. Severe traumatic brain injury in
Switzerland—feasibility and first results of a cohort study. Swiss Med Wkly
2008;138(23-24):327–34.
20. Lippert-Gruner M, Maegele M, Haverkamp H, et al. Health-related quality
of life during the first year after severe brain trauma with and without
polytrauma. Brain Inj 2007;21(5):451–5.
21. Broessner G, Helbok R, Lackner P, et al. Survival and long-term functional
outcome in 1155 consecutive neurocritical care patients. Crit Care Med
2007;35(9):2025–30.
22. Sander A. The Extended Glasgow Outcome Scale. The Center for Outcome
Measurement in Brain Injury, 2002, http://www.tbims.org/combi/gose
(accessed September 24, 2012).
23. Collie A, Maruff P, Makdissi M, et al. CogSport: reliability and correlation
with conventional cognitive tests used in postconcussion medical
evaluations. Clin J Sport Med 2003;13:28–32.
24. McCrory P, Meeuwisse W, Johnston K, et al. Consensus statement on
concussion in sport: the 3rd International Conference on Concussion in
Sport, Zurich, November 2008. Br J Sports Med 2009;43(Suppl 1):i76–i90.
25. Iverson GL, Lovell MR, Collins MW. Validity of ImPACT for measuring
processing speed following sports-related concussion. J Clin Exp
Neuropsychol 2005;27(6):683–9.
26. Mayer SA, Brun NC, Begtrup K, et al. Recombinant activated factor VII for
acute intracerebral hemorrhage. N Engl J Med 2005;352:777–85.
27. Ponsford J, Draper K, Schonberger M. Functional outcome 10 years after
traumatic brain injury: its relationship with demographic, injury severity, and
cognitive and emotional status. J Int Neuropsychol Soc 2008;14(2):233–42.
28. Kirkness CJ, Burr RL, Mitchell PH. Intracranial and blood pressure
variability and long-term outcome after aneurysmal sub-arachnoid
hemorrhage. Am J Crit Care 2009;8(3):241–51.
29. Poon WS, Ng SC, Wong GK, et al. Chronic hydrocephalus that requires
shunting in aneurysmal subarachnoid haemorrhage: its impact on clinical
outcome. Acta Neurochir Suppl 2008;102:129–30.
30. World Health Organization (WHO). International Classification of
Impairments, Disabilities, and Handicaps: a manual of classification relating
to the consequences of disease. Geneva, 1980.
31. Teasdale G, Jeanette B. Assessment of coma and impaired consciousness:
practical scale. Lancet 1974;2:81–4.
32. Rappaport M, Hall KM, Hopkins K, et al. Disability rating scale for severe
head trauma: coma to community. Arch Phys Med Rehabil 1982;63:118–23.
33. Hall K, Cope N, Rappaport M. Glasgow outcome scale and disability rating
scale: comparative usefulness in following recovery in traumatic head injury.
Arch Phys Med Rehabil 1985;66:35–7.
34. Gouvier WD, Blanton PD, LaPorte KK, et al. Reliability and validity of the
disability rating scale and the levels of cognitive functioning scale in
monitoring recovery from severe head injury. Arch Phys Med Rehabil
1987;68(2):94–7.
35. Neese LE, Caroselli JS, Klaas R, et al. Neuropsychological assessment and the
disability rating scale (DRS): a concurrent validity study. Brain Inj
2000;14(8):719–24.
36. Rankin J. Cerebral vascular accidents in patients over the age of 60:
prognosis. Scott Med J 1957;2(5):200–15.
37. Van Swieten J, Koudstaal P, Visser M, et al. Interobserver agreement for the
assessment of handicap in stroke patients. Stroke 1988;19:604–7.
38. Shinohara Y, Minematsu K, Amano T, et al. Modified Rankin scale with
expanded guidance scheme and interview questionnaire: interrater
agreement and reproducibility of assessment. Cerebrovasc Dis
2006;21:271–8.
39. Quinn TJ, Lees KR, Hardemark HG, et al. Initial experience of a digital
training resource for modified Rankin scale assessment in clinical trials.
Stroke 2007;38:2257–61.
40. Gresham GE, Phillips TF, Labi ML. ADL status in stroke: relative merit s of
these standard indexes. Arch Phys Med Rehabil 1980;61(8):355–8.
41. Koch S, Romano JG, Forteza AM, et al. Rapid blood pressure reduction in
acute intracerebral hemorrhage: feasibility and safety. Neurocrit Care
2008;8:316–21.
42. Naval NS, Abdelhak TA, Zeballos P, et al. Prior statin use reduces mortality
in intracerebral hemorrhage. Neurocrit Care 2008;8:6–12.
43. Kwiatkowski TG, Libman RB, Frankel M, et al. Effects of tissue plasminogen
activator for acute ischemic stroke at one year. N Engl J Med
1999;40:1781–7.
44. Albers GW, Bates VE, Clark WM, et al. Intravenous tissue-type plasminogen
activator for treatment of acute stroke. JAMA 2000;283(9):1145–50.
45. Stead LG, Gilmore RM, Bellolio MF, et al. Hyperglycemia as an independent
predictor of worse outcome in non-diabetic patients presenting with acute
ischemic stroke. Neurocrit Care 2009;10(2):181–6.
46. Smith WS, Tsao JW, Billings ME, et al. Prognostic significance of
angiographically confirmed large vessel intracranial occlusion in patients
presenting with acute brain ischemia. Neurocrit Care 2006;4(1):14–7.
47. Murthy JM, Chowdary GV, Murthy TV, et al. Decompressive craniectomy
with clot evacuation in large hemispheric hypertensive intracerebral
hemorrhage. Neurocrit Care 2005;2(3):258–62.
48. Schmidt JM, Rincon F, Fernandez A, et al. Cerebral infarction associated
with acute subarachnoid hemorrhage. Neurocrit Care 2007;7(1):10–7.
49. Shah S, Vanclay F, Cooper B. Improving the sensitivity of the Barthel index
for stroke rehabilitation. J Clin Epidemiol 1989;42(8):703–9.
50. Loewen SC, Anderson BA. Predictors of stroke outcome using objective
measurement scales. Stroke 1990;21:78–81.
51. Granger CV, Dewis LS, Peters NC, et al. Stroke rehabilitation: analysis of
repeated Barthel index measures. Arch Phys Med Rehabil 1979;60(1):14–7.
52. Morgalla MH, Will BE, Roser F, et al. Do long-term results justify
decompressive craniectomy after severe traumatic brain injury? J Neurosurg
2008;109:685–90.
53. Powell J, Heslin J, Greenwood R. Community based rehabilitation after
severe traumatic brain injury: a randomized controlled trial. J Neurol
Neurosurg Psychiatry 2002;72:193–202.
54. Yarkony GM, Roth, EJ, Heinemann AW, et al. Benefits of rehabilitation for
traumatic spinal cord injury. Arch Neurol 1987;44(1):93–6.
55. Altman DG, Royston P: The cost of dichotomising continuous variables.
BMJ 2006;332:1080.
56. Maas AI, Steyerberg EW, Marmarou A, et al. IMPACT recommendations for
improving the design and analysis of clinical trials in moderate to severe
traumatic brain injury. Neurotherapeutics 2010;7:127–34.
113.e2 Section II—Clinical and Laboratory Assessment
57. McHugh GS, Butcher I, Steyerberg EW, et al. A simulation study evaluating
approaches to the analysis of ordinal outcome data in randomized controlled
trials in traumatic brain injury: results from the IMPACT project. Clin Trials
2010;7:44–57.
58. Murray GD, Barer D, Choi S, et al. Design and analysis of phase III trials
with ordered outcome scales: the concept of the sliding dichotomy.
J Neurotrauma 2005;22:511–17.
59. Roozenbeek B, Lingsma HF, Perel P, et al. IMPACT (International Mission
on Prognosis and Clinical Trial Design in Traumatic Brain Injury) study
group; CRASH (corticosteroid randomisation after significant head injury)
trial collaborators. The added value of ordinal analysis in clinical trials: an
example in traumatic brain injury. Crit Care 2011;15(3):R127.
60. Optimising Analysis of Stroke Trials (OAST) Collaboration, Bath PM, Gray
LJ, et al. Can we improve the statistical analysis of stroke trials? Statistical
reanalysis of functional outcomes in stroke trials. Stroke 2007;38:1911–15.
61. Findler M, Cantor J, Haddad L, et al. The reliability and validity of the SF-36
health survey questionnaire for use with individuals with traumatic brain
injury. Brain Inj 2001;15(8):715–23.
62. MacKenzie EJ, McCarthy ML, Ditunno JF, et al. Using the SF-36 for
characterizing outcome after multiple trauma involving head injury.
J Trauma 2002;52(3):527–34.
63. Timofeev I, Kirkpatrick PJ, Corteen E, et al. Decompressive craniectomy in
traumatic brain injury: outcome following protocol-driven therapy. Acta
Neurochir Suppl 2006;96:11–6.
64. Neugebauer E, Bouillon B, Bullinger M, et al. Quality of life after multiple
trauma: summary and recommendations of the consensus conference. Rest
Neurol Neurosci 2002;20: 161–7.
65. McNaughton HK, Weatherall M, McPherson KM. Functional measures
across neurologic disease states: analysis of factors in common. Arch Phys
Med Rehabil 2005;86(11):2184–8.
66. Nestvold K, Stavem K. Determinants of health-related quality of life 22 years
after hospitalization for traumatic brain injury. Brain Inj 2009;23(1):15–21.
67. Andelic N, Hammergren N, Bautz-Holter E, et al. Functional outcome and
health-related quality of life 10 years after moderate-to-severe traumatic
brain injury. Acta Neurol Scand 2008;120(1):16–23.
68. Jackson JC, Hart RP, Gordon SM, et al. Six-month neuropsychological
outcome of medical intensive care unit patients. Crit Care Med
2003:31(4):1226–34.
69. Cote I, Gregoire JP, Moisan J, et al. Quality of life in hypertension: the SF-12
compared to the SF-36. Can J Clin Pharmacol 2004;11(2):232–8.
70. Kiely JM, Grasel KJ, Guse CE, et al. Correlation of SF-12 and SF-36 in a
trauma population. J Surg Res 2006;132(2):214–18.
71. Riddle DL, Lee KT, Stratford PW. Use of SF-36 and SF-12 health status
measures: a quantitative comparison for groups versus individual patients.
Med Care 2001;39(8):867–78.
72. Wee CC, Davis RB, Hamel MB. Comparing the SF-12 and SF-36 health
status questionnaires in patients with and without obesity. Health Qual Life
Out 2008;30(6):6–11.
73. Nanda U, McLendon PM, Andresen EM, et al. The SIP68: an abbreviated
sickness impact profile for disability outcomes research. Qual Life Res
2003;12(5):583–95.
74. Lezak M. Relationships between personality disorders, social disturbances,
and physical disability following traumatic brain injury. J Head Trauma
Rehabil 1987;2(1):57–69.
75. Lezak M, O’Brien K. Longitudinal study of emotional, social, and physical
changes after traumatic brain injury. J Learn Disabil 1988;21(8):456–63.
76. Malec J. The Mayo Portland Adaptability Inventory. The Center for Outcome
Measurement in Brain Injury, 2005, http://www.tbims.org/combi/mpai
(accessed September 24, 2012).
77. Malec J. Comparability of Mayo-Portland adaptability inventory ratings by
staff, significant others and people with acquired brain injury. Brain Inj
2004;18(6):563–75.
78. Bezner JR, Hunter DL. Wellness perception in persons with traumatic brain
injury and its relation to functional independence. Arch Phys Med Rehabil
2001;82(6):787–92.
79. Wright J. The functional assessment measure. The Center for Outcome
Measurement in Brain Injury, 2000, http://www.tbims.org/combi/FAM
(accessed September 24, 2012).
80. Jackson JC, Obremskey W, Bauer R, et al. Long-term cognitive, emotional,
and functional outcomes in trauma intensive care unit survivors without
intracranial hemorrhage. J Trauma Inj Infect Crit Care 2007;62(1):80–8.
81. Folstein MF, Folstein SE, McHugh PR. Mini-mental state: a practical method
for grading the state of patients for the clinician. J Psychiatr Res
1975;12:189–98.
82. Cockrell JR, Folstein MF. Mini mental state examination (MMSE).
Psychopharmacology 1988;24(4):689–92.
83. Anthony JC, LeResche L, Niaz U, et al. Limits of the ‘mini-mental state’ as a
screening test for dementia and delirium among hospital patients. Psychol
Med 1982;12(2):397–408.
84. Tombaugh TN, McIntyre NJ. The mini-mental state examination: a
comprehensive review. J Am Geriatr Soc 1992;40(9):922–35.
85. Crum RM, Anthony JC, Bassett SS, et al. Population-based norms for the
mini-mental state examination by age and educational level. JAMA
1993;18:2386–91.
86. Nasreddine ZS, Phillips NA, Bedirian V, et al. The Montreal cognitive
assessment, MOCA: a brief screening tool for mild cognitive impairment.
J Am Geriatr Soc 2005;53(4):695–9.
87. Ruff RL, Ruff SS, Wang X-F. Headaches among operation Iraqi freedom/
operation enduring freedom veterans with mild traumatic brain injury
associated with exposures to explosions. J Rehabil Res Dev
2008;45(7):941–52.
88. Rabin LA, Barr WB, Burton LA. Assessment practices of clinical
neuropsychologists in the United States and Canada: a survey of INS, NAN
and APA division 40 members. Arch Clin Neuropsychol 2005;20:33–65.
89. Mueller J, Kiernan R, Langston JW. Cognistat manual. Fairfax, CA:
Cognistat, Inc.; 2009.
90. Goodglass H, Kaplan E, Barresi B. Boston diagnostic aphasia examination.
3rd ed. Lutz, FL: Psychological Assessment Resources; 2000.
91. Jurica PJ, Leitten CL, Mattis S. DRS-2: Dementia rating scale-2 professional
manual. Lutz, FL: Psychological Assessment Resources, Inc; 2001.
92. Lucas JA, Ivnik RJ, Smith GE, et al. Normative date for the Mattis demential
rating scale. J Clin Exper Neuropsychol 1998;20(40):536–47.
93. Ivnik RJ, Malec JF, Smith GE, et al. Mayo’s older Americans normative
studies: WAIS-R, WMS-R, AND AVLT norms for ages 56 through 97. Clin
Neuropsychol 1992;6(Suppl.):1–104.
94. Bank AL, Yochim BP, MacNeill SE, et al. Expanded normative data for the
Mattis dementia rating scale for use with urban, elderly medical patients.
Clin Neuropsychol 2000;14:149–56.
95. Schmidt KS, Mattis PJ, Adams J, et al. Alternate-form reliability of the
dementia rating scale-2. Arch Clin Neuropsychol 2005;20(4):435–41.
II
Chapter
13  
Brain Death
Farzana Tariq and Peter M. Black
Introduction
Brain death is one of the most important diagnoses to be
made in the neurocritical care unit (NCCU). The increasing
ability to maintain the heart and other vital organs in the face
of devastating brain injury has made it mandatory to recog-
nize circumstances in which further intervention is futile
because of widespread irreversible brain destruction. Imple-
menting criteria for brain death is important for:
1. Family and patient decision making
2. Effective use of expensive technology and resources
3. Potential organ donation to allow other patients a new life
Bioethicists, physicians, philosophers, religious leaders, and
those in the legal profession, among others, will continue to
discuss (and disagree) about how to best define human death.
These theories can range from the Harvard Criteria of Brain
Death that defines death as the cessation of all brain activity,
through the Irreversibility Standard that defines death when
the capacity for consciousness is lost forever, to the Cognitive
Standard that defines death as the loss of almost all core
mental properties (e.g., memory, self-consciousness, moral
agency, and the capacity for reason).1 This chapter emphasizes
the practical determination of brain death in the NCCU in
adults and relevance to organ donation, and discusses the
philosophical concepts of brain death. End-of-life care, brain
death criteria, and the concept and understanding of brain
death vary in different countries.2 In this chapter, the focus is
on how brain death is diagnosed in the United States.
Historical Background
Death was described for many centuries as cessation of circu-
lation and respiration. This definition was no longer valid
when modern medical devices, increasingly sophisticated
ICUs, and greater understanding of physiology in the face of
severe brain injury made it possible to maintain artificial res-
piration and circulation indefinitely despite massive brain
destruction.3 This brought a need for more precise criteria to
define brain death.
An early landmark was the publication of criteria for irre-
versible coma in 1968 by an ad hoc committee at Harvard
Medical School.4 This group described a condition known as
irreversible coma that they equated with cerebral death or
brain death. They created a set of practical steps that could be
used to make this diagnosis. Although the field has changed
114
in the 40 years since the report, the fundamental criteria have
not changed markedly—lack of brainstem reflexes, lack of
cortical activity as shown by electroencephalography (EEG)
and clinical examination, and irreversibility.
The initial development of this concept was marked by two
major refinements, both of considerable importance for
intensive care worldwide. The first was acceptance of the
concept of brain death. In most Western cultures, acceptance
of this concept is complete; in Japan and other Eastern tradi-
tions, this concept is increasingly being accepted. In the United
States, an important landmark was the drafting of a model act
in 1980 known as the Uniform Determination of Death Act,5
which states:
An individual who has sustained either irreversible cessa-
tion of circulatory and respiratory functions or irreversible
cessation of all function of the entire brain including the
brainstem, is dead. The determination of brain death must
be made in accordance with accepted medical standards.
The adoption of this statute by many states helped establish
the concept that brain death was death. Uniform acceptance
of the concept in case law in other states confirmed the idea
that brain death equates to human death. The concept of brain
death became standard. Although there are two explicit defini-
tions of death in this statute, cardiorespiratory function can be
said to be important only to perfuse the brain, so death of the
brain becomes the major criterion for death in both definitions.
Since the adoption of the Uniform Determination of Death
Act, all court rulings in which brain death has been challenged
in the United States have upheld the medical practice of death
determination using neurologic criteria according to state law.
Two major issues that arise in court rulings are consequences
of documentation of the time of brain death and family-
physician discord on withdrawal of intensive care support.5
The second area of increasing refinement has been the cri-
teria used to make the determination of death. With time,
increasing emphasis has been placed on blood flow determi-
nation to establish this diagnosis, although the fundamental
criteria of absent brainstem and cortical function and irrevers-
ibility remain the core criteria.6
Brain Death Medical Criteria
The American Academy of Neurology has provided evidence-
based practice parameters to determine brain death in
adults.7,8 However, there still is variance in how this is done in
© Copyright 2013 Elsevier Inc. All rights reserved.
 Section II—Clinical and Laboratory Assessment 115
different hospitals, states, and countries that can create poten-
tial pitfalls in the process.9 Parameters that commonly vary
include prerequisites to declare brain death, qualification
of physicians, acceptable body temperature when the exami-
nation is performed, number of required examinations, need
for a second physician to confirm the diagnosis, and docu-
mentation. Despite these variances, the basic medical criteria
for brain death include:
 Coma or unresponsiveness (with a known cause)
 Absence of purposeful motor responses in all extremities

Absence of brainstem reflexes including apnea

Demonstrated irreversibility
The details of the clinical testing are described later. It is
important to know the cause of the patient’s coma and to
determine that the condition is irreversible. This means
excluding conditions that may mimic brain death, such as
barbiturate intoxication, valproic acid intoxication, severe
alcohol intoxication, sedative overdose, organophosphate poi-
soning, baclofen overdose, high cervical cord injury, or hypo-
thermia. In rare instances, snake bites, Miller Fisher syndrome
(cranial nerve palsy, ataxia, areflexia), fulminant Guillain-
Barré syndrome that includes both central and peripheral
nervous system involvement, and brainstem encephalitis also
may mimic brain death. It also means observing the patient
for long enough that it is clear that the condition is not going
to resolve—a minimum of 6 hours in most current proto-
cols.10,11 Finally, some protocols require confirmation of brain
death by a neurologist or neurosurgeon. While there can be
potential mimics that neurointensivists should be aware of,
there are no published reports in adults of recovery of neuro-
logic function after a brain death diagnosis using the criteria
described in the 1995 American Academy of Neurology prac-
tice parameter.8 In children, brain death may be more compli-
cated to diagnose. Guidelines add a component of continued
unresponsiveness for up to 48 hours. Table 13.1 presents these
criteria in summary form.
Table 13.1  Criteria for the Clinical
Determination of Brain Death
Coma
Absence of reversible medical conditions:
Hypothermia <32°C
Drug overdose (barbiturate, benzodiazepines, narcotics)
Electrolyte anomalies (metabolic acidosis)
Absent pupillary light reflex
Absent corneal reflex
Absent caloric response
Absent gag reflex
Absent cough in response to tracheal suctioning
Absent sucking and rooting reflex
Absent respiratory drive at PaCO2 that is 60 mm Hg or 20 mm
Hg above normal baseline values
Interval between two evaluations is age dependent:
Term to 2 months old, 48 hr
>2 months to 1 year old, 24 hr
>1 year to <18 years old, 12 hr
>18 years old interval is optional, usually 6 hr
Determination of Brain Death
Clinical Examination
A careful well-documented clinical examination is central to
the diagnosis of brain death. The time and detailed set of tests
performed should be noted in the record; the apnea test needs
to be done only once, usually at the end of the time period
used for the determination (6 hours in most cases). Prerequi-
sites to the examination include the following: (1) establish
the cause and irreversible nature of the coma, (2) exclude the
use of any central nervous system depressant drugs or neuro-
muscular blockade, (3) ensure normal temperature (>36° C)
and systolic blood pressure (>100 mm Hg, and (4) ensure
laboratory values are close to normal.
Coma
Brain dead patients are in a state of deep coma. The cause of
coma should be determined and should be irreversible. If the
cause is not known, then longer observation is required.12
Based on the Glasgow Coma Scale (see Chapter 11), the
patient should have a score of 3; based on the FOUR score,13
the patient should have a score of E(0)M(0)B(0)R(0).
Motor Responses
There should be no purposeful motor response to painful
stimuli such as pressure on the supraorbital ridge or nail bed.
Pinching a patient on the chest is not an adequate way to elicit
these responses. Reflex motor responses that do not preclude
the diagnosis of brain death include the following: facial myo-
kymia (spontaneous twitching of the cheek); undulating toe
movements, reflecting L5 and S1 activity; the Lazarus sign
(movements of upper limb from spinal cord integrated reflex
activity); abdominal reflexes; flexion reflexes; and periodic leg
movements, especially if they are generated during apnea
testing, respiratory acidosis, or brisk neck flexion. It is impor-
tant to understand the origin of these movements because
they may pose concern to nurses, physicians, and family
members about whether there is whole brain destruction.
These spinal cord reflexes occur more often in young adults
than in older adults.14-18
Brainstem Reflexes
The following tests are performed to evaluate brainstem
function.
Pupillary response: The pupils may be round, oval, or irregu-
larly shaped. They are usually midsize (4 to 6 mm) but may
be dilated. Drugs and preexisting anatomic ocular abnormali-
ties may be a confounding factor.19 The pupillary light reflex
(testing the second and third cranial nerves) must be absent.
Pilocarpine supersensitivity due to denervation following
instillation of 0.06% of the drug also may be seen. In some
instances spontaneous asynchronous pupillary constriction
and dilation may occur, which should not be a confounding
factor for brain death diagnosis.20,21
Ocular movements: Both oculocephalic (doll’s eye response)
and vestibulo-ocular (caloric test) reflexes must be absent in
brain death patients. The oculocephalic reflex is elicited by
rapidly and vigorously turning the head to 90 degrees laterally
on both sides. A normal response is deviation of the eyes to
116 Section II—Clinical and Laboratory Assessment
the opposite side of the head turning; in brain death, there is
no deviation of eyes. This test should not be done if there is
a suspected fracture or instability of the spine. The vestibulo-
ocular reflex is elicited by raising the head to 30 degrees and
irrigating both tympanic membranes with 50 mL of iced
saline or water. In brain death patients, there is no eye devia-
tion. The patient should be observed for up to 1 minute fol-
lowing irrigation, with a 5-minute wait between testing for
each ear. Contraindications to the test include rupture of the
tympanic membrane.
Drugs including sedatives, aminoglycoside antibiotics, tri-
cyclic antidepressants, and some antiseizure agents can dimin-
ish the oculocephalic and vestibulo-ocular reflexes. Facial
trauma involving auditory canal and petrous bone also can
inhibit these reflexes.
Corneal and facial motor responses: Both corneal and facial
reflexes are absent in brain death patients. The corneal reflex
is elicited by touching the cornea of the patient with a cotton
swab; the normal response is a blink, which is absent in brain
death. The facial motor response is elicited by applying pres-
sure to the temporomandibular joint, supraorbital ridge, or
nail bed. A normal response is facial grimacing, which is not
seen in patients with brain death. Interpretation of these
maneuvers can be difficult in patients with facial trauma.
Pharyngeal and tracheal reflexes: Both of these reflexes are
absent in brain death patients. The pharyngeal reflex is elicited
by touching the posterior pharyngeal wall with a tongue blade
that normally results in a gag. The cough reflex is elicited by
using bronchial suctioning. These reflexes may be difficult to
evaluate in orally intubated patients but may be feasible with
deep suctioning in the orally intubated patient.22-24
Apnea testing: The medulla controls the central breathing
center via chemoreceptors. These receptors sense changes in
PaCO2. During apnea testing, target PaCO2 levels are up to
60 mm Hg. To do this test, the patient is removed from the
ventilator; 100% oxygen may be administered via the endo-
tracheal tube. In case of chronic hypercarbia (e.g., a patient
with chronic obstructive pulmonary disease), higher target
values of PaCO2 are required. Cardiac dysrhythmias and
hypotension are complications associated with apnea testing
and can be reduced by taking precautionary measures, such as
preoxygenation and adequate maintenance of the baseline sys-
tolic blood pressure up to 120 mm Hg with pressors. Both
hypocarbia and hypercarbia diminish viability of the organs
for organ donation; one suggestion is to administer CO2 exog-
enously after preoxygenation to shorten the interval for hyper-
carbia.25,26 Many patients who undergo apnea testing are on
vasopressors, and an apnea test is not possible in between 5%
and 10% of patients because of hemodynamic instability or
inadequate lung function.27 In hemodynamically stable
patients, it is rare that apnea testing is aborted (<5%). Patients
who fail completion of apnea testing tend to be younger, have
significantly greater alveolar arterial (A-a) gradients, and are
more acidotic.28 Confirmatory tests then are needed if apnea
testing is not possible or is aborted.
Unstable blood pressure and heart rate: Unstable blood pres-
sure and heart rate may provide an early indication of hypo-
thalamic disturbance and impending death. Patients who
subsequently meet the criteria for brain death have a higher
heart rate and less variability as a result of loss of vegetative
impulses from the brainstem. This may be a useful finding in
deeply comatose patients.29-31
Time of Observation
The clinical examination should be repeated after 6 hours
before labeling a person brain dead. The need for repeat exam-
ination and the timing of repeat examination may vary among
institutions. In some described criteria for brain death, the
examination should be done by two physicians.32-35 In practice,
however, the observation time to a second neurologic exami-
nation is often much greater. For example, in a recent review
of 1229 adult and 82 pediatric patients pronounced brain
dead in New York hospitals serviced by the New York Organ
Donor Network, the mean brain death declaration interval
between two examinations was 19.2 hours, that is three times
longer than the state guideline.36 Brain death examinations
were less frequent on weekends and the interval longer in
smaller hospitals. A longer interval was associated with a
decrease in organ donation. The authors found that none of
the patients declared brain dead regained brainstem function
upon repeat examination and so suggested that a single brain
death examination for patients older than 1 year may suffice
and so increase organ donation and reduce intensive care unit
(ICU) costs.36 Longer brain death duration (i.e., time between
declaration of brain death and organ harvesting), however,
may not adversely affect organ survival on transplant.37 Newer
practice parameters support a single examination rather than
dual examination; the impact is unclear, although early studies
suggest equivalence.38
Confirmatory Tests
Confirmatory tests are not mandatory unless clinical tests are
inconclusive. While there are several reports where the value
of ancillary radiologic imaging is described in decisions about
brain death,39 a comprehensive clinical evaluation performed
by skilled examiners is considered to have perfect diagnostic
accuracy.40 Possible confirmatory tests are listed in Table 13.2
and are presented briefly here as tests of cerebral blood flow
(CBF) and electrodiagnostic tests.
Tests of Cerebral Blood Flow
Cerebral angiography. Cerebral angiography is the “gold stan-
dard” to demonstrate absent CBF. It is the most reliable test
to confirm brain death, and can save time and reduce an
unnecessary stay in the ICU. However, it is cumbersome and
requires moving a critically ill patient into an angiography
suite for several hours. In brain death patients, cerebral circu-
lation is absent at the level of the carotid bifurcation, although
flow may be demonstrated in the external carotid circulation.
Extravasation of contrast may result from autolysis.41-44
Transcranial Doppler ultrasonography (TCD). TCD is an
important noninvasive confirmatory indirect blood flow test
that can be done daily in the NCCU. When criteria recom-
mended by the Task Force Group on Cerebral Death of the
Neurosonology Research Group of the World Federation of
Neurology are used, TCD is a reliable method to establish
brain death.45 Brain death patients may show the following
findings: (1) brief systolic flow and no diastolic flow, (2) no
flow at all in a patient who previously had documented flow,
and (3) brief systolic forward flow with systolic spikes and
diastolic reverse flow. In patients with open skull fractures,
external cerebrospinal fluid drainage, or large decompressive
 Section II—Clinical and Laboratory Assessment 117

Table 13.2  Confirmatory Tests That Can Be Electrodiagnostic Testing

Used to Help Confirm the Clinical Diagnosis
of Brain Death
Confirmatory Test
Cerebral angiography
Electroencephalogram
Transcranial Doppler
ultrasonography
Somatosensory and
auditory evoked potential
Technetium 99 scan
Xenon computed
tomography
Brain tissue oxygenation
Blood pressure and heart
rate variability
Results
No circulation at level of
carotid bifurcation and circle
of Willis
Patent external circulation
Electrocerebral silence for
30 minutes
Brief systolic flow and reverse
diastolic flow
Brief systolic flow and no
diastolic flow
No flow in patients with
previously documented flow
No response
No radionuclide uptake by
brain parenchyma
Average global flow of less
than 5 mL/100 mL/min
Persistent zero value for more
than 30 min
Absent variability
EEG. An EEG is usually performed for 30 minutes before
concluding brain death because of electrocerebral silence. It is
recommended that a minimum of eight scalp electrodes be
used with a distance between electrodes of at least 10 cm.
However, the EEG is not entirely reliable as a confirmatory test
to document brain death in that minimal electrical activity
may be seen in some patients who otherwise fulfill brain death
criteria, and the EEG may appear flat in cases of comatose
patients with intact brainstem functioning. According to the
American Encephalographic Society guidelines, electrocere-
bral silence is confirmed when there is absence of cortical
activity greater than 2 µV for 30 minutes. EEG results may
be contaminated by electromyographic artifacts from scalp
motor units.61,62
Somatosensory and brainstem auditory evoked potentials.
The somatosensory evoked potential (SSEP) is a bedside test
done with a portable instrument via stimulation of the median
nerve. When SSEP is used for brain death determination, it
presupposes that there is absent pathology in the intervening
structures between the peripheral nerve and the cortex (e.g.,
cervical spine injury). Patients with brainstem death have no
response. These tests are useful in patients in whom mislead-
ing factors such as depressant drugs, hypothermia, and meta-
bolic disturbances are present. These tests are less sensitive
and not recommended for children younger than 6 months
of age.63-65

craniotomies, oscillating flow can be seen, and TCD should be

carried out by an experienced person to avoid misinterpreta-
tion of collateral flow signals.46-50
Radionuclide studies. Cerebral planar scintigraphy or cere-
bral scintotomography with Tc-99 hexamethylpropylenea-
mine oxime is a safe, sensitive, and cost-effective test that may
be extremely helpful in patients with a skull defect or scalp
trauma in whom angiography may be difficult to interpret
or when there are other confounding factors that may make
diagnosis equivocal.51 Approximately 25 mCi is given intra­
venously and images are obtained with the help of a mobile
scintillation camera.52,53
Xenon-enhanced computed tomography (Xe-CT). Xe-CT
is an important tool to measure CBF arrest as a confirmatory
test for brain death. An average global flow of less than
5 mL/100 mL/min confirms brain death. This also is a clini-
cally useful tool to determine prognosis of the patient.54 Pres-
ently, the U.S. Food and Drug Administration limits the use
of xenon to research protocols only in the United States.
Brain tissue oxygenation and other tests. Monitoring of brain
tissue oxygenation in the comatose patient may help establish
the diagnosis of brain death. A sustained (>30 min) partial
pressure of brain tissue oxygen of zero is consistent with brain
death. This measurement may be helpful in pharmacologically
depressed patients, including children, to suggest when formal
brain death assessment should be performed.55-57 However,
brain oxygen levels have yet to be incorporated into defini-
tions of brain death. Newer diagnostic tests, such as bispectral
index, magnetic resonance imaging, computed tomography
angiography (CTA), and computed tomography perfusion
(CTP), have been suggested as useful,58-60 but there is insuffi-
cient evidence to determine whether newer ancillary tests can
confirm the cessation of brain function.8
Controversies in Brain Death
There remain several important philosophical and practical
issues in brain death despite its increasing acceptance world-
wide (see also Chapter 8). Some suggest that the definition of
brain death should be broadened further, and others suggest
it already is too broad. The central issue for both groups is the
concept of irreversible coma and particularly loss of respira-
tory drive. A patient who is unresponsive and who has no
corneal reflexes but who has spontaneous respiration is not
brain dead by present criteria. Although recent changes may
allow such patients to have support stopped, this is different
from declaring them dead by brain criteria. A similar issue
arises in anencephalic infants, in whom there is no cortex but
some of the brainstem is functioning including spontaneous
breathing. Even though the child can never gain conscious-
ness, he or she is not brain dead.10-12
Some authors have suggested that irreversible prolonged
loss of consciousness is adequate to establish brain death.
Proponents of this position suggest that:
1. Some clinically dead patients maintain residual vegetative
functions that are mediated by the brain or brainstem (e.g.,
hypothalamic function when they do not manifest diabe-
tes insipidus).
2. Some patients may retain slight cerebral electrical activity
on EEG despite having no brainstem reflexes.
3. Some patients have stereotyped movements in the form of
complex spinal reflexes that are difficult to tell from pur-
poseful movements.
4. Confirmatory tests are necessarily positive predictors or
may have ambiguity in their interpretation to reliably
establish death.
118 Section II—Clinical and Laboratory Assessment
Proponents of this view postulate that brain death criteria
should be based on the diagnosis of permanent loss of con-
sciousness rather than loss of total brain function.66-73 Their
new formulation has three components: definition, criteria,
and diagnosis:
Definition: Irreversible loss of consciousness, which is the
most integrating function of the organism
Criteria: Loss of cortical-subcortical connections to gener-
ate both components of consciousness (capacity and
cognition)
Diagnostic tests: No waking response to stimuli (capacity),
no cognitive or affective functions (content)74
A second idea is to turn time back to 1968 and reform the
concept of “irreversible coma” to “irreversible apneic coma”
and so abandon the idea of whole brain death but accept
instead lack of brainstem function as a sufficient criterion.
This was the principle behind British guidelines. These ideas
have also generated discussion about organ procurement and
the dead donor rule that currently states that organ donors
must be dead before donation. Donors when declared brain
dead are dead by neurologic criteria. Consequently there is
debate whether patients who have suffered severe and irrevers-
ible brain damage can become organ donors, even though
they are not yet dead and organ removal then becomes the
proximal cause of death.75 For the moment, none of these
points of view have enough proponents to make a decisive
change in the criteria already presented.
Since the ad hoc Harvard report was issued, the medical
and legal definitions of death have evolved and concepts
such as loss of integration of the whole organism, loss of
autonomy, and loss of personhood have generated discus-
sion. Functional neuroimaging studies also have provided
new insights into patients with severe brain injury and
coma.76 In addition, some scholars advocate return to the
traditional circulatory and respiratory criteria to diagnose
death since brain dead patients may manifest a large variety
of biologic function when mechanically ventilated.77 Recent-
ly the Institute of Medicine, the Joint Commission on the
Accreditation of Healthcare Organizations, and the United
Network for Organ Sharing have proposed organ donation
after cessation of circulation and respiration as a way to in-
crease organ procurement. In ICU care, this concept of
“timed death” and subsequent organ donation (i.e., con-
trolled organ donation after cardiac death) has generated
controversy about the limits of treatment that aid organ
transplant but that may hasten death, and the period of
cardiac arrest needed to declare death and commence organ
procurement.78 A critical factor to understand the meaning
of death using circulatory-respiratory tests is the distinction
between the “permanent” (will not reverse) and “irrevers-
ible” (cannot reverse) cessation of circulation, not heart-
beat.79,80 Legal statutes specify irreversible cessation of
circulatory and respiratory functions. However, the accepted
medical standard is permanent cessation because physicians
recognize that irreversible cessation inevitably and swiftly
follows once circulation no longer restores itself spontane-
ously and cannot be restored medically (i.e., permanent is a
surrogate for irreversibility). The revised Uniform Anatomi-
cal Gift Act of 2006 provides protection to those involved in
organ procurement. However, the courts have not provided
an opinion on whether health care providers may be held
liable for injuries that arise from the determination of
death.81 Interdisciplinary teamwork that includes nurses and
other clinicians in the ICU, palliative care, and the local
organ procurement organization, among others, is necessary
for organ donation in non–heart-beating donation and do-
nation after cardiac death. Debate or controversy about
death criteria can be healthy in a purely philosophical sense
but in health care can be harmful, for example, creating dis-
tress for donor families and health care staff, and so institu-
tional standards are necessary to ensure separation of death
criteria and organ donation.82
Family Discussions
The family should be told clearly that the patient is dead by
brain criteria when these rules are followed; subsequent
removal of the ventilator is not “termination of life support”
but rather recognizing that death has occurred. In general,
dead patients are not ventilated. How families respond to
brain death or organ donation is beyond the scope of this
chapter, and the reader is referred to other narratives.83
Brain Death and Organ Donation
One of the most important aspects to establishing clear crite-
ria for brain death is organ donation, although the number of
organ donors varies greatly across nations. In many U.S. states,
it is mandatory to raise the possibility of organ donation with
the family of a patient who may be imminently brain dead.
Transplant coordinators help with the subsequent steps, and
this along with an institutional multidisciplinary organ dona-
tion approach or even countrywide efforts can help increase
organ procurement.84-86 Current belief is that it is better for
the treating intensivist, neurosurgeon, or neurologist not to be
involved directly in the solicitation of organs. Their role is
simply to establish that brain death is imminent. Uniform
criteria that define imminent brain death and hence potential
organ donation are not defined in detail but in general will
include a mechanically ventilated deeply comatose patient,
with irreversible catastrophic brain injury with a clear etiology
of known origin, a Glasgow Coma Score of 3, and the progres-
sive absence of at least three out of six brainstem reflexes or a
FOUR score of zero.87
There remains a chronic shortage of organ donors. Family
consent represents one limiting factor for successful dona-
tion; less than two thirds of families agree to organ donation.
Associated factors in the United States include family
members of minority populations, older potential donors,
medical brain deaths, and a longer duration between brain
death declaration and discussion about organ donation.88 In
other countries, family refusal for organ donation or low
number of organ donations may include denial and rejection
of brain death, religious beliefs or fear about organ trade,89
and small hospital size.90 General public awareness is
required to support organ and tissue transplantation.74 In
addition, structured education and instruction on definitions
of brain and cardiac death and organ donation policies are
necessary in medical and nursing schools and even in sub-
sequent training.91-93
The care of the brain dead individual between declaration
of death and organ procurement and of donation after cardiac
death is beyond the scope of this chapter. There are however
 Section II—Clinical and Laboratory Assessment 119
many factors that become important to organ procurement
and subsequent recipient survival.94-99 The transplant team
rather then the neurointensivist is better equipped to address
this period of care, but the neurointensivist should be aware
of issues such as ventilatory status, inflammation, hemody-
namic instability, endocrine and metabolic disturbances, and
altered internal homeostasis that may jeopardize organs that
could be donated. Finally, there are rare circumstances that
lead to maternal brain death during pregnancy. Management
in this circumstance requires a multidisciplinary approach
that follows current guidelines and recommendations for the
fetus and for organ preservation of the potential donor.100
Withdrawal of Life Support
Although a separate issue, withdrawal of life-sustaining mea-
sures in the individual with catastrophic neurologic injury is
in part interwoven with the concept of brain death because
this decision incorporates “irreversibility” and “permanence.”
In addition, some of these patients may be suitable for organ
donation after cardiac death protocols. When care is with-
drawn in ventilated and comatose patients, up to half will die
within 60 minutes, usually from cardiac arrest. Factors such
as absent corneal and cough reflexes, extensor or absent motor
response, and poor oxygenation are associated with earlier
death.101 This information can be used to guide family and
health care provider expectations. In these patients, the pres-
ence and severity of pain cannot be assessed and so palliative
analgesia and sedation are necessary. Palliative sedation can be
complex and requires consideration about whether the medi-
cation contributes to covert euthanasia or acceleration of
death.
References
1. Manninen BA. Defining human death: an intersection of bioethics and
metaphysics. Rev Neurosci 2009;20(3-4):283–92.
2. Ball CG, Navsaria P, Kirkpatrick AW, et al. The impact of country and
culture on end-of-life care for injured patients: results from an international
survey. J Trauma 2010;69(6):1323–34.
3. Practice parameters for determining brain death in adults: summary
statement. Report of the quality Standards Subcommittee of the American
Academy of Neurology. In: Practice handbook: American Academy of
Neurology. St Paul, Minn: American Academy of Neurology; 1994.
4. A definition of irreversible coma: report of the Ad Hoc Committee of the
Harvard Medical School to Examine the Definition of the brain death.
JAMA 1968;205:337–40.
5. Burkle CM, Schipper AM, Wijdicks EF. Brain death and the courts.
Neurology 2011;76(9):837–41.
6. Guidelines for the determination of death: report of the medical
consultation on the diagnosis of death to the President’s Commission for
the Study of Ethical Problems in Medicine and Biomedical and Behavioral
Research. JAMA 1981;246:2184–6.
7. The Quality Standards Subcommittee of the American Academy of
Neurology. Practice parameters for determining brain death in adults
(summary statement). Neurology 1995;45:1012–14.
8. Wijdicks EF, Varelas PN, Gronseth GS, et al, American Academy of
Neurology. Evidence-based guideline update: determining brain death in
adults: report of the Quality Standards Subcommittee of the American
Academy of Neurology. Neurology 2010;74(23):1911–18.
9. Busl KM, Greer DM. Pitfalls in the diagnosis of brain death. Neurocrit Care
2009;11(2):276–87.
10. John J, Gane BD, Plakkal N, et al. Snake bite mimicking brain death. Case J
2008;1(1):16.
11. Miller RK. Fisher syndrome, a brainstem encephalitis, mimics brain death.
Clin Pediatr (Phila) 1993;32(11):685–7.
12. Singounas EG. Glasgow coma scale and brain death: a proposal. Acta
Neurochir (Wein) 1995;133(1-2):60.
13. Wijdicks EF, Bamlet WR, Maramattom BV, et al. Validation of a new coma
scale: the FOUR score. Ann Neurol 2005;58(4):585–93.
14. Jain S, DeGeorgia M. Brain death associated reflexes and automatism.
Neurocrit Care 2005;3(2):122–6.
15. Han SG, Kim GM, Lee KH, et al. Reflex movements in patients with brain
death: a prospective study in tertiary medical center. J Korean Med
Sci.2006;21(3):588–90.
16. Saposnik G, Maurino J, Saizar R. Facial myokymia in brain death. Eur J
Neurol 2001;8(3):227–30.
17. Saposnik G, Maurino J, Saizar R, et al. Undulating toe movements in brain
death. Eur J Neurol 2004;11(11):723–7.
18. Van den bent MJ, Ronday M, Oosterlee A. Unexpected movements in a brain
dead patient. Ned Tijdschr Geneeskd 1993;137(51):2653–5.
19. Ishiguro T, Tamagawa S, Ogawa H. Changes of pupil size in brain death
patient. Seishin Shinkeigaku Zasshi 1992;94(9):864–73.
20. Ishikawa S. Pupil recent development of investigation. Rinsho Shinkeigaku
1990;30(12):1329–33.
21. Shlugman D, Parulekar M, Elston JS, et al. Abnormal pupil activity in a brain
stem–dead patient. Br J Anaesth 2001;86(5):717–20.
22. Wang MY, Wallace P, Gruen JP. Brain death documentation: analysis and
issues. Neurosurgery 2002;15(3):731–6.
23. Thomke F, Weilmann LS. Current concepts in diagnosing brain death in
Germany. Med Klin (Munich) 2000;95(2):85–9.
24. Nathanson M, Bergman PS, Anderson PJ. Significance of oculocephalic
and caloric response in the unconscious patient. Neurology 1957;7(12):
829–32.
25. Sharpe MD, Young GB, Harris C. The apnea test for brain death
determination: an alternative approach. Neurocrit Care 2004;1(3):363–6.
26. Wu XL, Fang Q, Li L, et al. Complication associated with the apnea test
in the determination of the brain death. Chin Med J (Engl) 2008;121(13):
1169–72.
27. Wijdicks EF, Rabinstein AA, Manno EM, et al. Pronouncing brain death:
contemporary practice and safety of the apnea test. Neurology 2008;71(16):
1240–4.
28. Yee AH, Mandrekar J, Rabinstein AA, et al. Predictors of apnea test failure
during brain death determination. Neurocrit Care 2010;12(3):352–5.
29. Conci F, Di Rienzo M, Castiglioni P. Blood pressure and heart variability and
baroreflex sensitivity before and after brain death. J Neurosurg Psychiatry
2001;71(5):621–31.
30. Freitas J, Puig J, Rocha AP, et al. Heart rate variability in brain death. Clin
Auton Res 1996;6(3):141–6.
31. Knuttgen D, Zur Nieden K, Muller-Georges MR, et al. Computer aided
analysis of heart rate variability in brain death. Int J Clin Monit Comput
1997;14(1):37–42.
32. Ashwal S, Scheneider S. Brain death in the newborn. Pediatrics 1989;84(3):
429–37.
33. Villani A, Onofri A, Bianchi R, et al. Determination of brain death in
intensive pediatric therapy. Pediatr Med Chir 1998;20(1):19–23.
34. Parker BL, Frewen TC, Levine SD, et al. Declaring pediatric brain death:
current practice in a Canadian pediatric critical care unit. CMAJ 1995;
153(7):909–16.
35. Religious, cultural, legal, and ethical considerations. Crit Care Med 1993;
21(12):1951–65.
36. Schimizu N, Shemie S, Miyasaka E, et al. Preliminary report: use of clinical
criteria for the determination of pediatric brain death and confirmation by
radionuclide cerebral blood flow. Masi 2000;49(10):1126–32.
37. Lustbader D, O’Hara D, Wijdicks EF, et al. Second brain death examination
may negatively affect organ donation. Neurology 2011;76(2):119–24.
38. Nijboer WN, Moers C, Leuvenink HG, et al. How important is the duration
of the brain death period for the outcome in kidney transplantation? Transpl
Int 2011;24(1):14–20.
39. Varelas PN, Rehman M, Abdelhak T, et al. Single brain death examination
is equivalent to dual brain death examinations. Neurocrit Care 2011;15(3):
547–53.
40. Roberts DJ, MacCulloch KA, Versnick EJ, et al. Should ancillary brain blood
flow analyses play a larger role in the neurological determination of death?
Can J Anaesth 2010;57(10):927–35.
41. Wijdicks EF. The case against confirmatory tests for determining brain death
in adults. Neurology 2010;75(1):77–83.
42. Young GB, Lee D. A critique of ancillary tests for brain death. Neurocrit Care
2004;1(4):499–508.
43. Young GB, Shemie SD, Doig J, et al. Brief review: the role of ancillary
tests in the neurological determination of death. Can Anaesth 2006;53(6):
540–3.
44. Wijdicks EFM. Determining brain death in adults. Neurology 1995;45:
1003–11.
120 Section II—Clinical and Laboratory Assessment
45. Alvarez LA, Lipto RB, Hirschfeld A, et al. Brain death determination by
angiography in the setting of a skull defect. Arch Neurol 1988;45(2):225–7.
46. Sharma D, Souter MJ, Moore AE, et al. Clinical experience with transcranial
Doppler ultrasonography as a confirmatory test for brain death: a
retrospective analysis. Neurocrit Care 2011;14(3):370–6.
47. Poularas J, Karakitsos D, Kouraklis G, et al. Comparison between
transcranial color Doppler ultrasonography and angiography in the
confirmation of brain death. Transplant Proc 2006;38(5):1213–17.
48. Valentin A, Karnik R, Winkler WB, et al. Transcranial Doppler for early
identification of potential organ transplant donors. Wien Klin Wochenschr
1997;109(21):836–9.
49. Kuo JR, Chen CF, Chico CC, et al. Time dependant validity in the diagnosis
of brain death using transcranial Doppler sonography. J Neurolsurg
Psychiatry 2006;77(5):646–9.
50. Calleja S,Tembl JI, Sequra T. Sociedad Espanola Neurosonologia.
Recommendations of the use of transcranial Doppler to determine the
existence of cerebral circulatory arrest as diagnostic support of brain death.
Neurolgia 2007;22(7):441–7.
A complete list of references for this chapter can be found online at
www.expertconsult.com.
 Section II—Clinical and Laboratory Assessment 120.e1
References
1. Manninen BA. Defining human death: an intersection of bioethics and
metaphysics. Rev Neurosci 2009;20(3-4):283–92.
2. Ball CG, Navsaria P, Kirkpatrick AW, et al. The impact of country and
culture on end-of-life care for injured patients: results from an
international survey. J Trauma 2010;69(6):1323–34.
3. Practice parameters for determining brain death in adults: summary
statement. Report of the quality Standards Subcommittee of the American
Academy of Neurology. In: Practice handbook: American Academy of
Neurology. St Paul, Minn: American Academy of Neurology; 1994.
4. A definition of irreversible coma: report of the Ad Hoc Committee of the
Harvard Medical School to Examine the Definition of the brain death.
JAMA 1968;205:337–40.
5. Burkle CM, Schipper AM, Wijdicks EF. Brain death and the courts.
Neurology 2011;76(9):837–41.
6. Guidelines for the determination of death: report of the medical
consultation on the diagnosis of death to the President’s Commission for
the Study of Ethical Problems in Medicine and Biomedical and Behavioral
Research. JAMA 1981;246:2184–6.
7. The Quality Standards Subcommittee of the American Academy of
Neurology. Practice parameters for determining brain death in adults
(summary statement). Neurology 1995;45:1012–4.
8. Wijdicks EF, Varelas PN, Gronseth GS, et al. American Academy of
Neurology. Evidence-based guideline update: determining brain death in
adults: report of the Quality Standards Subcommittee of the American
Academy of Neurology. Neurology 2010;74(23):1911–18.
9. Busl KM, Greer DM. Pitfalls in the diagnosis of brain death. Neurocrit Care
2009;11(2):276–87.
10. John J, Gane BD, Plakkal N, et al. Snake bite mimicking brain death. Case J
2008;1(1):16.
11. Miller RK. Fisher syndrome, a brainstem encephalitis, mimics brain death.
Clin Pediatr (Phila) 1993;32(11):685–7.
12. Singounas EG. Glasgow coma scale and brain death: a proposal. Acta
Neurochir (Wein) 1995;133(1-2):60.
13. Wijdicks EF, Bamlet WR, Maramattom BV, et al. Validation of a new coma
scale: the FOUR score. Ann Neurol 2005;58(4):585–93.
14. Jain S, DeGeorgia M. Brain death associated reflexes and automatism.
Neurocrit Care 2005;3(2):122–6.
15. Han SG, Kim GM, Lee KH, et al. Reflex movements in patients with brain
death: a prospective study in tertiary medical center. J Korean Med Sci
2006;21(3):588–90.
16. Saposnik G, Maurino J, Saizar R. Facial myokymia in brain death. Eur J
Neurol 2001;8(3):227–30.
17. Saposnik G, Maurino J, Saizar R, et al. Undulating toe movements in brain
death. Eur J Neurol 2004;11(11):723–7.
18. Van den bent MJ, Ronday M, Oosterlee A. Unexpected movements in a
brain dead patient. Ned Tijdschr Geneeskd 1993;137(51):2653–5.
19. Ishiguro T, Tamagawa S, Ogawa H. Changes of pupil size in brain death
patient. Seishin Shinkeigaku Zasshi 1992;94(9):864–73.
20. Ishikawa S. Pupil recent development of investigation. Rinsho Shinkeigaku
1990;30(12):1329–33.
21. Shlugman D, Parulekar M, Elston JS, et al. Abnormal pupil activity in a
brain stem–dead patient. Br J Anaesth 2001;86(5):717–20.
22. Wang MY, Wallace P, Gruen JP. Brain death documentation: analysis and
issues. Neurosurgery 2002;15(3):731–6.
23. Thomke F, Weilmann LS. Current concepts in diagnosing brain death in
Germany. Med Klin (Munich) 2000;95(2):85–9.
24. Nathanson M, Bergman PS, Anderson PJ. Significance of oculocephalic
and caloric response in the unconscious patient. Neurology 1957;7(12):
829–32.
25. Sharpe MD, Young GB, Harris C. The apnea test for brain death
determination: an alternative approach. Neurocrit Care 2004;1(3):363–6.
26. Wu XL, Fang Q, Li L, et al. Complication associated with the apnea test
in the determination of the brain death. Chin Med J (Engl) 2008;
121(13):1169–72.
27. Wijdicks EF, Rabinstein AA, Manno EM, et al. Pronouncing brain death:
contemporary practice and safety of the apnea test. Neurology
2008;71(16):1240–4.
28. Yee AH, Mandrekar J, Rabinstein AA, et al. Predictors of apnea test failure
during brain death determination. Neurocrit Care 2010;12(3):352–5.
29. Conci F, Di Rienzo M, Castiglioni P. Blood pressure and heart variability
and baroreflex sensitivity before and after brain death. J Neurosurg
Psychiatry 2001;71(5):621–31.
30. Freitas J, Puig J, Rocha AP, et al. Heart rate variability in brain death. Clin
Auton Res 1996;6(3):141–6.
31. Knuttgen D, Zur Nieden K, Muller-Georges MR, et al. Computer aided
analysis of heart rate variability in brain death. Int J Clin Monit Comput
1997;14(1):37–42.
32. Ashwal S, Scheneider S. Brain death in the newborn. Pediatrics
1989;84(3):429–37.
33. Villani A, Onofri A, Bianchi R, et al. Determination of brain death in
intensive pediatric therapy. Pediatr Med Chir 1998;20(1):19–23.
34. Parker BL, Frewen TC, Levine SD, et al. Declaring pediatric brain death:
current practice in a Canadian pediatric critical care unit. CMAJ
1995;153(7):909–16.
35. Religious, cultural, legal, and ethical considerations. Crit Care Med
1993;21(12):1951–65.
36. Schimizu N, Shemie S, Miyasaka E, et al. Preliminary report: use of clinical
criteria for the determination of pediatric brain death and confirmation by
radionuclide cerebral blood flow. Masi 2000;49(10):1126–32.
37. Lustbader D, O’Hara D, Wijdicks EF, et al. Second brain death examination
may negatively affect organ donation. Neurology 2011;76(2):119–24.
38. Nijboer WN, Moers C, Leuvenink HG, et al. How important is the duration
of the brain death period for the outcome in kidney transplantation?
Transpl Int 2011;24(1):14–20.
39. Varelas PN, Rehman M, Abdelhak T, et al. Single brain death examination
is equivalent to dual brain death examinations. Neurocrit Care
2011;15(3):547–53.
40. Roberts DJ, MacCulloch KA, Versnick EJ, et al. Should ancillary brain blood
flow analyses play a larger role in the neurological determination of death?
Can J Anaesth 2010;57(10):927–35.
41. Wijdicks EF. The case against confirmatory tests for determining brain
death in adults. Neurology 2010;75(1):77–83.
42. Young GB, Lee D. A critique of ancillary tests for brain death. Neurocrit
Care 2004;1(4):499–508.
43. Young GB, Shemie SD, Doig J, et al. Brief review: the role of ancillary tests
in the neurological determination of death. Can Anaesth 2006;53(6):540–3.
44. Wijdicks EFM. Determining brain death in adults. Neurology 1995;45:
1003–11.
45. Alvarez LA, Lipto RB, Hirschfeld A, et al. Brain death determination by
angiography in the setting of a skull defect. Arch Neurol 1988;45(2):225–7.
46. Sharma D, Souter MJ, Moore AE, et al. Clinical experience with transcranial
Doppler ultrasonography as a confirmatory test for brain death: a
retrospective analysis. Neurocrit Care 2011;14(3):370–6.
47. Poularas J, Karakitsos D, Kouraklis G, et al. Comparison between
transcranial color Doppler ultrasonography and angiography in the
confirmation of brain death. Transplant Proc 2006;38(5):1213–17.
48. Valentin A, Karnik R, Winkler WB, et al. Transcranial Doppler for early
identification of potential organ transplant donors. Wien Klin Wochenschr
1997;109(21):836–9.
49. Kuo JR, Chen CF, Chico CC, et al. Time dependant validity in the diagnosis
of brain death using transcranial Doppler sonography. J Neurolsurg
Psychiatry 2006;77(5):646–9.
50. Calleja S,Tembl JI, Sequra T. Sociedad Espanola Neurosonologia.
Recommendations of the use of transcranial Doppler to determine the
existence of cerebral circulatory arrest as diagnostic support of brain death.
Neurolgia 2007;22(7):441–7.
51. Burger R. Value of transcranial Doppler ultrasonography compared with
scintigraphic techniques and EEG in brain death. Zentrailbl Neurochir
2000;61(10):7–13.
52. Bertagna F, Barozzi O, Puta E, et al. Residual brain viability, evaluated by
(99m)Tc-ECD SPECT, in patients with suspected brain death and with
confounding clinical factors. Nucl Med Commun 2009;30(10):815–21.
53. Conrad GR, Sinha P. Scintography as a confirmatory test of brain death.
Semin Nucl Med 2003;33(4):312–23.
54. Kurtek RW, Lia KK, Tauxe WN, et al. Tc 99 hexamethylpropylene amino
oxime scintography in the diagnosis of brain death and its implications
in the harvesting of organs used for transplant. Clin Nucl Med 2000;
25(1):7–10.
55. Pistoia F, Johnson DW, Darby JM, et al. The role of xenon CT
measurements of cerebral blood flow in the clinical determination of brain
death. AJNR Am J Neurodial 1991;12(1):97–103.
56. Figaji AA, Kent SJ. Brain tissue oxygenation in children diagnosed with
brain death. Neurocrit Care 2010;12(1):56–61.
57. Smith ML, Counelis GJ, Maloney-Welinsky E, et al. Brain tissue oxygen
tension in clinical brain death: a case series. Neurol Res 2007;29(7):755–9.
58. Palmer S, Badar MK. Brain tissue oxygenation in brain death. Neurocrit
Care 2005;2(1):17–22.
59. Bohatyrewicz R, Sawicki M, Walecka A, et al. Computed tomographic
angiography and perfusion in the diagnosis of brain death. Transplant Proc
2010;42(10):3941–6.
120.e2 Section II—Clinical and Laboratory Assessment
60. Escudero D, Otero J, Marqués L, et al. Diagnosing brain death by CT
perfusion and multislice CT angiography. Neurocrit Care 2009;11(2):
261–71.
61. Berenguer CM, Davis FE, Howington JU. Brain death confirmation:
comparison of computed tomographic angiography with nuclear medicine
perfusion scan. J Trauma 2010;68(3):553–9.
62. Rimmele T, Malhicheikh A, Boselli E, et al. The electroencephalogram is
not an adequate test to confirm the diagnosis of brain death. Can J Anaesth
2008;55(3):188–90.
63. Sethi NK, Sethi PK, Torgounick J. EMG artifact in brain death
electroencephalogram, is it a cry of medullary death? Clin Neurol
Neurosurg 2008;110(7):729–31.
64. Guerit JM. Evoked potential, a safe brain death confirmatory tool? Eur J
Med 1992;1(4):233–43.
65. Wagner W. Scalp, ear lobe and nasopharyngeal recording of the median
nerve somatosensory evoked P14 potential in coma and brain death patient
and amplitude analysis. Brain 1996;119(Pt 5):1507–21.
66. Jarddim M. Brainstem auditory evoked potential as a model to assist
diagnosis of brain death. Pro Fono 2008;20(2):123–8.
67. Bacigalupo F. The debate about death: an imperishable discussion. Biol Res
2007;40(4):523–34.
68. Nojszewska M. Determination of brain death. Neurol Neurochir Pol
2002;36(1):91–104.
69. Boobes Y, Al Daker N. What it means to die in Islam and modern medicine.
Saudi J Kidney Dis Transpl 1996;7(2):121–7.
70. Bryne PA, O’Reilly S, Quary PM. Brain death: an opposing viewpoint.
JAMA 1979;242(18):1985–90.
71. Trueba J. Clinical death, a diagnosis and a testimony. An Sist Sanit Nava
2007;30(Suppl 3):57–70.
72. Aker J, Pearson KS. AANA Journal Course. Update for nurse anesthetists
brain death: terminology, clinical criteria, and diagnostic testing. AANA J
1997;65(2):129–35.
73. Truog RD, Fackler JC. Rethinking brain death. Crit Care Med
1992(12):1705–13.
74. Walker AE, Diamond EL, Moseley J. The neuropathological findings in
irreversible coma: a critique of the respirator. J Neuropathol Exp Neurol
1975;34(4):295–323.
75. Machado-Curbelo C. A new formulation of death, definition, criteria and
diagnostic tests. Rev Neurol 1998;26(154):1040–7.
76. Collins M. Reevaluating the dead donor rule. J Med Philos 2010;35(2):
154–79.
77. Cavanna AE, Cavanna SL, Servo S, et al. The neural correlates of impaired
consciousness in coma and unresponsive states. Discov Med
2010;9(48):431–8.
78. Miller FG, Truog RD. Decapitation and the definition of death. J Med
Ethics 2010;36(10):632–4.
79. Souter M, Van Norman G. Ethical controversies at end of life after
traumatic brain injury: defining death and organ donation. Crit Care Med
2010;38(9 Suppl):S502–9.
80. Thomas AG. Continuing the definition of death debate: the report of the
President’s Council on Bioethics on Controversies in the Determination of
Death. Bioethics 2012;26(2):101–7.
81. Veatch RM. Transplanting hearts after death measured by cardiac criteria:
the challenge to the dead donor rule. J Med Philos 2010;35(3):313–29.
82. Verheijde JL, Rady MY, McGregor J. Presumed consent for organ
preservation in uncontrolled donation after cardiac death in the United
States: a public policy with serious consequences. Philos Ethics Humanit
Med 2009;4:15.
83. Dubois JM. The ethics of creating and responding to doubts about death
criteria. J Med Philos 2010;35(3):365–80.
84. Manuel A, Solberg S, MacDonald S. Organ donation experiences of family
members. Nephrol Nurs J 2010;37(3):229–37.
85. Kong AP, Barrios C, Salim A, et al. A multidisciplinary organ donor council
and performance improvement initiative can improve donation outcomes.
Am Surg 2010;76(10):1059–62.
86. Bleakley G. Implementing minimum notification criteria for organ donation
in an acute hospital’s critical care units. Nurs Crit Care 2010;15(4):185–91.
87. Matesanz R, Marazuela R, Dominguez-Gil B, et al. The 40 donors per
million population plan: an action plan for improvement of organ
donation and transplantation in Spain. Transplant Proc 2009;41(8):3453–6.
88. de Groot YJ, Jansen NE, Bakker J, et al. Imminent brain death: point of
departure for potential heart-beating organ donor recognition. Intensive
Care Med 2010;36(9):1488–94.
89. Brown CV, Foulkrod KH, Dworaczyk S, et al. Barriers to obtaining family
consent for potential organ donors. J Trauma 2010;68(2):447–51.
90. Ghorbani F, Khoddami-Vishteh HR, et al. Causes of family refusal for
organ donation. Transplant Proc 2011;43(2):405–6.
91. Formanek M, Schöffski O. Difficulties with the organ donation process in
small hospitals in Germany. Transplant Proc 2010;42(5):1445–8.
92. Anker AE, Feeley TH, Friedman E, et al. Teaching organ and tissue
donation in medical and nursing education: a needs assessment. Prog
Transplant 2009;19(4):343–8.
93. Lima CX, Lima MV, Cerqueira RG, et al. Organ donation: cross-sectional
survey of knowledge and personal views of Brazilian medical students and
physicians. Transplant Proc 2010;42(5):1466–71.
94. Rykhoff ME, Coupland C, Dionne J, et al. A clinical group’s attempt to raise
awareness of organ and tissue donation. Prog Transplant 2010;20(1):33–9.
95. Murugan R, Venkataraman R, Wahed AS, et al, HIDonOR Study
Investigators. Increased plasma interleukin-6 in donors is associated with
lower recipient hospital-free survival after cadaveric organ transplantation.
Crit Care Med 2008;36(6):1810–16.
96. Dictus C, Vienenkoetter B, Esmaeilzadeh M, et al. Critical care management
of potential organ donors: our current standard. Clin Transplant
2009;23(Suppl 21):2–9.
97. Mascia L, Pasero D, Slutsky AS, et al. Effect of a lung protective strategy
for organ donors on eligibility and availability of lungs for transplantation:
a randomized controlled trial. JAMA 2010;304(23):2620–7.
98. Dubbeld J, van Hoek B, Ringers J. Use of a liver from donor after cardiac
death: is it appropriate for the sick or the stable? Curr Opin Organ
Transplant 2011;16(2):239–42.
99. Frontera JA. How I manage the adult potential organ donor: donation after
cardiac death (part 2). Neurocrit Care 2010;12(1):111–16.
100. Esmaeilzadeh M, Dictus C, Kayvanpour E, et al. One life ends, another
begins: management of a brain-dead pregnant mother—a systematic
review. BMC Med 2010;8:74.
101. Yee AH, Rabinstein AA, Thapa P, et al. Factors influencing time to death
after withdrawal of life support in neurocritical patients. Neurology
2010;74(17):1380–5.

Glucose and Nutrition
Sarice L. Bassin and Thomas P. Bleck
Introduction
Hyperglycemia is common in acutely ill patients. In the past,
stress hyperglycemia was considered to be a beneficial adaptive
response. However, prospective observational and clinical
trial data show a consistent and clear association between
hyperglycemia and increased morbidity and mortality in all
critically ill populations and in particular those with acute
brain injury, stroke, and subarachnoid hemorrhage. Glycemic
control therefore has become an important aspect of critical
care. However, the optimal level of glycemic control in critical
illness remains in debate since Van den Berghe et al. published
their study conclusions in 2001 showing that strict glucose
control was associated with reduced mortality.1,2 In these ran-
domized trials intensive insulin therapy (IIT) was used to
target a blood glucose of 80 to 110 mg/dL. Studies since then
have not been able to duplicate these results (Normoglycemia
in Intensive Care Evaluation-Survival Using Glucose Algo-
rithm Regulation [NICE-SUGAR]),3 and so a more liberal
glucose target of 140 to 180 mg/dL is favored. In addition, an
association between hypoglycemia that can be caused by IIT,
and high glycemic variability and poor outcome also is appar-
ent, further emphasizing the importance of glucose control.4-7
The safety and efficacy of IIT may be influenced by patient-
related and intensive care unit–related variables. In particular,
nutrition is closely associated with glycemic control.8,9 Fur-
thermore, guidelines recommend use of early nutritional
support, preferably enteral nutrition. However, optimal energy
requirements and the timetable of nutritional support still are
not fully resolved, in part because energy expenditure needs
may not be met with enteral nutrition only, and there are
different recommendations for parental nutrition in Europe
and North America.10-15 This chapter addresses the impor-
tance of glucose control: how to monitor glucose levels in ICU
patients, practical aspects of glucose control in ICUs, nutrition
therapy, and methods to calculate nutritional requirements in
critical care.
Glucose
Plasma glucose usually is between 60 and 100 mg/dL before
meals and up to 150 mg/dL after a meal. Hypoglycemia is
defined as a blood glucose less than 50 mg/dL. Several factors
such as malnutrition, cirrhosis, renal failure, and alcoholic
ketoacidosis or drugs such as pentamidine, haloperidol,
bactrim, or salicylates increase the risk of developing hypogly-
cemia. Traditionally, acute hyperglycemia was defined as a
random glucose concentration greater than 200 mg/dL. In
© Copyright 2013 Elsevier Inc. All rights reserved.
Chapter
14  
II
2010, the American Diabetes Association (ADA) proposed a
threshold of 140 mg/dL (7.8 mmol/L; ADA Standards of
Medical Care in Diabetes–2010 http://care.diabetesjournals.
org/content/33/Supplement_1/S11.full). Hyperglycemia may
be considered as mild (110-180 mg/dL), moderate (180-
400 mg/dL) or severe (>400 mg/dL). Table 14.1 lists the
diagnostic criteria for diabetic ketoacidosis or hyperos-
molar hyperglycemic state. In the hospitalized patient stress
and medication (e.g., steroids) are factors associated with
elevation of blood glucose. In addition, there can be relative
insulin resistance after administration of catecholamines or
glucocorticoids.
Assessment of Glucose in the Intensive
Care Unit
Blood to measure glucose can be obtained from a variety of
sites (e.g., a fingerstick device, venipuncture, or through an
intra-arterial or venous catheter). Fingersticks may be inac-
curate when there is edema, hypoperfusion, or anemia,
whereas capillary blood glucose monitoring may not be accu-
rate in shock.16 In addition, care to prevent contamination
by intravenous solutions is needed when using samples from
arterial or venous catheters because pseudohyperglycemia
may be present if a blood sample is drawn from a line where
the patient is receiving dextrose 5% in water (D5W) or total
parenteral nutrition (TPN). Laboratory analysis of plasma
is the best means to analyze blood glucose levels. However,
this technique may be too slow for care of critically ill patients.
Many hospitals now have laboratories operated by licensed
medical technologists in, or close to, the ICU that allow
rapid assessment of blood gases, electrolytes, hemoglobin,
prothrombin time (PT), and glucose through a blood gas
analyzer or “stat” platform (e.g., i-STAT [Abbott Point of
Care, Princeton, NJ], Radiometer ABL 825, or RapidLab
865).17 These devices generally are very accurate and can be
used interchangeably with core laboratory devices,18 but turn-
around time can still be a limiting factor.
Point-of-Care Testing
Glucose testing using point-of-care (POC) devices has become
increasingly common in hospital and ICU settings as well as
in the operating room and in the outpatient environment.19
In addition, there are many different self-monitoring blood
glucose (SMBG) devices available for patients to use, for
example, Accu-Chek Active and Accu-Chek Performa (Roche
Diagnostics, Indianapolis, IN), OneTouch Ultra2 (LifeScan
121
122 Section II—Clinical and Laboratory Assessment
Table 14.1  Diagnostic Criteria for Diabetic Ketoacidosis and Hyperosmolar Hyperglycemic State
Criteria Diabetic Ketoacidosis
Mild Moderate
Plasma glucose (mg/dL) >250 >250
Arterial pH 7.25-7.30 7.00-7.24
Serum bicarbonate (mEq/L) 15-18 10 to <15
Urine ketones Positive Positive
Serum ketones Positive Positive
Effective serum osmolality* Variable Variable
Anion gap †
>10 >12
Level of consciousness Alert Alert/drowsy
Adapted from Kitabachi AE, Umpierrez GE, Murphy MB, et al. Hyperglycemic crisis in adult patients with diabetes: a consensus statement from the American
Diabetes Association. Diabetes Care 2006;29:2739–48.
*
Effective serum osmolality = 2[measured Na+] + glucose/18.
†
Anion gap = Na+ − (Cl− + HCO3−)
Inc., Milpitas, CA), MediSense Optium or Xceed (Abbott Dia-
betes Care Inc., Alameda, CA), and others. These devices all
are subject to strict regulatory standards (i.e., International
Organization for Standardization. In vitro diagnostic test
systems: requirements for blood glucose monitoring systems
for self-testing in managing diabetes mellitus. [ISO 15197:2003
http://www.iso.org/iso/home/store/catalogue_tc/catalogue_
detail.htm?csnumber=26309; accessed Oct. 19, 2012.)
Although POC devices are used at bedside, they are licensed
under laboratory medical directors and governed by several
regulatory bodies such as Health Care Financing Administra-
tion (HCFA), The Joint Commission, Clinical Laboratory
Improvements Amendments of 1988 (CLIA 88), the College
of American Pathologists (CAP), and individual state depart-
ments of health, hospital policies, and procedure manuals for
each device. This means use of the devices must follow regula-
tion, and there needs to be proficiency testing for accuracy and
testing of unknown samples provided by CAP or commercial
sources for comparison to reference range values to calibrate
and verify the device. Staff who use the devices also are
expected to receive annual competency testing. POC monitors
are reasonably accurate, particularly when blood from a cath-
eter is used. However, POC glucometers can be inaccurate
(20%) when patients have low blood glucose levels17 and can
differ from laboratory values20; not all achieve ISO 15197
accuracy criteria.21,22 This can limit their use when tight
glucose control is targeted.23 Consequently it may be better to
examine blood glucose levels using central lab or blood gas
analyzers rather than POC devices24 and when POC devices
are used careful clinical evaluation of each system for glucose
measurement is critical.
i-STAT Analyzer
This device can measure blood gases, electrolytes, glucose, and
certain coagulation parameters through single-use sensors
constructed using thin-film technology. Because the sensor
film is thin, it can be used immediately and unlike thick film
sensors does not require a calibration period. Waste, however,
is a potential disadvantage of use.
Hyperosmolar Hyperglycemic State
Severe
>250 >600
<7.00 >7.30
<10 >15
Positive Small
Positive Small
Variable >320
>12 Variable
Stupor/coma Stupor/coma
GEM 3000
The GEM 3000 (Instrumentation Laboratories, Bedford, MA)
has thick film sensors or electrodes in strips that can measure
glucose, lactate, and blood gases. The sensors contain reagents
and calibrators packed into small cartridges that are placed in
the body of the analyzer. The cartridges can test multiple
samples and the sensors are reusable. Samples with increased
osmolalities, high protein, or high lipid volumes adversely
affect the function. In addition, gallamine triethiodide,
ethanol, sodium fluoride, potassium oxalate, acetaminophen,
isoniazide, or thiocyanate may affect glucose or lactate levels.
Minimally Invasive Devices
Continuous monitoring sensors (e.g., CDI 500 Parameter
Monitoring System, Terumo Cardiovascular Systems, Ann
Arbor, MI) that are inserted into the bloodstream are available
but their benefit still needs to be elucidated. These monitors
based on optical technology are primarily used to measure
blood gases and most commonly have been used in cardiac
surgery. However, electrochemical sensors that can monitor
glucose are available.
Microdialysis
In 2009 the Eirus system (Dipylon, Solna, Sweden) was CE-
marked for continuous glucose and lactate monitoring in
blood, and in 2011, the Eirus TLC catheter received its CE
mark. This technology is based on microdialysis so no blood
is removed from the patient, and it does not require a separate
line because the device includes a central venous catheter. It
allows second-by-second monitoring of glucose and lactate.
The glucose value is presented on a monitor as a numeric
value and a trend curve is provided. Studies have verified its
technical feasibility, accuracy, and performance in the ICU.25,26
However, as of 2012 the device has not yet become commer-
cially available.
 Section II—Clinical and Laboratory Assessment 123
Why Is Glycemic Control Important in the
Neurocritical Care Unit?
In critical illness, release of stress hormones (e.g., epinephrine
and cortisol), medications such as exogenous glucocorticoids
or catecholamines, and release of inflammatory mediators
(e.g., in sepsis or trauma) all inhibit insulin release and action.
This results in stress hyperglycemia, defined as an increase in
blood glucose in the presence of acute illness,27 through
enhanced gluconeogenesis, inhibition of glycogen synthesis,
and impaired insulin-mediated glucose uptake by tissues. In
addition, intravenous dextrose (e.g., in parenteral nutrition or
in antibiotic solutions), can exacerbate or contribute to hyper-
glycemia. Overall, hyperglycemia may be present in 40% of
critically ill patients; 80% of ICU patients with hyperglycemia
have no history of diabetes before admission.28
Hyperglycemia during critical illness is associated with
adverse outcome and an increased risk for infection, myocar-
dial infarction, and neurologic and renal damage, particularly
in patients who are not diabetic before admission.29 The effect
of hyperglycemia also depends in part on patient pathology
(i.e., ICU admission diagnosis).30 However, the impact of
hyperglycemia on patients with preexisting diabetes is less
clear even though diabetic patients are at greater risk for com-
plications.31 In fact, in diabetic patients with elevated hemo-
globin A1c (HbA1c) levels (>7%), glucose levels that are
desirable in other patients appear to be “unsafe” in diabetic
patients with chronic hyperglycemia.32 Although the likeli-
hood of hyperglycemia is greater in sicker patients, it appears
that hyperglycemia itself is an independent factor associated
with outcome particularly in stroke, both hemorrhagic and
ischemic, and acute coronary syndromes.33-40 In addition, sur-
vival is inversely associated with the duration of hyperglyce-
mia.29 However, the exact threshold at which elevated blood
glucose is deleterious remains uncertain. The relationship
between hyperglycemia and increased risk of worse outcome
has driven research into glucose control in the ICU, the value
and structure of which remain debated, in part because tight
glucose control is associated with an increased risk for hypo-
glycemia that in turn is associated with adverse outcome and
mortality. Furthermore, glucose variability also is associated
with an increased risk of poor outcome. For example, Cueni-
Villoz et al.7 observed that increased blood glucose variability
during therapeutic hypothermia was associated with in-hos-
pital mortality after cardiac arrest, independent of injury
severity and mean blood glucose levels.
Consistent evidence from experimental and clinical studies
demonstrates that hyperglycemia in the setting of acute neu-
rologic injury is associated with worse outcomes; this concept
is widely accepted in the management of critical neurologic
illness in adults as well as children.41,42 Hyperglycemia at
admission is present in almost half of acute stroke patients,
and is an independent risk factor for mortality at 30 days, 1
year, and 6 years after stroke.33 Baird et  al. found that serum
glucose greater than 130  mg/dL during the first 72 hours
after a stroke was associated with significant increase in infarct
size as measured by magnetic resonance imaging (MRI) 3 to
6 days after the ictus.34 Likewise, increased blood glucose is
a risk factor for death and disability after aneurysmal sub-
arachnoid hemorrhage (SAH).35,36,43 For example, McGirt
et  al. observed that patients with blood glucose greater than
200  mg/dL for 2 or more days were seven times more likely
to have a poor outcome at 10 months after SAH.37 Others
have found that mean glucose concentrations greater than
140  mg/dL are associated with the development of symp-
tomatic vasospasm after SAH.38,44 In traumatic brain injury
(TBI) hyperglycemia also is associated with decreased sur-
vival, higher intracranial hypertension, and increased length
of hospital stay.45,46
Whether hyperglycemia in the setting of acute neurologic
injury is simply an epiphenomenon or actually serves a pro-
tective purpose remains controversial. The stress response
associated with neurologic injury can lead to hyperglycemia
through liberation of catecholamines that induce gluconeo-
genesis and insulin resistance. This may be an adaptive and
protective response because glucose is the primary source of
energy in the brain. However, some have postulated that in the
setting of anaerobic metabolism, hyperglycemia will result in
the accumulation of lactate and an intracellular acidosis, a
potential mediator of secondary injury. Conversely, if the
supply of glucose, the main energy substrate of the brain, is
inadequate, there could be insufficient fuel for glycolysis in
already ischemic tissue. Termed the glucose paradox of cerebral
ischemia, this conundrum makes determination of an exact
target for plasma glucose concentrations in patients with
severe neurologic injury difficult.47 The conundrum is further
complicated by accumulating evidence that in some circum-
stances after acute brain injury lactate is used as a fuel in the
brain and may even be protective.48,49
Insulin Therapy in Critical Care
In 2001 and 2006 two single center randomized clinical
trials1,2 demonstrated the association between IIT and
improved outcomes in critically ill patients in surgical ICUs
(SICU) and medical ICUs (MICU). In these trials patients
were randomly assigned to IIT (target glucose 80-110 mg/dL
[4.4-6.1 mmol/L]) or standard care (target glucose 180-
200 mg/dL [9.9-11 mmol/L]). In SICU patients (n = 1548),
Van den Berghe et al. observed that an IIT protocol that
treated hyperglycemia by administration of intravenous infu-
sion insulin adjusted to maintain blood glucose between 80
and110 mg/dL was associated with reduced mortality during
intensive care (8% for conventional therapy vs. 4.6% for IIT)
especially among patients who were in the ICU more than 5
days. Moreover, IIT reduced in-hospital mortality (10.9% for
conventional care vs. 7.2% for IIT), bloodstream infection,
renal failure, critical-illness polyneuropathy, and transfu-
sions.1 The greatest reduction in mortality was observed for
deaths associated with multiple-organ failure with a proven
septic focus. A follow-up MICU study2 (n = 1200 patients)
showed IIT was associated with prevention of newly acquired
kidney injury, accelerated weaning from mechanical ventila-
tion, and quicker discharge from the ICU and the hospital. In
patients who received IIT, mortality was greater among those
who were in the ICU less than 3 days but significantly decreased
in those patients with MICU stays longer than 3 days.2 The
mechanism by which IIT lowered mortality remains poorly
understood and probably involves both glycemic and non­
glycemic pathways. The potential nonglycemic benefits of
insulin include myocardial protection, partial correction of
abnormal serum lipid profiles, and the prevention of excessive
inflammation.50
124 Section II—Clinical and Laboratory Assessment
Table 14.2  Guidelines for Glycemic Control in the Intensive Care Unit
Patient
Year Organization Population
2009 American Association of ICU patients
Clinical Endocrinologists
and American Diabetes
Association
2009 Surviving Sepsis Campaign ICU patients
2009 Institute for Healthcare ICU patients
Improvement
2008 American Heart Association Acute
coronary
syndromes
2007 European Society of ICU patients
Cardiology and European with
Association for the Study cardiac
of Diabetes disorders
Modified from Kavanagh BP, McCowen KC. Clinical practice: glycemic control in the ICU. N Engl J Med 2010;363(26):2540–6.
ICU, Intensive care unit.
However, other studies have not been able to duplicate the
results of Van den Berghe et al.1,2 and in a meta-analysis that
included 29 randomized trials with 8432 patients Wiener
et al.3 found that tight glucose control (IIT) increased the risk
of critical hypoglycemia in critically ill adult patients but did
not reduce hospital mortality. In turn, hypoglycemia increases
the risk of mortality.5 Although IIT was popular in the years
immediately before and after 2000, recommendations (Table
14.2) now target a blood glucose concentration between 140
and 180 mg/dL (8-10 mmol/L) according to NICE-SUGAR,
published in 2009.51 In this randomized clinical trial 6104
patients were randomized to IIT (target blood glucose
81-108 mg/dL (4.5-6 mmol/L), or conventional glucose
control (target <180 mg/dL [<10 mmol/L]). Ninety-day mor-
tality was significantly less in the conventional glucose control
group. In addition, severe hypoglycemia (<40 mg/dL) was rare
when using conventional glucose control (0.5% vs. 6.8% in
the IIT group). Other investigators also have observed that
hypoglycemia is rare using the current American Association
of Clinical Endocrinologists and American Diabetes Associa-
tion (AACE/ADA) consensus recommendations that use a
blood glucose target of 140 to 180 mg/dL in the ICU.52
However, there are differences, in particular glycemic control
superimposed on different nutritional strategies, between the
trials in Leuven1,2 and NICE-SUGAR that may limit direct
comparison. Indeed it is conceivable that the safety and effi-
cacy of IIT may be influenced by patient-related and ICU
setting–related variables.
Insulin Therapy in Neurocritical Care
The potential mechanisms by which IIT is theorized to benefit
the critically ill—attenuation of the cortisol stress response,
improved immune cell function, reduced endothelial dysfunc-
tion, correction of abnormal lipids, prevention of toxicity to
mitochondria, and reduced hyperglycemic axonal damage53—
may be relevant in patients suffering from acute neurologic
illness. In a preplanned subanalysis of SICU patients, 63 of
Updated Since
Treatment Target Glucose Definition of NICE-SUGAR
Threshold mg/dL Hypoglycemia Trial
180 140-180 <70 Yes
180 150 Not stated Yes
180 <180 <40 Yes
180 90-140 Not stated No
Not stated Strict (refers to Not stated No
80-110 mg/dL)
whom had isolated brain injury, Van den Berge et al. also
observed that IIT reduced both the incidence of critical illness
polyneuropathy and time on the ventilator.54 In the patients
with isolated brain injury, 1-year functional outcomes were
significantly improved in those patients in the IIT group,
despite the admission Glasgow Coma Scale (GCS) score being
higher in the conventional insulin therapy group. For MICU
patients who needed intensive care for at least 7 days, IIT
reduced the incidence of critical illness myopathy and neu-
ropathy and was protective against prolonged mechanical ven-
tilation.55 Subsequent studies suggest glycemic control can
help reduce critical illness polyneuropathy or myopathy,
although this may depend on the diagnostic criteria used.56
Several studies have focused on glucose control specifically
in patients with primary neurologic diseases (e.g., TBI and
stroke); overall it does not appear that IIT improves outcome,
whereas its use is associated with more frequent episodes of
hypoglycemia that in some patients aggravates outcome.57-64
The UK Glucose Insulin in Stroke Trial (GIST) randomized
patients with acute ischemic stroke to either conventional
insulin therapy or IIT (72 and 126 mg/dL); 90-day mortality
was similar in the two groups.57 Bilotta et al. randomized 78
SAH patients to conventional or IIT. The patients randomized
to the IIT group had a significant decrease in infection but no
difference in mortality or neurologic outcome.58 The same
researchers randomized 97 patients with severe TBI to either
conventional treatment for glucose greater than 220 mg/dL or
IIT for (80-120 mg/dL). The patients treated with IIT had
more episodes of hypoglycemia but rates of infection, poor
outcome, and mortality were similar in the two groups.59
Two randomized phase II trials of IIT for acute ischemic
stroke were completed in the United States. The Treatment of
Hyperglycemia in Ischemic Stroke trial aimed for glucose con-
centrations less than 130 mg/dL in the tight control group,
and less than 200 mg/L in the loose control group. In the
Glucose Regulation in Acute Stroke trial, the goals were 80 to
120 mg/dL and 80 to 180 mg/dL, respectively. In both, the
investigators were able to achieve their targets for tight glucose
 Section II—Clinical and Laboratory Assessment 125
control without significant hypoglycemia or other major
adverse events.65,66
Other data in primary neurologic injury raise concerns
about the effects of IIT on cerebral metabolism. Microdialysis
studies in patients with severe TBI or SAH suggest that tight
systemic glucose control is associated with reduced cerebral
extracellular glucose and increased risk of brain energy
crisis.67-71 In addition, Vespa et al. reported that low extracel-
lular cerebral glucose after TBI was associated with poor
outcomes, even in the absence of increased microdialysate
lactate.68 Similarly in patients with severe SAH, Oddo et al.70
observed that increased insulin dose independently predicted
brain energy crisis after adjustment for intracranial pressure
(ICP) and cerebral perfusion pressure (CPP) and that this in
turn was associated with increased mortality. In addition, the
relationship between brain interstitial and systemic glucose
concentrations is complex and may change when the systemic
glucose concentration is greater than 180 mg/dL.72 Other
microdialysis studies suggest administration of insulin could
impede the glucose supply of the brain73 and that brain energy
metabolic crisis (elevated lactate-to-pyruvate ratio) may occur
with acute reduction in serum glucose even when those levels
are normal.74 Together these various data suggest that IIT in
patients with already marginal serum glucose levels may be
detrimental, and the optimal goal for a glucose level in the
setting of acute neurologic injury remains to be elucidated.
The available evidence suggests a less restrictive target for
systemic glucose control (8-10 mmol/L; 144-180 mg/dL) may
be reasonable. Furthermore the data emphasize the impor-
tance of monitoring blood glucose and the effects of therapy
and that an approach that is perhaps more individualized to
avoid both hypoglycemia and hyperglycemia is warranted
until specific glycemic control guidelines for patients with
severe brain injury in the neurocritical car unit (NCCU) are
better defined.43 An example of a continuous insulin protocol
is provided in Table 14.3.
Table 14.3  An Example of a Continuous
Insulin Infusion Therapy Protocol
 Regular insulin in 0.9% sodium chloride (1 U/mL)
administered through large-bore peripheral or central
venous catheter.
 The insulin infusion is adjusted based on every-hour
Accu-Cheks to maintain target blood glucose.
 Accu-Chek measurements performed every 2 hours when
blood glucose is controlled for 8 hours or more and then
switch to subcutaneous insulin therapy.
Blood Glucose Initial Insulin Infusion
Level (mg/dL) (U/hr)
>351 14 (plus 0.1 U/kg IV bolus)
280-350 10
240-279 8 Nutritional status often deteriorates further because of poor
200-239 6 intake associated with reduced level of consciousness, dyspha-
170-199 4 gia, and weakness that limits the ability to self-feed. Therefore
140-169 3
111-139 2 closely associated with glycemic control such that balanced
IV, Intravenous.
Beyond a Glucose Threshold
There are several areas of research beyond just the threshold
at which to treat elevated blood glucose that are relevant to
glycemic control: (1) treatment algorithms and patient safety
mechanisms (see also Chapter 6),75 (2) use of computerized
decision support systems to deliver insulin or development
of continuous monitors with automated closed-loop systems,
(3) glucose variability, (4) whether target levels should differ
according to the indication for ICU admission or the stage of
acute illness (i.e., a population-specific glycemic control strat-
egy), and (5) the relationship between glucose, glycemic
control, and nutrition. In addition, attention to ICU processes
is important. For example, the risk of hypoglycemia associated
with IIT is associated with nursing schedules and fatigue,76
whereas when nurses override clinical decision support
systems for IIT they tend to use lower-than-recommended
doses to avoid hypoglycemia.77 Finally bioinformatics systems
can be used to facilitate integration of glucose control and
nutrition.78
Glucose variability, that is, changes in the measured level of
blood glucose, may be of particular importance because it
appears to be an independent or even casual factor associated
with adverse outcome.7,79-81 This in part may explain the
limited effect of IIT because IIT seems to increase glucose
variability.80 On the other hand IIT also is associated with
hypoglycemia, the severity of which is independently associ-
ated with increased mortality.80,82 However, there are no
prospective studies that examine whether reduced glucose
variability is associated with better outcome, and so insulin
therapy is for the moment targeted to a glucose threshold
rather than reduced glucose variability.
Nutrition
Since the 1990s, research has verified the importance of nutri-
tion in the critically ill. In addition to maintaining weight,
providing energy, and ensuring a positive nitrogen balance,
nutrition, particularly successful enteral nutrition, can help
augment immune function and is associated with fewer noso-
comial infections,83,84 reduced mortality,85 and improved neu-
rologic outcomes.86 However, it appears that the mean time to
start enteral nutrition (EN) in critically ill patients is 3 days,87
and average caloric intake is only 50% of American College of
Chest Physicians (ACCP) goals.88 Despite national guidelines
that support early and aggressive nutritional support,89,90
patients in the ICU are often underfed.
Supplying adequate nutrition to those patients with severe
neurologic injury is particularly problematic. Many of those
admitted to the hospital are already malnourished, including
one fifth of patients with acute stroke.91 Acute neurologic
injury is associated with profound hypermetabolic and
hypercatabolic states. Patients therefore can lose nitrogen in
quantities sufficient to reduce weight by 15% per week.89
optimizing nutritional status is an essential but often over-
looked aspect of neurocritical care. In addition, nutrition is
enteral nutrition may reduce the risk of hypoglycemia associ-
ated with IIT.8 For example, a meta-analysis suggests that IIT
126 Section II—Clinical and Laboratory Assessment
is associated with reduced mortality only when a high propor-
tion of calories is provided parenterally.9 Consistent with this
parenteral dextrose was an important source of energy in the
first Leuven study, which showed that IIT was associated with
reduced mortality.1 Furthermore, the rates of hypoglycemia1
were less than those in NICE-SUGAR51 in which intravenous
dextrose rarely was used and nutritional support was less
aggressive. Hence the benefit of IIT may depend on nutritional
support or adequate calorie supply to avoid the hypoglycemia
of IIT. Consistent with this, microdialysis studies show that
brain glucose is favorably influenced by enteral nutrition
without an adverse effect on other markers such as glutamate
or the lactate-to-pyruvate ratio.92
Initiation and Optimization of Feeding
Studies that address the appropriate timing, amount, and
method of feeding are difficult to generalize because of small
sample sizes, variable definitions of “early” and “late” feeding,
and inconsistent adherence to guidelines. Overall, the observa-
tions support a recommendation of “early” enteral feeding.93-96
Marik and Zaloga93 performed a meta-analysis of 15 random-
ized controlled trials that included 753 critically ill patients.
Early feeding, defined as nutrition that commenced within 36
hours of admission, was associated with a 50% decrease in
infectious complications and a 2.2-day reduction in length of
stay (LOS). The reduction in LOS was most pronounced in
those patients with trauma, head injury, and burns. Rubinson
et al. followed 138 patients after MICU admission and found
that failing to provide more than or equal to 25% of the rec-
ommended calories was associated with a higher risk of
bloodstream infection.94 Likewise, Villet et al. reported that
in 48 critically ill patients, a negative energy balance was
associated with an increased number of infectious and
noninfectious complications.95
Studies in neurocritical care also demonstrate an associa-
tion between lack of nutrition and adverse outcomes. For
example, Hartl et al.97 evaluated the effect of early nutrition in
nearly 800 patients with severe TBI. Patients who were not fed
within 5 days of injury had twice the incidence of death than
those fed early. Every 10 kcal/kg decrease in caloric intake was
associated with a 30% to 40% increase in mortality, even after
controlling for hypotension, age, pupillary status, initial GCS,
and computed tomography scan findings. Taylor et al. in a
prospective study found that TBI patients randomized to a
tube-feeding regimen that supplied full energy requirements
from the day of admission had nearly 25% fewer infections
and decreased markers of inflammation than patients who
were started on tube feeds that were gradually increased to
goal.98 In patients with acute stroke, protein-energy malnutri-
tion is associated with increased infection and worse outcome,
independent of age or baseline nutritional status.99
Route of Acute Nutritional Delivery
There is limited information to recommend whether EN or
parenteral nutrition (PN) is superior for patients in the
NCCU. Many ICU practitioners were trained by the motto,
“If the gut works, use it.” Lack of EN increases susceptibility
to infection by impairing the delicate systems of mucosal
immunity, gut cytokines, and endothelial markers of inflam-
mation.100 However, EN may be associated with high gastric
residual volumes—retention, diarrhea, and aspiration—that
may limit enteral access for nutrition. Moreover, EN may be
interrupted by frequent displacement of access tubes, discon-
tinued feeds during incompatible medicine administration
and procedures, and inadequate ordering by staff. For
example, McClave et  al. observed that only 14% of patients
reached 90% of goal feeding within 72 hours of the start
of EN.101
Successful tube feeding requires adequate gastric emptying,
which is often impaired by the primary neurologic condition,
use of opiates, and limited mobility. A high residual volume is
the most frequent cause to discontinue enteral tube feeding,87,102
but the threshold volume for which tube feeds should be
stopped varies between practices. ACCP guidelines recom-
mend that a gastric residual greater than 150 mL should
prompt a decrease in the infusion rate, consideration of PN,
or use of small bowel feeding.83 However, McClave et al.
studied the incidence of aspiration over a wide range of resid-
ual volumes from 200 to 400 mL and found no relation
between the residual volume and aspiration.102 The authors
suggest that feeds should not be stopped when residual
volumes are less than 400 to 500 mL unless there are other
signs of intolerance. A strict bowel regimen that includes pro-
motility agents, stool softeners, and liberalization of activity
orders may help improve tolerance to enteral feeding.
Both jejunal and gastric tubes are used routinely to provide
EN. Although the gastric route is associated with a higher
incidence of high gastric residual volumes, it is safe and well
tolerated in most patients.103 Montejo studied the efficacy and
complication rate of gastric and jejunal feeding in 101 ran-
domized patients. There were no differences in the incidence
of hospital-acquired pneumonia, length of stay, multiple-
organ failure score, or mortality between groups, despite the
gastric feeding group having a significantly higher incidence
of high gastric residuals.87 Early data suggesting that TBI
patients benefit from jejunal over gastric feeding have been
criticized because the gastric group received no nutrition until
bowel sounds returned, whereas the jejunal group received
nutrition within 36 hours of admission.104 Furthermore,
gastric tubes are easy to place, and generally eliminate the need
for multiple radiographs to document correct placement.
Jejunal tubes have been linked to a higher caloric intake,105
perhaps associated with decreased cessation of EN secondary
to high gastric residual volumes.
PN is advised when the enteral route is either unusable or
unable to provide adequate calories. When PN is used, criti-
cally ill patients should be given a mixture that consists of
amino acids (between 0.8 and 1.5 g/kg/day), carbohydrates
(about 60% of the nonprotein energy) and fat (about 40% of
the nonprotein energy), electrolytes, and micronutrients.12
Close monitoring is required when PN is used for (1) toler-
ance of electrolytes and macronutrients, (2) glycemia control,
and (3) specific micronutrients.106 PN may be used alone or
in combination with EN; therefore, even if the patient requires
supplemental PN to meet calorie and nitrogen requirements,
the enteral route should be used unless it is contraindicated.
When PN and EN are used together, late (8 days) may be
preferable to early (48 hours) initiation of PN.107 A potential
advantage of PN is that nutritional goals may be reached more
quickly with PN, especially in trauma patients who may
require 130% to 150% of predicted energy expenditure for
many weeks after injury.108 However, some studies suggest PN
 Section II—Clinical and Laboratory Assessment 127
is associated with increased risk of infection, particularly hos-
pital-acquired candidemia,109 and increased cost. By contrast
other studies suggest the risk of infection is not increased. For
example, Woodcock et al. evaluated 500 patients who received
adjuvant nutritional support and found no evidence that PN
was associated with a higher incidence of infection.110 Simi-
larly, in a small study that evaluated PN and EN in 45 TBI
patients, Hadley et al. observed that infection was similar in
the two groups.111
Vasopressor use is considered a contraindication to EN. The
American Society of Parenteral and Enteral Nutrition (ASPEN)
2002 guidelines state that EN should be deferred until the
patient is hemodynamically stable.10 However, vasopressors in
the NCCU do not necessarily indicate hemodynamic instabil-
ity because they often are used to manage cerebral vasospasm
or maintain cerebral perfusion pressure. Furthermore, the
data that support the recommendation to limit EN during
vasopressor use are based largely on expert opinions and con-
tradictory literature. On the other hand Doshi and Divgiovine
found that early enteral feeding was associated with a 20%
decrease in ICU mortality and a 25% decrease in hospital
mortality among 1174 patients who required mechanical ven-
tilation and vasopressors.112 Similarly Berger et al. found that
no patient with circulatory failure who received EN and
required norepinephrine or dopamine after cardiac bypass
experienced any serious gastrointestinal complication.113
These data suggest that although EN should be used with
caution in patients who are hypotensive, it is well tolerated
and likely of benefit to patients on vasopressors.
How to Determine Energy Needs
There are several methods to calculate calorie needs, but the
best method to determine nutritional needs in hospitalized
patients remains debated. There are multiple predictive equa-
tions, but these may over- or underestimate caloric needs.114
Indirect calorimetry may be a more useful measure of energy
needs in hospitalized patients. A metabolic cart, which mea-
sures the oxygen consumed and the carbon dioxide produced
by the patient and then calculates the energy expenditure, is
felt to be particularly beneficial for patients who are obese, in
a pharmacologically induced coma, or being treated for neu-
rotrauma.115 General nutrient requirements for ICU patients
recommended by the ACCP are listed in Table 14.4.90 These
guidelines should be adapted to the specific disease state and
baseline nutritional status of each individual patient. For
example, a high-protein, hypocaloric regimen may be prefer-
able in the obese patient, although specific guidelines for obese
patients are limited.116,117 This means understanding each
patient’s premorbid nutritional state, basal metabolic rate,
activity level, and stress factors that contribute to greater
energy needs.
Indirect Calorimetry
Assessment of nutritional status is multidimensional and
requires consideration of several factors, including nutrition
status, nutritional intake, and energy expenditure. The refer-
ence standard to measure energy expenditure and caloric
requirements in the ICU is indirect calorimetry.118 For example,
the Deltatrac II MBM-200 Metabolic Monitor (Datex-Eng-
strom Division, Instrumentation Corp., Datex-Engstrom,
Table 14.4  ACCP Recommendations for
Nutrition in ICU Patients
Total calories  25 total kilocalories per kilogram usual
body weight per day
 1 mL of water per kilocalorie
Glucose  30%-70% of total calories per day
Fats  15%-30% of total calories per day
 7% of total calories administered as
omega-6 PUFA triglycerides
 Monitoring of triglyceride clearance is
recommended.
 Serum triglyceride levels above
500 mg/dL should initiate consideration
for a decrease in total calories and/or
omega-6 PUFA triglycerides.
Protein  15%-20% of total calories per day
 Initiate therapy at 1.2 to 1.5 g/kg/day;
consider increase dosing in patients
with trauma or burns.
 BUN level exceeding 100 mg/dL should
initiate consideration for a decrease in
dosing.
 Nitrogen balance should be assessed
every 5-7 days and protein dosing
adjusted accordingly.
Micronutrients  Potassium, magnesium, zinc, and
phosphate should be provided in doses
adequate to maintain normal serum
levels.
ACCP, American College of Chest Physicians; BUN, blood urea nitrogen; ICU,
intensive care unit; PUFA, polyunsaturated fatty acid.
Finland) can be used to perform indirect calorimetry with
most commonly used ventilators. Indirect calorimetry is a
technique that measures oxygen consumption and carbon
dioxide production to calculate resting energy expenditure
and respiratory quotient (RQ) and to measure the actual
energy expenditure (measured energy expenditure [MEE]) of
a patient. Because more than 95% of the energy generated
requires oxygen, there is a direct proportion between oxygen
consumption and the metabolic rate. One liter of oxygen con-
sumed generates 3.8 kcal (16.32 kJ): 1 L of carbon dioxide
produced generates 1.1 kcal (4.60 kJ). Ideally the patient
should be assessed in a thermoneutral environment, at rest for
more than 30 minutes, and without food for 2 hours for
precise measurements. A stress factor to account for injury is
not necessary because the MEE accounts for the effects of the
disease. However, because the measurement is at rest, it is
necessary to multiply by an “activity” factor of 1 to 1.3,
depending on whether the patient is intubated, at bed rest,
paralyzed, or out of bed.
Formulas to Determine Caloric Requirements
Instead of indirect calorimetry, which is not widely avail-
able, there are a number of predictive equations that can
be used to derive nutritional needs based on the degree of
injury and patient weight among other variables.90 These
various equations and nutritional status indices, however,
may be less accurate than indirect calorimetry.119
The Maastricht Index (MI)120 and the Nutritional Risk
Index (NRI),121 can be used to assess the nutritional status of
128 Section II—Clinical and Laboratory Assessment
a patient to provide a sense of the individual’s premorbid
condition. The MI is calculated as follows:
 MI = [20.68 − (0.24 × albumin, g/L) − (19.21 ×
serum transthyretin, g/L) − (1.86 × lymphocytes, 106/L)
− (0.04 × ideal weight)]
Nutritional status is graded as either malnourished or not
malnourished with the MI; patients with a MI greater than
0 are considered malnourished. Use of the MI is limited
to patients younger than 70 years of age.The NRI was devel-
oped to assess malnutrition of surgical patients. There also
are geriatric-specific NRIs.
 NRI = (1.519 × serum albumin, g/dL) +
(41.7 × present weight [kg]/ideal body weight [kg])
A value of NRI greater than100 indicates that the subject is
not malnourished, 97.5 to 100 indicates the subject is mildly
malnourished, 83.5 to 97.5 indicates that the subject is mod-
erately malnourished, and NRI less than 83.5 indicates that
the subject is severely malnourished.
Basal metabolic rate (BMR) is a simple and accurate system
that can be used to estimate caloric needs is according to the
following formula:
 Energy requirements = BMR ×
activity level (usually 1.25) × stress factor
BMR describes the energy required to maintain cell integrity
in the resting state and at thermoneutrality. A significant
portion of energy expenditure goes into thermogenesis, the
maintenance of normal body temperature through heat pro-
duction. Thermoneutrality usually is close to 80° F [28°  C]
when the heat loss and the need for increased heat produc-
tion to maintain the body temperature are minimal.  The
BMR for man has been defined by direct calorimetry. Body
size is a key factor, and so BMR is expressed in terms of
square meters of body surface or body weight. Other vari-
ables include age and sex. The average adult BMR value is
1500 to 1800  kcal per day (850 to 950  kcal/m2/day or 20 to
24  kcal/kg/day).
The Harris-Benedict equation can be used to calculate
resting energy expenditure (REE).122
 For men, REE = 66.5 + (13.75 × weight [kg]) +
(5.003 × height [cm]) − (6.775 × age [yr])
 Or, REE = 65 + (6.2 × weight [lb]) +
(12.7 × height [in]) − (6.8 × age [yr])
 For women, REE = 655.1 + (9.563 × weight [kg]) +
(1.850 × height [cm]) − (4.676 × age [yr])
 Or, REE = 655 + (4.3 × weight [lb]) +
(4.7 × height [in]) − (4.7 × age [yr])
A variety of multiplication factors (activity, stress, tempera-
ture that range from 1.2 to >2) are introduced to calculate
adequate calorie intake.
 Total caloric requirements = REE × by the sum of the stress
and activity factors, for example, total calorie require-
ments/day = 1.25 × REE. For each 1° C more than 37° C
add 10% extra allowance.
The activity level reflects energy used by muscles for average
daily activity. In the ICU, patient activity is usually limited and
is usually considered to be no greater than an additional 25%
of the BMR. On the other hand excessive muscle activity such
as from excessive work of breathing or a combative disori-
ented state increases the energy expenditure.
Stress-Induced Energy Needs
The stress factor (SF) defines the hypermetabolism induced
by illness and is a multiple of the BMR (normalized to 1). The
value takes into consideration the type of injury and the
average metabolic response with this degree of injury. For
example, an elective operation increases needs about 10% (SF
= 1.1), major infection about 25% (SF = 1.25), and multiple
blunt trauma about 50% (SF = 1.5). Burn injury leads to the
greatest increase in metabolic demands.
Anthropometric Measurements
Anthropometric measurements such as patients’ height and
weight to calculate body mass index (BMI), triceps skinfold
thickness (TSF), and midarm circumference (MAC) to calcu-
late muscle area among others may be used to help calculate
energy and nutritional needs.
 BMI = weight (kg)/height m2
 Arm muscle area (mm2) = (MAC [mm] −
[TSF —3.14])2/4π
A BMI less than 18.5 is underweight; a BMI between 25 and
29.9 is overweight; a BMI greater than or equal to 30 is obese,
and a BMI greater than or equal to 40 is morbidly obese. These
values may need to be factored in when calculating calorie and
protein needs.
Biochemical Measurements
While in the ICU, biochemical measurements including serum
albumin and prealbumin, transferrin, retinol binding protein,
magnesium, phosphorus, total lymphocyte count (TLC), and
a 24-hour urine collection to measure urine urea nitrogen
(UUN) and creatinine can help guide nutritional therapy. The
creatinine height index is calculated from 24-hour urinary
creatinine excretion.123 The nitrogen balance is derived from
the daily volume of infusate minus the daily output of urinary
nitrogen:
 Nitrogen balance (g/day) = protein intake (grams)/6.25 −
(UUN + 4g obligatory loss)
 Or, nitrogen balance = nitrogen intake − nitrogen output
UUN is measured in grams of nitrogen excreted in the
urine over a 24-hour period. The following values are con-
sidered normal (although this may vary between laborato-
ries): serum albumin, 3.5 to 5  g/dL; prealbumin, 18 to
43mg/dL; total lymphocyte count (TLC) greater than 1500/
mm3; and creatinine height index greater than 90%.
Patients admitted to the ICU should be weighed on admis-
sion and then at least twice weekly. Ideal body weights (IBWs)
then can be used to calculate caloric goals:
 IBW males = 106 lb for 5'0" + 6 lb/inch >5'
(or IBW = 50 kg + 2.3 kg for each inch >5 feet)
 IBW females = 100 lb for 5'0" + 5 lb/inch >5'
(or IBW = 45.5 kg + 2.3 kg for each inch over 5 feet)
 Section II—Clinical and Laboratory Assessment 129
Conclusion
Initial caloric goals are 25 to 35 kcal/kg of IBW. An exception
may be patients with spinal cord injury who are paralyzed and
who may require fewer initial calories that are inversely related
to the level of injury (i.e., the higher the lesion, the lower the
energy expenditure). The protein requirement is 1.5 to 1.8 g/
kg/day using IBW for all patients. However, protein require-
ments are altered by conditions such as acute pancreatitis and
hepatic encephalopathy in which protein is restricted to 0.6 g/
kg/day for 3 days and then increased to less than or equal to
1.2 g/kg/day or reduced if the mental status deteriorates. In
addition, protein needs may need to be modified for renal
failure (creatinine [Cr] >2.5). For example, the initial protein
intake may need to be reduced to 1 to 1.3 g/kg/day and the
daily blood urea nitrogen (BUN) followed. If the patient is
oliguric, protein intake may need to be started at less than or
equal to 0.6 g/kg/day and the daily BUN followed. Specific
enteral nutritional supplements are available for these various
conditions (e.g., Magnacal, Peptinex, NutriHep).
References
1. Van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in
critically ill patients. N Engl J Med 2001;345:1359–67.
2. Van den Berghe G, Wilmer A, Hermans G, et al. Intensive insulin therapy in
medical intensive care patients. N Engl J Med 2006;354:449–61.
3. Wiener RS, Wiener DC, Larson RJ. Benefits and risks of tight glucose control
in critically ill adults: a meta-analysis. JAMA 2008;300(8):933–44.
4. Jacka MJ, Torok-Both CJ, Bagshaw SM. Blood glucose control among
critically ill patients with brain injury. Can J Neurol Sci 2009;36(4):
436–42.
5. Krinsley JS, Schultz MJ, Spronk PE, et al. Mild hypoglycemia is
independently associated with increased mortality in the critically ill.
Crit Care 2011;15(4):R173.
6. Duning T, van den Heuvel I, Dickmann A, et al. Hypoglycemia aggravates
critical illness-induced neurocognitive dysfunction. Diabetes Care 2010;
33(3):639–44. Epub 2009, Dec 23.
7. Cueni-Villoz N, Devigili A, Delodder F, et al. Increased blood glucose
variability during therapeutic hypothermia and outcome after cardiac arrest.
Crit Care Med 2011;39(10):2225–31.
8. Kauffmann RM, Hayes RM, Jenkins JM, et al. Provision of balanced
nutrition protects against hypoglycemia in the critically ill surgical patient.
JPEN J Parenter Enteral Nutr 2011;35(6):686–94. Epub 2011, Jul 12.
9. Marik PE, Preiser JC. Toward understanding tight glycemic control in the
ICU: a systematic review and metaanalysis. Chest 2010;137:544–51.
10. ASPEN Board of Directors and the Clinical Guidelines Task Force.
Guidelines for the use of parenteral and enteral nutrition in adult and
pediatric patients. J Parenter Enteral Nutr 2002;26S:1SA–138SA.
11. Kreymann KG, Berger MM, Deutz NE, et al. ESPEN guidelines on enteral
nutrition: intensive care. Clin Nutr 2006;25:210–23.
12. Kreymann G, Adolph M, Druml W, et al. Intensive medicine: guidelines on
parenteral nutrition. Ger Med Sci 2009;7:doc14.
13. Berger MM, Chiolero RL. Hypocaloric feeding: pros and cons. Curr Opin
Crit Care 2007;13:180–6.
14. Heyland DK, Dhaliwal R, Drover JW, et al. Canadian clinical practice
guidelines for nutrition support in mechanically ventilated, critically ill adult
patients. JPEN J Parenter Enteral Nutr 2003;27:355–73.
15. McClave SA, Martindale RG, Vanek VW, et al. Guidelines for the provision
and assessment of nutrition support therapy in the adult critically ill patient:
Society of Critical Care Medicine (SCCM) and American Society for
Parenteral and Enteral Nutrition (ASPEN). JPEN J Parenter Enteral Nutr
2009;33(3):277–316.
16. Juneja D, Pandey R, Singh O. Comparison between arterial and capillary
blood glucose monitoring in patients with shock. Eur J Intern Med
2011;22(3):241–4. Epub 2011, Feb 16.
17. Corstjens AM, Ligtenberg JJ, van der Horst IC, et al. Accuracy and feasibility
of point-of-care and continuous blood glucose analysis in critically ill ICU
patients. Crit Care 2006;10(5):R135.
18. Leino A, Kurvinen K. Interchangeability of blood gas, electrolyte and
metabolite results measured with point-of-care, blood gas and core
laboratory analyzers. Clin Chem Lab Med 2011;49(7):1187–91. Epub 2011,
Apr 20.
19. Penfornis A, Personeni E, Borot S. Evolution of devices in diabetes
management. Diabetes Technol Ther 2011;13(Suppl 1):S93–102.
20. Shearer A, Boehmer M, Closs M, et al. Comparison of glucose point-of-care
values with laboratory values in critically ill patients. Am J Crit Care
2009;18(3):224–30.
21. Roth-Kleiner M, Stadelmann Diaw C, Urfer J, et al. Evaluation of different
POCT devices for glucose measurement in a clinical neonatal setting.
Eur J Pediatr 2010;169(11):1387–95. Epub 2010, Jun 24.
22. Meynaar IA, van Spreuwel M, Tangkau PL, et al. Accuracy of AccuChek
glucose measurement in intensive care patients. Crit Care Med 2009;37(10):
2691–6.
23. Egi M, Finfer S, Bellomo R. Glycemic control in the ICU. Chest
2011;140(1):212–20.
24. Ichai C, Preiser JC, Société Française d’Anesthésie-Réanimation, et al.
International recommendations for glucose control in adult non diabetic
critically ill patients. Crit Care 2010;14(5):R166. Epub 2010, Sep 14.
25. Rooyackers O, Blixt C, Mattsson P, Wernerman J. Continuous glucose
monitoring by intravenous microdialysis. Acta Anaesthesiologica
Scandinavica 2010;54(7):841–7.
26. Möller F, Liska J, Franco-Cereceda A. Evaluation of a continuous blood
glucose monitoring system using central venous microdialysis. Crit Care
2011;15(Suppl 1):P405. Published online 2011 March 11. doi: 10.1186/
cc9825. PMCID: PMC3068334.
27. Dungan KM, Braithwaite SS, Preiser JC. Stress hyperglycaemia. Lancet
2009;373:1798–807.
28. Farrokhi F, Smiley D, Umpierrez GE. Glycemic control in non-diabetic
critically ill patients. Best Pract Res Clin Endocrinol Metab
2011;25(5):813–24.
29. Egi M, Bellomo R, Stachowski E, et al. Blood glucose concentration and
outcome of critical illness: the impact of diabetes. Crit Care Med 2008;36:
2249–55.
30. Falciglia M, Freyberg RW, Almenoff PL, et al. Hyperglycemia-related
mortality in critically ill patients varies with admission diagnosis.
Crit Care Med 2009;37(12):3001–9.
31. Siegelaar SE, Hoekstra JB, DeVries JH. Special considerations for the diabetic
patient in the ICU: targets for treatment and risks of hypoglycaemia.
Best Pract Res Clin Endocrinol Metab 2011;25(5):825–34.
32. Egi M, Bellomo R, Stachowski E, et al. The interaction of chronic and acute
glycemia with mortality in critically ill patients with diabetes. Crit Care Med
2011;39(1):105–11.
33. Williams LS, Rotich J, Qi R, et al. Effects of admission hyperglycemia on
mortality and costs in acute ischemic stroke. Neurology 2002;59:67–71.
34. Baird TA, Parsons MW, Phanh T, et al. Persistent post stroke hyperglycemia
is independently associated with infarct expansion and worse clinical
outcomes. Stroke 2003;34:2208–14.
35. Juvela S, Siironen J, Kuhmonen J. Hyperglycemia, excess weight, and history
of hypertension as risk factors for poor outcome and cerebral infarction
after aneurysmal subarachnoid hemorrhage. J Neurosurg
2005;102(6):998–1003.
36. Frontera JA, Fernandez A, Claassen J, et al. Hyperglycemia after SAH:
predictors, associated complications, and impact on outcome. Stroke
2006;37(1):199–203.
37. McGirt MJ, Woodworth GF, Ali M, et al. Persistent perioperative
hyperglycemia as an independent predictor of poor outcome after
aneurysmal subarachnoid hemorrhage. J Neurosurg 2007;107(6):1080–5.
38. Badjatia N, Topcuoglu MA, Buonanno FS, et al. Relationship between
hyperglycemia and symptomatic vasospasm alter subarachnoid hemorrhage.
Crit Care Med 2005;33:1603–9.
39. Worthley MI, Shrive FM, Anderson TJ, et al. Prognostic implication of
hyperglycemia in myocardial infarction and primary angioplasty. Am J Med
2007;120(7):643.e1–e7.
40. Fogelholm R, Murros K, Rissanen A, et al. Admission blood glucose and
short term survival in primary intracerebral haemorrhage: a population
based study. J Neurol Neurosurg Psychiatry 2005;76:349–53.
41. Salim A, Hadjizacharia P, Dubose J, et al. Persistent hyperglycemia in severe
traumatic brain injury: an independent predictor of outcome. Am Surg
2009;75(1):25–9.
42. Smith RL, Lin JC, Adelson PD, et al. Relationship between hyperglycemia
and outcome in children with severe traumatic brain injury. Pediatr Crit
Care Med 2012;13(1):85–91.
130 Section II—Clinical and Laboratory Assessment
43. Schmutzhard E, Rabinstein AA, Participants in the International
Multi-Disciplinary Consensus Conference on the Critical Care Management
of Subarachnoid Hemorrhage. Spontaneous subarachnoid hemorrhage and
glucose management. Neurocrit Care 2011;15(2):281–6.
44. Dumont T, Rughani A, Silver J, et al. Diabetes mellitus increases risk of
vasospasm following aneurysmal subarachnoid hemorrhage independent of
glycemic control. Neurocrit Care 2009;11(2):183–9. Epub 2009, May 27.
45. Vogelzang M, Nijboer JMM, van der Horst ICC, et al. Hyperglycemia has a
stronger relation with outcome in trauma patients than in other critically ill
patients. J Trauma 2006;60:873–9.
46. Jeremitsky E, Omert LA, Dunham M, et al. The impact of hyperglycemia on
patients with severe brain injury. J Trauma 2005;58:47–50.
47. McCormick MT, Muir KW, Gray CS, et al. Management of hyperglycemia in
acute stroke: how, when, and for whom? Stroke 2008;39:2177–85.
48. Oddo M, Levine JM, Frangos S, et al. Brain lactate metabolism in humans
with subarachnoid hemorrhage. Stroke 2012. Epub ahead of print.
49. Wyss MT, Jolivet R, Buck A, et al.. In vivo evidence for lactate as a neuronal
energy source. J Neurosci 2011;31(20):7477–85.
50. Vanhorebeek I, Langouche L, Van den Berghe G. Glycemic and nonglycemic
effects of insulin: how do they contribute to a better outcome of critical
illness? Curr Opin Crit Care 2005;11:304–11.
A complete list of references for this chapter can be found online at
www.expertconsult.com.
 Section II—Clinical and Laboratory Assessment 130.e1
References
1. Van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in
critically ill patients. N Engl J Med 2001;345:1359–67.
2. Van den Berghe G, Wilmer A, Hermans G, et al. Intensive insulin therapy
in medical intensive care patients. N Engl J Med 2006;354:449–61.
3. Wiener RS, Wiener DC, Larson RJ. Benefits and risks of tight glucose
control in critically ill adults: a meta-analysis. JAMA 2008;300(8):933–44.
4. Jacka MJ, Torok-Both CJ, Bagshaw SM. Blood glucose control among
critically ill patients with brain injury. Can J Neurol Sci 2009;36(4):
436–42.
5. Krinsley JS, Schultz MJ, Spronk PE, et al. Mild hypoglycemia is
independently associated with increased mortality in the critically ill. Crit
Care 2011;15(4):R173.
6. Duning T, van den Heuvel I, Dickmann A, et al. Hypoglycemia aggravates
critical illness-induced neurocognitive dysfunction. Diabetes Care 2010;
33(3):639–44. Epub 2009, Dec 23.
7. Cueni-Villoz N, Devigili A, Delodder F, et al. Increased blood glucose
variability during therapeutic hypothermia and outcome after cardiac
arrest. Crit Care Med 2011;39(10):2225–31.
8. Kauffmann RM, Hayes RM, Jenkins JM, et al. Provision of balanced
nutrition protects against hypoglycemia in the critically ill surgical patient.
JPEN J Parenter Enteral Nutr 2011;35(6):686–94. Epub 2011, Jul 12.
9. Marik PE, Preiser JC. Toward understanding tight glycemic control in the
ICU: a systematic review and metaanalysis. Chest 2010;137:544–51.
10. ASPEN Board of Directors and the Clinical Guidelines Task Force.
Guidelines for the use of parenteral and enteral nutrition in adult and
pediatric patients. J Parenter Enteral Nutr 2002;26S:1SA–138SA.
11. Kreymann KG, Berger MM, Deutz NE, et al. ESPEN guidelines on enteral
nutrition: intensive care. Clin Nutr 2006;25:210–23.
12. Kreymann G, Adolph M, Druml W, et al. Intensive medicine: guidelines on
parenteral nutrition. Ger Med Sci 2009;7:doc14.
13. Berger MM, Chiolero RL. Hypocaloric feeding: pros and cons. Curr Opin
Crit Care 2007;13:180–6.
14. Heyland DK, Dhaliwal R, Drover JW, et al. Canadian clinical practice
guidelines for nutrition support in mechanically ventilated, critically ill
adult patients. JPEN J Parenter Enteral Nutr 2003;27:355–73.
15. McClave SA, Martindale RG, Vanek VW, et al. Guidelines for the provision
and assessment of nutrition support therapy in the adult critically ill
patient: Society of Critical Care Medicine (SCCM) and American Society
for Parenteral and Enteral Nutrition (ASPEN). JPEN J Parenter Enteral
Nutr 2009;33(3):277–316.
16. Juneja D, Pandey R, Singh O. Comparison between arterial and capillary
blood glucose monitoring in patients with shock. Eur J Intern Med
2011;22(3):241–4. Epub 2011, Feb 16.
17. Corstjens AM, Ligtenberg JJ, van der Horst IC, et al. Accuracy and
feasibility of point-of-care and continuous blood glucose analysis in
critically ill ICU patients. Crit Care 2006;10(5):R135.
18. Leino A, Kurvinen K. Interchangeability of blood gas, electrolyte and
metabolite results measured with point-of-care, blood gas and core
laboratory analyzers. Clin Chem Lab Med 2011;49(7):1187–91. Epub 2011,
Apr 20.
19. Penfornis A, Personeni E, Borot S. Evolution of devices in diabetes
management. Diabetes Technol Ther 2011;13(Suppl 1):S93–102.
20. Shearer A, Boehmer M, Closs M, et al. Comparison of glucose point-of-
care values with laboratory values in critically ill patients. Am J Crit Care
2009;18(3):224–30.
21. Roth-Kleiner M, Stadelmann Diaw C, Urfer J, et al. Evaluation of different
POCT devices for glucose measurement in a clinical neonatal setting.
Eur J Pediatr 2010;169(11):1387–95. Epub 2010, Jun 24.
22. Meynaar IA, van Spreuwel M, Tangkau PL, et al. Accuracy of AccuChek
glucose measurement in intensive care patients. Crit Care Med 2009;
37(10):2691–6.
23. Egi M, Finfer S, Bellomo R. Glycemic control in the ICU. Chest 2011;
140(1):212–20.
24. Ichai C, Preiser JC, Société Française d’Anesthésie-Réanimatio, et al.
International recommendations for glucose control in adult non diabetic
critically ill patients. Crit Care 2010;14(5):R166. Epub 2010, Sep 14.
25. Rooyackers O, Blixt C, Mattsson P, Wernerman J. Continuous glucose
monitoring by intravenous microdialysis. Acta Anaesthesiologica
Scandinavica 2010;54(7):841–7.
26. Möller F, Liska J, Franco-Cereceda A. Evaluation of a continuous blood
glucose monitoring system using central venous microdialysis. Crit Care
2011;15(Suppl 1):P405. Published online 2011 March 11. doi: 10.1186/
cc9825. PMCID: PMC3068334.
27. Dungan KM, Braithwaite SS, Preiser JC. Stress hyperglycaemia. Lancet
2009;373:1798–807.
28. Farrokhi F, Smiley D, Umpierrez GE. Glycemic control in non-diabetic
critically ill patients. Best Pract Res Clin Endocrinol Metab
2011;25(5):813–24.
29. Egi M, Bellomo R, Stachowski E, et al. Blood glucose concentration and
outcome of critical illness: the impact of diabetes. Crit Care Med
2008;36:2249–55.
30. Falciglia M, Freyberg RW, Almenoff PL, et al. Hyperglycemia-related
mortality in critically ill patients varies with admission diagnosis.
Crit Care Med 2009;37(12):3001–9.
31. Siegelaar SE, Hoekstra JB, DeVries JH. Special considerations for the
diabetic patient in the ICU: targets for treatment and risks of
hypoglycaemia. Best Pract Res Clin Endocrinol Metab 2011;25(5):
825–34.
32. Egi M, Bellomo R, Stachowski E, et al. The interaction of chronic and acute
glycemia with mortality in critically ill patients with diabetes. Crit Care
Med 2011;39(1):105–11.
33. Williams LS, Rotich J, Qi R, et al. Effects of admission hyperglycemia on
mortality and costs in acute ischemic stroke. Neurology 2002;59:67–71.
34. Baird TA, Parsons MW, Phanh T, et al. Persistent post stroke hyperglycemia
is independently associated with infarct expansion and worse clinical
outcomes. Stroke 2003;34:2208–14.
35. Juvela S, Siironen J, Kuhmonen J. Hyperglycemia, excess weight, and history
of hypertension as risk factors for poor outcome and cerebral infarction
after aneurysmal subarachnoid hemorrhage. J Neurosurg
2005;102(6):998–1003.
36. Frontera JA, Fernandez A, Claassen J, et al. Hyperglycemia after SAH:
predictors, associated complications, and impact on outcome. Stroke
2006;37(1):199–203.
37. McGirt MJ, Woodworth GF, Ali M, et al. Persistent perioperative
hyperglycemia as an independent predictor of poor outcome after
aneurysmal subarachnoid hemorrhage. J Neurosurg 2007;107(6):1080–5.
38. Badjatia N, Topcuoglu MA, Buonanno FS, et al. Relationship between
hyperglycemia and symptomatic vasospasm alter subarachnoid
hemorrhage. Crit Care Med 2005;33:1603–9.
39. Worthley MI, Shrive FM, Anderson TJ, et al. Prognostic implication of
hyperglycemia in myocardial infarction and primary angioplasty. Am J Med
2007;120(7):643.e1–e7.
40. Fogelholm R, Murros K, Rissanen A, et al. Admission blood glucose and
short term survival in primary intracerebral haemorrhage: a population
based study. J Neurol Neurosurg Psychiatry 2005;76:349–53.
41. Salim A, Hadjizacharia P, Dubose J, et al. Persistent hyperglycemia in severe
traumatic brain injury: an independent predictor of outcome. Am Surg
2009;75(1):25–9.
42. Smith RL, Lin JC, Adelson PD, et al. Relationship between hyperglycemia
and outcome in children with severe traumatic brain injury. Pediatr Crit
Care Med 2012;13(1):85–91.
43. Schmutzhard E, Rabinstein AA. Participants in the International
Multi-Disciplinary Consensus Conference on the Critical Care
Management of Subarachnoid Hemorrhage. Spontaneous
subarachnoid hemorrhage and glucose management. Neurocrit Care
2011;15(2):281–6.
44. Dumont T, Rughani A, Silver J, et al. Diabetes mellitus increases risk of
vasospasm following aneurysmal subarachnoid hemorrhage independent of
glycemic control. Neurocrit Care 2009;11(2):183–9. Epub 2009, May 27.
45. Vogelzang M, Nijboer JMM, van der Horst ICC, et al. Hyperglycemia has a
stronger relation with outcome in trauma patients than in other critically
ill patients. J Trauma 2006;60:873–9.
46. Jeremitsky E, Omert LA, Dunham M, et al. The impact of hyperglycemia
on patients with severe brain injury. J Trauma 2005;58:47–50.
47. McCormick MT, Muir KW, Gray CS, et al. Management of hyperglycemia
in acute stroke: how, when, and for whom? Stroke 2008;39:2177–85.
48. Oddo M, Levine JM, Frangos S, et al. Brain lactate metabolism in humans
with subarachnoid hemorrhage. Stroke 2012; Epub ahead of print.
49. Wyss MT, Jolivet R, Buck A, et al. In vivo evidence for lactate as a neuronal
energy source. J Neurosci 2011;31(20):7477–85.
50. Vanhorebeek I, Langouche L, Van den Berghe G. Glycemic and nonglycemic
effects of insulin: how do they contribute to a better outcome of critical
illness? Curr Opin Crit Care 2005;11:304–11.
51. NICE-SUGAR Study Investigators, Finfer S, Chittock DR, et al. Intensive
versus conventional glucose control in critically ill patients. N Engl J Med
2009;360(13):1283–97. Epub 2009, Mar 24.
52. Shetty S, Inzucchi SE, Goldberg PA, et al. Adapting to the New Consensus
Guidelines for Managing Hyperglycemia During Critical Illness: the
130.e2 Section II—Clinical and Laboratory Assessment
Updated Yale Insulin Infusion Protocol. Endocr Pract 2011;2:1–19. Epub
ahead of print.
53. Van den Berghe G. How does blood glucose control with insulin save lives
in intensive care? J Clin Invest 2004;114:1187–95.
54. Ven den Berghe G, Schoonheydt K, Becx P, et al. Insulin therapy protects
the central and peripheral nervous system of intensive care patients.
Neurology 2005;64:1348–53.
55. Hermans G, Wilmer A, Meersseman W, et al. Impact of intensive insulin
therapy on neuromuscular complications and ventilator dependency in
the medical intensive care unit. Am J Respir Crit Care Med 2007;175:
480–98.
56. Wieske L, Harmsen RE, Schultz MJ, et al. Is critical illness neuromyopathy
and duration of mechanical ventilation decreased by strict glucose control?
Neurocrit Care 2011;14(3):475–81.
57. Gray CS, Hildreth AJ, Sandercock PA, et al. Glucose-potassium-insulin
infusions in the management of post-stroke hyperglycemia: the UK Glucose
Insulin in Stroke Trial (GIST-UK). Lancet Neurol 2007;6:397–406.
58. Bilotta F, Spinelli A, Giovannini F, et al. The effect of intensive insulin
therapy ion infection rate, vasospasm, neurologic outcome, and mortality
in the neurointensive care unit after intracranial aneurysm clipping in
patients with acute subarachnoid hemorrhage: a randomized prospective
pilot trial. J Neurosurg Anesthesiol 2007;19:156–60.
59. Bilotta F, Caramia R, Cernak I, et al. Intensive insulin therapy after severe
traumatic brain injury: a randomized clinical trial. Neurocrit Care
2008;9(2):159–66.
60. Green DM, O’Phelan KH, Bassin SL, et al. Intensive versus conventional
insulin therapy in critically ill neurologic patients. Neurocrit Care
2010;13(3):299–306.
61. Graffagnino C, Gurram AR, Kolls B, et al. Intensive insulin therapy in the
neurocritical care setting is associated with poor clinical outcomes.
Neurocrit Care 2010;13(3):307–12.
62. Tiemessen CA, Hoedemaekers CW, van Iersel FM, et al. Intensive insulin
therapy increases the risk of hypoglycemia in neurocritical care patients.
J Neurosurg Anesthesiol 2011;23(3):206–14.
63. Shan L, Hao PP, Chen YG. Efficacy and safety of intensive insulin therapy
for critically ill neurologic patients: a meta-analysis. J Trauma
2011;71(5):1460–4.
64. Coester A, Neumann CR, Schmidt MI. Intensive insulin therapy in severe
traumatic brain injury: a randomized trial. J Trauma 2010;68(4):904–11.
65. Bruno A, Kent TA, Coull BM, et al. Treatment of hyperglycemia in ischemic
stroke (THIS): a randomized pilot trial. Stroke 2008;39:384–9.
66. Johnston KC, Hall CL, Kissela BM, et al. Glucose Regulation in Acute
Stroke Patients (GRASP) trial: a randomized pilot trial. Stroke
2009;40(12):3804–9.
67. Graetz D, Nagel A, Schlenk F, et al. Role of hyperglycemia and insulin
treatment for cerebral metabolism in aneurysmal SAH. Meeting abstract.
59th Annual Meeting of the German Society of Neurosurgery, Wurzberg,
Germany, 2008.
68. Vespa PM, McArthur D, O’Phelan K, et al. Persistently low extracellular
glucose correlates with poor outcome 6 months after human traumatic
brain injury despite a lack of increased lactate: a microdialysis study.
J Cerebral Blood Flow Metab 2003;23:865–77.
69. Vespa P, Boonyaputthikul R, McArthur DL, et al. Intensive insulin therapy
reduces microdialysis glucose values without altering glucose utilization or
improving the lactate/pyruvate ratio after traumatic brain injury. Crit Care
Med 2006;34:850–6.
70. Oddo M, Schmidt JM, Carrera C, et al. Impact of tight glycemic control on
cerebral glucose metabolism after severe brain injury: a microdialysis study.
Crit Care Med 2008;36:3233–8.
71. Schlenk F, Vajkoczy P, Sarrafzadeh A. Inpatient hyperglycemia following
aneurysmal subarachnoid hemorrhage: relation to cerebral metabolism and
outcome. Neurocrit Care 2009;11(1):56–63. Epub 2009, May 6.
72. Liogier LE, Macyszyn L, Kosty JA, et al. Brain and blood glucose after
subarachnoid hemorrhage: a change in relationship that depends on blood
glucose levels. New Orleans: AANS/CNS Joint Section on Cerebrovascular
Disease Annual Meeting; 2012.
73. Zetterling M, Hillered L, Enblad P, et al. Relation between brain interstitial
and systemic glucose concentrations after subarachnoid hemorrhage.
J Neurosurg 2011;115(1):66–74. Epub 2011, Apr 8.
74. Helbok R, Schmidt JM, Kurtz P, et al. Systemic glucose and brain energy
metabolism after subarachnoid hemorrhage. Neurocrit Care
2010;12(3):317–23.
75. Davidson PC, Steed RD, Bode BW. Glucommander: a computer-directed
intravenous insulin system shown to be safe, simple, and effective in
120,618 h of operation. Diabetes Care 2005;28:2418–23.
76. Louie K, Cheema R, Dodek P, et al. Intensive nursing work schedules and
the risk of hypoglycaemia in critically ill patients who are receiving
intravenous insulin. Qual Saf Health 2010;19(6):e42. Epub 2010,
Aug 4.
77. Campion TR Jr, May AK, Waitman LR, et al. Characteristics and effects of
nurse dosing over-rides on computer-based intensive insulin therapy
protocol performance. J Am Med Inform Assoc 2011;18(3):251–8. Epub
2011, Mar 14.
78. Berger MM, Que YA. Bioinformatics assistance of metabolic and nutrition
management in the ICU. Curr Opin Clin Nutr Metab Care 2011;14(2):
202–8.
79. Krinsley JS. Glycemic variability: a strong independent predictor of
mortality in critically ill patients. Crit Care Med 2008;36:3008–13.
80. Meyfroidt G, Keenan DM, Wang X, et al. Dynamic characteristics of blood
glucose time series during the course of critical illness: effects of intensive
insulin therapy and relative association with mortality. Crit Care Med
2010;38:1021–9.
81. Matsushima K, Peng M, Velasco C, et al. Glucose variability negatively
impacts long-term functional outcome in patients with traumatic brain
injury. J Crit Care 2011. Epub ahead of print.
82. Egi M, Bellomo R, Stachowski E, et al. Hypoglycemia and outcome in
critically ill patients. Mayo Clin Proc 2010;85:217–24.
83. Rubinson L, Diette GB, Song X, et al. Low caloric intake is associated with
nosocomial blood stream infections in patients in the medical intensive
care unit. Crit Care Med 2004;(32):350–7.
84. Heyland DK, Stephens KE, Day AG, et al. The success of enteral nutrition
and ICU-acquired infections: a multicenter observational study. Clin Nutr
2011;30(2):148–55. Epub 2010, Oct 25.
85. Serón-Arbeloa C, Puzo-Foncillas J, Garcés-Gimenez T, et al. A retrospective
study about the influence of early nutritional support on mortality and
nosocomial infection in the critical care setting. Clin Nutr 2011;30(3):
346–50. Epub 2010, Dec 4.
86. Hartl R, Gerber LM, Ni Q, et al. Effect of early nutrition on deaths due to
severe traumatic brain injury. J Neurosurg 2008;109:50–6.
87. Montejo JC. Enteral nutrition-related gastrointestinal complications in
critically ill patients: a multicenter study. Crit Care Med 1999;27:
1447–53.
88. Krishnan JA, Parce PB, Martinz A, et al. Caloric intake in medical ICU
patients: consistency of care with guidelines and relationship to clinical
outcomes. Chest 2003;124:297–305.
89. Brain Trauma Foundation, American Association of Neurological Surgeons,
Congress of Neurological Surgeons. Guidelines for the management of
severe traumatic brain injury. Nutrition J Neurotrauma 2007;24(Suppl 1):
S77–82.
90. Cerra F, Benitez MR, Blackburn, et al. Applied nutrition in ICU patients:
a consensus statement of the American College of Chest Physicians. Chest
1997;111:769–78.
91. Dennis M. Nutrition after stroke. Br Med Bull 2000;56:466–75.
92. Kinoshita K, Moriya T, Utagawa A, et al. Change in brain glucose after
enteral nutrition in subarachnoid hemorrhage. J Surg Res 2010;162(2):
221–4. Epub 2009, Sep 22.
93. Marik PE, Zaloga GP. Early enteral nutrition in acutely ill patients:
a systematic review. Crit Care Med 2001;29:2264–70.
94. Rubinson L, Diette GB, Song X, et al. Low caloric intake is associated with
nosocomial bloodstream infections in patients in the medical intensive care
unit. Crit Care Med 2004;32:350–7.
95. Villet S, Chiolero RL, Bollmann MD, et al. Negative impact of hypocaloric
feeding and energy balance on clinical outcome in ICU patients. Clin Nutr
2005;24:502–9.
96. Nguyen NQ, Besanko LK, Burgstad C, et al. Delayed enteral feeding impairs
intestinal carbohydrate absorption in critically ill patients. Crit Care Med
2012;40(1):50–4.
97. Hartl R, Gerber LM, Ni Q, et al. Effect of early nutrition on deaths due to
severe traumatic brain injury. J Neurosurg 2008;109:50–6.
98. Taylor S, Fettes S, Jewkes C, et al. Prospective, randomized, controlled trial
to determine the effect of early enhanced enteral nutrition on clinical
outcome in mechanically ventilated patients suffering head in-jury.
Crit Care Med 1999;27:2525–31.
99. Davalos A, Ricart W, Gonzalex-Huix F. Effect of malnutrition after acute
stroke on clinical outcome. Stroke 1996;27:1028–32.
100. Kudsk KA. Effect of route and type of nutrition on intestine-derived
inflammatory responses. Am J Surg 2003;185:16–21.
101. McClave SA, Sexton LK, Spain DA, et al. Enteral tube feeding in the
intensive care unit: factors impeding adequate delivery. Crit Care Med
1999;27:1252–6.
 Section II—Clinical and Laboratory Assessment 130.e3
102. McClave SA, Lukan JK, Stefater JA, et al. Poor validity of residual volumes
as a marker for risk of aspiration in critically ill patients. Crit Care Med
2005;33:324–30.
103. Rice TW, Swope T, Bozeman S, et al. Variation in enteral nutrition delivery
in mechanically ventilated patients. Nutrition 2005;21:786–92.
104. Grahm T, Zadroxny D, Harrington T. The benefits of early jejunal
hyperalimentation in the head-injured patients. Neurosurgery 1989;25:
729–35.
105. Montecalvo MA, Steger KA, Farber HW, et al. Nutritional outcome and
pneumonia in critical care patients randomized to gastric versus jejunal
tube feedings. The Critical Care Research Team. Crit Care Med
1992;20(10):1377–87.
106. Peterson S, Chen Y. Systemic approach to parenteral nutrition in the ICU.
Curr Drug Saf 2010;5(1):33–40.
107. Casaer MP, Mesotten D, Hermans G, et al. Early versus late parenteral
nutrition in critically ill adults. N Engl J Med 2011;365(6):506–17. Epub
2011, Jun 29.
108. Borzotta AP, Pennings J, Papasadero B, et al. Enteral versus
parenteral nutrition after severe closed head injury. J Trauma
1994;37:459–68.
109. Snydman DR. Shifting patterns in the epidemiology of nosocomial Candida
infections. Chest 2003;123:500S–3S.
110. Woodcock, NP, Zeigler D, Palmer D, et al. Enteral versus parenteral
nutrition: a pragmatic study. Nutrition 2001;17:1–12.
111. Hadley M, Grahn T, Harrington T, et al. Nutritional support and
neurotrauma: a critical review of early nutrition in forty-five acute head
injury patients. Neurosurgery 1986;19:367–73.
112. Doshi P, Divgiovine B. Effects of early enteral feeding on the outcome of
critically ill mechanically ventilated medical patients on vasopressors. Chest
2006;130 (Supp):101S.
113. Berger MM, Revelly JP, Cayeux MC, et al. Enteral nutrition in critically ill
patients with severe hemodynamic failure after cardiopulmonary bypass.
Clin Nutr 2005;24:124–32.
114. Boullata J, Williams J, Cottrell F, et al. Accurate determination of energy
needs in hospitalized patients. J Am Diet Assoc 2007;107:393–401.
115. American Association for Respiratory Care. AARC clinical practice
guidelines. Metabolic measurement using indirect calorimetry during
mechanical ventilation. Respir Care 1994;39(12):1170–5.
116. McClave SA, Kushner R, Van Way CW 3rd, et al. Nutrition therapy of
the severely obese, critically ill patient: summation of conclusions and
recommendations. JPEN J Parenter Enteral Nutr 2011;35(5 Suppl):88S–96S.
117. Martindale RG, DeLegge M, McClave S, et al. Nutrition delivery for obese
ICU patients: delivery issues, lack of guidelines, and missed opportunities.
JPEN J Parenter Enteral Nutr 2011;35(5 Suppl):80S–7S.
118. Flancbaum L, Choban PS, Sambucco S, et al. Comparison of indirect
calorimetry, the Fick method, and prediction equations in estimating the
energy requirements of critically ill patients. Am J Clin Nutr 1999;69:461–6.
119. Frankenfield DC, Rowe WA, Smith JS, et al. Validation of several established
equations for resting metabolic rate in obese and nonobese people. J Am
Diet Assoc 2003;103:1152–9.
120. de Jong PCM, Wesdorp RIC, Volovics A, et al. The value of objective
measurements to select patients who are malnourished. Clin Nutr
1985;4:61–6.
121. Hall JC. The use of internal validity in the construct of an index of
undernutrition. J Parenter Enteral Nutr 1990;14:582–7.
122. Harris J, Benedict F. A biometric study of basal metabolism in man.
Washington, DC: Carnegie Institute of Washington; 1919.
123. Bistrian BR, Blackburn GL, Sherman M, et al. Therapeutic index of
nutritional depletion in hospitalized patients. Surg Gynecol Obstet
1975;141:512–16.

Hematology and Coagulation
Monisha A. Kumar
Introduction
Bleeding and thrombosis are common complications in the
intensive care unit (ICU); their impact on neurologically ill
patients is of particular concern because these complications
may result in permanent neurologic disability. Sequelae from
unrecognized bleeding can lead to cerebral ischemia, organ
hypoperfusion, multiorgan failure, and death. Thrombotic
complications, such as deep vein thrombosis (DVT), are
common in neurologically injured patients given the associ-
ated limb weakness and complicated by the altered mental
state. Multimodality monitoring in the neurologic ICU may
preempt and thus mitigate the consequences of these common
conditions.
The first part of this chapter briefly reviews the physiology
of erythropoises and the consequences of anemia and throm-
bocytopenia, explores the complications of blood transfusion,
and discusses alternatives to blood product administration in
neurologically critically ill patients. The clinical utility and
application of new monitoring techniques available to evalu-
ate anemia and thrombocytopenia in the ICU are discussed.
The latter portion of this chapter is devoted to the diagnosis,
monitoring, and management of venous thromboembolism.
Hematology
Erythropoiesis and Blood Counts
Erythropoiesis is catalyzed by the hormone erythropoietin
and leads to production of mature erythrocytes. Reticulocytes
are immature erythrocytes that are released into the circula-
tion. Within 1 day, reticulocytes transform into mature red
blood cells that circulate for 100 to 120 days. Ultimately, senes-
cent red blood cells (RBCs) are removed from circulation by
macrophages in the spleen and other reticuloendothelial
tissues. On average, the number of erythrocytes formed is
equivalent to the number destroyed.
The complete blood count (CBC) provides accurate infor-
mation about hemoglobin (Hgb) concentration and cell
counts as well as calculated indices of erythrocyte size and
Hgb content.1 The Hgb concentration varies by both age and
gender (Table 15-1). In the average adult male, the Hgb con-
centration varies between 13 and 17.5 g/dL; in women it
ranges from 12 to 16 g/dL. A high white blood cell (WBC)
count, lipemia, or a precipitating monoclonal protein can
result in a spuriously high Hgb. High altitude will result in an
increased Hgb concentration; the rise in Hgb is proportional
to the magnitude of elevation. Women of childbearing age
© Copyright 2013 Elsevier Inc. All rights reserved.
Chapter
15  
II
have an average Hgb 10% lower than men of the same age.1
Because of the wide range of normal Hgb and hematocrit
(Hct) values, it often is difficult to diagnose mild anemia.
The RBC count is the number of RBCs per unit volume. It
generally ranges between 4.2 and 5.9 million cells per micro-
liter.1,2 RBCs are the most common cell type present in blood.
They are smaller than white blood cells but larger than plate-
lets. A normal platelet count is between 150,000 and
400,000/µL. Thrombocytopenia is defined as a platelet count
less than 150,000/µL.2
The Hct is the ratio of RBCs to blood volume. With modern
laboratory equipment, the Hct is calculated and not measured
directly. The mean concentration of Hgb within the red cell
population (mean corpuscular hemoglobin concentration
[MCHC] is the quotient of the Hgb divided by the Hct. The
red cell distribution width (RDW), another parameter of red
cell size, measures the variability in red cell size of circulating
erythrocytes.3 For example, the etiology of anemia with a low
MCV and a high RDW is most likely iron deficiency; even
though the anemia is microcytic, the variability of RBC size is
high. In contrast, a macrocytic anemia would yield both a high
MCV and RDW. Together with the reticulocyte count, the
MCV and RDW can narrow the differential diagnosis of
anemia.
The reticulocyte count (RC) measures the number of
immature RBCs in circulation and is a marker of RBC produc-
tion. This count may be inaccurate in the presence of nucle-
ated RBCs or nuclear debris in the peripheral blood. When
the RC is reported as a percentage, it should be adjusted for
the total number of RBCs present. This correction can be
made by multiplying the reticulocyte count by the Hct and
divided by age and gender specific normative Hct values.
Alternatively, the percent of reticulocytes can be multipied by
the red cell count to determine the absolute RC.2 The normal
absolute reticulocyte count (ARC) is between 25,000 and
75,000/µL.2 In the presence of anemia, an ARC less than
75,000/µL indicates a hypoproliferative process, whereas an
ARC greater than 100,000/µL indicates hemolysis or an appro-
priate erythropoietin response. RCs between 75,000 and
100,000/µL must be interpreted for the degree of anemia
present.2
Review of the peripheral blood smear remains an informa-
tive diagnostic tool. Not only can it confirm findings of an
automated CBC, but it also may suggest a marrow disease or
hemolytic process. Microcytic, hypochromic cells may suggest
iron deficiency or thalassemia, whereas macrocytosis with
ovalocytes may suggest a megaloblastic anemia. The periph-
eral smear may demonstrate echinocytes, seen in uremia;
131
132 Section II—Clinical and Laboratory Assessment

Table 15.1  Reference Hematologic Values in Children and Adults*

Hemoglobin Hematocrit Red Blood Cell
(g/dL) (%) Count (1012/L) MCV (fL) MCH (pg) MCHC (g/dL)
Age Mean −2 SD Mean −2 SD Mean −2 SD Mean −2 SD Mean −2 SD Mean −2 SD
Birth (cord blood) 16.5 13.5 51 42 4.7 3.9 108 98 34 31 33 30
1-3 days (capillary) 18.5 14.5 56 45 5.3 4.0 108 95 34 31 33 29
1 week 17.5 13.5 54 42 5.1 3.9 107 88 34 28 33 28
2 weeks 16.5 12.5 51 39 4.9 3.6 105 86 34 28 33 28
1 month 14.0 10.0 43 31 4.2 3.0 104 85 34 28 33 29
2 months 11.5 9.0 35 28 3.8 2.7 96 77 30 26 33 29
3-6 months 11.5 9.5 35 29 3.8 3.1 91 74 30 25 33 30
0.5-2 years 12.0 10.5 36 33 4.5 3.7 78 70 27 23 33 30
2-6 years 12.5 11.5 37 34 4.6 3.9 81 75 27 24 34 31
6-12 years 13.5 11.5 40 35 4.6 4.0 86 77 29 25 34 31
12-18 YEARS
Female 14.0 12.0 41 36 4.6 4.1 90 78 30 25 34 31
Male 14.5 13.0 43 37 4.9 4.5 88 78 30 25 34 31
18-49 YEARS
Female 14.0 12.0 41 36 4.6 4.0 90 80 30 26 34 31
Male 15.5 13.5 47 41 5.2 4.5 90 80 30 26 34 31

Data from Dallman PR: Blood-forming tissues. In: Rudolph A, editor. Pediatrics. 16th ed. New York: Appleton-Century-Crofts; 1977. p. 1111.
MCH, Mean corpuscular hemoglobin; MCHC, mean corpuscular hemoglobin concentration; MCV, mean corpuscular volume.
*These data have been compiled from several sources. Emphasis is given to studies employing electronic counters and to the selection of populations that are likely
to exclude individuals with iron deficiency. The mean −2 SD can be expected to include 95% of the observations in a normal population.

acanthocytes, observed in liver disease; or schistocytes, found
in disseminated intravascular coagulation (DIC). Special
stains of the peripheral smear may assist in the diagnosis of
anemia, particularly when there is rouleaux formation and
automated counters may be inaccurate.
Early after acute hemorrhage, red cell mass and plasma
volume contract at the same rate; therefore, there may not be
any observed reduction in Hgb concentration or Hct. The
platelet count often increases after acute blood loss and there
often is an increase in reticulocyte count. Because young
erythrocytes are larger than mature ones, the MCV is usually
slightly elevated. Bleeding is usually accompanied by an
increase in indirect bilirubin, which reflects the catabolism of
heme from extravasated RBCs.
mechanical ventilation and may be particularly detrimental
for patients in the neurocritical care unit (NCCU) with
reduced cerebrovascular reserve.
Anemia is associated with increased ICU length of stay and
increased mortality in critically ill patients.5 In surgical ICU
patients with cardiac disease, anemia is associated with
increased risk of death.8 Anemia also is associated with the
development of left ventricular hypertrophy in patients with
end-stage renal disease, and its correction is associated with
fewer cardiovascular events.9 Preoperative anemia is associ-
ated with worse surgical outcomes10 and preprocedural
anemia among patients undergoing percutaneous coronary
intervention is associated with increased adverse in-hospital
outcomes.11

Anemia Causes of Anemia in the Intensive

Anemia in the Intensive Care Unit
Anemia is a common problem among critically ill patients.
According to the World Health Organization, anemia is clin-
ically defined as an Hgb concentration less than or equal to
12 g/dL for women and 13 g/dL for men.4 Nearly two thirds
of patients are anemic on admission to the ICU5 and
between 70% and 95% of patients become anemic by day
3.6,7 Anemia is potentially deleterious because lower Hgb
levels decrease oxygen carrying capacity of the blood and
may reduce tissue oxygenation. This is detrimental for
patients with critical illness who often have increased meta-
bolic demand for oxygen from infection, sepsis, fever, and
Care Unit
The causes of anemia in the ICU are varied and may occur
concurrently. These include sepsis, trauma, burns, infections,
hemolysis, immune-associated iron deficiency, decreased
endogenous erythropoietin production, renal failure, liver
disease, malignancy, collagen vascular disease, nutritional
deficiency, gastrointestinal bleeding, drug-related causes, and
phlebotomy. Phlebotomy is the most significant modifiable
factor associated with anemia in the ICU and is a predictor of
blood transfusion12; it accounts for 40 to 70 mL per day of
blood loss early in the ICU period.13,14 Smoller observed that
the amount of blood drawn per day varied from 12 mL per
day on the ward, to 33 mL a day in the ICU in patients without
 Section II—Clinical and Laboratory Assessment 133
an arterial line, to 74 mL a day in ICU patients with an arterial
line.15 For every 100 mL of blood removed by phlebotomy,
Hgb and Hct levels fall by 0.7 g/dL and 1.9%, respectively.12
ICU patients with evidence of inflammation and multior-
gan failure develop anemia despite a lack of active bleeding.
This “anemia of critical illness” results from underproduction
of erythrocytes and is similar to anemia of chronic disease,
which is thought to be due to inflammation. Inflammatory
cytokines, such as tumor necrosis factor-α (TNFα),
interferon-α (IFNα), and interleukin-1 (IL-1), suppress eryth-
ropoiesis and are central in the pathogenesis of anemia in
these patients.16 These inflammatory mediators inhibit
hypoxia-induced (as opposed to anemia-induced) erythro-
poietin (EPO) production by up to 89%.17
Proinflammatory cytokines not only induce bone marrow
suppression of erythropoiesis but also aggravate bleeding and
disrupt iron metabolism. These cytokines exacerbate intesti-
nal bleeding through increased permeability of the intestinal
wall. They alter iron metabolism by stimulating phagocytosis
of RBCs. More than 90% of ICU patients have low levels of
serum iron (Fe), total iron binding capacity (TIBC), and Fe-
to-TIBC ratio, but have normal or elevated serum ferritin.18,19
Despite low levels of circulating iron, EPO levels remain only
modestly increased and are not markedly elevated. This
blunted EPO response results from the inhibition of EPO
transcription by inflammatory mediators. Rogiers et al. found
that at the same serum level of Hgb, ambulatory anemic
patients had an eight times higher level of EPO than did septic
ICU patients.18
The etiology of anemia in the ICU is multifactorial and can
be difficult to determine. The diagnosis of anemia in the ICU
is complicated because serial monitoring of hematologic pro-
files (i.e., phlebotomy) can exacerbate the condition.
Monitoring Anemia in the Intensive
Care Unit
Modern laboratory testing and the availability of automated
cell counters provide a wealth of diagnostic information for
interpretation.
The extended differential count (EDC) quantifies other cell
types in the peripheral blood including nucleated RBCs
(NRBCs). Although NRBCs typically are only seen in neo-
nates, their presence has been identified in adults, notably in
patients with septicemia, massive hemorrhage, and severe
hypoxia.20,21 When identified in adults, the presence and
burden of NRBC may correlate with mortality and poor
outcome. For example, Stachon observed a mortality rate of
44% in surgical patients with NRBCs compared with 4.2% for
patients without.21
Other parameters of cellular analysis available through
automated analyzers include the immature reticulocyte frac-
tion (IRF), mean reticulocyte hemoglobin content (CHr),
mean reticulocyte volume (MCVr), fragmented RBC (FRBC)
count, and the immature platelet fraction (IPF). IRF is an early
and sensitive index of erythropoiesis. When characterized
with total reticulocyte count, it can discriminate between
anemias with increased erythropoiesis (increased total reticu-
locyte count and IRF) such as hemolytic anemias; decreased
marrow function (decreased total reticulocyte count and IRF);
and conditions such as acute infections and myelodysplastic
syndrome with dissociation of total reticulocyte count and
IRF (total reticulocyte count low, IRF high).20 It also is useful
to identify resumption of marrow activity after bone marrow
transplantation or chemotherapy.
The CHr and the MCVr are reticulocyte indices that reflect
the adequacy of iron stores available for erythropoiesis. The
CHr is important because it can help identify iron-deficiency
anemia in the face of the acute phase response when total iron
stores, as evidenced by an elevated ferritin and transferrin, are
adequate. This may be seen in anemia of chronic disease. The
MCVr has not been well studied, but has been used to evaluate
possible EPO abuse in sports.22 The main limitations of these
new indices include limited availability of the machines and
variable sensitivity of the analyses.
FRBCs can be found in a variety of diseases from microan-
giopathies to cardiovascular disorders (valve prostheses and
endocarditis). Microangiopathies need immediate diagnosis
and treatment, and expedient identification and quantifica-
tion is paramount. The sensitivity of the new analyzers to
diagnose microangiopathy is very high (>90%), but the speci-
ficity is low (20%-50%).23,24 Therefore in certain populations,
the FRBC count may be a good screening test for microangi-
opathies, but further study is warranted.
In addition to new laboratory indices of red cell size and
function, new monitoring devices permit measurements of
total Hgb concentration. In contrast to the standard cyano-
methemoglobin method of Hgb determination, co-oximeters
use spectrophotometry to determine Hgb concentration. The
methodology of Hgb measurement is similar to that of oxy-
hemoglobin measurement by conventional pulse oximetry,
except that instead of two wavelengths of light, multiple wave-
lengths are transmitted through the finger to measure the light
absorbance characteristics of Hgb.25 Point-of-care co-oximeters,
such as the HemoCue (Angelholm, Sweden), allow for capil-
lary measurement of Hgb concentration at the bedside,
whereas noninvasive co-oximeters, such as the Radical-7
(Masimo Corp, Irvine, CA) device, provide continuous Hgb
measurements. The accuracy of these devices compare well
against standard laboratory assays, although the Radical-7
may give lower readings than the HemoCue during surgery.26,27
Hgb measurements with these devices have the additional
benefit of providing rapid results that may facilitate monitor-
ing in the operating room, ICU, and emergency department.
Anemia in the Neurocritical Care Unit
Anemia is common in the NCCU and, depending on the defi-
nition, occurs in nearly 40% of patients.28 It has been identi-
fied as an independent factor associated with death or severe
disability in several disorders including subarachnoid hemor-
rhage (SAH), particularly in patients who develop delayed
cerebral ischemia, traumatic brain injury (TBI), or intracere-
bral hemorrhage (ICH).28-34 However, there remains variation
in how anemia is managed.
In the uninjured brain, vasodilation of the cerebral arteries
compensates for anemia-related reduction in oxygen carrying
capacity; therefore, frank brain hypoxia does not typically
occur at Hgb concentrations greater than 6 g/dL.35 The ability
of cerebral vessels to vary their caliber to maintain constant
perfusion is termed cerebral autoregulation. With impairment
of cerebral autoregulation, such as may occur after SAH or
TBI, tissue hypoxia may occur at higher Hgb concentrations.
Consistent with this, fewer ischemic events are observed in
134 Section II—Clinical and Laboratory Assessment
SAH patients with higher Hgb concentrations.33 However, the
Hgb threshold below which cerebral metabolic dysfunction
occurs is only beginning to be elucidated.
A fundamental tenet of neurocritical care is that ischemia
is an important cause of brain injury. Anemia and compro-
mised oxygen delivery may exacerbate ischemia; therefore
anemia may represent a therapeutic target in the NCCU.
However, the ramifications of anemia may vary by the type of
brain injury. The consequences of anemia after SAH, particu-
larly among patients with severe vasospasm or delayed cere-
bral ischemia (DCI), may be dire. In contrast, the significance
of cerebral ischemia in other neorocritically ill subpopulations
is less clear; therefore, anemia may be of less concern.
Optimal Hemoglobin Threshold in the
Neurocritical Care Unit Population
Several physiology studies in neurocritically ill patients have
sought to define an Hgb threshold by measuring markers of
cerebral metabolism.35-37 Brain tissue oxygen tension (PbtO2)
monitoring and lactate-to-pyruvate ratio (LPR) measure-
ments from cerebral microdialysis demonstrated increased
brain hypoxia and cell energy dysfunction when Hgb levels
were less than 9 g/dL in SAH patients.37 However, the physi-
ologic data in TBI are conflicting.37-39 Although TBI patients
frequently have pathologic evidence of cerebral infarction,
premortem evidence of ischemia is debatable.40-42 Early evi-
dence demonstrated compromised cerebral blood flow sug-
gestive of ischemia; however, it was later demonstrated that
the altered blood flow correlated with decreased cerebral
metabolism.40,43,44 Regardless, anemia appears to be a predictor
of poor outcome, although transfusion of PRBCs does not
always appear to mitigate its effect.45
The treatment of anemia in the NCCU is unclear in part
because of the heterogeneous pathophysiology of diseases
treated therein. Moreover, transfusion trials in general ICUs
included very few or excluded neurologically critically ill
patients and therefore results cannot be easily extrapolated
from them.46,47 Transfusion studies in patients with active
cardiac ischemia have not demonstrated consistent benefit
from a liberal transfusion approach.5,8 It is unclear whether
this is analogous to patients with cerebral ischemia.
Thrombocytopenia
Thrombocytopenia in the Critically Ill
The importance of platelets and their role in thrombus forma-
tion is clear, but their significance in sepsis is being eluci-
dated.48 Thrombocytopenia occurs in 13% to 58% of ICU
patients49-53; 23% have at least one platelet count less than
100,000/mm3 and 10% had counts less than 50,000 mm3.50
Sepsis is considered the major independent risk factor for
thrombocytopenia in the ICU. Other factors associated with
thrombocytopenia are episodes of bleeding or transfusion,
longer ICU stays, and Acute Physiology and Chronic Health
Evaluation II (APACHE II) scores greater than 15.52 The devel-
opment of thrombocytopenia during the ICU stay is associ-
ated with decreased ICU survival and may have a greater
implication for prognosis than admission thrombocytope-
nia.51,53,54 For example, Cawley et al. observed that a drop in
platelet count to less than 50% of admission levels was
associated with greater mortality than admission APACHE II,
Simplified Acute Physiology Score (SAPS), or multiple organ
dysfunction syndrome (MODS) scores.49
ICU-related thrombocytopenia is primarily caused by
peripheral loss, destruction, and margination of platelets
rather than bone marrow hypoplasia or nutritional deficiency.
Sepsis accounts for 48% of cases. Other causes include liver
disease, hypersplenism, consumption, platelet destruction,
DIC, medications, immune phenomena, and intravascular
devices.55 The cause of a low platelet count in the ICU can be
difficult to determine and in more than 25% of critically ill
patients the cause is multifactorial. Pseudothrombocytopenia
occurs when platelets clump in association with an ethylene-
diamine tetra-acetic acid (EDTA)–dependent platelet aggluti-
nin.52 An accurate platelet count can be determined by
collecting the sample in a citrated collection tube or by using
a heparinized blood sample. Postresuscitative hemodilution
can result in thrombocytopenia (i.e., after blood transfusion
or with crystalloid or colloid replacement therapy for blood
loss). The post-transfusion reduction in platelet count is asso-
ciated with both splenic sequestration of platelets and a reduc-
tion in viable platelets in stored blood.
Platelets play a complex role in sepsis and multiorgan dys-
function. Platelets of septic patients may display increased
adhesion and aggregation.56,57 Bacteria and bacterial products
may affect platelet function. Lipopolysaccharide increases
platelet aggregation in some animal models, but has the oppo-
site effect on human platelets in vitro.48 Lipoteichoic acid, a
cellular membrane component of gram-positive bacteria,
inhibits platelet aggregation in human platelets.48 Escherichia
coli endotoxin may indirectly reduce platelet responsiveness.48
Drug-Induced Thrombocytopenia
Drug-induced thrombocytopenia (DIT) is a common
nonimmune-mediated disorder, associated with bone marrow
suppression and accelerated platelet destruction and defined
as a platelet count between 50 and 150,000 × 106/L with any
degree of spontaneous bleeding.58 It is observed in 19% to
25% of ICU patients59,60 and associated with medications such
as cytotoxic chemotherapeutic agents,61,62 thiazide diuretics,
ethanol, tolbutamide, and bleomycin. DIT typically develops
2 to 3 days after a previously administered medication, and 7
days after a virgin drug. Discontinuation of the medication
normalizes the platelet count by 5 to 8 days.63 A major bleed-
ing event can occur in 4% of patients,63 and these patients may
benefit from intravenous immunoglobulin (IVIg), plasma-
pheresis, or platelet transfusion64 but not corticosteroids.65
Drug-Induced Immune Thrombocytopenia
Drug-induced immune thrombocytopenia (DITP) is a disor-
der characterized by the development of drug-dependent
antibodies that bind to glycoproteins on the surface of plate-
lets and accelerate their destruction in the presence of the
specific drug. This type of disorder may be more severe than
DIT. Hundreds of drugs have been associated with this type
of thrombocytopenia; the most common are heparin, quinine
(cinchona alkaloids), penicillin, sulfonamides, nonsteroidal
anti-inflammatory drugs (NSAIDs), anticonvulsants, anti-
rheumatics, oral hypoglycemics, gold salts, diuretics, rifampi-
cin, ranitidine, and the glycoprotein (GP) IIb/IIIa inhibitors

(e.g., abciximab, tirofiban, or eptifibatide).58 The precise
mechanisms by which various drugs incite the immune
response are beyond the scope of this chapter. Some patients
can make both drug-independent and drug-dependent anti-
bodies during the course of medication exposure. The drug-
independent antibodies (autoantibodies) usually are transient
but if they persist, can result in a chronic autoimmune throm-
bocytopenic purpura. When agents such as GPIIb/IIIa inhibi-
tors are used, platelet counts should be frequently monitored.
If the platelet count is less than 50,000/µL, the drug should be
discontinued. If the platelet count is less than 10,000/µL and
there is severe bleeding or if an invasive procedure is required,
then platelet transfusion should be considered.66 Readminis-
tration of GPIIb/IIIa at a later date is not recommended
because the rechallenge might prove more severe than the
initial event.
Heparin-Induced Thrombocytopenia
Heparin-induced thrombocytopenia (HIT) is an anticoagulant-
induced prothrombotic immune disorder caused by heparin-
dependent platelet-activating IgG antibodies that recognize
complexes of platelet factor 4 (PF4) bound to heparin.67,68 HIT
should be suspected when the platelet count falls to less than
50% of the baseline level or an absolute number of less than
150,000/µL between days 5 and 14 of heparin exposure.
Several factors are associated with the development of HIT:
(1) the dose or duration of heparin use; (2) the type of heparin
administered; (3) the patient population; and (4) patient
gender69 (Table 15.2). A greater risk of HIT is observed in
patients exposed to a high dose of heparin or a long duration
of treatment, with unfractionated heparin rather than low
molecular weight heparin (LMWH), with surgical rather than
medical patients, and in females.
PF4 is a tetrameric protein, involved with hemostasis,
immunoregulation, and angiogenesis. Heparin binding
induces a conformational change in PF4. HIT differs from
most other drug-induced immune thrombocytopenias in that
the induced antibodies engage the Fc receptor (not Fab region)
and thereby activate platelets (i.e., they promote thrombosis
and a hypercoagulable state, which is characteristic of HIT).
PF4 also neutralizes the anticoagulant effect of heparin, which
exacerbates the prothrombotic milieu.
Table 15.2  Risk Factors for Immune Heparin-Induced Thrombocytopenia
Risk Factor Odds Ratio
Duration of heparin use ∼20-100
(>1 week vs. <1 day)
Type of Heparin ∼10-15
UFH > LMWH >
fondaparinux
Type of patient (surgery ∼3-4
> medical > pregnancy)
Gender (female > male) ∼1.5-2.0
Used with permission from Warkentin TE. Heparin-induced thrombocytopenia. Hematol Oncol Clin North Am 2007;21(4):592.
HIT, Heparin-induced thrombocytopenia; LMWH, low molecular weight heparin; UFH, unfractionated heparin.
Section II—Clinical and Laboratory Assessment 135
Clinical complications of HIT include both venous and
arterial thromboses.55 Arterial thrombosis is more common in
those with known cardiac disease, whereas DVT is seen in
postoperative patients. Cerebral sinus thrombosis, venous
limb gangrene, and hemorrhagic adrenal infarction also are
reported.67 Concomitant prothrombotic risk factors include
diabetes mellitus, malignancy, systemic lupus erythematosus
(SLE), antiphospholipid antibody syndrome, indwelling cath-
eters, and trauma.55 The incidence of stroke is 3.1% among
patients with HIT and is more common in females, in patients
with more severe thrombocytopenia, and early in the course
of HIT. Myocardial infarction can occur in 3% to 5% of
patients and amputation in as many as 20% of patients with
HIT and limb thrombosis. In SAH, HIT may occur in up to
15% of patients, and its risk is increased in Fisher grade 3
patients, females, and perhaps endovascular treatments.70 The
incidence of hypodensity on computed tomography (CT)
scans and worse outcome is increased.
PF4 antibodies are transient and are undetectable a median
of 50 to 85 days after an episode of HIT.71 In some patients
antibodies remain detectable at low levels for several months.
If heparin is readministered to a patient with high levels of
antibodies, development of thrombocytopenia can be rapid.
This is unlikely more than 100 days from exposure. The “4
T’s”—Thrombocytopenia, Timing, Thrombosis, and oTher
plausible explanations for the clinical scenario—is an eight-
point clinical scoring system that evaluates the likelihood of
developing HIT (Table 15.3).72 It has a high negative predic-
tive value and can help rule out HIT. If HIT is suspected all
heparin sources must be stopped, preferably before the dem-
onstration of HIT antibodies. The direct thrombin inhibitor
family of anticoagulants should replace heparin because treat-
ment with LMWH does not fully eradicate the risk of HIT.
Thrombotic Microangiopathies: Thrombotic
Thrombocytopenic Purpura
Thrombotic thrombocytopenic purpura (TTP) is a rare but
potentially fatal disorder that mostly affects adult women. It
is characterized by red cell fragmentation (microangiopathy),
hemolytic anemia, and thrombosis (consumptive thrombocy-
topenia) that lead to ischemic organ damage.73 The predilec-
tion for thrombosis is a consequence of abnormal numbers of
Comment
Odds ratio (OR) estimated as follows: risk for HIT postcardiac surgery (with
UFH prophylaxis for >1 week) ∼2% versus risk for “delayed-onset HIT”
(without prophylaxis) ∼0.02%-0.1%
Difference in risk for HIT between heparin types is established for postsurgical
thromboprophylaxis (UFH vs. LMWH) and is more pronounced in women.
ORs shown are for UFH versus LMWH (woman, postsurgical prophylaxis) [18]
Highest reported frequencies of HIT are in postsurgical thromboprophylaxis
(OR shown is for surgery vs. medical) [18]
Difference in risk for HIT between women and men has only been established
for UFH thromboprophylaxis
136 Section II—Clinical and Laboratory Assessment

Table 15.3  Estimating the Pretest Probability of Heparin-Induced Thrombocytopenia:

The “4 T’s” Scoring System
Points (0, 1, or 2 for each of 4 categories: maximum possible score = 8)
Date : 2 1 0
Thrombocytopenia* >50% platelet decrease to 30%-50% platelet count decrease <30% platelet
Score = _____________ nadir ≥20 × 109/L (or >50% directly resulting from decrease or
surgery) or nadir 10 − 19 × 109/L nadir <10 × 109/L
Timing† of platelet count Days 5-10 onset† or ≤1 day Consistent with days 5-10 decrease, Platelet count
decrease, thrombosis, or (with recent heparin but not clear (e.g. , missing decrease ≥4 days
other sequelae (first day of exposure within past 5-30 platelet counts), or ≤1 day (heparin without recent
heparin course = day 0) days) exposure within past 31-100 days), heparin exposure
Score = ________ or platelet decrease after day 10
Thrombosis (including adrenal Proven new thrombosis, or Progressive or recurrent thrombosis None
infarction) or other sequelae skin necrois (at injection or erythematous skin lesions
(e.g., skin lesions) site), or post-IV heparin (at injection sites), or suspected
Score = ____________ bolus anaphylactoid reaction thrombosis (not proven)
OTher cause for No explantion for platelet Possible other cause is evident Definite other
thrombocytopenia count decrease is evident cause is present
Score = ____________
Total score = _______________ Pretest probability score: 6-8 = high, 4-5 = intermediate, 0-3 = low

IV, Intravenous.
*Changes to score can occur, based on new information (e.g., further decrease in platelets, new thrombosis, other causes for platlet decrease).
†
First day of immunizing heparin exposure considered day 0; the day the platelet count begins to decrease is considered the day of onset of thrombocytopenia (it
generally takes 1 to 3 more days until an arbitrary threshold that defines thrombocytopenia is passed. Usually, heparin administered at or near surgery is the most
immunizing situation).

von Willebrand multimers (µL-vWF) in the circulation due Disseminated Intravascular

to altered vWF homeostasis. vWF is normally released in
response to prothrombotic stimuli, and persistence of these
multimers leads to platelet activation and clumping with sub-
sequent microvascular thrombosis. The protease, ADAMTS13,
a disintegrin and metalloproteinase with a thrombospondin
type 1 motif, member 13 normally cleaves the µL-vWF into
smaller peptides.73 When the protease activity is decreased or
absent there is accumulation of the large multimers. The activ-
ity of ADAMTS13 can be decreased by many different factors
but most cases of acquired adult TTP result from autoanti-
bodies to ADAMTS13. Congenital TTP results from complete
absence of this protease.
TTP often is seen with infection, pregnancy, or estrogen-
containing compounds. General clinical features include fever,
anemia, thrombocytopenia, renal failure, and neurologic
symptoms from multiorgan damage due to microvascular
thrombi. Patients present with complaints of weakness,
malaise, nausea, vomiting, abdominal pain, hallucinations,
and seizures. Diagnostic laboratory tests for TTP include
hemoglobin, platelet count, peripheral smear (to document
fragmentation), lactate dehydrogenase (LDH), bilirubin, and
reticulocyte count. ADAMTS13 activity level and antibody
tests may be helpful. The differential diagnosis for TTP
includes DIC, the HELLP syndrome (hemolytic anemia, ele-
vated liver enzymes, low platelets), hemolytic uremic syn-
drome (HUS), malignant hypertension, preeclampsia, and
some infections. Treatment is aimed at prompt removal of
circulating antibodies (by plasma exchange), replacement of
ADAMTS13 (by administration of FFP, cryoprecipitate) and
immunosuppression (with steroids, cyclosporine, vincristine,
cyclophosphamide, or rituximab).73
Coagulation
Disseminated intravascular coagulation (DIC) is a syndrome
of thrombosis and systemic hemorrhage caused by an abnor-
mally activated extrinsic clotting cascade or tissue factor
system. Tissue factor (TF) is a tissue membrane glycoprotein,
which binds to factor VII. Normally, the TF/factor VIIa
complex activates factors X and XI that initiate clotting. In the
face of systemic inflammation, circulating cytokines, IL-1,
IL-6, and TNFα cause the constitutive release of tissue factor
and down-regulate thrombomodulin, an endothelial cell
surface receptor that activates anticoagulants and inhibits
factors Va and VIIa.74 The down-regulation of thrombomodu-
lin converts the antithrombotic blood vessel endothelial
surface to a prothrombotic one. The early thrombotic phase
causes widespread fibrin deposition, which is followed by con-
sumption of platelets and clotting factors that culminate in
fibrinolysis and hemorrhage.
The diagnosis of DIC should always be suspected if the
patient has acrocyanosis, purpura, dusky cyanosis of the nose,
ears, and genitalia, or is oozing blood from venipuncture or
arterial catheter sites. Neurologic complications of DIC
include coma, encephalopathy, SAH, small and large vessel
stroke, embolic infarction from nonbacterial thrombotic
emboli, and intracerebral hemorrhage.73 DIC is a major cause
of stroke in medical ICUs, is a frequent complication in ter-
minally ill patients, and may aggravate neurologic deficit after
TBI.74 Several scoring systems help diagnose DIC. One such
system developed by the International Society of Thrombosis
and Haemostasis combines platelet count, the ratio of fibrin
monomers to fibrin degradation products, a prolonged
 Section II—Clinical and Laboratory Assessment 137
prothrombin time (PT), and fibrinogen level to determine a
DIC score.75 The sensitivity and specificity of this scoring
system is greater than 90%. Other scoring systems includes a
D-dimer test and can help predict mortality.76
Monitors of Platelet Function
Circulating platelets vary in size, metabolism, and activity.
Newly released platelets contain RNA and owing to the simi-
larity to reticulocytes, are called reticulated platelets. The
largest are more reactive and secrete a greater quantity of
thrombogenic factors.20 Automated counters can quantify
reticulated platelets and determine platelet size variability
and mean platelet volume. Fluorescent dyes and cytometers
have allowed quantification of reticulated platelets. Increased
presence of reticulated platelets may be seen in peripheral
platelet destruction and acute blood loss, whereas a de-
creased number may be seen in marrow hypoplasia or cyto-
toxic chemotherapy. The increase in immature platelet
fraction (IPF) may precede the rise in platelet count by
days,77 and may serve to limit prophylactic platelet trans­
fusion after chemotherapy.78 Moreover, an increase in reticu-
lated platelets may be associated with an increased risk of
thrombosis in thrombocytosis and myeloproliferative
disorders.85
New automated counters can provide the mean platelet
volume (MPV) and platelet distribution width (PDW), which
are analogous to the erythrocyte indices. There is a nonlinear
inverse relationship between MPV and platelet concentra-
tion; MPV decreases with higher platelet counts.80 Therefore
the MPV should be assessed relative to platelet number. MPV
is also a known marker of platelet activation. MPV elevation
is associated with increased risk of myocardial infarction
(MI) in patients with coronary artery disease (CAD) and
increased mortality after acute MI.81,82 Studies have demon-
strated MPV elevations in acute ischemic stroke82-85; more-
over, some studies suggest that increased MPV is related to
poor functional outcome, although this remains controver-
sial.86,87 Although these parameters provide novel informa-
tion, there are many limitations to the use of platelet indices.
The anticoagulant used for CBC determination incites non-
uniform structural changes in platelets. As a consequence, the
MPV may increase or decrease depending on the method of
measurement. There is no mathematical correction for this
phenomenon, and therefore platelet indices are still consid-
ered experimental.
Platelet Function Assays
Platelet aggregation with a platelet aggregometer is the gold
standard to test platelet function. However, this method is
labor intensive and requires technologic expertise; therefore,
it is unsuitable for bedside monitoring. Since 2000, devices
that can be used in the laboratory or at the point of care
(POC) have become available; however, the results do not
always correlate with the gold standard, aggregometry, and
more studies need to define the role of these new assays.
A variety of POC testing (POCT) devices are available for
platelet function. These devices have been developed primar-
ily to measure the effect of antiplatelet medications and not
necessarily to diagnose platelet function defects. The Platelet
Function Analyzer 100 (PFA-100, Siemans Corporation, NY)
is an optical detection device that measures closure time, or
time to the cessation of the blood flow through a channel
coated with collagen and a platelet activator, either adenosine
diphosphate (ADP) or epinephrine. Eight hundred microliters
of citrated whole blood is required for the analysis. Although
PFA-100 results are abnormal in both congenital platelet
abnormalities88 and moderate-severe von Willebrand’s disease
(vWD), it is not sensitive enough to serve as an adequate
screening test to exclude all congenital abnormalities or mild
vWD. A normal ADP closure time has been found to be a
good negative predictor of bleeding89; however, an elevated
closure time has a poor positive predictive value. Other POCT
is intended primarily to monitor antiplatelet medications such
as aspirin, clopidogrel, or glycoprotein IIb/IIIa inhibitors.
Correlation with platelet aggregometry has been poor; one
study found the sensitivity of the PFA-100 and the Verify-Now
(Accumetrics, San Diego, CA) to be 62% and 39%, respec-
tively. Another study of 100 patients treated with 75 to 100 mg
of aspirin for CAD demonstrated variable levels of aspirin
resistance (12%-22%) when measured by four tests of platelet
aggregability.90 Only 2% of patients were aspirin resistant by
all tests, and correlation between tests was poor.
Transfusion in the Critically Ill
Packed Red Blood Cell Transfusion
Blood transfusion in the ICU is common due in large part to
the increased severity of illness, the burden of chronic diseases,
improved ICU management, and blood-intensive surgeries.91
Annually, approximately 15 millions units of blood are
donated, and 13 million are transfused.92,93 Up to 85% of
patients remaining in the ICU more than 1 week are trans-
fused, with an average total transfusion of 9.5 units.6 Overall
packed red blood cell (PRBC) transfusion rate among ICU
patients is between 37% and 50%.7,13,14
Since 2000, concern about the safety of PRBC transfusion,
including immunosuppressive and microcirculatory compli-
cations, has led to a reappraisal of transfusion practices. In
1999, the Transfusion Requirements in Critical Care (TRICC)
trial, a randomized study, examined whether a restrictive
(Hgb threshold of 7 g/dL) and liberal (Hgb threshold of 10 g/
dL) transfusion strategy in the ICU was associated with mor-
tality.5 The primary endpoint of 30-day mortality was similar
between two groups and in subgroups, including in patients
with cardiac disease, severe infections or septic shock, or
trauma. However, in patients with an APACHE II score less
than 20, or those younger than 55 years of age, a restrictive
approach was associated with less mortality. The authors rec-
ommended that critically ill patients with isovolemic anemia
receive PRBC transfusion when the Hgb concentration is less
than 7 g/dL and that Hgb levels be maintained between 7 and
9 g/dL. Many observational studies are consistent with this
observation.43 The most appropriate transfusion strategies
remain unclear for patients in the NCCU; although guide-
lines regarding transfusion practice have been published for
general critical care,94 neurocritical care,95 TBI,96 and SAH.46
Efficacy of Transfusion
RBC transfusions are intended to augment oxygen delivery
and improve tissue oxygenation. It is presumed that increased
138 Section II—Clinical and Laboratory Assessment
Hgb concentration increases oxygen carrying capacity and
provides more oxygen to delivery-dependent tissue. However,
whether blood transfusion successfully achieves improved
tissue oxygenation remains unclear. In the NCCU population
several studies have examined the effect of transfusion on
brain oxygen. In summary, transfusion increases brain oxygen
in about 75% of patients.96-98 However, even when transfusion
increases brain oxygen it may not always correct other markers
of cellular distress (e.g., the LPR).35
Sequelae of blood storage, termed the storage lesion,
may be responsible for the observed lack of efficacy of RBC
transfusion (RBCT). First, during storage, RBCs undergo a
predictable change from biconcave disks to deformed spher-
oechinocytes, resulting in loss of deformability and decreased
ability to navigate the microcirculation (vessels of 3-8 µm).99
Second, increased osmotic fragility may lead to decreased
RBC survival.100,101 Third, adenosine triphosphate (ATP) is
depleted.102 Fourth, the p50, a measure of Hgb affinity for
oxygen, is lower in stored blood, which results in less oxygen
unloading to tissues.103 Fifth, after 7 to 14 days, stored blood
is depleted of 2,3-diphosphoglycerate (2,3-DPG)104; this shifts
the oxygen dissociation curve to the left and reduces the
amount of oxygen available for tissue consumption. Sixth,
prolonged storage results in adhesion of RBCs to endothelial
cells, which can compromise microvascular circulation.105,106
Finally, storage for more than 42 days may cause vasoconstric-
tion upon transfusion due to free Hgb and lysophosphatidyl
choline species released from the cellular membrane of senes-
cent RBCs.107
Potential Side Effects of Transfusion
A variety of deleterious side effects are associated with
PRBC transfusion. First, although rare, risks are associated
with crossmatching and the transmission of viral and prion
diseases. The risk of transmission of human immunodefi-
ciency virus (HIV), hepatitis C virus, and hepatitis B virus
are 1 in 1,900,000, 1 in 600,000, and 1 in 220,000 respec-
tively.108 Cytomegalovirus (CMV) is present in 4% of trans-
fusions from healthy donors due to the reactivation of latent
CMV in leukocytes.109,110 Other transfusion-transmitted
infections are thought to be due to contaminated leuko-
cytes.111 Contaminated leukocytes111 may also contribute to
infection or febrile nonhemolytic transfusion reactions
(FNHTR) and hemolytic transfusion reactions.112 In addi-
tion, leukocyte activity on RBC membranes during storage
and degradation during storage that releases oxygen free
radicals and proteases, may incite inflammation in the trans-
fusion recipient.113,114
Transfusion may have detrimental effects on the immune
system. Transfusion-related immunomodulation may lead to
alloimmunization,91 a heightened immune response as in
transfusion reactions, or tolerance induction, a suppression of
the immune response that predisposes to hospital-acquired
infections. Alloimmunization and consequent induction of
human leukocyte antigen (HLA) antibodies and T cell activa-
tion, results in several clinical syndromes, including transfu-
sion reactions, transfusion-associated graft-versus-host
disease, transfusion-related acute lung injury (TRALI), and
perhaps various autoimmune diseases.91 TRALI is defined as
a new episode of respiratory distress not explained by an alter-
nate etiology that occurs during, or within 6 hours of, a
completed transfusion.115 The most common clinical features
include bilateral pulmonary edema, hypoxemia, fever, dyspnea,
and hypotension in the presence of normal cardiac func-
tion.115 Plasma-rich blood components (FFP and platelets)
and high-volume transfusion may predispose to TRALI,
although it has been associated with all blood product com-
ponents including IVIg and cryoprecipitate.115,116
Tolerance induction after transfusion is associated with a
decrease in natural killer cell function, defective antigen pre-
sentation, and a reduction in helper/suppressor T-lymphocyte
ratio. Consequently transfusion is linked to increased predis-
position to hospital-acquired and postoperative infections in
variety of disorders and ICU patients, multiorgan dysfunc-
tion, acute respiratory distress syndrome (ARDS), systemic
inflammatory response system (SIRS), and even to cancer
recurrence.117-129 A recent meta-analysis of transfusion in the
ICU130 suggests that the pooled odds ratio for developing an
infectious complication was 1.8 (95% CI, 1.5-2.2).
The deleterious effects of transfusion are thought to
occur in large part from inflammatory cytokines (e.g.,
IL-1, IL-6, IL-8), inflammatory mediators (e.g., bactericidal
permeability-increasing protein [BPI] and TNFα)131; neutro-
phil activation132 and the accumulation of histamine, lipids,
cytokines, cellular membranes, and HLA class antigens.
Some of these compounds are derived from WBCs,
which are present in the cellular components of standard-
preparation blood products. Consequently although it seems
reasonable to expect that leukoreduction of stored blood
would mitigate the immunomodulatory effects of transfused
RBCs, its clinical impact and cost effectiveness have not
been fully elucidated.133-138 There also may be non–leukocyte
mediated mechanisms of immunosuppression139 (i.e., RBC
lysis leads to release of arginase, an enzyme that degrades
arginine, an amino acid that normally stimulates lymphatic
function).140
Strategies to Prevent Anemia
Anemia can be detrimental to ICU patients, but so is its cor-
rection with blood transfusion. One way to reduce the inci-
dence of anemia in the ICU is to decrease the amount of
phlebotomized blood. Strategies to decrease phlebotomy
include the use of pediatric collection tubes in adults,
recycling of discard volumes, POC technology, and new
paradigms in laboratory-ordering practices. The use of
pediatric-size laboratory tubes can reduce blood loss by
almost 50%.141 Discard volumes required for “clearing the
line” amount to almost a third of daily blood loss in the ICU.19
Recycling of discard volumes through the use of blood con-
servatory devices can minimize iatrogenic blood loss. POC
technologies employ rapid bedside analyzers of whole blood
that provide accurate analyses of multiple analytes with
minimal blood sampling. One milliliter of whole blood is suf-
ficient for bedside analysis of pH measurements, arterial
blood gases, selected electrolytes, glucose, and hemato-
crit.142,143 Use of another POC device, namely thromboelas-
tography, has been shown to reduce not only administration
of blood components but also bleeding.144,145 Education
of physicians and nurses to promote judicious laboratory
ordering and testing is paramount. Documentation of phle-
botomy volume on ICU flow sheets has been shown to effec-
tively decrease iatrogenic blood loss.146 Discontinuation of
 Section II—Clinical and Laboratory Assessment 139
laboratory panels and standing laboratory orders may further
reduce unnecessary blood draws. Batching of blood draws can
also minimize the wastage associated with each phlebotomy
attempt.
Alternatives to Transfusion
Because RBCT can worsen outcome, alternative strategies
have been tried, including stimulation of red cell production
by recombinant human EPO (rhuEPO), Hgb-based oxygen
carriers, and perfluorocarbons. A detailed discussion of Hgb-
based oxygen carriers and perfluorocarbons is beyond the
scope of this chapter but is available elsewhere.147-149 The use
of EPO is of particular interest to the NCCU patient popu-
lation because both EPO and its receptor (EPOR) are present
in neural and vascular tissue. Furthermore, several lines of
evidence suggest that EPO may be neuroprotective.150-152
Whether rhuEPO is useful in the ICU remains unclear,
because some studies suggest that although transfusion
needs may be reduced153-156 there is a significant increase in
the incidence of thrombotic vascular events.157 In addition,
a variety of side effects including seizures, hypertension,
headache, tachycardia, nausea, and clotted vascular access
are described.158 Enthusiasm for the use of rhuEPO therefore
has waned and many studies of erythropoiesis-stimulating
agents (ESAs) have been halted. In 2008, the U.S. Food and
Drug Administration (FDA) published a black box warning
on all ESAs for increased risk of mortality, cardiovascular
and thrombotic events, and tumor progression or recur-
rence. Because of its neuroprotective effects, EPO studies
in ischemic stroke, SAH, and TBI continue (clinicaltrials.
gov/sho/NCT00313716 and www.controlled-trials.com/mrct/
trial/661629/bellomo). There are promising results from pre-
liminary human randomized controlled trials (RCTs) in
SAH.159 However, mortality was actually greater with EPO
in a recent multicenter RCT of acute ischemic stroke
patients.160
Coagulation
Hemostasis
The coagulation system is normally quiescent. Endothelial
cells that line the intimal surface of blood vessels promote
smooth blood flow by inhibiting platelet function and fibrin
accumulation.161 Upon vascular injury, the antithrombotic
properties of the endothelial monolayer are lost, and pro-
thrombotic substances contained within the subendothelium
are exposed, inciting platelet adhesion. This is mediated by
vWF, which binds platelets at its receptor in the platelet
membrane and attaches to the damaged vessel wall. Adherent
platelets release the contents of the storage granules, includ-
ing ADP and thromboxane A2, thus activating circulating
platelets to bind the damaged vessel wall. These activated
platelets expose binding sites for fibrinogen and lead to fibrin
deposition, the final step in the formation of the hemostatic
plug. Fibrin is converted from plasma fibrinogen by throm-
bin. Thrombin is formed from prothrombin, its zymogen
precursor, and activated factors X and V (the common
pathway). Factor X is most often activated by the tissue factor
(extrinsic) pathway, but may be activated by the contact acti-
vation (intrinsic) pathway. Fibrin is ultimately digested by
the fibrinolytic system, which is mediated by plasmin.
Plasmin is formed from endothelial-derived plasminogen
activators.
Four bleeding tests have traditionally been used to evaluate
the bleeding patient: the PT, activated partial thromboplastin
time (aPTT), bleeding time, and platelet count.161 The PT
measures the integrity of the extrinsic and common pathways,
including factors VII, X, and V; prothrombin; and fibrinogen;
whereas the aPTT measures the integrity of the intrinsic
and common pathways, including high molecular weight kal-
likrein (HMWK); prekallikrein; factors XII, XI, IX, VIII, X,
and V; prothrombin; and fibrinogen.161 Bleeding time is a
screening test of platelet-endothelial interactions. It measures
the time to cessation of bleeding after a standard incision in
the volar aspect of the forearm. It is prolonged in thrombocy-
topenia, vWD, platelet abnormalities, and primary vascular
disorders; it is usually not affected by coagulation factor defi-
ciencies. This test, however, lacks reproducibility and has poor
sensitivity and specificity.
Normal results of these screening tests essentially exclude
clinically significant bleeding. Notable exceptions include
factor XIII deficiency and vWD, which may have normal
results to these screening tests. If the bleeding time is pro-
longed and thrombocytopenia is not present, evaluation for
medications (e.g., aspirin) or diseases (e.g., end-stage renal
disease [ESRD]) that may affect platelet function is warranted.
Prolongation of the PT or partial thromboplastin time (PTT)
usually indicates a factor deficiency or presence of an inhibi-
tor. A mixing study can discriminate between these etioloiges
because a deficiency should correct with addition of normal
plasma, whereas the presence of an inhibitor would not correct
the prolonged time.
Venous Thromboembolism
Venous Thromboembolism in the Intensive
Care Unit
Venous thromboembolism (VTE) is composed of DVT and
pulmonary embolism (PE) and occurs in 117 to 145 per
100,000 individuals each year.162 In the United States, the fre-
quency of DVT approaches 2 million people per year,163 with
fatal PE in 50,000 to 200,000 patients.164,165 The 1-year mortal-
ity rate is as high as 21% for DVT and 39% for PE,161 and PE
accounts for 5% to 10% of potentially preventable hospital
deaths.166 Risk factors for VTE include: (1) age greater than 40
years; (2) prolonged immobility; (3) prior VTE; (4) malig-
nancy; (5) myeloproliferative disorders; (6) cardiac dysfunc-
tion; (7) stroke; (8) respiratory failure; (9) pregnancy; (10)
obesity; (11) varicose veins; (12) estrogen use; (13) inflamma-
tory bowel disease; (14) nephritic syndrome; (15) femoral vein
catheterization; (16) sepsis; (17) major surgery; (18) trauma;
and (19) inherited or acquired thrombophilias. Among hos-
pitalized patients the risk is much greater in ICU than other
hospitalized patients167,168; in particular, patients who remain
in the ICU more than 72 hours and who have end-stage renal
failure, personal or family history of VTE, platelet transfusion,
and vasopressor therapy are at increased risk.165 Despite
routine thromboprophylaxis, 10% of ICU patients develop
lower extremity DVT, and VTE remains one of the most
common unsuspected autopsy findings in critically ill
patients.169
140 Section II—Clinical and Laboratory Assessment
Venous Thromboembolism
in Neurosurgical Patients
Neurosurgical patients have several unique risks for VTE
including: (1) treatment-related immobilization, (2) disease-
related or postoperative paralysis, (3) prolonged surgical pro-
cedures (>4 hours), (4) delayed initiation of activity, (5)
corticosteroid use, and (6) pharmacologic dehydration. In
particular patients with TBI, malignant tumors, advanced age
and those who undergo craniotomy (in contrast to spinal
surgery) are at greater risk.170,171 Kaolin-activated thrombo-
elastography has identified a transient hypercoagulable state
during and after neurosurgical operations; indices of hyper-
coagulability were more pronounced with longer operative
times and cranial rather than spinal procedures.172 Brain tissue
is rich in tissue factor (factor VII), which may incite a pro-
thrombotic state. After surgical procedures there is increased
inhibition of tissue plasminogen activator (tPA), a potent clot-
dissolving enzyme that converts plasminogen to plasmin.173
Diminished tPA results in a reduced capacity to clear thrombi.
Local hypoxemia in venous channels that may occur with
lengthy neurosurgical procedures or with prolonged inactivity
can promote clot formation in as little as 2 hours.174
Among neurosurgical patients after craniotomy, the
reported rates of VTE vary widely from zero to 60%.175-182
Although the prevalence of PE is between 1.5% and 8.4%
among neurosurgical patients, the risk of mortality from PEs
can be as high as 50% in this population.183,184 The greatest
risk for DVT is in patients with brain tumors (28%-43%),
followed by those undergoing craniotomy (25%), and those
with TBI (20%).185,186
The overall risk of developing DVT in patients with brain
tumors is 3% to 60%.185-187 The incidence of DVT is generally
considered highest with malignant gliomas, although some
meningiomas, especially those that adhere to the sinuses, can
be associated with higher DVT rates.188-191 The pathogenesis
of VTE in brain tumor patients is thought to be associated
with the high concentration of TF, the cell surface receptor of
factor VII/VIIa, which plays a pivotal role in the initiation of
the coagulation cascade. High-grade tumors may express
higher concentrations of TF, procoagulants, and fibrinolytic
inhibitors than low-grade tumors.192 Elevated levels of
D-dimer, homocysteine, lipoprotein (a) (lp[a]), vascular
endothelial growth factor (VEGF), tPA and plasminogen acti-
vator inhibitor (PAI) are found in patients with malignant
glioma and likely contribute to the observed hypercoagulable
state.193,194 Chemotherapeutic agents used to treat malignancy,
including nitrosurea and cisplatin, may also increase the risk
of DVT in these patients.195
Thromboprophylaxis in
Neurosurgical Patients
There is a general reluctance to use pharmacologic prophy-
laxis for VTE in neurosurgical patients because the conse-
quences of bleeding may be devastating. Several studies have
attempted to determine whether the risk of pharmacoprophy-
laxis outweighs the risk of VTE. Danish et al. sought to deter-
mine the best method of thromboprophylaxis in neurosurgical
patients by use of a decision analysis model196 that incorpo-
rated the incidence, prevalence, and likelihood ratios of VTE
and ICH. The addition of heparin lowered the incidence of
VTE but increased the incidence of ICH. The impact of
intracranial bleeding was greater than the benefits of heparin-
ization, so they concluded that overall outcomes were best
with mechanical prophylaxis alone. In a single center study the
combination of low-dose unfractionated heparin and twice-
weekly surveillance Doppler ultrasound in neurosurgical
patients was associated with an incidence of PE of 0.09% and
of DVT of 2.6%, a rate of VTE lower than that in the other
ICUs within the same institution.197 In a randomized trial of
307 patients undergoing elective neurosurgical procedures,
17% of patients with compression stocking therapy and
enoxaparin developed DVT, whereas 32% who received only
compression devices had DVTs.175 The incidence of major
bleeding was similar. The ninth American College of Chest
Physicians (ACCP) conference on antithrombotic and throm-
bolytic therapy recommends mechanical prophylaxis, prefer-
ably with intermittent pneumatic compression (IPC) or
pharmacolagic prophylaxis be used over no prophylaxis for
patients undergoing craniotomy.198 The risk of a DVT in
patients who undergo elective spinal surgery is low. The 9th
edition ACCP guidelines recommend mechanical prophylaxis,
preferably with IPC, over no prophylaxis (Grade 2c), unfrac-
tionated heparin (Grade 2c), or LMWH (Grade 2c) unless
they have additional risk factors such as advanced age, known
malignancy, presence of a neurologic deficit, previous VTE, or
an anterior surgical approach.199 There also is debate on what
to do after TBI or spinal cord injury (SCI). The 9th edition
ACCP guidelines suggest that in the absence of a contraindica-
tion, all TBI patients receive LDUH (Grade 2c), LMWH
(Grade 2c) or mechanical prophylaxis, preferably with IPC
(Grade 2c).198 If LMWH is contraindicated, then IPC is rec-
ommended. Inferior vena cava (IVC) filters are not recom-
mended as primary prophylaxis in trauma patients. In patients
with acute SCI, which carries the highest risk of VTE, the
ACCP guidelines recommend VTE prophylaxis with the use
of IPC and either low-dose unfractionated heparin (LDUH)
or LWMH, once primary hemostasis has been achieved.198
Venous Thromboembolism Prophylaxis
in Stroke Patients
VTE is a common complication of stroke, similar to that for
surgical patients, and is associated with increased mortality
and morbidity.185,198 A meta-analysis of more than 23,000 isch-
emic stroke patients suggested that low-dose LMWH therapy
rather than unfractionated heparin (UFH) or placebo had the
best risk-benefit profile, decreasing the risk of both DVT and
PE without an increased risk of spontaneous intracerebral
hemorrhage (sICH).200 The PREVAIL study, a randomized
trial of 1762 ischemic stroke patients, found that 40 mg of
once-daily enoxaparin reduced the risk of VTE by 43% com-
pared with UFH (5000 units subcutaneously twice daily)
including inpatients with a National Institutes of Health
Stroke Scale (NIHSS) score greater than or less than 14.201 The
rate of symptomatic ICH was similar but there was a small
increase in the rate of extracranial bleeding in the enoxaparin
group (1% versus 0%).
There is less evidence about optimal VTE prophylaxis in
patients with hemorrhagic strokes, although the risk of
DVT may be up to four times greater in ICH than in patients
with ischemic stroke. The increased risk is likely related to
reduced rates of pharmacoprophylaxis and a higher degree of
neurologic impairment. Abrupt cessation and reversal of oral
anticoagulation also may increase this risk, although one study
 Section II—Clinical and Laboratory Assessment 141
found that patients with oral anticoagulation–related hemor-
rhages were not at increased risk of thromboembolic events.202
Review of data from the National Hospital Discharge Survey
demonstrates that of 14 million patients with ischemic stroke
and 1.6 million patients with ICH the incidence of PE
and DVT were 0.51% and 0.74% and 0.68% and 1.1%,
respectively.177
The incidence of VTE after ICH may be reduced with the
addition of intermittent pneumatic compression to elastic
stockings203 and may be further reduced with administration
of low-dose UFH administered three times a day when started
2 days after ICH.204 Pharmacoprophylaxis with either UFH or
LMWH may safely be initiated 2 days after ictus.204,205 The
American Heart Association and American Stroke Association
(AHA-ASA) guidelines for ICH management recommend
intermittent pneumatic compression in addition to elastic
stockings for prevention of VTE (class I, level B evidence) and
consideration of low-dose subcutaneous LMWH or UFH after
1 to 4 days from onset in patients with immobility and docu-
mented cessation of bleeding.206,207
Diagnosis and Surveillance
for Venous Thromboembolism
The presence of clinical symptoms is not a reliable marker
of DVT, and hospital-acquired VTE is usually clinically
silent.199 Traditional clinical features such as a tender, warm,
and swollen calf; a positive Homans’ sign; and venous dila-
tion are unreliable.196 The vast majority of patients with PE
exhibit no stigmata of DVT.168 The highest risk of pulmo-
nary embolism is associated with thromboses of the proxi-
mal veins above the popliteal fossa. In less than 1% of
patients, calf vein thrombi cause PE. DVTs of the proximal
leg contribute to 90% of pulmonary emboli, whereas the
remainder is thought to arise from the pelvis or the upper
extremities.
The gold standard diagnostic test to identify a DVT remains
contrast venography; however, it is being rapidly replaced with
Duplex ultrasonography (DUS) given the invasive nature,
technical demands, costs, and potential risks of venogra-
phy.208,209 For proximal DVT, ultrasonography has diagnostic
sensitivities of 89% to 96% and specificities of 94% to
99%.209,210 The sensitivity and specificity of DUS is lower for
symptomatic calf DVT, 73% and 93%, respectively.211 It should
be noted that for asymptomatic patients DUS has a sensitivity
of only 50%; therefore DUS should only be performed when
there is a clinical suspicion of DVT. Radiologic imaging is
necessary for the diagnosis of PE in patients with at least an
intermediate pretest probability of PE.
D-dimer is protein produced when cross-linked fibrin is
degraded by plasmin, the principal enzyme involved in fibri-
nolysis. D-dimer has been studied extensively.211 Although
D-dimer levels coupled with a low pretest probability of DVT
may reliably exclude DVT in some populations,212 its utility in
hospitalized or critically ill patients is limited.213,214 In some
patients with confirmed DVT, the D-dimer level may be
normal.215
Routine ultrasound surveillance of lower extremities has
been proposed to decrease the PE rate, especially in neurologi-
cally injured patients often with altered mental states and limb
immobility.197,216 A decision and cost effectiveness analysis
demonstrated that adherence to pharmacoprophylaxis resulted
in fewer DVTs and was eight times less costly per quality
adjusted life-year.217 The authors concluded that optimizing
thromboprophylaxis was a better value in terms of cost and
health gains when compared with routine ultrasound
screening.
Cerebral Venous Thrombosis
Thrombosis of the cerebral veins and sinuses is a rare but
serious disorder characterized by headache, seizures, focal
neurologic signs, papilledema, and occasionally coma and
death.218 A prothrombotic factor is found in nearly two thirds
of cases.219 Cavernous sinus thrombosis is usually caused by
infection and is characterized by periorbital edema, proptosis,
chemosis, papilledema, and ocular paresis.218 Occlusion of the
cerebral veins leads to brain infarction in a nonarterial distri-
bution, petechial hemorrhage, and localized edema, whereas
occlusion of the venous sinuses can cause decreased venous
outflow and increased ICP. Etiologies include antecedent head
trauma, hypercoagulable states, ear and sinus infections (espe-
cially otitis media or mastoiditis with hypoplasia of the con-
tralateral transverse sinus), collagen vascular and hematologic
diseases, oral contraceptive use, pregnancy and the puerpe-
rium, neurosurgical procedures, and dehydration. Lumbar
puncture also may lead to sinus thrombosis by traction placed
on cortical veins associated with downward shift of the brain
after spinal fluid removal.220
The frequency of puerperal cerebral venous thrombosis
(CVT) is only slightly less than the incidence of puerperal
arterial strokes. Incidence estimates for CVT in women during
pregnancy and the puerperal period range from 1 per 2500 to
1 per 10,000 deliveries in the United States.221,222 CVT does not
preclude future pregnancy; however, formal consultation with
a hematologist or maternal-fetal medicine specialist is reason-
able.219 Pregnant patients with CVT are preferentially treated
with LMWH rather than UFH. For pregnant women with a
history of CVT, prophylactic anticoagulation during future
pregnancies may be beneficial.219
In the acute setting, diagnosis should include routine blood
studies (CBC, chemistry panel, PT, aPTT), erythrocyte sedi-
mentation rate, and screening for oral contraceptive use and
inflammatory or infectious diseases.219 Testing for prothrom-
botic conditions including protein C, protein S, antithrombin
deficiency, antiphospholipid syndrome, prothrombin gene
mutation, and factor V Leiden can be beneficial for the man-
agement of patients and is generally indicated 2 to 4 weeks
after anticoagulation has been discontinued. A normal
D-dimer level, by sensitive immunoassay or enzyme-linked
immunosorbent assay (ELISA), may be considered to identify
patients with a low probability of CVT; a normal level should
not preclude diagnosis in a patient with a high clinical suspi-
cion.219,223,224 Magnetic resonance imaging with venography is
a highly sensitive diagnostic tool. Both T1- and T2-weighted
images demonstrate hyperintense signal from the thrombosed
sinuses. Thrombi are isointense on T1-weighted images early
in the course of the disease (<5 days). Angiography can be
particularly helpful when there is cortical vein thrombosis in
the absence of sinus thrombosis. The mainstay of treatment
is anticoagulation regardless of presence of ICH; duration of
treatment varies according to etiology (i.e., 3-6 months for
transient risk factors compared with lifelong treatment
for those with severe thrombophilias).225,226 Treatment for
refractory CVT includes endovascular thrombectomy,
thrombolysis, or thromborrhexis.
142 Section II—Clinical and Laboratory Assessment
Thrombophilias
Testing for hypercoagulable states in unselected populations
is low yield, although a thrombophilic defect can be demon-
strated in more than 50% of the patients who present with a
VTE.227 Thrombophilias increase the risk of first-time VTE,
but it is unclear whether they increase the risk of recurrent
VTE . Approximately 1% to 4% of ischemic strokes are caused
by thrombophilias.228
Protein C, Protein S, Antithrombin
Deficiency
Activated proteins C and S regulate the coagulation cascade
by inactivating factor Va. Protein C deficiency is relatively rare,
with a prevalence of approximately 0.2% in the general popu-
lation and 2.5% to 6% in patients with first-time VTE.228 Ret-
rospective analyses of family studies indicate that the risk of
VTE in patients with protein C deficiency is approximately 10
times that of nondeficient patients. Recurrent thromboses in
these patients have been estimated between 4.8% and 17% per
year.229 The prevalence of protein S deficiency, an autosomal
dominant defect, is about 0.03% to 1.6% in the general popu-
lation.230 Among patients with demonstrated VTE, the preva-
lence is 1% to 2%. One third of patients with protein S
deficiency will have a VTE event by the age of 60 years. Anti-
thrombin deficiency occurs 0.5% to 7.5% with a first-time
VTE.230 The inheritance pattern is also autosomal dominant
and almost all patients are heterozygous. It should be sus-
pected when unusually high doses of heparin are required to
increase bleeding time.
Factor V Leiden and Prothrombin
Gene Mutation (G20210A)
Factor V Leiden and prothrombin gene mutations are “gain of
function” lesions that result in enhanced thrombosis and pro-
coagulation. Inactivation of factor Va occurs by cleavage at
three sites. Failure of cleavage by activated protein C at the
first site, position 506, is known as factor V Leiden (FVL).
Factor Va is inactivated 10 times more slowly than normal in
FVL carriers.231 FVL is responsible for at least 90% of acti-
vated protein C–resistant conditions. It is the most prevalent
thrombophilia and is found in 5% to 12% of the general
population.231
The second most prevalent thrombophilia in the general
population is the prothrombin gene mutation. It is found in
1% of the population and is implicated as a cause in 4% to
8% of patients with first-time VTE. The risk of first thrombo-
sis is 0.46% for a person carrying this gene.232 A prothrombin
level higher than 115% (90th percentile) results in a 2.1-fold
increased VTE risk.233,234 Data on recurrent events in these
patients are not as compelling as they are for deficiencies of
proteins C and S and antithrombin III.
Elevated Levels of Coagulation Factors
Elevated levels of coagulation factors VIII, IX, XI and fibrino-
gen increase the risk for VTE. Of these, factor VIII is the most
prothrombotic and in a dose-dependent manner.235-237 Every
10% increase in factor VIII results in a 10% increased risk for
VTE. Elevated factor IX levels are associated with a twofold
increased risk of VTE.237 Fibrinogen levels of greater than
500 mg/dL are associated with approximately a fourfold
increased risk of VTE.235 The remaining factors have not been
shown to be prothrombotic.
Plasminogen Activator Inhibitor 1 and
Lipoprotein (a)
Strokes may be caused by alterations in genetic regulation of
fibrinolysis by plasminogen activator inhibitor (PAI)-1 and
lp(a).227,232 PAI-1 inhibits tissue plasminogen activator activity
and suppresses fibrinolysis. Lp(a) is a modified low-density
lipoprotein that exhibits structural homology to plasminogen,
a plasma protein important in dissolving blood clots.
Lp(a) competes with plasminogen, which interferes with
fibrinolysis.
Antiphospholipid Syndrome
The antiphospholipid syndrome (APS) is defined by the com-
bination of thrombosis and/or pregnancy-related morbidity
and medium to high titiers of persistent circulating antiphos-
pholipid antibodies (aPL): lupus anticoagulant (IA), anticar-
diolipin antibodies (aCL), and/or anti-b2 glycoprotein I
antibodies (anti-b2 GPI).238 aPL antibodies are a heteroge-
neous family of antibodies to membrane phospholipids that
are primarily associated with venous thrombosis and fetal loss,
although stroke is the most common arterial thrombotic
manifestation. The combination of aPL antibodies and
tobacco or oral contraceptive use significantly increases the
risk of thrombosis.239
Evaluation and Treatment
of Thrombophilias
The inherited thrombophilias rarely contribute to strokes in
adults; therefore, screening for these states is not routinely
recommended.226 For those patients with known inherited
thrombophilia who present with stroke, full evaluation for
alternative etiologies (atherosclerosis, nonvalvular atrial fibril-
lation) should be performed. However, evaluation for DVT
should be considered because presence of DVT may alter
treatment (e.g., anticoagulation). Antiplatelet therapy may be
reasonable treatment in stroke patients with low-titer aPL
antibodies; for patients with APS, treatment with anticoagula-
tion to a goal target international normalized ratio (INR) of
2 to 3 or even >3 may be warranted.240
Hyperhomocysteinemia
Homocystinuria is a rare condition associated with elevated
levels of homocysteine (>100 µmol/L), marfanoid features,
mental retardation, and premature arterial and venous
thromboses. It is caused by mutations in either cystathionine-
β-synthase or methyl tetrahydrofolate reductase. Contrary to
homocystinuria, mild hyperhomocysteinemia is common
and occurs in about 5% to 10% of the general population.233
Its clinical relevance and contribution to stroke risk are
unknown. Folate supplementation may be considered in
patients with hyperhomocysteinemia; however, there is no
compelling evidence to suggest that reducing homocysteine
levels abrogates stroke risk.
 Section II—Clinical and Laboratory Assessment 143
Anticoagulants
Parenteral anticoagulants may be broadly divided into
two groups: indirect and direct anticoagulants. Indirect
anticoagulants rely on plasma cofactors for thrombin inacti-
vation, whereas direct anticoagulants do not. The indirect
anticoagulants include UFH, LMWHs, fondaparinux, and
danaparoid. The direct agents target thrombin and include
recombinant hirudins, bivalirudin and argatroban.
Heparin
Heparin is a mucopolysaccharide of heterogeneous molecu-
lar size; it can vary in weight from 3,000 to 30,000 kDa and
averages approximately 45 saccharide units.241-243 Its major
anticoagulant effect is mediated by its interaction with anti-
thrombin (AT), but it also inhibits factors Xa, IXa, XIa, and
XIIa. Heparin binds to AT and converts it from a slow to a
fast inactivator of thrombin. Heparin binds to AT through a
glucosamine unit, which contains a unique pentasaccharide
sequence. One third of heparin chains are of sufficient length
to bridge the heparin/AT to thrombin, forming a ternary AT/
heparin/thrombin complex, which potentiates inhibition.
Heparin then dissociates from thrombin and is available for
reuse. Smaller chains that contain the high affinity pentasac-
charide can inhibit factor Xa, although thrombin is 10-fold
more sensitive to inhibition than factor Xa.244 Heparin also
has anticoagulant effects independent of the pentasaccharide
sequence. It binds to heparin cofactor II and catalyzes the
inactivation of factor IIa; however, this effect is chain length
dependent, requiring heparin chains of at least 24 saccharide
units.245
Heparin is administered intravenously or subcutaneously.
Subcutaneous administration requires higher dosing given the
reduced bioavailability. Intravenous (IV) bolus administra-
tion can provide an immediate effect. Heparin binds plasma
proteins, endothelial cells, and macrophages246,247; this phe-
nomenon results in an unpredictable anticoagulant response
and may explain some of the apparent heparin resistance.
Heparin resistance may also be due to AT deficiency,248
increased heparin clearance,249,250 elevated factor VIII,251,252
and elevated fibrinogen.252
The risk of bleeding associated with heparin increases at
higher doses and with concomitant administration of fibri-
nolytic agents253,254 or GPIIb/IIIa inhibitors.255 Recent surgery,
trauma, invasive procedures, and defects of hemostasis are
also risk factors for bleeding.244 Monitoring of heparin and
dose adjustment has become standard practice given the
variable anticoagulant response. When given in therapeutic
doses, the anticoagulant effect is monitored by the aPTT. The
activated clotting time (ACT) is used to monitor high
heparin doses administered to patients, such as during endo-
vascular neurosurgical procedures or percutaneous coronary
interventions.
Unlike the evidence supporting a target range for the INR
for warfarin monitoring, the data to adjust the dose of
heparin to target a therapeutic range are weak. An aPTT ratio
of 1.5 to 2.5 times control was suggested by a retrospective
study in the 1970s that demonstrated lower risk of VTE; this
practice has been widely adopted despite a lack of strong evi-
dence in its support.256 However, neither has this practice
been rigorously studied by randomized control trials, nor are
the same instruments used to measure the aPTT.257-261 There-
fore the therapeutic range for the aPTT is laboratory spe-
cific.261,262 Although many heparin nomograms exist, none is
applicable to all aPTT reagents. For patients who require
high levels of heparin administration (>35,000 units/day
UFH), it may be reasonable to monitor anti-Xa levels and
adjust the dose accordingly because this practice lessens the
amount of heparin administered without compromising clin-
ical outcomes.251
The anticoagulant effects of heparin can be reversed with
IV protamine sulfate, a protein that complexes with heparin
to form a stable salt: 1 mg of IV protamine will neutralize 100
units of IV heparin.244 The half-life of heparin is 60 to 90
minutes; therefore, when reversing with protamine, only
heparin administered over the last few hours needs to be con-
sidered when calculating the dose. Caution is needed when
administering protamine because it can cause anaphylaxis.244
The main side effects of heparin therapy are HIT (see earlier)
and osteoporosis.
Low Molecular Weight Heparin
LMWHs are derived from UFH and are less active against
thrombin relative to factor Xa.263,264 The anti–Xa-to-anti–IIa
ratio of LWMHs is 2 to 4 : 1 (based on molecular size distribu-
tion), whereas the anti–Xa-to-anti–IIa ratio of UFH is 1 : 1.244
LMWHs are one third the molecular weight of UFH, which
roughly corresponds to 15 saccharide units. They have a more
predictable pharmacokinetic profile than UFH due to a
reduced binding affinity for cells and proteins; this also extends
their plasma half-life.265 The lower incidence of HIT also is
related to the decrease in platelet binding. The various LMWHs
have different anticoagulant profiles and are thus not inter-
changeable. Routine laboratory monitoring generally is not
recommended for most patients treated with LMWH. The
ACCP recommends anti-Xa monitoring in some special pop-
ulations, such as in pregnant women who require treatment
with LMWH.244 Others suggest monitoring in obese patients
and those with renal insufficiency. The ACCP recommends
weight-based dosing in the treatment or prophylaxis of obese
patients and preferential use of UFH in patients with severe
renal dysfunction.126-128
Although protamine sulfate incompletely neutralizes the
anticoagulant effect of LMWHs, it should be administered at
a dose of 1 mg per 100 anti-Xa units of LMWH in the acutely
bleeding patient if it was administered during the preceding 8
hours.244 A second dose of protamine sulfate at 0.5 mg per 100
units of anti-Xa should be given if bleeding continues. Side
effects of LMWH include HIT and osteoporosis. The fre-
quency of HIT is threefold lower with LMWH; however,
LMWHs can form complexes with PF4 that are capable of
binding HIT antibodies and therefore LMWHs should not be
used in the treatment of HIT.
Fondaparinux
Fondaparinux is a synthetic analog of the AT-binding penta-
saccharide found in both UFH and LMWH that has higher
anti-Xa activity than LMWH and a longer half-life (~17
hours).266,267 It does not increase the rate of thrombin inhibi-
tion by AT. Due to its predictable pharmacokinetics and neg-
ligible nonspecific binding, fondaparinux can be administered
144 Section II—Clinical and Laboratory Assessment
subcutaneously once daily in fixed doses without laboratory
monitoring. It is not intended for use in patients with renal
insufficiency. Routine monitoring is not recommended, but if
needed the anticoagulant activity can be measured with
anti-Xa assays. The anticoagulant effects of fondaparinux are
not reversible with protamine sulfate; recombinant factor VIIa
may be effective.268 There have been no reports of HIT with
fondaparinux, and it has been used to safely treat patients with
this disorder.269
Danaparoid
Danaparoid is a mixture of glycosaminoglycans (heparan
sulfate, dermatan sulfate, and chondroitin sulfate); its antico-
agulant properties are derived from inhibition of factor Xa in
an AT-dependent manner.244 Its use is limited to treatment of
HIT. An advantage of danaparoid is that it does not prolong
INR, which is beneficial when transitioning patients to vitamin
K antagonists. However, the 25-hour half-life is a significant
disadvantage because there is no antidote for it.
Direct Anticoagulants
In contrast to the indirect anticoagulants, the direct thrombin
inhibitors bind to thrombin and block its enzymatic activity.
The aPTT is used to monitor the anticoagulant effects of
direct thrombin inhibitors; however, this test is not always
accurate. The dose response is not linear and the aPTT pla-
teaus at high doses of these drugs. The ecarin clotting time
may be more accurate, but lack of availability and standardiza-
tion generally precludes its routine use.
Hirudin
Hirudin, derived from the salivary gland of the medicinal
leech, is available in recombinant form as lepirudin or desi-
rudin. They both irreversibly inactivate thrombin. They have
an identical pharmacokinetic profile and mechanism of
action; the amino-terminal domain interacts with active site
of thrombin, and the carboxy-terminus binds to the
substrate-binding site. The plasma half-life of the hirudins is
60 minutes when administered IV and 120 minutes when
given subcutaneously.270 Hirudins are renally excreted and so
are contraindicated in patients with renal failure.271 When
used to treat HIT, the anticoagulant effect of lepirudin is
monitored by the aPTT with the goal ratio between 1.5 and
2.5. When administered for thromboprophylaxis, routine
monitoring is unnecessary.244
Antibodies against hirudin develop in 40% of patients
treated with lepirudin.244 These antibodies generally do not
have much clinical impact; however, analphylaxis may occur
if patients with antibodies are reexposed to hirudin.
Bivalirudin
Bivalirudin is a hirudin analog that reversibly binds to the
active site of thrombin. Bivalirudin has a plasma half-life
of 25 minutes after IV injection. Bivalirudin is licensed as
an alternative to heparin for patients undergoing percuta-
neous coronary interventions (PCIs) in patients with HIT.
In contrast to hirudin, bivalirudin is not immunogenic.
The anticoagulant effect may be monitored with the aPTT
and the ACT, but the results are variable and unreli-
able.244,272 New anti-factor IIa assays hold promise to better
monitor the effects of this anticoagulant without a veritable
antidote.272
Argatroban
Argatroban is a reversible thrombin inhibitor with a 45-
minute plasma half-life. It is hepatically metabolized via the
cytochrome p450 system and caution must be used in the
face of liver dysfunction.244 It is indicated for the treatment
of HIT and as an alternative to heparin for PCI in patients
with HIT. All of the direct thrombin inhibitors increase the
INR, but this is especially true of argatroban. This compli-
cates the transition from argatroban to vitamin K antago-
nists. Some target a goal INR greater than 4 when vitamin K
antagonists (VKAs) are given in conjunction with direct
thrombin inhibitors; however, monitoring a chromogenic
assay of factor X levels may be a safer practice.244
Oral Direct Thrombin Inhibitors
The need for a fixed-dose oral anticoagulant with a broader
therapeutic window and lower inter- and intrapatient vari-
ability that does not require serial laboratory testing and dose
adjustment led to the novel anticoagulants dabigatran, rivar-
oxaban, apixaban, and edoxaban. These agents inhibit throm-
bin (dabigatran) or factor Xa (rivaroxaban, apixaban, and
edoxaban) directly rather than through a cofactor. Although
they reversibly bind their substrate, their anticoagulant effect
is irreversible because there is no veritable antidote. These
agents provide rapid onset to therapeutic effect in contrast to
warfarin. They all demonstrate low variability in pharmaco-
dynamic characteristics, thus obviating the need for routine
coagulation monitoring.
Apixaban has demonstrated efficacy and safety for the pre-
vention of venous thromboembolism after elective hip or knee
replacement273,274 and to prevent stroke in patients with non-
valvular atrial fibrillation (AF).275 The FDA has approved riva-
roxaban for the prevention of stroke secondary to AF.
Dabigatran has been compared with warfarin to prevent
stroke and systemic embolization in more than 18,000 patients
with AF.276 At a dose of 110 mg twice daily, dabigatran proved
noninferiority to warfarin and demonstrated lower rates of
symptomatic ICH. A higher dose of 150 mg twice daily was
associated with lower rates of stroke but similar rates of symp-
tomatic hemorrhage compared with warfarin. Although these
oral anticoagulants have proven effective at reducing second-
ary stroke due to AF, concerns remain about the lack of revers-
ibility of its anticoagulant effect.
Vitamin K Antagonists
Warfarin is a vitamin K antagonist (VKA) that produces an
anticoagulant effect by obstructing the cyclic interconversion
of vitamin K and its 2,3 epoxide, thereby modulating
γ-carboxylation of glutamate residues on the N-terminal
regions of vitamin K–dependent proteins. Not only do VKAs
have anticoagulant properties due to inhibition of vitamin
K–dependent coagulation factors II, VII, IX, and X, but they
also may have procoagulant properties by inhibiting carbox-
ylation of the regulatory anticoagulant proteins C, S, and Z.277
 Section II—Clinical and Laboratory Assessment 145
VKAs are used to treat venous thromboembolism, atrial fibril-
lation, and valvular heart disease.198 They are complex agents
with a narrow therapeutic index and many food and drug
interactions. VKAs require routine monitoring and frequent
dose adjustments to maintain their safety and efficacy.
Advances in warfarin management have been made through
dedicated anticoagulation clinics and patient self-testing and
management. Despite these advances, in 2007 more than
60,000 emergency department visits were related to warfarin
therapy, the majority of which were for acute hemorrhage.278
Health care resources dedicated to warfarin management sub-
stantially exceed the cost of the drug itself.
Point of Care Testing
POCT is diagnostic testing performed with rapid generation
of results at or near the site of the patient in order to optimize
patient care. These assays often do not require laboratory
support, are easy to perform, and efficiently deliver results.
Some of these tests use capillary blood and others use venous
blood or citrated samples. Careful attention to manufacturers’
recommendations is necessary because erroneous values may
be obtained if the inappropriate sample type is used. This
discussion focuses on those POCTs with a self-contained
reagent system. Much of this literature has been driven by use
of POCT in cardiac surgery, because these patients have plate-
let dysfunction and high bleeding risk and require antiplatelet
and anticoagulant medications.
Some POCTs do not contain a self-contained reagent system
and may require samples to be pipetted or centrifuged. Quality
assurance may be limited because there are no formal external
quality assessment measures for many of these devices,
although some commercially available quality control samples
with established reference ranges are available. Internal quality
controls also must be performed routinely to ensure reliability
of results. Maintenance of current standard operating proce-
dures specifying sample collection, handling and storing prac-
tices, selection of appropriate reagents, and generation of
adequate records and reporting are mandatory for optimal
utility. The quality of the results also depends on personnel
and frequency of practice. The combination of internal and
external quality assurance is imperative to provide both
precise and accurate results.
Practical limitations of POCT devices include (1) difficulty
in obtaining sufficient sample volume; (2) difficulty in access-
ing test strip or application reservoir; (3) insufficient time to
deliver blood sample; (4) inaccuracies based on presence of
heparin or other medications/drugs; and (5) ineligibility for
patients with blood abnormalities such as presence of lupus
anticoagulant or extremes of Hct.
Prothrombin Time/International
Normalized Ratio
Historically, blood product replacement after surgery has been
based on clinical judgment. The POCT PT/INR can guide
blood product administration expediently in patients who
undergo major surgery. Different devices are based on differ-
ent principles to generate the INR value. Some use light reflec-
tance properties as metal particles move through a magnetic
field, whereas others monitor movement of blood in capillary
tubes, changes in electrochemical impedence, or thrombin
generation through cleavage of a peptide substrate.143 Inac-
curacies may be predicted by the principle underlying the
technology. For example, POCT monitors that measure capil-
lary blood flow may give lower INR values for samples with a
high Hct.
Activated Partial Thromboplastin Time
The aPTT may be used to screen preoperatively for abnor-
malities in hemostasis (especially when used with the PT) or
to guide heparin dosing. A number of studies have shown that
POCT devices can accurately measure the aPTT and thus
guide management of patients on UFH during a variety of
procedures.279-284 However, the correlation between POCT
aPTT and central laboratory aPTT remains controversial.285-287
Moreover, POC aPTT may be more susceptible to effects of
other medications such as aprotinin or epsilon aminocaproic
acid.288 The demand for POCT aPTT has been low and the
reliability of the results remains suspect. Only the Hemochron
device remains available.289
Activated Clotting Time
The ACT is exclusively used to monitor anticoagulation with
high doses of UFH (e.g., during cardiac bypass surgery or
interventional vascular procedures) because high concentra-
tions of UFH result in an unclottable aPTT. A baseline ACT
should be obtained on all patients before UFH administration.
The ACT may be affected by the choice of activator, hypother-
mia, hemodilution, or medications. Currently available POCT
devices for the aPTT include the Hepcon Haemostasis Man-
agement System (HMS), the Hemochron Response device, the
Actalyke MAX-ACT, the Hemochron Junior II and the i-Stat
analyzer.
Estimations and Modifications
of the Thrombin Time
Historically, the thrombin time was used to monitor patients
on UFH; however, there is a poor correlation between heparin
concentration and thrombin time. Furthermore, at high con-
centrations of heparin, the thrombin time becomes unclot-
table and therefore is unsuitable to monitor patients treated
with high UFH doses. The high dose thrombin time (HiTT,
Hemochron) is more specific to heparin effects than the
ACT.290 In contrast to the ACT, the HiTT is not affected by
hypothermia or hemodilution. However, there is poor correla-
tion between the HiTT and the ACT, which has been previ-
ously validated in the cardiac surgery population. The Hepcon
HMS establishes whole blood heparin concentration using an
automated heparin-protamine titration method. There are
some concerns that HMS may overestimate the heparin dose
required to maintain adequate anticoagulation as compared
with the ACT method.291
Thromboelastography
Thromboelastography (TEG) and rotational thromboelas-
tometry (ROTEM) are POCTs that measure clot formation,
clot strength, and clot dissolution. It has been suggested that
this technology better reflects in vivo hemostasis and thus may
provide more information than standard coagulation assays,
146 Section II—Clinical and Laboratory Assessment
especially in the perioperative setting. TEG analyzes hemostasis
using whole blood, which better accounts for the effect of cel-
lular elements in the coagulation process than the traditional
“cascade” model. The cascade model of coagulation is helpful
to understand hemostasis in vitro but does not fully explain
it in vivo. TEG has been shown to significantly reduce the use
of blood component therapy144,145 and overall blood loss.292
TEG yields several parameters of clotting.293 The reaction
time, R, represents the time until to initial fibrin formation
after mechanical stress and reflects the ability to generate
thrombin. Prolongation of this time reflects anticoagulant
activity, whereas a shortened R time suggests hypercoagulabil-
ity. The alpha, or angle parameter, represents the rate of fibrin
generation and cross linkage. The faster the rate of fibrin
generation, the larger is the measured angle. The K value also
measures clot kinetics; specifically it assesses the time it takes
to reach amplitude of 20 mm. The maximum amplitude
(MA) represents the tensile strength of the clot; the stronger
the clot, the larger the MA. Platelets contribute most to clot
strength and therefore contribute most significantly (80%) to
the MA, whereas fibrinogen, which binds platelets, contrib-
utes about 20%. The G value is an exponential expression of
the MA; it is more sensitive to changes in platelet function
than the MA at higher values. The coagulation index (CI)
represents the linear relationship between the R value, the K
value, the angle, and MA. A CI greater than 3 represents
hypercoagulability. The estimated percent lysis (EPL) and
LY30 (lysis after 30 minutes) are measures of clot dissolution.
The EPL is the percent change in the MA over time and the
LY30 is the amplitude 30 minutes after the MA divided by the
MA. An EPL greater than 7.5% and an LY30 greater than 15%
represent fibrinolysis; if associated with a CI less than 1 is it
primary fibrinolysis with a tendency toward bleeding, and if
CI is greater than 3 it is secondary fibrinolysis, not character-
ized by bleeding.
The analysis was originally performed on whole blood
without an activator. Subsequent modifications have been
developed that allow faster result generation and test stan-
dardization. The choice of activator depends on the clinical
setting. Rapid-thromboelastography (r-TEG), uses TF as the
activator; it allows for faster curve generation but compro-
mises some of the results (e.g., the R value). Without activa-
tion TEG takes 30 to 60 minutes to provide all information.
Quality assurance needs to be performed daily.
Since 2000, a variety of POC testing (POCT) devices have
become available to examine platelet function and measure
the effect of antiplatelet medications (see previous section on
Platelet Function Assays).
D-Dimer Assays
Combined with a pretest probability score, D-dimer assays can
be used to reliably exclude a diagnosis of VTE. Some D-dimer
assays use capillary blood whereas others use heparinized
blood or citrated plasma. The sensitivity of POC D-dimer tests
appears to be less sensitive than the best available clinical labo-
ratory tests.211,294-296
Conclusion
Anemia, bleeding, and thrombosis occur frequently in the
NCCU, and their treatment can be complicated. The diagnosis
of anemia, through serial laboratory testing, exacerbates the
condition. Anemic patients may be at higher risk of cerebral
ischemia. Neurosurgical patients are among the most frequent
of surgical populations to develop DVT and fatal PE given
postoperative immobility, hemiparesis, and altered mental
state. Combined modalities of thromboprophylaxis and inten-
sive surveillance may prove the best method to identify and
treat VTE. Understanding the pathogenesis of these various
conditions, suspicion that they may exist, and multimodality
monitoring in the NCCU may identify these disorders earlier,
thereby minimizing the devastating impact of these common
ailments.
References
1. Bunn HF. Approach to the anemias. 24th ed. Philadelphia: Elsevier; 2012.
2. Marks PW, Glader B. Approach to anemia in the adult and child. 5th ed.
Philadelphia: Churchill Livingstone; 2009.
3. Ginder G. Microcytic and hypochromic anemias. In: Goldman L, Schafer A,
editors. Goldman’s Cecil Textbook of Medicine. 24th ed. Philadelphia:
Elsevier; 2012. p. 1039–44.
4. Nutritional anaemias. Report of a WHO scientific group. World Health
Organ Tech Rep Ser 1968;405:5–37.
5. Hebert PC, Wells G, Tweeddale M, et al. Does transfusion practice affect
mortality in critically ill patients? Transfusion Requirements in Critical Care
(TRICC) investigators and the Canadian Critical Care Trials Group. Am J
Respir Crit Care Med 1997;155(5):1618–23.
6. Corwin HL. Blood transfusion in the critically ill patient. Dis Mon
1999;45(10):409–26.
7. Groeger JS, Guntupalli KK, Strosberg M, et al. Descriptive analysis of critical
care units in the United States: patient characteristics and intensive care unit
utilization. Crit Care Med 1993;21(2):279–91.
8. Carson JL, Duff A, Poses RM, et al. Effect of anaemia and cardiovascular
disease on surgical mortality and morbidity. Lancet 1996;348(9034):
1055–60.
9. Spiess BD, Ley C, Body SC, et al. Hematocrit value on intensive care
unit entry influences the frequency of Q-wave myocardial infarction after
coronary artery bypass grafting. The Institutions of the Multicenter Study of
Perioperative Ischemia (McSPI) research group. J Thorac Cardiovasc Surg
1998;116(3):460–7.
10. Kuriyan M, Carson JL. Anemia and clinical outcomes. Anesthesiol Clin
North America 2005;23(2):315–25, vii.
11. McKechnie RS, Smith D, Montoye C, et al. Prognostic implication of anemia
on in-hospital outcomes after percutaneous coronary intervention.
Circulation 2004;110(3):271–7.
12. Chant C, Wilson G, Friedrich JO. Anemia, transfusion, and phlebotomy
practices in critically ill patients with prolonged ICU length of stay: a cohort
study. Crit Care 2006;10(5):R140.
13. Corwin HL, Gettinger A, Pearl RG, et al. The CRIT Study: anemia and blood
transfusion in the critically ill—current clinical practice in the United States.
Crit Care Med 2004;32(1):39–52.
14. Vincent JL, Baron JF, Reinhart K, et al. Anemia and blood transfusion in
critically ill patients. JAMA 2002;288(12):1499–507.
15. Smoller BR, Kruskall MS. Phlebotomy for diagnostic laboratory tests in
adults: pattern of use and effect on transfusion requirements. N Engl J Med
1986;314(19):1233–5.
16. Shander A. Anemia in the critically ill. Crit Care Clin 2004;20(2):159–78.
17. Faquin WC, Schneider TJ, Goldberg MA. Effect of inflammatory cytokines
on hypoxia-induced erythropoietin production. Blood 1992;79(8):1987–94.
18. Rogiers P, Zhang H, Leeman M, et al. Erythropoietin response is blunted in
critically ill patients. Intensive Care Med 1997;23(2):159–62.
19. Rodriguez RM, Corwin HL, Gettinger A, et al. Nutritional deficiencies and
blunted erythropoietin response as causes of the anemia of critical illness.
J Crit Care 2001;16(1):36–41.
20. Buttarello M, Plebani M. Automated blood cell counts: state of the art. Am J
Clin Pathol 2008;130(1):104–16.
21. Stachon A, Kempf R, Holland-Letz T, et al. Daily monitoring of nucleated
red blood cells in the blood of surgical intensive care patients. Clin Chim
Acta 2006;366:329–35.
22. Parisotto R, Wu M, Ashenden MJ, et al. Detection of recombinant human
erythropoietin abuse in athletes utilizing markers of altered erythropoiesis.
Haematologica. 2001;86:128–37.
 Section II—Clinical and Laboratory Assessment 147
23. Zini G, DiMario A, Garzia M. Clinical usefulness of red cell fragments
identified by the Sysmex XE-2100 hematological analyser [lecture].
Proceeding of the Sysmex European Haematology Symposium. Lisbon,
Portugal; June 13-14, 2007.
24. Lesesve JF, Salignac S, Alla F, et al. Comparative evaluation of schistocyte
counting by an automated method and by microscopic determination. Am J
Clin Pathol 2004;121:739–45.
25. Berkow L, Rotolo S, Mirski E. Continuous noninvasive hemoglobin
monitoring during complex spine surgery. Anesth Analg 2011;113(6):
1396–402.
26. Frasca D, Dahyot-Fizelier C, Catherine K, et al. Accuracy of a continuous
noninvasive hemoglobin monitor in intensive care unit patients. Crit Care
Med 2011;39(10):2277–82.
27. Lamhaut L, Apriotesei R, Combes X, et al. Comparison of the accuracy of
noninvasive hemoglobin monitoring by spectrophotometry (SpHb) and
HemoCue with automated laboratory hemoglobin measurement.
Anesthesiology 2011;115(3):548–54.
28. Wartenberg KE, Schmidt JM, Claassen J, et al. Impact of medical
complications on outcome after subarachnoid hemorrhage. Crit Care Med
2006;34(3):617–23; quiz 24.
29. Kramer AH, Gurka MJ, Nathan B, et al. Complications associated with
anemia and blood transfusion in patients with aneurysmal subarachnoid
hemorrhage. Crit Care Med 2008;36(7):2070–5.
30. Van Beek JG, Mushkudiani NA, Steyerberg EW, et al. Prognostic value of
admission laboratory parameters in traumatic brain injury: results from the
IMPACT study. J Neurotrauma 2007;24(2):315–28.
31. Salim A, Hadjizacharia P, DuBose J, et al. Role of anemia in traumatic brain
injury. J Am Coll Surg 2008;207(3):398–406.
32. Steyerberg EW, Mushkudiani N, Perel P, et al. Predicting outcome after
traumatic brain injury: development and international validation of
prognostic scores based on admission characteristics. PLoS Med
2008;5(8):e165; discussion e.
33. Kramer AH, Zygun DA, Bleck TP, et al. Relationship between hemoglobin
concentrations and outcomes across subgroups of patients with aneurysmal
subarachnoid hemorrhage. Neurocrit Care 2009;10(2):157–65.
34. Kimberly WT, Wu O, Arsava EM, et al. Lower hemoglobin correlates with
larger stroke volumes in acute ischemic stroke. Cerebrovasc Dis Extra
2011;1(1):44–53.
35. Zygun DA, Nortje J, Hutchinson PJ, et al. The effect of red blood cell
transfusion on cerebral oxygenation and metabolism after severe traumatic
brain injury. Crit Care Med 2009;37(3):1074–8.
36. Kurtz P, Schmidt JM, Claassen J, et al. Anemia is associated with metabolic
distress and brain tissue hypoxia after subarachnoid hemorrhage. Neurocrit
Care 2010;13(1):10–6.
37. Oddo M, Milby A, Chen I, et al. Hemoglobin concentration and cerebral
metabolism in patients with aneurysmal subarachnoid hemorrhage. Stroke
2009;40:1275–81.
38. Leal-Noval SR, Munoz-Gomez M, Murillo-Cabezas F. Optimal hemoglobin
concentration in patients with subarachnoid hemorrhage, acute ischemic
stroke and traumatic brain injury. Curr Opin Crit Care 2008;14(2):
156–62.
39. Sahuquillo J, Poca MA, Garnacho A, et al. Early ischaemia after severe head
injury. Preliminary results in patients with diffuse brain injuries. Acta
Neurochir (Wien) 1993;122(3-4):204–14.
40. Vespa P, Bergsneider M, Hattori N, et al. Metabolic crisis without brain
ischemia is common after traumatic brain injury: a combined microdialysis
and positron emission tomography study. J Cereb Blood Flow Metab
2005;25(6):763–74.
41. Menon DK. Brain ischaemia after traumatic brain injury: lessons from
15O2 positron emission tomography. Curr Opin Crit Care 2006;12(2):
85–9.
42. Graham DI, Adams JH, Doyle D. Ischaemic brain damage in fatal
non-missile head injuries. J Neurol Sci 1978;39(2-3):213–34.
43. Vespa PM. The implications of cerebral ischemia and metabolic dysfunction
for treatment strategies in neurointensive care. Curr Opin Crit Care
2006;12(2):119–23.
44. McLaughlin MR, Marion DW. Cerebral blood flow and vasoresponsivity
within and around cerebral contusions. J Neurosurg 1996;85(5):871–6.
45. Carlson AP, Schermer CR, Lu SW. Retrospective evaluation of anemia and
transfusion in traumatic brain injury. J Trauma 2006;61(3):567–71.
46. Diringer MN, Bleck TP, Claude Hemphill J, 3rd, et al. Critical care
management of patients following aneurysmal subarachnoid hemorrhage:
recommendations from the Neurocritical Care Society’s Multidisciplinary
Consensus Conference. Neurocrit Care 2011;15(2):211–40.
47. Le Roux PD. Anemia and transfusion after subarachnoid hemorrhage.
Neurocrit Care 2011;15(2):342–53.
48. Vincent JL, Yagushi A, Pradier O. Platelet function in sepsis. Crit Care Med
2002;30(5 Suppl):S313–17.
49. Cawley MJ, Wittbrodt ET, Boyce EG, et al. Potential risk factors associated
with thrombocytopenia in a surgical intensive care unit. Pharmacotherapy
1999;19(1):108–13.
50. Baughman RP, Lower EE, Flessa HC, et al. Thrombocytopenia in the
intensive care unit. Chest 1993;104(4):1243–7.
A complete list of references for this chapter can be found online at
www.expertconsult.com.
 Section II—Clinical and Laboratory Assessment 147.e1
References
1. Bunn HF. Approach to the anemias. 24th ed. Philadelphia: Elsevier; 2012.
2. Marks PW, Glader B. Approach to anemia in the adult and child. 5th ed.
Philadelphia: Churchill Livingstone; 2009.
3. Ginder G. Microcytic and hypochromic anemias. In: Goldman L, Schafer A,
editors. Goldman’s Cecil Textbook of Medicine. 24th ed. Philadelphia:
Elsevier; 2012. p. 1039–44.
4. Nutritional anaemias. Report of a WHO scientific group. World Health
Organ Tech Rep Ser 1968;405:5–37.
5. Hebert PC, Wells G, Tweeddale M, et al. Does transfusion practice affect
mortality in critically ill patients? Transfusion Requirements in Critical
Care (TRICC) investigators and the Canadian Critical Care Trials Group.
Am J Respir Crit Care Med 1997;155(5):1618–23.
6. Corwin HL. Blood transfusion in the critically ill patient. Dis Mon
1999;45(10):409–26.
7. Groeger JS, Guntupalli KK, Strosberg M, et al. Descriptive analysis of
critical care units in the United States: patient characteristics and intensive
care unit utilization. Crit Care Med 1993;21(2):279–91.
8. Carson JL, Duff A, Poses RM, et al. Effect of anaemia and cardiovascular
disease on surgical mortality and morbidity. Lancet 1996;348(9034):
1055–60.
9. Spiess BD, Ley C, Body SC, et al. Hematocrit value on intensive care unit
entry influences the frequency of Q-wave myocardial infarction after
coronary artery bypass grafting. The Institutions of the Multicenter Study
of Perioperative Ischemia (McSPI) research group. J Thorac Cardiovasc
Surg 1998;116(3):460–7.
10. Kuriyan M, Carson JL. Anemia and clinical outcomes. Anesthesiol Clin
North America 2005;23(2):315–25, vii.
11. McKechnie RS, Smith D, Montoye C, et al. Prognostic implication of
anemia on in-hospital outcomes after percutaneous coronary intervention.
Circulation 2004;110(3):271–7.
12. Chant C, Wilson G, Friedrich JO. Anemia, transfusion, and phlebotomy
practices in critically ill patients with prolonged ICU length of stay: a
cohort study. Crit Care 2006;10(5):R140.
13. Corwin HL, Gettinger A, Pearl RG, et al. The CRIT Study: anemia and
blood transfusion in the critically ill—current clinical practice in the
United States. Crit Care Med 2004;32(1):39–52.
14. Vincent JL, Baron JF, Reinhart K, et al. Anemia and blood transfusion in
critically ill patients. JAMA 2002;288(12):1499–507.
15. Smoller BR, Kruskall MS. Phlebotomy for diagnostic laboratory tests in
adults: pattern of use and effect on transfusion requirements. N Engl J Med
1986;314(19):1233–5.
16. Shander A. Anemia in the critically ill. Crit Care Clin 2004;20(2):159–78.
17. Faquin WC, Schneider TJ, Goldberg MA. Effect of inflammatory cytokines
on hypoxia-induced erythropoietin production. Blood 1992;79(8):1987–94.
18. Rogiers P, Zhang H, Leeman M, et al. Erythropoietin response is blunted in
critically ill patients. Intensive Care Med 1997;23(2):159–62.
19. Rodriguez RM, Corwin HL, Gettinger A, et al. Nutritional deficiencies and
blunted erythropoietin response as causes of the anemia of critical illness.
J Crit Care 2001;16(1):36–41.
20. Buttarello M, Plebani M. Automated blood cell counts: state of the art. Am
J Clin Pathol 2008;130(1):104–16.
21. Stachon A, Kempf R, Holland-Letz T, et al. Daily monitoring of nucleated
red blood cells in the blood of surgical intensive care patients. Clin Chim
Acta 2006;366:329–35.
22. Parisotto R, Wu M, Ashenden MJ, et al. Detection of recombinant human
erythropoietin abuse in athletes utilizing markers of altered erythropoiesis.
Haematologica. 2001;86:128–37.
23. Zini G, DiMario A, Garzia M. Clinical usefulness of red cell fragments
identified by the Sysmex XE-2100 hematological analyser [lecture].
Proceeding of the Sysmex European Haematology Symposium. Lisbon,
Portugal; June 13-14, 2007.
24. Lesesve JF, Salignac S, Alla F, et al. Comparative evaluation of schistocyte
counting by an automated method and by microscopic determination. Am
J Clin Pathol 2004;121:739–45.
25. Berkow L, Rotolo S, Mirski E. Continuous noninvasive hemoglobin
monitoring during complex spine surgery. Anesth Analg
2011;113(6):1396–402.
26. Frasca D, Dahyot-Fizelier C, Catherine K, et al. Accuracy of a continuous
noninvasive hemoglobin monitor in intensive care unit patients. Crit Care
Med 2011;39(10):2277–82.
27. Lamhaut L, Apriotesei R, Combes X, et al. Comparison of the accuracy of
noninvasive hemoglobin monitoring by spectrophotometry (SpHb) and
HemoCue with automated laboratory hemoglobin measurement.
Anesthesiology 2011;115(3):548–54.
28. Wartenberg KE, Schmidt JM, Claassen J, et al. Impact of medical
complications on outcome after subarachnoid hemorrhage. Crit Care Med
2006;34(3):617–23; quiz 24.
29. Kramer AH, Gurka MJ, Nathan B, et al. Complications associated with
anemia and blood transfusion in patients with aneurysmal subarachnoid
hemorrhage. Crit Care Med 2008;36(7):2070–5.
30. Van Beek JG, Mushkudiani NA, Steyerberg EW, et al. Prognostic value of
admission laboratory parameters in traumatic brain injury: results from the
IMPACT study. J Neurotrauma 2007;24(2):315–28.
31. Salim A, Hadjizacharia P, DuBose J, et al. Role of anemia in traumatic brain
injury. J Am Coll Surg 2008;207(3):398–406.
32. Steyerberg EW, Mushkudiani N, Perel P, et al. Predicting outcome after
traumatic brain injury: development and international validation of
prognostic scores based on admission characteristics. PLoS Med
2008;5(8):e165; discussion e.
33. Kramer AH, Zygun DA, Bleck TP, et al. Relationship between hemoglobin
concentrations and outcomes across subgroups of patients with aneurysmal
subarachnoid hemorrhage. Neurocrit Care 2009;10(2):157–65.
34. Kimberly WT, Wu O, Arsava EM, et al. Lower hemoglobin correlates with
larger stroke volumes in acute ischemic stroke. Cerebrovasc Dis Extra
2011;1(1):44–53.
35. Zygun DA, Nortje J, Hutchinson PJ, et al. The effect of red blood cell
transfusion on cerebral oxygenation and metabolism after severe traumatic
brain injury. Crit Care Med 2009;37(3):1074–8.
36. Kurtz P, Schmidt JM, Claassen J, et al. Anemia is associated with metabolic
distress and brain tissue hypoxia after subarachnoid hemorrhage. Neurocrit
Care 2010;13(1):10–6.
37. Oddo M, Milby A, Chen I, et al. Hemoglobin concentration and cerebral
metabolism in patients with aneurysmal subarachnoid hemorrhage. Stroke
2009;40:1275–81.
38. Leal-Noval SR, Munoz-Gomez M, Murillo-Cabezas F. Optimal hemoglobin
concentration in patients with subarachnoid hemorrhage, acute ischemic
stroke and traumatic brain injury. Curr Opin Crit Care 2008;14(2):156–62.
39. Sahuquillo J, Poca MA, Garnacho A, et al. Early ischaemia after severe head
injury. Preliminary results in patients with diffuse brain injuries. Acta
Neurochir (Wien) 1993;122(3-4):204–14.
40. Vespa P, Bergsneider M, Hattori N, et al. Metabolic crisis without brain
ischemia is common after traumatic brain injury: a combined microdialysis
and positron emission tomography study. J Cereb Blood Flow Metab
2005;25(6):763–74.
41. Menon DK. Brain ischaemia after traumatic brain injury: lessons from
15O2 positron emission tomography. Curr Opin Crit Care 2006;12(2):
85–9.
42. Graham DI, Adams JH, Doyle D. Ischaemic brain damage in fatal
non-missile head injuries. J Neurol Sci 1978;39(2-3):213–34.
43. Vespa PM. The implications of cerebral ischemia and metabolic
dysfunction for treatment strategies in neurointensive care. Curr Opin Crit
Care 2006;12(2):119–23.
44. McLaughlin MR, Marion DW. Cerebral blood flow and vasoresponsivity
within and around cerebral contusions. J Neurosurg 1996;85(5):871–6.
45. Carlson AP, Schermer CR, Lu SW. Retrospective evaluation of anemia and
transfusion in traumatic brain injury. J Trauma 2006;61(3):567–71.
46. Diringer MN, Bleck TP, Claude Hemphill J, 3rd, et al. Critical care
management of patients following aneurysmal subarachnoid hemorrhage:
recommendations from the Neurocritical Care Society’s Multidisciplinary
Consensus Conference. Neurocrit Care 2011;15(2):211–40.
47. Le Roux PD. Anemia and transfusion after subarachnoid hemorrhage.
Neurocrit Care 2011;15(2):342–53.
48. Vincent JL, Yagushi A, Pradier O. Platelet function in sepsis. Crit Care Med
2002;30(5 Suppl):S313–17.
49. Cawley MJ, Wittbrodt ET, Boyce EG, et al. Potential risk factors associated
with thrombocytopenia in a surgical intensive care unit. Pharmacotherapy
1999;19(1):108–13.
50. Baughman RP, Lower EE, Flessa HC, et al. Thrombocytopenia in the
intensive care unit. Chest 1993;104(4):1243–7.
51. Stephan F, Hollande J, Richard O, et al. Thrombocytopenia in a surgical
ICU. Chest 1999;115(5):1363–70.
52. Hanes SD, Quarles DA, Boucher BA. Incidence and risk factors of
thrombocytopenia in critically ill trauma patients. Ann Pharmacother
1997;31(3):285–9.
53. Moreau D, Timsit JF, Vesin A, et al. Platelet count decline: an early
prognostic marker in critically ill patients with prolonged ICU stays. Chest
2007;131(6):1735–41.
54. Nijsten MW, ten Duis HJ, Zijlstra JG, et al. Blunted rise in platelet count in
critically ill patients is associated with worse outcome. Crit Care Med
2000;28(12):3843–6.
147.e2 Section II—Clinical and Laboratory Assessment
55. Napolitano LM, Warkentin TE, Almahameed A, et al. Heparin-induced
thrombocytopenia in the critical care setting: diagnosis and management.
Crit Care Med 2006;34(12):2898–911.
56. Gawaz M, Dickfeld T, Bogner C, et al. Platelet function in septic
multiple organ dysfunction syndrome. Intensive Care Med 1997;23(4):
379–85.
57. Mavrommatis AC, Theodoridis T, Orfanidou A, et al. Coagulation system
and platelets are fully activated in uncomplicated sepsis. Crit Care Med
2000;28(2):451–7.
58. Visentin GP, Liu CY. Drug-induced thrombocytopenia. Hematol Oncol Clin
North Am 2007;21(4):685–96, vi.
59. Bonfiglio MF, Traeger SM, Kier KL, et al. Thrombocytopenia in intensive
care patients: a comprehensive analysis of risk factors in 314 patients. Ann
Pharmacother 1995;29(9):835–42.
60. Shalansky SJ, Verma AK, Levine M, et al. Risk markers for
thrombocytopenia in critically ill patients: a prospective analysis.
Pharmacotherapy 2002;22(7):803–13.
61. Carey PJ. Drug-induced myelosuppression: diagnosis and management.
Drug Saf 2003;26(10):691–706.
62. Curtis BR, Kaliszewski J, Marques MB, et al. Immune-mediated
thrombocytopenia resulting from sensitivity to oxaliplatin. Am J Hematol
2006;81(3):193–8.
63. Pedersen-Bjergaard U, Andersen M, Hansen PB. Drug-specific
characteristics of thrombocytopenia caused by non-cytotoxic drugs. Eur J
Clin Pharmacol 1998;54(9-10):701–6.
64. Nguyen TC, Stegmayr B, Busund R, et al. Plasma therapies in thrombotic
syndromes. Int J Artif Organs 2005;28(5):459–65.
65. Pedersen-Bjergaard U, Andersen M, Hansen PB. Drug-induced
thrombocytopenia: clinical data on 309 cases and the effect of
corticosteroid therapy. Eur J Clin Pharmacol 1997;52(3):183–9.
66. Llevadot J, Coulter SA, Giugliano RP. A practical approach to the diagnosis
and management of thrombocytopenia associated with glycoprotein IIb/
IIIa receptor inhibitors. J Thromb Thrombolysis 2000;9(2):175–80.
67. Arepally GM, Ortel TL. Clinical practice. Heparin-induced
thrombocytopenia. N Engl J Med 2006;355(8):809–17.
68. Warkentin TE. Heparin-induced thrombocytopenia: pathogenesis and
management. Br J Haematol 2003;121(4):535–55.
69. Warkentin TE. Heparin-induced thrombocytopenia. Hematol Oncol Clin
North Am 2007;21(4):589–607, v.
70. Hoh BL, Aghi M, Pryor JC, et al. Neurosurgery 2005;57:243–48.
71. Warkentin TE, Kelton JG. Temporal aspects of heparin-induced
thrombocytopenia. N Engl J Med 2001;344(17):1286–92.
72. Lo GK, Juhl D, Warkentin TE, et al. Evaluation of pretest clinical score
(4 T’s) for the diagnosis of heparin-induced thrombocytopenia in two
clinical settings. J Thromb Haemost 2006;4:759–65.
73. Zheng XL, Sadler JE. Pathogenesis of thrombotic microangiopathies. Ann
Rev Pathol Mech Dis 2008;3:249–77.
74. Schwartzman RJ, Hill JB. Neurologic complications of disseminated
intravascular coagulation. Neurology 1982;32:791–7.
75. Voves C, Waillemin WA, Zeerleder S. International Society on Thrombosis
and Haemostasis score for overt disseminated intravascular coagulation
predicts organ dysfunction and fatality in sepsis patients. Blood Coagul
Fibrinolysis 2006;17:445–51.
76. Angstwurm MW, Dempfle CE, Spannagl M. New disseminated
intravascular coagulation score: a useful tool to predict mortality in
comparison with Acute Physiology and Chronic Health Evaluation II
and Logistic Organ Dysfunction scores. Crit Care Med 2006;34(2):314–20;
quiz 28.
77. Stohlawetz P, Stiegler G, Knobl P, et al. The rise of reticulated platelets after
intensive chemotherapy for AML reduces the need for platelet transfusion.
Ann Hematol 1999;78:271–3.
78. Chaoui D, Chakroun T, Robert F, et al. Reticulated platelets: a reliable
measure to reduce prophylactic platelet transfusions after intensive
chemotherapy. Transfusion 2005;45:766–72.
79. Rinder HM, Schuster JE, Rinder CS, et al. Correlation of thrombosis with
increased platelet turnover in thrombocytosis. Blood 1998;91:1288–94.
80. Bessman JD, Williams LJ, Gilmer PR. Mean platelet volume: the inverse
relation of platelet size and count in normal subjects, and an artifact of
other particles. Am J Clin Pathol 1981;76:289–93.
81. Endler G, Klimesch A, Sunder-Plassmann H, et al. Mean platelet volume is
an independent risk factor for myocardial infarction but not for coronary
artery disease. Br J Haematol 2002;117:399–404.
82. Arevalo-Lorido JC, Carretero-Gomez J, Alvarez-Oliva A, et al. Mean platelet
volume in acute phase of ischemic stroke, as predictor of mortality and
functional outcome after 1 year. J Stroke Cerebrovasc Dis 2011 Oct 15
(Epub ahead of print).
83. Ntaios G, Gurer O, Faouzi M, et al. Hypertension is an independent
predictor of mean platelet volume in patients with acute ischaemic stroke.
Intern Med J 2011;41(9):691–5.
84. Ha SI, Choi DH, Ki YJ, et al. Stroke prediction using mean platelet volume
in patients with atrial fibrillation. Platelets 2011;22(6):408–14.
85. Ilhan D, Ozbabalik D, Gulcan E, et al. Evaluation of platelet activation,
coagulation, and fibrinolytic activation in patients with symptomatic
lacunar stroke. Neurologist 2010;16(3):188–91.
86. D’Erasmo E, Aliberti G, Celi FS, et al. Platelet count, mean platelet volume
and their relation to prognosis in cerebral infarction. J Intern Med
1990;227(1):11–4.
87. O’Malley T, Langhorne P, Elton RA, et al. Platelet size in stroke patients.
Stroke 1995;26(6):995–9.
88. Hayward CP, Harrison P, Cattaneo M, et al. Platelet function analyzer
(PFA)-100 closure time in the evaluation of platelet disorders and platelet
function. J Thromb Haemost 2006;4(2):312–19.
89. Slaughter TF, Sreeram G, Sharma AD, et al. Reversible shear-mediated
platelet dysfunction during cardiac surgery as assessed by the PFA-100
platelet function analyzer. Blood Coagul Fibrinolysis 2001;12(2):85–93.
90. Harrison P, Segal H, Blasbery K, et al. Screening for aspirin responsiveness
after transient ischemic attack and stroke: comparison of 2 point-of-care
platelet function tests with optical aggregometry. Stroke 2005;36(5):
1001–5.
91. Raghavan M, Marik PE. Anemia, allogenic blood transfusion, and
immunomodulation in the critically ill. Chest 2005;127(1):295–307.
92. Becker C. Bled dry: fiscal woes threaten supply data from National Blood
Data Resource Center. Mod Healthc 2003;33(2):12–3.
93. Sullivan MT, Wallace EL. Blood collection and transfusion in the United
States in 1999. Transfusion 2005;45(2):141–8.
94. Napolitano LM, Kurek S, Luchette FA, et al. Clinical practice guideline: red
blood cell transfusion in adult trauma and critical care. Crit Care Med
2009;37(12):3124–57.
95. Kramer AH, Zygun DA. Anemia and red blood cell transfusion in
neurocritical care. Crit Care 2009;13:1–22.
96. Utter GH, Shahlaie K, Zwienenberg-Lee M, et al. Anemia in the setting of
traumatic brain injury: the arguments for and against liberal transfusion.
J Neurotrauma 2011;28:155–65.
97. Smith ML, Counelis GJ, Maloney-Wilensky E, et al. Brain tissue oxygen
tension in clinical brain death: a case series. Neurol Res 2007;29:755–9.
98. Dhar R, Zazulia AR, Videen TO, et al. Red blood cell transfusion increases
cerebral oxygen delivery in anemic patients with subarachnoid hemorrhage.
Stroke 2009;40:3039–44.
99. Card RT, Mohandas N, Perkins HA, et al. Deformability of stored red blood
cells. Relationship to degree of packing. Transfusion 1982;22(2):96–101.
100. Beutler E, Kuhl W, West C. The osmotic fragility of erythrocytes after
prolonged liquid storage and after reinfusion. Blood 1982;59(6):1141–7.
101. Card RT, Mohandas N, Mollison PL. Relationship of post-transfusion
viability to deformability of stored red cells. Br J Haematol
1983;53(2):237–40.
102. Tinmouth A, Fergusson D, Yee IC, et al. Clinical consequences of red cell
storage in the critically ill. Transfusion 2006;46(11):2014–27.
103. Napolitano LM, Corwin HL. Efficacy of red blood cell transfusion in the
critically ill. Crit Care Clin 2004;20(2):255–68.
104. Apstein CS, Dennis RC, Briggs L, et al. Effect of erythrocyte storage and
oxyhemoglobin affinity changes on cardiac function. Am J Physiol
1985;248(4 Pt 2):H508–15.
105. Hovav T, Yedgar S, Manny N, et al. Alteration of red cell aggregability and
shape during blood storage. Transfusion 1999;39(3):277–81.
106. Berezina TL, Zaets SB, Morgan C, et al. Influence of storage on red blood
cell rheological properties. J Surg Res 2002;102(1):6–12.
107. Knight JA, Voorhees RP, Martin L, et al. Lipid peroxidation in stored red
cells. Transfusion 1992;32(4):354–7.
108. Goodnough LT. Risks of blood transfusion. Crit Care Med 2003;31(12 Supp
l):S678–86.
109. Soderberg-Naucler C, Fish KN, et al. Reactivation of latent human
cytomegalovirus by allogeneic stimulation of blood cells from healthy
donors. Cell 1997;91(1):119–26.
110. Bowden RA, Sayers M, Flournoy N, et al. Cytomegalovirus immune
globulin and seronegative blood products to prevent primary
cytomegalovirus infection after marrow transplantation. N Engl J Med
1986;314(16):1006–10.
111. Chu RW. Leukocytes in blood transfusion: adverse effects and their
prevention. Hong Kong Med J 1999;5(3):280–4.
112. Phelan HA, Gonzalez RP, Patel HD, et al. Prestorage leukoreduction
ameliorates the effects of aging on banked blood. J Trauma
2010;69(2):330–7.
 Section II—Clinical and Laboratory Assessment 147.e3
113. Silliman CC, Moore EE, Kelher MR, et al. Identification of lipids that
accumulate during the routine storage of prestorage leukoreduced red
blood cells and cause acute lung injury. Transfusion 2011;51(12):2549–54.
114. Silliman CC, Clay KL, Thurman GW, et al. Partial characterization of lipids
that develop during the routine storage of blood and prime the neutrophil
NADPH oxidase. J Lab Clin Med 1994;124(5):684–94.
115. Shander A, Popovsky MA. Understanding the consequences of transfusion-
related acute lung injury. Chest 2005;128(5 Suppl 2):598S–604S.
116. Reese Jr EP, McCullough JJ, Craddock PR. An adverse pulmonary reaction
to cryoprecipitate in a hemophiliac. Transfusion 1975;15(6):583–8.
117. Offner PJ, Moore EE, Biffl WL, et al. Increased rate of infection associated
with transfusion of old blood after severe injury. Arch Surg
2002;137(6):711–16; discussion 6–7.
118. Edna TH, Bjerkeset T. Association between blood transfusion and infection
in injured patients. J Trauma 1992;33(5):659–61.
119. Moore FA, Moore EE, Sauaia A. Blood transfusion: an independent risk
factor for postinjury multiple organ failure. Arch Surg 1997;132(6):620–4;
discussion 4–5.
120. Agarwal N, Murphy JG, Cayten CG, et al. Blood transfusion increases the
risk of infection after trauma. Arch Surg 1993;128(2):171–6; discussion 6–7.
121. Claridge JA, Sawyer RG, Schulman AM, et al. Blood transfusions correlate
with infections in trauma patients in a dose-dependent manner. Am Surg
2002;68(7):566–72.
122. Malone DL, Dunne J, Tracy JK, et al. Blood transfusion, independent of
shock severity, is associated with worse outcome in trauma. J Trauma
2003;54(5):898–905; discussion 905-7.
123. Dunne JR, Malone DL, Tracy JK, et al. Allogenic blood transfusion in the
first 24 hours after trauma is associated with increased systemic
inflammatory response syndrome (SIRS) and death. Surg Infect (Larchmt)
2004;5(4):395–404.
124. Wobbes T, Bemelmans BL, Kuypers JH, et al. Risk of postoperative septic
complications after abdominal surgical treatment in relation to
perioperative blood transfusion. Surg Gynecol Obstet 1990;171(1):59–62.
125. Chelemer SB, Prato BS, Cox Jr PM, et al. Association of bacterial infection
and red blood cell transfusion after coronary artery bypass surgery. Ann
Thorac Surg 2002;73(1):138–42.
126. Leal-Noval SR, Jara-Lopez I, Garcia-Garmendia JL, et al. Influence of
erythrocyte concentrate storage time on postsurgical morbidity in cardiac
surgery patients. Anesthesiology 2003;98(4):815–22.
127. Leal-Noval SR, Rincon-Ferrari MD, Garcia-Curiel A, et al. Transfusion of
blood components and postoperative infection in patients undergoing
cardiac surgery. Chest 2001;119(5):1461–8.
128. Shorr AF, Jackson WL, Kelly KM, et al. Transfusion practice and blood
stream infections in critically ill patients. Chest 2005;127(5):1722–8.
129. Taylor RW, Manganaro L, O’Brien J, et al. Impact of allogenic packed red
blood cell transfusion on nosocomial infection rates in the critically ill
patient. Crit Care Med 2002;30(10):2249–54.
130. Marik PE, Corwin HL. Efficacy of red blood cell transfusion in the critically
ill: a systematic review of the literature. Crit Care Med 2008;36(9):2667–74.
131. Fransen E, Maessen J, Dentener M, et al. Impact of blood transfusions on
inflammatory mediator release in patients undergoing cardiac surgery.
Chest 1999;116(5):1233–9.
132. Zallen G, Moore EE, Ciesla DJ, et al. Stored red blood cells selectively
activate human neutrophils to release IL-8 and secretory PLA2. Shock
2000;13(1):29–33.
133. Stec N, Shirey RS, Smith B, et al. The efficacy of performing red cell elution
studies in the pretransfusion testing of patients with positive direct
antiglobulin tests. Transfusion 1986;26(3):225–6.
134. Heddle NM, Klama LN, Griffith L, et al. A prospective study to identify the
risk factors associated with acute reactions to platelet and red cell
transfusions. Transfusion 1993;33(10):794–7.
135. Hebert PC, Fergusson D, Blajchman MA, et al. Clinical outcomes following
institution of the Canadian universal leukoreduction program for red blood
cell transfusions. JAMA 2003;289(15):1941–9.
136. Dzik WH, Anderson JK, O’Neill EM, et al. A prospective, randomized
clinical trial of universal WBC reduction. Transfusion 2002;42(9):
1114–22.
137. Watkins TR, Rubenfeld GD, Martin TR, et al. Effects of leukoreduced blood
on acute lung injury after trauma: a randomized controlled trial. Crit Care
Med 2008;36(5):1493–9.
138. Nathens AB, Nester TA, Rubenfeld GD, et al. The effects of leukoreduced
blood transfusion on infection risk following injury: a randomized
controlled trial. Shock 2006;26(4):342–7.
139. Hogman CF, Meryman HT. Storage parameters affecting red blood cell
survival and function after transfusion. Transfus Med Rev
1999;13(4):275–96.
140. Prins HA, Houdijk AP, Nijveldt RJ, et al. Arginase release from red blood
cells: possible link in transfusion induced immune suppression? Shock
2001;16(2):113–15.
141. Smoller BR, Kruskall MS, Horowitz GL. Reducing adult phlebotomy blood
loss with the use of pediatric-sized blood collection tubes. Am J Clin Pathol
1989;91(6):701–3.
142. Khor B, Van Cott EM. Laboratory evaluation of hypercoagulability. Clin
Lab Med 2009;29(2):339–66.
143. Perry DJ, Fitzmaurice DA, Kitchen S, et al. Point-of-care testing in
haemostasis. Br J Haematol 2010;150(5):501–14.
144. Ak K, Isbir CS, Tetik S, et al. Thromboelastography-based transfusion
algorithm reduces blood product use after elective CABG: a prospective
randomized study. J Card Surg 2009;24(4):404–10.
145. Shore-Lesserson L, Manspeizer HE, DePerio M, et al.
Thromboelastography-guided transfusion algorithm reduces transfusions in
complex cardiac surgery. Anesth Analg 1999;88(2):312–19.
146. Foulke GE, Harlow DJ. Effective measures for reducing blood loss from
diagnostic laboratory tests in intensive care unit patients. Crit Care Med
1989;17(11):1143–5.
147. Natanson C, Kern SJ, Lurie P, et al. Cell-free hemoglobin-based blood
substitutes and risk of myocardial infarction and death: a meta-analysis.
JAMA 2008;299(19):2304–12.
148. Moore EE, Moore FA, Fabian TC, et al. Human polymerized hemoglobin
for the treatment of hemorrhagic shock when blood is unavailable: the
USA multicenter trial. J Am Coll Surg 2009;208(1):1–13.
149. Harrington JP, Wollocko H. Pre-clinical studies using OxyVita hemoglobin,
a zero-linked polymeric hemoglobin: a review. J Artif Organs
2010;13(4):183–8.
150. Mammis A, McIntosh TK, Maniker AH. Erythropoietin as a
neuroprotective agent in traumatic brain injury. Surg Neurol
2009;71(5):527–31; discussion 31.
151. Assaraf MI, Diaz Z, Liberman A, et al. Brain erythropoietin receptor
expression in Alzheimer disease and mild cognitive impairment. J
Neuropathol Exp Neurol 2007;66(5):389–98.
152. Ehrenreich H, Kästner A, Weissenborn K, et al. Circulating damage marker
profiles support a neuroprotective effect of erythropoietin in ischemic
stroke patients. Mol Med 2011;17(11–12):1306–10.
153. Corwin HL, Gettinger A, Rodriguez RM, et al. Efficacy of recombinant
human erythropoietin in the critically ill patient: a randomized, double-
blind, placebo-controlled trial. Crit Care Med 1999;27(11):2346–50.
154. Corwin HL, Gettinger A, Pearl RG, et al. Efficacy of recombinant human
erythropoietin in critically ill patients: a randomized controlled trial. JAMA
2002;288(22):2827–35.
155. Gabriel A, Kozek S, Chiari A, et al. High-dose recombinant human
erythropoietin stimulates reticulocyte production in patients with multiple
organ dysfunction syndrome. J Trauma 1998;44(2):361–7.
156. van Iperen CE, Gaillard CA, Kraaijenhagen RJ, et al. Response of
erythropoiesis and iron metabolism to recombinant human erythropoietin
in intensive care unit patients. Crit Care Med 2000;28(8):2773–8.
157. Corwin HL, Gettinger A, Fabian TC, et al. Efficacy and safety of epoetin
alfa in critically ill patients. N Engl J Med 2007;357(10):965–76.
158. Epoetin alfa [package insert]. Thousand Oaks, CA. 2008; Amgen Inc.
159. Tseng MY, Hutchinson PJ, Richards HK, et al. Acute systemic erythropoietin
therapy to reduce delayed ischemic deficits following aneurysmal
subarachnoid hemorrhage: a phase II randomized, double-blind,
placebo-controlled trial. Clinical article. J Neurosurg 2009;111(1):171–80.
160. Ehrenreich H, Weissenborn K, Prange H, et al. Recombinant human
erythropoietin in the treatment of acute ischemic stroke. Stroke
2009;40(12):e647–56.
161. Schafer AI. Approach to the patient with bleeding and thrombosis. In:
Goldman I, Schafer A, editors. Goldman’s Cecil Textbook of Medicine. 24th
ed. Philadelphia: Elsevier; 2012. p. 1121–4.
162. Lim W, Crowther MA, Ginsberg JS. Venous thromboembolism. In:
Hoffman R, Furie B, Benz EJ, et al, editors. Hematology. 5th ed.
Philadelphia: Churchill Livingstone; 2009. p. 2043–54.
163. Haines ST. Venous thromboembolism: pathophysiology and clinical
presentation. Am J Health Syst Pharm 2003;60(22 Suppl 7):S3–5.
164. Becker DM. Venous thromboembolism: epidemiology, diagnosis,
prevention. J Gen Intern Med 1986;1(6):402–11.
165. Anderson Jr FA, Wheeler HB, Goldberg RJ, et al. A population-based
perspective of the hospital incidence and case-fatality rates of deep vein
thrombosis and pulmonary embolism. The Worcester DVT Study. Arch
Intern Med 1991;151(5):933–8.
166. Cohen AT, Tapson VF, Bergmann JF, et al. Venous thromboembolism risk
and prophylaxis in the acute hospital care setting (ENDORSE study): a
multinational cross-sectional study. Lancet 2008;371(9610):387–94.
147.e4 Section II—Clinical and Laboratory Assessment
167. Schonhofer B, Kohler D. Prevalence of deep-vein thrombosis of the leg in
patients with acute exacerbation of chronic obstructive pulmonary disease.
Respiration 1998;65(3):173–7.
168. Harris LM, Curl GR, Booth FV, et al. Screening for asymptomatic deep
vein thrombosis in surgical intensive care patients. J Vasc Surg 1997;26(5):
764–9.
169. Twigg SJ, McCrirrick A, Sanderson PM. A comparison of post mortem
findings with post hoc estimated clinical diagnoses of patients who die in a
United Kingdom intensive care unit. Intensive Care Med 2001;27(4):706–10.
170. Bostrom S, Holmgren E, Jonsson O, et al. Post-operative thromboembolism
in neurosurgery: a study on the prophylactic effect of calf muscle
stimulation plus dextran compared to low-dose heparin. Acta Neurochir
(Wien) 1986;80(3-4):83–9.
171. Quevedo JF, Buckner JC, Schmidt JL, et al. Thromboembolism in patients
with high-grade glioma. Mayo Clin Proc 1994;69(4):329–32.
172. Abrahams JM, Torchia MB, McGarvey M, et al. Perioperative assessment of
coagulability in neurosurgical patients using thromboelastography. Surg
Neurol 2002;58(1):5–11; discussion 11-2.
173. Juhan-Vague I, Conte-Devolx B, Aillaud MF, et al. Effects of DDAVP and
venous occlusion on the release of tissue-type plasminogen activator and
von Willebrand factor in patients with panhypopituitarism. Thromb Res
1984;33(6):653–9.
174. Hamer JD, Malone PC, Silver IA. The PO2 in venous valve pockets: its
possible bearing on thrombogenesis. Br J Surg 1981;68(3):166–70.
175. Agnelli G, Piovella F, Buoncristiani P, et al. Enoxaparin plus compression
stockings compared with compression stockings alone in the prevention of
venous thromboembolism after elective neurosurgery. N Engl J Med
1998;339(2):80–5.
176. Auguste KI, Quinones-Hinojosa A, Berger MS. Efficacy of mechanical
prophylaxis for venous thromboembolism in patients with brain tumors.
Neurosurg Focus 2004;17(4):E3.
177. Raslan AM, Fields JD, Bhardwaj A. Prophylaxis for venous thrombo-
embolism in neurocritical care: a critical appraisal. Neurocrit Care 2010;
12(2):297–309.
178. Cerrato D, Ariano C, Fiacchino F. Deep vein thrombosis and low-dose
heparin prophylaxis in neurosurgical patients. J Neurosurg 1978;49(3):
378–81.
179. Powers SK, Edwards MS. Prophylaxis of thromboembolism in the
neurosurgical patient: a review. Neurosurgery 1982;10(4):509–13.
180. Smith SF, Simpson JM, Sekhon LH. Prophylaxis for deep venous
thrombosis in neurosurgical oncology: review of 2779 admissions over a
9-year period. Neurosurg Focus 2004;17(4):E4.
181. Swann KW, Black PM. Deep vein thrombosis and pulmonary emboli in
neurosurgical patients: a review. J Neurosurg 1984;61(6):1055–62.
182. Valladares JB, Hankinson J. Incidence of lower extremity deep vein
thrombosis in neurosurgical patients. Neurosurgery 1980;6(2):138–41.
183. Hamilton MG, Hull RD, Pineo GF. Prophylaxis of venous thromboembolism
in brain tumor patients. J Neurooncol 1994;22(2):111–26.
184. Altschuler E, Moosa H, Selker RG, et al. The risk and efficacy of
anticoagulant therapy in the treatment of thromboembolic complications
in patients with primary malignant brain tumors. Neurosurgery
1990;27(1):74–6; discussion 7.
185. Hamilton MG, Hull RD, Pineo GF. Venous thromboembolism in
neurosurgery and neurology patients: a review. Neurosurgery 1994;34(2):
280–96; discussion 96.
186. Sawaya R, Zuccarello M, Elkalliny M, et al. Postoperative venous
thromboembolism and brain tumor. Part I. Clinical profile. J Neurooncol
1992;14(2):119–25.
187. Marras LC, Geerts WH, Perry JR. The risk of venous thromboembolism is
increased throughout the course of malignant glioma: an evidence-based
review. Cancer 2000;89(3):640–6.
188. Sawaya R, Hammoud M, Schoppa D, et al. Neurosurgical outcomes in a
modern series of 400 craniotomies for treatment of parenchymal tumors.
Neurosurgery 1998;42(5):1044–55; discussion 55–6.
189. Sawaya R, Decourteen-Meyers G, Copeland B. Massive preoperative
pulmonary embolism and suprasellar brain tumor: case report and review
of the literature. Neurosurgery 1984;15(4):566–71.
190. Sawaya R, Cummins CJ, Kornblith PL. Brain tumors and plasmin
inhibitors. Neurosurgery 1984;15(6):795–800.
191. Turpie AG, Hirsh J, Gent M, et al. Prevention of deep vein thrombosis in
potential neurosurgical patients: a randomized trial comparing graduated
compression stockings alone or graduated compression stockings plus
intermittent pneumatic compression with control. Arch Intern Med
1989;149(3):679–81.
192. Hamada K, Kuratsu J, Saitoh Y, et al. Expression of tissue factor correlates
with grade of malignancy in human glioma. Cancer 1996;77(9):1877–83.
193. Sciacca FL, Ciusani E, Silvani A, et al. Genetic and plasma markers of
venous thromboembolism in patients with high grade glioma. Clin Cancer
Res 2004;10(4):1312–17.
194. Chan AT, Atiemo A, Diran LK, et al. Venous thromboembolism occurs
frequently in patients undergoing brain tumor surgery despite prophylaxis.
J Thromb Thrombolysis 1999;8(2):139–42.
195. Muchmore JH, Dunlap JN, Culicchia F, Kerstein MD. Deep vein
thrombophlebitis and pulmonary embolism in patients with malignant
gliomas. South Med J 1989;82(11):1352–6.
196. Danish SF, Burnett MG, Ong JG, et al. Prophylaxis for deep venous
thrombosis in craniotomy patients: a decision analysis. Neurosurgery
2005;56(6):1286–92; discussion 92–4.
197. Misra M, Roitberg B, Ebersole K, et al. Prevention of pulmonary embolism
by combined modalities of thromboprophylaxis and intensive surveillance
protocol. Neurosurgery 2004;54(5):1099–102; discussion 102–3.
198. Guyatt GH, Akl EA, Crowther M, et al. Executive summary: antithrombotic
therapy and prevention of thrombosis, 9th ed: American College of chest
physicians evidence based clinical practice guidelines. Chest
2012;141:7S–47S.
199. Geerts WH, Bergqvist D, Pineo GF, et al. Prevention of venous
thromboembolism. American College of Chest Physicians
Evidence-Based Clinical Practice Guidelines (8th ed). Chest 2008;
133(6 Suppl):381S–453S.
200. Kamphuisen PW, Agnelli G. What is the optimal pharmacological
prophylaxis for the prevention of deep-vein thrombosis and pulmonary
embolism in patients with acute ischemic stroke? Thromb Res
2007;119(3):265–74.
201. Sherman DG, Albers GW, Bladin C, et al. The efficacy and safety of
enoxaparin versus unfractionated heparin for the prevention of venous
thromboembolism after acute ischaemic stroke (PREVAIL study): an
open-label randomised comparison. Lancet 2007;369(9570):1347–55.
202. Goldstein JN, Fazen LE, Wendell L, et al. Risk of thromboembolism
following acute intracerebral hemorrhage. Neurocrit Care 2009;10(1):
28–34.
203. Lacut K, Bressollette L, Le Gal G, et al. Prevention of venous thrombosis in
patients with acute intracerebral hemorrhage. Neurology 2005;65(6):
865–9.
204. Boeer A, Voth E, Henze T, et al. Early heparin therapy in patients with
spontaneous intracerebral haemorrhage. J Neurol Neurosurg Psychiatry
1991;54(5):466–7.
205. Tetri S, Hakala J, Juvela S, et al. Safety of low-dose subcutaneous
enoxaparin for the prevention of venous thromboembolism after primary
intracerebral haemorrhage. Thromb Res 2008;123(2):206–12.
206. Broderick J, Connolly S, Feldmann E, et al. Guidelines for the management
of spontaneous intracerebral hemorrhage in adults, 2007 update: a
guideline from the American Heart Association/American Stroke
Association Stroke Council, High Blood Pressure Research Council, and the
Quality of Care and Outcomes in Research Interdisciplinary Working
Group. Stroke 2007;38(6):2001–23.
207. Morgenstern LB, Hemphill 3rd JC, Anderson C, et al. Guidelines for the
management of spontaneous intracerebral hemorrhage: a guideline for
healthcare professionals from the American Heart Association/American
Stroke Association. Stroke 2010;41(9):2108–29.
208. Palareti G, Cosmi B, Legnani C. Diagnosis of deep vein thrombosis. Semin
Thromb Hemost 2006;32(7):659–72.
209. Bates SM, Ginsberg JS. Clinical practice. Treatment of deep-vein
thrombosis. N Engl J Med 2004;351(3):268–77.
210. Segal JB, Eng J, Tamariz LJ, et al. Review of the evidence on diagnosis of
deep venous thrombosis and pulmonary embolism. Ann Fam Med
2007;5(1):63–73.
211. Heim SW, Schectman JM, Siadaty MS, et al. D-dimer testing for deep
venous thrombosis: a metaanalysis. Clin Chem 2004;50(7):1136–47.
212. Wells PS, Anderson DR, Rodger M, et al. Evaluation of D-dimer in the
diagnosis of suspected deep-vein thrombosis. N Engl J Med
2003;349(13):1227–35.
213. Crowther MA, Cook DJ, Griffith LE, et al. Neither baseline tests of
molecular hypercoagulability nor D-dimer levels predict deep venous
thrombosis in critically ill medical-surgical patients. Intensive Care Med
2005;31(1):48–55.
214. Engelberger RP, Aujesky D, Calanca L, et al. Comparison of the diagnostic
performance of the original and modified Wells score in inpatients and
outpatients with suspected deep vein thrombosis. Thromb Res
2011;127(6):535–9.
215. Soto MJ, Grau E, Gadelha T, et al. Clinical significance of a negative
D-dimer level in patients with confirmed venous thromboembolism:
findings from the RIETE Registry. J Thromb Haemost 2011;9(2):407–10.
 Section II—Clinical and Laboratory Assessment 147.e5
216. Flinn WR, Sandager GP, Silva MB, Jr., et al. Prospective surveillance for
perioperative venous thrombosis: experience in 2643 patients. Arch Surg
1996;131(5):472–80.
217. Sud S, Mittmann N, Cook DJ, et al. Screening and prevention of venous
thromboembolism in critically ill patients: a decision analysis and
economic evaluation. Am J Respir Crit Care Med 2011;184(11):1289–98.
218. Stam J. Thrombosis of the cerebral veins and sinuses. N Engl J Med
2005;352(17):1791–8.
219. Saposnik G, Barinagarrementeria F, Brown RD Jr., et al. Diagnosis and
management of cerebral venous thrombosis: a statement for healthcare
professionals from the American Heart Association/American Stroke
Association. Stroke 2011;42(4):1158–92.
220. Wilder-Smith E, Kothbauer-Margreiter I, Lammle B, et al. Dural puncture
and activated protein C resistance: risk factors for cerebral venous sinus
thrombosis. J Neurol Neurosurg Psychiatry 1997;63(3):351–6.
221. Wilterdink JL, Easton JD. Cerebral ischemia in pregnancy. Adv Neurol
2002;90:51–62.
222. Lanska DJ, Kryscio RJ. Peripartum stroke and intracranial venous
thrombosis in the National Hospital Discharge survey. Obstet Gynecol
1997;89(3):413–18.
223. Cucchiara B, Messe S, Taylor R, et al. Utility of D-dimer in the diagnosis of
cerebral venous sinus thrombosis. J Thromb Haemost 2005;3(2):387–9.
224. Kosinski CM, Mull M, Schwarz M, et al. Do normal D-dimer levels reliably
exclude cerebral sinus thrombosis? Stroke 2004;35(12):2820–5.
225. Stam J, De Bruijn SF, DeVeber G. Anticoagulation for cerebral sinus
thrombosis. Cochrane Database Syst Rev 2002;(4):CD002005.
226. Furie KL, Kasner SE, Adams RJ, et al. Guidelines for the prevention of
stroke in patients with stroke or transient ischemic attack: a guideline for
healthcare professionals from the American Heart Association/American
Stroke Association. Stroke 2011;42(1):227–76.
227. Cohn DM, Roshani S, Middeldorp S. Thrombophilia and venous
thromboembolism: implications for testing. Semin Thromb Hemost
2007;33(6):573–81.
228. Fields MC, Levine SR. Thrombophilias and stroke: diagnosis, treatment,
and prognosis. J Thromb Thrombolysis 2005;20(2):113–26.
229. Van Cott EM, Soderberg BL, Laposata M. Activated protein C resistance,
the factor V Leiden mutation, and a laboratory testing algorithm. Arch
Pathol Lab Med 2002;126(5):577–82.
230. van den Belt AG, Sanson BJ, Simioni P, et al. Recurrence of venous
thromboembolism in patients with familial thrombophilia. Arch Intern
Med 1997;157(19):2227–32.
231. Griffin JH, Heeb MJ, Kojima Y, et al. Activated protein C resistance:
molecular mechanisms. Thromb Haemost 1995;74(1):444–8.
232. Rahemtullah A, Van Cott EM. Hypercoagulation testing in ischemic stroke.
Arch Pathol Lab Med 2007;131(6):890–901.
233. Poort SR, Rosendaal FR, Reitsma PH, Bertina RM. A common genetic
variation in the 3’-untranslated region of the prothrombin gene is
associated with elevated plasma prothrombin levels and an increase in
venous thrombosis. Blood 1996;88(10):3698–703.
234. Folsom AR, Cushman M, Tsai MY, et al. A prospective study of venous
thromboembolism in relation to factor V Leiden and related factors. Blood
2002;99(8):2720–5.
235. Koster T, Blann AD, Briet E, et al. Role of clotting factor VIII in effect of
von Willebrand factor on occurrence of deep-vein thrombosis. Lancet
1995;345(8943):152–5.
236. van Hylckama Vlieg A, van der Linden IK, Bertina RM, et al. High levels of
factor IX increase the risk of venous thrombosis. Blood
2000;95(12):3678–82.
237. Meijers JC, Tekelenburg WL, Bouma BN, et al. High levels of coagulation
factor XI as a risk factor for venous thrombosis. N Engl J Med
2000;342(10):696–701.
238. Brey RL, Muscal E, Chapman J. Antiphospholipid antibodies and the brain:
a consensus report. Lupus 2011;20:153–7.
239. Urbanus RT, Siegerink B, Roest M, et al. Antiphospholipid antibodies and
risk of myocardial infarction and ischaemic stroke in young women in the
RATIO study: a case-control study. Lancet Neurol 2009;8:998–1005.
240. Ruiz-Irastorza G, Cuadrado MJ, Ruiz-Arruza I, et al. Evidence-based
recommendations for the prevention and long-term management of
thrombosis in antiphospholipid antibody-positive patients: report of a task
force at the 13th International Congress on Antiphospholipid Antibodies.
Lupus 2011;20:206–18.
241. Andersson LO, Barrowcliffe TW, Holmer E, et al. Anticoagulant properties
of heparin fractionated by affinity chromatography on matrix-bound
antithrombin III and by gel filtration. Thromb Res 1976;9(6):575–83.
242. Johnson EA, Mulloy B. The molecular-weight range of mucosal-heparin
preparations. Carbohydr Res 1976;51(1):119–27.
243. Harenberg J. Pharmacology of low molecular weight heparins. Semin
Thromb Hemost 1990;(16 Suppl):12–8.
244. Hirsh J, Bauer KA, Donati MB, et al. Parenteral anticoagulants: American
College of Chest Physicians Evidence-Based Clinical Practice Guidelines
(8th ed.). Chest 2008;133(6 Suppl):141S–59S.
245. Tollefsen DM, Majerus DW, Blank MK. Heparin cofactor II. Purification
and properties of a heparin-dependent inhibitor of thrombin in human
plasma. J Biol Chem 1982;257(5):2162–9.
246. Young E, Prins M, Levine MN, et al. Heparin binding to plasma proteins,
an important mechanism for heparin resistance. Thromb Haemost
1992;67(6):639–43.
247. Whitfield LR, Lele AS, Levy G. Effect of pregnancy on the relationship
between concentration and anticoagulant action of heparin. Clin
Pharmacol Ther 1983;34(1):23–8.
248. Olson JD, Arkin CF, Brandt JT, et al. College of American Pathologists
conference XXXI on laboratory monitoring of anticoagulant therapy:
laboratory monitoring of unfractionated heparin therapy. Arch Pathol Lab
Med 1998;122(9):782–98.
249. Marks PW. Treatment of venous thromboembolism. In: Colman RW,
Marder VJ, Clowes AW, et al, editors. Thrombosis and Hemostasis: Basic
Principles and Clinical Practice. Philadelphia: Lippincott Williams &
Wilkins; 2006. p. 1307–16.
250. Hirsh J, van Aken WG, Gallus AS, et al. Heparin kinetics in venous
thrombosis and pulmonary embolism. Circulation 1976;53(4):691–5.
251. Levine MN, Hirsh J, Gent M, et al. A randomized trial comparing activated
thromboplastin time with heparin assay in patients with acute venous
thromboembolism requiring large daily doses of heparin. Arch Intern Med
1994;154(1):49–56.
252. Edson JR, Krivit W, White JG. Kaolin partial thromboplastin time: high
levels of procoagulants producing short clotting times or masking
deficiencies of other procoagulants or low concentrations of anticoagulants.
J Lab Clin Med 1967;70(3):463–70.
253. Antman EM. Hirudin in acute myocardial infarction: safety report from the
Thrombolysis and Thrombin Inhibition in Myocardial Infarction (TIMI)
9A Trial. Circulation 1994;90(4):1624–30.
254. GUSTO Investigators. Randomized trial of intravenous heparin versus
recombinant hirudin for acute coronary syndromes: the Global Use of
Strategies to Open Occluded Coronary Arteries (GUSTO) IIa investigators.
Circulation 1994;90:1631–7.
255. EPIC Investigators. Use of a monoclonal antibody directed against the
platelet glycoprotein IIb/IIIa receptor in high-risk coronary angioplasty: the
EPIC investigation. N Engl J Med 1994;330:956–61.
256. Basu D, Gallus A, Hirsh J, Cade J. A prospective study of the value of
monitoring heparin treatment with the activated partial thromboplastin
time. N Engl J Med 1972;287(7):324–7.
257. Rosborough TK. Comparing different lots of activated partial
thromboplastin time reagent: analysis of two methods. Am J Clin Pathol
1998;110(2):173–7.
258. Baker BA, Adelman MD, Smith PA, et al. Inability of the activated partial
thromboplastin time to predict heparin levels: time to reassess guidelines
for heparin assays. Arch Intern Med 1997;157(21):2475–9.
259. Bates SM, Grand’Maison A, Johnston M, et al. A latex D-dimer reliably
excludes venous thromboembolism. Arch Intern Med 2001;161(3):447–53.
260. Bates SM, Weitz JI, Johnston M, et al. Use of a fixed activated partial
thromboplastin time ratio to establish a therapeutic range for
unfractionated heparin. Arch Intern Med 2001;161(3):385–91.
261. Brandt JT, Triplett DA. Laboratory monitoring of heparin. Effect of
reagents and instruments on the activated partial thromboplastin time. Am
J Clin Pathol 1981;76(4 Suppl):530–7.
262. van den Besselaar AM, Meeuwisse-Braun J, Bertina RM. Monitoring
heparin therapy: relationships between the activated partial thromboplastin
time and heparin assays based on ex-vivo heparin samples. Thromb
Haemost 1990;63(1):16–23.
263. Holmer E, Kurachi K, Soderstrom G. The molecular-weight dependence of
the rate-enhancing effect of heparin on the inhibition of thrombin, factor
Xa, factor IXa, factor XIa, factor XIIa and kallikrein by antithrombin.
Biochem J 1981;193(2):395–400.
264. Holmer E, Soderberg K, Bergqvist D, et al. Heparin and its low molecular
weight derivatives: anticoagulant and antithrombotic properties.
Haemostasis 1986;16(Suppl 2):1–7.
265. Hirsh J, Weitz JI. New antithrombotic agents. Lancet 1999;353(9162):
1431–6.
266. Choay J, Petitou M, Lormeau JC, et al. Structure-activity relationship in
heparin: a synthetic pentasaccharide with high affinity for antithrombin III
and eliciting high anti-factor Xa activity. Biochem Biophys Res Commun
1983;116(2):492–9.
147.e6 Section II—Clinical and Laboratory Assessment
267. Choay J. Biologic studies on chemically synthesized pentasaccharide and
tetrasaccharide fragments. Semin Thromb Hemost 1985;11(2):81–5.
268. Bijsterveld NR, Moons AH, Boekholdt SM, et al. Ability of recombinant
factor VIIa to reverse the anticoagulant effect of the pentasaccharide
fondaparinux in healthy volunteers. Circulation 2002;106(20):2550–4.
269. Kuo KH, Kovacs MJ. Fondaparinux: a potential new therapy for HIT.
Hematology 2005;10(4):271–5.
270. Wallis RB. Hirudins: from leeches to man. Semin Thromb Hemost
1996;22(2):185–96.
271. Lefevre G, Duval M, Gauron S, et al. Effect of renal impairment on the
pharmacokinetics and pharmacodynamics of desirudin. Clin Pharmacol
Ther 1997;62(1):50–9.
272. Salemi A, Agrawal YP, Fontes MA. An assay to monitor bivalirudin levels on
cardiopulmonary bypass. Ann Thorac Surg 2011;92(1):332–4.
273. Connolly SJ, Eikelboom J, Joyner C, et al. Apixaban in patients with atrial
fibrillation. N Engl J Med 2011;364:806–17.
274. Connolly SJ, Ezekowitz MD, Yusuf S, et al. Dabigatran versus warfarin in
patients with atrial fibrillation. N Engl J Med 2009;361(12):1139–51.
275. Ansell J, Hirsh J, Hylek E, et al. Pharmacology and management of the
vitamin K antagonists. American College of Chest Physicians Evidence-
Based Clinical Practice Guidelines (8th ed.). Chest 2008;133:160S–98S.
276. Shehab N, Sperling LS, Kegler SR, et al. National estimates of emergency
department visits for hemorrhage-related adverse events from clopidogrel
plus aspirin and from warfarin. Arch Intern Med 2010;170:1926–33.
277. Lassen MR, Gallus A, Raskob GE, et al. Apixaban versus enoxaparin for
thromboprophylaxis after hip replacement. N Engl J Med 2010;363:
2487–98.
278. Lassen MR, Raskob GE, Gallus A, et al. Apixaban versus enoxaparin for
thromboprophylaxis after knee replacement (ADVANCE-2): a randomised
double-blind trial. Lancet 2010;375:807–15.
279. Chavez JJ, Weatherall JS, Strevels SM, et al. Evaluation of a point-of-care
coagulation analyzer on patients undergoing cardiopulmonary bypass
surgery. J Clin Anesth 2004;16(1):7–10.
280. Merlani PG, Chenaud C, Cottini S, et al. Point of care management of
heparin administration after heart surgery: a randomized, controlled trial.
Intensive Care Med 2006;32(9):1357–64.
281. Zabel KM, Granger CB, Becker RC, et al. Use of bedside activated partial
thromboplastin time monitor to adjust heparin dosing after thrombolysis
for acute myocardial infarction: results of GUSTO-I. Global Utilization of
Streptokinase and TPA for Occluded Coronary Arteries. Am Heart J
1998;136(5):868–76.
282. Schroeder AP, Knudsen LL, Husted SE, et al. Bedside coagulometry during
intravenous heparin therapy after coronary angioplasty. J Thromb
Thrombolysis 2001;12(2):157–63.
283. Eiswirth G, Walch S, Bommer J. New bedside test for monitoring
anticoagulation during hemodialysis. Artif Organs 1998;22(4):346–8.
284. Reiner JS, Coyne KS, Lundergan CF, et al. Bedside monitoring of heparin
therapy: comparison of activated clotting time to activated partial
thromboplastin time. Cathet Cardiovasc Diagn 1994;32(1):49–52.
285. Smythe MA, Koerber JM, Westley SJ, et al. Use of the activated partial
thromboplastin time for heparin monitoring. Am J Clin Pathol
2001;115(1):148–55.
286. Ferring M, Reber G, de Moerloose P, et al. Point of care and central
laboratory determinations of the aPTT are not interchangeable in surgical
intensive care patients. Can J Anaesth 2001;48(11):1155–60.
287. Reiss RA, Haas CE, Griffis DL, et al. Point-of-care versus laboratory
monitoring of patients receiving different anticoagulant therapies.
Pharmacotherapy 2002;22(6):677–85.
288. Choi TS, Greilich PE, Shi C, et al. Point-of-care testing for prothrombin
time, but not activated partial thromboplastin time, correlates with
laboratory methods in patients receiving aprotinin or epsilon-aminocaproic
acid while undergoing cardiac surgery. Am J Clin Pathol 2002;117(1):74–8.
289. Jaryno S, Bennett K, Loder C, et al. Validation of a new whole blood
coagulation monitoring system. J Extra Corpor Technol 2002;34(4):
271–5.
290. Wang JS, Lin CY, Karp RB. Comparison of high-dose thrombin time with
activated clotting time for monitoring of anticoagulant effects of heparin in
cardiac surgical patients. Anesth Analg 1994;79(1):9–13.
291. Despotis GJ, Levine V, Joist JH, et al. Antithrombin III during cardiac
surgery: effect on response of activated clotting time to heparin and
relationship to markers of hemostatic activation. Anesth Analg
1997;85(3):498–506.
292. Royston D, von Kier S. Reduced haemostatic factor transfusion using
heparinase-modified thrombelastography during cardiopulmonary bypass.
Br J Anaesth 2001;86(4):575–8.
293. Gonzalez E, Pieracci FM, Moore EE, et al. Coagulation abnormalities in the
trauma patient: the role of point of care thromboelastography. Semin
Thromb Hemost 2010;36:723–37.
294. Baker J, Quick H, Hullsiek KH, et al. Interleukin-6 and d-dimer levels are
associated with vascular dysfunction in patients with untreated HIV
infection. HIV Med 2010;11(9):608–9.
295. Dempfle CE, Suvajac N, Elmas E, et al. Performance evaluation of a new
rapid quantitative assay system for measurement of D-dimer in plasma and
whole blood: PATHFAST D-dimer. Thromb Res 2007;120(4):591–6.
296. Dempfle CE, Korte W, Schwab M, et al. Sensitivity and specificity of a
quantitative point of care D-dimer assay using heparinized whole blood, in
patients with clinically suspected deep vein thrombosis. Thromb Haemost
2006;96(1):79–83.
II
Chapter
16  
Monitoring Inflammation
Alejandro M. Spiotta, Alan Siu, and J. Javier Provencio
Introduction
The basic tenet of neurocritical care is prevention of secondary
injury. Secondary injuries following the initial ictus include
global increases in intracranial pressure (ICP), alterations in
cerebral perfusion, hydrocephalus, and seizures, among others.
In addition, systemic insults such as infections, hypotension,
thromboemboli, hyperglycemia, hyperthermia, malnutrition,
and other complications can affect patients in the neurocritical
care unit (NCCU) adversely. Accumulating evidence suggests
that inflammation can exacerbate secondary injury in brain
disease and in particular subarachnoid hemorrhage, vaso-
spasm, traumatic brain injury (TBI), and stroke; it can even
be epileptogenic.1-6 Monitoring neuroinflammation therefore
may allow better understanding of the impact of therapies
aimed at preventing or minimizing secondary injury. Whether
prevention of inflammation will improve patient outcome fol-
lowing brain injury is still to be defined because although
pharmacologic therapies to attenuate inflammation can ame-
liorate pathology, microglial activation that is central to the
inflammatory response also plays a role in recovery and regen-
eration (i.e., inflammation may have a dual role).
Monitoring Neuroinflammation
As the understanding of the pathophysiologic process follow-
ing acute brain injury including subarachnoid hemorrhage
(SAH), stroke, and TBI has increased, awareness about the
deleterious effects of inflammatory activation as both a trigger
and exacerbating factor of secondary injury has created new
opportunities in monitoring neuroinflammation. For example,
findings from clinical studies suggest that neutrophils play a
critical role in ischemia-reperfusion injury, because tissue
damage is associated with extent of neutrophil infiltration.7-9
The critical role that systemic as well as local inflammatory
activators play in determining the extent of central nervous
system (CNS) injury following an ictus is highlighted by
the emergence and widespread use of anti-inflammatory
pharmaceuticals, such as statins, in neurocritical care.10-14
3-Hydroxy-3-methylglutaryl coenzyme A (HMG CoA) reduc-
tase inhibitors, commonly known as statins, have lipid-
lowering properties, and animal models of acute brain injury
have demonstrated their anti-inflammatory properties.15-18
There is mounting evidence for the role of neuroinflammation
as a contributor to secondary brain tissue injury in stroke,
148
neurotrauma, SAH, and various chronic CNS diseases. Inflam-
matory mediators are indeed elevated in the cerebrospinal
fluid (CSF) and serum of patients with acute and chronic
brain injury, and are associated with poor outcome.19-25 Various
animal models also point to potential protective mechanisms
with the inhibition of proinflammatory mediators. As the role
of neuroinflammation in secondary brain injury becomes
more evident, the need to monitor its activity in the NCCU
becomes important.
The knowledge that may be gained by monitoring neuro-
inflammation and its clinical utility can be organized into
three main areas:
1. Identifying risk factors to predict secondary injury
2. Providing endpoints to monitor responses to specific ther-
apies aimed at modulating the inflammatory state
3. Gaining insights into the pathophysiology of the disease
process
Monitoring of neuroinflammation thus may provide a more
sensitive and earlier detection of secondary injury and in turn
widen the therapeutic window. Furthermore, studying the
response of the various measured parameters to conventional
and experimental medical and interventional therapies should
augment our knowledge of the disease course and potentially
lead to new therapeutic strategies.
Basic Science of Inflammation
The brain was once considered to be an “immune-privileged”
organ system, inaccessible and independent from systemic
inflammation. Although the CNS has unique features, this
view has largely fallen out of favor because there is accumulat-
ing evidence of crosstalk between the CNS and the immune
system outside of the CNS. For example, CNS cells have been
shown to modulate peripheral immune function in response
to peripheral disease and injury through the systemic release
of various cytokines (i.e., interleukin-1 [IL-1], IL-6, tumor
necrosis factor-α [TNFα]) and vagal efferents.26,27
Cerebral edema has many potential causes but can be con-
sidered an inflammatory process. Indeed IL-1 neutralizing
antibodies in animal models can attenuate brain edema.28
Because the skull is functionally a closed box, significant
development of edema occupies the excess space in the
cranium, after which pressure increases and so can damage
neurons and glia. Cerebral edema occurs at different times in
© Copyright 2013 Elsevier Inc. All rights reserved.
 Section II—Clinical and Laboratory Assessment 149
different cerebral injuries. In TBI and SAH, cerebral edema
develops quickly and can cause increased ICP in many patients.
Conversely, in ischemic stroke, cerebral edema develops slowly
over the first 72 hours after the initial ictus. The reason for the
differences in these time courses is not well defined; however,
the course of cerebral edema could be predicted by monitor-
ing inflammation.
Inflammatory processes contribute to CNS injury and neu-
rodegeneration.20,27 In addition there is accumulating evidence
that inflammation and in particular leukocyte-endothelial cell
interactions, play a critical role in the pathogenesis of vaso-
spasm after SAH2,29 and that inflammatory processes contrib-
ute to formation of cerebral aneurysms.30-32 Various types of
acute and chronic brain injuries result in gliosis, a reaction
characterized by the activation, proliferation, and hypertrophy
of mononuclear phagocytic-type cells,33-35 which occurs in a
well-defined manner to promote a systematic restoration of
the brain parenchyma in the setting of neuronal injury. This
process is mediated by the resident astrocytes and microglia,
which produce various chemical mediators to recruit periph-
erally derived immune cells to further enhance the inflamma-
tory response. The inhibition of these chemical mediators
(i.e., TNFα, IL-1) and the key components that modulate
leukocyte trafficking across the blood-brain barrier (BBB)
have shown great potential to improve outcome in neuroin-
flammatory diseases such as multiple sclerosis and hold the
same potential in acute brain injury (natalizumab therapy).
Biology of the Brain’s
Immune System
Inflammatory processes within the CNS are different from the
periphery because the CNS parenchyma lacks resident den-
dritic cells, with inflammation instead being initiated by the
resident microglia.36 The exact role of microglia remains
under debate because they release a number of factors that
modulate both secondary injury and recovery after injury,
including pro- and anti-inflammatory cytokines, chemokines,
nitric oxide, prostaglandins, growth factors, and superoxide
species.37 The most unique quality, however, is the presence of
a BBB that regulates the subsequent immune amplification by
the peripheral inflammatory cells. The BBB, along with the
neuroepithelial cells, modulate the inflammatory mediators
and ultimately the peripheral leukocytes that participate in a
neuroinflammatory state.36,38-41 These key mediators, namely
the cytokines and adhesion molecules, are the focus of current
drug development.42
CNS injury can result in acute or chronic neurodegenera-
tion. In the acute setting, conditions such as TBI and cerebral
ischemia lead to parenchymal injury that is characterized by
substantial nerve cell loss from excitotoxicity (i.e., glutamate
release), oxidative stress, and regional electrolyte disturbances
among other pathophysiologic processes. Proinflammatory
cytokines are elevated in human CSF and serum after TBI. In
addition, increased microglial activation can be detected in
vivo using the positron emission tomography (PET) ligand
[11C](R)PK11195 even years after an acute injury.43 Animal
models of acute injury have further implicated these proin-
flammatory cytokines as contributors to the extent of brain
injury because their inhibition can reduce the extent of cell
death. Extensive investigations into various conditions of
chronic neurodegeneration such as multiple sclerosis and
Alzheimer’s disease have also implicated a role for inflamma-
tion,44,45 because cytokines such as TNFα and interferon-γ
(IFNγ) are elevated in the CSF. In addition, in vitro studies
point to inflammation’s damaging effects on neurons, whereas
cytokine inhibition in animal models slows the disease
process.46 Interestingly, immune responses also are necessary
for neuroregeneration, suggesting there is an innate balance
between adaptive and deleterious immune activation or
immunomodulation.47
The entry of leukocytes into the CNS parenchyma depends
on the expression of specific molecules by the endothelium
(intracellular adhesion molecules [ICAMs], integrins, selec-
tins, and chemokines) along the capillary and postcapillary
venules in the brain. The production of chemokines and the
activation of the endothelium are initiated by the activated
parenchymal microglia, which produce the cytokines and
other inflammatory mediators to further stimulate the inflam-
matory cascade.36
The activated endothelium is marked by an up-regulation
of cell-adhesion molecules (i.e., selectins and integrins) that
act to increase interactions to anchor peripheral leukocytes
to promote transmigration. In this multistep paradigm of
leukocyte-endothelial interactions to cross the BBB, the first
step is tethering, which is characterized by a transient contact
between the glycosylated ligands on leukocytes and selectins
on the endothelium. The next step is rolling, which is medi-
ated by the shear forces of flowing blood to promote locomo-
tion of the leukocyte and increase exposure to activating
factors (i.e., cytokines, chemokines) immobilized on the
endothelial luminal surface. Further activation of the leuko-
cytes is characterized by conformational changes of integrins
into a higher affinity state for stronger interactions with endo-
thelial cell-adhesion molecules (ICAM-1, vascular cell adhe-
sion molecule 1 [VCAM-1]), resulting in leukocyte arrest onto
the endothelial surface. The arrested leukocytes then can
extend protrusions enriched in chemokine receptors, and
extravasate across the BBB by either a paracellular or transcel-
lular route. Various matrix metalloproteinases (MMPs) are
secreted to facilitate the transmigration process.
The Febrile State: Monitoring Brain
and Body Temperature
Fever is the oldest and most easily measured marker of an
underlying inflammatory condition, with careful records of
elevated temperature and its relationship to disease course
dating back to antiquity.48 Raised core body temperature is
associated with worse outcome following a variety of acute
neurologic insults.49-54 The exact mechanism by which hyper-
thermia exacerbates CNS injury is not yet fully elucidated.
Animal infarct models demonstrate that hyperthermia can
enhance the depth and extent of injury by one of two mecha-
nisms: increased cytokine release and inflammation or
increased energy expenditure and a widened metabolic gap.55-
59
However, hyperthermia in a brain-injured patient has many
disparate causes. First, pharmaceutical and infectious etiolo-
gies are common.60-62 Second, there may be direct hypotha-
lamic injury. Third, factors such as heat shock protein (HSP)
synthesis, release of proinflammatory cytokines such as IL-1,
IL-6, TNFα, and interferon, or increased glutamate63-65 each,
150 Section II—Clinical and Laboratory Assessment
or in combination, may act as the stimulus to trigger fever
after TBI (for review see reference 62). Factors that may aggra-
vate neurologic outcome in the setting of hyperthermia
include increased glucocorticoid production, increased meta-
bolic expenditure, altered oxygen consumption, glutamate or
nitric oxide release, increased myeloid cell and cytokine activ-
ity, or increased vascular permeability.58,66-70 Given the complex
interplay between factors for fever following SAH, it is not
surprising that fever has been found to be an independent risk
factor for the development of vasospasm following SAH in
some studies,53,54 whereas others have not found this relation-
ship.71 Fever at admission also is a risk factor for vasospasm
after TBI72 and is associated with poor clinical grade after
aneurysm rupture.73
It is now possible to routinely monitor brain temperature
(BT; see Chapter 37). Although brain and body temperature
may correlate after SAH and TBI, average BT is usually greater
than the average rectal temperature by about 1° C. The differ-
ence between brain and body temperature becomes greater as
a patient becomes febrile.74-77 The effect of BT on brain physi-
ology is not well understood. Some studies have demonstrated
that fever may be associated with increased ICP,75 while others
suggest that brain hyperthermia may not affect brain oxygen78
or may even increase it.79 In part these different results may
be associated with changes in cerebral blood volume or cere-
bral blood flow. On the other hand, microdialysis studies
suggest that fever control can attenuate markers of cerebral
energy dysfunction in the brain.80 However, exactly what the
relationship between BT and energy dysfunction is, is not
clear. The clinical utility of BT monitoring, and in particular
the relationship of BT with cerebral metabolism requires
more study.
Serum Inflammatory Markers
Systemic indices of inflammation revolve around markers of
acute phase reactants such as serum white blood cell (WBC)
count and C reactive protein (CRP). Elevation in WBCs can
occur in the setting of infection, inflammation, and stress or
tissue necrosis. The limitation of these parameters is their
nonspecific nature in CNS injury. This can limit an interpreta-
tion about CNS inflammation because elevation of these
parameters may reflect other organ injury or secondary infec-
tion. Despite these limitations, several converging lines of
evidence suggest that monitoring acute phase reactants may
help develop models to predict secondary injury following
brain injury.81
SAH is a disease in which inflammation monitoring may be
the most advantageous and in particular to evaluate vaso-
spasm.2,82-84 For example, leukocytosis has been found to be an
independent risk factor for delayed cerebral ischemia (DCI)
following aneurysmal SAH, with each increase of 1000 in the
serum WBC associated with a 9% increase in the likelihood
of developing DCI.73,85 Furthermore, a peak serum WBC
greater than 15,000 is associated with a 3.3-fold greater inci-
dence of DCI,73 whereas a CSF neutrophil content of greater
than 62% on the third day after SAH is an independent factor
associated with later vasospasm.86
The relationship between CRP, vasospasm and the develop-
ment of DCI is still to be fully elucidated.85,87 For example,
Rothoerl et al.87 observed that CRP values were higher in
patients who developed DCI 5 to 8 days after aneurysm
rupture. This likely represents the response to ischemia. There
was a trend observed for higher CRP values on the preceding
days but these did not achieve significance. Serum CRP also
was greater in poor-grade patients, which makes it difficult to
conclude a “causal” relationship between a heightened acute
phase response and vasospasm, or that CRP simply reflects the
severity of the initial brain injury.
In ischemic stroke, CRP and leukocytosis also are indepen-
dent factors associated with outcome including cognitive
outcome among survivors.88-90 In addition, several lines of
evidence demonstrate a relationship between progression of
infarction and worse neurologic outcome and elevated serum
cytokines IL-1, IL-6, and TNFα.89,91-95 What is unclear is
whether there is a causal relationship.
Cerebrospinal Fluid
Inflammatory Markers
Cytokines and downstream activators of the inflammatory
response can be measured in the CSF. This is better than brain
biopsy but there are limitations to CSF sampling including:
(1) sampling error when relative concentrations of proteins
and molecules must diffuse into the CSF from a heteroge-
neously injured brain (i.e., the measured values likely reflect
a weighted average from surrounding tissue and its proximity
to the subarachnoid space); (2) CSF levels of individual cyto-
kines may be related to their diffusion properties in fluids and
distance from the source; (3) it is restricted to intermittent
(e.g., daily) sampling; and (4) different half-lives of these
inflammatory markers both in the CSF and collecting vials. In
addition, it should be remembered that most cytokines act
over short distances to effect local cells. It may therefore be
desirable in some patients to obtain tissue and fluid closest to
the injury, to allow for more specific sampling of brain
inflammation.
Several clinical studies suggest that CSF and blood are
functionally separate compartments when cytokine concen-
trations are measured. For example, Shiozaki et  al.96 com-
pared CSF and serum concentrations of several cytokines
from patients suffering from multiorgan trauma and TBI to
those with isolated TBI. The concentrations of the proinflam-
matory cytokines TNFα and IL-1 were greater in CSF than
serum in those with isolated brain injury. In patients with
multiorgan trauma, serum levels were greater than in CSF,
whereas CSF concentrations of both proinflammatory and
anti-inflammatory cytokines were similar to those with iso-
lated CNS injury. There was an association between Injury
Severity Score, an index of systemic injury burden, and
tumor necrosis factor-α receptor 1 (TNFr1). Patients with
elevated ICP had higher concentrations of CSF IL-1, IL-10,
IL-Ira, and TNFr1 than those with normal ICP. These results
suggest that CSF cytokine monitoring will more accurately
reflect CNS pathology than serum cytokines.
Hydrocephalus that requires external ventricular drainage
is common following SAH; hence CSF samples often are avail-
able. SAH has been found to cause an increase in cytokines
IL-1, IL-6, transforming growth factor-β (TGFβ) and TNFα
released from leukocytes in CSF21,97-99 and so monitoring the
inflammatory response and in particular IL-6 maybe a useful
tool to help guide care of SAH patients.84 However, in
SAH patients who develop systemic inflammatory response
 Section II—Clinical and Laboratory Assessment 151
syndrome (SIRS) ventricular CSF cytokine levels do not
appear to correlate with serum levels of IL-1 and TNFα.97 In
addition, although SIRS is associated with poor outcome, it
may not always be associated with vasospasm.100,101 In general
SIRS is more common in patients in poor clinical grade, and
some but not all studies suggest it is associated with how an
aneurysm is occluded.100,101
IL-1 and IL-6 levels consistently have been observed to be
greater in poor-grade than good-grade patients although an
initial inflammatory response is observed in patients of all
clinical grades.102 There also appears to be a relationship with
DCI and inflammatory markers particularly in the CSF.102,103
For example, Mathiesen et al.21 analyzed CSF daily following
SAH and found IL-1 to increase during delayed ischemic neu-
rologic deficit (DIND). Elevated IL-1 and TNFα levels also
correlated with poor outcome.21 Kwon and Jeon found that
admission IL-6 concentration correlated with delayed isch-
emic deficits.98 SAH also appears to induce the production of
nitric oxide metabolites, as inferred by the association with
increased levels of nitric oxide breakdown products, nitrite
and nitrate (NOx), in CSF.104 However, nitric oxide metabo-
lites in CSF appear to be different from serum concentra-
tions.104,105 In patients who develop vasospasm and DCI, CSF
concentrations of NOx are greater between 2 and 8 days after
aneurysm rupture.106
In TBI, CSF cytokine concentrations are elevated.65,107-109
The highest concentrations usually are measured during the
first 24 hours, after which there is a decline.110 However, there
is a difference between CSF and serum concentrations of
various cytokines.96,111,112 The difference between plasma and
CSF levels of cytokines does not appear to be associated with
BBB breakdown. This suggests that the cytokines are released
by microglia, astrocytes, and neurons112 rather than simply
reflecting “spill-over” from the blood. There are conflicting
data on what cytokine levels in the CSF mean. Some studies
show that IL-1 concentration in CSF is associated with unfa-
vorable outcome96,108 following TBI. Other studies, however,
suggest that peak CSF IL-6 levels are associated with improved
outcome,113 IL-6 deficiency is associated with poor outcome,114
and up-regulation of IL-10 production may play a neuro­
protective role by attenuating TNFα release.111,115
Methods to Sample Cytokines
and Monitor Neuroinflammation
in Clinical Practice
Immune system cells and brain cells produce cytokines that
act as intercellular signaling molecules. Depending on their
concentration, cytokines can have toxic or trophic effects and
can be pro- or anti-inflammatory.116 In addition several cyto-
kines can have synergistic effects, and therefore a full under-
standing of the interactions between both deleterious and
protective cytokines rather than just the absolute concentra-
tion of a particular cytokine may be needed to derive mean-
ingful interpretation of clinical data. Several other factors
make clinical interpretation of human cytokine data complex.
First, each cytokine is produced during a defined time period
and so results may depend on when the samples are obtained.
Second, cytokines are produced in different amounts within
different compartments—blood, brain, extracellular fluid
(ECF), and CSF—and hence what is sampled will influence
the findings. It appears that IL-1b and vascular endothelial
growth factor (VEGF) levels are greater in brain ECF than in
CSF, whereas IL-6, TNF, IL-8 and IL-1ra are greater in CSF
than in microdialysate.3 Third, there are complex interactions
of the immune system with the BBB and blood-CSF barriers
that influence the findings. Finally, it is unclear whether it is
the peak cytokine level, the area under the curve, or the spe-
cific cytokine production relative to other mediators that is
most important to the biologic effect.
In clinical practice there are four basic approaches to
examine the role of inflammatory mediators: (1) blood sam-
pling (both arterial and jugular venous), (2) CSF sampling,
(3) microdialysis, and (4) direct tissue sampling from ex vivo
or postmortem tissue. In addition, PET studies may be used
to examine microglial activation. Blood sampling is the easiest
technique because it is readily available, can be easily repeated,
requires no specialized equipment, and cytokines can be
screened for with multiplex assay techniques. However, cyto-
kines in arterial blood samples may indicate both intracranial
and extracranial pathology. Paired arterial and jugular venous
sampling can be used to help examine the relative contribu-
tions of the brain or the systemic inflammatory response.
However, detection of a gradient is not always feasible because
the absolute concentrations of some cytokines are close to the
limit of sensitivity of the assay techniques.
External ventricular drainage when used provides a useful
method to evaluate CSF inflammation because CSF produc-
tion of cytokines is thought to represent brain production.
However, the concentration of a cytokine may depend on how
the CSF is drained (i.e., intermittently or continuously)117 and
CSF production. In addition, it is unclear whether CSF drain-
age is a method to eliminate various cytokines (i.e., is “thera-
peutic”). This then raises the question of the biologic relevance
of any immune mediator detected in the CSF.
Microdialysis continuously samples the brain interstitial
space (see Chapter 36). This may be the most biologically
relevant compartment because inflammatory mediators are
secreted here and act on their surface membrane target recep-
tors. It can be difficult to assay cytokines and inflammatory
mediators using microdialysis because these proteins have a
greater molecular weight than the metabolic intermediaries
(glucose, lactate, pyruvate) commonly examined with this
technique. Consequently cytokines diffuse across the micro-
dialysis membrane at a slower rate, and a smaller proportion
crosses into the catheter (i.e., the relative recovery is low).
Several techniques can be used to address this, including use
of catheters with larger molecular weight cutoff membranes
(e.g., 100 kDa), addition of colloids (such as dextrans or
albumin) to the perfusate, and use of sensitive multiplex tech-
niques to assay the cytokines (e.g., the Luminex xMAP plat-
form). In the laboratory setting cytokine binders (e.g., heparin)
have been added to the perfusion fluid. This has not been used
clinically because it is conceivable that heparin is lost from the
perfusate into the ECF.118
Future Investigation
Further Understanding of the Dual Effects
of Cytokines Following Acute Brain Injury
Acute brain injury is followed by a cytokine-triggered influx
of lymphocytes and myeloid cells into the injured region.
152 Section II—Clinical and Laboratory Assessment
These infiltrating cells are further stimulated in an autocrine
and paracrine fashion to produce an inflammatory environ-
ment, which may contribute to further tissue injury. However,
some signals are adaptive and confer neuroprotection as well
as promote tissue repair.119 For example, IL-6 is both a neuro-
tropic and inflammatory factor.120 The complex interplay of
the various factors and downstream effectors involved in an
inflammatory response in the setting of acute brain injury is
only beginning to be elucidated.
The major challenge when monitoring cytokines as markers
of inflammation is that many cytokines can exert different
effects that depend on the local environment. This also pres-
ents an obstacle when interpreting animal and clinical studies
that relate inflammatory mediators and extent of brain injury
or outcome. It is likely a balance between the various impli-
cated cytokines and their intracellular signaling cascades that
determines whether the inflammatory response is adaptive or
maladaptive. Thus, monitoring cytokines in isolation may not
be useful unless its role is incorporated into the greater context
of the inflammatory environment and individualized to each
particular patient.
Further Investigation on the Interplay
Between Systemic and CNS
Inflammatory Pathways
The relationship between cerebral inflammatory response and
the systemic, extracerebral response needs to be better under-
stood. They are distinct but not independent pathways, and it
is likely that systemic factors influence those occurring in the
CNS and that the neuroinflammatory pathway modulates the
systemic cascade.
Continuous, Bedside Monitoring of
Neuroinflammation
It is feasible to measure intrathecal interleukins at bedside
using point-of-care testing with an immunoassay chip-test.
These bedside results are similar to those obtained with labo-
ratory testing.81 Microdialysis techniques to sample inflamma-
tory mediators in the tissue most at risk for secondary injury,
may further advance the ability to monitor neuroinflamma-
tion.110 For example, the balance between members of the IL-1
cytokine family, in particular between IL-1b and its endoge-
nous inhibitor IL-1ra, may be more important than absolute
levels of IL-1b. In TBI patients, the microdialysate IL-1ra:IL-1b
ratio is greater in patients with favorable rather than unfavor-
able outcomes.121 The critical factor is whether following
trends in neuroinflammation heralds the onset of secondary
injury in a timely and predictable manner to alter manage-
ment, or whether it is a late sign at a time when intervention
is not likely to affect the outcome.
Conclusion
Inflammatory processes, initiated by the resident microglia
within the CNS, occur after a variety of acute neurological
insults. These processes may have a dual role, that is, inflam-
matory responses can play a role in recovery whereas accumu-
lating evidence indicates that inflammation can aggravate
secondary injury in conditions such as TBI, stroke, or SAH.
Consistent with this, animal studies suggest that the inhibition
of proinflammatory mediators can provide neuroprotection.
Monitoring of inflammatory markers in the CNS that may be
measured in the serum, CSF, or brain interstitial fluid using
microdialysis therefore will likely play an increasing role in the
NCCU.
References
1. Mashaly HA, Provencio JJ. Inflammation as a link between brain injury and
heart damage: the model of subarachnoid hemorrhage. Cleve Clin J Med
2008;75(Suppl 2):S26–30.
2. Pradilla G, Chaichana KL, Hoang S, et al. Inflammation and cerebral
vasospasm after subarachnoid hemorrhage. Neurosurg Clin N Am 2010;
21(2):365–79.
3. Helmy A, De Simoni MG, Guilfoyle MR, et al. Cytokines and innate
inflammation in the pathogenesis of human traumatic brain injury. Prog
Neurobiol 2011;95(3):352–72. Epub 2011 Sep 16.
4. Brough D, Tyrrell PJ, Allan SM. Regulation of interleukin-1 in acute brain
injury. Trends Pharmacol Sci 2011;32(10):617–22. Epub 2011 Jul 23.
5. Ravizza T, Balosso S, Vezzani A. Inflammation and prevention of
epileptogenesis. Neurosci Lett 2011;497(3):223–30. Epub 2011 Feb 26.
6. Denes A, Thornton P, Rothwell NJ, et al. Inflammation and brain injury:
acute cerebral ischaemia, peripheral and central inflammation. Brain Behav
Immun 2010;24:708–23.
7. Ott L, McClain CJ, Gillespie M, et al. Cytokines and metabolic dysfunction
after severe head injury. J Neurotrauma 1994;11:447–72.
8. Ross SA, Halliday MI, Campbell GC, et al. The presence of tumour necrosis
factor in CSF and plasma after severe head injury. Br J Neurosurg 1994;8:
419–25.
9. Shohami E, Novikov M, Bass R, et al. Closed head injury triggers early
production of TNF alpha and IL-6 by brain tissue. J Cereb Blood Flow
Metab 1994;14:615–19.
10. Chou SH, Smith EE, Badjatia N, et al. A randomized, double-blind,
placebo-controlled pilot study of simvastatin in aneurysmal subarachnoid
hemorrhage. Stroke 2008;39:2891–3.
11. Lynch JR, Wang H, McGirt MJ, et al. Simvastatin reduces vasospasm after
aneurysmal subarachnoid hemorrhage: results of a pilot randomized clinical
trial. Stroke 2005;36:2024–6.
12. McGirt MJ, Blessing R, Alexander MJ, et al. Risk of cerebral vasospasm after
subarachnoid hemorrhage reduced by statin therapy: a multivariate analysis
of an institutional experience. J Neurosurg 2006;105:671–4.
13. Sillberg VA, Wells GA, Perry JJ. Do statins improve outcomes and reduce the
incidence of vasospasm after aneurysmal subarachnoid hemorrhage: a
meta-analysis. Stroke 2008;39:2622–6.
14. Tseng MY, Czosnyka M, Richards H, et al. Effects of acute treatment with
pravastatin on cerebral vasospasm, autoregulation, and delayed ischemic
deficits after aneurysmal subarachnoid hemorrhage: a phase II randomized
placebo-controlled trial. Stroke 2005;36:1627–32.
15. Chen SF, Hung TH, Chen CC, et al. Lovastatin improves histological and
functional outcomes and reduces inflammation after experimental traumatic
brain injury. Life Sci 2007;81:288–98.
16. Lu D, Goussev A, Chen J, et al. Atorvastatin reduces neurological deficit and
increases synaptogenesis, angiogenesis, and neuronal survival in rats
subjected to traumatic brain injury. J Neurotrauma 2004;21:21–32.
17. Lu D, Qu C, Goussev A, et al. Statins increase neurogenesis in the dentate
gyrus, reduce delayed neuronal death in the hippocampal CA3 region, and
improve spatial learning in rats after traumatic brain injury. J Neurotrauma
2007;24:1132–46.
18. Turkoglu OF, Eroglu H, Okutan O, et al. Atorvastatin efficiency after
traumatic brain injury in rats. Surg Neurol 2009;72:146–52.
19. Feuerstein G, Wang X, Barone FC. Cytokines in brain ischemia—the role of
TNF alpha. Cell Mol Neurobiol 1998;18:695–701.
20. Lucas SM, Rothwell NJ, Gibson RM. The role of inflammation in CNS injury
and disease. Br J Pharmacol 2006;147(Suppl 1):S232–40.
21. Mathiesen T, Edner G, Ulfarsson E, et al. Cerebrospinal fluid interleukin-1
receptor antagonist and tumor necrosis factor-alpha following subarachnoid
hemorrhage. J Neurosurg 1997;87:215–20.
22. Rodriguez-Yanez M, Castillo J. Role of inflammatory markers in brain
ischemia. Curr Opin Neurol 2008;21:353–7.
23. Stefini R, Catenacci E, Piva S, et al. Chemokine detection in the cerebral
tissue of patients with posttraumatic brain contusions. J Neurosurg 2008;
108:958–62.
 Section II—Clinical and Laboratory Assessment 153
24. Waje-Andreassen U, Krakenes J, Ulvestad E, et al. An early marker for
outcome in acute ischemic stroke. Acta Neurol Scand 2005;111:360–5.
25. Welsh P, Lowe GD, Chalmers J, et al. Associations of proinflammatory
cytokines with the risk of recurrent stroke. Stroke 2008;39:2226–30.
26. Borovikova LV, Ivanova S, Zhang M, et al. Vagus nerve stimulation attenuates
the systemic inflammatory response to endotoxin. Nature 2000;405:458–62.
27. Rothwell NJ, Luheshi GN. Interleukin 1 in the brain: biology, pathology and
therapeutic target. Trends Neurosci 2000;23:618–25.
28. Clausen F, Hanell A, Israelsson C, et al. Neutralization of interleukin-1beta
reduces cerebral edema and tissue loss and improves late cognitive outcome
following traumatic brain injury in mice. Eur J Neurosci 2011;34:110–23.
29. Chaichana KL, Pradilla G, Huang J, et al. Role of inflammation (leukocyte-
endothelial cell interactions) in vasospasm after subarachnoid hemorrhage.
World Neurosurg 2010;73(1):22–41. Epub 2009 Aug 6.
30. Kataoka H, Aoki T. Molecular basis for the development of intracranial
aneurysm. Expert Rev Neurother 2010;10(2):173–17.
31. Tulamo R, Frösen J, Hernesniemi J, et al. Inflammatory changes in the
aneurysm wall: a review. J Neurointerv Surg 2010;2(2):120–30. Epub 2010
Mar 12.
32. Aoki T, Nishimura M. Targeting chronic inflammation in cerebral
aneurysms: focusing on NF-kappaB as a putative target of medical therapy.
Expert Opin Ther Targets 2010;14(3):265–73.
33. Norton WT, Aquino DA, Hozumi I, et al. Quantitative aspects of reactive
gliosis: a review. Neurochem Res 1992;17:877–85.
34. O’Callaghan JP. Quantitative features of reactive gliosis following toxicant-
induced damage of the CNS. Ann N Y Acad Sci 1993;679:195–210.
35. Perry VH, Gordon S. Macrophages and the nervous system. Int Rev Cytol
1991;125:203–44.
36. Town T, Nikolic V, Tan J. The microglial “activation” continuum: from innate
to adaptive responses. J Neuroinflammation 2005;2:24.
37. Loane DJ, Byrnes KR. Role of microglia in neurotrauma. Neurotherapeutics
2010;7(4):366–77.
38. Engelhardt B, Ransohoff RM. The ins and outs of T-lymphocyte trafficking
to the CNS: anatomical sites and molecular mechanisms. Trends Immunol
2005;26:485–95.
39. Man S, Ubogu EE, Ransohoff RM. Inflammatory cell migration into the
central nervous system: a few new twists on an old tale. Brain Pathol 2007;
17:243–50.
40. Pachter JS, de Vries HE, Fabry Z. The blood-brain barrier and its role in
immune privilege in the central nervous system. J Neuropathol Exp Neurol
2003;62:593–604.
41. Ransohoff RM, Kivisakk P, Kidd G. Three or more routes for leukocyte
migration into the central nervous system. Nat Rev Immunol 2003;3:569–81.
42. Simard JM, Geng Z, Woo SK, et al. Glibenclamide reduces inflammation,
vasogenic edema, and caspase-3 activation after subarachnoid hemorrhage.
J Cereb Blood Flow Metab 2009;29:317–30.
43. Ramlackhansingh AF, Brooks DJ, Greenwood RJ, et al. Inflammation after
trauma: microglial activation and traumatic brain injury. Ann Neurol
2011;70(3):374–83. Epub 2011 Jun 27.
44. De Santi L, Polimeni G, Cuzzocrea S, et al. Neuroinflammation and
neuroprotection: an update on (future) neurotrophin-related strategies in
multiple sclerosis treatment. Curr Med Chem 2011;18(12):1775–84.
45. Amor S, Puentes F, Baker D, van der Valk P. Inflammation in
neurodegenerative diseases. Immunology 2010;129(2):154–69.
46. Laskowitz DT, Vitek MP. Apolipoprotein E and neurological disease:
therapeutic potential and pharmacogenomic interactions.
Pharmacogenomics 2007;8:959–69.
47. Iannotti CA, Clark M, Horn KP, et al. A combination immunomodulatory
treatment promotes neuroprotection and locomotor recovery after contusion
SCI. Exp Neurol 2011;230(1):3–15. Epub 2010 Mar 23.
48. Thompson HJ. Fever: a concept analysis. J Adv Nurs 2005;51:484–92.
49. Busto R, Dietrich WD, Globus MY, et al. Small differences in intraischemic
brain temperature critically determine the extent of ischemic neuronal
injury. J Cereb Blood Flow Metab 1987;7:729–38.
50. Diringer MN, Reaven NL, Funk SE, et al. Elevated body temperature
independently contributes to increased length of stay in neurologic intensive
care unit patients. Crit Care Med 2004;32:1489–95.
A complete list of references for this chapter can be found online at
www.expertconsult.com.
 Section II—Clinical and Laboratory Assessment 153.e1
References
1. Mashaly HA, Provencio JJ. Inflammation as a link between brain injury and
heart damage: the model of subarachnoid hemorrhage. Cleve Clin J Med
2008;75(Suppl 2):S26–30.
2. Pradilla G, Chaichana KL, Hoang S, et al. Inflammation and cerebral
vasospasm after subarachnoid hemorrhage. Neurosurg Clin N Am 2010;
21(2):365–79.
3. Helmy A, De Simoni MG, Guilfoyle MR, et al. Cytokines and innate
inflammation in the pathogenesis of human traumatic brain injury. Prog
Neurobiol 2011;95(3):352–72. Epub 2011 Sep 16.
4. Brough D, Tyrrell PJ, Allan SM. Regulation of interleukin-1 in acute
brain injury. Trends Pharmacol Sci 2011;32(10):617–22. Epub 2011
Jul 23.
5. Ravizza T, Balosso S, Vezzani A. Inflammation and prevention of
epileptogenesis. Neurosci Lett 2011;497(3):223–30. Epub 2011 Feb 26.
6. Denes A, Thornton P, Rothwell NJ, et al. Inflammation and brain injury:
acute cerebral ischaemia, peripheral and central inflammation. Brain Behav
Immun 2010;24:708–23.
7. Ott L, McClain CJ, Gillespie M, et al. Cytokines and metabolic dysfunction
after severe head injury. J Neurotrauma 1994;11:447–72.
8. Ross SA, Halliday MI, Campbell GC, et al. The presence of tumour necrosis
factor in CSF and plasma after severe head injury. Br J Neurosurg 1994;8:
419–25.
9. Shohami E, Novikov M, Bass R, et al. Closed head injury triggers early
production of TNF alpha and IL-6 by brain tissue. J Cereb Blood Flow
Metab 1994;14:615–19.
10. Chou SH, Smith EE, Badjatia N, et al. A randomized, double-blind,
placebo-controlled pilot study of simvastatin in aneurysmal subarachnoid
hemorrhage. Stroke 2008;39:2891–3.
11. Lynch JR, Wang H, McGirt MJ, et al. Simvastatin reduces vasospasm after
aneurysmal subarachnoid hemorrhage: results of a pilot randomized
clinical trial. Stroke 2005;36:2024–6.
12. McGirt MJ, Blessing R, Alexander MJ, et al. Risk of cerebral vasospasm after
subarachnoid hemorrhage reduced by statin therapy: a multivariate analysis
of an institutional experience. J Neurosurg 2006;105:671–4.
13. Sillberg VA, Wells GA, Perry JJ. Do statins improve outcomes and reduce
the incidence of vasospasm after aneurysmal subarachnoid hemorrhage: a
meta-analysis. Stroke 2008;39:2622–6.
14. Tseng MY, Czosnyka M, Richards H, et al. Effects of acute treatment with
pravastatin on cerebral vasospasm, autoregulation, and delayed ischemic
deficits after aneurysmal subarachnoid hemorrhage: a phase II randomized
placebo-controlled trial. Stroke 2005;36:1627–32.
15. Chen SF, Hung TH, Chen CC, et al. Lovastatin improves histological and
functional outcomes and reduces inflammation after experimental
traumatic brain injury. Life Sci 2007;81:288–98.
16. Lu D, Goussev A, Chen J, et al. Atorvastatin reduces neurological deficit
and increases synaptogenesis, angiogenesis, and neuronal survival in rats
subjected to traumatic brain injury. J Neurotrauma 2004;21:21–32.
17. Lu D, Qu C, Goussev A, et al. Statins increase neurogenesis in the dentate
gyrus, reduce delayed neuronal death in the hippocampal CA3 region, and
improve spatial learning in rats after traumatic brain injury. J Neurotrauma
2007;24:1132–46.
18. Turkoglu OF, Eroglu H, Okutan O, et al. Atorvastatin efficiency after
traumatic brain injury in rats. Surg Neurol 2009;72:146–52.
19. Feuerstein G, Wang X, Barone FC. Cytokines in brain ischemia—the role of
TNF alpha. Cell Mol Neurobiol 1998;18:695–701.
20. Lucas SM, Rothwell NJ, Gibson RM. The role of inflammation in CNS
injury and disease. Br J Pharmacol 2006;147 Suppl 1:S232–40.
21. Mathiesen T, Edner G, Ulfarsson E, et al. Cerebrospinal fluid interleukin-1
receptor antagonist and tumor necrosis factor-alpha following
subarachnoid hemorrhage. J Neurosurg 1997;87:215–20.
22. Rodriguez-Yanez M, Castillo J. Role of inflammatory markers in brain
ischemia. Curr Opin Neurol 2008;21:353–7.
23. Stefini R, Catenacci E, Piva S, et al. Chemokine detection in the cerebral
tissue of patients with posttraumatic brain contusions. J Neurosurg
2008;108:958–62.
24. Waje-Andreassen U, Krakenes J, Ulvestad E, et al. An early marker for
outcome in acute ischemic stroke. Acta Neurol Scand 2005;111:360–5.
25. Welsh P, Lowe GD, Chalmers J, et al. Associations of proinflammatory
cytokines with the risk of recurrent stroke. Stroke 2008;39:2226–30.
26. Borovikova LV, Ivanova S, Zhang M, et al. Vagus nerve stimulation
attenuates the systemic inflammatory response to endotoxin. Nature 2000;
405:458–62.
27. Rothwell NJ, Luheshi GN. Interleukin 1 in the brain: biology, pathology
and therapeutic target. Trends Neurosci 2000;23:618–25.
28. Clausen F, Hanell A, Israelsson C, et al. Neutralization of interleukin-1beta
reduces cerebral edema and tissue loss and improves late cognitive outcome
following traumatic brain injury in mice. Eur J Neurosci 2011;34:110–23.
29. Chaichana KL, Pradilla G, Huang J, et al. Role of inflammation (leukocyte-
endothelial cell interactions) in vasospasm after subarachnoid hemorrhage.
World Neurosurg 2010;73(1):22–41. Epub 2009 Aug 6.
30. Kataoka H, Aoki T. Molecular basis for the development of intracranial
aneurysm. Expert Rev Neurother 2010;10(2):173–17.
31. Tulamo R, Frösen J, Hernesniemi J, et al. Inflammatory changes in the
aneurysm wall: a review. J Neurointerv Surg 2010;2(2):120–30. Epub 2010
Mar 12.
32. Aoki T, Nishimura M. Targeting chronic inflammation in cerebral
aneurysms: focusing on NF-kappaB as a putative target of medical therapy.
Expert Opin Ther Targets 2010;14(3):265–73.
33. Norton WT, Aquino DA, Hozumi I, et al. Quantitative aspects of reactive
gliosis: a review. Neurochem Res 1992;17:877–85.
34. O’Callaghan JP. Quantitative features of reactive gliosis following
toxicant-induced damage of the CNS. Ann N Y Acad Sci 1993;679:195–210.
35. Perry VH, Gordon S. Macrophages and the nervous system. Int Rev Cytol
1991;125:203–44.
36. Town T, Nikolic V, Tan J. The microglial “activation” continuum: from
innate to adaptive responses. J Neuroinflammation 2005;2:24.
37. Loane DJ, Byrnes KR. Role of microglia in neurotrauma. Neurotherapeutics
2010;7(4):366–77.
38. Engelhardt B, Ransohoff RM. The ins and outs of T-lymphocyte trafficking
to the CNS: anatomical sites and molecular mechanisms. Trends Immunol
2005;26:485–95.
39. Man S, Ubogu EE, Ransohoff RM. Inflammatory cell migration into the
central nervous system: a few new twists on an old tale. Brain Pathol
2007;17:243–50.
40. Pachter JS, de Vries HE, Fabry Z. The blood-brain barrier and its role in
immune privilege in the central nervous system. J Neuropathol Exp Neurol
2003;62:593–604.
41. Ransohoff RM, Kivisakk P, Kidd G. Three or more routes for leukocyte
migration into the central nervous system. Nat Rev Immunol 2003;3:
569–81.
42. Simard JM, Geng Z, Woo SK, et al. Glibenclamide reduces inflammation,
vasogenic edema, and caspase-3 activation after subarachnoid hemorrhage.
J Cereb Blood Flow Metab 2009;29:317–30.
43. Ramlackhansingh AF, Brooks DJ, Greenwood RJ, et al. Inflammation after
trauma: microglial activation and traumatic brain injury. Ann Neurol
2011;70(3):374–83. Epub 2011 Jun 27.
44. De Santi L, Polimeni G, Cuzzocrea S, et al. Neuroinflammation and
neuroprotection: an update on (future) neurotrophin-related strategies in
multiple sclerosis treatment. Curr Med Chem 2011;18(12):1775–84.
45. Amor S, Puentes F, Baker D, van der Valk P. Inflammation in
neurodegenerative diseases. Immunology 2010;129(2):154–69.
46. Laskowitz DT, Vitek MP. Apolipoprotein e and neurological disease:
therapeutic potential and pharmacogenomic interactions.
Pharmacogenomics 2007;8:959–69.
47. Iannotti CA, Clark M, Horn KP, et al. A combination immunomodulatory
treatment promotes neuroprotection and locomotor recovery after
contusion SCI. Exp Neurol 2011;230(1):3–15. Epub 2010 Mar 23.
48. Thompson HJ. Fever: a concept analysis. J Adv Nurs 2005;51:484–92.
49. Busto R, Dietrich WD, Globus MY, et al. Small differences in intraischemic
brain temperature critically determine the extent of ischemic neuronal
injury. J Cereb Blood Flow Metab 1987;7:729–38.
50. Diringer MN, Reaven NL, Funk SE, et al. Elevated body temperature
independently contributes to increased length of stay in neurologic
intensive care unit patients. Crit Care Med 2004;32:1489–95.
51. Hajat C, Hajat S, Sharma P. Effects of poststroke pyrexia on stroke
outcome: a meta-analysis of studies in patients. Stroke 2000;31:410–14.
52. Naidech AM, Bendok BR, Bernstein RA, et al. Fever burden and functional
recovery after subarachnoid hemorrhage. Neurosurgery 2008;63:212–17;
discussion 217–18.
53. Oliveira-Filho J, Ezzeddine MA, Segal AZ, et al. Fever in subarachnoid
hemorrhage: relationship to vasospasm and outcome. Neurology 2001;56:
1299–304.
54. Weir B, Disney L, Grace M, et al. Daily trends in white blood cell count and
temperature after subarachnoid hemorrhage from aneurysm. Neurosurgery
1989;25:161–5.
55. Chen H, Chopp M, Welch KM. Effect of mild hyperthermia on the
ischemic infarct volume after middle cerebral artery occlusion in the rat.
Neurology 1991;41:1133–5.
56. Dietrich WD. The importance of brain temperature in cerebral injury.
J Neurotrauma 1992;9 Suppl 2:S475–85.
153.e2 Section II—Clinical and Laboratory Assessment
57. Morimoto T, Ginsberg MD, Dietrich WD, et al. Hyperthermia enhances
spectrin breakdown in transient focal cerebral ischemia. Brain Res 1997;
746:43–51.
58. Whalen MJ, Carlos TM, Clark RS, et al. The relationship between brain
temperature and neutrophil accumulation after traumatic brain injury in
rats. Acta Neurochir Suppl 1997;70:260–1.
59. Matthews DS, Bullock RE, Matthews JN, et al. Temperature response to
severe head injury and the effect on body energy expenditure and cerebral
oxygen consumption. Arch Dis Child 1995;72:507–15.
60. Stocchetti N, Rossi S, Zanier ER, et al. Pyrexia in head-injured patients
admitted to intensive care. Intensive Care Med 2002;28:1555–62.
61. Stubbe HD, Greiner C, Van Aken H, et al. Cerebral vascular and metabolic
response to sustained systemic inflammation in ovine traumatic brain
injury. J Cereb Blood Flow Metab 2004;24:1400–8.
62. Thompson HJ, Tkacs NC, Saatman KE, et al. Hyperthermia following
traumatic brain injury: a critical evaluation. Neurobiol Dis 2003;12:163–73.
63. Allan SM, Rothwell NJ. Cytokines and acute neurodegeneration. Nat Rev
Neurosci 2001;2:734–44.
64. Huang WT, Tsai SM, Lin MT. Involvement of brain glutamate release in
pyrogenic fever. Neuropharmacology 2001;41:811–18.
65. Morganti-Kossmann MC, Rancan M, Otto VI, et al. Role of cerebral
inflammation after traumatic brain injury: a revisited concept. Shock
2001;16:165–77.
66. Adachi H, Fujisawa H, Maekawa T, et al. Changes in the extracellular
glutamate concentrations in the rat cortex following localized by
hyperthermia. Int J Hyperthermia 1995;11:587–99.
67. Chatzipanteli K, Alonso OF, Kraydieh S, et al. Importance of posttraumatic
hypothermia and hyperthermia on the inflammatory response after fluid
percussion brain injury: biochemical and immunocytochemical studies.
J Cereb Blood Flow Metab 2000;20:531–42.
68. Rovlias A, Kotsou S. The influence of hyperglycemia on neurological
outcome in patients with severe head injury. Neurosurgery 2000;46:335–42;
discussion 342–3.
69. Saper CB, Breder CD. The neurologic basis of fever. N Engl J Med
1994;330:1880–6.
70. Young B. Nutritional and metabolic management of the head-injured
patient. Neurotrauma 1996:345–63.
71. McGirt MJ, Mavropoulos JC, McGirt LY, et al. Leukocytosis as an
independent risk factor for cerebral vasospasm following aneurysmal
subarachnoid hemorrhage. J Neurosurg 2003;98:1222–6.
72. Shahlaie K, Keachie K, Hutchins IM, et al. Risk factors for posttraumatic
vasospasm. J Neurosurg 2011;115(3):602–11. Epub 2011 Jun 10.
73. Hanafy KA, Morgan Stuart R, Fernandez L, et al. Cerebral inflammatory
response and predictors of admission clinical grade after aneurysmal
subarachnoid hemorrhage. J Clin Neurosci 2010;17(1):22–5. Epub 2009
Dec 8.
74. Gupta AK, Al-Rawi PG, Hutchinson PJ, et al. Effect of hypothermia on
brain tissue oxygenation in patients with severe head injury. Br J Anaesth
2002;88:188–92.
75. Rossi S, Zanier ER, Mauri I, et al. Brain temperature, body core
temperature, and intracranial pressure in acute cerebral damage. J Neurol
Neurosurg Psychiatry 2001;71(4):448–54.
76. Rumana CS, Gopinath SP, Uzura M, et al. Brain temperature exceeds
systemic temperature in head-injured patients. Crit Care Med 1998;26:
562–7.
77. Soukup J, Zauner A, Doppenberg EM, et al. The importance of brain
temperature in patients after severe head injury: relationship to intracranial
pressure, cerebral perfusion pressure, cerebral blood flow, and outcome.
J Neurotrauma 2002;19:559–71.
78. Spiotta AM, Stiefel MF, Heuer GG, et al. Brain hyperthermia after
traumatic brain injury does not reduce brain oxygen. Neurosurgery
2008;62:864–72; discussion 872.
79. Stocchetti N, Protti A, Lattuada M, et al. Impact of pyrexia on
neurochemistry and cerebral oxygenation after acute brain injury. J Neurol
Neurosurg Psychiatry 2005;76(8):1135–9.
80. Oddo M, Frangos S, Milby A, et al. Induced normothermia attenuates
cerebral metabolic distress in patients with aneurysmal subarachnoid
hemorrhage and refractory fever. Stroke 2009;40(5):1913–6.
81. Dengler J, Schefold JC, Graetz D, et al. Point-of-care testing for
interleukin-6 in cerebro spinal fluid (CSF) after subarachnoid
haemorrhage. Med Sci Monit 2008;14(12):BR265–8.
82. Maiuri F, Gallicchio B, Donati P, et al. The blood leukocyte count and its
prognostic significance in subarachnoid hemorrhage. J Neurosurg Sci
1987;31:45–8.
83. Rovlias A, Kotsou S. The blood leukocyte count and its prognostic
significance in severe head injury. Surg Neurol 2001;55:190–6.
84. Muroi C, Mink S, Seule M, et al. Monitoring of the inflammatory response
after aneurysmal subarachnoid haemorrhage in the clinical setting: review
of literature and report of preliminary clinical experience. Acta Neurochir
Suppl 2011;110(Pt 1):191–6.
85. Kasius KM, Frijns CJ, Algra A, et al. Association of platelet and
leukocyte counts with delayed cerebral ischemia in aneurysmal
subarachnoid hemorrhage. Cerebrovasc Dis 2010;29(6):576–83. Epub
2010 Apr 8.
86. Provencio JJ, Fu X, Siu A, et al. CSF neutrophils are implicated in the
development of vasospasm in subarachnoid hemorrhage. Neurocrit Care
2010;12(2):244–51.
87. Rothoerl RD, Axmann C, Pina AL, et al. Possible role of the C-reactive
protein and white blood cell count in the pathogenesis of cerebral
vasospasm following aneurysmal subarachnoid hemorrhage. J Neurosurg
Anesthesiol 2006;18:68–72.
88. Muir KW, Weir CJ, Alwan W, et al. C-reactive protein and outcome after
ischemic stroke. Stroke 1999;30:981–5.
89. Smith CJ, Emsley HC, Gavin CM, et al. Peak plasma interleukin-6 and
other peripheral markers of inflammation in the first week of ischaemic
stroke correlate with brain infarct volume, stroke severity and long-term
outcome. BMC Neurol 2004;4:2.
90. Rothenburg LS, Herrmann N, Swardfager W, et al. The relationship
between inflammatory markers and post stroke cognitive impairment.
J Geriatr Psychiatry Neurol 2010;23(3):199–205. Epub 2010 Jul 2.
91. Blanco M, Castellanos M, Rodriguez-Yanez M, et al. High blood pressure
and inflammation are associated with poor prognosis in lacunar
infarctions. Cerebrovasc Dis 2006;22:123–9.
92. Castellanos M, Castillo J, Garcia MM, et al. Inflammation-mediated damage
in progressing lacunar infarctions: a potential therapeutic target. Stroke
2002;33:982–7.
93. Leira R, Rodriguez-Yanez M, Castellanos M, et al. Hyperthermia is a
surrogate marker of inflammation-mediated cause of brain damage in
acute ischaemic stroke. J Intern Med 2006;260:343–9.
94. Oto J, Suzue A, Inui D, et al. Plasma proinflammatory and anti-
inflammatory cytokine and catecholamine concentrations as predictors
of neurological outcome in acute stroke patients. J Anesth 2008;22:
207–12.
95. Vila N, Castillo J, Davalos A, et al. Proinflammatory cytokines and early
neurological worsening in ischemic stroke. Stroke 2000;31:2325–9.
96. Shiozaki T, Hayakata T, Tasaki O, et al. Cerebrospinal fluid concentrations
of anti-inflammatory mediators in early-phase severe traumatic brain
injury. Shock 2005;23:406–10.
97. Gruber A, Rossler K, Graninger W, et al. Ventricular cerebrospinal fluid and
serum concentrations of STNFR-I, IL-1RA, and IL-6 after aneurysmal
subarachnoid hemorrhage. J Neurosurg Anesthesiol 2000;12:297–306.
98. Kwon KY, Jeon BC. Cytokine levels in cerebrospinal fluid and delayed
ischemic deficits in patients with aneurysmal subarachnoid hemorrhage.
J Korean Med Sci 2001;16:774–80.
99. Takizawa T, Tada T, Kitazawa K, et al. Inflammatory cytokine cascade
released by leukocytes in cerebrospinal fluid after subarachnoid
hemorrhage. Neurol Res 2001;23:724–30.
100. Tam AK, Ilodigwe D, Mocco J, et al. Impact of systemic inflammatory
response syndrome on vasospasm, cerebral infarction, and outcome after
subarachnoid hemorrhage: exploratory analysis of CONSCIOUS-1
database. Neurocrit Care 2010;13(2):182–9.
101. Dhar R, Diringer MN. The burden of the systemic inflammatory response
predicts vasospasm and outcome after subarachnoid hemorrhage. Neurocrit
Care 2008;8(3):404–12.
102. Sarrafzadeh A, Schlenk F, Gericke C, et al. Relevance of cerebral
interleukin-6 after aneurysmal subarachnoid hemorrhage. Neurocrit Care
2010;13(3):339–46.
103. Zanier ER, Brandi G, Peri G, et al. Cerebrospinal fluid pentraxin 3 early
after subarachnoid hemorrhage is associated with vasospasm. Intensive
Care Med 2011;37(2):302–9. Epub 2010 Nov 12.
104. Suzuki Y, Osuka K, Noda A, et al. Nitric oxide metabolites in the cisternal
cerebral spinal fluid of patients with subarachnoid hemorrhage.
Neurosurgery 1997;41(4):807–11.
105. Rejdak K, Petzold A, Sharpe MA, et al. Serum and urine nitrate and nitrite
are not reliable indicators of intrathecal nitric oxide production in acute
brain injury. J Neurol Sci 2003;208:1–7.
106. Woszczyk A, Deinsberger W, Boker DK. Nitric oxide metabolites in cisternal
CSF correlate with cerebral vasospasm in patients with a subarachnoid
haemorrhage. Acta Neurochir (Wien) 2003;145(4):257–63.
107. Bell MJ, Kochanek PM, Doughty LA, et al. Interleukin-6 and interleukin-10
in cerebrospinal fluid after severe traumatic brain injury in children.
J Neurotrauma 1997;14:451–7.
 Section II—Clinical and Laboratory Assessment 153.e3
108. Hayakata T, Shiozaki T, Tasaki O, et al. Changes in CSF S-100B and
cytokine concentrations in early-phase severe traumatic brain injury. Shock
2004;22:102–7.
109. Kossmann T, Stahel PF, Lenzlinger PM, et al. Interleukin-8 released into the
cerebrospinal fluid after brain injury is associated with blood-brain barrier
dysfunction and nerve growth factor production. J Cereb Blood Flow
Metab 1997;17:280–9.
110. Perez-Barcena J, Ibáñez J, Brell M, et al. Lack of correlation among
intracerebral cytokines, intracranial pressure, and brain tissue oxygenation
in patients with traumatic brain injury and diffuse lesions. Crit Care Med
2011;39(3):533–40.
111. Csuka E, Morganti-Kossmann MC, Lenzlinger PM, et al. IL-10 levels in
cerebrospinal fluid and serum of patients with severe traumatic brain
injury: relationship to IL-6, TNF-alpha, TGF-beta1 and blood-brain barrier
function. J Neuroimmunol 1999;101:211–21.
112. Maier B, Schwerdtfeger K, Mautes A, et al. Differential release of
interleukines 6, 8, and 10 in cerebrospinal fluid and plasma after traumatic
brain injury. Shock 2001;15:421–6.
113. Singhal A, Baker AJ, Hare GM, et al. Association between cerebrospinal
fluid interleukin-6 concentrations and outcome after severe human
traumatic brain injury. J Neurotrauma 2002;19:929–37.
114. Ley EJ, Clond MA, Singer MB, et al. IL6 deficiency affects function after
traumatic brain injury. J Surg Res 2011;170(2):253–6. Epub 2011 Mar 29.
115. de Waal Malefyt R, Abrams J, Bennett B, et al. Interleukin 10(IL-10) inhibits
cytokine synthesis by human monocytes: an autoregulatory role of il-10
produced by monocytes. J Exp Med 1991;174:1209–20.
116. McCombe PA, Read SJ. Immune and inflammatory responses to stroke:
good or bad? Int J Stroke 2008;3:254–65.
117. Suzuki S, Tanaka K, Suzuki N. Ambivalent aspects of interleukin-6 in
cerebral ischemia: Inflammatory versus neurotrophic aspects. J Cereb Blood
Flow Metab 2009;29:464–79.
118. Viviani B, Bartesaghi S, Corsini E, et al. Cytokines role in neurodegenerative
events. Toxicol Lett 2004;149:85–9.
119. Shore PM, Thomas NJ, Clark RS, et al. Continuous versus intermittent
cerebrospinal fluid drainage after severe traumatic brain injury in children:
effect on biochemical markers. J Neurotrauma 2004;21:1113–22.
120. Duo J, Stenken JA. In vitro and in vivo affinity microdialysis sampling of
cytokines using heparin-immobilized microspheres. Anal Bioanal Chem
2011;399:783–93.
121. Hutchinson PJ, O’Connell MT, Rothwell NJ, et al. Inflammation in human
brain injury: intracerebral concentrations of IL-1alpha, IL-1beta, and their
endogenous inhibitor IL1-ra. J of Neurotrauma 2007;24:1545–57.
II
Chapter
17  
Infection
Barnett R. Nathan and John J. Stern
Introduction
Infections in the intensive care setting are common, both as
the reason for the intensive care unit (ICU) admission and
as a nosocomial complication of critical illness. Nosocomial
infections, now referred to as hospital acquired infections
(HAIs), are thought to complicate between one quarter and
one half of all ICU admissions. In the Extended Prevalence
of Infection in Intensive Care (EPIC II) study, a 1-day, pro-
spective, point prevalence study, 51% of 13,796 adult (older
than 18 years) patients in 1265 participating ICUs from 75
countries were considered infected; 9084 (71%) were receiv-
ing antibiotics.1 The ICU mortality rate of the infected
patients was twofold greater than that of noninfected
patients (25% vs. 11%). Although central nervous system
(CNS) infections are far less common as admission diagno-
ses or as complications of critical care, these infections,
including the hospital-acquired complications of neurosurgi-
cal intervention and the community-acquired infections of
meningitis, encephalitis, and brain abscess are the focus of
this chapter. Each of these infections is discussed separately
to provide insights on what tests are needed to make the
diagnosis. Infections such as catheter-associated bloodstream
infections (BSIs) and ventilator-associated pneumonia (VAP)
are reviewed in Chapter 6. Finally, the role of infection sur-
veillance, prevention, and control is reviewed.
Bacterial Infections
Bacterial Central Nervous System
Space-Occupying Lesions
Examples of space-occupying lesions of the CNS include
parenchymal abscesses, epidural abscesses, and empyemas.
The most common sources of brain abscesses are from hema-
tologic spread (most common lung), direct extension from a
parameningeal source (including the sinuses, middle ear, or
after dental procedures, head trauma, or surgery), suppurative
lung disease, or congenital heart disease (e.g., patent foramen
ovale or patent ductus arteriosus, i.e., right to left shunting).
Rare causes of brain abscess have been reported after tongue
piercing or endovascular occlusion of aneurysms using Gug-
lielmi detachable coils.2 Infective endocarditis occasionally
presents first with a brain abscess.
154
Clinical Presentation of Bacterial Central
Nervous System Space-Occupying Lesions
Parenchymal brain abscess may present with many symp-
toms. Although the classic triad of fever, headache, and focal
neurologic signs is helpful, this occurs in less than half of
patients, and the presentation may be more of a mass lesion
than an infection. Other manifestations, including symptoms
and signs of the original infection (otitis or sinusitis), may
be present and more impressive. Although the average course
of brain abscesses from the time of symptom presentation to
hospital admission can be as short as 5 days,3 the course may
be more indolent. Subdural empyema also may present with
the triad of sinusitis, fever, and neurologic deficit. However,
signs of cortical irritation such as seizures (50%), raised
intracranial pressure (headache, vomiting, and papilledema
[50%]), or focal deficits (75%) are a frequent presentation.
Spinal epidural abscess is a neurosurgical emergency, and the
typical presenting features include back pain, malaise, and
fever. Neurologic deficits associated with spinal epidural
abscesses often are associated with venous thrombophlebitis
and so may evolve rapidly or be greater than the lesion size
suggests.
Epidemiology
The frequency of brain abscesses in the United States ranges
from 0.3 to 1.3 cases per 100,000 persons per year, with a
predominance occurring in men between the ages of 30 to 40.
Twenty five percent of all brain abscesses occur in children
between the ages of 4 to 7, resulting from either cyanotic heart
disease or extension from an otic source. The organism causing
the infection depends in part on the source of the infection
and the patient’s age. Common organisms in adults include
anaerobic organisms, in particular Bacteroides and Peptostrep-
tococcus in some series, whereas in others, aerobic organisms
especially Staphylococcus, Streptococcus (most common),
Enterobacteriaceae, and Haemophilus are more common. In
infants Proteus and Citrobacter are the most frequent cause,
whereas in children (3-5 years old) Haemophilus, Streptococcus
pneumoniae, and Bacteroides fragilis (otogenic spread) are
common. Staphylococcus, S. pneumoniae, and Haemophilus
influenzae are common from sinus spread and Streptococcus
or Staphylococcus aureus from cyanotic heart disease. Since
© Copyright 2013 Elsevier Inc. All rights reserved.
 Section II—Clinical and Laboratory Assessment 155
1981 the incidence of human immunodeficiency virus (HIV)–
related brain abscesses due to toxoplasmosis has become the
most common cause of a protozoal space-occupying brain
lesion and must be considered in any patient presenting with
a ring-enhancing brain lesion. These lesions may be single or
multiple and should lead to expeditious HIV serologic testing.
Aspergillus is the most common fungal organism to cause a
brain abscess.
Subdural empyemas (SDEs)—a collection of purulence
located between the arachnoid and dura—account for 15% to
20% of all space-occupying infections and result from exten-
sion of a suppurative process in a paranasal sinus or otorhi-
nologic infection. Less commonly, subdural empyemas result
from head trauma or after a neurosurgical procedure. Tewari
et al., found the majority of SDEs are identified in infancy
through the third decade of life (60%).4
Spinal epidural abscesses usually occur as the result of
hematogenous seeding of the epidural space from a distant
suppurative focus such as infective endocarditis, from an
infected central line catheter, or from the injection of intrave-
nous drugs. S. aureus, and less commonly aerobic and anaero-
bic streptococci, and gram-negative rods cause the vast
majority of epidural infections.
Diagnosis of Central Nervous System
Space-Occupying Lesions
Fever with or without a peripheral white count and symptoms
of new CNS deficit suggest an intracranial infection. Other
signs and symptoms such as headache, vomiting, and papill-
edema are not specific to these diagnoses. Cranial imaging with
computed tomography (CT) or magnetic resonance imaging
(MRI) are sensitive tests; MRI is able to provide more infor­
mation and better resolution.5 On MRI diffusion-weighted
imaging (DWI) an intraparenchymal abscess demonstrates
high signal on DWI and low apparent diffusion coefficient
(ADC) in the abscess cavity.6,7 In subdural empyema, there
may be a disproportionate amount of underlying cortical and
white matter edema and enhancement relative to the size of
the fluid collection. In patients with spinal epidural abscess
there may evidence of diskitis and vertebral body infection.
However, even imaging does not have the sensitivity required
to definitively diagnose all infections. Lumbar puncture should
not be performed in those with brain abscesses given the risk
of herniation.
Laboratory and Microbiologic Analysis
Laboratory Studies
The most useful blood test is the complete blood count with
a manual differential. Although an elevated white blood cell
(WBC) count is always helpful, patients presenting with a
significant suppurative process may have a normal peripheral
WBC count. In these patients bandemia may be the only
abnormal finding to guide the clinician toward making the
diagnosis of an infectious process. It is imperative that the
treating team request a manual differential on presentation.
Furthermore, a markedly elevated platelet count such as one
greater than 500,000 can also help suggest the presence of a
suppurative CNS process. An elevated erythrocyte sedimenta-
tion rate (ESR) and C-reactive protein (CRP) may be useful
but are nonspecific.
Microbiologic Analysis
Blood cultures should always be obtained on admission to
identify a potential hematogenous source for the CNS or epi-
dural infection. CNS-obtained material must be sent to a labo-
ratory for Gram stain, routine aerobic and anaerobic culture,
and in select patients for mycobacterial and fungal cultures.
In the immunocompromised patient, a silver stain should be
requested to identify Aspergillus species along with a request
for modified acid-fast bacillus (AFB) staining to potentially
identify a Nocardia species. In an HIV-infected patient patho-
logic specimens should be sent to a lab to look for Toxoplasma
gondii tachyzoites. In the event that routine cultures fail to
grow after 4 to 5 days, the clinician must consider whether he
or she is dealing with a routine anaerobic process that failed
to grow or a more unusual pathogen such as a Nocardia
species, fungal or mycobacterial infection, or T. gondii in an
HIV-infected patient. In these patients, consultation with an
infectious diseases expert and the microbiology laboratory is
advised.
Follow-up
The most effective approach to evaluate a patient’s response
to therapy is to follow the patient’s temperature and WBC
trend and the patient’s mental status and to obtain CNS
imaging on a regular basis to assess structural improvement.
Management of Space-Occupying
Infections
Treatment of brain abscess is both surgical and medical.
Abscesses greater than 2.5 cm in diameter or those associ-
ated with mass effect should be excised or aspirated. Aspira-
tion may be performed with image guidance and through a
twist drill or burr hole. Brain imaging should be used to
follow the treatment response, with repeat scans every 1 to 2
weeks.
In patients with ventriculitis, meningitis, or hydrocephalus
that requires cerebrospinal fluid (CSF) drainage, or those with
inaccessible abscesses or early abscess formation (cerebritis),
medical treatment alone can be attempted preferably after
microbiologic analysis on blood or CSF. Broad-spectrum
antibiotics are started and refined when culture results are
available. Empiric drug regimens for immunocompetent
patients should include coverage for methicillin-resistant
S. aureus (MRSA), anaerobes, and gram-negative bacilli
(including Pseudomonas coverage in the setting of trauma or
post neurosurgery) with a regimen such as vancomycin, met-
ronidazole, and cefotaxime, typically for 6 to 8 weeks. The
treatment response can be followed initially with weekly brain
imaging. This can be spaced out to every 2 weeks during the
remainder of the 6- to 8-week antibiotic course, with follow-up
scans every 2 to 4 months for the following year to assess for
recurrences.
The treatment of subdural empyema is surgical either
through a craniotomy or burr holes. Early treatment (within
72 hours of symptom onset or sooner) is preferable.8 Broad-
spectrum empiric antibiotics (similar to that described for
brain abscess) should be started early. Once the cause of the
infection is defined from culture analysis, antibiotics can be
tailored to the organism and continued for 3 to 4 weeks
after drainage. Spinal epidural abscess is a neurosurgical
156 Section II—Clinical and Laboratory Assessment
emergency. Without treatment, severe and permanent spinal
cord damage can occur in just a few hours in part because of
venous thrombosis rather than compression alone. Treatment
includes surgical decompression and drainage, a search for
the infectious source (if possible), and antibiotics (empiric
as described earlier, then specific antibiotics based on the
pathogens).
Outcome
The mortality of patients with brain abscess is between
20% and 30%9 and greater in those with a depressed level
of consciousness or in coma at admission and distant meta-
static focus of infection. Factors such as age, multiple abscesses,
and steroid use do not appear to alter outcomes. Mortality
in subdural empyema is between 10% and 40%. Younger
age and early treatment is associated with better outcome.4,8
For example, Renaudin and Frazee observed that patients
treated in less than 72 hours had less than 10% mortality,
whereas 70% of patients died or were disabled when treated
more than 72 hours after symptom onset.8 Outcome also
is associated with neurologic status at presentation and
whether sinus thrombophlebitis occurs and its extent.
Outcome after spinal epidural abscess is associated with
how quickly the diagnosis is made and surgical decompression
is achieved and the severity of the spinal cord symptoms
at the time of decompression. The more symptoms at the
time of decompression, the worse the outcome, although
compared with similar neurologic presentations in traumatic
spinal cord injury, those with epidural abscess tend to do
better.10
Bacterial Meningitis
Epidemiology
S. pneumoniae (or pneumococcus) and Neisseria meningitidis
remain the most common etiology for community-
acquired bacterial meningitis. With the widespread use of
H. influenzae–conjugated vaccine in children, the incidence of
H. influenzae meningitis has decreased.11 S. pneumoniae (or
pneumococcus) is now the most common cause of meningi-
tis in the United States, although its incidence too is decreas-
ing with the use of the multivalent pneumococcal vaccine.
Group B streptococcus is the most common cause of bacterial
meningitis in newborns but rare after the neonatal period,
N. meningitidis in children, teens, and young adults and
pneumococcus is the most common cause of bacterial men-
ingitis in adults.
Clinical Features
The typical presenting signs and symptoms of bacterial
meningitis are headache, fever, stiff neck, and altered mental
status that each occur in 80% of patients. Kernig’s sign and
Brudzinski’s sign are found in 50% of patients. In most
cases of bacterial meningitis there also is associated inflam-
mation of the underlying brain tissue (cerebritis). In addi-
tion, secondary venous sinus thrombosis, vasculitis, or
cranial nerve inflammation may occur. Therefore focal neu-
rologic deficits are seen in about 20% of patients and sei-
zures in about 30% of patients. Papilledema at presentation
is uncommon (<1%).
Diagnosis
The diagnosis of bacterial meningitis is made based on the
clinical presentation and CSF examination. At the time of a
lumbar puncture many patients have an increased opening
pressure (>15 cm H2O). CSF total WBC count is in the 1000
to 10,000 range, and almost all patients with bacterial menin-
gitis have a predominance of neutrophils. CSF glucose is
usually decreased, although the CSF-to-serum ratio is more
accurate. The CSF protein is almost always elevated (>100 g/
dL). C-reactive protein (CRP), either in the serum or the CSF,
or procalcitonin can help differentiate bacterial meningitis
from viral meningitis. The specific diagnosis is made on CSF
Gram stain and culture.
The Gram stain, developed by Hans Christian Gram in
Berlin in the late 19th century, colors bacteria by using a
crystal violet stain to colorize what are referred to as gram-
positive organisms blue, namely streptococci and staphylo-
cocci. In a subsequent step the specimen is then decolorized
with alcohol and counter-stained with the red dye, safranin.
Gram-negative bacteria such as coliforms are unable to hold
their crystal violet stain but take up the safranin stain, render-
ing these bacteria red. These procedures are now done exclu-
sively in a microbiology laboratory. This procedure can guide
antibiotic therapy until culture and sensitivity results are
available.
A head CT scan is typically not required and not needed to
diagnosis bacterial meningitis. However, it can be useful to
exclude mass lesions that are a contradiction to lumbar punc-
ture (LP). The literature supporting the necessity of doing a
CT scan of the brain prior to LP is minimal. Between 3% and
4% of patients with meningitis develop herniation syn-
dromes12,13; about 1% of these patients herniate within several
hours of the LP (1%). However, a normal CT scan does not
eliminate the risk for herniation nor does an abnormal CT
scan predict herniation after LP.14 In 2004 Tunkel et al.15 wrote
the “Practice Guidelines for the Management of Bacterial
Meningitis.” A head CT is recommended before LP in patients
with suspected bacterial meningitis who (a) have a history of
CNS disease, (b) are immunocompromised, (c) present with
a seizure, or have (d) papilledema, (e) an abnormal level of
consciousness, or (f) a focal neurologic deficit. Furthermore,
in the setting of suspected bacterial meningitis the decision to
begin antimicrobial therapy should be made expeditiously
and independent of when and if a head CT is performed
because earlier treatment is associated with better outcome.
Management
Treatment of bacterial meningitis should not be delayed while
waiting for a head CT scan. Empiric treatment of the patient
should be provided based on the patient’s relative risk for
specific organisms. Most immunocompetent adults are at
highest risk for S. pneumoniae. Those greater than 50 years old
are also at risk for Listeria monocytogenes. Teens and young
adults are also susceptible to N. meningitidis (meningococ-
cus). Empiric antibiotics should be administered based on
these likely causes. Beta-lactam–resistant pneumococcus is
becoming more prevalent in the community (up to 15% in
some metropolitan areas in the United States) and so vanco-
mycin should be used as an empiric treatment in the appropri-
ate patient population. Once the Gram stain is obtained
and organisms identified by morphology and staining
 Section II—Clinical and Laboratory Assessment 157
characteristics the antibiotics can be modified to be more
specific. The antibiotics can be changed once again if neces-
sary once the organism has been grown and identified in
culture and sensitivities have been obtained. Antibiotic treat-
ment durations are based on the organism found. Both
meningococcus and H. influenzae can be treated successfully
for 7 days. S. pneumoniae requires 2 weeks of treatment,
whereas Listeria and other gram-negative organisms require
at least 3 weeks of treatment. In addition, an infectious disease
consultant should guide antibiotic therapy.
The role of beta corticosteroids has been controversial
since the early 1990s. A randomized clinical trial of dexa-
methasone (10 mg of dexamethasone every 6 hours for 4
days) that included 301 patients with bacterial meningitis has
been published.16 The first dose of steroid was given before
antibiotic administration. Outcome was better in those who
received dexamethasone. This improvement in outcome was
almost exclusively in the patients who had S. pneumoniae
meningitis; however, due to a small sample size, conclusions
regarding the effectiveness of steroids in H. influenzae and
meningococcal meningitis are unclear. Because at the time of
presentation the offending organism is unknown, the dexa-
methasone can be started before antibiotics. If an organism
other than S. pneumoniae is found to be the offending patho-
gen, clinical guidelines for treatment of bacterial meningitis
recommend against continued routine use of corticoste-
roids.15 Good supportive intensive care medicine is important
in these patients. In bacterial meningitis increased intracra-
nial pressure does occur, yet it is unclear whether monitoring
of intracranial pressure or treatment with hypertonic agents
such as mannitol or hypertonic saline is useful in this patient
population.
Herpes Simplex Encephalitis
Epidemiology
Herpes simplex viruses (HSVs) are distributed worldwide, and
humans are the sole reservoir of this virus. Between 25% and
60% of humans are seropositive for herpes simplex virus by
adulthood. Herpes simplex encephalitis (HSE) is the most
common cause of sporadic fatal encephalitis in the United
States. It is estimated to occur in 1/250,000 to 1/500,000 in the
United States or approximately 250 to 500 cases per year.
Pathophysiology
It is unclear whether herpes simplex encephalitis represents a
primary or recurrent infection. The virus can lie dormant in
neurons and ganglia once the patient has been infected.
However, it has been demonstrated that patients with HSE
and concomitant cutaneous herpes simplex may have two
different strains of HSV. No specific triggers for HSE have
been identified. Immunosuppression does not predispose
patients to HSE. However, patients who are immunosup-
pressed are at risk for a more aggressive disease with a worse
outcome.
HSE is always a cortical infection, and no cases of HSE have
been reported to involve the brainstem. The temporal lobe is
the most commonly infected brain region, although other
lobes can develop encephalitis. The virus causes a hemor-
rhagic and necrotizing encephalitis.
Clinical Features
The common clinical manifestations of HSE are a change in
personality, seizures, and decreased level of consciousness.
Fever is common, and focal or generalized seizures occur in
two thirds of patients. Focal neurologic findings such aphasia
or neglect associated with a cortical abnormality or dysphagia
and hemiparesis associated with cortical dysfunction may be
seen. Headache, nausea, vomiting, and papilledema also are
common. The clinical presentation may be subtle and some-
what slow moving.
Diagnosis
The diagnosis of HSE is made based on the clinical course,
imaging (CT and MRI), and laboratory evaluation, in particu-
lar CSF analysis and CSF PCR. Electroencephalography and
brain biopsy may be helpful in some patients, but brain biop-
sies are now infrequently performed in this setting. The CSF
WBC count in patients with HSE is in the 10s to the 100s but
values of 1000 to 2000 cell/mm3 may be observed. Most of the
WBCs are lymphocytes. However, 10% to 25% of patients can
have high numbers of neutrophils in their CSF early in the
disease course. Red cells in the range of 10s to 1000s/mm3 are
seen in the CSF of about 50% of patients and xanthochromia
is common. CSF protein may be mild to moderately elevated
in approximately 50% of patients. Glucose is usually normal
but may be slightly reduced. Opening pressure is elevated in
one third of patients. The CSF PCR for herpes simplex is now
the test of choice for diagnosis and has a sensitivity of 96%
and specificity of 99%.17 Very early in the disease (<24 hours)
the test may be negative, and with treatment the sensitivity
may decrease.
On MRI imaging, temporal lobe abnormalities, often hem-
orrhagic, are frequently found. Hypodensities in the temporal
lobe or frontal lobes are found on CT in a severe case of herpes
encephalitis. On MRI there are hyperintensities on T2 and
gadolinium enhancement around the lesion in the lobe that is
infected. Electroencephalograms (EEGs) are sensitive (84%)
but nonspecific (32% specificity) for HSE. The characteristic
EEG demonstrates spike and slow activity with periodic later-
alized epileptiform discharge (PLED). Today brain biopsy
rarely is required.
Treatment
Acyclovir (10 mg/kg every 8 hours for 21 days assuming
normal renal function) is the drug of choice to treat HSE.18
Higher doses and a longer treatment duration may help
prevent relapse.18 Outcome for HSE has improved signifi-
cantly since the introduction of acyclovir, but mortality
remains 19% at 6 months and 28% at 18 months.19 Predictors
of poor outcome include age greater than 30 years and coma
on presentation. In these patients mortality is nearly 70%.19
Significant neurologic sequelae are often observed in those
that survive.
West Nile Viral Encephalitis
Characteristics of the West Nile Virus
The West Nile virus (WNV) is a single-stranded RNA virus
that belongs to the family Flaviviridae. This family of viruses
158 Section II—Clinical and Laboratory Assessment
includes the Japanese encephalitis virus, the St. Louis
encephalitis virus, and the Kunjin virus. There is serologic
cross-reactivity with all of these viruses, and so those vac-
cinated against one may be seropositive against another.
This virus was thought to have originated in the West Nile
valley of Uganda. Genetic linkage of the WNV epidemic
in the United States suggests that the viral serotype virus
comes from the Middle East. Clinically, only the United
States and Israeli WNV infections have caused severe human
disease.
Epidemiology
WNV infection is now considered epidemic in the United
States.20 Between 1999 and 2008 there were more than 31,000
cases and 1150 deaths (≈3.5% mortality). Of these cases nearly
12,000 (≈35%) were West Nile virus encephalitis (WNVE).21, 22
The incidence of WNVE increases in patients greater than 50
years old; there is a 10-times higher risk of meningitis or
encephalitis in those who are 50 to 59 years old and a 43-times
higher risk of encephalitis in patients older than 80 years of
age. There also is an increased risk of encephalitis in patients
who are immunocompromised secondary to malignancy,
transplant, or HIV.
Clinical Presentation
There is a clinical spectrum of West Nile viral infections that
ranges from mild, almost asymptomatic cases to the most
severe version, encephalitis. WNV infection without enceph-
alitis is typically asymptomatic or a “flulike illness” with
fever, malaise, myalgias, and headache. In those patients who
develop WNVE, fever, fatigue, nausea, vomiting, headache,
myalgias and altered mental status are common. Half of the
patients may develop objective weakness23 that appears to be
associated with anterior horn cell injury in the spinal cord
appearing similar to a polio presentation.24 These patients
also may develop compromised respiratory function or frank
respiratory failure. A subset of WNVE patients can present
with movement disorders associated with basal ganglia
involvement.25
Diagnosis
The diagnosis of WNV is made by examination of immuno-
globulin M (IgM) antibodies in the serum or CSF. Because
IgM does not cross the blood-brain barrier, the finding of
IgM antibodies specific for WNV in the CSF indicates an
acute infection with the virus. Vaccines to yellow fever and
Japanese encephalitis are cross-reactive with the antibodies
to WNV; however, this is IgG rather than IgM. The remain-
ing CSF analysis is nonspecific: up to 2000 white cells, pre-
dominantly lymphocytes, a mild to moderately elevated
protein, and normal glucose are found.23,26 Peripheral blood
analysis typically is nonspecific; the WBC count is normal
or slightly elevated, or lymphopenia may be observed.26
Patients who develop encephalitis may have hyponatremia
associated with the syndrome of inappropriate antidiuretic
hormone (SIADH) secretion. Imaging studies are nonspe-
cific. Head CTs often are normal. However, one third of
patients have enhancement of the meninges or periventricu-
lar areas on brain MRI. Abnormalities that involve the
cortex, brainstem, basal ganglia, or spinal cord also may be
seen on MRI.27,28
Treatment and Prognosis
The care of patients who develop WNVE can be very resource
intensive, particularly in those who develop motor weakness
and use up many ICU patient-days.29 There is currently no
proven treatment for WNV infection, although there are small
case series using ribavirin, interferon alpha-2b, hyperimmune
gammaglobulin,30,31 and a monoclonal antibody to the WNV.
The mortality in those patients who develop WNVE (usually
the elderly) is 10% to 18%29 and about half the survivors do
not return to their premorbid function. In those who develop
motor neuron damage up to two thirds have continued weak-
ness. One year after discharge, 55% of patients report that they
were not fully recovered; symptoms include fatigue, weakness,
gait difficulty, and memory problems.32
Infections After Neurosurgical
Procedures
Background
Infections after neurosurgical procedures, including those
associated with intracranial monitors or drains (e.g., external
ventricular drains or lumbar drains), or after cranial or spinal
surgical procedures are rare but when they occur contribute
to prolonged length of stay in the ICU or hospital and adversely
affect outcome.33
External Ventricular Drains
The external ventricular drain or ventriculostomy was devel-
oped in the early 1950s. During the 1960s Lundberg reported
positive CSF cultures in 6% of patients but no clinical infec-
tions in patients with ventriculostomies.34 Most reports that
describe infections associated with external ventricular drains
(EVDs) are retrospective: the incidence is between zero and
22%. Lozier et al.35 in a literature review of ventriculostomy-
related infections, reported a combined infection rate between
8% and 9%. However, the manner in which ventriculostomy-
associated infections are described varies greatly and not every
paper defines what an “infection” is. There are several possible
scenarios based on the ranked certainty of infectious ventricu-
litis: (1) ventriculostomy-related infection, (2) suspected
ventriculostomy-related infection, (3) ventriculitis, (4) ven-
triculostomy colonization, and (5) contamination (Table
17.1).35 According to this classification a positive CSF culture
from an EVD means colonization or contamination when
other CSF parameters are normal. When the CSF WBC count
and protein also are abnormal, then there is infection in the
face of a positive CSF culture. Infections associated with
lumbar drain infection are about 3%, although fewer studies
have addressed this question.36
Risk factors for infection associated with ventriculostomies
include intraventricular hemorrhage, subarachnoid hemor-
rhage, open depressed skull fracture, basilar cranial fracture
with CSF leak, neurosurgical operation, ventriculostomy irri-
gation, systemic infection, and longer duration of catheteriza-
tion.35 However, the role of insertion duration remains unclear
and it seems that if no infection has developed by 9 days, then
 Section II—Clinical and Laboratory Assessment 159

Table 17.1  Classification Scheme of Ventriculostomy-Related Infections

Suspected
Ventriculostomy- Ventriculostomy- Ventriculostomy
Parameters Ventriculitis Related Infection Related Infection Colonization Contamination
Cultures/GS Variable One or more Absence Multiple positive Isolated culture
culture or GS cultures or GS or GS
Glucose Low Declining Declining Normal/expected Normal/expected
Protein High Increasing Increasing Normal/expected Normal/expected
CSF WBC Pleocytosis Increasing Increasing Normal/expected Normal/expected
pleocytosis
Clinical Fever, meningeal signs, Fever None Sometimes fever None
photophobia,
altered mental status

Adapted from Lozier AP, Sciacca RR, Romagnoli MF, et al. Ventriculostomy-related infections: a critical review of the literature. Neurosurgery 2002;51(1):170–81;
discussion 181–2.
CSF, Cerebrospinal fluid; GS, Gram stain; WBC, white blood cell.

the risk may decline. Most studies suggest that gram-negative
bacilli and gram-positive cocci are the most common species
that infect ventricular drains.37,38
Intraparenchymal Pressure Monitors
The infection rate associated with intracranial pressure moni-
tors (“bolts”) is reported to be between zero and 4%, much
lower than that associated with ventriculostomies.39-42 In part
this may result from the limited, if any, manipulation of a bolt
once inserted, whereas ventriculostomies may be sampled and
drugs injected into them.
Craniotomy Infections
Infections including meningitis or ventriculitis complicate
about 4% of craniotomies.43,44 The bacteria associated with
these infections are typically skin flora such as Staphylococcus,
Streptococcus, and less frequently gram-negative bacilli. Risk
factors for a craniotomy infection include emergency surgery,
longer duration of surgery, reoperation during the same
admission, bleeding complications, CSF leak, or the surgeon
who performs the craniotomy.43
Diagnosis
The diagnosis of true ventriculostomy-associated infection
has the following characteristics: (1) progressive decline in
CSF glucose, (2) increasing CSF protein, (3) progressive CSF
pleocytosis, (4) one or more positive CSF cultures or Gram
stain, and (5) fever.36 Similar findings are evident when there
is ventriculitis, but the patient is far sicker. A reduced CSF-to-
serum glucose ratio can help in diagnosis. A ratio of 0.4 has a
sensitivity of 77% and a specificity of 87%.45 Other methods
including CSF lactate, cytokine analysis, and cell index
studies45-47 are useful but lack sensitivity and specificity.48 The
routine analysis of CSF to help predict infection does not
appear to be helpful and instead the frequent access to the
system may increase the risk of contamination or infec-
tion.35,48,49 Furthermore, in a prospective patient cohort in
which CSF was sampled every 3 days compared with a histori-
cal control group sampled more frequently, Williams et al.50
demonstrated a significant reduction in ventriculitis with less
frequent sampling.
Prevention
The data supporting the role of antibiotic prophylaxis or
changing the catheter to help prevent EVD associated infec-
tions are conflicting. In a randomized trial Poon et al. com-
pared continuous prophylactic ampicillin/sulbactam plus
aztreonam to a single dose at the time of the procedure.51
Continuous antibiotic prophylaxis was associated with fewer
infections but greater mortality if an infection occurred. Ret-
rospective data suggest that the risk of infection increases once
an EVD has been in place between days 5 and 935 but that
routine changes may not help prevent this.38,52 For example
Wong et al. in a randomized trial of 103 patients observed a
similar infection rate whether catheters were changed every
5 days or not.53 Some institutions use prophylactic vancomy-
cin or other antibiotics in this setting but data as to its efficacy
are lacking.
Treatment
When there is evidence of infection, treatment should begin
immediately. If cultures are available, the antibiotics should be
directed at the appropriate organisms. Without available cul-
tures or Gram stains, antibiotic choice should be directed at
gram-negative bacilli (including Pseudomonas) and both
Streptococcus and Staphylococcus (including S. aureus).
Surveillance and Evaluating
the Febrile Neurocritical
Care Unit Patient
Often a patient in the neurocritical care unit (NCCU) (e.g., a
postoperative neurosurgical patient), receives both steroids
and antipyretics, making an assessment of the patient’s poten-
tial for an infection challenging. Of paramount importance
is a careful daily examination of the patient, paying close
attention to level of consciousness, lung sounds, wounds,
skin, peripheral and central line insertion sites, presence of
160 Section II—Clinical and Laboratory Assessment
diarrhea, and urine clarity. Given the frequency of intubation
and ventilator support in these patients, the risk of a VAP is
relatively high, and careful evaluation of the integrity and
position of the endotracheal tube (ET) should be made daily,
along with continual vigilance regarding the color and consis-
tency of any sputum aspirated by the nursing staff. Should a
patient develop a fever (>38o C), a chest x-ray is indicated.
Modern ICU care of the intubated patient now includes
careful mouth care using chlorhexidine-containing mouth
products, inline suctioning, and elevation of the head of the
bed to 30 degrees, all in an effort to minimize VAPs (see
Chapter 6). In a patient destined for prolonged ventilator
support, data now support early tracheotomy placement.
In the presence of an unexplained fever, blood cultures
also are essential to evaluate for a potential central venous
catheter (CVC)–associated infection. Although CVCs often
are silver impregnated to help reduce the incidence of
catheter-related infections, these infections remain common
and a threat to an ICU patient. Blood cultures should always
be obtained in a febrile patient as soon as possible and pref-
erably during the febrile episode to increase the yield of cap-
turing the offending organism. Customary practice is to
draw aerobic and anaerobic cultures both peripherally and
through the central line. This method increases the yield,
and, when a blood culture is positive from the central line
source yet negative from the periphery, points toward a
central line source and early removal of the offending line.
Occasionally Candida bloodstream infections occur in an
ICU patient (6.9 per 1000 patients), in particular in those
whom have been on broad-spectrum antibiotics for pro-
longed periods, on steroids, or are receiving TPN.54 The
source is most often from a central venous catheter (CVC)
and responds to removal of the catheter followed by a 10-day
course of an appropriate antifungal agent such as fluconazole
or caspofungin. Antigen detection and PCR assays could
provide an alternative to microscopy, culture, and serology to
detect fungal infections, but further study and validation still
are required.
Other potential sources for fever are deep venous thrombo-
sis with or without pulmonary emboli that can present with
tachycardia, fever, hypotension, and a sepsis-like syndrome.
Careful examination of the patient and his or her extremities
can lead to a timely duplex ultrasound of the legs or arms if
indicated. Numerous medications used in the NCCU patient
can produce fever including phenytoin, carbamazepine, van-
comycin, beta-lactam antibiotics, and blood products. Looking
carefully for a new rash may assist in determining that the
fever is indeed a medication-related problem. A careful review
of the medication list is mandatory to evaluate a febrile post-
operative or NCCU patient. Serum procalcitonin, and use
of scores such as the Simplified Acute Physiology Score II
(SAPS II), Acute Physiology and Chronic Health Evaluation
(APACHE), or Sequential Organ Failure Assessment (SOFA)*
may differentiate between infectious and noninfectious fever
in the ICU.55
Clostridium difficile–induced colitis is an ever-expanding
complication of hospitalization and primarily associated with
antibiotic use but occasionally is associated with chemothera-
peutic agents.56 Recently identified strains of hypervirulent
C. difficile (PCR ribotype 027 or BI/NAP1/027) have entered
*Note: SOFA also refers to Sepsis-Related Organ Failure Assessment.
the hospital environment, producing toxic megacolon, and
occasional deaths are now being reported in increasing
numbers particularly in the elderly.57 C. difficile colitis presents
with a fever and diarrhea but with few other clinical signs and
symptoms. The diarrhea often has a unique odor. The diag-
nosis of C. difficile requires identification of C. difficile toxin
A or B in diarrheal stool. However, the accuracy of these diag-
nostic tests is limited and the optimal diagnostic testing algo-
rithm is still being elucidated; for example, it is difficult to
predict and prevent development of severe or relapsing C.
difficile infection in patients who initially present with mild
symptoms.58 The use of probiotics such as Lactobacillus plan-
tarum 299/299v plus fiber (LAB), to prevent and or help with
treatment of C. difficile is conflicting.59
Urinary tract infections (UTIs) and in particular catheter-
associated UTIs, remain a common source of HAIs and
account for 20% to 50% of all HAIs in the ICU.60 Longer
duration of Foley catheterization (>5 days) and poor urinary
catheter management, specifically disconnection of the closed
system, are significant factors associated with hospital-
acquired UTIs.61 In a patient with an indwelling Foley catheter
(FC), simple evaluation of the urine may be confusing. Fur-
thermore, nearly all patients with an FC in place for more than
4 weeks have pyuria and bacteriuria. In the febrile patient with
cloudy urine and a fever, it is incumbent on the team of physi-
cians to rule out other potential sources of fever before attrib-
uting the fever to the urine. Treatment of UTIs with antibiotics
is more likely to clear bacteriuria and relieve symptoms but
also selects for resistant uropathogens and commensal bacte-
ria and can adversely affect the gut and vaginal microbiota.
Uropathogens are increasingly becoming resistant to currently
available antibiotics and the case-fatality rate of hospital-
acquired UTIs remains high.62 Consequently, alternative strat-
egies including avoidance of inappropriate FC use, removal of
a Foley, intermittent catheterization, and condom catheters
are necessary to manage UTIs.63
Sepsis
Sepsis is associated with increased morbidity, mortality, and
costs of care64 and remains the leading cause of death in criti-
cally ill patients.65 The definition of sepsis is provided in Table
17.2. The initial stage of sepsis is referred to as systemic
inflammatory response syndrome (SIRS) and is defined as a
temperature greater than 38o C or less than 36o C, pulse more
than 90 beats per minute, respiratory rate greater than 20 per
minute, and a WBC more than 12,000 or less than 4000.
However, there are no specific laboratory studies at present to
help guide practitioners in defining an early septic patient
although serum procalcitonin levels may provide some insight
into severe infections.66,67 Instead careful attention to the
patient is the bedrock of early detection and treatment of an
infected patient, including a postoperative neurosurgical
patient in the NCCU. Table 17.3 lists common laboratory
findings in septic patients. In addition, PCR testing provides
increased sensitivity for bloodstream infections.68
The Surviving Sepsis Campaign (SSC) was launched in
2002 as a collaborative initiative between the European
Society of Intensive Care Medicine (ESICM), the Interna-
tional Sepsis Forum (ISF), and the Society of Critical
Care Medicine (SCCM) to address the poor outcome associ-
ated with sepsis. The SSC developed evidence-based
 Section II—Clinical and Laboratory Assessment 161

management guidelines that were published in 2004 and others—have been identified to help evaluate quality of
updated in 200869 with the goal of improving outcomes in sepsis care and improve delivery of evidence-based care in
sepsis and septic shock. Since the development of these the ICU.73 Finally, ICU patients may be followed for disease
guidelines, more than 20,000 patients have been entered into progression using a number of techniques, for example, the
the SSC database; analysis of this showed that participation SOFA system (Table 17.4) that originally was developed by
in the SSC was associated with continuous quality improve- the ESICM to describe organ failure in sepsis.74,75 The SOFA
ment in sepsis care and a 5.4% absolute survival benefit.70,71 is a six-organ dysfunction/failure score that measures multi-
Compliance with the overall sepsis bundle and its individual ple organ failure daily. Each organ is graded from 0 (normal)
elements is important. This can be facilitated by weekly feed- to 4 (the most abnormal), providing a daily score of 0 to 24
back sessions.72 In addition, a variety of potential measures points. Sequential assessment of organ dysfunction during
of quality of care for septic patients—vancomycin adminis- the first few days of ICU care is a good indicator of progno-
tration and its timing, steroid administration, blood culture sis. The mean and highest SOFA scores are associated with
collection, activated protein C administration, and broad- outcome. In addition an increase in SOFA score during the
spectrum antibiotic administration and its timing, among first 48 hours in the ICU predicts a mortality rate of at least
50%.74-77 SOFA use has been validated in the NCCU and
traumatic brain injury (TBI).78
Table 17.2  Consensus Definition of SIRS
and Sepsis by the American College
of Chest Physicians and the Society Infection Prevention
of Critical Care Medicine in the Intensive Care Unit
SIRS Two or more of the following criteria: HAIs are common (about 5% of all hospitalized patients), and
—Temperature <36° C or >38° C
—HR >90
in particular catheter-related bloodstream infection, VAP, sur-
—RR >20 or PaCO2 <32 gical site infection, and catheter-associated urinary tract infec-
—WBC >12,000 <4000, or >10% tions increase the cost of care and patient length of stay and
immature (band) forms adversely affect outcome.79-81 In patients with acute neurologic
Sepsis Documented infection together with 2 disorders (e.g., TBI or stroke) or those admitted to the ICU
or more SIRS criteria above the incidence of HAIs is greater—30%.82-84 For example, Wes-
Severe sepsis Sepsis associated with organ dysfunction tendorp et al. in a meta-analysis of 87 studies (8 of which were
restricted to patients admitted to the ICU) that included
Septic shock Sepsis with refractory hypotension or
hypoperfusion abnormalities in spite 137,817 patients after stroke observed that infection compli-
of adequate fluid resuscitation cated 30% of these patients.83 In ICUs today multiresistant
gram-negative bacteria are becoming more frequent as the
Modified from: Bone RC, Balk RA, Cerra FB, et al. Definitions for sepsis and
organ failure and guidelines for the use of innovative therapies in sepsis. The cause of HAIs.85 However, several lines of evidence demon-
ACCP/SCCM Consensus Conference Committee. American College of Chest strate that hand hygiene, isolation, antibiotic control mea-
Physicians/Society of Critical Care Medicine. Chest 1992;101(6):1644–55 sures, and compliance with protocols designed to prevent
HR, Heart rate; RR, respiratory rate; SIRS, Systemic inflammatory response
syndrome; WBC, white blood cell.
specific infections are associated with a reduction in HAIs
including central line–associated bloodstream infections and

Table 17.3  Common Laboratory Findings in Sepsis

Laboratory Test Findings Comments
White blood cell count Leukocytosis or leukopenia Endotoxemia may cause early leukopenia.
Platelet count Thrombocytosis or Early in the disease course a high platelet count may be
thrombocytopenia observed as an acute-phase response.
Low platelet counts found in DIC.
Coagulation Protein C deficiency Coagulation abnormalities may be identified before onset
Antithrombin deficiency of organ failure and without frank bleeding.
Elevated D-dimer level
Prolonged INR (PT) and PTT
Creatinine Increase from baseline Doubling indicates acute renal injury.
Lactic acid >4 mmol/L (36 mg/dL) Consistent with tissue hypoxia
Liver function and Elevated bilirubin, alkaline Hypoperfusion can contribute to acute hepatocellular injury.
enzymes phosphatase, AST, and ALT
Serum phosphate Hypophosphatemia Inversely correlated with proinflammatory cytokine levels
CRP Elevated Acute-phase response
Procalcitonin Elevated May help differentiate infectious from noninfectious SIRS

ALT, Alanine aminotransferase; AST, aspartate transaminase; CRP, C-reactive protein; DIC, disseminated intravascular coagulation; INR, international normalized ratio;
PT, prothrombin time; PTT, partial thromboplastin time; SIRS, systemic inflammatory response syndrome.
162 Section II—Clinical and Laboratory Assessment

Table 17.4  The SOFA (Sepsis-Related Organ Failure Assessment)* Score to Describe Organ
Dysfunction or Failure
Organ  0  1 2 3 4
Respiration †
>400 <400 <300 <200 <100
PaO2/FiO2 (mm Hg) SaO2/FiO2 221-301 142–220 67-141 <67
Coagulation >150 <150 <100 <50 <20
platelets 103/mm3
Liver <1.2 1.2-1.9 2-5.9 6-11.9 >12
bilirubin (mg/dL)
Cardiovascular‡ No hypotension MAP <70 Dopamine ≤5 or Dopamine >5 or Dopamine >15 or
hypotension dobutamine (any) norepinephrine norepinephrine
≤0.1 >0.1
CNS 15 13-14 10-12 6-9 <6
Glasgow Coma score
Renal <1.2 1.2-1.9 2-3.4 3.5-4.9 or <500 >5 or <200
creatinine (mg/dL)
or urine output
(mL/day)

Modified from Vincent JL, Moreno R, Takala J, et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of
the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med 1996;22(7):707–10.
*The SOFA is a six-organ dysfunction/failure score measuring multiple organ failure daily. Each organ is graded from 0 (normal) to 4 (the most abnormal), providing
a daily score of 0 to 24 points.
†
PaO2/FiO2 ratio was used preferentially. If not available, the SaO2/FiO2 ratio was used.
‡
Vasoactive medications administered for at least 1 hr (dopamine and norepinephrine µg/kg/min).
CNS, Central nervous system; FiO2, inspired oxygen fraction; MAP, mean arterial pressure; PaO2, arterial oxygen tension; SaO2, arterial oxygen saturation.

VAP86-92 (see also Chapter 6). The most important aspect of
infection prevention in the ICU including of MRSA and
vancomycin-resistant enterococcus (VRE) is compliance with
handwashing.93,94 This can be monitored using World Health
Organization hand hygiene observation methods. For routine
care alcohol-based gels are easy to use and extremely effective.
However, for patients with C. difficile infections, alcohol gels
have no activity against the C. difficile spore and careful hand-
washing with soap is required. Hands should be washed exten-
sively with soap and water for at least 15 seconds, but preferably
30 seconds. A strong environmental cleaning program also is
needed to help reduce the risk of HAIs.95 For example, a previ-
ous room occupant with C. difficile is an independent signifi-
cant risk factor for C. difficile acquisition in the ICU.96 Patient
visitation does not appear to increase the risk for infection in
the ICU,97 whereas conversion to single rooms in the ICU can
reduce the rate at which patients develop infections while in
the ICU.98
Appropriate isolation of infected patients also is required to
reduce the colonization of health care workers and the spread
of these organisms to other patients. All health care workers
must wear barriers, including gown and gloves, when they
enter the room of a patient colonized or infected with MRSA,
VRE, or C. difficile. These items must be removed before
exiting the room and not reused unless washed in the hospital
laundry. Currently all patients who enter the hospital environ-
ment transferred from another health care institution includ-
ing a rehabilitation facility are screened for nasal carriage for
MRSA and isolated if necessary. This may be important
because 11% of patients who enter the ICU have a nares
sample culture result positive for MRSA.99 This form of active
surveillance remains controversial but has become standard
across the United States and other industrialized nations in an
effort to reduce the spread of MRSA within hospitals.100
Whether universal screening rather than targeted screening
and expanded use of barrier precautions is cost effective is still
being elucidated.101-105 However, some form of screening and
use of general quality measures is better than no screening to
decrease the incidence of endemic MRSA in ICUs; same-day
PCR testing for high-risk patients may provide the greatest
cost savings.106-108
There are no data or recommendations on the screening of
health care workers for the presence of MRSA nasal carriage.
In the event that a particular surgeon or surgical team experi-
ences a higher than anticipated number of epidemiologically
related postoperative infections, the hospital infection preven-
tion team should get involved to identify the problem.109 These
evaluations often include culturing the team’s nares, axillae,
and groin for MRSA; carefully evaluating the perioperative
procedures in place including the timing and appropriateness
of the preoperative antibiotic, and identifying any breach in
sterile technique within the operative suite. In the event a
surgical team member is found to be carrying MRSA, a variety
of antimicrobial agents are available to eliminate carriage.
These same principles may be used to examine clustering of
infections in the ICU.110
Health Care Worker–Acquired
Infections
Human Immunodeficiency Virus
The prevalence of HIV in the United States is approximately
1 million infected individuals with 20% currently not aware
of their status and not in care. The overall rate of HIV trans-
mission through percutaneous inoculation (i.e., a needle or
 Section II—Clinical and Laboratory Assessment 163
other instrument that pierces the skin) is 0.3%. In the event
of a sharps or significant splash injury in the ICU, it is impera-
tive that the health care worker seek immediate attention at
the hospital’s employee health department or emergency
department. Early postexposure prophylaxis of an HIV-
infected body fluid with highly active antiretroviral therapy is
both effective and well tolerated.111 Guidelines for prophylaxis
after occupational exposure are available from the Centers for
Disease Control and Prevention (CDC),91 the Department of
Health and Human Services (DHHS), and the World Health
Organization (WHO).112-114
Hepatitis B
Since wide spread mandatory use of the hepatitis B vaccine,
hepatitis B is far less common as an HAI. However, hepatitis
C remains a problem, with more than 4 million infected indi-
viduals in the United States alone. The risk for hepatitis C
transmission following a needlestick with contaminated blood
is approximately 3%, but there are currently no effective pro-
phylactic agents for these injuries or exposure. Care, avoiding
fatigue, and patience in the operating room or ICU are the
most effective methods to avoid a sharps injury.
Conclusion
Infection is a common reason for admission to the ICU in
general and in the NCCU HAIs such as catheter-related blood-
stream infections, UTIs, VAP, surgical site infections (includ-
ing those associated with a ventriculostomy), and C. difficile
may complicate a third of all admissions. Several factors such
as poor nutrition and hyperglycemia may increase the risk of
an HAI, which when it develops has significant effects on the
cost of care and may adversely affect patient outcome. Strate-
gies such as hand hygiene, environmental cleaning, appropri-
ate hospital staffing, and education and use of evidence-based
guidelines among others can reduce NCCU infection rates.
Infection-control committees can assist in implementing
these policies, the success of which requires careful monitor-
ing of the individual patient and the ICU as a whole.115
References
1. Vincent JL, Rello J, Marshall J, et al. International study of the prevalence
and outcomes of infection in intensive care units. JAMA 2009;302(21):
2323–9.
2. Kirollos RW, Bosma JJ, Radhakrishnan J, et al. Endovascularly treated
cerebral aneurysm using Guglielmi detachable coils acting as a nidus for
brain abscess formation secondary to Salmonella bacteremia: case report.
Neurosurgery 2002;51(1):234–7; discussion 237–8.
3. Cavusoglu H, Kaya RA, Turkmenoglu ON, et al. Brain abscess: analysis of
results in a series of 51 patients with a combined surgical and medical
approach during an 11-year period. Neurosurg Focus 2008;24(6):E9.
4. Tewari MK, Sharma RR, Shiv VK, et al. Spectrum of intracranial subdural
empyemas in a series of 45 patients: current surgical options and outcome.
Neurol India 2004;52(3):346–9.
5. Falcone S, Post MJ. Encephalitis, cerebritis, and brain abscess:
pathophysiology and imaging findings. Neuroimaging Clin N Am 2000;
10(2):333–53.
6. Ebisu T, Tanaka C, Umeda M, et al. Discrimination of brain abscess from
necrotic or cystic tumors by diffusion-weighted echo planar imaging. Magn
Reson Imaging 1996;14(9):1113–6.
7. Cartes-Zumelzu FW, Stavrou I, Castillo M, et al. Diffusion-weighted imaging
in the assessment of brain abscesses therapy. AJNR Am J Neuroradiol 2004;
25(8):1310–7.
8. Renaudin JW, Frazee J. Subdural empyema: importance of early diagnosis.
Neurosurgery 1980;7(5):477–9.
9. Qureshi HU, Habib AA, Siddiqui AA, et al. Predictors of mortality in brain
abscess. J Pak Med Assoc 2002;52(3):111–6.
10. Koo DW, Townson, AF, Dvorak MF, et al. Spinal epidural abscess: a 5-year
case-controlled review of neurologic outcomes after rehabilitation. Arch
Phys Med Rehabil 2009;90(3):512–16.
11. Schuchat A, Robinson K, Wenger JD, et al. Bacterial meningitis in the United
States in 1995. N Engl J Med 1997;337(14):970–6.
12. Rennick G, Shann F, de Campo J. Cerebral herniation during bacterial
meningitis in children [see comments]. BMJ 1993;306(6883):953–5.
13. Durand ML, Calderwood SB, Weber DJ, et al. Acute bacterial meningitis in
adults: a review of 493 episodes. N Engl J Med 1993;328:21–8.
14. Hasbun R, Abrahams J, Jekel J, et al. Computed tomography of the head
before lumbar puncture in adults with suspected meningitis. N Engl J Med
2001;345(24):1727–33.
15. Tunkel AR, Hartman BJ, Kaplan SL, et al. Practice guidelines for the
management of bacterial meningitis. Clin Infect Dis 2004;39(9):1267–84.
16. de Gans J, van de Beek D. Dexamethasone in adults with bacterial
meningitis. N Engl J Med 2002;347(20):1549–56.
17. Lakeman FD, Whitley RJ. Diagnosis of herpes simplex encephalitis:
application of polymerase chain reaction to cerebrospinal fluid from
brain-biopsied patients and correlation with disease. National Institute of
Allergy and Infectious Diseases Collaborative Antiviral Study Group [see
comments]. J Infect Dis 1995;171(4):857–63.
18. Tunkel AR, Glaser CA, Bloch KC, et al. The management of encephalitis:
clinical practice guidelines by the Infectious Diseases Society of America.
Clin Infect Dis 2008;47(3):303–27.
19. Whitley RJ, Alford CA, Hirsch MS, et al. Vidarabine versus acyclovir therapy
in herpes simplex encephalitis. N Engl J Med 1986;314(3):144–9.
20. Provisional surveillance summary of the West Nile virus epidemic: United
States, January-November 2002. MMWR Morb Mortal Wkly Rep 2002;
51(50):1129–33.
21. Division of Vector Borne Infectious Diseases. CDC West Nile virus
homepage. 2009 [cited; Available at www.cdc.gov/ncidod/dvbid/westnile/
index.htm.
22. Acute flaccid paralysis syndrome associated with West Nile virus
infection: Mississippi and Louisiana, July-August 2002. MMWR Morb
Mortal Wkly Rep 2002;51(37):825–8.
23. Jeha LE, Sila CA, Lederman RJ, et al. West Nile virus infection: a new acute
paralytic illness. Neurology 2003;61(1):55–9.
24. Leis AA, Stokic DS, Polk JL, et al. A poliomyelitis-like syndrome from West
Nile virus infection. N Engl J Med 2002;347(16):1279–80.
25. Robinson RL, Shahida S, Madan N, et al. Transient parkinsonism in West
Nile virus encephalitis. Am J Med 2003;115(3):252–3.
26. Guarner J, Shieh WJ, Hunter S, et al. Clinicopathologic study and laboratory
diagnosis of 23 cases with West Nile virus encephalomyelitis. Hum Pathol
2004;35(8):983–90.
27. Petropoulou KA, Gordon SM, Prayson RA, et al. West Nile virus
meningoencephalitis: MR imaging findings. AJNR Am J Neuroradiol
2005;26(8):1986–95.
28. Ali M, Safriel Y, Sohi J, et al. West Nile virus infection: MR imaging findings
in the nervous system. AJNR Am J Neuroradiol 2005;26(2):289–97.
29. Pepperell C, Rau N, Krajden S, et al. West Nile virus infection in 2002:
morbidity and mortality among patients admitted to hospital in
southcentral Ontario. CMAJ 2003;168(11):1399–405.
30. Makhoul B, Braun E, Herskovitz M, et al. Hyperimmune gammaglobulin for
the treatment of West Nile virus encephalitis. Isr Med Assoc J
2009;11(3):151–3.
31. Lewis M, Amsden JR. Successful treatment of West Nile virus infection after
approximately 3 weeks into the disease course. Pharmacotherapy
2007;27(3):455–8.
32. Gottfried K, Quinn R, Jones T. Clinical description and follow-up
investigation of human West Nile virus cases. South Med J 2005;98(6):
603–6.
33. Beer R, Pfausler B, Schmutzhard E. Infectious intracranial complications in
the neuro-ICU patient population. Curr Opin Crit Care 2010;Mar 18. Epub
ahead of print.
34. Lundberg N. Continuous recording and control of ventricular fluid pressure
in neurosurgical practice. Acta Psychiatr Scand Suppl 1960;36(149):1–193.
35. Lozier AP, Sciacca RR, Romagnoli MF, et al. Ventriculostomy-related
infections: a critical review of the literature. Neurosurgery 2002;51(1):170–
81; discussion 181–2.
36. Coplin WM, Avellino Am, Kim DK, et al. Bacterial meningitis associated
with lumbar drains: a retrospective cohort study. J Neurol Neurosurg
Psychiatry 1999;67(4):468–73.
37. Sundbarg G, Nordstrom CH, Soderstrom S. Complications due to prolonged
ventricular fluid pressure recording. Br J Neurosurg 1988;2(4):485–95.
164 Section II—Clinical and Laboratory Assessment
38. Arabi Y, Memish ZA, Balkhy HH, et al. Ventriculostomy-associated
infections: incidence and risk factors. Am J Infect Control
2005;33(3):137–43.
39. Flibotte JJ, Lee KE, Koroshetz WJ, et al. Continuous antibiotic prophylaxis
and cerebral spinal fluid infection in patients with intracranial pressure
monitors. Neurocrit Care 2004;1(1):61–8.
40. Rebuck JA, Murry KR, Rhoney DH, et al. Infection related to intracranial
pressure monitors in adults: analysis of risk factors and antibiotic
prophylaxis. J Neurol Neurosurg Psychiatry 2000;69(3):381–4.
41. Martinez-Manas RM, Santamarta D, de Campos JM, et al. Camino
intracranial pressure monitor: prospective study of accuracy and
complications. J Neurol Neurosurg Psychiatry 2000;69(1):82–6.
42. Yablon JS, Lantner HJ, McCormack TM, et al. Clinical experience with a
fiberoptic intracranial pressure monitor. J Clin Monit 1993;9(3):171–5.
43. Blomstedt GC. Infections in neurosurgery: a retrospective study of 1143
patients and 1517 operations. Acta Neurochir (Wien) 1985;78(3-4):81–90.
44. Korinek AM, Golmard JL, Elcheick A, et al. Risk factors for neurosurgical
site infections after craniotomy: a critical reappraisal of antibiotic
prophylaxis on 4,578 patients. Br J Neurosurg 2005;19(2):155–62.
45. Leib SL, Boscacci R, Gratzl O, et al. Predictive value of cerebrospinal fluid
(CSF) lactate level versus CSF/blood glucose ratio for the diagnosis of
bacterial meningitis following neurosurgery. Clin Infect Dis
1999;29(1):69–74.
46. Pfausler B, Beer R, Engelhardt K, et al. Cell index: a new parameter for the
early diagnosis of ventriculostomy (external ventricular drainage)-related
ventriculitis in patients with intraventricular hemorrhage? Acta Neurochir
(Wien) 2004;146(5):477–81.
47. Lopez-Cortes LF, Marquez-Arbizu R, Jimenez-Jimenez LM, et al.
Cerebrospinal fluid tumor necrosis factor-alpha, interleukin-1beta,
interleukin-6, and interleukin-8 as diagnostic markers of cerebrospinal fluid
infection in neurosurgical patients [see comments]. Crit Care Med
2000;28(1):215–9.
48. Schade RP, Schinkel J, Roelandse FW, et al. Lack of value of routine analysis
of cerebrospinal fluid for prediction and diagnosis of external drainage-
related bacterial meningitis. J Neurosurg 2006;104(1):101–8.
49. Hader WJ, Steinbok P. The value of routine cultures of the cerebrospinal
fluid in patients with external ventricular drains. Neurosurgery
2000;46(5):1149–53; discussion 1153–5.
50. Williams TA, Leslie GD, Dobb GJ, et al. Decrease in proven ventriculitis by
reducing the frequency of cerebrospinal fluid sampling from extraventricular
drains. J Neurosurg 2011;115(5):1040–6. Epub 2011, Jul 29.
A complete list of references for this chapter can be found online at
www.expertconsult.com.
 Section II—Clinical and Laboratory Assessment 164.e1
References
1. Vincent JL, Rello J, Marshall J, et al. International study of the prevalence
and outcomes of infection in intensive care units. JAMA 2009;302(21):
2323–9.
2. Kirollos RW, Bosma JJ, Radhakrishnan J, et al. Endovascularly treated
cerebral aneurysm using Guglielmi detachable coils acting as a nidus for
brain abscess formation secondary to Salmonella bacteremia: case report.
Neurosurgery 2002;51(1):234–7; discussion 237–8.
3. Cavusoglu H, Kaya RA, Turkmenoglu ON, et al. Brain abscess: analysis of
results in a series of 51 patients with a combined surgical and medical
approach during an 11-year period. Neurosurg Focus 2008;24(6):E9.
4. Tewari MK, Sharma RR, Shiv VK, et al. Spectrum of intracranial subdural
empyemas in a series of 45 patients: current surgical options and outcome.
Neurol India 2004;52(3):346–9.
5. Falcone S, Post MJ. Encephalitis, cerebritis, and brain abscess:
pathophysiology and imaging findings. Neuroimaging Clin N Am 2000;
10(2):333–53.
6. Ebisu T, Tanaka C, Umeda M, et al. Discrimination of brain abscess from
necrotic or cystic tumors by diffusion-weighted echo planar imaging. Magn
Reson Imaging 1996;14(9):1113–6.
7. Cartes-Zumelzu FW, Stavrou I, Castillo M, et al. Diffusion-weighted
imaging in the assessment of brain abscesses therapy. AJNR Am J
Neuroradiol 2004;25(8):1310–7.
8. Renaudin JW, Frazee J. Subdural empyema: importance of early diagnosis.
Neurosurgery 1980;7(5):477–9.
9. Qureshi HU, Habib AA, Siddiqui AA, et al. Predictors of mortality in brain
abscess. J Pak Med Assoc 2002;52(3):111–6.
10. Koo DW, Townson, AF, Dvorak MF, et al. Spinal epidural abscess: a 5-year
case-controlled review of neurologic outcomes after rehabilitation. Arch
Phys Med Rehabil 2009;90(3):512–16.
11. Schuchat A, Robinson K, Wenger JD, et al. Bacterial meningitis in the
United States in 1995. N Engl J Med 1997;337(14):970–6.
12. Rennick G, Shann F, de Campo J. Cerebral herniation during bacterial
meningitis in children [see comments]. BMJ 1993;306(6883):953–5.
13. Durand ML, Calderwood SB, Weber DJ, et al. Acute bacterial meningitis in
adults: a review of 493 episodes. N Engl J Med 1993;328:21–8.
14. Hasbun R, Abrahams J, Jekel J, et al. Computed tomography of the head
before lumbar puncture in adults with suspected meningitis. N Engl J Med
2001;345(24):1727–33.
15. Tunkel AR, Hartman BJ, Kaplan SL, et al. Practice guidelines for the
management of bacterial meningitis. Clin Infect Dis 2004;39(9):1267–84.
16. de Gans J, van de Beek D. Dexamethasone in adults with bacterial
meningitis. N Engl J Med 2002;347(20):1549–56.
17. Lakeman FD, Whitley RJ. Diagnosis of herpes simplex encephalitis:
application of polymerase chain reaction to cerebrospinal fluid from
brain-biopsied patients and correlation with disease. National Institute of
Allergy and Infectious Diseases Collaborative Antiviral Study Group [see
comments]. J Infect Dis 1995;171(4):857–63.
18. Tunkel AR, Glaser CA, Bloch KC, et al. The management of encephalitis:
clinical practice guidelines by the Infectious Diseases Society of America.
Clin Infect Dis 2008;47(3):303–27.
19. Whitley RJ, Alford CA, Hirsch MS, et al. Vidarabine versus acyclovir
therapy in herpes simplex encephalitis. N Engl J Med 1986;314(3):
144–9.
20. Provisional surveillance summary of the West Nile virus epidemic: United
States, January-November 2002. MMWR Morb Mortal Wkly Rep 2002;
51(50):1129–33.
21. Division of Vector Borne Infectious Diseases. CDC West Nile virus
homepage. 2009 [cited; Available at www.cdc.gov/ncidod/dvbid/westnile/
index.htm.
22. Acute flaccid paralysis syndrome associated with West Nile virus
infection: Mississippi and Louisiana, July-August 2002. MMWR Morb
Mortal Wkly Rep 2002;51(37):825–8.
23. Jeha LE, Sila CA, Lederman RJ, et al. West Nile virus infection: a new acute
paralytic illness. Neurology 2003;61(1):55–9.
24. Leis AA, Stokic DS, Polk JL, et al. A poliomyelitis-like syndrome from West
Nile virus infection. N Engl J Med 2002;347(16):1279–80.
25. Robinson RL, Shahida S, Madan N, et al. Transient parkinsonism in West
Nile virus encephalitis. Am J Med 2003;115(3):252–3.
26. Guarner J, Shieh WJ, Hunter S, et al. Clinicopathologic study and
laboratory diagnosis of 23 cases with West Nile virus encephalomyelitis.
Hum Pathol 2004;35(8):983–90.
27. Petropoulou KA, Gordon SM, Prayson RA, et al. West Nile virus
meningoencephalitis: MR imaging findings. AJNR Am J Neuroradiol
2005;26(8):1986–95.
28. Ali M, Safriel Y, Sohi J, et al. West Nile virus infection: MR imaging findings
in the nervous system. AJNR Am J Neuroradiol 2005;26(2):289–97.
29. Pepperell C, Rau N, Krajden S, et al. West Nile virus infection in 2002:
morbidity and mortality among patients admitted to hospital in
southcentral Ontario. CMAJ 2003;168(11):1399–405.
30. Makhoul B, Braun E, Herskovitz M, et al. Hyperimmune gammaglobulin
for the treatment of West Nile virus encephalitis. Isr Med Assoc J 2009;
11(3):151–3.
31. Lewis M, Amsden JR. Successful treatment of West Nile virus infection after
approximately 3 weeks into the disease course. Pharmacotherapy
2007;27(3):455–8.
32. Gottfried K, Quinn R, Jones T. Clinical description and follow-up
investigation of human West Nile virus cases. South Med J 2005;98(6):
603–6.
33. Beer R, Pfausler B, Schmutzhard E. Infectious intracranial complications in
the neuro-ICU patient population. Curr Opin Crit Care 2010;Mar 18. Epub
ahead of print.
34. Lundberg N. Continuous recording and control of ventricular fluid
pressure in neurosurgical practice. Acta Psychiatr Scand Suppl 1960;
36(149):1–193.
35. Lozier AP, Sciacca RR, Romagnoli MF, et al. Ventriculostomy-related
infections: a critical review of the literature. Neurosurgery 2002;51(1):170–
81; discussion 181–2.
36. Coplin WM, Avellino Am, Kim DK, et al. Bacterial meningitis associated
with lumbar drains: a retrospective cohort study. J Neurol Neurosurg
Psychiatry 1999;67(4):468–73.
37. Sundbarg G, Nordstrom CH, Soderstrom S. Complications due to
prolonged ventricular fluid pressure recording. Br J Neurosurg 1988;2(4):
485–95.
38. Arabi Y, Memish ZA, Balkhy HH, et al. Ventriculostomy-associated
infections: incidence and risk factors. Am J Infect Control 2005;33(3):
137–43.
39. Flibotte JJ, Lee KE, Koroshetz WJ, et al. Continuous antibiotic prophylaxis
and cerebral spinal fluid infection in patients with intracranial pressure
monitors. Neurocrit Care 2004;1(1):61–8.
40. Rebuck JA, Murry KR, Rhoney DH, et al. Infection related to intracranial
pressure monitors in adults: analysis of risk factors and antibiotic
prophylaxis. J Neurol Neurosurg Psychiatry 2000;69(3):381–4.
41. Martinez-Manas RM, Santamarta D, de Campos JM, et al. Camino
intracranial pressure monitor: prospective study of accuracy and
complications. J Neurol Neurosurg Psychiatry 2000;69(1):82–6.
42. Yablon JS, Lantner HJ, McCormack TM, et al. Clinical experience with a
fiberoptic intracranial pressure monitor. J Clin Monit 1993;9(3):171–5.
43. Blomstedt GC. Infections in neurosurgery: a retrospective study of 1143
patients and 1517 operations. Acta Neurochir (Wien) 1985;78(3-4):81–90.
44. Korinek AM, Golmard JL, Elcheick A, et al. Risk factors for neurosurgical
site infections after craniotomy: a critical reappraisal of antibiotic
prophylaxis on 4,578 patients. Br J Neurosurg 2005;19(2):155–62.
45. Leib SL, Boscacci R, Gratzl O, et al. Predictive value of cerebrospinal fluid
(CSF) lactate level versus CSF/blood glucose ratio for the diagnosis of
bacterial meningitis following neurosurgery. Clin Infect Dis 1999;29(1):
69–74.
46. Pfausler B, Beer R, Engelhardt K, et al. Cell index: a new parameter for the
early diagnosis of ventriculostomy (external ventricular drainage)-related
ventriculitis in patients with intraventricular hemorrhage? Acta Neurochir
(Wien) 2004;146(5):477–81.
47. Lopez-Cortes LF, Marquez-Arbizu R, Jimenez-Jimenez LM, et al.
Cerebrospinal fluid tumor necrosis factor-alpha, interleukin-1beta,
interleukin-6, and interleukin-8 as diagnostic markers of cerebrospinal fluid
infection in neurosurgical patients [see comments]. Crit Care Med
2000;28(1):215–9.
48. Schade RP, Schinkel J, Roelandse FW, et al. Lack of value of routine analysis
of cerebrospinal fluid for prediction and diagnosis of external drainage-
related bacterial meningitis. J Neurosurg 2006;104(1):101–8.
49. Hader WJ, Steinbok P. The value of routine cultures of the cerebrospinal
fluid in patients with external ventricular drains. Neurosurgery
2000;46(5):1149–53; discussion 1153–5.
50. Williams TA, Leslie GD, Dobb GJ, et al. Decrease in proven ventriculitis by
reducing the frequency of cerebrospinal fluid sampling from
extraventricular drains. J Neurosurg 2011;115(5):1040–6. Epub 2011, Jul 29.
51. Poon WS, Ng S, Wai S. CSF antibiotic prophylaxis for neurosurgical
patients with ventriculostomy: a randomised study. Acta Neurochir Suppl
1998;71:146–8.
52. Holloway KL, Barnes T, Choi S, et al. Ventriculostomy infections: the effect
of monitoring duration and catheter exchange in 584 patients. J Neurosurg
1996;85(3):419–24.
164.e2 Section II—Clinical and Laboratory Assessment
53. Wong GK, Poon WS, Wai S, et al. Failure of regular external ventricular
drain exchange to reduce cerebrospinal fluid infection: result of a
randomised controlled trial. J Neurol Neurosurg Psychiatry 2002;73(6):
759–61.
54. Kett DH, Azoulay E, Echeverria PM, et al. Extended Prevalence of Infection
in ICU Study (EPIC II) Group of Investigators. Candida bloodstream
infections in intensive care units: analysis of the extended prevalence of
infection in intensive care unit study. Crit Care Med 2011;39(4):665–70.
55. Jeon EJ, Jung JW, Choi JC, et al. The value of procalcitonin and the SAPS II
and APACHE III scores in the differentiation of infectious and non-
infectious fever in the ICU: a prospective, cohort study. J Korean Med Sci
2010;25(11):1633–7. Epub 2010, Oct 26.
56. Ananthakrishnan AN. Clostridium difficile infection: epidemiology, risk
factors and management. Nat Rev Gastroenterol Hepatol 2011;8(1):17–26.
Epub 2010, Nov 30.
57. Freeman J, Bauer MP, Baines SD, et al. The changing epidemiology of
Clostridium difficile infections. Clin Microbiol Rev 2010;23(3):529–49.
58. Bobo LD, Dubberke ER, Kollef M. Clostridium difficile in the ICU: the
struggle continues. Chest 2011;140(6):1643–53.
59. Oudhuis GJ, Bergmans DC, Dormans T, et al. Probiotics versus antibiotic
decontamination of the digestive tract: infection and mortality. Intensive
Care Med 2011;37(1):110–7. Epub 2010, Aug 19.
60. Chenoweth CE, Saint S. Urinary tract infections. Infect Dis Clin North Am
2011;25(1):103–15. Epub 2010, Dec 18.
61. Markovic-Denic L, Mijovic B, Jankovic S. Risk factors for hospital-acquired
urinary tract infection: a case-control study. Int Urol Nephrol
2011;43(2):303–8. Epub 2010, Sep 17.
62. Chang R, Greene MT, Chenoweth CE, et al. Epidemiology of hospital-
acquired urinary tract-related bloodstream infection at a university
hospital. Infect Control Hosp Epidemiol 2011;32(11):1127–9. Epub 2011,
Oct 6.
63. Knoll BM, Wright D, Ellingson L, et al. Reduction of inappropriate urinary
catheter use at a Veterans Affairs hospital through a multifaceted quality
improvement project. Clin Infect Dis 2011;52(11):1283–90.
64. Levinson AT, Casserly BP, Levy MM. Reducing mortality in severe sepsis
and septic shock. Semin Respir Crit Care Med 2011;32(2):195–205. Epub
2011, Apr 19.
65. Angus D, Linde-Zwirble W, Lidicker J, et al. Epidemiology of severe sepsis
in the United States: analysis of incidence, outcome, and associated costs of
care. Crit Care Med 2001;29:1303–10
66. Bloos F, Marshall JC, Dellinger RP, et al. Multinational, observational study
of procalcitonin in ICU patients with pneumonia requiring mechanical
ventilation: a multicenter observational study. Crit Care 2011;15(2):R88.
Epub 2011, Mar 7.
67. Giamarellos-Bourboulis EJ, Tsangaris I, Kanni T, et al. Procalcitonin as an
early indicator of outcome in sepsis: a prospective observational study.
J Hosp Infect 2011;77(1):58–63. Epub 2010, Dec 4.
68. Lehmann LE, Hunfeld KP, Steinbrucker M, et al. Improved detection of
blood stream pathogens by real-time PCR in severe sepsis. Intensive Care
Med 2010;36(1):49–56. Epub 2009, Sep 15.
69. Dellinger RP, Levy MM, Carlet JM, et al. Surviving Sepsis Campaign:
international guidelines for management of severe sepsis and septic shock:
2008. Crit Care Med 2008;36(1):296–327.
70. Marshall JC, Dellinger RP, Levy M. The Surviving Sepsis Campaign: a
history and a perspective. Surg Infect (Larchmt) 2010;11(3):275–81.
71. Levy MM, Dellinger RP, Townsend SR, et al. The Surviving Sepsis
Campaign: results of an international guideline-based performance
improvement program targeting severe sepsis. Intensive Care Med 2010;
36(2):222–31. Epub 2010, Jan 13.
72. Schramm GE, Kashyap R, Mullon JJ, et al. Septic shock: a multidisciplinary
response team and weekly feedback to clinicians improve the process of
care and mortality. Crit Care Med 2011;39(2):252–8.
73. Berenholtz SM, Pronovost PJ, Ngo K, et al. Developing quality measures
for sepsis care in the ICU. Jt Comm J Qual Patient Saf 2007;33(9):
559–68.
74. Vincent JL, Moreno R, Takala J, et al. The SOFA (Sepsis-related Organ
Failure Assessment) score to describe organ dysfunction/failure. On behalf
of the Working Group on Sepsis-Related Problems of the European Society
of Intensive Care Medicine. Intensive Care Med 1996;22(7):707–10.
75. Vincent JL, de Mendonça A, Cantraine F, et al. Use of the SOFA score to
assess the incidence of organ dysfunction/failure in intensive care units:
results of a multicenter, prospective study. Working group on “sepsis-related
problems” of the European Society of Intensive Care Medicine. Crit Care
Med 1998;26(11):1793–800.
76. Ferreira FL, Bota DP, Bross A, et al. Serial evaluation of the SOFA score to
predict outcome in critically ill patients. JAMA 2001;286(14):1754––8
77. Jones AE, Trzeciak S, Kline JA. The Sequential Organ Failure Assessment
score for predicting outcome in patients with severe sepsis and evidence of
hypoperfusion at the time of emergency department presentation. Crit
Care Med 2009;37(5):1649–54.
78. Zygun D, Berthiaume L, Laupland K, et al. SOFA is superior to MOD score
for the determination of non-neurologic organ dysfunction in patients with
severe traumatic brain injury: a cohort study. Crit Care 2006;10(4):R115.
79. Roberts RR, Scott RD 2nd, Hota B, et al. Costs attributable to healthcare-
acquired infection in hospitalized adults and a comparison of economic
methods. Med Care 2010;48(11):1026–35.
80. Hassan M, Tuckman HP, Patrick RH, et al. Cost of hospital-acquired
infection. Hosp Top 2010;88(3):82–9.
81. Waters HR, Korn R Jr, Colantuoni E, et al. The business case for quality:
economic analysis of the Michigan Keystone Patient Safety Program in
ICUs. Am J Med Qual 2011;26(5):333–9.
82. Kourbeti IS, Vakis AF, Papadakis JA, et al. Infections in traumatic brain
injury patients. Clin Microbiol Infect 2011;Jun 30. Epub ahead of print.
83. Westendorp WF, Nederkoorn PJ, Vermeij JD, et al. Post-stroke infection: a
systematic review and meta-analysis. BMC Neurol 2011;11:110.
84. Cairns S, Reilly J, Booth M. Prevalence of healthcare-associated infection in
Scottish intensive care units. J Hosp Infect 2010;76(4):308–10. Epub 2010,
Oct 23.
85. Doyle JS, Buising KL, Thursky KA, et al. Epidemiology of infections
acquired in intensive care units. Semin Respir Crit Care Med
2011;32(2):115–38. Epub 2011, Apr 19.
86. Ramirez P, Bassi GL, Torres A. Measures to prevent nosocomial infections
during mechanical ventilation. Curr Opin Crit Care 2012;18(1):86–92.
87. Bouadma L, Deslandes E, Lolom I, et al. Long-term impact of a
multifaceted prevention program on ventilator-associated pneumonia in a
medical intensive care unit. Clin Infect Dis 2010;51(10):1115–22. Epub
2010, Oct 11.
88. Sihler KC, Chenoweth C, Zalewski C, et al. Catheter-related vs. catheter-
associated blood stream infections in the intensive care unit: incidence,
microbiology, and implications. Surg Infect (Larchmt) 2010;11(6):529–34.
89. Hospenthal DR, Green AD, Crouch HK, et al. Infection prevention and
control in deployed military medical treatment facilities. J Trauma
2011;71(2 Suppl 2):S290–8.
90. Cherry-Bukowiec JR, Denchev K, Dickinson S, et al. Prevention of
catheter-related blood stream infection: back to basics? Surg Infect
(Larchmt) 2011;12(1):27–32. Epub 2010, Dec 20.
91. Centers for Disease Control and Prevention (CDC). Vital signs: central
line-associated blood stream infections—United States, 2001, 2008, and
2009. MMWR Morb Mortal Wkly Rep 2011;60(8):243–8.
92. Longmate AG, Ellis KS, Boyle L, et al. Elimination of central-venous-
catheter-related bloodstream infections from the intensive care unit. BMJ
Qual Saf 2011;20(2):174–80.
93. Lin MY, Hayden MK. Methicillin-resistant Staphylococcus aureus and
vancomycin-resistant enterococcus: recognition and prevention in intensive
care units. Crit Care Med 2010;38(8 Suppl):S335–44.
94. Monistrol O, Calbo E, Riera M, et al. Impact of a hand hygiene educational
programme on hospital-acquired infections in medical wards. Clin
Microbiol Infect 2011;Nov 22. Epub ahead of print.
95. Kleypas Y, McCubbin D, Curnow ES. The role of environmental cleaning in
health care-associated infections. Crit Care Nurs Q 2011;34(1):11–7.
96. Shaughnessy MK, Micielli RL, DePestel DD, et al. Evaluation of hospital
room assignment and acquisition of Clostridium difficile infection. Infect
Control Hosp Epidemiol 2011;32(3):201–6.
97. Adams S, Herrera A 3rd, Miller L, Soto R. Visitation in the intensive care
unit: impact on infection prevention and control. Crit Care Nurs Q 2011;
34(1):3–10.
98. Teltsch DY, Hanley J, Loo V, et al. Infection acquisition following intensive
care unit room privatization. Arch Intern Med 2011;171(1):32–8.
99. Nair N, Kourbatova E, Poole K, et al. Molecular epidemiology of
methicillin-resistant Staphylococcus aureus (MRSA) among patients
admitted to adult intensive care units: the STAR*ICU trial. Infect Control
Hosp Epidemiol 2011;32(11):1057–63. Epub 2011, Sep 20.
100. Prabaker K, Weinstein RA. Trends in antimicrobial resistance in intensive
care units in the United States. Curr Opin Crit Care 2011;17(5):472–9.
101. Leonhardt KK, Yakusheva O, Phelan D, et al. Clinical effectiveness and cost
benefit of universal versus targeted methicillin-resistant Staphylococcus
aureus screening upon admission in hospitals. Infect Control Hosp
Epidemiol 2011;32(8):797–803.
102. Kohlenberg A, Schwab F, Behnke M, et al. Screening and control of
methicillin-resistant Staphylococcus aureus in 186 intensive care units:
different situations and individual solutions. Crit Care 2011;15(6):R285.
Epub ahead of print.
 Section II—Clinical and Laboratory Assessment 164.e3
103. Huskins WC, Huckabee CM, O’Grady NP, et al. Intervention to reduce
transmission of resistant bacteria in intensive care. N Engl J Med 2011;
364(15):1407–18.
104. Kypraios T, O’Neill PD, Huang SS, et al. Assessing the role of undetected
colonization and isolation precautions in reducing methicillin-resistant
Staphylococcus aureus transmission in intensive care units. BMC Infect Dis
2010;10:29.
105. Martinez-Capolino C, Reyes K, Johnson L, et al. Impact of active
surveillance on meticillin-resistant Staphylococcus aureus transmission and
hospital resource utilisation. J Hosp Infect 2010;74(3):232–7. Epub 2010,
Feb 12.
106. Olchanski N, Mathews C, Fusfeld L, et al. Assessment of the influence of
test characteristics on the clinical and cost impacts of methicillin-resistant
Staphylococcus aureus screening programs in US hospitals. Infect Control
Hosp Epidemiol 2011;32(3):250–7.
107. Byrnes MC, Adegboyega T, Riggle A, et al. Nasal swabs collected routinely
to screen for colonization by methicillin-resistant Staphylococcus aureus in
intensive care units are a sensitive screening test for the organism in clinical
cultures. Surg Infect (Larchmt) 2010;11(6):511–5.
108. Ananda-Rajah MR, McBryde ES, Buising KL, et al. The role of general
quality improvement measures in decreasing the burden of endemic MRSA
in a medical-surgical intensive care unit. Intensive Care Med
2010;36(11):1890–8. Epub 2010, Aug 6.
109. Cook E, Marchaim D, Kaye KS. Building a successful infection prevention
program: key components, processes, and economics. Infect Dis Clin North
Am 2011;25(1):1–19.
110. Patterson JE, Malani PN, Maragakis LL. Infection control in the intensive
care unit: progress and challenges in systems and accountability. Crit Care
Med 2010;38(8 Suppl):S265–8.
111. Raphael J, Landovitz MD, Currier JS. Postexposure prophylaxis for HIV
infection. N Engl J Med 2009;361:1768–75.
112. Updated U.S. Public Health Service guidelines for the management of
occupational exposures to HIV and recommendations for postexposure
prophylaxis. Atlanta: Centers for Disease Control and Prevention; 2005.
113. Updated information regarding antiretroviral agents used as HIV
postexposure prophylaxis for occupational HIV exposures. MMWR Morb
Mortal Wkly Rep 2007;56:1291–2.
114. Post-exposure prophylaxis to prevent HIV infection: joint WHO/ILO
guidelines on post-exposure prophylaxis (PEP) to prevent HIV infection.
Geneva: World Health Organization; 2007.
115. Barsanti MC, Woeltje KF. Infection prevention in the intensive care unit.
Infect Dis Clin North Am 2009;23(3):703–25.

Biomarkers
Robert G. Siman
Introduction
Acute brain injuries triggered by head trauma, intracranial
aneurysm rupture, cardiac arrest, and stroke among others are
an enormous medical, social, and economic problem. Acute
traumatic brain injury (TBI) is the most common cause of
death and disability in children and young adults.1 Cardiac
arrest kills about 300,000 individuals annually in the United
States, and even successful resuscitation of out-of-hospital
cases leads to neurologic dysfunction in the vast majority of
instances (www.ctsnet.org/doc/2988). Approximately 800,000
coronary artery bypass surgeries are performed yearly in the
United States, and it is estimated that as many as 80% are
associated with postsurgical impairments in neurologic or
cognitive function.2,3 More than 750,000 strokes occur in the
U.S. annually, and treatment with clot-dissolving drugs is
indicated for only a small proportion of cases.4 About 30,000
cases of aneursymal subarachnoid hemorrhage (SAH) occur
per year, and severe aneurysmal SAH carries a high risk of
brain damage and neurologic dysfunction, especially when
accompanied by cerebral vasospasm.
Treatments conducted in the neurocritical care unit
(NCCU) and emergency department are directed at the
pathophysiologic processes that occur with acute brain inju-
ries. Progress is being made, in part through the use of neu-
romonitoring and neuroimaging to better understand these
pathophysiologic processes, identify deleterious risk factors,
improve preservation of brain tissue, and reduce mortality,
morbidity, and long-term brain dysfunction. However, it has
been difficult to assess in a rigorous fashion the relative merits
of the many treatment options available in the NCCU, and
considerable strides still need to be made to identify patients
at risk for irreversible brain damage, detect early and poten-
tially treatable stages before the onset of irreversible injury,
rapidly identify patients who most benefit from specific treat-
ment measures to reduce subsequent neurodegeneration and
brain dysfunction, and optimize treatment parameters for
reducing brain damage and improving outcome. Further-
more, despite 25 years of extensive basic and clinical research
into intraneuronal signaling mechanisms for neuron death
and pharmacologic strategies for neuroprotection, there still
are no U.S. Food and Drug Administration (FDA)–approved
neuroprotective therapies for any form of acute brain injury.
Recent multidisciplinary workshops (e.g., Classification of
TBI for Targeted Therapies, October 2007; Common Data
Elements workshop for Research and Psychological Health
in TBI, March 2009, Comparative Effectiveness Research in
TBI, June 2010) have identified important reasons for the
© Copyright 2013 Elsevier Inc. All rights reserved.
Chapter
18  
II
discordance between laboratory advances and clinical success
particularly in TBI. Included among the recommendations to
address this is the application of emerging technologies (e.g.,
biomarkers, genomics, and advanced magnetic resonance
imaging [MRI]) to better understand pathophysiology, help
classify disease heterogeneity, and examine treatment effects.
Simple diagnostic and prognostic tests sensitive to the onset
and magnitude of brain damage could have widespread and
far-reaching applications in the NCCU, and in particular help
identify patients at risk of irreversible brain injury and dys-
function. In TBI, computed tomography (CT) and MRI can
confirm the diagnosis of moderate to severe brain injury.
However, CT scanning does not efficiently diagnose mild or
repetitive mild TBI, which can lead to neurodegeneration and
neurologic dysfunction, and MRI has suboptimal sensitivity.5
Similarly, the impact of mild stroke or coronary bypass surgery
remains difficult to quantify with routine neuroimaging
methods.6 Prognostic tests that define patients at high risk for
death or disability and sensitive tests to detect secondary inju-
ries also could guide the direction and aggressiveness of patient
care. There is a variable risk for ischemic brain damage after
cardiac arrest, and currently there are no methods to identify
the subset of patients susceptible to irreversible brain injury
that would benefit most from neuroprotective treatment. The
significance of this shortcoming is heightened with the success
of induced hypothermia after cardiac arrest. Although this
helps preserve short- and long-term brain function, the treat-
ment is complex, expensive to administer, efficacious for only
a subset of resuscitated patients, and dependent on treatment
variables that have yet to be optimized.
Following severe SAH or TBI, monitoring often generates
data about the onset, severity, and location of secondary
pathologic events through measures of blood flow, energy
metabolism, local oxygen partial pressure, and electroenceph-
alographic activity. However, there are no direct indices or
surrogate measures for the onset and magnitude of irrevers-
ible brain damage. There are many treatment interventions
that still need to be evaluated fully or optimized in controlled
clinical trials, for example, the relative merits of mannitol or
hyperosmotic saline; determining the optimal osmolarity or
target pressures when using hyperosmolar therapy; the role of
induced normothermia; finding the optimal glucose or hemo-
globin; different ventilation strategies; and how best to prevent
venous thrombosis, among many others that have to be
answered in the NCCU. These questions can be difficult to
study in a controlled trial, in part because of the sample size
needed, the complexity of patient care, and the insensitivity
and imprecision of commonly used functional outcome
165
166 Section II—Clinical and Laboratory Assessment
measures. Surrogate markers for irreversible acute brain
damage could play a major role in helping to answer these
questions in the NCCU. The clinical evaluation of potential
neuroprotectant drugs also often requires large and expensive
trials that can present a barrier for pharma­ceutical and bio-
technology companies with promising but unproven experi-
mental therapeutics. Surrogate measures for efficacy that
reduce the complexity and cost while increasing the discrimi-
native power of early-stage drug trials would facilitate clinical
research for novel neuroprotectants and help drug discovery
and development.7 In addition, use of state-of-the-art bio-
marker platforms may help during the recovery phase of any
acute brain injury to identify effective therapies to enhance
recovery in a targeted and individualized fashion.8 This
concept may apply not only to strategies used in rehabilitation
after moderate or severe TBI but also to managing the cogni-
tive or neurobehavioral sequelae even after mild injury with
techniques such as evaluation of micro-ribonucleic acid
(RNA) species in peripheral blood mononuclear cells.9
There are many potential uses for surrogate markers for
brain damage in preclinical studies. This is especially true for
markers that relate directly and mechanistically to the under-
lying molecular pathways for neurodegeneration. Rather than
relying solely on neurologic tests that reflect in complex and
often indirect ways the anatomy, pathology and physiology of
brain damage, or histopathologic methods that restrict analy-
ses to the time of sacrifice, it may be possible to examine
mechanisms of neurodegeneration and assess their temporal
characteristics by serial sampling in living animals using sur-
rogate markers. These markers also could help identify optimal
dosing, evaluate specific neurodegenerative mechanisms, or
define efficacy of neuroprotective therapeutics over time in
living animals. Furthermore, a surrogate marker with the sen-
sitivity to detect mild brain damage may facilitate mechanistic
and experimental therapeutic studies of both single and repet-
itive mild insults, and of factors that regulate the threshold for
brain damage. Finally, comparative analyses of a panel of
potential surrogate markers for brain damage in human clini-
cal samples and preclinical experimental models should
provide new information about the clinical relevance of
various animal model systems, none of which have been vali-
dated pharmacologically because of the lack of clinically
proven neuroprotective drug treatments.
Desirable Characteristics of a
Surrogate Marker for Brain Damage
Biomarker (biologic marker) was introduced in 1989 as a
Medical Subject Heading (MeSH) term. In 2001 a National
Institutes of Health (NIH) working group standardized the
definition of a biomarker as “a characteristic that is objectively
measured and evaluated as an indicator of normal biologic
processes, pathogenic processes, or pharmacologic responses
to a therapeutic intervention.” The working group also
described different types of biomarkers including that mea-
sured on a biosample (e.g., blood, urine, cerebrospinal fluid
[CSF], or tissue), a recording obtained from a person (e.g.,
blood pressure, electrocardiogram [ECG]), or an imaging
study (e.g., MRI or CT scan). This chapter focuses on biosam­
ples. Several recent developments have stimulated the bio-
marker field: first, the completion of the Human Genome
Project and the growth of microarrays, proteomics, and
nanotechnology provides new methods to develop sensitive
high-throughput assays. Second, advances in bioinformatics
and cross-disciplinary collaborations have enhanced the
ability to retrieve and analyze large quantities of data. Third,
there now is understanding that diseases develop from the
dynamic dysregulation of several gene regulatory networks,
proteins, and metabolic alterations that reflect complex per-
turbations (genetic and environmental) of the “biologic
system.” Best practice guidelines to standardize the quality and
accessibility of biosamples and so realize the full potential and
accelerate genomic, proteomic, and metabolomic research in
acute brain injury have been published.10 It is beyond the
scope of this chapter to review the many technical, financial,
legal, and regulatory hurdles that have to be overcome to
develop clinically acceptable and validated biomarkers that are
commercially produced and FDA proved.11
To evaluate patients with potential acute brain injuries, the
measurement of biochemical markers from a CSF source are
useful in many settings, whereas serum measures would have
broader utility. Studies also suggest that use of microdialysis
(i.e., from brain interstitial fluid) to recover some biomarkers
may be feasible.12-14 An ideal CSF or serum marker should have
several characteristics,15,16 including: (1) high degrees of sen-
sitivity and specificity for brain damage, (2) appearance only
after irreversible degeneration of brain tissue, (3) rapid transit
to the bloodstream in the case of serum markers, and (4) cor-
relations with the time course and extent of brain injury.
There should be low variability in the measurement of an ideal
marker to guarantee a predictable relationship between its
concentration in CSF, serum, or urine and the degree of brain
injury. The ideal marker should provide information that is
unavailable from other clinical data alone, for example, rapid
diagnosis of at-risk patients, and levels should be responsive
to treatment interventions that improve brain function,
thereby providing a pharmacodynamic measure of irreversible
brain damage. Finally, there should be a reliable and facile
assay for the ideal marker with the simplicity, speed, and low
cost needed to quickly inform and direct patient care and
achieve widespread and routine application.
Current Status of
Surrogate Markers
A surrogate measure for brain damage that meets the stringent
requirements of the ideal marker has not emerged yet.
However, several biochemical markers have demonstrated
promise in preclinical and clinical studies and so raise the
possibility that a simple cerebrospinal fluid, blood, or urine
test for acute brain damage is feasible. Examples of clinical
studies are summarized in Table 18.1. Conceptually, proteins
that are expressed primarily in the nervous system could
appear outside the central nervous system (CNS) preferen-
tially under pathologic neurodegenerative conditions that are
accompanied by disruption of the blood-brain and blood-
cerebrospinal fluid barriers. Consistent with this concept, it is
well-established that TBI, ischemia, stroke, or SAH among
others cause localized, reversible breakdown of the blood-
brain barrier that, although transient, can persist for several
hours or occur even days following the insult.17,18 In addition,
certain nervous system–enriched proteins are detectable in
 Section II—Clinical and Laboratory Assessment 167

Table 18.1  Examples of Clinical Studies of Surrogate Markers for Acute Brain Damage
Biomarkers Source Key Finding Reference
TRAUMATIC BRAIN INJURY
S100β, NSE Serum Prognostic for poor neurologic outcome 22
Tau CSF Prognostic for long-term outcome 120
α-Spectrin CSF Related to CT findings, long-term outcome 75
Cleaved tau CSF Prognostic for elevated intracranial pressure and long-term outcome 58
SUBARACHNOID HEMORRHAGE
NFL CSF Prognostic for long-term outcome 70
S100β, GFAP Serum Associated with hemorrhage type, neurologic dysfunction, intracranial pressure 50
pNFH CSF, serum Associated with vasospasm 67
CARDIAC ARREST WITH RESUSCITATION
S100β Serum Prognostic for mortality 24
S100β, NSE Serum Prognostic for mortality (both) and memory dysfunction (S100β) 121
STROKE
S100β, GFAP CSF Predictive of infarct volume 25
S100β Serum Prognostic for long-term neurologic outcome and related to infarct volume 122
Tau CSF Prognostic for long-term neurologic outcome and related to infarct volume 123

CSF, Cerebrospinal fluid; CT, computed tomography; GFAP, glial fibrillary acidic protein; NFL, neurofilament light chain; NSE, neuron-specific enolase;
pNFH, hypophospho-neurofilament H.

human CSF and serum following brain injury or in associa-
tion with chronic neurodegenerative disease. So far the devel-
opment of clinically useful surrogate markers has been most
successful in the prion disorder Creutzfeldt-Jakob disease, for
which CSF measurement of the brain-enriched 14-3-3 protein
family is accepted into clinical practice as a diagnostic marker
for the disorder.19,20
Glial Proteins
S100β
Several non-neuronal proteins whose expression either is
brain enriched or induced in the brain in response to neuronal
injury have been evaluated in preclinical and clinical studies
as potential surrogate markers for acute brain damage. The
most extensively studied is S100β, an astrocyte-enriched
protein. S100β expression is increased with the astroglial
hypertrophic response to brain injury,21 and increases are
detectable based on two-antibody sandwich immunoassay in
CSF and serum of patients following TBI, SAH, cardiopulmo-
nary bypass surgery, cardiac arrest with resuscitation, carbon
monoxide poisoning, and ischemic or hemorrhagic stroke.22-33
Many of these studies demonstrate a prognostic relationship
between serum or CSF level of S100β and neurologic outcome,
mortality, or secondary cerebral insults (e.g., elevated intra-
cranial pressure),28 although there is no consensus on cutoff
values or time of sampling. Furthermore, there is a suggestion
that S100β elevation may be an indicator of impending dete-
rioration, for example, an increase in intracranial pressure
(ICP)28 or of when to obtain a head CT scan in mild TBI.34-36
In severe TBI, maximal serum levels of S100β are associated
with poor long-term function assessed by the Glasgow Coma
Scale,37 and some studies suggest that after multivariable
analysis models that include mass lesion, pupils, S100β
and another astrocyte marker, glial fibrillary acidic protein
(GFAP), best predict death and unfavorable outcome.29
In SAH, admission serum level of S100β is associated with
short-term survival, injury severity as assessed by neuroimag-
ing, and poor outcome at 1 year, whereas therapeutic inter­
vention with external ventricular drainage is associated with
reduced serum S100β level.38
However, not all studies have found that use of S100β is
reliable to discriminate severity of injury or prognosis.39-41
Several limitations also are associated with S100β as a bio-
marker for acute brain injury. First, high serum concentra-
tions of S100β are observed in association with chronic
diseases unrelated to acute brain damage42,43; this can occur in
humans even in the absence of brain injury and may be
released after injury by tissue other than the brain.44,45 Second,
not all types of brain damage are associated with measurable
increases in serum concentration of S100β46 Third, S100β
elevations are not always predictive of long-term neurologic
dysfunction47 and have shown inconsistent prognostic value
in mild forms of acute brain damage.48 In children, these
inconsistencies may be associated with patient age.34 Fourth,
the data that demonstrate a positive relationship between
serum and CSF levels of S100β and patient prognosis have
been largely derived from retrospective analysis, that is, cutoff
biomarker levels that relate to distinct outcomes all have been
selected from observing the data post-hoc. Consequently,
multiple, reproducible studies have yet to establish definitively
the clinical utility of S100β as a diagnostic or prognostic
measure of acute brain injuries, or as a pharmacodynamic tool
to measure treatment effectiveness. Finally, methodologic
considerations add variance to the clinical measure of S100β
that will require both independent confirmation and inter-
laboratory standardization to promote the prognostic utility
168 Section II—Clinical and Laboratory Assessment
of the biomarker.49 In summary, although certainly worthy of
study, S100β is not yet established as either a specific and
broadly applicable surrogate marker for acute brain damage,
or a highly reproducible and consistent index of short-term
and long-term prognosis that may be used in a defined
clinical application in a prospective fashion across multiple
laboratories.
Other Glial Proteins
There have been fewer studies on other glial-specific proteins
that show promise as candidate biomarkers for acute brain
injuries; these include the astroglial GFAP and the oligoden-
droglial myelin basic protein (MBP). In SAH, serum levels of
GFAP on hospital arrival are associated with scales of neuro-
logic dysfunction and CT Fisher grade scores, and increase in
patients with rebleeding or ischemic events.50,51 Serum GFAP
levels at admission are associated with abnormal head CT
findings, mortality, and poor long-term outcome following
severe TBI; 20- to 30-fold increases are observed in patients
with poor outcome.29,52,53 In pediatric TBI, initial serum MBP
and S100β levels correlate with poor outcome, although the
relationship is stronger for children 4 years or younger.54
Serum and CSF GFAP levels also are elevated in the severe
pediatric TBI.55 Although this elevation is associated with
Pediatric Cerebral Performance Category scores at 6 months,
the GFAP increase does not discriminate admission injury
severity or differentiate therapeutic strategies. In children who
receive extracorporeal membrane oxygenation, peak GFAP
levels are higher in children with brain injury than those
without.56 It remains uncertain, however, whether these glial
markers can guide therapy.
Neuronal Proteins
Several nervous system–enriched proteins expressed predomi-
nantly or exclusively in neurons have been studied as potential
surrogate markers for acute brain injuries, including tau,
neuron-specific enolase (NSE), neurofilament polypeptides,
and α-spectrin (see Table 18.1).
Tau
The microtubule-associated protein tau is increased in CSF
and serum of severely head-injured patients primarily as tau
proteolytic fragments of approximately 30 to 50 kDa of inde-
terminate sequence rather than the intact protein and may
serve as indicators of outcome.57-59 Zemlan and colleagues
developed a monoclonal antibody preferentially reactive with
these unidentified tau fragments and established a two-
antibody immunoassay to provide evidence for measurable
elevations in serum tau concentrations in association with
human TBI.57 However, serum-cleaved tau was not detected
in several head-injury cases, and without an unambiguous
characterization of these tau fragments, it is difficult to relate
their appearance in body fluids to mechanisms that underlie
neurodegeneration.
Neuron-Specific Enolase
NSE is another neuronal protein that has been studied exten-
sively as a surrogate marker for acute brain damage. The
results, however, are mixed. In some studies the magnitude of
serum NSE elevation is associated with long-term brain func-
tional outcome following cardiopulmonary resuscitation.60,61
However, the cutoff serum NSE level for maximal prognostic
sensitivity and specificity was selected in a post-hoc fashion.
Thus it remains to be determined whether serum NSE mea-
sures have sufficient reliability and robustness for prospective
applications in routine clinical practice. In severe pediatric
TBI, serum NSE changes have a poor association with
outcome in children less than 4 years old or in victims of
inflicted TBI.62 In severe adult TBI, both NSE and S100β
serum levels correlated with contusion volumes or episodes
of reduced cerebral perfusion pressure (CPP), but their
respective release patterns over time differ for cases of primary
cortical contusion, diffuse axonal injury, and cerebral edema
without focal mass lesions.28,63 Increases in NSE have been
observed in boxers with predominantly blows to the head
rather than the body64 and after an extended resting period
that suggests repetitive head trauma results in sustained
release of this brain-specific protein to the peripheral circula-
tion.65 In SAH, serum NSE levels are not always elevated.50
Consequently, whereas NSE has received increased attention
as a marker for brain damage associated with cardiac arrest
and resuscitation, its use in other acute brain injuries appears
thus far to be limited.
Neurofilaments
Neurofilaments are the intermediate filaments of neurons,
and because of their unique protein composition the constitu-
ent neurofilament triplet polypeptides neurofilament heavy
chain (NFH), neurofilament middle chain (NFM), and neu-
rofilament light chain (NFL) have been evaluated as markers
for acute brain damage in several studies. In SAH, CSF levels
of a particular phosphoform of NFH are increased in patients
with poor 3-month outcome,66 whereas serum NFH phospho-
form levels increase when there is cerebral vasospasm.67 A
particular hypophosphorylated form of NFH is expressed in
neurons predominantly within axons, and CSF alterations in
this phosphoform have been proposed as an index of axonal
damage.66 CSF NFL levels also have been shown to increase
following severe TBI or SAH68 and reportedly correlate with
long-term outcome following both cardiac arrest and resusci-
tation69 and SAH.70 However, other studies suggest that
although CSF NFL concentrations are elevated in SAH, espe-
cially those with acute ischemia, there is no association with
outcome suggesting that neurofilament as a biomarker may be
better suited to TBI when there is axonal injury.71 Consistent
with this, the level of neurofilament protein heavy chain in
serum is not associated with brain ischemia on routine brain
imaging techniques after carotid endarectomy.72 Proteolytic
breakdown products of neurofilament heavy chains can be
recovered using microdialysis, and these brain extracellular
fluid markers might be of prognostic value.13,21
Spectrin
Another cytoskeletal protein that is expressed in neurons and
shows promise as a systemic marker for acute brain damage
is the actin-binding protein spectrin. The alpha subunit of the
nonerythroid form of spectrin, referred to as αII, although
expressed in a diverse number of cell types, is expressed at high
 Section II—Clinical and Laboratory Assessment 169
levels in neurons.73 Levels of αII-spectrin and spectrin break-
down products (SBDP145 produced by calpain, and SBDP120
produced by caspase-3)74 increase in CSF following severe
TBI,75 SAH,76 and cardiac arrest with resuscitation,77 and the
increases are associated with various measures of patient
outcome. In addition is a correlation between patients with
longer ICP elevations and CSF αII-SBDPs.78 The specific
nature of these breakdown products and their relevance to the
pathophysiologic processes that underlie acute neurodegen-
eration are described further in the following section.
In summary, there are a number of glial- and neuron-
enriched proteins and markers from brain endothelial cells
that have shown promise as diagnostic or prognostic indica-
tors of acute brain damage deserving of further study, but
unfortunately none has emerged yet as a validated and reliable
prospective marker with routine clinical applicability.79,80 In
addition, whether a combination of markers is more valuable
than single markers is still to be validated.81,82 Moreover, the
CSF and serum elevations in the most widely studied bio-
markers have yet to be related to underlying pathophysiologic
brain processes or brain regions and neural pathways being
impacted, and with the exception of a single NFH phospho-
form, to neuronal subcellular elements that may be preferen-
tially affected.
Proteolytic Mechanisms for Acute
Neurodegeneration That Could
Be Exploited to Develop
Surrogate Markers
The biochemical signaling pathways that underlie neurode-
generation offer the prospect to develop mechanism-based
markers for brain damage potentially measurable in accessible
body fluids. In particular, apoptotic and necrotic neurodegen-
erative signaling are characterized by the activation of distinct
families of cysteine proteases, and the proteolytic “finger-
prints” of degraded protein substrates for these cell death–
activated proteases have been used widely as tissue markers
for experimental neurodegeneration. Most modes of apop-
totic neuronal death are triggered by the activation of cas-
pases, cysteine proteases with the unusual specificity for
cleavage of protein substrates on the carboxyl side of aspartate
residues. Caspase activation and neuronal apoptosis have been
demonstrated as contributing factors in numerous studies of
acute brain damage after TBI, global and focal cerebral isch-
emia, and hypoxia-hypoglycemia.83,84 Eleven human caspases
have been identified, at least 8 of which are expressed in the
brain and may be activated during apoptosis. These proteases
are synthesized as inactive zymogens and require proteolytic
processing and oligomeric assembly for their activation. Con-
sequently, cleavage site-specific antibodies have been prepared
that react preferentially with activated, processed subunits of
many of the caspases, and these have facilitated the biochemi-
cal quantitation and histochemical localization of the activa-
tion of caspases 3, 6, 8, and 9 that occur during acute brain
damage. Furthermore, the substrate specificity of caspases is
determined primarily by short domains of four or five amino
acids that end in an aspartate residue, making it possible to
design and prepare cleavage site-specific antibodies that react
specifically with preferred caspase recognition motifs. Cleav-
age site-specific antibodies reactive with caspase-derived
proteolytic fragments of lamin A, αII-spectrin, polyADP-
ribose polymerase, actin, and other caspase substrates have
been useful tissue markers for caspase-mediated apoptosis.85-87
Thus far the only caspase substrates evaluated as potential
CSF and serum markers for acute brain injury have been
the caspase-1 products interleukin-1β and interleukin-18.88,89
However, specificity of these proinflammatory cytokines as
markers for brain damage has been limited by their abundant
expression in hematopoietic cells that mediate diverse inflam-
matory processes throughout the body. Recent clinical studies
however suggest serum interleukin-1β and to a lesser extent
tumor necrosis factor alpha (TNFα) may show promise as
markers of poor outcome or intracranial hypertension after
TBI.90 The identification of additional caspase substrate pro-
teins that are expressed at high levels predominantly in the
brain, degraded during apoptotic neurodegeneration in asso-
ciation with acute brain damage, and leaked into the CSF or
blood in an injury-specific manner could lead to the identifi-
cation of a mechanism-based surrogate marker for the apop-
totic component of acute brain injury.
Activation of the calpain family of calcium-activated cyste-
ine proteases has been linked in numerous studies to necrotic
neurodegeneration.91-94 Calpains are inactive at normal, physi-
ologic intraneuronal free calcium concentrations, but are acti-
vated by the intraneuronal calcium overload that accompanies
excitotoxicity; global and focal cerebral ischemia; traumatic
injury to the brain, spinal cord, and peripheral nerves; hypoxia-
hypoglycemia; status epilepticus; oxidative and nitrative stress;
and mitochondrial respiratory failure. Calpains also are acti-
vated in some instances of apoptosis, and very brief episodes
of calpain activation can occur without neurodegeneration.
However, strong and sustained calpain activation without an
accompanying caspase activation is diagnostic for neuronal
necrosis.95-97 One of the most abundant high-affinity calpain
substrate proteins is the alpha subunit of spectrin, and cleav-
age site–specific antibodies reactive with calpain-derived frag-
ments of α-spectrin, but not with caspase-derived fragments
or the spectrin holoprotein, have been used widely as cell
and tissue markers for calpain activity. They have linked sus-
tained calpain activation with the necrotic neurodegeneration
that accompanies TBI, global and focal cerebral ischemia,
and hypoxia-hypoglycemia.87,95,98-101 Similar to the caspase-
derived protein fragments, the immunodetection of calpain
substrate protein fragments that are generated from brain-
enriched proteins during neuronal necrosis and enter the CSF
or bloodstream with acute brain damage could be valuable
mechanism-based surrogate markers for the necrotic compo-
nent of acute brain injuries.
Novel Approaches to Identify
Potential Surrogate Markers
for Brain Damage
Enormous strides have been made with proteomics methods
to identify proteins in complex biologic samples, and these
methods now are being applied to define CSF and serum
proteomes and analyses of changes in the levels of specific
proteins in association with disease states. Protein separation
170 Section II—Clinical and Laboratory Assessment
and identification are performed by either two-dimensional
polyacrylamide gel electrophoresis (2D PAGE), liquid chro-
matography (LC), mass spectrometry (MS), or combinations
of 2D PAGE, LC, and MS.102,103 Although it may be theoreti-
cally possible to measure a protein within protein-rich serum
whose levels change in relation to the time course and extent
of brain damage, no such marker protein has been identified
yet by proteomics methods. The sensitivity of detection by 2D
PAGE or MS approaches may need to improve to detect the
expected minute (attomolar [10–18 molar]) quantity of brain-
enriched proteins that may circulate in the blood post injury.
The complexity of methods for MS and their considerable
expense are further barriers to their adoption into routine
clinical practice. Although proteomics approaches to identify
and quantitate surrogate markers for acute brain damage have
promise, it remains to be seen whether they offer sufficient
simplicity, sensitivity, and specificity to identify and quantify
surrogate markers for brain damage that will have broad clini-
cal applicability.
Unlike proteomics methods, immunoassays may offer the
simplicity, sensitivity, and specificity that would be required
of a clinically useful surrogate marker for brain damage.
Rather than evaluating target antigens one by one as potential
bases for a surrogate measure for brain damage in accessible
body fluids, as has been conducted so far, a biologically based
approach to identify numerous potential surrogate markers
could expedite the development and validation of a clinically
useful immunoassay. The Siman laboratory initiated a global,
systematic analysis of protein release by degenerating neurons
maintained in primary culture to establish a neurobiologic
approach to identify new potential surrogate markers.97,104,105
The value of this neurobiologic approach is supported by the
finding that proteins released preferentially by degenerating
cultured neurons also are increased markedly in CSF and
serum following TBI or cerebral ischemia in rodent experi-
mental models (see following section). This neurobiologic
approach can identify new markers and provide immunoas-
says for them with the potential to aid in diagnosis, prognosis,
and therapeutic evaluation of acute brain damage.
The identification of a large number of novel candidate
markers for acute brain injuries would foster development of
a panel of markers that in turn can be expected to have diag-
nostic and prognostic sensitivity and specificity beyond that
of individual markers. Indeed, clinical practices in oncology
and cardiology have benefited considerably from the avail-
ability of multiple markers to classify tumors and detect and
characterize myocardial infarction. In addition, a neurobio-
logic approach to identify surrogate markers for brain damage
based on underlying proteolytic pathways for necrotic and
apoptotic neurodegeneration would be valuable to define
further the biochemical mechanisms for brain damage both
in patients and preclinical experimental models.
Neuron-Enriched Proteins Released
During Neurodegeneration Are
Markers for Acute Brain Damage
in Experimental Models
As a first step to develop and validate a panel of biomarkers
for acute brain damage, the neurobiologic approach men-
tioned earlier examined the most prevalent polypeptides
released from degenerating cultured cortical neurons using
well-established proteomics methods of two-dimensional
electrophoresis and mass spectrometry. It identified 14 of the
most abundant polypeptides released preferentially during
neuronal necrosis.97 Seven are known to be expressed pri-
marily in the brain, and are present in the adult CNS: 14-3-
3β and ζ isoforms, ubiquitin C-terminal hydrolase L1
(UCH-L1), collapsin response mediator proteins 2 and 4
(CRMP-2 and CRMP-4), GAP-43 and creatine kinase B.106-110
Because these 7 proteins are brain-enriched proteins released
in response to neurodegeneration, they are candidate surro-
gate markers for brain injury potentially measurable in acces-
sible body fluids. A second set of candidate markers was
identified by examining abundant cytoskeletal proteins that
are expressed predominantly or exclusively in neurons. This
approach demonstrated that dying neurons release several
intact structural proteins and numerous proteolytic frag-
ments, including NFH, the microtubule-associated proteins 2
(MAP2) and tau, the α-subunit of the actin-binding protein
spectrin, calpain-derived fragments of α-spectrin and tau
and caspase derivatives of α-spectrin and the nuclear matrix
protein lamin A.
To address whether proteins and their proteolyic frag-
ments released from degenerating cultured neurons may be
markers for brain damage, CSF and serum samples were
evaluated from rats subjected to lateral fluid-percussion TBI
or transient global cerebral ischemia. It was found that two
sets of cytoskeletal proteolytic fragments and several intact
proteins increase markedly in CSF following experimental
TBI in the rat.97,104 Tau is detectable as five calpain deriva-
tives that co-migrate with tau fragments released during
necrotic neurodegeneration in vitro. Additionally, α-spectrin
fragments of 120 to 150 kDa and intact 14-3-3 protein iso-
forms β and ζ and NFH also increase in rat CSF following
TBI. Cleavage site-specific antibodies reactive with calpain-
derived α-spectrin fragments confirm the appearance of
approximately 150 and 145 kDa calpain derivatives of spec-
trin. The additional set of α-spectrin derivatives of approxi-
mately 120 kDa that also increase after brain injury, albeit to
a lesser degree, are similar in size to an α-spectrin fragment
formed by caspase proteolysis93,111 and likely are caspase
derived. The biomarker results are consistent with histopa-
thology studies that indicate TBI triggers brain cell death
both by necrosis and apoptosis, with necrosis predominating
under most circumstances.100,112,113 These findings demon-
strate that several proteins released from dying cultured
neurons increase in an accessible body fluid after experi-
mental TBI, and suggest that different pathophysiologic
mechanisms that contribute to brain damage may be distin-
guishable in living animals by mechanistically distinct bio-
chemical markers for neurodegeneration.
To determine whether the injury-induced increases in intact
neuronal proteins and proteolytic fragments reflect general-
ized changes in CSF protein levels, by 2D PAGE and silver
staining the proteome for rat CSF proteins following moder-
ate TBI was examined. The 2D patterns of the most abundant
rat CSF polypeptides were similar with or without injury.96 It
was concluded that the TBI-induced elevations in CSF levels
of the neuron-enriched proteins detectable by immunoassay
are not accompanied by generalized changes in the levels of
major CSF proteins. The results highlight the sensitivity of
immunoassay methods to detect brain injury–induced altera-
tions in neuronal polypeptides present in relatively low abun-
dance in an accessible body fluid.
 Section II—Clinical and Laboratory Assessment 171

Surrogate markers could have important clinical uses to

Proteins Released from
Degenerating Neurons as
Pharmacodynamic Markers for
Experimental Acute Brain Damage
A surrogate marker for acute brain damage that has preclinical
and clinical use should be sensitive to the injury magnitude.
“Mild” TBI in the rat resembles mild human TBI because
there is relatively sparse neurodegeneration and injury is
without accompanying mortality, apnea, or seizures. Even
mild TBI in the rat leads to increased CSF levels of 14-3-3β,
14-3-3ζ, calpain-cleaved tau, NFH, UCH-L1 and calpain-
cleaved α-spectrin (Fig. 18.1). At all time points examined,
CSF concentrations of the six markers are greater following
moderate TBI than mild TBI.
To assess whether CSF levels of proteins released from
degenerating neurons could serve as markers for ischemic
brain injury and reflect its severity, CSF was examined 48
hours after an episode of forebrain ischemia varying from 3
to 10 minutes.88,104 Levels of 14-3-3β, 14-3-3ζ, and α-spectrin
fragments are below the limit of detection in CSF of sham rats
subjected to surgical procedures only, but are readily measur-
able following ischemia and increase by up to 20-fold. Quan-
titative Western blot analysis demonstrates that CSF levels of
14-3-3β increase with increasing ischemia duration, and cor-
relate with the severity of acute cerebral histopathology as
assessed by Fluoro-Jade and silver impregnation labeling of
acutely degenerating cells (Fig. 18.2; see also reference 68).
When cases of ischemia are analyzed separate from the insult
duration, CSF levels of 14-3-3β and the calpain-cleaved
17 kDa tau fragment correlate significantly with the numbers
of acutely degenerating neurons. This relationship is not ob-
served for all CSF markers because calpain-cleaved α-spectrin
levels have a limited relationship with injury severity. These
results support the hypothesis that neuron-enriched proteins
such as 14-3-3β change dynamically in CSF in relation to the
degree of acute brain injury, and raise the possibility that brain
injuries of mild severity that may be difficult to diagnose with
routine neuroimaging, may be detectable instead using sur-
rogate markers for neurodegeneration.
identify patients at risk for acute brain injury, stratify patients
based on an early indicator of the severity of acute brain
damage, and provide relatively fast, simple, and inexpensive
assessment of candidate neuroprotectant treatments in early
stage trials. To investigate whether proteins released from
degenerating neurons may be pharmacodynamic markers that
are responsive to neuroprotectant therapy, 14-3-3 isoform
levels were studied in CSF of rats subjected to moderate TBI
and treated with magnesium that is a known neuroprotectant
in the rat.114-117 As shown on the right of Figure 18.2, magne-
sium treatment significantly attenuated the TBI-induced
increase in 14-3-3β by 30% and the increase in 14-3-3ζ by
37%. The reduced marker levels were observed at 6 hours post
injury. This early attenuation of biomarker increases occurs
well before the full temporal development of injury-induced
histopathology, and supports the potential of diagnostic,
prognostic, and therapeutic applications of surrogate markers
in the clinic.
Novel CSF Neurodegeneration
Markers Provide Evidence for Acute
Central Nervous System Damage
in Humans
Some patients who undergo cardiopulmonary bypass develop
lasting cognitive and neurologic problems post surgery.2,3
Proteins released from degenerating neurons obtained from
patients’ lumbar CSF during thoracoabdominal surgeries to
repair aortic aneurysms were analyzed; some of the patients
underwent cardiopulmonary bypass with complete circula-
tion arrest. As shown in Figure 18.3, a subset of patients
exhibit time-dependent increases in 14-3-3β, 14-3-3ζ, calpain-
cleaved α-spectrin, a hypophosphorylated form of NFH
reactive with a phosphoform-specific monoclonal antibody
(pNFH), and UCH-L1, manifesting in most instances between
2 and 12 hours after circulation arrest. All seven cases of com-
plete circulation arrest exhibit time-dependent increases in
multiple novel markers for neurodegeneration.77 More recent
experiments have increased the biomarker panel to eight

24 6 6 24 24 Sham Ischemic
Sham Mild Mod Mild Mod

Intact
Intact 150 kDa α-spectrin
150 kDa α-spectrin 120 kDa
120 kDa

30 kDa 14-3-3 ζ 14-3-3 ζ

30 kDa 14-3-3 β 14-3-3 β

A B

Fig. 18.1  Proteins released by degenerating neurons increase in cerebrospinal fluid (CSF) of rats following traumatic brain injury (TBI) (A) or cerebral
ischemia (B). TBI of varying severity was produced in a lateral fluid-percussion rat model. At 6 or 24 hours following TBI or sham surgery, CSF was
collected and subjected to Western blot analysis for α-spectrin, 14-3-3β, and 14-3-3ζ. Compared to sham controls, TBI rapidly increased intact
α-spectrin, a ≈150 kDa calpain cleavage product, and to a lesser degree a ≈120 kDa caspase cleavage product, and the two 14-3-3 isoforms. CSF
levels of all these proteins varied in relation to injury severity. In B, a similar analysis was conducted in rats subjected to either sham surgery or 10
minutes of global forebrain ischemia with 48-hour reperfusion. In both models the calpain derivative of α-spectrin predominates but the caspase
derivative also is sometimes increased, that is, both necrotic and apoptotic mechanisms contribute to the acute brain injuries. (Modified from Siman R,
McIntosh TK, Soltesz KM, et al. Proteins released from degenerating neurons are surrogate markers for acute brain damage in the rat. Neurobiol Dis 2004;16:311–20.)
172 Section II—Clinical and Laboratory Assessment

21,000 14-3-3β 14-3-3β 14-3-3ζ

18,000 R2 = 0.8046

Protein level (U/µL)

15,000 * **
12,000
9000 400
600
6000

CSF protein level (U/µL)

3000
0 300
0 10,000 20,000 30,000 40,000 50,000 400
Fluoro-Jade Positive Cells
200

5000 Calpain-cleaved spectrin 200

R2 = 0.0184
Protein level (U/µL)

100
4000

3000

2000

Magnesium

Magnesium
Vehicle

Vehicle
1000

0
0 10,000 20,000 30,000 40,000 50,000
Fluoro-Jade Positive Cells
Fig. 18.2  14-3-3 protein levels in rat cerebrospinal fluid (CSF) reflect the magnitude of experimental brain injury and are pharmcodynamic
measures of neuroprotectant therapy. Left, CSF content of 14-3-3β but not cleaved α-spectrin correlates with global ischemic neurodegeneration
irrespective of the duration of the ischemia. The total numbers of acutely degenerating neurons at 48 hours post injury were determined from Fluoro-
Jade staining and compared with CSF 14-3-3β and calpain-cleaved α-spectrin content from rats given episodes of global forebrain ischemia ranging
from 3 to 10 minutes. Degenerating cell counts are from the major vulnerable regions: hippocampus, parietal cortex, striatum, and inferior colliculus.
Right, The increase in CSF 14-3-3β and 14-3-3ζ following experimental traumatic brain injury (TBI) in the rat are attenuated by a neuroprotectant
treatment. Levels of the two 14-3-3 variants are shown 6 hours after lateral fluid percussion TBI in the rat, either in the absence or presence of
treatment with magnesium chloride (n = 12/group), an established neuroprotectant in this TBI model. *, p < 0.05; **, p < 0.01. (Left, modified from
Siman R, Zhang C, Roberts VL, et al. Novel surrogate markers for acute brain damage: cerebrospinal fluid levels correlate with severity of ischemic neurodegeneration in
the rat. J Cereb Blood Flow Metab 2005;25:1433–44.)

Cleaved
14-3-3β 14-3-3ζ α-spectrin pNFH UCH-L1
Case # 0 24 0 24 0 24 0 24 0 24
Circ arrest
1
5
13
16
18
38

No arrest
3
6
8
9
11
12
14
31
37
45
49

Fig. 18.3  Increases in multiple cerebrospinal fluid (CSF) markers occur in patients who undergo aortic surgery using cardiopulmonary
bypass with deep hypothermic circulatory arrest (DHCA). CSF levels of 14-3-3β, 14-3-3ζ, cleaved α-spectrin, a hypophosphorylated form of
neurofilament H (pNFH), and UCH-L1 were quantified before cardiopulmonary bypass with DHCA and 24 hours after reestablishment of circulation,
and divided into four quadrants: low (blue), moderate (yellow), high (orange), and very high (red) levels. The aortic surgical cases are shown separated
into groups that either were subjected to DHCA or not. Note that all cases of DHCA are accompanied by time-dependent CSF increases in at least
four of the proteins. In addition, 3 of 12 cases that did not involve cardiopulmonary bypass with DHCA also show time-dependent CSF increases in
at least four of the proteins. All three of these cases had acute neurologic complications indicative of CNS injury. (From Siman R, Roberts VL, McNeil E,
et al. Biomarker evidence for mild central nervous system injury after surgically-induced circulation arrest. Brain Res 2008;1213:1–11.)
 Section II—Clinical and Laboratory Assessment 173
members, as NSE and two additional phosphoforms of NFH
(dephosphorylated NFH and hyperphosphorylated NFH) also
increase following clinically induced circulation arrest. Among
12 cases of aneurysm repair that did not require complete
circulation arrest, increased CSF biomarkers for neurodegen-
eration were restricted to the subset that developed acute neu-
rologic complications.77 The findings indicate that candidate
markers identified initially in cultured rat cortical neurons can
be elevated in an accessible body fluid in humans, and raise
the possibility that CSF detection of a panel of neurodegen-
eration markers may have clinical use to identify and guide
treatment of ischemic CNS injury. A clinical utility in ischemic
brain damage is supported by a recent study of CSF neurode-
generation biomarkers in aneurysmal subarachnoid hemor-
rhage patients, in which elevations in the marker panel
were associated with the incidence and severity of cerebral
vasospasm, brain infarction, and long-term dysfunctional
outcomes.118
The same neuron-enriched protein panel that serves as a
CSF biomarker for ischemic central nervous system injury
following circulation arrest also may have clinical use in TBI.
The Siman laboratory examined a panel of eight biomarker
candidates in CSF samples derived from patients with severe
TBI, and presurgical and normal pressure hydrocephalus con-
trols. By quantitative Western blot (Fig. 18.4) and fluorescence
enzyme immunoassays (data not shown), each of the eight
proteins is near or below the detection limit in controls, but
all eight are detected readily and elevated consistently in CSF
TBI-3
TB1-10
NPH-1
TB1-8
24 48 72 96
14-3-3β
14-3-3ζ
UCH-L1
NSE
Intact α-spectrin
Calpain fragment
Caspase fragment
Total NFH
Fig. 18.4  A panel of biomarkers for human traumatic brain injury
(TBI) based on proteins released from degenerating neurons.
Western blot analysis is shown here of ventricular cerebrospinal fluid
(CSF) samples taken from cases of severe TBI (Glasgow Coma Scale <9)
or normal pressure hydrocephalus (NPH).All six proteins released from
degenerating neurons increase in human CSF after TBI. As exemplified
by case TBI-3, neurodegeneration markers are elevated by 24 hours and
reach peak levels typically between 72 and 96 hours post injury. NFH,
Neurofilament heavy chain; NSE, neuron-specific enolase; UCH, ubiquitin
C-terminal hydrolase. (From Siman R, Toraskar N, Dang A, et al. A panel of
neuron-enriched proteins as markers for traumatic brain injury in humans. J
Neurotrauma 2009; 26:1867–77.)
following TBI. Peak CSF levels of these neuron-enriched pro-
teins are reached typically between 72 and 96 hours after
severe TBI. The same panel of neuron-enriched proteins also
may be sensitive, specific, and clinically useful serum markers
for acute brain damage because severe TBI increases serum
levels of the three distinct neurofilament H phosphoforms.
The results identify eight neuron-enriched proteins that may
serve as a panel of surrogate markers for the minimally inva-
sive clinical detection, management, and therapeutic evalua-
tion of TBI and ischemic brain injuries, including mild forms
of damage.
Future Directions
Despite recent research efforts that have increased the number
of promising candidate surrogate markers for acute brain
damage, there remain several significant hurdles to bring these
biomarkers to widespread clinical practice. First will be the
establishment of validated biomarker detection methods along
with prospectively defined cutoff levels that can provide valu-
able diagnostic or prognostic information as part of routine
clinical practice. Precedents are available from other fields for
the successful development and clinical application of bio-
marker immunoassays. However, it still is uncertain whether
surrogate markers for acute brain damage can be identified
for which CSF or serum levels change consistently and can be
measured with sufficient speed to direct the course of patient
care, and with sufficient reliability to apply across multiple
hospitals. Second, additional research is required to determine
if either the time course or magnitude of biomarker altera-
tions, or both, has value in the diagnosis, prognosis, and thera-
peutic treatment of brain injuries. There are only a handful of
replicated studies that demonstrate relationships between bio-
marker alterations and neurologic, behavioral, or neuroradio-
logic measures of short- and long-term brain dysfunction, and
biomarker alterations have yet to demonstrate clinical signifi-
cance when obtained in a prospective fashion, as would be
required of a clinical tool for the diagnostic, prognostic, and
therapeutic evaluation of acute brain damage. Finally, whereas
immunoassay methods are in place to measure many candi-
date biomarkers at the laboratory bench with high sensitivity
and specificity that may be completed within a period of
hours, clinical application of these biomarkers would benefit
from quantitation that can be completed within minutes and
with extremely high reliability to provide the rapid diagnostic
or prognostic information needed at the bedside.
To date, studies of biomarkers for acute brain injury have
been conducted mostly with CSF or serum. Brain interstitial
fluid, measured through intracerebral microdialysis, is another
potential source to measure clinically meaningful biomarker
changes. Studies suggest the feasibility of microdialysis12-14 to
assess protein biomarkers for acute brain damage. It is con-
ceivable that by using probes that allow macromolecules such
as moderately sized proteins and their proteolytic fragments
to be collected, it may be possible to use brain interstitial
fluids to detect biomarkers that may help assess, on a near-
continuous basis, the onset, severity, and localization of sec-
ondary pathophysiologic processes and how management
affects these processes.
Another potential application to be explored is the contin-
ued development of biomarkers that distinguish different
mechanisms of brain injury, for example, apoptotic and
174 Section II—Clinical and Laboratory Assessment
necrotic neurodegeneration. The identification of specific
proteolytic fragments of brain-enriched proteins generated
by either calpains or caspases raises the possibility of develop-
ing an approach to assess the relative contributions of calpain-
driven necrosis and caspase-mediated apoptosis in brain
damage. Such patient-specific information may help target
pharmacologic and other interventions to the appropriate
patient most likely to benefit from particular treatments that
differentially impact pathophysiologic processes. A panel of
brain-enriched protein biomarkers also may provide clinically
useful information on the relative involvement of particular
brain regions or neuronal elements. For example, a panel of
biomarkers may be differentially expressed across the neur-
axis, raising the possibility that differential changes in them
may provide clinically useful information on the relative
involvement of particular brain regions in evolving degenera-
tive pathology.104 Petzold66 has proposed that the CSF and
serum alterations in an NFH phosphoform that is concen-
trated in axons reflects axonal damage. It is possible that with
knowledge about markers such as somatodendritic neuronal
proteins (e.g., MAP2, astroglial proteins [e.g., S100β], and
oligodendroglial proteins [e.g., MBP]), it may be possible to
distinguish pathophysiologic processes that affect particular
and subcellular elements and so may respond differently to
pharmacologic intervention.
Conclusion
An important future application of biomarkers is the potential
to serve as pharmacodynamic measures of injury. Only a
limited number of clinical studies have evaluated the impact
of treatment on the magnitude and time course of biomarker
alterations that accompany acute brain injuries, and even
fewer examples of biomarker levels that serve as pharmacody-
namic measures that respond to intervention (e.g., Baker
et al).119 Such pharmacodynamic markers could serve as end-
point measures in early stage clinical trials to assess the com-
parative efficacies of various interventions, assess the beneficial
effects of experimental treatments, and guide current strate-
gies of management, for example, determining the optimal
CPP or allow practitioners to identify not only whether
patients are at risk of damage, but also whether they are still
able to benefit from therapy. Pharmacodynamic markers that
correlate with the severity of an evolving injury can be simply
and rapidly measured and could reduce the complexity and
expense and increase the discriminative power of early stage
intervention trials. This is an important potential application
for surrogate markers given the large size and expense of con-
ducting clinical trials and the failure of many trials of experi-
mental agents in the past that relied on clinical outcome
measures. In addition, knowledge biomarkers may help relate
the neurologic and neurobehavioral endpoint measures to the
underlying pathophysiologic processes and neuroanatomic
correlates for brain dysfunction. Basic science research has
increased our knowledge about the mechanisms involved in
brain injury and led to the discovery of new biomarkers.
However, translating this research to patients’ benefits still
remains a formidable challenge.
Acknowledgments
The research described in this chapter from the Siman laboratory was
conducted with many colleagues: Drs. Chen Zhang, Robert Neumar,
Michael McGarvey, M. Sean Grady, and Tracy McIntosh, along with
Eileen Maloney, Justin Plaum, Victoria Roberts, Lindsay Dressler, Eliza-
beth McNeil, Antony Dang, and Nikhil Toraskar. This research was sup-
ported by R01 NS048234 (RS) and P50 NS08803 (TKM, MSG).
References
1. Goldstein M. Traumatic brain injury: a silent epidemic. Ann Neurol
1990;27:327.
2. Roche GW, Kanchuger M, Mangano CM, et al. Adverse cerebral outcomes
after coronary bypass surgery. N Engl J Med 1996;335:1857–64.
3. Van Dijk D, Jansen WE, Hijman R, et al. Cognitive outcome after off-pump
and on-pump coronary artery bypass graft surgery: a randomized trial.
JAMA 2002;287:1405–12.
4. Williams GR, Jiang JG, Matchar DB, et al. Incidence and occurrence of total
(first-ever and recurrent) stroke. Stroke 1999;30:2523–8.
5. Teasdale G, Teasdale E, Hadley D. Computed tomographic and magnetic
resonance imaging classification of head injury. J Neurotrauma
1992;9(Suppl. 1):S249–57.
6. Guadagno JV, Calautti C, Baron JC. Progress in imaging stroke: emerging
clinical applications. Br Med Bull 2003;65:145–57.
7. Zhang Z, Larner SF, Kobeissy F, et al. Systems biology and theranostic
approach to drug discovery and development to treat traumatic brain injury.
Methods Mol Biol 2010;662:317–29.
8. Wagner AK. TBI translational rehabilitation research in the 21st century:
exploring a Rehabilomics research model. Eur J Phys Rehabil Med 2010;
46(4):549–56.
9. Pasinetti GM, Fivecoat H, Ho L. Personalized medicine in traumatic brain
injury. Psychiatr Clin North Am 2010;33(4):905–13.
10. Manley GT, Diaz-Arrastia R, Brophy M, et al. Common data elements for
traumatic brain injury: recommendations from the biospecimens and
biomarkers working group. Arch Phys Med Rehabil 2010;91(11):1667–72.
11. Oli MW, Hayes RL, Robinson G, et al. Traumatic brain injury biomarkers:
from pipeline to diagnostic assay development. Methods Mol Biol 2009;
566:293–302.
12. Afinowi R, Tisdall M, Keir G, et al. Improving the recovery of S100β protein
in cerebral microdialysis: implications for multimodal monitoring in
neurocritical care. J Neurosci Methods 2009;181(1):95–9. Epub 2009 Mar 14.
13. Petzold A, Tisdall MM, Girbes AR, et al. In vivo monitoring of neuronal loss
in traumatic brain injury: a microdialysis study. Brain 2011;134(Pt 2):
464–83.
14. Marklund N, Blennow K, Zetterberg H, et al. Monitoring of brain interstitial
total tau and beta amyloid proteins by microdialysis in patients with
traumatic brain injury. J Neurosurg 2009;110(6):1227–37.
15. Qureshi AI. Evaluation of brain injury using a blood test: is there a role in
clinical practice? Crit Care Med 2002;30:2778–9.
16. Ingebrigtsen T, Romner B. Biochemical serum markers of traumatic brain
injury. J Trauma 2002;52:798–808.
17. Betz A, Dietrich W. Blood-brain barrier dysfunction in cerebral ischemia.
In: Ginsberg MD, Bogousslavsky J, editors. Cerebrovascular disease. Oxford,
England: Blackwell; 1998. p. 358–70.
18. Del Zoppo GJ, Hallenbeck JM. Advances in the vascular pathophysiology of
ischemic stroke. Thromb Res 2000;98:73–81.
19. Wiltfang J, Otto M, Baxter HC, et al. Isoform pattern of 14-3-3 proteins in
the cerebrospinal fluid of patients with Creutzfeldt-Jakob disease.
J Neurochem 1997;73:2485–90.
20. Green AJE. Cerebrospinal fluid brain-derived proteins in the diagnosis of
Alzheimer’s disease and Creutzfeldt-Jakob disease. Neuropathol Appl
Neurobiol 2002;28:427–40.
21. Donato R. S100: a multigenic family of calcium-modulated proteins of the
EF-hand type with intracellular and extracellular functional roles. Int J
Biochem Cell Biol 2001;33:637–68.
22. McKeating E, Andrews P, Mascia L. Relationship of neuron specific enolase
and S100 concentrations in systemic and jugular venous system to injury
severity and outcome after traumatic brain injury. Acta Neurochir 1998;
71:117–9.
23. Blomquist S, Johnsson P, Luhrs C, et al. The appearance of S100 protein in
serum during and immediately after cardiopulmonary bypass surgery:
a possible marker for cerebral injury. J Cardiotharac Vasc Anesth 1997;11:
699–703.
24. Rosen H. Increased serum levels of the S-100 protein are associated with
hypoxic brain damage after cardiac arrest. Stroke 1998;29:473–7.
25. Aurell A, Rosengren, I, Karlsson B, et al. Determination of S-100 and glial
fibrillary acidic protein concentration in cerebrospinal fluid after brain
infarction. Stroke 1991;22:1254–8.
 Section II—Clinical and Laboratory Assessment 175
26. Mussack T, Biberthaler P, Kanz K-G, et al. Serum S100β and interleukin-8 as
predictive markers for comparative neurologic outcome analysis of patients
after cardiac arrest and severe traumatic brain injury. Crit Care Med 2002;
30:2669–74.
27. Wunderlich MT, Ebert AD, Kratz T, et al. Early neurobehavioral outcome
after stroke is related to release of neurobiochemical markers of brain
damage. Stroke 1999;30:1190–5.
28. Stein DM, Kufera JA, Lindell A, et al. Association of CSF biomarkers and
secondary insults following severe traumatic brain injury. Neurocrit Care
2011;14(2):200–7.
29. Vos PE, Jacobs B, Andriessen TM, et al. GFAP and S100β are biomarkers of
traumatic brain injury: an observational cohort study. Neurology 2010;
75(20):1786–93.
30. Wojtczak-Soska K, Lelonek M. S-100β protein: an early prognostic marker
after cardiac arrest. Cardiol J 2010;17(5):532–6.
31. Murillo-Cabezas F, Muñoz-Sánchez MA, Rincón-Ferrari MD, et al. The
prognostic value of the temporal course of S100beta protein in post-acute
severe brain injury: a prospective and observational study. Brain Inj 2010;
24(4):609–19.
32. Cakir Z, Aslan S, Umudum Z, et al. S-100β and neuron-specific enolase levels in
carbon monoxide-related brain injury. Am J Emerg Med 2010;28(1):61–7.
33. James ML, Blessing R, Phillips-Bute BG, et al. S100β and brain natriuretic
peptide predict functional neurological outcome after intracerebral
haemorrhage. Biomarkers 2009;14(6):388–94.
34. Filippidis AS, Papadopoulos DC, Kapsalaki EZ, et al. Role of the S100β
serum biomarker in the treatment of children suffering from mild traumatic
brain injury. Neurosurg Focus 2010;29(5):E2.
35. Hallén M, Karlsson M, Carlhed R, et al. S-100β in serum and urine after
traumatic head injury in children. J Trauma 2010;69(2):284–9.
36. Undén J, Romner B. Can low serum levels of S100β predict normal CT
findings after minor head injury in adults?: an evidence-based review and
meta-analysis. J Head Trauma Rehabil 2010;25(4):228–40.
37. Nylen K, Ost M, Csajbok LZ, et al. Serum levels of S100B, S100A1B and
S100BB are all related to outcome after severe traumatic brain injury. Acta
Neurochir 2008;150:221–7.
38. Stranjalis G, Korfias S, Psachoulia C, et al. The prognostic value of serum
S-100β protein in spontaneous subarachnoid haemorrhage. Acta Neurochir
2007;149:231–7.
39. Geyer C, Ulrich A, Gräfe G, et al. Diagnostic value of S100β and neuron-
specific enolase in mild pediatric traumatic brain injury. J Neurosurg Pediatr
2009;4(4):339–44.
40. Bechtel K, Frasure S, Marshall C, et al. Relationship of serum S100β levels
and intracranial injury in children with closed head trauma. Pediatrics
2009;124:e697–704.
41. Morochovic R, Rácz O, Kitka M, et al. Serum S100β protein in early
management of patients after mild traumatic brain injury. Eur J Neurol
2009;16(10):1112–7.
42. Green A, Harvey R, Thompson E, et al. Increased S100 beta in the
cerebrospinal fluid of patients with frontotemporal dementia. Neurosci Lett
1997;235:5–8.
43. Otto M, Wiltfang J, Schutz E, et al. Diagnosis of Creutzfeldt-Jakob disease by
measurement of S100 protein in serum: prospective case-control study. BMJ
1998;316:577–82.
44. Anderson RE, Hansson LO, Nilsson O, et al. High serum S100β levels for
trauma patients without head injuries. Neurosurgery 2001;48:1255–58.
45. Savola O, Pyhtinen J, Leino TK, et al. Effects of head and extracranial
injuries on serum protein S100B levels in trauma patients. J Trauma
2004;56:1229–34.
46. Beems T, Simons KS, van Geel WJA, et al. Serum and CSF concentrations
of brain-specific proteins in hydrocephalus. Acta Neurochir 2003;145:
37–43.
47. Piazza O, Storti MP, Cotena S, et al. S100β is not a reliable prognostic index
in paediatric TBI. Pediatr Neurosurg 2007;43:258–64.
48. Bazarian JJ, Zemlan FP, Mookerjee S, et al. Serum S100β and cleaved tau are
poor predictors of long-term outcome after mild traumatic brain injury.
Brain Inj 2006;20:759–65.
49. Muller K, Elverland A, Romner B, et al. Analysis of protein S-100β in serum:
a methodological study. Clin Chem Lab Med 2006;44:1111–4.
50. Vos PE, van Gils M, Beems T, et al. Increased GFAP and S100β but not NSE
serum levels after subarachnoid haemorrhage are associated with clinical
severity. Eur J Neurol 2006;13:632–8.
A complete list of references for this chapter can be found online at
www.expertconsult.com.
 Section II—Clinical and Laboratory Assessment 175.e1
References
1. Goldstein M. Traumatic brain injury: a silent epidemic. Ann Neurol
1990;27:327.
2. Roche GW, Kanchuger M, Mangano CM, et al. Adverse cerebral outcomes
after coronary bypass surgery. N Engl J Med 1996;335:1857–64.
3. Van Dijk D, Jansen WE, Hijman R, et al. Cognitive outcome after off-pump
and on-pump coronary artery bypass graft surgery: a randomized trial.
JAMA 2002;287:1405–12.
4. Williams GR, Jiang JG, Matchar DB, et al. Incidence and occurrence of total
(first-ever and recurrent) stroke. Stroke 1999;30:2523–8.
5. Teasdale G, Teasdale E, Hadley D. Computed tomographic and magnetic
resonance imaging classification of head injury. J Neurotrauma 1992;
9(Suppl. 1):S249–57.
6. Guadagno JV, Calautti C, Baron JC. Progress in imaging stroke: emerging
clinical applications. Br Med Bull 2003;65:145–57.
7. Zhang Z, Larner SF, Kobeissy F, et al. Systems biology and theranostic
approach to drug discovery and development to treat traumatic brain
injury. Methods Mol Biol 2010;662:317–29.
8. Wagner AK. TBI translational rehabilitation research in the 21st century:
exploring a Rehabilomics research model. Eur J Phys Rehabil Med 2010;
46(4):549–56.
9. Pasinetti GM, Fivecoat H, Ho L. Personalized medicine in traumatic brain
injury. Psychiatr Clin North Am 2010;33(4):905–13.
10. Manley GT, Diaz-Arrastia R, Brophy M, et al. Common data elements
for traumatic brain injury: recommendations from the biospecimens and
biomarkers working group. Arch Phys Med Rehabil 2010;91(11):1667–72.
11. Oli MW, Hayes RL, Robinson G, et al. Traumatic brain injury biomarkers:
from pipeline to diagnostic assay development. Methods Mol Biol
2009;566:293–302.
12. Afinowi R, Tisdall M, Keir G, et al. Improving the recovery of S100β
protein in cerebral microdialysis: implications for multimodal monitoring
in neurocritical care. J Neurosci Methods 2009;181(1):95–9. Epub 2009
Mar 14.
13. Petzold A, Tisdall MM, Girbes AR, et al. In vivo monitoring of neuronal
loss in traumatic brain injury: a microdialysis study. Brain 2011;134(Pt 2):
464–83.
14. Marklund N, Blennow K, Zetterberg H, et al. Monitoring of brain
interstitial total tau and beta amyloid proteins by microdialysis in patients
with traumatic brain injury. J Neurosurg 2009;110(6):1227–37.
15. Qureshi AI. Evaluation of brain injury using a blood test: is there a role in
clinical practice? Crit Care Med 2002;30:2778–9.
16. Ingebrigtsen T, Romner B. Biochemical serum markers of traumatic brain
injury. J Trauma 2002;52:798–808.
17. Betz A, Dietrich W. Blood-brain barrier dysfunction in cerebral ischemia.
In: Ginsberg MD, Bogousslavsky J, editors. Cerebrovascular disease. Oxford,
England: Blackwell; 1998. p. 358–70.
18. Del Zoppo GJ, Hallenbeck JM. Advances in the vascular pathophysiology of
ischemic stroke. Thromb Res 2000;98:73–81.
19. Wiltfang J, Otto M, Baxter HC, et al. Isoform pattern of 14-3-3 proteins in
the cerebrospinal fluid of patients with Creutzfeldt-Jakob disease.
J Neurochem 1997;73:2485–90.
20. Green AJE. Cerebrospinal fluid brain-derived proteins in the diagnosis of
Alzheimer’s disease and Creutzfeldt-Jakob disease. Neuropathol Appl
Neurobiol 2002;28:427–40.
21. Donato R. S100: a multigenic family of calcium-modulated proteins of the
EF-hand type with intracellular and extracellular functional roles. Int J
Biochem Cell Biol 2001;33:637–68.
22. McKeating E, Andrews P, Mascia L. Relationship of neuron specific enolase
and S100 concentrations in systemic and jugular venous system to injury
severity and outcome after traumatic brain injury. Acta Neurochir 1998;
71:117–9.
23. Blomquist S, Johnsson P, Luhrs C, et al. The appearance of S100 protein in
serum during and immediately after cardiopulmonary bypass surgery:
a possible marker for cerebral injury. J Cardiotharac Vasc Anesth 1997;11:
699–703.
24. Rosen H. Increased serum levels of the S-100 protein are associated with
hypoxic brain damage after cardiac arrest. Stroke 1998;29:473–7.
25. Aurell A, Rosengren, I, Karlsson B, et al. Determination of S-100 and glial
fibrillary acidic protein concentration in cerebrospinal fluid after brain
infarction. Stroke 1991;22:1254–8.
26. Mussack T, Biberthaler P, Kanz K-G, et al. Serum S100β and interleukin-8
as predictive markers for comparative neurologic outcome analysis of
patients after cardiac arrest and severe traumatic brain injury. Crit Care
Med 2002;30:2669–74.
27. Wunderlich MT, Ebert AD, Kratz T, et al. Early neurobehavioral outcome
after stroke is related to release of neurobiochemical markers of brain
damage. Stroke 1999;30:1190–5.
28. Stein DM, Kufera JA, Lindell A, et al. Association of CSF biomarkers and
secondary insults following severe traumatic brain injury. Neurocrit Care
2011;14(2):200–7.
29. Vos PE, Jacobs B, Andriessen TM, et al. GFAP and S100β are biomarkers of
traumatic brain injury: an observational cohort study. Neurology 2010;
75(20):1786–93.
30. Wojtczak-Soska K, Lelonek M. S-100β protein: an early prognostic marker
after cardiac arrest. Cardiol J 2010;17(5):532–6.
31. Murillo-Cabezas F, Muñoz-Sánchez MA, Rincón-Ferrari MD, et al. The
prognostic value of the temporal course of S100beta protein in post-acute
severe brain injury: a prospective and observational study. Brain Inj 2010;
24(4):609–19.
32. Cakir Z, Aslan S, Umudum Z, et al. S-100beta and neuron-specific enolase
levels in carbon monoxide-related brain injury. Am J Emerg Med 2010;
28(1):61–7.
33. James ML, Blessing R, Phillips-Bute BG, et al. S100β and brain natriuretic
peptide predict functional neurological outcome after intracerebral
haemorrhage. Biomarkers 2009;14(6):388–94.
34. Filippidis AS, Papadopoulos DC, Kapsalaki EZ, et al. Role of the S100β
serum biomarker in the treatment of children suffering from mild
traumatic brain injury. Neurosurg Focus 2010;29(5):E2.
35. Hallén M, Karlsson M, Carlhed R, et al. S-100β in serum and urine after
traumatic head injury in children. J Trauma 2010;69(2):284–9.
36. Undén J, Romner B. Can low serum levels of S100β predict normal CT
findings after minor head injury in adults?: an evidence-based review and
meta-analysis. J Head Trauma Rehabil 2010;25(4):228–40.
37. Nylen K, Ost M, Csajbok LZ, et al. Serum levels of S100B, S100A1B and
S100BB are all related to outcome after severe traumatic brain injury. Acta
Neurochir 2008;150:221–7.
38. Stranjalis G, Korfias S, Psachoulia C, et al. The prognostic value of serum
S-100β protein in spontaneous subarachnoid haemorrhage. Acta Neurochir
2007;149:231–7.
39. Geyer C, Ulrich A, Gräfe G, et al. Diagnostic value of S100β and
neuron-specific enolase in mild pediatric traumatic brain injury.
J Neurosurg Pediatr 2009;4(4):339–44.
40. Bechtel K, Frasure S, Marshall C, et al. Relationship of serum S100β levels
and intracranial injury in children with closed head trauma. Pediatrics
2009;124:e697–704.
41. Morochovic R, Rácz O, Kitka M, et al. Serum S100β protein in early
management of patients after mild traumatic brain injury. Eur J Neurol
2009;16(10):1112–7.
42. Green A, Harvey R, Thompson E, et al. Increased S100 beta in the
cerebrospinal fluid of patients with frontotemporal dementia. Neurosci Lett
1997;235:5–8.
43. Otto M, Wiltfang J, Schutz E, et al. Diagnosis of Creutzfeldt-Jakob disease
by measurement of S100 protein in serum: prospective case-control study.
BMJ 1998;316:577–82.
44. Anderson RE, Hansson LO, Nilsson O, et al. High serum S100β levels
for trauma patients without head injuries. Neurosurgery 2001;48:
1255–58.
45. Savola O, Pyhtinen J, Leino TK, et al. Effects of head and extracranial
injuries on serum protein S100β levels in trauma patients. J Trauma
2004;56:1229–34.
46. Beems T, Simons KS, van Geel WJA, et al. Serum and CSF concentrations
of brain-specific proteins in hydrocephalus. Acta Neurochir 2003;145:
37–43.
47. Piazza O, Storti MP, Cotena S, et al. S100β is not a reliable prognostic index
in paediatric TBI. Pediatr Neurosurg 2007;43:258–64.
48. Bazarian JJ, Zemlan FP, Mookerjee S, et al. Serum S100β and cleaved tau
are poor predictors of long-term outcome after mild traumatic brain
injury. Brain Inj 2006;20:759–65.
49. Muller K, Elverland A, Romner B, et al. Analysis of protein S-100β in
serum: a methodological study. Clin Chem Lab Med 2006;44:1111–4.
50. Vos PE, van Gils M, Beems T, et al. Increased GFAP and S100beta but not
NSE serum levels after subarachnoid haemorrhage are associated with
clinical severity. Eur J Neurol 2006;13:632–8.
51. Nylen K, Csajbok LZ, Ost M, et al. Serum glial fibrillary acidic protein is
related to focal brain injury and outcome after aneurysmal subarachnoid
hemorrhage. Stroke 2007;38:1489–94.
52. Vos PE, Lamers KJ, Hendriks JC, et al. Glial and neuronal proteins in serum
predict outcome after severe traumatic brain injury. Neurology 2004;62:
303–10.
175.e2 Section II—Clinical and Laboratory Assessment
53. Honda M, Tsuruta R, Kaneko T, et al. Serum glial fibrillary acidic protein is
a highly specific biomarker for traumatic brain injury in humans compared
with S-100β and neuron-specific enolase. J Trauma 2010;69(1):104–9.
54. Berger RP, Beers SR, Richichi R, et al. Serum biomarker concentrations and
outcome after pediatric traumatic brain injury. J Neurotrauma
2007;24:1793–801.
55. Fraser DD, Close TE, Rose KL, et al. Severe traumatic brain injury in
children elevates glial fibrillary acidic protein in cerebrospinal fluid and
serum. Pediatr Crit Care Med 2011;12(3):319–24.
56. Bembea MM, Savage W, Strouse JJ, et al. Glial fibrillary acidic protein as a
brain injury biomarker in children undergoing extracorporeal membrane
oxygenation. Pediatr Crit Care Med 2010; Nov 4. Epub ahead of print.
57. Zemlan FP, Rosenberg WS, Luebbe PA, et al. Quantification of axonal
damage in traumatic brain injury: affinity purification and characterization
of cerebrospinal fluid tau levels. J Neurochem 1999;72:741–50.
58. Shaw GJ, Jauch EC, Zemlan FP. Serum cleaved tau protein levels and
clinical outcome in adult patients with closed head injury. Ann Emerg Med
2002;39:254–57.
59. Liliang PC, Liang CL, Weng HC, et al. Tau proteins in serum predict
outcome after severe traumatic brain injury. J Surg Res 2010;160(2):302–7.
Epub 2009 Jan 10.
60. Zingler VC, Krumm B, Bertsch T, et al. Early prediction of neurological
outcome after cardiopulmonary resuscitation: a multimodal approach
combining neurobiochemical and electrophysiological investigations may
provide high prognostic certainty in patients after cardiac arrest. Eur
Neurol 2003;49:79–84.
61. Reisinger J, Hollinger K, Lang W, et al. Prediction of neurological outcome
after cardiopulmonary resuscitation by serial determination of serum
neuron-specific enolase. Eur Heart J 2007;28:52–8.
62. Shore PM, Berger RP, Varma S, et al. Cerebrospinal fluid biomarkers versus
Glasgow Coma Scale and Glasgow Outcome Scale in pediatric traumatic
brain injury: the role of young age and inflicted injury. J Neurotrauma
2007;24:75–86.
63. Herrmann M, Jost S, Kutz S, et al. Temporal profile of release of
neurobiochemical markers of brain damage after traumatic brain injury is
associated with intracranial pathology as demonstrated in cranial
computerized tomography. J Neurotrauma 2000;17:113–22.
64. Graham MR, Myers T, Evans P, et al. Direct hits to the head during amateur
boxing is associated with a rise in serum biomarkers for brain injury. Int J
Immunopathol Pharmacol 2011;24(1):119–25.
65. Zetterberg H, Tanriverdi F, Unluhizarci K, et al. Sustained release of
neuron-specific enolase to serum in amateur boxers. Brain Inj 2009;
23(9):723–6.
66. Petzold A, Keir G, Kay A, et al. Axonal damage and outcome in
subarachnoid haemorrhage. J Neurol Neurosurg Psychiatry 2006;77:753–9.
67. Lewis SB, Wolper RA, Miralia L, et al. Detection of phosphorylated NF-H
in the cerebrospinal fluid and blood of aneurysmal subarachnoid
hemorrhage patients. J Cereb Blood Flow Metab 2008;28:1261–71.
68. Van Geel WJ, Rosengren LE, Verbeek MM. An enzyme immunoassay to
quantify neurofilament light chain in cerebrospinal fluid. J Immunol
Methods 2005;296:179–85.
69. Rosen H, Karlsson JE, Rosengren L. CSF levels of neurofilament is a
valuable predictor of long-term outcome after cardiac arrest. J Neurol Sci
2004;221:19–24.
70. Nylen K, Csajbok LZ, Ost M, et al. CSF neurofilament correlates with
outcome after aneurysmal subarachnoid hemorrhage. Neurosci Lett 2006;
404:132–6.
71. Zanier ER, Refai D, Zipfel GJ, et al. Neurofilament light chain levels in
ventricular cerebrospinal fluid after acute aneurysmal subarachnoid
haemorrhage. J Neurol Neurosurg Psychiatry 2011;82(2):157–9.
72. Sellner J, Petzold A, Sadikovic S, et al. The value of the serum
neurofilament protein heavy chain as a biomarker for peri-operative brain
injury after carotid endarterectomy. Neurochem Res 2009;34(11):1969–74.
Epub 2009 May 9.
73. Glenney Jr JR, Glenney P. Fodrin is the general spectrin-like protein found
in most cells whereas spectrin and the TW protein have a restricted
distribution. Cell 1983;34:503–12.
74. Mondello S, Robicsek SA, Gabrielli A, et al. αII-Spectrin breakdown
products (SBDPs): diagnosis and outcome in severe traumatic brain injury
patients. J Neurotrauma 2010;27(7):1203–13.
75. Pineda JA, Lewis SB, Valadka AB, et al. Clinical significance of alphaII-
spectrin breakdown products in cerebrospinal fluid after severe traumatic
brain injury. J Neurotrauma 2007;24:354–66.
76. Lewis SB, Velat GJ, Miralia L, et al. Alpha-II spectrin breakdown products
in aneurysmal subarachnoid hemorrhage: a novel biomarker of proteolytic
injury. J Neurosurg 2007;107:792–6.
77. Siman R, Roberts VL, McNeil E, et al. Biomarker evidence for mild central
nervous system injury after surgically-induced circulation arrest. Brain Res
2008;1213:1–11.
78. Brophy GM, Pineda JA, Papa L, et al. AlphaII-spectrin breakdown product
cerebrospinal fluid exposure metrics suggest differences in cellular injury
mechanisms after severe traumatic brain injury. J Neurotrauma
2009;26(4):471–9.
79. Ning M, Sarracino DA, Kho AT, et al. Proteomic temporal profile of human
brain endothelium after oxidative stress. Stroke 2011;42(1):37–43. Epub
2010 Dec 16.
80. Foerch C, Montaner J, Furie KL, et al. Invited article: searching for oracles?
Blood biomarkers in acute stroke. Neurology 20094;73(5):393–9.
81. Lo TY, Jones PA, Minns RA. Combining coma score and serum biomarker
levels to predict unfavorable outcome following childhood brain trauma.
J Neurotrauma 2010;27(12):2139–45. Epub 2010 Nov 2.
82. Turck N, Vutskits L, Sanchez-Pena P, et al. A multiparameter panel method
for outcome prediction following aneurysmal subarachnoid hemorrhage.
Intensive Care Med 2010;36(1):107–15. Epub 2009 Sep 17.
83. Springer JE. Apoptotic cell death following traumatic injury to the central
nervous system. J Biochem Mol Biol 2002;35:94–105.
84. Ferrer I. Apoptosis: future targets for neuroprotective strategies.
Cerebrovasc Dis 2006;21 (Suppl. 2):9–20.
85. Yang F, Sun X, Beech W, et al. Antibody to caspase-cleaved actin detects
apoptosis in differentiated neuroblastoma and plaque-associated
neurons and microglia in Alzheimer’s disease. Am J Pathol 1998;152:
379–89.
86. Siman R, Bozyczko-Coyne D, Meyer SL, et al. Immunolocalization of
caspase proteolysis in situ: evidence for widespread caspase-mediated
apoptosis of neurons and glia in the postnatal rat brain. Neuroscience
1999;92:1425–42.
87. Zhang C, Siman R, Xu YA, et al. Comparison of calpain and caspase
activities in adult rat brain after transient forebrain ischemia. Neurobiol Dis
2002;10:289–305.
88. Yatsiv I, Morganti-Kossmann MC, Perez D, et al. Elevated intracranial IL-18
in humans and mice after traumatic brain injury and evidence of
neuroprotective effects of IL-18 binding protein after experimental closed
head injury. J Cereb Blood Flow Metab 2002;22:971–8.
89. Hayakata T, Shiozaki T, Tasaki O, et al. Changes in CSF S100beta and
cytokine concentrations in early-phase severe traumatic brain injury. Shock
2004;22:102–7.
90. Stein DM, Lindell A, Murdock KR, et al. Relationship of serum and
cerebrospinal fluid biomarkers with intracranial hypertension and cerebral
hypoperfusion after severe traumatic brain injury. J Trauma 2011;70(5):
1096–103.
91. Roberts-Lewis JM, Siman R. Spectrin proteolysis in the hippocampus:
a biochemical marker for neuronal injury and neuroprotection. Ann N Y
Acad Sci 1993;679:78–86.
92. Siman R, Bozyczko-Coyne D, Savage MJ, et al. The calcium-activated
protease calpain I and ischemia-induced neurodegeneration. Adv Neurol
1996;71:167–74.
93. Wang KKW. Calpain and caspase: can you tell the difference? Trends
Neurosci 2000;23:20–6.
94. Saatman KE, Creed J, Raghupathi R. Calpain as a therapeutic target in
traumatic brain injury. Neurotherapeutics 2010;7(1):31–42.
95. Siman R, Reaume AG, Savage MJ, et al. Presenilin-1 P264L knock-in
mutation: differential effects on Abeta production, amyloid deposition, and
neuronal vulnerability. J Neurosci 2000;20:8717–26.
96. Neumar RW, Xu YA, Gada H, et al. Cross-talk between calpain and caspase
proteolytic systems during neuronal apoptosis. J Biol Chem 2003;278:
14162–67.
97. Siman R, McIntosh TK, Soltesz KM, et al. Proteins released from
degenerating neurons are surrogate markers for acute brain damage in the
rat. Neurobiol Dis 2004;16:311–20.
98. Saido TC, Yokota M, Nagao S, et al. Spatial resolution of fodrin proteolysis
in postischemic brain. J Biol Chem 1993;268:25239–43.
99. Roberts-Lewis JM, Savage MJ, Marcy VR, et al. Immunolocalization of
calpain I-mediated spectrin degradation to vulnerable neurons in the
ischemic gerbil brain. J Neurosci 1994;14:3934–44.
100. Saatman KE, Bozyczko-Coyne D, Marcy V, et al. Prolonged calpain-
mediated spectrin breakdown occurs regionally following experimental
brain injury in the rat. J Neuropathol Exp Neurol 1996;55:850–60.
101. Yamashima T, Saido TC, Takita M, et al. Transient brain ischemia provokes
Ca2+, PIP2 and calpain responses prior to delayed neuronal death in
monkeys. Eur J Neurosci 1996;8:1932–44.
102. Aebersold R, Mann M. Mass spectrometry-based proteomics. Nature
2003;422:198–207.
 Section II—Clinical and Laboratory Assessment 175.e3
103. McDonald WH, Yates 3rd JR. Shotgun proteomics and biomarker discovery.
Dis Markers 2002;18:99–105.
104. Siman R, Zhang C, Roberts VL, et al. Novel surrogate markers for acute
brain damage: cerebrospinal fluid levels correlate with severity of ischemic
neurodegeneration in the rat. J Cereb Blood Flow Metab 2005;25:1433–44.
105. Siman R, Toraskar N, Dang A, et al. A panel of neuron-enriched proteins as
markers for traumatic brain injury in humans. J Neurotrauma
2009;26(11):1867–77.
106. De Graan PN, Oestreicher AB, Schrama LH, et al. Phosphoprotein B-50
(GAP-43): localization and function. Prog Brain Res 1986;69:37–50.
107. Hemmer W, Wallimann T. Functional aspects of creatine kinase in brain.
Dev Neurosci 1993;15:249–60.
108. Aitken A, Jones D, Soneji Y, et al. 14-3-3 Proteins: biological function and
domain structure. Biochem Soc Trans 2003;23:605–11.
109. Nacher J, Rosell DR, McEwen BS. Widespread expression of rat collapsin
response-mediated protein 4 in the telencephalon and other areas of the
adult rat central nervous system. J Comp Neurol 2000; 424:628–39.
110. Dahmen N, Jungblut PR, Fischer V, et al. Detection of TOAD-64 in adult
rat brain as revealed by two-dimensional protein gel electrophoresis
followed by MALDI mass spectrometry: possible modulatory effect of
chronic haloperidol treatment. Prog Neuropsychopharmacol Biol Psychiatry
2002;26:693–8.
111. Pike BR, Zhao X, Newcomb JK, et al. Temporal relationships between de
novo protein synthesis, calpain and caspase 3-like protease activation, and
DNA fragmentation during apoptosis in septo-hippocampal cultures.
J Neurosci Res 1998;52:505–20.
112. Yakovlev AG, Knoblach SM, Fan L, et al. Activation of CPP32-like caspases
contributes to neuronal apoptosis and neurological dysfunction after
traumatic brain injury. J Neurosci 1997;17:7415–24.
113. Raghupathi R, Graham DI, McIntosh TK. Apoptosis after traumatic brain
injury. J Neurotrauma 2000;17:927–38.
114. Smith DH, Okiyama K, Gennarelli TA, et al. Magnesium and ketamine
attenuate cognitive dysfunction following experimental brain injury.
Neurosci Lett 1993;157:211–4.
115. Feldman Z, Gurevitch B, Artru AA, et al. Effect of magnesium given 1 hour
after head trauma on brain edema and neurological outcome. J Neurosurg
1996;85:131–7.
116. Bareyre FM, Saatman KE, Raghupathi R, et al. Postinjury treatment with
magnesium chloride attenuates cortical damage after traumatic brain injury
in rats. J Neurotrauma 2000;17:1029–39.
117. Browne KD, Leoni MJ, Iwata A, et al. Acute treatment with MgSO4
attenuates long-term hippocampal tissue loss after brain trauma in the rat.
J Neurosci Res 2004;77:878–83.
118. Siman R, Giovannone N, Toraskar N, et al. Evidence that a panel of
neurodegeneration biomarkers predicts vasospasm, infarction, and outcome
in aneurysmal subarachnoid hemorrhage. PLoS One 2011;12:e28938.
119. Baker AJ, Rhind SG, Morrison LJ, et al. Resuscitation with hypertonic
saline-dextran reduces serum biomarker levels and correlates with outcome
in severe traumatic brain injury patients. J Neurotrauma
2009;26(8):1227–40.
120. Ost M, Nylen K, Csajbok L, et al. Initial CSF total tau correlates with 1 year
outcome in patients with traumatic brain injury. Neurology
2006;67:1600–4.
121. Grubb NR, Simpson C, Sherwood RA, et al. Prediction of cognitive
dysfunction after resuscitation from out-of-hospital cardiac arrest using
serum neuron-specific enolase and protein S-100. Heart 2007;93:1268–73.
122. Wunderlich MT, Wallesch CS, Goertler M. Release of neurobiochemical
markers of brain damage is related to the neurovascular status on
admission and the site of arterial occlusion in acute ischemic stroke.
J Neurol Sci 2004;227:49–53.
123. Bitsch A, Horn C, Kemmling Y, et al. Serum tau protein level as a marker of
axonal damage in acute ischemic stroke. Eur Neurol 2002;47:45–51.
II
Chapter
19  
Volume Status and
Cardiac Function
Jose L. Pascual, Jiri Horak, Vicente H. Gracias, and Patrick J. Neligan
Introduction
The injured brain is exquisitely sensitive to global changes in
blood flow and oxygen delivery. Brain injury rarely occurs in
the absence of systemic injury, either as a primary (trauma)
or secondary process (subarachnoid hemorrhage). Hence,
volume and circulatory abnormalities may induce or worsen
brain injuries, and brain injuries may induce circulatory
abnormalities. In the setting of cerebral autoregulatory dys-
function, large swings in cardiac output or peripheral resis-
tance may have an adverse effect on brain tissue perfusion.
Furthermore, manipulation of volume status and blood pres-
sure is central to the management of conditions such as sub-
arachnoid hemorrhage (SAH) and traumatic brain injury
(TBI) to maintain cerebral perfusion pressure (CPP) or
prevent delayed cerebral ischemia (DCI). Whereas neurocriti-
cal care focuses on neurologic monitoring, cardiovascular
monitoring is performed on all neurologically injured patients
and the data derived are central to all therapeutic decisions.
However, there is no single set of optimal hemodynamic or
cardiac values that applies to all patients. For example, an
adequate blood pressure to avoid cerebral hypoperfusion
likely is different in a young individual with no preexisting
disease compared with an older patient who has hypertension
and atherosclerotic disease. Consequently, an understanding
of the methods of cardiovascular monitoring, the principles
behind them, and interpretation of hemodynamic data is a
core competence for the neurointensivist. This chapter dis-
cusses techniques to monitor volume status and cardiac func-
tion. The evaluation of cardiac ischemia, acute coronary
syndromes, and use of biomarkers in these disorders is beyond
the scope of this chapter, and the reader is referred to recent
reviews on these topics.1,2
Physical Examination and
Laboratory Determination
of Effective Circulating Volume
Evaluation of volume status, preload, and fluid responsiveness
is important in neurocritical care because hypotension is a
well-known secondary cerebral insult. In addition, volume
176
expansion or vasopressors frequently are used to augment
cerebral blood flow (CBF). However, positive fluid balance or
induced hypertension also can have adverse effects on outcome
or intracranial physiology.3-5 Hence, an understanding of a
patient’s volume status is fundamental to his or her care. Acute
and chronic intravascular fluid deficits may manifest with
varying degrees of clinical findings. Patients with chronic
volume deficits may present with a decreased skin turgor,
weight loss, sunken eyes, hypothermia, oliguria, orthostatic
hypotension, and chronic low-grade tachycardia. On the other
hand, those with acute volume deficits may present clinically
with changes in vital signs, with tachycardia and hypotension,
oliguria, and anuria, but also with more subtle changes in
mental status, such as restlessness and confusion. Clinical
signs of volume excess include weight gain, pulmonary con-
gestion or edema, and peripheral edema.
Basic laboratory tests should be obtained to help determine
effective circulating volume. An increase in all plasma electro-
lyte concentrations may indicate volume loss, but creatinine
and blood urea nitrogen (BUN) elevations may be more spe-
cific. In particular, a BUN-to-creatinine ratio greater than 15
may be more sensitive to indicate an intravascular fluid deficit.
In systemic volume loss, renal sodium retention results in
urinary sodium concentration (UNa) that typically drops
below 20 mEq/L. The fractional excretion of sodium (FENa)
is considered more sensitive and can be calculated by the
formula
UNa × PCr/PNa × UCr ×100%
where U and P are urine and plasma measured concentra-
tions of sodium and creatinine. A FENa less than 1% indi-
cates a reduced effective circulating volume seen by the
kidney provided no diuretics have been administered in the
previous 24 hours.
Although these clinical and laboratory evaluations usually
are adequate to monitor the surgical or medical patient in the
outpatient or ward setting, they may often be insufficient and
inconclusive in the critically ill patient. Instead, more sophis-
ticated, continuous and real-time methods of hemodynamic
cardiovascular monitoring become necessary to help deter-
mine adequate global tissue perfusion.
© Copyright 2013 Elsevier Inc. All rights reserved.
 Section II—Clinical and Laboratory Assessment 177
Blood Pressure Monitoring
Blood pressure, universally measured in intensive care units
(ICUs), reflects myriad factors that maintain the systemic cir-
culation. These include, but are not limited to, circulating
blood volume (both stressed and unstressed) venous return,
splanchnic venular and arteriolar tone, vascular resistance
within several microcirculations, heart rate, cardiac compli-
ance, right ventricular afterload, airway pressure, left ventricu-
lar contractility, core body temperature, baroreceptor reflexes
and adrenal medullary output. Distilling these factors into a
single figure—the mean arterial pressure—to guide therapy is
fraught with hazard. In simple terms, blood pressure reflects
stroke volume and peripheral resistance. However, compensa-
tory reflexes in one of these parameters may balance variabil-
ity in the other. An example is vasoconstriction in the setting
of hypovolemia. The result may be misleadingly “normal”
blood pressure despite severe systemic disease. Convention-
ally, blood pressure is measured two ways: noninvasively, using
manual techniques or oscillometry, and invasively, using direct
arterial cannulation.
Noninvasive
Manual Intermittent Techniques
To measure both systolic and diastolic arterial pressure, the
most widely used intermittent manual method is the ausculta-
tory technique, originally described by Scipione et al.6 but
popularized by Korotkoff in 1905. Using a sphygmomanom-
eter, cuff, and stethoscope, Korotkoff measured blood pressure
by auscultation of the sounds generated by arterial blood flow.
These sounds are a complex series of audible frequencies pro-
duced by turbulent flow, instability of the arterial wall, and
shock wave formation as external occluding pressure on the
artery is reduced. The pressure at which the first Korotkoff
sound is heard generally is accepted as systolic pressure (phase
I). The sound character progressively changes (phases II and
III), becomes muffled (phase IV), and is finally absent (phase
V). Diastolic pressure is recorded at phase IV or V. However,
phase V may never occur in certain pathophysiologic states
such as aortic regurgitation. In 1905, Harvey Cushing7 recom-
mended the routine use of the noninvasive mercury sphygmo-
manometer in intraoperative management of patients.
The sphygmomanometer method has several shortcom-
ings. In particular, it is important that measurements are con-
sistent from observer to observer in critical care. However, if
the cuff is released too rapidly or if the observer fails to iden-
tify sounds, then incorrect measurements may be recorded.
The sounds may be particularly difficult to hear in shock states
or with tachyarrhythmias, making inherent interobserver vari-
ability relevant in this situation. Moreover, if the patient is
shivering or edematous, the blood pressure may appear higher
than normal. A slight bend in the limb on which the cuff is
placed also may lead to falsely elevated recordings, whereas
leaks in the pneumatic system may lead to falsely lower record-
ings. Finally, cuff size and fit are also of critical importance.
The width of the cuff should be 20% greater than arm diam-
eter. If the cuff is too narrow or too short the blood pressure
will be overestimated. Many of these problems also are associ-
ated with continuous blood pressure measurement using
oscillometry.
Oscillometry
Many limitations of manual intermittent blood pressure mea-
surement have been overcome by automated noninvasive
blood pressure (NIBP) devices. NIBP devices measure mean
arterial pressure and use an algorithm to estimate the systolic
and diastolic blood pressure. The major advantages of auto-
mated blood pressure monitoring are reliability, reproduc-
ibility, and built-in low- and high-pressure alarms. In addition,
in modern devices data are transferred directly into the
monitor or information system, freeing the bedside caregiver
for other duties. This may result in more frequent sampling
but perhaps less attentiveness to the results.
Most automated NIBP devices are based on oscillometry.
This monitoring system consists of pneumatic and electronic
components. Modern oscillometers use a single cuff to
occlude the artery and sense the arterial signal. The cuff is
inflated above systolic pressure and then deflates either con-
tinuously or in a step-wise manner. As the cuff pressure
decreases, at a rate about 4 mm Hg per second, below occlu-
sive pressure, blood starts flowing through the artery and
causes detectable oscillation. The pulse pressure wave and the
gauge pressure in the occluding cuff are detected and con-
verted into an electronic signal by a transducer. The pressure
at which the peak amplitude of arterial pulsations occurs cor-
responds closely to directly measured mean arterial pressure
(MAP), and values for systolic and diastolic pressure are
derived from proprietary formulas that examine the rate of
change of the pressure pulsations. A microprocessor uses an
algorithm to determine systolic, diastolic, and mean arterial
pressures. Systolic pressure is identified as the pressure at
which pulsations are increasing and are at 25% to 50% of
maximum. Diastolic pressure is the most difficult value to
determine by oscillometry and is commonly recorded when
the pulse amplitude has declined from the peak value by
80%.8 Oscillometers are generally safe and reliable; however,
similar to manual intermittent techniques, care must be taken
to ensure proper cuff size.
Invasive
Continuous, invasive blood pressure is the gold standard used
to monitor blood pressure in critically ill patients. First
described 60 years ago,9 it is now considered the most
reliable method, especially in hypotensive and vasoconstricted
patients.10,11 The beat-to-beat and continuity of measurement
allow early diagnosis and intervention.
This technique requires an intra-arterial cannula, fluid-
filled tubing, a pressure transducer and a microprocessor with
a display. The pressure transducer is the key component of the
system. It is a low-compliance diaphragm that moves in
response to pressure changes transmitted through fluid filled
tubing. This movement is converted to electrical signals, most
commonly using a strain gauge set-up, based on the principle
that electrical resistance of a wire changes with changes of its
length. There are four wires or strain gauges that form a elec-
trical circuit based on the Wheatstone bridge. In this set-up,
with any movement of the diaphragm two wires are com-
pressed and the two other stretched, thus increasing the
sensitivity to diaphragm movement four times. Modern tech-
nology uses semiconductors or silicone crystals instead of
wires. The MAP is calculated by integrating the area under the
178 Section II—Clinical and Laboratory Assessment
waveform, and the systolic and diastolic pressures calculated
using an algorithm. The arterial pressure waveform is made
of multiple sine waveforms with different frequencies that can
potentially match the natural oscillatory frequency of the
system and resonance and distort the signal. The monitoring
system is designed to keep its frequency above the arterial
waveform frequencies for accurate measurement.
The major problems associated with invasive blood pres-
sure monitoring are dampening and resonance; these can
affect accuracy of the blood pressure waveform and measure-
ment. Dampening decreases the rate of signal change and
so leads to low pulse pressure with low systolic and high
diastolic pressure readings. Arterial cannula occlusion, air
bubbles, and soft tubing are common causes of an over-
dampened signal. Interestingly, MAP accuracy is preserved
because of energy conservation; hence, if the waveform is
dampened, the mean pressure is accurate but the systolic and
diastolic are not.
Strengths, Limitations, and Clinical
Indications for Noninvasive and
Invasive Blood Pressure Monitors
Both invasive and noninvasive blood pressure monitors may
damage local tissue. Although noninvasive automated blood
pressure measurement techniques are considered relatively
safe, repetitive cuff inflation may cause patient discomfort and
persistent arterial occlusion that may result in pain, petechiae
and ecchymoses, limb edema, venous stasis and thrombophle-
bitis, peripheral neuropathy, and rarely compartment syn-
drome (Table 19.1). These events occur more often after
prolonged periods of excessively frequent cuff inflation-
deflation cycling and result from local trauma. Other factors
that may contribute to complications include cuff misplace-
ment across a joint, or repeated attempts to determine blood
pressure in the presence of an artifact-producing condition
such as involuntary muscle tremors.
Most invasive arterial lines are placed in the radial artery
because of easy access and protective collateral flow by the
ulnar artery. This makes digital ischemia an uncommon com-
plication.12 Although other arterial sites are described, the
femoral artery is the most common alternative, preferred by
some over the radial site because it may better reflect central
pressures and its luminal pressure is perhaps less affected by
peripheral vascular disease or arteriosclerosis.13,14 The brachial
and axillary arteries also may be used but have less collateral
circulation to counteract end limb ischemia. The major com-
plication of arterial lines is local thrombosis. In the majority
of arterial lines some degree of nonocclusive thrombosis
occurs without flow compromise. This is seen in more than
Table 19.1  Complications of Noninvasive
Blood Pressure Measurement
Pain
Petechiae and ecchymoses
Limb edema
Venous stasis and thrombophlebitis
Peripheral neuropathy
Compartment syndrome
50% of radial lines, with up to 15% resulting in complete
occlusion.15 The main risk factors for occlusive thrombosis are
catheter size relative to vessel size, the number of arterial
punctures needed to insert the cannula, length of catheter,
indwelling time, presence of underlying atherosclerosis, and
use of vasopressors.16 Arterial catheterization also may result
in arterial injury and lead to local hematoma or pseudoaneu-
rysm formation.17 The risk of pseudoaneurysm formation is
increased with local infection. Infectious complications of
arterial lines are rare but may be related to cannulation site,
and in particular the femoral site, where infection may occur
in up to 18% of cases.18 Therefore all arterial cannulation
insertions should be conducted with full sterile technique.
Because the oscillometric method is not standardized and
measuring algorithms are protected by each manufacturer,
it is difficult to make accurate comparisons between both
methods. Several studies have shown discrepant results
between methods in critically ill patients. Oscillometry gener-
ally underestimates intra-arterial systolic blood pressure,19,20
as opposed to diastolic blood pressure, which is overestimated,
particularly in hypertensive patients.21 MAP measurement is
more reliable.22
The invasive blood pressure monitor must be zeroed and
leveled, in the phlebostatic axis (at the level of the right
atrium), regardless of the site of the arterial cannula. Inade-
quate or incorrect zeroing results in incorrect diagnoses and
inappropriate therapy. The bedside clinician should always
check the “zero” before initiating antihypertensive or vaso-
pressor therapy.
Pulse Pressure and Stroke
Volume Variability
A major reason to use invasive rather than noninvasive blood
pressure monitoring in the ICU is to use respiratory cycle–
varying measurements of stroke volume variability (SVV) and
pulse pressure variation (PPV) during mechanical ventilation
to help predict fluid responsiveness.23 Indeed some studies
suggest that SVV may be the most accurate predictor of fluid
responsiveness, and therapy that is based on SVV appears to
be associated with improved hemodynamic stability.24,25 Arte-
rial contour analysis has been validated against indirect mea-
sures of cardiac output using thermodilution or dye dilution
techniques.26 Positive intrathoracic pressure increases right
ventricle (RV) afterload and reduces RV preload. Hence, on
inspiration, RV stroke volume falls and results in a decrease
in the subsequent left ventricle (LV) stroke volume, which
manifests during the next expiration. In addition, LV stroke
volume may increase slightly during inspiration as a conse-
quence of increased LV filling from enhanced pulmonary
venous return, increased LV compliance because of decreased
RV dimensions, decreased LV afterload, and external pressure
on the LV.27 Thus respiratory variations in arterial pressure
reflect in large part the corresponding cyclic variations in LV
stroke volume. This relationship has been demonstrated in
directly measured (with transesophageal echocardiography
[TEE]) SVV28 or through measurement of arterial PPV, a sur-
rogate of SVV.29 PPV is a good indicator of fluid responsive-
ness as can be demonstrated when plotting stroke volume
against end diastolic volume on a Frank-Starling curve. Varia-
tions are exaggerated when the LV is functioning on the steep
 Section II—Clinical and Laboratory Assessment 179
portion of the curve, whereby small changes in preload induce
large changes in stroke volume.27 Patients on the flat portion
of the curve are insensitive to these cyclic respiratory-related
changes in preload (Fig. 19.1).30
In the mechanically ventilated patient, the effect of respira-
tion on blood pressure can be measured by adjusting the speed
of the arterial pressure to match the end-tidal carbon dioxide
(etCO2) (or airway pressure) trace on the monitor. A 15%
decrease in pulse pressure appears to be a sensitive indicator
of fluid responsiveness.29 PPV or SVV has been shown to
predict fluid responsiveness in mechanically ventilated
patients,31 in cardiac surgery patients,27,32 in septic shock,29,33
with hemorrhage,34 and in elective brain surgery.35 However,
these findings are not universal and other studies have failed
to demonstrate a clear relationship between SVV correlates
and fluid responsiveness.36-38
Unlike bedside echocardiography, arterial pulse contour
analysis does not require an experienced operator and its use
is associated with less patient discomfort. However, the arterial
c
d
a to flow, and pressure divided by flow equals resistance.
Stroke volume
b surrogate for tissue flow. For example, the kidney, an auto-
End diastolic volume
Fig. 19.1  Stroke volume variability (SVV): on the steep part of the Frank-
Starling curve, the SV falls considerably on inspiration, from a to b,
reflecting significant SVV. This is not the case in the “fuller heart,” on
the flat part of the curve, where the fall in SV, from c to d is much less.
Venous side Heart Arterial side
BP 110/70 mm Hg
Normovolemia
SvO2 = 70%
Normal NORMAL VASCULAR TONE
SV = 70 mL
BP 110/70 mm Hg
Hypovolemia
SvO2 = 40%
Low INTENSE VASOCONSTRICTION
SV = 70 mL
pressure contour may not always predict stroke volume
because the pulse contour analysis omits the contributions of
changing vascular tone and vascular bed distribution.39 In
addition, several limitations summarized by the acronym SOS
need to be considered when using SVV: (1) S, small tidal
volume, for example, lung protection strategies and spontane-
ous breathing activity; (2) O, open chest conditions; and (3)
S, sustained cardiac arrhythmia24,40 because all may influence
the accuracy of SVV evaluation of volume status or fluid
responsiveness. Thus respiratory cycle-varying measurements
related to cardiac function may be useful tools in the manage-
ment of neurocritical care unit (NCCU) patients, particularly
those that have large fluctuations in circulating volume (e.g.
patients that have diabetes insipidus or those receiving aggres-
sive osmotherapy with resultant diuresis).
Monitoring Flow and Volume
To maintain homeostasis, organs require blood flow sufficient
to match metabolic demands. There are two components to
flow: volume and pressure. There must be sufficient volume
in the circulation to ensure that the entire organ tissue receives
distribution, and this must be driven by adequate pressure to
ensure its timeliness. So in simple terms, flow is volume
divided by time. In general, perfusion pressure is proportional
However, if only pressure (in this case blood pressure) is mea-
sured, misleading conclusions about flow may arise: high
resistance results in normal pressure but low flow (Fig. 19.2).
Hence, blood pressure alone should be used with caution as a
regulated organ, is highly sensitive to hypoperfusion, which is
manifested as oliguria. Hence, the combination of a “normal”
blood pressure and a urinary output greater than 0.5 mL/kg/
hr suggests adequate blood flow. However, oliguria may be
associated with the stress response, and polyuria can result
from diuretic administration, particularly mannitol.
Evidence of end organ perfusion is helpful if present but
means to measure it are relatively limited in routine care.
What are the alternatives? Tissue blood flow can be measured
Tissues
O2 delivery
1200 mL/min
Fig. 19.2  Impact of vasoconstriction on tissue
blood flow. When there is a decrease in cardiac
output there is intense vaso- and venoconstriction
O2 delivery
that results in apparently normal blood pressure.
500 mL/min
However, tissue blood flow has fallen considerably,
manifested by an increased oxygen-to-extraction
ratio, and a fall in venous oxygen saturation
(SvO2). BP, Blood pressure.
180 Section II—Clinical and Laboratory Assessment
using either specific flow monitors or laboratory analysis of
end products of metabolism. Flow monitors measure stroke
volume and cardiac output or analogous markers of cardiac
function. Laboratory markers include by-products of metabo-
lism, such as lactate, or oxygen consumption using, for
example, mixed venous oxygen saturation (SVO2). Other
monitor types that measure tissue perfusion (such as tissue
oxygen electrodes) or metabolic activity (such as carbon
dioxide: capnometry) also may give information about flow.
Principles of Cardiac
Output Monitoring
Because blood pressure monitoring allows for limited infer-
ence about blood flow, there is a widespread interest in mea-
suring stroke volume and cardiac output (CO). This, at least
theoretically, provides information about circulating volume,
cardiac preload, afterload, and contractility and its response
to volume loading, changes in pressure, and diuresis. This may
be particularly useful in the brain-injured patient in whom
cardiovascular dysfunction may accompany neurologic injury.
Furthermore, the cardiovascular system matches blood flow
to the body’s demands that may vary among patients and
within the same patient over time. Therefore there is no single
“normal” value of CO, and decisions about the adequacy of
CO need to be considered with other variables (e.g., tissue
perfusion and metabolic demand) in mind.
There are two types of cardiac output monitors: those that
use thermodilution or the Fick principle to measure right
ventricular stroke volume and those that analyze arterial
waveforms to calculate CO. The thermodilution method uses
a change in pulmonary (or femoral) arterial pressure to cal-
culate CO, associated with injection of below body tempera-
ture fluid or the use of a heating coil. A thermistor in the artery
generates a temperature change curve, the area under which
is proportional to the CO, which can be calculated. This rep-
resents the gold standard method to measure CO; it is simple
and reproducible. Thermodilution CO is the method used to
measure cardiac output in pulmonary artery (flotation) cath-
eters (PACs, Swan-Ganz catheters) and the PAC-based con-
tinuous thermodilution methods (Vigilance, Edwards Life
Sciences [Irvine, CA]; Opti-Q, ICU Medical [San Clemente,
CA]; and TruCCOMS, AorTech [Rogers, MN]). Less invasive
devices also using thermodilution (e.g., PiCCO [Pulsion
Medical Systems, Irving, TX] and VolumeView [Edwards Life
Sciences, Irvine, CA]), ultrasound (e.g., COstatus [Transonic
Systems Inc., Ithaca, NY] or lithium chloride as an indicator
(e.g., LiDCO [LiDCO Ltd., Cambridge, UK]).41 These devices
use central venous (for calibration) and peripheral arterial
cannulas (sensor) instead of needing pulmonary artery can-
nulation. The PiCCO and VolumeView systems require a
femoral artery catheter. Application of all dilution methods
assumes three major conditions: (1) complete mixing of blood
and indicator, (2) no loss of indicator between place of injec-
tion and place of detection, and (3) constant blood flow. The
errors made are primarily related to the violation of these
conditions. Of the mentioned methods the transpulmonary
indicator dilution methods are, as the so-called continuous
cardiac output thermodilution methods, reasonably accepted
in clinical practice42,43 and have been used in neurocritical
care. For example, Mutoh et al.44 compared PAC CO
measurements to transpulmonary thermodilution using
PiCCO-derived CO in patients with SAH. Pooled analysis of
data showed close agreement between PAC and PiCCO-
derived values of cardiac index.
According to the Fick equation, CO (in liters per minute)
equals oxygen consumption (milliliters of O2 per minute)
divided by the arteriovenous O2 difference (milliliters of O2
per liter of blood). The NiCO (noninvasive cardiac output,
Novametrix) device uses the Fick equation by calculating CO
through changes in capnometry associated with re-breathing
CO2 through a dead space circuit within the breathing system.
Pulse contour analysis is based on the concept that the
contour of the arterial waveform is proportional to the stroke
volume, which can be estimated by the integral of the change
in pressure from end systole to end diastole over time. Most
arterial contour methods relate an arterial pressure or pressure
difference to a flow or volume change (PiCCO [Pulsion],
PulseCO [LiDCO] and Modelflow [TNO/BMI]). Because the
estimated stroke volume is determined by the elastic and resis-
tance patterns of the aorta, the devices are usually calibrated
using transpulmonary thermodilution, typically with a
femoral arterial catheter and a central venous line. Output of
these pulse contour systems is calculated beat-to-beat, but
presentation of the data is typically with a 30-second window.
The most sophisticated arterial waveform analysis tool is
the esophageal Doppler monitor (EDM), which is placed tran-
sorally in the esophagus. The Doppler ultrasound measures
the extent of distension of the thoracic aorta during systole,
from which stroke volume and CO are derived.
Central Venous Pressure:
What It Does and Does Not Do
Central venous catheters (CVCs) are 8- to 20-cm flexible cath-
eters that are placed in a central vein (subclavian, internal
jugular, or femoral), the tips of which lie in the superior vena
cava (SVC), inferior vena cava (IVC), or right atrium. They
are inserted for a variety of reasons: the administration of
vasopressors and hypertonic dextrose solutions (total paren-
teral nutrition), long-term venous access, and measurement
of central venous pressure (CVP) among them. Peripherally
inserted central catheters (PICCs) are longer lines typically
inserted in antecubital veins with their tips also in the SVC.
PICC lines have replaced CVCs for drug and nutrition deliv-
ery, but they are considered unreliable to monitor caval pres-
sures. Nonetheless, even CVP measurements obtained from
CVCs remain a controversial method to estimate effective
circulating volume.45 There is no “normal” CVP, and small
changes in its magnitude may result from large changes in
circulating volume. However, CVP measurement may still be
valuable because it may indicate a change in patient physiol-
ogy, and changes in the waveforms may reflect intrinsic cardiac
abnormalities (Fig. 19.3). CVP is useful as a trend and should
be combined with some form of flow monitoring such as
SVO2, stroke volume, urine output, or mentation. In this para-
digm, the CVP is used to construct Starling curves.
Ventricular performance from volume loading usually
reaches a plateau around a CVP of 10 mm Hg.46 When CVP
is greater than 10 mm Hg, hypervolemia may be the cause,
but other causes such as acute pulmonary hypertension,
arrhythmia, RV dysfunction, cardiac tamponade, increased
 Section II—Clinical and Laboratory Assessment 181
R arterial compliance is neither consistent nor constant and
P T
ECG
tracing
Q CO LiDCO (LiDCO+) (LiDCO Ltd.), PiCCO (Pulsion Medical
mm Hg
S Sciences)—combine peripheral arterial pulse contour analysis
a c
v
y
Jugular x
tracing
Fig. 19.3  Central venous pressure (CVP) waveforms. The central
venous waveform seen on the monitor reflects the events of cardiac
contraction; the central venous catheter transduces slight variations in
pressure that occur in the right atrium during the cardiac cycle and
transmits them as a characteristic waveform.
a wave: Increased atrial pressure during right atrial contraction. It
correlates with the P wave on an electrocardiogram (ECG).
c wave: Early ventricular contraction. The tricuspid valve bulges into
the right atrium. It correlates with the end of the QRS segment on an
ECG.
x descent: Atrial relaxation. It occurs before the T wave on an ECG.
v wave: Atrial filling. The right atrium progressively fills while the
tricuspid valve is closed. It occurs as the T wave is ending on an ECG.
y descent: The tricuspid valve opens in diastole; blood flowing into the
right ventricle and the atrium empties. It occurs before the P wave on
an ECG.
intrathoracic or intra-abdominal pressure (auto–positive
end-expiratory pressure [PEEP], pneumothorax, abdominal
compartment syndrome) should be considered and
evaluated.
When CVP is elevated, interpretation of CVP waveforms
can be useful. It is important to use electrocardiography to
identify or time CVP waveforms. For example, the Y descent
represents the decrease in atrial pressure at the beginning of
diastole. It follows the V wave. The right atrial Y descent usually
is increased in restrictive physiology, pericardial constriction,
volume-overloaded right heart or right heart ischemia, or
infarction. In cardiac tamponade it is small. A large V wave is
seen in tricuspid regurgitation; this may be associated with RV
failure secondary to volume overload or intense hypoxic pul-
monary vasoconstriction. In atrial fibrillation the A wave dis-
appears. In atrioventricular dissociation or junctional rhythm
a tall cannon A wave is produced. In patients with diastolic
dysfunction these arrhythmias can cause sudden hypotension,
and analysis of CVP waveform can be diagnostic.
Devices Used to Monitor Cardiac
Output: Strengths and Weaknesses
Pulse Contour Methods
Systolic ejection results in the propulsion of a stoke volume
into the arterial tree and distention of the aorta and distal
arteries. A characteristic waveform is produced that reflects
the stroke volume and elastic properties of the arterial wall.
The shape of the pulse waveform and the area under the curve
can be mathematically related to the cardiac output. However,
there is great inter- and intrapatient variability. In addition,
aortic valve regurgitation may decrease the accuracy of arterial
pressure waveform analysis. Because compliance is the math-
ematical relationship between pressure and volume, external
calibration of the pressure signal with an alternative cardiac
output technique is required. Pulse contour devices—Pulse
Systems, Munich, Germany), and VolumeView (Edwards Life
to calculate stroke volume and use dilution or thermodilution
cardiac output measurement to calibrate the system. Devices
that analyze pulse waveforms also analyze and display pulse
pressure variability, which may be used to assess dynamic
preload and fluid responsiveness (in mechanically ventilated
patients).
PiCCO and VolumeView calculate cardiac output continu-
ously from pulse contour analysis of the aortic waveform via
femoral arterial cannula using transpulmonary thermodilu-
tion. Both require a CVC (internal jugular or subclavian vein)
to calibrate cardiac output after ice cold fluid injections cen-
trally being sensed by the thermistor in the femoral artery.
The major advantage of this system over a PAC is that there
is no requirement to catheterize the right heart. The PiCCO
device measures the area under the aortic waveform; the sys-
tolic area is identified as that part of the waveform proximal
to the dicrotic notch, and this is proportional to the stroke
volume. Although beat-to-beat volumes are measured, these
are averaged over 30 seconds to avoid inaccuracy associated
with anomalous waveforms, interference, and extrasystoles.
The continued accuracy of these monitors depends on the
frequency of calibration using thermodilution. The manufac-
turer recommends recalibrations every 8 hours, and more
frequently in unstable patients.47
In PiCCO and VolumeView monitors the temperature dif-
ferential detected using the arterial thermistor is composed of
a series of exponential decay curves as the cold injectate passes
through the various compartments of the circulatory system.
Unlike the PAC, the receiver is located at a significant distance
from the injector. Consequently, a series of exponential decay
curves are constructed as the injectate passes along. Because
the injectate is administered centrally and the temperature
difference is measured in a proximal artery, the majority of
the temperature change occurs in the intathoracic compart-
ment. As a consequence, it is possible to obtain an intratho-
racic blood volume index (ITBVI) and extravascular lung
water index (EVLWI). This is a significant advantage to the
PiCCO device because pulmonary edema commonly accom-
panies brain injury from acute lung injury, heart failure, or of
neurogenic origin. Finally, in addition to SVV, the device also
purports to measure global end-diastolic volume, and so
permits the construction of Starling curves with volume titra-
tion. To date, this particular device appears to correlate very
well with thermodilution techniques.26,48-50
Lithium Dilution
Lithium chloride in low doses is a nontoxic substance that is
not metabolized. When injected, its concentration is easily
measured using a lithium selective electrode. Lithium dilution
cardiac output is calculated from the area under the
concentration–time curve when injected from a central line
182 Section II—Clinical and Laboratory Assessment
and measured peripherally. Injection through the antecubital
vein appears to be as accurate as a central line. Pulse CO
LiDCO (LiDCO+) combines pulse contour analysis with
lithium dilution calibration.
The major disadvantage of LiDCO+ is the injection of
lithium and the requirement for calibration of cardiac output
at least every 8 hours. This is not required in PiCCO. In addi-
tion, in patients that are hyponatremic or have recently
received neuromuscular blocking agents, the calibration data
may be inaccurate. Data are unreliable with aortic valve disease
or balloon counterpulsation therapy. The major advantage of
LCO+ is that no specialized central or arterial line is required.
Noncalibrated Pulse Contour   ultrasound to measure aortic blood flow velocity, which allows
Waveform Analysis
FloTrac/Vigileo (F/V, Edwards Lifesciences) calculates con-
tinuous cardiac output (CCO), stroke volume (SV), SVV, and
systemic vascular resistance (SVR) using pulse contour analy-
sis obtained from a single sensor attached to an arterial line
at any peripheral site. Unlike PiCCO or VolumeView, it does
not require external calibration, a central line, or specialized
catheter. Since its introduction in 2006, the device has seen
several software upgrades, with progressive improvement in
accuracy. There were concerns that early software versions
underestimated cardiac output when there was increased arte-
rial compliance and vasodilatory shock,51,52 but third-
generation software was recently validated in septic patients.53,54
F/V combines rapid analysis in real time of the arterial pres-
sure waveform with demographic data (such as sex, age,
weight, and height) applied to an evolving algorithm to cal-
culate cardiac output. Arterial pulsatility is directly propor-
tional to stroke volume. The device appears capable of
correcting for dynamic changes in vascular tone and compli-
ance by analyzing skewness and kurtosis of the arterial wave-
form. These correction variables are updated every 60 seconds
and the arterial waveform is analyzed and averaged over 20
seconds, thus eliminating artifacts, jitter, and premature con-
tractions. Cardiac output is calculated using the arterial wave-
form and the heart rate yielding apparently reliable measures
of SVV and hence fluid responsiveness.
Mayer et al.55 performed a meta-analysis of studies that
have investigated F/V. Except for patients with irregular heart
rhythms and aortic stenosis, cardiac output measurements by
F/V are similar to those obtained using transthoracic echocar-
diography in critically ill patients.56 Earlier studies demon-
strated poor correlation between F/V and thermodilution
methods; with newer software the correlation has improved
and with each software update accuracy continues to improve.57
It should be remembered that thermodilution methods,
although considered the gold standard, are not ideal devices
to compare with F/V; measurement intervals and averaging
times are substantially longer with all thermodilution methods.
Hence it is possible that F/V is more sensitive to dynamic
changes in cardiovascular activity. Conversely, it is likely that
the F/V is severely limited in patients with aortic valve disease,
those with indwelling intra-aortic balloon pumps, those
rewarming from hypothermia, and patients with intracardiac
shunts. It is conceivable that this particular device will become
the dominant product in the market in future because of its
simplicity of placement, lack of observer error, continuity of
data, and the lack of need for secondary vascular access.
Currently, however, F/V is recommended in situations in
which alternative monitors are unavailable. Another similar,
noncalibrated monitor, the MostCare system (Vytech, Padova,
Italy) uses the pressure recording analytical method (PRAM)
to analyze the arterial wave contour peripherally. Similar to
the other systems MostCare uses a proprietary algorithm to
derive cardiac and flow variables from contour analysis.58
Esophageal Doppler
Esophageal Doppler has been widely used in perioperative
medicine (e.g., cardiac surgery), but has made minimal
inroads into ICUs, particularly NCCUs. The thoracic aorta is
located in proximity to the esophagus. The device uses Doppler
derivation of SV and CO. Insertion of the probe is rapid, and
clinically useful derived data are available within seconds. Sub-
stantial published data support the use of esophageal Doppler
to guide goal-directed therapy.59 There are nonetheless several
potential drawbacks: (1) patient positioning is awkward,
(2) there is a steep learning curve and significant interobserver
variability, and (3) the esophageal Doppler, although small in
diameter and pliable, cannot be inserted into nonsedated and
nonintubated patients or patients with known esophageal
disease. This device is unlikely to find a place during neuro-
surgical procedures because the head may be inaccessible
during surgery. To be used in NCCUs, placement problems
and observer error need to be overcome.
Other Devices
NiCO (noninvasive CO [Novametrix; Wallingford, CT]) is a
device that uses partial rebreathing of CO2 to derive cardiac
output in mechanically ventilated patients. The effects of
unknown ventilation or perfusion inequality in certain
patients may explain why the performance of this method
shows a lack of agreement between thermodilution and CO2-
rebreathing cardiac output. Furthermore, it may be unsuitable
for neurologically injured patients because its use may cause
fluctuation in CBF.
COstatus (Transonic Systems Inc., Ithaca, NY) calculates
cardiac output by using ultrasound technology to measure
changes in blood ultrasound velocity and blood flow follow-
ing an injection of warm saline solution. Some studies have
demonstrated that CO values measured with ultrasound indi-
cator dilution techniques correlate well with PAC, and the
COstatus system provides derived variables including total
end diastolic volume index.41,60
Pulse dye densiometry (Nihon Kohden, Japan) is a tech-
nique that uses transpulmonary dye dilution with transcuta-
neous signal detection to measure CO. The dye is injected
intravenously, and although the system is relatively noninva-
sive, supporting data for this technology is limited. This
approach cannot be currently recommended.
Electrical bioimpedance uses constant electrical current
stimulation to identify thoracic impedance variations associ-
ated with blood flow. Bioreactance has evolved from bio-
impedance and measures changes in frequency of electrical
currents across the chest wall. These concepts are noninvasive
and elegant in design, but to date there are few data to support
the use of this approach in critical care.61,62 In addition, electri-
cal interference may limit their potential use in the ICU.
 Section II—Clinical and Laboratory Assessment 183
Right Heart Catheterization
and Pulmonary Artery
Pressure Monitoring
The balloon tipped flow-directed PAC was introduced in the
early 1970s and quickly became widely used in critical care,
under the moniker of its inventors Swan and Ganz.63 The
bedside ability to “float” a catheter into the heart originated
on a serendipitous observation by cardiologist H. J. Swan when
he observed a sailboat and imagined a parachute-like device
catching blood flow and dragging a catheter with it. Swan
subsequently worked with William Ganz to incorporate cardiac
output computation capability using thermodilution into the
catheter itself.64 Despite decades of controversy65,66 about
whether information provided by a PAC can help influence
outcomes in ICU patients, use of the PAC has declined signifi-
cantly, particularly outside of North America. Nevertheless the
PAC remains the gold standard hemodynamic monitor in
clinical use to assess cardiac output, and despite its limitations
most other devices are compared to it.67,68 In addition a PAC
also can provide information about pulmonary artery pres-
sures, right-sided and left-sided filling pressures, and SVO2. In
general, its use should be reserved for more complex patients.
The PAC directly measures intrathoracic, intravascular, and
intracardiac pressures and is able to display their pressure
waveforms. It also provides measurement of CO and SVO2.
Measurement of these hemodynamic indices is continuous
and mostly accurate and clearly superior to the clinical assess-
ment of these indices by physical examination in critically ill
patients.65
The PAC can be inserted through a wide bore introducer in
any central vein, and “floated” through the right atrium and
ventricle into one of the pulmonary arteries using a balloon
attached to the distal part of the catheter. Usually the device
is advanced until the balloon is “wedged” into a proximal
arteriole, at which point flow ceases and the port distal to the
balloon is able to measure the static pressure in the fluid
column that equilibrates across the pulmonary capillaries and
left atrium. Thus because there are no valves between the
pulmonary capillaries and the left atrium, this pulmonary
capillary wedge pressure (PCWP) is a reflection of left atrial
pressure (Fig. 19.4).
However, measured pulmonary artery pressures (PAPs) also
vary depending on the anatomic position of the PAC tip. The
lung is anatomically divided in three West zones. The West
lung zone 1, or the upper, least dependent area or middle area
(zone 2) has alveolar pressures that are greater than PAPs.
Thus in these lung regions blood flow may be partially or
completely interrupted and thus prevent pressure equilibra-
tion across the pulmonary capillary system during wedging.
Only in zone 3 does PAP always surpass alveolar pressure,
making blood flow constant, thus making it the optimal ana-
tomic position for the PAC tip to be placed in and wedged.
In addition, using PCWP as a surrogate of left atrial pressure
assumes a pulmonary capillary and venular resistance of zero.
This assumption is faulty because the capillary accounts for
up to 40% of total pulmonary vascular resistance. In critically
ill patients in whom pulmonary constriction occurs secondary
to hypoxemia, this value may rise to even higher levels. In
conditions such as shock, respiratory failure, sepsis, and
acute respiratory distress syndrome (ARDS), the various
CHARACTERISTIC INTRACARDIAC PRESSURE WAVE FORMS
DURING PASSAGE THROUGH THE HEART
40
mm Hg
RA RV PA PCW
20
0
Fig. 19.4  Placement of a pulmonary artery catheter. The different
pressure waveforms (upper panel) guide the clinician to the location of
the tip of the catheter. PA, pulmonary artery; PCW, pulmonary capillary
wedge; RA, Right atrium; RV, right ventricle.
assumptions may not be valid and so can make the use of
PCWP estimation of preload faulty.69
Nonetheless, a PAC can reliably measure PAPs and cardiac
output using the thermodilution method, and is the only
device that accurately measures true core temperature and
mixed SVO2. Moreover, measurements are independent of
cardiac pathology, particularly of the aortic valve and cardio-
vascular anomalies distal to it. Thus PAPs also may provide
some indication of volume status on their own. Normal sys-
tolic pulmonary pressures mirror RV systolic pressures in the
absence of pulmonic valve disease and normally range between
20 and 30 mm Hg. The normal mean pulmonary artery pres-
sure (mPAP) at rest is 14 plus or minus 3 mm Hg. Elevation
in PAPs may result from hypervolemic states but also from
pulmonary hypertension, LV failure, and mitral valve disease.70
In the correct hands the PAC is an invaluable tool to diagnose
and monitor treatment of pulmonary hypertension and right
heart failure (Fig. 19.5). However, the correct interpretation
of PAC-derived hemodynamic data and the subsequent thera-
peutic intervention remain major challenges and are the main
reasons for complications associated with these devices.71
Many clinicians believe that PCWP reliably reflects preload,
and is useful to construct Starling curves. This is unlikely.72
The pressure-volume relationship of the LV changes dynami-
cally depending on clinical circumstances, and pressure figures
are altered by changes in ventricular and atrial compliance,
ventricular systolic and diastolic function, valvular function,
rhythm and heart rate, afterload, intrathoracic pressures, and
abdominal pressures. They also change with therapeutic inter-
ventions.45,72,73 Furthermore, PAC pressures alone should not
be used to guide volume resuscitation, and the inappropriate
use of these measures may have resulted in the lack of pub-
lished beneficial outcome data associated with this device
to date.
Two types of PAC are available in clinical practice: intermit-
tent bolus (older and less expensive) and continuous cardiac
output devices. The noncontinuous model requires intermit-
tent injection of cooled crystalloid to measure cardiac output
184 Section II—Clinical and Laboratory Assessment

Hemodynamically with a thermistor located downstream within the pulmonary

unstable patient artery. The data are averaged over time (about 10 minutes) to
produce an accurate series of measurements; hence there may
be a delay of several minutes before the device records major
Is the patient hypoxic hemodynamic changes (i.e., this approach may be limited in
or is RV dysfunction its use in a very unstable patient). Further, the data may be
suspected? misleading when there are rapid changes of body temperature
such as following cardiopulmonary bypass.
Yes No PAP and PCWP measurements can provide useful informa-
tion. Dynamic PAP changes reflect changes in patient condi-
Is the patient at risk for Yes Is femoral arterial tion or changes induced by therapeutic intervention (e.g., a
malignant arryhthmias? cannulation safe?
pulmonary vasodilator). In the NCCU the PAC may be useful
No to measure global hemodynamic function and the pulmonary
Yes No
arterial response to vasopressor therapy. CBF depends on LV
Insert a pulmonary Insert PiCCO+ Attach FloTrac/
stroke volume, particularly when there is autoregulatory dys-
artery catheter or Lidco+ Vigileo function. Vasopressors increase both systemic and pulmonary
vascular resistance, and profound pulmonary hypertension
Fig. 19.5  Decision tree for placement of a hemodynamic monitor. may result in RV dysfunction, secondary to excessive afterload.
RV, Right ventricle. Thus in patients who require high doses of vasopressors to
maintain CPP, pulmonary arterial catheterization can be used
to guide both fluid and vasopressor therapy. Finally, the pres-
sure waveforms on the PAC can be used to diagnose cardiac
pathology. For example, a large “v” wave on a PAP tracing may
indicate acute mitral regurgitation.

Bedside Cardiac Imaging

Fig. 19.6  Swan-Ganz catheter (note the lack of visible distal heating coil,
which would be used for the continuous cardiac output [CCO] method).
using the area under the thermodilution curve (Fig. 19.6) and
despite its limitations is still considered the standard reference
for other devices.
Usually three injections are performed, and an average of
the three is recorded. The cold bolus is injected into the right
atrium, and the temperature change in the pulmonary artery
is used to calculate cardiac output. This system is highly accu-
rate, and there is surprisingly little interobserver variability.
SVO2 can be measured but requires a laboratory analysis of
the collected sample. This measurement is also highly accurate
and reproducible.
Continuous CO PACs were introduced to reduce the work-
load of the bedside nurse and provide “continuous data.” This
is misleading because they use a random sequence of tem­
perature changes generated by thermal filament (Vigilance,
Edwards Life Sciences) or thermal coil (OptiQ, ICU Medical,
San Clemente, CA) that is located in the superior vena cava,
The Current Role of Echocardiography  
in Critical Care
Echo reflection, the basis of echocardiography technology was
first described by Lazzaro Spallanzani when evaluating the
effect of sound on navigation practices of bats.74 Langevin
developed sound navigation and ranging (SONAR) based on
the same concept in the search for methods to detect enemy
submarines in World War I.74 In 1946 the French physicist
André Denier suggested that ultrasound could be used to look
at internal organs.75 Inge Edlerm, a Swedish internist, described
the first use of ultrasound on the heart when attempting to
distinguish coexisting mitral regurgitation in stenotic valves.76
After collaborating with local physicist Hellmuth Hertz for
several years, they were able to generate the first moving
images of the heart, which were presented at the Lund Royal
Physiological Society in 1954.77 Multiple incremental improve-
ments in ultrasound techniques have occurred since: two-
dimensional (B-mode), motion (M-mode), Doppler (velocity
of motion), and color Doppler (direction of motion) technol-
ogy. Frazin et al. first introduced transesophageal echocar-
diography (TEE) in 1976.78
In critically ill populations echocardiography now greatly
expands the capabilities of intensivists to accurately and
rapidly diagnose the cause of hemodynamic instability. With
advances in technology, ultrasound units now are smaller,
multipurpose with improved image quality, and less expen-
sive. In the ICU echocardiography is a complementary diag-
nostic tool that, to be fully understood and appropriately used,
requires training and experience, but in recent years the use
of and training in echocardiography by intensivists has
increased79-81 and appropriate use guidelines have been pub-
lished.82 Furthermore, there is a general agreement among
critical care societies worldwide that general critical care ultra-
sound and “basic” critical care echocardiography should be
 Section II—Clinical and Laboratory Assessment 185
part of the curriculum of ICU physicians.83 This is important
because even basic knowledge of bedside echocardiography
may provide life-saving diagnosis in certain conditions (e.g.,
severe heart dysfunction, large pericardial effusion, or severe
hypovolemia).84
The role of echocardiography in neurocritical care is just
beginning to be elucidated, although it is a safe, noninvasive
bedside test that provides information about cardiac struc-
ture, function, and hemodynamics. Different physiologic
conditions and use of vasopressors make critical care echocar-
diography interpretation challenging. Furthermore, most of
the quantitative echocardiographic evaluations have been
validated only in cardiac surgery and the cardiology popula-
tion. In these patients echocardiography plays a key role to
help assess LV and RV function, chamber size, and cardiac
valve dysfunction, and in cardiac surgery and high-risk non-
cardiac surgery, abnormal echocardiographic findings are an
independent predictor of poor outcome.85 Echocardiography
also may be helpful in preload assessment when intravascular
volume determination is difficult. In the ICU echocardio-
graphic evaluation of unexplained hypoxemia, intracardiac
versus intrapulmonary shunt, or diagnosis of cardiac tampon-
ade and aortic dissection is becoming a first-line, bedside
diagnostic tool.
In critical care up to 25% of bedside studies, particularly of
transthoracic echocardiography (TTE), may be of limited
diagnostic quality or nondiagnostic. Contrast echocardiogra-
phy, using microspheres coupled with contrast-specific ultra-
sound imaging modalities, may help overcome this limitation
and so avoid the need for TEE.86 Furthermore, the concept of
limited TTE has evolved in the last few years and this has
greatly facilitated the use of echocardiography to guide goal-
directed therapy in the ICU.87,88 Focused, goal-directed echo-
cardiography (TTE or TEE) usually can be performed in a few
minutes and may include qualitative assessment of LV and RV
function, an estimate of aortic valve gradient, RV systolic pres-
sure, and intravascular volume status among other variables.
Transthoracic Versus Transesophageal
Echocardiography
TTE involves the placement of an ultrasound transducer on
the epigastrium and between ribs to obtain sonographic
images of the heart, aorta, and vena cava. Because of its safety,
reliability, and rapidity, TTE should be the first-line modality
used in most ICU patients to assess hemodynamic instabil-
ity.89 Poor or limited acoustic windows may limit image
quality. However, new technology, harmonic imaging, and
new echo contrast products have significantly improved echo
signal acquisition.90 TEE is conducted through a miniaturized
transducer mounted laterally on the tip of an endoscopic
shaft. Under direct vision and with continuous electrocardi-
ography, the endoscope is inserted into the esophagus and
positioned behind and under the heart to acquire ultrasono-
graphic images of the heart in different angles and planes.
Most TEE transducers can rotate within their casing by 180
degrees (multiplanar) and the shaft may be flexed laterally or
anteroposteriorly. TEE is indicated when a TTE study is not
adequate or in special circumstances, such as when evaluating
aortic dissection or endocarditis of prosthetic valves, or in
excluding intracardiac thrombus presence before elective car-
dioversion. TEE can be used easily in ICUs because the
majority of patients are already intubated and can be sedated
for the procedure.91 However, the possibility of esophageal
disease, recent esophageal or gastric surgery, and an unstable
cervical spine are contraindications to TEE.
Utility of Echocardiography  
in Intensive Care Units
Ventricular Function
LV dysfunction in critically ill patients is common and may
be caused by ischemia, sepsis, or hyperadrenergic states (“neu-
rogenic shock”). When the LV becomes dysfunctional, its
diameter progressively increases and the diastolic volume–to–
stroke volume ratio increases: expressed as a percentage, this
means that the fraction of end-diastolic blood that is ejected
per cycle is reduced. Ejection fraction (EF) can only be reliably
measured by echocardiography, and it is useful to correlate
filling pressures and stroke volume, using a hemodynamic
monitor (such as a PAC) with EF. MAP may appear normal
despite a severe reduction in ventricular function because this
may be impaired further by vasoconstrictor use to increase
CPP. The size of the ventricle is measured using either M-mode
or two-dimensional (2D)-mode. Hence, in neurologic disease,
when vasodilators cannot be safely used, echocardiography
may guide inotropic therapy. Echocardiography also may
unveil regional wall motion abnormalities, usually associated
with myocardial ischemia.
RV dysfunction also is very common in critically ill patients.
Pulmonary embolism (PE) and ARDS are the main causes.92
In addition, excessive mean airway pressure, fat embolism, and
elevated pulmonary vascular resistance also cause increased
RV afterload and RV strain. The right ventricle’s complex
pyramidal shape makes quantitative and qualitative evalua-
tion of systolic function very difficult. Although the PAC can
be useful to diagnose RV failure, echocardiography is required
to diagnose the underlying cause. 2D echocardiography evalu-
ation of RV size, shape, kinetics of septum, and RV free wall
along with the evaluation of severity of tricuspid regurgitation
(TR) by color Doppler are integral components of a complete
evaluation. RV size is generally compared to LV size. In a four-
chamber view the ratio between RV and LV end-diastolic area
is measured. A diastolic ventricular ratio greater than 0.6 sug-
gests moderate, and a ratio greater than 1, severe dilation.93
Hypoxemia and Intracardiac Shunt
The incidence of patent foramen ovale is about 30%. In this
setting of RV dysfunction right atrial pressure can exceed the
left atrial pressure. A foramen ovale may open, causing the
anatomic substrate for significant intracardiac shunt that can
be detected by contrast or agitated saline echocardiogram.
After injection of agitated saline in a peripheral or central vein,
the appearance of contrast or microbubbles in the left atrium
within three cardiac cycles is diagnostic of a right-to-left
shunt.
Assessment of Cardiac Output
Thermodilution CO measurement is not always accurate in
critically ill patients. Very low or very high CO, severe TR,
rapid temperature changes, or intracardiac shunt can cause
186 Section II—Clinical and Laboratory Assessment
false readings. In these conditions echocardiography can rela-
tively reliably measure SV and thus CO.94 The most common
technique is Doppler-derived instantaneous blood flow mea-
surement through a conduit (LV outflow tract, pulmonic or
mitral valve). For example, if the ultrasound beam is directed
along the aorta, part of the signal is reflected back by the
moving red blood cells. The frequency is different, and this
Doppler shift then is used to calculate the flow velocity and
volume and cardiac output. SV is equal to the product of
cross-sectional area (CSA) of the conduit, determined by 2D
echo, and integration of instantaneous blood flow, velocity
time integral (VTI), through the conduit.
CSA = Diameter of conduit (D) squared X (π/4)
SV = CSA × VTI
SV × Heart rate (HR) = CO
CO = CSA × VTI × HR
The technique does not allow continuous monitoring and
may not be valid when there is an intra-aortic balloon pump.
Volume Status
Knowledge about volume status and fluid responsiveness is
essential in the NCCU. However, traditional measures of these
variables—static preload filling-pressure markers such as CVP
or pulmonary artery occlusion pressure (PAOP)—may not be
ideal tools to guide fluid management.31 ICU studies show that
echocardiography, including a transthoracic-focused rapid
echocardiographic examination can help estimate volume
status or fluid responsiveness or guide fluid therapy such as in
septic shock.95-97 Systolic obliteration of the LV is a good indi-
cator of inadequate preload. Besides hypovolemia, right heart
dysfunction can be the culprit. On the other hand, ventricle
dilation, defined by left ventricular end-diastolic LV diameter,
can be just a sign of chronic heart failure and some patients
can still respond to volume challenge.98 LV dimensions are
measured directly to provide the primary preload parameters
of left ventricular end-diastolic area (LVEDA) or volume
(LVEDV) with the latter considered a superior determinant of
fluid status. The modified Simpson formula is one of two
methods to calculate LVEDV by using the assumption that the
cardiac chamber is a sum of multiple cylinders in a truncated
ellipse. The alternative method of LVEDV calculation uses
semiautomated disk summation to calculate volume from
multiple diameters perpendicular to a longitudinal ventricular
axis. The latter method is more expeditious with fewer required
variations in probe position (short-axis view only) and is the
method of choice for volume calculation in the presence of
distorted ventricular anatomy.99 Several studies have demon-
strated a persistent underestimation of LV volumes by the disk
summation method.100,101 Not all reports favor LVEDV as an
indicator of fluid responsiveness. Whereas some studies indi-
cate significant LVEDV differences in responders and nonre-
sponders to a fluid challenge,102,103 others find no difference
compared with those measured with right arterial pressure or
PCWP.28,104,105 In ventilated patients, superior vena cava col-
lapsibility also should be systematically gauged as an indicator
of fluid responsiveness. With positive intrathoracic pressure,
the vena cava demonstrates greater collapsibility or inspira-
tory diameter decrease when volume status is low than when
effective circulating volume is adequate.106,107 Similarly, respi-
ratory variation of inferior vena cava diameter (dIVC) mea-
sured by bedside ultrasonography can help indicate fluid
responsiveness in mechanically ventilated patients. For
example, in SAH patients Moretti and Pizzi108 observed that a
cutoff dIVC greater than16% was 70% sensitive and 100%
specific as a predictor of fluid responsiveness. Esophageal
Doppler echocardiography also has yielded indicators of
preload. However, stroke volume as measured by esophageal
Doppler may not be reliable because of significant interob-
server variability.109 On the other hand, flow time correction
(FTc) or the duration of the aortic velocity signal corrected
for heart rate is considered a promising static indicator of
cardiac preload.110 In a study of neurosurgical patients, FTc
values were predictive of greater than10% increases in stroke
volume index from a fluid challenge.111 Other studies, however,
are less clear on the correlation between FTc and fluid
status.112,113
Laboratory Analysis of Global
Cardiovascular Function
SVO2
When blood flows through tissues, oxygen is removed and
venous blood returns in an oxygen-depleted state to the right
heart. The quantity of oxygen in this mixed venous blood has
been used as a surrogate marker of tissue blood flow, and is
most easily measured by using an optical probe that continu-
ously measures mixed SVO2. Ideally SVO2 that reflects the
balance between oxygen delivery (DO2) and consumption
(VO2) (i.e., the “adequacy” of tissue oxygenation) is measured
with a PAC. If there is no PAC, a CVC can be used to measure
oxygen saturation in the superior vena cava (ScVO2). There
are, however, important differences between SVO2 and
ScVO2.114 Furthermore, the accuracy of ScVO2 depends on the
position of the CVC tip; when located high in the superior
vena cava it is less reliable, whereas when adjacent to the right
atrium it is very close to SVO2. Normally, venous blood is 70%
to 80% saturated with oxygen. When there is increased oxygen
uptake by tissues, or inadequate blood flow, the SVO2 falls.
Hence where there is a clinical suggestion of tissue hypoperfu-
sion in the absence of hypotension (cold clammy extremities,
oliguria), SVO2 is a robust marker of tissue blood flow. The
use of venous oximetry in shock and severe sepsis is a feature
of goal-directed resuscitation and has been validated by Rivers,
Shoemaker, and others.115-117 However, the role of therapy to
increase mixed or central venous oxygen saturation as a
marker of adequate tissue perfusion or fluid responsiveness or
as a guide to fluid therapy in ICU patients still needs to be
fully elucidated.118-121
Lactate
The presence or absence of metabolic acidosis has been widely
used as a surrogate endpoint of resuscitation for critically ill
and trauma patients since the 1970s, when it was observed that
hypothermia, coagulopathy, and acidosis constituted a “triad
of death” in trauma victims.122 The mechanism of acidosis is
presumed to be cellular oxygen debt secondary to hypoperfu-
sion or hypoxia. The anion implicated in this acidosis is
lactate. The ability to measure serum lactate was not
 Section II—Clinical and Laboratory Assessment 187
universally available until the 1990s, and so a surrogate, the
base deficit,123 has been used widely. Care should be taken
using this tool, and inferences about its validity should be
made in the context of quantitative acid-base chemistry.124-127
In the setting of acute trauma there may be multiple acidifying
and alkalinizing processes that occur simultaneously depend-
ing on the volume and nature of intravenous fluids.128 A
“normal” base deficit or excess may hide potentially dangerous
acid-base disturbances, and base deficit does not reliably
reflect lactate in the emergency setting.124,129,130
Lactic acidosis on admission to the emergency department
is a marker of illness severity. The magnitude of acidosis and
of serum lactate elevation is associated with patient out-
comes.122,131-133 The speed of lactate clearance from the circula-
tion also is a prognostic indicator.131,134-136 This led to the
concept of hyperaggressive volume resuscitation to “wash out
lactate,” under the assumption that serum lactate is a marker
of tissue hypoperfusion. For example McNelis et al. observed
that lactate clearance was more rapid in surgical intensive care
unit (SICU) survivors. However, oxygen delivery and con-
sumption was similar in survivors and nonsurvivors.135 This
indicates that although admission lactate and lactate clearance
may be independent factors associated with mortality, the
mechanism of hyperlactemia may not necessarily be hypoper-
fusion. In addition, several lines of evidence suggest that
lactate itself may be harmless137-139 or a necessary intermediate
in wound healing and repair because it may enhance collagen
production and deposition and angiogenesis.140 Consistent
with this, oxygen levels appear to have little effect on wound
lactate137 and whether using serum lactate as a marker of tissue
perfusion to guide therapy helps improve outcome is still to
be fully elucidated.118 Similarly, the role of muscle or subcuta-
neous microdialysis of lactate, pyruvate, and the lactate-to-
pyruvate ratio to guide resuscitation requires further study.141,142
There is emerging evidence that lactic acidosis in trauma
and critical illness may not be associated with hypoxemia but
also may result from epinephrine-driven aerobic glycoly-
sis.139,143 This results from cyclic adenosine monophosphate
(cAMP)–medicated Na+-K+ ATPase activity that is driven
by beta-adrenoceptor activation.144 Administration of beta-
adrenergic agents increases tissue lactate levels; administration
of antiadrenergic agents reduces tissue lactate levels.139
Hyperglycemia can aggravate ischemic brain damage
because of increased anaerobic glycolysis, lactate production,
and tissue acidosis.145 However, the brain is glucose depen-
dent, and glycolysis supplies energy for metabolism during
hypoxemia. In addition, lactate may be a source of energy
during the postischemia state. Thus there is a “glucose paradox
of cerebral ischemia” in which increased energy delivery
appears to be associated with worsened outcomes.146 However,
there also is evidence to suggest that elevated brain lactate can
be associated with better recovery after acute brain injury
predominantly because of hyperglycolysis rather than
hypoxia.147
Alternatively the brain injury may be associated with ele-
vated serum cortisol because elevated blood glucose and
lactate is associated with stress and stress is associated with
increased blood corticosteroid levels.146 In animal studies the
timing of hyperglycemia before and after ischemic brain inju-
ries alters outcomes. The application of corticosteroid inhibi-
tors may reduce, whereas the application of steroids may
increase the severity of brain injuries.146
Conclusion
Monitoring of volume status and cardiac function plays a fun-
damental role in the care of critically ill patients including
those managed in the NCCU, in part because a variety of acute
neurologic disorders can lead to cardiac instability (e.g., Takot-
subo cardiomyopathy) after aneurysm rupture or changes in
volume status (e.g., diabetes insipidus). In addition, hemody-
namic manipulation (e.g., induced hypertension or hypervol-
emia) is an important intervention to regulate CPP and CBF.
However, fluid therapy or induced hypertension can aggravate
outcome through exacerbation of lung function that already
may be compromised.3,4,148 There are a variety of methods,
both invasive and noninvasive, that can be used to monitor
volume and cardiac status including: clinical evaluation and
laboratory analysis, arterial pressure pulse contour and wave-
form analysis, CVCs, PACs, transpulmonary thermodilution,
and bedside echocardiography or ultrasound among others.
There has been a trend to use less invasive devices to measure
hemodynamic variables.41 In addition, TTE and even TEE,
although traditionally the domain of cardiologists, are used
more frequently in critical care and in the operating room in a
focused goal-directed manner.88 These various monitors may
be used alone or in combination preferably with integration of
the variables. For example, a hypotensive patient with low
cardiac output differs in diagnostic and therapeutic possibili-
ties from a hypotensive patient with increased cardiac output.
However, the optimal system to monitor a patient may vary
over the patient’s hospital course and depends on the indi-
vidual patient, his or her pathology, and the problems that
may arise. More important, as with any monitor, the informa-
tion obtained needs to be interpreted and applied correctly.149
Key Points
1. Cardiovascular monitoring is an essential component of
neurocritical care.
2. Blood pressure monitoring can be achieved using oscil-
lometry or by invasive arterial line placement. Both types
of monitoring can produce errors in measurement. Because
of continuity and accuracy of monitoring very low blood
pressures, the intra-arterial measurement represents the
gold standard in critically ill patients.
3. Arterial lines reliably measure mean arterial pressure and
can be used to estimate fluid responsiveness through pulse
pressure variability. However problems of resonance and
dampening may limit interpretation.
4. Blood pressure alone is not always a good measure of tissue
blood flow. Some form of tissue flow monitoring is
required in patients at risk of low flow state.
5. Several stroke volume monitors that perform pulse contour
analysis are available. Each of these devices appears to be
reliable and have varying levels of invasiveness.
6. The pulmonary artery catheter remains the gold standard
device for hemodynamic monitoring. All hemodynamic
devices require a high level of expertise to correctly inter-
pret acquired data.
7. Echocardiography is an important diagnostic tool in hemo-
dynamically unstable patients. Transesophageal echocar-
diography provides better image quality than transthoracic,
but it is more invasive.
188 Section II—Clinical and Laboratory Assessment
8. Mixed venous oxygen saturation is an easily performed
and useful measure of tissue blood flow.
9. The presence of tissue acidosis can help to identify
critically ill patients, and apportion risk. However, it is
important to avoid over interpreting the mechanism of
hyperlactatemia.
References
1. Ramasamy I. Biochemical markers in acute coronary syndrome. Clin Chim
Acta 2011;412(15-16):1279–96. Epub 2011, Apr 8.
2. Noveanu M, Mebazaa A, Mueller C. Cardiovascular biomarkers in the ICU.
Curr Opin Crit Care 2009;15(5):377–83.
3. Contant CF, Valadka AB, Gopinath SP, et al. Adult respiratory distress
syndrome: a complication of induced hypertension after severe head injury.
J Neurosurg 2001;95:560–8.
4. Boyd JH, Forbes J, Nakada TA, et al. Fluid resuscitation in septic shock: a
positive fluid balance and elevated central venous pressure are associated
with increased mortality. Crit Care Med 2011;39(2):259–65.
5. Fletcher JJ, Bergman K, Blostein PA, et al. Fluid balance, complications, and
brain tissue oxygen tension monitoring following severe traumatic brain
injury. Neurocrit Care 2010;13(1):47–56.
6. Riva-Rocci S. Un Nuovo sfigmomanometro. Gazzeta Medica di Torino 1896;
47:981–96.
7. Cushing H. On routine determination of arterial tension in operating room
and clinic. Boston Med Surg J 1903;148:250.
8. Bigatello LM, Schmidt U. Arterial blood pressure monitoring. Minerva
Anestesiol 2003;69(4):201–9.
9. Peterson L, Dripps R, Risman G. A method for recording the arterial pulse
and blood pressure in man. Am Heart J 1949;37:771.
10. Cohn JN. Blood pressure measurement in shock. Mechanism of inaccuracy
in ausculatory and palpatory methods. JAMA 1967;199(13):118–22.
11. O’Rourke MF. Impact pressure, lateral pressure, and impedance in the
proximal aorta and pulmonary artery. J ApplPhysiol 1968;25(5):533–41.
12. Slogoff S, Keats AS, Arlund C. On the safety of radial artery cannulation.
Anesthesiology 1983;59(1):42–7.
13. Chauhan S, Saxena N, Mehrotra S, et al. Femoral artery pressures are more
reliable than radial artery pressures on initiation of cardiopulmonary bypass.
J Cardiothorac Vasc Anesth 2000;14(3):274–6.
14. Mignini MA, Piacentini E, Dubin A. Peripheral arterial blood pressure
monitoring adequately tracks central arterial blood pressure in critically ill
patients: an observational study. Crit Care 2006;10(2):R43.
15. Martin C, Saux P, Papazian L, et al. Long-term arterial cannulation in ICU
patients using the radial artery or dorsalis pedis artery. Chest 2001;119(3):
901–6.
16. Clark VL, Kruse JA. Arterial catheterization. Crit Care Clin 1992;8(4):687–97.
17. Ganchi PA, Wilhelmi BJ, Fujita K, et al. Ruptured pseudoaneurysm
complicating an infected radial artery catheter: case report and review of the
literature. Ann Plast Surg 2001;46(6):647–50.
18. Band JD, Maki DG. Infections caused by aterial catheters used for
hemodynamic monitoring. Am J Med 1979;67(5):735–41.
19. Araghi A, Bander JJ, Guzman JA. Arterial blood pressure monitoring in
overweight critically ill patients: invasive or noninvasive? Crit Care 2006;
10(2):R64.
20. Bur A, Herkner H, Vlcek M, et al. Factors influencing the accuracy of
oscillometric blood pressure measurement in critically ill patients.
Crit Care Med 2003;31(3):793–9.
21. Manios E, Vemmos K, Tsivgoulis G, et al. Comparison of noninvasive
oscillometric and intra-arterial blood pressure measurements in hyperacute
stroke. Blood Press Monit 2007;12(3):149–56.
22. Muecke S, Bersten A, Plummer J. The mean machine: accurate non-invasive
blood pressure measurement in the critically ill patient. J Clin Monit
Comput 2009;23(5):283–97.
23. Michard F, Boussat S, Chemla D, et al. Relation between respiratory changes
in arterial pulse pressure and fluid responsiveness in septic patients with
acute circulatory failure. Am J Respir Crit Care Med 2000;162:134–8.
24. Michard F. Stroke volume variation: from applied physiology to improved
outcomes. Crit Care Med 2011;39(2):402–3.
25. Benes J, Chytra I, Altmann P, et al. Intraoperative fluid optimization using
stroke volume variation in high risk surgical patients: results of prospective
randomized study. Crit Care 2010;14(3):R118. Epub 2010, Jun 16.
26. Goedje O, Hoeke K, Lichtwarck-Aschoff M, et al. Continuous cardiac output
by femoral arterial thermodilution calibrated pulse contour analysis:
comparison with pulmonary arterial thermodilution. Crit Care Med 1999;
27(11):2407–12.
27. Kramer A, Zygun D, Hawes H, et al. Pulse pressure variation predicts fluid
responsiveness following coronary artery bypass surgery. Chest 2004;126(5):
1563–8.
28. Feissel M, Michard F, Mangin I, et al. Respiratory changes in aortic blood
velocity as an indicator of fluid responsiveness in ventilated patients with
septic shock. Chest 2001;119(3):867–73.
29. Michard F, Boussat S, Chemla D, et al. Relation between respiratory changes
in arterial pulse pressure and fluid responsiveness in septic patients with
acute circulatory failure. Am J Respir Crit Care Med 2000;162(1):134–8.
30. Michard F, Lopes MR, Auler JO Jr. Pulse pressure variation: beyond the fluid
management of patients with shock. Crit Care 2007;11(3):131.
31. Marik PE, Cavallazzi R, Vasu T, et al. Dynamic changes in arterial waveform
derived variables and fluid responsiveness in mechanically ventilated
patients: a systematic review of the literature. Crit Care Med 2009;37(9):
2642–7. Review.
32. Auler JO Jr, Galas F, Hajjar L, et al. Online monitoring of pulse pressure
variation to guide fluid therapy after cardiac surgery. Anesth Analg 2008;
106(4):1201–6, table of contents.
33. Marx G, Cope T, McCrossan L, et al. Assessing fluid responsiveness by stroke
volume variation in mechanically ventilated patients with severe sepsis. Eur J
Anaesthesiol 2004;21(2):132–8.
34. Ornstein E, Eidelman LA, Drenger B, et al. Systolic pressure variation
predicts the response to acute blood loss. J Clin Anesth 1998;10(2):
137–40.
35. Berkenstadt H, Margalit N, Hadani M, et al. Stroke volume variation as a
predictor of fluid responsiveness in patients undergoing brain surgery.
Anesth Analg 2001;92(4):984–9.
36. Wiesenack C, Prasser C, Rodig G, et al. Stroke volume variation as an
indicator of fluid responsiveness using pulse contour analysis in
mechanically ventilated patients. Anesth Analg 2003;96(5):1254–7.
37. de Waal EE, Rex S, Kruitwagen CL, et al. Stroke volume variation obtained
with FloTrac/Vigileo fails to predict fluid responsiveness in coronary artery
bypass graft patients. Br J Anaesth 2008;100(5):725–6.
38. Sander M, von Heymann C, Foer A, et al. Pulse contour analysis after
normothermic cardiopulmonary bypass in cardiac surgery patients. Crit
Care 2005;9(6):R729–34.
39. Pinsky MR. Probing the limits of arterial pulse contour analysis to predict
preload responsiveness. Anesth Analg 2003;96(5):1245–7.
40. Mahjoub Y, Pila C, Friggeri A, et al. Assessing fluid responsiveness in
critically ill patients: false-positive pulse pressure variation is detected by
Doppler echocardiographic evaluation of the right ventricle. Crit Care Med
2009;37(9):2570–5.
41. Vincent JL, Rhodes A, Perel A, et al. Clinical review: update on
hemodynamic monitoring—a consensus of 16. Crit Care 2011;15(4):229.
42. Benington S, Ferris P, Nirmalan M. Emerging trends in minimally invasive
haemodynamic monitoring and optimization of fluid therapy. Eur J
Anaesthesiol 2009;26(11):893–905. Review.
43. Jabot J, Monnet X, Bouchra L, et al. Cardiac function index provided by
transpulmonary thermodilution behaves as an indicator of left ventricular
systolic function. Crit Care Med 2009;37(11):2913–8.
44. Mutoh T, Kazumata K, Ishikawa T, et al. Performance of bedside
transpulmonary thermodilution monitoring for goal-directed
hemodynamic management after subarachnoid hemorrhage. Stroke 2009;
40(7):2368–74.
45. Marik PE, Baram M, Vahid B. Does central venous pressure predict fluid
responsiveness? A systematic review of the literature and the tale of seven
mares. Chest 2008;134(1):172–8.
46. Magder S, Bafaqeeh F. The clinical role of central venous pressure
measurements. J Intensive Care Med 2007;22(1):44–51.
47. Hamzaoui O, Monnet X, Richard C, et al. Effects of changes in vascular tone
on the agreement between pulse contour and transpulmonary
thermodilution cardiac output measurements within an up to 6-hour
calibration-free period. Crit Care Med 2008;36(2):434–40.
48. Gust R, Gottschalk A, Bauer H, et al. Cardiac output measurement by
transpulmonary versus conventional thermodilution technique in intensive
care patients after coronary artery bypass grafting. J Cardiothorac Vasc
Anesth 1998;12:519–22. doi: 10.1016/S1053-0770(98)90093–3.
49. Ritter S, Rudiger A, Maggiorini M. Transpulmonary thermodilution-derived
cardiac function index identifies cardiac dysfunction in acute heart failure
and septic patients: an observational study. Crit Care 2009;13(4):R133. doi:
10.1186/cc7994.
50. Sakka SG, Ruhl CC, Pfeiffer UJ, et al. Assessment of cardiac preload and
extravascular lung water by single transpulmonary thermodilution. Intensive
Care Med 2000;26:180–7. doi: 10.1007/s001340050043.
A complete list of references for this chapter can be found online at
www.expertconsult.com.
 Section II—Clinical and Laboratory Assessment 188.e1
References
1. Ramasamy I. Biochemical markers in acute coronary syndrome. Clin Chim
Acta 2011;412(15-16):1279–96. Epub 2011, Apr 8.
2. Noveanu M, Mebazaa A, Mueller C. Cardiovascular biomarkers in the ICU.
Curr Opin Crit Care 2009;15(5):377–83.
3. Contant CF, Valadka AB, Gopinath SP, et al. Adult respiratory distress
syndrome: a complication of induced hypertension after severe head injury.
J Neurosurg 2001;95:560–8.
4. Boyd JH, Forbes J, Nakada TA, et al. Fluid resuscitation in septic shock: a
positive fluid balance and elevated central venous pressure are associated
with increased mortality. Crit Care Med 2011;39(2):259–65.
5. Fletcher JJ, Bergman K, Blostein PA, et al. Fluid balance, complications, and
brain tissue oxygen tension monitoring following severe traumatic brain
injury. Neurocrit Care 2010;13(1):47–56.
6. Riva-Rocci S. Un Nuovo sfigmomanometro. Gazzeta Medica di Torino
1896;47:981–96.
7. Cushing H. On routine determination of arterial tension in operating room
and clinic. Boston Med Surg J 1903;148:250.
8. Bigatello LM, Schmidt U. Arterial blood pressure monitoring. Minerva
Anestesiol 2003;69(4):201–9.
9. Peterson L, Dripps R, Risman G. A method for recording the arterial pulse
and blood pressure in man. Am Heart J 1949;37:771.
10. Cohn JN. Blood pressure measurement in shock. Mechanism of inaccuracy
in ausculatory and palpatory methods. JAMA 1967;199(13):118–22.
11. O’Rourke MF. Impact pressure, lateral pressure, and impedance in the
proximal aorta and pulmonary artery. J ApplPhysiol 1968;25(5):533–41.
12. Slogoff S, Keats AS, Arlund C. On the safety of radial artery cannulation.
Anesthesiology 1983;59(1):42–7.
13. Chauhan S, Saxena N, Mehrotra S, et al. Femoral artery pressures are more
reliable than radial artery pressures on initiation of cardiopulmonary
bypass. J Cardiothorac Vasc Anesth 2000;14(3):274–6.
14. Mignini MA, Piacentini E, Dubin A. Peripheral arterial blood pressure
monitoring adequately tracks central arterial blood pressure in critically ill
patients: an observational study. Crit Care 2006;10(2):R43.
15. Martin C, Saux P, Papazian L, et al. Long-term arterial cannulation in ICU
patients using the radial artery or dorsalis pedis artery. Chest 2001;119(3):
901–6.
16. Clark VL, Kruse JA. Arterial catheterization. Crit Care Clin 1992;8(4):
687–97.
17. Ganchi PA, Wilhelmi BJ, Fujita K, et al. Ruptured pseudoaneurysm
complicating an infected radial artery catheter: case report and review of
the literature. Ann Plast Surg 2001;46(6):647–50.
18. Band JD, Maki DG. Infections caused by aterial catheters used for
hemodynamic monitoring. Am J Med 1979;67(5):735–41.
19. Araghi A, Bander JJ, Guzman JA. Arterial blood pressure monitoring in
overweight critically ill patients: invasive or noninvasive? Crit Care 2006;
10(2):R64.
20. Bur A, Herkner H, Vlcek M, et al. Factors influencing the accuracy of
oscillometric blood pressure measurement in critically ill patients. Crit Care
Med 2003;31(3):793–9.
21. Manios E, Vemmos K, Tsivgoulis G, et al. Comparison of noninvasive
oscillometric and intra-arterial blood pressure measurements in hyperacute
stroke. Blood Press Monit 2007;12(3):149–56.
22. Muecke S, Bersten A, Plummer J. The mean machine: accurate non-invasive
blood pressure measurement in the critically ill patient. J Clin Monit
Comput 2009;23(5):283–97.
23. Michard F, Boussat S, Chemla D, et al. Relation between respiratory
changes in arterial pulse pressure and fluid responsiveness in septic patients
with acute circulatory failure. Am J Respir Crit Care Med 2000;162:134–8.
24. Michard F. Stroke volume variation: from applied physiology to improved
outcomes. Crit Care Med 2011;39(2):402–3.
25. Benes J, Chytra I, Altmann P, et al. Intraoperative fluid optimization using
stroke volume variation in high risk surgical patients: results of prospective
randomized study. Crit Care 2010;14(3):R118. Epub 2010, Jun 16.
26. Goedje O, Hoeke K, Lichtwarck-Aschoff M, et al. Continuous cardiac
output by femoral arterial thermodilution calibrated pulse contour analysis:
comparison with pulmonary arterial thermodilution. Crit Care Med 1999;
27(11):2407–12.
27. Kramer A, Zygun D, Hawes H, et al. Pulse pressure variation predicts fluid
responsiveness following coronary artery bypass surgery. Chest 2004;126(5):
1563–8.
28. Feissel M, Michard F, Mangin I, et al. Respiratory changes in aortic blood
velocity as an indicator of fluid responsiveness in ventilated patients with
septic shock. Chest 2001;119(3):867–73.
29. Michard F, Boussat S, Chemla D, et al. Relation between respiratory
changes in arterial pulse pressure and fluid responsiveness in septic patients
with acute circulatory failure. Am J Respir Crit Care Med 2000;162(1):
134–8.
30. Michard F, Lopes MR, Auler JO Jr. Pulse pressure variation: beyond the
fluid management of patients with shock. Crit Care 2007;11(3):131.
31. Marik PE, Cavallazzi R, Vasu T, et al. Dynamic changes in arterial waveform
derived variables and fluid responsiveness in mechanically ventilated
patients: a systematic review of the literature. Crit Care Med 2009;37(9):
2642–7. Review.
32. Auler JO Jr, Galas F, Hajjar L, et al. Online monitoring of pulse pressure
variation to guide fluid therapy after cardiac surgery. Anesth Analg
2008;106(4):1201–6, table of contents.
33. Marx G, Cope T, McCrossan L, et al. Assessing fluid responsiveness by
stroke volume variation in mechanically ventilated patients with severe
sepsis. Eur J Anaesthesiol 2004;21(2):132–8.
34. Ornstein E, Eidelman LA, Drenger B, et al. Systolic pressure variation
predicts the response to acute blood loss. J Clin Anesth 1998;10(2):
137–40.
35. Berkenstadt H, Margalit N, Hadani M, et al. Stroke volume variation as a
predictor of fluid responsiveness in patients undergoing brain surgery.
Anesth Analg 2001;92(4):984–9.
36. Wiesenack C, Prasser C, Rodig G, et al. Stroke volume variation as an
indicator of fluid responsiveness using pulse contour analysis in
mechanically ventilated patients. Anesth Analg 2003;96(5):1254–7.
37. de Waal EE, Rex S, Kruitwagen CL, et al. Stroke volume variation obtained
with FloTrac/Vigileo fails to predict fluid responsiveness in coronary artery
bypass graft patients. Br J Anaesth 2008;100(5):725–6.
38. Sander M, von Heymann C, Foer A, et al. Pulse contour analysis after
normothermic cardiopulmonary bypass in cardiac surgery patients. Crit
Care 2005;9(6):R729–34.
39. Pinsky MR. Probing the limits of arterial pulse contour analysis to predict
preload responsiveness. Anesth Analg 2003;96(5):1245–7.
40. Mahjoub Y, Pila C, Friggeri A, et al. Assessing fluid responsiveness in
critically ill patients: false-positive pulse pressure variation is detected by
Doppler echocardiographic evaluation of the right ventricle. Crit Care Med
2009;37(9):2570–5.
41. Vincent JL, Rhodes A, Perel A, et al. Clinical review: update on
hemodynamic monitoring—a consensus of 16. Crit Care 2011;15(4):229.
42. Benington S, Ferris P, Nirmalan M. Emerging trends in minimally invasive
haemodynamic monitoring and optimization of fluid therapy. Eur J
Anaesthesiol 2009;26(11):893–905. Review.
43. Jabot J, Monnet X, Bouchra L, et al. Cardiac function index provided by
transpulmonary thermodilution behaves as an indicator of left ventricular
systolic function. Crit Care Med 2009;37(11):2913–8.
44. Mutoh T, Kazumata K, Ishikawa T, et al. Performance of bedside
transpulmonary thermodilution monitoring for goal-directed
hemodynamic management after subarachnoid hemorrhage. Stroke 2009;
40(7):2368–74.
45. Marik PE, Baram M, Vahid B. Does central venous pressure predict fluid
responsiveness? A systematic review of the literature and the tale of seven
mares. Chest 2008;134(1):172–8.
46. Magder S, Bafaqeeh F. The clinical role of central venous pressure
measurements. J Intensive Care Med 2007;22(1):44–51.
47. Hamzaoui O, Monnet X, Richard C, et al. Effects of changes in vascular
tone on the agreement between pulse contour and transpulmonary
thermodilution cardiac output measurements within an up to 6-hour
calibration-free period. Crit Care Med 2008;36(2):434–40.
48. Gust R, Gottschalk A, Bauer H, et al. Cardiac output measurement by
transpulmonary versus conventional thermodilution technique in intensive
care patients after coronary artery bypass grafting. J Cardiothorac Vasc
Anesth 1998;12:519–22. doi: 10.1016/S1053-0770(98)90093–3.
49. Ritter S, Rudiger A, Maggiorini M. Transpulmonary thermodilution-
derived cardiac function index identifies cardiac dysfunction in acute heart
failure and septic patients: an observational study. Crit Care
2009;13(4):R133. doi: 10.1186/cc7994.
50. Sakka SG, Ruhl CC, Pfeiffer UJ, et al. Assessment of cardiac preload and
extravascular lung water by single transpulmonary thermodilution.
Intensive Care Med 2000;26:180–7. doi: 10.1007/s001340050043.
51. Biancofiore G, Critchley LA, Lee A, et al. Evaluation of an uncalibrated
arterial pulse contour cardiac output monitoring system in cirrhotic
patients undergoing liver surgery. Br J Anaesth 2009;102:47–54.
52. Monnet X, Anguel N, Naudin B, et al. Arterial pressure-based cardiac
output in septic patients: different accuracy of pulse contour, uncalibrated
pressure waveform devices. Crit Care 2010;14:R109.
188.e2 Section II—Clinical and Laboratory Assessment
53. De Backer D, Marx G, Tan A, et al. Arterial pressure-based cardiac output
monitoring: a multicenter validation of the third- generation software in
septic patients. Intensive Care Med 2011;37(2):233–40.
54. Monnet X, Lahner D. Can the ‘ ‘FloTrac’ ’ really track flow in septic patients?
Intensive Care Med 2011;37(2):183–5.
55. Mayer J, Boldt J, Poland R, et al. Continuous arterial pressure waveform-
based cardiac output using the FloTrac/Vigileo: a review and meta-analysis.
J Cardiothorac Vasc Anesth 2009;23(3):401–6.
56. McLean AS, Huang SJ, Kot M, et al. Comparison of cardiac output
measurements in critically ill patients: FloTrac/Vigileo vs transthoracic
Doppler echocardiography. Anaesth Intensive Care 2011;39(4):590–8.
57. Mayer J, Boldt J, Beschmann R, et al. Uncalibrated arterial pressure
waveform analysis for less-invasive cardiac output determination in obese
patients undergoing cardiac surgery. Br J Anaesth 2009;103(2):185–90.
58. Maj GLG, Landoni G, Bignami E, et al. PRAM (pressure recording
analytical method) for cardiac index measurement in unstable patients.
Minerva Anestesiol 2008;74:60.
59. Funk DJ, Moretti EW, Gan TJ. Minimally invasive cardiac output
monitoring in the perioperative setting. Anesth Analg 2009;108(3):
887–97.
60. Tsutsui M, Matsuoka N, Ikeda T, et al. Comparison of a new cardiac output
ultrasound dilution method with thermodilution technique in adult
patients under general anesthesia. J Cardiothorac Vasc Anesth 2009;23(6):
835–40.
61. Squara P, Denjean D, Estagnasie P, et al. Noninvasive cardiac output
monitoring (NICOM): a clinical validation. Intensive Care Med 2007;33:
1191–4.
62. Marque S, Cariou A, Chiche JD, et al. Comparison between Flotrac-Vigileo
and Bioreactance, a totally noninvasive method for cardiac output
monitoring. Crit Care 2009;13:R73.
63. Swan HJ, Ganz W, Forrester J, et al. Catheterization of the heart in man
with use of a flow-directed balloon-tipped catheter. N Engl J Med 1970;
283(9):447–51.
64. Wagner JG, Leatherman JW. Right ventricular end-diastolic volume as a
predictor of the hemodynamic response to a fluid challenge. Chest 1998;
113(4):1048–54.
65. Connors AF Jr, Speroff T, Dawson NV, et al. The effectiveness of right heart
catheterization in the initial care of critically ill patients. SUPPORT
Investigators. JAMA 1996;276(11):889–97.
66. Sandham JD, Hull RD, Brant RF, et al. A randomized, controlled trial of the
use of pulmonary-artery catheters in high-risk surgical patients. N Engl J
Med 2003;348(1):5–14.
67. Vincent JL, Pinsky MR, Sprung CL. The pulmonary artery catheter: in
medio virtus. Crit Care Med 2008;36(11):3093–6. Review.
68. Squara P, Cecconi M, Rhodes A, et al. Tracking changes in cardiac output:
methodological considerations for the validation of monitoring devices.
Intensive Care Med 2009;35:1801–8.
69. Pinsky MR. Pulmonary artery occlusion pressure. Intensive Care Med
2003;29(1):19–22.
70. Gayat E, Mebazaa A. Pulmonary hypertension in critical care. Curr Opin
Crit Care 2011;17(5):439–48.
71. Gnaegi A, Feihl F, Perret C. Intensive care physicians’ insufficient knowledge
of right-heart catheterization at the bedside: time to act? Crit Care Med
1997;25(2):213–20.
72. Kumar A, Anel R, Bunnell E, et al. Pulmonary artery occlusion pressure and
central venous pressure fail to predict ventricular filling volume, cardiac
performance, or the response to volume infusion in normal subjects. Crit
Care Med 2004;32(3):691–9.
73. Michard F, Teboul JL. Predicting fluid responsiveness in ICU patients: a
critical analysis of the evidence. Chest 2002;121(6):2000–8.
74. Singh S, Goyal A. The origin of echocardiography: a tribute to Inge Edler.
Tex Heart Inst J 2007;34(4):431–8.
75. Edler I, Lindstrom K. The history of echocardiography. Ultrasound Med
Biol 2004;30(12):1565–644.
76. Nilsson J, Westling H. Ultrasound in Lund: three world premieres. Clin
Physiol Funct Imaging 2004;24(3):137–40.
77. Edler I, Hertz CH. The use of ultrasonic reflectoscope for the continuous
recording of the movements of heart walls. 1954. Clin Physiol Funct
Imaging 2004;24(3):118–36.
78. Frazin L, Talano JV, Stephanides L, et al. Esophageal echocardiography.
Circulation 1976;54(1):102–8.
79. Mayo PH. Training in critical care echocardiography. Ann Intensive Care
2011;30(1):36.
80. Quintard H, Philip I, Ichai C. French survey on current use of ultrasound
in the critical care unit: ECHOREA. Ann Fr Anesth Reanim 2011;Oct 4.
Epub ahead of print.
81. Vignon P, Mücke F, Bellec F, et al. Basic critical care echocardiography:
validation of a curriculum dedicated to noncardiologist residents. Crit Care
Med 2011;39:636–42.
82. American College of Cardiology Foundation Appropriate Use Criteria Task
Force, American Society of Echocardiography, American Heart Association,
et al. ACCF/ASE/AHA/ASNC/HFSA/HRS/SCAI/SCCM/SCCT/SCMR 2011
Appropriate Use Criteria for Echocardiography. A Report of the American
College of Cardiology Foundation Appropriate Use Criteria Task Force,
American Society of Echocardiography, American Heart Association,
American Society of Nuclear Cardiology, Heart Failure Society of America,
Heart Rhythm Society, Society for Cardiovascular Angiography and
Interventions, Society of Critical Care Medicine, Society of Cardiovascular
Computed Tomography, and Society for Cardiovascular Magnetic
Resonance Endorsed by the American College of Chest Physicians. J Am
Coll Cardiol 2011;57(9):1126–66.
83. International expert statement on training standards for critical care
ultrasonography. Expert Round Table on Ultrasound in ICU. Intensive Care
Med 2011;37(7):1077–83. Epub 2011, May 26.
84. Cholley BP, Vieillard-Baron A, Mebazaa A. Echocardiography in the ICU:
time for widespread use! Intensive Care Med 2006;32(1):9–10.
85. Matyal R, Skubas NJ, Shernan SK, et al. Perioperative assessment of
diastolic dysfunction. Anesth Analg 2011;113(3):449–72. Epub 2011, Aug 3.
86. Platts DG, Fraser JF. Contrast echocardiography in critical care: echoes of
the future? A review of the role of microsphere contrast echocardiography.
Crit Care Resusc 2011;13(1):44–55.
87. Faris JG, Veltman MG, Royse CF. Limited transthoracic echocardiography
assessment in anaesthesia and critical care. Best Pract Res Clin Anaesthesiol
2009;23(3):285–98.
88. Cowie BS. Focused transthoracic echocardiography in the perioperative
period. Anaesth Intensive Care 2010;38(5):823–36.
89. Joseph MX, Disney PJ, Da CR, et al. Transthoracic echocardiography to
identify or exclude cardiac cause of shock. Chest 2004;126(5):1592–7.
90. Lepper W, Belcik T, Wei K, et al. Myocardial contrast echocardiography.
Circulation 2004;109(25):3132–5.
91. Colreavy FB, Donovan K, Lee KY, et al. Transesophageal echocardiography
in critically ill patients. Crit Care Med 2002;30(5):989–96.
92. Vieillard-Baron A, Page B, Augarde R, et al. Acute cor pulmonale in massive
pulmonary embolism: incidence, echocardiographic pattern, clinical
implications and recovery rate. Intensive Care Med 2001;27(9):1481–6.
93. Jardin F, Dubourg O, Bourdarias JP. Echocardiographic pattern of acute cor
pulmonale. Chest 1997;111(1):209–17.
94. Feinberg MS, Hopkins WE, vila-Roman VG, et al. Multiplane
transesophageal echocardiographic doppler imaging accurately determines
cardiac output measurements in critically ill patients. Chest
1995;107(3):769–73.
95. Griffee MJ, Merkel MJ, Wei KS. The role of echocardiography in
hemodynamic assessment of septic shock. Crit Care Clin 2010;26(2):365–
82, table of contents.
96. Ferrada P, Murthi S, Anand RJ, et al. Transthoracic focused rapid
echocardiographic examination: real-time evaluation of fluid status in
critically ill trauma patients. J Trauma 2011;70(1):56–62; discussion 62–4.
97. Noritomi DT, Vieira ML, Mohovic T, et al. Echocardiography for
hemodynamic evaluation in the intensive care unit. Shock 2010;34
(Suppl 1):59–62.
98. Tousignant CP, Walsh F, Mazer CD. The use of transesophageal
echocardiography for preload assessment in critically ill patients. Anesth
Analg 2000;90(2):351–5.
99. Hofer CK, Ganter MT, Rist A, et al. The accuracy of preload assessment by
different transesophageal echocardiographic techniques in patients
undergoing cardiac surgery. J Cardiothorac Vasc Anesth 2008;22(2):
236–42.
100. Bernard Y, Meneveau N, Boucher S, et al. Lack of agreement between left
ventricular volumes and ejection fraction determined by two-dimensional
echocardiography and contrast cineangiography in postinfarction patients.
Echocardiography 2001;18(2):113–22.
101. Jensen-Urstad K, Bouvier F, Hojer J, et al. Comparison of different
echocardiographic methods with radionuclide imaging for measuring left
ventricular ejection fraction during acute myocardial infarction treated by
thrombolytic therapy. Am J Cardiol 1998;81(5):538–44.
102. Cheung AT, Savino JS, Weiss SJ, et al. Echocardiographic and hemodynamic
indexes of left ventricular preload in patients with normal and abnormal
ventricular function. Anesthesiology 1994;81(2):376–87.
103. Reuter DA, Felbinger TW, Schmidt C, et al. Stroke volume variations for
assessment of cardiac responsiveness to volume loading in mechanically
ventilated patients after cardiac surgery. Intensive Care Med 2002;28(4):
392–8.
 Section II—Clinical and Laboratory Assessment 188.e3
104. Bennett-Guerrero E, Kahn RA, Moskowitz DM, et al. Comparison of
arterial systolic pressure variation with other clinical parameters to predict
the response to fluid challenges during cardiac surgery. Mt Sinai J Med
2002;69(1-2):96–100.
105. Greim CA, Roewer N, Apfel C, et al. Relation of echocardiographic preload
indices to stroke volume in critically ill patients with normal and low
cardiac index. Intensive Care Med 1997;23(4):411–6.
106. Barbier C, Loubieres Y, Schmit C, et al. Respiratory changes in inferior vena
cava diameter are helpful in predicting fluid responsiveness in ventilated
septic patients. Intensive Care Med 2004;30(9):1740–6.
107. Vieillard-Baron A, Chergui K, Rabiller A, et al. Superior vena caval
collapsibility as a gauge of volume status in ventilated septic patients.
Intensive Care Med 2004;30(9):1734–9.
108. Moretti R, Pizzi B. Inferior vena cava distensibility as a predictor of fluid
responsiveness in patients with subarachnoid hemorrhage. Neurocrit Care
2010;13(1):3–9.
109. Roeck M, Jakob SM, Boehlen T, et al. Change in stroke volume in response
to fluid challenge: assessment using esophageal Doppler. Intensive Care
Med 2003;29(10):1729–35.
110. Singer M, Bennett ED. Noninvasive optimization of left ventricular filling
using esophageal Doppler. Crit Care Med 1991;19(9):1132–7.
111. Lee JH, Kim JT, Yoon SZ, et al. Evaluation of corrected flow time in
oesophageal Doppler as a predictor of fluid responsiveness. Br J Anaesth
2007;99(3):343–8.
112. Monnet X, Rienzo M, Osman D, et al. Esophageal Doppler monitoring
predicts fluid responsiveness in critically ill ventilated patients. Intensive
Care Med 2005;31(9):1195–201.
113. Singer M. The FTc is not an accurate marker of left ventricular preload.
Intensive Care Med 2006;32(7):1089; author reply 91.
114. Reinhart K, Kuhn HJ, Hartog C, et al. Continuous central venous and
pulmonary artery oxygen saturation monitoring in the critically ill.
Intensive Care Med 2004;30:1572–8.
115. Gattinoni L, Brazzi L, Pelosi P, et al. A trial of goal-oriented hemodynamic
therapy in critically ill patients. SvO2 Collaborative Group. N Engl J Med
1995;333(16):1025–32. Epub 1995/10/19. doi: 10.1056/
NEJM199510193331601.
116. Rivers E, Nguyen B, Havstad S, et al. Early goal-directed therapy in the
treatment of severe sepsis and septic shock. N Engl J Med 2001;345(19):
1368–77.
117. Shoemaker WC, Appel PL, Kram HB, et al. Prospective trial of supranormal
values of survivors as therapeutic goals in high-risk surgical patients. Chest
1988;94(6):1176–86. Epub 1988, Dec 1.
118. Gurgel ST, do Nascimento P Jr. Maintaining tissue perfusion in high-risk
surgical patients: a systematic review of randomized clinical trials. Anesth
Analg 2011;112(6):1384–91. Epub 2010, Dec 14.
119. Velissaris D, Pierrakos C, Scolletta S, et al. High mixed venous oxygen
saturation levels do not exclude fluid responsiveness in critically ill septic
patients. Crit Care 2011;15(4):R177. Epub ahead of print.
120. Giraud R, Siegenthaler N, Gayet-Ageron A, et al. ScvO2 as a marker to
define fluid responsiveness. J Trauma 2011;70(4):802–7.
121. Gattinoni L, Brazzi L, Pelosi P, et al. A trial of goal-oriented hemodynamic
therapy in critically ill patients. SvO2 collaborative group. N Engl J Med
1995;333:1025–32.
122. Eddy VA, Morris JA Jr, Cullinane DC. Hypothermia, coagulopathy, and
acidosis. Surg Clin North Am 2000;80(3):845–54. Epub 2000, Jul 18.
123. Runciman WB, Skowronski GA. Pathophysiology of haemorrhagic shock.
Anaesth Intensive Care 1984;12(3):193–205. Epub 1984, Aug 1.
124. Brill SA, Stewart TR, Brundage SI, et al Base deficit does not predict
mortality when secondary to hyperchloremic acidosis. Shock
2002;17(6):459–62. Epub 2002, Jun 19.
125. Fencl V, Leith DE. Stewart’s quantitative acid-base chemistry: applications
in biology and medicine. Respir Physiol 1993;91(1):1–16. Epub 1993, Jan 1.
126. Stewart PA. Modern quantitative acid-base chemistry. Can J Physiol
Pharmacol 1983;61(12):1444–61. Epub 1983, Dec 1.
127. Schlichtig RGA, Severinghaus JW. Current status of acid-base quantitation
in physiology and medicine. Anesthesiol Clin North Am 1998;16:211–33.
128. Fencl V, Jabor A, Kazda A, et al. Diagnosis of metabolic acid-base
disturbances in critically ill patients. Am J Respir Crit Care Med 2000;
162(6):2246–51. Epub 2000, Dec 9.
129. Kaplan LJ, Kellum JA. Initial pH, base deficit, lactate, anion gap, strong ion
difference, and strong ion gap predict outcome from major vascular injury.
Crit Care Med 2004;32(5):1120–4. Epub 2004, Jun 12.
130. Martin MJ, FitzSullivan E, Salim A, et al. Discordance between lactate and
base deficit in the surgical intensive care unit: which one do you trust? Am
J Surg 2006;191(5):625–30. Epub 2006, May 2. doi: 10.1016/j.amjsurg.
2006.02.014.
131. Husain FA, Martin MJ, Mullenix PS, et al. Serum lactate and base deficit as
predictors of mortality and morbidity. Am J Surg 2003;185(5):485–91.
Epub 2003, May 3.
132. Vitek V, Cowley RA. Blood lactate in the prognosis of various forms of
shock. Ann Surg 1971;173(2):308–13. Epub 1971, Feb 1.
133. Guyette F, Suffoletto B, Castillo JL, et al. Prehospital serum lactate as a
predictor of outcomes in trauma patients: a retrospective observational
study. J Trauma 2011;70(4):782–6.
134. Abramson D, Scalea TM, Hitchcock R, et al. Lactate clearance and survival
following injury. J Trauma 1993;35(4):584–8; discussion 8-9. Epub 1993,
Oct 1.
135. McNelis J, Marini CP, Jurkiewicz A, et al. Prolonged lactate clearance is
associated with increased mortality in the surgical intensive care unit. Am J
Surg 2001;182(5):481–5. Epub 2002, Jan 5.
136. Nguyen HB, Rivers EP, Knoblich BP, et al. Early lactate clearance is
associated with improved outcome in severe sepsis and septic shock. Crit
Care Med 2004;32(8):1637–42. Epub 2004, Aug 3.
137. Gladden LB. Lactate metabolism: a new paradigm for the third millennium.
J Physiol 2004;558(Pt 1):5–30. Epub 2004, May 8. doi: 10.1113/jphysiol.
2003.058701.
138. James JH, Luchette FA, McCarter FD, et al. Lactate is an unreliable
indicator of tissue hypoxia in injury or sepsis. Lancet 1999;354(9177):505–
8. Epub 1999, Aug 28. doi: 10.1016/S0140-6736(98)91132-1.
139. Luchette FA, Jenkins WA, Friend LA, et al. Hypoxia is not the sole cause of
lactate production during shock. J Trauma 2002;52(3):415–9. Epub 2002,
Mar 20.
140. Constant JS, Feng JJ, Zabel DD, et al. Lactate elicits vascular endothelial
growth factor from macrophages: a possible alternative to hypoxia. Wound
Repair Regen 2000;8(5):353–60. Epub 2000, Dec 15.
141. Ohashi H, Kawasaki N, Fujitani S, et al. Utility of microdialysis to detect
the lactate/pyruvate ratio in subcutaneous tissue for the reliable monitoring
of hemorrhagic shock. J Smooth Muscle Res 2009;45(6):269–78.
142. van den Heuvel I, Vlasselaers D, Wouters PJ, et al. Serial lactate
measurements using microdialysis of interstitial fluid do not correlate with
plasma lactate in children after cardiac surgery. Pediatr Crit Care Med 2009;
10(1):66–70.
143. James JH, Wagner KR, King JK, et al. Stimulation of both aerobic glycolysis
and Na(+)-K(+)-ATPase activity in skeletal muscle by epinephrine or
amylin. Am J Physiol 1999;277(1 Pt 1):E176-86. Epub 1999, Jul 17.
144. Clausen T. Na+-K+ pump regulation and skeletal muscle contractility.
Physiol Rev 2003;83(4):1269–324. Epub 2003, Sep 25. doi: 10.1152/
physrev.00011.2003.
145. Schurr A. Lactate, glucose and energy metabolism in the ischemic brain
(Review). Int J Mol Med 2002;10(2):131–6. Epub 2002, Jul 18.
146. Payne RS, Tseng MT, Schurr A. The glucose paradox of cerebral ischemia:
evidence for corticosterone involvement. Brain Res 2003;971(1):9–17. Epub
2003, Apr 15.
147. Oddo M, Levine JM, Frangos S, et al. Brain lactate metabolism in humans
with subarachnoid hemorrhage. Stroke 2012;43(5):1418–21.
148. Mascia L. Acute lung injury in patients with severe brain injury: a double
hit model. Neurocrit Care 2009;11(3):417–26.
149. Ospina-Tascon GA, Cordioli RL, Vincent JL. What type of monitoring has
been shown to improve outcomes in acutely ill patients? Intensive Care
Med 2008;34:800–20.

Ventilation and
Pulmonary Function
Maurizio Cereda and Patrick J. Neligan
Introduction
Respiratory monitoring is an important aspect of managing
patients admitted to the intensive care unit (ICU), including
those with neurologic problems. Obviously, detecting gas
exchange abnormalities is essential to properly manage many
acute intracranial processes but neurologic patients often have
coexisting pulmonary diseases and up to a third of patients
admitted to a neurocritical care unit (NCCU) can develop
acute lung injury (ALI) or acute respiratory distress syndrome
(ARDS).1 Some possible causes for ARDS are listed in Table
20.1. These may require complex ventilator management,
which is facilitated by the availability of various types of pul-
monary physiologic data. Although gas exchange monitoring
is a well-established practice, the bedside measurement of
respiratory mechanics is still relatively underused in the ICU.
This stems from clinicians’ poor understanding of respiratory
physiology, and because the evidence about the impact of this
type of monitoring on outcomes is often confusing. This
chapter provides a physiologic background on pulmonary
function monitoring in the ICU, focusing on bedside mea-
surement of gas exchange and respiratory mechanics, and
highlights how these data can be used to support sound ven-
tilator management choices. In addition how to decide who
should be ventilated and criteria to determine liberation from
mechanical ventilation is discussed briefly.
Basic Respiratory Physiology
and What It Means for
Pulmonary Assessment
There are two compartments to the respiratory system: the
lung and the chest wall. The chest wall includes the abdomen
because abdominal pathology, including obesity or abdomi-
nal compartment syndrome can alter respiratory mechanics.
To inflate the respiratory system requires pressure to be gen-
erated because of its resistive and elastic properties. Elasticity
is expressed as compliance (volume change or pressure
change). Esophageal pressure and airway pressure both should
be measured to assess how each compartment of the respira-
tory system influences respiratory mechanics, particularly
© Copyright 2013 Elsevier Inc. All rights reserved.
Chapter
20  
II
compliance. Most resistance comes from the airways, but
small amounts are associated with tissue resistance, stress
relaxation, and gas distribution.
Respiratory mechanics can be evaluated by several tech-
niques. Among these the rapid airway occlusion technique that
estimates alveolar elastic recoil pressure by measuring the
inspiratory plateau airway pressure (Pplat) is a simple bedside
test for ICU use. Intrinsic positive end-expiratory pressure
(PEEPi) is another important variable that usually is measured
using end-expiratory airway occlusion. PEEPi has important
cardiopulmonary effects including decreased cardiac output,
alveolar overdistention, increased work of breathing, and
effects on compliance.
Imaging and Diagnostic Studies
A variety of imaging and diagnostic tools are available to
evaluate pulmonary disease in the ICU, to identify high-risk
populations, and as diagnostic tools to optimize mechanical
ventilation.2-7 Basic tests include the bedside chest x-ray
(CXR) and arterial blood gas analysis.8 Venous blood gas or
evaluation of electrolytes and in particular serum bicarbon-
ate may be useful for ventilator management when arterial
blood gas analysis is not available. In patients who are diffi-
cult to wean, assessment of nutritional (e.g., phosphate) and
endocrine status (e.g., cortisol and thyroid function) can be
helpful. Noninvasive tests include computed tomography
(CT), positron emission tomography (PET) or CT/PET,9,10
CT angiography for pulmonary embolus,11 electrical imped-
ance tomography (EIT), and lung ultrasound (LUS). Invasive
tests for diagnosis include bronchoscopy, bronchoalveolar
lavage, transbronchial lung biopsy, and open lung biopsies.
These invasive tests can be safely performed even in patients
with ARDS. Common complications include transient hypox-
emia, respiratory acidosis, and pneumothorax.12 Formal pul-
monary function tests such as spirometry, diffusing capacity
of the lung for carbon monoxide (DLCO), exercise testing, or
bronchial challenges are infrequently performed on critical
care patients.13 In the future biomarkers and genomic analy-
sis may become useful tools in diagnosis or identification
of patients at risk for pulmonary disorders such as ALI
or ARDS.14
189
190 Section II—Clinical and Laboratory Assessment

Table 20.1  Some Causes of Acute Table 20.2  American-European Consensus

Respiratory Distress Syndrome Definitions of ALI and ARDS Developed  
in 1994
Direct Lung Injury Indirect Lung Injury
Common Pneumonia Sepsis ALI CRITERIA
Aspiration of gastric Severe trauma with Timing: acute onset
contents shock and multiple Oxygenation: PaO2/FiO2 ≤300 mm Hg (regardless of PEEP level)
blood transfusions Chest radiograph: bilateral infiltrates on frontal chest
Less common Pulmonary Drug overdose radiograph
contusion Acute pancreatitis Pulmonary artery wedge pressure: ≤18 mm Hg when
Fat emboli Blood transfusion measured or no clinical evidence of left atrial hypertension
Near-drowning Pregnancy-related ARDS CRITERIA
Inhalational injury ARDS
Reperfusion Same as ALI except oxygenation: PaO2/FiO2 200 mm Hg
pulmonary edema (regardless of PEEP level)

Adapted from Ware LB, Matthay MA. The acute respiratory distress Adapted from Bernard GR, Artigas A, Brigham KL, et al. The American-
syndrome. N Engl J Med 2000;342:1334–48. European consensus conference on ARDS: definitions, mechanisms, relevant
ARDS, Acute respiratory distress syndrome. outcomes, and clinical trial coordination. Am J Respir Crit Care Med
1994;149:818–24.
ALI, Acute lung injury; ARDS, acute respiratory distress syndrome; FiO2,
inspiratory oxygen fraction; PaO2, arterial oxygen tension; PEEP, positive
end-expiratory pressure.
The bedside CXR can reveal abnormalities that may not be
clinically evident. In addition, a CXR can detect the malposi-
tion of tubes, including an endotracheal tube and catheters
and identify complications associated with these devices.15
Today quantitative CT is the gold standard to assess lung mor-
phology, recruitment, and hyperinflation of lung tissue at
Table 20.3  The Murray Lung Injury Score*
different inflation pressures. LUS is a useful, radiation-free, Variable Score
noninvasive bedside lung-imaging tool.16 Like CT it is primar- 1. Chest roentgenogram score
ily a tool to detect changes in density. EIT and PET provide No alveolar consolidation 0
more functional data.9 For example, EIT can provide a non- Alveolar consolidation confined to 1 quadrant 1
invasive continuous image of pulmonary impedance. This can Alveolar consolidation confined to 2 quadrants 2
be useful during mechanical ventilation because it indicates Alveolar consolidation confined to 3 quadrants 3
Alveolar consolidation in all 4 quadrants 4
the distribution of ventilation.3 PET also can be used to assess
lung inflammation10 that along with impaired gas exchange is 2. Hypoxemia score
PaO2/FiO2 >300 0
observed in ARDS, ALI, or ventilator-induced lung injury PaO2/FiO2 225-299 1
(VILI). PaO2/FiO2 175-224 2
PaO2/FiO2 100-174 3
PaO2/FiO2 <100 4
ARDS and ALI
3. PEEP score (when ventilated) (cm H2O)
Many patients in the ICU can develop ALI or ARDS, and PEEP ≤5 0
patients with severe acute brain injury often are at risk for PEEP 6-8 1
developing these conditions. Mortality from ALI and ARDS PEEP 9-11 2
PEEP 12-14 3
remains high and their development adversely affects intra- PEEP >15 4
cranial physiology.17-21 It is important then to know how to
4. Respiratory system compliance score (when
define these conditions, recognize who is at risk for them, and available) (mL/cm H2O)
diagnose their onset (Tables 20.1, 20.2, 20.3, and 20.4). ARDS Compliance >80 0
is a clinical-radiologic diagnosis and includes severe hypox- Compliance 60-79 1
emia assessed by an arterial oxygen tension/fraction of inspired Compliance 40-59 2
oxygen ratio less than 200 and bilateral infiltrates on a CXR Compliance 20-39 3
Compliance <19 4
in the absence of left atrial hypertension. The definition of
ARDS and ALI is provided in Table 20.2.22-24 Other sets of Adapted from Murray JF, Matthay MA, Luce JM, et al. An expanded
definition criteria have been proposed (see Tables 20.3 and definition of the adult respiratory distress syndrome. Am Rev Respir Dis
1988;138:720–3.
20.4). In children ALI and ARDS often are defined by the *The final score is calculated by the addition of the component parts:
oxygenation index defined as the product of mean airway score 0, no lung injury; score 1 to 2.5, mild to moderate lung injury;
pressure (MAP) × FiO2 × 100/PaO2.25 However, the sensitivity score >2.5, severe lung injury.
and specificity of the clinical diagnosis may depend on the FiO2, Inspiratory oxygen fraction; PaO2, arterial oxygen tension; PEEP, positive
end-expiratory pressure.
underlying etiology. Other diagnostic tests are used in large
part to exclude conditions that may mimic ALI or ARDS or
to guide therapy. Chest CT can help evaluate the extent of the
disease and guide ventilation and prognosis. Current therapy
is centered on lung-protective mechanical ventilation and a
restrictive fluid management strategy.
 Section II—Clinical and Laboratory Assessment 191

realization that these tracings have profound physiologic

Table 20.4  Comparative Analyses of meanings and that their analysis provides useful clinical
Commonly Used Definitions for ALI   information in ventilated patients.29 A more recent addition
and ARDS is the development of volumetric capnography, which is the
Definition Pros Cons analysis of CO2 over volume waveforms and provides further
pathophysiologic insight.30 Similar to pulse oximetry, cap-
AECC Simple and easy Acute onset: not defined
to use PAOP often not
nometry and capnography have become part of the clinical
Differentiates ALI measured PEEP/ routine in the OR, although their use is still somewhat limited
and ARDS compliance/MAP not in ICU patients.
Prognostic considered
capability based Risk factors not
on ARMA study emphasized Engineering and Technology
Murray score Takes PEEP/ Does not include MAP Pulse oximeters determine oxygen saturation (SpO2) using
compliance into Does not exclude heart
consideration failure
the spectrophotometric characteristics of pulsatile arterial
Differentiates mild Does not identify blood. Oxyhemoglobin absorbs light in the infrared spectrum
to moderate individual risk factors at 940 nm, whereas deoxyhemoglobin absorbs light preferen-
from severe Prognostic ability not tially at 660 nm, corresponding to red light. The pulse oxim-
lung injury validated eter probe consists of two diodes, each emitting red or
Radiologic criteria
more specific infrared light into tissue, and of sensors that measure the
emerging amount of radiation. SpO2 is obtained by process-
Delphi Defines criteria for Excludes P/F >200 and
onset (<72 hr) <300
ing the relative absorption of light by oxygenated and non-
Risk factor Does not include oxygenated hemoglobin. The device reports only the pulsatile
emphasized compliance or MAP component of the recorded signal, and the final result is a
Takes PEEP into continuous measurement of the patient’s oxyhemoglobin
consideration status and pulse rate. Calibration is done in volunteers who
Objectively rules
out heart failure breathe hypoxic mixtures of gas. Standard, bi-wavelength
pulse oximetry does not detect the presence of dyshemoglo-
Oxygenation Takes MAP into Does not take PEEP and
index consideration compliance into
binemias, but multiwave pulse oximeters recently have been
Prognostic ability consideration developed. These instruments also can estimate total hemo-
validated Does not exclude heart globin concentration together with carboxyhemoglobin and
failure Does not methemoglobin levels.31
evaluate radiologic Clinical capnometry is performed using a source of infrared
signs
light and measuring the absorption by CO2 at various wave-
Adapted from Raghavendran K, Napolitano LM. Definition of ALI/ARDS. Crit lengths, using a photodetector within a sampling cell. The
Care Clin 2011;27(3):429–37. measured light absorption is then compared with a CO2-free
AECC, American European Consensus Conference; ALI, acute lung injury;
ARDS, acute respiratory disorder syndrome; ARMA, acute respiratory
reference, and finally CO2 is expressed either as partial
management in ALI/ARDS; MAP, mean arterial pressure; PAOP, pulmonary pressure or as concentration. There are two types of capnom-
artery occlusion pressure; PEEP, positive end-expiratory pressure; P/F, PaO2/FiO2 eters: (1) mainstream capnometers measure CO2 directly in a
ratio.
cuvette placed in the ventilator circuit, close to the endotra-
cheal tube, and (2) sidestream capnometers aspirate gas from
the airway opening using a small-bore tubing line; CO2 is then
Gas Exchange Monitoring measured in a chamber placed within the monitor. Thus CO2
measurements lag behind the respiratory cycle in sidestream
History capnometry. All modern capnometers also provide a capnog-
Blood gas analysis was introduced by Severinghaus in 1958 and raphy tracing on the display screen. Volumetric capnographs
rapidly became an essential part of the clinical routine.26 This combine a mainstream capnometer with a flowmeter within
technique provides direct measurement of arterial blood PO2, the same device,30 thus providing measurements of CO2
PaCO2, and pH and also allows for calculation of serum bicar- production and dead space. In this technique, flow and CO2
bonate, base deficit-excess, and oxyhemoglobin saturation. signals need to be synchronized for measurements to be
The last can be directly measured using co-oximetry, which meaningful.
also measures carboxyhemoglobin and methemoglobin.
The history of pulse oximetry started in the 1930s, with
the first studies on the absorption of light by oxygenated Indications for Use
blood. However, the first clinically acceptable instruments Pulse oximetry is currently part of the basic ICU monitoring
became available only in the 1970s.27 Their introduction and is applied to almost all patients irrespective of their clini-
made bedside monitoring of arterial oxygenation possible in cal status: in fact SpO2 is often referred to as the “fifth vital
the ICU, in the operating room (OR), and in many other sign.” The main purpose of SpO2 monitoring is to provide an
hospital locations. Capnometry, the measurement of carbon early warning of hypoxemia and to avoid its complications. In
dioxide (CO2) in the expired gas, was first made possible by ventilated patients, SpO2 also is used to titrate oxygen therapy32
infrared photometry in the 1940s. A subsequent step was the and positive end-expiratory pressure (PEEP). This decreases
introduction of capnography, which is the analysis of expired the need for frequent arterial blood sampling.33 Capnometry
CO2 versus time waveforms.28 This advancement followed the and capnography are considered a standard in the OR, where
192 Section II—Clinical and Laboratory Assessment
they are mainly used to assess the adequacy of the airway and
of ventilation. In particular, capnometry is the most effective
method to verify successful intubation.34 In the ICU, capnom-
etry is mostly used for select patient populations, such as those
with intracranial hypertension, where it provides a surrogate
for continuous monitoring of arterial PaCO2, the control of
which can be important in management of increased intra-
cranial pressure (ICP). Capnometry also can be used to facili-
tate ventilator weaning and at the same time limit blood
sampling.35 However, there are limitations of this monitor as
an estimate of arterial PaCO2 that should not be disregarded
in patients with lung diseases (see following text).
Information Provided  
and Troubleshooting
Pulse oximetry is used to measure arterial saturation nonin-
vasively and accurately predicts arterial PO2.36 However, the
oxyhemoglobin dissociation curve has a sigmoid shape, and
on its flat portion large PO2 decreases may cause only minimal
changes in SpO2 and be underestimated. Furthermore, the
accuracy and precision of pulse oximetry measurements
decrease when hemoglobin saturation is less than 90%.37 For
these two reasons, any decrease in the SpO2 below 90% must
be considered a significant clinical event and assumed to
predict hypoxemia. Saturations of less than 80% do not convey
any meaningful information other than that the patient is
severely hypoxemic, because no calibration data on human
subjects are available in this range. The use of standard pulse
oximeters is limited in the presence of dyshemoglobinemias,
and thus co-oximetry should be performed whenever carbon
monoxide poisoning is suspected. Hypotension, vasoconstric-
tion, and hypothermia reduce the pulsatility of capillary blood
and the ability of the pulse oximeter to identify a tracing. In
addition, dark skin pigmentation decreases the accuracy of
SpO2 measurement.32 This same study also suggested that a
higher SpO2 goal should be used when titrating oxygen therapy
in black or dark-skinned rather than fair-skinned patients.
Capnography tracings are characterized by stages that
correspond to specific moments of the respiratory cycle
(Fig. 20.1, A). At the start of exhalation, measured PaCO2 is
negligible because the airways contain only fresh gas (point
1). The volume of the airways is called anatomic dead space
because the gas they contain does not participate in gas
exchange. As expiration proceeds, CO2-rich alveolar gas enters
the airways, causing a rapid increase in PaCO2 (points 2 to 3).
When the dead space has been completely washed out by
alveolar gas, the exhaled PaCO2 curve reaches a plateau (points
3 to 4). During inspiration, fresh gas enters the airways and
PaCO2 rapidly drops to zero again (point 5). The highest mea-
sured PaCO2, identified at the end of expiration (point 4) is
called end-tidal CO2 (etCO2) and is indicative of the PaCO2 in
the alveoli. Because the alveolar gas is in equilibrium with the
pulmonary capillary blood, etCO2 is also very close to arterial
PaCO2 in healthy subjects and it can be used to adjust ventila-
tion. However, etCO2 may underestimate arterial PaCO2 in the
presence of diseases that cause significant alveolar dead space,
such as chronic obstructive pulmonary disease (COPD), pul-
monary embolism, and ARDS.38 Alveolar dead space is the
wasted fraction of inspired gas that does not participate in
CO2 exchange because of the presence of alveolar units with
low or absent perfusion relative to ventilation. In this case, gas
PaCO2
3 4
1 2 5
A
PaCO2
B
Fig. 20.1  A, Schematic representation of the lungs (top), with the
alveolar spaces filled with CO2-rich gas (dark blue) and the airways
containing fresh gas (light blue); bottom: normal capnometry tracing
(blue line) with arterial PaCO2 (red line). The stages of the capnometry
tracing are marked with numbers (see text for explanation). B, Schematic
representation of the lungs (top), where low-CO2 gas exhaled by
hypoperfused alveolar units mixes with CO2-rich gas from well perfused
units, causing a decrease in the expired PaCO2 to levels that are
significantly lower than in the arterial blood (bottom).
exhaled by such units contains very low or negligible CO2, thus
causing etCO2 to be lower than the arterial PaCO2 (Fig. 20.1,
B). The difference between the two variables is proportional
to the fraction of alveolar dead space. Thus in patients with
lung disease, etCO2 should never be assumed to accurately
represent arterial PaCO2 unless an association between the
two variables can be confirmed by arterial blood gas analysis.
When there is a discrepancy, the trend in etCO2 still can be
used. Volumetric capnography allows assessment of both ana-
tomic and alveolar dead space, together with CO2 production.
These variables can be used to titrate mechanical ventilation,
to detect changes in hemodynamic status and pulmonary
embolism, and to confirm diagnoses and prognoses.30
Analysis of capnographic tracings provides further infor-
mation, and the clinician should be aware of a number of
abnormal patterns. In patients with COPD and asthma, the
upstroke of expired CO2 is slower than normal and a plateau
stage is never reached. In this case, etCO2 and arterial PaCO2
cannot be determined. This phenomenon results from severe
dishomogeneity of expiration kinetics among different alveo-
lar units. An exponential decrease in etCO2 over minutes is
an ominous sign, because it indicates a rapid decrease in pul-
monary perfusion, which can be due to pulmonary embolism,
sudden onset of shock, or impending cardiopulmonary arrest.
Capnography also can be used to detect leaks in the breathing
 Section II—Clinical and Laboratory Assessment 193
circuit, disconnection from the ventilator, and a patient’s inspi-
ratory attempts.
Therapy Based on Gas Exchange
Monitoring and Its Effect on Outcome
Both pulse oximetry and capnometry became widely accepted
for clinical use in the mid- to late 1980s. Since then, multiple
medical societies have recommended the use of these moni-
tors, particularly in the OR but also in other settings. It is very
likely that the introduction of SpO2 and capnometry has con-
tributed to improved perioperative morbidity and mortality.
This statement is based mostly on an analysis of closed anes-
thetic malpractice claims that suggested that a majority of
intraoperative respiratory mishaps would have been avoided
had pulse oximetry and capnometry been available.39 The
same dataset also showed a historic trend of reduction in
respiratory-related claims following introduction of these
monitors. However, the strength of clinical evidence that sup-
ports the use of capnometry and pulse oximetry in the peri-
operative setting and in the ICU is disappointing, although
there is evidence to suggest use of pulse oximetry makes a
difference to patient care. First, randomized studies conducted
in the perioperative period show that the use of pulse oxim-
etry increases the rate of detection of hypoxemia.33,40 Second,
Moller et al.40 in a randomized study that included 28,802
patients observed that patients who received SpO2 monitoring
had a lower incidence of intraoperative electrocardiographic
ischemic changes than control patients who did not have a
SpO2 monitor. However, the overall rate of perioperative com-
plications and the mortality were similar in the two groups.
In a smaller study from the same group, the rate of cognitive
dysfunction was similar in patients who were or were not
monitored with pulse oximetry.41 A 2003 Cochrane review
concluded that although pulse oximetry improved detection
of hypoxemia, its utility and cost effectiveness as a tool to
improve outcomes is questionable, at least in the perioperative
period.42 This observation was confirmed in a subsequent
Cochrane analysis of five randomized trials that included
22,992 patients and compared pulse oximetry or no pulse
oximetry during the perioperative period (i.e., there was no
difference in mortality although the incidence of hypoxemia
in the recovery room was less in the pulse oximeter group).43
However, given the wide acceptance of this monitoring tech-
nique, it is unlikely that further randomized studies will be
performed in the future.
Capnometry, and particularly volumetric capnography
provides helpful physiologic information that can be used to
guide mechanical ventilation and weaning. Dead space is also
an independent predictor of mortality in ARDS patients.38
However, there is a lack of outcome evidence that this form
of monitoring is helpful in the ICU, other than for the confir-
mation of intubation.34
Risks
Pulse oximetry and capnometry are essentially risk-free tech-
niques, from the point of view of device safety. Similar to any
other type of monitoring, it can be argued that patient harm
may result from poor interpretation of the data and lack of
understanding of the technique, rather than from the monitor
itself. An example of a common misunderstanding is the
assumption that a good SpO2 value also reflects adequate ven-
tilation. In reality, oxygen therapy in a hypoventilated patient
may effectively correct hypoxemia while masking severe
hypercapnia, thus giving clinicians a false sense of security.
This phenomenon could be the cause of adverse respiratory
events during procedural sedation and can be prevented if
capnometry is performed with pulse oximetry.
Respiratory Mechanics
and Ventilator Variables
History
The conceptual background to respiratory mechanics moni-
toring is provided by classic pulmonary physiology. Studies on
pressure-volume loops obtained during inflation and defla-
tion of surfactant deficient lungs44 showed characteristic
abnormalities45 that later in time also were observed in patients
with ARDS.45,46 However, it was the introduction of the rapid
flow interruption technique46,47 that brought the measurement
of respiratory system compliance and resistance to the bedside.
Respiratory mechanics have become more relevant in critical
care since the demonstration that excessive alveolar inflation
leads to ventilator-associated lung injury (VALI)48 and can
worsen patient outcomes.49
The use of esophageal pressure (Pes) to measure work of
breathing and to separately study the mechanics of the lung
and of the chest wall was initially tested in studies on healthy
volunteers and in patients with COPD.50 In the 1990s, this
technique was extended to ICU patients and led to an
understanding of the importance of chest wall mechanics51,52
and patient-ventilator dyssynchrony (PVD) that can increase
the work of breathing.53 Impulse oscillometry can be used
as a noninvasive method to assess pulmonary compliance
rather than using the esophageal pressure method, which is
invasive.54
The advent of CT of the chest in ARDS patients led to the
demonstration that ventilation is not homogeneous and that
atelectasis is relevant in this syndrome.55 In particular, CT is
able to reveal discrepancies between bedside chest x-rays and
various clinical and physiologic parameters in ARDS patients.2
For a long time, PEEP was used to achieve goal oxygenation
levels through alveolar recruitment in ARDS and in other
forms of acute respiratory failure. The discovery of cyclic
alveolar collapse and reopening in human ARDS,56 and animal
evidence that these phenomena contribute to VALI57 led to the
formulation of the “open lung concept.” According to this
idea, optimization of alveolar recruitment may improve
patient outcomes and should be a goal of ventilator manage-
ment, independent from the oxygenation response. Although
the results of randomized studies on this topic have been
contradictory,58,59 the open lung concept has renewed the
interest in bedside clinical tools to measure alveolar recruit-
ment, lung volumes,60 and regional patterns of ventilation.61
Engineering and Technology
Monitoring respiratory mechanics requires pneumotacho-
graphs and electromechanical transducers to measure gas
flows and airway pressures (Paw). Pneumotachographs quan-
tify flow from the pressure drop generated by the passage of
gas through a known resistance, which typically includes a
194 Section II—Clinical and Laboratory Assessment

thin metal grid to maintain laminar gas flow. Other designs cmH2O
25
are available, although this is the most frequently used flow-
meter. A differential pressure transducer converts the pres- 20
sure difference across the resistance into an electrical signal,
which then is amplified, calibrated, and processed. Volumes 15
are obtained by mathematical integration of the measured

Paw
10
flow. Transducers typically are embedded in the ventilator
machinery, remotely from the patient and serve the dual 5
purpose of feeding the ventilator servocontrols and of pro-
0
viding data for display. Modern ventilators also display curves,
facilitating detection and interpretation of events. In certain –5
ventilators, flows and pressures are measured at the proximal
opening of the endotracheal tube, bypassing the ventilator Lpm
tubing. This design renders the measurements more accurate
but exposes the transducers to vapor condensation and 100
mucus accumulation. 50
Many ventilators allow the clinician to perform a rapid flow
interruption maneuver at the end of inspiration or expiration 0

Flow
by pushing a dedicated button on the ventilator console. –50
During this maneuver both the inspiratory and the expiratory
valves of the ventilator remain closed for a short time and so –100
provide conditions of zero flow. This maneuver is fundamen- –150
tal to the measurement of respiratory mechanics variables at
the bedside. An algorithm that automatically performs repeti- –200
tive, short-flow interruptions and provides frequently updated
measurements is available, but it has been created specifically L
to be implemented in the proportional assist ventilation 1.50
(PAV+) mode.62 1.25
Pes is measured using a thin, 10-cm-long balloon partially 1.00
filled with air and inserted into the lower third of the esopha-
Volume

gus.50 The use of central venous pressure as a surrogate for Pes
also has been described.63 Pes measurement is included in
pulmonary monitoring systems provided with disposable
esophageal catheters and pneumotachographs.
Pressure volume loops are obtained by step-wise or con-
tinuous inflations-deflation cycles of the lungs, which are
performed using either a large-volume syringe or a ventilator
with dedicated software.64 Graphic or mathematical analysis65
is used to extrapolate clinically useful information from
these loops.
EIT is a relatively new bedside imaging technique that
constructs an image of gas distribution within the chest
during the different phases of respiration using a series of
electrode pairs placed around the circumference of the
chest3,61 Potential applications for this technique include
titration of PEEP, recruitment maneuver, and assessment of
regional patterns of ventilation.66-69 Although commercial
devices are available, their clinical application has still to be
fully elucidated.5 Techniques that measure lung volumes
through gas dilution or washout techniques are in the process
of being tested and may be used in the future to help opti-
mize ventilator parameters.60
Indications for Use
Monitoring of Paw, tidal volume (TV), expired minute volume
(VE), and other basic ventilator variables is essential to the
care of all ventilated patients. These data document the effec-
tiveness of ventilation, they signal disconnections from the
ventilator and airway obstruction, and they provide early
warnings of changes in patient condition. In many institutions
these are the only variables that are collected during routine
0.75
0.50
0.25
0
2
–0.25
Fig. 20.2  An example of patient-ventilator asynchrony. From top
to bottom: airway pressure (Paw), flow, and volume versus time tracings
during assist control ventilation. In this example the ventilator does not
match the patient’s demand for flow and volume. This causes Paw to
decrease and to have a characteristic indentation during inspiration. This
phenomenon can cause patient discomfort unless inspiratory flow and
volume settings are adjusted. L, Liters; Lpm, liters per minute.
patient management and are probably sufficient in the major-
ity of patients. However, slightly more advanced monitoring
is easy to perform and can be very useful in many cases. First,
visual inspection and interpretation of Paw and flow versus
time tracings can help detect asynchronies between patient
and ventilator70 (Fig. 20.2). Furthermore, the flow interrup-
tion method should be used periodically in every ventilated
patient if the machinery allows it. This technique is associated
with no additional risk or cost and allows the measurement
of compliance, inspiratory resistance, and intrinsic PEEP
(PEEPi), all variables that can rapidly reveal changes in patient
conditions.
Monitoring inspired volumes and pressures can help detect
the presence of excessive alveolar inflation. This is important
because alveolar inflation has outcome implications.49 Evi-
dence suggests that integrating Pes in pulmonary monitoring
 Section II—Clinical and Laboratory Assessment 195

V′i Paw − Palv

Ri =
V′i
Paw
∆V
Crs =
∆Palv
Chest
wall PI = Palv − Pes

Palv ∆V ∆V
CI = Ccw =
Pes ∆PI Pes
A

Pip
Pplat
Paw
Paw
PEEPi
V′i PEEP
Flow

TV Flow
Volume

B 1 2 Time C Time
Fig. 20.3  A, The respiratory system is schematically represented as two compartments in series: the lungs and the chest wall. During the delivery of
inspiratory flow (V′i), airway pressure (Paw) is higher than the alveolar pressure (Palv), due to inspiratory resistance (Ri). Palv allows calculation of the
compliance of the respiratory system. If esophageal pressure is known (Pes) it is possible to calculate transpulmonary pressure (Pl) and, finally, the
compliance of the lung (Cl) and of the chest wall (Ccw). B, Paw, flow, and volume tracings during a rapid flow interruption maneuver are shown.
At time 1 the peak inspiratory pressure (Pip) is reached and the inspiratory flow is stopped. After a brief transient stage Paw reaches a plateau pressure
(Pplat) at time 2 and approximates Palv. C, Paw, flow and Palv (blue dashed line) versus time tracings in the presence of intrinsic positive end-expiratory
pressure (PEEPi). At the end of expiration, Palv is abnormally higher than the applied level of PEEP, as also suggested by the presence of residual end-
expiratory flow. When an expiratory flow interruption is performed, Paw equilibrates with Palv and rapidly increases. Intrinsic PEEP (PEEPi) is quantified
as the difference between Paw after the interruption and the applied PEEP. TV, Tidal volume.

provides more accurate estimates of alveolar distention than
Paw alone,71,72 particularly when chest wall mechanics are
abnormal. Pressure volume curves analysis is used in certain
institutions to identify the optimal level of PEEP in ARDS
patients and has been used in clinical studies that evaluate
ventilation strategies in ARDS.58,73 However, the precise physi-
ologic meaning of the information obtained is still to be
defined.
Information Provided   sis, pneumonia, and cardiogenic pulmonary edema. In many
and Troubleshooting
The respiratory system requires the generation of pressure for
its inflation, as a result of its resistive and elastic properties
(Fig. 20.3, A). Inspiratory resistance (Ri) is generated mainly
in the endotracheal tube and in the bronchial tree, with an
additional viscoelastic or maldistributive component.46 Ri is
calculated as the ratio between the pressure drop across the
airways and the inspiratory flow (V′i). Ri varies with the size
of the endotracheal tube and with V′i, but values higher than
10 cm H2O/L/second typically are abnormal. Ri is abnormally
increased in COPD74 and asthma, in the presence of proximal
airway and endotracheal tube obstruction, and, to a lesser
degree, in ARDS.52 The elastic recoil of the respiratory system
is quantified as compliance (Crs), which is the ratio between
inspired volume and the corresponding increase of alveolar
pressure (Palv), or as its reciprocal, elastance. Because the lung
and the chest wall are compartments in series, the actual dis-
tending pressure of the lung is the transpulmonary pressure
(Pl), the difference between Palv and pleural pressure. The
latter can be estimated by measuring Pes. Furthermore the
value of Crs is determined by the compliance of both the lung
(Cl) and the chest wall (Ccw). In pulmonary physiology, the
chest wall also includes the abdominal compartment.
Normal Crs values are 70 to 80 mL/cm H2O; in normal
individuals Crs is affected by body size. A Crs value less than
50 mL/cm H2O is typical in ARDS and can be used to define
this syndrome and to grade its severity.75 Crs also can be lower
than normal in many other acute conditions such as atelecta-
patients the change in Crs results from a decrease in Cl.76
However, in ARDS many patients also have a decrease in
Ccw.52 This finding typically is from abnormally high intra-
abdominal pressure and may be related to a nonpulmonary
etiology of respiratory failure.77 Chest wall compliance is
decreased in patients with COPD exacerbations.78
PEEPi is another important measurement in ventilated
patients. This is the pressure that remains in the alveoli at
the end of expiration, above the applied level of PEEP.79
PEEPi is caused by prolonged expiration, by flow-limitation
phenomena that are common in COPD patients, by short
expiratory times, or more commonly by a combination of
these factors.74
During pressure-controlled modes of mechanical ventila-
tion, inspiratory Paw is preset and it is important to monitor
volumes to detect respiratory mechanical abnormalities.
During volume-controlled ventilation modes, it is TV that is
fixed, and any alterations of respiratory mechanics manifest
196 Section II—Clinical and Laboratory Assessment
themselves with changes of Paw. In particular, the peak inspi-
ratory pressure (Pip) is the first variable to alert the clinician
that the patient has a mechanical problem. However, Pip is
measured in the proximal airway and its measurement does
not allow for distinction between elastic and resistive prob-
lems because Palv cannot be directly measured. However, Palv
can be estimated using the rapid flow interruption technique,47
which provides conditions of zero flow at the end of inspira-
tion and a stable plateau pressure (Plat) (Fig. 20.3, B). In these
conditions, gas is in equilibrium throughout the airways and
Pplat is close to Palv. Pplat is a function of the elastic recoil of
the respiratory system, if respiratory muscles are inactive.
Knowing Pplat allows the computation of Crs because:
Crs = TV/[Pplat − PEEP + PEEPi]
PEEPi should not be disregarded because it leads to falsely low
compliance values if not accounted for. PEEPi is quantified as
the difference between the pressure measured during an expi-
ratory flow interruption and the level of PEEP set on the
ventilator (Fig. 20.3, C). Pplat is important because it is the
only bedside method to estimate alveolar distention at the end
of inspiration. If Pes is available, it is also possible to measure
Ccw and Cl with the same maneuver. It must be remembered
that Pes can have a bias in the supine position because the
weight of the heart increases its value. However, tidal changes
in Pes still are valid.
It is possible to assume that the difference between Pip and
Pplat (see Fig. 20.3, B) is the resistive pressure drop across the
airways and that it can be used to estimate Ri because Ri =
(Pip − Pplat)/V′i. It is important that different measurements
of Ri are obtained at comparable V′i values because its Ri flow
dependent, due to turbulence. Although Crs can be measured
with any type of ventilator mode and any pattern of flow
delivery, it is preferable to use a volume-controlled mode with
a constant flow waveform to measure Ri.
The flow interruption method does not require muscle
paralysis as long as a constant Paw is observed for at least 0.25
second.62 When this is not possible, one can temporarily
increase TV or inspiratory flow to minimize respiratory drive
at the end of inspiration. Ventilator software applications
automatically calculate Crs and Ri, but they may not recognize
a patient’s muscle activity or gas leaks that render the Pplat
measurement void. Thus visual inspection of the tracings and
manual computation of the variables are advisable. The flow
interruption maneuver should be repeated multiple times and
the results averaged.
Volume-pressure curves in patients with ARDS have a char-
acteristic biphasic slope (Fig. 20.4): pressure application at low
lung volume yields little increase in volume until a lower
inflection point (LIP) is reached. At this point compliance
improves dramatically and the slope of the curve increases. As
pressure continues to increase, an upper inflection point
(UIP) is reached, where the curve flattens again. The LIP is
thought to represent the pressure required for alveolar recruit-
ment,80 which is the volume of intrapulmonary gas that is
gained by reopening alveolar units. The UIP is likely related
to excessive alveolar distention.81 These two variables are used
to identify optimal PEEP and TV settings, with the goal of
maximizing alveolar recruitment and minimizing alveolar dis-
tention.82 However, the value of the LIP may not correspond
to the PEEP required to maximize recruitment, and the
Volume
UIP
LIP
Pressure
Fig. 20.4  Schematic representation of volume-pressure curves in a
healthy subject (blue line) and in a patient with acute respiratory distress
syndrome (ARDS) (red line). LIP, Lower inflection point; UIP, upper
inflection point.
pressure volume curves do not seem to predict pulmonary
volume during mechanical ventilation.83
Therapy Based on Respiratory  
Mechanics and Ventilator Variables  
and the Effect on Outcome
Respiratory mechanics measurements can be used to diagnose
new conditions, to follow their clinical course, and to monitor
the response to therapy. There is also evidence that Crs is
better than oxygenation deficit, as a predictor of mortality in
ARDS.38 In daily practice, respiratory mechanics help distin-
guish between processes that have radically different manage-
ment approaches, particularly when multiple abnormalities
coexist (i.e., bronchospasm in a patient with ARDS). In addi-
tion, the measurement of Ri can be used to monitor the
response to bronchodilator therapy.84 The detection of signifi-
cant levels of PEEPi is very important because this can cause
hemodynamic impairment, elevated work of breathing, and a
patient’s inability to trigger the ventilator.85 If not recognized,
PEEPi can lead to very poor clinical choices (i.e., unnecessary
fluid administration in a hypotensive patient). PEEPi can be
addressed with changes in ventilator settings, such as decreased
respiratory rate and inspiratory times, and only in refractory
cases with increased sedation.
Pplat is used to monitor the level of alveolar distention
during mechanical ventilation. A TV reaching a Pplat of 30 cm
H2O should inflate alveoli approximately to the level of total
lung capacity. Beyond this value the alveolar walls receive a
pathologic stress that generates mechanical and then inflam-
matory injury that eventually leads to VALI.86 Patients with
ARDS are particularly exposed to VALI because of the typical
decrease in lung volumes, from alveolar edema or atelectasis.55
Thus ventilation is distributed to a smaller than normal popu-
lation of alveolar units, and each inspiration may cause exces-
sive alveolar distention and stress, even with TVs that are
acceptable in healthy patients. Maintaining Pplat below 30 cm
H2O should guarantee that inspiratory alveolar inflation is not
greater than “physiologic.” In fact, Pplat and TV limitation is
part of the lung protective strategy that was shown to have
 Section II—Clinical and Laboratory Assessment 197
outcome benefits in ARDS.49 Several points deserve discus-
sion. First, a safety limit for Pplat may not exist because an
outcome benefit from TV limitation is observed even when
Pplat is less than 30 cm H2O.87 Second, elevated TVs probably
are harmful in non-ARDS populations and even in patients
with healthy lungs, irrespective of their Pplat values.88 Third,
Cl and Ccw both affect Pplat. When Pes and Pl are measured,
excessive alveolar distention can be identified even in the pres-
ence of acceptable Pplat, as suggested by a large change in Pl.72
In these patients further reductions in TV and Pplat may be
necessary. By contrast, patients with low Ccw, which is typi-
cally due to intra-abdominal hypertension, can have high
Pplat with acceptable lung distention and may not need ven-
tilator adjustments.77
It is still unclear how alveolar recruitment should be mea-
sured and whether any of the available bedside tools can be
considered acceptable for clinical practice. Early studies sug-
gested that Crs could be used to define the optimal PEEP
setting.89 However, for reasons that are still incompletely
understood, other studies did not always detect a correlation
between compliance and recruitment when PEEP was
changed90. The appeal of the pressure-volume curve has
increased since it was used to set PEEP in clinical trials that
have tested the open lung concept, although there are concerns
about the methodology of these studies.58,73 Other techniques
to assess recruitment at bedside, such as EIT and gas dilution
or washout techniques are promising but have not yet been
validated in outcome studies. An innovative strategy that
includes Pes measurement in the setting of PEEP has been
proposed.91 Its rationale is that Pes and Pl can be used to
identify the level of PEEP that guarantees adequate expiratory
lung distention. With this approach the authors demonstrated
a benefit in improved oxygenation and Crs compared with a
control group in which PEEP was set using a conventional
PEEP/FiO2 nomogram.
Risks
Monitoring respiratory mechanics and the use of the flow
interruption technique is associated with little risk to the
patient and can be performed in any condition. Measuring Pes
requires the insertion of esophageal catheters. However, these
are typically thin and soft tipped, and to these authors’ knowl-
edge no cases of morbidity from this method have been
reported. Measuring volume-pressure curves may require dis-
connection from the ventilator and so may expose the patient
to hypoxemia and minor hypercarbia, but the procedure is
overall well tolerated by most patients.92
Decisions About Intubation,
Weaning, and Liberation from
Mechanical Ventilation
Ventilation and pulmonary function are monitored in the
ICU, in large part to assess the severity of pulmonary disease,
its progress, and how a patient responds to therapy. In the ICU
this often means making decisions about intubation and
mechanical ventilation.
Which patient should be intubated? The specific indications
for endotracheal intubation are difficult to define, and a
variety of parameters need to be considered. In general, the
indications can be divided into three broad groups: (1) hypoxic
respiratory failure, (2) hypercarbic ventilatory failure (includ-
ing cardiac arrest), and (3) impaired consciousness and airway
protection.93 In addition, patients in the NCCU may require
intubation to facilitate specific therapies (e.g., to help reduce
ICP or manage status epilepticus). The decision to intubate
a patient may be complicated because there are few data
to support specific guidelines, and in recent years a variety
of noninvasive forms of ventilation have become available.
Nevertheless, the adage “the indication for intubation and
mechanical ventilation is thinking of it”94 is a simple guide to
remember.
In the NCCU intubation and mechanical ventilation are
indicated in the following conditions: (1) depressed con-
sciousness and in particular a Glasgow Coma Scale score less
than or equal to 8; (2) inability to protect the airway or clear
secretions; (3) need to reduce ICP by ventilation control; (4)
PaO2 less than 60 mm Hg despite supplemental O2; (5) PaCO2
greater than 50 mm Hg, or pH less than 7.2; (6) respiratory
rate greater than 40 breaths per minute or less than 10 per
minute; (7) muscle fatigue; (8) airway compromise (e.g., a
neck hematoma after anterior cervical surgery); and (9)
hemodynamic instability.
Although pulse oximetry can be used to assess for hypox-
emia, it is best assessed using blood gas analysis because oxim-
etry measures the saturation of hemoglobin and not PaO2, and
it can be unreliable in patients with severe anemia, carbon
monoxide poisoning, methemoglobinemia, sepsis, or periph-
eral vasoconstriction.95 Furthermore, normal oxygen satura-
tion does not exclude hypoxemia, particularly in patients on
a high FiO2. In these patients evaluation of the alveolar-arterial
oxygen gradient can help because a widening alveolar-arterial
oxygen gradient indicates worsening hypoxemia. Bedside
monitors to detect hypercarbia are frequently used in the OR
but are less commonly available in the ICU.35 This can be
important because oxygen saturation obtained on pulse oxim-
etry can be normal when there is significant hypoventilation.
Instead the diagnosis of hypoventilation depends on clinical
findings and on arterial blood gas analysis.
Weaning and Liberation from
Mechanical Ventilation
Weaning often is used to describe the transition from intuba-
tion and mechanical ventilation to a spontaneous breathing
patient who has a protected airway, but strictly speaking it
refers to the progressive reduction in the amount of support
provided by a mechanical ventilator. This and liberation from
mechanical ventilation are fundamental aspects of critical care
and are important because a longer duration of mechanical
ventilation is associated with increased cost and mortal-
ity.19,96,97 In addition, up to half of the time spent on mechani-
cal ventilation happens after the weaning process is started.19,97
Recent consensus guidelines suggest that the weaning process
should begin shortly after intubation and be considered in six
discrete steps: (1) treatment of acute respiratory failure, (2)
clinical judgment that weaning may be possible, (3) assess-
ment of the readiness to wean, (4) a spontaneous breathing
trial, (5) extubation, and (6) possible reintubation.98
ICU patients may be classified as: (1) simple to wean (i.e.,
extubated on the first attempt; about 60% of general ICU
198 Section II—Clinical and Laboratory Assessment
Table 20.5  Factors Associated with Readiness
to Wean Ventilator Support in NCCU Patients
CLINICAL ASSESSMENT
• Adequate cough
• Absence of excessive tracheobronchial secretions
• Resolution of disease for which the patient was intubated
• Normal intracranial pressure (ICP)
• Patient awake and cooperative
• Chest wall pain well controlled
OBJECTIVE MEASUREMENTS
• Clinical stability
• Stable cardiovascular status (i.e., heart rate ≤140 beats/
min; systolic BP 90-160 mm Hg, no or minimal
vasopressors)
• Stable metabolic status
• Adequate oxygenation
• SaO2 >90% on FiO2 ≤0.4
• PaO2 ≥50-60 mm Hg on FiO2 ≤0.5 (or PaO2/FiO2
≥150 mm Hg)
• PAO2-PaO2 gradient <350 mm Hg
• PEEP ≤8 mm Hg
• Adequate pulmonary function
• Respiratory rate ≤35 breaths/min
• Maximal inspiratory pressure ≤ –20 to –25 cm H2O
• VT >5 mL/kg
• VC >10 mL/kg
• Rapid shallow breathing index (respiratory rate/VT) <105
breaths/min/L
• No significant respiratory acidosis
Adapted from Boles J, Bion J, Connors A, et al. Weaning from mechanical
ventilation. Eur Respir J 2007; 29:1033–56.
BP, Blood pressure; FiO2, inspiratory oxygen fraction; NCCU, neurocritical
care unit; PaO2, arterial oxygen tension; PAO2, alveolar oxygen tension;
PEEP, positive end-expiratory pressure; SaO2, arterial oxygen saturation;
VC, vital capacity; VT, tidal volume.
patients), (2) difficult to wean (i.e., up to three attempts or up
to 7 days from the onset of weaning), or (3) prolonged wean
(i.e., more than three attempts or more than 7 days from the
onset of weaning).98 Decisions about the readiness to wean
requires two interdependent steps: assessment of predictors of
weaning and successful completion of a spontaneous breath-
ing trial. Factors associated with readiness to wean are listed in
Table 20.5.98,99 No single finding is specific for weaning success.
Instead the factors listed in Table 20.5 identify patients who
are not suitable for weaning (i.e., a spontaneous breathing trial
in these patients may have adverse effects). All other patients
should undergo a spontaneous breathing trial that may be
accomplished using a T-tube or T-piece, pressure support ven-
tilation (PSV), or automatic tube compensation (ATC). Each
of these may or may not be used with continuous positive
airway pressure (CPAP). The criteria that indicate failure of a
spontaneous breathing trial are listed in Table 20.6.98,100,101
The next step in liberation from mechanical ventilation is
extubation. This requires an assessment of the patient’s ability
to protect and maintain a patent airway based on: (1) level of
consciousness, (2) cough strength, (3) frequency of secretions,
and (4) airway patency. A Glasgow Coma Scale score greater
than 8 is associated with a greater likelihood of successful
extubation. A number of tests evaluate cough strength such as
card moistening102 or spirometry103 and assess airway patency
such as determination of a cuff leak.104 However, the cuff leak
test is neither sensitive nor specific and most physicians make
Table 20.6  Spontaneous Breathing Trials:
Criteria That Indicate Failure
CLINICAL ASSESSMENT AND SUBJECTIVE INDICES
• Agitation and anxiety
• Deterioration in mental status
• Diaphoresis
• Cyanosis
• Evidence of increasing effort
• Increased accessory muscle activity
• Dyspnea
OBJECTIVE MEASUREMENTS
• Agitation and anxiety
• PaO2 ≤50-60 mm Hg on FiO2 ≥0.5 or SaO2 <90%
• PaCO2 >50 mm Hg or an increase in PaCO2 by >8 mm Hg
• pH <7.32 or a decrease in pH ≥0.07 pH unit
• Respiratory frequency >35 breath/min or increased by >50%
• Rapid shallow breathing index >105 breath/min/L
• HR >140 breaths/min or increased by >20%
• Systolic BP >180 mm Hg or increased by >20%
• Systolic BP <90 mm Hg
• Cardiac arrhythmias
Adapted from Boles J, Bion J, Connors A, et al. Weaning from mechanical
ventilation. Eur Respir J 2007;29:1033–56.
BP, Blood pressure; FiO2, inspiratory oxygen fraction; HR, heart rate; PaCO2,
arterial carbon dioxide tension; PaO2, arterial oxygen tension; SaO2, arterial
oxygen saturation.
a subjective determination of cough strength. When a patient
is difficult to wean (i.e., fails a spontaneous breathing trial or
extubation), a search should be made for factors that reduce
the chances of successful weaning. Management then has to
balance the need for adequate ventilator support (to reduce
respiratory fatigue) against the need to minimize support
(increase respiratory autonomy) to improve the likelihood of
successful weaning. The various available strategies are beyond
the scope of this chapter but include assist control ventilation
(ACV), synchronized intermittent mechanical ventilation
(SIMV), PSV, T-piece trials, use of noninvasive ventilation or
facilitation therapy, and tracheostomy.100,105-110
Conclusions
Mechanical ventilation is crucial to life support and its correct
management has significant effects on patient outcome. An
in-depth understanding of basic pulmonary physiology prin-
ciples is therefore essential to the care of neurologic critically
ill patients, with and without coexisting pulmonary problems.
Respiratory monitoring allows assessment of the physiologic
effects of ventilator settings and, even more importantly, pro-
vides the ability to minimize the iatrogenic injury by mechani-
cal ventilation.
References
1. Hoesch RE, Lin E, Young M, et al. Acute lung injury in critical neurological
illness. Crit Care Med 2012;40(2):587–93.
2. Pelosi P, Rocco PR, de Abreu MG. Use of computed tomography scanning to
guide lung recruitment and adjust positive-end expiratory pressure. Curr
Opin Crit Care 2011;17(3):268–74.
3. Luecke T, Corradi F, Pelosi P. Lung imaging for titration of mechanical
ventilation. Curr Opin Anaesthesiol 2012;25(2):131–40.
4. Bellani G, Mauri T, Pesenti A. Imaging in acute lung injury and acute
respiratory distress syndrome. Curr Opin Crit Care 2012;18(1):29–34.
 Section II—Clinical and Laboratory Assessment 199
5. Lundin S, Stenqvist O. Electrical impedance tomography: potentials and
pitfalls. Curr Opin Crit Care 2012;18(1):35–41.
6. Strumwasser A, Chu E, Yeung L, et al. A novel CT volume index score
correlates with outcomes in polytrauma patients with pulmonary contusion.
J Surg Res 2011;170(2):280–5. Epub 2011, Apr 5.
7. Stark P, Greene R, Kott MM, et al. Improved control of image optical density
with low-dose digital and conventional radiography in bedside imaging.
CT-findings in ARDS. Radiologe 1987;27(8):367−9.
8. Eisenhuber E, Schaefer-Prokop CM, Prosch H, et al. Bedside chest
radiography. Respir Care 2012;57(3):427–43.
9. Musch G. Positron emission tomography: a tool for better understanding of
ventilator-induced and acute lung injury. Curr Opin Crit Care 2011;17(1):
7–12.
10. de Prost N, Tucci MR, Melo MF. Assessment of lung inflammation with
18F-FDG PET during acute lung injury. AJR Am J Roentgenol
2010;195(2):292–300.
11. Magaña M, Bercovitch R, Fedullo P. Diagnostic approach to deep venous
thrombosis and pulmonary embolism in the critical care setting. Crit Care
Clin 2011;27(4):841–67, vi.
12. Terminella L, Sharma G. Diagnostic studies in patients with acute
respiratory distress syndrome. Semin Thorac Cardiovasc Surg 2006;18(1):
2–7.
13. Ruppel GL, Enright PL. Pulmonary function testing. Respir Care 2012;57(1):
165–75. doi: 10.4187.respcare.01640.edu
14. Barnett N, Ware LB. Biomarkers in acute lung injury: marking forward
progress. Crit Care Clin 2011;27(3):661–83.
15. Raoof S, Feigin D, Sung A, et al. Interpretation of plain chest roentgenogram.
Chest 2012;141(2):545–58.
16. Xirouchaki N, Magkanas E, Vaporidi K, et al. Lung ultrasound in critically ill
patients: comparison with bedside chest radiography. Intensive Care Med
2011;37(9):1488–93. Epub 2011, Aug 2.
17. Monchi M, Bellenfant F, Cariou A, et al. Early predictive factors of survival
in the acute respiratory distress syndrome: a multivariate analysis. Am J
Respir Crit Care Med 1998;158(4):1076–81.
18. Raghavendran K, Napolitano LM. Definition of ALI/ARDS. Crit Care Clin
2011;27(3):429–37.
19. Esteban A, Anzueto A, Frutos F, et al: Characteristics and outcomes in adult
patients receiving mechanical ventilation: a 28-day international study.
JAMA 2002;287:345–55.
20. Seeley E, McAuley DF, Eisner M, et al. Predictors of mortality in acute lung
injury during the era of lung protective ventilation. Thorax 2008;63(11):
994–8.
21. Oddo M, Nduom E, Frangos S, et al. Lung function and brain oxygen after
severe traumatic brain injury. Neurosurgery 2010;67:338–44.
22. Bernard GR, Artigas A, Brigham KL, et al. The American-European
consensus conference on ARDS: definitions, mechanisms, relevant
outcomes, and clinical trial coordination. Am J Respir Crit Care Med 1994;
149:818–24.
23. Murray JF, Matthay MA, Luce JM, et al. An expanded definition of the adult
respiratory distress syndrome. Am Rev Respir Dis 1988;138:720–3.
24. Ferguson ND, Davis AM, Slutsky AS, et al. Development of a clinical
definition for acute respiratory distress syndrome using the Delphi
technique. J Crit Care 2005;20(2):147–54.
25. Trachsel D, McCrindle BW, Nakagawa S, et al. Oxygenation index predicts
outcome in children with acute hypoxemic respiratory failure. Am J Respir
Crit Care Med 2005;172:206–11.
26. Severinghaus JW. The invention and development of blood gas analysis
apparatus. Anesthesiology 2002;97:253–6.
27. Tremper KK, Barker SJ. Pulse oximetry. Anesthesiology 1989;70:98–108.
28. Thompson JE, Jaffe MB. Capnographic waveforms in the mechanically
ventilated patient. Respir Care 2005;50:100–8; discussion 108–9.
29. Carlon GC, Ray C, Jr, Miodownik S, et al. Capnography in mechanically
ventilated patients. Crit Care Med 1988;16:550–6.
30. Anderson CT, Breen PH. Carbon dioxide kinetics and capnography during
critical care. Crit Care 2000;4:207–15.
31. Barker SJ, Badal JJ. The measurement of dyshemoglobins and total
hemoglobin by pulse oximetry. Curr Opin Anaesthesiol 2008;21:805–10.
32. Jubran A, Tobin MJ. Reliability of pulse oximetry in titrating supplemental
oxygen therapy in ventilator-dependent patients. Chest 1990;97:1420–5.
33. Bierman MI, Stein KL, Snyder JV. Pulse oximetry in the postoperative care of
cardiac surgical patients: a randomized controlled trail. Chest 1992;102:
1367–70.
34. Knapp S, Kofler J, Stoiser B, et al. The assessment of four different methods
to verify tracheal tube placement in the critical care setting. Anesth Analg
1999;88:766–70.
35. Husain T, Gatward JJ, Hambidge OR, et al. Strategies to prevent airway
complications: a survey of adult intensive care units in Australia and New
Zealand. Br J Anaesth 2012;108(5):800–6. Epub 2012, Mar 13.
36. Wouters PF, Gehring H, Meyfroidt G, et al. Accuracy of pulse oximeters: the
European multi-center trial. Anesth Analg 2002;94:S13–6.
37. Van de Louw A, Cracco C, Cerf C, et al. Accuracy of pulse oximetry in the
intensive care unit. Intensive Care Med 2001;27:1606–13.
38. Nuckton TJ, Alonso JA, Kallet RH, et al. Pulmonary dead-space fraction as a
risk factor for death in the acute respiratory distress syndrome. N Engl J
Med 2002;346:1281–6.
39. Tinker JH, Dull DL, Caplan RA, et al. Role of monitoring devices in
prevention of anesthetic mishaps: a closed claims analysis. Anesthesiology
1989;71:541–6.
40. Moller JT, Johannessen NW, Espersen K, et al. Randomized evaluation of
pulse oximetry in 20,802 patients. II. Perioperative events and postoperative
complications. Anesthesiology 1993;78:445–53.
41. Moller JT, Svennild I, Johannessen NW, et al. Perioperative monitoring with
pulse oximetry and late postoperative cognitive dysfunction. Br J Anaesth
1993;71:340–7.
42. Pedersen T, Dyrlund Pedersen B, et al. Pulse oximetry for perioperative
monitoring. Cochrane Database Syst Rev 2003;(3):CD002013.
43. Pedersen T, Møller AM, Hovhannisyan K. Pulse oximetry for perioperative
monitoring. Cochrane Database Syst Rev 2009;(4):CD002013.
44. Mead J, Whittenberger JL, Radford EP, Jr. Surface tension as a factor in
pulmonary volume-pressure hysteresis. J Appl Physiol 1957;10:191–6.
45. Matamis D, Lemaire F, Harf A, et al. Total respiratory pressure-volume
curves in the adult respiratory distress syndrome. Chest 1984;86:58–66.
46. Bates JH, Rossi A, Milic-Emili J. Analysis of the behavior of the respiratory
system with constant inspiratory flow. J Appl Physiol 1985;58:1840–8.
47. D’Angelo E, Calderini E, Torri G, et al. Respiratory mechanics in
anesthetized paralyzed humans: effects of flow, volume, and time. J Appl
Physiol 1989;67:2556–64.
48. Dreyfuss D, Saumon G. Ventilator-induced lung injury: lessons from
experimental studies. Am J Respir Crit Care Med 1998;157:294–323.
49. The Acute Respiratory Distress Syndrome Network. Ventilation with lower
tidal volumes as compared with traditional tidal volumes for acute lung
injury and the acute respiratory distress syndrome. the acute respiratory
distress syndrome network. N Engl J Med 2000;342:1301–8.
50. Benditt JO. Esophageal and gastric pressure measurements. Respir Care
2005;50:68–75; discussion 75–7.
A complete list of references for this chapter can be found online at
www.expertconsult.com.
 Section II—Clinical and Laboratory Assessment 199.e1
References
1. Hoesch RE, Lin E, Young M, et al. Acute lung injury in critical neurological
illness. Crit Care Med 2012;40(2):587–93.
2. Pelosi P, Rocco PR, de Abreu MG. Use of computed tomography scanning
to guide lung recruitment and adjust positive-end expiratory pressure. Curr
Opin Crit Care 2011;17(3):268–74.
3. Luecke T, Corradi F, Pelosi P. Lung imaging for titration of mechanical
ventilation. Curr Opin Anaesthesiol 2012;25(2):131–40.
4. Bellani G, Mauri T, Pesenti A. Imaging in acute lung injury and acute
respiratory distress syndrome. Curr Opin Crit Care 2012;18(1):29–34.
5. Lundin S, Stenqvist O. Electrical impedance tomography: potentials and
pitfalls. Curr Opin Crit Care 2012;18(1):35–41.
6. Strumwasser A, Chu E, Yeung L, et al. A novel CT volume index score
correlates with outcomes in polytrauma patients with pulmonary
contusion. J Surg Res 2011;170(2):280–5. Epub 2011, Apr 5.
7. Stark P, Greene R, Kott MM, et al. Improved control of image optical
density with low-dose digital and conventional radiography in bedside
imaging. CT-findings in ARDS. Radiologe 1987;27(8):367−9.
8. Eisenhuber E, Schaefer-Prokop CM, Prosch H, et al. Bedside chest
radiography. Respir Care 2012;57(3):427–43.
9. Musch G. Positron emission tomography: a tool for better understanding
of ventilator-induced and acute lung injury. Curr Opin Crit Care 2011;
17(1):7–12.
10. de Prost N, Tucci MR, Melo MF. Assessment of lung inflammation with
18F-FDG PET during acute lung injury. AJR Am J Roentgenol 2010;195(2):
292–300.
11. Magaña M, Bercovitch R, Fedullo P. Diagnostic approach to deep venous
thrombosis and pulmonary embolism in the critical care setting. Crit Care
Clin 2011;27(4):841–67, vi.
12. Terminella L, Sharma G. Diagnostic studies in patients with acute
respiratory distress syndrome. Semin Thorac Cardiovasc Surg 2006;18(1):
2–7.
13. Ruppel GL, Enright PL. Pulmonary function testing. Respir Care 2012;
57(1):165–75. doi: 10.4187.respcare.01640.edu
14. Barnett N, Ware LB. Biomarkers in acute lung injury: marking forward
progress. Crit Care Clin 2011;27(3):661–83.
15. Raoof S, Feigin D, Sung A, et al. Interpretation of plain chest
roentgenogram. Chest 2012;141(2):545–58.
16. Xirouchaki N, Magkanas E, Vaporidi K, et al. Lung ultrasound in critically
ill patients: comparison with bedside chest radiography. Intensive Care Med
2011;37(9):1488–93. Epub 2011, Aug 2.
17. Monchi M, Bellenfant F, Cariou A, et al. Early predictive factors of survival
in the acute respiratory distress syndrome: a multivariate analysis. Am J
Respir Crit Care Med 1998;158(4):1076–81.
18. Raghavendran K, Napolitano LM. Definition of ALI/ARDS. Crit Care Clin
2011;27(3):429–37.
19. Esteban A, Anzueto A, Frutos F, et al: Characteristics and outcomes in adult
patients receiving mechanical ventilation: a 28-day international study.
JAMA 2002;287:345–55.
20. Seeley E, McAuley DF, Eisner M, et al. Predictors of mortality in acute lung
injury during the era of lung protective ventilation. Thorax 2008;63(11):
994–8.
21. Oddo M, Nduom E, Frangos S, et al. Lung function and brain oxygen after
severe traumatic brain injury. Neurosurgery 2010;67:338–44.
22. Bernard GR, Artigas A, Brigham KL, et al. The American-European
consensus conference on ARDS: definitions, mechanisms, relevant
outcomes, and clinical trial coordination. Am J Respir Crit Care Med 1994;
149:818–24.
23. Murray JF, Matthay MA, Luce JM, et al. An expanded definition of the
adult respiratory distress syndrome. Am Rev Respir Dis 1988;138:720–3.
24. Ferguson ND, Davis AM, Slutsky AS, et al. Development of a clinical
definition for acute respiratory distress syndrome using the Delphi
technique. J Crit Care 2005;20(2):147–54.
25. Trachsel D, McCrindle BW, Nakagawa S, et al. Oxygenation index predicts
outcome in children with acute hypoxemic respiratory failure. Am J Respir
Crit Care Med 2005;172:206–11.
26. Severinghaus JW. The invention and development of blood gas analysis
apparatus. Anesthesiology 2002;97:253–6.
27. Tremper KK, Barker SJ. Pulse oximetry. Anesthesiology 1989;70:98–108.
28. Thompson JE, Jaffe MB. Capnographic waveforms in the mechanically
ventilated patient. Respir Care 2005;50:100–8; discussion 108–9.
29. Carlon GC, Ray C, Jr, Miodownik S, et al. Capnography in mechanically
ventilated patients. Crit Care Med 1988;16:550–6.
30. Anderson CT, Breen PH. Carbon dioxide kinetics and capnography during
critical care. Crit Care 2000;4:207–15.
31. Barker SJ, Badal JJ. The measurement of dyshemoglobins and total
hemoglobin by pulse oximetry. Curr Opin Anaesthesiol 2008;21:805–10.
32. Jubran A, Tobin MJ. Reliability of pulse oximetry in titrating
supplemental oxygen therapy in ventilator-dependent patients. Chest 1990;
97:1420–5.
33. Bierman MI, Stein KL, Snyder JV. Pulse oximetry in the postoperative care
of cardiac surgical patients: a randomized controlled trail. Chest 1992;102:
1367–70.
34. Knapp S, Kofler J, Stoiser B, et al. The assessment of four different methods
to verify tracheal tube placement in the critical care setting. Anesth Analg
1999;88:766–70.
35. Husain T, Gatward JJ, Hambidge OR, et al. Strategies to prevent airway
complications: a survey of adult intensive care units in Australia and New
Zealand. Br J Anaesth 2012;108(5):800–6. Epub 2012, Mar 13.
36. Wouters PF, Gehring H, Meyfroidt G, et al. Accuracy of pulse oximeters: the
European multi-center trial. Anesth Analg 2002;94:S13–6.
37. Van de Louw A, Cracco C, Cerf C, et al. Accuracy of pulse oximetry in the
intensive care unit. Intensive Care Med 2001;27:1606–13.
38. Nuckton TJ, Alonso JA, Kallet RH, et al. Pulmonary dead-space fraction as
a risk factor for death in the acute respiratory distress syndrome. N Engl J
Med 2002;346:1281–6.
39. Tinker JH, Dull DL, Caplan RA, et al. Role of monitoring devices in
prevention of anesthetic mishaps: a closed claims analysis. Anesthesiology
1989;71:541–6.
40. Moller JT, Johannessen NW, Espersen K, et al. Randomized evaluation of
pulse oximetry in 20,802 patients. II. Perioperative events and postoperative
complications. Anesthesiology 1993;78:445–53.
41. Moller JT, Svennild I, Johannessen NW, et al. Perioperative monitoring with
pulse oximetry and late postoperative cognitive dysfunction. Br J Anaesth
1993;71:340–7.
42. Pedersen T, Dyrlund Pedersen B, et al. Pulse oximetry for perioperative
monitoring. Cochrane Database Syst Rev 2003;(3):CD002013.
43. Pedersen T, Møller AM, Hovhannisyan K. Pulse oximetry for perioperative
monitoring. Cochrane Database Syst Rev 2009;(4):CD002013.
44. Mead J, Whittenberger JL, Radford EP, Jr. Surface tension as a factor in
pulmonary volume-pressure hysteresis. J Appl Physiol 1957;10:191–6.
45. Matamis D, Lemaire F, Harf A, et al. Total respiratory pressure-volume
curves in the adult respiratory distress syndrome. Chest 1984;86:58–66.
46. Bates JH, Rossi A, Milic-Emili J. Analysis of the behavior of the respiratory
system with constant inspiratory flow. J Appl Physiol 1985;58:1840–8.
47. D’Angelo E, Calderini E, Torri G, et al. Respiratory mechanics in
anesthetized paralyzed humans: effects of flow, volume, and time. J Appl
Physiol 1989;67:2556–64.
48. Dreyfuss D, Saumon G. Ventilator-induced lung injury: lessons from
experimental studies. Am J Respir Crit Care Med 1998;157:294–323.
49. The Acute Respiratory Distress Syndrome Network. Ventilation with lower
tidal volumes as compared with traditional tidal volumes for acute lung
injury and the acute respiratory distress syndrome. the acute respiratory
distress syndrome network. N Engl J Med 2000;342:1301–8.
50. Benditt JO. Esophageal and gastric pressure measurements. Respir Care
2005;50:68–75; discussion 75–7.
51. Loring SH, O’Donnell CR, Behazin N, et al. Esophageal pressures in acute
lung injury: do they represent artifact or useful information about
transpulmonary pressure, chest wall mechanics, and lung stress? J Appl
Physiol 2010;108(3):515–22. Epub 2009, Dec 17.
52. Pelosi P, Cereda M, Foti G, et al. Alterations of lung and chest wall
mechanics in patients with acute lung injury: effects of positive end-
expiratory pressure. Am J Respir Crit Care Med 1995;152:531–7.
53. Unroe M, MacIntyre N. Evolving approaches to assessing and
monitoring patient-ventilator interactions. Curr Opin Crit Care 2010;16(3):
261–8.
54. Galetke W, Randerath WJ, Feier C, et al. Esophageal pressure method and
impulse oscillometry to assess mechanical properties of the respiratory
system in healthy men. Med Sci Monit 2009;15(8):CR429–35.
55. Gattinoni L, Caironi P, Pelosi P, et al. What has computed tomography
taught us about the acute respiratory distress syndrome? Am J Respir Crit
Care Med 2001;164:1701–11.
56. Crotti S, Mascheroni D, Caironi P, et al. Recruitment and derecruitment
during acute respiratory failure: a clinical study. Am J Respir Crit Care Med
2001;164:131–40.
57. Muscedere JG, Mullen JB, Gan K, et al. Tidal ventilation at low airway
pressures can augment lung injury. Am J Respir Crit Care Med 1994;149:
1327–34.
58. Amato MB, Barbas CS, Medeiros DM, et al. Effect of a protective-
ventilation strategy on mortality in the acute respiratory distress syndrome.
N Engl J Med 1998;338:347–54.
199.e2 Section II—Clinical and Laboratory Assessment
59. Brower RG, Lanken PN, MacIntyre N, et al. Higher versus lower positive
end-expiratory pressures in patients with the acute respiratory distress
syndrome. N Engl J Med 2004;351:327–36.
60. Chiumello D, Cressoni M, Chierichetti M, et al. Nitrogen washout/washin,
helium dilution and computed tomography in the assessment of end
expiratory lung volume. Crit Care 2008;12:R150.
61. Bodenstein M, David M, Markstaller K. Principles of electrical impedance
tomography and its clinical application. Crit Care Med 2009;37(2):713–24.
62. Younes M, Webster K, Kun J, et al. A method for measuring passive
elastance during proportional assist ventilation. Am J Respir Crit Care Med
2001;164:50–60.
63. Chieveley-Williams S, Dinner L, Puddicombe A, et al. Central venous and
bladder pressure reflect transdiaphragmatic pressure during pressure
support ventilation. Chest 2002;121:533–8.
64. Sydow M, Burchardi H, Zinserling J, et al. Improved determination of static
compliance by automated single volume steps in ventilated patients.
Intensive Care Med 1991;17:108–14.
65. Harris RS, Hess DR, Venegas JG. An objective analysis of the pressure-
volume curve in the acute respiratory distress syndrome. Am J Respir Crit
Care Med 2000;161:432–9.
66. Grivans C, Lundin S, Stenqvist O, et al. Positive end-expiratory pressure-
induced changes in end-expiratory lung volume measured by spirometry
and electric impedance tomography. Acta Anaesthesiol Scand
2011;55(9):1068–77. doi:10.1111/j.1399–6576.2011.02511.x. Epub 2011,
Sep 7.
67. Pulletz S, Adler A, Kott M, et al. Regional lung opening and closing
pressures in patients with acute lung injury. J Crit Care 2011;Oct 25. Epub
ahead of print.
68. Moerer O, Hahn G, Quintel M. Lung impedance measurements to monitor
alveolar ventilation. Curr Opin Crit Care 2011;17(3):260–7.
69. Lowhagen K, Lindgren S, Odenstedt H, et al. A new non-radiological
method to assess potential lung recruitability: a pilot study in ALI patients.
Acta Anaesthesiol Scand 2011;55(2):165–74.
70. Georgopoulos D, Prinianakis G, Kondili E. Bedside waveforms
interpretation as a tool to identify patient-ventilator asynchronies. Intensive
Care Med 2006;32:34–47.
71. Talmor D, Sarge T, O’Donnell CR, et al. Esophageal and transpulmonary
pressures in acute respiratory failure. Crit Care Med 2006;34:1389–94.
72. Chiumello D, Carlesso E, Cadringher P, et al. Lung stress and strain during
mechanical ventilation for acute respiratory distress syndrome. Am J Respir
Crit Care Med 2008;178:346–55.
73. Villar J, Kacmarek RM, Perez-Mendez L, et al. A high positive end-
expiratory pressure, low tidal volume ventilatory strategy improves
outcome in persistent acute respiratory distress syndrome: a randomized,
controlled trial. Crit Care Med 2006;34:1311–18.
74. Broseghini C, Brandolese R, Poggi R, et al. Respiratory mechanics during
the first day of mechanical ventilation in patients with pulmonary edema
and chronic airway obstruction. Am Rev Respir Dis 1988;138:355–61.
75. Murray JF, Matthay MA, Luce JM, et al. An expanded definition of the
adult respiratory distress syndrome. Am Rev Respir Dis 1988;138:720–3.
76. Polese G, Rossi A, Appendini L, et al. Partitioning of respiratory mechanics
in mechanically ventilated patients. J Appl Physiol 1991;71:2425–33.
77. Gattinoni L, Pelosi P, Suter PM, et al. Acute respiratory distress syndrome
caused by pulmonary and extrapulmonary disease. different syndromes?
Am J Respir Crit Care Med 1998;158:3–11.
78. Musch G, Foti G, Cereda M, et al. Lung and chest wall mechanics in
normal anaesthetized subjects and in patients with COPD at different PEEP
levels. Eur Respir J 1997;10:2545–52.
79. Pepe PE, Marini JJ. Occult positive end-expiratory pressure in mechanically
ventilated patients with airflow obstruction: the auto-PEEP effect. Am Rev
Respir Dis 1982;126:166–70.
80. Gattinoni L, Pesenti A, Avalli L, et al. Pressure-volume curve of total
respiratory system in acute respiratory failure. computed tomographic scan
study. Am Rev Respir Dis 1987;136:730–6.
81. Roupie E, Dambrosio M, Servillo G, et al. Titration of tidal volume and
induced hypercapnia in acute respiratory distress syndrome. Am J Respir
Crit Care Med 1995;152:121–8.
82. Richard JC, Brochard L, Vandelet P, et al. Respective effects of end-
expiratory and end-inspiratory pressures on alveolar recruitment in acute
lung injury. Crit Care Med 2003;31:89–92.
83. Jonson B, Richard JC, Straus C, et al. Pressure-volume curves and
compliance in acute lung injury: evidence of recruitment above the lower
inflection point. Am J Respir Crit Care Med 1999;159:1172–8.
84. Dhand R, Duarte AG, Jubran A, et al. Dose-response to bronchodilator
delivered by metered-dose inhaler in ventilator-supported patients. Am J
Respir Crit Care Med 1996;154:388–93.
85. MacIntyre NR, Cheng KC, McConnell R. Applied PEEP during pressure
support reduces the inspiratory threshold load of intrinsic PEEP. Chest
1997;111:188–93.
86. Marini JJ, Gattinoni L. Ventilatory management of acute respiratory distress
syndrome: a consensus of two. Crit Care Med 2004;32:250–5.
87. Hager DN, Krishnan JA, Hayden DL, et al. Tidal volume reduction in
patients with acute lung injury when plateau pressures are not high. Am J
Respir Crit Care Med 2005;172:1241–5.
88. Gajic O, Dara SI, Mendez JL, et al. Ventilator-associated lung injury in
patients without acute lung injury at the onset of mechanical ventilation.
Crit Care Med 2004;32:1817–24.
89. Suter PM, Fairley B, Isenberg MD. Optimum end-expiratory airway
pressure in patients with acute pulmonary failure. N Engl J Med 1975;292:
284–9.
90. Katz JA, Ozanne GM, Zinn SE, et al. Time course and mechanisms of
lung-volume increase with PEEP in acute pulmonary failure.
Anesthesiology 1981;54:9–16.
91. Talmor D, Sarge T, Malhotra A, et al. Mechanical ventilation guided by
esophageal pressure in acute lung injury. N Engl J Med 2008;359:2095–104.
92. Lee WL, Stewart TE, MacDonald R, et al. Safety of pressure-volume curve
measurement in acute lung injury and ARDS using a syringe technique.
Chest 2002;121:1595–601.
93. American College of Surgeons. ATLS: advanced trauma life support for
doctors. 8th ed. Chicago: ACS; 2010.
94. Marino PL. Principles of mechanical ventilation. In: Marino PL, editor. The
ICU book. Philadelphia: Lippincott Williams & Wilkins; 2007. p. 461.
95. Wilson BJ, Cowan HJ, Lord JA, et al. The accuracy of pulse oximetry in
emergency department patients with severe sepsis and septic shock: a
retrospective cohort study. BMC Emerg Med 2010;10:9.
96. Cooper LM, Linde-Zwirble WT. Medicare intensive care unit use: analysis
of incidence, cost, and payment. Crit Care Med 2004;32:2247–53.
97. Ely EW, Baker AM, Dunagan DP, et al. Effect on the duration of mechanical
ventilation of identifying patients capable of breathing spontaneously. N
Engl J Med 1996;335:1864–9.
98. Boles JM, Bion J, Connors A, et al. Weaning from mechanical ventilation.
Eur Respir J 2007;29:1033–56.
99. Monaco F, Drummond GB, Ramsay P, et al. Do simple ventilation and gas
exchange measurements predict early successful weaning from respiratory
support in unselected general intensive care patients? Br J Anaesth 2010;
105(3):326–33. Epub 2010, Jul 22.
100. Esteban A, Frutos F, Tobin MJ, et al. A comparison of four methods of
weaning patients from mechanical ventilation. Spanish Lung Failure
Collaborative Group. N Engl J Med 1995;332:345–50.
101. Yang KL, Tobin MJ. A prospective study of indexes predicting the outcome
of trials of weaning from mechanical ventilation. N Engl J Med 1991;324:
1445–50.
102. Khamiees M, Raju P, DeGirolamo A, et al. Predictors of extubation
outcome in patients who have successfully completed a spontaneous
breathing trial. Chest 2001;120:1262–70.
103. Smina M, Salam A, Khamiees M, et al. Cough peak flows and extubation
outcomes. Chest 2003;124:262–8
104. Fisher MM, Raper RF. The “cuff-leak” test for extubation. Anaesthesia 1992;
47:10–2.
105. Brochard L, Rauss A, Benito S, et al. Comparison of three methods of
gradual withdrawal from ventilatory support during weaning from
mechanical ventilation. Am J Respir Crit Care Med 1994;150:896–903.
106. Trevisan CE, Vieira SR. Noninvasive mechanical ventilation may be useful
in treating patients who fail weaning from invasive mechanical ventilation:
a randomized clinical trial. Crit Care 2008;12:R51.
107. Agarwal R, Aggarwal AN, Gupta D, et al Role of noninvasive positive-
pressure ventilation in postextubation respiratory failure: a meta-analysis.
Respir Care 2007;52:1472–9.
108. Burns KE, Adhikari NK, Meade MO. A meta-analysis of noninvasive
weaning to facilitate liberation from mechanical ventilation. Can J Anaesth
2006;53:305–15.
109. Rizk EB, Patel AS, Stetter CM, et al. Impact of tracheostomy timing on
outcome after severe head injury. Neurocrit Care 2011;15(3):481–9.
110. Cabello B, Thille AW, Roche-Campo F, et al. Physiological comparison of
three spontaneous breathing trials in difficult-to-wean patients. Intensive
Care Med 2010;36(7):1171–9. Epub 2010, Mar 30.
II
Chapter
21  
Endocrine Evaluation
Matthew R. Sanborn and Carrie A. Sims
Introduction
The neuroendocrine system is important for body homeosta-
sis but can undergo significant changes in response to critical
illness. During the acute phase of critical illness or trauma,
anabolic hormonal pathways are down-regulated in favor
of fulfilling more immediate needs. With prolonged critical
illness, the hypothalamic-pituitary axis (HPA) is subject to
reduced pulsatile secretion of pituitary hormones, reduced
peripheral activity, and global hormonal suppression. If
unrecognized, HPA dysfunction can lead to problems in both
the short and long term. However, monitoring the endocrine
status of critically ill patients can be a challenge. First, pituitary
dysfunction often can mimic deficits associated with the neu-
rologically injured patient (Table 21.1). In addition, critical
illness may induce endocrine dysfunction that is not observed
outside the intensive care unit (ICU). Therefore a high index
of suspicion or a screening protocol is necessary. Second,
many of the signs and symptoms of endocrine dysfunction are
subtle, may be delayed in onset, or may be masked by inter-
ventions common in the acute setting. Third, assays used in
healthy patients may underestimate hormone levels in the
acutely ill, and standards used to define appropriate normal
values may not be applicable. Although there are few widely
adopted management guidelines, protocol-driven screening of
the at-risk patient may avoid the potential for misdiagnosis
and so influence outcome. There are, however, two important
questions in management: (1) does endocrine intervention
improve outcome when there is critical illness–induced endo-
crine dysfunction, and (2) does endocrine intervention help
critically ill patients without endocrine dysfunction (e.g.,
steroid use) to improve hemodynamics?
This chapter reviews endocrine deficiencies frequently
encountered in patients in the neurocritical care unit (NCCU)
and discusses the endocrine consequences of traumatic brain
injury (TBI) and subarachnoid hemorrhage (SAH). This is
important because nearly all endocrine axes have close links
with the central nervous system including regulation of pitu-
itary hormone production and direct innervation of endo-
crine glands. Glycemic control and nutrition are addressed in
Chapter 14.
Practicalities of Endocrine
Assessment in the Critical Care Unit
Several practical difficulties are associated with endocrine
assessment in the critical care unit. First, patients in the NCCU,
200
particularly those with severe neurologic disorders, sepsis, or
multiorgan dysfunction, are likely to have physiologic changes
rarely, if ever, seen in patients in an outpatient setting. Second,
it is unclear whether standard tests of endocrine function that
primarily are designed to assess patients in the outpatient
setting (i.e., not under “stress”) are valid in critical care. For
example, basal levels of hormones and regulation of hormone
responsiveness (i.e., feedback regulation) cannot be assumed
to be the same in the typical patient undergoing outpatient
evaluation and one in the ICU. Third, although hormone
levels in the circulation depend in part on synthesis, they also
are influenced by levels of binding proteins1 and tissue-specific
metabolism,2 both of which can be altered by critical illness.
Fourth, it is unclear whether adequate serum levels equate
with tissue levels. For example, how some hormone receptors
respond to their ligands can be altered during sepsis and
hypoxia,3 and hormone metabolism, in particular that of
thyroid and cortisol, is altered at the tissue level in critically ill
patients.2,4 Fifth, endocrine assessment depends on measure-
ment of basal hormone levels and often dynamic tests (i.e.,
stimulation or inhibition of hormone synthesis). These
various tests normally are performed under standard condi-
tions such as fasting without the interference of other medica-
tions, and at a particular time. These conditions may be
difficult to replicate in the ICU. Furthermore, several dynamic
tests (e.g., insulin tolerance test or metyrapone test) used in
the outpatient setting may be dangerous to use in the critical
care patient. Sixth, there can be a delay in turnaround of
hormone assay results, and hence empiric treatment may be
needed. Seventh, any critical illness is a dynamic process, and
physiologic responses vary and evolve over time. Consistent
with this, cortisol levels are well known to vary in the same
patient during a 24-hour period. This implies that frequent
samples may be needed but with that comes an increased risk
of false-positive results. Finally, various medications and ther-
apies given to patients in the ICU may alter the accuracy of
laboratory testing of endocrine function5-7 (Table 21.2). In
addition, results may differ across laboratories because stan-
dardization of assays may vary. Together these limitations
mean that the clinician in the NCCU must pay careful atten-
tion to the whole patient rather than simply interpreting indi-
vidual hormone levels.
The challenges facing laboratory staff, including preana-
lytic, analytic, postanalytic and those associated with assay
methodology are reviewed elsewhere.5 A key issue, despite
considerable effort, is that units used for reporting assay
results and assay standardizations vary around the world.
© Copyright 2013 Elsevier Inc. All rights reserved.
 Section II—Clinical and Laboratory Assessment 201

Table 21.1  Signs and Symptoms of Hypothalamic-Pituitary Axis Dysfunction

Hormone Deficiency Symptoms Signs
Adrenocorticotropic Weakness, nausea, fever, shock, abdominal Hypotension, hypoglycemia, hyponatremia,
hormone (ACTH) pain, headache, adrenal crisis myopathy, anemia, eosinophilia
Thyroid hormone Anergy, neuropsychiatric symptoms, cold Bradycardia, hypotension, neuropathy, myopathy
intolerance, memory problems, weight gain
Growth hormone Anergy, poor quality of life Osteoporosis, reduced lean body mass, dyslipidemia
Follicle-stimulating and Oligomenorrhea or amenorrhea, sexual Reduced muscle mass and exercise tolerance,
luteinizing hormone dysfunction, mood disorders osteoporosis, infertility, loss of secondary hair

superior hypophyseal arteries. The anterior pituitary contains

Table 21.2  Common Drugs Used in the glandular cells that respond to inhibitory and releasing factors
Intensive Care Unit That May Affect secreted by the arcuate, periventricular, medial preoptic, and
Endocrine Function paraventricular nuclei from the hypothalamus. The axons of
Adrenal Dysfunction Thyroid Dysfunction the neurons located within these nuclei project to the median
eminence, where they release secretory and inhibitory factors
Heparins Heparins
into a capillary plexus. This median eminence capillary plexus
Warfarin and other Iodinated contrast dye then drains into a second capillary plexus within the anterior
anticoagulants pituitary via the hypophyseal portal veins. The hypothalamic
Phenobarbital, phenytoin Phenobarbital, phenytoin, factors then stimulate or inhibit the glandular cells of the
carbamazepine anterior pituitary and modify secretion of hormones into the
Azole antifungals Amiodarone secondary capillary plexus. Ultimately, the hypophyseal portal
Etomidate Corticosteroids vein drains to the cavernous sinus and to the systemic circula-
tion through the internal jugular vein.
Glucocorticoid therapy Dobutamine, dopamine
Opiates and benzodiazepines Furosemide
Statins Octreotide Corticotroph Dysfunction
Rifampin Lithium, atypical antipsychotics
Rifampin
Etiology
Decreased or inappropriate cortisol secretion is the hall-
mark of adrenal insufficiency. Adrenal insufficiency may be
due to processes that affect: (1) the adrenal gland itself
Furthermore, there can be variation in assay performance (primary adrenal insufficiency), (2) the pituitary gland and
based on the manufacturer. Consequently there are interna- secretion of adrenocorticotropic hormone (ACTH) (second-
tional scientific efforts to standardize assays in which assays ary adrenal insufficiency), or (3) the hypothalamus and
are compared against a candidate reference measurement pro- its secretion of corticotropin-releasing hormone (tertiary
cedure. This means that a given substance, for example, free adrenal insufficiency).
cortisol, measured in a given system such as serum is collected Acute primary adrenal insufficiency may result from
under defined conditions, and the units used to report the bilateral adrenal hemorrhage or infarction secondary to
levels are traceable to a reference measurement procedure trauma, coagulopathy, thromboembolic disease, or sepsis
and a reference preparation that are certified by such organi- (e.g., Waterhouse-Friderichsen syndrome). Waterhouse-
zations as the Joint Committee for Traceability in Laboratory Friderichsen syndrome was first described in the setting of
Medicine. meningococcemia, but also has been reported in several
other systemic bacterial infections including those caused
by Pseudomonas aeruginosa, Streptococcus pneumoniae, Neis-
Anterior Pituitary Dysfunction seria gonorrhoeae, Escherichia coli, Haemophilus influenzae,
and Staphylococcus aureus.8 Several drugs can cause adrenal
Anatomy insufficiency by inhibiting cortisol biosynthesis (e.g., etomi-
The pituitary gland is divided into two functionally, physio- date, ketoconazole, metyrapone, suramin) or by accelerating
logically, and developmentally distinct units: the anterior pitu- its metabolism (e.g., phenytoin, barbiturates, rifampin).9-14
itary, or adenohypophysis, and the posterior pituitary, or Any disease process or medication that interferes with
neurohypophysis. The anterior pituitary develops from ecto- ACTH secretion can cause secondary adrenal insufficiency.
dermal cells that overlie the pharynx that invaginate to form Infection, TBI, tumors, infiltrative diseases, hemorrhage, and
Rathke’s pouch. The posterior wall of Rathke’s pouch also infarction each may damage the pituitary gland.15 Glucocor-
forms a small intermediate lobe that secretes melanocyte- ticoid use is the most common cause of drug-induced second-
stimulating hormone in the fetus, but this regresses in the ary adrenal insufficiency. Exogenous steroids suppress the
adult. The pituitary gland derives its blood supply from two HPA that leads to atrophy of ACTH-producing cells. Second-
branches of the internal carotid artery: the inferior and ary insufficiency can occur when glucocorticoid therapy is
202 Section II—Clinical and Laboratory Assessment
Table 21.3  Recommendations for Perioperative Steroid Coverage
Degree of Surgical Stress
Minor (e.g., inguinal herniorrhaphy)
Moderate (total knee replacement, abdominal hysterectomy, open
cholecystectomy)
Major (pancreaticoduodenectomy, cardiac surgery with bypass)
Adapted from Salem M, Tainsh RE, Jr., Bromberg J, et al. Perioperative glucocorticoid coverage: a reassessment 42 years after emergence of a problem. Ann Surg
1994;219:416–25.
*Hydrocortisone supplementation should be administered in divided daily doses.
tapered or when physiologic stress exceeds the maintenance
dose. Consequently, patients who have been on chronic ste-
roids require a “stress-dose” perioperatively or when admitted
to the NCCU (Table 21.3).
Any process that inhibits the hypothalamic production of
corticotropin-releasing hormone (CRH), such as tumors,
infiltrative diseases, e.g., sarcoidosis, cranial irradiation, and
chronic high-dose glucocorticoid therapy16 will result in ter-
tiary adrenal insufficiency. Without the ACTH-secretagogue
action of CRH on the anterior pituitary, ACTH levels decline,
endogenous cortisol levels fall, and the adrenal glands
atrophy.
Dysfunction of the HPA in the critically ill patient in the
absence of glucocorticoid treatment or structural damage to
the adrenal glands, pituitary, or hypothalamus is an increas-
ingly recognized phenomenon. The condition has been
described as relative adrenal insufficiency (RAI), or critical
illness-related corticosteroid insufficiency (CIRCI). These dis-
orders are common in sepsis and are the basis of trials of
glucocorticoids in septic shock. The diagnosis is based on
plasma cortisol profiles, but there are currently no tests that
clearly identify which patient is truly glucocorticoid “defi-
cient” at a cellular level. A consensus statement defined CIRCI
as inadequate cellular corticosteroid activity for the severity of
the patient’s illness.17 The pathophysiology of this form of
adrenal “insufficiency” is not well defined, but critical illness
may alter the HPA and down-regulate its feedback responses
resulting in impaired cortisol production and tissue resistance
to glucocorticoids.18 Although the stress of critical illness is
associated with high circulating cortisol levels, there often is a
blunted response to ACTH. For example, supplemental hydro-
cortisone may improve hemodynamic stability in septic shock,
suggesting a relative shortage of steroids or a down-regulation
of available receptors.19-21
Clinical Presentation
Symptoms of acute adrenal insufficiency are often nonspecific
and include vomiting, lethargy, abdominal pain, fever, and
mental status changes. Severe hypotension and shock then
may develop. This presentation can be confused with, or exac-
erbate preexisting, septic shock. In some patients, the degree
of cardiovascular collapse is associated with mineralocorticoid
rather than glucocorticoid insufficiency—patients who receive
glucocorticoids may still develop an adrenal crisis if their min-
eralocorticoid requirements are not met.22 Outcome can be
Dose* Duration
25 mg hydrocortisone daily 1 day
50-75 mg hydrocortisone daily 1 to 2 days
100-150 mg hydrocortisone daily 2 to 3 days
poor if glucocorticoid and mineralocorticoid insufficiency is
not diagnosed and treated.23,24
Monitoring the Hypothalamic-
Pituitary-Adrenal Axis in the ICU
Considerable controversy surrounds the utility, optimal
method, and interpretation of hypothalamic-pituitary-
adrenal axis testing in the critically ill patient. In the healthy,
non-ICU patient the initial diagnostic tests for adrenal insuf-
ficiency include morning plasma ACTH and serum cortisol
levels. In general, primary adrenal insufficiency can be dis-
tinguished from secondary disease by an elevation in plasma
ACTH levels. A cortisol level less than 100 nmol/L (or
3.6 µg/dL) indicates adrenal insufficiency, whereas a value
greater than 500 nmol/L (or 18 µg/dL) suggests an intact
hypothalamic-pituitary-adrenal axis.25 However, in the ICU
patient a normal or even elevated cortisol does not mean
there will be an appropriate response to the stress of surgery
or critical illness. The appropriate cortisol level for critically
ill patients is unknown, and tests used to diagnose adreno-
cortical dysfunction have not been validated in the ICU pop-
ulation.26,27 Consensus guidelines suggest a random cortisol
of less than 275 nmol/L (10 µg/dL) or a change in cortisol
levels of less than 250 nmol/L (9 µg/dL) following 250 µg of
cosyntropin are the best tests available to diagnose adrenal
insufficiency in the critically ill patient.17
Several factors make diagnosis in critical care more chal-
lenging. First, cortisol immunoassays measure protein-bound
cortisol. Both cortisol-binding globulin and albumin are neg-
ative acute-phase reactants and their levels decrease in the
setting of stress.28 This makes the relationship between total
and free cortisol unpredictable and may give the appearance
of adrenal insufficiency, even though free cortisol levels may
be appropriate. Second, saturation of the corticotropin recep-
tor, with a corresponding limit of cortisol release, may con-
found dynamic measurements of glucocorticoid function.
Consequently a low, unstimulated free plasma cortisol level
has been proposed as a better marker of adrenal insufficiency.29
Although this makes free cortisol an attractive target for
testing, the assay is not routinely available. Furthermore, total
cortisol has been shown to correlate with free cortisol in
treatment-resistant hypotension in critical illness.30
Hyponatremia and hyperkalemia develop in acute adrenal
insufficiency, so assessment of electrolytes can be a useful
 Section II—Clinical and Laboratory Assessment 203
adjunct in the patient who develops hypotension or other
symptoms of acute adrenal insufficiency.
Adrenal Insufficiency Management
An acute adrenal crisis can rapidly progress to death; there-
fore, suspected cases should be immediately treated with
100 mg of intravenous (IV) hydrocortisone followed by 100
to 200 mg per day. Because there is relatively little risk associ-
ated with short-term steroid use, treatment of acute adrenal
insufficiency should not be delayed pending diagnostic evalu-
ation. Hydrocortisone provides both glucocorticoid and min-
eralocorticoid supplementation without sustained suppression
of the HPA and is the preferred agent to provide “stress dose”
steroids. Fluid resuscitation with isotonic fluid and vasopres-
sors also are often required. Electrolyte abnormalities and
hypoglycemia should be corrected.
Treatment of CIRCI in severe sepsis may be beneficial
but remains controversial. In patients with severe septic
shock, Annane et al. demonstrated corticosteroid replacement
reduced the duration of vasopressor support and improved
survival in patients with RAI.31 However, in another ran­
domized clinical trial, corticosteroid therapy of septic shock
(CORTICUS), steroid use was associated with improved
hemodynamic stability but had no effect on overall survival.
In addition, the cosyntropin stimulation test could not help
differentiate who would benefit from steroid supplementa-
tion.32 Several factors including differences in patient acuity,
timing of enrollment, and the addition of fludrocortisone may
explain the different outcome results in these two trials.31,32
The 2008 Surviving Sepsis Campaign guidelines do not rec-
ommend using a cosyntropin stimulation test to diagnose
relative adrenal insufficiency but suggest hydrocortisone
(<300 mg/day) may be used in any patient with septic shock
that is refractory to fluid and vasopressor support.33 Similarly,
the consensus statements from an international task force by
the American College of Critical Care Medicine recommend
treatment of critically ill patients based on clinical criteria
rather than the results of adrenal function testing.17
Because hydrocortisone has mineralocorticoid activity, the
benefit of adding fludrocortisone as described by Annane
et al. remains unclear.31 Nonetheless, oral fludrocortisone
(50 µg daily) is considered an optional therapy in the Surviv-
ing Sepsis Campaign guidelines. Supplemental steroids should
be tapered when vasopressors are no longer required to mini-
mize side effects including the risk of secondary adrenal
suppression.
Thyrotroph Dysfunction
Physiology
Thyroid hormone secretion is tightly regulated by the
hypothalamic-pituitary-thyroid axis. Thyrotropin-releasing
hormone (TRH) is released from the hypothalamus and
stimulates the synthesis and secretion of thyroid-stimulating
hormone (TSH). TSH, in turn, controls the synthesis and
secretion of the thyroid hormones, thyroxine (T4) and triio-
dothyronine (T3). More than 99.5% of T4 and T3 are protein
bound in the serum and are metabolically inactive. The small
percentage of free T4 and T3 influence metabolic function
and modulate the release of both TRH and TSH using
a negative feedback system. The thyroid gland primarily pro-
duces T4, a biologically inactive hormone that must be con-
verted to the active hormone T3 in peripheral tissues. Because
thyroid hormone is critical for cellular metabolism, both
overproduction and undersecretion can affect all organ
systems.
Hyperthyroidism and Thyroid Storm
Excess T4 levels characterize hyperthyroidism. An overproduc-
tive thyroid nodule or gland is the most common etiology, and
TSH production is suppressed. Less commonly, excess TSH
causes elevated thyroid hormone levels. Thyrotoxic crisis, or
thyroid storm, is a rare but potentially life-threatening state
that most frequently occurs in patients with untreated or par-
tially treated hyperthyroidism. Trauma or an acute illness can
precipitate this transition from stable hyperthyroidism to a
decompensated crisis. Thyroid function tests are not helpful
to distinguish the two states. Instead the diagnosis of thyroid
storm is made on clinical grounds. Signs and symptoms of
thyroid storm are exaggerated features of hyperthyroidism
and include fever (>38.5° C), tachycardia, hypertension, agita-
tion, delirium, lethargy, seizures, coma, nausea, vomiting,
diarrhea, and, in severe cases, jaundice.
Treating Thyrotoxicosis
The therapeutic goals of treating thyroid storm are to: (1)
decrease hormone production and secretion, (2) block the
conversion of T4 to T3, and (3) antagonize the catecholamin-
ergic effects of thyroid hormone. Thionamides such as
propylthiouracil and methimazole can block new thyroid
hormone synthesis within hours of administration. Propyl-
thiouracil is preferred because it also inhibits the peripheral
conversion of T4 to T3. These agents, however, do not prevent
the release of stored hormone. High-dose iodine (e.g., Lugol’s
solution or saturated solution of potassium iodide) can acutely
block the release of T4 and T3, but should be given only after
thyroid synthesis has been blocked. If synthetic function is not
adequately suppressed, the iodine bolus will enhance thyroid
hormone synthesis and thus exacerbate the thyrotoxicosis.
Glucocorticoids can reduce the peripheral conversion of T4 to
T3 and, given a high risk of concomitant adrenal insufficiency,
are considered standard of care in severe thyrotoxicosis. Beta-
adrenergic blockade, in particular propranolol that has anti-
adrenergic effects and inhibits the peripheral conversion of T4
to T3, remains a mainstay of therapy. Alternatively, esmolol
can be used when rapid titration is needed to reduce potential
side effects. Clinical improvement usually is rapid following
the initiation of β-blocker therapy. In addition to the expected
cardiac effects, there is marked improvement in agitation, con-
fusion, psychosis, diaphoresis, diarrhea, and fever.34
Myxedema Coma
Etiology and Presentation
Myxedema coma is the result of severe, decompensated hypo-
thyroidism that can occur insidiously as the result of severe
long-standing hypothyroidism or it can be precipitated by
an acute event such as infection, myocardial infarction,
head injury, cold exposure, or certain medications (e.g.,
opiates, lithium, amiodarone). Patients with myxedema coma
204 Section II—Clinical and Laboratory Assessment
demonstrate the classic features of severe hypothyroidism
including dry skin, thin hair, periorbital swelling, nonpitting
edema of the hands and feet, a hoarse voice, macroglossia, and
delayed deep tendon reflexes. Progression to myxedema coma,
however, is characterized by a depressed mental status, hypo-
tension, and hypothermia. Focal and generalized seizures also
have been reported.35 Depressed myocardial contractility and
bradycardia result in low cardiac output, profound hypoten-
sion, and diminished cerebral perfusion.36 Mortality can be
high (30%-60%) even with early diagnosis and appropriate
therapy37,38 and is associated with the degree of hypothermia.
A core temperature less than 32° C is associated with a worse
prognosis.
Laboratory Findings in Myxedema Coma
Hyponatremia and hypoglycemia are common. Low serum
sodium results from excessive vasopressin secretion and
impaired free water excretion.39,40 Significant hyponatremia
(≤125 mEq/L) is common and may contribute to mental
status changes. Decreased gluconeogenesis and hypoglycemia
may occur with hypothyroidism alone or with adrenal insuf-
ficiency.41 Infection is frequently a precipitating cause of
myxedema coma. However, an elevated white blood cell
(WBC) count often is absent, although a left shift may be
observed.
Management of Myxedema Coma
When clinically suspected, therapy should be instituted
without waiting for laboratory confirmation. However,
thyroid function tests (serum TSH and free T4) should be
drawn before starting therapy. A cortisol level also should be
obtained to investigate the possibility of concurrent adrenal
insufficiency. Appropriate hormonal supplementation nor-
malizes the basal metabolic rate and reverses all symptoms
and signs of hypothyroidism.42 Hydrocortisone should be
given with thyroid replacement for several days and then
tapered based on clinical improvement and assessment of
adrenal function because thyroid replacement may unmask
coexisting adrenal insufficiency and precipitate an adrenal
crisis.43
The deiodinase conversion of T4 to T3 is impaired in severe
hypothyroidism and so IV T3 may be preferable given its
greater biologic availability. T3 rapidly achieves effective
tissue levels44 and crosses the blood-brain barrier more readily
than T4 and can hasten the improvement of neurologic symp-
toms.45 The rapid onset of T3, however, also may increase the
risk of treatment-associated complications such as myocar-
dial infarction or arrhythmias. Treatment with IV T4 is associ-
ated with a slower onset of action because the patient’s
inherent tissue deiodinase activity is required to convert T4 to
T3 and clinical improvement may take 1 to 3 days. However,
the slower onset of action may decrease the likelihood of
cardiac complications. A third treatment option is to supple-
ment both T4 and T3. This provides T3 at a subtherapeutic
dose for immediate action and a loading dose of T4. T3 is
given as an initial IV dose of 10 µg, followed by 10 µg every 8
to 12 hours until there is clinical improvement and the patient
is able to take maintenance oral doses of T4. In addition, a
loading dose of 200 to 300 µg of T4 is given intravenously,
followed by 100 µg 24 hours later, and then a daily dose of
50 µg. The daily maintenance dose should then be adjusted
based on laboratory findings and clinical improvement. Ini-
tially, TSH and free T4 levels should be closely followed to
prevent overtreatment. After a stable daily dose is adopted,
patients should have their thyroid function reevaluated in 4
to 6 weeks.
Nonthyroidal Illness Syndrome
Abnormal circulating levels of thyroid hormone are associated
with many acute illnesses and do not necessarily indicate
pathology at the level of the hypothalamus or pituitary (Figure
21.1). Under normal conditions, T4 is converted by tissue
5′-monodeiodinase to T3, the metabolically active thyroid
hormone. During periods of carbohydrate deprivation or
illness, an alternative pathway of deiodination predominates
and rT3, a biologically inactive hormone, is created.46,47 The
clearance of rT3 also may be impaired because of the inhibited
5′-monodeiodinase activity.48 High circulating levels of proin-
flammatory cytokines and increased serum cortisol con­
centrations may promote this shift in deiodination.49-52
Medications frequently used in the ICU (e.g., amiodarone,
glucocorticoids, propranolol) also can contribute to depressed
T3 concentrations by inhibiting tissue 5′-monodeiodinase
activity. With mild illness, only T3 levels decline but with
increasing severity, a decrease in both T3 and T4 is observed
without a concomitant elevation in TSH. These changes once
were considered an adaptive response and the term euthyroid
sick syndrome was adopted. Increasing evidence suggests that
the thyroid abnormalities seen in critical illness reflect a state
of central hypothyroidism.53 The designation of euthyroid sick
syndrome has been abandoned in preference of the more
metabolically neutral term nonthyroidal illness syndrome
(NTIS), which also is referred to as the low T3 syndrome.4
TRH
Glucocorticoids
Dopamine
Dobutamine
TSH
Cytokines, glucocorticoids,
T4 amiodarone, propranolol inhibit
T3 rT3
Cytokines, glucocorticoids,
amiodarone, propranolol
inhibit
T2
Fig. 21.1  Changes in hypothalamic-pituitary-thyroid axis in critical
illness. rT3, Reverse triiodothyronine; TRH, thyrotropin-releasing
hormone; TSH, thyroid-stimulating hormone; T2, 3,5-Diiodo-L-thyronine;
T3, triiodothyronine; T4, thyroxine.
 Section II—Clinical and Laboratory Assessment 205
Low T3 and T4 levels and in particular a progressive decline
in T4 levels54 in critically ill patients may be associated with
higher mortality. However, replacement of low T4 levels in
NTIS with levothyroxine may not improve outcome.55 Van
den Berghe et al. suggest the thyroid function abnormalities
seen in prolonged critical illness reflect a maladaptive depres-
sion of the neuroendocrine drive that should be treated at the
hypothalamic level. In 20 critically ill adult patients, they
observed that coinfusion of TRH and growth hormone–
releasing peptide (GHRP)-2 was associated with the return of
pulsatile TSH secretion and the normalization of thyroid
hormone concentrations.56 Further clinical studies are needed
to establish the efficacy and benefit of this treatment strategy
in NTIS. Patients with elevated TSH levels, however, have a
“central” hypothyroidism and should be treated with thyroid
replacement therapy. Medications such as glucocorticoids,
dopamine, and dobutamine suppress TSH production and
may contribute to a state of central hypothyroidism.57
Untreated central hypothyroidism may produce symptoms
(e.g., decreased energy, neuropsychiatric symptoms, weight
gain) that mimic those seen during long-term recovery after
TBI, and a high index of suspicion is required to appropriately
diagnose hypothyroidism in these patients. In addition, the
low T3 syndrome may be associated with the need for pro-
longed mechanical ventilation.58 Recovery from the underly-
ing illness is associated with a normalization of thyroid
function with TSH values recovering before an improvement
in T4 levels is observed.59
Monitoring the Hypothalamic-
Pituitary-Thyroid Axis in the ICU
Routine thyroid function surveillance is not recommended in
the ICU because the majority of critically ill patients only
have transient or nonspecific abnormalities. Thyroid function
testing should be reserved for at-risk patients or those whose
clinical picture suggests thyroid dysfunction. There is signifi-
cant overlap among the laboratory values seen in various
states of thyroid dysfunction (Table 21.4) and so laboratory
abnormalities must be interpreted within the clinical context.
TSH and free T4 levels can be measured to monitor thyroid
function in ICU patients (see Figure 21.1). In general, an
elevated T4 level suggests hyperthyroidism, whereas a
depressed T4 concentration suggests hypothyroidism. TSH
levels typically are suppressed in hyperthyroidism but can
occasionally be elevated in central hyperthyroidism. TSH
Table 21.4  Diagnosis of Thyroid Axis
Dysfunction
Disorder T3 T4 TSH
Thyrotoxicosis High High Low
Hyperthyroidism High High Low (<0.01 mIU/L)
Myxedema Low Low High (rarely
normal or low)
NTIS Low Normal to low Normal to low
(0.01-0.1 mIU/L)
NTIS, Nonthyroidal illness syndrome; TSH, thyroid-stimulating hormone;
T3, triiodothyronine; T4, thyroxine.
concentrations are generally elevated in hypothyroidism.
Although not routinely measured, T3 levels can be useful
when the diagnosis of NTIS is entertained. Although T3 levels
are universally low in NTIS, T4 and TSH values are variable.
The diagnosis of thyroid dysfunction therefore should be
made in the context of the T4-TSH relationship. TSH should
be measured using a third-generation assay with a functional
sensitivity limit of 0.01 to 0.02 mIU/L to differentiate NTIS
from underlying thyroid dysfunction.60 Using this more sen-
sitive assay, most hospitalized patients with hyperthyroidism
will have a TSH value less than 0.01 mIU/L and those with
NTIS will have values between 0.01 and 0.1 mIU/L.61,62 Both
total T4 and free T4 have been used to evaluate NTIS. Assays
for free T4 theoretically provide information about the con-
centration of biologically active hormone; however, the tech-
nical difficulties of measuring free hormone levels in critical
illness may limit its usefulness.62 T4 levels in NTIS range from
normal to markedly suppressed.
Somatotroph and
Gonadotroph Dysfunction
During the acute phase of critical illness the pattern of growth
hormone (GH) secretion changes. In healthy people, GH is
released in a pulsatile fashion with peaks alternating with
nearly undetectable troughs. In the acutely ill ICU patient, the
baseline level of GH increases, as does the frequency of peaks
whereas serum levels of insulin-like growth factor (IGF)-1,
growth hormone–binding protein (GHBP), and the expres-
sion of GH receptor in peripheral tissues are decreased. It is
postulated that this is an adaptive and desirable response to
acute illness. The direct effects of the increased levels of GH
include lipolysis, antagonization of insulin, and stimulation
of the immune system and the decreased levels of IGF-1 and
GH receptor attenuate the somatotropic and anabolic effects
of GH.63
Dysfunction of the somatotrophs and gonadotrophs do not
produce acute life-threatening illness. However, there is accu-
mulating evidence that somatotroph dysfunction may play a
role in the “wasting syndrome” of prolonged critical illness in
which protein wasting occurs despite adequate caloric intake.
In prolonged ICU stays, GH pulsatility remains abnormal and
secretion decreases relative to the acute stage. IGF-1 levels also
remain low.64 Chronic GH deficiency may lead to a lack of
vigor, decreased exercise tolerance, a decrease in lean body
mass with redistribution of fat, and decreased social function-
ing. This GH deficiency then may limit rehabilitation poten-
tial in patients who survive their ICU stay. Although not
acutely life threatening, chronic dysfunction of luteinizing or
follicle-stimulating hormone secretion may result in sexual
dysfunction, mood disorders, osteoporosis, and alterations in
menstrual function in women.
Monitoring Somatotroph and Gonatroph
Function in the ICU
Somatotrophs or gonadotrophs are not routinely tested in the
ICU, in part because some of the observed changes may be
adaptive and there does not appear to be a benefit to correct
dysfunction in the acute setting. Guidelines suggest that GH
testing should be delayed until other pituitary defects have
206 Section II—Clinical and Laboratory Assessment
been corrected for a period of at least 3 to 6 months.65 Ongoing
research into the role of GH in prolonged critical illness may
support a role for testing in the chronic phases of ICU treat-
ment, although there is not enough evidence to support this
recommendation.
Treating Somatotroph and Gonadotroph
Dysfunction in the ICU
Currently there is no justification for GH replacement or
supplementation in the acute ICU setting. Although several
small studies have demonstrated improved nitrogen balance
and increased IGF-1 levels with GH replacement in prolonged
illness, two large, randomized trials demonstrated a significant
increase in mortality, length of stay, and length of ventilator
dependence with GH supplementation.66 Intervening at the
hypothalamic level, however, may offer some benefit in
patients with prolonged critical illness. Short term (2-5 days)
infusion of GHRP-2 and TRH in chronically ill ICU patients
has been shown to reestablish pulsatility, restore GH and TSH
secretion, improve peripheral responsiveness, and initiate a
shift toward anabolic pathways. Although encouraging, further
studies are needed to look at the effects of this treatment on
morbidity and mortality.67
Posterior Pituitary Dysfunction
Physiology
The posterior pituitary, or neurohypophysis, develops from an
evagination of the embryonic ventricular system. Unlike the
anterior pituitary, the posterior pituitary does not contain
glandular cells but is composed of terminal axons that origi-
nate in the supraoptic and paraventricular nuclei of the hypo-
thalamus. The posterior pituitary secretes two hormones:
vasopressin and oxytocin. Vasopressin or antidiuretic hormone
(ADH) is derived from a molecule synthesized in the supra-
optic and paraventricular nuclei of the hypothalamus. The
precursor molecule is packaged, cleaved to its active form, and
transported down axons to the posterior pituitary, where it is
stored and released into the systemic circulation via a capillary
plexus. ADH secretion is tightly regulated by changes in
plasma osmotic pressure, and plasma volume and acts on the
kidney to reduce excretion of free water and on the vasculature
to improve vascular tone. At physiologic levels the antidiuretic
effect is prominent. ADH has also been implicated in regulat-
ing social behaviors such as aggression and pair bonding. Oxy-
tocin is a hormone structurally similar to ADH but synthesized
in distinct neurons within the supraoptic and paraventricular
nuclei. It is responsible for uterine contraction and the lacta-
tion let-down reflex.
Disorders of Vasopressin
in the NCCU
Disorders of vasopressin are encountered frequently in the
NCCU. By contrast, oxytocin has little known physiologic
relevance in the care of NCCU patients. Disorders of vasopres-
sin may result from either too little (diabetes insipidus) or too
much (syndrome of inappropriate antidiuretic hormone
[SIADH]) secretion of the hormone. Central diabetes
insipidus (DI) is a consequence of diminished or absent ADH
secretion and results in the excretion of large amounts of free
water and should be differentiated from nephrogenic DI,
which results from resistance to the actions of ADH at the
kidney. Central DI is frequently encountered in the NCCU.
Approximately 30% of cases are idiopathic, 25% are associated
with tumors of the brain or pituitary, 20% result from intra-
cranial surgery, and 16% are post-traumatic. SIADH, on the
other hand, occurs when the hypothalamus releases too much
ADH. This results in the retention of free water, hyponatremia,
volume expansion, and inappropriately concentrated urine.
Monitoring Vasopressin Dysfunction  
in the ICU
Vasopressin levels are difficult to measure because the mole-
cule is unstable, extensively bound to platelets, and rapidly
cleared. These difficulties have prevented its routine clinical
measurement. Copeptin, on the other hand, is a stable mole-
cule co-synthesized with vasopressin and mirrors vasopressin
release. Copeptin has been shown to correlate with vasopres-
sin levels across a range of serum osmolarities in healthy vol-
unteers and can be useful to help discriminate DI from other
similar clinical presentations and SIADH from sodium-
wasting states.68-70 Stimulated copeptin levels measured 45
minutes after an insulin infusion appear also to be helpful.71
However, further validation studies are necessary and as such,
vasopressin dysfunction remains a clinical diagnosis.
Diabetes Insipidus
A diagnosis of DI can be made when urine output is high
(>250 mL/hr or >3 mL/kg/hr in a pediatric population) and
the urine is dilute (urine osmolality 50-150 mOsm/L or spe-
cific gravity 1.001-1.005). A diagnosis of DI also requires
serum sodium to be normal or high. DI cannot occur in the
face of primary adrenal dysfunction, because some degree of
mineralocorticoid activity is obligatory for the kidney to make
free water. In rare cases in which the diagnosis remains unclear,
a water deprivation test may be used to provoke ADH release.
All IVs are stopped and the patient is allowed nothing per
mouth. Urine osmolality is checked every hour. A decrease
in urine output and a urine osmolality between 600 and
850 mOsm/L indicate a normal response and excludes DI.
The test is terminated after 6 to 8 hours or when urine osmo-
lality plateaus (<30 mOsm change in 3 hours) indicating an
abnormal response. If the patient does not respond normally
to the water deprivation test, then exogenous ADH (5 U
aqueous vasopressin [Pitressin] subcutaneously) is adminis-
tered. Exogenous ADH should increase urine osmolality to
more than 300 mOsm/L. A less than 5% increase in urine
osmolality following vasopressin indicates normal ADH levels,
a 6% to 67% increase indicates a partial ADH deficiency, and
a greater than 67% increase indicates severe ADH deficiency.
The patient must be monitored closely during the water
deprivation test because fatal dehydration can result.
Syndrome of Inappropriate
Antidiuretic Hormone
SIADH is a clinical diagnosis that does not have a single con-
firmatory test. Bartter and Schwarz proposed the classic
 Section II—Clinical and Laboratory Assessment 207
diagnostic criteria in 1967 that includes: (1) hyponatremia
with corresponding hypo-osmolality, (2) continued renal
excretion of sodium, (3) urine that is less than maximally
dilute, (4) absence of clinical evidence of volume depletion,
(5) absence of other causes of hyponatremia, and (6) correc-
tion of hyponatremia by fluid restriction.72 More specifically,
serum sodium is generally less than 134 mEq/L, serum osmo-
lality is less than 280 mOsm, and urinary sodium is greater
than 18 mEq/L but can often be as high as 50 to 150 mEq/L.
If asymptomatic and mild, acute SIADH may be treated by
fluid restriction (<1 L/day). More severe cases may require
correction with hypertonic saline and furosemide. The rate of
correction should be monitored closely because central
pontine myelinolysis may occur with too rapid a correction of
serum sodium. If the SIADH is chronic in nature, then a more
generous fluid restriction of 1200 to 1800 mL/day with furo-
semide may be appropriate. Demeclocyline also may be used
to partially antagonize the effects of ADH on the renal tubules.
Cerebral Salt Wasting
It is important to differentiate SIADH from cerebral salt
wasting (CSW). Like SIADH, CSW occurs in the setting
of intracranial disease and produces hyponatremia. Unlike
SIADH, CSW results in a decrease in extracellular fluid
volume. Because laboratory values may be similar or unreveal-
ing, volume status is the principal method to differentiate
SIADH from CSW. Volume status may be measured by central
venous pressure, a pulmonary capillary wedge pressure, or a
nuclear medicine study of plasma volume. Low volume sup-
ports the diagnosis of CSW. A high potassium in the setting
of hyponatremia is incompatible with the diagnosis of SIADH
and supports a diagnosis of CSW.73 Fluid restriction exacer-
bates CSW and should be avoided. The primary goals of treat-
ment in CSW are sodium and volume replacement. Salt may
be replaced orally with salt tabs or with IV hypertonic saline.
Rapid correction should be avoided to reduce the risk of
central pontine myelinolysis.
Endocrine Consequences of Specific
Diseases Managed in the NCCU
Endocrine Consequences   centage of DI cases may be persistent.82,84,85
of Pituitary Surgery
Surgery on or near the pituitary gland can impact endocrine
function. For example, Fatemi et al. observed a 5% incidence
of worse anterior pituitary function, whereas 24% of patients
had improved anterior pituitary function following trans­
sphenoidal resection of pituitary adenomas.74 Pituitary surgery
also can impact posterior pituitary function and cause shifts
in water and electrolyte balance. Isolated DI is the most
common deficit seen after transsphenoidal surgery and occurs
in up to 39% of patients who develop hypernatremia postop-
eratively. Diuresis is typically maximal in the first and second
postoperative days. Isolated hyponatremia also can occur, and
up to 21% of patients develop SIADH. The development of
hyponatremia, however, often is delayed, reaching a nadir on
postoperative day 9 or 10. In the majority of cases, SIADH is
transient and responds to simple fluid restriction. Approxi-
mately 16% of patients have both DI and hyponatremia
during their postoperative recovery.75
Endocrine Consequences  
of Traumatic Brain Injury
Hypopituitarism is a recognized complication of TBI. Al-
though first reported in 1918, hypopituitarism following
TBI was long thought to be a relatively uncommon compli-
cation. More recent literature, however, suggests that the
prevalence of post-traumatic hypopituitarism (specifically
anterior pituitary hormone deficiencies) following TBI
ranges between 28% and 68%. These deficits may manifest
either in the acute period following injury or in a delayed
fashion. Some cases of pituitary dysfunction that present in
the acute phase spontaneously resolve within 6 months post
injury. Similarly, the delayed onset of anterior pituitary dys-
function is generally evident within 6 months following
head injury.76,77
Endocrine dysfunction in TBI may be associated with
anterior lobe necrosis, posterior lobe hemorrhage, or stalk
laceration or vascular injury to the hypothalamus. In
addition TBI is hypercatabolic, and nutritional status may
influence endocrine function.78 GH deficiency is the most
common hormone deficiency seen following TBI. GH-
secreting somatotroph cells are particularly vulnerable to vas-
cular insult because they are located in the lateral aspect of
the pituitary gland and derive their blood supply from the
hypothalamo-pituitary portal vessels. In contrast ACTH- and
TSH-producing cells are more protected because of their
ventral and medial location and their dual blood supply from
the portal vessels and branches of the anterior pituitary
artery.79,80 However, abnormalities in the ACTH axis may still
complicate between 4% and 50% of TBI patients. The inci-
dence depends in part on the method used to assess the
ACTH axis and when the tests are performed because hor-
monal function is a dynamic process after TBI.81 Suppression
of gonadotropins is common after TBI but most levels recover
relatively quickly and nearly all within a year.82 However,
patients who are hypogonadal when they start rehabilitation
have lower Functional Independence Measure (FIM) scores at
discharge.83 Posterior pituitary dysfunction in the form of
either DI or SIADH also is seen in TBI but seems unrelated to
the incidence of anterior hypopituitarism. Although DI or
SIADH may be present transiently in the acute phase of
injury, SIADH almost always resolves, whereas a small per-
How best to screen patients with moderate or severe TBI
for endocrine dysfunction is still being elucidated. In the acute
period following TBI, ACTH deficiency can lead to acute
adrenal insufficiency and life-threatening complications. Con-
sequently, some centers screen all patients with moderate or
severe TBI with morning cortisol levels for the first 5 post-
injury days.86 If a deficiency of ACTH is discovered, screening
for pituitary hormone deficiencies seems reasonable. There is
no evidence, however, to support the routine screening for
thyroid, GH, or gonadal hormone deficiencies in the acute
period. Rather, anterior pituitary dysfunction should be
assessed at either 3 or 6 months after moderate or severe
injury.76 Guidelines released in 2005 suggest prospective and
retrospective screening of moderate and severe TBI patients
with the subsequent replacement with vasopressin, glucocor-
ticoid and thyroid hormones. Gonadal and GH replacement
should be postponed until the existence of these deficits is
confirmed following the correction of other hormone
208 Section II—Clinical and Laboratory Assessment
deficiencies. If gonadal and GH deficiencies are proven, then
replacement may help improve rehabilitation by increasing
muscle mass, energy levels, and emotional stability and reduc-
ing the sense of social isolation.80 The role for GH replacement
after TBI, however, is still to be elucidated.81
Endocrine Consequences of   Immunol 2005;174:2250–7.
Subarachnoid Hemorrhage
In the acute phase after SAH hypogonadotropic hypogonad-
ism is common, but rates of ACTH and TSH deficiency are
rare (7.1% and 1.8%, respectively).87,88 Long-term hypopitu-
itarism is a known complication following SAH, and anterior
pituitary dysfunction is observed in between 37.5% and 55%
of patients following SAH, although in part this depends on
the evaluation methods used.77 Persistent posterior pituitary
dysfunction, however, is less common. However, few studies
have looked specifically at the effect of treating hypopitu­
itarism after SAH. Preliminary data suggest that chronic
neuroendocrine dysfunction contributes to depression, and
disturbances in quality of life and sleep in SAH survivors.89
Sodium imbalance in SAH associated with the SIADH
secretion or CSW syndrome are frequent after SAH. However,
there are conflicting reports regarding the incidence of HPA
dysfunction in the acute setting following SAH.90 Whereas
some authors have reported a decrease in cortisol and
ACTH23,91 others have found no change92 or an increase93 in
cortisol levels. In the subgroup of SAH patients that are refrac-
tory to vasopressor therapy, however, the incidence of CIRCI
after SAH is reported to be as high as 68%.94 In addition there
is some evidence to suggest levels of cortisol or ACTH may be
associated with global cerebral edema or cerebral ischemia in
the first week after SAH.95 This may be particularly important
in management of vasospasm and delayed cerebral ischemia.
In some patients it may be difficult or impossible to maintain
adequate perfusion pressure to prevent ischemia because of
adrenal insufficiency.96 As such, testing for adrenal insuffi-
ciency in patients with SAH with hypotension refractory to
vasopressors should be considered. There may be modest
outcome benefits when using fludrocortisones in some
patients.97 There is limited study on the role of corticosteroids
in SAH; the few published studies suggest that routine corti-
costeroid use is not indicated.98,99
Conclusion
Endocrine abnormalities often may be overlooked and under-
diagnosed in the NCCU. Symptoms, both acute and chronic,
may be masked by or mimic sequelae of neurologic injury. It
is therefore important that a high index of suspicion be main-
tained in both the acute setting and in follow-up care of the
NCCU population. TBI patients with moderate to severe
injury should undergo testing of the pituitary axis both in the
acute setting and at follow-up. Vasopressin, thyroid hormone,
and cortisol deficiencies must be addressed immediately.
Although there is no evidence to support replacing GH and
sex-steroid hormones in the acute setting, deficiencies in these
hormones may impair rehabilitation potential and should be
considered after other pituitary function has been optimized
at follow-up. However, further research is needed to establish
the benefits and risks of endocrine manipulation in the criti-
cally ill.
References
1. Cooper MS, Stewart PM. Adrenal insufficiency in critical illness. J Intensive
Care Med 2007;22:348–62.
2. Cohen J, Venkatesh B. Assessment of tissue cortisol activity. Crit Care Resusc
2009;11:287–9.
3. Leonard MO, Godson C, Brady HR, et al. Potentiation of glucocorticoid
activity in hypoxia through induction of the glucocorticoid receptor. J
4. Mebis L, Van den Berghe G. Thyroid axis function and dysfunction in
critical illness. Best Pract Res Clin Endocrinol Metab 2011 Oct;25(5):745–57.
5. Clark PM, Gordon K. Challenges for the endocrine laboratory in critical
illness. Best Pract Res Clin Endocrinol Metab 2011;25(5):847–59.
6. Thomas Z, Bandali F, McCowen K, et al. Drug-induced endocrine disorders
in the intensive care unit. Crit Care Med 2010;38(6 Suppl):S219–30.
Review.
7. Vuong C, Van Uum SH, O’Dell LE, et al. The effects of opioids and opioid
analogs on animal and human endocrine systems. Endocr Rev
2010;31:98–132.
8. Chin R. Adrenal crisis. Crit Care Clin 1991;7:23–42.
9. Cuthbertson BH, Sprung CL, Annane D, et al. The effects of etomidate on
adrenal responsiveness and mortality in patients with septic shock. Intensive
Care Med 2009;35:1868–76.
10. Sonino N. The use of ketoconazole as an inhibitor of steroid production.
N Engl J Med 1987;317:812–8.
11. Ashby H, DiMattina M, Linehan WM, et al. The inhibition of human
adrenal steroidogenic enzyme activities by suramin. J Clin Endocrinol Metab
1989;68:505–8.
12. Kyriazopoulou V, Parparousi O, Vagenakis AG. Rifampicin-induced adrenal
crisis in addisonian patients receiving corticosteroid replacement therapy.
J Clin Endocrinol Metab 1984;59:1204–6.
13. Wagner RL, White PF, Kan PB, et al. Inhibition of adrenal steroidogenesis by
the anesthetic etomidate. N Engl J Med 1984;310:1415–21.
14. Elias AN, Gwinup G. Effects of some clinically encountered drugs on steroid
synthesis and degradation. Metabolism 1980;29:582–95.
15. Arlt W, Allolio B. Adrenal insufficiency. Lancet 2003;361:1881–93.
16. Schmiegelow M, Feldt-Rasmussen U, Rasmussen AK, et al. Assessment of the
hypothalamo-pituitary-adrenal axis in patients treated with radiotherapy
and chemotherapy for childhood brain tumor. J Clin Endocrinol Metab
2003;88:3149–54.
17. Marik PE, Pastores SM, Annane D, et al. Recommendations for the diagnosis
and management of corticosteroid insufficiency in critically ill adult patients:
consensus statements from an international task force by the American
College of Critical Care Medicine. Crit Care Med 2008;36:1937–49.
18. Meduri GU, Yates CR. Systemic inflammation-associated glucocorticoid
resistance and outcome of ARDS. Ann N Y Acad Sci 2004;1024:24–53.
19. Annane D, Sebille V, Charpentier C, et al. Effect of treatment with low doses
of hydrocortisone and fludrocortisone on mortality in patients with septic
shock. JAMA 2002;288:862–71.
20. Ligtenberg JJ, Zijlstra JG. The relative adrenal insufficiency syndrome
revisited: which patients will benefit from low-dose steroids? Curr Opin Crit
Care 2004;10:456–60.
21. Sprung CL, Annane D, Keh D, et al. Hydrocortisone therapy for patients
with septic shock. N Engl J Med 2008;358:111–24.
22. Cronin CC, Callaghan N, Kearney PJ, et al. Addison disease in patients
treated with glucocorticoid therapy. Arch Intern Med 1997;157:456–8.
23. Dimopoulou I, Tsagarakis S, Kouyialis AT, et al. Hypothalamic-pituitary-
adrenal axis dysfunction in critically ill patients with traumatic brain injury:
incidence, pathophysiology, and relationship to vasopressor dependence and
peripheral interleukin-6 levels. Crit Care Med 2004;32:404–8.
24. Dimopoulou I, Tsagarakis S, Douka E, et al. The low-dose corticotropin
stimulation test in acute traumatic and non-traumatic brain injury:
incidence of hypo-responsiveness and relationship to outcome. Intensive
Care Med 2004;30:1216–9.
25. Glowniak JV, Loriaux DL. A double-blind study of perioperative steroid
requirements in secondary adrenal insufficiency. Surgery 1997;121:123–9.
26. Cooper MS, Stewart PM. Corticosteroid insufficiency in acutely ill patients.
N Engl J Med 2003;348:727–34.
27. Nieboer P, van der Werf TS, Beentjes JA, et al. Catecholamine dependency in
a polytrauma patient: relative adrenal insufficiency? Intensive Care Med
2000;26:125–7.
28. Beishuizen A, Thijs LG, Vermes I. Patterns of corticosteroid-binding globulin
and the free cortisol index during septic shock and multitrauma. Intensive
Care Med 2001;27:1584–91.
29. Klaff LS, Wisse BE. Current controversy related to glucocorticoid and insulin
therapy in the intensive care unit. Endocr Pract 2007;13:542–9.
 Section II—Clinical and Laboratory Assessment 209
30. Molenaar N, Johan Groeneveld AB, Dijstelbloem HM, et al. Assessing
adrenal insufficiency of corticosteroid secretion using free versus total
cortisol levels in critical illness. Intensive Care Med 2011;37:1986–93.
31. Annane D, Sebille V, Charpentier C, et al. Effect of treatment with low doses
of hydrocortisone and fludrocortisone on mortality in patients with septic
shock. JAMA 2002;288:862–71.
32. Sprung CL, Annane D, Keh D, et al. Hydrocortisone therapy for patients
with septic shock. N Engl J Med 2008;358:111–24.
33. Dellinger RP, Levy MM, Carlet JM, et al. Surviving Sepsis Campaign:
international guidelines for management of severe sepsis and septic shock:
2008. Crit Care Med 2008;36:296–327.
34. Burch HB, Wartofsky L. Life-threatening thyrotoxicosis. thyroid storm.
Endocrinol Metab Clin North Am 1993;22:263–77.
35. Sanders V. Neurologic manifestations of myxedema. N Engl J Med 1962;266:
599–603 concl.
36. Klein I, Ojamaa K. Thyroid hormone and the cardiovascular system. N Engl
J Med 2001;344:501–9.
37. Hylander B, Noree LO, Rosenqvist U, et al. Myxedema coma: a treatment
schedule. Lakartidningen 1980;77:1833–5.
38. Wartofsky L. Myxedema coma. Endocrinol Metab Clin North Am 2006;35:
687–98, vii–viii.
39. Iwasaki Y, Oiso Y, Yamauchi K, et al. Osmoregulation of plasma vasopressin
in myxedema. J Clin Endocrinol Metab 1990;70:534–9.
40. Skowsky WR, Kikuchi TA. The role of vasopressin in the impaired water
excretion of myxedema. Am J Med 1978;64:613–21.
41. Fliers E, Wiersinga WM. Myxedema coma. Rev Endocr Metab Disord
2003;4:137–41.
42. Zulewski H, Muller B, Exer P, et al. Estimation of tissue hypothyroidism by a
new clinical score: evaluation of patients with various grades of
hypothyroidism and controls. J Clin Endocrinol Metab 1997;82:771–6.
43. Bigos ST, Ridgway EC, Kourides IA, et al. Spectrum of pituitary alterations
with mild and severe thyroid impairment. J Clin Endocrinol Metab 1978;46:
317–25.
44. Pereira VG, Haron ES, Lima-Neto N, et al. Management of myxedema coma:
report on three successfully treated cases with nasogastric or intravenous
administration of triiodothyronine. J Endocrinol Invest 1982;5:331–4.
45. Chernow B, Burman KD, Johnson DL, et al. T3 may be a better agent than
T4 in the critically ill hypothyroid patient: evaluation of transport across the
blood-brain barrier in a primate model. Crit Care Med 1983;11:99–104.
46. Harris AR, Fang SL, Vagenakis AG, et al. Effect of starvation, nutriment
replacement, and hypothyroidism on in vitro hepatic T4 to T3 conversion in
the rat. Metabolism 1978;27:1680–90.
47. Davidson MB, Chopra IJ. Effect of carbohydrate and noncarbohydrate
sources of calories on plasma 3,5,3´-triiodothyronine concentrations in man.
J Clin Endocrinol Metab 1979;48:577–81.
48. Chopra IJ. An assessment of daily production and significance of thyroidal
secretion of 3, 3´, 5´-triiodothyronine (reverse T3) in man. J Clin Invest
1976;58:32–40.
49. Nagaya T, Fujieda M, Otsuka G, et al. A potential role of activated NF-kappa
B in the pathogenesis of euthyroid sick syndrome. J Clin Invest
2000;106:393–402.
50. Monig H, Arendt T, Meyer M, et al. Activation of the hypothalamo-
pituitary-adrenal axis in response to septic or non-septic diseases:
implications for the euthyroid sick syndrome. Intensive Care Med 1999;25:
1402–6.
A complete list of references for this chapter can be found online at
www.expertconsult.com.
 Section II—Clinical and Laboratory Assessment 209.e1
References
1. Cooper MS, Stewart PM. Adrenal insufficiency in critical illness. J Intensive
Care Med 2007;22:348–62.
2. Cohen J, Venkatesh B. Assessment of tissue cortisol activity. Crit Care Resusc
2009;11:287–9.
3. Leonard MO, Godson C, Brady HR, et al. Potentiation of glucocorticoid
activity in hypoxia through induction of the glucocorticoid receptor.
J Immunol 2005;174:2250–7.
4. Mebis L, Van den Berghe G. Thyroid axis function and dysfunction in
critical illness. Best Pract Res Clin Endocrinol Metab 2011 Oct;25(5):745–57.
5. Clark PM, Gordon K. Challenges for the endocrine laboratory in critical
illness. Best Pract Res Clin Endocrinol Metab 2011;25(5):847–59.
6. Thomas Z, Bandali F, McCowen K, et al. Drug-induced endocrine disorders
in the intensive care unit. Crit Care Med 2010;38(6 Suppl):S219–30. Review.
7. Vuong C, Van Uum SH, O’Dell LE, et al. The effects of opioids and opioid
analogs on animal and human endocrine systems. Endocr Rev 2010;31:
98–132.
8. Chin R. Adrenal crisis. Crit Care Clin 1991;7:23–42.
9. Cuthbertson BH, Sprung CL, Annane D, et al. The effects of etomidate on
adrenal responsiveness and mortality in patients with septic shock. Intensive
Care Med 2009;35:1868–76.
10. Sonino N. The use of ketoconazole as an inhibitor of steroid production.
N Engl J Med 1987;317:812–8.
11. Ashby H, DiMattina M, Linehan WM, et al. The inhibition of human
adrenal steroidogenic enzyme activities by suramin. J Clin Endocrinol Metab
1989;68:505–8.
12. Kyriazopoulou V, Parparousi O, Vagenakis AG. Rifampicin-induced adrenal
crisis in addisonian patients receiving corticosteroid replacement therapy.
J Clin Endocrinol Metab 1984;59:1204–6.
13. Wagner RL, White PF, Kan PB, et al. Inhibition of adrenal steroidogenesis by
the anesthetic etomidate. N Engl J Med 1984;310:1415–21.
14. Elias AN, Gwinup G. Effects of some clinically encountered drugs on steroid
synthesis and degradation. Metabolism 1980;29:582–95.
15. Arlt W, Allolio B. Adrenal insufficiency. Lancet 2003;361:1881–93.
16. Schmiegelow M, Feldt-Rasmussen U, Rasmussen AK, et al. Assessment of the
hypothalamo-pituitary-adrenal axis in patients treated with radiotherapy
and chemotherapy for childhood brain tumor. J Clin Endocrinol Metab
2003;88:3149–54.
17. Marik PE, Pastores SM, Annane D, et al. Recommendations for the diagnosis
and management of corticosteroid insufficiency in critically ill adult patients:
consensus statements from an international task force by the American
College of Critical Care Medicine. Crit Care Med 2008;36:1937–49.
18. Meduri GU, Yates CR. Systemic inflammation-associated glucocorticoid
resistance and outcome of ARDS. Ann N Y Acad Sci 2004;1024:24–53.
19. Annane D, Sebille V, Charpentier C, et al. Effect of treatment with low doses
of hydrocortisone and fludrocortisone on mortality in patients with septic
shock. JAMA 2002;288:862–71.
20. Ligtenberg JJ, Zijlstra JG. The relative adrenal insufficiency syndrome
revisited: which patients will benefit from low-dose steroids? Curr Opin Crit
Care 2004;10:456–60.
21. Sprung CL, Annane D, Keh D, et al. Hydrocortisone therapy for patients
with septic shock. N Engl J Med 2008;358:111–24.
22. Cronin CC, Callaghan N, Kearney PJ, et al. Addison disease in patients
treated with glucocorticoid therapy. Arch Intern Med 1997;157:456–8.
23. Dimopoulou I, Tsagarakis S, Kouyialis AT, et al. Hypothalamic-pituitary-
adrenal axis dysfunction in critically ill patients with traumatic brain injury:
incidence, pathophysiology, and relationship to vasopressor dependence and
peripheral interleukin-6 levels. Crit Care Med 2004;32:404–8.
24. Dimopoulou I, Tsagarakis S, Douka E, et al. The low-dose corticotropin
stimulation test in acute traumatic and non-traumatic brain injury:
incidence of hypo-responsiveness and relationship to outcome. Intensive
Care Med 2004;30:1216–9.
25. Glowniak JV, Loriaux DL. A double-blind study of perioperative steroid
requirements in secondary adrenal insufficiency. Surgery 1997;121:123–9.
26. Cooper MS, Stewart PM. Corticosteroid insufficiency in acutely ill patients.
N Engl J Med 2003;348:727–34.
27. Nieboer P, van der Werf TS, Beentjes JA, et al. Catecholamine dependency in
a polytrauma patient: relative adrenal insufficiency? Intensive Care Med
2000;26:125–7.
28. Beishuizen A, Thijs LG, Vermes I. Patterns of corticosteroid-binding globulin
and the free cortisol index during septic shock and multitrauma. Intensive
Care Med 2001;27:1584–91.
29. Klaff LS, Wisse BE. Current controversy related to glucocorticoid and insulin
therapy in the intensive care unit. Endocr Pract 2007;13:542–9.
30. Molenaar N, Johan Groeneveld AB, Dijstelbloem HM, et al. Assessing
adrenal insufficiency of corticosteroid secretion using free versus total
cortisol levels in critical illness. Intensive Care Med 2011;37:1986–93.
31. Annane D, Sebille V, Charpentier C, et al. Effect of treatment with low doses
of hydrocortisone and fludrocortisone on mortality in patients with septic
shock. JAMA 2002;288:862–71.
32. Sprung CL, Annane D, Keh D, et al. Hydrocortisone therapy for patients
with septic shock. N Engl J Med 2008;358:111–24.
33. Dellinger RP, Levy MM, Carlet JM, et al. Surviving Sepsis Campaign:
international guidelines for management of severe sepsis and septic shock:
2008. Crit Care Med 2008;36:296–327.
34. Burch HB, Wartofsky L. Life-threatening thyrotoxicosis: thyroid storm.
Endocrinol Metab Clin North Am 1993;22:263–77.
35. Sanders V. Neurologic manifestations of myxedema. N Engl J Med
1962;266:599–603 concl.
36. Klein I, Ojamaa K. Thyroid hormone and the cardiovascular system.
N Engl J Med 2001;344:501–9.
37. Hylander B, Noree LO, Rosenqvist U, et al. Myxedema coma: a treatment
schedule. Lakartidningen 1980;77:1833–5.
38. Wartofsky L. Myxedema coma. Endocrinol Metab Clin North Am 2006;35:
687–98, vii–viii.
39. Iwasaki Y, Oiso Y, Yamauchi K, et al. Osmoregulation of plasma vasopressin
in myxedema. J Clin Endocrinol Metab 1990;70:534–9.
40. Skowsky WR, Kikuchi TA. The role of vasopressin in the impaired water
excretion of myxedema. Am J Med 1978;64:613–21.
41. Fliers E, Wiersinga WM. Myxedema coma. Rev Endocr Metab Disord 2003;
4:137–41.
42. Zulewski H, Muller B, Exer P, et al. Estimation of tissue hypothyroidism by a
new clinical score: evaluation of patients with various grades of
hypothyroidism and controls. J Clin Endocrinol Metab 1997;82:771–6.
43. Bigos ST, Ridgway EC, Kourides IA, et al. Spectrum of pituitary alterations
with mild and severe thyroid impairment. J Clin Endocrinol Metab 1978;
46:317–25.
44. Pereira VG, Haron ES, Lima-Neto N, et al. Management of myxedema coma:
report on three successfully treated cases with nasogastric or intravenous
administration of triiodothyronine. J Endocrinol Invest 1982;5:331–4.
45. Chernow B, Burman KD, Johnson DL, et al. T3 may be a better agent than
T4 in the critically ill hypothyroid patient: evaluation of transport across the
blood-brain barrier in a primate model. Crit Care Med 1983;11:99–104.
46. Harris AR, Fang SL, Vagenakis AG, et al. Effect of starvation, nutriment
replacement, and hypothyroidism on in vitro hepatic T4 to T3 conversion in
the rat. Metabolism 1978;27:1680–90.
47. Davidson MB, Chopra IJ. Effect of carbohydrate and noncarbohydrate
sources of calories on plasma 3,5,3´-triiodothyronine concentrations in man.
J Clin Endocrinol Metab 1979;48:577–81.
48. Chopra IJ. An assessment of daily production and significance of thyroidal
secretion of 3, 3´, 5´-triiodothyronine (reverse T3) in man. J Clin Invest
1976;58:32–40.
49. Nagaya T, Fujieda M, Otsuka G, et al. A potential role of activated NF-kappa
B in the pathogenesis of euthyroid sick syndrome. J Clin Invest
2000;106:393–402.
50. Monig H, Arendt T, Meyer M, et al. Activation of the hypothalamo-
pituitary-adrenal axis in response to septic or non-septic diseases:
implications for the euthyroid sick syndrome. Intensive Care Med
1999;25:1402–6.
51. Corssmit EP, Heyligenberg R, Endert E, et al. Acute effects of interferon-
alpha administration on thyroid hormone metabolism in healthy men.
J Clin Endocrinol Metab 1995;80:3140–4.
52. Chopra IJ, Williams DE, Orgiazzi J, et al. Opposite effects of
dexamethasone on serum concentrations of 3,3´,5´-triiodothyronine (reverse
T3) and 3,3´5-triiodothyronine (T3). J Clin Endocrinol Metab 1975;41:
911–20.
53. DeGroot LJ. “Non-thyroidal illness syndrome” is functional central
hypothyroidism, and if severe, hormone replacement is appropriate in light
of present knowledge. J Endocrinol Invest 2003;26:1163–70.
54. Meyer S, Schuetz P, Wieland M, et al. Low triiodothyronine syndrome: a
prognostic marker for outcome in sepsis? Endocrine 2011;39(2):167–74.
Epub 2011, Jan 6.
55. Brent GA, Hershman JM. Thyroxine therapy in patients with severe
nonthyroidal illnesses and low serum thyroxine concentration. J Clin
Endocrinol Metab 1986;63:1–8.
56. Van den Berghe G, de Zegher F, Baxter RC, et al. Neuroendocrinology of
prolonged critical illness: effects of exogenous thyrotropin-releasing
hormone and its combination with growth hormone secretagogues. J Clin
Endocrinol Metab 1998;83:309–19.
209.e2 Section II—Clinical and Laboratory Assessment
57. Roberts CG, Ladenson PW. Hypothyroidism. Lancet 2004;363:793–803.
58. Bello G, Pennisi MA, Montini L, et al. Nonthyroidal illness syndrome and
prolonged mechanical ventilation in patients admitted to the ICU. Chest
2009;135:1448–54.
59. Hamblin PS, Dyer SA, Mohr VS, et al. Relationship between thyrotropin and
thyroxine changes during recovery from severe hypothyroxinemia of critical
illness. J Clin Endocrinol Metab 1986;62:717–22.
60. Demers LM. Thyroid disease: pathophysiology and diagnosis. Clin Lab Med
2004;24:19–28.
61. Spencer CA, LoPresti JS, Patel A, et al. Applications of a new
chemiluminometric thyrotropin assay to subnormal measurement. J Clin
Endocrinol Metab 1990;70:453–60.
62. Stockigt J. Assessment of thyroid function: towards an integrated
laboratory: clinical approach. Clin Biochem Rev 2003;24:109–22.
63. Van den Berghe G. Novel insights into the neuroendocrinology of critical
illness. Eur J Endocrinol 2000;143:1–13.
64. Van den Berghe G, de Zegher F, Bouillon R. Clinical review 95: acute and
prolonged critical illness as different neuroendocrine paradigms. J Clin
Endocrinol Metab 1998;83:1827–34.
65. Sonksen PH, Christiansen JS. Consensus guidelines for the diagnosis and
treatment of adults with growth hormone deficiency. Growth Hormone
Research Society. Growth Horm IGF Res 1998;8(Suppl B):89–92.
66. Takala J, Ruokonen E, Webster NR, et al. Increased mortality associated with
growth hormone treatment in critically ill adults. N Engl J Med 1999;
341:785–92.
67. Van den Berghe G, Wouters P, Weekers F, et al. Reactivation of pituitary
hormone release and metabolic improvement by infusion of growth
hormone-releasing peptide and thyrotropin-releasing hormone in
patients with protracted critical illness. J Clin Endocrinol Metab
1999;84:1311–23.
68. Balanescu S, Kopp P, Gaskill MB, et al. Correlation of plasma copeptin and
vasopressin concentrations in hypo-, iso-, and hyperosmolar states. J Clin
Endocrinol Metab 2011;96:1046–52.
69. Morgenthaler NG, Struck J, Alonso C, et al. Assay for the measurement of
copeptin, a stable peptide derived from the precursor of vasopressin. Clin
Chem 2006;52:112–9.
70. Fenske W, Stork S, Blechschmidt A, et al. Copeptin in the differential
diagnosis of hyponatremia. J Clin Endocrinol Metab 2009;94:123–9.
71. Katan M, Morgenthaler NG, Dixit KC, et al. Anterior and posterior pituitary
function testing with simultaneous insulin tolerance test and a novel
copeptin assay. J Clin Endocrinol Metab 2007;92:2640–3.
72. Bartter FC, Schwartz WB. The syndrome of inappropriate secretion of
antidiuretic hormone. Am J Med 1967;42:790–806.
73. Harrigan MR. Cerebral salt wasting syndrome: a review. Neurosurgery
1996;38:152–60.
74. Fatemi N, Dusick JR, Mattozo C, et al. Pituitary hormonal loss and recovery
after transsphenoidal adenoma removal. Neurosurgery 2008;63:
709–18; discussion 18–19.
75. Kristof RA, Rother M, Neuloh G, et al. Incidence, clinical manifestations,
and course of water and electrolyte metabolism disturbances following
transsphenoidal pituitary adenoma surgery: a prospective observational
study. J Neurosurg 2009;111(3):555–62.
76. Behan LA, Agha A. Endocrine consequences of adult traumatic brain injury.
Horm Res 2007;68(Suppl 5):18–21.
77. Schneider HJ, Kreitschmann-Andermahr I, Ghigo E, et al.
Hypothalamopituitary dysfunction following traumatic brain injury and
aneurysmal subarachnoid hemorrhage: a systematic review. JAMA 2007;
298:1429–38.
78. Chourdakis M, Kraus MM, Tzellos T, et al. Effect of early compared with
delayed enteral nutrition on endocrine function in patients with traumatic
brain injury: an open-labeled randomized trial. JPEN J Parenter Enteral Nutr
2012;36(1):108–16. Epub 2011, Sep 30.
79. Kelly DF, Gonzalo IT, Cohan P, et al. Hypopituitarism following traumatic
brain injury and aneurysmal subarachnoid hemorrhage: a preliminary
report. J Neurosurg 2000;93:743–52.
80. Ghigo E, Masel B, Aimaretti G, et al. Consensus guidelines on screening for
hypopituitarism following traumatic brain injury. Brain Inj 2005;19:
711–24.
81. Hannon MJ, Sherlock M, Thompson CJ. Pituitary dysfunction following
traumatic brain injury or subarachnoid haemorrhage: in “Endocrine
Management in the Intensive Care Unit.” Best Pract Res Clin Endocrinol
Metab 2011;25(5):783–98.
82. Agha A, Phillips J, O’Kelly P, et al. The natural history of post-traumatic
hypopituitarism: implications for assessment and treatment. Am J Med
2005;118:1416.
83. Carlson NE, Brenner LA, Wierman ME, et al. Hypogonadism on admission
to acute rehabilitation is correlated with lower functional status at admission
and discharge. Brain Injury 2009;23(4):336–44.
84. Schneider HJ, Schneider M, Saller B, et al. Prevalence of anterior pituitary
insufficiency 3 and 12 months after traumatic brain injury. Eur J Endocrinol
2006;154:259–65.
85. Aimaretti G, Ambrosio MR, Di Somma C, et al. Residual pituitary function
after brain injury-induced hypopituitarism: a prospective 12-month study.
J Clin Endocrinol Metab 2005;90:6085–92.
86. Cohan P, Wang C, McArthur DL, et al. Acute secondary adrenal insufficiency
after traumatic brain injury: a prospective study. Crit Care Med
2005;33:2358–66.
87. Klose M, Brennum J, Poulsgaard L, et al. Hypopituitarism is uncommon
after aneurysmal subarachnoid haemorrhage. Clinical Endocrinology (Oxf)
2010;73(1):95–101.
88. Parenti G, Cecchi PC, Ragghianti B, et al. Evaluation of the anterior pituitary
function in the acute phase after spontaneous subarachnoid hemorrhage.
J Endocrinol Invest 2010;34(5):361–5.
89. Kreitschmann-Andermahr I, Poll E, Hutter BO, et al. Quality of life and
psychiatric sequelae following aneurysmal subarachnoid haemorrhage: does
neuroendocrine dysfunction play a role? Clin Endocrinol (Oxf) 2007;66:
833–7.
90. Vespa P. Endocrine function following acute SAH. Neurocrit Care 2011;15:
361–4.
91. Kelly DF, Gonzalo IT, Cohan P, et al. Hypopituitarism following traumatic
brain injury and aneurysmal subarachnoid hemorrhage: a preliminary
report. J Neurosurg 2000;93:743–52.
92. Savaridas T, Andrews PJ, Harris B. Cortisol dynamics following acute severe
brain injury. Intensive Care Med 2004;30:1479–83.
93. Bendel S, Koivisto T, Ruokonen E, et al. Pituitary-adrenal function in
patients with acute subarachnoid haemorrhage: a prospective cohort study.
Crit Care 2008;12:R126.
94. Weant KA, Sasaki-Adams D, Dziedzic K, et al. Acute relative adrenal
insufficiency after aneurysmal subarachnoid hemorrhage. Neurosurgery
2008;63(4):645–9; discussion 649–50.
95. Zetterling M, Engström BE, Hallberg L, et al. Cortisol and
adrenocorticotropic hormone dynamics in the acute phase of subarachnoid
haemorrhage. Br J Neurosurg 2011;25(6):684–92.
96. Weant KA, Sasaki-Adams D, Dziedzic K, et al. Acute relative adrenal
insufficiency after aneurysmal subarachnoid hemorrhage. Neurosurgery
2008;63:645–9; discussion 9–50.
97. Hasan D, Lindsay KW, Wijdicks EF, et al. Effect of fludrocortisone acetate in
patients with subarachnoid hemorrhage. Stroke 1989;20:1156–61.
98. Gomis P, Graftieaux JP, Sercombe R, et al. Randomized, double-blind,
placebo-controlled, pilot trial of high-dose methylprednisolone in
aneurysmal subarachnoid hemorrhage. J Neurosurg 2010;112:681–8.
99. Feigin VL, Anderson N, Rinkel GJ, et al. Corticosteroids for aneurysmal
subarachnoid haemorrhage and primary intracerebral haemorrhage.
Cochrane Database Syst Rev 2005;3:CD004583.
II
Chapter
22  
Renal, Electrolyte, and
Acid-Base Assessment
Guy M. Dugan
Introduction
Between 25% and 75% of critically ill patients are affected by
acute kidney injury (AKI), and AKI is an independent risk
factor for mortality in critically ill patients,1-4 including those
with neurologic disorders such as traumatic brain injury (TBI)
or subarachnoid hemorrhage (SAH).5 Even small changes in
serum creatinine (SC) are associated with an increased risk of
death.5 In addition, the incidence of end-stage renal disease
(ESRD) is increasing. These patients, who have significant
comorbidities, represent an important group of patients who
require critical care. This chapter reviews renal physiology
and AKI and its assessment, including consensus definitions
(RIFLE: risk, injury, failure, loss, end-stage disease) from the
Acute Kidney Network (AKIN). In addition, an approach to
acid-base disorders is described. Assessment of volume status
and sodium balance and its disorders are reviewed in Chapters
19 and 21, respectively.
Normal Renal Physiology
The kidney maintains the extracellular environment by excret-
ing the waste products of metabolism (urea, creatinine, and
uric acid) while simultaneously regulating the excretion of
water and solutes (Na+, K+, H+) by causing changes in tubular
reabsorption or secretion. The glomerular filtrate is more than
60 times the volume of plasma, thus almost all of this fluid
must be returned to the systemic circulation by tubular reab-
sorption. Conversely, solutes move from the peritubular capil-
lary through the cell and into the urine in a process called
tubular secretion. Na+, Cl−, and H2O are reabsorbed, H+ is
secreted, K+ and uric acid are both absorbed and secreted, and
filtered creatinine is excreted virtually unchanged.6
Hypovolemia is monitored by baroreceptors (located in the
carotid sinuses and afferent arterioles of the glomerulus), and
plasma volume is regulated by the resorption or excretion
of sodium and water. The baroreceptors signal to increase
intravascular volume by three mechanisms: (1) renin-
angiotensin-aldosterone system (RAAS); (2) sympathetic acti-
vation with the nerves releasing epinephrine, norepinephrine,
210
and dopamine; and (3) antidiuretic hormone (ADH). Tech-
niques to monitor volume status are discussed in Chapter 19.
Renin-Angiotensin-Aldosterone
System
With hypovolemia, the juxtaglomerular apparatus releases
renin that converts angiotensinogen from the liver to angio-
tensin I. Angiotensin I is converted to angiotensin II by
angiotensin-converting enzyme (ACE), primarily in the lung
(Fig. 22.1). Angiotensin II is also produced by the zona glo-
merulosa of the adrenal gland in response to hypovolemia or
hyperkalemia and has the following functions: (1) it increases
Na+ and water resorption in the kidney at the proximal con-
voluted tubule (PCT); (2) it increases the production and
release of aldosterone from the adrenal gland that in turn
causes an increase in Na+ resorption in the collecting tubule;
and (3) it is the most potent vasoconstrictor in the body and
increases blood pressure through its direct vasoconstrictive
effects.
Aldosterone acts on two types of cells in the collecting
tubule—the principal cell and the intercalated cell. At the
basolateral membrane of the principal cell, aldosterone
increases the Na+-K+-adenosine triphosphate (ATPase) pump
that creates a concentration gradient between the tubular
lumen and the principal cell (Fig. 22.2). This allows Na+ to
flow down the concentration gradient with the secretion of
K+.7 The potassium-sparing diuretics, amiloride and triam-
terene, act by directly closing Na+ channels, and spirono­
lactone competes with aldosterone. At the level of the
intercalated cell, aldosterone acts at the H+-ATPase pump to
increase the secretion of H+ into the tubule lumen (Fig. 22.3).
As H+ is secreted into the lumen, HCO3− is formed inside the
cell and then resorbed into the plasma. Thus an increase in
aldosterone can result in a metabolic alkalosis and decreased
activity results in type 4 renal tubular acidosis.
Sympathetic activity increases plasma volume by increas-
ing Na+ resorption in the PCT and by stimulating renin
release.
© Copyright 2013 Elsevier Inc. All rights reserved.
 Section II—Clinical and Laboratory Assessment 211

Alternative Third
pathways ventricle
Angiotensinogen 1
e.g. chymase 4

Hypophysiotropic 2
area
Angiotensin I Hypothalamus
Renin Mammillary
Bradykinin body
Negative ACE Median eminence O.C.
Inactive
feedback Portal capillaries Stalk
fragments
INCREASED BLOOD PRESSURE
Anterior
Angiotensin II Posterior lobe
Sodium lobe
retention
AT2 receptor
AVP Pituitary
Aldosterone Oxytocin
(storage) 3

AT1 receptors
in adrenal
gland AT1 receptors in blood LH ACTH Prolactin
vessels → vasoconstriction FSH
TSH GH
Fig. 22.1  Diagram showing the renin-angiotensin pathway.
Fig. 22.4  Illustration of the hypothalamic pituitary axis.

Peritubular
Tubular Na 3Na
capillary Antidiuretic Hormone-Arginine
lumen ATPase
Vasopressin
2K
The hypothalamus regulates osmolality (water content)
through the mechanisms of thirst and antidiuretic hormone
Aldo-R Aldo (ADH). ADH is the generic name given to what is now
K known as arginine vasopressin that is synthesized in the
supraoptic and paraventricular nuclei of the hypothalamus
ANP-R and stored in the posterior lobe of the pituitary (Fig. 22.4).
There are three stimuli for ADH secretion: (1) hyperosmolal-
ity—as little as a 1% increase in osmolality will stimulate
ADH; (2) low circulating volume more than a 10% decrease
Fig. 22.2  Diagram illustrating the effect of aldosterone at the basolateral in volume or more than 10 to 15 mm Hg decrease in blood
membrane of the principal cell where it increases the NA+−K+−ATPase
pressure; and (3) inappropriate release (i.e., syndrome of
pump.
inappropriate antidiuretic hormone [SIADH]) that is dis-
cussed further in Chapter 21.
Tubular Peritubular
lumen capillary Plasma Osmolality
Plasma osmolality refers to the concentration of all the solutes
(electrolytes and nonelectrolytes) in the plasma and is nor-
ATPase H+
mally between 285 and 295 mmol/L. The kidney exerts control
over osmolality at several locations. In the proximal convo-
H2O Cl– luted tubule, sodium and water are resorbed in an isotonic
manner, so the osmolality of the fluid leaving is isotonic. At
the loop of Henle (LOH), Na+-K+-2 Cl− pumps remove elec-
OH– + CO2 HCO3
CA trolytes from the tubular fluid and in the process generate
K+ both a dilute urine and a hypertonic medullary interstitium.
ATPase ADH inserts preformed water channels into the collecting
H+
tubule that allows water to be osmotically reabsorbed and
to concentrate the urine (Fig. 22.5). Osmolality is estimated
by calculating the concentration of three solutes:
Fig. 22.3  Diagram illustrating aldosterone effects at the level of the
intercalated cell, where it acts at the H+−ATPase pump to increase the Osmolality = 2 × Na+ (mmol/L) + Blood urea nitrogen
secretion of H+ into the tubule lumen. (BUN) (mg/dL)/2.8 plus (glucose mg/dL)/18.
212 Section II—Clinical and Laboratory Assessment

Osmolarity of
300 interstitial fluid
300 100 (mosm/L)
300
100
300 300
Cortex H2O NaCl H2O
400 200
H2O H2O
NaCl 400 400
Outer H2O
NaCl H2O
medulla
H2O NaCl H2O
600 400 600 600
H2O
H2O NaCl
Urea
H2O
Inner H2O NaCl Urea
medulla 900 700 900
H2O
H2O NaCl Urea
1200 1200
1200

Active
transport
Passive
transport

Fig. 22.5  Illustration of the Counter Current Mechanism in the kidney and how the kidney affects osmolality at several locations.

BUN and glucose are measured in mg/dL and must be
converted to mmol/L. To convert mg to mmol, divide the mg
by the MW (MW = mg/mmol). Thus the conversion of BUN:
BUN = 18 mg/dL
(18 mg/dL/(28 mg/mmol) × 10 dL/L = 6.4 mmol/L.
Similar osmolality calculations can be made for commonly
infused intravenous (IV) solutions. D5W has:
5 g/100 mL (or 5000 mg/dL/180 mg/mmol) × 10 dL/L
= 278 mmol/L.
Also:
0.9% normal saline (NS) = 0.9 g/dL = (900 mg/dL/58.5 mg/
mmol) × 10 dL/L = 154 mmol/L.
This is balanced by an equal number of Cl−, so the osmolal-
ity is 154 mmol/L + 154 mmol/L = 308 mmol/L, that is, iso-
tonic. The units of osmolality are expressed as either mOsm/
kg in conventional units or mmol/L in SI units.
Because Mg+, Ca+, bilirubin, and albumin are not included
in the calculation, there will always be a difference between
the measured and calculated osmolality. This osmolar gap is
less than 10 mmol/L, and a greater value indicates the accu-
mulation of osmoles that are either endogenous (lactic acid,
ketones) or exogenous (ethanol, methanol, ethylene glycol,
isopropyl alcohol). In clinical practice it can be useful to
know that ethanol has a molecular weight of 46, so a level of
230 mg/dL is 50 mmol/L (230 mg/dL/46 mg/mmol) ×10 dL/L
= 50 mmol/L).
Urine concentration is measured by osmolality and by spe-
cific gravity. Osmolality is the number of particles per liter and
is determined by freezing point depression. It is used in evalu-
ation of hyponatremia and the investigation of polydipsia and
polyuria. Specific gravity is the mass (in grams) of 1 mL of
solution. Thus if there are particularly heavy particles such as
glucose or radiocontrast dye, the specific gravity will overes-
timate the urine concentration compared with the osmolality.
There are several circumstances in which the measurement of
urine osmolality may be important8:
1. Distinguishing between prerenal AKI and acute tubu-
lar necrosis (ATN). A urine osmolality greater than
500  mosmol/kg suggests prerenal disease, in which
lower values are not helpful.
2. To distinguish between primary polydipsia and other
causes of hyponatremia. A urine osmolality less than
100 mosmol/kg indicates normal suppression of ADH and
therefore the presence of primary polydipsia. A higher
urine osmolality suggests the presence of at least some
ADH usually due to effective volume depletion or SIADH.
3. A decreasing urinary osmolality suggests that a spontane-
ous water diuresis has begun to correct hyponatremia. This
helps to avoid inadvertent overcorrection in hyponatremic
 Section II—Clinical and Laboratory Assessment 213
patients with reversible causes of water retention such as
volume depletion, pain, nausea, or drugs.
4. Urine osmolality helps distinguish between primary
polydipsia and diabetes insipidus (DI) as a cause of poly-
uria. After the plasma osmolality is increased by water
restriction to stimulate ADH release, the urine becomes
concentrated with primary polydipsia. The urinary
osmolality remains relatively dilute with both forms of
DI, though patients with central DI have some concen-
tration after administration of desmopressin (DAVP).
5. In patients who are hypernatremic, a urine osmolality less
than plasma osmolality suggests the presence of DI.
AKI in the Intensive Care Unit
Overview
Acute renal failure is common in the intensive care unit (ICU),
with a prevalence of up to 25% to 75%, and it is an independent
risk factor for poor outcome with mortality rates from 28% to
90%. These wide ranges reflect that it is a complex, poorly
understood process with more than 30 definitions in the litera-
ture. The following section provides a template for evaluation
and management. A comprehensive review of therapy, however,
is beyond the scope of this chapter, and the reader is referred
to recent reviews on this topic for further information.3,9-14
Epidemiology
Standardization of the definition of AKI has resulted in a
better understanding of its incidence. Population-based
studies suggest it is nearly as common as myocardial infarc-
tion.15,16 The causes of AKI are several. For example, Liano et
al. collected data on 748 cases of AKI in a prospective, multi-
center study from Madrid. The following causes were identi-
fied: ATN, 45%; prerenal, 21%; acute on chronic renal failure,
13%; urinary tract obstruction, 10%; glomerulonephritis or
vasculitis, 4%; acute interstitial nephritis, 2%; and atheroem-
boli, 1%.17 In the United States, epidemiologic data from the
Program to Improve Care in Acute Renal Disease (PICARD)
examined the etiology of AKI in 618 patients from five medical
centers. The most common causes were ischemic acute tubular
necrosis (including sepsis and hypotension), prerenal disease
(hypovolemia, hemorrhage), nephrotoxicity (radiocontrast
media [RCM], rhabdomyolysis), AKI with cardiac disease
(heart failure and shock), AKI with liver disease (hepatorenal
syndrome, cirrhosis), and multifactorial etiologies.18 Multi-
center epidemiologic studies suggest that overall sepsis is now
the most common cause of AKI in the ICU.1,3
Diagnosis
Conceptually AKI refers to the acute decline in glomerular and
tubular function that results in azotemia (the accumulation of
nitrogenous waste products), and the inability to maintain
fluid and electrolyte homeostasis. The diagnosis of renal
failure is based on a history and physical examination along
with general supporting chemistries, the urinalysis, assess-
ment of the glomerular filtration rate (GFR; primarily the
BUN and SC), and urine output. However, not all elevations
in BUN and creatinine indicate renal injury,19 and oliguria
(<400 mL/24 hr) is not a universal finding (Table 22.1).
Table 22.1  Causes of Nonrenal Elevations of
Blood Urea Nitrogen and Creatinine
Nonrenal causes of elevated urea
• Corticosteroids
• Hyperalimentation
• Gastrointestinal bleeding
• Tetracyclines
Nonrenal causes of elevated creatinine
• Rhabdomyolysis-increased creatinine release from skeletal
muscle
• Interference with creatinine assay
• Acetone
• Cefoxitin
• Flucytosine
• Methyldopa
Blocked tubular creatinine secretion
• Cimetidine
• Trimethoprim
RIFLE and AKIN Criteria
The understanding of AKI has been hindered by the numer-
ous definitions used in various studies, and an important
contribution to conceptualization was the development of a
consensus definition by the Acute Dialysis Quality Initiative
(ADQI) in which they introduced the RIFLE criteria.20 This
stratifies renal failure into:
Risk of renal dysfunction: a 1.5 times’ increase in SC;
urine output less than 0.5 mL/kg/hr times 6 hours
(GFR decrease >25%)
Injury to the kidneys: twofold increase in SC; urine
output less than 0.5 mL/kg/hr times 12 hours
(GFR decrease >50%)
Failure of kidney function: threefold increase in SC;
urine output less than 0.5 mL/kg/hr times 24 hours
(GFR decrease 75%)
Loss of kidney function: defined as the need for renal
replacement therapy (RRT) for more than 4 weeks
End-stage kidney disease: need for dialysis for longer than
3 months
A modification of the RIFLE criteria was subsequently pro-
posed by the AKIN, which included representatives from the
ADQI group and from other nephrology and intensive care
societies. Their proposal for kidney failure resulted in a three-
stage classification that corresponds to risk (stage 1), injury
(stage 2), and failure (stage 3) of the RIFLE criteria. Loss and
ESRD are removed form the staging system and defined as
outcomes.
Stage 1: Increase in SC of 0.3 mg/dL or an increase in 1.5
to 2 from baseline; urine output less than 0.5 mL/
kg/hr for more than 6 hours
Stage 2: Increase in SC more than two- to threefold from
baseline; urine output less than 0.5 mg/kg/hr for
more than 12 hours
Stage 3: Increase in SC more than threefold from baseline
or SC greater than or equal to 4 mg/dL with an
acute increase of at least 0.5 mg/dL; urine output
less than 0.5 mg/kg/hr for 24 hours or anuria for
12 hours
The addition of an absolute change in SC of more than or
equal to 0.3 mL/kg/hr is based on data that show an 80%
214 Section II—Clinical and Laboratory Assessment
increase in mortality associated with changes in SC concen-
tration of as little as 0.3 to 0.5 mL/kg/hr.21 The last two cri-
teria are identical to the RIFLE risk criteria. The criteria
should be applied after volume resuscitation and urinary
tract obstruction excluded if oliguria is the sole diagnostic
criterion. These criteria may have their greatest applicability
in helping to standardize definitions in clinical studies.
AKIN also recommended replacing the term acute renal
failure with acute kidney injury to represent the entire spec-
trum of acute renal failure.
However, these modifications may still be misleading in
assessing AKI. The SC does not accurately reflect the GFR in
a patient who is not in steady state. Thus a patient may be
developing AKI with a normal creatinine, and a rising creati-
nine may not imply ongoing kidney injury. Furthermore, cre-
atinine is removed by dialysis, so it is not possible to assess
kidney function by measuring it once dialysis is initiated.
Moreover, AKI may exist despite a normal urine output.
This has led to investigation into a number of biomarkers—
N-acetyl-glucosaminidase (NAG), neutrophil gelatinase–
associated lipocalin (NGAL), cystatin C, or kidney injury
molecule 1 (KIM-1) among others—that may better indicate
the state of kidney injury.3,22-26 Finally, RIFLE and AKIN cri-
teria rely in large part on SC and measurement of urine output
for the diagnosis of AKI. Changes in SC may be delayed and
in most ICUs urine output is measured manually, so these
measurements may be subject to human error. Newer tech-
niques that provide continuous minute-to-minute monitor-
ing of urine output may allow more accurate evaluation of
urine output and hence early identification of AKI.27
Classification of AKI
A standard strategy to classify AKI is to divide it into prerenal
disease, postrenal disease, and intrinsic disease based on the
primary site of injury—the tubules, glomerulus, interstitium,
or vessels. Prerenal disease may progress to tubular injury, and
ATN may result from ischemic or toxic causes. These catego-
ries may have some clinical utility in organizing thinking, but
they imply an insight into pathophysiology (prerenal) or his-
tology (ATN) that is either greatly simplified or misleading.28
A complementary approach is to examine the injury in differ-
ent disease states. An integrative approach using both of these
strategies is used in this outline.
Prerenal AKI and ATN
Prerenal AKI may result from intravascular volume depletion,
an alteration of glomerular hemodynamics, or vascular com-
promise (Table 22.2). Prerenal AKI may evolve into an entity
traditionally referred to as ATN; the distinctions based on
urine chemistries are outlined in Table 22.3. Physiologically,
the reduction in blood pressure activates cardiac and arterial
receptors that increases sympathetic neural tone (improving
blood pressure but causing renal vasoconstriction), and
releasing both renin, which leads to the production of angio-
tensin II, and ADH. Angiotensin II and adrenergic activation
stimulate the proximal reabsorption of sodium, and aldoste-
rone increases the sodium reabsorption in the distal tubule.
This results in a urinary sodium concentration of less than
20 mEq/L. There is enhanced reabsorption of urea from the
medullary collecting ducts, resulting in a disproportionate
Table 22.2  Causes of Prerenal Acute
Kidney Injury
Intravascular volume depletion
Hemorrhage, GI volume loss (diarrhea, vomiting), skin losses
(sweat, burns), renal losses (diuretics, osmotic diuresis),
third space fluid; e.g., pancreatitis, bowel surgery, crush
injury
Decreased “effective” circulating volume
CHF, nephritic syndrome, cirrhosis, sepsis
Renal vasoconstriction
Liver failure, hypercalcemia, ACE inhibitors, ARB, NSAIDs
Increased renal vein pressure/IVC obstruction
Abdominal compartment syndrome
Hypotension from shock—cardiogenic, septic
Selective ischemia—bilateral renal artery stenosis, often in the
context of ACEI or ARB
ACE, Angiotensin-converting enzyme; ACEI, angiotensin-converting enzyme
inhibitor; ARB, angiotensin receptor blocker; CHF, congestive heart failure;
GI, gastrointestinal; IVC, inferior vena cava; NSAIDs, nonsteroidal
anti-inflammatory drugs.
Table 22.3  Laboratory Findings That Help
Differentiate Prerenal AKI and ATN
Laboratory Prerenal AKI ATN
Urine specific gravity >1.020 ≈1.012
Urine sodium, mEq/L <20 >30
BUN/SC ratio 20 10
Urine osmolality >350 ≈300
Urine/plasma osmolality >1.5 1.0
FENa <1% >1%
FE urea <35% >50%
AKI, Acute kidney injury; ATN, acute tubular necrosis; BUN, blood urea
nitrogen; FE, fractional excretion; FENa, fractional excretion of sodium;
SC, serum creatinine
increase of BUN compared with creatinine with a BUN-
to-SC ratio of greater than 10 : 1 to 15 : 1. As tubular function
is preserved, there is a high urine osmolality, greater than
350 mOsm/L.
Of the tests available (see Table 22.3), the fractional excre-
tion of sodium (FENa) may be the preferred screening test to
differentiate between prerenal AKI and ATN (Table 22.4).
FENa = (urine Na × serum creatinine)/
(serum Na × urine creatinine) ×100
Classically, a FENa less than 1% is characteristic of prerenal
AKI, whereas a FENa greater than 1% to 2% signifies ATN.
However, a low FENa is not unique to prerenal disease. It can
occur in disorders associated with normal tubular function
but a low GFR such as acute glomerulonephritis (GN), vascu-
litis, and acute urinary tract obstruction. It also can be seen
when ATN is superimposed on a chronic sodium-retaining
state as may occur with aminoglycoside therapy, congestive
heart failure (CHF), or cirrhosis.29 The confounding effects of
diuretics may be obviated by the use of the fractional excretion
of urea, which when less than 35% accurately indicates prer-
enal azotemia.30
 Section II—Clinical and Laboratory Assessment 215
Table 22.4  Interpretation of the Fractional
Excretion of Sodium
FENa <1%
Prerenal AKI
Low GFR with normal tubular function
Acute glomerulonephritis
Vasculitis
Acute urinary tract obstruction
ATN
Chronic disease with sodium-retaining states
Aminoglycosides
Cirrhosis
CHF
Rhabdomyolysis
Contrast nephropathy
FENa >1%-2%
ATN
Prerenal AKI with:
Glucosuria
Diuretics
Acute on chronic disease
Mannitol
AKI, Acute kidney injury; ATN, acute tubular necrosis; CHF, congestive heart
failure; FENa, fractional excretion of sodium; GFR, glomerular filtration rate.
Fig. 22.6  Urine sediment on urinalysis showing multiple muddy brown
granular casts that suggest acute tubular necrosis. (Courtesy Harvard
Medical School.)
The urinalysis in prerenal disease is normal (this may
include granular casts), whereas ATN has muddy brown casts31
(Fig. 22.6). Pathologically, tubular necrosis is usually not seen
with ATN, and there is often limited histologic evidence of
injury despite marked functional impairment.32 ATN may be
induced by those factors that cause prerenal AKI as well as by
drugs, rhabdomyolysis, tumor lysis, and vascular insults.
Diuretics may convert ATN from an oliguric to a nonoliguric
state and may be tried for volume control, but they have no
effect on renal recovery or survival.33,34 Loop diuretics work by
inhibition of a Na+−K+−2Cl− transporter in the loop of Henle.
A secretory isoform of this transporter is present in the inner
ear endolymph, and thus high-dose diuretics may lead to per-
manent hearing loss.35 Dopamine does not provide protection
in early ATN and may cause harm by reducing renal blood
flow in postischemic ATN.36-38 Prevention of ATN by optimiz-
ing volume status and avoiding nephrotoxic agents remain the
mainstay of management.
Fig. 22.7  Urine sediment showing free red cells and a red cell cast
that is tightly packed with red cells. Red cell casts are virtually
diagnostic of glomerulonephritis or vasculitis. (Courtesy Harvard Medical
School.)
Glomerular Disease in AKI
Acute glomerulonephritis may be the cause of AKI in the
ICU due to bacterial endocarditis, staphylococcal sepsis,
visceral abscesses, hepatitis B, systemic lupus erythematosus
(SLE), Goodpasture syndrome, or rapidly progressive glo-
merulonephritis (RPGN). Red cells and red blood cell casts,
and proteinuria are seen on urinalysis31 (Fig. 22.7). An alter-
ation of glomerular hemodynamics also may contribute to
AKI. Afferent vasoconstriction may be seen in hepatorenal
syndrome and efferent vasodilation from ACE inhibitors.
Vasodilators such as nitroprusside and nifedipine may con-
tribute to AKI by altering intrarenal hemodynamics without
systemic hypotension.39
Interstitial Disease in AKI
Drugs most often induce acute interstitial nephritis, but infec-
tions such as from Legionella, Leptospira, and streptococcal
organisms, and autoimmune disorders may also be responsi-
ble. The most commonly implicated drugs are beta-lactam
antibiotics, sulfonamides, rifampin, and nonsteroidal anti-
inflammatory drugs (NSAIDs). The complete clinical spec-
trum presents as fever, rash, arthralgias, and eosinophilia, and
a urine analysis shows sterile pyuria with white blood cell
casts, and eosinophiluria (except NSAIDs). Hematuria and
proteinuria also are common. The FENa usually is greater
than 1%, indicating tubular damage, but lower values may be
seen in mild disease.40
AKI in Cirrhosis
Patients with cirrhosis or acute hepatitis can develop renal
failure without proteinuria and with a reduced urinary
sodium. The hallmark is intense renal vasoconstriction with
peripheral vasodilation. The kidneys in patients with hepa-
torenal syndrome (HRS) are normal on autopsy and func-
tioned normally when transplanted into patients without
cirrhosis. Moreover, HRS can be reversed following liver
transplantation.41-43 In 1996 the International Ascites Club
published a consensus paper that subdivided HRS into two
216 Section II—Clinical and Laboratory Assessment

types. Type 1 HRS is characterized by a rapid decline in renal

function with a doubling of SC to more than 2.5 mg/dL or
having a creatinine clearance to less than 20 mL/min within
2 weeks. In type 2 HRS the presentation is of stable renal
failure in a patient with refractory ascites in which the SC
increases to more than 1.5 mg/dL or a creatinine clearance of
less than 40 mL/min.44 Untreated type 1 has mortality as high
as 80% in 2 weeks, whereas type 2 has a median survival of
approximately 6 months.
The precipitating factors for HRS have been identified
as bacterial infection (48%), gastrointestinal (GI) bleeding
(33%), aggressive paracentesis (27%), and drugs (diuretics,
aminoglycosides, nonsteroidal anti-inflammatory drugs
[NSAIDs], and angiotensin-converting enzyme inhibitors
[ACEIs]/angiotensin II receptor blockers [ARBs]). However,
24% develop type 1 HRS without an obvious precipitating
factor.45 A high index of suspicion is needed to assess AKI in
the context of advanced liver disease because reduction in
muscle may render a SC within the normal range even in the
context of a decreased GFR, and GI bleeding and the amount
of protein in the diet may affect the BUN. Sepsis should be
suspected in any cirrhotic patient with AKI even in the absence
of leukocytosis and fever. Clinically the syndrome presents
with prerenal physiology (i.e., oliguria, a benign urine sedi-
ment, and low urine sodium).
Treatment may include midodrine (an alpha agonist) and
octreotide (a long-acting form of somatostatin that inhibits
endogenous vasodilator release), with albumin, and terlipres-
Fig. 22.8  Cutaneous manifestations of cholesterol emboli syndrome.
sin (a vasopressin analog used in Europe).46,47 Dialysis should
be offered only if there is a chance for liver transplantation in
the short term. Transjugular intrahepatic portosystemic shunt
(TIPS) involves placing a stent between the hepatic vein and 22.8). Anticoagulation is dangerous in patients with CES but
the intrahepatic portion of the portal vein. It is primarily may be necessary in patients with a thrombotic event.50
designed to treat bleeding varices, but it has been used for
HRS, though it may worsen encephalopathy. It may serve as a
bridge to liver transplantation, which is the only effective Abdominal Compartment Syndrome
treatment.48,49 Abdominal compartment syndrome (ACS) refers to symp-
tomatic organ dysfunction arising from an increase in intra-
abdominal pressure (IAP; See also Chapter 23). A pressure
Vascular Diseases Causing AKI of 20 with organ dysfunction is the operational definition
A number of vascular disorders of both the large and small used in studies, but in clinical practice a single IAP cannot
vessels may give a prerenal picture of AKI. Large vessel disease predictably diagnose ACS in all patients. ACS arises from
may arise from operative arterial cross-clamping, renal artery tissue edema or third-spaced fluid and may occur in diverse
thrombosis or embolism, or renal artery stenosis. Renal infarc- clinical states such as abdominal surgery, pancreatitis, and
tion may present with fever, hematuria, acute flank pain, ileus, bleeding. Hypotension may arise from the ensuing inferior
and leukocytosis that may mimic an acute abdomen. Hyper- vena cava (IVC) obstruction, and be accompanied by a high
tensive crises may disrupt the vascular endothelium and be pulmonary artery occlusion pressure or elevated femoral vein
associated with thrombocytopenia and microangiopathy, reti- pressure despite a reduced venous return. Clinically the kidney
nopathy, and AKI. Systemic vasculitis such as Wegener’s gran- dysfunction looks similar to prerenal kidney disease with a low
ulomatosis, polyarteritis nodosa, hypersensitivity vasculitis, urinary sodium, and an increase in plasma renin, aldosterone,
and Henoch-Schönlein purpura often cause AKI. and ADH level. Oliguria may develop at an IAP of 15 mm Hg
Cholesterol emboli syndrome (CES) arises following aortic and anuria at 30 mm Hg.51 A high index of suspicion is needed
manipulation from percutaneous intervention, surgery, or because this condition may look like evolution of the underly-
systemic anticoagulation. The crystals incite an inflammatory ing disease process with hypotension, acute respiratory dis-
reaction and adventitial fibrosis may obliterate the vessel tress syndrome (ARDS), and AKI from multiorgan failure.52
lumen. Hypocomplementemia and eosinophilia are seen, and
any organ may be affected including the kidney, the GI tract
with bleeding from microinfarcts, the skin, and the central Drugs and AKI
nervous system (CNS) with stroke. The cutaneous manifesta- Many medications can induce kidney failure and may lead to
tions may include a blue toe or livedo reticularis that results a picture of prerenal AKI, ATN, or acute interstitial nephritis.
from a superficial infarct and compensatory dilation of the A complete discussion is beyond the scope of this chapter, but
surrounding vessels to give a livedo or “lacelike” pattern (Fig. some common drugs are discussed.
 Section II—Clinical and Laboratory Assessment 217
NSAIDs and Acetaminophen-Induced AKI
Prostacyclin and prostaglandin E2 preserve renal blood flow
and glomerular filtration and inhibition of prostaglandin syn-
thesis with a NSAID can lead to a reversible AKI with an SC
increase within the first 3 to 7 days.53,54 This hemodynamically
mediated injury may be less with aspirin because there is only
partial inhibition of glomerular cyclooxygenase.55 Acute inter-
stitial nephritis and the nephritic syndrome also may occur
with NSAIDs and it presents with hematuria, pyuria, white
cell casts,31 and proteinuria along with an acute rise in SC (Fig.
22.9). The complete syndrome of fever, rash, eosinophilia, and
eosinophiluria may be absent. Spontaneous recovery usually
occurs within weeks to months after discontinuation of
therapy.56 Membranous nephropathy with proteinuria has
been reported to occur with diclofenac, but probably any
NSAID may be involved.57
Acetaminophen-induced acute renal failure may occur with
an overdose, but some alcoholics are at increased risk from
therapeutic doses. The pathologic changes reveal ATN, and the
uric acid shows granular and epithelial cell casts31 (Fig. 22.10).
Fig. 22.9  White cell cast in which blue-stained white cells (arrow) are
contained within a granular cast. (Courtesy Frances Andrus, BA, Victoria
Hospital, London, Ontario, Canada.)
Fig. 22.10  Epithelial cell cast with free epithelial cells (arrow) in
the urine sediment. Renal tubular epithelial cells are larger than white
cells and have a single, large central nucleus. (Courtesy Frances Andrus, BA,
Victoria Hospital, London, Ontario, Canada.)
Recovery to baseline generally occurs spontaneously within
1 to 4 weeks, and acetylcysteine has no protective effect on the
kidney as it does with the liver.58 Drugs that increase cyto-
chrome P450 activity such as phenytoin, phenobarbital,
rifampin, and carbamazepine enhance the risk of acetamino-
phen nephrotoxicity.59
Radiocontrast Media–Induced AKI
Radiocontrast media (RCM) are either ionic or nonionic and
are of variable osmolality. Risk factors for RCM–induced AKI
include underlying renal insufficiency (SC >1.5 mg/dL), dia-
betic nephropathy, heart failure, volume depletion, PCI (pro-
motes the development of atheroemboli), and the dose and
tonicity of the RCM agent. Renal injury is observed within
the first 12 to 24 hours with recovery typically beginning
within 3 to 5 days.60,61 Patients with diabetes and heart failure
are at increased risk of AKI from RCM, possibly due to
impaired nitric oxide (NO) generation.62 Many of the com-
monly used RCM agents induce a false-positive result when
either a dipstick or sulfosalicylic acid is used to detect pro-
teinuria, and this should not be tested for at least 24 hours
after a contrast study.63 ATN may result; however, the FENa is
often less than 1%. Prevention is best accomplished with
hydration, often with sodium bicarbonate. N-acetylcysteine
may also have protective benefit.64
Aminoglycosides
These are concentrated in the proximal tubular cells, and AKI
can become evident several days after the drug has been
stopped. Distal tubular damage also occurs with a loss in
renal concentrating ability resulting in polyuria and hypo-
magnesemia. Risk factors include volume contraction, age,
hypokalemia, and a short dosing interval. The routine moni-
toring of peak and tough levels does not decrease the likeli-
hood of nephrotoxicity.
Lithium
Lithium may cause a nephrogenic DI by causing ADH resis-
tance in up to 20% of patients. Some data suggest lithium
levels may increase with concomitant ACE inhibitors.65,66
ACE Inhibitors
The primary side effects occur by one of two mechanisms. A
reduction in angiotensin II formation may cause hypoten-
sion, AKI, hyperkalemia, and problems during pregnancy.
Cough, edema, and anaphylactoid reactions are related to
increased kinins because ACE is also a kininase. The side
effects related to reduced angiotensin II, but not those related
to kinins, are also seen with the angiotensin receptor block-
ers. AKI is seen predominantly in patients with bilateral renal
artery stenosis, hypertensive nephrosclerosis, CHF, polycystic
kidney disease, or chronic renal failure. In each of these the
GFR is partially maintained by efferent arteriolar vasocon-
striction mediated by angiotensin II.67 The rise in
SC usually occurs in 3 to 5 days. Hyperkalemia may result
from ACEI/ARBs because angiotensin II increases aldoste-
rone that is the major hormonal stimulus to urinary potas-
sium excretion.
218 Section II—Clinical and Laboratory Assessment
Fig. 22.11  Skin involvement is characteristic with scleromyxedema-like fibrosis in nephrogenic systemic fibrosis (NSF) that can complicate Gadolinium
administration.
Fig. 22.12  Uric acid crystals seen on the urinalysis. These crystals
are pleomorphic, most often appearing as rhombic plates or rosettes.
They are yellow or reddish-brown and form only in an acid urine. (Courtesy
Harvard Medical School.)
Amphotericin B
Amphotericin B results in renal insufficiency with K+ and
Mg++ wasting and metabolic acidosis due to a type 1 or distal
RTA and polyuria due to a nephrogenic DI.
Gadolinium
Gadolinium may be contraindicated in patients with a GFR
less than 30 mL/min or any degree of renal insufficiency in
patients with hepatorenal syndrome because its use is associ-
ated with nephrogenic systemic fibrosis (NSF). The latency
of NSF is usually 2 to 4 weeks, but may be between 2 days
and 18 months. Skin involvement is characteristic with
scleromyxedema-like fibrosis that may lead to disabling joint
contractures (Fig. 22.11). Systemic involvement may include
the lungs, pleura, myocardium, pericardium, and dura. Scleral
plaques are common.68
Acute Tumor Lysis Syndrome, Acute Uric
Acid Nephropathy, and AKI
Acute tumor lysis syndrome consists of hyperphosphatemia
(>8 mg/dL), hypocalcemia, hyperkalemia, and hyperuricemia
Table 22.5  Causes of Rhabdomyolysis
• Trauma and compartment syndromes
• Seizures, dystonic reactions
• Impaired energy supply—carbon monoxide poisoning,
mitochondrial myopathies
• Malignant hyperthermia and neuroleptic malignant
syndrome
• Near-drowning with hypothermia
• Drugs
Anticholinergics due to the loss of impaired sweating
Statins, colchicine—direct myotoxins
Macrolides, cyclosporine, gemfibrozil, and protease
inhibitors interfere with statin clearance and potentiate
toxicity
• Cocaine-induced hyperthermia
• Carbon monoxide—insufficient energy production
• Infections—ehrlichiosis, falciparum malaria
• Electrolyte disorders—hypokalemia, hypophosphatemia (in
alcoholics receiving total parenteral nutrition without
phosphate supplementation)
(>15 mg/dL) in patients with tumors associated with high cell
turnover or with chemotherapy (e.g., after treatment of leu-
kemias and lymphoma, or other malignancies including
medulloblastomas). AKI results from tubular obstruction by
precipitated uric acid crystals or depiction of calcium phos-
phate complexes in the tubular lumen and interstitium.69
Patients have decreased urine output with uric acid crystals
seen on the urinalysis31 (Fig. 22.12). Treatment includes ade-
quate hydration and allopurinol. The role of urinary alkalini-
zation is controversial.70
Pigment-Induced AKI
Myoglobin from rhabdomyolysis and hemoglobin from intra-
vascular hemolysis are heme pigments that may induce AKI.
Hemolysis leading to AKI is uncommon and usually arises
from a transfusion reaction due to ABO incompatible blood.
Rhabdomyolysis is more common and may arise from a
variety of causes (Table 22.5). Patients with rhabdomyolysis
present with the triad of pigmented granular casts, elevated
creatine kinase (CK), and myoglobinuria—a red to brown
urine sediment that tests positive for heme but without red
blood cells. Myoglobin is rapidly cleared, so one may see
 Section II—Clinical and Laboratory Assessment 219
elevated CK levels without myoglobinuria. The plasma is
normal color unless its excretion is impaired from renal insuf-
ficiency. Patients with hemolysis have a similar finding in their
urine but have a reduced haptoglobin and abnormal periph-
eral smear with red plasma.71
The renal injury results from obstruction of the tubules
from heme pigment casts, tubular injury from free chelatable
iron, and volume depletion. In contrast to other forms of
ATN, the FENa is often less than 1%. In rhabdomyolysis,
abnormalities in serum electrolytes are common and include
metabolic acidosis, hyperuricemia, hyperkalemia, and hyper-
phosphatemia. Initial hypocalcemia, caused from the deposi-
tion of calcium in damaged muscle and decreased bone
responsiveness to parathyroid hormone (PTH), may subse-
quently be followed by hypercalcemia during recovery when
this calcium is released and secondary hyperparathyroidism
from AKI.72
Treatment is with saline, potentially up to 1 to 2 L/hr to
maintain a urine output of 200 to 300 mL/hr. After diuresis is
established, urinary alkalinization with 75 mmol of sodium
bicarbonate in 1 L 0.45 NS with a goal of a urine pH above
6.5 can be initiated. This may mitigate some of the deleteri-
ous effect of myoglobin (or hemoglobin), but it may induce
hypocalcemia by precipitating calcium phosphate with
ensuing tetany, seizures, or arrhythmias. Close monitoring of
serum bicarbonate, calcium, potassium, and urine pH is
required. A loop diuretic or mannitol may be added if diure-
sis is not established after volume repletion. Mannitol may be
beneficial with a CK greater than 20,000 to 30,000 U/L, but it
is recommended that the plasma osmolality and osmolal gap
be measured every 4 to 6 hours. Therapy is continued until
the urine discoloration clears and the plasma CK is less than
5000 to 10,000 U/L or if the urine pH does not increase
above 6.5 after 4 hours or the osmolal gap rises above
55 mosmol/kg.73
AKI in Pregnancy
AKI may occur in association with microangiopathic hemo-
lytic anemia and thrombocytopenia in late pregnancy. Two
entities of principal concern are thrombotic thrombocytope-
nic purpura and hemolytic uremic syndrome (TTP-HUS)
and severe preeclampsia, usually with the HELLP syndrome
(hemolysis, elevated liver enzymes, low platelets). TTP-HUS
probably exists as a continuum, but TTP is usually seen ante-
partum and is diagnosed when neurologic abnormalities pre-
dominate, whereas HUS is characterized by more severe renal
failure and is seen postpartum. These distinctions are fre-
quently unclear and may not be important for manage-
ment.74 The therapy of TTP-HUS in pregnancy is plasma
infusion with or without plasma exchange as in the nonpreg-
nant patient. Delivery is indicated for the HELLP variant of
preeclampsia.
Acute fatty liver of pregnancy results from fatty infiltration
of hepatocytes with inflammation and is associated with AKI
in 60% of cases. Patients have elevated liver function tests
(LFTs) with disseminated intravascular coagulation (DIC)
and delivery is indicated. Bilateral renal cortical necrosis is a
complication of pregnancy caused by abruptio placentae, pla-
centa previa, intrauterine fetal death, or amniotic fluid embo-
lism. Patients present with a triad of anuria, gross hematuria,
and flank pain. There is no specific therapy.
Table 22.6  Causes of Postrenal Failure
Bladder outlet obstruction
Prostate enlargement
Pelvic pathology
Crystalluria
Acyclovir
Indinavir
Uric acid
Papillary tip necrosis
Diabetes mellitus with pyelonephritis
Analgesic abuse
Sickle cell disease
Postrenal AKI
Although an uncommon cause of AKI in the ICU, postrenal
obstruction represents a reversible form of renal injury.
Imaging of the renal system is required for proper manage-
ment of AKI as well as measuring postvoid residuals (>50 mL
is abnormal). As intratubular pressure increases, glomerular
filtration pressure decreases. Postrenal AKI can be divided into
renal and extrarenal causes (Table 22.6). Intrarenal causes
include crystal deposition that may occur with ethylene glycol,
uric acid nephropathy that may occur in tumor lysis syn-
drome, or tubular obstruction from light chain casts that may
occur with multiple myeloma. Extrarenal causes may include
prostate disease, pelvic malignancy, and retroperitoneal disor-
ders.75 It is important to recognize that patients with a partial
obstruction may have a normal or increased urine output due
to a secondary concentrating defect.
The commonly discussed “postobstructive” diuresis is
usually a physiologic response to excrete the fluid retained
during the obstruction, and assiduous replacement of this can
lead to volume overload. There may be the need for some
fluid above baseline due to a mild sodium-wasting tendency
and a mild concentrating defect due to the down-regulation
of water channels. Half normal saline at 75 mL/hr, with atten-
tion to volume status, is suggested as an appropriate initial
therapy.76
Renal Replacement Therapy in AKI
The acute indications for RRT include volume overload, life-
threatening metabolic abnormalities such as hyperkalemia
and acidosis, acute pericarditis, and encephalopathy. A detailed
description of RRT is beyond the scope of this chapter; the
reader is referred to recent reviews for further informa-
tion.10,12,77-80 The objectives of RRT are to remove fluid by
ultrafiltration and control azotemia by removing excess solute.
Although peritoneal dialysis is still used, RRT is almost always
accomplished by intravascular access in the ICU. Intermittent
hemodialysis (IHD) is the most frequently used form of RRT
in the ICU and is accomplished by a central venous double-
lumen catheter that passes the blood through a dialysis mem-
brane. Studies have demonstrated improved clinical outcomes
with the use of synthetic biocompatible membranes (polysul-
fone, polyamide, polyacrylonitrile, and polymethyl methacry-
late) compared to cellulose-based membranes (cellulose
acetate, cuprophane.) Hypoxemia with a PaO2 decrease of
10 torr may occur, but the major complication of IHD is
hypotension that may prompt consideration for different
continuous renal replacement therapy (CRRT) strategies.
220 Section II—Clinical and Laboratory Assessment

However, there is no survival advantage with CRRT compared

with IHD, and the RRT choice may depend on a variety of Table 22.7  Antibiotic Dosing
factors. When rapid solute control is necessary such as hyper- Recommendations in CRRT85
kalemia, IHD is most suitable. Slow continuous ultrafiltration Drug CVVH CVVHD (HDF)
(SCUF) may be preferred if volume overload without
Amikacin 7.5 mg/kg q24h Same
azotemia is present. Continuous veno-venous hemofiltration
(CVVH) is indicated with hepatic failure, head trauma, or Amphotericin
septic shock.19   Deoxycholate 0.4-1 mg/kg q24h Same
  Lipid complex 3-5 mg/kg q24h Same
Neurologic complications can be common in patients   Liposomal 3-5 mg/kg q24h Same
undergoing hemodialysis. The dialysis disequilibrium syn-
Acyclovir 5-7.5 mg/kg q24h Same
drome refers to headache, nausea, delirium, and asterixis that
may progress to seizures, coma, or death. It is thought to result Ampicillin-sulbactam 3 g q12h 3 g q8h
from cerebral edema, and risk factors include older age, severe Aztreonam 1-2 g q12h 2 g q12h
acidosis, and new hemodialysis patients with markedly ele- Cefazolin 1-2 g q12h 2 g q12h
vated BUN. Milder symptoms such as muscle cramps and
anorexia are part of this syndrome. Erythropoietin may cause Cefepime 1-2 g q12h 2 g q12r
a rapid increase in blood pressure and result in encephalopa- Cefotaxime 1-2 g q12h 2 g q12h
thy and seizures. Other medications may precipitate seizures Ceftazidime 1-2 g q12 h 2 g q12h
in the hemodialysis population including beta-lactam antibi-
Ceftriaxone 2 g q12-24h Same
otics, meperidine, metoclopramide, and acyclovir. Hypoten-
sion or a reactive hypertension from a reactive vasoconstriction Cefuroxime 1.5 g q8-12h 1.5 g q8h
from a reactive hypotension may precipitate seizures. Dialysis Clindamycin 600-900 mg q8h Same
patients are at increased risk for intracerebral hemorrhage Ciprofloxacin 200 mg q12h 200-400 mg q12h
(ICH) and SAH, and subdural hematoma and ischemic
Colistin 2.5 mg/kg q48h Same
strokes.81 CRRT may affect the dosing of antibiotics, and Table
22.7 reflects the changes that may be needed. Daptomycin 4 or 6 mg/kg q48h Same
Doxycycline 100-200 mg q12h Same

Prognosis in AKI Fluconazole 200-400 mg q24h 400-800 mg q24h

Nonoliguric AKI, as well as AKI that is associated with ami- Imipenem-cilastatin 250 mg q6h or 250 mg q6h or
500 mg q8h 500 mg q8h
noglycosides and RCM, may have a mortality of 25% to 30%. 500 mg q6h
AKI occurring in the context of septic shock and burns has a
Itraconazole No change No change
mortality rate of 70% to 90%, usually as a result of infection.50
If the patient survives, oliguria averages 10 to 14 days. Over Levofloxacin 250 mg q24h Same
the next 3 to 7 days is an increase in urinary volume, but a Linezolid 600 mg q12h Same
sustained rise in the BUN and creatinine. Fluid administration Meropenem 1 g q12h Same
is often needed, and dialysis can usually be discontinued. Most
survivors regain function within 30 days, although rarely Metronidazole 500 mg q6-8h Same
recovery may extend to 90 days.19 Moxifloxacin 400 mg q24h Same
Nafcillin or oxacillin 2 g q4-6h Same
Acid-Base Balance in Critical Care Piperacillin- 2.25 g q6h 2.25-3.375 g q6h
tazobactam
A number of rules should be learned, but an important prin-
Rifampin 300-600 mg q12h Same
ciple is that all acid-base assessment must be accompanied
by a clinical history because this will help elucidate which Ticarcillin-clavulanate 2 g q6-8h 3.1 g q6h
of several possible patterns reflect clinical reality.7,82,83 An Tobramycin and Load 2 mg/kg then
approach to analysis of acid-base disorders is provided in gentamicin 1.5-2 mg/kg q24h
Table 22.8. Also, because an arterial blood gas (ABG) is and check levels
obtained in the evaluation of these disorders, a brief statement Vancomycin 1 g q48h (check 1 g q24h (check
is made about normal gas exchange. Monitoring ventilation levels) levels)
and pulmonary function is discussed in Chapter 20. Voriconazole 4 mg/kg PO q12h Same

CRRT, Continuous renal replacement therapy; CVVH, continuous veno-venous

Normal Gas Exchange hemofiltration; CVVHD, continuous veno-venous hemodialysis.

Respiration is a complex process involving integration between

the pons, medulla, spinal cord, carotid body, and muscles.
Increases in PaCO2 are sensed in the medulla, causing stimula-
tion of respiration. Hypoxemia is detected in the carotid body
(located at the bifurcation of the internal and external carotid
artery). In chronic respiratory acidosis the pH is nearly
normal, despite a high PaCO2, and the medulla is no longer
sensitive to elevations in PaCO2. This makes hypoxemia the
primary stimulus for respiration. Delivering high concentra-
tions of oxygen to these patients may suppress the hypoxic
respiratory drive. The alveolar-arterial gradient (A-a gradient)
is important in evaluating disorders of respiration (Table
22.9). The A-a gradient tells us that the presence of carbon
dioxide (CO2) in the alveoli reduces the alveolar oxygen
content. Traditionally, when a clinician is confronted with
 Section II—Clinical and Laboratory Assessment 221

Table 22.8  An Approach to Acid-Base Disorders (59)

STAGE 1: IDENTIFY THE PRIMARY ACID-BASE DISORDER
Rule 1
Examine the pHa and PaCO2. If either is abnormal, an acid-base abnormality is present.
Rule 2
If they are both abnormal, a primary disorder is present. If they are abnormal in the same direction, the primary disorder is
metabolic. If they are abnormal in opposite directions, the primary disorder is respiratory.
Example: 7.23/23 Both are decreased, so this is a primary metabolic acidosis.
Rule 3
If only one is abnormal, a mixed disorder is present.
Example: 7.37/55 Because the PaCO2 is elevated and there is a normal pH, a mixed disorder is present: a respiratory acidosis with a
metabolic alkalosis. Compensatory processes do not normalize the pH.
STAGE 2: EVALUATE THE COMPENSATORY RESPONSE
If only one of either the PaCO2 or pH is abnormal, a mixed disorder is present; skip to stage 3 for further analysis. When the pH
and PaCO2 are both abnormal and a primary disorder is present, evaluate the degree of compensation to determine if another
process is present.
Rule 4
If the pH and PaCO2 change in the same direction, a metabolic process is present, and the HCO3 is used to determine the
appropriate PaCO2.
Compensation for metabolic acidosis:  PaCO2 = (1.5 × HCO3) + 8 ± 2
Compensation for metabolic alkalosis  PaCO2 = (0.7 × HCO3) + 20 ± 5
If the PaCO2 is greater than expected, a concomitant respiratory acidosis is present, whereas if it is lower than predicted, a
respiratory alkalosis exists.
Example: pH = 7.23, PaCO2 = 23, HCO3 = 15 mEq/L
When the pH and PaCO2 are both abnormal in the same lower direction, this indicates a primary metabolic acidosis. The expected
PaCO2 would be 30.5 ± 2. Because the PaCO2 is lower than this expected value, a superimposed respiratory alkalosis is present.
Rule 5
If the pH and PaCO2 change in opposite directions, a respiratory disorder is present. Using the history and the equations for the
expected pH and HCO3, determine if the condition is acute, partially compensated, or fully compensated.
Example: pH = 7.54, PaCO2 = 23
If the history suggests an acute process, the equation for the expected change in pH is:
Expected pH = 7.40 + (0.008 × [ΔPaCO2])
The calculated value, 7.54, is the same as the measured value, so this represents an acute respiratory alkalosis. If the pH were
higher, a metabolic alkalosis would also be present. If the pH were lower, a metabolic acidosis would be present.
STAGE 3: CHECK THE ANION GAP AND GAP–GAP TO EVALUATE METABOLIC ACIDOSIS
An anion gap (AG) acidosis may be accompanied by a normal AG acidosis or a metabolic alkalosis.
This is called the gap–gap and is detected by examining the ratio:
AG excess/HCO3 deficit = (AG-12)/(24-HCO3)
If the increase in the AG due to added H+ is balanced by the decrease in HCO3, the ratio is 1, and only an AG acidosis exists.
However, if a hyperchloremic acidosis also is present, the decrease in HCO3 will be greater than the increase in the AG, and the
ratio will be <1. Alternatively, if a metabolic alkalosis is present, as may occur with nasogastric suctioning, the decrease in HCO3 is
less than the increase in AG, and the ratio is >1.
Interpretation of gap–gap:
<1: Additional non-AG acidosis
1-2: Typical pattern in high anion gap acidosis
>2: Additional metabolic alkalosis, or perhaps a coexisting chronic respiratory acidosis, which has resulted in a compensatory
elevation of HCO3

the application of 100% oxygen). The A-a gradient increases

Table 22.9  Alveolar Gas Equation with age according to the following formula:
PAO2 = PiO2 − (PaCO2/R), where PiO2 = FiO2 (PB − PH2O)
or using common values: Normal A-a gradient = (Age + 10) / 4.
PAO2 = (FiO2 × [760 − 47]) − (PaCO2/0.8) = 150 − PaCO2/0.8

Abbreviations: PiO2 = partial pressure of O2 in the central airways;

FiO2 (fraction of inspired oxygen); FiO2 on room air = 0.21, PaCO2 (value from
your arterial blood gas [ABG]); PB = barometric pressure (760 mm Hg at sea
level, PB = PN2 + PO2 + PaCO2 + PH2O); PH2O = water vapor pressure (47 mm Hg at
37° C); R = respiratory quotient (ratio of carbon dioxide production to oxygen
consumption) = VCO2/VO2 = 0.8 (usual).
hypoxemia, the A-a gradient is calculated. If it is normal, the
low arterial oxygen tension (PaO2) is caused by hypoventila-
tion. If it is increased, it may be the result of ventilation-
perfusion (V/Q) mismatching (PaO2 increases with application
of 100% oxygen) or of shunting (PaO2 does not increase with
Respiratory Acid-Base Disorders
Acute Respiratory Acidosis
Respiratory acidosis normally is buffered by an increase in
HCO3−. In an acute respiratory acidosis, the HCO3 will increase
by 1 mEq/L for each 10 mm Hg elevation in PaCO2 greater
than 40 mm Hg. The expected HCO3 is described by the fol-
lowing equation:
HCO3 = 24 + ([ΔPaCO2]/10)
If the PaCO2 increases from 40 to 60 mm Hg, the HCO3
therefore will increase from 24 to 26 mm Hg. An increase
222 Section II—Clinical and Laboratory Assessment
greater than this suggests the concomitant presence of a meta-
bolic alkalosis. The expected change in pH during this process
is expressed by the following equation:
Expected pH = 7.40 − (0.008 × [ ∆PaCO2 ])
Hence if the PaCO2 increases acutely from 40 to 60 mm Hg,
the pH will decrease to 7.24 (7.40 − [0.008 × 20]).
Chronic Respiratory Acidosis
In a chronic respiratory acidosis that develops over 2 to 3 days,
the HCO3− will increase by 4 mEq/L for every 10 mm Hg
change in PaCO2 described by the following equation:
Expected HCO3− = 24 + 4([ΔPaCO2]/10)
If the PaCO2 increases from 40 to 60 mm Hg, the HCO3−
will increase from 24 to 32 mm Hg. The pH is similarly
affected, changing 0.003 unit for each 1 mm Hg change in
PaCO2 according to the following equation:
Expected pH = 7.40 − (0.003 × [ΔPaCO2])
In a patient who has chronic obstructive pulmonary disease
(COPD) with a PaCO2 that equals 60 mm Hg, the expected
pH is 7.34.
In addition to the elevated HCO3−, there may also be a lower
chloride level. As new HCO3− is generated in the collecting
duct from hydrolysis, hydrogen is secreted into the tubular
lumen. Na+ flows in to maintain electrical neutrality. This
influx leaves less Na+ available to be resorbed with Cl− and so
Cl− excretion is increased.
Acute Respiratory Alkalosis
Respiratory alkalosis is buffered by a decrease in HCO3−. In an
acute respiratory alkalosis the HCO3− decreases by 2 mEq/L
for each 10 mm Hg decrease in PaCO2 less than 40 mm Hg
according to the following equation:
Expected HCO3− = 24 − 2(ΔPaCO2/10)
If the PaCO2 decreases from 40 to 20 mm Hg, the HCO3
will decrease from 24 to 20 mm Hg. In practice, a HCO3− of
less than 18 mEq/L indicates a coexisting metabolic acidosis.
The expected change in pH during this process is expressed
by the following equation:
Expected pH = 7.40 + (0.008 × [ΔPaCO2])
If the PaCO2 acutely decreases from 40 to 15 mm Hg, the
pH will rise to 7.56.
Chronic Respiratory Alkalosis
In chronic respiratory alkalosis the HCO3− will decrease by
5  mEq/L for every 10  mm  Hg decrease in PaCO2 Described
by the equation:
Expected HCO3− = 24 − 5(ΔPaCO2/10)
Thus if the PaCO2 decreases from 40 to 20 mmHg, the HCO3−
will decrease from 24 to approximately 14 mm Hg. The HCO3
generally does not fall below 12 for compensation. The change
in pH can be expressed by:
Expected pH = 7.40 + (0.003 × [ΔPaCO2])
Cerebral blood flow can be partially regulated by CO2-
induced changes in the pH. An alkalosis causes cerebral vaso-
constriction and there may be a 40% decrease in cerebral
blood flow with a decrease in the PaCO2 from 40 to 20 mm Hg.
This technique to decrease intracranial pressure (ICP) is mod-
ulated by pH and not the PaCO2.
Metabolic Acid-Base Disorders
The physiology underlying acid-base disorders is discussed
below but the equations for compensation are summarized
first. In metabolic disorders, compensation results in an ele-
vated PaCO2 for a metabolic alkalosis and a lower PaCO2 in a
metabolic acidosis. The expected changes can be calculated by
the following equations:
Metabolic alkalosis:
Expected PaCO2 = (0.7 × HCO3) + 20 ± 5
Metabolic acidosis:
Expected PaCO2 = (1.5 × HCO3) + 8 ± 2
Metabolic Alkalosis
Metabolic alkalosis can result from nasogastric (NG) suction-
ing or diuretics from the loss of chloride. For every molecule
of Cl− that is excreted, one molecule of HCO3− is reabsorbed.
Though there is a loss of both H+ and Cl−, it is chloride deple-
tion, stimulating HCO3− reabsorption, which is the major con-
tributing factor to the alkalosis. Thiazide and loop diuretics
promote natriuresis and Cl− is excreted with the sodium.
Additionally, volume depletion promotes metabolic alkalosis
in two ways. The increase in Na+ resorption is accompanied
by an increase in bicarbonate reabsorption. Volume depletion
also stimulates renin, promoting the formation of aldosterone,
which causes H+ excretion in the distal tubules. Volume resus-
citation is not effective in reversing the alkalosis unless Cl− is
also replaced. Hypokalemia is associated with alkalosis and a
shift of H+ into the cells and an increase in H+ secretion in the
distal tubules. Aldosterone and other mineralocorticoids act
on the collecting tubule to stimulate the resorption of Na+,
accompanied by the excretion of H+ and K+, which gives rise
to hypokalemia, hypertension, and hypernatremia. Metabolic
alkalosis also may result from the administration of exogenous
HCO3 such as HCO3 infusions, from acetate in TPN, or from
the administration of antacids in renal failure. Citrate in
banked blood may produce a metabolic alkalosis, but a
minimum of 8 units must be transfused.
Clinically, metabolic alkalosis can be divided into two
categories: (1) saline-responsive, associated with volume de-
pletion; and (2) saline-resistant, associated with excess miner-
alocorticoid activity. In saline-responsive metabolic alkalosis,
volume depletion stimulates the resorption of Na+ that also
causes the retention of Cl− and HCO3−. However, some extra
HCO3− is excreted with Na+ to maintain electrical neutrality,
so a urinary Cl− less than 20 mEq/L is a better indicator of
volume depletion in metabolic alkalosis.
Anion Gap
A normal anion-gap acidosis occurs because the lost HCO3− is
replaced by a Cl−. Causes for this are listed in Table 22.10. The
anion gap (AG) is defined as:
 Section II—Clinical and Laboratory Assessment 223
Table 22.10  Causes for Normal
Anion-Gap Acidosis
• Gastrointestinal loss—secretions from below the stomach
are rich in bicarbonate because they neutralize the acid
from the stomach.
• Renal loss of HCO3−—proximal RTA
• Early renal failure of a distal RTA
• Ingestions—HCO3−—acetate in TPN
• Carbonic anhydrase inhibitors—acetazolamide
• Recovery of ketoacidosis when given normal saline
RTA, Renal tubular acidosis; TPN, total parenteral nutrition.
Table 22.11  Causes of an Anion-
Gap Acidosis
Propylene glycol Ethanol
Lactic acidosis Ethylene glycol
Uremia Diabetic ketoacidosis (DKA)
Methanol Starvation
Salicylate Isoniazid
AG = Na+ − (Cl− + HCO3−)
An anion gap exists because there are anions such as sulfate,
phosphate, organic anions, and proteins that are not measured
on the routine chemistry panel, where there are very few
unmeasured cations. The normal AG is between 7 and
13 mEq/L, but newer autoanalyzers report a higher Cl−, so the
normal may be between 3 and 11 mEq/L. Hypoalbuminemia
may spuriously lower the AG and mask an underlying anion-
gap acidosis. Causes of an anion-gap acidosis are listed in
Table 22.11 and each is briefly described.
Propylene Glycol
Propylene glycol is an alcohol used to increase the water solu-
bility of lorazepam, diazepam, esmolol, nitroglycerin, and
phenytoin, and it is metabolized in the liver to lactate and
pyruvate. Signs of toxicity are agitation, seizures, coma, tachy-
cardia, and hypotension. The lactate may exceed 10 mEq/L
and clinically mimic sepsis.
Lactic Acidosis
A normal AG may accompany lactic acidosis. Iberti et al.
observed that 50% of patients with a lactate level between 5
and 9.9 mmol/L had AG less than 12 mEq/L.84 Lactic acidosis
results from mitochondrial dysfunction and has three forms.
(1) Type A arises from tissue hypoxia—septic, cardiogenic, or
hemorrhagic shock, or respiratory failure. (2) Type B is caused
by impaired mitochondrial oxygen utilization, such as from
toxins; that is, there is no systemic hypoperfusion. These
toxins can include: metformin (Glucophage); cyanide poison-
ing, for example, a nitroprusside infusion; thiamine (B1) defi-
ciency that can be associated with loss of body stores, increased
excretion by furosemide (Lasix), or magnesium depletion that
results in a functional form of thiamine deficiency; malignan-
cies, alcoholism, or AZT (zidovudine). (3) Type D occurs in
patients with jejunoileal bypass or in patients with short
bowel syndrome. The shortened small intestine is unable to
absorb dietary sugar, leaving carbohydrates to be metabolized
by bacteria in the colon. Lactobacilli metabolism of carbohy-
drate produces D lactic acid. It presents with neurologic
changes (confusion, ataxia, slurred speech, memory loss) sim-
ulating intoxication without alcohol ingestion after a high-
carbohydrate meal.
Ketoacidosis
Glucose and ketones are the two fuels available for the body.
In the fed state, glucose is the primary fuel source; ketones
assume this role in the fasting state. Ketones are produced
in the liver from triglycerides, and there are three types:
(1) B-hydroxybutyrate, the predominant ketone of diabetic
ketoacidosis (DKA) and alcoholic ketoacidosis (AKA), but not
measured by the standard urine assay; (2) acetoacetate, mea-
sured by the lab, but accounts for only 25% or less of the total
ketone load; and (3) acetone, which does not cause an acidosis.
Ketoacidosis becomes clinically significant when insulin activ-
ity is reduced in the following conditions: (1) DKA, the body
is unable to produce or release insulin; (2) AKA, alcohol
induces lipolysis (increases the supply of fatty acids for ketone
production) and consumes nicotinamide adenine dinucleo-
tide (NAD+) (decreasing gluconeogenesis, forcing the liver
to produce ketones); and (3) starvation-insulin release is
suppressed.
In AKA and starvation ketoacidosis, the treatment is glucose,
which stimulates endogenous insulin production to suppress
ketogenesis. In DKA are insulin deficiency, hyperglycemia,
ketosis (causing AG), and hyponatremia (often pseudohypo-
natremia), and treatment with insulin and fluids is needed.
Hyperglycemic hyperosmolar nonketotic (HHNK) coma is
seen in people with type 2 diabetes and is characterized
by extremely high blood sugar (≈1000 mg/dL) but without
ketosis (or acidosis) because insulin is present. The primary
treatment is volume replacement. Uremia causes an anion gap
acidosis from an impaired ability to clear the daily acid load
of daily metabolism, but the GFR is usually less than 40 mL/
min before this occurs.
Alcohols
The metabolites of alcohols, generated by alcohol and alde-
hyde dehydrogenase are responsible for the toxicity associated
with alcohol ingestion. Thus it can take 12 to 24 hours for
acidosis and organ failure to occur, and this may be delayed
further when there also is ethanol ingestion. Consequently,
management is to prevent metabolite formation until the
parent compound can be excreted or dialyzed. Methanol and
ethylene glycol intoxication present with a high anion gap
acidosis, but the measurements of these may be off-site and
will not be available to assist immediate clinical decision
making. The osmolal gap defined as the difference between
the calculated osmolality (osmolality = 2 × Na+[mmol/L] +
BUN [mg/dL]/2.8 + [glucose mg/dL]/18) and the measured
osmolality may help clinical decisions. The normal osmolal
gap is 10 mOsm/kg H2O or less. Causes of an elevated osmolal
gap are listed in Table 22.12.
As one integrates the history and laboratory findings, it is
important to remember that the alcohols contribute to the
osmolal gap and the toxic metabolites contribute to the
acidosis. Thus toxicity is not excluded on a gap less than
224 Section II—Clinical and Laboratory Assessment

10 mOsm/kg. A small elevation isn’t necessarily from a toxic

Table 22.12  Causes of an Elevated Osmolal metabolite, but a gap greater than 25 is presumptive evidence
Gap with an Anion-Gap Acidosis in the appropriate setting. 100 mg/dL of ethanol (MW = 46)
Ethylene glycol Elevated osmolal gap without acidosis
adds 22 mOsm/kg. Isopropyl alcohol does not cause an acido-
Methanol Ethanol sis, but its primary metabolite, acetone, manifests by positive
Formaldehyde Isopropanol serum and urinary ketones.
ESRD without dialysis Diethyl ether
Paraldehyde Mannitol
DKA Severe hyperproteinemia Methanol
AKA Severe hyperlipidemia
Lactic acidosis Methanol poisoning occurs with the ingestion of shellac,
varnish, wiper fluids and deicing solutions, or denatured
AKA, Alcoholic ketoacidosis; DKA, diabetic ketoacidosis; ESRD, end-stage
renal disease. gelled alcohol (Sterno). It is converted by alcohol dehydroge-
nase into the toxic formaldehyde and then formic acid.

A B

C D

Fig. 22.13  Brain MRI obtained on day 15 after methanol intoxication. A, T2-weighted image showed high signal abnormalities in bilateral basal
ganglia (arrows), frontal, and occipital subcortical white matter (arrowheads), consistent with edematous change. B, T2-weighted image showed
edematous change involving bilateral optic tracts and optic radiations (arrows). High signal edematous change was also noted in the optic disc of
the left eye (arrowheads). C, T1-weighted image showed slightly high signal component in bilateral basal ganglia, indicating the hemorrhage (arrows).
D, T1-weighted image with gadolinium administration showed marginal enhancement in bilateral putamen, indicating breakdown of the blood-brain
barrier.
 Section II—Clinical and Laboratory Assessment 225
Malaise, nausea, and headache occur within 12 to 24 hours of
ingestion, followed by acidosis, visual disturbances (formalde-
hyde’s retinal toxicity), seizures, and coma. Serious toxicity
occurs with one teaspoon—50 mL may be lethal. Optic nerve
edema and basal ganglia infarcts can be observed (Fig. 22.13).
Ethylene Glycol
Ethylene glycol is found in antifreeze and solvents, and inges-
tion manifests initially by neurologic symptoms, including
cranial nerve palsies and tetany (from oxalate-induced hypo-
calcemia) and later by pulmonary edema. Renal failure is a late
finding due to the effects of oxalate crystals. Lactic acidosis
can be seen with ethylene glycol, but it is insufficient to
account for the degree of acidosis.
Treatment of Alcohol or Methanol Toxicity
For each of these disorders time is crucial, and important
decisions may need to be made without all the data. Therapy
includes sodium bicarbonate and fomepizole. Fomepizole
(or ethanol) inhibits alcohol dehydrogenase and blocks
metabolite production, but early treatment is crucial because
it does not prevent toxicity if metabolism to an acid species
has already occurred. It is useful for both methanol and eth-
ylene glycol poisoning. Hemodialysis should be used if meta-
bolic acidosis is present or there is evidence of end-organ
damage. All methanol intoxications should also receive cofac-
tor therapy: either folinic acid (50 mg IV) or folic acid (50 mg
every 6 hours IV) and probably thiamine (100 mg IV).
Salicylate
Aspirin overdose is characterized by the accumulation of sal-
icylic acid, the active metabolite of acetylsalicylic acid toxic-
ity is characterized by two primary acid-base disorders.
Respiratory alkalosis initially ensues from direct CNS stimu-
lation and is followed by a metabolic acidosis. Salicylic acid
itself contributes little to the anion gap (due to its high
molecular weight), but the anion gap results from the
buildup of lactic acid and ketones by unclear mechanisms.
The treatment is alkalinization of the urine. Increasing the
pH from 7.2 to 7.5 halves the tissue level of salicylic acid.
Dialysis is indicated for patients in coma or with levels
80 mg/dL or greater.
Conclusion
Electrolyte and acid-base abnormalities and acute and chronic
kidney injury are ubiquitous in neurocritical care. An under-
standing of basic renal physiology assists in providing a frame-
work for understanding these disorders and a context for their
management.
References
1. Piccinni P, Cruz DN, Gramaticopolo S, et al. Prospective multicenter study
on epidemiology of acute kidney injury in the ICU: a critical care
nephrology Italian collaborative effort (NEFROINT). Minerva Anestesiol
2011;77(11):1072–83. Epub 2011, May 11.
2. Clec’h C, Gonzalez F, Lautrette A, et al. Multiple-center evaluation of
mortality associated with acute kidney injury in critically ill patients: a
competing risks analysis. Crit Care 2011;15(3):R128. Epub 2011, May 17.
3. Bellomo R. Acute renal failure. Semin Respir Crit Care Med 2011;32(5):639–
50. Epub 2011, Oct 11.
4. Mandelbaum T, Scott DJ, Lee J, et al. Outcome of critically ill patients with
acute kidney injury using the Acute Kidney Injury Network criteria. Crit
Care Med 2011;39(12):2659–64.
5. Li N, Zhao WG, Zhang WF. Acute kidney injury in patients with severe
traumatic brain injury: implementation of the acute kidney injury network
stage system. Neurocrit Care 2011;14(3):377–81.
6. Rose BD, Post TW. Introduction to renal function. Available at www.upto
date.com (accessed August 2009).
7. Rose BD, Post TW. Collecting tubules. Available at www.uptodate.com
(accessed August 2009).
8. Rose BD. Urine osmolality versus specific gravity. Available at www.uptodate
.com (accessed August 2009).
9. Pasticci F, Fantuzzi AL, Pegoraro M, et al. Nutritional management of stage 5
chronic kidney disease. J Ren Care 2012;38(1):50–8.
10. Patel P, Nandwani V, McCarthy PJ, et al. Continuous renal replacement
therapies: a brief primer for the neurointensivist. Neurocrit Care 2010;13(2):
286–94.
11. Arulkumaran N, Montero RM, Singer M. Management of the dialysis patient
in general intensive care. Br J Anaesth 2012;108(2):183–92. Epub 2012, Jan 4.
12. Basu RK, Wheeler DS, Goldstein S, et al. Acute renal replacement therapy in
pediatrics. Int J Nephrol 2011;2011:785392. Epub 2011, Jun 1.
13. Chuasuwan A, Kellum JA. Acute kidney injury and its management. Contrib
Nephrol 2011;171:218–25. Epub 2011, May 23.
14. Claure-Del Granado R, Mehta RL. Withholding and withdrawing renal
support in acute kidney injury. Semin Dial 2011;24(2):208–14. doi:10.1111/
j.1525-139X.2011.00832.x.
15. Srisawat N, Kellum JA. Acute kidney injury: definition, epidemiology, and
outcome. Curr Opin Crit Care 2011;17(6):548–55.
16. Singbartl K, Kellum JA. AKI in the ICU: definition, epidemiology, risk
stratification, and outcomes. Kidney Int 2011;Oct 5. Epub ahead of print.
17. Liano F, Pascual J. Madrid Acute Renal Failure Study Group. Epidemiology
of acute renal failure: a prospective, multicenter, community-based study.
Kidney Int 1996;50:811.
18. Mehta RL, Pascual MT, Soroko S, et al. Spectrum of acute renal failure in the
intensive care unit: the 1 PICARD experience. Kidney Int 2004;66:1613.
19. Muther RS. Acute renal failure in the critically ill. In: ACCP, editor. ACCP
Critical Care Board Review 2007: Course Syllabus. Northbrook, Ill: American
College of Chest Physicians; 2007.
20. Bellomo R, Ronco C, Kellum JA, et al. Acute renal failure-definition,
outcome measures, animal models, fluid therapy and information
technology needs: the Second International Consensus Conference of the
Acute Dialysis Quality Initiative ADQI. Group Crit Care 2004;8:R204–12.
21. Chertow GM, Burick E, Honour M, et al. Acute kidney injury, mortality,
length of stay, and costs in hospitalized patients. J Am Soc Nephrol 2005;
16:3365–70.
22. Palevsky PM. Definition of acute kidney injury (acute renal failure).
Available at www.uptodate.com (accessed August 2009).
23. Waring WS, Moonie A. Earlier recognition of nephrotoxicity using novel
biomarkers of acute kidney injury. Clin Toxicol (Phila) 2011;49(8):
720–8.
24. Payen D, Legrand M. Can we identify prerenal physiology and does it
matter? Contrib Nephrol 2011;174:22–32. Epub 2011, Sep 9.
25. Srisawat N, Wen X, Lee M, et al. Urinary biomarkers and renal recovery in
critically ill patients with renal support. Clin J Am Soc Nephrol
2011;6(8):1815–23. Epub 2011, Jul 14.
26. Doi K, Negishi K, Ishizu T, et al. Evaluation of new acute kidney injury
biomarkers in a mixed intensive care unit. Crit Care Med 2011;39(11):
2464–9.
27. Panagiotou A, Garzotto F, Gramaticopolo S, et al. Continuous real-time
urine output monitoring for early detection of acute kidney injury. Contrib
Nephrol 2011;171:194–200. Epub 2011, May 23.
28. Bellomo R. Defining, quantifying, and classifying acute renal failure. Crit
Care Clin 2005;21:223–7.
29. Muther RS. Drug interference with renal function tests. Am J Kidney Dis
1983;3:118–20.
30. Kaplan AA, Kohn OF. Fractional excretion of urea as a guide to renal
dysfunction. Am J Nephrol 1992:12:54–9.
31. Post TW, Rose BD. Urinalysis in the diagnosis of renal disease. Available at
www.uptodate.com (accessed August 2009).
32. Bohle A, Christensen J, Kokot F, et al. Acute renal failure in man: new aspects
concerning pathogenesis. A morphometric study. Am J Nephrol 1990;10:
374–88.
33. Lameire N, Vanholder R, Van Biesen W. Loop diuretics for patients with
acute renal failure: helpful or harmful. JAMA 2002;288:2599.
226 Section II—Clinical and Laboratory Assessment
34. Cantarovich F, Rangoonwala B, Lorenz H, et al. High-dose furosemide for
established ARF: a prospective, randomized, double-blind, placebo-
controlled, multicenter trial. Am J Kidney Dis 2004;44:402.
35. Delpire E, Lu J, England R, et al. Deafness and imbalance associated with
inactivation of the secretory Na-K-2Cl co-transporter. Nat Genet 1999;
22:192.
36. Marik PE, Iglesias J. Low-dose dopamine does not prevent acute renal failure
in patients with septic shock and oliguria. NORASEPT II Study Investigators.
Am J Med 1999;107:387–90.
37. Bellomo R, Chapman M, Finfer S, et al. Low-dose dopamine in patients
with early renal dysfunction: a placebo-controlled randomized trial. Lancet
2000;356:2139.
38. Lauschke A, Teichgraber UK, Frei U, et al. Low-dose dopamine worsens
renal perfusion in patients with acute renal failure. Kidney Int 2006;69:
1669.
39. Reid GM, Muther RS. Nitroprusside-induced acute azotemia. Am J Nephrol
1987;7:313–15.
40. Rose B, Appel GB. Clinical manifestations and diagnosis of acute interstitial
nephritis. Available at www.uptodate.com (accessed August 2009).
41. Hecker R, Sherlock S. Electrolyte and circulatory changes in terminal liver
failure. Lancet 1956;2:1221–5.
42. Koppel MH, Coburn JN, Mims MM, et al. Transplantation of cadaveric
kidneys from patients with hepatorenal syndrome. Evidence for the
functional nature of renal failure in advanced liver disease. N Engl J Med
1969;280:1267–71.
43. Iwatsuke S, Popovtzer MM, Corman JI, et al. Recovery from hepatorenal
syndrome after orthotopic liver transplantation. N Engl J Med 1973;289:
1155–9.
44. Arroyo V, Gines P, Gerbes AL, et al. Definition and diagnostic criteria of
refractory ascites and hepatorenal syndrome in cirrhosis. Int Ascites Club
Hepatol 1996;23:164.
45. Watt K, Uhanova J, Minuk GY. Hepatorenal syndrome: diagnostic accuracy,
clinical features, and outcome in a tertiary care centre. Am J Gastroenterol
2002;97:2046–50.
46. Esrailian E, Pantangco ER, Kyulo NL, et al. Octreotide/Midodrine therapy
significantly improves renal function and 30-day survival in patients with
type 1 hepatorenal syndrome: a large retrospective series. Dig Dis Sci
2007;52:742.
47. Uriz J, Gines P, Cardenas A, et al. Terlipressin plus albumin infusion: an
effective and safe therapy of hepatorenal syndrome. J Hepatol 2000;22:43–8.
48. Wong F, Pantea L, Sniderman K. Midodrine, octreotide, albumin, and TIPS
in selected patients with cirrhosis and type 1 hepatorenal syndrome.
Hepatology 2004;40:55–64.
49. Olson JC, Wendon JA, Kramer DJ, et al. Intensive care of the patient with
cirrhosis. Hepatology 2011;54(5):1864–72. doi:10.1002/hep.24622.
50. McCarthy JT. Prognosis of patients with acute renal failure in the intensive
care unit: a tale of two eras. Mayo Clin Proc 1996;7:117–26.
A complete list of references for this chapter can be found online at
www.expertconsult.com.
 Section II—Clinical and Laboratory Assessment 226.e1
References
1. Piccinni P, Cruz DN, Gramaticopolo S, et al. Prospective multicenter study
on epidemiology of acute kidney injury in the ICU: a critical care
nephrology Italian collaborative effort (NEFROINT). Minerva Anestesiol
2011;77(11):1072–83. Epub 2011, May 11.
2. Clec’h C, Gonzalez F, Lautrette A, et al. Multiple-center evaluation of
mortality associated with acute kidney injury in critically ill patients: a
competing risks analysis. Crit Care 2011;15(3):R128. Epub 2011, May 17.
3. Bellomo R. Acute renal failure. Semin Respir Crit Care Med 2011;32(5):639–
50. Epub 2011, Oct 11.
4. Mandelbaum T, Scott DJ, Lee J, et al. Outcome of critically ill patients with
acute kidney injury using the Acute Kidney Injury Network criteria. Crit
Care Med 2011;39(12):2659–64.
5. Li N, Zhao WG, Zhang WF. Acute kidney injury in patients with severe
traumatic brain injury: implementation of the acute kidney injury network
stage system. Neurocrit Care 2011;14(3):377–81.
6. Rose BD, Post TW. Introduction to renal function. Available at
www.uptodate.com (accessed August 2009).
7. Rose BD, Post TW. Collecting tubules. Available at www.uptodate.com
(accessed August 2009).
8. Rose BD. Urine osmolality versus specific gravity. Available at www.uptodate.
com (accessed August 2009).
9. Pasticci F, Fantuzzi AL, Pegoraro M, et al. Nutritional management of stage 5
chronic kidney disease. J Ren Care 2012;38(1):50–8.
10. Patel P, Nandwani V, McCarthy PJ, et al. Continuous renal replacement
therapies: a brief primer for the neurointensivist. Neurocrit Care
2010;13(2):286–94.
11. Arulkumaran N, Montero RM, Singer M. Management of the dialysis patient
in general intensive care. Br J Anaesth 2012;108(2):183–92. Epub 2012, Jan 4.
12. Basu RK, Wheeler DS, Goldstein S, et al. Acute renal replacement therapy in
pediatrics. Int J Nephrol 2011;2011:785392. Epub 2011, Jun 1.
13. Chuasuwan A, Kellum JA. Acute kidney injury and its management. Contrib
Nephrol 2011;171:218–25. Epub 2011, May 23.
14. Claure-Del Granado R, Mehta RL. Withholding and withdrawing renal
support in acute kidney injury. Semin Dial 2011;24(2):208–14. doi:10.1111/j
.1525-139X.2011.00832.x.
15. Srisawat N, Kellum JA. Acute kidney injury: definition, epidemiology, and
outcome. Curr Opin Crit Care 2011;17(6):548–55.
16. Singbartl K, Kellum JA. AKI in the ICU: definition, epidemiology, risk
stratification, and outcomes. Kidney Int 2011;Oct 5. Epub ahead of print.
17. Liano F, Pascual J, Madrid Acute Renal Failure Study Group. Epidemiology
of acute renal failure: a prospective, multicenter, community-based study.
Kidney Int 1996;50:811.
18. Mehta RL, Pascual MT, Soroko S, et al. Spectrum of acute renal failure in the
intensive care unit: the 1 PICARD experience. Kidney Int 2004;66:1613.
19. Muther RS. Acute renal failure in the critically ill. In: ACCP, editor. ACCP
Critical Care Board Review 2007: Course Syllabus. Northbrook, Ill: American
College of Chest Physicians; 2007.
20. Bellomo R, Ronco C, Kellum JA, et al. Acute renal failure-definition,
outcome measures, animal models, fluid therapy and information
technology needs: the Second International Consensus Conference of the
Acute Dialysis Quality Initiative ADQI. Group Crit Care 2004;8:R204–12.
21. Chertow GM, Burick E, Honour M, et al. Acute kidney injury, mortality,
length of stay, and costs in hospitalized patients. J Am Soc Nephrol 2005;
16:3365–70.
22. Palevsky PM. Definition of acute kidney injury (acute renal failure).
Available at www.uptodate.com (accessed August 2009).
23. Waring WS, Moonie A. Earlier recognition of nephrotoxicity using novel
biomarkers of acute kidney injury. Clin Toxicol (Phila) 2011;49(8):720–8.
24. Payen D, Legrand M. Can we identify prerenal physiology and does it
matter? Contrib Nephrol 2011;174:22–32. Epub 2011, Sep 9.
25. Srisawat N, Wen X, Lee M, et al. Urinary biomarkers and renal recovery in
critically ill patients with renal support. Clin J Am Soc Nephrol 2011;6(8):
1815–23. Epub 2011, Jul 14.
26. Doi K, Negishi K, Ishizu T, et al. Evaluation of new acute kidney injury
biomarkers in a mixed intensive care unit. Crit Care Med 2011;39(11):2464–9.
27. Panagiotou A, Garzotto F, Gramaticopolo S, et al. Continuous real-time
urine output monitoring for early detection of acute kidney injury. Contrib
Nephrol 2011;171:194–200. Epub 2011, May 23.
28. Bellomo R. Defining, quantifying, and classifying acute renal failure. Crit
Care Clin 2005;21:223–7.
29. Muther RS. Drug interference with renal function tests. Am J Kidney Dis
1983;3:118–20.
30. Kaplan AA, Kohn OF. Fractional excretion of urea as a guide to renal
dysfunction. Am J Nephrol 1992:12:54–9.
31. Post TW, Rose BD. Urinalysis in the diagnosis of renal disease. Available at
www.uptodate.com (accessed August 2009).
32. Bohle A, Christensen J, Kokot F, et al. Acute renal failure in man: new aspects
concerning pathogenesis. A morphometric study. Am J Nephrol 1990;10:
374–88.
33. Lameire N, Vanholder R, Van Biesen W. Loop diuretics for patients with
acute renal failure: helpful or harmful. JAMA 2002;288:2599.
34. Cantarovich F, Rangoonwala B, Lorenz H, et al. High-dose furosemide for
established ARF: a prospective, randomized, double-blind, placebo-
controlled, multicenter trial. Am J Kidney Dis 2004;44:402.
35. Delpire E, Lu J, England R et al. Deafness and imbalance associated with
inactivation of the secretory Na-K-2Cl co-transporter. Nat Genet 1999;
22:192.
36. Marik PE, Iglesias J. Low-dose dopamine does not prevent acute renal failure
in patients with septic shock and oliguria. NORASEPT II Study Investigators.
Am J Med 1999;107:387–90.
37. Bellomo R, Chapman M, Finfer S, et al. Low-dose dopamine in patients with
early renal dysfunction: a placebo-controlled randomized trial. Lancet
2000;356:2139.
38. Lauschke A, Teichgraber UK, Frei U, et al. Low-dose dopamine worsens renal
perfusion in patients with acute renal failure. Kidney Int 2006;69:1669.
39. Reid GM, Muther RS. Nitroprusside-induced acute azotemia. Am J Nephrol
1987;7:313–15.
40. Rose B, Appel GB. Clinical manifestations and diagnosis of acute interstitial
nephritis. Available at www.uptodate.com (accessed August 2009).
41. Hecker R, Sherlock S. Electrolyte and circulatory changes in terminal liver
failure. Lancet 1956;2:1221–5.
42. Koppel MH, Coburn JN, Mims MM, et al. Transplantation of cadaveric
kidneys from patients with hepatorenal syndrome. Evidence for the
functional nature of renal failure in advanced liver disease. N Engl J Med
1969;280:1267–71.
43. Iwatsuke S, Popovtzer MM, Corman JI, et al. Recovery from hepatorenal
syndrome after orthotopic liver transplantation. N Engl J Med 1973;289:
1155–9.
44. Arroyo V, Gines P, Gerbes AL, et al. Definition and diagnostic criteria of
refractory ascites and hepatorenal syndrome in cirrhosis. Int Ascites Club
Hepatol 1996;23:164.
45. Watt K, Uhanova J, Minuk GY. Hepatorenal syndrome: diagnostic accuracy,
clinical features, and outcome in a tertiary care centre. Am J Gastroenterol
2002;97:2046–50.
46. Esrailian E, Pantangco ER, Kyulo NL, et al. Octreotide/midodrine therapy
significantly improves renal function and 30-day survival in patients with
type 1 hepatorenal syndrome: a large retrospective series. Dig Dis Sci
2007;52:742.
47. Uriz J, Gines P, Cardenas A, et al. Terlipressin plus albumin infusion:
an effective and safe therapy of hepatorenal syndrome. J Hepatol 2000;22:
43–8.
48. Wong F, Pantea L, Sniderman K. Midodrine, octreotide, albumin, and TIPS
in selected patients with cirrhosis and type 1 hepatorenal syndrome.
Hepatology 2004;40:55–64.
49. Olson JC, Wendon JA, Kramer DJ, et al. Intensive care of the patient with
cirrhosis. Hepatology 2011;54(5):1864–72. doi:10.1002/hep.24622.
50. McCarthy JT. Prognosis of patients with acute renal failure in the intensive
care unit: a tale of two eras. Mayo Clin Proc 1996;7:117–26.
51. Richards WO, Scovill W, Shin B, et al. Acute renal failure associate with
increased intra-abdominal pressure. Ann Surg 1983;197:183.
52. Mohmand H, Goldfarb S. Renal dysfunction associated with intra-
abdominal hypertension and the abdominal compartment syndrome. J Am
Soc Nephrol 2011;22(4):615–21. Epub 2011, Feb 10.
53. Oates JA, Fitzgerald GA, Branch RA, et al. Clinical implication of
prostaglandin and thromboxane A2 formation. N Engl J Med 1988;319:761.
54. Huerta C, Castellsague J, Baras-Lorenzo C, et al. Nonsteroidal anti-
inflammatory drugs and risk of ARF in the general population. Am J Kidney
Dis 2005;45:531.
55. Patrono C, Dunn MJ. The clinical significance of inhibition of renal
prostaglandin synthesis. Kidney Int 1987;32:1.
56. Clive DM, Stoff JS. Renal syndromes associated with nonsteroidal
anti-inflammatory drugs. N Engl J Med 1984;310:563.
57. Radford MG, Holley KE, Grande JP, et al. Reversible membranous
nephropathy associated with the use of nonsteroidal anti-inflammatory
drugs. JAMA 1996;276:466.
58. Kaysen GA, Pond SM, Roper MH, et al. Combined hepatic and renal injury
in alcoholics during therapeutic use of acetaminophen. Arch Intern Med
1985;145:2019–23.
59. Blakely P, McDonald BR. Acute renal failure de to acetaminophen ingestion:
A case report and review of the literature. J Am Soc Nephrol 1995;6:48.
226.e2 Section II—Clinical and Laboratory Assessment
60. Weisbord SD, Palevsky PM. Radiocontrast-induced acute renal failure.
J Intensive Care Med 2005;20:63.
61. Lakhal K, Ehrmann S, Chaari A, et al. Acute Kidney Injury Network
definition of contrast-induced nephropathy in the critically ill: incidence and
outcome. J Crit Care 2011;26(6):593–9. Epub 2011, Jul 6.
62. Agmon Y, Peleg H, Greenfield A, et al. Nitric oxide and prostanoids protect
the renal outer medulla from radiocontrast toxicity in the rat. J Clin Invest
1994;94:1069.
63. Morcos DK, El-Nahas AM, Broen P, et al. Effect of iodinated water soluble
contrast media on urinary protein assays. BMJ 1992;305:29.
64. Tepel M, van der Giet M, Schwarzfeld C, et al. Prevention of radiographic-
contrast-agent-induces reductions in renal function by acetylcysteine. N Engl
J Med 2000;343:180–4.
65. Boton R, Gaviria M, Batlle DC. Prevalence, pathogenesis, and treatment of
renal dysfunction associated with chronic lithium therapy. Am J Kidney Dis
1987;10:329.
66. Lehmann K, Ritz E. Angiotensin-converting enzyme inhibitors may cause
renal dysfunction in patients on long-term lithium treatment. Am J Kidney
Dis 1995;25:82.
67. Kaplan NM, Rose BD. Major side effects of ACI inhibitors. Available at
www.uptodate.com (accessed August 2009).
68. Miskulin D, Gul A, Rudnick MR, et al. Nephrogenic systemic fibrosis/
nephrogenic fibrosing dermopathy in advanced renal failure. Available at
www.uptodate.com (accessed August 2009).
69. Haas M, Ohler L, Watzke H, et al. The spectrum of acute renal failure in
tumor lysis syndrome. Nephrol Dial Transplant 1999;13:776–9.
70. Fernandez PC, Larson RA, Agus ZS. Tumor lysis syndrome. Available at
www.uptodate.com (accessed August 2009).
71. Gabow PA, Kaehny WD, Kelleher SP. The spectrum of rhabdomyolysis.
Medicine (Baltimore) 1982;61:141.
72. Miller ML. Rhabdomyolysis. Available at www.uptodate.com (accessed
August 2009).
73. Rose BD. Clinical features and prevention of heme pigment-induced acute
tubular necrosis. Available at www.uptodate.com (accessed August 2009).
74. August P, George JN. Acute renal failure in pregnancy. www.uptodate.
com (accessed August 2009).
75. Poole BD, Schrier RW. Acute renal failure. In Fink MP, Abraham E, et al.
Textbook of critical care, ed 5. Elsevier, New York, 2005, p. 1139–50.
76. Rose BD. Urine output in urinary tract obstruction and postobstructive
diuresis. Available at www.uptodate.com (accessed August 2009).
77. Negash DT, Dhingra VK, Copland M, et al. Intensity of continuous renal
replacement therapy in acute kidney injury in the intensive care unit: a
systematic review and meta-analysis. Vasc Endovascular Surg
2011;45(6):504–10. Epub 2011, Jun 5.
78. Graham P, Lischer E. Nursing issues in renal replacement therapy:
organization, manpower assessment, competency evaluation and quality
improvement processes. Semin Dial 2011;24(2):183–7. doi:10.1111/j.1525-
139X.2011.00835.x.
79. Prowle JR, Schneider A, Bellomo R. Clinical review: optimal dose of
continuous renal replacement therapy in acute kidney injury. Crit Care
2011;15(2):207. Epub 2011, Mar 18.
80. Lines SW, Cherukuri A, Murdoch SD, et al. The outcomes of critically ill
patients with acute kidney injury receiving renal replacement therapy. Int J
Artif Organs 2011;34(1):2–9.
81. Barri YM, Golper TA, Seizures in patients undergoing hemodialysis.
Available at www.uptodate.com (accessed August 2009).
82. Marino PL. The ICU book. 3rd ed. Lippincott Williams & Wilkins,
Philadelphia, 2007.
83. Gauthier PM, Szerlip HM. Metabolic acidosis in the intensive care unit.
Critical care clinics. Crit Care Clin 2002;18(2):289–308.
84. Iberti TJ, Leibowitz AB, Papadakos PJ. Low sensitivity of the anion gap as a
screen to detect hyperlactatemia in critically ill patients. Crit Care Med
1990;18:275–7.
85. Trotman RL, Williamson JC, Shoemaker DM, Salzer WL. Antibiotic dosing
in critically ill adult patients receiving continuous renal replacement therapy.
Clin Infect Dis 2005;41(8):1159–66.

Gastrointestinal and
Hepatic Disorders
Christiana E. Hall and Aashish R. Patel
Introduction
Gastrointestinal (GI) issues and maladies and in particular GI
bleeding, liver dysfunction, and pancreatitis can complicate
the primary course of critically ill neurologic patients. In addi-
tion, patients with hepatic encephalopathy may be cared for
in the neurocritical care unit (NCCU). The neurointensivist
therefore must be prepared to recognize, coordinate specialty
care, and provide local monitoring and support when these
issues arise. This chapter reviews disorders of the GI system
including the liver and pancreas that are of importance in the
NCCU. A comprehensive review of GI and hepatic function is
beyond the scope of this chapter. Nutrition is reviewed in
Chapter 14.
Gastrointestinal Bleeding
GI bleeding is relatively common in acutely ill hospitalized
patients, and the risk may be augmented in the NCCU associ-
ated with severe sympathetic stress such as in acute subarach-
noid hemorrhage (SAH) or neurotrauma patients, including
both traumatic brain injury (TBI) and spinal cord injury
(SCI) and in patients in whom steroids are administered. The
bleeding source may be from the upper or lower GI tract. The
ligament of Treitz is the landmark that distinguishes the upper
from lower GI tract. Common causes for upper GI bleeding
(UGIB) are peptic ulcer disease, including stress-induced ulcer
bleeding,1 variceal hemorrhage, and esophagitis. Common
causes for lower GI bleeding (LGIB) are diverticulitis, angio-
dysplasia, colitis that can have several causes, tumors, and local
anorectal pathologies. UGIB presents with hematemesis,
melena, hematochezia, and occasionally hypotension when
more severe. In patients with a nasogastric tube in place, the
bleeding may manifest in the drainage contents (e.g., coffee-
ground gastric contents). LGIB presents with hematochezia
and melena, and can be a reason for anemia.
When GI bleeding occurs the initial focus is to ensure
hemodynamic stability with resuscitation using crystalloid or
blood products as appropriate. Patient monitoring therefore
is aimed at recognition and avoidance of hemorrhagic shock
using reliable real-time continuous blood pressure and heart
rate monitoring. In more severe GI bleeding, resuscitation can
be facilitated using central venous pressure (CVP), mixed
venous oxygen saturation (ScvO2), and newer noninvasive
© Copyright 2013 Elsevier Inc. All rights reserved.
Chapter
23  
II
hemodynamic monitoring techniques.2 These noninvasive
devices can provide data on cardiac performance, preload, and
systemic vascular resistance, similar to that provided by Swan-
Ganz catheterization (see Chapter 19).
Laboratory assessment includes hemoglobin and hemato-
crit, and to exclude an occult bleeding diathesis platelet count,
coagulation indices, and if necessary disseminated intravascu-
lar coagulation (DIC). The brain contains high levels of tissue
factor (TF), and in severe injury or associated with a pro-
longed or complex surgery (e.g., for a large arteriovenous
malformation [AVM]), significant TF release may occur and
induce the DIC cascade. The optimal platelet count after an
injury to or a disease of the nervous system may depend on
pathology and time after the initial insult. There is general
agreement that a platelet count greater than 100,000 is a rea-
sonable goal after TBI, after neurosurgery, or for anticipated
intervention or instrumentation of the central nervous system
(CNS). This includes for placement of an intracranial monitor
or external ventricular drain. By contrast for other patients
without CNS disorders or likely surgery, a platelet count
greater than 50,000 is reasonable perhaps even when there is
GI bleeding. For the neurointensivist this means that platelet
support may differ depending on whether GI bleeding occurs
in a patient with a disorder such as myasthenia gravis or severe
TBI or is a postoperative neurosurgical patient. The various
laboratory indices should be repeated at intervals until they
are stabilized. This typically includes assessment of complete
blood count with platelets and if necessary coagulation studies
every 6 hours.
GI bleeding requires an assessment of the source of bleed-
ing and should include an early consultation with a gastroen-
terology service. Endoscopy typically is the initial means to
identify the source of UGIB, determine whether active bleed-
ing still is present, and guide the frequency of monitoring,
including repeat endoscopy or other diagnostic tests. In addi-
tion, the source of bleeding may be treated through endo-
scopic techniques. An intravenous bolus of proton pump
inhibitors followed by an infusion is recommended to prepare
the patient for endoscopy associated with nonvariceal bleed-
ing.3,4 Somatostatin or octreotide can be considered for vari-
ceal bleeding. The Blatchford score, which is based on several
factors including urea and hemoglobin levels, systolic blood
pressure, heart rate, hepatic disease, cardiac failure, melena,
227
228 Section II—Clinical and Laboratory Assessment
and presentation with syncope, can be used to determine
which patients require urgent endoscopy.5 Once endoscopy is
complete the Rockall score, which includes the endoscopic
diagnosis and evidence for hemorrhage among other vari-
ables, can be used to stratify patients into high or low risk for
recurrent hemorrhage or mortality.6 Bedside colonoscopy is
Table 23.1  Summary of Recommendations from 2010 Guidelines of the International Consensus
Upper Gastrointestinal Bleeding Conference Group*
Endoscopic management:
1. Institution-specific protocols for multidisciplinary UGIB
are recommended; availability of an endoscopist trained
in endoscopic hemostasis is recommended.
2. Urgent availability of endoscopic support staff
3. Endoscopy within 24 hours for most cases
4. Hemostatic treatment not indicated for low-risk lesions
5. Clots in the ulcer bed should be irrigated with the goal
of revealing the underlying lesion for treatment.
6. Removing adherent clots is controversial; consider
endoscopic therapy vs. intensive PPI therapy.
7. Hemostatic endoscopic treatment is indicated for
high-risk lesions.
8. Epinephrine injection alone is suboptimal: best used as
part of a combined therapy
9. Endoscopic coaptive therapy methods are equivalent.
10. Clips, thermocoagulation, or sclerosant injection should
be used for high-risk lesions, alone or combined with
epinephrine.
11. Second-look endoscopy as a routine is not recommended.
12. In case of rebleeding, a second endoscopic therapy
attempt is generally recommended.
Nonendoscopic and nonpharmacologic in-hospital
management:
1. Following endoscopy, patients at low risk may be fed
within 24 hours.
2. Patients with high-risk lesions should remain hospitalized
for 72 hours after endoscopy.
3. Surgical consultation is advised when endoscopic therapy
has failed.
4. Percutaneous embolization may provide an alternative to
surgery when endoscopic therapy has failed.
5. Patients with bleeding ulcers require testing for
Helicobacter pylori and should receive eradication therapy
if positive; eradication should be confirmed with a
negative test result.
6. Negative acute testing for H. pylori should be repeated.
Adapted from Barkun AN, Bardou M, Kuipers EJ, et al. International consensus recommendations on the management of patients with nonvariceal upper
gastrointestinal bleeding. Ann Intern Med 2010;152:101–13; Barkun A, Bardou M, Marshall JK, et al. Consensus recommendations for managing patients with
nonvariceal upper gastrointestinal bleeding. Ann Intern Med 2003;139:843–57.
*Nonvariceal UGIB.
ASA, Acetylsalicylic acid; COX, cyclooxygenase; Hgb, hemoglobin; NSAIDs, nonsteroidal anti-inflammatory drugs; PRBC, packed red blood cell; PPI, proton pump
inhibitor; UGIB, upper gastrointestinal bleeding.
feasible when LGIB is identified,7 but whether urgent colonos-
copy is associated with better outcome than a standard care
algorithm based on angiography and expectant colonoscopy
is unclear.8 Current guidelines for evaluation, monitoring, and
management of UGIB, variceal bleeding, and LGIB are pro-
vided in Table 23.19,10; Table 23.211; and Figure 23.1.12
Resuscitation, risk assessment, and pre-endoscopy management:
1. Urgent/emergent evaluation and initiation of resuscitation
2. Use appropriate prognostic scales for mortality and rebleeding
to stratify patients as low or high risk.
3. Consider nasogastric tube placement.
4. PRBC transfusion for Hgb 7 g/dL
5. Correct coagulopathy, but do not delay endoscopy.
6. Promotility agents are not recommended as routine for
pre-endoscopy.
7. Selected patients determined at low risk for rebleed after acute
ulcer bleeding may be considered for discharge after endoscopy.
8. Pre-endoscopic PPI therapy may be useful to downstage the
lesion and decrease need for endoscopic management, but its
administration should not delay endoscopic evaluation.
Pharmacologic management:
1. Histamine receptor antagonists are not recommended for
treatment in acute UGIB.
2. Somatostatin and octreotide are not recommended for
treatment in acute UGIB.
3. PPI infusion protocol (bolus plus continuous infusion) should be
given post–successful endoscopic treatment to all high-risk
lesion patients to reduce rebleeding and mortality.
4. Patients should be discharged on once-daily regimen of oral PPI
for a duration appropriate to the treated etiology.
Postdischarge ASA and NSAIDs:
1. For patients with history of ulcer bleeding who require chronic
treatment with NSAIDs, it must be recognized that whether
treatment is with traditional NSAID plus PPI or with COX-2
inhibitor alone, the risk for recurrent ulcer bleeding remains
clinically significant.
2. For patients with history of ulcer bleeding, the combination of
a COX-2 antagonist plus PPI is recommended to reduce risk of
rebleed compared with COX-2 agent alone.
3. For patients on low-dose ASA for cardiovascular risk who then
develop UGIB, the ASA therapy should be resumed as soon as
risk for cardiovascular complication is thought to outweigh risk
for rebleed.
4. In patients who require cardiovascular prophylaxis after ulcer
bleeding, it should be noted that clopidogrel alone has a higher
risk for rebleeding than ASA plus PPI.
 Section II—Clinical and Laboratory Assessment 229

Table 23.2  Summary of Recommendations for Acute Variceal Hemorrhage from the Practice
Parameters Committee of the American College of Gastroenterology
HEMORRHAGE FROM ESOPHAGEAL VARICES
1. Acute UGIB in patients with cirrhosis requires prompt attention with judicious intravascular volume support and blood
transfusion aimed at maintenance of Hgb at ≈8 g/dL.
2. Short, up to 1 week of antibiotic prophylaxis of SBP and other infections is recommended for all cirrhotic patients with
gastrointestinal hemorrhage: oral norfloxacin 400 mg bid or intravenous (IV) ciprofloxacin or IV ceftriaxone 1 g/day in advanced
cirrhotic patients or when quinolone resistance may be suspected.
3. Pharmacologic therapy using somatostatin or its analogs octreotide, vapreotide and terlipressin (latter not available in United
States) should be initiated immediately on suspicion of variceal bleeding and continued 3 to 5 days after diagnosis.
4. Endoscopy should be performed within 12 hours to diagnose variceal hemorrhage and treat by endoscopic variceal ligation (EVL)
or sclerotherapy.
5. Transjugular intrahepatic portosystemic shunt (TIPS) is indicated for patients with uncontrolled bleeding, or who have failed
combined pharmacologic and endoscopic treatment.
6. Balloon tamponade should be used as a temporizing measure maximally for 24 hours in patients with uncontrollable bleeding
while awaiting definitive endoscopic or TIPS treatment.
HEMORRHAGE FROM GASTRIC VARICES
1. In patients bleeding from gastric fundal varices, endoscopic variceal obturation using tissue adhesives such as cyanoacrylate is
preferred if available; EVL is an option.
2. TIPS is indicated for patients with uncontrolled bleeding or who have failed combined pharmacologic and endoscopic treatment.

Adapted from Garcia-Tsao G, Sanyal AJ, Grace ND, et al. Prevention and management of gastroesophageal varices and variceal hemorrhage in cirrhosis. Am J
Gastroenterol 2007;102:2086–102.
Hgb, Hemoglobin; SBP, spontaneous bacterial peritonitis; UGIB, upper gastrointestinal bleeding.

Bloody diarrhea
Hematochezia

Evaluate, resuscitate • Stool culture

consult GI endoscopist • Stool E. coli O157:H7
• Stool C. diff toxin
• Immunocompromised
consider CMV
NGT aspirate • Consider IBD
• Treat as appropriate
to identified cause
• If pain consider
ischemia

No blood Any blood

normal gastric/ or concern for UGIB, or
duodenal fluids hemodynamic instability

Bowel prep →
(–) (+)
Treated as
Colonoscopy EGD
appropriate

Source No source Technically

identified identified impossible

Treated as Continued Severe Arteriography

Fig. 23.1  Algorithm for lower gastrointestinal
appropriate bleeding Nuclear scanning
CT/CTA bleeding (LGIB). C. diff, Clostridium difficile; CMV,
Surgical consult cytomegalovirus; CT, computed tomography; CTA,
No or lesser Small bowel
bleeding suspected computed tomography angiography; E. coli,
Escherichia coli; EGD, esophagogastroduodenoscopy;
GI, gastrointestinal; IBD, inflammatory bowel disease;
Small Bowel Studies MRI, magnetic resonance imaging; NGT, nasogastric
Capsule enteroscopy tube; UGIB, upper gastrointestinal bleeding. (Modified
Push enteroscopy from Barnert J, Messmann H. Diagnosis and management of
Double balloon enteroscopy
lower gastrointestinal bleeding. Nat Rev Gastroenterol
CT or MRI enterography techniques
Hepatol 2009;6:637–46.)
230 Section II—Clinical and Laboratory Assessment
Hepatic Failure
Few patients with hepatic failure are cared for in the NCCU.
However, patients with hepatic encephalopathy and cerebral
edema frequently require the expertise of neurointensivists
and neurosurgeons and management in the NCCU, for
example, when therapeutic hypothermia is used for increased
intracranial pressure (ICP) associated with hepatic encepha-
lopathy.13,14 The pathology of hepatic encephalopathy is
incompletely understood but appears to be associated with
nitrogenous GI products and in particular ammonia that
bypass liver metabolism in acute liver failure (ALF). These
substances cross the blood-brain barrier, where they are
thought to increase gama-aminobutyric acid (GABAergic)
transmission and injure astrocytes that subsequently swell
(i.e., cytotoxic edema). For this reason, corticosteroids are of
no use. On the other hand, several lines of evidence suggest
that proinflammatory mechanisms are involved in the patho-
genesis of brain edema in ALF; this proposed mechanism
provides a basis for using induced hypothermia to bridge
patients with hepatic encephalopathy to transplant.15,16
ALF, also know as fulminant liver failure, is the abrupt loss
of liver function in a patient without previous liver disease. It
typically is associated with coagulopathy (international nor-
malized ratio [INR] >0.5) and encephalopathy. In the United
States, acetaminophen accounts for approximately 50% of
ALF cases: other causes include hepatitis, drug-induced liver
injury, and viral or autoimmune hepatitis. Initial evaluation
includes a liver function panel, prothrombin time [PT]/INR,
complete blood count, fibrinogen, D-dimer, acetaminophen
level, toxicology screen, electrolytes, and creatinine. In addi-
tion, a search for an etiology is necessary; this may include
alpha-fetoprotein, ceruloplasmin, serum protein electropho-
resis, virology (cytomegalovirus; Epstein-Barr virus; hepatitis
A, B, or C virus antigens; or antibodies), or antinuclear anti-
body among others. An abdominal computed tomography
(CT) scan to evaluate liver volume is useful, but if not feasible
bedside ultrasound may be obtained. A head CT scan is neces-
sary for patients with grade III or IV encephalopathy (see
following text). During an ICU stay, patient follow-up evalu-
ation may include arterial blood gas, arterial lactate, arterial
blood oxygen (ABO) analysis (two separate tests), and every-
6-hour PT/INR, transaminase level, total and direct bilirubin,
and serum sodium. Frequent assessment of serum osmolarity
or osmolality gap (when mannitol is used), and coagulation
status also is required.
There are two important approaches to classification of ALF
patients: (1) grading of hepatic encephalopathy and (2) deter-
mining whether a patient is eligible for transplant and whether
transplant is needed to prevent death. There are several clas-
sification scales for hepatic encephalopathy, but the West
Haven Simplified Criteria17 (Table 23.3) is used most fre-
quently. The severity of encephalopathy is the main barometer
of disease severity and grades III and IV often mark the
threshold for escalation of therapies. Determining who
requires a liver transplant to prevent death is complex and
requires consideration of ethical factors and an anticipation
of future needs to ensure that definitive therapy is available in
time to rescue the patient. The King’s College Criteria18 and
the Model for End-Stage Liver Disease (MELD)19 are widely
used scales to predict mortality in ALF and to help triage
patients with severe liver disease for transplant (Table 23.4).
Table 23.3  West Haven Criteria for
Semiquantitative Grading of Mental State
in Acute Liver Failure
GRADE I
• Trivial lack of awareness
• Euphoria or anxiety
• Shortened attention span
• Impaired performance of addition
GRADE II
• Lethargy or apathy
• Minimal disorientation for time or place
• Subtle personality change
• Inappropriate behavior
• Impaired performance of subtraction
GRADE III
• Somnolence to semistupor, but responsive to verbal stimuli
• Confusion
• Gross disorientation
GRADE IV
• Coma (unresponsive to verbal or noxious stimuli)
Adapted from Atterbury CE, Maddrey WC, Conn HO. Neomycin-sorbitol and
lactulose in the treatment of acute portal-systemic encephalopathy: a
controlled, double-blind clinical trial. Am J Dig Dis 1978;23:398–406.
The Acute Physiology and Chronic Health Evaluation
(APACHE) II score or Sequential Organ Failure Assessment
(SOFA) score also can help predict outcome in acetaminophen-
induced ALF.20,21 These scores (SOFA, MELD) and Simplified
Acute Physiology Score (SAPS) II also can be used to predict
outcome among patients with chronic liver insufficiency and
cirrhosis who require ICU admission.22 Some studies suggest
that ICU prognostic models such as SOFA may perform better
than liver specific scores such as Child-Turcotte-Pugh (CTP)
or MELD for outcome prediction among cirrhotic patients
who are admitted to the ICU.23
The average overall mortality in ALF is 80%. By contrast,
60% to 80% survive after transplant. Thus once a patient is
considered eligible for a transplant, the goals of care are to
provide a bridge until this occurs. In 2007, the U.S. Acute
Liver Failure Study Group (ALFSG) published guidelines to
codify ICU management of ALF patients in their participat-
ing centers.24 Key points associated with the neurologic
aspects and in particular patient monitoring of fulminant
liver failure is summarized later in this chapter.14 Other
aspects of ICU care—hemodynamic support, mechanical
ventilation, renal support, prevention of infection, and nutri-
tion and management of end-stage liver disease—are dis-
cussed elsewhere.25-27
Hepatic Encephalopathy
and Hyperammonemia
Hepatic encephalopathy (HE) is a feature of ALF. However,
patients with chronic liver failure who acutely deteriorate
(acute on chronic liver failure) may also experience HE.28 In
these patients HE manifests by confusion, poor judgment, or
personality change among other clinical features that may be
exacerbated by the disease that requires ICU admission or
the admission itself. The terminology associated with HE is

Table 23.4  Common Prognostic Scales for Mortality Risk Assessment in Acute Liver Failure
King’s College Criteria18
Non-APAP all cause prothrombin time (PT) >100 seconds
(INR >6.5)
Or
Any three of the following:
Age <10 or >40 years
Etiology: non-A, non-B hepatitis, halothane hepatitis,
idiosyncratic drug reactions
Duration of jaundice before onset of encephalopathy and
>7 days
PT >50 seconds
Serum bilirubin >18 mg/dL (all of scale is irrespective of
the grade of encephalopathy)
PPV mortality = 98%
NPV mortality = 82%
APAP, N-acetyl-p-aminophenol; INR, international normalized ratio; Ln, natural logarithm; NPV, negative predictive value; PPV, positive predictive value.
defined in a report of the Hepatic Encephalopathy Consensus
Group at the World Congress of Gastroenterology in 1998.29,30
Brain astrocytes internally detoxify ammonia to glutamine,
an osmotically active moiety. Although increased brain am­
monia and its detoxification product, glutamine, appear to
play a role in the pathogenesis of HE and brain edema patho-
genesis,31 the evidence for ammonia treatment is insufficient
for a high-level recommendation in consensus guidelines.
Similarly, the role of probiotics or rifaximin for HE is still
being elucidated.32,33 If ammonia-reducing agents are admin-
istered, lactulose is preferred. Neomycin also may be used but
generally is not recommended because its use is associated
with a risk for nephrotoxicity. Potential adverse effects of
inhibitors of brain glutamine synthesis (e.g., methionine sulf-
oximine) to prevent brain edema also have limited their use
in ALF. When lactulose is used, the following should be moni-
tored: (1) stool output and (2) the grade of HE and develop-
ment of harmful side effects including abdominal distention,
increased risk for aspiration, and dehydration. Interval assess-
ment of serum ammonia levels may aid in the optimal titra-
tion of the therapy, although the relationship between
ammonia levels and severity of HE across patients varies
because of differences in ammonia metabolism and analytic
methods. Newer methods such as quantification of ammo-
nium in the breath have been described.34
Seizures
Seizures, including subtle manifestations are common in ALF
patients with grade III or IV HE.35 Although the available data
are conflicting, they do not support routine phenytoin pro-
phylaxis. Sedation using benzodiazepines or propofol can
confer antiepileptic benefits. Electroencephalography (EEG) is
recommended for grade III or IV HE when there is: (1) an
unexplained episodic change in neurologic examination, (2)
myoclonus, and (3) a need to titrate barbiturate coma or guide
induced hypothermia to control increased ICP.36,37
Section II—Clinical and Laboratory Assessment 231
Model for End-Stage
Modified King’s College18 Liver Disease (MELD)19
Acetaminophen-induced disease MELD score =
Arterial pH <7.3 (irrespective of 3.78 [Ln serum bilirubin mg/dL]
grade of encephalopathy) +
Or 11.2 [Ln INR]
Grade III or IV encephalopathy +
and 9.57 [Ln serum creatinine mg/dL]
PT >100 seconds (INR >6.5) +
6.43
Serum creatinine >3.4 mg/dL
____________________________
Interpretation:
40 or more → 71.3% mortality
30-39 → 52.6% mortality
20-29 → 19.6% mortality
10-19 → 6.0% mortality
<9 → 1.9% mortality
PPV mortality = 84% 3-month mortality
NPV mortality = 86%
Cerebral Edema
Cerebral edema is most prominent in patients with the fastest
onset of ALF and is associated with increased morbidity and
mortality. Cerebral edema also may be observed in patients
with severe chronic liver disease or with portosystemic shunts.
For all HE grade III or IV patients and for those with an acute
change in mental status, a baseline brain CT is recommended.
A normal CT scan does not exclude the likelihood of increased
ICP. Intracranial hypertension is more likely in patients
with hyperacute and acute causes of ALF, grade III or IV HE,
those who develop pupillary abnormalities, seizures, systemic
inflammation, an arterial ammonia greater than 150 µmol/L,
hyponatremia, and those on vasopressor support.38 Whether
to place an ICP monitor is controversial largely because of
bleeding concerns and a lack of randomized trials to guide
clinicians. However, ICP monitoring can help guide manage-
ment of cerebral edema. The ALFSG group24 recommends an
ICP monitor be considered for grade III or IV HE patients
being bridged to liver transplant, and those with a potential
for survival without transplant who receive medical manage-
ment. Because the bleeding risks are 10% to 20%, mild HE
(grade I and II) patients should not have an invasive ICP
monitor. Subarachnoid and intraparenchymal fiber-optic ICP
monitors are preferred because the risk associated with ven-
triculostomy use is greater, whereas epidural monitors have
limited accuracy. Treatments to optimize coagulation should
be given immediately before the procedure. Alternatively non-
invasive ICP monitors may be considered or used to decide
about the value of invasive monitors (see Chapters 34 and 46).
The optimal ICP or cerebral perfusion pressure (CPP) asso-
ciated with increased survival and recovery in grade III or IV
HE is not known; however, recommended targets are to main-
tain ICP less than 25 and CPP from 50 to 80 mm Hg. Patient
volume status needs to be monitored to help maintain nor-
motension and to guide correction of hypovolemia where
present. Vasopressors may be used as a second-line therapy
when systolic blood pressure (SBP) is less than 90 mm Hg or
232 Section II—Clinical and Laboratory Assessment
mean arterial pressure (MAP) is less than 65 mm Hg to help
keep CPP between 50 and 80 mm Hg.
ICP management in HE is similar to that for other condi-
tions that cause increased ICP. The following general features
should be monitored: (1) head of bed position, (2) head in
line position, (3) level of agitation (see Chapter 11), (4)
PaCO2 (ideally 30-40 mm Hg, and (5) systemic temperature
to target normothermia and avoid hyperpyrexia. When ICP is
monitored, these variables can be optimized against the ICP
measurement. Cerebral hyperemia is considered to be one of
several factors associated with cerebral edema in ALF. The
role of therapies such as hyperventilation, N-acetylcysteine,
thiopentone sodium, and propofol still has to be fully
elucidated.39
The ALFSG recommends mannitol as the first-line osmotic
agent for elevated ICP in ALF. Whether a scheduled or
as-needed dose is preferred is not indicated but because of a
better side effect profile, lower bolus doses where effective
(e.g., 0.25-0.50 g/kg) are recommended. These bolus doses
should continue while ICP is greater than 25 provided serum
osomolarity is less than 320 mOsm/L; this should be checked
every 6 hours. The interval calculation of the osmolar gap may
help determine timing of subsequent doses to sustain a thera-
peutic osmotic effect. Hypertonic saline as a bolus treatment
or a continuous infusion also may be an option to help control
elevated ICP. When used, serum sodium should be checked
every 6 hours to target serum sodium of 145 to 155.
Second-line therapies for increased ICP include induced
hypothermia or barbiturate coma. Cooling does lower ICP in
HE and can help bridge a patient to liver transplant; an ICP
monitor should guide the depth of cooling.13 Temperature
needs to be closely monitored with a target temperature in the
mild to moderate hypothermia range, but not less than 32° C.
Modern indwelling catheter-based systems or surface cooling
methods that have tight thermostatic regulation in a feedback
loop from a reliable core temperature sensor work best. Older
devices are more difficult to regulate and often yield undesir-
ably large temperature fluctuations. When induced hypother-
mia is used, electrolytes and coagulation parameters should
be assessed frequently and urine output and hydration status
must be closely monitored because “cold diuresis” can occur.
Initiation and maintenance of barbiturate coma require
intensive hemodynamic monitoring to titrate therapies for
hemodynamic stability. Continuous EEG looking for a burst
suppression pattern helps to titrate the barbiturate dose. For
example, a progressive increase in the time between bursts
may indicate a higher than necessary dose and thus a need to
reduce the infusion rate of a barbiturate maintenance dose.
Daily serum levels of pentobarbital should be followed.
However, the adequacy of therapy is demonstrated by success-
ful reduction of ICP as indicated by an ICP monitor.
Pancreatitis
Pancreatitis, usually in its milder forms, occasionally may
complicate the course of patients under care for primary
neurologic illness. Guidelines provide an overview of the dis-
tribution of severity in acute pancreatitis.40 Simple interstitial
pancreatitis accounts for 85% of cases, whereas 15% are nec-
rotizing and about a third of these develop infected necrosis.
One in 10 patients with interstitial pancreatitis develops any
organ failure, which tends to be reversible. In necrotizing
Table 23.5  Ranson Criteria in Acute
Pancreatitis
Prognosis for
At Admission Initial 48 Hours Mortality
Age >55 years Fall in Hct >10% <3 signs = 1%
WBC >16,000/mm3 Calcium <8 mg/dL 3-4 signs = 16%
Glucose >200 mg/dL PaO2 <60 mm Hg 5-6 signs = 40%
LDH >350 IU/L Base deficit >4 mEq/L >7 signs = 100%
SGOT >250 SF units BUN increase by
>5 mg/dL
Third space losses >6 L
From Ranson JH, Rifkind KM, Roses DF, et al. Prognostic signs and the
role of operative management in acute pancreatitis. Surgery, Gynecology &
Obstetrics 1974;139(1):69-81
BUN, Blood urea nitrogen; Hct, hematocrit; LDH, lactate dehydrogenase;
PaO2, arterial oxygen tension; SGOT, serum glutamic-oxaloacetic transaminase;
WBC, white blood cell.
pancreatitis the median prevalence for organ failure is about
50%. Overall mortality for pancreatitis is 5% with a range
between 3% for interstitial and 17% for necrotizing pancre-
atitis. In the absence of organ failure, mortality is zero but
increases to 3% for single-organ failure and up to 50% for
multiple-organ failure.41
In patients with pancreatitis initial laboratory evaluation
should include complete blood count (CBC) with differential,
complete serum chemistry including calcium, liver function,
lactate dehydrogenase, amylase, lipase, and triglycerides. The
Ranson Criteria, particularly those based on 48-hour labora-
tory data can be used to estimate mortality risk and for risk
stratification (Table 23.5). In addition, APACHE II scores are
useful: scores that increase during the first 48 hours are associ-
ated with worse course, whereas improving scores are associ-
ated with a benign course. Hemoconcentration at admission
may be a poor prognostic indicator and particularly if the
hematocrit then increases. In the absence of hemoconcentra-
tion, pancreatic necrosis is unlikely. Serum C-reactive protein
(CRP) levels tend to peak at 36 to 72 hours, and when greater
than 150 mg/L may suggest pancreatic necrosis.
Except where there is need to evaluate for alternative diag-
nostic considerations, a CT of the abdomen is more useful
when obtained after a delay of 2 to 4 days than at admission.
By that time a more severe clinical course has declared itself,
and the CT then can distinguish between simple interstitial
disease or the presence of necrosis and microcirculation dis-
ruption. This distinction requires contrast administration,
and the study is often called a dynamic CT of the pancreas or
contrast-enhanced multidetector CT (MDCT). The timing of
this study should be considered carefully to avoid contrast
with a rising serum creatinine and evolving renal failure.
When findings of necrotizing pancreatitis are present, inter-
mittent serial CT examinations may be required to monitor
for secondary intra-abdominal complications, pseudocyst
development, organized necrosis, and vascular complications
such as thrombosis or pseudoaneurysms. However, the cumu-
lative dose of radiation with serial CT scans needs to be con-
sidered.42 The Balthazar CT criteria for acute pancreatitis43
divide the severity of pancreatitis into five grades, A through
E: normal, enlarged pancreas, inflamed and or peripancreatic
fat, single fluid collection, and two or more fluid collections.
Grade A indicates the least and grade E the most severe form
 Section II—Clinical and Laboratory Assessment 233
of pancreatitis. Most patients with severe pancreatitis have one
or several pancreatic fluid collections (grades D and E).
Increasing grade is associated with increasing morbidity and
mortality. The CT severity grade is an attempt to improve the
prognostic accuracy of CT. In each grade patients are assigned
points for the amount of necrosis; higher numerical scores
indicate greater extent of necrosis and worse prognosis.44,45 In
general, CT grading scales more accurately diagnose clinically
severe disease and better identify pancreatic infection or the
need for intervention than do clinical criteria.46
Patients with more severe pancreatitis or those with altered
vital signs, diminishing urine output (UOP), increasing
volume requirements, tachypnea or tachycardia, encephalopa-
thy, or any indication of early organ failure require ICU care.
Initial care focuses on adequate fluid resuscitation and avoid-
ance of hypoxemia. In more severe cases third spacing can be
profound, and aggressive fluid support may be required.
Emesis, diaphoresis, and vascular permeability all contribute
to the volume challenges, and the development of hypovole-
mia increases the risk of a necrotizing process if the microcir-
culation is inadequately supplied. Fluid status therefore should
be monitored closely using continuous monitoring of blood
pressure, heart rate, and urine output. In more severe cases,
management may be best guided by CVP and ScvO2 monitor-
ing and noninvasive monitoring techniques of cardiac func-
tion that provide data on cardiac performance, preload, and
systemic vascular resistance (see Chapter 19).
Infected pancreatic necrosis occurs in about one third of
patients with necrotizing pancreatitis and generally develops
after the first week. Clinically, patients with or without infected
necrotizing pancreatitis appear similar. An abdominal CT may
help make the diagnosis because air bubbles may be observed
in the retroperitoneum with infection but not in sterile necro-
sis. Diagnosis is made by CT-guided percutaneous aspiration
that if negative may be repeated every 5 to 7 days when sys-
temic toxicity persists. Surgical debridement, often repeated,
is the gold standard for treatment.
Abdominal Compartment Syndrome
Increased abdominal pressure may arise in many disorders,
but some more common conditions include cirrhosis with
ascites, following major trauma, pancreatitis, after massive
resuscitation or massive transfusion, sepsis, acidosis, or respi-
ratory failure that requires high positive end-expiratory pres-
sure (PEEP) levels for treatment. A high index of clinical
suspicion in the appropriate circumstances, and willingness to
initiate IAP monitoring, usually by bladder pressure measure-
ment, is needed to diagnose dangerous, but often externally
occult, elevations in intra-abdominal pressure (IAP).47,48
Failure to recognize this condition and failure to monitor it
can compromise respiratory function and intra-abdominal
organ function and lead to multiorgan failure and death.49
IAP is the steady-state pressure inside the abdominal cavity.
It is measured in millimeters of mercury (mm Hg). The usual
method to measure IAP is through fluid column pressure
transduction from the port of a Foley catheter into which
25 mL of sterile saline are instilled into the bladder with the
Foley catheter clamped just distal to the measurement port.
Steady-state pressure is obtained at end expiration in a relaxed
patient with the transducer zeroed at the midaxillary line.
Continuous measurement protocols are described, but
intermittent pressure transduction—every hour in the at-risk
population—is used most frequently.50,51
Abdominal perfusion pressure (APP) is described by:
APP = MAP − IAP
Abdominal compartment syndrome and organ function
compromise occur when IAP approaches MAP and visceral
perfusion is compromised. The optimal target APP is greater
than or equal to 60 mm Hg. Normal IAP in critically ill
patients is thought to be 5 to 7 mm Hg. It may be near zero
in the healthy population. IAP is graded according to sever-
ity50,51: grade I, 12-15 mm Hg; grade II, 16 to 20 mm Hg;
grade III, 21 to 25 mm Hg; grade IV, greater than 25 mm Hg.
There is no single pressure threshold at which reductions in
microcirculatory blood flow sufficient to initiate organ dys-
function, are certain to occur. Oliguria is one of the first
outward signs of organ compromise.52-54 Critical IAP levels
seem to begin in the range of 10 to 15 mm Hg. Abdominal
compartment syndrome therefore should not defined by a
single number but rather as sustained IAP greater than
20 mm Hg (with or without APP <60 mm Hg) that is associ-
ated with new organ dysfunction or failure.50 Management of
intra-abdominal hypertension (abdominal compartment syn-
drome) includes: (1) correction of positive fluid balance, (2)
evacuation of intraluminal contents, (3) percutaneous drain-
age of peritoneal free fluid, (4) improvement of abdominal
wall compliance such as prone position and alternating (asyn-
chronous) ventilation, or (5) decompression of the abdomen
by laparotomy.55,56
Conclusion
Gastrointestinal (GI) disorders, GI bleeding, hepatic dysfunc-
tion, pancreatitis, and intra-abdominal compartment syn-
drome can develop in patients in the NCCU. In addition,
some patients with hepatic encephalopathy require care in the
NCCU to manage increased ICP (e.g., using therapeutic
hypothermia).13,14 The neurointensivist therefore should have
knowledge of these disorders and how to care for the various
conditions. A number of grading scales (e.g., the West Haven
Criteria,17 MELD criteria,19 or Ranson Criteria)57 among
others, that are specific to GI, liver, and pancreatic disease or
general ICU scales (e.g., APACHE, SOFA, or SAPs)20-22 can
help guide care. In addition, groups such as the U.S. Acute
Liver Failure Study Group (ALFSG) and the Practice Param-
eters Committee of the American College of Gastroenterology
have published guidelines to codify ICU management of
patients with GI, hepatic, or pancreatic disease.9-12,24
References
1. Ali T, Harty RF. Stress-induced ulcer bleeding in critically ill patients.
Gastroenterol Clin North Am 2009;38(2):245–65.
2. Vincent JL, Rhodes A, Perel A, et al. Clinical review: update on
hemodynamic monitoring: a consensus of 16. Crit Care 2011;15(4):229.
3. Barkun A, Bardou M, Marshall JK, et al. Consensus recommendations for
managing patients with nonvariceal upper gastrointestinal bleeding. Ann
Intern Med 2003;139(10):843–57.
4. Greenspoon J, Barkun A, Bardou M, et al. Management of patients with
nonvariceal upper gastrointestinal bleeding. Clin Gastroenterol Hepatol
2012;10(3):234–9. Epub 2011, Aug 4.
5. Stanley AJ, Ashley D, Dalton HR, et al. Outpatient management of patients
with low-risk upper-gastrointestinal haemorrhage: multi-centre validation
and prospective evaluation. Lancet 2009;373:42–7.
234 Section II—Clinical and Laboratory Assessment
6. Packham CJ, Rockall TA, Logan RF. Outpatient care for selected patients
with acute upper gastrointestinal bleeding. Lancet 1995;345:659–60.
7. Kim BC, Cheon JH, Kim TI, et al. Risk factors and the role of bedside
colonoscopy for lower gastrointestinal hemorrhage in critically ill patients.
Hepatogastroenterology 2008;55(88):2108–11.
8. Green BT, Rockey DC, Portwood G, et al. Urgent colonoscopy for evaluation
and management of acute lower gastrointestinal hemorrhage: a randomized
controlled trial. Am J Gastroenterol 2005;100:2395–402.
9. Barkun AN, Bardou M, Kuipers EJ, et al. International consensus
recommendations on the management of patients with nonvariceal upper
gastrointestinal bleeding. Ann Intern Med 2010;152:101–13.
10. Barkun A, Bardou M, Marshall JK, et al. Consensus recommendations for
managing patients with nonvariceal upper gastrointestinal bleeding. Ann
Intern Med 2003;139:843–57.
11. Garcia-Tsao G, Sanyal AJ, Grace ND, et al. Prevention and management of
gastroesophageal varices and variceal hemorrhage in cirrhosis. Am J
Gastroenterol 2007;102:2086–102.
12. Barnert J, Messmann H. Diagnosis and management of lower
gastrointestinal bleeding. Nat Rev Gastroenterol Hepatol 2009;6:637–46.
13. Stravitz RT, Larsen FS. Therapeutic hypothermia for acute liver failure. Crit
Care Med 2009;37(7 Suppl):S258–64.
14. Frontera JA, Kalb T. Neurological management of fulminant hepatic failure.
Neurocrit Care 2011;14(2):318–27.
15. Jiang W, Desjardins P, Butterworth RF. Direct evidence for central
proinflammatory mechanisms in rats with experimental acute liver failure:
protective effect of hypothermia. J Cereb Blood Flow Metab 2009;29(5):
944–52. Epub 2009, Mar 4.
16. Stravitz RT, Lee WM, Kramer AH, et al. Therapeutic hypothermia for acute
liver failure: toward a randomized, controlled trial in patients with advanced
hepatic encephalopathy. Neurocrit Care 2008;9(1):90–6.
17. Atterbury CE, Maddrey WC, Conn HO. Neomycin-sorbitol and lactulose in
the treatment of acute portal-systemic encephalopathy: a controlled,
double-blind clinical trial. Am J Dig Dis 1978;23:398–406.
18. O’Grady JG, Alexander GJ, Hayllar KM, et al. Early indicators of prognosis
in fulminant hepatic failure. Gastroenterology 1989;97:439–55.
19. Weisner R, Edwards E, Freeman R, et al. Model for end-stage liver disease
(MELD) and allocation of donor livers. Gastroenterology 2003;124:91–6.
20. Larson AM, Polson J, Fontana RJ, et al. Acetaminophen-induced acute liver
failure: results of a United States multicenter, prospective study. Hepatology
2005;42:1364–72.
21. Craig DG, Reid TW, Wright EC, et al. The sequential organ failure
assessment (SOFA) score is prognostically superior to the model for
end-stage liver disease (MELD) and MELD variants following paracetamol
(acetaminophen) overdose. Aliment Pharmacol Ther 2012;35(6):705–13.
doi: 10.1111/j.1365-2036.2012.04996.x. Epub 2012, Jan 20.
22. Levesque E, Hoti E, Azoulay D, et al. Prospective evaluation of the prognostic
scores for cirrhotic patients admitted to an intensive care unit. J Hepatol
2012;56(1):95–102. Epub 2011, Aug 9.
23. Cholongitas E, Senzolo M, Patch D, et al. Review article: scoring systems for
assessing prognosis in critically ill adult cirrhotics. Aliment Pharmacol Ther
2006 1;24(3):453–64.
24. Stravitz RT, Kramer AH, Davern T, et al. Intensive care of patients with acute
liver failure: recommendations of the U.S. Acute Liver Failure Study Group.
Crit Care Med 2007;35:2498–508.
25. Findlay JY, Fix OK, Paugam-Burtz C, et al. Critical care of the end-stage liver
disease patient awaiting liver transplantation. Liver Transpl 2011;17(5):
496–510. doi: 10.1002/lt.22269.
26. Al-Khafaji A, Huang DT. Critical care management of patients with
end-stage liver disease. Crit Care Med 2011;39(5):1157–66.
27. Bémeur C, Desjardins P, Butterworth RF. Role of nutrition in the
management of hepatic encephalopathy in end-stage liver failure. J Nutr
Metab 2010;489823. Epub 2010, Dec 22.
28. Wlodzimirow KA, Eslami S, Abu-Hanna A, et al. A systematic review on
prognostic indicators of acute on chronic liver failure and their predictive
value for mortality. Liver Int 2012 Mar 19. Epub ahead of print.
29. Ferenci P, Lockwood A, Mullen K, et al. Hepatic encephalopathy: definition,
nomenclature, diagnosis, and quantification: final report of the working
party at the 11th World Congresses of Gastroenterology, Vienna, 1998.
Hepatology 2002;35(3):716–21.
30. Mullen KD, Prakash RK. New perspectives in hepatic encephalopathy. Clin
Liver Dis 2012;16(1):1–5. Epub 2012, Jan 16.
31. Desjardins P, Du T, Jiang W, et al. Pathogenesis of hepatic encephalopathy
and brain edema in acute liver failure: role of glutamine redefined.
Neurochem Int 2012 Feb 21. Epub ahead of print.
32. McGee RG, Bakens A, Wiley K, et al. Probiotics for patients with hepatic
encephalopathy. Cochrane Database Syst Rev 2011;11:CD008716.
33. Bass NM, Mullen KD, Sanyal A, et al. Rifaximin treatment in hepatic
encephalopathy. N Engl J Med 2010;362(12):1071–81.
34. Blanco Vela CI, Bosques Padilla FJ. Determination of ammonia
concentrations in cirrhosis patients: still confusing after all these years? Ann
Hepatol 2011;10(Suppl 2):S60–5.
35. Ellis AJ, Wendon JA, Williams R: Subclinical seizure activity and prophylactic
phenytoin infusion in acute liver failure: a controlled clinical trial.
Hepatology 2000;32:536–41.
36. Kaplan PW, Rossetti AO. EEG patterns and imaging correlations in
encephalopathy: encephalopathy part II. J Clin Neurophysiol
2011;28(3):233–51.
37. Marchetti P, D’Avanzo C, Orsato R, et al. Electroencephalography in
patients with cirrhosis. Gastroenterology 2011;141(5):1680–9.e1-2. Epub
2011, Jul 18.
38. Shawcross DL, Wendon JA. The neurological manifestations of acute liver
failure. Neurochem Int 2011 Nov 2. Epub ahead of print.
39. Mpabanzi L, Jalan R. Neurological complications of acute liver failure:
pathophysiological basis of current management and emerging therapies.
Neurochem Int 2011 Nov 13. Epub ahead of print.
40. Banks PA, Freeman ML, the Practice Guidelines Committee of the American
College of Gastroenterology. Practice guidelines in acute pancreatitis. Am J
Gastroenterol 2006;101:2379–400.
41. Vege SS, Gardner TB, Chari ST, et al. Low mortality and high morbidity in
severe acute pancreatitis without organ failure: a case for revising the Atlanta
classification to include “moderately severe acute pancreatitis.” Am J
Gastroenterol 2009;104(3):710–5. Epub 2009, Feb 3.
42. Ball CG, Correa-Gallego C, Howard TJ, et al. Radiation dose from computed
tomography in patients with necrotizing pancreatitis: how much is too
much? J Gastrointest Surg 2010;14(10):1529–35. Epub 2010, Sep 8.
43. Balthazar EJ, Ranson JHC, Naidich DP, et al. Acute pancreatitis: prognostic
value of CT. Radiology 1985;156:767–72.
44. Balthazar EJ, Robinson DL, Megibow AJ, et al. Acute pancreatitis: value of
CT in establishing prognosis. Radiology 1990;174:331–6.
45. Balthazar EJ. Acute pancreatitis: assessment of severity with clinical and CT
evaluation. Radiology 2002;223(3):603–13. Review.
46. Bollen TL, Singh VK, Maurer R, et al. Comparative evaluation of the
modified CT severity index and CT severity index in assessing severity of
acute pancreatitis. AJR Am J Roentgenol 2011;197(2):386–92.
47. Malbrain ML, Chiumello D, Pelosi P, et al. Incidence and prognosis
of intraabdominal hypertension in a mixed population of critically ill
patients: a multiple-center epidemiological study. Crit Care Med 2005;33:
315–22.
48. Malbrain ML. Abdominal pressure in the critically ill: measurement and
clinical relevance. Intensive Care Med 1999;25:1453–8.
49. Hedenstierna G, Larsson A. Influence of abdominal pressure on respiratory
and abdominal organ function. Curr Opin Crit Care 2012;18(1):80–5.
Review.
50. Malbrain MLNG, Cheatham ML, Kirkpatrick A, et al. Results from the
international conference of experts on intra-abdominal hypertension and
compartment syndrome. I Definitions. Intensive Care Med 2006;32:
1722–32.
A complete list of references for this chapter can be found online at
www.expertconsult.com.
 Section II—Clinical and Laboratory Assessment 234.e1
References
1. Ali T, Harty RF. Stress-induced ulcer bleeding in critically ill patients.
Gastroenterol Clin North Am 2009;38(2):245–65.
2. Vincent JL, Rhodes A, Perel A, et al. Clinical review: update on
hemodynamic monitoring: a consensus of 16. Crit Care 2011;15(4):229.
3. Barkun A, Bardou M, Marshall JK, et al. Consensus recommendations for
managing patients with nonvariceal upper gastrointestinal bleeding. Ann
Intern Med 2003;139(10):843–57.
4. Greenspoon J, Barkun A, Bardou M, et al. Management of patients with
nonvariceal upper gastrointestinal bleeding. Clin Gastroenterol Hepatol
2012;10(3):234–9. Epub 2011, Aug 4.
5. Stanley AJ, Ashley D, Dalton HR, et al. Outpatient management of patients
with low-risk upper-gastrointestinal haemorrhage: multi-centre validation
and prospective evaluation. Lancet 2009;373:42–7.
6. Packham CJ, Rockall TA, Logan RF. Outpatient care for selected patients
with acute upper gastrointestinal bleeding. Lancet 1995;345:659–60.
7. Kim BC, Cheon JH, Kim TI, et al. Risk factors and the role of bedside
colonoscopy for lower gastrointestinal hemorrhage in critically ill patients.
Hepatogastroenterology 2008;55(88):2108–11.
8. Green BT, Rockey DC, Portwood G, et al. Urgent colonoscopy for evaluation
and management of acute lower gastrointestinal hemorrhage: a randomized
controlled trial. Am J Gastroenterol 2005;100:2395–402.
9. Barkun AN, Bardou M, Kuipers EJ, et al. International consensus
recommendations on the management of patients with nonvariceal upper
gastrointestinal bleeding. Ann Intern Med 2010;152:101–13.
10. Barkun A, Bardou M, Marshall JK, et al. Consensus recommendations for
managing patients with nonvariceal upper gastrointestinal bleeding. Ann
Intern Med 2003;139:843–57.
11. Garcia-Tsao G, Sanyal AJ, Grace ND, et al. Prevention and management of
gastroesophageal varices and variceal hemorrhage in cirrhosis. Am J
Gastroenterol 2007;102:2086–102.
12. Barnert J, Messmann H. Diagnosis and management of lower
gastrointestinal bleeding. Nat Rev Gastroenterol Hepatol 2009;6:637–46.
13. Stravitz RT, Larsen FS. Therapeutic hypothermia for acute liver failure. Crit
Care Med 2009;37(7 Suppl):S258–64.
14. Frontera JA, Kalb T. Neurological management of fulminant hepatic failure.
Neurocrit Care 2011;14(2):318–27.
15. Jiang W, Desjardins P, Butterworth RF. Direct evidence for central
proinflammatory mechanisms in rats with experimental acute liver failure:
protective effect of hypothermia. J Cereb Blood Flow Metab 2009;29(5):
944–52. Epub 2009, Mar 4.
16. Stravitz RT, Lee WM, Kramer AH, et al. Therapeutic hypothermia for acute
liver failure: toward a randomized, controlled trial in patients with advanced
hepatic encephalopathy. Neurocrit Care 2008;9(1):90–6.
17. Atterbury CE, Maddrey WC, Conn HO. Neomycin-sorbitol and lactulose in
the treatment of acute portal-systemic encephalopathy: a controlled,
double-blind clinical trial. Am J Dig Dis 1978;23:398–406.
18. O’Grady JG, Alexander GJ, Hayllar KM, et al. Early indicators of prognosis
in fulminant hepatic failure. Gastroenterology 1989;97:439–55.
19. Weisner R, Edwards E, Freeman R, et al. Model for end-stage liver disease
(MELD) and allocation of donor livers. Gastroenterology 2003;124:91–6.
20. Larson AM, Polson J, Fontana RJ, et al. Acetaminophen-induced acute liver
failure: results of a United States multicenter, prospective study. Hepatology
2005;42:1364–72.
21. Craig DG, Reid TW, Wright EC, et al The sequential organ failure assessment
(SOFA) score is prognostically superior to the model for end-stage liver
disease (MELD) and MELD variants following paracetamol
(acetaminophen) overdose. Aliment Pharmacol Ther 2012;35(6):705–13. doi:
10.1111/j.1365-2036.2012.04996.x. Epub 2012, Jan 20.
22. Levesque E, Hoti E, Azoulay D, et al. Prospective evaluation of the prognostic
scores for cirrhotic patients admitted to an intensive care unit. J Hepatol
2012;56(1):95–102. Epub 2011, Aug 9.
23. Cholongitas E, Senzolo M, Patch D, et al. Review article: scoring systems for
assessing prognosis in critically ill adult cirrhotics. Aliment Pharmacol Ther
2006 1;24(3):453–64.
24. Stravitz RT, Kramer AH, Davern T, et al. Intensive care of patients with acute
liver failure: recommendations of the U.S. Acute Liver Failure Study Group.
Crit Care Med 2007;35:2498–508.
25. Findlay JY, Fix OK, Paugam-Burtz C, et al. Critical care of the end-stage liver
disease patient awaiting liver transplantation. Liver Transpl 2011;17(5):
496–510. doi: 10.1002/lt.22269.
26. Al-Khafaji A, Huang DT. Critical care management of patients with
end-stage liver disease. Crit Care Med 2011;39(5):1157–66.
27. Bémeur C, Desjardins P, Butterworth RF. Role of nutrition in the
management of hepatic encephalopathy in end-stage liver failure. J Nutr
Metab 2010;489823. Epub 2010, Dec 22.
28. Wlodzimirow KA, Eslami S, Abu-Hanna A, et al. A systematic review on
prognostic indicators of acute on chronic liver failure and their predictive
value for mortality. Liver Int 2012 Mar 19. Epub ahead of print.
29. Ferenci P, Lockwood A, Mullen K, et al. Hepatic encephalopathy: definition,
nomenclature, diagnosis, and quantification: final report of the working
party at the 11th World Congresses of Gastroenterology, Vienna, 1998.
Hepatology 2002;35(3):716–21.
30. Mullen KD, Prakash RK. New perspectives in hepatic encephalopathy. Clin
Liver Dis 2012;16(1):1–5. Epub 2012, Jan 16.
31. Desjardins P, Du T, Jiang W, et al. Pathogenesis of hepatic encephalopathy
and brain edema in acute liver failure: role of glutamine redefined.
Neurochem Int 2012. Epub ahead of print.
32. McGee RG, Bakens A, Wiley K, et al. Probiotics for patients with hepatic
encephalopathy. Cochrane Database Syst Rev 2011;11:CD008716.
33. Bass NM, Mullen KD, Sanyal A, et al. Rifaximin treatment in hepatic
encephalopathy. N Engl J Med 2010;362(12):1071–81.
34. Blanco Vela CI, Bosques Padilla FJ. Determination of ammonia
concentrations in cirrhosis patients: still confusing after all these years? Ann
Hepatol 2011;10(Suppl 2):S60–5.
35. Ellis AJ, Wendon JA, Williams R: Subclinical seizure activity and prophylactic
phenytoin infusion in acute liver failure: a controlled clinical trial.
Hepatology 2000;32:536–41.
36. Kaplan PW, Rossetti AO. EEG patterns and imaging correlations in
encephalopathy: encephalopathy part II. J Clin Neurophysiol
2011;28(3):233–51.
37. Marchetti P, D’Avanzo C, Orsato R, et al. Electroencephalography in patients
with cirrhosis. Gastroenterology 2011;141(5):1680–9.e1-2. Epub 2011, Jul 18.
38. Shawcross DL, Wendon JA. The neurological manifestations of acute liver
failure. Neurochem Int 2011 Nov 2. Epub ahead of print.
39. Mpabanzi L, Jalan R. Neurological complications of acute liver failure:
pathophysiological basis of current management and emerging therapies.
Neurochem Int 2011 Nov 2. Epub ahead of print.
40. Banks PA, Freeman ML, the Practice Guidelines Committee of the American
College of Gastroenterology. Practice guidelines in acute pancreatitis. Am J
Gastroenterol 2006;101:2379–400.
41. Vege SS, Gardner TB, Chari ST, et al. Low mortality and high morbidity in
severe acute pancreatitis without organ failure: a case for revising the Atlanta
classification to include “moderately severe acute pancreatitis.” Am J
Gastroenterol 2009;104(3):710–5. Epub 2009, Feb 3.
42. Ball CG, Correa-Gallego C, Howard TJ, et al. Radiation dose from computed
tomography in patients with necrotizing pancreatitis: how much is too
much? J Gastrointest Surg 2010;14(10):1529–35. Epub 2010, Sep 8.
43. Balthazar EJ, Ranson JHC, Naidich DP, et al. Acute pancreatitis: prognostic
value of CT. Radiology 1985;156:767–72.
44. Balthazar EJ, Robinson DL, Megibow AJ, et al. Acute pancreatitis: value of
CT in establishing prognosis. Radiology 1990;174:331–6.
45. Balthazar EJ. Acute pancreatitis: assessment of severity with clinical and CT
evaluation. Radiology 2002;223(3):603–13. Review.
46. Bollen TL, Singh VK, Maurer R, et al. Comparative evaluation of the
modified CT severity index and CT severity index in assessing severity of
acute pancreatitis. AJR Am J Roentgenol 2011;197(2):386–92.
47. Malbrain ML, Chiumello D, Pelosi P, et al. Incidence and prognosis of
intraabdominal hypertension in a mixed population of critically ill patients:
a multiple-center epidemiological study. Crit Care Med 2005;33:315–22.
48. Malbrain ML. Abdominal pressure in the critically ill: measurement and
clinical relevance. Intensive Care Med 1999;25:1453–8.
49. Hedenstierna G, Larsson A. Influence of abdominal pressure on respiratory
and abdominal organ function. Curr Opin Crit Care 2012;18(1):80–5.
Review.
50. Malbrain MLNG, Cheatham ML, Kirkpatrick A, et al. Results from the
international conference of experts on intra-abdominal hypertension and
compartment syndrome. I Definitions. Intensive Care Med 2006;32:1722–32.
51. Malbrain MLNG, Cheatham ML, Kirkpatrick A, et al. Results from the
international conference of experts on intra-abdominal hypertension and
compartment syndrome. II. Recommendations. Intensive Care Med
2007;33:951–62.
52. Sugrue M, Jones F, Janjua KJ, et al. Temporary abdominal closure: a
prospective evaluation of its effects on renal and respiratory physiology.
J Trauma 1998;45:914–21.
53. Sugrue M, Jones F, Deane SA, et al. Intra-abdominal hypertension is an
independent cause of postoperative renal impairment. Arch Surg
1999;134:1082–5.
234.e2 Section II—Clinical and Laboratory Assessment
54. Malbrain MLNG, Deeren DH, De Potter T, et al. Abdominal compartment
syndrome following rectus sheath hematoma: bladder-to-gastric pressure
difference as a guide to treatment. ANZ J Surg 2005;75:A8.
55. Anand RJ, Ivatury RR. Surgical management of intra-abdominal
hypertension and abdominal compartment syndrome. Am Surg
2011;77(Suppl 1):S42–5.
56. De Keulenaer BL, De Waele JJ, Malbrain ML. Nonoperative management of
intra-abdominal hypertension and abdominal compartment syndrome:
evolving concepts. Am Surg 2011;77(Suppl 1):S34–41.
57. Ranson JH, Rifkind KM, Roses DF, et al. Prognostic signs and the role of
operative management in acute pancreatitis. Surgery, Gynecology &
Obstetrics 1974;139(1):69–81.
III
Chapter
24  
Evoked Potentials
Emmanuel Carrera, Ronald G. Emerson, and Jan Claassen
Introduction
Neuromonitoring can help determine the extent of neurologic
injury, can detect secondary injury at a potentially reversible
stage, can help assess the effect of clinical care on an individual
patient, and may aid in predicting outcome. This can be par-
ticularly important in the neurocritical care unit (NCCU),
where the neurologic examination may be difficult because of
depressed consciousness, hypothermia, or use of sedatives.
Somatosensory evoked potentials (SSEPs) have been used in
clinical practice since they were first described in the late 1940s
in patients with myoclonus,1 and after electroencephalogra-
phy (EEG) are the most frequently used neurophysiologic tool
in the NCCU. In general, SSEPs are less affected by sedation
or hypothermia than EEG.2 Brainstem auditory evoked poten-
tials (BAEPs) have been used since the 1960s to study nonin-
vasively and objectively the function of the brainstem auditory
pathways.3 Other tools such as motor evoked potentials and
visual evoked potentials are less frequently used in the NCCU.
Use of evoked potential (EP) monitoring in neurocritical care
practice varies regionally and even between intensive care
units (ICUs), and the ready availability of neuroimaging has
limited the application of EPs in many ICUs. However, the
diagnostic and prognostic value of EPs is embraced by national
and international organizations.4 Advantages of EPs include:
(1) they are noninvasive, (2) they may be used serially, (3) they
give objective quantitative values that can be followed over
time and compared between patients, (4) they are relatively
stable in the presence of mild hypothermia and relatively resis-
tant to many commonly used sedatives, and (5) they are inex-
pensive.5 In addition, modern invasive neuromonitoring and
treatment devices such as intravascular cooling catheters can
limit patient transport or use of magnetic resonance imaging
(MRI). In the absence of portable imaging devices, diagnostic
tools such as EPs that can be brought to the patient’s bedside
provide an alternative diagnostic tool. The technical feasibility
of continuous EP monitoring along with continuous EEG
monitoring has been demonstrated.6
Engineering and Technology
Somtatosensory Evoked Potentials
SSEP testing assesses the integrity of the dorsal column–
lemniscal system.5 This pathway projects via the dorsal column
of the spinal cord to the cuneate nucleus in the lower
236
brainstem, to the ventroposterior lateral thalamus, to the
primary somatosensory receiving cortex, and then to a wide
network of cortical areas involved in somatosensory process-
ing (secondary somatosensory cortex, posterior parietal
cortex, posterior and mid-insula and mid-cingulate cortex).
Median and tibial nerves are most often stimulated in SSEP
testing, although others (e.g., ulnar, peroneal) may be used
when appropriate.
Stimulation
The stimulus for SSEPs is a brief electrical pulse delivered by
a pair of electrodes placed on the skin above the nerve. The
median nerve is stimulated at the lateral-anterior wrist; the
tibial nerve is stimulated at the ankle, with the cathode
between the Achilles tendon and the medial malleolus. To
reduce artifact produced by electrical stimulation, a ground
electrode is placed between the stimulation site and the
recording site. The stimulus is delivered as a constant mono-
phasic square-wave pulse that lasts approximately 100 to
200 msec, at a rate of typically 4 per second. The intensity of
the stimulation is two to four times the sensory threshold as
evidenced by the contraction of the muscle innervated by the
stimulated nerve.
Recording
Both standard surface disk electrodes and needle electrodes
can be used. Recording electrodes for upper limb SSEPs are
placed at the clavicle between the heads of the sternocleido-
mastoid muscles (Erb’s point) and on the skin overlying cer-
vical bodies 6 to 7. On the scalp, electrodes are placed at
CP3 and CP4 (respectively midway between C3 and C4) of
the international 10-20 system. Bipolar recording between
CP3 and CP4 detects the cortical component of the SSEP;
referential recording, to earlobe or noncephalic locations
(e.g., the shoulder), is used to detect subcortial (brainstem)
components. For tibial SSEPs, electrodes are placed in the
popliteal fossa (peripheral component) and over the lumbar
vertebra (lumbar component). At least two bipolar channels
(e.g., CPz-Fpz and CP3-CP4) (see Chapter 25 for a review of
EEG montage nomenclature) should be used to record the
cortical component; as with the median SSEPs, referential
scalp recording is used to detect subcortical components.
The SSEP waveform is obtained by averaging typically from
© Copyright 2013 Elsevier Inc. All rights reserved.
 Section III—Electrophysiology 237

500 to 2000 stimuli. It is is necessary to repeat at least two

independent averages to demonstrate reproducibility. For Table 24.1  Normal Values for Evoked
clinical purposes, analysis times of 50 msec (median nerve) Potentials from Healthy Volunteers
or 100 msec (tibial nerve) are used. Latency, Latency,
Mean Upper Limit
Recording Site (msec) (msec)
Filters
MEDIAN SSEP
SSEPs are recorded using a broad pass-band with high-pass
N9 Erb’s point 9.8 11.5
and low-pass filters set typically to 30 Hz and 2000 Hz,
respectively. Notch filters to eliminate power line noise (50 or N13 Cervical spine (C7) 13.3 14.5
60 Hz) should be used with caution because of their ten- N20 Contralateral cortex 19.8 23.0
dency to create “ringing” oscillatory artifact. Digital postpro- (CP3 or CP4)
cessing of the recorded signal may be appropriate in specific INTERVALS MEDIAN
situations. CCT (P14-N20) — 5.6 6.6
Tibial SSEP
Brainstem Auditory Evoked Potentials N8 Popliteal fossa 8.5 10.5
BAEPs are generated using brief auditory stimuli that activate N22 Lumbar spine (L1) 21.8 25.2
the cochlea, the auditory nerve, and then the brainstem audi-
P30 Fz-Cv7 29.2 34.7
tory pathways.7
P39 Cz-Fz 38.0 43.9
INTERVALS TIBIAL
Stimulation
N22-P30 — 7.4 10.2
The stimulus used to generate a BAEP consists of a brief (less
P30-P39 — 8.7 13.4
than 200 μsec) “click,” typically delived at a rate of approxi-
mately 10 Hz. The stimuli are presented to one ear at a time CCT, Central conduction time; SSEP, somatosensory evoked potential.
using headphones or ear inserts while the nonstimualted ear
is “masked” with white hissing noise to prevent stimulation by
sound conducted through the cranium.

indentified initially based on the effects of clinical lesions and

Recording
Either standard surface electrodes or needle electrodes may be
used. BAEPs are typically recorded using a two-channel
montage: ipsilateral ear to vertex (channel 1) and contralateral
ear to vertex (channel 2). Averaging of 2000 to 4000 stimuli is
typically required; at least two independent averages must be
recorded to prove reproducibility of waveforms.
Filters
BAEPs are recorded using a broad pass-band, with the high-
pass filter typically set to 100 Hz or 150 Hz and the low-pass
filter set to 3000 Hz.
Information Provided and What
the Information Means
Somtatosensory Evoked Potentials
SSEP recordings consist of signals recorded from different
sites along the afferent sensory pathways. By convention, nega-
tive singals are displayed as upward deflections, and positive
signals downward. Physiologic signals are recorded with dif-
ferential amplifiers, devices that amplify the difference between
two inputs. A negative signal in the first, or “noninverting”
input produces an upward inflection; a negative signal in the
second, or “inverting” input produces a downward inflection.
The time between stimulation and the recorded potential is
called the latency, and is usually expressed in milliseconds. A
normal median SSEP waveform is shown in Figure 24.1.
Generator sources for specific components of the SSEP were
direct recording from intracranial structures; more recently,
source modeling based on three-dimensional MRI has been
used to revisit these generators.8
Clinical interpretation is based primarily on the analysis of
“short latency” SSEP signals, up to and including those cor-
responding to the activation of the somatosensory receiving
cortex (Table 24.1). The N9 is a compound action potential
reflecting the peripheral afferent volley and is used as a refer-
ence point from which to calculate conduction time for
the centrally recorded potentials. Peripheral neuropathy may
increase the latency and decrease the amplitude of this wave-
form. N13 is generated by dorsal horn cells of the cervical
spinal cord. P14 is generated in the caudal medial lemniscus
within the lower medulla, with possible additional presynaptic
contribution from the upper cervical cord. N18 is attributed
to postsynaptic activity in multiple brainstem gray matter
structures. N20 reflects activation of the primary somatosen-
sory receiving area, located in the posterior bank of the
rolandic fissure in Brodmann’s area 3b.9,10 Although actually
measuring a composite of white and gray matter events in the
large fiber sensory system between the cervicomedullary junc-
tion and somatosensory cortex, the P14 to N20 interpeak
latency (calculated by subtracting the P14 peak latency
from the N20 peak latency) is commonly referred to as the
central conduction time (CCT).5 Absent or delayed central
conduction with present peripheral conduction may permit
lesions to be localized by SSEPs.
Principal components of the tibial SSEP include the N8
recorded from the popliteal fossa, the N22 generated by the
dorsal gray matter of the lumbosacral cord, the P30 generated
at the cervicomedullary junction, and the P39 generated in the
238 Section III—Electrophysiology

N20

N18

P14

5 msec

Fig. 24.1  Normal median nerve somatosensory evoked potential (SSEP) recording. The four anatomical cross-sections (from bottom to top
of illustration) are as follows: cervical spinal cord, brainstem, midbrain, and brain (cerebral cortex, coronal section).

somatosensory cortex.5 In contrast to the median nerve SSEP, I III

the P39 cortical signal9 is positive in polarity and variably
maximal in amplitude either near the vertex or over the cen-
troparietal scalp ipsilateral to the stimulated leg. This interin- V
dividual variability in the scalp topography of P39, due to the A2-Cz
variations in the normal location of the leg and foot area
within the interhemispheric fissure, necessitates use of at least 0.2 uV
two bipolar scalp-scalp channels, one in the midline and
another using laterally placed electrodes, to ensure reliable 1 msec
detection.
Middle- and long-latency SSEP components are not com- A1-Cz
monly used clinically in the ICU because they are less reliable
and more susceptible than short-latency SSEPs to the effects
of medication and to depressed levels of consciousness.5

Fig. 24.2  Normal brainstem auditory evoked potential (BAEP) bilaterally

Brainstem Auditory Evoked Potentials
Clinical interpretation of the BAEP is based on signals gener-
ated within approximately 7 milliseconds following the stim-
ulus (Fig. 24.2). Studies suggest that BAEPs have multiple
generators, with nonsequential interaction between second-
and third-order ipsilateral and contralateral neuronal cells
and axons. Although the exact identity of short-latency BAEP
waveforms remains somewhat uncertain, commonly recog-
nized generators include the distal auditory nerve (wave I),
the auditory nerve as it exits the porus acousticus or the
cochlear nucleus (wave II), the cochlear nucleus or ipsilateral
superior olivary nucleus (wave III), the superior olivary
(click 90 db, noise 60 db; recorded from the right ear) in a 49-year-old
woman with a right frontal brain tumor.
nucleus or axons of the lateral lemniscus (wave IV), and the
inferior colliculus and ventral lateral lemniscus (wave V).11
Because waves II and IV are less reliably recorded across indi-
viduals, clinical interpretation is based primarily on assess-
ment of waves I, III, and V (see Fig. 24.2).
BAEPs are resistant to the effects of sedative medication and
to a patient’s level of consciousness depression, adding to their

utility in the ICU setting. Although the amplitude and latency
of these waves are affected substantially by click stimulus
intensity, the I to V interpeak latency (as well as I to III and
III to V interpeak latencies) is relatively unaffected by intensity
changes and interpeak latency and is therefore relied on for
clinical interpretion.
Delays in wave V with a normal wave I latency suggest
a conduction abnormality somewhere central to the distal
portion of cranial nerve VIII. Absence of wave I with pre-
served wave V could reflect a problem with the peripheral
hearing apparatus or with the auditory nerve, but commonly
reflects technical difficulty recording wave I. Functional dis-
ruption of the brainstem may cause loss of waves II to V with
preservation of wave I.12 Unilaterally abnormal BAEPs most
often reflect ipsilateral brainstem damage.
More rostral structures, including generators within the
primary auditory cortex and surrounding areas, are respon-
sible for longer latency components.13 However, as with SSEPs,
the later components of the BAEP are affected by conscious-
ness state and attention,14 and are less commonly used clini-
cally in the ICU setting.
Indications for Use
In the ICU, EPs are primarily used for prognostication, par-
ticularly in patients after cardiac arrest and those with trau-
matic brain injury (TBI). Diagnostically, EPs have largely been
replaced by imaging studies (see Chapters 26 and 28), invasive
neuromonitoring (e.g., brain oxygen monitoring), and con-
tinuous EEG (see Chapter 25). However, serial and even con-
tinuous EP monitoring is feasible2,6 and may be considered
when continuous noninvasive neuromonitoring is warranted
and EEG may be of limited value, such as in evolving spinal
cord or brainstem injury. In addition EPs may be used to
complement or supplement the information provided by
imaging or other monitors. Consensus recommendations to
guide optimal use of electrophysiologic tests including SSEP
for diagnosis, follow-up, and prognosis in the ICU have been
published.15
CP3-CP4
CP4-chin
SC5-FPZ
Right Erbs-chin
0.7 uV
A 5 msec
Fig. 24.3  Somatosensory evoked potentials (SSEPs) after cardiac arrest in a 65-year-old-man with about 15 minutes of pulseless electrical
activity who underwent hypothermia. The tracings show SSEPs recorded after (A) right and (B) left median nerve stimulation. There are bilaterally
present peripheral and cervical and subcortical median nerve SSEP potentials and absent cortical responses.
Section III—Electrophysiology 239
Postanoxic Coma (Hypoxic-Ischemic
Encephalopathy)
After cardiopulmonary resuscitation, the neurologic examina-
tion does not reliably predict the presence or absence of
specific SSEP patterns, particularly absent or present N20
responses.16 Nevertheless, EPs can help predict outcome,
particularly poor outcome rather than good outcome, after
cardiac arrest. Normal SSEP responses have less predictive
power than abnormal or absent responses. It does not appear
that induced hypothermia after cardiac arrest (e.g., 33° C for
24 hours) affects the predictive power of SSEPs, although
withdrawal of care may bias these observations.17,18 In particu-
lar, bilaterally absent cortical SSEP responses 3 days after
resuscitation are associated with a poor prognosis.19-30 Short-
latency SSEPs (N20) may be the most reliable method to
predict poor outcome in anoxic-ischemic encephalopathy.
This is included in practice parameters on outcome prediction
in comatose patients after cardiopulmonary resuscitation
from the Quality Standards Subcommittee of the American
Academy of Neurology and regarded as Level II evidence (i.e.,
probably disease specific).27 For example, in a prospective
study of 407 cardiac arrest patients, Zandbergen et al. observed
bilaterally absent N20s in 45% of patients that were comatose
at 72 hours; all these had poor outcome19 (Fig. 24.3). In a
subsequent systematic literature review that included 4500
anoxic patients, Zandbergen et al. observed that bilaterally
absent N20s within the first week were 100% specific for
poor outcome.27 However, among those patients with poor
outcome, 46% may have preserved N20 responses within the
first week of cardiac arrest28 (Fig. 24.4). SSEPs perform favor-
ably when compared with other outcome predictors such as
the pupillary light response, the Glasgow Coma Scale (GCS),
or EEG.26 Combining SSEPs with other outcome predictors
such as EEG or serum markers of neuronal injury provides
high prognostic accuracy.24-26,31 Although SSEPs may allow the
clinician to predict poor prognosis with some degree of accu-
racy, good outcome is much more difficult to predict. Discrep-
ancies between SSEP findings and outcome (i.e., recovery of
CP4-CP3
CP3-chin
SC5-FPZ
0.7 uV
5 msec
Left Erbs-chin
B
240 Section III—Electrophysiology
C4'-C3'
C3'-right Erbs
0.7 uV
5 msec
Left-right Erbs
A B
Fig. 24.4  Somatosensory evoked potentials (SSEPs) in a 59-year-old man with multiple cardiac arrests and possible hypoxic brain injury.
Right (A) and left (B) median nerve SSEPs show present cortical responses and mildly prolonged central conduction times bilaterally.
SSEPs but absence of clinical improvement)19,32 may be
explained by hypoxic injury sparing the somatosensory tracts,
or secondary brain injury acquired after the SSEPs were
obtained (i.e., recurring cardiac arrests or cardiac output
failure).
Several other SSEP parameters including the CCT, the N20-
to-P25 amplitude ratio,26,33 and the N70 latency34 are associ-
ated with outcome but are less robust than absent cortical
SSEP responses. When the N20s are present or questionable,
late SSEP components such as the N70 (absent or delayed
>130 msec) can help identify patients likely to have a poor
outcome.34 However, false positives are common (between 4%
and 15%) and make it impractical to base treatment decisions
on this finding. Late components may be better associated
with long-term cognitive outcome than short-latency compo-
nents35 but are not routinely used for patients with cardiac
arrest. The role of BAEPs after cardiac arrest is less well studied.
However, middle-latency auditory evoked responses are absent
in patients who die or remain in a persistent vegetative state.22
Finally, EPs can be used to guide care in neonatal encepha-
lopathy or childhood asphyxia.36
Traumatic Brain Injury
Following TBI, SSEPs can help predict short-term mortality
and long-term outcome.34,35,37-44 Bilaterally absent cortical
SSEP responses with preserved peripheral and spinal poten-
tials are associated with poor outcome.37-40 In general, SSEPs
more accurately predict poor outcome than good outcome.41
For example, in a review of 44 studies, Carter and Butt41
observed that 71% of patients with normal SSEPs (N = 553)
have a favorable outcome (normal or moderate disability;
Fig. 24.5). By contrast when SSEPs are absent bilaterally (n =
777), 99% have an unfavorable outcome (severe disability,
vegetative state, or death). The positive predictive value and
sensitivity are 71% and 59%, respectively, for normal SSEP
and a favorable outcome; and 99% and 46.2%, respectively,
for bilaterally absent SSEP and an unfavorable outcome.
Patients who have a favorable outcome despite bilaterally
absent SSEPs (N = 12) are more commonly children, have
C3'-C4'
C4'-left Erbs
0.7 uV
5 msec
Right-left Erbs
focal lesions, have subdural or extradural fluid collections, or
have undergone decompressive craniotomy in the 48 hours
before the recording. Overall, SSEPs perform favorably when
compared with other TBI outcome predictors such as the
GCS, pupillary or motor responses, computed tomography
(CT), and EEG findings,42 and the the false-positive rate of
bilaterally absent SSEPs after TBI is less than 0.5%.
Repeated SSEPs can help detect secondary injury (e.g.,
hematoma enlargement or increased intracranial presure
[ICP]), brainstem herniation, or cerebral ischemia.38,45,46 In
some patients a change in SSEPs may precede an ICP increase.46
Serial SSEPs also may suggest recovery6,43 (Fig. 24.6). For
example, reduction of latency and normalization of amplitude
may occur earlier than recovery in the clinical examination.43
Poor long-term functional outcome (death or vegetative state
at 2 years) is seen in comatose patients with traumatic brain-
stem lesions and absent N20s that do not recover within 48
hours of the injury. However, recovery of the N20 does not
necessarily mean recovery of brain function.47 Prolongation
or absent CCT is associated with poor 1-year cognitive and
behavioral function.48
Although rare, reappearance of previously absent bilateral
SSEPs may occur after TBI. This is described in some patients
with good outcome after infratentorial hemorrhage or follow-
ing hemicraniectomy.49-51 In addition, circumscribed bilateral
sensory tract contusions in the brainstem may mimic bilateral
SSEP loss from supratentorial injury. This may not necessarily
mean a fatal prognosis.50 Finally, marked attenuation of the
cortical N20-P25 must be distinguished from persistent loss
of these signals.43 BAEPs are used less frequently than SSEPs
for TBI outcome prediction,52-54 but when performed, BAEP
abnormalities are more frequently seen in TBI patients with
poor outcome.52,55
Acute Ischemic Stroke
Diagnosis of acute ischemic stroke (AIS), is done by neuro-
logic examination and neuroimaging. EPs, however, may be
used in a patient with infratentorial ischemia if MRI is not
available or in a patient with a pacemaker who cannot have
 Section III—Electrophysiology 241

A B

C3-C4
C4-C3
C4-left Erbs

C4-right Erbs C3-left Erbs

C3-right Erbs
0.5 uV CS6-left Erbs

5 msec
CS5-right Erbs
Right-left Erbs

0.5 uV
Left-right Erbs
5 msec
C D
Fig. 24.5  Somatosensory evoked potential (SSEPs) in a comatose 76-year-old man with a traumatic right subdural hematoma (A) and contralateral
parietal contusion (B). Symmetric bilateral median nerve cortical and subcortial SSEPs are present. The left median nerve SSEP (C) (Erb’s 11 msec, P14
15 msec, N20 22 msec) and the right median nerve SSEP (D) (Erb’s 11 msec, P14 16 msec, N20 23 msec) are present.

an MRI, and can be used for continuous monitoring in select
patients at risk for herniation (e.g., a large cerebellar stroke).
Ischemia of the pontine tegmentum results in both BAEP (loss
of wave V) and SSEP abnormalities (absent N20). Ischemia
restricted to the cerebellar peduncles only affects the BAEPs.56
False-negative EP results may occur—brainstem strokes may
have normal SSEPs with medial or lateral strokes in up to 60%
and 75% patients, respectively.57 EPs may be useful in patients
with locked-in syndrome to provide objective evidence of
brainstem involvement although there is no specific pattern
to the abnormalities,58,59 and depending on the affected struc-
tures, BAEP and SSEP findings may vary from unilaterally
normal to bilaterally absent.58-60 Finally, BAEP obtained within
24 hours of large middle cerebral artery infarction may help
identify patients at risk for malignant edema61 and thus the
need for a hemicraniectomy.
The role of EPs in outcome prediction after AIS is less well
studied than TBI or anoxia and is mainly limited to case series
of patients with posterior fossa or massive hemispheric infarc-
tion. Bilateral BAEP and SSEP abnormalities are seen in
patients with poor outcome after pontine and mesencephalic
infarction,62 and those with basilar thrombosis63 and can be
particularly helpful when making decisions about prognosis
when consciousness is impaired. Good outcome (moderate
disability or better) is more likely when BAEPs and SSEPs are
normal.63 EP abnormalities are of indeterminate predictive
value in patients with locked-in syndrome.60 Cerebellar
infarcts associated with coma may be accompanied by prolon-
gation of the I to V interpeak latencies,64 although this abnor-
mality may persist despite clinical improvement.
There are conflicting data on the association between EP
findings and outcome after supratentorial AIS. On the one
hand, normal BAEPs but not SSEPs obtained before a hemi-
craniectomy are associated with survival and good functional
outcome (Barthel Index ≥60) after large hemispheric strokes.65
On the other hand, SSEP findings after small internal capsule
or corona radiate strokes66 do not correlate with 3-month
outcome (Barthel Index). Others have found that serial SSEPs
and BAEPs are associated with 30-day Glasgow Outcome
Scale independent of stroke type and location, with the
exception that BAEPs are not associated with outcome in
those with infratentorial lesions.67 In part these differences in
242 Section III—Electrophysiology

Vertebra 7 Left cortical response Vertebra 7 Right cortical response

Day 1 Day 1
N13 N13
N20

1 µV
1 µV P25

Day 10 Day 10
N13 N13
N20

P25
N20 P25

Day 304 Day 304

N13 N13

N20 N20

P25

10 20 30 40 msec 10 20 30 40 msec 10 20 30 40 msec 10 20 30 40 msec

Fig. 24.6  Recovery of somatosensory evoked potential (SSEP) after traumatic brain injury (TBI). This 14-year-old boy was admitted in coma
(Glasgow Coma Scale score 3). He developed elevated intracranial pressure, and a clinical improvement was first seen on day 13 after the injury. He
did not move his left side purposefully until day 26. The initial median nerve SSEP is abnormal. Left cortical recordings were absent, and right
hemispheric responses were prolonged with a severely decreased amplitude. For the left cortical projections, the first indication of recovery was
observed with SSEPs on day 10. Over the next few days, these SSEP findings improved. Twenty-three days after injury, latencies of the right cortical
projection were within normal limits, whereas amplitudes remained reduced. One year after the accident, he had a minimal residual speech impairment
but had returned to his former activities. (Reprinted with permission from Claassen J, Hansen HC. Early recovery after closed traumatic head injury: somatosensory
evoked potentials and clinical findings. Crit Care Med 2001;29:494–502.)

results may stem from differences in study design and small
sample size.
Several EP observations are associated with stroke location.
AIS of the ventroposterior thalamus leads to an absent
response, decrease in amplitude, delay in peak latency, or
attenuation of median N20-P25 and tibial P40 components.68
Lateral ventroposterior lesions preferentially affect tibial
SSEPs, whereas medial lesions affect median SSEPs. Strokes
that affect the thalamocortical pathways lead to less consistent
SSEP abnormalities. Although large strokes of the corona
radiate may lead to similar SSEP abnormalities as thalamic
strokes,68 smaller lesions of the corona radiate or internal
capsule often have an inconsistent relationship between EP
findings and sensory symptoms.66 This inconsistency is likely
from incomplete involvement of the thalamocortical pathways
in small strokes.
Intracerebral Hemorrhage
Evoked potentials (EPs) have little if any role in the diagno-
sis of an intracerebral hemorrhage (ICH). In addition EP
localization value is lim­ited after thalamic or putaminal
ICH.69 However, BAEP abnormalities may provide informa-
tion about small posterior circulation bleeds, particularly
when tegmental structures are affected.70 Serial SSEPs may
provide an alternative to frequent brain imaging to examine
potential for neurologic recovery, particularly for ICHs in
the posterior circulation. Recovery of motor function after
ICH in the putamen or thalamus generally parallels the nor-
malization of SSEP components.69
SSEPs and BAEPs can provide prognostic information in
patients with an ICH located in the brainstem, particularly in
the pons70,71 or in deep supratentorial structures including the
putamen72,73 and thalamus.72 In particular, prolongation of I
to V interpeak latencies in the BAEP and bilaterally absent
N20s in the SSEP are are associated with poor prognosis.70,74
SSEPs may have a lower false-negative rate for poor outcome
than BAEPs after putaminal ICHs.73 Good functional outcome
is likely when BAEPs are normal and SSEPs are normal after
putaminal ICH.72 Following a pontine ICH, good outcome
may be seen if normal amplitudes and latencies of waves I to
V in BAEPs are at least unilaterally present.70
Subarachnoid Hemorrhage
There is an unclear relationship between the severity of sub-
arachnoid hemorrhage (SAH) and EP findings; some studies
suggest an association between CCT prolongation75 or absence
of SSEP and BAEP responses76 and the Hunt Hess grade,
whereas others found no association between CCT abnor-
malities and clinical severity.77,78 Xenon single photon emis-
sion computed tomography (SPECT)79,80 studies suggest CCT
prolongation is associated with impaired cerebral blood flow
(CBF) after SAH, particularly when CBF is less than
30 mL/100 g/min.81 However, EP studies do not have a role in
vasospasm diagnosis in clinical practice.

Studies in the 1970s did not find an association between EP
findings and outcome.82 Subsequent studies describe an asso-
ciation between BAEP and SSEP abnormalities including
bilateral CCT prolongation (>6 msec) or bilaterally absent
cortical responses and poor outcome or death after SAH.77,83,84
Tibial rather than median nerve SSEP may be more robust in
outcome prediction.84 Patients with normal SSEPs have the
potential for a good outcome.77 These differences may be asso-
ciated with advances in SAH management particularly for
poor-grade SAH patients.
Spinal Cord Injury
EPs have an established role during spinal surgery but are not
frequently used to monitor patients with spinal cord injury
(SCI) in the ICU.85 Although EP abnormalities after SCI,
including spinal cord ischemia, are associated with long-term
outcome,86-88 they do not appear to add any prognostic accu-
racy to the neurologic examination but can be used to supple-
ment it.86,89,90 SSEPs may have a role in injury assessment for
comatose patients with suspected SCI.88 In these patients the
extent of neurologic deficits is associated with the degree of
SSEP abnormalities: N9 and N20 latencies for the upper limbs
and P40 and P60 for the lower limbs.91
Brain Death
There is no single method to confirm brain death, the diag-
nosis of which depends in large part on the clinical examina-
tion (see Chapter 13) and strict protocols. However, Wijdicks
in a review conducted a decade ago, found there was a lack of
consensus on the exact criteria in many published practice
parameters and guidelines and that 28 of 70 (40%) of the
practice guidelines he studied mandated the use of confirma-
tory testing. Many of these recommended use of EPs.4 When
repeat studies are performed, patients who progress to brain
death lose subcortical SSEP and all BAEP responses.12,92 Loss
of the median nerve SSEP noncephalic P14 and of its cephalic
referenced reflection N14 and the N18 is seen in brain
death.92,93 No specific BAEP pattern is seen in patients who
progress to brain death.94,95 Diagnostic accuracy is improved
by combining BAEPs and SSEPs.96 EPs can also be helpful in
patients when there is doubt about the clinical findings or
cardiopulmonary instability limits the ability to perform an
apnea test.
Coma Associated with Metabolic
or Toxic Encephalopathy
The International Society for Hepatic Encephalopathy and
Nitrogen Metabolism (ISHEN) published guidelines for the
use of neurophysiologic investigations in hepatic encepha-
lopathy.97 There are three main recommendations. First, the
choice of neurophysiologic test depends in part on the severity
of encephalopathy. Quantitative EEG may be used for mild
forms, but SSEPs should be considered for more severe
encephalopathy. Second, when N20 responses are bilaterally
absent, further evaluation is recommended to determine the
extent of irreversible brain injury before liver transplantation.
Third, SSEPs and BAEPs may be used to help make decisions
about ICP and its management, for example, when an ICP
monitor is not feasible or to complement the information of
Section III—Electrophysiology 243
an ICP monitor (i.e., determining the consequence of an ICP
increase).46
Outside the Intensive Care Unit
A large body of literature describes the use of intraoperative
neurophysiologic monitoring including use of SSEPs, BAEPs,
transcranial motor evoked potentials (tcMEPs), and direct
electromygraphy, and evidence-based guidelines were pub-
lished on the use of intraoperative monitoring.98 The reader
is referred to reviews on the topic.99-101 SSEPs can: (1) help
identify cerebral ischemia during cerebrovascular surgery
including carotid endarterectomy or aneurysm surgery,102 or
during repair of the thoracic aorta,103 (2) identify spinal cord
injury during spine surgery,104 or (3) help guide resection of
epileptic foci,105 tumors,106 or arteriovenous malformations
(AVMs)107 in select cortical areas or intramedullary spinal
lesions,108 whereas BAEPs can be useful during posterior fossa
surgery.109 In general, neurologic abnormalities such as para-
plegia after spinal surgery never or very rarely occur if there
are no changes in EPs during intraoperative monitoring. EPs
also are used in studies of cognition and behavior and to
evaluate conditions such as multiple sclerosis, Friedreich’s
ataxia, and Alzheimer’s disease.
How to Use Evoked Potential
Monitoring, and Troubleshooting
Electrical Artifact
Electrical artifact (“noise”) may frequently interfere with EP
recordings in the ICU setting. This is of particular concern in
SSEP recordings because notch filters are discouraged. The
noise level influences the accuracy of the recording and thus
interobserver agreement. This can limit the use of SSEPs in
prognosis decisions.110 Therefore every effort should be made
to minimize the artifact level (<0.25 microV). If this is not
possible, studies should be interpreted with caution. Neuro-
muscular junction blockade also may be considered when
noise is excessive.111
Medication Effects
Subcortical EP components are relatively unaffected by most
medications that are used in the ICU. By contrast, cortical EP
components are more susceptible to medication effects. Seda-
tive medications, including benzodiazepines, propofol, barbi-
turates, nitrous oxide, and halogenated inhalational agents all
depress the cortical components of EPs in a dose-related
manner.112-115 However, in these circumstances cortical EP
components are more robust than the EEG, and consequently
SSEPs may be used to monitor patients with barbiturate-
induced coma and an isoelectric EEG.114
A severe overdose with central nervous system (CNS)
depressants (e.g., amitriptyline, barbiturates, meprobamate,
or nitrazepam) may affect SSEP and BAEP latencies. These
abnormalities may include CCT prolongation and N20 dis-
persion of the SSEP, and interpeak delays of the BAEP.35
Pheny­toin may increase latencies and decrease amplitudes of
brainstem BAEP components and abnormal BAEPs, includ-
ing increasing latency of wave I, and I to V and I to III inter-
peak latencies may be observed in up to a third of patients
244 Section III—Electrophysiology
taking phenytoin, particularly with supratherapeutic phenyt-
oin levels.116 In phenytoin intoxication, all waves except BAEP
waves I and V may be lost.117
Hypothermia-Induced Changes
Therapeutic hypothermia is increasingly used in the ICU, and
so the effects of temperature on EP studies require careful
consideration. Much of the knowledge about EPs during
hypothermia comes from its use during cardiac and neurovas-
cular surgery. However, reports are variable and depend in
part on what was measured and at what temperature. There is
a linear relationship between decreasing temperature and
increasing cortical and peripheral SSEP latencies (N10, P14,
and N19).118 The N19 is consistently recordable at tempera-
tures greater than 26° C, usually disappearing between 20° C
and 25° C. In general the more peripherally generated compo-
nents, N10 and P14, are more robust to the effect of hypother-
mia. P14 is elicitable at more than 21° C, whereas N10 persists
even when the temperature is less than 20° C. Subcortical
SSEP components (P13 and N14) become more consistently
recognized at profound hypothermia (<20° C) compared with
normothermia, whereas cortical components disappear.119
Hypothermia increases BAEP latency of wave I interpeak
intervals.113 There is a linear increase of latencies and inter-
peak intervals by approximately 7% per 1° C decrease in tem-
perature.120 At 26° C, BAEP values are approximately doubled.
These temperature-induced changes are reversible with
rewarming, but during the rewarming phase latencies may
differ compared with the cooling phase (“hysteresis”).118 BAEP
amplitudes may initially increase when temperature is between
25° C and 30° C and then decrease at lower temperatures.120-122
Longer latency components may disappear with moderate
degrees of hypothermia, but ultimately all components disap-
pear when temperature is less than 14° C to 20° C.120,123
Induced hypothermia is commonly used after cardiac
arrest. When SSEP and BAEPs are studied between 24 and 28
hours after the arrest and hypothermia (33° C),17 established
EP parameters associated with outcome (i.e., poor outcome
with bilaterally absent N20 in SSEPs) remain reliable. BAEPs
had no additional value in outcome prediction.
Risks
BAEP and SSEP studies are safe. However, several ICU-
specific considerations about electrical safety should be kept
in mind.124 In particular, ICU patients may be at increased
risk for electrical injury, in particular cardiac arrhythmias.
Fluid paths, especially central lines, may allow currents to
enter a patient’s body. Weak or comatose patients are unable
to withdraw from current sources. To reduce these risks, all
electrical equipment used in the patient area should be the
three-wire grounded type with “hospital grade” plugs. Equip-
ment should be tested on a regular schedule for excessive
current leakage and electrical outlets tested for integrity of the
ground connection. Devices connected to one patient should
be plugged into the same group of outlets. Equipment or
outlets that show any sign of malfunction or damage should
not be used. Likewise patient ground connections should
never be improvised; patient connections on clinical equip-
ment use special isolators or current limiters to reduce the
risk of shock.
Conclusion
Clinical neurophysiology including SSEPs and BAEPs can be
useful in the NCCU for diagnosis (e.g., brain death or neuro-
muscular disorders), for follow-up, and as a guide to manage-
ment (e.g., hepatic encephalopathy), and in outcome prediction
(e.g., anoxic-ischemic encephalopathy) after cardiac arrest,
TBI, and toxic and metabolic encephalopathy. In general, the
information provided by EPs is better able to predict poor
outcome than good outcome, and the significance of the find-
ings depends in part on the patient’s pathology.
Acknowledgments
We thank Noeleen Ostapkovich for her careful review of this manuscript
and Edward Gallo, REEGT, for help with reproducing figures.
References
1. Dawson G. Investigations on a patient subject to myoclonic seizure after
stimulation. J Neurol Neurosurg Psychiatry 1947;10:141–62.
2. Moulton RJ, Brown JI, Konasiewicz SJ. Monitoring severe head injury: a
comparison of EEG and somatosensory evoked potentials. Can J Neurol Sci
1998;25:S7–11.
3. Celesia G, Puletti F. Auditory cortical areas of man. Neurology
1969;19:211–20.
4. Wijdicks EF. Brain death worldwide: accepted fact but no global consensus
in diagnostic criteria. Neurology 2002;58:20–5.
5. Cruccu G, Aminoff MJ, Curio G, et al. Recommendations for the clinical use
of somatosensory-evoked potentials. Clin Neurophysiol 2008;119:1705–19.
6. Fossi S, Amantini A, Grippo A, et al. Continuous EEG-SEP monitoring of
severely brain injured patients in NICU: methods and feasibility.
Neurophysiol Clin 2006;36:195–205.
7. Celesia GC, Brigell MG. Auditory evoked potentials. In: Niedermeyer E,
Lopes da Silva F, editors. Electroencephalography: basic principles, clinical
applications, and related fields. Baltimore: Lippincott Williams & Wilkins;
2005. p. 1137–64.
8. Mauguiere F. Somatosensory evoked potentials: normal responses, abnormal
waveforms, and clinical applications in neurological diseases. In:
Niedermeyer E, Lopes da Silva F, editors. Electroencephalography: basic
applications, and related fields. Baltimore: Lippincott Williams & Wilkins;
2005. p. 1067–119.
9. Goff GD, Matsumiya Y, Allison T, et al. The scalp topography of human
somatosensory and auditory evoked potentials. Electroencephalogr Clin
Neurophysiol 1977;42:57–76.
10. Desmedt JE, Ozaki I. SEPs to finger joint input lack the N20-P20 response
that is evoked by tactile inputs: contrast between cortical generators in areas
3b and 2 in humans. Electroencephalogr Clin Neurophysiol 1991;80:513–21.
11. Scherg M, von Cramon D. A new interpretation of the generators of BAEP
waves I-V: results of a spatio-temporal dipole model. Electroencephalogr
Clin Neurophysiol 1985;62:290–9.
12. Starr A. Auditory brain-stem responses in brain death. Brain
1976;99:543–54.
13. Yvert B, Crouzeix A, Bertrand O, et al. Multiple supratemporal sources of
magnetic and electric auditory evoked middle latency components in
humans. Cereb Cortex 2001;11:411–23.
14. Linden DE. The p300: where in the brain is it produced and what does it tell
us? Neuroscientist 2005;11:563–76.
15. Guerit JM, Amantini A, Amodio P, et al. Consensus on the use
of neurophysiological tests in the intensive care unit (ICU):
electroencephalogram (EEG), evoked potentials (EPs), and
electroneuromyography (ENMG). Neurophysiol Clin 2009;39:71–83.
16. Bouwes A, Binnekade JM, Verbaan BW, et al. Predictive value of neurological
examination for early cortical responses to somatosensory evoked potentials
in patients with postanoxic coma. J Neurol 2012;259(3):537–41. Epub 2011,
Sep 2.
17. Tiainen M, Kovala TT, Takkunen OS, et al. Somatosensory and brainstem
auditory evoked potentials in cardiac arrest patients treated with
hypothermia. Crit Care Med 2005;33:1736–40.
18. Bouwes A, Binnekade JM, Zandstra DF, et al. Somatosensory evoked
potentials during mild hypothermia after cardiopulmonary resuscitation.
Neurology 2009;73:1457–61.

19. Zandbergen EG, Hijdra A, Koelman JH, et al. Prediction of poor outcome
within the first 3 days of postanoxic coma. Neurology 2006;66:62–8.
20. Madl C, Kramer L, Domanovits H, et al. Improved outcome prediction in
unconscious cardiac arrest survivors with sensory evoked potentials
compared with clinical assessment. Crit Care Med 2000;28:721–6.
21. Gendo A, Kramer L, Hafner M, et al. Time-dependency of sensory evoked
potentials in comatose cardiac arrest survivors. Intensive Care Med 2001;27:
1305–11.
22. Logi F, Fischer C, Murri L, et al. The prognostic value of evoked responses
from primary somatosensory and auditory cortex in comatose patients. Clin
Neurophysiol 2003;114:1615–27.
23. Berek K, Lechleitner P, Luef G, et al. Early determination of neurological
outcome after prehospital cardiopulmonary resuscitation. Stroke 1995;26:
543–9.
24. Young GB, Doig G, Ragazzoni A. Anoxic-ischemic encephalopathy: clinical
and electrophysiological associations with outcome. Neurocrit Care
2005;2:159–64.
25. Chen R, Bolton CF, Young B. Prediction of outcome in patients with anoxic
coma: a clinical and electrophysiologic study. Crit Care Med 1996;24:672–8.
26. Bassetti C, Bomio F, Mathis J, et al. Early prognosis in coma after cardiac
arrest: a prospective clinical, electrophysiological, and biochemical study of
60 patients. J Neurol Neurosurg Psychiatry 1996;61:610–5.
27. Zandbergen EG, de Haan RJ, Stoutenbeek CP, et al. Systematic review of
early prediction of poor outcome in anoxic-ischaemic coma. Lancet 1998;
352:1808–12.
28. Wijdicks EF, Hijdra A, Young GB, et al. Practice parameter: prediction of
outcome in comatose survivors after cardiopulmonary resuscitation (an
evidence-based review): report of the Quality Standards Subcommittee of
the American Academy of Neurology. Neurology 2006;67:203–10.
29. Zhang Y, Su YY, Haupt WF, et al. Application of electrophysiologic
techniques in poor outcome prediction among patients with severe focal and
diffuse ischemic brain injury. J Clin Neurophysiol 2011;28(5):497–503.
30. Rossetti AO, Oddo M, Liaudet L, et al. Predictors of awakening from
postanoxic status epilepticus after therapeutic hypothermia. Neurology 2009;
72(8):744–9.
31. Meynaar IA, Oudemans-van Straaten HM, van der Wetering J, et al. Serum
neuron-specific enolase predicts outcome in post-anoxic coma: a prospective
cohort study. Intensive Care Med 2003;29:189–95.
32. Krieger DW. Evoked potentials not just to confirm hopelessness in anoxic
brain injury. Lancet 1998;352:1796–7.
33. Walser H, Mattle H, Keller HM, et al. Early cortical median nerve
somatosensory evoked potentials: prognostic value in anoxic coma. Arch
Neurol 1985;42:32–8.
34. Zandbergen EG, Koelman JH, de Haan RJ, et al. SSEPs and prognosis in
postanoxic coma: only short or also long latency responses? Neurology 2006;
67:583–6.
35. Prohl J, Rother J, Kluge S, et al. Prediction of short-term and long-term
outcomes after cardiac arrest: a prospective multivariate approach combining
biochemical, clinical, electrophysiological, and neuropsychological
investigations. Crit Care Med 2007;35:1230–7.
Section III—Electrophysiology 245
36. Lori S, Bertini G, Molesti E, et al. The prognostic role of evoked potentials
in neonatal hypoxic-ischemic insult. J Matern Fetal Neonatal Med 2011;
24 Suppl 1:69–71. Epub 2011, Aug 31.
37. Sleigh JW, Havill JH, Frith R, et al. Somatosensory evoked potentials in
severe traumatic brain injury: a blinded study. J Neurosurg 1999;91:
577–80.
38. Rumpl E, Prugger M, Battista HJ, et al. Short latency somatosensory evoked
potentials and brain-stem auditory evoked potentials in coma due to CNS
depressant drug poisoning. Preliminary observations. Electroencephalogr
Clin Neurophysiol 1988;70:482–9.
39. Procaccio F, Polo A, Lanteri P, et al. Electrophysiologic monitoring in
neurointensive care. Curr Opin Crit Care 2001;7:74–80.
40. Chiappa KH, Hoch DB. Electrophysiologic monitoring. In: Ropper AH,
editor. Neurological and neurosurgical intensive care. New York: Raven Press;
1993. p. 147–83.
41. Carter BG, Butt W. Review of the use of somatosensory evoked potentials in
the prediction of outcome after severe brain injury. Crit Care Med 2001;29:
178–86.
42. Carter BG, Butt W. Are somatosensory evoked potentials the best predictor
of outcome after severe brain injury? A systematic review. Intensive Care
Med 2005;31:765–75.
43. Claassen J, Hansen HC. Early recovery after closed traumatic head injury:
somatosensory evoked potentials and clinical findings. Crit Care Med 2001;
29:494–502.
44. Houlden DA, Taylor AB, Feinstein A, et al. Early somatosensory evoked
potential grades in comatose traumatic brain injury patients predict
cognitive and functional outcome. Crit Care Med 2010;38:167–74.
45. Moulton RJ, Shedden PM, Tucker WS, et al. Somatosensory evoked potential
monitoring following severe closed head injury. Clin Invest Med 1994;17:
187–95.
46. Amantini A, Fossi S, Grippo A, et al. Continuous EEG-SSEP monitoring in
severe brain injury. Clin Neurophysiol 2009;39:85–93.
47. Christophis P. The prognostic value of somatosensory evoked potentials in
traumatic primary and secondary brain stem lesions. Zentralbl Neurochir
2004;65:25–31.
48. Furlonger AJ, Sleigh JW, Havill JH, et al. Cognitive and psychosocial
outcome in survivors of severe traumatic brain injury: correlations with
cerebral perfusion pressure, frontal lobe damage and somatosensory evoked
potentials. Crit Care Resusc 2000;2:246–52.
49. Rumpl E, Prugger M, Gerstenbrand F, et al. Central somatosensory
conduction time and short latency somatosensory evoked potentials in
post-traumatic coma. Electroencephalogr Clin Neurophysiol 1983;56:
583–96.
50. Pohlmann-Eden B, Dingethal K, Bender HJ, et al. How reliable is the
predictive value of SEP (somatosensory evoked potentials) patterns in severe
brain damage with special regard to the bilateral loss of cortical responses?
Intensive Care Med 1997;23:301–8.
A complete list of references for this chapter can be found online at
www.expertconsult.com.

References
1. Dawson G. Investigations on a patient subject to myoclonic seizure after
stimulation. J Neurol Neurosurg Psychiatry 1947;10:141–62.
2. Moulton RJ, Brown JI, Konasiewicz SJ. Monitoring severe head injury: a
comparison of EEG and somatosensory evoked potentials. Can J Neurol Sci
1998;25:S7–11.
3. Celesia G, Puletti F. Auditory cortical areas of man. Neurology 1969;19:
211–20.
4. Wijdicks EF. Brain death worldwide: accepted fact but no global consensus
in diagnostic criteria. Neurology 2002;58:20–5.
5. Cruccu G, Aminoff MJ, Curio G, et al. Recommendations for the clinical
use of somatosensory-evoked potentials. Clin Neurophysiol 2008;119:
1705–19.
6. Fossi S, Amantini A, Grippo A, et al. Continuous EEG-SEP monitoring of
severely brain injured patients in NICU: methods and feasibility.
Neurophysiol Clin 2006;36:195–205.
7. Celesia GC, Brigell MG. Auditory evoked potentials. In: Niedermeyer E,
Lopes da Silva F, editors. Electroencephalography: basic principles, clinical
applications, and related fields. Baltimore: Lippincott Williams & Wilkins;
2005. p. 1137–64.
8. Mauguiere F. Somatosensory evoked potentials: normal responses,
abnormal waveforms, and clinical applications in neurological diseases. In:
Niedermeyer E, Lopes da Silva F, editors. Electroencephalography: basic
applications, and related fields. Baltimore: Lippincott Williams & Wilkins;
2005. p. 1067–119.
9. Goff GD, Matsumiya Y, Allison T, et al. The scalp topography of human
somatosensory and auditory evoked potentials. Electroencephalogr Clin
Neurophysiol 1977;42:57–76.
10. Desmedt JE, Ozaki I. SEPs to finger joint input lack the N20-P20 response
that is evoked by tactile inputs: contrast between cortical generators in
areas 3b and 2 in humans. Electroencephalogr Clin Neurophysiol 1991;80:
513–21.
11. Scherg M, von Cramon D. A new interpretation of the generators of BAEP
waves I-V: results of a spatio-temporal dipole model. Electroencephalogr
Clin Neurophysiol 1985;62:290–9.
12. Starr A. Auditory brain-stem responses in brain death. Brain
1976;99:543–54.
13. Yvert B, Crouzeix A, Bertrand O, et al. Multiple supratemporal sources of
magnetic and electric auditory evoked middle latency components in
humans. Cereb Cortex 2001;11:411–23.
14. Linden DE. The p300: where in the brain is it produced and what does it
tell us? Neuroscientist 2005;11:563–76.
15. Guerit JM, Amantini A, Amodio P, et al. Consensus on the use of
neurophysiological tests in the intensive care unit (ICU):
electroencephalogram (EEG), evoked potentials (EPs), and
electroneuromyography (ENMG). Neurophysiol Clin 2009;39:71–83.
16. Bouwes A, Binnekade JM, Verbaan BW, et al. Predictive value of
neurological examination for early cortical responses to somatosensory
evoked potentials in patients with postanoxic coma. J Neurol 2012;259(3):
537–41. Epub 2011, Sep 2.
17. Tiainen M, Kovala TT, Takkunen OS, et al. Somatosensory and brainstem
auditory evoked potentials in cardiac arrest patients treated with
hypothermia. Crit Care Med 2005;33:1736–40.
18. Bouwes A, Binnekade JM, Zandstra DF, et al. Somatosensory evoked
potentials during mild hypothermia after cardiopulmonary resuscitation.
Neurology 2009;73:1457–61.
19. Zandbergen EG, Hijdra A, Koelman JH, et al. Prediction of poor outcome
within the first 3 days of postanoxic coma. Neurology 2006;66:62–8.
20. Madl C, Kramer L, Domanovits H, et al. Improved outcome prediction in
unconscious cardiac arrest survivors with sensory evoked potentials
compared with clinical assessment. Crit Care Med 2000;28:721–6.
21. Gendo A, Kramer L, Hafner M, et al. Time-dependency of sensory evoked
potentials in comatose cardiac arrest survivors. Intensive Care Med 2001;
27:1305–11.
22. Logi F, Fischer C, Murri L, et al. The prognostic value of evoked responses
from primary somatosensory and auditory cortex in comatose patients.
Clin Neurophysiol 2003;114:1615–27.
23. Berek K, Lechleitner P, Luef G, et al. Early determination of neurological
outcome after prehospital cardiopulmonary resuscitation. Stroke 1995;26:
543–9.
24. Young GB, Doig G, Ragazzoni A. Anoxic-ischemic encephalopathy: clinical
and electrophysiological associations with outcome. Neurocrit Care 2005;2:
159–64.
25. Chen R, Bolton CF, Young B. Prediction of outcome in patients with anoxic
coma: a clinical and electrophysiologic study. Crit Care Med 1996;24:672–8.
Section III—Electrophysiology 245.e1
26. Bassetti C, Bomio F, Mathis J, et al. Early prognosis in coma after cardiac
arrest: a prospective clinical, electrophysiological, and biochemical study of
60 patients. J Neurol Neurosurg Psychiatry 1996;61:610–5.
27. Zandbergen EG, de Haan RJ, Stoutenbeek CP, et al. Systematic review of
early prediction of poor outcome in anoxic-ischaemic coma. Lancet 1998;
352:1808–12.
28. Wijdicks EF, Hijdra A, Young GB, et al. Practice parameter: prediction of
outcome in comatose survivors after cardiopulmonary resuscitation (an
evidence-based review): report of the Quality Standards Subcommittee of
the American Academy of Neurology. Neurology 2006;67:203–10.
29. Zhang Y, Su YY, Haupt WF, et al. Application of electrophysiologic
techniques in poor outcome prediction among patients with severe focal
and diffuse ischemic brain injury. J Clin Neurophysiol 2011;28(5):497–503.
30. Rossetti AO, Oddo M, Liaudet L, et al. Predictors of awakening from
postanoxic status epilepticus after therapeutic hypothermia. Neurology
2009;72(8):744–9.
31. Meynaar IA, Oudemans-van Straaten HM, van der Wetering J, et al. Serum
neuron-specific enolase predicts outcome in post-anoxic coma: a
prospective cohort study. Intensive Care Med 2003;29:189–95.
32. Krieger DW. Evoked potentials not just to confirm hopelessness in anoxic
brain injury. Lancet 1998;352:1796–7.
33. Walser H, Mattle H, Keller HM, et al. Early cortical median nerve
somatosensory evoked potentials: prognostic value in anoxic coma. Arch
Neurol 1985;42:32–8.
34. Zandbergen EG, Koelman JH, de Haan RJ, et al. SSEPs and prognosis in
postanoxic coma: only short or also long latency responses? Neurology
2006;67:583–6.
35. Prohl J, Rother J, Kluge S, et al. Prediction of short-term and long-term
outcomes after cardiac arrest: a prospective multivariate approach
combining biochemical, clinical, electrophysiological, and
neuropsychological investigations. Crit Care Med 2007;35:1230–7.
36. Lori S, Bertini G, Molesti E, et al. The prognostic role of evoked potentials
in neonatal hypoxic-ischemic insult. J Matern Fetal Neonatal Med 2011;
24 Suppl 1:69–71. Epub 2011, Aug 31.
37. Sleigh JW, Havill JH, Frith R, et al. Somatosensory evoked potentials in
severe traumatic brain injury: a blinded study. J Neurosurg 1999;91:
577–80.
38. Rumpl E, Prugger M, Battista HJ, et al. Short latency somatosensory evoked
potentials and brain-stem auditory evoked potentials in coma due to CNS
depressant drug poisoning. Preliminary observations. Electroencephalogr
Clin Neurophysiol 1988;70:482–9.
39. Procaccio F, Polo A, Lanteri P, et al. Electrophysiologic monitoring in
neurointensive care. Curr Opin Crit Care 2001;7:74–80.
40. Chiappa KH, Hoch DB. Electrophysiologic monitoring. In: Ropper AH,
editor. Neurological and neurosurgical intensive care. New York: Raven
Press; 1993. p. 147–83.
41. Carter BG, Butt W. Review of the use of somatosensory evoked potentials
in the prediction of outcome after severe brain injury. Crit Care Med 2001;
29:178–86.
42. Carter BG, Butt W. Are somatosensory evoked potentials the best predictor
of outcome after severe brain injury? A systematic review. Intensive Care
Med 2005;31:765–75.
43. Claassen J, Hansen HC. Early recovery after closed traumatic head injury:
somatosensory evoked potentials and clinical findings. Crit Care Med 2001;
29:494–502.
44. Houlden DA, Taylor AB, Feinstein A, et al. Early somatosensory evoked
potential grades in comatose traumatic brain injury patients predict
cognitive and functional outcome. Crit Care Med 2010;38:167–74.
45. Moulton RJ, Shedden PM, Tucker WS, et al. Somatosensory evoked
potential monitoring following severe closed head injury. Clin Invest Med
1994;17:187–95.
46. Amantini A, Fossi S, Grippo A, et al. Continuous EEG-SSEP monitoring in
severe brain injury. Clin Neurophysiol 2009;39:85–93.
47. Christophis P. The prognostic value of somatosensory evoked potentials in
traumatic primary and secondary brain stem lesions. Zentralbl Neurochir
2004;65:25–31.
48. Furlonger AJ, Sleigh JW, Havill JH, et al. Cognitive and psychosocial
outcome in survivors of severe traumatic brain injury: correlations with
cerebral perfusion pressure, frontal lobe damage and somatosensory evoked
potentials. Crit Care Resusc 2000;2:246–52.
49. Rumpl E, Prugger M, Gerstenbrand F, et al. Central somatosensory
conduction time and short latency somatosensory evoked potentials in
post-traumatic coma. Electroencephalogr Clin Neurophysiol 1983;56:
583–96.
50. Pohlmann-Eden B, Dingethal K, Bender HJ, et al. How reliable is the
predictive value of SEP (somatosensory evoked potentials) patterns in
245.e2 Section III—Electrophysiology
severe brain damage with special regard to the bilateral loss of cortical
responses? Intensive Care Med 1997;23:301–8.
51. Krieger D, Jauss M, Schwarz S, et al. Serial somatosensory and brainstem
auditory evoked potentials in monitoring of acute supratentorial mass
lesions. Crit Care Med 1995;23:1123–31.
52. Cant BR, Hume AL, Judson JA, et al. The assessment of severe head injury
by short-latency somatosensory and brain-stem auditory evoked potentials.
Electroencephalogr Clin Neurophysiol 1986;65:188–95.
53. Lindsay KW, Carlin J, Kennedy I, et al. Evoked potentials in severe head
injury—analysis and relation to outcome. J Neurol Neurosurg Psychiatry
1981;44:796–802.
54. Lindsay K, Pasaoglu A, Hirst D, et al. Somatosensory and auditory brain
stem conduction after head injury: a comparison with clinical features in
prediction of outcome. Neurosurgery 1990;26:278–85.
55. Rappaport M, Hemmerle AV, Rappaport ML. Short and long latency
auditory evoked potentials in traumatic brain injury patients. Clin
Electroencephalogr 1991;22:199–202.
56. Shimbo Y, Sakata M, Hayano M, et al. Topographical relationships between
the brainstem auditory and somatosensory evoked potentials and the
location of lesions in posterior fossa stroke. Neurol Med Chir (Tokyo)
2003;43:282–91. discussion 92.
57. Kalita J, Misra UK. Motor and sensory evoked potential studies in
brainstem strokes. Electromyogr Clin Neurophysiol 1997;37:379–83.
58. Gutling E, Isenmann S, Wichmann W. Electrophysiology in the locked-in-
syndrome. Neurology 1996;46:1092–101.
59. Towle VL, Maselli R, Bernstein LP, et al. Electrophysiologic studies on
locked-in patients: heterogeneity of findings. Electroencephalogr Clin
Neurophysiol 1989;73:419–26.
60. Bassetti C, Mathis J, Hess CW. Multimodal electrophysiological studies
including motor evoked potentials in patients with locked-in syndrome:
report of six patients. J Neurol Neurosurg Psychiatry 1994;57:1403–6.
61. Burghaus L, Liu WC, Dohmen C, et al. Evoked potentials in acute ischemic
stroke within the first 24 h: possible predictor of a malignant course.
Neurocrit Care 2008;9:13–6.
62. Stern BJ, Krumholz A, Weiss HD, et al. Evaluation of brainstem stroke
using brainstem auditory evoked responses. Stroke 1982;13:705–11.
63. Krieger D, Adams HP, Rieke K, et al. Prospective evaluation of the
prognostic significance of evoked potentials in acute basilar occlusion. Crit
Care Med 1993;21:1169–74.
64. Krieger D, Adams HP, Rieke K, et al. Monitoring therapeutic efficacy of
decompressive craniotomy in space occupying cerebellar infarcts using
brain-stem auditory evoked potentials. Electroencephalogr Clin
Neurophysiol 1993;88:261–70.
65. Steiner T, Jauss M, Krieger DW. Hemicraniectomy for massive cerebral
infarction: evoked potentials as presurgical prognostic factors. J Stroke
Cerebrovasc Dis 1998;7:132v8.
66. Kalita J, Misra UK. Somatosensory evoked potential studies in internal
capsule and corona radiata infarction. J Neurol 1998;245:531–6.
67. Haupt WF, Birkmann C, Halber M. Serial evoked potentials and outcome in
cerebrovascular critical care patients. J Clin Neurophysiol 2000;17:326–30.
68. Chu NS. Median and tibial somatosensory evoked potentials. Changes in
short- and long-latency components in patients with lesions of the
thalamus and thalamo-cortical radiations. J Neurol Sci 1986;76:199–219.
69. Kato H, Sugawara Y, Ito H, et al. Somatosensory evoked potentials
following stimulation of median and tibial nerves in patients with localized
intracerebral hemorrhage: correlations with clinical and CT findings. J
Neurol Sci 1991;103:172–8.
70. Ferbert A, Buchner H, Bruckmann H. Brainstem auditory evoked potentials
and somatosensory evoked potentials in pontine haemorrhage. Correlations
with clinical and CT findings. Brain 1990;113(Pt 1):49–63.
71. Lumenta CB. Measurements of brain-stem auditory evoked potentials in
patients with spontaneous intracerebral hemorrhage. J Neurosurg 1984;60:
548–52.
72. Liu CW, Chu NS, Ryu SJ. CT somatosensory and brainstem auditory
evoked potentials in the early prediction of functional outcome in
putaminal hemorrhage. Acta Neurol Scand 1991;84:28–32.
73. Shigemori M, Yuge T, Kawasaki K, et al. Evaluation of brain dysfunction in
hypertensive putaminal hemorrhage with multimodality evoked potentials.
Stroke 1987;18:72–6.
74. Facco E, Behr AU, Munari M, et al. Auditory and somatosensory evoked
potentials in coma following spontaneous cerebral hemorrhage: early
prognosis and outcome. Electroencephalogr Clin Neurophysiol 1998;107:
332–8.
75. Wang AD, Cone J, Symon L, et al. Somatosensory evoked potential
monitoring during the management of aneurysmal SAH. J Neurosurg
1984;60:264–8.
76. Haupt WF, Hojer C, Pawlik G. Prognostic value of evoked potentials and
clinical grading in primary subarachnoid haemorrhage. Acta Neurochir
(Wien) 1995;137:146–50. discussion 50.
77. Ritz R, Schwerdtfeger K, Strowitzki M, et al. Prognostic value of SSEP in
early aneurysm surgery after SAH in poor-grade patients. Neurol Res
2002;24:756–64.
78. Szabo S, Miko L, Novak L, et al. Correlation between central
somatosensory conduction time, blood flow velocity, and delayed cerebral
ischemia after aneurysmal subarachnoid hemorrhage. Neurosurg Rev 1997;
20:188–95.
79. Rosenstein J, Wang AD, Symon L, et al. Relationship between hemispheric
cerebral blood flow, central conduction time, and clinical grade in
aneurysmal subarachnoid hemorrhage. J Neurosurg 1985;62:25–30.
80. Ikeda K, Yamashita J. Postoperative central conduction time and cerebral
blood flow in patients with aneurysmal subarachnoid hemorrhage:
relationship with prognosis and ischemic conditions. Surg Neurol 1992;38:
445–53.
81. Fazl M, Houlden DA, Weaver K. Correlation between cerebral blood flow,
somatosensory evoked potentials, CT scan grade and neurological grade
in patients with subarachnoid hemorrhage. Can J Neurol Sci 1991;18:
453–7.
82. Symon L, Hargadine J, Zawirski M, et al. Central conduction time as an
index of ischaemia in subarachnoid haemorrhage. J Neurol Sci
1979;44:95–103.
83. Schick U, Dohnert J, Meyer JJ, et al. Prognostic significance of SSEP, BAEP
and serum S-100B monitoring after aneurysm surgery. Acta Neurol Scand
2003;108:161–9.
84. Wachter D, Christophis P, Stein M, et al. Use of multimodal
electrophysiological monitoring to predict outcome after subarachnoid
hemorrhage? A prospective series. J Neurosurg Sci 2011;55(3):179–87.
85. Niimi Y, Deltis V, Berenstein A. Neurophysiological monitoring during
spinal endovascular procedures. In: Nuwer M, editor. Handbook of clinical
neurophysiology. Amsterdam: Elsevier; 2008. p. 651–70.
86. Iseli E, Cavigelli A, Dietz V, et al. Prognosis and recovery in ischaemic and
traumatic spinal cord injury: clinical and electrophysiological evaluation.
J Neurol Neurosurg Psychiatry 1999;67:567–71.
87. Jacobs SR, Yeaney NK, Herbison GJ, et al. Future ambulation prognosis as
predicted by somatosensory evoked potentials in motor complete and
incomplete quadriplegia. Arch Phys Med Rehabil 1995;76:635–41.
88. Curt A, Dietz V. Ambulatory capacity in spinal cord injury: significance of
somatosensory evoked potentials and ASIA protocol in predicting outcome.
Arch Phys Med Rehabil 1997;78:39–43.
89. Katz RT, Toleikis RJ, Knuth AE. Somatosensory-evoked and dermatomal-
evoked potentials are not clinically useful in the prognostication of acute
spinal cord injury. Spine 1991;16:730–5.
90. Zörner B, Blanckenhorn WU, Dietz V, et al. Clinical algorithm for improved
prediction of ambulation and patient stratification after incomplete spinal
cord injury. J Neurotrauma 2010;27(1):241–52.
91. Cheliout-Heraut F, Loubert G, Masri-Zada T, et al. Evaluation of early
motor and sensory evoked potentials in cervical spinal cord injury.
Neurophysiol Clin 1998;28:39–55.
92. Buchner H, Ferbert A, Hacke W. Serial recording of median nerve
stimulated subcortical somatosensory evoked potentials (SEPs) in
developing brain death. Electroencephalogr Clin Neurophysiol
1988;69:14–23.
93. Sonoo M, Tsai-Shozawa Y, Aoki M, et al. N18 in median somatosensory
evoked potentials: a new indicator of medullary function useful for the
diagnosis of brain death. J Neurol Neurosurg Psychiatry 1999;67:374–8.
94. Goldie WD, Chiappa KH, Young RR, et al. Brainstem auditory and
short-latency somatosensory evoked responses in brain death. Neurology
1981;31:248–56.
95. Machado C, Valdes P, Garcia-Tigera J, et al. Brain-stem auditory evoked
potentials and brain death. Electroencephalogr Clin Neurophysiol
1991;80:392–8.
96. Facco E, Munari M, Gallo F, et al. Role of short latency evoked potentials in
the diagnosis of brain death. Clin Neurophysiol 2002;113:1855–66.
97. Guerit JM, Amantini A, Fischer C, et al. Neurophysiological investigations
of hepatic encephalopathy: ISHEN practice guidelines. Liver Int
2009;29:789–96.
98. Nuwer MR, Emerson RG, Galloway G, et al. Evidence-based guideline
update: intraoperative spinal monitoring with somatosensory and
transcranial electrical motor evoked potentials. J Clin Neurophysiol 2012;
29(1):101–8.
99. Gonzalez AA, Jeyanandarajan D, Hansen C, et al. Intraoperative
neurophysiological monitoring during spine surgery: a review. Neurosurg
Focus 2009;27(4):E6.

100. Resnick DK, Anderson PA, Kaiser MG, et al. Electrophysiological
monitoring during surgery for cervical degenerative myelopathy and
radiculopathy. J Neurosurg Spine 2009;11(2):245–52.
101. McGarvey ML, Cheung AT. Neuologic intraoperative electrophysiologic
monitoring. In: Reich DL, Kahn RA, Mittnacht AJC, et al, editors.
Monitoring in anesthesia and perioperative care. New York: Cambridge
University Press; 2011. p. 199–217.
102. Wicks RT, Pradilla G, Raza SM, et al. Impact of changes in intraoperative
somato-sensory evoked potentials (SSEPs) on stroke rates after clipping of
intracranial aneurysms. Neurosurgery 2011;Nov 3. Epub ahead of print.
103. Shinzawa M, Yoshitani K, Minatoya K, et al. Changes of motor evoked
potentials during descending thoracic and thoracoabdominal aortic surgery
with deep hypothermic circulatory arrest. J Anesth 2011;Dec 27. Epub
ahead of print.
104. Quraishi NA, Lewis SJ, Kelleher MO, et al. Intraoperative multimodality
monitoring in adult spinal deformity: analysis of a prospective series of one
hundred two cases with independent evaluation. Spine 2009;34(14):1504–12.
105. Simon MV, Cole AJ, Chang EC, et al. An intraoperative multimodal
neurophysiologic approach to successful resection of precentral gyrus
epileptogenic lesions. Epilepsia 2012;Feb 6. Epub ahead of print.
106. Krieg SM, Shiban E, Droese D, et al. Predictive value and safety of
intraoperative neurophysiological monitoring using motor evoked
potentials in glioma surgery. Neurosurgery 2011;Nov 3. Epub ahead of print.
107. Lepski G, Honegger J, Liebsch M, et al. Safe resection of arteriovenous
malformations in eloquent motor areas aided by functional imaging and
intraoperative monitoring. Neurosurgery 2011;Sep 23. Epub ahead of print.
108. Constantini S, Miller DC, Allen JC, et al. Radical excision of intramedullary
spinal cord tumors: surgical morbidity and long-term follow-up evaluation
in 164 children and young adults. J Neurosurg 2000;93(2 Suppl):183–93.
109. Simon MV. Neurophysiologic intraoperative monitoring of the
vestibulocochlear nerve. J Clin Neurophysiol 2011;28(6):566–81.
110. Zandbergen EG, Hijdra A, de Haan RJ, et al. Interobserver variation in the
interpretation of SSEPs in anoxic-ischaemic coma. Clin Neurophysiol
2006;117:1529–35.
111. Hakimi K, Kinney G, Kraft G, et al. Reliability in interpretation of median
somatosensory evoked potentials in the setting of coma: factors and
implications. Neurocrit Care 2009;11(3):353–61.
Section III—Electrophysiology 245.e3
112. Sloan TB, Ronai AK, Toleikis JR, et al. Improvement of intraoperative
somatosensory evoked potentials by etomidate. Anesth Analg 1988;67:
582–5.
113. Legatt A. BAEP in surgery. In: Nuwer M, editor. Intraoperative monitoring
of neural function handbook of clinical neurophysiology. Elsevier; 2008.
p. 334–49.
114. Drummond JC, Todd MM, Schubert A, et al. Effect of the acute
administration of high dose pentobarbital on human brain stem auditory
and median nerve somatosensory evoked responses. Neurosurgery 1987;20:
830–5.
115. Liu EH, Wong HK, Chia CP, et al. Effects of isoflurane and propofol on
cortical somatosensory evoked potentials during comparable depth of
anaesthesia as guided by bispectral index. Br J Anaesth 2005;94:193–7.
116. Chan YW, Woo E, Yu YL. Chronic effects of phenytoin on brain-stem
auditory evoked potentials in man. Electroencephalogr Clin Neurophysiol
1990;77:119–26.
117. Hirose G, Kitagawa Y, Chujo T, et al. Acute effects of phenytoin on
brainstem auditory evoked potentials: clinical and experimental study.
Neurology 1986;36:1521–4.
118. Markand ON, Warren C, Mallik GS, et al. Effects of hypothermia on short
latency somatosensory evoked potentials in humans. Electroencephalogr
Clin Neurophysiol 1990;77:416–24.
119. Guerit JM, Soveges L, Baele P, Dion R. Median nerve somatosensory evoked
potentials in profound hypothermia for ascending aorta repair.
Electroencephalogr Clin Neurophysiol 1990;77:163–73.
120. Markand ON, Lee BI, Warren C, et al. Effects of hypothermia on brainstem
auditory evoked potentials in humans. Ann Neurol 1987;22:507–13.
121. Rodriguez RA, Audenaert SM, Austin EH, 3rd, et al. Auditory evoked
responses in children during hypothermic cardiopulmonary bypass: report
of cases. J Clin Neurophysiol 1995;12:168–76.
122. Kusakari J, Inamura N, Sakurai T, et al. Effect of hypothermia upon the
electrocochleogram and auditory evoked brainstem response. Tohoku J Exp
Med 1984;143:351–9.
123. Rosenblum SM, Ruth RA, Gal TJ. Brain stem auditory evoked potential
monitoring during profound hypothermia and circulatory arrest. Ann Otol
Rhinol Laryngol 1985;94:281–3.
124. Seaba P. Electrical safety. Am J EEG Technol 1980;20:1–13.
III
Chapter
25  
Electroencephalography
Peter Horn, Mauro Oddo, and Sarah E. Schmitt
Introduction
The German psychiatrist Hans Berger made the first recording
of the electric field of the human brain in 1924 in Jena. He
gave this recording the name electroencephalogram (EEG).1
His crucial observations served as the basis for the later appli-
cation of the EEG in clinical practice and promoted the fast
development of clinical neurology and neurophysiology. The
EEG itself represents the spontaneous electrical activity of the
brain. Most brain electrical activity in a healthy individual is
between 20 and 40 microvolts, although it may reach up to
100 microvolts. There are differences when measured on the
scalp or directly on the brain surface. The bandwidth of the
signal ranges between 0.4 and about 50 Hz and is highly vari-
able. In contrast to spontaneous EEG background activity,
evoked potentials (EPs) represent the components of the EEG
that arise in response to a certain stimulus (e.g., visual, tactile,
or auditory) (see Chapter 24). These signals are usually below
the noise level and are thus not readily distinguished in the
EEG background activity. Therefore one must use repetition
of the stimuli and signal averaging to improve the signal-to-
noise ratio to get appropriate information.2
In clinical practice, the recording of EEG background activ-
ity and the application of EPs are broadly accepted. Thus
EEG, including both traditional recordings and modifications
such as compressed spectral analysis, is a cornerstone of
neurologic diagnostics, clinical neurophysiology, and related
research. Before 1996, EEGs were performed on analog
machines with paper printouts. Technical advances since 2000
have led to a step-wise introduction of the EEG into neuro-
critical care. The era of digitization has broadened the
spectrum for EEG recordings on the neurocritical care unit
(NCCU) (EEG information can be recorded continuously
and stored digitally).3 EEG and EPs have become common
and valuable tools in the NCCU. In particular, different
EEG-derived parameters can help predict individual clinical
prognosis and outcome.4,5 Continuous EEG (cEEG) allows
real-time monitoring of the underlying cerebral physiology
noninvasively and cost effectively6 in critically ill patients, and
it can be used to gauge the efficacy of certain medications or
therapeutic maneuvers. For example, cEEG can be used for
seizure diagnosis in comatose patients after traumatic brain
injury (TBI), detection of delayed cerebral ischemia (DCI),
after subarachnoid hemorrhage (SAH), and titration of thera-
pies for status epilepticus or barbiturate-induced coma among
246
others. However, the integration of EEG recording into the
environment of neurocritical care requires a refined technical
setup and highly trained technical and medical staff. These
necessities require a mutidisciplinary team. This chapter
focuses on the application of intermittent (iEEG) or cEEG
monitoring in the NCCU.
Physiologic Electroencephalogram
Background Activity
In a healthy individual, typical EEG background activity
consists of a well-organized and predictable spectrum of
waveforms. However, numerous factors may influence the fre-
quency and organization of these waveforms. There exist
ultra-slow and ultra-fast frequencies that in general do not
play a significant role in clinical practice in the intensive care
unit (ICU). However, ultra-fast frequencies can be clinically
useful in predicting outcomes from epilepsy surgery. Very low-
frequency changes usually are filtered out and although they
may be observed in patients in coma or preterminal condi-
tions, they usually do not provide clinically useful informa-
tion. In clinical practice the frequency range of the EEG is
broken down as follows:
Delta rhythm: 0.1 to 3.5 Hz
Theta rhythm: 4 to 7.5 Hz
Alpha rhythm: 8 to 13 Hz
Beta rhythm: 14 to 30 Hz
Gamma rhythm: >30 Hz (often filtered out on scalp EEG)
Several aspects are essential for EEG interpretation: (1) fre-
quency, (2) amplitude, (3) rhythmicity, (4) morphology, and
(5) spatial distribution. These aspects are highly variable
under normal conditions and show a strong dependence
on numerous factors such as age, level of consciousness
(alertness), and physical or mental activity. In addition, many
external or intrinsic stimuli can influence EEG background
activity.7,8
Pathologic Electroencephalogram
Activity
EEG background activity is very sensitive to changes in brain
homeostasis and physiology. EEG may show manifestations of
© Copyright 2013 Elsevier Inc. All rights reserved.

Fig. 25.1  Right temporal seizure with evolution in morphology, amplitude, frequency, and spatial distribution.
numerous conditions, including metabolic encephalopathy,
systemic hypoxia, brain trauma, ischemic and hemorrhagic
stroke, brain tumor, and infectious disease to name a few.
Any change in electrical brain activity, either focal or gen-
eralized, that leads to a failure of local compensation may
cause epileptic discharges. The extent and duration of EEG
alterations are highly variable. An electrographic seizure is
defined as a pattern lasting more than 10 seconds with (1)
epileptiform waveforms at 3 Hz or greater, (2) epileptiform
waveforms at less than 3 Hz with clear evidence of evolution
in time or space and (3) evidence of clinical or electrographic
improvement following administration of an antiepileptic
drug (AED. Fig. 25.1). Furthermore, a seizure is likely to be
present if the pathologic EEG is characterized by an evolution
in frequency, morphology, and location.3
Types of Seizures, Related Phenomena,
and Differential Diagnosis
In the clinical setting many seizures are easily recognized and
therefore amenable to direct intervention and treatment. Con-
vulsive seizures (either primary or secondarily generalized)
are usually clinically obvious and do not require an EEG for
diagnosis. However, many focal seizures are associated with
few or no outward clinical manifestations. These seizures,
which are referred to as nonconvulsive seizures or subclinical
seizures require an EEG for their diagnosis, including among
Section III—Electrophysiology 247
patients with impaired consciousness in whom clinical diag-
nosis of a seizure may be difficult. In this group of patients
EEG also can help assess status epilepticus (SE), nonconvul-
sive status epilepticus (NCSE), and seizure-like activity such
as periodic epileptiform discharges (PEDs) or generalized
periodic epileptiform discharges (G-PEDs).9-11
SE is a neurologic emergency with high mortality and mor-
bidity. There are several types of SE, and consequently it has
several definitions. The International League Against Epilepsy
defines SE as “a seizure that persists for a sufficient length of
time or repeated frequently enough that recovery between
attacks does not occur.”11a SE can be subdivided based on clini-
cal and EEG criteria into generalized convulsive status (tonic-
clonic, tonic, clonic, and myoclonic), generalized nonconvulsive
(absence) status, elementary partial status (with several sub-
types), and complex partial status. Official definitions have
not included the exact duration of time, but based on animal
studies that demonstrate significant brain damage after 30
minutes, convulsive status epilepticus (CSE) traditionally is
considered as one continuous unremitting seizure that lasts
greater than 30 minutes. CSE may be observed in between 1%
and 30% of ICU patients.12-14 However, it is impractical and
unsafe to wait 30 minutes before initiating treatment and
because it is rare for generalized convulsive seizures to spon-
taneously cease after 5 minutes, some suggest that a duration
of 5 or 10 minutes may be a reasonable operational defini-
tion.15,16 NCSE is defined as an impairment of consciousness
248 Section III—Electrophysiology
or behavior from baseline state for at least 30 minutes without
convulsive movements, and the presence of one or more of
the following: (1) repetitive focal or generalized epileptiform
activity (spikes, sharp waves, spike-and-wave, sharp-and-slow
wave complexes) or rhythmic theta or delta activity at more
than two per second; (2) the EEG patterns described in 1 at
less than one per second, but with improvement after intrave-
nous (IV) AED injection (e.g., a benzodiazepine); or (3) a
temporal evolution of epileptiform or rhythmic activity
greater than 1 Hz with change in location or frequency over
time and normalization (either clinically or electrographi-
cally) in response to AED administration. If a patient experi-
ences frequent recurrent seizures without a clear return to
baseline mental status between seizures, then he or she is in
NCSE by definition. Many of the patients with NCSE have
subtle motor manifestations; these manifestations may be
focal or generalized.17,18 NCSE may be classified according to
patient age or subdivided into: (1) complex partial SE (CPSE)
or (2) generalized nonconvulsive SE.
Nonconvulsive seizures (NCS) are common in critically ill
patients, and many studies suggest that the majority of sei-
zures in these patients have limited clinical manifestations and
can only be recognized with EEG monitoring.19 After primary
brain injury, NCS occur in 11% to 55% of patients.20,21 The
variable prevalence of NCS reported in the studies may reflect
different injury types and different levels of sedation present
during the EEG recordings. The criteria for diagnosis are listed
in Table 25.1. The clinical spectrum of NCS is broad, and
without prolonged monitoring the majority of NCS will not
be identified. However, it is unclear what constitutes an
adequate duration of monitoring. Several studies suggest that
the continuous monitoring of EEG background activity is
Table 25.1  Diagnostic Criteria for a
Nonconvulsive Seizure
Any pattern lasting at least 10 seconds satisfying any one of
the following three primary criteria:
Primary criteria
Repetitive generalized or focal spikes, sharp waves,
spike-and-wave complexes at ≥3/second
Repetitive generalized or focal spikes, sharp waves,
spike-and-wave or sharp-and-slow wave complexes at convulsive seizure (GCS) (21.6%), or generalized status epi-
<3/second with a secondary criterion
Sequential rhythmic, periodic, or quasi-periodic waves at
≥1/second and clear evolution in frequency (increasing or
decreasing by at least 1/second), morphology, or location
(gradual spread into or out of a region that involves at
least two electrodes). Change in sharpness without
another morphology change is not enough to be
considered evolution in morphology. Evolution in
amplitude alone is not sufficient for diagnosis.
Secondary criterion
Significant improvement in clinical state or appearance of
previously absent normal EEG patterns (such as posterior-
dominant “alpha” rhythm) that is temporally associated
with administration of a rapidly acting AED. Resolution of
the “epileptiform” discharges that leaves diffuse slowing
without clinical improvement and without appearance of
previously absent normal EEG patterns would not satisfy
the secondary criterion.
Modified from Chong DJ, Hirsch LJ. Which EEG patterns warrant treatment in
the critically ill? Reviewing the evidence for treatment of periodic epileptiform
discharges and related patterns. J Clin Neurophysiol 2005;22:79–91.
AED, Antiepileptic drug; EEG, electroencephalogram.
necessary for at least 48 hours to detect 85% of the CSE and
NCSE in comatose ICU patients. Thus the latency for the
detection of NCSE or CSE might be as long as 48 to 72 hours,
or even up to 120 hours.10,22
Up to 50% of individuals with a history of generalized
tonic-clonic seizures who fail to return to baseline despite
adequate treatment have evidence of NCSE on EEG.23 There
is evidence that NCSE may exacerbate neuronal injury. For
example, NCSE is known to lead to significant hippocampal
atrophy after TBI indicative for severe neuronal injury.24 It
can also exacerbate midline shift in intracerebral hemorrhage
(ICH) patients.25 In addition, nonconvulsive electrographic
post-traumatic seizures have been observed to result in
increases in ICP and microdialysis markers of cellular distress
(e.g., lactate pyruvate ratio or glutamate).26,27
A variety of other periodic EEG patterns can be observed
in critically ill patients such as triphasic waves, periodic later-
alized epileptiform discharges (PLEDs), or burst suppression.
PLEDs are considered traditionally an interictal epileptiform
disturbance (Fig. 25.2). They are observed most frequently
with acute focal destructive lesions (e.g., infarct, infection, or
tumor). The presence of PLEDs may predict a delayed time to
first seizure among ICU patients who undergo cEEG monitor-
ing; that is, if PLEDs are present a longer duration of cEEG
may be indicated. A spontaneous burst-suppression pattern
indicates severe brain dysfunction particularly in comatose
patients.18 This condition is associated with bursts of cortical
activity that alternate with near complete suppression of back-
ground activity. The most common context in which burst
suppression occurs is in pharmacologically induced comas for
the control of elevated ICP (see following text) or refractory
status epilepticus.28 However, a nonpharmacologically induced
burst-suppression pattern may be seen in patients with severe
diffuse brain injury, particularly in severe diffuse anoxic brain
injury. In addition, focal or generalized nonseizure activity,
like the aforementioned rhythmic or periodic sharp tran-
sients, is frequently observed in critically ill patients. The inci-
dence of G-PED or PLED ranges between 50% and 58%.10,23
Patients with PEDs are at high risk of seizures. G-PEDS rep-
resent a rare phenomenon that was observed in 37 of 3000
EEG recordings only in one center.29 In this series, 89% of 37
patients had either a myoclonic seizure (35.2%), a generalized
lepticus (GSE) (32.4%) within 48 hours of the detection of
the G-PEDs. The occurrence of G-PEDS and its variants is
associated with subacute sclerosing panencephalitis (SSPE)
and sporadic Creutzfeldt-Jakob disease but most frequently
observed acutely in patients with diffuse anoxic brain injury.30
The application of high-frequency recording by means
of subdural grid electrodes allows the detection of self-
propagating waves of neural and glial depolarization. This
slow progressing wave that spreads at 2 to 5 mm/minute over
the cortex is known as cortical spreading depression (CSD).
In the normal brain the EEG shows a period of transient
depression of cortical electrical activity that is associated
with an increase in regional CBF and oxygen delivery (intact
neurovascular coupling). This phenomenon has a multifac-
torial etiology and is known to occur in migraine. Under
pathologic conditions, however, neurovascular coupling is
impaired and the depolarization wave causes a reduction in
rCBF (spreading ischemia, neurovascular uncoupling). The
increased energy demands imposed by CSD may worsen

Fig. 25.2  Left central temporal periodic lateralized epileptiform discharges (PLEDs) at approximately 1 Hz.
neuronal injury under these conditions.31 Thus after aneu-
rysmal SAH the number and duration of clusters of CSD are
temporally linked to the development of delayed infarct.32,33
The number of clusters is highly variable and may last
between 3 and 60 hours.
Electroencephalogram
Methodology
Duration of Electroencephalogram
Monitoring
The occurrence and time course of seizures, NCS, NCSE, and
PLEDs among others is highly variable and disease or pathol-
ogy dependent. Routine iEEG recordings, which last less than
30 minutes, fail to capture a seizure in more than half of all
seizing patients. However, up to 50% of the seizures occur
within the first 60 minutes of EEG recording2,34,35 and 80% of
comatose patients have seizures within 24 hours of monitor-
ing. Thus in comatose patients, exclusion of NCS or NCSE
requires at least 24 to 48 hours of recording time, whereas the
duration of routine iEEG should be at least 30 to 60 minutes
long for seizure detection. When EEG is used to detect isch-
emia and in particular DCI after aneurysmal SAH, monitoring
should continue for the whole period of risk, up to 14 days
after insult.36,37
Section III—Electrophysiology 249
Electrode Placement and Montage
Scalp pathology or other cranial injuries can limit electrode
placement38,39 and data quality. For monitoring of generalized
activity the standard 21 electrodes40 according to the 10-20
system may be adequate. A few studies41 suggest 10 electrode
positions and a structured 14-channel montage with referen-
tial and bipolar derivations for monitoring in patients with
TBI. For cEEG monitoring Labar et al.42 successfully used 5
electrodes and 2 channels to analyze trends in compressed
spectral arrays in SAH patients. However, it is generally
accepted that reducing the number of electrodes limits resolu-
tion of waveform morphology and distribution. Therefore
focal abnormalities may be more difficult to recognize. Fur-
thermore, full electrode configuration improves the ability to
distinguish brain signals from artifacts, aids in spatial localiza-
tion of pathologic activity, and provides redundancy.
Alternative methods are amplitude-integrated EEG (aEEG)
monitoring using two electrodes for seizure detection, a tech-
nique used in purpose-built devices that are common in neo-
natal ICUs. Most studies suggest a sensitivity of about 60%,
but in some studies using this technique a third or fewer of
seizures identified using conventional electrode arrangements
are detected with aEEG.43,44 Similarly, a six-electrode configu-
ration placed at the hairline and compared to a full electrode
montage may miss a third of seizures and up to one half of
250 Section III—Electrophysiology
PLEDs. Because ictal arrhythmias, including ictal asystole
and ictal bradycardia, are relatively common phenomena, a
channel dedicated to electrocardiographic monitoring, prefer-
ably using two noncephalic electrodes, is nearly always needed
and may reveal important changes in the cardiac rhythm that
may be seen in association with some seizures.45-47 A nonce-
phalic electrocardiographic channel can help distinguish
between certain artifacts and rhythmic EEG discharges. In
daily practice, skin breakdown can be a problem, particularly
when cup electrodes are used for prolonged periods. Alterna-
tively needle electrodes can be used, but there are risks associ-
ated with this such as needlesticks or broken subdermal wires.
Recently, intracortical electrodes that are placed stereotacti-
cally into deep brain regions have been used in clinical
research. Using this approach, Stuart et al., were able to dem-
onstrate the potential value of depth electrodes to detect DCI
associated with vasospasm in poor-grade SAH patients.48
Data Analysis
Recording of EEG background activity requires the inspec-
tion of the raw EEG. The visual analysis remains the stan-
dard method of EEG assessment. It is the best method to
Fig. 25.3  Changes in color spectral array correspond to left temporal seizure activity in this patient with recurrent left temporal complex partial
seizures.
recognize seizures and to detect reasonably rapid (seconds to
minutes) changes in EEG content on an expert level. The
obvious disadvantages are the time efforts needed for data
review and the low sensitivity to gradual trends in the EEG
(loss of faster frequencies or changes in variability of fre-
quency content). In addition, this method is highly subjec-
tive, and interpretation of EEG patterns can vary between
different electroencephalographers. Therefore visual analysis
of the raw EEG signal alone limits the application and bene-
ficial impact of cEEG monitoring technology in the ICU.49
Various approaches to process the raw EEG improve the
diagnostic accuracy and cut down the time needed for data
interpretation. The aEEG method depicts time-compressed
and rectified EEG amplitude derived from raw EEG on a
semi-logarithmic scale, whereas frequency domain methods
of EEG present an alternative approach of signal simplifica-
tion.50,51 The aEEG provides an accurate measure of EEG
background activity52 that can be useful for clinical applica-
tion. Color density spectral array (CDSA) applies fast-Fourier
transformation (FFT) to convert raw EEG into a time-
compressed and color-coded display (Fig. 25.3).53 Clinical
applications of quantitative EEG such as CDSA and related
techniques include monitoring the depth of sedation,54

detection of cerebral ischemia,55 and identification of sei-
zures in adults, children, and neonates.56,57 The use of cEEG
may be most useful in patients that require long-term moni-
toring, for example, after poor-grade SAH when patients are
at risk for both seizures and ischemia. Claassen et al. reported
that 19% of patients who had cEEG after poor-grade SAH
(n = 108) had seizures, and 95% of these seizures were NCS
without any detectable clinical correlate. The application of
quantitative analysis to the cEEG (relative alpha variability,
post-stimulation alpha-to-delta ratio [ADR]) enhances its
use in detection of DCI and increases its sensitivity for isch-
emia detection.55 However, this does not improve the sensi-
tivity for seizure detection, but its use can accelerate the
process of screening the EEG for seizures. When using quan-
titative cEEG, raw EEG data should be available for individ-
ual review. The application of video EEG recordings (vEEG)
can help increase the likelihood for seizure detection on the
ICU,58 particularly in patients with suspected NCSE.59
Electroencephalogram Artifacts
and Artifact Detection
Typical artifacts of the EEG are either non-neural physiologic
activities of the subject or caused by external technical
problems.60 Physiologic artifact sources are eye blinks, eye
movements, muscle activity in the vicinity of the head (e.g.,
face muscles, jaws, tongue, neck), heartbeat, pulse, and Mayer
waves. Most problematic in clinical practice are ocular (electro-
oculogram [EOG]) and muscle (electromyogram [EMG])
artifacts. EOG activity is caused by lateral eye movements or
by eye blinks (f = 20 times/min).61 This results in low-frequency
activity most prominent over the anterior head regions (f max
= 4 Hz), whereas EMG activity usually shows as high-frequency
activity (>20 Hz), with a wide range of amplitudes.62 A wide
variety of technical artifacts can be seen, including induction
artifact, electrostatic artifact, capacitative artifact, and 60 Hz
artifact. In particular, the presence of an electrical appliance
running on an AC circuit can create a 60-Hz artifact.60 Artifact
rejection remains challenging in practice where hand-
optimized, semiautomatic, and fully automatic approaches
exist.61 Semiautomatic approaches, although less time con-
suming, require user interaction for ambiguous or outlier
components.63 Fully automated methods are proposed for the
classification of eye artifacts but are used only in some non-
standardized research protocols.64 These methods, although
robust for the detection of eye artifacts, are not easily usable
for non–eye artifacts and may require the additional recording
of the EOG. Future developments include EEGLAB plug-ins
and support vector machine (SVM)–based systems that in
a fully automatic mode recognize and reject major artifacts
like eye blinks, eye movements, and heartbeats; the detection
of muscular or more subtle artifacts has not been reported
so far.65-67
Indications for
Electroencephalogram Monitoring
General Considerations
EEG is increasingly recognized as a useful tool to detect sei-
zures and seizure-like activity in different pathologies involv-
ing the central nervous system (CNS) and to serve as a
Section III—Electrophysiology 251
diagnostic tool for coma of unknown etiology. In addition,
EEG recordings help monitor and direct therapeutic maneu-
vers in the NCCU. This includes the monitoring of the burst-
suppression pattern during treatment of refractory intracranial
hypertension, the monitoring of treatment efficacy for NCSE
and CSE, and the detection of DCI.4,68-70
Traumatic Brain Injury
In TBI, seizures are identified more frequently when EEG is
performed than through clinical observation. However, the
exact incidence varies according to the methodology used—
AEDs, sedatives, and duration of monitoring. For example, in
a prospective study of 94 severe TBI patients who underwent
cEEG (on average for 7 days), Vespa et al. observed seizures in
22% of the subjects, half of which were NCS: 50% of seizures
occurred within day 1, despite phenytoin prophylaxis.71 Brain
contusions and skull fractures were not associated with
increased seizures. When no antiepileptic prophylaxis was
given, Ronne-Engstrom and Winkler observed NCS in 28%
and NCSE in 11% of TBI patients (n = 70) who had cEEG
monitoring more than 24 hours.72 NCS were associated with
extracerebral or intracerebral hematomas and older age (>60
years). The majority of these events lasted less than 30 minutes
and all responded to benzodiazepines and phenytoin. On the
other hand, Amantini et al. observed far fewer seizures in TBI
patients (3%). These differences may be associated with use
of a much greater dose of medications such as midazolam,
propofol, thiopental, or a combination, that can have antisei-
zure activity.73
EEG recordings have been used to predict outcome in TBI
patients. Several factors are associated with outcome, includ-
ing (1) an increase or decrease in the absolute and relative
amplitudes in the alpha and theta bands in the EEG can help
differentiate survivors from nonsurvivors74; (2) a delta EEG
pattern is associated with death and poor functional outcome75;
(3) a reduced percentage of alpha variability is associated with
poor prognosis41; and (4) the presence of EEG reactivity to
painful stimuli is associated with a good outcome.76 However,
prognostic errors based on EEG may be observed in 20% of
patients.74
Aneurysmal Subarachnoid Hemorrhage
The incidence of seizures after aneurysmal SAH (aSAH) varies
from 4% to 24%. Most cases occur at the time of aneurysm
rupture or shortly after the initial hemorrhage.77 Risk factors
for seizures include a history of epilepsy, middle cerebral
artery aneurysms, a cerebral infarct, and an intracerebral
hematoma.78 NCS also are identified after SAH. The incidence
varies with the method of detection. For example, the inci-
dence of NCS is 3%, when using repeated standard (i.e.,
iEEG).79 When using cEEG, and in particular for more than
24 hours, seizures are detected in 19% of patients, and the
majority (95%) of them are NCS.34 Up to 8% of patients have
NCSE, despite antiepileptic prophylaxis (fosphenytoin).80 Risk
factors for NCSE include (1) older age, (2) female sex, (3)
poor-grade SAH, (4) the amount of cisternal blood, (5) hydro-
cephalus, and (6) the presence of structural brain lesions
including focal or global cerebral edema. NCSE can be refrac-
tory to therapy, and many of these patients have a poor
outcome. In many of these series cEEG was used in patients
252 Section III—Electrophysiology
Bilateral frontal ischemia — decrease in alpha variability
Fig. 25.4  Decrease in alpha variability in subarachnoid hemorrhage (SAH) associated with bifrontal ischemia.
Table 25.2  Relationship Between Cerebral
Blood Flow and Electroencephalogram
Changes
CBF mL/100 g/min EEG Change Reversibility
35-70 Normal No injury
25-35 Loss of beta Reversible
18-25 Theta slowing Reversible
12-18 Delta slowing Reversible
<8-10 Suppression Irreversible
CBF, Cerebral blood flow; EEG, electroencephalogram.
with convulsive seizure activity, unexplained neurologic dete-
rioration, or persistent and unexplained coma, but NCS are
still common even without any detectable clinical correlate.34
EEG including iEEG, aEEG, and cEEG recordings can help
detect DCI.42,81-83 For example, Rivierez et al. studied 151
SAH patients on day 1 and day 5 after SAH with standard
EEG, and found that patients (n = 78) with bilateral bursts of
slow waves were more likely to have DCI.81 Artifacts and the
need for manual postprocessing of the raw data can limit the
use of EEG to detect DCI. However, quantitative EEG analy-
sis (qEEG) holds promise as a method to detect DCI (Fig.
25.4). A decrease in the ADR extracted from the cEEG shows
a strong association with DCI in poor-grade SAH patients.83
The median decrease of ADR for patients with DCI was 24%,
whereas an increase of 3% was observed in patients without
DCI associated with vasospasm.42,82,83 In a cohort of 32
mainly good-grade patients monitored with cEEG and trend-
ing of a quantitative measure, the relative alpha variability,
Vespa et al.82 found that in all 19 patients with angiographi-
cally proven vasospasm, relative alpha variability was signifi-
cantly decreased and preceded the diagnosis of vasospasm by
2 to 3 days. Table 25.2 lists the relation between EEG changes
and the corresponding changes in regional cerebral blood
flow (rCBF); these changes can be used to help differentiate
reversible from irreversible ischemia. When the EEG wave-
form pattern is suppressed, rCBF is less than 8 to 10 mL/100 g/
minute and indicates irreversible tissue damage.
In 2009, the occurrence of CSD after aSAH was demon-
strated.84,85 This may be present in more than 70% of the
patients studied and often is associated with epileptic seizures.
There is growing evidence that the presence of CSD, the
number of CSD clusters, and their duration may precede or
lead to DCI and potentially infarction.31,86
There are conflicting data on the effect of seizures on SAH
patients’ outcome.79,87 Relatively limited data exist on the
prognostic value of EEG after SAH; the presence of PEDs, the
absence of normal sleep architecture and reactivity to stimuli,
and NCSE on cEEG are associated with poor outcome in this
population.78
Intracerebral Hemorrhage
Seizures can be identified in 31% of patients with ICH. NCS
occur in between 10% and 30% of patients and may be
detected within 48 hours of cEEG initiation, regardless of
antiepileptic phrophylaxis.25,88 This information may be
important because there is a described independent associa-
tion between phenytoin prophylaxis and worse cognitive
outcome after ICH.89
Hypoxic-Ischemic Encephalopathy
EEG can be used for seizure and NCS detection and for
outcome prediction after hypoxic-ischemic encephalopathy
(HIE). In adult cardiac arrest (CA) patients, EEG seizures are
identified in 17% to 36% of patients.13,90,91 Electrographic sei-
zures also are observed when therapeutic hypothermia (TH)
is used and occur both during TH and after rewarming.92,93
Because TH and sedation have antiepileptic properties, the
occurrence of electrical seizures during TH may reflect wide-
spread severe brain injury, and prolonged seizures that occur

during TH are associated with poor outcome.91,94 However, a
subset of patients who develop postanoxic seizures can have a
favorable outcome.95
A dichotomized scaling of postanoxic EEG patterns, “malig-
nant” versus “benign” was described before the use of TH.
Malignant patterns include nonreactive background and
spontaneous burst suppression and are a powerful EEG
pattern (false-positive rate, 0%) to identify patients with poor
outcome.96 For example, a nonreactive EEG pattern is associ-
ated with in-hospital mortality,91,97 whereas the presence of a
reactive EEG background98 or continuous activity90 predicts
awakening from postanoxic coma. These data, however, are
not validated in patients who receive TH. In the posthypother-
mia era, there are also some data to suggest that the absence
of an N20 cortical response on somatosensory evoked poten-
tials (SSEPs) in a postanoxic patient is still associated with a
very poor prognosis.91
In neonates with HIE, therapeutic hypothermia can help
improve clinical outcome.99 EEG can be used to detect seizures
during and after therapeutic hypothermia. For example,
Wusthoff et al. observed seizures in 65% of neonates (n = 26)
with HIE who were studied; NCS were present in 47% and SE
in 23%. Seizure onset was observed within 48 hours in 76%
of the patients.100 Unlike in adults, the background pattern
does not enhance the predictive value of EEG among infants
with HIE.101
Brain Tumors
Brain tumors are a common cause of epilepsy, accounting for
10% to 15% of all causes of seizures in adult patients.102
However, the exact incidence of seizures in patients with brain
tumors admitted to the ICU is poorly understood, in part
because cEEG monitoring often is not part of a prospective
protocol but rather is obtained for suspected cases based on
clinical judgment. Among 158 patients admitted to an NCCU,
Bladin et al. observed seizures in patients, 6 of whom had
brain tumors. Most of the patients (5/6) had clinical sei-
zures.103 Claassen et al. observed seizures in 23% of patients
with brain tumors admitted to an NCCU who had cEEG; all
presented as NCS.34 Seizures were detected during the first 24
hours of monitoring in 88% of patients, whereas cEEG
detected seizures more than 24 hours after cEEG monitoring
was started in patients who were comatose at the time moni-
toring began. Patients with seizures: (1) were younger, (2)
were more likely to be comatose, and (3) had a more frequent
history of seizures than patients without seizures at cEEG. It
is unclear if risk factors for seizure development in all brain
tumors (e.g., cortical localization or low-grade tumors) may
have the same effect on seizure occurrence in those patients
admitted in the ICU. Therefore EEG monitoring may be con-
sidered in brain tumor patients with altered mental status or
coma when admitted to the ICU. The duration of monitoring
should be 24 hours in noncomatose patients without evidence
of ictal activity and more than 48 to 72 hours in comatose
patients.
Meningoencephalitis
Seizures can complicate bacterial, viral, or tuberculous men-
ingitis or encephalitis.104 Malaria also can cause seizures. In a
prospective cohort of adult acute bacterial meningitis (ABM)
Section III—Electrophysiology 253
in Europe, Zoons et al. observed seizures in 17% of patients.
These were associated with severe CNS and systemic inflam-
mation, structural CNS lesions and pneumococcal meningitis.
Overall mortality rate of patients with epilepsy was 41%.105 In
a retrospective review of NCCU patients who had cEEG,
Claassen et al. observed that 29% of those with CNS infections
had seizures. Of those, half were NCS, which accounted for
10% of all NCS in this population.34 Other studies show that
6% to 12% of adult patients with ABM have seizures, but the
relationship between outcome and seizures is unclear.106,107 In
infants with bacterial meningitis, abnormal EEG findings,
including a continuous low voltage or flat trace on aEEG, EEG
background activity and overall EEG description are associ-
ated with adverse outcome.108 Infants with normal or mildly
abnormal EEGs generally have a good neurologic status at
discharge.109
Sepsis and Metabolic Disorders
Multiple-organ failure can lead to seizures, and NCS may be
particularly common in these patients.110 For example, Oddo
et al., in a retrospective analysis of 201 patients without a
history of brain injury who were admitted to a medical ICU,
observed NCS in 10% of the patients.36 Sepsis was an inde-
pendent risk factor for NCS. In patients with liver failure, EEG
can help detect seizures, particularly NCS, which are frequent
in this setting.111 After toxic-metabolic encephalopathy, NCS
are observed in 20% of patients admitted to the ICU who
undergo cEEG.34
Electroencephalographic
Monitoring to Guide Therapy
The recording of EEG background activity can be used to
assess and guide the depth of sedation in patients with
pharmacologically induced coma, for example, the induction
of “burst-suppression-EEG” to control ICP and or control
refractory status epilepticus (RSE). In these patients monitor-
ing is recommended to guide dosing, with careful titration to
maintain a sufficient burst-suppression pattern on the EEG
while minimizing adverse drug effects. It is not clear the
number of bursts per minute considered standard; however,
values of 3 to 5 or 4 to 6 bursts per minute are targeted during
barbiturate therapy. Alternatively, the bispectral index (BIS)
monitor may be used to titrate barbiturate therapy because
the BIS and suppression ratio (SR) values correlate with the
standard EEG-based method.112-116
EEG also may serve, in combination with other studies and
clinical exam findings, as a surrogate marker of clinical
outcome in patients after CA and after TH, although current
practice parameters do not recommend the use of EEG alone
for predicting prognosis. In cardiac arrest patients, a continu-
ous or reactive EEG background is associated with a good
recovery.117 On the other hand, irreversible coma is usually
inevitable with a spontaneous burst-suppression EEG pattern,
an unresponsive EEG pattern, or absent N20 on SSEPs.117,118
Similar observations have been made in children.119
Refractory Status Epilepticus
Status epilepticus resistant to benzodiazepines and at least
one antiepileptic drug is defined as refractory, and guidelines
254 Section III—Electrophysiology
advise coma induction with midazolam, propofol, or barbitu-
rates.120 Coma induction itself may cause severe side effects.37
So far, there are no randomized trials that show an outcome
benefit for patients with RSE who undergo EEG monitoring;
the data are derived from prospective cohort trials, observa-
tional databases, and expert opinion. Nevertheless, iEEG
and cEEG can help detect seizure activity and manage the level
of sedation. cEEG monitoring in RSE helps obtain early
seizure suppression, maintain burst-suppression pattern if it
is refractory, and withdraw anesthetic drugs safely while mon-
itoring the possibility of seizure reappearance.3 It is not clear
whether complete suppression of cerebral electrogenesis,
simple seizure suppression, or burst-suppression pattern is
more effective.121,122 Some suggest it is advisable to maintain
a burst-to-suppression ratio of approximately 1 : 4. Others
advise a burst-suppression pattern of about 10 seconds of
suppression for each burst of activity, ideally for at least 24
hours.37,120,123 Barbiturates tend to be associated with higher
likelihood of achieving burst suppression on EEG than pro-
pofol or midazolam, but their use is associated with longer
mechanical ventilation and hospital stay.121
Monitoring of Sedation Depth
and Burst Suppression
The BIS monitor was developed to assess the hypnotic com-
ponent of general anesthesia.124 A good correlation between
the cEEG burst rate per minute and the BIS and suppression
ratio to assess the depth of propofol anesthesia in the treat-
ment of RSE is described, but the BIS monitor does recognize
regional epileptic activity and epileptic activity during the
burst-suppression pattern. Therefore cEEG should be consid-
ered the primary monitoring technique to assess the depth of
anesthesia in RSE. If cEEG is not available, a BIS monitor may
be considered an alternative to assess the level of sedation
although its use in RSE still is controversial.69,124 BIS monitor-
ing has been used to guide sedation depth in patients with
severe trauma. However, despite a high specificity (90%) for
the detection of sedation changes, the interpretation of the BIS
readings remains subjective and appears to be even more
complex in TBI patients.125
Logistics and Costs
The routine application of EEG monitoring and especially
continuous monitoring in the NCCU is still limited by the
high costs for technical equipment and personnel. Because
high-quality data recording and interpretation are necessary
for any clinical application, a specialized EEG technician and
an experienced electroencephalographer are essential.126 In
addition the integration of EEG monitoring into ICU care
may trigger EEG-driven clinical decision making and there-
fore reduce the need for other diagnostic or imaging modali-
ties.4 Therefore cEEG implementation requires specific
training for nurses, technicians, and physicians.6 Because any
intervention can influence the EEG signal and there are
numerous sources for artifacts, the medical staff at bedside
should be trained to be able to detect and interpret common
artifacts and to recognize basic EEG patterns and their dynam-
ics. Developments such as magnetic resonance imaging
(MRI)–compatible electrodes127 and smaller recording and
data storage systems and Web-based data access and data
Table 25.3  Advantages and Disadvantages
of Electroencephalography as a Monitor in
the Intensive Care Unit
Advantages Disadvantages
Sensitive to changes in Nonspecific as to etiology of
neurologic functioning abnormality technical issues
from variety of causes Difficult to implement
Structural Requires skilled personnel
Metabolic 24 hr/day
Large amount of data (data
storage and review)
Good spatial resolution Susceptible to artifacts
Monitors many brain areas Electrical
simultaneously Eye movements, ECG, EMG
Excellent temporal resolution Electrode placement
2-4 ms
Bedside, noninvasive
ECG, Electrocardiogram; EMG, electromyogram.
review tools will further facilitate the integration of EEG mon-
itoring in the ICU. In addition, there are efforts to develop
practical definitions of clinical relevant EEG findings in the
ICU.18 Table 25.3 summarizes the typical advantages and dis-
advantages of EEG methodology as a monitor in the ICU.
Conclusion
In summary, EEG and especially cEEG is a promising and
evolving technique to assess neurologic status in the critically
ill. It allows the noninvasive assessment of brain function over
a large surface area. Based on the knowledge gathered so far,
EEG should be considered in those with acute brain injury,
impaired mental status, and unexplained alteration in con-
sciousness. Bedside EEG can help titrate therapy and repre-
sents a valuable tool to detect secondary insults, such as
cerebral ischemia in at-risk patients. However, EEG monitor-
ing in the ICU is still an expensive cost-intensive technology
that requires well-trained, dedicated medical and technical
staff. Continued education and research are necessary to
determine whether EEG leads to management and treatment
decisions that are associated with outcomes. Advances in
information technology may soon make real-time bedside
monitoring of brain activity using cEEG possible.
References
1. Tudor M, Tudor L, Tudor KI. [Hans Berger (1873-1941)—The history of
electroencephalography]. Acta Med Croatica 2005;59(4):307–13.
2. Claassen J, Mayer SA. Continuous electroencephalographic monitoring in
neurocritical care. Curr Neurol Neurosci Rep 2002;2:534–40.
3. Friedman D, Claassen J, Hirsch LJ. Continuous electroencephalogram
monitoring in the intensive care unit. Anesth Analg 2009;109:506–23.
4. Vespa PM, Nenov V, Nuwer MR. Continuous EEG monitoring in the
intensive care unit: early findings and clinical efficacy. J Clin Neurophysiol
1999;16:1–13.
5. Horst ter HJ, van Olffen M, Remmelts HJ, et al. The prognostic value of
amplitude integrated EEG in neonatal sepsis and/or meningitis. Acta
Paediatr 2010;99:194–200.
6. Kull LL, Emerson RG. Continuous EEG monitoring in the intensive care
unit: technical and staffing considerations. J Clin Neurophysiol 2005;22:
107–18.
7. Brenner RP, Schaul N. Periodic EEG patterns: classification, clinical
correlation, and pathophysiology. J Clin Neurophysiol 1990;7:249–67.

8. Andraus MEC, Andraus CF, Alves-Leon SV. Periodic EEG patterns:
importance of their recognition and clinical significance. Arq Neuropsiquiatr
2012;70:145–51.
9. Holtkamp M, Meierkord H. Nonconvulsive status epilepticus: a diagnostic
and therapeutic challenge in the intensive care setting. Ther Adv Neurol
Disord 2011;4:169–81.
10. García-Morales I, García MT, Galán-Dávila L, et al. Periodic lateralized
epileptiform discharges: etiology, clinical aspects, seizures, and evolution in
130 patients. J Clin Neurophysiol 2002;19:172–7.
11. Snodgrass SM, Tsuburaya K, Ajmone-Marsan C. Clinical significance of
periodic lateralized epileptiform discharges: relationship with status
epilepticus. J Clin Neurophysiol 1989;6:159–72.
11a.  International League Against Epilepsy. Proposal for revised clinical
and electroencephalographic classification of epileptic seizures. From the
Commission on Classification and Terminology of the International League
Against Epilepsy. Epilepsia 1981;22(4):489–501.
12. Rossetti AO, Kaplan PW. Seizure semiology: an overview of the ‘inverse
problem’. Eur Neurol 2010;63:3–10.
13. Rossetti AO, Urbano LA, Delodder F, et al. Prognostic value of continuous
EEG monitoring during therapeutic hypothermia after cardiac arrest. Crit
Care 2010;14:R173.
14. Meierkord H, Holtkamp M. Status epilepticus: an independent outcome
predictor after cerebral anoxia. Neurology 2008;70:2015–6; author reply
2015–6.
15. Alldredge BK, Gelb AM, Isaacs SM, et al. A comparison of lorazepam,
diazepam, and placebo for the treatment of out-of-hospital status
epilepticus. N Engl J Med 2001;345:631–7.
16. Treiman DM, Meyers PD, Walton NY, et al. A comparison of four treatments
for generalized convulsive status epilepticus. N Engl J Med 1998;339:792–8.
17. Brenner RP. The interpretation of the EEG in stupor and coma. Neurologist
2005;11:271–84.
18. Hirsch LJ, Brenner RP, Drislane FW, et al. The ACNS Subcommittee On
Research Terminology for Continuous EEG Monitoring. Proposed
standardized terminology for rhythmic and periodic EEG patterns
encountered in critically ill patients. J Clin Neurophysiol 2005;22:128–35.
19. Jirsch J, Hirsch LJ. Nonconvulsive seizures: developing a rational approach to
the diagnosis and management in the critically ill population. Clin
Neurophysiol 2007;118:1660–70.
20. Jordan KG. Neurophysiologic monitoring in the neuroscience intensive care
unit. Neurol Clin 1995;13:579–626.
21. Lowenstein DH, Aminoff MJ. Clinical and EEG features of status epilepticus
in comatose patients. Neurology 1992;42:100–4.
22. Pohlmann-Eden B, Hoch DB, Cochius JI, et al. Periodic lateralized
epileptiform discharges: a critical review. J Clin Neurophysiol
1996;13:519–30.
23. Walker M, Cross H, Smith S, et al. Nonconvulsive status epilepticus:
Epilepsy Research Foundation workshop reports. Epileptic Disord
2005;7(3):253–96.
24. Vespa PM, McArthur DL, Xu Y, et al. Nonconvulsive seizures after traumatic
brain injury are associated with hippocampal atrophy. Neurology 2010;75:
792–8.
25. Vespa PM, O’Phelan K, Shah M, et al. Acute seizures after intracerebral
hemorrhage: a factor in progressive midline shift and outcome. Neurology
2003;60:1441–6.
26. Vespa P, Martin NA, Nenov V, et al. Delayed increase in extracellular
glycerol with post-traumatic electrographic epileptic activity: support for
the theory that seizures induce secondary injury. Acta Neurochir Suppl 2002;
81:355–7.
27. Vespa PM, Miller C, McArthur D, et al. Nonconvulsive electrographic
seizures after traumatic brain injury result in a delayed, prolonged increase
in intracranial pressure and metabolic crisis. Crit Care Med 2007;35(12):
2830–6.
Section III—Electrophysiology 255
28. Oertel M, Kelly DF, Lee JH, et al. Metabolic suppressive therapy as a
treatment for intracranial hypertension: why it works and when it fails. Acta
Neurochir Suppl 2002;81:69–70.
29. Yemisci M, Gurer G, Saygi S, et al. Generalised periodic epileptiform
discharges: clinical features, neuroradiological evaluation and prognosis in
37 adult patients. Seizure 2003;12:465–72.
30. Chong DJ, Hirsch LJ. Which EEG patterns warrant treatment in the critically
ill? Reviewing the evidence for treatment of periodic epileptiform discharges
and related patterns. J Clin Neurophysiol 2005;22:79–91.
31. Strong AJ, Hartings JA, Dreier JP. Cortical spreading depression: an adverse
but treatable factor in intensive care? Curr Opin Crit Care 2007;13:126–33.
32. Dreier JP: The role of spreading depression, spreading depolarization and
spreading ischemia in neurological disease. Nat Med 2011;1–9.
33. Fabricius M, Fuhr S, Willumsen L, et al. Association of seizures with cortical
spreading depression and peri-infarct depolarisations in the acutely injured
human brain. Clin Neurophysiol 2008;119:1973–84.
34. Claassen J, Mayer SA, Kowalski RG, et al. Detection of electrographic
seizures with continuous EEG monitoring in critically ill patients. Neurology
2004;62:1743–8.
35. Hirsch LJ. Classification of EEG patterns in patients with impaired
consciousness. Epilepsia 2011;52 (Suppl. 8):S21–4.
36. Oddo M, Carrera E, Claassen J, et al. Continuous electroencephalography in
the medical intensive care unit. Crit Care Med 2009;37:2051–6.
37. Rossetti AO, Oddo M. The neuro-ICU patient and electroencephalography
paroxysms: if and when to treat. Curr Opin Crit Care 2010;16:105–9.
38. Jordan D, Schneider G, Hock A, et al. EEG parameters and their
combination as indicators of depth of anaesthesia. Biomed Tech (Berl)
2006;51:89–94.
39. Jordan KG. Continuous EEG monitoring in the neuroscience intensive care
unit and emergency department. J Clin Neurophysiol 1999;16:14–39.
40. Klem GH, Lüders HO, Jasper HH, et al. The ten-twenty electrode system
of the International Federation. The International Federation of Clinical
Neurophysiology. Electroencephalogr Clin Neurophysiol Suppl 1999;52:3–6.
41. Vespa PM, Boscardin WJ, Hovda DA, et al. Early and persistent impaired
percent alpha variability on continuous electroencephalography monitoring
as predictive of poor outcome after traumatic brain injury. J Neurosurg
2002;97:84–92.
42. Labar DR, Fisch BJ, Pedley TA, et al. Quantitative EEG monitoring for
patients with subarachnoid hemorrhage. Electroencephalogr Clin
Neurophysiol 1991;78:325–32.
43. Wusthoff CJ, Shellhaas RA, Clancy RR. Limitations of single-channel EEG
on the forehead for neonatal seizure detection. J Perinatol 2009;29:237–42.
44. Shellhaas RA, Soaita AI, Clancy RR. Sensitivity of amplitude-integrated
electroencephalography for neonatal seizure detection. Pediatrics
2007;120:770–7.
45. Opherk C, Coromilas J, Hirsch LJ. Heart rate and EKG changes in 102
seizures: analysis of influencing factors. Epilepsy Res 2002;52:117–27.
46. Opherk C, Hirsch LJ. Ictal heart rate differentiates epileptic from non-
epileptic seizures. Neurology 2002;58:636–8.
47. Zijlmans M, Flanagan D, Gotman J. Heart rate changes and ECG
abnormalities during epileptic seizures: prevalence and definition of an
objective clinical sign. Epilepsia 2002;43:847–54.
48. Stuart RM, Waziri A, Weintraub D, et al. Intracortical EEG for the detection
of vasospasm in patients with poor-grade subarachnoid hemorrhage.
Neurocrit Care 2010;13:355–8.
49. Scheuer ML, Wilson SB. Data analysis for continuous EEG monitoring in the
ICU: seeing the forest and the trees. J Clin Neurophysiol 2004;21:353–78.
50. Rampil IJ. Elements of EEG signal processing. Int J Clin Monit Comput
1987;4:85–98.
A complete list of references for this chapter can be found online at
www.expertconsult.com.

References
1. Tudor M, Tudor L, Tudor KI. [Hans Berger (1873-1941)—The history of
electroencephalography]. Acta Med Croatica 2005;59(4):307–13.
2. Claassen J, Mayer SA. Continuous electroencephalographic monitoring in
neurocritical care. Curr Neurol Neurosci Rep 2002;2:534–40.
3. Friedman D, Claassen J, Hirsch LJ. Continuous electroencephalogram
monitoring in the intensive care unit. Anesth Analg 2009;109:506–23.
4. Vespa PM, Nenov V, Nuwer MR. Continuous EEG monitoring in the
intensive care unit: early findings and clinical efficacy. J Clin Neurophysiol
1999;16:1–13.
5. Horst ter HJ, van Olffen M, Remmelts HJ, et al. The prognostic value of
amplitude integrated EEG in neonatal sepsis and/or meningitis. Acta
Paediatr 2010;99:194–200.
6. Kull LL, Emerson RG. Continuous EEG monitoring in the intensive care
unit: technical and staffing considerations. J Clin Neurophysiol 2005;22:
107–18.
7. Brenner RP, Schaul N. Periodic EEG patterns: classification, clinical
correlation, and pathophysiology. J Clin Neurophysiol 1990;7:249–67.
8. Andraus MEC, Andraus CF, Alves-Leon SV. Periodic EEG patterns:
importance of their recognition and clinical significance. Arq
Neuropsiquiatr 2012;70:145–51.
9. Holtkamp M, Meierkord H. Nonconvulsive status epilepticus: a diagnostic
and therapeutic challenge in the intensive care setting. Ther Adv Neurol
Disord 2011;4:169–81.
10. García-Morales I, García MT, Galán-Dávila L, et al. Periodic lateralized
epileptiform discharges: etiology, clinical aspects, seizures, and evolution in
130 patients. J Clin Neurophysiol 2002;19:172–7.
11. Snodgrass SM, Tsuburaya K, Ajmone-Marsan C. Clinical significance of
periodic lateralized epileptiform discharges: relationship with status
epilepticus. J Clin Neurophysiol 1989;6:159–72.
11a.  International League Against Epilepsy. Proposal for revised clinical
and electroencephalographic classification of epileptic seizures. From the
Commission on Classification and Terminology of the International League
Against Epilepsy. Epilepsia 1981;22(4):489–501.
12. Rossetti AO, Kaplan PW. Seizure semiology: an overview of the ‘inverse
problem’. Eur Neurol 2010;63:3–10.
13. Rossetti AO, Urbano LA, Delodder F, et al. Prognostic value of continuous
EEG monitoring during therapeutic hypothermia after cardiac arrest. Crit
Care 2010;14:R173.
14. Meierkord H, Holtkamp M. Status epilepticus: an independent outcome
predictor after cerebral anoxia. Neurology 2008;70:2015–6; author reply
2015–6.
15. Alldredge BK, Gelb AM, Isaacs SM, et al. A comparison of lorazepam,
diazepam, and placebo for the treatment of out-of-hospital status
epilepticus. N Engl J Med 2001;345:631–7.
16. Treiman DM, Meyers PD, Walton NY, et al. A comparison of four
treatments for generalized convulsive status epilepticus. N Engl J Med
1998;339:792–8.
17. Brenner RP. The interpretation of the EEG in stupor and coma. Neurologist
2005;11:271–84.
18. Hirsch LJ, Brenner RP, Drislane FW, et al. The ACNS Subcommittee
on Research Terminology for Continuous EEG Monitoring. Proposed
standardized terminology for rhythmic and periodic EEG patterns
encountered in critically ill patients. J Clin Neurophysiol 2005;22:
128–35.
19. Jirsch J, Hirsch LJ. Nonconvulsive seizures: developing a rational approach
to the diagnosis and management in the critically ill population. Clin
Neurophysiol 2007;118:1660–70.
20. Jordan KG. Neurophysiologic monitoring in the neuroscience intensive care
unit. Neurol Clin 1995;13:579–626.
21. Lowenstein DH, Aminoff MJ. Clinical and EEG features of status
epilepticus in comatose patients. Neurology 1992;42:100–4.
22. Pohlmann-Eden B, Hoch DB, Cochius JI, et al. Periodic lateralized
epileptiform discharges: a critical review. J Clin Neurophysiol 1996;13:
519–30.
23. Walker M, Cross H, Smith S, et al. Nonconvulsive status epilepticus:
Epilepsy Research Foundation workshop reports. Epileptic Disord
2005;7(3):253–96.
24. Vespa PM, McArthur DL, Xu Y, et al. Nonconvulsive seizures after
traumatic brain injury are associated with hippocampal atrophy. Neurology
2010;75:792–8.
25. Vespa PM, O’Phelan K, Shah M, et al. Acute seizures after intracerebral
hemorrhage: a factor in progressive midline shift and outcome. Neurology
2003;60:1441–6.
Section III—Electrophysiology 255.e1
26. Vespa P, Martin NA, Nenov V, et al. Delayed increase in extracellular
glycerol with post-traumatic electrographic epileptic activity: support for
the theory that seizures induce secondary injury. Acta Neurochir Suppl
2002;81:355–7.
27. Vespa PM, Miller C, McArthur D, et al. Nonconvulsive electrographic
seizures after traumatic brain injury result in a delayed, prolonged increase
in intracranial pressure and metabolic crisis. Crit Care Med 2007;35(12):
2830–6.
28. Oertel M, Kelly DF, Lee JH, et al. Metabolic suppressive therapy as a
treatment for intracranial hypertension: why it works and when it fails.
Acta Neurochir Suppl 2002;81:69–70.
29. Yemisci M, Gurer G, Saygi S, et al. Generalised periodic epileptiform
discharges: clinical features, neuroradiological evaluation and prognosis in
37 adult patients. Seizure 2003;12:465–72.
30. Chong DJ, Hirsch LJ. Which EEG patterns warrant treatment in the
critically ill? Reviewing the evidence for treatment of periodic epileptiform
discharges and related patterns. J Clin Neurophysiol 2005;22:79–91.
31. Strong AJ, Hartings JA, Dreier JP. Cortical spreading depression: an adverse
but treatable factor in intensive care? Curr Opin Crit Care 2007;13:126–33.
32. Dreier JP. The role of spreading depression, spreading depolarization and
spreading ischemia in neurological disease. Nat Med 2011;1–9.
33. Fabricius M, Fuhr S, Willumsen L, et al. Association of seizures with
cortical spreading depression and peri-infarct depolarisations in the acutely
injured human brain. Clin Neurophysiol 2008;119:1973–84.
34. Claassen J, Mayer SA, Kowalski RG, et al. Detection of electrographic
seizures with continuous EEG monitoring in critically ill patients.
Neurology 2004;62:1743–8.
35. Hirsch LJ. Classification of EEG patterns in patients with impaired
consciousness. Epilepsia 2011;52 (Suppl. 8):S21–4.
36. Oddo M, Carrera E, Claassen J, et al. Continuous electroencephalography
in the medical intensive care unit. Crit Care Med 2009;37:2051–6.
37. Rossetti AO, Oddo M. The neuro-ICU patient and electroencephalography
paroxysms: if and when to treat. Curr Opin Crit Care 2010;16:105–9.
38. Jordan D, Schneider G, Hock A, et al. EEG parameters and their
combination as indicators of depth of anaesthesia. Biomed Tech (Berl)
2006;51:89–94.
39. Jordan KG. Continuous EEG monitoring in the neuroscience intensive care
unit and emergency department. J Clin Neurophysiol 1999;16:14–39.
40. Klem GH, Lüders HO, Jasper HH, et al. The ten-twenty electrode system of
the International Federation. The International Federation of Clinical
Neurophysiology. Electroencephalogr Clin Neurophysiol Suppl 1999;52:3–6.
41. Vespa PM, Boscardin WJ, Hovda DA, et al. Early and persistent impaired
percent alpha variability on continuous electroencephalography monitoring
as predictive of poor outcome after traumatic brain injury. J Neurosurg
2002;97:84–92.
42. Labar DR, Fisch BJ, Pedley TA, et al. Quantitative EEG monitoring for
patients with subarachnoid hemorrhage. Electroencephalogr Clin
Neurophysiol 1991;78:325–32.
43. Wusthoff CJ, Shellhaas RA, Clancy RR. Limitations of single-channel EEG
on the forehead for neonatal seizure detection. J Perinatol 2009;29:237–42.
44. Shellhaas RA, Soaita AI, Clancy RR. Sensitivity of amplitude-integrated
electroencephalography for neonatal seizure detection. Pediatrics 2007;120:
770–7.
45. Opherk C, Coromilas J, Hirsch LJ. Heart rate and EKG changes in 102
seizures: analysis of influencing factors. Epilepsy Res 2002;52:117–27.
46. Opherk C, Hirsch LJ. Ictal heart rate differentiates epileptic from
non-epileptic seizures. Neurology 2002;58:636–8.
47. Zijlmans M, Flanagan D, Gotman J. Heart rate changes and ECG
abnormalities during epileptic seizures: prevalence and definition of an
objective clinical sign. Epilepsia 2002;43:847–54.
48. Stuart RM, Waziri A, Weintraub D, et al. Intracortical EEG for the detection
of vasospasm in patients with poor-grade subarachnoid hemorrhage.
Neurocrit Care 2010;13:355–8.
49. Scheuer ML, Wilson SB. Data analysis for continuous EEG monitoring in
the ICU: seeing the forest and the trees. J Clin Neurophysiol 2004;21:
353–78.
50. Rampil IJ. Elements of EEG signal processing. Int J Clin Monit Comput
1987;4:85–98.
51. Rampil IJ. A primer for EEG signal processing in anesthesia. Anesthesiology
1998;89:980–1002.
52. Naqeeb al N, Edwards AD, Cowan FM, et al. Assessment of neonatal
encephalopathy by amplitude-integrated electroencephalography. Pediatrics
1999;103:1263–71.
53. Wilson SB, Scheuer ML, Emerson RG, et al. Seizure detection: evaluation of
the Reveal algorithm. Clin Neurophysiol 2004;115:2280–91.
255.e2 Section III—Electrophysiology
54. Bricolo A, Turazzi S, Faccioli F, et al. Clinical application of compressed
spectral array in long-term EEG monitoring of comatose patients.
Electroencephalogr Clin Neurophysiol 1978;45:211–25.
55. Claassen J, Mayer SA, Hirsch LJ. Continuous EEG monitoring in
patients with subarachnoid hemorrhage. J Clin Neurophysiol 2005;22:
92–8.
56. Abend NS, Dlugos D, Herman S. Neonatal seizure detection using
multichannel display of envelope trend. Epilepsia 2008;49:349–52.
57. Friedman DE, Schevon C, Emerson RG, et al. Cyclic electrographic seizures
in critically ill patients. Epilepsia 2008;49:281–7.
58. Pandian JD, Cascino GD, So EL, et al. Digital video-electroencephalographic
monitoring in the neurological-neurosurgical intensive care unit: clinical
features and outcome. Arch Neurol 2004;61:1090–4.
59. Sutter R, Fuhr P, Grize L, et al. Continuous video-EEG monitoring
increases detection rate of nonconvulsive status epilepticus in the ICU.
Epilepsia 2011;52:453–7.
60. Winkler I, Haufe S, Tangermann M. Automatic classification of artifactual
ICA-components for artifact removal in EEG signals. Behav Brain Funct
2011;7:30.
61. Iwasaki M, Kellinghaus C, Alexopoulos AV, et al. Effects of eyelid closure,
blinks, and eye movements on the electroencephalogram. Clin
Neurophysiol 2005;116:878–85.
62. Goncharova II, McFarland DJ, Vaughan TM, et al. EMG contamination of
EEG: spectral and topographical characteristics. Clin Neurophysiol 2003;
114:1580–93.
63. Barbati G, Porcaro C, Zappasodi F, et al. Optimization of an independent
component analysis approach for artifact identification and removal in
magnetoencephalographic signals. Clin Neurophysiol 2004;115:1220–32.
64. Gao JF, Yang Y, Lin P, et al. Automatic removal of eye-movement and blink
artifacts from EEG signals. Brain Topogr 2010;23:105–14.
65. Choi D, Ryu Y, Lee Y, et al. Performance evaluation of a motor-imagery-
based EEG-Brain computer interface using a combined cue with
heterogeneous training data in BCI-naive subjects. Biomed Eng Online
2011;10:91.
66. Furdea A, Ruf CA, Halder S, et al. A new (semantic) reflexive brain-
computer interface: in search for a suitable classifier. J Neurosci Methods
2012;203:233–40.
67. Brunner C, Billinger M, Vidaurre C, et al. A comparison of univariate,
vector, bilinear autoregressive, and band power features for brain-computer
interfaces. Med Biol Eng Comput 2011;49:1337–46.
68. Scheuer ML. Continuous EEG monitoring in the intensive care unit.
Epilepsia 2002;43(Suppl. 3):S114–27.
69. Musialowicz T, Mervaala E, Kälviäinen R, et al. Can BIS monitoring be used
to assess the depth of propofol anesthesia in the treatment of refractory
status epilepticus? Epilepsia 2010;51:1580–6.
70. Shellhaas RA, Gallagher PR, Clancy RR. Assessment of neonatal
electroencephalography (EEG) background by conventional and two
amplitude-integrated EEG classification systems. J Pediatr 2008;153:
369–74.
71. Vespa PM, Nuwer MR, Nenov V, et al. Increased incidence and impact of
nonconvulsive and convulsive seizures after traumatic brain injury as
detected by continuous electroencephalographic monitoring. J Neurosurg
1999;91:750–60.
72. Ronne-Engstrom E, Winkler T. Continuous EEG monitoring in patients
with traumatic brain injury reveals a high incidence of epileptiform
activity. Acta Neurol Scand 2006;114:47–53.
73. Amantini A, Fossi S, Grippo A, et al. Continuous EEG-SEP monitoring in
severe brain injury. Neurophysiol Clin 2009;39:85–93.
74. Steudel WI, Krüger J. Using the spectral analysis of the EEG for prognosis
of severe brain injuries in the first post-traumatic week. Acta Neurochir
Suppl 1979;28:40–2.
75. Beridze M, Khaburzania M, Shakarishvili R, et al. Dominated EEG patterns
and their prognostic value in coma caused by traumatic brain injury.
Georgian Med News 2010;28–33.
76. Gütling E, Gonser A, Imhof HG, et al. EEG reactivity in the prognosis of
severe head injury. Neurology 1995;45:915–18.
77. Butzkueven H, Evans AH, Pitman A, et al. Onset seizures independently
predict poor outcome after subarachnoid hemorrhage. Neurology 2000;55:
1315–20.
78. Claassen J, Hirsch LJ, Frontera JA, et al. Prognostic significance of
continuous EEG monitoring in patients with poor-grade subarachnoid
hemorrhage. Neurocrit Care 2006;4:103–12.
79. Little AS, Kerrigan JF, McDougall CG, et al. Nonconvulsive status
epilepticus in patients suffering spontaneous subarachnoid hemorrhage.
J Neurosurg 2007;106:805–11.
80. Dennis LJ, Claassen J, Hirsch LJ, et al. Nonconvulsive status epilepticus
after subarachnoid hemorrhage. Neurosurgery 2002;51:1136–43; discussion
1144.
81. Rivierez M, Landau-Ferey J, Grob R, et al. Value of electroencephalogram in
prediction and diagnosis of vasospasm after intracranial aneurysm rupture.
Acta Neurochir 1991;110:17–23.
82. Vespa PM, Nuwer MR, Juhász C, et al. Early detection of vasospasm after
acute subarachnoid hemorrhage using continuous EEG ICU monitoring.
Electroencephalogr Clin Neurophysiol 1997;103:607–15.
83. Claassen J, Hirsch LJ, Kreiter KT, et al. Quantitative continuous EEG
for detecting delayed cerebral ischemia in patients with poor-grade
subarachnoid hemorrhage. Clin Neurophysiol 2004;115:2699–710.
84. Dreier JP, Major S, Manning A, et al. Cortical spreading ischaemia is a
novel process involved in ischaemic damage in patients with aneurysmal
subarachnoid haemorrhage. Brain 2009;132:1866–81.
85. Mutch WAC. New concepts regarding cerebral vasospasm: glial-centric
mechanisms. Can J Anesth/J Can Anesth 2010;57:479–89.
86. Dreier JP, Woitzik J, Fabricius M, et al. Delayed ischaemic neurological
deficits after subarachnoid haemorrhage are associated with clusters of
spreading depolarizations. Brain 2006;129:3224–37.
87. Rosengart AJ, Huo JD, Tolentino J, et al. Outcome in patients with
subarachnoid hemorrhage treated with antiepileptic drugs. J Neurosurg
2007;107:253–60.
88. Claassen J, Jetté N, Chum F, et al. Electrographic seizures and periodic
discharges after intracerebral hemorrhage. Neurology 2007;69:1356–65.
89. Messé SR, Sansing LH, Cucchiara BL, et al. Prophylactic antiepileptic drug
use is associated with poor outcome following ICH. Neurocrit Care 2009;
11:38–44.
90. Rundgren M, Westhall E, Cronberg T, et al. Continuous amplitude-
integrated electroencephalogram predicts outcome in hypothermia-treated
cardiac arrest patients. Critical Care Medicine 2010;38:1838–44.
91. Rossetti AO, Oddo M, Logroscino G, et al. Prognostication after cardiac
arrest and hypothermia: a prospective study. Ann Neurol 2010;67:
301–7.
92. Legriel S, Bruneel F, Sediri H, et al. Early EEG monitoring for detecting
postanoxic status epilepticus during therapeutic hypothermia: a pilot study.
Neurocrit Care 2009;11:338–44.
93. Abend NS, Topjian A, Ichord R, et al. Electroencephalographic monitoring
during hypothermia after pediatric cardiac arrest. Neurology 2009;72:
1931–40.
94. Rossetti AO, Logroscino G, Liaudet L, et al. Status epilepticus: an
independent outcome predictor after cerebral anoxia. Neurology 2007;69:
255–60.
95. Rossetti AO, Oddo M, Liaudet L, et al. Predictors of awakening from
postanoxic status epilepticus after therapeutic hypothermia. Neurology
2009;72:744–9.
96. Rothstein TL, Thomas EM, Sumi SM. Predicting outcome in hypoxic-
ischemic coma: a prospective clinical and electrophysiologic study.
Electroencephalogr Clin Neurophysiol 1991;79:101–7.
97. Fugate JE, Wijdicks EFM, Mandrekar J, et al. Predictors of neurologic
outcome in hypothermia after cardiac arrest. Ann Neurol 2010;68:
907–14.
98. Kaplan PW, Genoud D, Ho TW, et al. Etiology, neurologic correlations,
and prognosis in alpha coma. Clin Neurophysiol 1999;110:205–13.
99. Simbruner G, Mittal RA, Rohlmann F, et al. Systemic hypothermia after
neonatal encephalopathy: outcomes of neo.nEURO.network RCT. Pediatrics
2010;126:e771–8.
100. Wusthoff CJ, Dlugos DJ, Gutierrez-Colina A, et al. Electrographic seizures
during therapeutic hypothermia for neonatal hypoxic-ischemic
encephalopathy. J Child Neurol 2011;26:724–8.
101. Shankaran S, Pappas A, McDonald SA, et al. Predictive value of an early
amplitude integrated electroencephalogram and neurologic examination.
Pediatrics 2011;128:e112–20.
102. Brogna C, Gil Robles S, Duffau H: Brain tumors and epilepsy. Expert Rev
Neurother 2008;8:941–55.
103. Bladin CF, Alexandrov AV, Bellavance A, et al. Seizures after stroke:
a prospective multicenter study. Arch Neurol 2000;57:1617–22.
104. Edmond K, Clark A, Korczak VS, et al. Global and regional risk of disabling
sequelae from bacterial meningitis: a systematic review and meta-analysis.
Lancet Infect Dis 2010;10:317–28.
105. Zoons E, Weisfelt M, de Gans J, et al. Seizures in adults with bacterial
meningitis. Neurology 2008;70:2109–15.
106. Vibha D, Bhatia R, Prasad K, et al. Clinical features and independent
prognostic factors for acute bacterial meningitis in adults. Neurocrit Care
2010;13:199–204.

107. Flores-Cordero JM, Amaya-Villar R, Rincón-Ferrari MD, et al. Acute
community-acquired bacterial meningitis in adults admitted to the
intensive care unit: clinical manifestations, management and prognostic
factors. Intensive Care Med 2003;29:1967–73.
108. Horst ter HJ, Jongbloed-Pereboom M, van Eykern LA, et al. Amplitude-
integrated electroencephalographic activity is suppressed in preterm
infants with high scores on illness severity. Early Hum Dev 2011;87:
385–90.
109. Klinger G, Chin CN, Otsubo H, et al. Prognostic value of EEG in neonatal
bacterial meningitis. Pediatr Neurol 2001;24:28–31.
110. Bergen DC, Ristanovic R, Gorelick PB, et al. Seizures and renal failures. Int
J Artif Organs 1994;17:247–51.
111. Guerit J-M, Amantini A, Fischer C, et al. Neurophysiological investigations
of hepatic encephalopathy: ISHEN practice guidelines. Liver Int
2009;29:789–96.
112. Seder DB, Fraser GL, Robbins T, et al. The bispectral index and suppression
ratio are very early predictors of neurological outcome during therapeutic
hypothermia after cardiac arrest. Intensive Care Med 2010;36:281–8.
113. Chollet-Xémard C, Combes X, Soupizet F, et al. Bispectral index
monitoring is useless during cardiac arrest patients’ resuscitation.
Resuscitation 2009;80:213–16.
114. Young GB, Jordan KG, Doig GS. An assessment of nonconvulsive seizures
in the intensive care unit using continuous EEG monitoring: an
investigation of variables associated with mortality. Neurology
1996;47:83–9.
115. Riker RR, Fraser GL, Wilkins ML. Comparing the bispectral index and
suppression ratio with burst suppression of the electroencephalogram
during pentobarbital infusions in adult intensive care patients.
Pharmacotherapy 2003;23:1087–93.
Section III—Electrophysiology 255.e3
116. Tobias JD. Bispectral index monitoring documents burst suppression
during pentobarbital coma. J Intensive Care Med 2008;23:258–62.
117. Oddo M, Rossetti AO. Predicting neurological outcome after cardiac arrest.
Current Opinion in Critical Care 2011;17:254–9.
118. Rabinstein AA, Wijdicks EFM. The value of EEG monitoring after cardiac
arrest treated with hypothermia. Neurology 2012;78:774–5.
119. Kessler SK, Topjian AA, Gutierrez-Colina AM, et al. Short-term outcome
prediction by electroencephalographic features in children treated with
therapeutic hypothermia after cardiac arrest. Neurocrit Care 2011;14:37–43.
120. Meierkord H, Boon P, Engelsen B, et al. EFNS guideline on the
management of status epilepticus. Eur J Neurol 2006;13:445–50.
121. Rossetti AO, Logroscino G, Bromfield EB. Refractory status epilepticus:
effect of treatment aggressiveness on prognosis. Arch Neurol
2005;62:1698–702.
122. Krishnamurthy KB, Drislane FW. Depth of EEG suppression and outcome
in barbiturate anesthetic treatment for refractory status epilepticus.
Epilepsia 1999;40:759–62.
123. Meierkord H, Boon P, Engelsen B, et al. EFNS guideline on the
management of status epilepticus in adults. Euro J Neurol 2010;17(3):
348–55.
124. Johansen JW. Update on bispectral index monitoring. Best Pract Res Clin
Anaesthesiol 2006;20:81–99.
125. Ogilvie MP, Pereira BMT, Ryan ML, et al. Bispectral index to monitor
propofol sedation in trauma patients. J Trauma 2011;71:1415–21.
126. Young GB. Continuous EEG monitoring in the ICU. Acta Neurol Scand
2006;114:67–8.
127. Mirsattari SM, Lee DH, Jones D, et al. MRI compatible EEG electrode
system for routine use in the epilepsy monitoring unit and intensive care
unit. Clin Neurophysiol 2004;115:2175–80.
IV
Chapter
26  
Computed Tomography
Asako Miyakoshi and Wendy A. Cohen
Introduction
Computed tomography (CT)–based imaging technology con-
tinues to be the principal imaging modality in patients who
require neurocritical care. CT imaging can monitor intracra-
nial events, CT angiography (CTA) can demonstrate intracra-
nial and extracranial vasculature, and CT perfusion (CTP) can
provide measures of intracranial blood flow. All of these
studies are done on multislice CT scanners using commer-
cially available software and can be performed on portable
scanners housed in the neurocritical care unit (NCCU) (i.e.,
point-of-care assessment).1 Noncontrast head CT shows ana-
tomic intracranial injury: hemorrhage, hydrocephalus, contu-
sion, mass effect, shifts, ischemia, infarction, shear injury, and
diffuse axonal injury (DAI). Contrast-enhanced head CT facil-
itates identification of medium to large vascular abnormalities
(although typically not aneurysms), masses, and blood-brain
barrier disruption. Contrast also is required for CTA and CTP.
CTA may show vascular occlusion in a patient with an acute
infarct, the source of an intraparenchymal hemorrhage, or
vasospasm following subarachnoid hemorrhage (SAH). CTP
is used to evaluate infarct core versus penumbra in a patient
with an acute infarct, to show areas of reduced blood flow in
a patient with vasospasm, and to demonstrate the presence or
absence of autoregulation following traumatic brain injury
(TBI).
A Brief Comparison
with Other Techniques
Other imaging and monitoring techniques can provide similar
information to that obtained by CT (See also Chapters
27-29). This information can complement or supplement the
CT data. The presence of an ischemic infarct or the extent of
injury with DAI is more accurately determined using
magnetic resonance imaging (MRI). Disadvantages of MRI
when compared with CT include the limitations in identi­
fication of acute intracranial hemorrhage, difficulty moni­
toring an unstable patient in the MRI suite because of
requirements for nonferromagnetic life-support devices,
longer imaging times with MRI, and the relatively greater
expense of MRI equipment. The additional information
from an MRI does not usually change acute management in
patients with TBI.2 Magnetic resonance angiography (MRA)
258
is a standard tool to image intracranial and extracranial vas-
culature in nontraumatized patients and is frequently used in
patients with acute ischemic infarcts who also undergo MRI
when being evaluated for urgent intra-arterial thrombolysis.
In these patients magnetic resonance perfusion (MRP) is
used to define areas of abnormal perfusion not yet infarcted
(diffusion-perfusion mismatch).3 In addition, MRA can be
useful to assess vascular dissections.4 However, MRA and MRP
have the technical limitations of MRI in a patient on life
support.
Cerebral perfusion also is measured with nonradioactive
xenon-CT (Xe-CT). The advantage of stable Xe-CT is the
possibility to measure regional cerebral blood flow in a quan-
titative manner. During a Xe-CT, xenon gas is inhaled in
combination with oxygen over 5 minutes during CT scan-
ning. Xenon is lipid-soluble and passes freely though the
blood-brain barrier. Therefore brain, venous, and end-tidal
xenon concentrations are assumed to be similar. Xenon is
radiodense, and the density of xenon on a CT image has
been shown to correlate with end-tidal xenon concentration.
Calculation of cerebral blood flow (CBF) is based on use of
the Fick equation. Accuracy of CBF measurement using
stable xenon has been confirmed by comparison to radiola-
beled microsphere studies.5,6 Xe-CT has some inherent limi-
tations. Side effects of xenon inhalation include sedation,
bronchospasm, and respiratory depression. Xenon blood
flow measurements require specialized equipment to deliver
the gas and to measure the end-tidal xenon concentration.
Because the gas is sedating, some patients have difficulty
lying still for the 5-minute scan acquisition. Availability of
xenon gas is variable in the United States and at the time of
publication is used under research protocols.7,8
Transcranial Doppler (TCD) is used to evaluate cerebral
perfusion by measurement of flow velocities in first- and
second-order vessels. Access depends on adequate acoustic
windows and accuracy may be operator dependent.7 An
advantage to TCD is ease of use (typically bedside) and
repeatability.
Single photon emission computed tomography (SPECT)
using Tc-99m ethyl cysteinate dimer (ECD) or Tc-99m hexa-
methyl propyleneamine oxime (HMPAO) provides a qualita-
tive evaluation of cerebral perfusion. The isotope is bound on
the first pass through the brain. Evaluation of supratentorial
flow is made in relation to cerebellar flow. Repeat evaluation
cannot occur until after the technetium isotope has decayed.
© Copyright 2013 Elsevier Inc. All rights reserved.

Techniques
Computed Tomography (Nonenhanced CT)
Cross-sectional images of the head from a nonenhanced head
CT scan show the major intracranial anatomy (ventricles, cis-
terns, sulci, gray matter vs. white matter borders) and the skull
anatomy. A head CT demonstrates the presence of acute intra-
cranial hemorrhage more reliably than any other imaging
examination. Other pathologic processes that can be identi-
fied include midline shift, mass effect, hydrocephalus, intra-
cranial edema, and encephalomalacia. Skull and facial fractures
are shown from the same scan acquisition using windows and
reconstruction algorithms maximized for bone. The only con-
traindication to a head CT is an inability of the patient to lie
on the CT table. In the setting of acute intracranial decom-
pensation, pregnancy or pediatric age become relative consid-
erations or contraindications because CT requires the use of
ionizing radiation. A head CT can be acquired in 1 to 2
minutes using current-generation scanners. Life support
equipment, including intracranial monitoring systems, is
easily accommodated. Scans are typically displayed in an axial
projection but can be reconstructed into other planes, usually
coronal or sagittal.
A head CT does not show vascular structures, may not
show masses, may not easily aid identification of subacute-
to-chronic hemorrhage, may not demonstrate demyelination,
and may not identify intracranial infection. Small infarcts or
new infarcts in patients with prior ischemic areas may be hard
to visualize. Depending on the clinical question CT with iodin-
ated contrast, MRI without or with gadolinium-based contrast,
or vascular imaging using either CTA, MRA, or catheter angi-
ography may be a more informative imaging procedure.
Computed Tomography Angiography
CTA scans follow the first pass of a contrast load as it moves
through the vascular tree of the scanned anatomy. The scan
acquisition is typically timed to be 1 to 2 seconds behind the
arterial peak of the contrast bolus. The smallest available slice
is used for the scan acquisition to improve detection of small
vascular abnormalities and to improve the three-dimensional
(3-D) reformations of the vascular volumes. Scans can be
viewed as the initial acquisition (source images), as maximum
intensity projections (MIPs) often thicker than the initial scan
acquisition, or as true 3-D volumes. MIPs usually include the
surrounding brain and bone. This is a disadvantage when
evaluating the cavernous segments of the internal carotid
arteries or the vertebral arteries within the foramen transver-
saria of the cervical vertebrae. MIPs facilitate visualization of
longer segments of the vessel because the entire diameter of
the vessel is included in the volume and MIPs can be obtained
in any desired orientation. Intracranial vessels are well seen
with a combination of coronal, axial, and sagittal projections.
Extracranial carotid and vertebral arteries may be better
evaluated using coronal, oblique, and sagittal reformations.
Three-dimensional volume reformations eliminate the sur-
rounding bone and brain and are rotated in space, either as
single images or as a cine loop. Fuzzy connectedness also may
be used to eliminate bony structures at the skull base.9 However,
the diagnostic accuracy of bone-subtracted CTA often is no
better than careful analysis of 3-D volume-rendered CTA
when source data are assessed in multiplanar reformats.10
Section IV—Radiology 259
Evaluation of CTA examinations should include the source
images as well as the MIPs and the 3-D reformations.11,12
Source images show small vascular wall abnormalities. The
reformations show the anatomic relationships of the vessels.
Atherosclerotic narrowing of a vessel may be more easily
appreciated on the source image and may be averaged out on
an MIP or 3-D reformation. Conversely the caudal-pointing
dome of an anterior communicating artery aneurysm or
cranial-directed dome of a basilar tip aneurysm may be dif-
ficult to define on axial source images but will be clearly
shown on the reformations. The common artifacts in a CTA
examination arise from patient motion, which causes misreg-
istration in the MIPs and in the 3-D reconstructions. An inad-
equate contrast bolus caused by a mistimed contrast injection
or due to poor cardiac output can occur but is less common.
Numerous studies have documented the accuracy of CTA for
aneurysms greater than 4 to 5 mm in diameter. Aneurysms
3 mm or less are missed 20% of the time.13,14
Computed Tomography Perfusion
Calculation of CTP parameters is based on the central volume
principle, represented by the following equation:
CBF = CBV/MTT
where CBF equals cerebral blood flow (mL/100 g/min−1),
CBV equals cerebral blood volume (mL/100 g), and MTT
equals mean transit time (seconds, time of the bolus passing
through a volume of tissue). One commonly used method
uses deconvolution, a mathematical process, to derive these
parameters from the time density curve generated by a first-
pass measurement of contrast through the brain. MTT is cal-
culated from the arterial input time-density curve, CBV is
calculated as the area under time-density curve in the tissue
of interest, and CBF is determined using the central volume
principle.15,16 The density of the iodinated contrast is corre-
lated to the volume of contrast in a region. Another CT-based
technique uses a maximum slope model from this calcula-
tion.17 There is also variation between commercial vendors in
the implementation of the deconvolution algorithm. This
results in variability of the quantitative results from CT per-
fusion calculations.18 Older techniques to measure cerebral
perfusion used tracers that were freely diffusible across the
blood-brain barrier and not limited to the intravascular space
(such as xenon in radioactive and nonradioactive forms).
Iodinated contrast in contradistinction to xenon remains
intravascular under normal circumstances. Actual perfusion
of brain parenchyma is not shown. Although data obtained
from Xe-CT cannot be directly applied to CTP due to techni-
cal differences, generally good correlation between CTP and
Xe-CT is reported.19
The advantage of CTP is widespread availability and rela-
tively low cost. CTP can be obtained on any multislice CT
scanner with appropriate software availability. A disadvantage
of CTP is the limited anatomic coverage of the brain using
16- to 128-slice scanners. The length of the CT detector deter-
mines the volume of brain evaluated on a single CTP acquisi-
tion. Attempts to move the table during the CTP acquisition
are difficult to implement and have not been generally adopted
(toggle table). The posterior fossa is difficult to evaluate during
a CTP study because of artifact from the bone in the skull base
and because it is difficult to image the posterior fossa without
260 Section IV—Radiology
scanning through the lens of the eye. Gantry angulation can
eliminate the direct dose to the eyes while including the top
of the cerebellum and the occipital lobes in the imaging plane.
This angulation is usually not able to include the lower por-
tions of the posterior fossa.
Although CTP is described as quantitative, postprocessing
techniques can affect the measurements. Issues of input vessel
selection, vessel opacification in the area being evaluated, and
variation in acquisition software remain.20-22 Postprocessing
can be done manually or automatically. Interobserver agree-
ment in measurements of CBF, CBV, MTT and volume of
infarct core and penumbra is better with automatically post-
processed data.23,24 CTP calculations also are based on the
assumption of an intact blood-brain barrier. Many studies
use internal comparisons to evaluate CBF, CBV, and MTT
rather than the numerical values and reproducibility of
quantitative assessment has been questioned, especially in
processes such as acute ischemic infarct.25
Blood flow in normal mixed gray and white matter is 50 to
80 mL/100 g brain tissue/min. CBF is higher in the gray
matter compared with white matter. CBF is kept relatively
constant by autoregulation mechanisms. In the absence of
changes to the autoregulatory system CBF changes with altera-
tions in PaCO2, functional requirements, and administration
of vasoactive substances. Under normal conditions CBF is not
changed by changes in systemic blood pressure. Low-flow
states are of concern because of the risk of stroke. Nerve cell
function deteriorates at CBF less than 20 mL/100 g/min. Cell
death occurs at CBF less than 5-10 mL/100 g/min.26
Patient Safety
Radiation dose and contrast load are significant patient safety
issues with CT-based imaging. CTA scans are similar to other
CT studies in that the scanner acquires single slices at multiple
levels. The usual issue for these patients is the number of
diagnostic cranial CT exams obtained per patient rather than
the dose delivered per study.
CTP studies require acquisition of 40 to 50 scans over 1
minute at a single anatomic level. The scanning techniques
used for a perfusion CT are lower than those for a diagnostic
head study, typically 80 to 100 kVp/200 mA for CTP com-
pared with 120 to 130 kVp/100 to 250 mA for a head CT. The
lower kVp of CTP results in a lower radiation dose and
improved visualization of the iodinated contrast. Radiation
dose measures ionizing energy absorbed per unit of mass and
is measured in gray (Gy) or milligray (mGy); 1 Gy = 1 joule
per kilogram. Alternative units are sieverts (Sv) or millisieverts
(mSv). For x-ray radiation in the ranges produced by CT scan-
ners, 1 mSv = 1 mGy. When CTP is done at the basal ganglia
level, avoiding the orbits, intracranial radiation dose is 180 to
363 mGy, surface dose is 327 to 713 mGy, and lens dose (from
scatter) is 11 m to 22 mGy.27 This is higher than the average
surface dose of 66 to 83 mGy at the tuber frontale by conven-
tional head CT scan using 130 to 135 kVp and 105 to 270 mA.28
If CTP is performed at multiple levels, the dose is increased
in proportion to number of acquisitions. The threshold dose
for cataract formation is 2 to 10 Gy. The overall effective dose
equivalent for CTP is 2 plus or minus 2.5 mSv. The dose from
a CTA of the brain is 2 to 4 mSv.27,29,30 The dose from other
blood flow measurement techniques, such as positron emis-
sion tomography (PET) and single photon emission computed
tomography (SPECT), is higher.31,32 For comparison, back-
ground radiation dose over a year in approximately 3 mSv, the
dose from a chest x-ray is 0.1 mSv, and the dose during a
roundtrip flight from Los Angeles to New York is 0.03 mSv.
Iodinated contrast intravenously administered increases
serum creatinine.33 Patients with compromised renal function
may be at greater risk.34 CTA usually requires 60 to 100 mL
of contrast, CTP 40 to 50 mL per anatomic level evaluated.
The significance of these contrast doses depends on the
patient’s underlying renal function, total amount of contrast
given, and the timing of those doses. For example, if a patient
has an acute SAH, the following studies may happen in rapid
succession: CTA and diagnostic catheter angiography without
or with aneurysm coiling, each with a significant dye load. In
a patient admitted for polytrauma, admission studies may
include CTA of the cranial and extracranial cervical vessels,
CTA of aorta and peripheral vessels, abdominal CT with con-
trast to evaluate for abdominal injury, and possibly cranial
CTP to evaluate the presence of absence of cerebral autoregu-
lation. A patient with an acute infarct is likely to have a more
limited set of studies: CTA and CTP. Only the minority who
might benefit from intra-arterial treatment are likely to have
angiography. Additionally, in a stroke patient, MRI, MRA, and
MRP may be used acutely, possibly diminishing the volume
of iodinated contrast. This is, however, the patient population
most likely to have compromised renal function secondary to
systemic processes such as diabetes and hypertension.
Indications
Routine Head Computed Tomography Scan
The indications for noncontrast head CT are well established
and described, and the reader is referred to recent reviews or
textbooks that describe this it in detail. Here is provided a brief
synopsis of how head CT can aid the neurointensivist.
A head CT is the initial emergent study for patients with
acute neurologic injury, whether stroke, trauma, confusion, or
unknown neurologic event. It also is the primary study to
follow the neurologically compromised patient in an intensive
care unit (ICU) setting. A typical example is the patient
with an acute intracranial hemorrhage. The acute and sub-
acute management questions can involve resolution of the
hemorrhage with diminishing mass effect, increasing paren-
chymal vasogenic edema in response to intraparenchymal
blood products, increased volume of blood if delayed hemor-
rhage occurs, or hydrocephalus if blood within the arachnoid
granulations causes diminished cerebrospinal fluid (CSF)
resorption. Other issues addressed by head CT include iden-
tification of generalized brain swelling such as might occur in
a patient with DAI, midbrain compression (unilateral from
uncal herniation or bilateral from diffuse swelling), midline
shift, new or evolving brain infarction, postoperative issues, or
new fluid collections. Identification of possible infection or
vascular injury requires a contrast study.
Imaging Appearence
INTRACRANIAL HEMORRHAGE
Acute intraparenchymal hemorrhage is denser than brain with
CT. The density reflects the concentration of the protein com-
ponent of the hemoglobin molecule and diminishes with
 Section IV—Radiology 261

A B C
Fig. 26.1  Computed tomography (CT) and computed tomography angiography (CTA) images to illustrate an intraparenchymal hemorrhage
and spot sign on the CTA. A, Axial noncontrast head CT. Hemorrhage is present in the left temporal lobe with extension into the occipital horn of
the lateral ventricle. B, CTA source image. A punctate focus of enhancement on the lateral margin of the hemorrhage is seen (spot sign); this suggests
ongoing hemorrhage (arrow). C, Delayed contrast enhanced CT. Irregular collection of contrast in the lateral margin of the clot is caused by ongoing
hemorrhage (arrow). This scan was obtained 1 to 2 minutes after the contrast injection.

breakdown of the hemoglobin molecule. An uncomplicated
clot within the brain becomes more lucent on its margins,
continues to decrease in density over time (2-3 weeks de­
pending on the initial size of the clot), and ultimately
resorbs, leaving an area of encephalomalacia (low density with
focal volume loss). When contrast is given, contrast extravasa-
tion is associated with subsequent intracranial hemorrhage
growth.35 This is similar to the “spot sign,” a bright spot on
CTA source images that predicts hematoma growth (Fig. 26.1).
SAH, unlike intraparenchymal hemorrhage, remains within
the sulci and cisterns. It does not undergo the breakdown
process seen with intraparenchymal clots. Density is depen-
dent on the concentration of blood in the subarachnoid space
(i.e., the volume of the hemorrhage within the volume of
CSF). Small hemorrhages may be poorly seen because of
mixing of the blood with CSF. Resolution of subarachnoid
blood is dependent on clearing by CSF circulatory processes.
Attempts have been made to quantitate the volume of SAH in
relation to the likelihood of developing vasospasm.36,37 The
Fisher scale was an early method to correlate the volume and
distribution of subarachnoid blood to the development of
vasospasm.38 Subsequent modifications or the development of
newer grading scales have improved the predictive reliability.
In the Hijdra scale the volume of blood in each of 10 basal
cisterns and four ventricles is separately evaluated and summed
for a total score.39
Acute blood within a more chronic fluid collection may be
seen as a blood-fluid level with the denser blood (both in
terms of visual appearance and true fluid density) in a depen-
dent part of the collection. This often is seen as blood in
the occipital horns (blood layering in the dependent part of
the ventricular system in a supine patient), but also occurs
in chronic subdural hematoma (SDH) and in other cystic
cavities.
ACUTE SUBDURAL HEMATOMA
Acute SDH is found within the subdural space, superficial to
the arachnoid. On imaging, acute SDH conforms to the cur-
vature of the skull, is confined by dural boundaries (does not
cross the midline), may extend between the leaves of the dura
in the falx and tentorium, and may extend across suture lines.
An acute SDH tends to be higher density than brain. Chronic
SDH is of lower density than brain because of hemoglobin
breakdown similar to the process in intraparenchymal hemor-
rhage. Acute bleeding into a chronic SDH results in mixed
density and blood-fluid levels. More chronic SDH may not
conform as clearly to the curve of the skull.
ACUTE EPIDURAL HEMATOMA
An acute epidural hematoma (EDH) is hyperdense to brain,
bows away from the curve of the skull (sometimes described
as lens shaped), and does not cross sutures but can cross the
midline (crosses dural sinuses). An EDH often is associated
with skull fractures, classically in the squamous temporal bone
across the path of the middle meningeal artery. EDHs, which
lie superficial to dura, are less likely than SDH to be associated
with an adjacent brain contusion.
INCREASED INTRACRANIAL PRESSURE
Increased intracranial pressure (ICP) causes a decrease in the
size of CSF spaces in the brain followed by increasing severity
of herniation. Diagnostic clues include a generalized decrease
in the size of sulci and cisterns. In particular, compression or
absence of the perimesencephalic cisterns is an indication that
ICP may be elevated. However, head CT findings do not always
reliably predict intracranial physiology (e.g., ICP and brain
oxygen).40 If the increase in ICP is due to diffuse swelling, the
ventricles also may decrease in size. If it is due to hydrocepha-
lus, the ventricles may increase in size. When it is associated
with local mass effect and herniation, there may be compres-
sion of the ventricles on the side of the lesion, displacement
of midline away from the lesion, and effacement of cisterns.
An example is subfalcine and uncal herniation in a patient
with a subacute, large, middle cerebral artery infarct with
marked cytotoxic edema.
HERNIATION
The appearance of herniation depends on the location of the
mass lesion or mass effect within the brain. Uncal herniation
and subfalcine herniation occur with mass lesions on one side
262 Section IV—Radiology

A B

C D
Fig. 26.2  Head computed tomography (CT) scan illustrating subfalcine and uncal herniation associated with an acute-on-chronic subdural
hematoma (SDH). A, Uncal herniation. Right sided holohemispheric mixed density SDH. There is right uncal herniation with flattening of the
suprasellar cistern and midbrain compression. B, Comparison: normal brain CT at the level of the midbrain. C, Subfalcine herniation. A more superior
slice in the same patient that shows compression of the right lateral ventricle and subfalcine herniation. D, Comparison: normal brain CT at the level
of the basal ganglia or lateral ventricles.

of the brain (Fig. 26.2). The uncus is displaced toward the
opposite side of the brain from the intracranial mass, the
suprasellar cistern is distorted, and the midbrain is displaced
away from the mass and compressed. Subfalcine herniation is
usually part of this constellation with compression of the ipsi-
lateral ventricle, compression of the third ventricle, and dis-
placement of the third ventricle and septum pellucidum across
midline away from the mass lesion. In general, when there is
more than 5 mm of midline shift, a surgical solution for the
increase in ICP often is necessary. Uncal herniation can be
bilateral. This might occur if the patient had diffuse swelling.
In this case all supratentorial ventricles are compressed, the
suprasellar cistern and perimesencephalic cisterns are effaced,
and the midbrain is flattened.
Posterior fossa mass effect can be seen as compression and
displacement of the fourth ventricle, compression of the
cerebellopontine-angle cisterns and cisterns anterior to the
brainstem, and displacement of the vermis upward with flat-
tening and distortion of the quadrigeminal plate cistern
(tectum) (i.e., “upward herniation”; Fig. 26.3). Also possible,
but more difficult to appreciate using CT, is downward dis-
placement of the cerebellar tonsils into the foramen magnum.
INFARCT
Ischemic injury to the brain on CT (Fig. 26.4) is seen as areas
of low density with poor to absent identification of the differ-
ence between gray matter and white matter (loss of gray-white
differentiation). Examples are the insular ribbon sign with
 Section IV—Radiology 263

A B

C D
Fig. 26.3  Head computed tomography (CT) scan that illustrates progressive upward herniation. A, Admission head CT scan that demonstrates
and early infarct (low density) in the left cerebellum. The fourth ventricle is midline. B, More superior slice from the admission head CT scan of the
same patient. The quadrigeminal plate cistern (arrow) and suprasellar cistern (arrowhead) are open. C, Day 1 head CT scan. There is partial effacement
of the quadrigeminal plate cistern (arrow). The suprasellar cistern and temporal horns are normal (arrowheads). D, Head CT scan obtained on day 2.
The quadrigeminal plate cistern is effaced (arrow) and the temporal horns are dilated (arrowheads). There is increased mass effect in the posterior
fossa caused by swelling of the subacute cerebellar infarct.

absence of a visible insula on the CT scan, generalized loss of
gray-white differentiation in an arterial vascular territory, or
loss of gray-white differentiation in the hemispheres in a
patient with diffuse anoxic injury. Acutely there may be
minimal mass effect. With time (12-48 hours) there can be
marked swelling of the areas of cytotoxic edema with possible
herniation. Occasionally hemorrhagic transformation (i.e.,
the development of hemorrhage within an acute or subacute
infarct) may occur.
The appearance of a venous infarct differs from that of an
arterial infarct in the distribution of the low-density region
and the propensity toward hemorrhage. The areas of low
density do not correspond to conventional arterial distribu-
tions but may involve deep structures, the temporal lobe,
and paramidline structures. The diagnosis depends on
appreciating high density in a venous sinus, absence of
contrast within a venous sinus on a contrast-enhanced CT
(delta sign), or absence of a sinus on a magnetic resonance
venogram (Fig. 26.5). Venous sinus thrombosis is more reli-
ably diagnosed than is the thrombosis of a superficial vein
because of the variation in the distribution of superficial
cerebral veins.
Computed Tomography Angiography
Indications for CTA include diagnosis and evaluation of intra-
cranial aneurysm, either ruptured or unruptured, diagnosis of
cerebral vasospasm, diagnosis of intracranial vascular malfor-
mations, evaluation of extracranial and intracranial vascular
structures in a patient with ischemic infarct or chronic cere-
brovascular disease, evaluation of underlying pathology in a
patient with acute intracranial hemorrhage, diagnosis of
264 Section IV—Radiology

A B

C D

Fig. 26.4  Computed tomography (CT) imaging showing evolution of acute to subacute infarct. A, Noncontrast head CT scan obtained 2
hours after symptom onset. B, Noncontrast head CT scan obtained 9 hours later demonstrates loss of gray-white differentiation in the left hemisphere.
C, Noncontrast head CT scan obtained 3 days after symptom onset illustrating completed left middle cerebral artery infarct. D, Coronal image of
computed tomography angiography shows occlusion of left middle cerebral artery.

vascular injury following blunt or penetrating trauma, and
diagnosis of cerebral venous thrombosis (CVT).
Computed Tomography Perfusion
CTP is a developing technology, and its use is still to be fully
elucidated. Indications include diagnosis of acute infarction,
evaluation of patients with acute infarction for possible inter-
vention, diagnosis of vasospasm, assessment of autoregulation
in TBI patients, evaluation of blood-brain barrier permeabil-
ity in acute infarction to assess for the risk of hemorrhagic
transformation, measurement of cerebral vascular reserve in
patients with chronic vascular occlusions, and evaluation of
tumor vascularity.41,42
Subarachnoid Hemorrhage
The sensitivity of noncontrast-enhanced CT for SAH within
12 hours of ictus is 98% to 100%. If there is a high suspicion
for SAH or if the patient has a subacute hemorrhage, lumbar
puncture may still be required to exclude the uncommon
but possible false-negative CT. Some suggest when there is
strong clinical suspicion of SAH that the combination of CT
and CTA may offer a less invasive diagnostic paradigm than
lumbar puncture.43 The quality of CTA to detect aneurysms is
almost comparable to digital subtraction angiography (DSA).
However, DSA remains superior to CTA to detect small aneu-
rysms (<5 mm), especially with 3-D rotational angiography.
Three-dimensional rotational angiography also allows for
more precise evaluation and measurement of aneurysms.13,44
CTA can assist in the rapid diagnosis or exclusion of aneu-
rysms and arteriovenous malformations in the acutely ill
patient, such as the patient with a large intracerebral hema-
toma who is rapidly deteriorating. When used as a first screen-
ing test for intracranial hemorrhage, CTA is often as accurate
as DSA for more proximal lesions but has a lower sensitivity
for smaller more distal lesions and dural arterio-venous
fistulas.45
CT angiography can be used in the initial work-up of
patients with a suspicion of an intracranial aneurysm
 Section IV—Radiology 265

A B

Fig. 26.5  Computed tomography (CT) and magnetic resonance imaging (MRI) for comparison illustrating venous sinus thrombosis.
A, Contrast-enhanced CT. A thrombosis of the right sigmoid sinus is illustrated with a delta sign (nonopacification of the sinus) (arrow). There is
contrast enhancement of the left sigmoid sinus (arrowhead). B, T1-weighted noncontrast axial MRI. Subacute clot seen in the right sigmoid sinus is
high signal (arrow). There is a flow void in the left sigmoid sinus (arrowhead). The high signal material in the mastoid resection cavity (anterior to the
sinus thrombosis) is surgically implanted fat.

(Fig. 26.6). Romijn et al. compared CTA with bone elimina-
tion to DSA and 3-D rotational angiography in 108 patients
with clinically suspected SAH. Specificity, sensitivity, positive
predictive value, and negative predictive value of CTA with
bone removal was 0.99, 0.90, 0.98, and 0.95, respectively. Sen-
sitivity was 0.99 for aneurysms greater than or equal to 3 mm
and 0.38 for aneurysms less than 3 mm.44 Other authors have
found similar results without bone removal.13,46-48 When CTA
is negative the false-negative rate is reported by some to be
zero after DSA.49 In particular, negative CTA findings are reli-
able to exclude an aneurysm with a perimesencephalic distri-
bution of blood. When the hemorrhage is diffuse or more
peripheral, DSA should be considered when CTA is negative.50
However, the sensitivity of on-call resident interpretation of a
CTA after SAH is lower than an attending,51 and among
patients who undergo subsequent digital subtraction angiog-
raphy there is a treatment plan change in about 20%.52 CTA
may be used to decide whether to occlude the aneurysm using
endovascular or surgical techniques, and in some cases, guide
operative management. The need for catheter angiography
before decisions of management (coil vs. clip) and to deter-
mine surgical approach are a balance between the immediate
surgical concerns, the requirement for rapid intervention,
and the experience of the involved physicians.50,53,54 Discrepan-
cies between CTA and DSA frequently are associated with
location, for example, cavernous vs. supraclinoid, or differen-
tiation of a vascular loop from a small aneurysm. Use of CTA
to evaluate aneurysms after occlusion (clip or coil) is less clear.
The artifact from the metal coil or clip can prevent accurate
visualization of possible residual aneurysm.55,56 Despite this,
there are reports of CTA use to evaluate treated aneurysms
with 87% to 97% sensitivity and specificity, although intrao-
bserver concordance was less robust.57
Symptomatic vasospasm occurs in 17% to 40% of SAH
patients.37,59 Diagnosis and management of cerebral vaso-
spasm can be facilitated using CTA.59,60 Neurologic decline in
a patient with aneurysmal SAH has many potential causes,
both intracranial (cerebral vasospasm and ischemia, hydro-
cephalus, increasing edema, new hemorrhage) and systemic
(oxygenation, infection, metabolic abnormalities). Diagnosis
of cerebral vasospasm can be made with many different
imaging modalities, TCD, CTA, SPECT, CTP, and catheter
angiography. Catheter angiography, the diagnostic gold stan-
dard, is invasive and has a stroke risk, although in the SAH
population this is very rare.61,62 SPECT correlates with angio-
graphically demonstrated spasm 65% of the time.63,64 TCD
may not demonstrate distal spasm but is a very useful tool to
identify spasm in proximal or basal vessels (see Chapter 30).
CTA is complementary to TCD, which is performed bedside.
In various studies CTA has 75% to 90% accuracy to detect
cerebral vasospasm when compared with catheter angiogra-
phy. The diagnosis of vasospasm is concordant between CTA
and catheter angiography for no spasm (96%) and severe
spasm (100%) and usually adequate when the patient is symp-
tomatic.65 However, the accuracy using CTA for mild and
moderate spasm is 57% and 64%. Concordance between CTA
and catheter angiography is greater for proximal than for
distal vessels.65-67 Catheter angiography is necessary if intra-
arterial treatment of vasospasm is contemplated. It is helpful,
if using CTA to diagnose vasospasm, to have a baseline study
for comparison of vessel size. CTA measures of vascular diam-
eter slightly overestimate the size of larger vessels and under-
estimate the size of the smaller distal branches.68 Other authors
suggest the correlation between CTA and DSA measured
diameters is between 0.45 and 0.76 and that underestimation
of diameter of large and overestimation of diameter of small
arteries occurs with CTA. The major limitation of CTA,
however, is the inability to measure some vessels because of
artifact.68 In addition, CTA cannot demonstrate changes in
flow dynamics that are visible with catheter angiography.
CTA shows the anatomy of the vascular tree. Perfusion
methods (i.e., CTP) show areas of abnormal CBF. Timely
evaluation of low perfusion status is important because angio-
plasty of vasospasm can be performed to prevent secondary
infarction when low-flow states are identified. SPECT, TCD,
PET, Xe-CT, and CTP all can show reduced CBF associated
with vasospasm. SPECT also has been used to evaluate brain
perfusion. For SPECT, patients are required to lie still longer
266 Section IV—Radiology

A B

C D E

F G
Fig. 26.6  Basilar bifurcation aneurysm imaged and displayed using seven different methods. A, Noncontrast head computed tomography
(CT). Diffuse subarachnoid hemorrhage. Lower density focus in the suprasellar cistern is the basilar aneurysm (arrow). B, Computed tomography
angiography (CTA)—coronal maximum intensity projection (MIP). The basilar bifurcation aneurysm projects superior to the posterior cerebral arteries. 
C, CTA—sagittal MIP. The basilar bifurcation aneurysm projects superior to the dorsum sella. D, CTA—three-dimensional (3-D) reconstruction.
E, Catheter angiogram. The left vertebral artery injection (AP projection) shows the basilar aneurysm. F, 3-D MIP reconstruction of the rotational
angiographic acquisition-catheter angiogram. The left vertebral artery was injected. The small human figure in the lower right shows the patient
orientation. G, 3-D MIP reconstruction of the rotational angiographic acquisition-catheter angiogram. The hollow-vessel reconstruction can help guide
intra-arterial (endovascular) treatment.

than for CTP, and it can only be repeated after 24 hours. MRP
is not commonly used to evaluate cerebral vasospasm in the
critically ill patient following SAH because of difficulty moni-
toring the patient’s physiologic status in the vicinity of a high
field magnet, the distortion caused by metal within the head
(clip, coil), and the qualitative nature of the study.
CTP studies in patients with intracranial vasospasm have
examined CBF, CBV, and MTT. In broad terms MTT is very
sensitive to changes in cerebral perfusion but is not specific
for infarction. CBV is specific for areas of brain with irrevers-
ible ischemic damage. CBF and CBV are significantly lower
and MTT longer in patients with moderate to severe

vasospasm. CBF and CBV show biphasic decrease at 1 to 3
days and at 10 to 17 days after ictus, more apparent in the
more vascular gray matter than in white matter. In one study,
MTT showed early areas of diminished perfusion, but could
not be used to separate mild, moderate, and severe vaso-
spasm.29 CBF reduction is closely related to clinical outcome.
Sviri et al. found that CBF reduction and prolongation of
MTT were closely correlated with delayed ischemic deficit,
severity of vasospasm seen by catheter angiography, vaso-
spasm seen on TCD, perfusion estimated on SPECT, and clini-
cal course.69,70 In the presence of delayed cerebral ischemia
(DCI), threshold values for CTP-determined CBF and MTT
are considered to be 36.5 mL/100 g/min (95% sensitivity, 70%
specificity) and 5.4 seconds (78% sensitivity, 70% specificity),
respectively. However, absolute threshold values may not be
generalizable because of scanner equipment and postprocess-
ing differences.71 Others have confirmed the prolongation of
MTT and its correlation to CBF and vasospasm, and some
suggest that in DCI qualitative assessment of CTP may be a
better imaging modality than head CT or CTA for quick
decision making.72-76 Because of some questions about the
reliability of absolute flow values in a CTP calculation, Van de
Schaaf used side-to-side ratios to evaluate territorial abnor-
malities in perfusion. CBF ratio of 0.72 for abnormal to
normal appeared to best correlate with delayed ischemic
changes.77 Noncontrast CT should be done as necessary to
exclude other causes of clinical deterioration including recur-
rent hemorrhage, hydrocephalus, and edema. Relative perfu-
sion changes on admission CTP studies, that is, before DCI
develops (CBF, MTT, and time to peak) may help identify
patients who will go on to delayed ischemia following SAH.78-
80
The absolute flow values at admission may not always be
helpful. Good correlation is seen between MRP and CTP for
MTT.3 Finally, changes in CTP, particularly in poor-grade
patients, may be used as a surrogate marker to monitor the
success of treatment strategies for vasospasm81 or guide man-
agement decisions in the comatose SAH patient.82
Acute Cerebral Infarction
Acute stroke imaging starts with CT. The first consideration is
differentiation of ischemic infarct from intracranial hemor-
rhage. If ischemic infarct is present and the patient is within
the time window for emergent treatment (3 hours for intra-
venous tissue plasminogen activator [tPA] and 6 to 8 hours
for intra-arterial methods in the anterior circulation) the size
of the infarct, the anatomic location, and the presence of
occluded vessels in the neck or brain should be defined.83,84 If
there is a hemorrhage, the location and source of the bleed is
an immediate question as is midline shift and possible hernia-
tion. CT is the primary imaging modality to diagnose acute
intracranial hemorrhage. CT is less sensitive than MRI for
diagnosis of acute ischemic infarct, particularly in the poste-
rior fossa and brainstem. Despite this, protocols around the
use and timing of MRI in a patient with an acute infarct seen
within the treatment window vary among centers. Protocols
may include immediate CT only, immediate CT, CTA plus or
minus CTP, CT followed by MRI, MRA, and MRP or any
number of permutations of this. CTA, CTP, MRA, and MRP
provide similar information in the acute period.85-87 The com-
bination of CT, CTA, and CTP does not appear to adversely
increase the time to thrombolysis in patients with acute
Section IV—Radiology 267
ischemic stroke88,89 or increase the risk of contrast-induced
nephropathy90 and is cost effective91 (i.e., multimodal CT
imaging [CT, CTA, and CTP] can play a significant role in
acute stroke neuroimaging).42,92,93 In addition, combined CT
and CTP studies may be used to select some patients
for endovascular reperfusion94 or intravenous thrombolysis95
rather than selection solely on time-based criteria, or if the
time of stroke onset is not known combined CT and CTP can
be used to help predict the likelihood of hemorrhagic trans-
formation or development of malignant edema.96
CTA for acute ischemic infarcts should include the extra-
cranial as well as intracranial vessels, typically the aortic
arch through the top of the head. CTA may be more spe-
cific for identification of small, slow flow but not occluded
vessels than is MRA. Neither is completely specific. However,
if acute occlusion is suspected, either study can be used to
guide therapy, including of suspected basilar artery throm-
bosis.97 CTA also can be used as an indicator of severity
of ischemic injury. Smith et  al. in 72 patients with acute
stroke found that baseline National Institutes of Health
(NIH) stroke score and intracranial large vessel occlusion
predicted poor final neurologic outcome at discharge.
Similar outcome is found with posterior circulation occlu-
sions86,98 and when large vessel occlusion is seen on CTA
after transient ischemic attack (TIA), a decline in neurologic
function is more likely.99
Early imaging in the patient with an acute ischemic infarct
is directed to identify tissue at risk of infarction compared with
tissue with irreversible injury (penumbra vs. core). The under-
lying assumption is that acute intervention may be more effec-
tive if there is tissue at risk but does not have irreversible injury.
Perfusion studies can be used to make this distinction.100,101
For CTP the parameters used to define penumbra versus isch-
emic core remain under discussion. In most series the penum-
bra is considered to have reduced CBF, normal or elevated
CBV due to cerebral autoregulatory response, and elevated
MTT (Fig. 26.7). Tan et al. evaluated 113 patients, 55 of whom
went on to experience infarction. CBV was the most accurate
predictor of final infarct volume (97% specificity), whereas
increased MTT was predictive of tissue at risk for infarction
in the presence of vascular occlusion.90 Others have found
similar results. The irreversible core will have decreased CBF
and CBV with elevated MTT.15,16,102 Volume of the infarct core
on CTP correlates well with volume of the infarct on diffusion-
weighted MRI. However, CBV is not always low in patients
with acute infarction103 and some suggest CTP-measured CBF
corresponds with the acute diffusion-weighted imaging lesion
better than CBV.104
There have been attempts to quantify the measures of CBF,
CBV, and MTT in patients with acute infarction. For preserva-
tion of cerebral function CBF needs to be greater than 18 to
19 mL/100 g/min; minimum CBF to preserve tissue viability
is 10 to 15 mL/100 g/min.105,106 Viability of tissue within the
penumbra with diminished blood flow is time dependent,
although clear time thresholds have not been established. CTP
has been used as a predictor of thrombolytic therapy and
clinical outcome.102,107-109 Wintermark et al. reported using a
34% decrease in CBF relative to clinically normal areas as a
threshold for ischemia and infarction, and a CBV threshold of
2.5 mL/100 g to differentiate infarction from penumbra.
There was good correlation of the results of CT perfusion and
final infarcted area after recanalization, which reflected
268 Section IV—Radiology

A B

C D E
Fig. 26.7  Computed tomography perfusion study used to demonstrate the ischemic penumbra. The images, all obtained at the level of the
centrum semiovale, show mismatch of cerebral blood flow (CBF) or mean transit time to cerebral blood volume. A, CBF map. Large area of diminished
CBF in the distribution of the middle cerebral artery (blue is low flow). B, Mean transit time (MTT) map. Large area of prolonged MTT in the distribution
of the middle cerebral artery (blue is slow flow). C, Cerebral blood volume map. Cortical blood volume is normal to minimally increased (blue is
low blood volume). D, Noncontrast CT. Mild diffuse sulcal effacement is illustrated in the right hemisphere with a focal area of low density (arrow).
E, magnetic resonance imaging (MRI) diffusion-weighted image. Punctate areas of restricted diffusion in a watershed distribution on the right.

recovery of the penumbra after arterial recanalization.109
Indeed reperfusion (with an MTT reperfusion index >75%)
may be a more accurate predictor of infarct volume than
recanalization.110 However, generalized clinical application of
numerical values requires further study, because the values
may vary with different perfusion acquisition techniques and
pretreatment CTP prognostic maps may not always be associ-
ated with recovery or not follow early reperfusion.111 Whether
threshold perfusion values might be used in other disorders,
including TBI, remains unclear. In addition, threshold values
may not be generalizable because of differences in equipment
and postprocessing methods.
primary diagnostic imaging study. DAI is the one process that
can be difficult to diagnose with CT, because hemorrhage may
not be present and the small foci of abnormal low density may
be difficult to see with CT. The progression of these processes
in the acute phase and the secondary complications (e.g.,
swelling, shift, hydrocephalus, delayed hemorrhage, infarct)
provide the indications for follow-up CT scans. Routine CT
also can be used to grade the head injury through the Mar-
shall112 or Rotterdam scales.113 There is good interobserver
reproducibility between neuroradiologists and neurosurgeons
when calculating these scales,114 both of which can be prog-
nostically important.115,116

Traumatic Brain Injury Computed Tomography Angiography

Routine Computed Tomography
Traumatic intracranial injury includes contusion, subarach-
noid or intraventricular hemorrhage, hematoma (intraparen-
chymal/SDH/EDH), and DAI. This can occur from a variety
of mechanisms, have secondary injury from local cellular
process or from associated vascular injury with infarct, and be
seen in combination. Because most cranial injury is associated
with hemorrhage, mass lesions, or shift, CT has become the
CTA in trauma is of most use to screen for or identify vascular
injury in blunt or penetrating trauma4,117-119 although DSA
may still be needed in some patients.120 Incidence of injury to
the carotid or vertebral arteries following blunt trauma is
approximately 30% in a subset of patients who have specific
findings, including cervical spine fracture or dislocation, frac-
ture through the foramen transversarium, upper cervical
spine injury with a high-force mechanism, LeFort II and
LeFort III facial injuries, bilateral mandibular condyle

fractures, fractures through the carotid canal in the skull base,
expanding hematoma in the neck, and unexplained acute
infarct. This compares an incidence of less than 1% in the
larger population of patients with blunt injury.121-125 Penetrat-
ing injury also places vascular structures at risk and requires
either exploration if in zone 2 or imaging if in zone 1 or zone
3. Although some discussion remains, in the majority of these
patients CTA is the initial study of choice because of rapidity
of acquisition, low risk, and accuracy.126-130 As in the other
applications of CTA, the development of x-ray tubes that have
sufficient heat capacity to image larger areas of the body (i.e.,
arch to vertex), thin isovoxel slices, and rapid, flexible refor-
mations has expedited the conversation from catheter angiog-
raphy to CTA.
Computed Tomography Perfusion
Use of cerebral perfusion techniques to modify treatment of
TBI patients, especially closed head injury, is less uniformly
embraced than is CTA to identify vascular injuries or CTP to
evaluate an acute ischemic infarct. Xe-CT and CTP have been
used to estimate CBF in relation to cerebral perfusion pressure
(CPP), evaluate the status of autoregulatory functions, and
possibly predict the extent of intracranial injury. Conceptu-
ally, elevation of mean arterial pressure does not change CPP
if autoregulation is intact. In contrast, if autoregulation is
absent, elevation of mean arterial pressure will increase CBF
and cerebral perfusion. The aim of the treatment is to main-
tain CPP to minimize secondary intracranial injury.
In general, target values should be ICP less than 20 mm Hg
and CPP of 50 to 70 mm Hg. CPP values less than 50 mm Hg
carry a two- to three-times higher risk of cerebral ischemia
and cerebral hypoperfusion, whereas CPP values greater than
70 mm Hg may be associated with an increased incidence of
systemic complications such as cardiorespiratory failure. CPP-
directed therapy versus ICP-directed therapy depends on the
state of the individual’s cerebral vascular autoregulation. If
autoregulation is intact, a CPP-directed therapy could be used
with a higher probability of favorable outcome, whereas if
autoregulation is absent ICP-guided therapy may yield better
results.
Cerebral contusion can enlarge and cause increased ICP
and neurologic deterioration during a patient’s ICU course.
Early identification of the patient at risk for contusion enlarge-
ment could prevent further tissue damage. CTP may allow
better evaluation of tissue viability than noncontrast CT.
Compared with noncontrast routine CT, the areas with
decreased CBF, CBV, and prolonged MTT correlate with
development of brain damage seen on later follow-up CTs.131
Wintermark et al. also reported a good correlation between
the extent of CTP abnormalities with CPP. Preservation
of autoregulation may be a good predictor of the patients’
outcome.132,133
Brain Death
CTA can be used to aid in the diagnosis of brain death,
although it may not always be reliable.134 In one study of 21
patients, the sensitivity of CTA compared with electroenceph-
alography was 52.4%. Brain death on CTA was defined as an
absence of intracranial vascular opacification.135 Divergence
between catheter angiography and CTA was 30.2% in another
Section IV—Radiology 269
study where brain death was defined as absence of intracranial
flow on catheter angiography.136
Portable Computed Tomography
Mobile CT scanners for cranial imaging have evolved since
the late 1990s. This point of care technology has the advantage
of bringing the scanner to the ICU patient. These are the
patients who are most difficult to transport because of ventila-
tory requirements, extensive physiologic monitoring, and
uncertainty of physiologic changes that may occur during
transport.137 The best images from a mobile scanner are
similar in detail to the average image from fixed CT scanners,
and the radiation dose is comparable.1,138-140 If needed, extra
shielding can be added when the device is used in an open
ward environment to restrict radiation exposure to others.141
The scanners have less flexibility in angulation of the image
with respect to the patient’s head. There also are potential
limitations in anatomy that can be imaged. For example, the
length of the patient’s neck determines the most caudal slice
on one scanner available in 2011. This may limit evaluation
of the patient’s neck or cervical spine. Portable CT scanners
are slower than conventional fixed CT. However, the major
advantage is that patients do not have to be removed from
their beds. This means patients can be continually monitored
and ventilated during the scan without having first to be
disconnected.142
Conclusion
Noncontrast CT is an important and common tool to evaluate
patients with acute neurologic disorders and those in the
NCCU. CTA and CTP both provide noninvasive imaging.
CTA is more firmly established as a method to evaluate the
intracranial vasculature. CTP is a developing technology and
requires further evaluation to fully define its use and accuracy.
Both can provide useful information that aids care of patients
in the NCCU. In an era of increasing concern with radiation
dosage, the use of either has to be weighed against the biologic
risks, particularly when multiple follow-up CT scans are per-
formed. The advent of portable CT scans has allowed CT
imaging to be performed in NCCU.
References
1. Peace K, Maloney E, Frangos S, et al. The use of a portable head CT scanner
in the ICU. J Neurosci Nurs 2010;42(2):109–16.
2. Manolakai D, Velmahos GC, Spaniolas K, et al. Early magnetic resonance
imaging is unnecessary in patients with traumatic brain injury. J Trauma
2009;66(4):1008–12.
3. Eastwood JD, Lev MH, Wintermark M, et al. Correlation of early dynamic
CT perfusion imaging with whole-brain MR diffusion and perfusion
imaging in acute hemispheric stroke. Am J Neuroradiol 2003;24(9):1869–75.
4. Heuer G, Hurst R, Le Roux P. Traumatic carotid artery injury. In: Winn HR,
editor. Youmans neurological surgery. 6th ed. Philadelphia: Elsevier; 2011.
5. Gur D, Yonas H, Jackson DL, et al. Simultaneous measurements of cerebral
blood flow by the xenon/CT method and the microsphere method: a
comparison. Invest Radiol 1985;20(7):672–7.
6. Yonas H, Pindzola RP, Johnson DW. Xenon/computed tomography cerebral
blood flow and its use in clinical management. Neurosurg Clin N Am
1996;7(4):605–16.
7. Coles JP. Imaging of cerebral blood flow and metabolism. Curr Opin
Anesthesiol 2006;19:473–80.
8. Horn P, Vajkoczy P, Thome C, et al. Xenon-induced flow activation in
patients with cerebral insult who undergo xenon-enhanced CT blood flow
studies. Am J Neuroradiol 2001;22:1543–9.
270 Section IV—Radiology
9. Abrahams J, Saha P, Hurst R, et al. Three dimensional bone-free rendering of
the cerebral circulation by use of computed tomography and fuzzy
connectedness. Neurosurgery 2002;51:264–9.
10. Ramasundara S, Mitchell PJ, Dowling RJ. Bone subtraction CT angiography
for the detection of intracranial aneurysms. J Med Imaging Radiat Oncol
2010;54(6):526–33. doi:10.1111/j.1754-9485.2010.02211.x.
11. Anderson GB, Steinke DE, Petruk KC, et al. Computed tomographic
angiography versus digital subtraction angiography for the diagnosis and
early treatment of ruptured intracranial aneurysms. Neurosurgery
1999;45(6):1315–20.
12. van der Schaaf IC, Velthuis BK, Wermer MJH, et al. New detected aneurysms
on follow-up screening in patients with previously clipped intracranial
aneurysms comparison with DSA or CTA at the time of SAH. Stroke
2005;36:1753–8.
13. Dammert S, Krings T, Moller-Hartmann W, et al. Detection of intracranial
aneurysms with multislice CT: comparison with conventional angiography.
Neuroradiology 2004;46(6):427–34.
14. Agid R, Lee SK, Willinsky RA, et al. Acute subarachnoid hemorrhage: using
64-slice multidetector CT angiography to “triage” patients’ treatment.
Neuroradiology 2006;48(11):787–94.
15. Nabavi DG, Cenic A, Craen RA, et al. Quantitative assessment of cerebral
hemodynamics using computed tomography: stability, accuracy, and
precision studies in dogs. J Comput Assist Tomogr 1999;23:506–15.
16. Wintermark M, Maeder P, Thiran JP, et al. Quantitative assessment of
regional cerebral blood flow by perfusion CT studies at low injection rates:
a critical review of the underlying theoretical models. Eur Radiol 2001;11:
1220–30.
17. Abels B, Klotz E, Tomandl BF, et al. Perfusion CT in acute ischemic stroke:
a qualitative and quantitative comparison of deconvolution and maximum
slope approach. Am J Neuroradiol 2010;31:1690–8.
18. Kudo K, Sasaki M, Yamada K, et al. Differences in CT perfusion maps
generated by different commercial software: quantitative analysis by using
identical source data of acute stroke patients. Radiology 2010;254:200–9.
19. Sase S, Honda M, Machida K, et al. Comparison of cerebral blood flow
between perfusion computed tomography and xenon-enhanced computed
tomography for normal subjects: territorial analysis. J Comput Assist
Tomogr 2005;29(2):270–7.
20. Sanelli PC, Lev MH, Eastwood JD, et al. The effect of varying user-selected
input parameters on quantitative values in CT perfusion maps. Acad Radiol
2004;11(10):1085–92.
21. Wintermark M, Lau BC, Chien J, et al. The anterior cerebral artery is an
appropriate arterial input function for perfusion-CT processing in patients
with acute stroke. Neuroradiology 2008;50:227–36.
22. Waaijera A, van der Schaaf IC, Velthuisa BK, et al. Reproducibility of
quantitative CT brain perfusion measurements in patients with symptomatic
unilateral carotid artery stenosis. Am J Neuroradiol 2007;28:927–32.
23. Fiorella D, Heiserman J, Prenger E, et al. Assessment of the reproducibility of
postprocessing dynamic CT perfusion data. Am J Neuroradiol 2004;25(1):
97–107.
24. Soares BP, Dankbaar JW, Bredno J, et al. Automated versus manual post-
processing of perfusion-CT data in patients with acute cerebral ischemia:
influence on interobserver variability. Neuroradiology 2009;51(7):445–51.
25. Imielinska C, Liu X, Rosiene J, et al. Toward objective quantification of
perfusion-weighted computed tomography in subarachnoid hemorrhage:
quantification of symmetry and automated delineation of vascular
territories. Acad Radiol 2005;12(7):874–87.
26. Ito H, Kanno I, Kato C, et al. Database of normal human cerebral blood
flow, cerebral blood volume, cerebral oxygen extraction fraction and cerebral
metabolic rate of oxygen measured by positron emission tomography
with 15O-labelled carbon dioxide or water, carbon monoxide and oxygen:
a multicentre study in Japan. Eur J Nucl Med Mol Imaging 2004;31(5):
635–43.
27. Hirata M, Sugawara Y, Fukutomi Y, et al. Measurement of radiation dose in
cerebral CT perfusion study. Radiat Med 2005;23(2):97–103.
28. Cohnen M, Fischer H, Hamacher J, et al. CT of the head by use of reduced
current and kilovoltage: relationship between image quality and dose
reduction. Am J Neuroradiol 2000;21(9):1654–60.
29. Nabavi DG, LeBlanc LM, Baxter B, et al. Monitoring cerebral perfusion after
subarachnoid hemorrhage using CT. Neuroradiology 2001;43:7–16.
30. Lee EJ, Lee SK, Agid R, et al. Comparison of image quality and radiation
dose between fixed tube current and combined automatic tube current
modulation in craniocervical CT angiography. Am J Neuroradiol 2009;30:
1754–9.
31. Huda W, Sandison GA. The use of the effective dose equivalents, HE, for
99mTc labelled radiopharmaceuticals. Eur J Nucl Med 1989;15(4):174–9.
32. Huda W, Sandison GA. Estimates of the effective dose equivalent, HE, in
positron emission tomography studies. Eur J Nucl Med 1990;17(3-4):116–20.
33. Langner S, Stumpe S, Kirsch M, et al. No increased risk for contrast-induced
nephropathy after multiple CT perfusion studies of the brain with a
nonionic, dimeric, iso-osmolal contrast medium. Am J Neuroradiol
2008;29(8):1525–9.
34. Newhouse JH, Kho D, Qasim A, et al. Frequency of serum creatinine changes
in the absence of iodinated contrast material: implications for studies of
contrast nephrotoxicity. Am J Radiol 2008;191:376–82.
35. Hallevi H, Abraham AT, Barreto AD, et al. The spot sign in intracerebral
hemorrhage: the importance of looking for contrast extravasation. Stroke
2010;41(1):e34–40. Epub 2009, Nov 12.
36. Rosen DS, Macdonald RL. Subarachnoid hemorrhage grading scales: a
systematic review. Neurocrit Care 2005;2:110–18.
37. Frontera JA, Claassen J, Schmidt JM, et al. Prediction of symptomatic
vasospasm after subarachnoid hemorrhage: the modified fisher scale.
Neurosurgery 2006;59(1):21–7.
38. Fisher CM, Kistler JP, Davis JM. Relation of cerebral vasospasm to
subarachnoid hemorrhage visualized by computerized tomographic
scanning. Neurosurgery 1980;6:1–9.
39. Hijdra A, Brouwers PJ, Vermeulen M, et al. Grading the amount of blood on
computed tomograms after subarachnoid hemorrhage. Stroke 1990;21:
1156–61.
40. Katsnelson M, Mackenzie L, Frangos S, et al. Are initial radiographic and
clinical scales associated with subsequent intracranial pressure and brain
oxygen levels after severe traumatic brain injury? Neurosurgery 2011;Nov 9.
Epub ahead of print.
41. Hoeffner EG, Case I, Jain R, et al. Cerebral perfusion CT: technique and
clinical applications. Radiology 2004;231:632–44.
42. Leiva-Salinas C, Wintermark M, Kidwell CS. Neuroimaging of cerebral
ischemia and infarction. Neurotherapeutics 2011;8(1):19–27.
43. McCormack RF, Hutson A. Can computed tomography angiography of the
brain replace lumbar puncture in the evaluation of acute-onset headache
after a negative noncontrast cranial computed tomography scan? Acad
Emerg Med 2010;17(4):444–51.
44. Romijn M, Gratama van Andel HA, van Walderveen MA, et al. Diagnostic
accuracy of CT angiography with matched mask bone elimination for
detection of intracranial aneurysms: comparison with digital subtraction
angiography and 3D rotational angiography. Am J Neuroradiol 2008;29(1):
134–9.
45. Yeung R, Ahmad T, Aviv RI, et al. Comparison of CTA to DSA in
determining the etiology of spontaneous ICH. Can J Neurol Sci 2009;36(2):
176–80.
46. Nijjar S, Patel B, McGinn G, et al. Computed tomographic angiography as
the primary diagnostic study in spontaneous subarachnoid hemorrhage.
J Neuroimaging 2007;17(4):295–9.
47. Westerlaan HE, Gravendeel J, Fiore D, et al. Multislice CT angiography in
the selection of patients with ruptured intracranial aneurysms suitable for
clipping or coiling. Neuroradiology 2007;49(12):997–1007.
48. Mohan S, Lee W, Tan JT, et al. Multi-detector computer tomography
angiography in the initial assessment of patients acutely suspected of having
intracranial aneurysm rupture. Ann Acad Med (Singaore) 2009;38(9):769–73.
49. Prestigiacomo CJ, Sabit A, He W, et al. Three dimensional CT angiography
versus digital subtraction angiography in the detection of intracranial
aneurysms in subarachnoid hemorrhage. J Neurointerv Surg 2010;2(4):
385–9. Epub 2010, Jun 24.
50. Agid R, Andersson T, Almqvist H, et al. Negative CT angiography findings
in patients with spontaneous subarachnoid hemorrhage: when is digital
subtraction angiography still needed? AJNR Am J Neuroradiol 2010;31(4):
696–705. Epub 2009, Nov 26.
A complete list of references for this chapter can be found online at
www.expertconsult.com.

References
1. Peace K, Maloney E, Frangos S, et al. The use of a portable head CT
scanner in the ICU. J Neurosci Nurs 2010;42(2):109–16.
2. Manolakai D, Velmahos GC, Spaniolas K, et al. Early magnetic resonance
imaging is unnecessary in patients with traumatic brain injury. J Trauma
2009;66(4):1008–12.
3. Eastwood JD, Lev MH, Wintermark M, et al. Correlation of early dynamic
CT perfusion imaging with whole-brain MR diffusion and perfusion
imaging in acute hemispheric stroke. Am J Neuroradiol 2003;24(9):
1869–75.
4. Heuer G, Hurst R, Le Roux P. Traumatic carotid artery injury. In: Winn
HR, editor. Youmans neurological surgery. 6th ed. Philadelphia: Elsevier;
2011.
5. Gur D, Yonas H, Jackson DL, et al. Simultaneous measurements of cerebral
blood flow by the xenon/CT method and the microsphere method: a
comparison. Invest Radiol 1985;20(7):672–7.
6. Yonas H, Pindzola RP, Johnson DW. Xenon/computed tomography cerebral
blood flow and its use in clinical management. Neurosurg Clin N Am
1996;7(4):605–16.
7. Coles JP. Imaging of cerebral blood flow and metabolism. Curr Opin
Anesthesiol 2006;19:473–80.
8. Horn P, Vajkoczy P, Thome C, et al. Xenon-induced flow activation in
patients with cerebral insult who undergo xenon-enhanced CT blood flow
studies. Am J Neuroradiol 2001;22:1543–9.
9. Abrahams J, Saha P, Hurst R, et al. Three dimensional bone-free rendering
of the cerebral circulation by use of computed tomography and fuzzy
connectedness. Neurosurgery 2002;51:264–9.
10. Ramasundara S, Mitchell PJ, Dowling RJ. Bone subtraction CT angiography
for the detection of intracranial aneurysms. J Med Imaging Radiat Oncol
2010;54(6):526–33. doi:10.1111/j.1754-9485.2010.02211.x.
11. Anderson GB, Steinke DE, Petruk KC, et al. Computed tomographic
angiography versus digital subtraction angiography for the diagnosis and
early treatment of ruptured intracranial aneurysms. Neurosurgery
1999;45(6):1315–20.
12. van der Schaaf IC, Velthuis BK, Wermer MJH, et al. New detected
aneurysms on follow-up screening in patients with previously clipped
intracranial aneurysms comparison with DSA or CTA at the time of SAH.
Stroke 2005;36:1753–8.
13. Dammert S, Krings T, Moller-Hartmann W, et al. Detection of intracranial
aneurysms with multislice CT: comparison with conventional angiography.
Neuroradiology 2004;46(6):427–34.
14. Agid R, Lee SK, Willinsky RA, et al. Acute subarachnoid hemorrhage: using
64-slice multidetector CT angiography to “triage” patients’ treatment.
Neuroradiology 2006;48(11):787–94.
15. Nabavi DG, Cenic A, Craen RA, et al. Quantitative assessment of cerebral
hemodynamics using computed tomography: stability, accuracy, and
precision studies in dogs. J Comput Assist Tomogr 1999;23:506–15.
16. Wintermark M, Maeder P, Thiran JP, et al. Quantitative assessment of
regional cerebral blood flow by perfusion CT studies at low injection rates:
a critical review of the underlying theoretical models. Eur Radiol
2001;11:1220–30.
17. Abels B, Klotz E, Tomandl BF, et al. Perfusion CT in acute ischemic stroke:
a qualitative and quantitative comparison of deconvolution and maximum
slope approach. Am J Neuroradiol 2010;31:1690–8.
18. Kudo K, Sasaki M, Yamada K, et al. Differences in CT perfusion maps
generated by different commercial software: quantitative analysis by
using identical source data of acute stroke patients. Radiology 2010;254;
200–9.
19. Sase S, Honda M, Machida K, et al. Comparison of cerebral blood flow
between perfusion computed tomography and xenon-enhanced computed
tomography for normal subjects: territorial analysis. J Comput Assist
Tomogr 2005;29(2):270–7.
20. Sanelli PC, Lev MH, Eastwood JD, et al. The effect of varying user-selected
input parameters on quantitative values in CT perfusion maps. Acad Radiol
2004;11(10):1085–92.
21. Wintermark M, Lau BC, Chien J, et al. The anterior cerebral artery is an
appropriate arterial input function for perfusion-CT processing in patients
with acute stroke. Neuroradiology 2008;50:227–36.
22. Waaijera A, van der Schaaf IC, Velthuisa BK, et al. Reproducibility of
quantitative CT brain perfusion measurements in patients with
symptomatic unilateral carotid artery stenosis. Am J Neuroradiol 2007;28:
927–32.
23. Fiorella D, Heiserman J, Prenger E, et al. Assessment of the reproducibility
of postprocessing dynamic CT perfusion data. Am J Neuroradiol
2004;25(1):97–107.
Section IV—Radiology 270.e1
24. Soares BP, Dankbaar JW, Bredno J, et al. Automated versus manual
post-processing of perfusion-CT data in patients with acute cerebral
ischemia: influence on interobserver variability. Neuroradiology 2009;51(7):
445–51.
25. Imielinska C, Liu X, Rosiene J, et al. Toward objective quantification of
perfusion-weighted computed tomography in subarachnoid hemorrhage:
quantification of symmetry and automated delineation of vascular
territories. Acad Radiol 2005;12(7):874–87.
26. Ito H, Kanno I, Kato C, et al. Database of normal human cerebral blood
flow, cerebral blood volume, cerebral oxygen extraction fraction and
cerebral metabolic rate of oxygen measured by positron emission
tomography with 15O-labelled carbon dioxide or water, carbon monoxide
and oxygen: a multicentre study in Japan. Eur J Nucl Med Mol Imaging
2004;31(5):635–43.
27. Hirata M, Sugawara Y, Fukutomi Y, et al. Measurement of radiation dose in
cerebral CT perfusion study. Radiat Med 2005;23(2):97–103.
28. Cohnen M, Fischer H, Hamacher J, et al. CT of the head by use of reduced
current and kilovoltage: relationship between image quality and dose
reduction. Am J Neuroradiol 2000;21(9):1654–60.
29. Nabavi DG, LeBlanc LM, Baxter B, et al. Monitoring cerebral perfusion
after subarachnoid hemorrhage using CT. Neuroradiology 2001;43:
7–16.
30. Lee EJ, Lee SK, Agid R, et al. Comparison of image quality and radiation
dose between fixed tube current and combined automatic tube current
modulation in craniocervical CT angiography. Am J Neuroradiol
2009;30:1754–9.
31. Huda W, Sandison GA. The use of the effective dose equivalents, HE,
for 99mTc labelled radiopharmaceuticals. Eur J Nucl Med 1989;15(4):
174–9.
32. Huda W, Sandison GA. Estimates of the effective dose equivalent, HE, in
positron emission tomography studies. Eur J Nucl Med 1990;17(3-4):
116–20.
33. Langner S, Stumpe S, Kirsch M, et al. No increased risk for contrast-
induced nephropathy after multiple CT perfusion studies of the brain with
a nonionic, dimeric, iso-osmolal contrast medium. Am J Neuroradiol
2008;29(8):1525–9.
34. Newhouse JH, Kho D, Qasim A, et al. Frequency of serum creatinine
changes in the absence of iodinated contrast material: implications for
studies of contrast nephrotoxicity. Am J Radiol 2008;191:376–82.
35. Hallevi H, Abraham AT, Barreto AD, et al. The spot sign in intracerebral
hemorrhage: the importance of looking for contrast extravasation. Stroke
2010;41(1):e34–40. Epub 2009, Nov 12.
36. Rosen DS, Macdonald RL. Subarachnoid hemorrhage grading scales: a
systematic review. Neurocrit Care 2005;2:110–18.
37. Frontera JA, Claassen J, Schmidt JM, et al. Prediction of symptomatic
vasospasm after subarachnoid hemorrhage: the modified Fisher scale.
Neurosurgery 2006;59(1):21–7.
38. Fisher CM, Kistler JP, Davis JM. Relation of cerebral vasospasm to
subarachnoid hemorrhage visualized by computerized tomographic
scanning. Neurosurgery 1980;6:1–9.
39. Hijdra A, Brouwers PJ, Vermeulen M, et al. Grading the amount of blood
on computed tomograms after subarachnoid hemorrhage. Stroke
1990;21:1156–61.
40. Katsnelson M, Mackenzie L, Frangos S, et al. Are initial radiographic and
clinical scales associated with subsequent intracranial pressure and brain
oxygen levels after severe traumatic brain injury? Neurosurgery 2011;Nov 9.
Epub ahead of print.
41. Hoeffner EG, Case I, Jain R, et al. Cerebral perfusion CT: technique and
clinical applications. Radiology 2004;231:632–44.
42. Leiva-Salinas C, Wintermark M, Kidwell CS. Neuroimaging of cerebral
ischemia and infarction. Neurotherapeutics 2011;8(1):19–27.
43. McCormack RF, Hutson A. Can computed tomography angiography of the
brain replace lumbar puncture in the evaluation of acute-onset headache
after a negative noncontrast cranial computed tomography scan? Acad
Emerg Med 2010;17(4):444–51.
44. Romijn M, Gratama van Andel HA, van Walderveen MA, et al. Diagnostic
accuracy of CT angiography with matched mask bone elimination for
detection of intracranial aneurysms: comparison with digital subtraction
angiography and 3D rotational angiography. Am J Neuroradiol 2008;29(1):
134–9.
45. Yeung R, Ahmad T, Aviv RI, et al. Comparison of CTA to DSA in
determining the etiology of spontaneous ICH. Can J Neurol Sci 2009;
36(2):176–80.
46. Nijjar S, Patel B, McGinn G, et al. Computed tomographic angiography as
the primary diagnostic study in spontaneous subarachnoid hemorrhage.
J Neuroimaging 2007;17(4):295–9.
270.e2 Section IV—Radiology
47. Westerlaan HE, Gravendeel J, Fiore D, et al. Multislice CT angiography in
the selection of patients with ruptured intracranial aneurysms suitable for
clipping or coiling. Neuroradiology 2007;49(12):997–1007.
48. Mohan S, Lee W, Tan JT, et al. Multi-detector computer tomography
angiography in the initial assessment of patients acutely suspected of
having intracranial aneurysm rupture. Ann Acad Med (Singaore)
2009;38(9):769–73.
49. Prestigiacomo CJ, Sabit A, He W, et al. Three dimensional CT angiography
versus digital subtraction angiography in the detection of intracranial
aneurysms in subarachnoid hemorrhage. J Neurointerv Surg 2010;2(4):
385–9. Epub 2010, Jun 24.
50. Agid R, Andersson T, Almqvist H, et al. Negative CT angiography findings
in patients with spontaneous subarachnoid hemorrhage: when is digital
subtraction angiography still needed? AJNR Am J Neuroradiol
2010;31(4):696–705. Epub 2009, Nov 26.
51. Hochberg AR, Rojas R, Thomas AJ, et al. Accuracy of on-call resident
interpretation of ct angiography for intracranial aneurysm in subarachnoid
hemorrhage. AJR Am J Roentgenol 2011;197(6):1436–41.
52. Tomycz L, Bansal NK, Hawley CR, et al. “Real-world” comparison of
non-invasive imaging to conventional catheter angiography in the
diagnosis of cerebral aneurysms. Surg Neurol Int 2011;2:134. Epub 2011,
Sep 30.
53. Kershenovich A, Rappaport ZH, Maimon S. Brain computed tomography
angiographic scans as the sole diagnostic examination for excluding
aneurysms in patients with perimesencephalic subarachnoid hemorrhage.
Neurosurgery 2006;59(4):798–801.
54. Miley JT, Taylor RA, Janardhan V, et al. The value of computed tomography
angiography in determining treatment allocation for aneurysmal
subarachnoid hemorrhage. Neurocrit Care 2008;9(3):300–6.
55. Wermer MJH, van der Schaaf IC, Velthuis BK, et al. Follow-up screening
after subarachnoid haemorrhage: frequency and determinants of new
aneurysms and enlargement of existing aneurysms. Brain 2005 (128):
2421–9.
56. Wallace RC, Karis JP, Partovi S, et al. Noninvasive imaging of treated
cerebral aneurysms, part II: CT angiographic follow-up of surgically
clipped aneurysms. Am J Neuroradiol 2007;28:1207–12.
57. Uysal E, Ozel A, Erturk SM, et al. Comparison of multislice computed
tomography angiography and digital subtraction angiography in the
detection of residual or recurrent aneurysm after surgical clipping with
titanium clip. Acta Neurochir (Wien) 2009;151(2):131–5.
58. Charpentier C, Audibert G, Guillemin F, et al. Multivariate analysis of
predictors of cerebral vasospasm occurrence after aneurysmal subarachnoid
hemorrhage. Stroke 1999;30(7):1402–8.
59. Diringer MN, Bleck TP, Claude Hemphill J 3rd, et al. Critical care
management of patients following aneurysmal subarachnoid
hemorrhage: recommendations from the Neurocritical Care Society’s
Multidisciplinary Consensus Conference. Neurocrit Care 2011;Jul 20. Epub
ahead of print.
60. Washington CW, Zipfel GJ, Participants in the International Multi-
disciplinary Consensus Conference on the Critical Care Management of
Subarachnoid Hemorrhage. Detection and monitoring of vasospasm and
delayed cerebral ischemia: a review and assessment of the literature.
Neurocrit Care 2011;15(2):312–7.
61. Willensky RA, Taylor SM, terBrugge K, et al. Neurologic complications of
cerebral angiography: prospective analysis of 2,899 procedures and review
of the literature. Radiology 2003;227:522–8.
62. Le Roux P, Elliott JP, Eskridge JM, et al. Risks and benefits of diagnostic
angiography following aneurysm surgery: a retrospective analysis of 597
studies. Neurosurgery 1998;42:1248–55.
63. Soucy JP, McNamara D, Mohr G, et al. Evaluation of vasospasm secondary
to subarachnoid hemorrhage with technetium 99m hexamethyl-
propyleneamine oxime (HMPAO) tomoscintigraphy. J Nucl Med 1990;31:
972–7.
64. Rajendran JG, Lewis DH, Newell DW, et al. Brain SPECT used to evaluate
vasospasm after subarachnoid hemorrhage: correlation with angiography
and transcranial Doppler. Clin Nucl Med 2001;26(2):125–30.
65. Shankar JJ, Tan IY, Krings T, et al. CT angiography for evaluation of
cerebral vasospasm following acute subarachnoid haemorrhage.
Neuroradiology 2011;May 4. Epub ahead of print.
66. Anderson GB, Ashforth R, Steinke DE, et al. CT angiography for the
detection of cerebral vasospasm in patients with acute subarachnoid
hemorrhage. Am J Neuroradiol 2000;21(6):1011–15.
67. Otawara Y, Ogasawara K, Ogawa A, et al. Evaluation of vasospasm after
subarachnoid hemorrhage by use of multislice computed tomographic
angiography. Neurosurgery 2002;51(4):939–42.
68. Ferguson SD, Rosen DS, Bardo D, et al. Arterial diameters on catheter and
computed tomographic angiography. Surg Neurol 2010;73(3):165–73.
69. Sviri GE, Britz GW, Lewis DH, et al. Dynamic perfusion computed
tomography in the diagnosis of cerebral vasospasm. Neurosurgery 2006;59:
319–25.
70. Sviri GE, Mesiwala AH, Lewis DH, et al. Dynamic perfusion computerized
tomography in cerebral vasospasm following aneurysmal subarachnoid
hemorrhage: a comparison with technetium-99m–labeled ethyl cysteinate
dimer–single-photon emission computerized tomography. J Neurosurg
2006;104:404–10.
71. Sanelli PC, Ugorec I, Johnson CE, et al. Using quantitative CT perfusion for
evaluation of delayed cerebral ischemia following aneurysmal subarachnoid
hemorrhage. AJNR Am J Neuroradiol 2011;Sep 29. Epub ahead of print.
72. Harrigan MR, Magnano CR, Guterman LR, et al. Computed tomographic
perfusion in the management of aneurysmal subarachnoid hemorrhage:
new application of an existent technique. Neurosurgery 2005;56(2):304–17.
73. Kanazawa R, Kato M, Ishikawab K, et al. Convenience of the computed
tomography perfusion method for cerebral vasospasm detection after
subarachnoid hemorrhage. Surg Neurol 2007;67(6):604–11.
74. Pham M, Johnson A, Bartsch AJ, et al. CT perfusion predicts secondary
cerebral infarction after aneurysmal subarachnoid hemorrhage. Neurology
2007;69:762–5.
75. Dankbaar JW, Rijsdijk M, van der Schaaf IC, et al. Relationship between
vasospasm, cerebral perfusion, and delayed cerebral ischemia after
aneurysmal subarachnoid hemorrhage. Neuroradiology
2009;51(12):813–19.
76. Dankbaar JW, de Rooij NK, Velthuis BK, et al. Diagnosing delayed cerebral
ischemia with different CT modalities in patients with subarachnoid
hemorrhage with clinical deterioration. Stroke 2009;40(11):3493–8. Epub
2009, Sep 17.
77. Van de Schaaf IC, Wermer MJ, van der Graaf Y, et al. CT after subarachnoid
hemorrhage: relation of cerebral perfusion to delayed cerebral ischemia.
Neurology 2006;66:1533–8.
78. van der Schaaf IC, Wermer MJ, van der Graaf YA, et al. Prognostic value of
cerebral perfusion-computed tomography in the acute stage after
subarachnoid hemorrhage for the development of delayed cerebral
ischemia. Stroke 2006;37:409–13.
79. Dankbaar JW, de Rooij NK, Smit EJ, et al. Changes in cerebral perfusion
around the time of delayed cerebral ischemia in subarachnoid hemorrhage
patients. Cerebrovasc Dis 2011;32(2):133–40. Epub 2011, Jul 19.
80. Sanelli PC, Jou A, Gold R, et al. Using CT perfusion during the early
baseline period in aneurysmal subarachnoid hemorrhage to assess for
development of vasospasm. Neuroradiology 2011;53(6):425–34. Epub 2010,
Aug 7.
81. Nogueira RG, Lev MH, Roccatagliata L, et al. Intra-arterial nicardipine
infusion improves CT perfusion-measured cerebral blood flow in patients
with subarachnoid hemorrhage-induced vasospasm. AJNR Am J
Neuroradiol 2009;30(1):160–4. Epub 2008, Oct 22.
82. Huang AP, Arora S, Wintermark M, et al. Perfusion computed
tomographic imaging and surgical selection with patients after poor-grade
aneurysmal subarachnoid hemorrhage. Neurosurgery 2010;67(4):964–74;
discussion 975.
83. The National Institute of Neurological Disorders and Stroke rt-PA Stroke
Study Group. Tissue plasminogen activator for acute ischemic stroke.
N Engl J Med 1995;333:1581–8.
84. Hacke W, Kaste M, Fieschi C, et al. Randomised double-blind placebo-
controlled trial of thrombolytic therapy with intravenous alteplase in acute
ischaemic stroke (ECASS II). Lancet 1998;352:1245–51.
85. Schramm P, Schellinger PD, Klotz E, et al. Comparison of perfusion
computed tomography and computed tomography angiography source
images with perfusion-weighted imaging and diffusion-weighted imaging
in patients with acute stroke of less than 6 hours duration. Stroke 2004;35:
1652–8.
86. Smith WS, Tsao JW, Billings ME, et al. Prognostic significance of
angiographically confirmed large vessel intracranial occlusion in patients
presenting with acute brain ischemia. Neurocrit Care 2006;4(1):14–7.
87. Tan JC, Dillon WP, Liu S, et al. Systematic comparison of perfusion-CT
and CT-angiography in acute stroke patients. Ann Neurol 2007;61(6):
533–43.
88. Salottolo KM, Fanale CV, Leonard KA, et al. Multimodal imaging does not
delay intravenous thrombolytic therapy in acute stroke. AJNR Am J
Neuroradiol 2011;32(5):864–8. Epub 2011, Feb 10.
89. Gentile NT, Cernetich J, Kanamalla US, et al. Expedited computed
tomography perfusion and angiography in acute ischemic stroke: a
feasibility study. J Emerg Med 2011;Apr 15. Epub ahead of print.

90. Lima FO, Lev MH, Levy RA, et al. Functional contrast-enhanced CT for
evaluation of acute ischemic stroke does not increase the risk of contrast-
induced nephropathy. AJNR Am J Neuroradiol 2010;31(5):817–21. Epub
2009, Dec 31.
91. Young KC, Benesch CG, Jahromi BS. Cost-effectiveness of multimodal CT
for evaluating acute stroke. Neurology 2010;75(19):1678–85. Epub 2010,
Oct 6.
92. Ledezma CJ, Wintermark M. Multimodal CT in stroke imaging: new
concepts. Radiol Clin North Am 2009;47(1):109–16.
93. Leiva-Salinas C, Wintermark M. Imaging of acute ischemic stroke.
Neuroimaging Clin N Am 2010;20(4):455–68.
94. Turk A, Magarik JA, Chaudry I, et al. CT perfusion-guided patient selection
for endovascular treatment of acute ischemic stroke is safe and effective.
J Neurointerv Surg 2011;Sep 19. Epub ahead of print.
95. Michel P, Ntaios G, Reichhart M, et al. Perfusion-CT guided intravenous
thrombolysis in patients with unknown-onset stroke: a randomized,
double-blind, placebo-controlled, pilot feasibility trial. Neuroradiology
2011;Aug 2. Epub ahead of print.
96. Hom J, Dankbaar JW, Soares BP, et al. Blood-brain barrier permeability
assessed by perfusion CT predicts symptomatic hemorrhagic
transformation and malignant edema in acute ischemic stroke. AJNR Am J
Neuroradiol 2011;32(1):41–8. Epub 2010, Oct 14.
97. Bonatti G, Ferro F, Haglmüller T, et al. Basilar artery thrombosis: imaging and
endovascular therapy. Radiol Med 2010;115(8):1219–33. Epub 2010, Jul 31.
98. Sylaja PN, Puetz V, Dzialowski I, et al. Prognostic value of CT angiography
in patients with suspected vertebrobasilar ischemia. J Neuroimaging
2008;18(1):46–9.
99. Poisson SN, Nguyen-Huynh MN, Johnston SC, et al. Intracranial large
vessel occlusion as a predictor of decline in functional status after transient
ischemic attack. Stroke 2011;42(1):44–7. Epub 2010, Nov 24.
100. Konstas AA, Wintermark M, Lev MH. CT perfusion imaging in acute
stroke. Neuroimaging Clin N Am 2011;21(2):215–38, ix.
101. Lui YW, Tang ER, Allmendinger AM, et al. Evaluation of CT perfusion in
the setting of cerebral ischemia: patterns and pitfalls. AJNR Am J
Neuroradiol 2010;31(9):1552–63. Epub 2010, Feb 25.
102. Schaefer PW, Roccatagliata L, Ledezma C, et al. First-pass quantitative CT
perfusion identifies thresholds for salvageable penumbra in acute stroke
patients treated with Intra-arterial therapy. Am J Neuroradiol 2006;27(1):
20–5.
103. Eastwood JD, Lev MH, Azhari T, et al. CT perfusion scanning with
deconvolution analysis: pilot study in patients with acute middle cerebral
artery stroke. Radiology 2002;222(1):227–36.
104. Campbell BCV, Christensen S, Levi CR, et al. Cerebral blood flow is the
optimal CT perfusion parameter for assessing infarct core. Stroke 2011;42:
3435–40.
105. Hossman KA. Neuronal survival and revival during and after cerebral
ischemia. Am J Emerg Med 1983;1(2):191–7.
106. Powers WJ, Grubb RL Jr., Darriet D, et al. Cerebral blood flow and cerebral
metabolic rate of oxygen requirements for cerebral function and viability in
humans. J Cereb Blood Flow Metab 1985;5(4):600–8.
107. Silvennoinen HM, Hamberg LM, Lindsberg PJ, et al. CT perfusion identifies
increased salvage of tissue in patients receiving intravenous recombinant
tissue plasminogen activator within 3 hours of stroke onset. Am J
Neuroradiol 2008;29(6):1118–23.
108. Knoepfli AS, Sekoranja L, Bonvin C, et al. Evaluation of perfusion CT and
TIBI grade in acute stroke for predicting thrombolysis benefit and clinical
outcome. J Neuroradiol 2009;36(3):131–7.
109. Wintermark M, Reichhart M, Cuisenaire O, et al. Comparison of admission
perfusion computed tomography and qualitative diffusion- and perfusion-
weighted magnetic resonance imaging in acute stroke patients. Stroke
2002;33(8):2025–31.
110. Soares BP, Tong E, Hom J, et al. Reperfusion is a more accurate predictor of
follow-up infarct volume than recanalization: a proof of concept using CT
in acute ischemic stroke patients. Stroke 2010;41(1):e34–40. Epub 2009,
Nov 12.
111. Zhao L, Barlinn K, Bag AK, et al. Computed tomography perfusion
prognostic maps do not predict reversible and irreversible neurological
dysfunction following reperfusion therapies. Int J Stroke 2011;6(6):544–6.
doi:10.1111/j.1747-4949.2011.00681.x.
112. Marshall LF, Bowers-Marshall S, Klauber MR, et al. A new classification of
head injury based on computerized tomography. J Neurosurg 1991;75
(Suppl):S14–20.
113. Maas AI, Hukkelhoven CW, Marshall LF, et al. Prediction of outcome in
traumatic brain injury with computed tomographic characteristics: a
comparison between the computed tomographic classification and
Section IV—Radiology 270.e3
combinations of computed tomographic predictors. Neurosurgery
2005;57(6):1173–82.
114. Chun KA, Manley GT, Stiver SI, et al. Interobserver variability in the
assessment of CT imaging features of traumatic brain injury.
J Neurotrauma 2010;27(2):325–30.
115. Lingsma HF, Roozenbeek B, Steyerberg EW, et al. Early prognosis in
traumatic brain injury: from prophecies to predictions. Lancet Neurol
2010;9(5):543–54.
116. Maas AI, Steyerberg EW, Butcher I, et al. Prognostic value of computerized
tomography scan characteristics in traumatic brain injury: results from the
IMPACT study. J Neurotrauma 2007;24(2):303–14.
117. Burlew CC, Biffl WL. Blunt cerebrovascular trauma. Curr Opin Crit Care
2010;Aug 31. Epub ahead of print.
118. Löhrer L, Vieth V, Nassenstein I, et al. Blunt cerebrovascular injuries in
acute trauma care: a screening protocol. Eur Spine J 2011;Sep 6. Epub
ahead of print.
119. Kaye D, Brasel KJ, Neideen T, et al. Screening for blunt cerebrovascular
injuries is cost-effective. J Trauma 2011;70(5):1051–6; discussion
1056–7.
120. DiCocco JM, Emmett KP, Fabian TC, et al. Blunt cerebrovascular injury
screening with 32-channel multidetector computed tomography: more
slices still don’t cut it. Ann Surg 2011;253(3):444–50.
121. York G, Barboriak D, Petrella J, et al. Association of internal carotid artery
injury with carotid canal fractures in patients with head trauma. Am J
Roentgenol 2005;184(5):1672–8.
122. Cothren CC, Moore EE, Ray CE Jr., et al. Screening for blunt
cerebrovascular injuries is cost-effective. Am J Surg 2005;190(6):845–9.
123. Cothren CC, Moore EE, Ray CE Jr., et al. Cervical spine fracture patterns
mandating screening to rule out blunt cerebrovascular injury. Surgery
2007;141(1):76–82.
124. Biffl WL, Moore EE, Ryu RK, et al. The unrecognized epidemic of blunt
carotid arterial injuries: early diagnosis improves neurologic outcome. Ann
Surg 1998;228(4):462–70.
125. Biffl WL, Moore EE, Offner PJ, et al. Optimizing screening for blunt
cerebrovascular injuries. Am J Surg 1999;178(6):517–22.
126. Berne JD, Reuland KS, Villarreal DH, et al. Sixteen-slice multi-detector
computed tomographic angiography improves the accuracy of screening for
blunt cerebrovascular injury. J Trauma 2006;60(6):1204–9.
127. Biffl WL, Egglin T, Benedetto B, et al. Sixteen-slice computed tomographic
angiography is a reliable noninvasive screening test for clinically significant
blunt cerebrovascular injuries. J Trauma 2006;60(4):745–51.
128. Eastman AL, Chason DP, Perez CL, et al. Computed tomographic
angiography for the diagnosis of blunt cervical vascular injury: is it ready
for primetime? J Trauma 2006;60(5):925–9.
129. Utter GH, Hollingworth W, Hallam DK, et al. Sixteen-slice CT angiography
in patients with suspected blunt carotid and vertebral artery injuries. J Am
Coll Surg 2006;203(6):838–48.
130. Malhotra AK, Camacho M, Ivatury RR, et al. Computed tomographic
angiography for the diagnosis of blunt carotid/vertebral artery injury: a
note of caution. Ann Surg 2007;246:632–43.
131. Soustiel JF, Mahamid E, Goldsher D, et al. Perfusion-CT for early
assessment of traumatic cerebral contusions. Neuroradiology
2008;50:189–96.
132. Wintermark M, Chioléro R, van Melle G, et al. Relationship between
brain perfusion computed tomography variables and cerebral perfusion
pressure in severe head trauma patients. Crit Care Med 2004;32(7):
1579–87.
133. Wintermark M, Chiolero R, Van Melle G, et al. Cerebral vascular
autoregulation assessed by perfusion CT in severe head trauma patient.
J Neuroradiol 2006;33(1):27–37.
134. Berenguer CM, Davis FE, Howington JU. Brain death confirmation:
comparison of computed tomographic angiography with nuclear medicine
perfusion scan. J Trauma 2010;68(3):553–9.
135. Quesnel C, Fulgencio JP, Adrie C, et al. Limitations of computed
tomographic angiography in the diagnosis of brain death. Intensive Care
Med 2007;33(12):2129–35.
136. Combes JC, Chomel A, Ricolfi F, et al. Reliability of computed tomographic
angiography in the diagnosis of brain death. Transplant Proc
2007;39(1):16–20.
137. Swanson E, Mascitelli J, Stiefel M, et al. The effect of patient transport on
brain oxygen in comatose patients. Neurosurgery 2010;66:925–32.
138. Larovere KL, Brett MS, Tasker RC, et al. Head computed tomography
scanning during pediatric neurocritical care: diagnostic yield and the
utility of portable studies. Neurocrit Care 2011;Sep 21. Epub ahead of
print.
270.e4 Section IV—Radiology
139. Carlson AP, Yonas H. Portable head computed tomography scanner-
technology and applications: experience with 3421 scans.
J Neuroimaging 2011;Jun 23. doi:10.1111/j.1552-6569.2011.00621.x. Epub
ahead of print.
140. Rumboldt Z, Huda W, All JW. Review of portable CT with assessment of a
dedicated head CT scanner. AJNR Am J Neuroradiol 2009;30(9):1630–6.
Epub 2009, Aug 6.
141. Stevens GC, Rowles NP, Foy RT, et al. The use of mobile computed
tomography in intensive care: regulatory compliance and radiation
protection. J Radiol Prot 2009;29(4):483–90. Epub 2009, Nov 18.
142. Peace K, Maloney E, Frangos S, et al. Portable head CT scan and its effect
on intracranial pressure, cerebral perfusion pressure and brain oxygen.
J Neurosurgery 2011;114:1479–84.

Xenon-Enhanced Computed
Tomography
Andrew P. Carlson and Howard Yonas
Introduction
Xenon-enhanced computed tomography (Xe-CT) is a power-
ful physiologic imaging tool that provides a clinician with
quantitative cerebral blood flow (CBF) data, allows the physi-
ologic response to various interventions to be measured,
and provides prognostic data to guide patient management.
Though the data provided are in a tomographic format,
it provides quantitative information for critical decision
making when needed. In this way Xe-CT CBF is similar to the
CBF monitors discussed in subsequent chapters rather than
imaging technologies. This chapter discusses the theory and
physics behind the data provided, the practical aspects of
Xe-CT, and the clinical decision-making process in several
neurologic conditions encountered in the neurocritical care
unit (NCCU) based on quantitative CBF data.
Technology
Theory and Physiology
CBF can be measured by a number of diffusible tracers. Xenon
133 was an ideal early tracer because xenon is highly lipid
soluble and its 133 isotope can be readily tracked with external
scintillation counters. Using this technology, early studies
showed that hemispheric and then superficial regional flow
could be measured in mL/100 g/min. However, because radio-
isotope precautions were needed and resolution in the depth
of the brain was limited, routine intensive care unit (ICU)
application was not possible. In the late 1970s, once it was
reported that stable xenon was radiodense, similar to iodine,
CT imaging techniques and the ability to tomographically
image CBF were reported.1,2 Tens of thousands of Xe-CT CBF
studies have been done since then without significant compli-
cations. When first available clinically, use of xenon in the
United States received “grandfather” approval as a contrast
agent from the U.S. Food and Drug Administration (FDA).
When Praxair, the company that supplied medical-grade
xenon, withdrew from the rare-gas market, xenon lost clinical
approval, and currently its use in the United States occurs
under investigational new drug (IND) and research protocols.
A reapplication for clinical approval has been submitted to
the FDA.
© Copyright 2013 Elsevier Inc. All rights reserved.
Chapter
27  
IV
Studies with xenon 133 use the “wash-out” curve of this
tracer, whereas studies with stable xenon focus on the “wash-
in” curve. Use of a wash-in curve reduces any theoretical
alteration of flow measurement due to an activation of CBF
after xenon inhalation and has been validated in other quan-
titative CBF studies.3-9 In 2000 the dose of xenon was reduced
to 28% because newer CT scanners had a lower noise level.
This helped reduce the few sensorial effects associated with
33% xenon and reduced cost. With 28% xenon, studies can be
routinely performed in cooperative outpatients and in venti-
lated ICU patients.
To calculate CBF using Xe-CT requires a modification of
the Kety-Schmidt equation. Two variables are needed to solve
the equation: (1) the arterial concentration curve of the agent
and (2) the extent and time course of tissue arrival that
depends on the blood flow and the blood-brain partition
coefficient, or lambda. The arterial concentration is measured
indirectly from the end-tidal xenon concentration; this is a
very good approximation of a highly diffusible gas such as
xenon. The brain partition coefficient and flow is calculated
for each CT pixel by solving the Kety-Schmidt for both vari-
ables using an iterative mathematical approach:
Cxe br (t) = λk o ∫ Cxe Art (u) e − k (t −u )du
t
where Cxebr (t) is the time dependent brain xenon concentra-
tion, λ is the blood-brain partition coefficient, k is the brain
uptake flow rate constant, CxeArt(u) is the time-dependent
arterial xenon concentration. This calculation is performed
typically at 4 CT levels; however, 8 to 10 levels can be used
with newer multidetector scanners.
The calculated CBF values then are represented graphically
in a standardized color-coded fashion for each pixel (Figs.
27.1 to 27.3). A bell-shaped filter is applied to aid visual
display. To obtain useful regional values, the cortical mantle
can automatically be divided into 20 2-cm-deep regions of
interest (ROI), and the average for each region displayed.
Alternatively, any ROI of regular or irregular shape and size
can be selected by the user on either the CT or the CBF image
(see Fig. 27.1). Additional analysis can be performed to assess
the validity and further manipulate the data (Figs. 27.2 to
27.7). For mixed cortical sources, CBF in the range of 54 plus
or minus 10 mL/100 g/min is normal. Values less than 15 to
20 mL/100 g/min represent ischemia and are associated with
271
272 Section IV—Radiology

Fig. 27.1  Standard xenon-enhanced

computed tomography (Xe-CT)
study with quantitative cerebral
blood flow (CBF) values for each of
the approximately 24,000 CT voxels.
The baseline scan is displayed on the
left, and the CBF coded image on the
right. The image on the right is color
coded to the range of CBF values
displayed on the right of the image. The
cortical mantle can be automatically
divided into 20 regions of interest. The
mean CBF within each region is
displayed in the box below. The tissue
heterogeneity (TH) is a measure of the
variability of CBF values within the
region. The number of voxels (area) and
mean Hounsfield unit (mean HU)
enhancement also are displayed for
each region of interest. The user also
can define regions of interest as shown
in areas 21 (right deep frontal white
matter) and 22 (left basal ganglia).

Fig. 27.2  Xenon-enhanced computed tomography (Xe-CT) image showing cerebral blood flow (CBF) values at each of the four levels
typically studied. The top row shows the baseline images, the middle row visually displays the color-coded CBF values (with the scale on the right).
The bottom row is a measure of the data reliability. Low confidence indicates that movement or other factors may have interfered with data acquisition.
High confidence values (blue) indicate that the values obtained are valid.
 Section IV—Radiology 273

Fig. 27.3  This xenon-enhanced computed tomography (Xe-CT) display shows the same image divided into six regions of interest with
the average cerebral blood flow (CBF) values displayed in the left box. The upper left scan is the baseline image, the upper right image is
the CBF data, the lower right is the confidence of the study, and the graph displays the frequency of occurrence of each numeric CBF value in the
scan slice.

BL 1 Base BL 1 Base
BL 2 1.09% BL 2 1.04%
CT 1 4.09% CT 1 3.15%
CT 2 4.33% CT 2 4.60%
CT 3 5.28% CT 3 7.38%
CT 4 7.38% CT 4 9.83%
CT 5 13.37% CT 5 12.24%
CT 6 20.49% CT 6 11.88%

0 15 30 45 0 15 30 45
(%) (%)

Tolerance level
Below tolerance level
Above tolerance level

Fig. 27.4  To further ensure the validity of the cerebral blood flow (CBF) study, this graph displays the consistency in position of each
of the six serial scans required for the data acquisition (CT 1-6). The computer compares the relative position of the inner table from scan to
scan, and movement of less than 15% from the baseline images (BL1 and BL2) is considered within tolerance. The graph on the left shows significant
movement in the last image, making the data less reliable, whereas the graph on the right shows a study with low movement and good reliability.
274 Section IV—Radiology

IMG. INCLUDED: 1 2 3 4 5 6 160

15 HU 150
10 140
130
5
120
0 110
0 2 4
Time (min) 100
90
80
Fig. 27.5  The raw data acquisition data can be
reviewed by the user for any data point in the image 70
as a graph of increasing Hounsfield unit (HU) 60
15 HU
enhancement over time. The six data points represent 50
the six images acquired to make the cerebral blood flow 10 40
(CBF) calculation. The upper graph represents the normal 30
5
enhancement in a region of gray matter, as indicated by 20
the upper arrow, and the lower graph represents the lack 0 10
0 2 4
of enhancement as seen in the ventricle as indicated by the Time (min) 0
lower arrow.

Fig. 27.6  A comparison cerebral blood flow (CBF) study. The upper images are the two studies compared side by side. The first study is done
with the patient sedated and the PaCO2 35 mm Hg. In the second study, the patient was allowed to respire normally, which brought her PaCO2 to
28 mm Hg. This was done to determine whether to allow the patient to continue to hyperventilate or sedate her to bring her PaCO2 to a more
normal level. The lower right image shows quantitatively the difference between the first and second studies—the red values showing an increase in
CBF and the blue values showing a decrease. These data show that the patient had improved CBF with a lower PaCO2. She was therefore allowed
to continue with the lower values, avoiding the complications of sedation or paralytics, and objectively improving her CBF.

neurologic deficit and infarction.2,10-14 Normal gray matter
blood flow is 84 mL/100 g/min and white matter is 20 mL/
100 g/min. The mixed cortical flow and gray matter flow
decrease with age, but white matter flow remains constant
(Fig. 27.8).13
The half-life of arterial xenon concentration is about 30
seconds, and 99% of the agent is washed out of the body
within 5 minutes (see Fig. 27.7). This makes it possible to
repeat the study within 10 minutes while varying physiologic
parameters such as arterial carbon dioxide pressure (PaCO2)
or blood pressure (see Fig. 27.6). Today a four-level study,
which measures almost 100,000 flow values on four CT levels
requires less than 10 seconds. This quick study time allows
Xe-CT CBF–derived flow information to be used in prospec-
tive clinical decision making. For example, each mm Hg
increase of PaCO2 should increase CBF 3% in the normal
patient. The amount of flow augmentation associated with
changes in PaCO2 provides insight about the vasodilation
 Section IV—Radiology 275

CO2
60

45

30

15

0
–2 –1 0 1 2 3 4 5 6
Time (min)

XENON
40 Fig. 27.7  These two graphs are the respiratory data
followed during a xenon cerebral blood flow (CBF)
30 study. The top graph shows the end-tidal CO2 during the
study. The bottom graph shows the end-tidal xenon, with
20 the patient breathing the xenon gas starting at time 0. The
curve of xenon wash-in is a very good approximation of
arterial concentration of xenon, given its high diffusibility.
10
This makes obtaining this input function into the CBF
calculation possible without arterial sampling, and with
0
accurate data of the wash-in curve rather than relying on 
–2 –1 0 1 2 3 4 5 6
a standard curve. The xenon washes out very quickly 
Time (min) (~1 minute) after the gas is turned off.

120 PURE GRAY MATTER Standard deviation Devices and Engineering

100 The Xe-CT CBF data depend on several components: (1) soft-
90 ware that uses CT images to document the arrival of xenon
CBF (mL/100 g/min)

80 within the brain; (2) xenon end-tidal data recorded by thermal

*y = (–
70 .45) x conductivity as an indirect measure of arterial levels of xenon;
+ 106.5
MIXED CORTICAL
60 (3) a computer to handle the complex mathematical process;
50 and (4) hardware that maintains and delivers fixed concentra-
40 **y = (–.21) x tions of xenon and oxygen. Significant advances have been
+ 61.13 made in each of these components. In the early 1980s, the
30
PURE WHITE MATTER delivery system was a large bag filled with the desired gas
20
10 mixture. Today a closed loop system that can mix and main-
N.S. tain the desired concentrations while scrubbing CO2 is used
0
10 20 30 40 50 60 70 80
in part because of cost. This reduced the amount of xenon
used from 20 to 4.5 L/study. The initial computers required
Age (years) an hour to compute a few-level study and the transfer of data
*p < .0001 required moving reel-to-reel stored data. Today data transfer
**p < .0025
is instantaneous and the calculation of 24,000 flow values on
Fig. 27.8  Normative cerebral blood flow (CBF) values and the each of four computer tomography levels requires a few
variation with age are displayed. Gray matter CBF decreases with seconds.
age, whereas white matter CBF remains relatively constant. (From Yonas The evolution of CT scanners means that a multilevel high-
H, Darby JM, Marks EC, et al. CBF measured by Xe-CT: approach to analysis and
resolution CBF study is now possible with lower concentra-
normal values. J Cereb Blood Flow Metab 1991;11:716–25.)
tions of xenon (28% rather than 33%). Because there is
variation between CT scanners, a “phantom” has been devel-
oped to standardize data if needed (Fig. 27.9). The goal is to
status of brain regions being examined. This response may be maintain a signal-to-noise ratio of at least 8 : 1; the lower
altered in various clinical situations (see following). Anxiety xenon concentration is possible because modern scanners
and arrhythmias may result when CO2 is used as a vasodila- have significantly reduced the noise level that does not com-
tory agent. Consequently acetazolamide, which is a carbonic promise the confidence in the numerical validity of the tech-
anhydrase inhibitor that does not have as many side effects nology. The lowest concentration of xenon that provides valid
but that causes tissue acidosis similar to 5% carbon dioxide data is the goal because xenon, although it is an inert noble
(CO2) inhalation, is used more frequently in clinical practice. gas, has properties that may alter neuron function. At 80%
When tissue loses its ability to augment CBF in response to a xenon is a very safe and effective anesthetic agent with very
vasodilatory challenge, it implies maximal vasodilation and so rapid onset and offset of action. At 33% xenon may cause
no further reserve. This concept is helpful in deciding about dysesthesias and altered consciousness in about 3% to 5% of
hemodynamic reserve and subsequent stroke risk.15-23 patients. These complications are observed in less than 1% of
276 Section IV—Radiology

CBF Histogram CBF

5%
Frequency in percentage Confidence
4% Worst

3% 1%
3%
2%
6%
1% 90%

0% 0%
0 40 80 120 160
CC/100 grams/Min Best

Fig. 27.9  Xenon-enhanced computed tomography (Xe-CT) of a phantom device, allowing for standardization of cerebral blood flow (CBF) data with
any CT scanner.

patients with 28% xenon. Apnea is a concern with exposure

to 100% xenon, but medically significant apnea does not occur
using 33% xenon. Only brief episodes of respiratory irregular-
ity (<1% of patients) that is reversible with coaching may be
observed using 28% xenon. Patients occasionally describe
some anxiety or mild euphoria, which clears within minutes
of study cessation. Vital signs are not typically affected.
The only component of the Xe-CT CBF system that has
not changed over the past 20 years is the mathematical
approach to calculate CBF. This methodology has undergone
extensive validation studies with multiple technologies, for
example, microspheres,3 iodoantipyrine4 and in vivo xenon
133,9 positron emission tomography (PET),6 and thermal
dilution flow probes.8 Consequently the early literature that
describes the Xe-CT CBF method remains relevant to current
technology. This methodology also is being integrated within Fig. 27.10  Portable computed tomography (CT) scanner and xenon
modern portable CT scanning technology (e.g., CereTom- delivery system stored in hallway of the neurocritical care unit.
NeuroLogica: Danvers, MA; Fig. 27.10). Even though epi-
sodic, this bedside tomographic quantitative data can help
guide patient care in the NCCU24-26 and complement the con-
tinuous data from invasive monitors that provide more local (Fig. 27.11). The patient should be cooperative, amenable to
information.27 In particular, portable imaging eliminates intravenous (IV) sedation, or mechanically ventilated because
many of the risks associated with hospital transport of the the head position must be stable during the 4.5 minutes of
critically ill. The image quality is reliable.28-30 In addition, data acquisition. The xenon delivery system is attached to the
information from Xe-CT permits correlation of anatomy and patient’s mask or to the ventilator circuit. The end-tidal CO2
physiology. and end-tidal xenon concentrations are monitored continu-
ously and displayed on the system monitor. The baseline CT
levels for the study are identified and imaged before starting
Xe-CT Cerebral Blood Flow Procedure xenon inhalation. Next, xenon inhalation is started (28%) and
A typical Xe-CT study is described. Once the patient is identi- the scanner begins serial imaging at the same preset levels as
fied for the study (see the following text for indications), he the arterial xenon concentration increases. The first 2 minutes
or she is brought to the CT scanner, or a portable scanner is are the most critical phase of data acquisition because head
taken to the patient’s bedside. The xenon technology can be movement at this time will make the study quantitatively
used with most CT scanners with only minimal adaptation unreliable, while movement later allows for salvage of the

Fig. 27.11  A xenon-enhanced computed tomography (Xe-CT)
system at a patient’s bedside. The xenon computer is in the lower left
of the picture in front of the chair. The control console for the portable
CT scanner is in the foreground on the right. The scanner itself is seen
in the background in the upper left of the frame at the patient’s head.
The xenon delivery system is attached to the ventilator circuit (not seen).
study by exclusion of later poor quality data. Movement
during image acquisition degrades the study because the base-
line images are subtracted from the subsequent images to get
a measure of the change of Hounsfield units. The CT data are
then transferred to the xenon workstation that contains the
time-linked air (arterial) curve data.
At the workstation, the computer uses the Hounsfield
enhancement and the end-tidal xenon data to solve the inte-
gral equation for each pixel of the CT scan. The program also
analyzes for the numerical reliability of the data, which needs
to be factored into the clinical analysis of the CBF data. The
data then can be manipulated using the software. ROIs can be
studied where each pixel’s flow is averaged over the entire ROI.
Limitations
There are several potential limitations when Xe-CT scans are
performed. First are inherent risks of radiation exposure.
There are multiple serial slices; these should be directed above
the orbit and lens (the most radiosensitive structure of the
head). Second, ventilation may be altered, but at 28% xenon
rarely if ever causes any significant alteration of ventilation
pattern. Third, the study takes longer than a standard CT scan
to set up and requires experienced personnel to operate the
equipment. Fourth, data can be limited by motion artifact
during scanning. Fifth, the technology and equipment are not
available at many institutions. Finally, the FDA currently does
not approve xenon for clinical use other than in approved
research protocols. Recently, 80% xenon was reported to be an
ideal anesthetic agent because even after hours of inhalation
it was not associated with any physiologic side effects and had
rapid awakening.31 Other studies have demonstrated that
xenon can be neuroprotective because it can block N-methyl-
d-aspartate (NMDA) receptors.32 Carlson et al. examined the
safety of 28% xenon in 2003 Xe-CT examinations, including
1486 non­ventilated patients.33 Respiratory suppression (>20
seconds) occurred in 39 (1.9%) studies; all resolved spontane-
ously. No adverse event that resulted in a persistent neurologic
change or other sequelae were observed.
Section IV—Radiology 277
Neurocritical Care Unit Applications
Patient management in the NCCU can be guided by published
recommendations and guidelines.34,35 Ideally management
needs to be patient specific and targeted, and it is in these
circumstances that information provided by Xe-CT can be
useful. There are many potential clinical scenarios in which
Xe-CT can be useful, including identification of otherwise
unrecognized problems, confirmation of clinical diagnoses,
confirmation of information from other monitors, titration of
specific treatments, and avoidance of adverse treatment
effects.26 In a recent prospective study of 2003 Xe CBF studies
at seven centers, 93% of studies were considered clinically
useful.33 The majority of studies were performed for ischemic
stroke and occlusive vascular disease.
Subarachnoid Hemorrhage and Vasospasm
Following aneurysmal subarachnoid hemorrhage (SAH),
vasospasm, or more appropriately, delayed ischemic neuro-
logic deficit (DIND) can adversely affect patient outcome.
Many patients (~70%) develop some angiographic evidence
for vasospasm,36,37 but far fewer develop clinical ischemia.37,38
In these patients Xe-CT is the only imaging or blood flow
technique that can identify tissue at risk for infarction14 and
measure clinical response to an intervention. In addition,
Xe-CT has been used to assess and compare therapies for
DIND, for example, normovolemia or hypervolemia,39 or
cardiac output augmentation compared with hypertension
only.37,40 In particular Xe-CT is very predictive of infarction in
SAH patients with vasospasm. Those who demonstrate hemo-
dynamically significant vasospasm (i.e., CBF between 6 and
15 mL/100 g/min) likely will develop infarction in that terri-
tory,41,42 whereas in patients in whom CBF is greater than
19 mL/100 g/min, infarction is very rare.41
CBF may be reduced after SAH often independent of
intracranial pressure (ICP) and cerebral perfusion pressure
(CPP)43; this decrease, however, is usually associated with the
severity of SAH. In these patients, Xe-CT may also be used
to answer routine management questions such as identify
the optimal blood pressure or PaCO2 level. This allows the
physician to determine which interventions will be beneficial
in a specific patient and what the best “threshold” for treat-
ment is in an individual. This can be very useful in SAH
because large fluid volumes often may be given, and here
Xe-CT can help differentiate hyperemia from vasospasm-
causing ischemia because in both circumstances transcranial
Doppler flow velocities can be increased.44-47 Use of Xe-CT to
examine the efficacy of hypertonic saline48,49 and identify
patients in whom vasopressor-induced hypertension may
increase ischemia rather than help it50,51 also is described.
Because Xe-CT studies can be repeated in quick succession,
patients also may be examined to assess their response to
vasopressor agents.
Quantitative CBF data also may be used to assess the effi-
cacy of specific treatments for vasospasm including intra-
arterial infusion of papaverine, angioplasty or even intra-aortic
balloon counter pulsation.52 In these studies it was found that
papaverine, although it reduced vasospasm, increased CBF
in less than half the patients.53,54 By contrast Xe-CT studies
have shown that angioplasty (mechanical rather than phar­
macologic) improves CBF55 and can reverse ischemia.56 This
278 Section IV—Radiology
information can be used to guide decisions about the risks and
benefits of angioplasty.57
Carotid Endarterectomy or Arteriovenous
Malformation Surgery
Vascular hyperemia that can cause intracerebral hemorrhage
after a critically stenotic vessel is opened during carotid end-
arterectomy (CEA) is a rare but well-known complication.
Xe-CT studies with or without an acetazolamide challenge
may help identify and guide management of these patients.58,59
Following removal of arteriovenous malformations (AVMs)
various patterns of CBF redistribution may occur, including
increased regional CBF, or more commonly, decreased CBF,
which can be responsible for a neurologic deficit.60 A quantita-
tive CBF study may assist with management of postoperative
blood pressure in a patient with new neurologic deficit not
explained by other imaging.
Vascular Occlusion
Xe-CT has been used frequently to evaluate hemodynamic
sufficiency in patients with occlusive carotid vascular disease
or who may be undergoing surgery in which the carotid is at
risk or needs to be sacrificed and so predict the need for a
bypass or risk of stroke.61 In addition, patients with moya­
moya disease can be studied to examine direct or indirect
revascularization techniques and the status of collateral
perfusion.62-64 To do this the Xe-CT study is performed before
and after acetazolamide administration or as part of a balloon
test occlusion. Reduced CBF values less than 30 mL/100 g/
min, an absolute CBF decrease, and significant asymmetry in
middle cerebral artery CBF during balloon test occlusion
suggest a high risk of stroke during carotid occlusion.21 Lack
of an appropriate vasodilatory response to acetazolamide
indicates a vasculature that has already maximally dilated; this
decreased hemodynamic reserve is very predictive of subse-
quent infarct65 in patients with atherosclerotic occlusive
disease17 or those about to undergo surgery. The quantitative
values obtained with Xe-CT appear to be better able to guide
management in these patients than the relative changes seen
with modalities such as single photon emission computed
tomography (SPECT).21 For example, Chaer et al. found that
among 179 patients with carotid stenosis more than 50% who
underwent an acetazolamide-activated xenon CBF study,
those older than 70 years of age had reduced cerebrovascular
reserve, which may explain the increased risk of stroke in
older adult patients.66
Traumatic Brain Injury
Traumatic brain injury (TBI) management in the NCCU is
centered on the prevention of secondary cerebral insults, and
in part this depends on control of adequate CPP to maintain
CBF. Current guidelines recommend a CPP between 50 and
70 mm Hg; in this range Xe-CT studies show that CBF is fairly
constant, but that even lower CPP (50 mm Hg) can provide
adequate CBF in some patients.67 Autoregulation and CO2
reactivity68 frequently are abnormal in patients with severe
TBI and so a one-to-one relationship between CPP and CBF
is not always present. TBI pathophysiology is complex and
heterogeneous, and in these patients Xe-CT studies can help
establish the state of autoregulation and the impact of treat-
ments on brain physiology and validate the information pro-
vided by more local monitors such as an ICP or brain oxygen
monitor.69 Consistent with this, in a small pilot study it appears
that there is a reasonable correlation between xenon CBF
and oxygen saturation measured using an INVOS cerebral
oximeter (Somanetics; Troy, MI).70
Measurements with Xe-CT have provided insight into the
pathophysiology after TBI and may detect ischemia not oth-
erwise seen on CT, and which may not be associated with
other abnormalities; this may help direct early medical and
surgical management.71,72 For example, the CBF patterns
around traumatic intracerebral hemorrhage or contusions can
be used to determine who requires surgery73,74 or guide blood
pressure and CO2 levels to ensure adequate CBF to the peri-
contusional tissue.75,76 However, it does not appear that xenon
CBF studies can be used to predict which contusions will
enlarge.77 Xe-CT studies also may help with outcome predic-
tion in adults and children78 and in particular when there is
reduced brainstem CBF, outcome usually is poor.79 Robertson
et al. also have used xenon CBF to examine how genetic varia-
tion in endothelial nitric oxide synthase (NOS3), that plays a
role in CBF regulation, contributes to the brain’s response to
injury.80
In children the CBF response after TBI differs from adults.
Hyperemia with increased ICP can occur, and in these patients
quantitative CBF data will show no need to increase CPP.78 On
the other hand, the CBF data can help identify tissue at risk
for ischemia. This may be important because an ICP monitor
and plain head CT will not differentiate the cause of the ICP
increase.
Ischemic Stroke
Acute stroke is common, and apart from intravenous tissue
plasminogen activator (tPA) given within 3 to 4.5 hours of
symptom onset, there are few specific treatment options. Data
gained from Xe-CT and other CT modalities (e.g., CT angi-
ography) may help clarify patient physiology and identify a
group of patients who have potentially salvageable tissue.81-85
This is important because there is a great variability in the
proportion of the ischemic core compared with salvageable
penumbra that depends on individual vascular variability and
collateral flow (Fig. 27.12).86 Recanalization cannot re­cover
infracted core, but may recover penumbral tissue (i.e., it is
possible to predict patients who will respond based on the
amount of salvageable tissue rather than time frame alone).87
In addition, the Xe-CT identified abnormality can be useful
in outcome prediction.86 Furthermore, if a patient has a neu-
rologic deficit but a normal CBF study, he or she will be likely
to recover normal function without intervention. By contrast,
a patient with a very large volume of already infracted tissue
(large core with CBF values less than 9 mL/100 g/min) is at
risk for edema and herniation.88 The progression to life-
threatening swelling and herniation can be accurately pre-
dicted (e.g., early Xe-CT can predict patients at risk for
herniation after stroke and identify patients who may benefit
from decompressive craniectomy before herniation occurs).84,89
In addition, Xe-CT has been used to define supratentorial
hypoperfusion after brainstem strokes, particularly larger
pontine infarcts and guide therapy to prevent supratentorial
infarction.90
 Section IV—Radiology 279

100%

90%

80%

70%
% of MCA territory
60%

50%

40%

30%

20%

10%

0%
28 35 33 15 9 6 12 1 3 22 23 24 11 7 34 17 16 13 26 21 25 4 19 30 14 36 2 8 18 20 5 29 27 10 32 31

Patient #
Core Penumbra NC/NP

Fig. 27.12  A series of patients with ischemic middle cerebral artery (MCA) stroke, showing the variability in infarcted core. (From Jovin TG, Yonas H,
Gebel JM, et al: The cortical ischemic core and not the consistently present penumbra is a determinant of clinical outcome in acute middle cerebral artery occlusion. Stroke
2003;34:2426–33.)

Intracerebral Hemorrhage
Intracerebral hemorrhage (ICH), although less common than
ischemic stroke, is devastating but difficult to treat. In particu-
lar the role of surgery is unclear.91,92 Xe-CT studies may offer
a method to select patients who may benefit from surgical
evacuation of an ICH by demonstration of a persistent pen-
umbra around the ICH.93,94
Tumor
There are several potential applications for Xe-CT in the eval-
uation and treatment of intracranial tumors. First, blood flow
studies have identified CBF gradients that extend peripherally
from a tumor and heterogeneity of blood flow within various
tumor types. The CBF levels adjacent to the tumor, including
both malignant and benign tumors, may even be at ischemic
levels.95-97 In some patients dexamethasone may improve
regional CBF around malignant glial tumors, but in others it
may decrease. Surgical removal of the tumor on the other
hand reliably improves CBF.98 Second, Xe-CT CBF studies can
be used to help define tumor types or degree of malig-
nancy.99,100 However, these studies may only be required in
patients who are not able to undergo magnetic resonance
imaging (MRI) studies because advanced MRI techniques can
be used to define tumor pathology and anatomy.101
Epilepsy
During interictal periods CBF has been shown to increase
then return to normal in the postictal period.102 In status
epilepticus CBF can decrease and reach potentially ischemic
levels. This is followed by a period of several weeks of
increased CBF.103 These Xe-CT data may help guide possible
intervention.
Brain Death
Brain death is discussed in Chapter 13. It is a clinical diagnosis
but confirmatory tests (e.g., absence of flow on catheter angi-
ography or brain scintigraphy imaging) can be useful in select
patients. Xe-CT studies suggest that global CBF less than
5 mL/100 g/min is associated with brain death.104,105
Other Conditions
Xe-CT data can be used to assess patient physiology in a
variety of other conditions such as in ventilation response to
bronchodilators in asthma and chronic obstructive pulmo-
nary disease (COPD),106 to provide quantitative information
on pulmonary ventilation and function,107-110 as an objective
method to assess the state of the liver in chronic liver disease,111
and to assess patient physiology in carbon monoxide poison-
ing.112 For example, following cardiac arrest, return of reduced
CBF to normal levels during the first 48 hours is associated
with a good prognosis.113
Conclusion
Xe-CT CBF studies have a long history of use in patients with
cerebrovascular disease because the studies provide quantita-
tive CBF data and can be repeated easily to assess hemody-
namic reserve or treatment response. These physiologic data
can be used to guide management decisions in the NCCU,
where management often is focused on efforts to improve CBF
280 Section IV—Radiology
to prevent secondary brain injuries due to persistent or late-
onset ischemia. These studies may be best achieved with the
combination of a portable CT scanner and Xe-CT CBF analy-
sis. However, xenon currently needs to be used in research
protocols in the United States because it does not have FDA
approval for clinical use. Recent safety analysis studies in 2003
scans using 28% xenon suggest that Xe-CT should be made
widely available because it has a very low risk of adverse events
and no permanent morbidity while it provides useful clinical
information.33
References
1. Drayer BP, Gur D, Yonas H, et al. Abnormality of the xenon brain:blood
partition coefficient and blood flow in cerebral infarction: an in vivo
assessment using transmission computed tomography. Radiology 1980;135:
349–54.
2. Gur D, Good WF, Wolfson SK, Jr, et al. In vivo mapping of local cerebral
blood flow by xenon-enhanced computed tomography. Science 1982;215:
1267–8.
3. Gur D, Yonas H, Jackson DL, et al. Simultaneous measurements of cerebral
blood flow by the xenon/CT method and the microsphere method.
A comparison. Invest Radiol 1985;20:672–7.
4. Wolfson SK, Jr, Clark J, Greenberg JH, et al. Xenon-enhanced computed
tomography compared with [14C]iodoantipyrine for normal and low
cerebral blood flow states in baboons. Stroke 1990;21:751–7.
5. Tone O, Ito U, Tomita H, et al. Correlation between cerebral blood flow
values obtained by Xe-CT and Kety-Schmidt (N2O) methods. Acta Neurol
Scand Suppl 1996;166:18–21.
6. Nariai T. Comparison of measurement between Xe/CT CBF and PET in
cerebrovascular disease and brain tumor. Acta Neurol Scand Suppl 1996;
166:10–2.
7. Bergholt B, Ostergaard L, von Oettingen G, et al. Cerebral blood flow in
volunteers measured by PET and Xe CT/CBF: a comparison. Keio J Med
2000;49(Suppl 1):A55–7.
8. Valadka AB, Hlatky R, Furuya Y, et al. Brain tissue PO2: correlation with
cerebral blood flow. Acta Neurochir Suppl 2002;81:299–301.
9. Matsuda M, Lee H, Kuribayashi K, et al. Comparative study of regional
cerebral blood flow values measured by Xe CT and Xe SPECT. Acta Neurol
Scand Suppl 1996;166:13–6.
10. Gur D, Yonas H, Good WF. Local cerebral blood flow by xenon-enhanced
CT: current status, potential improvements, and future directions.
Cerebrovasc Brain Metab Rev 1989;1:68–86.
11. Johnson DW, Stringer WA, Marks MP, et al. Stable xenon CT cerebral blood
flow imaging: rationale for and role in clinical decision making. AJNR Am J
Neuroradiol 1991;12:201–13.
12. Yonas H. Cerebral blood measurements in vasospasm. Neurosurg Clin N Am
1990;1:307–18.
13. Yonas H, Darby JM, Marks EC, et al. CBF measured by Xe-CT: approach to
analysis and normal values. J Cereb Blood Flow Metab 1991;11:716–25.
14. Yonas H, Sekhar L, Johnson DW, et al. Determination of irreversible
ischemia by xenon-enhanced computed tomographic monitoring of cerebral
blood flow in patients with symptomatic vasospasm. Neurosurgery 1989;
24:368–72.
15. Gur AY, Bova I, Bornstein NM. Is impaired cerebral vasomotor reactivity a
predictive factor of stroke in asymptomatic patients? Stroke 1996;27:
2188–90.
16. Horowitz M, Yonas H, Albright AL. Evaluation of cerebral blood flow and
hemodynamic reserve in symptomatic moyamoya disease using stable
xenon-CT blood flow. Surg Neurol 1995;44:251–61; discussion 62.
17. Nemoto EM, Yonas H, Kuwabara H, et al. Identification of hemodynamic
compromise by cerebrovascular reserve and oxygen extraction fraction in
occlusive vascular disease. J Cereb Blood Flow Metab 2004;24:1081–9.
18. Pindzola RR, Sashin D, Nemoto EM, et al. Identifying regions of
compromised hemodynamics in symptomatic carotid occlusion by
cerebrovascular reactivity and oxygen extraction fraction. Neurol Res 2006;
28:149–54.
19. Tarr RW, Johnson DW, Rutigliano M, et al. Use of acetazolamide-challenge
xenon CT in the assessment of cerebral blood flow dynamics in patients
with arteriovenous malformations. AJNR Am J Neuroradiol 1990;11:
441–8.
20. Webster MW, Makaroun MS, Steed DL, et al. Compromised cerebral blood
flow reactivity is a predictor of stroke in patients with symptomatic carotid
artery occlusive disease. J Vasc Surg 1995;21:338–44; discussion 44–5.
21. Witt JP, Yonas H, Jungreis C. Cerebral blood flow response pattern during
balloon test occlusion of the internal carotid artery. AJNR Am J Neuroradiol
1994;15(5):847–56.
22. Yonas H, Kromer H, Jungreis C. Compromised vascular reserves does predict
subgroups with carotid occlusion and an increased stroke risk. J Stroke
Cerebrovasc Dis 1997;6:458.
23. Yonas H, Smith HA, Durham SR, et al. Increased stroke risk predicted by
compromised cerebral blood flow reactivity. J Neurosurg 1993;79:483–9.
24. Gunnarsson T, Theodorsson A, Karlsson P, et al. Mobile computerized
tomography scanning in the neurosurgery intensive care unit: increase in
patient safety and reduction of staff workload. J Neurosurg 2000;93:432–6.
25. Hillman J, Sturnegk P, Yonas H, et al. Bedside monitoring of CBF with
xenon-CT and a mobile scanner: a novel method in neurointensive care. Br J
Neurosurg 2005;19:395–401.
26. Sturnegk P, Mellergard P, Yonas H, et al. Potential use of quantitative bedside
CBF monitoring (Xe-CT) for decision making in neurosurgical intensive
care. Br J Neurosurg 2007;21:332–9.
27. Schroder ML, Muizelaar JP. Monitoring of regional cerebral blood flow
(CBF) in acute head injury by thermal diffusion. Acta Neurochir Suppl
(Wien) 1993;59:47–9.
28. Carlson AP, Yonas H. Portable head computed tomography scanner-
technology and applications: experience with 3421 scans. J Neuroimaging.
2011; Jun 23. Epub ahead of print.
29. Peace K, Maloney E, Frangos S, et al. The use of a portable head CT scanner
in the ICU. J Neurosci Nurs 2010;42(2):109–16.
30. Peace K, Maloney E, Frangos S, et al. Portable head CT scan and its effect on
intracranial pressure, cerebral perfusion pressure and brain oxygen.
J Neurosurgery 2011;114:1479–84.
31. Hecker K, Baumert JH, Horn N, et al. Xenon, a modern anaesthesia gas.
Minerva Anestesiol 2004;70:255–60.
32. Dickinson R, Peterson BK, Banks P, et al. Competitive inhibition at the
glycine site of the N-methyl-D-aspartate receptor by the anesthetics xenon
and isoflurane: evidence from molecular modeling and electrophysiology.
Anesthesiology 2007;107:756–67.
33. Carlson AP, Brown AM, Zager E, et al. Xenon-enhanced cerebral blood flow
at 28% xenon provides uniquely safe access to quantitative, clinically useful
cerebral blood flow information: a multicenter study. AJNR Am J
Neuroradiol 2011;32(7):1315–20. Epub 2011 Jun 23.
34. The Brain Trauma Foundation. The American Association of Neurological
Surgeons. The Joint Section on Neurotrauma and Critical Care. Initial
management. J Neurotrauma 2000;17:463–9.
35. Adelson PD, Bratton SL, Carney NA, et al. Guidelines for the acute medical
management of severe traumatic brain injury in infants, children, and
adolescents. Chapter 1: introduction. Pediatr Crit Care Med 2003;4:S2–4.
36. Haley EC, Jr, Kassell NF, Torner JC. The International Cooperative Study on
the Timing of Aneurysm Surgery. The North American experience. Stroke
1992;23:205–14.
37. Joseph M, Ziadi S, Nates J, et al. Increases in cardiac output can reverse flow
deficits from vasospasm independent of blood pressure: a study using xenon
computed tomographic measurement of cerebral blood flow. Neurosurgery
2003;53:1044–51; discussion 51–2.
38. Fisher CM, Kistler JP, Davis JM. Relation of cerebral vasospasm to
subarachnoid hemorrhage visualized by computerized tomographic
scanning. Neurosurgery 1980;6:1–9.
39. Lennihan L, Mayer SA, Fink ME, et al. Effect of hypervolemic therapy on
cerebral blood flow after subarachnoid hemorrhage: a randomized
controlled trial. Stroke 2000;31:383–91.
40. Tone O, Tomita H, Tamaki M, et al. Correlation between cardiac output and
cerebral blood flow following subarachnoid hemorrhage. Keio J Med
2000;49(Suppl 1):A151–3.
41. Tanaka A, Kumate S, Nakayama Y, et al. Postoperative subarachnoid clots
and the pattern of cerebral ischemia associated with symptomatic
vasospasm. Surg Neurol 1998;49:164–8; discussion 8–9.
42. Compagnone C, Tagliaferri F, Fainardi E, et al. Diagnostic impact of the
spectrum of ischemic cerebral blood flow thresholds in sedated
subarachnoid hemorrhage patients. Acta Neurochir Suppl 2006;96:53–6.
43. Schubert GA, Seiz M, Hegewald AA, et al. Acute hypoperfusion immediately
after subarachnoid hemorrhage: a xenon contrast-enhanced CT study.
J Neurotrauma 2009;26(12):2225–31.
44. Clyde BL, Resnick DK, Yonas H, et al. The relationship of blood velocity as
measured by transcranial Doppler ultrasonography to cerebral blood flow as
determined by stable xenon computed tomographic studies after aneurysmal
subarachnoid hemorrhage. Neurosurgery 1996;38:896–904.
45. Chieregato A, Sabia G, Tanfani A, et al. Xenon-CT and transcranial Doppler
in poor-grade or complicated aneurysmatic subarachnoid hemorrhage
patients undergoing aggressive management of intracranial hypertension.
Intensive Care Med 2006;32:1143–50.

46. Rothoerl RD, Woertgen C, Brawanski A. Hyperemia following aneurysmal
subarachnoid hemorrhage: incidence, diagnosis, clinical features, and
outcome. Intensive Care Med 2004;30:1298–302.
47. Firlik KS, Firlik AD, Yonas H. Xenon-enhanced computed tomography in the
management of cerebral vasospasm following aneurysmal subarachnoid
hemorrhage. Keio J Med 2000;49(Suppl 1):A148–50.
48. Tseng MY, Al-Rawi PG, Pickard JD, et al. Effect of hypertonic saline on
cerebral blood flow in poor-grade patients with subarachnoid hemorrhage.
Stroke 2003;34:1389–96.
Section IV—Radiology 281
49. Al-Rawi PG, Tseng MY, Richards HK, et al. Hypertonic saline in patients
with poor-grade subarachnoid hemorrhage improves cerebral blood flow,
brain tissue oxygen, and pH. Stroke 2010;41(1):122–8. Epub 2009 Nov 12.
50. Darby JM, Yonas H, Marks EC, et al. Acute cerebral blood flow response to
dopamine-induced hypertension after subarachnoid hemorrhage.
J Neurosurg 1994;80:857–64.
A complete list of references for this chapter can be found online at
www.expertconsult.com.

References
1. Drayer BP, Gur D, Yonas H, et al. Abnormality of the xenon brain:blood
partition coefficient and blood flow in cerebral infarction: an in vivo
assessment using transmission computed tomography. Radiology 1980;
135:349–54.
2. Gur D, Good WF, Wolfson SK, Jr, et al. In vivo mapping of local cerebral
blood flow by xenon-enhanced computed tomography. Science 1982;215:
1267–8.
3. Gur D, Yonas H, Jackson DL, et al. Simultaneous measurements of cerebral
blood flow by the xenon/CT method and the microsphere method. A
comparison. Invest Radiol 1985;20:672–7.
4. Wolfson SK, Jr, Clark J, Greenberg JH, et al. Xenon-enhanced computed
tomography compared with [14C]iodoantipyrine for normal and low
cerebral blood flow states in baboons. Stroke 1990;21:751–7.
5. Tone O, Ito U, Tomita H, et al. Correlation between cerebral blood flow
values obtained by Xe-CT and Kety-Schmidt (N2O) methods. Acta Neurol
Scand Suppl 1996;166:18–21.
6. Nariai T. Comparison of measurement between Xe/CT CBF and PET in
cerebrovascular disease and brain tumor. Acta Neurol Scand Suppl 1996;
166:10–2.
7. Bergholt B, Ostergaard L, von Oettingen G, et al. Cerebral blood flow in
volunteers measured by PET and Xe CT/CBF: a comparison. Keio J Med
2000;49(Suppl 1):A55–7.
8. Valadka AB, Hlatky R, Furuya Y, et al. Brain tissue PO2: correlation with
cerebral blood flow. Acta Neurochir Suppl 2002;81:299–301.
9. Matsuda M, Lee H, Kuribayashi K, et al. Comparative study of regional
cerebral blood flow values measured by Xe CT and Xe SPECT. Acta Neurol
Scand Suppl 1996;166:13–6.
10. Gur D, Yonas H, Good WF. Local cerebral blood flow by xenon-enhanced
CT: current status, potential improvements, and future directions.
Cerebrovasc Brain Metab Rev 1989;1:68–86.
11. Johnson DW, Stringer WA, Marks MP, et al. Stable xenon CT cerebral blood
flow imaging: rationale for and role in clinical decision making. AJNR Am J
Neuroradiol 1991;12:201–13.
12. Yonas H. Cerebral blood measurements in vasospasm. Neurosurg Clin N
Am 1990;1:307–18.
13. Yonas H, Darby JM, Marks EC, et al. CBF measured by Xe-CT: approach to
analysis and normal values. J Cereb Blood Flow Metab 1991;11:716–25.
14. Yonas H, Sekhar L, Johnson DW, et al. Determination of irreversible
ischemia by xenon-enhanced computed tomographic monitoring of
cerebral blood flow in patients with symptomatic vasospasm. Neurosurgery
1989;24:368–72.
15. Gur AY, Bova I, Bornstein NM. Is impaired cerebral vasomotor reactivity a
predictive factor of stroke in asymptomatic patients? Stroke 1996;27:
2188–90.
16. Horowitz M, Yonas H, Albright AL. Evaluation of cerebral blood flow and
hemodynamic reserve in symptomatic moyamoya disease using stable
xenon-CT blood flow. Surg Neurol 1995;44:251–61; discussion 62.
17. Nemoto EM, Yonas H, Kuwabara H, et al. Identification of hemodynamic
compromise by cerebrovascular reserve and oxygen extraction fraction in
occlusive vascular disease. J Cereb Blood Flow Metab 2004;24:1081–9.
18. Pindzola RR, Sashin D, Nemoto EM, et al. Identifying regions of
compromised hemodynamics in symptomatic carotid occlusion by
cerebrovascular reactivity and oxygen extraction fraction. Neurol Res 2006;
28:149–54.
19. Tarr RW, Johnson DW, Rutigliano M, et al. Use of acetazolamide-challenge
xenon CT in the assessment of cerebral blood flow dynamics in patients
with arteriovenous malformations. AJNR Am J Neuroradiol 1990;11:441–8.
20. Webster MW, Makaroun MS, Steed DL, et al. Compromised cerebral blood
flow reactivity is a predictor of stroke in patients with symptomatic carotid
artery occlusive disease. J Vasc Surg 1995;21:338–44; discussion 44–5.
21. Witt JP, Yonas H, Jungreis C. Cerebral blood flow response pattern during
balloon test occlusion of the internal carotid artery. AJNR Am J
Neuroradiol 1994;15(5):847–56.
22. Yonas H, Kromer H, Jungreis C. Compromised vascular reserves does
predict subgroups with carotid occlusion and an increased stroke risk.
J Stroke Cerebrovasc Dis 1997;6:458.
23. Yonas H, Smith HA, Durham SR, et al. Increased stroke risk predicted by
compromised cerebral blood flow reactivity. J Neurosurg 1993;79:483–9.
24. Gunnarsson T, Theodorsson A, Karlsson P, et al. Mobile computerized
tomography scanning in the neurosurgery intensive care unit: increase in
patient safety and reduction of staff workload. J Neurosurg 2000;93:432–6.
25. Hillman J, Sturnegk P, Yonas H, et al. Bedside monitoring of CBF with
xenon-CT and a mobile scanner: a novel method in neurointensive care. Br
J Neurosurg 2005;19:395–401.
Section IV—Radiology 281.e1
26. Sturnegk P, Mellergard P, Yonas H, et al. Potential use of quantitative
bedside CBF monitoring (Xe-CT) for decision making in neurosurgical
intensive care. Br J Neurosurg 2007;21:332–9.
27. Schroder ML, Muizelaar JP. Monitoring of regional cerebral blood flow
(CBF) in acute head injury by thermal diffusion. Acta Neurochir Suppl
(Wien) 1993;59:47–9.
28. Carlson AP, Yonas H. Portable head computed tomography scanner-
technology and applications: experience with 3421 scans. J Neuroimaging
2011; Jun 23. Epub ahead of print.
29. Peace K, Maloney E, Frangos S, et al. The use of a portable head CT
scanner in the ICU. J Neurosci Nurs 2010;42(2):109–16.
30. Peace K, Maloney E, Frangos S, et al. Portable head CT scan and its effect
on intracranial pressure, cerebral perfusion pressure and brain oxygen.
J Neurosurgery 2011;114:1479–84.
31. Hecker K, Baumert JH, Horn N, et al. Xenon, a modern anaesthesia gas.
Minerva Anestesiol 2004;70:255–60.
32. Dickinson R, Peterson BK, Banks P, et al. Competitive inhibition at the
glycine site of the N-methyl-D-aspartate receptor by the anesthetics xenon
and isoflurane: evidence from molecular modeling and electrophysiology.
Anesthesiology 2007;107:756–67.
33. Carlson AP, Brown AM, Zager E, et al. Xenon-enhanced cerebral blood flow
at 28% xenon provides uniquely safe access to quantitative, clinically useful
cerebral blood flow information: a multicenter study. AJNR Am J
Neuroradiol 2011;32(7):1315–20. Epub 2011 Jun 23.
34. The Brain Trauma Foundation. The American Association of Neurological
Surgeons. The Joint Section on Neurotrauma and Critical Care. Initial
management. J Neurotrauma 2000;17:463–9.
35. Adelson PD, Bratton SL, Carney NA, et al. Guidelines for the acute
medical management of severe traumatic brain injury in infants,
children, and adolescents. Chapter 1: introduction. Pediatr Crit Care Med
2003;4:S2–4.
36. Haley EC, Jr, Kassell NF, Torner JC. The International Cooperative Study on
the Timing of Aneurysm Surgery. The North American experience. Stroke
1992;23:205–14.
37. Joseph M, Ziadi S, Nates J, et al. Increases in cardiac output can reverse
flow deficits from vasospasm independent of blood pressure: a study using
xenon computed tomographic measurement of cerebral blood flow.
Neurosurgery 2003;53:1044–51; discussion 51–2.
38. Fisher CM, Kistler JP, Davis JM. Relation of cerebral vasospasm to
subarachnoid hemorrhage visualized by computerized tomographic
scanning. Neurosurgery 1980;6:1–9.
39. Lennihan L, Mayer SA, Fink ME, et al. Effect of hypervolemic therapy on
cerebral blood flow after subarachnoid hemorrhage: a randomized
controlled trial. Stroke 2000;31:383–91.
40. Tone O, Tomita H, Tamaki M, et al. Correlation between cardiac output
and cerebral blood flow following subarachnoid hemorrhage. Keio J Med
2000;49(Suppl 1):A151–3.
41. Tanaka A, Kumate S, Nakayama Y, et al. Postoperative subarachnoid clots
and the pattern of cerebral ischemia associated with symptomatic
vasospasm. Surg Neurol 1998;49:164–8; discussion 8–9.
42. Compagnone C, Tagliaferri F, Fainardi E, et al. Diagnostic impact of the
spectrum of ischemic cerebral blood flow thresholds in sedated
subarachnoid hemorrhage patients. Acta Neurochir Suppl 2006;96:53–6.
43. Schubert GA, Seiz M, Hegewald AA, et al. Acute hypoperfusion
immediately after subarachnoid hemorrhage: a xenon contrast-enhanced
CT study. J Neurotrauma 2009;26(12):2225–31.
44. Clyde BL, Resnick DK, Yonas H, et al. The relationship of blood velocity as
measured by transcranial Doppler ultrasonography to cerebral blood flow
as determined by stable xenon computed tomographic studies after
aneurysmal subarachnoid hemorrhage. Neurosurgery 1996;38:896–904.
45. Chieregato A, Sabia G, Tanfani A, et al. Xenon-CT and transcranial Doppler
in poor-grade or complicated aneurysmatic subarachnoid hemorrhage
patients undergoing aggressive management of intracranial hypertension.
Intensive Care Med 2006;32:1143–50.
46. Rothoerl RD, Woertgen C, Brawanski A. Hyperemia following aneurysmal
subarachnoid hemorrhage: incidence, diagnosis, clinical features, and
outcome. Intensive Care Med 2004;30:1298–302.
47. Firlik KS, Firlik AD, Yonas H. Xenon-enhanced computed tomography in
the management of cerebral vasospasm following aneurysmal subarachnoid
hemorrhage. Keio J Med 2000;49(Suppl 1):A148–50.
48. Tseng MY, Al-Rawi PG, Pickard JD, et al. Effect of hypertonic saline on
cerebral blood flow in poor-grade patients with subarachnoid hemorrhage.
Stroke 2003;34:1389–96.
49. Al-Rawi PG, Tseng MY, Richards HK, et al. Hypertonic saline in patients
with poor-grade subarachnoid hemorrhage improves cerebral blood flow,
brain tissue oxygen, and pH. Stroke 2010;41(1):122–8. Epub 2009 Nov 12.
281.e2 Section IV—Radiology
50. Darby JM, Yonas H, Marks EC, et al. Acute cerebral blood flow response to
dopamine-induced hypertension after subarachnoid hemorrhage.
J Neurosurg 1994;80:857–64.
51. Ract C, Vigue B. Comparison of the cerebral effects of dopamine and
norepinephrine in severely head-injured patients. Intensive Care Med
2001;27:101–6.
52. Nussbaum ES, Sebring LA, Ganz WF, et al. Intra-aortic balloon
counterpulsation augments cerebral blood flow in the patient with cerebral
vasospasm: a xenon-enhanced computed tomography study. Neurosurgery
1998;42:206–13; discussion 13–4.
53. Firlik KS, Kaufmann AM, Firlik AD, et al. Intra-arterial papaverine for the
treatment of cerebral vasospasm following aneurysmal subarachnoid
hemorrhage. Surg Neurol 1999;51:66–74.
54. Clyde BL, Firlik AD, Kaufmann AM, et al. Paradoxical aggravation of
vasospasm with papaverine infusion following aneurysmal subarachnoid
hemorrhage. Case report. J Neurosurg 1996;84:690–5.
55. Murai Y, Kominami S, Kobayashi S, et al. The long-term effects of
transluminal balloon angioplasty for vasospasms after subarachnoid
hemorrhage: analyses of cerebral blood flow and reactivity. Surg Neurol
2005;64:122–6; discussion 7.
56. Firlik AD, Kaufmann AM, Jungreis CA, et al. Effect of transluminal
angioplasty on cerebral blood flow in the management of symptomatic
vasospasm following aneurysmal subarachnoid hemorrhage. J Neurosurg
1997;86:830–9.
57. Linskey ME, Horton JA, Rao GR, et al. Fatal rupture of the intracranial
carotid artery during transluminal angioplasty for vasospasm induced by
subarachnoid hemorrhage. Case report. J Neurosurg 1991;74:985–90.
58. Penn AA, Schomer DF, Steinberg GK. Imaging studies of cerebral
hyperperfusion after carotid endarterectomy. Case report. J Neurosurg
1995;83:133–7.
59. Yamane K, Shima T, Nishida M, et al. Changes in cerebral blood flow after
carotid endarterectomy. Keio J Med 2000;49(Suppl 1):A80–2.
60. Van Roost D, Schramm J, Solymosi L, et al. Presence and removal of
arteriovenous malformation: impact of regional cerebral blood flow, as
assessed with xenon/CT. Acta Neurol Scand Suppl 1996;166:136–8.
61. Schubert GA, Weinmann C, Seiz M, et al. Cerebrovascular insufficiency as
the criterion for revascularization procedures in selected patients:
a correlation study of xenon contrast-enhanced CT and PWI. Neurosurg
Rev 2009;32(1):29–35; discussion 35–6. Epub 2008 Sep 13.
62. Czabanka M, Peña-Tapia P, Scharf J, et al. Characterization of direct and
indirect cerebral revascularization for the treatment of European patients
with moyamoya disease. Cerebrovasc Dis 2011;32(4):361–9. Epub 2011
Sep 15.
63. Zaharchuk G, Do HM, Marks MP, et al. Arterial spin-labeling MRI can
identify the presence and intensity of collateral perfusion in patients with
moyamoya disease. Stroke 2011;42(9):2485–91. Epub 2011 Jul 28.
64. Huang AP, Liu HM, Lai DM, et al. Clinical significance of posterior
circulation changes after revascularization in patients with moyamoya
disease. Cerebrovasc Dis 2009;28(3):247–57. Epub 2009 Jul 14.
65. Field M, Jungreis CA, Chengelis N, et al. Symptomatic cavernous sinus
aneurysms: management and outcome after carotid occlusion and selective
cerebral revascularization. AJNR Am J Neuroradiol 2003;24:1200–7.
66. Chaer RA, Shen J, Rao A, et al. Cerebral reserve is decreased in elderly
patients with carotid stenosis. J Vasc Surg 2010;52(3):569–74; discussion
574–5.
67. Chieregato A, Tanfani A, Compagnone C, et al. Global cerebral blood flow
and CPP after severe head injury: a xenon-CT study. Intensive Care Med
2007;33:856–62.
68. Kushi H, Moriya T, Saito T, et al. Importance of metabolic monitoring
systems as an early prognostic indicator in severe head injured patients.
Acta Neurochir Suppl 1999;75:67–8.
69. Menzel M, Doppenberg EM, Zauner A, et al. Cerebral oxygenation in
patients after severe head injury: monitoring and effects of arterial
hyperoxia on cerebral blood flow, metabolism and intracranial pressure.
J Neurosurg Anesthesiol 1999;11:240–51.
70. Shafer R, Brown A, Taylor C. Correlation between cerebral blood flow and
oxygen saturation in patients with subarachnoid hemorrhage and traumatic
brain injury. J Neurointerv Surg 2011;Mar 25. Epub ahead of print.
71. Bouma GJ, Muizelaar JP, Stringer WA, et al. Ultra-early evaluation of
regional cerebral blood flow in severely head-injured patients using
xenon-enhanced computerized tomography. J Neurosurg 1992;77:360–8.
72. Latchaw RE, Yonas H, Darby JM, et al. Xe-CT cerebral blood flow
determination following cranial trauma. Acta Radiol Suppl 1986;369:370–3.
73. Chieregato A, Fainardi E, Servadei F, et al. Centrifugal distribution of
regional cerebral blood flow and its time course in traumatic intracerebral
hematomas. J Neurotrauma 2004;21:655–66.
74. von Oettingen G, Bergholt B, Gyldensted C, et al. Blood flow and ischemia
within traumatic cerebral contusions. Neurosurgery 2002;50:781–8;
discussion 8–90.
75. Chieregato A, Fainardi E, Tanfani A, et al. Induced acute arterial
hypertension and regional cerebral flow in intracontusional low density
area. Acta Neurochir Suppl 2003;86:361–5.
76. McLaughlin MR, Marion DW. Cerebral blood flow and vasoresponsivity
within and around cerebral contusions. J Neurosurg 1996;85:871–6.
77. Depreitere B, Aviv R, Symons S, et al. Study of perfusion in and around
cerebral contusions by means of computed tomography. Acta Neurochir
Suppl 2008;102:259–62.
78. Adelson PD, Clyde B, Kochanek PM, et al. Cerebrovascular response in
infants and young children following severe traumatic brain injury:
a preliminary report. Pediatr Neurosurg 1997;26:200–7.
79. Ritter AM, Muizelaar JP, Barnes T, et al. Brain stem blood flow, pupillary
response, and outcome in patients with severe head injuries. Neurosurgery
1999;44:941–8.
80. Robertson CS, Gopinath SP, Valadka AB, et al. Variants of the endothelial
nitric oxide gene and cerebral blood flow after severe traumatic brain
injury. J Neurotrauma 2011;28(5):727–37. Epub 2011 May 4.
81. Kilpatrick MM, Yonas H, Goldstein S, et al. CT-based assessment of acute
stroke: CT, CT angiography, and xenon-enhanced CT cerebral blood flow.
Stroke 2001;32:2543–9.
82. Meyer JS, Rauch GM. Why emergency XeCT-CBF should become routine
in acute ischemic stroke before thrombolytic therapy. Keio J Med 2000;
49(Suppl 1):A25–8.
83. Firlik AD, Kaufmann AM, Wechsler LR, et al. Quantitative cerebral blood
flow determinations in acute ischemic stroke: relationship to computed
tomography and angiography. Stroke 1997;28:2208–13.
84. Firlik AD, Yonas H, Kaufmann AM, et al. Relationship between cerebral
blood flow and the development of swelling and life-threatening herniation
in acute ischemic stroke. J Neurosurg 1998;89:243–9.
85. Hughes RL, Yonas H, Gur D, et al. Cerebral blood flow determination
within the first 8 hours of cerebral infarction using stable xenon-enhanced
computed tomography. Stroke 1989;20:754–60.
86. Jovin TG, Yonas H, Gebel JM, et al. The cortical ischemic core and not the
consistently present penumbra is a determinant of clinical outcome in
acute middle cerebral artery occlusion. Stroke 2003;34:2426–33.
87. Gupta R, Jovin TG, Yonas H. Xenon CT cerebral blood flow in acute stroke.
Neuroimaging Clin N Am 2005;15:531–42, x.
88. Firlik AD, Yonas H. The utility of XeCT cerebral blood flow in the
management of acute stroke. Keio J Med 2000;49(Suppl 1):A129–30.
89. Kalia KK, Yonas H. An aggressive approach to massive middle cerebral
artery infarction. Arch Neurol 1993;50:1293–7.
90. Miyazawa N, Uchida M, Fukamachi A, et al. Xenon contrast-enhanced CT
imaging of supratentorial hypoperfusion in patients with brain stem
infarction. AJNR Am J Neuroradiol 1999;20:1858–62.
91. Mendelow AD, Gregson BA, Fernandes HM, et al. Early surgery versus
initial conservative treatment in patients with spontaneous supratentorial
intracerebral haematomas in the International Surgical Trial in
Intracerebral Haemorrhage (STICH): a randomised trial. Lancet
2005;365:387–97.
92. Bhattathiri PS, Gregson B, Prasad KS, et al. Intraventricular hemorrhage
and hydrocephalus after spontaneous intracerebral hemorrhage: results
from the STICH trial. Acta Neurochir Suppl 2006;96:65–8.
93. Kitahara T, Yamashita T, Kashiwagi S, et al. Hemodynamics of hypertensive
putaminal hemorrhage evaluated by xenon-enhanced computed
tomography and acetazolamide test. Acta Neurol Scand Suppl
1996;166:139–43.
94. Ding HY, Han X, Lv CZ, et al. Xenon-CT study of regional cerebral blood
flow around hematoma in patients with basal ganglia hemorrhage. Acta
Neurochir Suppl 2008;105:161–4.
95. Mirzai S, Samii M. Current status and future challenges in cerebral blood
flow mapping in intracranial tumors. Keio J Med 2000;49(Suppl 1):
A16–24.
96. Behrens PF, Ostertag CB, Warnke PC. Regional cerebral blood flow in
peritumoral brain edema during dexamethasone treatment: a xenon-
enhanced computed tomographic study. Neurosurgery 1998;43:235–40;
discussion 40–1.
97. Tatagiba M, Mirzai S, Samii M. Peritumoral blood flow in intracranial
meningiomas. Neurosurgery 1991;28:400–4.
98. Van Roost D, Hartmann A, Quade G. Changes of cerebral blood flow
following dexamethasone treatment in brain tumour patients: a Xe/CT
study. Acta Neurochir (Wien) 2001;143:37–43.
99. Nakatsuka M, Mizuno S. Blood flow study of meningothelial and fibrous
meningiomas by xenon-CT. Keio J Med 2000;49(Suppl 1):A164–7.

100. Nasel C, Rossler K, Heimberger K, et al. Tumour blood flow and partition
coefficients: correlation with grade of cerebral gliomas using xenon-
enhanced computed tomography. Neuroradiology 1997;39:627–32.
101. Wang S, Kim S, Chawla S, et al. Differentiation between glioblastomas,
solitary brain metastases, and primary cerebral lymphomas using diffusion
tensor and dynamic susceptibility contrast-enhanced MR imaging. AJNR
Am J Neuroradiol 2011;32(3):507–1.
102. Johnson DW, Hogg JP, Dasheiff R, et al. Xe-CT cerebral blood flow studies
during continuous depth electrode monitoring in epilepsy patients. AJNR
Am J Neuroradiol 1993;14:245–52.
103. Okuchi K, Nagata K, Otsuka H, et al. Regional cerebral blood flow after
status epilepticus. Keio J Med 2000;49(Suppl 1):A75–6.
104. Pistoia F, Johnson DW, Darby JM, et al. The role of xenon CT
measurements of cerebral blood flow in the clinical determination of brain
death. AJNR Am J Neuroradiol 1991;12:97–103.
105. Darby JM, Yonas H, Gur D, et al. Xenon-enhanced computed tomography
in brain death. Arch Neurol 1987;44:551–4.
106. Goo HW, Yu J. Redistributed regional ventilation after the administration
of a bronchodilator demonstrated on xenon-inhaled dual-energy CT in a
patient with asthma. Korean J Radiol 2011;12(3):386–9. Epub 2011 Apr 25.
107. Sase S, Nakano H, Suzuki H, et al. Subtraction lung image for evaluating
pulmonary ventilation in xenon-enhanced CT. Med Phys
2010;37(8):4464–74.
Section IV—Radiology 281.e3
108. Park EA, Goo JM, Park SJ, et al. Chronic obstructive pulmonary disease:
quantitative and visual ventilation pattern analysis at xenon ventilation CT
performed by using a dual-energy technique. Radiology 2010;256(3):985–
97. Epub 2010 Jul 22.
109. Thieme SF, Hoegl S, Nikolaou K, et al. Pulmonary ventilation and perfusion
imaging with dual-energy CT. Eur Radiol 2010;20(12):2882–9. Epub 2010
Jun 23.
110. Goo HW, Yang DH, Hong SJ, et al. Xenon ventilation CT using dual-source
and dual-energy technique in children with bronchiolitis obliterans:
correlation of xenon and CT density values with pulmonary function test
results. Pediatr Radiol 2010;40(9):1490–7. Epub 2010 Apr 22.
111. Takahashi H, Suzuki M, Ikeda H, et al. Evaluation of quantitative portal
venous, hepatic arterial, and total hepatic tissue blood flow using xenon CT
in alcoholic liver cirrhosis-comparison with liver cirrhosis related to
hepatitis C virus and nonalcoholic steatohepatitis. Alcohol Clin Exp Res
2010;34 Suppl 1:S7–S13. Epub 2009 Nov 5.
112. Sesay M, Bidabe AM, Guyot M, et al. Regional cerebral blood flow
measurements with Xenon-CT in the prediction of delayed encephalopathy
after carbon monoxide intoxication. Acta Neurol Scand Suppl
1996;166:22–7.
113. Inoue Y, Shiozaki T, Irisawa T, et al. Acute cerebral blood flow variations
after human cardiac arrest assessed by stable xenon enhanced computed
tomography. Curr Neurovasc Res 2007;4:49–54.
IV
Chapter
28  
MRI in Neurocritical Care
Damien Galanaud and Louis Puybasset
Introduction
Knowing the extent of brain damage in the context of acute
cerebral injury can be of paramount importance, especially
when the patient is at crossroads, when a decision should be
made about whether or not to continue care. Although clinical
or electrophysiologic examinations provide some informa-
tion, they are not always reliable because of the degree of
sedation these patients receive. Computed tomography (CT)
is the first line of imaging examination to rule out a neuro-
surgical emergency (e.g., extra- or subdural hematoma) or
suggest increased intracranial pressure (ICP). However, CT
cannot provide a reliable evaluation of the prognosis because
of its poor sensitivity (especially in the posterior fossa) and
lack of specificity. Magnetic resonance imaging (MRI) has
excellent sensitivity to demonstrate brain lesions and charac-
terize different pathologic processes and so can be the neuro-
imaging procedure of choice in many patients. This chapter
reviews MRI brain imaging, MRI findings in select patholo-
gies encountered in the neurocritical care unit (NCCU) (e.g.,
traumatic brain injury, stroke, infection, anoxic injury, and
hemorrhage among others), and use of MRI including
advanced techniques in estimating prognosis. MRI imaging of
the spine, extracranial vasculature, or other organ systems
(e.g., cardiac MRI) is beyond the scope of this chapter,
although all can be relevant to patient care.1-4
MRI in a Brain-Injured Patient:
the Procedure
Contraindications and Safety Rules
Performing an MRI in a critically brain-injured patient is a
challenging task. The MRI scanner is mostly a very powerful
magnet (1.5-3 tesla), inside which the subject is almost com-
pletely inaccessible during the 30- to 60-minute examination.
There are several consequences of these technical hurdles:
 All medical devices associated with the patient must be
MR compatible. In general, since 1995 implantable devices
(e.g., stents, cardiac valves, neurosurgical clips, or coils
among others) are almost always MR compatible at
1.5 tesla. At higher field strength, however, their safety has
282
not been established. MR examinations often are possible
at 1.5 tesla in patients with pacemakers, neurostimulators,
and implantable defibrillators but remain dangerous, and
require extra caution. The presence of ferromagnetic
metals inside the eye or central nervous system, such as
older aneurysm clips, is a contraindication to MRI.
 Patient monitoring is hampered because only limited
visual observation of the subject is possible, and the moni-
toring devices, designed for use in a high magnetic field
are not as reliable as their counterparts used in the inten-
sive care unit (ICU). Thus MRI examination should not
be attempted in respiratory or hemodynamically unstable
patients, unless the procedure is absolutely needed.
 The MRI examination should not be attempted in patients
with intracranial hypertension because a prolonged flat
position can increase ICP.
 Electric syringes do not operate properly in the magnetic
field and have to be located away from the patient and the
magnet. This requires long intravenous (IV) tubing that
may slow or alter drug or fluid delivery.
 MRI contrast media injections should be avoided in
patients with severe renal insufficiency because this may
lead to nephrogenic systemic fibrosis.
MRI Sequences
An ever-increasing number of sequences can be performed to
examine the brain. Because the duration of the examination
is limited, great care should be taken to select the most appro-
priate sequence for the clinical question. Routinely performed
MRI sequences include the following:
 T1-weighted sequences are the “anatomic” sequences.
They are used mostly to detect morphologic changes, such
as ventricle enlargement or brain atrophy, and to detect
subacute parenchymal blood (methemoglobin, areas of
high signal) and areas of contrast enhancement.
 T2-weighted sequences are used to demonstrate lesions
(contusions, ischemia, edema), which generally appear as
areas of increased signal. In neuroimaging the most widely
used T2-weighted sequence is the fluid-attenuated inver-
sion recovery (FLAIR) sequence in which the signal of the
© Copyright 2013 Elsevier Inc. All rights reserved.

cerebrospinal fluid (CSF) is suppressed. In addition to
its superb ability to demonstrate areas of parenchymal
increased signal, FLAIR plays an important role in dem-
onstrating acute subarachnoid hemorrhage (SAH).
 T2*-weighted sequences are specially designed to demon-
strate intraparenchymal blood, both acute (deoxyhemo-
globin) and chronic (hemosiderin), which appears as areas
of decreased signal, the latter persisting for years.
 MR angiography (MRA) can be performed to examine the
arteries of the circle of Willis or MR venography (MRV)
to evaluate the intracranial veins and sinuses.
 Diffusion-weighted imaging studies the random move-
ments of water molecules. It is represented as the apparent
diffusion coefficient (ADC), which is a physical character-
istic of the tissue. Only a very limited number of patho-
logic processes can decrease the ADC, the most widely
encountered being acute ischemia.
These sequences are available on all MR scanners and can be
routinely performed. More advanced acquisitions, where
available, can be used to better characterize the pathologic
processes and the extent of brain damage. Indeed, normally
appearing brain tissue on T2-weighted sequences can be either
normal or abnormal according to what other sequences
are used:
 Diffusion tensor imaging (DTI) is a variant of diffusion
imaging that studies the spatial restriction of diffusion.
The presence of white matter tracts “channels” the water
molecules, which move in a preferential direction. The
preferential restriction of diffusivity is measured by the
fractional anisotropy (FA). This sequence can also be used
to follow individual white matter tracts, such as the pyra-
midal and spinothalamic. This processing is called fiber
tracking.
 MR spectroscopy (MRS) studies brain metabolism. This
sequence quantifies molecules specific to some cell types
or physiologic processes that may assist in differentiating
brain tumors from radiation necrosis or from an abscess
among others.5 The most easily detected are as follows:
 N-acetyl aspartate (NAA) is a neuronal marker synthe-
sized in the mitochondria of neurons from the amino
acid aspartic acid, which decreases where there is
neuron damage or death. For example, a reduction in
NAA is observed after stroke reaching a nadir about 2
weeks after symptom onset.6
 Creatine (Cr), a ubiquitous molecule involved in ener-
getic metabolism. It is often used as a reference.
 Choline (Cho) increases in cases of membrane synthe-
sis (e.g., brain tumors) or catabolism.
 Blood oxygen level–dependent (BOLD) functional MRI
(fMRI) is less commonly used because it is difficult to
perform in a sedated patient. It has mostly been used
after the acute phase, in vegetative or minimally con-
scious patients, to search for the presence of residual
high order functional activities.7-11 The coherent sponta-
neous BOLD fluctuations, observed with the resting state
fMRI method, which do not require the patient’s cooper-
ation, have been shown to be still present under general
anesthesia and could possibly be useful to evaluate the
persistence of structured brain activities in comatose
patients.10,12
Section IV—Radiology 283
MRI of Major Pathologies
Ischemic Stroke
Acute ischemic stroke (AIS) rarely alters consciousness, except
in massive hemispheric infarcts with brain herniation or
basilar artery occlusion. However, the diagnosis of AIS should
be considered when an ICU patient develops a new neurologic
deficit, especially if an early CT scan is normal or precipitating
factors such as cardiac or vascular surgery are present, or a
brainstem or posterior fossa stroke is likely.13,14 MRI is the
examination of choice to confirm the diagnosis. A rapid (less
than an hour after symptom onset) decrease of the ADC of
the affected brain area is seen on MRI.15,16 FLAIR is normal
within the first 4 to 6 hours and then shows areas of increased
signal. MRA may demonstrate the level of vascular obstruc-
tion. The presence of an area of decreased ADC and normal
FLAIR is highly sensitive and specific of AIS (Fig. 28.1). MRI
findings in acute stroke are listed in Table 28.1.
The gold standard imaging technique to assess cerebral
hemodynamics (i.e., flow and metabolism) is positron emis-
sion tomography (PET) (see Chapter 29), but novel MRI
approaches provide methods to measure cerebral blood flow
(CBF), cerebral blood volume (CBV), and cerebral metabolic
rate of oxygen (CMRO2).17
 When perfusion-weighted imaging (PWI) is used with
diffusion-weighted imaging (DWI), the ischemic penum-
bra often can be identified.18-20 In PWI, protons in arterial
blood to the brain are magnetically tagged; this creates an
intra-arterial contrast with the intracranial contents and
so allows for calculation of perfusion maps (i.e., cerebral
blood flow). PWI is relatively time consuming to obtain,
but a PWI-DWI mismatch in which there is still viable
tissue but inadequate flow identifies tissue at risk for
infarction. This information can be used to guide throm-
bolytic therapy. Alternatively, a DWI-FLAIR mismatch can
be used to guide thrombolytic therapy.21 Compared with
PET studies MRI evaluation may overestimate the size of
the penumbra.22
 Advances in phase contrast MRA allow noninvasive assess-
ment of blood flow rates and flow direction through quan-
titative magnetic resonance angiography (qMRA). This
permits evaluation of CBF and collateral circulation in
patients with stroke, transient ischemic attacks (TIAs), or
occlusive cerebral vascular disease, or after bypass surgery.23
 Arterial spin labeling (ASL) MRI, including pulsed ASL
(PASL) or continuous ASL (CASL), remains the most
popular MRI method used to measure CBF.24 ASL does not
require exogenous contrast and is best used to measure
CBF in the cerebral cortex. Results are less robust in white
matter.25 A newer ASL technique, vessel-selective ASL (VS-
ASL) permits separate labeling of feeding arteries such that
perfusion territories can be measured.26
 CBV can be measured by modified ASL approaches.27 In
addition, CBV may be noninvasively measured using vas-
cular space occupancy (VASO) MRI and its variant “inflow
VASO” (iVASO),28 or through quantitative STAR labeling
of arterial regions (QUASAR).27
 Decisions about CBF require an understanding about
metabolism (CMRO2). For example, brain tissue remains
viable when CMRO2 is preserved through increased oxygen
284 Section IV—Radiology

A B C

D E F G
Fig. 28.1  This 52-year-old male presented with rapid onset of ataxia and then coma. Magnetic resonance imaging (MRI) in diffusion-weighted
(A, B, with corresponding apparent diffusion coefficient [ADC] maps, D, E) and fluid-attenuated inversion recovery (FLAIR) (C) sequences. MR
angiography (F) and conventional digital substracted angiography (G). Areas of increased signal on diffusion imaging in the pons, cerebellum, and
thalami, with a decreased ADC corresponding to an acute ischemic stroke are seen. The relatively normal FLAIR sequence indicates that this stroke
is recent (<3-4 hours). MR angiography and digital subtraction angiography (DSA) show occlusion of the basilar artery.

this technique can be applied in a quantitative fashion to

Table 28.1  Magnetic Resonance Imaging
Findings in Acute Stroke
Hyperintensity on diffusion-weighted imaging
Little or no signal change on T2 or fluid-attenuated inversion
recovery (FLAIR) sequences
Hypointensity on apparent diffusion coefficient
Hypointense artery sign of acute intravascular thrombus on
gradient-recalled echo sequences
Absent arterial flow void, i.e., vessel occlusion, on
T2-weighted or FLAIR sequences
Hyperintense vessel sign, which suggests slow or collateral
flow, on FLAIR sequences
Arterial occlusion on magnetic resonance angiography may be
visible
Reduced or absent contrast on dynamic perfusion-weighted
imaging source images
Reduced or absent perfusion on perfusion-weighted imaging
perfusion parameter maps
Modified from Merino JG, Warach S. Imaging of acute stroke. Nat Rev Neurol
2010;6(10):560–71. Epub 2010, Sep 14.
patients in the ICU.
Intracranial Hematoma
CT scan is the imaging modality of choice to diagnose an
acute intracranial hematoma. However, MRI can be per-
formed as a diagnostic tool to determine the etiology of the
intracranial hematoma30 and evaluate the extent of surround-
ing edema or as a prognostic tool to determine the extent of
brain injury (see later section on The Prognostic Value of
MRI). Hemorrhagic lesions are best detected with T2*-
weighted sequences, on which they appear as areas of decreased
signal. This sequence also may show associated small hypoin-
tensities scattered in the brain (or “blebs”), resulting from
previous microvascular injuries. FLAIR sequences can help
define the extent of other vascular risk factor–related brain
lesions (e.g., lacunar infarcts and leukoaraiosis).

Anoxic Brain Injury

extraction fraction (OEF) even when CBF is reduced. OEF
is best evaluated using PET, but BOLD provides an indirect
marker of neuronal activity and so can provide qualitative
results that relate CBF and CMRO2 in patients with chronic
cerebrovascular disease.29 More research is required before
Anoxic brain injury, for example cardiac arrest, is common.
Lesions can involve the pallidum, the cerebellum, the thala-
mus, the striatum, and the cortex (laminar necrosis), more
specifically the occipital and perirolandic areas (Figs. 28.2 and
28.3).31,32 The brainstem usually is spared. Lesions appear as
 Section IV—Radiology 285

A B C
Fig. 28.2  Intracranial hemorrhage complicated by severe cardiomyopathy; 3 weeks later the patient remains in coma. Magnetic resonance
imaging (MRI) in T1 (A), fluid-attenuated inversion recovery (FLAIR) (B), and diffusion-weighted (C) sequences. There is a diffuse hyperintensity of
the cortex on T1 and FLAIR sequences that suggest laminar necrosis. The atrophy and increased signal of the striatum seen on diffusion imaging also
are associated with the prolonged low cardiac output.

A B
1494 Units RM 1805 Units RM
1400
Control
1200
1500 NAA

1000
NAA Cho
800 1000

600 Cr

400
500
200

0
0
–200

–490 –400
4.30 3.71 3.09 2.48 1.87 1.25 0.64 0.39 ppm 4.30 3.71 3.09 2.48 1.87 1.25 0.64 0.39 ppm
C D
Fig. 28.3  A 42-year-old male with altered consciousness 1 week after cardiac arrest. Magnetic resonance imaging (MRI) fluid-attenuated
inversion recovery (FLAIR) sequence (A, B) and MR spectroscopy (box in A) on the thalamus (C) and of a control subject (D). There are lesions of the
lenticular nuclei and occipital and frontal motor (arrows) cortices on the FLAIR sequence. N-acetyl aspartate (NAA) in the thalamus is decreased
compared with the control subject. These findings suggest anoxic ischemic injury. Cho, Choline; Cr, creatine.
286 Section IV—Radiology
hyperintense first on FLAIR and diffusion imaging (usually a
decreased ADC), then on T1 sequences.33 The time course of
these changes is still poorly known; they can appear within the
first 24 hours or be delayed for as much as 1 week. The pres-
ence of hyperintensity in the cortex (particularly the motor
cortex) and striatum suggests poor neurologic prognosis.34
Diffusion MRI also may help predict long-term outcome after
cardiac arrest.35 This can be particularly valuable when induced
hypothermia is used.36 In the most severe forms, brain atrophy
is observed in the following weeks.37,38 MR spectroscopy, if
performed early, may show the presence of lactate.39,40 Later, a
decrease in NAA can be observed (see Fig. 28.3). Its prognostic
value has been well established in children, especially in cases
of near drowning and perinatal anoxia.41,42 Susceptibility-
weighted imaging also may be used to help predict outcome
after neonatal hypoxic-ischemic injuries.43
Subarachnoid Hemorrhage
CT is considered the imaging modality of choice to detect
acute SAH. MRI may be useful, depending on what sequences
are performed to diagnose SAH if CT is negative and in the
subacute phase. The presence of blood in the subarachnoid
spaces gives a hyperintensity on the FLAIR sequence.44 False
negatives may occur and a normal examination does not rule
out the diagnosis, especially if the MRI is performed several
days after the onset of symptoms45 because of CSF washing of
blood. In addition the diagnostic accuracy of FLAIR may be
reduced when the SAH is in compressed cisterns.46 Higher-
field magnets (3 tesla and more) tend to have a higher rate of
both false positives and negatives. MRA or MRI can be used
to demonstrate the causative aneurysm and can help evaluate
ischemic complications secondary to vasospasm. The results
of MRA are most accurate when there is severe arterial nar-
rowing but less reliable for moderate or mild vasospasm.47 In
addition, the dynamics of blood flow are altered when there
are multiple areas of vessel narrowing; this compromises the
accuracy of time-of-flight sequences. Finally, admission DWI
imaging may help guide management decisions in poor-grade
SAH patients; those with no or only spotty DWI lesions tend
to have a good outcome.48
Infectious Brain Diseases
MRI can be extremely useful in some infectious brain
diseases.
 The lesions of herpetic meningoencephalitis predominate
in the temporal lobes, insula, and basi-frontal areas. Con-
trast-enhanced, ischemic or hemorrhagic foci may be
present.
 Bacterial meningitis usually does not require an MRI
examination. This examination is sometimes performed in
patients who develop focal signs or altered consciousness,
in search for an abscess, ventriculitis, or vasculitis.
 Brain abscesses appear as cystic, contrast-enhancing
lesions. The capsule shows hypointensity on T2 or FLAIR
and slight hypersignal on T1-weighted sequences, the
content of the abscess has decreased ADC, and amino acids
can be detected on MR spectroscopy49,50 (Fig. 28.4).
 Both edematous and ischemic lesions of the white matter
may be seen in septic shock–induced brain dysfunction.
However, the potential prognostic value of MRI in this
context remains to be fully elucidated.
Posterior Reversible
Leukoencephalopathy Syndrome
Posterior reversible leukoencephalopathy syndrome (PRESS)
is a complication of uncontrolled hypertension and of some
treatments such as immunosuppressors or some chemothera-
peutic agents (e.g., intrathecal methotrexate). Its diagnosis is
easily made with MRI, which shows areas of subcortical edema
that predominate in the posterior areas. The lesions also can
involve the frontal white matter or posterior fossa. Contrast
enhancement is uncommon. Diffusion imaging should be
performed to evaluate PRESS: two different patterns can be
observed51,52:
 Areas of increased ADC correspond to vasogenic edema.
These lesions usually are reversible and associated with
good prognosis.
 Areas of decreased ADC correspond to cytotoxic edema;
these lesions generally are not reversible and suggest a
more reserved prognosis.
Traumatic Brain Injury
CT is the initial imaging study for traumatic brain injury
(TBI) because it permits rapid and accurate diagnosis of
intracranial hematoma and mass effect on which to base
management decisions. MRI is useful for follow-up evalua-
tion especially of diffuse axonal injury (DAI), examination
of brainstem pathology, and to aid in decisions about
prognosis.53-55 Newer MRI sequences (e.g., gradient-recalled
echo, susceptibility-weighted imaging (SWI), and DWI) have
greater sensitivity for diffuse brain injury, whereas diffusion
tensor imaging provides information about white matter
injury.53,55 Multiple mechanisms lead to cerebral injury in
TBI; hence the imaging pattern can be varied and heteroge-
neous. This chapter focuses on pathology observed in the
brain and what is most relevant to follow up during ICU
care. The reader is referred to recent reviews for a full dis-
cussion of imaging in TBI including of the skull and cranio-
facial structures and for the common data elements on TBI
imaging.53-56
 Cerebral contusions can be edematous or hemorrhagic.
They appear as areas of hyperintensity on the FLAIR
sequence, with associated hypointensity on the T2*
sequence when hematomas are present. Their most
frequent locations are the basi-frontal areas and the
temporal lobes.
 DAI (Fig. 28.5). This injury results usually from decelera-
tion and is associated with shearing of the white matter
fibers. Three different kinds of DAI can be observed:
 Hemorrhagic DAI is the most easily detected: the white
matter lesions are associated with lesions of the small
arteries, causing very small areas of bleeding, which
give hypointense dots (or blebs) on the T2* sequence.
They have been associated with persistent cognitive
impairment in patients with an otherwise normal
MRI.57 SWI is a very sensitive method to identify
microbleeds, which are a marker of traumatic axonal
injury.58-60
 Section IV—Radiology 287

A B C D
Aa

Aa

4 3 2 1 0 4 3 2 1 0
E F
Fig. 28.4  This 45-year-old patient with a history of intravenous drug abuse and alcoholic cirrhosis presented with fever and a rapidly
altering consciousness. Magnetic resonance imaging (MRI) images: T1 with gadolinium (A), fluid-attenuated inversion recovery (FLAIR) with
gadolinium (B), diffusion-weighted sequences (C), and apparent diffusion coefficient (ADC) map (D). MR spectroscopy at short (E) and long (F) echo
times. The contrast-enhancing cystic lesion is high signal on diffusion imaging with a decreased ADC, and the presence of amino acids (Aa) on MR
spectroscopy are typical of an intracranial abscess. The associated meningitis and ventriculitis are diagnosed by the presence of increased signal of
the sulci and the frontal horns of the lateral ventricles on the FLAIR sequence.

Ischemic DAI occurs when the lesions of the small

 where the compression and displacement of this structure
arteries lead to vascular occlusion and ischemia. They
are best detected with the diffusion sequence that
shows punctiform areas of decreased ADC.
 DAI without vascular lesions is the most difficult to
detect. These lesions do not give any sign on MR
sequences, except atrophy of relatively late onset. DTI
is probably the best method to detect them; interrup-
tion of the white matter tract leads to a decrease
in the fractional anisotropy of the area, or can be
detected by fiber tracking. However, although some
studies have shown marked difference in fractional
anisotropy values between TBI subjects with a good
and poor prognosis, the role of DTI to determine
prognosis of individual patients remains to be fully
defined.59
 Acute vascular lesions can arise from direct compression,
commonly the posterior cerebral artery between the tem-
poral lobe and the midbrain (e.g., during temporal hernia-
tion). Carotid or vertebral dissection can lead to secondary
ischemic strokes that may occur several days after the initial
injury.61 These mechanisms lead to “classic” strokes, easily
demonstrated with diffusion imaging. Other processes can
lead to a more insidious form of ischemia. Subdural and
extradural hematomas stretch the vascular structures, and
can lead to a devascularization of brain parenchyma. These
lesions may be difficult to detect, except in the midbrain,
give microhemorrhagic infarctions detectable on the T2*
sequence (Duret’s hemorrhages).62
 Intracranial hypertension (IH) is common in TBI patients
admitted to the NCCU. The underlying mechanism can
include disruption of the blood-brain barrier, space-
occupying lesions (hematomas, contusions), and cytotoxic
edema from ischemic lesions. Although indirect signs of
IH such as reduced ventricles and sulci can be detected on
CT or MRI, the severity of IH is difficult to quantify on
imaging. The consequences of IH can be evaluated at the
subacute phase with DTI; the global injury to the brain
parenchyma leads to a diffuse decrease of supratentorial
FA (see Fig. 28.5).
 Extracranial complications also may adversely affect the
brain parenchyma. The most common is hypotension or
hypovolemia, which is associated with lesions similar to
those observed in anoxia or ischemia.
 Although patients with mild TBI and concussion rarely
require care in the NCCU, there is continued debate on
the diagnosis and management and in particular whether
the extent of tissue-based impairments or other stressors
causes disability. Conventional MRI may be insensitive to
microstructural changes that may be present. Advanced
imaging techniques, including SWI that is very sensitive to
microbleeds, DTI using different measures (i.e., FA, mean
diffusivity [MD], and axial diffusivity [AD] to examine
288 Section IV—Radiology

A B C

Control

D E
Fig. 28.5  Prolonged intracranial hypertension in a 17-year-old female after head injury. The patient remained unconscious 1 month later.
Magnetic resonance imaging (MRI) images: fluid-attenuated inversion recovery (FLAIR) (A), T2* (B), diffusion-weighted (C), and diffusion tensor
(D, fractional anisotropy map) sequences in the patient and an fractional anisotropy (FA) map at the same level of a control subject (E). There are
patchy areas of hyperintensity on FLAIR sequence, with areas of decreased signal on T2* and increased signal on diffusion imaging that correspond
to ischemic and hemorrhagic diffuse axonal injury. The diffuse alteration of the white matter tracks on the fractional anisotropy map results from
both diffuse axonal injuries and prolonged intracranial hypertension.

white matter), MR spectroscopy, and resting state fMRI to
evaluate the default mode network (DMN), however, can
help in evaluation of mild TBI and provide objective bio-
markers that underlie long-term cognitive deficits.59,63-69
The DTI maps can be normalized to the International
Consortium for Brain Mapping atlas for atlas-based analy-
sis and can be assessed using global methods (i.e., white
matter histogram or voxel-based approaches) or a regional
approach (i.e., tractography).70
The Prognostic Value of MRI
Advanced MRI techniques can be used to help determine the
prognosis of the severely brain-injured patient, but this is still
under investigation. Anoxic-ischemic disease and TBI have
been best studied.71-73 In a prospective cohort of severe TBI, it
was observed in 106 patients that the use of quantitative DTI
in the acute setting enhanced the sensitivity and specificity of
12-month functional outcome prediction compared with the
International Mission for Prognosis and Analysis of Clinical
Trials in TBI (IMPACT) score alone74 in preparation. Given
the complexity of this examination, there are less available
data in some pathologies such as intracranial hemorrhage.
Establishing the prognosis of TBI patients is relatively complex,
and involves the determination of: (1) the reversibility of
lesions, (2) their bilaterality and symmetry, and (3) the
involvement of critical functional brain areas.
Reversibility of Traumatic
Brain Injury Lesions
On conventional MRI images, the evolution and reversibility
of most lesions is well described. Vasogenic edema, whether
local (e.g., in a contusion or surrounding a hematoma) or
diffuse generally resolves over time. The natural evolution of
the hematomas is spontaneous regression over several weeks,
with the ultimate persistence of a “scar” that is hyperintense
on FLAIR and hypointense on T2*. The reversibility of these
lesions means that although they may appear extensive on
the initial MRI, they may not always mean poor prognosis,
especially if they do not involve major functional areas. By
contrast, other lesions may be considered “irreversible”
(e.g., ischemic lesions, and most DAI lesions). In these “irre-
versible” lesions there is a slow progression toward brain
atrophy, increasing sometimes even months after the initial
event.75 Most structural brain imaging studies after TBI that
incorporate volumetric measures and a longitudinal design
demonstrate some degree of brain atrophy that progresses
over time.76 This pattern of atrophy often is regionally

selective and associated with the severity of the injury and
outcome.77
The reversibility of the alterations seen on DTI and MRS
remains to be fully elucidated. An NAA reduction on MRS
maybe reversible because it can be associated with transient
neuronal dysfunction rather than death. Whether there is a
threshold value in NAA that differentiates between reversible
and irreversible neuronal damages has yet to be established.
Similar limits exist with diffusion tensor imaging; whereas
major reductions in FA correspond to white matter destruc-
tion, late, localized, re-increases in FA values are described.
This FA increase may result from axonal regrowth or
remyelination.78
Involvement of Major Brain Structures
Evaluation of the integrity or impairment of major brain
structures is a critical step in the prognostic evaluation of TBI
patients. From a practical viewpoint these can be divided into
areas critical for regaining consciousness, and structures
involved in high-order brain functions. Alternatively, lesions
may be considered according to depth—hemisphere, central,
and brainstem injury. In severe TBI, bilateral brainstem lesions
are strongly predictive of poor outcome, but the relationship
of lesion location to outcome in moderate TBI is less clear.79
The mechanisms of consciousness and their imaging cor-
relates remain incompletely understood. Dehaene et al. and
Parvizi and Damasio among others, have developed and
studied the concept of the “global workspace,” an interconnec-
tion of brainstem nuclei, white matter tracts, deep brain struc-
tures, and cortical areas, the integrity of which is critical to
consciousness. The following network is considered to be
important for consciousness: the substantia reticulata, located
in the posterior pons and connected to the posterointernal
thalamic nuclei through the mesencephalon.80,81 The thalamic
nuclei in turn are associated with the hypothalamus, the basal
forebrain, and various cortical areas including (but not limited
to) the cingulum (see also Chapter 10 for a discussion on
consciousness). Lesions in these areas are associated with a
delayed recovery of consciousness, a vegetative state (VS), or
minimally conscious state (MCS). Consistent with this, studies
have demonstrated that functional connectivity within the
well-characterized default mode network is associated with
the level of arousal in subjects with severe brain injury.82,83 In
addition, arterial spin labeling (ASL) techniques to examine
CBF in different brain regions may provide further insight
into functional recovery84 in the subacute stage. What remains
to be determined is exactly how the extent of lesions in these
areas of the brain relates to poor prognosis.
 Hemorrhagic lesions of the posterior pons, bilateral tha-
lamic lesions (either ischemic or hemorrhagic) that are
visible on conventional MR sequences, appear to be associ-
ated with poor prognosis.
 Bilateral lesions of the mesencephalon often are associated
with VS or MCS. These lesions in TBI often result from a
shearing mechanism that occurs during deceleration
because of the different mass between the telencephalon
and posterior fossa. They may be difficult to detect with
conventional MRI sequences, but are easily visible on dif-
fusion tensor images. However, quantitative assessment of
mesencephalic injury with DTI still is a challenge.
Section IV—Radiology 289
 The prognostic value of unilateral lesions of the thalamus,
pons or mesencephalon, and of metabolic alterations seen
on MR spectroscopy without associated lesions on conven-
tional sequences remains to be fully elucidated. However,
a significant decrease in NAA in these areas usually is asso-
ciated with a poor prognosis.85
Lesions of critical areas associated with high-order functions
do not per se impair consciousness but can lead to significant
deficits that must be taken into account when decisions about
prognosis and care are made. The deleterious effects of lesions
in cortical areas are relatively easy to predict. However, the
prognostic significance of alterations in the white matter
tracts that originate from or connect with these areas, is less
studied and remains a challenge. DTI, using fractional anisot-
ropy maps and fiber tracking algorithms, will probably be the
method of choice, but the application of this postprocessing
method to an individual patient’s diagnosis and to outcome
prediction is still in its infancy although promising in its utility
including for patients who remain unconsciousness several
days after injury.72,86-89 fMRI studies of functional connectivity
in the default mode network in the chronic phase after TBI
also may help in outcome prediction in the chronic phase after
TBI.90 The prognostic value of lesions in some cortical or
subcortical areas is as follows:
 Lesions of the primary motor cortex and pyramidal
tract are associated with hemiplegia or hemiparesia, but
extensive destruction of the motor cortex is uncommon.
The pyramidal tract is more commonly involved, by
hematomas of the centrum semiovale or internal capsule.
There can be delayed recovery in some patients of this
function.
 Lesions of the hippocampi and medial temporal lobe are
associated with short-term memory difficulties. Distinct
fMRI patterns during working memory also may provide
insights into recovery in the chronic stage.91
 Extensive lesions of the frontal lobes and cingular cortex
can cause behavioral problems such as akinetic mutism,
which may limit full recovery and integration into a
normal social life.
 Unilateral lesions, even extensive, may be relatively well
tolerated with late recovery, whereas this is not true if the
lesions are bilateral. Lesions of the language areas (Broca’s
and Wernicke’s) could carry a less favorable prognosis
related to asymmetry of this brain function.
Conclusion
MRI in critically ill NCCU patients can be a challenge, for
example, patient transport to the MRI facility or the pres-
ence of monitors that are not MRI compatible and need to
be replaced. However, MRI can provide very useful diagnos-
tic and prognostic information for patient management. The
imaging protocol should routinely include both conven-
tional (T1 and T2 weighted) and advanced (DTI, MR spec-
troscopy) imaging techniques. Imaging technologies and
techniques continue to evolve and in the future, multivariate
analysis of the information including that derived from
other advanced imaging techniques such as volumetric MRI
analysis, SWI, DTI, magnetization transfer imaging (MTI),
ASL, fMRI (including resting state and connectivity MRI),
290 Section IV—Radiology
and hyperpolarization scanning may help objectively and
reliably predict the outcome of the patient.
References
1. Looby S, Flanders A. Spine trauma. Radiol Clin North Am 2011;49(1):129–
63. Review.
2. Mughal MM, Khan MK, DeMarco JK, et al. Symptomatic and asymptomatic
carotid artery plaque. Expert Rev Cardiovasc Ther 2011;9(10):1315–30.
3. Chiribiri A, Ishida M, Nagel E, et al. Coronary imaging with cardiovascular
magnetic resonance: current state of the art. Prog Cardiovasc Dis
2011;54(3):240–52.
4. Jiji RS, Kramer CM. Cardiovascular magnetic resonance: applications in
daily practice. Cardiol Rev 2011;19(5):246–54. Review.
5. Elias AE, Carlos RC, Smith EA, et al. MR spectroscopy using normalized and
non-normalized metabolite ratios for differentiating recurrent brain tumor
from radiation injury. Acad Radiol 2011;18(9):1101–8.
6. Muñoz Maniega S, Cvoro V, Chappell FM, et al. Changes in NAA and lactate
following ischemic stroke: a serial MR spectroscopic imaging study.
Neurology 2008;71(24):1993–9.
7. Owen AM, Coleman MR, Menon DK, et al. Residual auditory function in
persistent vegetative state: a combined PET and fMRI study. Neuropsychol
Rehabil 2005;15(3–4):290–306.
8. Siva N. Patient in a coma plays tennis. Lancet Neurol 2006;5(11):906.
9. Schiff ND. fMRI reveals large-scale network activation in minimally
conscious patients. Neurology 2005;64(3):514–23.
10. Salvador R, Suckling J, Coleman MR, et al. Neurophysiological architecture
of functional magnetic resonance images of human brain. Cereb Cortex
2005;15(9):1332–42.
11. Owen AM, Coleman MR, Boly M, et al. Detecting awareness in the vegetative
state. Science 2006;313(5792):1402.
12. Vincent JL, Patel GH, Fox MD, et al. Intrinsic functional architecture in the
anaesthetized monkey brain. Nature 2007;447(7140):83–6.
13. Hwang DY, Silva GS, Furie KL, et al. Comparative sensitivity of computed
tomography vs. magnetic resonance imaging for detecting acute posterior
fossa infarct. J Emerg Med 2012;Feb 2. Epub ahead of print.
14. Merino JG, Warach S. Imaging of acute stroke. Nat Rev Neurol
2010;6(10):560–71. Epub 2010, Sep 14.
15. Schlaug G, Siewert B, Benfield A, et al. Time course of the apparent diffusion
coefficient (ADC) abnormality in human stroke. Neurology
1997;49(1):113–9.
16. Oppenheim C, Grandin C, Samson Y, et al. Is there an apparent diffusion
coefficient threshold in predicting tissue viability in hyperacute stroke?
Stroke 2001;32(11):2486–91.
17. Donahue MJ, Strother MK, Hendrikse J. Novel MRI approaches for assessing
cerebral hemodynamics in ischemic cerebrovascular disease. Stroke
2012;43(3):903–15. Epub 2012, Feb 16.
18. Nour M, Liebeskind DS. Brain imaging in stroke: insight beyond diagnosis.
Neurotherapeutics 2011;8(3):330–9.
19. Yoo AJ, González RG. Clinical applications of diffusion MR imaging for
acute ischemic stroke. Neuroimaging Clin N Am 2011;21(1):51–69, vii.
20. Copen WA, Schaefer PW, Wu O. MR perfusion imaging in acute ischemic
stroke. Neuroimaging Clin N Am 2011;21(2):259–83, x.
21. Thomalla G, Cheng B, Ebinger M, et al. DWI-FLAIR mismatch for the
identification of patients with acute ischaemic stroke within 4·5 h of
symptom onset (PRE-FLAIR): a multicentre observational study. Lancet
Neurol 2011;10(11):978–86. Epub 2011, Oct 4.
22. Heiss WD. The ischemic penumbra: correlates in imaging and implications
for treatment of ischemic stroke. The Johann Jacob Wepfer Award 2011.
Cerebrovasc Dis 2011;32(4):307–20. Epub 2011, Sep 15.
23. Amin-Hanjani S, Charbel FT, Malisch T, et al. Use of quantitative magnetic
resonance angiography to stratify stroke risk in symptomatic vertebrobasilar
disease. Stroke 2005;36:1140–5.
24. Hernandez DA, Bokkers RP, Mirasol RV, et al. Pseudocontinuous arterial
spin labeling quantifies relative cerebral blood flow in acute stroke. Stroke
2012;43(3):753–8. Epub 2012, Feb 16.
25. van Osch MJ, Teeuwisse WM, van Walderveen MA, et al. Can arterial spin
labeling detect white matter perfusion signal? Magn Reson Med
2009;62:165–73.
26. Wong EC. Vessel-encoded arterial spin-labeling using pseudocontinuous
tagging. Magn Reson Med 2007;58:1086–91.
27. Petersen ET, Lim T, Golay X. Model-free arterial spin labeling quantification
approach for perfusion MRI. Magn Reson Med 2006;55:219–32.
28. Hua J, Qin Q, Donahue MJ, et al. Inflow-based vascular-space occupancy
(IVASO) MRI. Magn Reson Med 2011;66:40–56.
29. Heyn C, Poublanc J, Crawley A, et al. Quantification of cerebrovascular
reactivity by blood oxygen level-dependent MR imaging and correlation
with conventional angiography in patients with Moyamoya disease. Am J
Neuroradiol 2010;31:862–7.
30. Kamel H, Navi BB, Hemphill JC 3rd. A rule to identify patients who require
magnetic resonance imaging after intracerebral hemorrhage. Neurocrit Care
2011;Jul 15. Epub ahead of print.
31. Wijdicks EF, Campeau NG, Miller GM. MR imaging in comatose survivors
of cardiac resuscitation. AJNR Am J Neuroradiol 2001;22(8):1561–5.
32. Tha KK, Terae S, Yamamoto T, et al. Early detection of global cerebral
anoxia: improved accuracy by high-b-value diffusion-weighted imaging with
long echo time. AJNR Am J Neuroradiol 2005;26(6):1487–97.
33. Arbelaez A, Castillo M, Mukherji SK. Diffusion-weighted MR imaging of
global cerebral anoxia. AJNR Am J Neuroradiol 1999;20(6):999–1007.
34. Barrett KM, Freeman WD, Weindling SM, et al. Brain injury after
cardiopulmonary arrest and its assessment with diffusion-weighted magnetic
resonance imaging. Mayo Clin Proc 2007;82(7):828–35.
35. Wijman CA, Mlynash M, Caulfield AF, et al. Prognostic value of brain
diffusion-weighted imaging after cardiac arrest. Ann Neurol
2009;65(4):394–402.
36. Oddo M, Rossetti AO. Predicting neurological outcome after cardiac arrest.
Curr Opin Crit Care 2011;17(3):254–9.
37. Allen JS, Tranel D, Bruss J, et al. Correlations between regional brain
volumes and memory performance in anoxia. J Clin Exper Neuropsychol
2006;28(4):457–76.
38. Ammermann H, Kassubek J, Lotze M, et al. MRI brain lesion patterns in
patients in anoxia-induced vegetative state. J Neurol Sci
2007;260(1-2):65–70.
39. Geraghty MC, Torbey MT. Neuroimaging and serologic markers of
neurologic injury after cardiac arrest. Neurol Clin 2006;24(1):107–21, vii.
40. Berek K, Lechleitner P, Luef G, et al. Early determination of neurological
outcome after prehospital cardiopulmonary resuscitation. Stroke
1995;26(4):543–9.
41. Kreis R, Arcinue E, Ernst T, et al. Hypoxic encephalopathy after near-
drowning studied by quantitative 1H-magnetic resonance spectroscopy.
J Clin Invest 1996;97(5):1142–54.
42. Zarifi MK, Astrakas LG, Poussaint TY, et al. Prediction of adverse outcome
with cerebral lactate level and apparent diffusion coefficient in infants with
perinatal asphyxia. Radiology 2002;225(3):859–70.
43. Kitamura G, Kido D, Wycliffe N, et al. Hypoxic-ischemic injury: utility of
susceptibility-weighted imaging. Pediatr Neurol 2011;45(4):220–4.
44. Noguchi K, Seto H, Kamisaki Y, et al. Comparison of fluid-attenuated
inversion-recovery MR imaging with CT in a simulated model of acute
subarachnoid hemorrhage. AJNR Am J Neuroradiol 2000;21(5):923–7.
45. Mohamed M, Heasly DC, Yagmurlu B, et al. Fluid-attenuated inversion
recovery MR imaging and subarachnoid hemorrhage: not a panacea. AJNR
Am J Neuroradiol 2004;25(4):545–50.
46. Shimoda M, Hoshikawa K, Shiramizu H, et al. Problems with diagnosis by
fluid-attenuated inversion recovery magnetic resonance imaging in patients
with acute aneurysmal subarachnoid hemorrhage. Neurol Med Chir (Tokyo)
2010;50(7):530–7.
47. Hattingen E, Blasel S, Dumesnil R, et al. MR angiography in patients with
subarachnoid hemorrhage: adequate to evaluate vasospasm-induced vascular
narrowing? Neurosurg Rev 2010;33(4):431–9. Epub 2010, Jun 8.
48. Sato K, Shimizu H, Fujimura M, et al. Acute-stage diffusion-weighted
magnetic resonance imaging for predicting outcome of poor-grade
aneurysmal subarachnoid hemorrhage. J Cereb Blood Flow Metab
2010;30(6):1110–20. Epub 2010, Jan 6.
49. Grand S, Passaro G, Ziegler A, et al. Necrotic tumor versus brain abscess:
importance of amino acids detected at 1H MR spectroscopy-initial results.
Radiology 1999;21(3):785–93.
50. Luthra G, Parihar A, Nath K, et al. Comparative evaluation of fungal,
tubercular, and pyogenic brain abscesses with conventional and diffusion
MR imaging and proton MR spectroscopy. AJNR Am J Neuroradiol
2007;28(7):1332–8.
A complete list of references for this chapter can be found online at
www.expertconsult.com.

References
1. Looby S, Flanders A. Spine trauma. Radiol Clin North Am 2011;49(1):129–
63. Review.
2. Mughal MM, Khan MK, DeMarco JK, et al. Symptomatic and asymptomatic
carotid artery plaque. Expert Rev Cardiovasc Ther 2011;9(10):1315–30.
3. Chiribiri A, Ishida M, Nagel E, et al. Coronary imaging with cardiovascular
magnetic resonance: current state of the art. Prog Cardiovasc Dis
2011;54(3):240–52.
4. Jiji RS, Kramer CM. Cardiovascular magnetic resonance: applications in
daily practice. Cardiol Rev 2011;19(5):246–54. Review.
5. Elias AE, Carlos RC, Smith EA, et al. MR spectroscopy using normalized and
non-normalized metabolite ratios for differentiating recurrent brain tumor
from radiation injury. Acad Radiol 2011;18(9):1101–8.
6. Muñoz Maniega S, Cvoro V, Chappell FM, et al. Changes in NAA and lactate
following ischemic stroke: a serial MR spectroscopic imaging study.
Neurology 2008;71(24):1993–9.
7. Owen AM, Coleman MR, Menon DK, et al. Residual auditory function in
persistent vegetative state: a combined PET and fMRI study. Neuropsychol
Rehabil 2005;15(3–4):290–306.
8. Siva N. Patient in a coma plays tennis. Lancet Neurol 2006;5(11):906.
9. Schiff ND. fMRI reveals large-scale network activation in minimally
conscious patients. Neurology 2005;64(3):514–23.
10. Salvador R, Suckling J, Coleman MR, et al. Neurophysiological architecture
of functional magnetic resonance images of human brain. Cereb Cortex
2005;15(9):1332–42.
11. Owen AM, Coleman MR, Boly M, et al. Detecting awareness in the vegetative
state. Science 2006;313(5792):1402.
12. Vincent JL, Patel GH, Fox MD, et al. Intrinsic functional architecture in the
anaesthetized monkey brain. Nature 2007;447(7140):83–6.
13. Hwang DY, Silva GS, Furie KL, et al. Comparative sensitivity of computed
tomography vs. magnetic resonance imaging for detecting acute posterior
fossa infarct. J Emerg Med 2012;Feb 2. Epub ahead of print.
14. Merino JG, Warach S. Imaging of acute stroke. Nat Rev Neurol
2010;6(10):560–71. Epub 2010, Sep 14.
15. Schlaug G, Siewert B, Benfield A, et al. Time course of the apparent diffusion
coefficient (ADC) abnormality in human stroke. Neurology 1997;49(1):113–9.
16. Oppenheim C, Grandin C, Samson Y, et al. Is there an apparent diffusion
coefficient threshold in predicting tissue viability in hyperacute stroke?
Stroke 2001;32(11):2486–91.
17. Donahue MJ, Strother MK, Hendrikse J. Novel MRI approaches for assessing
cerebral hemodynamics in ischemic cerebrovascular disease. Stroke
2012;43(3):903–15. Epub 2012, Feb 16.
18. Nour M, Liebeskind DS. Brain imaging in stroke: insight beyond diagnosis.
Neurotherapeutics 2011;8(3):330–9.
19. Yoo AJ, González RG. Clinical applications of diffusion MR imaging for
acute ischemic stroke. Neuroimaging Clin N Am 2011;21(1):51–69, vii.
20. Copen WA, Schaefer PW, Wu O. MR perfusion imaging in acute ischemic
stroke. Neuroimaging Clin N Am 2011;21(2):259–83, x.
21. Thomalla G, Cheng B, Ebinger M, et al. DWI-FLAIR mismatch for the
identification of patients with acute ischaemic stroke within 4·5 h of
symptom onset (PRE-FLAIR): a multicentre observational study. Lancet
Neurol 2011;10(11):978–86. Epub 2011, Oct 4.
22. Heiss WD. The ischemic penumbra: correlates in imaging and implications
for treatment of ischemic stroke. The Johann Jacob Wepfer Award 2011.
Cerebrovasc Dis 2011;32(4):307–20. Epub 2011, Sep 15.
23. Amin-Hanjani S, Charbel FT, Malisch T, et al. Use of quantitative magnetic
resonance angiography to stratify stroke risk in symptomatic vertebrobasilar
disease. Stroke 2005;36:1140–5.
24. Hernandez DA, Bokkers RP, Mirasol RV, et al. Pseudocontinuous arterial
spin labeling quantifies relative cerebral blood flow in acute stroke. Stroke
2012;43(3):753–8. Epub 2012, Feb 16.
25. van Osch MJ, Teeuwisse WM, van Walderveen MA, et al. Can arterial spin
labeling detect white matter perfusion signal? Magn Reson Med
2009;62:165–73.
26. Wong EC. Vessel-encoded arterial spin-labeling using pseudocontinuous
tagging. Magn Reson Med 2007;58:1086–91.
27. Petersen ET, Lim T, Golay X. Model-free arterial spin labeling quantification
approach for perfusion MRI. Magn Reson Med 2006;55:219–32.
28. Hua J, Qin Q, Donahue MJ, et al. Inflow-based vascular-space occupancy
(IVASO) MRI. Magn Reson Med 2011;66:40–56.
29. Heyn C, Poublanc J, Crawley A, et al. Quantification of cerebrovascular
reactivity by blood oxygen level-dependent MR imaging and correlation
with conventional angiography in patients with Moyamoya disease. Am J
Neuroradiol 2010;31:862–7.
Section IV—Radiology 290.e1
30. Kamel H, Navi BB, Hemphill JC 3rd. A rule to identify patients who require
magnetic resonance imaging after intracerebral hemorrhage. Neurocrit Care
2011;Jul 15. Epub ahead of print.
31. Wijdicks EF, Campeau NG, Miller GM. MR imaging in comatose survivors
of cardiac resuscitation. AJNR Am J Neuroradiol 2001;22(8):1561–5.
32. Tha KK, Terae S, Yamamoto T, et al. Early detection of global cerebral
anoxia: improved accuracy by high-b-value diffusion-weighted imaging with
long echo time. AJNR Am J Neuroradiol 2005;26(6):1487–97.
33. Arbelaez A, Castillo M, Mukherji SK. Diffusion-weighted MR imaging of
global cerebral anoxia. AJNR Am J Neuroradiol 1999;20(6):999–1007.
34. Barrett KM, Freeman WD, Weindling SM, et al. Brain injury after
cardiopulmonary arrest and its assessment with diffusion-weighted magnetic
resonance imaging. Mayo Clin Proc 2007;82(7):828–35.
35. Wijman CA, Mlynash M, Caulfield AF, et al. Prognostic value of brain
diffusion-weighted imaging after cardiac arrest. Ann Neurol
2009;65(4):394–402.
36. Oddo M, Rossetti AO. Predicting neurological outcome after cardiac arrest.
Curr Opin Crit Care 2011;17(3):254–9.
37. Allen JS, Tranel D, Bruss J, et al. Correlations between regional brain
volumes and memory performance in anoxia. J Clin Exper Neuropsychol
2006;28(4):457–76.
38. Ammermann H, Kassubek J, Lotze M, et al. MRI brain lesion patterns in
patients in anoxia-induced vegetative state. J Neurol Sci
2007;260(1–2):65–70.
39. Geraghty MC, Torbey MT. Neuroimaging and serologic markers of
neurologic injury after cardiac arrest. Neurol Clin 2006;24(1):107–21, vii.
40. Berek K, Lechleitner P, Luef G, et al. Early determination of neurological
outcome after prehospital cardiopulmonary resuscitation. Stroke
1995;26(4):543–9.
41. Kreis R, Arcinue E, Ernst T, et al. Hypoxic encephalopathy after near-
drowning studied by quantitative 1H-magnetic resonance spectroscopy.
J Clin Invest 1996;97(5):1142–54.
42. Zarifi MK, Astrakas LG, Poussaint TY, et al. Prediction of adverse outcome
with cerebral lactate level and apparent diffusion coefficient in infants with
perinatal asphyxia. Radiology 2002;225(3):859–70.
43. Kitamura G, Kido D, Wycliffe N, et al. Hypoxic-ischemic injury: utility of
susceptibility-weighted imaging. Pediatr Neurol 2011;45(4):220–4.
44. Noguchi K, Seto H, Kamisaki Y, et al. Comparison of fluid-attenuated
inversion-recovery MR imaging with CT in a simulated model of acute
subarachnoid hemorrhage. AJNR Am J Neuroradiol 2000;21(5):923–7.
45. Mohamed M, Heasly DC, Yagmurlu B, et al. Fluid-attenuated inversion
recovery MR imaging and subarachnoid hemorrhage: not a panacea. AJNR
Am J Neuroradiol 2004;25(4):545–50.
46. Shimoda M, Hoshikawa K, Shiramizu H, et al. Problems with diagnosis by
fluid-attenuated inversion recovery magnetic resonance imaging in patients
with acute aneurysmal subarachnoid hemorrhage. Neurol Med Chir (Tokyo)
2010;50(7):530–7.
47. Hattingen E, Blasel S, Dumesnil R, et al. MR angiography in patients with
subarachnoid hemorrhage: adequate to evaluate vasospasm-induced vascular
narrowing? Neurosurg Rev 2010;33(4):431–9. Epub 2010, Jun 8.
48. Sato K, Shimizu H, Fujimura M, et al. Acute-stage diffusion-weighted
magnetic resonance imaging for predicting outcome of poor-grade
aneurysmal subarachnoid hemorrhage. J Cereb Blood Flow Metab
2010;30(6):1110–20. Epub 2010, Jan 6.
49. Grand S, Passaro G, Ziegler A, et al. Necrotic tumor versus brain abscess:
importance of amino acids detected at 1H MR spectroscopy-initial results.
Radiology 1999;21(3):785–93.
50. Luthra G, Parihar A, Nath K, et al. Comparative evaluation of fungal,
tubercular, and pyogenic brain abscesses with conventional and diffusion
MR imaging and proton MR spectroscopy. AJNR Am J Neuroradiol
2007;28(7):1332–8.
51. Covarrubias DJ, Luetmer PH, Campeau NG, Posterior reversible
encephalopathy syndrome: prognostic utility of quantitative diffusion-
weighted MR images. AJNR Am J Neuroradiol 2002;23(6):1038–48.
52. Lamy C, Oppenheim C, Méder JF, et al. Neuroimaging in posterior reversible
encephalopathy syndrome. J Am Soc Neuroimaging 2004;14(2):89–96.
53. Provenzale JM. Imaging of traumatic brain injury: a review of the recent
medical literature. AJR Am J Roentgenol 2010;194(1):16–9.
54. Gean AD, Fischbein NJ. Head trauma. Neuroimaging Clin N Am
2010;20(4):527–56.
55. Kubal WS. Updated imaging of traumatic brain injury. Radiol Clin North
Am 2012;50(1):15–41.
56. Haacke EM, Duhaime AC, Gean AD, et al. Common data elements in
radiologic imaging of traumatic brain injury. J Magn Reson Imaging
2010;32(3):516–43.
290.e2 Section IV—Radiology
57. Scheid R, Walther K, Guthke T, et al. Cognitive sequelae of diffuse axonal
injury. Arch Neurol 2006;63(3):418–24.
58. Colbert CA, Holshouser BA, Aaen GS, et al. Value of cerebral
microhemorrhages detected with susceptibility-weighted MR imaging for
prediction of long-term outcome in children with nonaccidental trauma.
Radiology 2010;256(3):898–905.
59. Sharp DJ, Ham TE. Investigating white matter injury after mild traumatic
brain injury. Curr Opin Neurol 2011;24(6):558–63.
60. Beauchamp MH, Ditchfield M, Babl FE, et al. Detecting traumatic brain
lesions in children: CT versus MRI versus susceptibility weighted imaging
(SWI). J Neurotrauma 2011;28(6):915–27. Epub 2011, Jun 9.
61. Tawil I, Stein DM, Mirvis SE, et al. Posttraumatic cerebral infarction:
incidence, outcome, and risk factors. J Trauma 2008;64(4):849–53.
62. Alexander E, Kushner J, Six EG. Brainstem hemorrhages and increased
intracranial pressure: from Duret to computed tomography. Surg Neurol
1982;17(2):107–10.
63. Niogi SN, Mukherjee P. Diffusion tensor imaging of mild traumatic brain
injury. J Head Trauma Rehabil 2010;25(4):241–55.
64. Kasahara K, Hashimoto K, Abo M, et al. Voxel- and atlas-based analysis of
diffusion tensor imaging may reveal focal axonal injuries in mild traumatic
brain injury: comparison with diffuse axonal injury. Magn Reson Imaging
2012;Jan 26. Epub ahead of print.
65. Gonzalez PG, Walker MT. Imaging modalities in mild traumatic brain injury
and sports concussion. PM R 2011;3(10 Suppl 2):S413–24.
66. Henry LC, Tremblay J, Tremblay S, et al. Acute and chronic changes in
diffusivity measures after sports concussion. J Neurotrauma
2011;28(10):2049–59. Epub 2011, Oct 4.
67. Henry LC, Tremblay S, Leclerc S, et al. Metabolic changes in concussed
American football players during the acute and chronic post-injury phases.
BMC Neurol 2011;11:105.
68. Johnson B, Zhang K, Gay M, et al. Alteration of brain default network in
subacute phase of injury in concussed individuals: resting-state fMRI study.
Neuroimage 2012;59(1):511–8. Epub 2011, Aug 7.
69. Mayer AR, Mannell MV, Ling J, et al. Functional connectivity in mild
traumatic brain injury. Hum Brain Mapp 2011;32(11):1825–35.
70. Marquez de la Plata CD, Yang FG, Wang JY, et al. Diffusion tensor imaging
biomarkers for traumatic axonal injury: analysis of three analytic methods.
J Int Neuropsychol Soc 2011;17(1):24–35. Epub 2010, Nov 12.
71. Tollard E, Galanaud D, Perlbarg V, et al. Experience of diffusion tensor
imaging and 1H spectroscopy for outcome prediction in severe traumatic
brain injury: preliminary results. Crit Care Med 2009;37:1448–55.
72. Perlbarg V, Puybasset L, Tollard E, et al. Relation between brain lesion
location and clinical outcome in patients with severe traumatic brain injury:
a diffusion tensor imaging study using voxel-based approaches. Hum Brain
Mapp 2009;30(12):3924–33.
73. Warner MA, Marquez de la Plata C, Spence J, et al. Assessing spatial
relationships between axonal integrity, regional brain volumes, and
neuropsychological outcomes after traumatic axonal injury. J Neurotrauma
2010;27:2121–30.
74. Galanaud G, Perlbarg V, Gupta R, et al. for the NICER (Neuro Imaging in
Coma Emergence and Recovery). Diffusion tensor imaging for outcome
prediction in severe traumatic brain injury: results of a prospective
multi-center trial cohort. N Engl J Med 2012 (submitted).
75. Tomaiuolo F, Worsley KJ, Lerch J, et al. Changes in white matter in
long-term survivors of severe non-missile traumatic brain injury: a
computational analysis of magnetic resonance images. J Neurotrauma
2005;22(1):76–82.
76. Ross DE. Review of longitudinal studies of MRI brain volumetry in patients
with traumatic brain injury. Brain Inj 2011;25(13–14):1271–8.
77. Warner MA, Youn TS, Davis T, et al. Regionally selective atrophy after
traumatic axonal injury. Arch Neurol 2010;67(11):1336–44. Epub 2010, Jul
12.
78. Voss HU, Uluc AM, Dyke JP, et al. Possible axonal regrowth in late recovery
from the minimally conscious state. J Clin Invest 2006;116(7):2005–11.
79. Skandsen T, Kvistad KA, Solheim O, et al. Prognostic value of magnetic
resonance imaging in moderate and severe head injury: a prospective study
of early MRI findings and one-year outcome. J Neurotrauma
2011;28(5):691–9. Epub 2011, Apr 26.
80. Dehaene S, Kerszberg M, Changeux JP. A neuronal model of a global
workspace in effortful cognitive tasks. Proc Natl Acad Sci U S A
1998;95(24):14529–34.
81. Parvizi J, Damasio AR. Neuroanatomical correlates of brainstem coma. Brain
2003;126(Pt 7):1524–36.
82. Soddu A, Vanhaudenhuyse A, Bahri MA, et al. Identifying the default-mode
component in spatial IC analyses of patients with disorders of consciousness.
Hum Brain Mapp 2012;33(4);778–96.
83. Vanhaudenhuyse A, Noirhomme Q, Tshibanda LJ, et al. Default network
connectivity reflects the level of consciousness in non-communicative
brain-damaged patients. Brain 2010;133:161–71.
84. Liu AA, Voss HU, Dyke JP, et al. Arterial spin labeling and altered cerebral
blood flow patterns in the minimally conscious state. Neurology
2011;77(16):1518–23. Epub 2011, Sep 21.
85. Carpentier A, Galanaud D, Puybasset L, et al. Early morphologic and
spectroscopic magnetic resonance in severe traumatic brain injuries can
detect “invisible brain stem damage” and predict “vegetative states.”
J Neurotrauma 2006;23(5):674–85.
86. Betz JF, Zhuo J, Roy A, et al. Prognostic value of diffusion tensor imaging
parameters in severe traumatic brain injury. J Neurotrauma 2012;Feb 24.
Epub ahead of print.
87. Wang JY, Bakhadirov K, Abdi H, et al. Longitudinal changes of structural
connectivity in traumatic axonal injury. Neurology 2011;77(9):818–26. Epub
2011, Aug 3.
88. Newcombe V, Chatfield D, Outtrim J, et al. Mapping traumatic axonal injury
using diffusion tensor imaging: correlations with functional outcome. PLoS
One 2011;6(5):e19214.
89. Matsushita M, Hosoda K, Naitoh Y, et al. Utility of diffusion tensor imaging
in the acute stage of mild to moderate traumatic brain injury for detecting
white matter lesions and predicting long-term cognitive function in adults.
J Neurosurg 2011;115(1):130–9. Epub 2011, Mar 18.
90. Sharp DJ, Beckmann CF, Greenwood R, et al. Default mode network
functional and structural connectivity after traumatic brain injury. Brain
2011;134(Pt 8):2233–47.
91. Kasahara M, Menon DK, Salmond CH, et al. Traumatic brain injury alters
the functional brain network mediating working memory. Brain Inj
2011;25(12):1170–87.

PET and SPECT
Thomas Geeraerts and David K. Menon
Introduction
Computed tomography (CT) and magnetic resonance imaging
(MRI) provide structural images of the brain. CT allows early
diagnosis of surgically treatable intracranial space-occupying
lesions in traumatic brain injury (TBI) and intracranial hem-
orrhage and exclusion of intracranial hemorrhage before
thrombolysis. MRI is very sensitive at detecting white matter
abnormalities, and provides early diagnostic and prognostic
information in stroke and TBI.1-3 However, therapeutic strate-
gies based solely on structural images are limited because
structural changes are late and irreversible. Moreover, conven-
tional imaging is poor at assessing functional recovery. Func-
tional imaging allows early documentation of reversible
ischemia in vivo, validation and calibration of bedside clinical
monitoring tools, characterization of pathophysiology in indi-
vidual patients, and early outcome prediction.
Functional CT and MRI Versus
Radioligand Techniques
for Imaging Physiology
Functional imaging approaches in clinical and experimental
stroke have traditionally used cerebral blood flow (CBF)
thresholds for ischemia and have succeeded in identifying
useful predictive values for tissue survival or death.
Although fewer studies are available in the context of head
injury, similar approaches now attempt to identify critical
physiologic thresholds for tissue survival in TBI.4-6 Xenon CT
can provide absolute quantification of CBF, but this technique
does not provide whole brain coverage, with imaging typically
being limited to two to four selected slices. Although absolute
quantification of physiologic parameters allows the most
rational application of these thresholds to clinical contexts,
clinically useful data can be obtained from assessment of rela-
tive changes in cerebrovascular physiology. Perfusion CT or
MRI with intravenous contrast media allows estimation of
relative CBF and volume (rCBF and rCBV, respectively). At a
qualitative level, functional MRI (fMRI) using blood oxygen
level dependent (BOLD) contrast allows identification of
regional brain activity and the assessment of functional con-
nectivity, but no direct measure of cerebrovascular adequacy.
Proton magnetic resonance spectroscopy (1H-MRS) studies
show changes in absolute or relative levels of metabolites, and
so can document the consequences of ischemia. Lactate
© Copyright 2013 Elsevier Inc. All rights reserved.
Chapter
29  
IV
increases and N-acetyl aspartate (NAA) reductions in normal-
appearing tissue in head injury7 probably represent ischemia
and neuronal loss, respectively,8 although acute, reversible
NAA reductions may represent mitochondrial dysfunction.9
Changes in choline, creatine, and myoinositol also have been
demonstrated, but their significance is less certain. NAA
reductions appear to be an important marker of poor
outcome,7,10 but their potential reversibility implies that late
studies may be required to define long-term significance. In
addition, although some high-resolution studies have been
published,11 current implementation of the technique is gen-
erally limited by relatively poor spatial resolution,12 limited
brain coverage, and long imaging times.11
The major advantage of radioligand techniques such as
positron emission tomography (PET) and single photon emis-
sion computed tomography (SPECT) is the opportunity to
specifically aim at a target pathophysiologic process, and espe-
cially with PET, the high sensitivity to detect physiologic
changes in brain physiology. To date, PET is probably the most
selective and sensitive method to visualize and measure in vivo
receptor density and oxidative metabolism.13 However, unlike
MRI, the main limitation of radioligand techniques is the
exposure to radiation. For PET and SPECT, the radiation
exposure typically ranges between 2 and 10 mSv (depending
on the specific radioligand use), corresponding to between
one and four times annual natural background.14 Another
limitation is the importance of partial volume effect for small
structure that limits the spatial resolution to 3 to 5 mm for
PET and to 10 mm for SPECT. Moreover, the short half-life
of positron emitters used for PET as 11C or 15O requires the
presence of a cyclotron in the site where it is used: this increases
costs.
Methodologic Issues
In both PET and SPECT, a radioactive isotope is administered
and external detectors measure brain radioactivity. To avoid
prolonged radioactive exposure, relatively short-lived isotopes
are used to image cerebral physiology and metabolism or to
label proteins, receptors, or neurotransmitters (Table 29.1).
The synthesized labeled compound must have the same meta-
bolic or pharmacokinetic proprieties as the unlabeled com-
pound. However, they typically are administered in such small
amounts that pharmacodynamic effects are not important.
Isotopes of universal atoms such as carbon or oxygen can be
used to label molecules without inducing any modification in
291
292 Section IV—Radiology

Table 29.1  Examples of Possible Applications of Radioligand Methods to Study Brain Physiology
Methods Atoms Isotopes Half-life for Emission Examples of Labeled Tracer Applications
11 11
PET Carbon C 20.3 min C-radopride D2 dopamine receptors
11
C-flumazenil GABA-A receptor
11
C-PK11195 Activated microglia
(brain inflammation)
11
C-PiB or 11C-SB13 Amyloid plaques
15
Oxygen O 2.1 min H215O, C15O, and 15O2 CBF, CBV, OEF and CMRO2
18 18
Fluorine F 109 min Fluorodeoxyglucose (FDG) Glucose metabolism
18
Fluoromisonidazole Tissue hypoxia
99m
SPECT Technetium Tc99m 6.01 hr Tc-hexamethylpropyleneamine CBF
oxide
123
Iodine I 13.1 hr Iodoamphetamine CBF

CBF, Cerebral blood flow; CBV, cerebral blood volume; CMRO2, cerebral metabolic rate of oxygen; GABA, gamma-aminobutyric acid; OEF, oxygen extraction fraction;
PET, positron emission tomography; PiB, Pittsburg compound; SPECT, single photon emission computed tomography.

POSITRON EMISSION TOMOGRAPHY SINGLE PHOTON EMISSION

(PET) COMPUTED TOMOGRAPHY (SPECT)

Detectors

Positron emitting
tracer
Single photon emitting
5 mm

β+
tracer
e–
γ ray photon e+
511 keV
γ ray photon
511 keV γ ray photon
140 keV

Scan reconstruction Scan reconstruction

Fig. 29.1  Schematic of detection in positron emission tomography (PET, left panel) and single photon emission computed tomography (SPECT using
99m
Tc, right panel).

the chemical structure. Importantly most PET and SPECT
studies are performed away from the neurocritical care unit
(NCCU), and so careful management and anesthetic tech-
nique if necessary are required because the study environment
can be unfamiliar.15
Positron Emission Tomography
PET is a technique that measures the accumulation of positron-
emitting radioisotopes within a three-dimensional object. The
production of positron-emitting tracer requires a cyclotron.
The positrons that are emitted are the antimatter equivalent
of electrons. The collision of an electron and a positron anni-
hilates both particles, resulting in a release of energy in the
form of two photons (gamma rays) released at an angle of 180
degrees to each other (Fig. 29.1). This annihilation energy can
be detected externally using coincidence detectors, and the
locus of each reaction localized within the object by computer
algorithms. The tracer distributes to different compartments,
according to its pharmacokinetic proprieties. Kinetic model-
ing that often requires measurement of radioactivity input in
plasma is used to derive quantitative measures of the cerebral
concentration. Typically this process requires placement of an
arterial catheter and rapid sampling of arterial blood during
the tracer injection and equilibration, so as to provide an
input function for modeling tracer distribution. With some
tracers, however, an alternative approach makes use of the
inhomogeneous distribution of the tracer in the brain. This
 Section IV—Radiology 293

Parametric
(functional)
Normalization map
Voxel-by-voxel to standardized
analysis space

25 Reference tissue models

20 or
15 Plasma input function
10 40
5 35
0
30
25

kB/mL
Plasma activity
20
Metaboltes corrected plasma activity
15
10
5
0
Select 0 1000 2000 3000 4000
pertinent Plasma sample time [sec]
radiotracer
Appropriate
pharmacokinetic model

Emission
scan

Scan reconstruction
Computer processing

Fig. 29.2  The complex process of analyzing an image from positron emission tomography using 11C-flumazenil.

property can be used to define a brain area without specific
tracer binding, which can be used as a reference area to provide
a tissue input function.16 For example 11C-flumazenil PET can
be used to estimate central benzodiazepine receptor gamma-
aminobutyric acid (GABA)-A density, with the pons as a
tissue reference without specific binding; this avoids the need
to sample arterial blood.17
Although PET often is referred to as the gold standard in
physiologic or metabolic imaging, the absolute values that it
produces result from complex detection and computing
systems, both of which are prone to error (Fig. 29.2). Perhaps
most important, the accuracy of figures derived from PET is
crucially dependent on the validity of kinetic models, many
of which have not been tested in the injured brain. Although
the data obtained using the technique are extremely valuable,
and probably not obtainable by other methods, it is essential
that the physiologic insights provided by the PET are assessed
critically in the context of these considerations. Although
PET clearly can define many complex aspects of cerebral
physiology and pathophysiology, it is a research tool that is
relatively expensive and not universally available.
Single Photon Emission
Computed Tomography
SPECT relies on the imaging of gamma-emitting tracers
in which a single photon is emitted (see Fig. 29.2).
Tc99m-hexamethylpropyleneamine oxide (HMPAO)18 and
123
I-amphetamine have been used to map CBF.19 SPECT does
not require a cyclotron to be produced. SPECT is a relatively
simple and inexpensive technique that can be used to assess
three-dimensional cerebral perfusion. However, the images
produced are of relatively low resolution and generally
nonquantitative.
Although recent papers have reported on quantitative
implementations of SPECT,20 this requires further validation.
294 Section IV—Radiology

rCBF CMRO2 OEF CMRgluc OGR

0-30 ml/100 g/min 0-100 µm/100 g/min 0-75% 0-40 µm/100 g/min 0-10

Fig. 29.3  Triple oxygen positron emission tomography (PET) at day 1 after removal of a traumatic left subdural hematoma. On the side
of subdural hematoma is reduced relative cerebral blood flow (rCBF) and cerebral metabolic rate of oxygen (CMRO2) with high oxygen extraction
fraction (OEF) that suggests compensation for ischemia. The decrease in oxygen glucose ratio (OGR) indicates a disproportionate reduction in oxidative
metabolism. Nonoxidative glucose use is inefficient at generating adenosine triphosphate (ATP); this helps explain the perilesional hyperglycolysis
(increased fluorodeoxyglucose [FDG] uptake). CMRgluc, Cerebral metabolic rate for glucose.

One interesting development has been the use of analysis tech-
niques such as statistical parametric mapping, derived from
functional brain imaging, to examine changes in patterns of
blood flow across populations of patients with head injury.21
Imaging of Cerebral Blood Flow
and Volume
Patient evaluation in neurocritical care often may be con-
founded by the use of sedative agents and by the metabolic
effects of trauma or injury,22 which may cause primary reduc-
tions in cerebral metabolism. A decrease in CBF that is coupled
with the depressed metabolism would not represent ischemia.
Under these circumstances the only true measure of the ade-
quacy of CBF is a measurement of the oxygen extraction frac-
tion (OEF). A robust identification of ischemia depends on
the identification of areas with increased OEF. Triple oxygen
PET (using 15O-labeled tracers in three separate scans (H215O,
C15O, and 15O2) provides valid quantitative data for CBF, cere-
bral blood volume (CBV), OEF, and the cerebral metabolic
rate of oxygen (CMRO2) (Fig. 29.3).23-25
PET can play an important role in understanding the
pathophysiologic interplay among CBF brain cellular function
in ischemic stroke, carotid artery disease, vascular dementia,
intracerebral hemorrhage, and aneurysmal subarachnoid
hemorrhage (SAH).26 In clinical stroke, PET identifies a core
area with very low CBF, CBV, and oxidative metabolism, that
corresponds to irreversible damage.27 The surrounding area
with normal or elevated CBV and relative CBF preservation
with an increase in OEF is termed the penumbra28 and repre-
sents tissue at risk, that may be rescued by physiologic or
pharmacologic intervention. The extent of the penumbra
measured with PET has been shown in stroke patients to help
predict tissue outcome.29,30
After SAH, PET and SPECT have helped to document the
modification of CBF and CBV that may show a significant
decrease during vasospasm.31-33 However, the sensitivity of
Tc99m-HMPAO SPECT to detect symptomatic vasospasm is
only 50% and perhaps less in predicting vessel narrowing in
some vascular distributions.34 Newer semiquantitative SPECT
studies using Hermes brain registration and analysis software
(BRASS) may improve on this but requires further study.35
PET CBF in symptomatic patients also has been found to be
less than that observed in asymptomatic patients.36 The cor-
relation between metabolic markers of brain ischemia mea-
sured by microdialysis and a decrease in PET CBF is not
excellent, suggesting other causes for metabolic disorders.36,37
Nevertheless, PET or the combination of techniques such as
PET, perfusion CT and microdialysis can provide insights into
the role of cellular hypoxia after SAH,38 or the response to
therapies such as normal saline boluses to induce hypervol-
emia39 or blood transfusion to improve oxygen delivery.40
Several groups describe the use of PET and SPECT to map
TBI pathophysiology. Perhaps the most interesting of these are
studies in the acute phase, which aim to elucidate early patho-
physiology. In patients with mild to moderate TBI within the
first 3 months after injury, CBF measured by SPECT is reduced
by 40% to 70% predominantly in the frontal and temporal
lobes, basal ganglia, and thalamus.41 This reduced CBF was
associated with unfavorable outcome and poor performance
on neuropsychological testing and had a better relationship
with long-term outcome than CT or conventional MRI.41-43 At
6 months after TBI the extent of regional brain atrophy cor-
relates best with PET measurements of CMRO2 and CBF and
less so lobar OEF.44 SPECT-detected improvements in CBF
also have been observed to occur when there is cognitive
recovery after concussion45 or can help explain flow patterns
associated with cognitive fatigue.46 However, SPECT as a
stand-alone test may not be sufficient to confirm that a head
injury did or did not occur.47,48
PET has been used to precisely define the ischemic thresh-
old after TBI. A recent whole brain voxel-based study demon-
strated the presence of distributed ischemia within 24 hours
of TBI, as defined by an increased OEF.49,50 In contrast, previ-
ous studies have found low global OEF values after TBI,

suggesting that global ischemia is uncommon.51,52 These dif-
ferences may reflect difficulties in demonstrating ischemia at
different times after injury, when early compensated ischemia
is associated with a high OEF. By contrast, progressive cell
death within the voxel results in progressive reductions in
OEF that culminate in very low values in voxels where most
neurons have died. Averaged CBF in areas that eventually
progressed to structural abnormality was significantly reduced
(median CBF 16.9 mL/100 g/min) compared with nonlesion
areas. These variations may be confounded by spatial averag-
ing of OEF within structurally defined regions or across the
whole brain, possibly due to the dilution of small regions
with increased OEF by larger regions with normal or low
OEF. It also is possible that there are mechanisms of ischemic
cell death that are not characterized by OEF increases, such as
microvascular ischemia49 and mitochondrial dysfunction.53,54
Indeed hypoperfused pericontusional regions where there is
oxygen hypometabolism in the subacute stage after TBI may
not always result in tissue necrosis, emphasizing the need to
know about CBF and metabolism (and from any monitoring
modality) before conclusions about ischemia are made.55
Both the extent and duration of ischemia contribute to the
evolution of irreversible tissue damage. This time dependence
may mean that it is not possible to predict tissue outcome
based on measurement of CBF, CMRO2, or OEF at a single
time point after injury. It also is likely that tissue survival will
be better predicted by complex interactions of physiologic
variables rather than simple univariate thresholds. Initial mul-
tivariate analysis appears to provide better separation of
damaged and nonlesion tissue, and further investigation may
involve using multivariate statistical analysis or neural network
techniques to determine if there are combinations of physio-
logic variables that better predict tissue outcome. Indeed tech-
niques using multivariate algorithms (using data from acute
perfusion-weighted and diffusion-weighted MRI scans), to
generate voxel-based maps predicting tissue outcome, have
been used with some success in stroke studies.56
PET studies can be repeated and used to assess changes in
physiology with therapeutic maneuvers, such as osmotherapy,
hyperventilation, hyperoxia, transfusion, and CPP augmenta-
tion.40,49,57,58 Triple-oxygen PET has been used to address the
metabolic effects of hyperventilation in TBI. Hyperventilation
to an arterial carbon dioxide tension (PaCO2) of 25 mm Hg
produces some increase in OEF, but does cause a significant
reduction in global CMRO252 or in regions of interest with low
baseline CBF values.51 It is interesting that these studies did
not detect regions with extremely high OEF values that would
suggest true ischemia, even within 24 hours of head injury.
Other groups have used PET to show that moderate reduc-
tions in PaCO2 (to 31 mm Hg) can result in increases in the
volume of brain tissue with CBF values that are less than
ischemic thresholds (<20 mL/100 g/min).32 Importantly, the
development of these ischemic areas that typically are peri-
contusional or in white matter may not be detected by reduc-
tions in jugular bulb oxygen saturations below commonly
accepted thresholds for ischemia (<55%). Subsequent studies
have shown that ischemia, defined as cerebral venous oxygen
content of less than 3.5 mL per 100 mL (that equates to OEF
values of 75%-80%), are common in early head injury and
may be observed beyond 24 hours post injury. The volume of
brain defined as ischemic using these criteria shows an increase
with hyperventilation associated with reduced CMRO2 in
Section IV—Radiology 295
some regions. Interestingly, these OEF increases are not only
from reduced oxygen delivery but also associated with increases
in oxygen demand from hyperventilation. The burden of early
ischemia measured using these techniques correlates well with
eventual outcome.32
Other studies have correlated PET-derived measures of
cerebrovascular physiology with bedside monitoring tools
such as microdialysis and brain tissue oxygen49,59 (Fig. 29.4).
These studies help validate and calibrate bedside monitors,
define the limits or reliability of a given technique, and provide
insights into pathophysiology.60
Blood flow and metabolism vary dramatically across the
traumatized brain, and different regions may require different
therapeutic approaches. This is difficult to achieve in the clini-
cal environment because medical therapies are “global” in
nature. However, it is best to measure the effect of common
therapeutic interventions in which benefit is demonstrated to
decide about continuing that therapy. A potential advantage
of PET and monitoring of brain metabolism may be detection
of abnormalities in brain regions without structural change
on other imaging modalities.
Imaging of Oxygen
and Glucose Metabolism
18-Fluorodeoxyglucose (18FDG) PET can be used to estimate
glucose metabolism. Brain-injured patients commonly dem-
onstrate evidence of global hypometabolism and metabolic
stress that fails to recover in patients who have a poor outcome.
Bergsneider et al.61 demonstrated increased global and perile-
sional 18FDG uptake in acute head injury using combined
microdialysis FDG PET. Comparison with global measure-
ments suggested that these represented hyperglycolysis (imply-
ing anaerobic glucose utilization) rather than simple
hypermetabolism. The pathophysiology underlying these
abnormalities is beginning to be elucidated, but astrocyte glu-
tamate reuptake, relative ischemia, and obligate hyperglycoly-
sis in inflammatory cells have all been considered as possible
causes. In addition ictal FDG PET can show increased hip-
pocampal glucose uptake in TBI patients with seizures or non-
convulsive seizures.62 Further insight into pathophysiology
can be obtained when the FDG PET studies are combined
with microdialysis studies of brain glucose.60 However, there
is some concern that changes in the handling of FDG in the
injured brain may make standard kinetic models for this
tracer inappropriate and therefore contribute to an apparent
FDG increase uptake that is an artifact.63
The oxygen-glucose ratio (OGR; CMRO2/CMRgluc, both
measured in µmol/100 g/min) is calculated using CMRO2 and
CMRgluc from triple oxygen and FDG PET.64 This ratio has
been used to support the concept of nonischemic metabolic
crisis.65 Reduction in OGR indicates a disproportionate reduc-
tion in oxidative metabolism, probably related to impaired
mitochondrial function and in TBI is abnormal in three quar-
ters of the patients during the early phase, suggesting that
metabolic crisis without ischemia is common after TBI.65
18
FDG PET also has been used to quantify recovery from
head injury. Although 18FDG uptake in the acute phase cor­
relates poorly with clinical state,22 sequential changes in
FDG uptake parallel functional recovery.66 For example, Garcia-
Panach et al.67 studied 49 severe TBI patients who remained
296 Section IV—Radiology

50
45 r = 0.69
40

Lactate/pyruvate ratio
35
30
25
20
15
10
5
0
0 10 20 30 40 50 60 70
B Oxygen extraction fraction (%)

5 r = 0.46

PbtO2 (kPa)
3

2
OEF
1

0
0-75% 20 25 30 35 40 45
A
C Oxygen extraction fraction (%)

Fig. 29.4  Positron emission tomography (PET) validation and calibration of bedside clinical monitoring tools. A region of interest that
corresponds to catheter location is identified (gray circle). Oxygen extraction fraction (OEF) measured by PET (A) correlates with metabolic parameter
of oxidative metabolism as lactate-to-pyruvate ratio measured by microdialysis (B) and brain tissue oxygen (PbtO2) measured with Neurotrend sensor
(C). (Redrawn from Johnston AJ, Steiner LA, Coles JP, et al. Effect of cerebral perfusion pressure augmentation on regional oxygenation and metabolism after head injury.
Crit Care Med 2005;33[1]:189–95; Hutchinson PJ, Gupta AK, Fryer TF, et al. Correlation between cerebral blood flow, substrate delivery, and metabolism in head injury:
a combined microdialysis and triple oxygen positron emission tomography study. J Cereb Blood Flow Metab 2002;22[6]:735–45.)

vegetative or minimally conscious, had recovered but had post-
traumatic amnesia, or were fully recovered, and 10 controls
using FDG PET. The patients with the least recovery had the
most severe hypometabolism. In addition, greater levels of
activation of the corticocortical connections were associated
with better neurologic recovery, and thalamic metabolism
was least in the vegetative and minimally conscious patients.67
Similarly, in boxers with suspected chronic TBI decreased
18-fluorodeoxyglucose (18FDG) uptake is observed in select
brain regions.68 In patients who remain in coma 3 months after
injury there appears to be good concordance between FDG
PET and fMRI.69
Other studies have used nonquantitative H215O PET with
cognitive paradigms to assess patterns of cortical activation in
the recovery period of head injury. These functional activation
studies do not address acute pathophysiology, but may provide
important information about the neural substrate of cognitive
and mnemonic deficits after head injury. One area where this
has the potential to be helpful is to differentiate structural or
functional brain damage from comorbid depression or post-
traumatic stress disorder (PTSD) in patients with repetitive
mild TBI, especially combat veterans.70 FDG PET studies show
decreased cerebral metabolic rate of glucose in the cerebellum,
vermis, pons, and medial temporal lobe in TBI patients with
postconcussive symptoms compared with normal controls. In
addition, functional imaging studies such as PET may have
important ethical implications for the care of brain-injured
patients, for example, enhance the ability to diagnose a vegeta-
tive or minimally conscious state.71,72
The use of FDG PET is not limited to understanding brain
metabolism. Impaired gas exchange and inflammation are
hallmarks of lung disorders such as acute respiratory distress
syndrome (ARDS), acute lung injury (ALI), and ventilator-
induced lung injury (VILI). PET can be used to examine the
relationship between the distributions of pulmonary perfu-
sion, ventilation and aeration, and the effect of positive end-
expiratory pressure, recruitment maneuvers, and prone
positioning on these pulmonary changes in ALI, whereas FDG
PET can be used to study regional neutrophil metabolic acti-
vation in ALI and VILI.73-75 In addition, SPECT studies can
help diagnose pulmonary emboli.76 PET and SPECT also can
be used to evaluate cardiovascular disease including early
detection of pathophysiologic processes before the disease
manifests,77 whereas FDG PET may play a role in the evalua-
tion of a fever of unknown origin in which it has a very high
negative predictive value.78 Finally, FDG PET has been used as
a surrogate endpoint in trials to evaluate therapies for
Alzheimer’s disease.79

Brain Receptor Imaging
Dopamine and Serotonin Receptors
Modification in dopaminergic neurotransmission is involved
in many neurodegenerative diseases, as well as in schizophre-
nia. Different subtypes of dopamine receptors can be imaged
with PET and SPECT.13 In vivo quantification of functional
effect of drugs like neuroleptics is of interest in such patholo-
gies, and is a major source of PET development. Abnormal
function of serotoninergic neurotransmission is involved in
depression and compulsive disorders. Radioligand-labeled
molecules with specific binding to serotonin receptor sub-
types using PET or SPECT appears to be of great interest
and allows study of pathophysiology and therapeutic inter­
ventions.80-82 The potential list of receptors that can be imaged
using radioligand techniques—cholinergic, opioid, and ade-
nosine receptors—is in constant evolution.13 This can provide
insight into pathophysiologic processes of specific diseases or
recovery; for example, Östberg et al.83 observed that choliner-
gic function assessed with [methyl-(11)C]N-methylpiperidyl-
4-acetate that reflects acetylcholinesterase (AChE) activity was
less in TBI survivors than controls, particularly when there
was cognitive impairment.
GABA-A Receptors
GABA receptors are expressed in the cortex and deep gray
matter and can be imaged with 11C-flumazenil (11C-FMZ)
PET, a neuronal benzodiazepine or GABA receptor ligand.84,85
11
C-FMZ binding is very low in white matter. This marker
can be used to measure neuronal integrity or loss.25,86-88
SPECT imaging with [123I] iomazenil (IMZ)) also suggests
that imaging with this radioligand for the central benzodiaz-
epine receptor in the cortex may have the potential to dis-
close a reversible vulnerability of neurons following TBI.89 In
ischemic stroke 11C-FMZ PET is very sensitive at detecting
irreversible tissue damage and can be used to select patients
who may benefit from reperfusion therapy.90,91 11C-FMZ also
is used with high sensitivity in epilepsy to study the modifi-
cations of GABAergic neurotransmission induced by the
pathology and by its treatment with antiepileptic drugs.92,93
When combined with FDG PET, 11C-FMZ PET can help
provide insights into the coma and the vegetative state.72
Specific Pathophysiologic Processes
PK11195 is a ligand for peripheral benzodiazepine receptors
expressed by microglia and up-regulated in case of microglia
activation.94 PK11195 PET has been used to image, quantify,
and localize microglia activation in experimental models of
ischemia95 and in humans in multiple sclerosis, Alzheimer’s
disease, and herpes encephalitis.96-98 Because microglia activa-
tion is thought to be involved in the many pathologic pro-
cesses that contribute to neurodegeneration, the quantification
of the effect of therapeutic intervention on reactive microglia
is of particular interest. For example, studies in TBI patients
using PK11195 PET show that increased microglial activation
may be present up to 17 years after TBI and that high PK
binding in the thalamus is associated with greater cognitive
impairment.99
More recently PET studies with carbon 11-labeled Pitts-
burgh compound B ([(11)C]PiB) have been used to measure
Section IV—Radiology 297
cerebral beta-amyloid in patients with Down syndrome and
Alzheimer’s disease.100-102 Currently [(18)F]3’-F-PiB (flute-
metamol), (18)F-AV-45 (florbetapir), and (18)F-AV-1 (florbe-
taben) are undergoing phase II and III clinical trials. Ideally
these ligands may be able to help detect amyloid plaque in vivo
in the brains of Alzheimer’s disease patients or subjects with
mild cognitive impairment.103
Conclusion
PET and less so SPECT are relatively expensive and not uni-
versally available, but they represent powerful research tools
for functional imaging of the brain of patients in the NCCU
and for diverse pathologies such as Parkinson’s disease,
Alzheimer’s disease, depression and other mood disorders,
cancer including evaluation of glioblastoma aggressiveness,
lung dysfunction including detection of pulmonary embolus,
cardiac disease, and fever.73,76-78,104-111 PET provides quantitative
and SPECT qualitative data, and in neurocritical care both are
reliable techniques to monitor brain physiology. Their excel-
lent sensitivity at detecting small changes in cerebral physiol-
ogy has helped increase knowledge about the processes of
brain injury and recovery in a variety of neurologic conditions
including TBI and stroke. Furthermore, use of these techniques
in relatively small patient samples has provided insight into
how commonly used therapeutic strategies (e.g., induced
hypertension or hyperventilation) affect brain physiology and
have so informed better therapeutic approaches.
Acknowledgments
Dr. Thomas Geeraerts is supported by grants from the Société Française
d’Anesthésie et de Réanimation (SFAR) and from Journées d’Enseignement
Post-Universitaire d’Anesthésie-Réanimation (JEPU)-Novo Nordisk.
Professor David K. Menon is part of the Neuroscience Theme at the
Cambridge Biomedical Research Centre, and is supported by grants from
the National Institute of Health Research UK, Medical Research Council
(UK), Royal College of Anaesthetists, Wellcome Trust and Queens’
College Cambridge.
References
1. Kampfl A, Franz G, Aichner F, et al. The persistent vegetative state after
closed head injury: clinical and magnetic resonance imaging findings in 42
patients. J Neurosurg 1998;88(5):809–16.
2. Kampfl A, Schmutzhard E, Franz G, et al. Prediction of recovery from
post-traumatic vegetative state with cerebral magnetic-resonance imaging.
Lancet 1998;351(9118):1763–7.
3. Lovblad KO, Baird AE, Schlaug G, et al. Ischemic lesion volumes in acute
stroke by diffusion-weighted magnetic resonance imaging correlate with
clinical outcome. Ann Neurol 1997;42(2):164–70.
4. Heiss WD, Huber M, Fink GR, et al. Progressive derangement of periinfarct
viable tissue in ischemic stroke. J Cereb Blood Flow Metab 1992;12(2):193–203.
5. Marchal G, Beaudouin V, Rioux P, et al. Prolonged persistence of substantial
volumes of potentially viable brain tissue after stroke: a correlative PET-CT
study with voxel-based data analysis. Stroke 1996;27(4):599–606.
6. Powers WJ, Grubb RL, Jr., Darriet D, et al. Cerebral blood flow and cerebral
metabolic rate of oxygen requirements for cerebral function and viability in
humans. J Cereb Blood Flow Metab 1985;5(4):600–8.
7. Garnett MR, Blamire AM, Rajagopalan B, et al. Evidence for cellular damage
in normal-appearing white matter correlates with injury severity in patients
following traumatic brain injury: a magnetic resonance spectroscopy study.
Brain 2000;123(Pt 7):1403–9.
8. Higuchi T, Fernandez EJ, Maudsley AA, et al. Mapping of lactate and
N-acetyl-L-aspartate predicts infarction during acute focal ischemia: in vivo
1H magnetic resonance spectroscopy in rats. Neurosurgery 1996;38(1):121–
9; discussion 129–30.
298 Section IV—Radiology
9. Signoretti S, Marmarou A, Tavazzi B, et al. N-Acetylaspartate reduction as a
measure of injury severity and mitochondrial dysfunction following diffuse
traumatic brain injury. J Neurotrauma 2001;18(10):977–91.
10. Friedman SD, Brooks WM, Jung RE, et al. Quantitative proton MRS predicts
outcome after traumatic brain injury. Neurology 1999;52(7):1384–91.
11. Beauchamp NJ, Jr, Barker PB, Wang PY, et al. Imaging of acute cerebral
ischemia. Radiology 1999;212(2):307–24.
12. Garnett MR, Corkill RG, Blamire AM, et al. Altered cellular metabolism
following traumatic brain injury: a magnetic resonance spectroscopy study.
J Neurotrauma 2001;18(3):231–40.
13. Heiss WD, Herholz K. Brain receptor imaging. J Nucl Med
2006;47(2):302–12.
14. Small GW, Bookheimer SY, Thompson PM, et al. Current and future uses of
neuroimaging for cognitively impaired patients. Lancet Neurol
2008;7(2):161–72.
15. Veenith T, Coles JP. Anaesthesia for magnetic resonance imaging and
positron emission tomography. Curr Opin Anaesthesiol 2011;24(4):451–8.
16. Gunn RN, Lammertsma AA, Hume SP, et al. Parametric imaging of
ligand-receptor binding in PET using a simplified reference region model.
Neuroimage 1997;6(4):279–87.
17. Klumpers UM, Veltman DJ, Boellaard R, et al. Comparison of plasma input
and reference tissue models for analysing [(11)C]flumazenil studies. J Cereb
Blood Flow Metab 2008;28(3):579–87.
18. Bullock R, Statham P, Patterson J, et al. The time course of vasogenic
oedema after focal human head injury—evidence from SPECT mapping of
blood brain barrier defects. Acta Neurochir Suppl (Wien) 1990;51:286–8.
19. Shiina G, Onuma T, Kameyama M, et al. Sequential assessment of cerebral
blood flow in diffuse brain injury by 123I-iodoamphetamine single-photon
emission CT. AJNR Am J Neuroradiol 1998;19(2):297–302.
20. Della Corte F, Giordano A, Pennisi MA, et al. Quantitative cerebral blood
flow and metabolism determination in the first 48 hours after severe head
injury with a new dynamic SPECT device. Acta Neurochir (Wien) 1997;
139(7):636–41; discussion 41–2.
21. Stamatakis EA, Wilson JT, Hadley DM, et al. SPECT imaging in head injury
interpreted with statistical parametric mapping. J Nucl Med 2002;43(4):
476–83.
22. Bergsneider M, Hovda DA, Lee SM, et al. Dissociation of cerebral glucose
metabolism and level of consciousness during the period of metabolic
depression following human traumatic brain injury. J Neurotrauma 2000;
17(5):389–401.
23. Frackowiak RS, Jones T, Lenzi GL, et al. Regional cerebral oxygen utilization
and blood flow in normal man using oxygen-15 and positron emission
tomography. Acta Neurol Scand 1980;62(6):336–44.
24. Frackowiak RS, Lenzi GL, Jones T, et al. Quantitative measurement of
regional cerebral blood flow and oxygen metabolism in man using 15O and
positron emission tomography: theory, procedure, and normal values.
J Comput Assist Tomogr 1980;4(6):727–36.
25. Sette G, Baron JC, Mazoyer B, et al. Local brain haemodynamics and oxygen
metabolism in cerebrovascular disease: positron emission tomography. Brain
1989;112(Pt 4):931–51.
26. Powers WJ, Zazulia AR. PET in cerebrovascular disease. PET Clin 2010;5(1):
83106.
27. Baron JC. Mapping the ischaemic penumbra with PET: implications for
acute stroke treatment. Cerebrovasc Dis 1999;9(4):193–201.
28. Moustafa RR, Baron JC. Clinical review: imaging in ischaemic stroke—
implications for acute management. Crit Care 2007;11(5):227.
29. Heiss WD. Best measure of ischemic penumbra: positron emission
tomography. Stroke 2003;34(10):2534–5.
30. Heiss WD, Kracht LW, Thiel A, et al. Penumbral probability thresholds of
cortical flumazenil binding and blood flow predicting tissue outcome in
patients with cerebral ischaemia. Brain 2001;124(Pt 1):20–9.
31. Leclerc X, Fichten A, Gauvrit JY, et al. Symptomatic vasospasm after
subarachnoid haemorrhage: assessment of brain damage by diffusion and
perfusion-weighted MRI and single-photon emission computed tomography.
Neuroradiology 2002;44(7):610–6.
32. Coles JP, Minhas PS, Fryer TD, et al. Effect of hyperventilation on cerebral
blood flow in traumatic head injury: clinical relevance and monitoring
correlates. Crit Care Med 2002;30(9):1950–9.
33. Jabre A, Babikian V, Powsner RA, et al. Role of single photon emission
computed tomography and transcranial Doppler ultrasonography in clinical
vasospasm. J Clin Neurosci 2002;9(4):400–3.
34. Kincaid MS, Souter MJ, Treggiari MM, et al. Accuracy of transcranial
Doppler ultrasonography and single-photon emission computed
tomography in the diagnosis of angiographically demonstrated cerebral
vasospasm. J Neurosurg 2009;110(1):67–72.
35. Cooke D, Koppula B, Seiler D, et al. Semiquantitative software SPECT
analysis in aneurysmal subarachnoid hemorrhage-related vasospasm. Nucl
Med Commun 2010;31(1):53–8.
36. Sarrafzadeh AS, Haux D, Ludemann L, et al. Cerebral ischemia in
aneurysmal subarachnoid hemorrhage: a correlative microdialysis-PET
study. Stroke 2004;35(3):638–43.
37. Enblad P, Valtysson J, Andersson J, et al. Simultaneous intracerebral
microdialysis and positron emission tomography in the detection of
ischemia in patients with subarachnoid hemorrhage. J Cereb Blood Flow
Metab 1996;16(4):637–44.
38. Sarrafzadeh AS, Nagel A, Czabanka M, et al. Imaging of hypoxic-ischemic
penumbra with (18)F-fluoromisonidazole PET/CT and measurement of
related cerebral metabolism in aneurysmal subarachnoid hemorrhage.
J Cereb Blood Flow Metab 2010;30(1):36–45. Epub 2009, Sep 23.
39. Jost SC, Diringer MN, Zazulia AR, et al. Effect of normal saline bolus on
cerebral blood flow in regions with low baseline flow in patients with
vasospasm following subarachnoid hemorrhage. J Neurosurg 2005;103(1):
25–30.
40. Dhar R, Zazulia AR, Videen TO, et al. Red blood cell transfusion increases
cerebral oxygen delivery in anemic patients with subarachnoid hemorrhage.
Stroke 2009;40(9):3039–44. Epub 2009, Jul 23.
41. Bavetta S, Nimmon CC, White J, et al. A prospective study comparing SPET
with MRI and CT as prognostic indicators following severe closed head
injury. Nucl Med Commun 1994;15(12):961–8.
42. Jacobs A, Put E, Ingels M, et al. Prospective evaluation of technetium-99m-
HMPAO SPECT in mild and moderate traumatic brain injury. J Nucl Med
1994;35(6):942–7.
43. Prayer L, Wimberger D, Oder W, et al. Cranial MR imaging and cerebral
99mTc HM-PAO-SPECT in patients with subacute or chronic severe closed
head injury and normal CT examinations. Acta Radiol 1993;34(6):593–9.
44. Xu Y, McArthur DL, Alger JR, et al. Early nonischemic oxidative metabolic
dysfunction leads to chronic brain atrophy in traumatic brain injury. J Cereb
Blood Flow Metab 2010;30(4):883–94. Epub 2009, Dec 23.
45. Amen DG, Wu JC, Taylor D, et al. Reversing brain damage in former NFL
players: implications for traumatic brain injury and substance abuse
rehabilitation. J Psychoactive Drugs 2011;43(1):1–5.
46. Hattori N, Swan M, Stobbe GA, et al. Differential SPECT activation patterns
associated with PASAT performance may indicate frontocerebellar functional
dissociation in chronic mild traumatic brain injury. J Nucl Med
2009;50(7):1054–61. Epub 2009, Jun 12.
47. Wortzel HS, Filley CM, Anderson CA, et al. Forensic applications of cerebral
single photon emission computed tomography in mild traumatic brain
injury. J Am Acad Psychiatry Law 2008;36(3):310–22.
48. Lewine JD, Davis JT, Bigler ED, et al. Objective documentation of traumatic
brain injury subsequent to mild head trauma: multimodal brain imaging
with MEG, SPECT, and MRI. J Head Trauma Rehabil 2007;22(3):141–55.
49. Johnston AJ, Steiner LA, Coles JP, et al. Effect of cerebral perfusion pressure
augmentation on regional oxygenation and metabolism after head injury.
Crit Care Med 2005;33(1):189–95.
50. Abate MG, Trivedi M, Fryer TD, et al. Early derangements in oxygen and
glucose metabolism following head injury: the ischemic penumbra and
pathophysiological heterogeneity. Neurocrit Care 2008;9(3):319–25.
A complete list of references for this chapter can be found online at
www.expertconsult.com.

References
1. Kampfl A, Franz G, Aichner F, et al. The persistent vegetative state after
closed head injury: clinical and magnetic resonance imaging findings in 42
patients. J Neurosurg 1998;88(5):809–16.
2. Kampfl A, Schmutzhard E, Franz G, et al. Prediction of recovery from
post-traumatic vegetative state with cerebral magnetic-resonance imaging.
Lancet 1998;351(9118):1763–7.
3. Lovblad KO, Baird AE, Schlaug G, et al. Ischemic lesion volumes in acute
stroke by diffusion-weighted magnetic resonance imaging correlate with
clinical outcome. Ann Neurol 1997;42(2):164–70.
4. Heiss WD, Huber M, Fink GR, et al. Progressive derangement of periinfarct
viable tissue in ischemic stroke. J Cereb Blood Flow Metab 1992;12(2):
193–203.
5. Marchal G, Beaudouin V, Rioux P, et al. Prolonged persistence of substantial
volumes of potentially viable brain tissue after stroke: a correlative PET-CT
study with voxel-based data analysis. Stroke 1996;27(4):599–606.
6. Powers WJ, Grubb RL, Jr., Darriet D, et al. Cerebral blood flow and cerebral
metabolic rate of oxygen requirements for cerebral function and viability in
humans. J Cereb Blood Flow Metab 1985;5(4):600–8.
7. Garnett MR, Blamire AM, Rajagopalan B, et al. Evidence for cellular
damage in normal-appearing white matter correlates with injury severity in
patients following traumatic brain injury: a magnetic resonance
spectroscopy study. Brain 2000;123(Pt 7):1403–9.
8. Higuchi T, Fernandez EJ, Maudsley AA, et al. Mapping of lactate and
N-acetyl-L-aspartate predicts infarction during acute focal ischemia: in vivo
1H magnetic resonance spectroscopy in rats. Neurosurgery 1996;38(1):121–
9; discussion 129–30.
9. Signoretti S, Marmarou A, Tavazzi B, et al. N-Acetylaspartate reduction as a
measure of injury severity and mitochondrial dysfunction following diffuse
traumatic brain injury. J Neurotrauma 2001;18(10):977–91.
10. Friedman SD, Brooks WM, Jung RE, et al. Quantitative proton MRS
predicts outcome after traumatic brain injury. Neurology
1999;52(7):1384–91.
11. Beauchamp NJ, Jr, Barker PB, Wang PY, et al. Imaging of acute cerebral
ischemia. Radiology 1999;212(2):307–24.
12. Garnett MR, Corkill RG, Blamire AM, et al. Altered cellular metabolism
following traumatic brain injury: a magnetic resonance spectroscopy study.
J Neurotrauma 2001;18(3):231–40.
13. Heiss WD, Herholz K. Brain receptor imaging. J Nucl Med
2006;47(2):302–12.
14. Small GW, Bookheimer SY, Thompson PM, et al. Current and future uses
of neuroimaging for cognitively impaired patients. Lancet Neurol
2008;7(2):161–72.
15. Veenith T, Coles JP. Anaesthesia for magnetic resonance imaging and
positron emission tomography. Curr Opin Anaesthesiol 2011;24(4):451–8.
16. Gunn RN, Lammertsma AA, Hume SP, et al. Parametric imaging of
ligand-receptor binding in PET using a simplified reference region model.
Neuroimage 1997;6(4):279–87.
17. Klumpers UM, Veltman DJ, Boellaard R, et al. Comparison of plasma input
and reference tissue models for analysing [(11)C]flumazenil studies. J Cereb
Blood Flow Metab 2008;28(3):579–87.
18. Bullock R, Statham P, Patterson J, et al. The time course of vasogenic
oedema after focal human head injury—evidence from SPECT mapping of
blood brain barrier defects. Acta Neurochir Suppl (Wien) 1990;51:286–8.
19. Shiina G, Onuma T, Kameyama M, et al. Sequential assessment of cerebral
blood flow in diffuse brain injury by 123I-iodoamphetamine single-photon
emission CT. AJNR Am J Neuroradiol 1998;19(2):297–302.
20. Della Corte F, Giordano A, Pennisi MA, et al. Quantitative cerebral blood
flow and metabolism determination in the first 48 hours after severe head
injury with a new dynamic SPECT device. Acta Neurochir (Wien) 1997;
139(7):636–41; discussion 41–2.
21. Stamatakis EA, Wilson JT, Hadley DM, et al. SPECT imaging in head injury
interpreted with statistical parametric mapping. J Nucl Med 2002;43(4):
476–83.
22. Bergsneider M, Hovda DA, Lee SM, et al. Dissociation of cerebral glucose
metabolism and level of consciousness during the period of metabolic
depression following human traumatic brain injury. J Neurotrauma 2000;
17(5):389–401.
23. Frackowiak RS, Jones T, Lenzi GL, et al. Regional cerebral oxygen utilization
and blood flow in normal man using oxygen-15 and positron emission
tomography. Acta Neurol Scand 1980;62(6):336–44.
24. Frackowiak RS, Lenzi GL, Jones T, et al. Quantitative measurement of
regional cerebral blood flow and oxygen metabolism in man using 15O and
positron emission tomography: theory, procedure, and normal values.
J Comput Assist Tomogr 1980;4(6):727–36.
Section IV—Radiology 298.e1
25. Sette G, Baron JC, Mazoyer B, et al. Local brain haemodynamics and
oxygen metabolism in cerebrovascular disease: positron emission
tomography. Brain 1989;112(Pt 4):931–51.
26. Powers WJ, Zazulia AR. PET in cerebrovascular disease. PET Clin 2010;5(1):
83106.
27. Baron JC. Mapping the ischaemic penumbra with PET: implications for
acute stroke treatment. Cerebrovasc Dis 1999;9(4):193–201.
28. Moustafa RR, Baron JC. Clinical review: imaging in ischaemic stroke—
implications for acute management. Crit Care 2007;11(5):227.
29. Heiss WD. Best measure of ischemic penumbra: positron emission
tomography. Stroke 2003;34(10):2534–5.
30. Heiss WD, Kracht LW, Thiel A, et al. Penumbral probability thresholds of
cortical flumazenil binding and blood flow predicting tissue outcome in
patients with cerebral ischaemia. Brain 2001;124(Pt 1):20–9.
31. Leclerc X, Fichten A, Gauvrit JY, et al. Symptomatic vasospasm after
subarachnoid haemorrhage: assessment of brain damage by diffusion and
perfusion-weighted MRI and single-photon emission computed
tomography. Neuroradiology 2002;44(7):610–6.
32. Coles JP, Minhas PS, Fryer TD, et al. Effect of hyperventilation on cerebral
blood flow in traumatic head injury: clinical relevance and monitoring
correlates. Crit Care Med 2002;30(9):1950–9.
33. Jabre A, Babikian V, Powsner RA, et al. Role of single photon emission
computed tomography and transcranial Doppler ultrasonography in
clinical vasospasm. J Clin Neurosci 2002;9(4):400–3.
34. Kincaid MS, Souter MJ, Treggiari MM, et al. Accuracy of transcranial
Doppler ultrasonography and single-photon emission computed
tomography in the diagnosis of angiographically demonstrated cerebral
vasospasm. J Neurosurg 2009;110(1):67–72.
35. Cooke D, Koppula B, Seiler D, et al. Semiquantitative software SPECT
analysis in aneurysmal subarachnoid hemorrhage-related vasospasm. Nucl
Med Commun 2010;31(1):53–8.
36. Sarrafzadeh AS, Haux D, Ludemann L, et al. Cerebral ischemia in
aneurysmal subarachnoid hemorrhage: a correlative microdialysis-PET
study. Stroke 2004;35(3):638–43.
37. Enblad P, Valtysson J, Andersson J, et al. Simultaneous intracerebral
microdialysis and positron emission tomography in the detection of
ischemia in patients with subarachnoid hemorrhage. J Cereb Blood Flow
Metab 1996;16(4):637–44.
38. Sarrafzadeh AS, Nagel A, Czabanka M, et al. Imaging of hypoxic-ischemic
penumbra with (18)F-fluoromisonidazole PET/CT and measurement of
related cerebral metabolism in aneurysmal subarachnoid hemorrhage.
J Cereb Blood Flow Metab 2010;30(1):36–45. Epub 2009, Sep 23.
39. Jost SC, Diringer MN, Zazulia AR, et al. Effect of normal saline bolus on
cerebral blood flow in regions with low baseline flow in patients with
vasospasm following subarachnoid hemorrhage. J Neurosurg
2005;103(1):25–30.
40. Dhar R, Zazulia AR, Videen TO, et al. Red blood cell transfusion increases
cerebral oxygen delivery in anemic patients with subarachnoid hemorrhage.
Stroke 2009;40(9):3039–44. Epub 2009, Jul 23.
41. Bavetta S, Nimmon CC, White J, et al. A prospective study comparing SPET
with MRI and CT as prognostic indicators following severe closed head
injury. Nucl Med Commun 1994;15(12):961–8.
42. Jacobs A, Put E, Ingels M, et al. Prospective evaluation of technetium-99m-
HMPAO SPECT in mild and moderate traumatic brain injury. J Nucl Med
1994;35(6):942–7.
43. Prayer L, Wimberger D, Oder W, et al. Cranial MR imaging and cerebral
99mTc HM-PAO-SPECT in patients with subacute or chronic severe
closed head injury and normal CT examinations. Acta Radiol 1993;34(6):
593–9.
44. Xu Y, McArthur DL, Alger JR, et al. Early nonischemic oxidative metabolic
dysfunction leads to chronic brain atrophy in traumatic brain injury.
J Cereb Blood Flow Metab 2010;30(4):883–94. Epub 2009, Dec 23.
45. Amen DG, Wu JC, Taylor D, et al. Reversing brain damage in former NFL
players: implications for traumatic brain injury and substance abuse
rehabilitation. J Psychoactive Drugs 2011;43(1):1–5.
46. Hattori N, Swan M, Stobbe GA, et al. Differential SPECT activation
patterns associated with PASAT performance may indicate frontocerebellar
functional dissociation in chronic mild traumatic brain injury. J Nucl Med
2009;50(7):1054–61. Epub 2009, Jun 12.
47. Wortzel HS, Filley CM, Anderson CA, et al. Forensic applications of
cerebral single photon emission computed tomography in mild traumatic
brain injury. J Am Acad Psychiatry Law 2008;36(3):310–22.
48. Lewine JD, Davis JT, Bigler ED, et al. Objective documentation of
traumatic brain injury subsequent to mild head trauma: multimodal
brain imaging with MEG, SPECT, and MRI. J Head Trauma Rehabil 2007;
22(3):141–55.
298.e2 Section IV—Radiology
49. Johnston AJ, Steiner LA, Coles JP, et al. Effect of cerebral perfusion pressure
augmentation on regional oxygenation and metabolism after head injury.
Crit Care Med 2005;33(1):189–95.
50. Abate MG, Trivedi M, Fryer TD, et al. Early derangements in oxygen and
glucose metabolism following head injury: the ischemic penumbra and
pathophysiological heterogeneity. Neurocrit Care 2008;9(3):319–25.
51. Diringer MN, Videen TO, Yundt K, et al. Regional cerebrovascular and
metabolic effects of hyperventilation after severe traumatic brain injury.
J Neurosurg 2002;96(1):103–8.
52. Diringer MN, Yundt K, Videen TO, et al. No reduction in cerebral
metabolism as a result of early moderate hyperventilation following severe
traumatic brain injury. J Neurosurg 2000;92(1):7–13.
53. Verweij BH, Muizelaar JP, Vinas FC, et al. Improvement in mitochondrial
dysfunction as a new surrogate efficiency measure for preclinical trials:
dose-response and time-window profiles for administration of the calcium
channel blocker Ziconotide in experimental brain injury. J Neurosurg 2000;
93(5):829–34.
54. Clausen T, Zauner A, Levasseur JE, et al. Induced mitochondrial failure in
the feline brain: implications for understanding acute post-traumatic
metabolic events. Brain Res 2001;908(1):35–48.
55. Kawai N, Nakamura T, Tamiya T, et al. Metabolic disturbance without brain
ischemia in traumatic brain injury: a positron emission tomography study.
Acta Neurochir Suppl 2008;102:241–5.
56. Jiang Q, Chopp M, Zhang ZG, et al. The temporal evolution of MRI tissue
signatures after transient middle cerebral artery occlusion in rat. J Neurol
Sci 1997;145(1):15–23.
57. Diringer MN, Scalfani MT, Zazulia AR, et al. Cerebral hemodynamic and
metabolic effects of equi-osmolar doses mannitol and 23.4% saline in
patients with edema following large ischemic stroke. Neurocrit Care 2011;
14(1):11–7.
58. Nortje J, Coles JP, Timofeev I, et al. Effect of hyperoxia on regional
oxygenation and metabolism after severe traumatic brain injury:
preliminary findings. Crit Care Med 2008;36(1):273–81.
59. Hutchinson PJ, Gupta AK, Fryer TF, et al. Correlation between cerebral
blood flow, substrate delivery, and metabolism in head injury: a combined
microdialysis and triple oxygen positron emission tomography study.
J Cereb Blood Flow Metab 2002;22(6):735–45.
60. Hutchinson PJ, O’Connell MT, Seal A, et al. A combined microdialysis and
FDG-PET study of glucose metabolism in head injury. Acta Neurochir
(Wien) 2009;151(1):51–61; discussion 61. Epub 2008, Dec 20.
61. Bergsneider M, Hovda DA, Shalmon E, et al. Cerebral hyperglycolysis
following severe traumatic brain injury in humans: a positron emission
tomography study. J Neurosurg 1997;86(2):241–51.
62. Vespa PM, McArthur DL, Xu Y, et al. Nonconvulsive seizures after
traumatic brain injury are associated with hippocampal atrophy. Neurology
2010;75(9):792–8.
63. Richards HK, Simac S, Piechnik S, et al. Uncoupling of cerebral blood flow
and metabolism after cerebral contusion in the rat. J Cereb Blood Flow
Metab 2001;21(7):779–81.
64. Wu HM, Huang SC, Hattori N, et al. Selective metabolic reduction in gray
matter acutely following human traumatic brain injury. J Neurotrauma
2004;21(2):149–61.
65. Vespa P, Bergsneider M, Hattori N, et al. Metabolic crisis without brain
ischemia is common after traumatic brain injury: a combined microdialysis
and positron emission tomography study. J Cereb Blood Flow Metab
2005;25(6):763–74.
66. Bergsneider M, Hovda DA, McArthur DL, et al. Metabolic recovery
following human traumatic brain injury based on FDG-PET: time course
and relationship to neurological disability. J Head Trauma Rehabil 2001;
16(2):135–48.
67. García-Panach J, Lull N, Lull JJ, et al. A voxel-based analysis of FDG-PET in
traumatic brain injury: regional metabolism and relationship between the
thalamus and cortical areas. J Neurotrauma 2011;28(9):1707–17.
68. Provenzano FA, Jordan B, Tikofsky RS, et al. F-18 FDG PET imaging of
chronic traumatic brain injury in boxers: a statistical parametric analysis.
Nucl Med Commun 2010;31(11):952–7.
69. Kremer S, Nicolas-Ong C, Schunck T, et al. Usefulness of functional MRI
associated with PET scan and evoked potentials in the evaluation of brain
functions after severe brain injury: preliminary results. J Neuroradiol 2010;
37(3):159–66. Epub 2009, Sep 24.
70. Peskind ER, Petrie EC, Cross DJ, et al. Cerebrocerebellar hypometabolism
associated with repetitive blast exposure mild traumatic brain injury in 12
Iraq war veterans with persistent post-concussive symptoms. Neuroimage
2011;54 Suppl 1:S76–82. Epub 2010, Apr 10.
71. Bernat JL. Ethical issues in the treatment of severe brain injury: the impact
of new technologies. Ann N Y Acad Sci 2009;1157:117–30.
72. Heiss WD. PET in coma and in vegetative state. Eur J Neurol 2011;Jul 28.
Epub ahead of print.
73. Musch G. Positron emission tomography: a tool for better understanding
of ventilator-induced and acute lung injury. Curr Opin Crit Care 2011;
17(1):7–12.
74. de Prost N, Tucci MR, Melo MF. Assessment of lung inflammation with
18F-FDG PET during acute lung injury. AJR Am J Roentgenol 2010;195(2):
292–300.
75. Bellani G, Messa C, Guerra L, et al. Lungs of patients with acute respiratory
distress syndrome show diffuse inflammation in normally aerated regions:
a [18F]-fluoro-2-deoxy-D-glucose PET/CT study. Crit Care Med 2009;
37(7):2216–22.
76. Morris TA, Gerometta M, Yusen RD, et al. Detection of Pulmonary
Emboli With 99mTc-labeled anti-D-dimer (DI-80B3)Fab’ fragments
(ThromboView). Am J Respir Crit Care Med 2011;Jun 16. Epub ahead of
print.
77. Dobrucki LW, Sinusas AJ. PET and SPECT in cardiovascular molecular
imaging. Nat Rev Cardiol 2010;7(1):38–47. Epub 2009, Nov 24.
78. Keidar Z, Gurman-Balbir A, Gaitini D, et al. Fever of unknown origin: the
role of 18F-FDG PET/CT. J Nucl Med 2008;49(12):1980–5. Epub 2008, Nov 7.
79. Tzimopoulou S, Cunningham VJ, Nichols TE, et al. A multi-center
randomized proof-of-concept clinical trial applying [18F]FDG-PET for
evaluation of metabolic therapy with rosiglitazone XR in mild to moderate
Alzheimer’s disease. J Alzheimers Dis 2010;22(4):1241–56.
80. Tauscher J, Bagby RM, Javanmard M, et al. Inverse relationship between
serotonin 5-HT(1A) receptor binding and anxiety: a [(11)C]WAY-100635
PET investigation in healthy volunteers. Am J Psychiatry 2001;158(8):
1326–8.
81. Stamatakis EA, Hetherington MM. Neuroimaging in eating disorders. Nutr
Neurosci 2003;6(6):325–34.
82. Parsey RV, Kegeles LS, Hwang DR, et al. In vivo quantification of brain
serotonin transporters in humans using [11C]McN 5652. J Nucl Med
2000;41(9):1465–77.
83. Östberg A, Virta J, Rinne JO, et al. Cholinergic dysfunction after traumatic
brain injury: preliminary findings from a PET study. Neurology
2011;76(12):1046–50.
84. Heiss WD, Kracht L, Grond M, et al. Early [(11)C]flumazenil/H(2)O
positron emission tomography predicts irreversible ischemic cortical
damage in stroke patients receiving acute thrombolytic therapy. Stroke
2000;31(2):366–9.
85. Heiss WD, Kracht LW, Thiel A, et al. Penumbral probability thresholds of
cortical flumazenil binding and blood flow predicting tissue outcome in
patients with cerebral ischaemia. Brain 2001;124(Pt 1):20–9.
86. Heiss WD, Graf R, Fujita T, et al. Early detection of irreversibly damaged
ischemic tissue by flumazenil positron emission tomography in cats. Stroke
1997;28(10):2045–51.
87. Baron JC. Mapping the ischaemic penumbra with PET: a new approach.
Brain 2001;124(Pt 1):2–4.
88. Geeraerts T, Coles JP, Aigbirhio FI, et al. Validation of reference tissue
modelling for [11C]flumazenil positron emission tomography following
head injury. Ann Nucl Med 2011;25(6):396–405. Epub 2011, Apr 2.
89. Koizumi H, Fujisawa H, Kurokawa T, et al. Recovered neuronal viability
revealed by Iodine-123-iomazenil SPECT following traumatic brain injury.
J Cereb Blood Flow Metab 2010;30(10):1673–81. Epub 2010, Aug 4.
90. Heiss WD, Grond M, Thiel A, et al. Permanent cortical damage detected by
flumazenil positron emission tomography in acute stroke. Stroke 1998;
29(2):454–61.
91. Heiss WD, Grond M, Thiel A, et al. Tissue at risk of infarction rescued by
early reperfusion: a positron emission tomography study in systemic
recombinant tissue plasminogen activator thrombolysis of acute stroke.
J Cereb Blood Flow Metab 1998;18(12):1298–307.
92. Savic I, Ingvar M, Stone-Elander S. Comparison of [11C]flumazenil and
[18F]FDG as PET markers of epileptic foci. J Neurol Neurosurg Psychiatry
1993;56(6):615–21.
93. Richardson MP, Koepp MJ, Brooks DJ, et al. Benzodiazepine receptors in
focal epilepsy with cortical dysgenesis: an 11C-flumazenil PET study. Ann
Neurol 1996;40(2):188–98.
94. Tai YF, Piccini P. Applications of positron emission tomography (PET) in
neurology. J Neurol Neurosurg Psychiatry 2004;75(5):669–76.
95. Rojas S, Martin A, Arranz MJ, et al. Imaging brain inflammation with [(11)
C]PK11195 by PET and induction of the peripheral-type benzodiazepine
receptor after transient focal ischemia in rats. J Cereb Blood Flow Metab
2007;27(12):1975–86.
96. Banati RB, Newcombe J, Gunn RN, et al. The peripheral benzodiazepine
binding site in the brain in multiple sclerosis: quantitative in vivo imaging
of microglia as a measure of disease activity. Brain 2000;123(Pt 11):2321–37.

97. Cagnin A, Brooks DJ, Kennedy AM, et al. In-vivo measurement of activated
microglia in dementia. Lancet 2001;358(9280):461–7.
98. Cagnin A, Myers R, Gunn RN, et al. In vivo visualization of activated glia
by [11C] (R)-PK11195-PET following herpes encephalitis reveals projected
neuronal damage beyond the primary focal lesion. Brain 2001;124
(Pt 10):2014–27.
99. Ramlackhansingh AF, Brooks DJ, Greenwood RJ, et al. Inflammation after
trauma: microglial activation and traumatic brain injury. Ann Neurol
2011;70(3):374–83.
100. Landt J, D’Abrera JC, Holland AJ, et al. Using positron emission
tomography and carbon 11-labeled Pittsburgh compound B to image brain
fibrillar β-amyloid in adults with down syndrome: safety, acceptability, and
feasibility. Arch Neurol 2011;68(7):890–6. Epub 2011, Mar 14.
101. Vallabhajosula S. Positron emission tomography radiopharmaceuticals for
imaging brain beta-amyloid. Semin Nucl Med 2011;41(4):283–99.
doi:10.1053/j.semnuclmed.2011.02.005.
102. Barthel H, Gertz HJ, Dresel S, et al. Cerebral amyloid-β PET with
florbetaben (18F) in patients with Alzheimer’s disease and healthy controls:
a multicentre phase 2 diagnostic study. Lancet Neurol 2011;10(5):424–35.
Epub 2011, Apr 8.
103. Koivunen J, Scheinin N, Virta JR, et al. Amyloid PET imaging in patients
with mild cognitive impairment: a 2-year follow-up study. Neurology
2011;76(12):1085–90. Epub 2011, Feb 16.
Section IV—Radiology 298.e3
104. Yu EY, Mankoff DA. Positron emission tomography imaging as a cancer
biomarker. Expert Rev Mol Diagn 2007;7(5):659–72.
105. Harris RS, Schuster DP. Visualizing lung function with positron emission
tomography. J Appl Physiol 2007;102(1):448–58. Epub 2006, Oct 12.
106. Chen Q, Liu W, Li H, et al. Molecular imaging in patients with mood
disorders: a review of PET findings. Eur J Nucl Med Mol Imaging
2011;38(7):1367–80. Epub 2011, Apr 12.
107. Brooks DJ, Pavese N. Imaging biomarkers in Parkinson’s disease. Prog
Neurobiol 2011;Aug 30. Epub ahead of print.
108. Sacher J, Neumann J, Fünfstück T, et al. Mapping the depressed brain: a
meta-analysis of structural and functional alterations in major depressive
disorder. J Affect Disord 2011;Sep 2. Epub ahead of print.
109. Melrose RJ, Ettenhofer ML, Harwood D, et al. Cerebral metabolism,
cognition, and functional abilities in Alzheimer disease. J Geriatr Psychiatry
Neurol 2011;24(3):127–34.
110. Yakushev I, Schreckenberger M, Müller MJ, et al. Functional implications of
hippocampal degeneration in early Alzheimer’s disease: a combined DTI
and PET study. Eur J Nucl Med Mol Imaging 2011;Jul 27. Epub ahead of
print.
111. Kawai N, Maeda Y, Kudomi N, et al. Correlation of biological aggressiveness
assessed by 11C-methionine PET and hypoxic burden assessed by
18F-fluoromisonidazole PET in newly diagnosed glioblastoma. Eur J Nucl
Med Mol Imaging 2011;38(3):441–50. Epub 2010, Nov 12.
V
Chapter
30  
Neurosonology: Transcranial
Doppler and Transcranial
Color-Coded Duplex
Sonography
Jaroslaw Krejza and Michal Arkuszewski
Introduction
Doppler ultrasonography is a technique to measure the veloc-
ity and pulsatility of blood flow within the intracranial and
the extracranial arteries. Transcranial Doppler (TCD) ultraso-
nography is frequently used for the clinical evaluation of cere-
bral vasospasm following subarachnoid hemorrhage (SAH),
to evaluate cerebral autoregulation (CA), detect major arterial
occlusion, monitor recanalization in acute stroke, monitor
cerebral circulation during major cardiovascular procedures,
and assess hemodynamics in patients with intracranial hyper-
tension among other indications.1,2 Technologic advances
such as transcranial color-coded duplex sonography (TCCS)
have made TCD more reliable and accurate in its detection of
hemodynamic abnormalities. Newer applications of this tech-
nology include therapeutic use of ultrasound in the acute
stroke setting.
Technical Aspects of Transcranial
Doppler
Basics of Ultrasound
Ultrasonic waves that enter human tissue are transmitted,
absorbed, reflected, and scattered.3,4 Transmission properties
of a tissue depend on its density and elasticity. Density and
speed of propagation of ultrasound (US) waves determine a
tissue’s acoustic impedance. The larger the difference in acous-
tic impedance between tissues, the more US waves are reflected.
Reflection further depends on the angle of insonation, and
stronger echoes are received when the angle of insonation is
zero. Structures imaged by US B-mode, the two-dimensional
visualization of the internal organs and structures on the gray-
scale image, are displayed proportionally to the intensity of
returning echoes.3,4
300
Transcranial Doppler and TCCS Techniques
Structures imaged by B-mode are displayed proportionally to
the intensity of returning echoes, whereas flow velocity is
detected in Doppler mode if there is a difference in frequency
between emitted and returning echoes.3 The magnitude of the
frequency shift (ΔF) depends on the ultrasound transmission
velocity in the insonated tissue (C), the relative velocities of
the reflector (blood, V), and the frequency of the source (Fo).
The observed frequency shift (ΔF) is ΔF = 2 VFo/C.3,5 The shift
is measured only for that component of motion occurring
along the axis of the ultrasound beam. Therefore absolute
velocity measurements require that a correction be made for
the angle (θ) between the vessel and the beam as follows: V =
ΔFC/(2 Fo cos θ). In pulsed wave Doppler, a transducer gener-
ates ultrasound pulses and detects returning echoes. Assuming
that the speed of transmission of ultrasound in human tissues
is a constant, the time delay between the emitted pulse and
the returning echo enables the sampled structure’s depth to
be determined. The pulse’s duration and repetition frequency
impose limits on the maximum velocity that can be measured.
Anatomy is not displayed, however. Identification of intracra-
nial arteries in conventional TCD relies therefore on the a
priori defined depths of the sample volume and the direction
of flow.
Duplex Doppler imaging combines pulsed-wave Doppler
with two-dimensional, real-time B-mode imaging, which
allows precise placement of the sample volume in the vessel.
Optimal angle correction for velocity calculations can be per-
formed, as the course of the vessel in relation to the US beam
is visually depicted. Color-coded duplex sonography is the
most commonly used technology today for carotid imaging
and TCCS (Fig. 30.1). Red indicates blood flow toward the
transducer, whereas blue represents flow away from the trans-
ducer. High-flow velocities are depicted with increasing
© Copyright 2013 Elsevier Inc. All rights reserved.
 Section V—Cerebral Blood Flow 301

Sonographic
M2 MCA A1 ACA M1 MCA probe Temporal
M2 MCA window
M1 MCA P2 PCA
A2 ACA

P2 PCA
A B
Fig. 30.1  A, A typical semiaxial image of M1 and M2 segments of the right MCA superimposed on a sector-shaped conventional B-mode image.
The sample volume is precisely placed on a green color spot related to an aliasing artifact, which indicates the site of highest flow acceleration in the
M1 segment. The angle between the course of the vessel in relation to the ultrasound beam is measured by an electronic cursor. This approach allows
the angle-corrected flow velocity measurements from the waveform displayed below the B-mode image to be obtained. B, The complex spatial
relationships between ultrasound beam, courses of M1 and M2 MCA, A2 ACA and P2 PCA, and the site of temporal window. ACA, Anterior cerebral
artery; MCA, middle cerebral artery; PCA, posterior cerebral artery.

brightness, thus the presence, direction, and flow disturbances
can be quickly identified.
The insonation angle is important in flow velocity mea-
surements with TCCS because trajectory of a particular vessel
in the intracranial space is unpredictable.6,7 Thus if the veloc-
ity measurements are adjusted for the angle, the velocity
values are closer to the true values because the vessel’s geo-
metric factor is accounted for. If a measured velocity is
170 cm/second and the angle is 20 degrees, the true velocity
is 181 cm/second; however, if the angle is 40 degrees, the true
velocity is 222 cm/second. Such difference, caused by errone-
ous flow velocity measurement, can change management of a
patient. TCD/TCCS cannot be performed in about 11% of
patients because the acoustic window in the temporal bone is
weak or absent.8 Also following a pterional craniotomy the
acoustic window may be degraded because of subtemporal
bone work or the presence of Gelfoam in the epidural space.
Frequencies used for neurosonology range between 1 and
3.5 MHz for TCD/TCCS and 7.5 and 16 MHz for carotid
imaging.
Bioeffects and Safety of Ultrasound
Diagnostic US can produce heat that may be hazardous to
sensitive organs.9 Nonthermal effects, such as pressure changes
and mechanical disturbances, in tissue have not been demon-
strated in humans.10 US used in therapy, however, can cause
both substantial temperature increase and mechanical damage
in the tissue.10-12
Thermal Effects
Denser tissue absorbs more heat from US, thus fluid does not
heat very much, whereas bone heats the most, and on bone
surface the heat accumulation can be up to 50 times faster
than in soft tissues. This heating effect can be of importance,
particularly with use of long-term TCD monitoring and its
therapeutic applications. The thermal index (TI), provided in
US machines estimates the risk from heat. When the TI is
greater than 1, the risks of US should be weighed against the
benefits.10 The exposure has to be minimized despite reports
that short-term monitoring does not increase temperature at
the temporal window in vivo.13,14
Nonthermal Effects
US can also produce various mechanical effects such as cavita-
tion, pressure amplitude, force, torque, and acoustic stream-
ing.9,15 Cavitation occurs when US passes through an area that
contains a cavity, such as a gas bubble. US can cause the bubble
to expand and contract rhythmically. When bubbles pulsate,
they send secondary US waves in all directions. These second-
ary waves can actually improve US imaging. If the bubbles
contract toward the point of collapsing, they can build up very
high temperatures and pressures for a few tens of nanosec-
onds. These high temperatures and pressures can produce free
radicals and other toxic compounds that, although unlikely,
could theoretically cause genetic damage.16 The rapid con­
traction of bubbles can also cause microjets of liquid that
can damage cells. The safety guidelines for diagnostic US
are designed to prevent cavitations to occur.9,10 US-induced
changes in pressure, force, torque, and streaming, in turn, can
cause audible sounds, electrical changes in cell membranes
that make them more permeable to large molecules, move-
ment and redistribution of cells in liquid, and cell damage.17
In the experimental setting using promyelocytic leukemia
cells, sonoporation may lead to antiproliferation effects
including cell-cycle arrest and apoptosis through disrupting
various cell signaling pathways.18 In liquids, US causes a type
of stirring action called acoustic streaming. As the acoustic
pressure of US increases, the flow of liquid speeds up. When
the streaming liquid comes near a solid object, shearing may
occur, which can damage platelets and lead to abnormal blood
clotting.
302 Section V—Cerebral Blood Flow

resulting decrease in extracellular pH causes vasodilation;

Effects of Ultrasound Contrast Agents
These agents usually take the form of stable gas-filled micro-
bubbles (MBs) that can produce cavitations or microstream-
ing, the risk of which increases with the mechanical index
value. Additionally, a radiation force produced by the acoustic
wave, considered as noncavitational and nonthermal mecha-
nisms of ultrasonic bioeffects, may act on particles and
microbubbles.19
Cerebral Vascular Reactivity and
Transcranial Ultrasound Testing of
Cerebral Blood Flow Regulation
Cerebral Autoregulation
CA is the ability of brain microvasculature to maintain cere-
bral blood flow (CBF) relatively constant despite wide varia-
tions in cerebral perfusion pressure (CPP). In the healthy
state CA keeps CBF relatively constant within the range of
mean arterial blood pressure (MAP) between 60 and 150 mm
Hg.20 However, the upper and lower limits of CA can be
shifted up or down by endogenous as well as exogenous
factors. Sympathetic nervous system stimulation and angio-
tensin II can shift the upper and lower limits of CA up toward
higher pressures, whereas antihypertensive medications have
opposite effects.21,22 Patients who have untreated hypertension
have limits of regulation set at a higher level compared with
healthy people21; thus overzealous antihypertensive treatment
may lead to a dangerous CBF reduction at a relatively high
MAP. The mechanisms of CA are complex and not fully
understood and appear to be maintained by three different
control pathways: vasogenic, metabolic, and neurogenic.23
The vasogenic mechanism of CA is based on the intrinsic
ability of cerebral vessels to respond to changes in the wall’s
shear stress and transluminal pressure.24 An increase in
transluminal pressure activates vascular smooth muscle
cells, leading to vasoconstriction,25,26 whereas increased shear
stress27 triggers vasodilation.24 In metabolic regulation, cere-
brovascular resistance (CVR) is modified by PaCO2, PaO2, and
release of adenosine and potassium ions from neurons in
response to insufficient blood supply.28 Hypercapnia and the
hypocapnia leads to vasoconstriction and CBF decrease. The
importance of neurogenic regulation in the control of CBF is
a matter of debate. The cerebral vessels are innervated by an
extrinsic system such as nerve fibers that originate in ganglia
belonging to sympathetic, parasympathetic, and sensory
ganglia, and an intrinsic system that is composed from nerves
originating in different parts of brain such as the nucleus
basalis, locus coeruleus, and raphe nucleus.29,30 Sympathetic
stimulation constricts large cerebral arteries, but CBF does not
decrease because of immediately compensated vasodilation
of resistance arterioles.31 Sympathetic innervation seems to
protect the brain against MAP increases because sympathetic
activation shifts the upper and lower limits of CA toward a
higher MAP. Activation of the parasympathetic vascular nerve
fibers causes vasodilation.32 Vascular fibers belonging to the
intrinsic system33-35 may modulate vascular tone directly or
through stimulation of perivascular interneurons and glial
cells.36,37 However, the significance of this system is being
elucidated.34,35
Assessment of Cerebral Blood Flow
Regulation with Ultrasound
The CA can be evaluated by measuring relative blood flow
changes in response to the change in the MAP using “static”
or “dynamic” approaches.2,38-41 In the static method, CA effi-
ciency (i.e., the overall change in cerebral arterial resistance
[CAR] induced by changes in MAP) is assessed under steady-
state conditions. In this method the first cerebral blood flow
velocity (CBFV) measurement obtained at a constant baseline
MAP is followed by another steady-state measurement that is
taken after the CAR to a change in MAP. CA functions prop-
erly if the MAP change within the lower and upper limits of
CA, estimated for healthy humans, does not affect CBFV.
Dynamic methods use rapid MAP changes and analyze the
instantaneous CBFV in the process.38 Systemic hemodynamic
changes may be induced using rapid leg cuff deflation, pro-
gressive lower body negative pressure, the Valsalva maneuver,
deep breathing, ergometric exercise, or head-down tilting
(Fig. 30.2). Dynamic methods are able to assess both efficiency
and latency of the CA response, that is, the overall change in

A B

Fig. 30.2  Transcranial Doppler (TCD) tracings that illustrate dynamic autoregulation testing based on flow velocity measurements in the
middle cerebral artery during head-down tilting in a healthy volunteer. A, In the supine position mean blood flow velocity is 70 cm/second;
pulsatility index is 0.80; systolic and diastolic blood pressure are 120 mm Hg and 78 mm Hg, respectively; and heart rate is 68 beats per minute. B, While
the patient is in an upright position, mean blood flow velocity decreased to 62 cm/second; the pulsatility index increased to 1.05; blood pressure
changed to 125 mm Hg and 70 mm Hg; and heart rate increased to 75 beats per minute. (Courtesy the Department of Neurology, Ernst Moritz Arndt University
of Greifswald, Germany.)
 Section V—Cerebral Blood Flow 303

CAR and the time in which the CAR change is achieved. The
use of systemic hypotension to challenge the CA response is
limited, however, because pharmacologically induced hypo-
tension could result in ischemic injury in patients with already
insufficient CBF.
Two commonly used, but somewhat less dynamic methods
of physiologic challenge of CA that carry much less risk of
ischemia include administration of acetazolamide and sys-
temic PaCO2 manipulations. Acetazolamide is a carbonic
anhydrase inhibitor that acts as a potent vasodilator when
administered intravenously in a dose of 15 to 18 mg/kg
of body weight (standard dose 1000 mg).42 Acetazolamide
slowly penetrates the blood-brain barrier, where it reversibly
inhibits carbonic anhydrase that catalyzes the conversion of
bicarbonate and hydrogen ion to water and carbon dioxide
(CO2). Acetazolamide decreases the production of bicarbon-
ate and results in a decrease in the extracellular pH in the
brain.43,44 This induced acidosis results in vasodilation,45,46 and
CBF increase,46 whereas MAP, heart rate (HR), and respiratory
rate are unaffected.47 The effect on CBF may be seen within
the first 5 minutes after administration.48 In healthy subjects,
peak CBF augmentation of about 30% to 60% occurs at
approximately 10 minutes after bolus intravenous administra-
tion and changes little over the next 20 minutes (Fig. 30.3).49
This response can be blunted in an older population.50 The
percentage of increase in CBF after acetazolamide administra-
tion may be used to define cerebrovascular reserve as follows:
CVR = (CBFpost − CBFpre )/CBFpreacetazolamide × 100%
Although a global, symmetric increase in CBF by about
30% is accepted as normal, a variety of criteria are used to
define an abnormal response to acetazolamide, including: (1)
greater than 5% decrement in absolute CBF, (2) a less than
10% increment in absolute CBF, (3) an absolute change of less
than 10 mL/100 g/minute, and (4) a value greater than two
standard deviations below control values. Acetazolamide is
generally well tolerated in a dose less than 1000 mg. The most
commonly reported side effects include transient circumoral
numbness, paresthesias, and headaches.47
Manipulation of systemic partial CO2 pressure (PaCO2)
also can be used to test CVR. Low-level hypercarbia results
in a reproducible CBF increase between 0.01 and 0.02 mL/g/
min for each 1 mm Hg PaCO2 increase. The effect is rapid
and quickly reversible.51 CO2 manipulation is performed
using several techniques, including: (1) breath holding or
hyperventilation, (2) re-breathing, or (3) inhalation of 3% to
5% CO2. The effect of altered CO2 tension on CBF is quanti-
fied either as a fixed effect for a particular CO2 manipulation
(similar to the quantification of acetazolamide effect) or as
the slope of the CO2-CBF relationship according to the fol-
lowing formula52:
CVR = [(CBFpost-CO2 − CBFpre-CO2 )/(CBFpre-CO2 )] ×
[100/(Pa pre-CO2 – Pa post-CO2 )]
The combined use of breath holding and hyperventilation
tests allows assessment of the full range of vasomotor reactivity
(VMRr), both during hypercarbia (vasodilation) and hypo­
carbia (vasoconstriction). The relative difference between
minimal and maximal CBF defines VMRr as follows (Fig. 30.4):
VMRr = [(breath holding CBF − hyperventilation CBF)/
rest CBF] × 100%

A B
Fig. 30.3  Blood flow velocity measurements taken from the right middle cerebral artery with transcranial color-coded duplex sonography before
(A) and after (B) administration of 1000 mg acetazolamide in a healthy volunteer. The studies show healthy autoregulation expressed by a significant
(71%) increase in the mean flow velocity.

MCA velocity at rest MCA velocity after MCA velocity after

A B hyperventilation C breath holding

Fig. 30.4  Transcranial Doppler (TCD) tracings that show vasomotor reactivity range (78%) in a healthy volunteer determined by TCD velocity
measurements in the middle cerebral artery (MCA) (A) at rest mean blood flow velocity (BFV) is 60 cm/second, (B) after hyperventilation mean BFV
is 40 cm/second, and (C) after breath holding, mean BFV is 86 cm/second. (Courtesy Dr. G. Kozyra, the Department of Neurology, Medical University of Gdańsk,
Poland.)
304 Section V—Cerebral Blood Flow
The CBF increase in hypercarbia also can by described as a
breath holding index (BHI):
BHI = [(breath holding CBF − rest CBF)/rest CBF)]/
time of breath holding CBF] ×100%
In clinical practice, the magnitude of CBF changes with CO2
manipulation is smaller than that achieved with acetazol-
amide. The average increase in CVR is 1.1% to 2.9%. With 5%
CO2 inhalation, the CBF response observed usually is approxi-
mately half of the response observed with acetazolamide.
However, the correlation between CBF changes produced by
each technique varies from poor to moderate.53-55 CO2 manip-
ulation also may itself alter systemic MAP; an increase of about
10 mm Hg that may be induced with 5% CO2 then may obfus-
cate evaluation of the CO2-mediated CBF response.53 However,
CO2 manipulation is easily performed, generally well toler-
ated, and has no long-term effects. The rapid response to CO2
is particularly suitable for TCD or TCCS measurements.
An important phenomenon that can affect the results of CA
testing is that of “steal,”49 which is created by the interaction
between regions of different cerebrovascular response and the
systemic effects of the agents used for physiologic or pharma-
cologic challenge. Areas that have normal CA will show vaso-
dilation, lowering the perfusion pressure in the larger blood
vessels supplying both normal and abnormal areas. This lower
perfusion pressure results in a decrease in blood flow in the
regions that are already maximally dilated.
Assessment of CA Functionality with TCD
and TCCS
Implementation of TCD/TCCS is based on the premise that
CBF changes will be reflected only by changes in CBFV in
major brain arteries. If CA is intact, changes in MAP do not
affect CBFV, because of compensatory change in peripheral
arterial resistance. Static measures of CA are the correlation
coefficient (r) between CBFV and mean MAP as well as the
“index of static autoregulation”(sARi), which is a ratio of
percentage of change in CVR to percentage of change in
MAP (sARi = % CVR/% MAP).56,57 For perfect CA the r
would equal 0 and sARi would be 1, whereas complete lack of
CA would yield an r and sARi of 1 and 0, respectively. The
most commonly proposed threshold between normal and
impaired CA for both r and sARi is 0.5.56,58 Evaluation of CA
in dynamic methods is based on a “dynamic autoregulation
index” (dARi). This is the dynamic equivalent of sARi and is
defined as dARi equals (ΔCVR/ΔT)/ ΔMAP, where ΔT is the
time when the CA response occurs.57,59 Thus dARi not only
evaluates the overall change in CBFV induced by a change in
MAP, but also quantifies the time dependence of a CA
response. A value of dARi that equals 0 represents the absence
of CA, whereas a value of 9 corresponds to an efficient CA
response. The proposed threshold between normal and
impaired CA is 5.57,60 Static and dynamic methods of evalua-
tion of CA yield similar results.57
In the transient hyperemic response (THR) test, a brief
compression (3-10 seconds) of the ipsilateral common
carotid artery (CCA) results in a sudden reduction in the
middle cerebral artery (MCA) velocity due to distal vasodila-
tion, if CA is intact. Subsequently a transient increase in the
MCA velocity is seen on release of the compression. Two
CA indices have been described: THRR, which is the ratio
between the MCA velocity after release of compression, and
the velocity before onset of compression and SA, which is
the strength of CA calculated by normalizing the THRR for
changes in the MCA CBFV at the onset of compression. In
theory, the THR test assesses both the gradient and the limits
of the CA plateau without differentiating between the two.
The test has been used to study anesthetic agents and in crit-
ical care. The coefficient of variation less than 10% is much
lower than other CA tests, making it suitable for compari-
sons. The main advantages are reproducibility, simplicity,
and lack of pharmacologic intervention. There is a very small
risk of detaching carotid artery atheroma, but various studies
have demonstrated a good safety record even in at-risk
patients.
Influence of Anesthetic Agents
on Intracranial Blood Velocity
Induction Agents
Intravenous (IV) induction agents depress cerebral metabolic
rate of oxygen (CMRO2) without affecting coupling between
flow and metabolism, thus CBF and subsequently flow veloc-
ity in the MCA decreases in a dose-dependent fashion.61,62
Propofol has little effect on CA or cerebrovascular reactivity
to carbon dioxide (CRCO2), although hypercapnia may impair
CA at around 8 kPa.63-65 Ketamine tends to increase MCA
velocity because of an “excitatory effect”66; however, when the
effect is blunted by preexisting anesthesia, ketamine appears
to decrease MCA velocity. Volatile anesthetic agents can
decrease MCA velocity by reducing CMRO2, but also can
increase velocity by direct vasodilation.67-69 The net effect
depends largely on the agent and the dose used. When equi-
potent clinical doses are used, sevoflurane has little effect on
MCA velocity and CA; desflurane has the greatest effect, and
isoflurane an intermediate effect.70-74 It should be noted that
it is a dose-related phenomenon.
Nitrous oxide at concentrations of 30% to 60% increases
flow velocity in MCA and CBF.64,75 It may also disrupt CA
when added to volatile anesthetic agents or propofol.76,77 This
has been thought to be a result of excitation associated with
low dose anesthesia. However Matta et al64 observed increased
MCA velocity when nitrous was added to isoelectric EEG
doses of propofol. Muscle relaxants appear not to change flow
velocity in MCA in patients with neurologic injury.78,79
Opioids at moderate doses have no significant effect on MCA
velocity; however, at high doses they reduce flow velocity by a
presumed central mechanism80,81 unless they cause seizures,
which would presumably be associated with hyperemia. Vaso-
active agents such as norepinephrine,82 dobutamine, and
dopexamine83 have been shown to have no effect on CA and
CRCO2 in volunteers and in patients anaesthetized with iso-
flurane and propofol.84 Glyceryl trinitrate appears not to
affect CA.85
Limitations of TCD and TCCS
TCD measures velocities in large arteries, thus mapping of
regional impairment within the brain is not possible. Regional
areas of impaired CA in an artery distribution may be missed
when averaged with areas that are better perfused. Further-
more, correlation between flow velocities and CBF is weak in
 Section V—Cerebral Blood Flow 305

some patients, mostly because of collateral circulation.86,87
Assumptions about the changes in flow only hold if the arte-
rial diameter and angle of insonation remains constant.
Applications of Transcranial
Doppler Ultrasonography in
Neurocritical Care Monitoring
Diagnosis and Monitoring
of Cerebral Vasospasm
Cerebral vasospasm (VSP) contributes to morbidity and mor-
tality among patients after SAH.88 Medical treatment of VSP
with induced hypertension, hemodilution, and iso- or hyper-
volemia (3-H) can help improve CBF. However, use of these
therapies can be associated with increased cerebral edema or
hemorrhagic transformation in areas of infarction, congestive
heart failure, or pulmonary edema and so are best used when
the diagnosis of VSP is made rather than prophylactically.88,89
When medical treatment is unsuccessful, endovascular meth-
ods (e.g., intra-arterial infusion of vasodilative agents, or bal-
loon angioplasty) may restore flow. The outcome of patients
who develop VSP and in particular delayed cerebral ischemia
(DCI) associated with VSP is improved with early treatment.
However, when to initiate various interventions often may be
uncertain because the diagnosis and monitoring of VSP may
be difficult and unreliable when based solely on the neurologic
examination. Furthermore, other complications common in
SAH patients, such as recurrent hemorrhage, hydrocephalus,
metabolic disorders, and seizures can produce similar neuro-
logic symptoms.88
Digital subtraction angiography (DSA) is the gold standard
test to diagnose cerebral VSP, but this method is an impractical
monitoring tool because it is invasive and carries a small risk
of stroke, renal injury, and other complications.90 Other
methods, such as computed tomography (CT) and magnetic
resonance angiography, xenon-enhanced computed tomogra-
phy (Xe-CT), and CT perfusion are discussed in Chapters 26
through 28. These techniques may provide useful information
but there are practical limitations to their daily use as a moni-
toring tool.91,92
The knowledge that blood flow velocity is increased in a
constricted vessel has led to the use of TCD to detect and
monitor cerebral VSP at bedside.2, 93 However, TCD studies are
operator dependent, and the accuracy of TCD may differ from
one study to another. These variations have raised concerns
about the utility of TCD to reliably detect VSP, particularly in
vessels other than MCA.94,95 Generally mild or severe vaso-
spasm in the MCA can be consistently diagnosed, and the
diagnostic accuracy can be improved using TCCS rather than
TCD.95-99 When using TCCS, 80% of DSA-diagnosed cases of
mild narrowing and 92% of cases of severe MCA narrowing
can be properly classified. Age and sex correction of flow
velocity further increases TCCS performance, especially for
the diagnosis of less severe MCA spasm.96 The improved accu-
racy of TCCS is associated with its ability to image arteries,

A B

C D
Fig. 30.5  Transcranial color-coded duplex sonography (TCCS) and transcranial Doppler (TCD) tracings that show the importance of the
angle of insonation, effect of proper placement of the sample volume on flow velocity measurements, and diagnostic value of the ratio
of intracranial middle cerebral artery (MCA) to the ipsilateral extracranial internal carotid artery in a subarachnoid hemorrhage (SAH)
patient. A, The sample volume is placed in the M2 segment of the right MCA at a distance of 34 mm from the probe and an angle of 6 degrees;
the mean velocity is 217 cm/second. B, The sample is placed at the distance of 36 mm in the same artery segment, and the angle is 43 degrees; the
velocity obtained is 393 cm/second, an increase of 181%. In both images the sample is placed on the site of aliasing artifact indicated by blue. With
conventional TCD the change in blood flow velocity may be missed if the sample is placed at the distance of 45 to 55 mm, usually used in MCA
interrogation. C, The mean flow velocity in the extracranial internal carotid artery is 27 cm/second; thus the velocity ratios are 8.04, and 14.56,
respectively for A and B measurements. These findings are consistent with severe vasospasm identified in the M2 segment of the MCA during
angiography (D, solid arrow). There also is severe vasospasm of A1 segment of the ACA (dashed arrow).
306 Section V—Cerebral Blood Flow
to position the sample volume in the specific site of the
investigated artery, and to determine the angle between the
artery and the US beam (Fig. 30.5).99,100 Unlike the MCA,
the accuracy of TCD in the anterior cerebral artery (ACA),
posterior cerebral artery (PCA), internal carotid artery (ICA),
basilar artery (BA), and vertebral artery (VA) is not as well
studied and accuracy appears low.
TCD or TCCS evaluations are valuable in the day-to-day
care of SAH patients to detect VSP and evaluate therapeutic
effects.101-103 A baseline study should be obtained at admission
when the probability of large artery spasm is still relatively low,
to establish reference values to which subsequent measure-
ments can be compared and then be repeated daily, par­
ticularly during the period of high risk for VSP. Serial
measurements are useful because a rapid increase in flow
velocity (>65 cm/second over 24 hours) is associated with a
poorer outcome, and is suggested as an indication to consider
induced hypertension (i.e., symptomatic vasospasm or DCI is
likely). In these patients the velocity ratios (VMCA : VICA,
VtICA : VICA), or Lindegaard ratio that compares intra- to extra-
cranial blood flow velocities, should be calculated because the
velocity ratios are not altered by the therapy and so reflect
arterial narrowing. A ratio of less than 3 is rarely found in
patients with VSP, and ratios of greater than 6 may distinguish
moderate from severe MCA VSP.104 Several factors including
age and sex, course of the artery, increased ICP and CPP, MAP,
hematocrit (Hct), PaCO2, temperature, presence and pattern
of collateral flow, therapeutic interventions (e.g., hypervol-
emia or use of vasopressors with impaired CA), presence of
impaired CA, and multiple segment narrowing influence flow
velocities and should be considered. For example, correction
for PaCO2 and Hct may be required if artificial ventilation or
hemodilution is in progress, and when interpreting Doppler
results, global and local ICP and CA state should be taken into
account. High impedance indices may suggest the presence of
localized or generalized increased ICP, and usually are associ-
ated with poorer outcome.98 CA also may be impaired in many
patients with VSP (Fig. 30.6).105,106 Finally, the pulsatility index
(PI), which provides information about distal vascular
Breath
holding
30 sec
Vasospasm
Control
Fig. 30.6  Transcranial Doppler (TCD) waveforms from the middle
cerebral artery obtained from a patient with vasospasm after subarachnoid
hemorrhage (SAH) (upper panel), and a healthy volunteer (lower panel)
before and after breath holding. In the SAH patient mean blood flow
velocity (BFV) increased by 5 cm/second to 175 cm/second after 30
seconds of breath holding, which is consistent with impaired
autoregulation (breath holding index = 0.01). In the healthy subject
BFV increased by 30 cm/second to 88 cm/second (breath holding index
= 1.72). (Courtesy the Department of Neurology, Ernst Moritz Arndt University of
Greifswald, Germany.)
resistance, may be low when there is symptomatic large vessel
vasospasm (i.e., there is compensatory distal vasodilation for
the hypoperfusion).107
Middle Cerebral Artery Vasospasm
The accuracy of TCD and TCCS to detect VSP is best deter-
mined for MCA; TCCS perhaps shows better agreement with
DSA than TCD.95-99,108 A mean blood flow velocity (BFV)
greater than 200 cm/second, a rapid rise in flow velocities
(>50 cm/second/day), or Lindegaard ratio (VMCA : VICA) greater
than 6 indicate severe arterial narrowing.94,109,110 VSP is unlikely
if the mean BFV is less than 120 cm/second. However, about
60% of patients have velocities that fall between these two
values. In addition, negative results do not exclude the pres-
ence of VSP. The best Doppler parameter is peak-systolic
velocity, and a threshold of 182 cm/second corresponds to
maximal diagnostic accuracy. TCCS efficiency, sensitivity,
specificity, positive predictive value (PPV), and negative pre-
dictive value (NPV) are reported to be 92%, 86%, 93%, 73%,
and 97%, respectively.95 The VMCA : VICA ratio (Lindegaard
index104) can help differentiate patients with high intracranial
flow velocities associated with hyperemia from those with a
BFV increase from VSP (see Fig. 30.5). Studies suggest that the
overall accuracy of the VMCA : VICA ratio in the diagnosis of mild
and moderate to severe MCA narrowing is better than the
respective accuracy of velocity measurements alone.98 A value
3.6 of the ratio appears to be threshold to diagnose mild (up
to 25% narrowing) spasm in the M1 segment of the MCA,
whereas a threshold of 4.4 indicates moderate to severe spasm
(>25% artery narrowing).98 The thresholds are higher than the
upper normal reference limits of the VMCA : VICA ratio, calcu-
lated on the basis of the mean velocity.111 Neural networks can
be used to improve TCCS diagnostic performance; classifica-
tion accuracy is 92% in moderate to severe spasm detection,
and 87% in spasm of other grades.112
Vasospasm of Other Arteries
TCD-based diagnosis for vessels other than the MCA is less
well defined and generally less accurate.108 Visualization of the
ACA with TCD or TCCS is more difficult than that of the
MCA.6 TCD studies based on relatively small sample sizes
show a sensitivity of 42% and specificity of 76% when 120 cm/
second mean velocity is used as a threshold to detect ACA
spasm. A TCCS threshold of 75 cm/second mean BFV is
associated with 71% sensitivity and 85% specificity.97 False-
negative results for the ACA may be explained by collateral
flow through the anterior communicating artery (ACoA) and
difficulties with angiographic differentiation of ACA hypopla-
sia from vasospasm. The VACA : VICA ratio can be helpful to
differentiate spasm from the normal artery; the ratio VACA : VICA
varies between 0.54 and 2.55.104,111 In practice, however, a diag-
nosis of unilateral ACA spasm is not always a necessity because
the hemodynamic consequences may be reduced if the ACoA
is patent. However, bilateral ACA spasm may reduce flow to
the distal ACA segments. In these patients TCCS can detect
increases in velocity that involve at least one artery.
There are few published data on TCD diagnosis of ICA,
PCA, VA, and BA spasm.94, 113 TCD sensitivity is reported to
be 25%, 48%, 44%, and 77%, respectively, whereas specificity
is 91%, 69%, 88%, and 79%, respectively.94 The sensitivity

(100%) and specificity (95%) for BA spasm are greater when
the mean BFV in the BA is compared with the extracranial VA
greater than 2 is considered a threshold for spasm.113 Normal
reference ranges of the velocity ratio VPCA : VVA (0.76-2.90) also
can help interpret abnormal velocity results in the posterior
circulation.111 TCD is not useful to detect VSP in more distal
branches of intracranial arteries.114,115 Data on TCCS in this
context are lacking.
Ischemic Cerebrovascular Disease
Acute Stroke
TCD and TCCS can be used in acute stroke and to evaluate
recanalization therapies as a tool to assess occlusive disease in
the main segments of the intracranial arteries, to evaluate early
recanalization, and to guide therapeutic decisions.2,116-122 The
Thrombolyis in Brain Ischemia (TIBI) scores based on TCD
findings can be used to assess initial hemodynamics in acute
stroke and early recanalization.123-125 This scale has six levels
of hemodynamic abnormalities; grades 0 to 3 are associated
with worse outcome,126 whereas grades 4 and 5 (i.e., normal
flow and hyperperfusion) are associated with favorable
outcome. Because TIBI grades are frequently dichotomized as
grades 0-3 or grades 4-5, a Consensus on Grading Intracranial
Flow (COGIF) obstruction score is suggested to better reflect
hemodynamic changes based on Doppler findings.
 Grade 1. This indicates no flow and corresponds to an
occlusion of the M1 segment of the MCA or the carotid
bifurcation. The main diagnostic criterion is the absence
of a color Doppler flow signal and Doppler spectrum in
the location of the proximal MCA segment.127-129 To make
a reliable diagnosis requires sufficient visibility of other
arteries (e.g., the ACA A1-segment, C1-segment of intra-
cranial ICA, ipsilateral A2 or of the contralateral anterior
circulation) or veins (deep middle cerebral vein) of the
anterior circulation. When insonation conditions are
unsatisfactory and a reliable diagnosis is not possible, use
of an US contrast agent is required.
 Grade 2. This represents low flow in the M1 segment and
may be found in (1) an upstream carotid obstruction that
reduces downstream flow in the MCA, (2) intracranial
ICA disease with compensatory flow diversion to the con-
tralateral hemisphere and an increase in the asymmetry
index,130,131 (3) partial recanalization of the M1 segment
of the MCA or the carotid bifurcation, and (4) down-
stream obstruction (i.e., MCA branch occlusion), when
the asymmetry index is high. Upstream and downstream
obstruction have to be differentiated from partial M1
recanalization.
 Grade 3. This represents High-flow velocities. A focal
increase in velocity indicates intracranial stenosis (grade
3A), whereas increased flow velocities along the complete
arterial segment indicate hyperperfusion (grade 3B) and
may be found in the initial stages of recanalization or with
collateral flow across the involved artery. The degree of
intracranial stenosis may be incorrectly estimated when
there is hyperperfusion.
 Grade 4. Normal flow (i.e., flow velocities) is within the
normal reference range.
Follow-up studies can be classified as (1) reflow—partial
recanalization is assumed when there is an improvement by
Section V—Cerebral Blood Flow 307
one or more COGIF and complete recanalization when flow
within a previously occluded vessel is now grade 3B or 4;
(2) no change—the grade observed at baseline persists; and
(3) worsening when the measured flow decreases one or more
grades.
In acute stroke TCCS also can be used to examine the posi-
tion of the third ventricle and a potential midline shift.117,132-134
Intracerebral hemorrhage, including hemorrhagic transfor-
mation of an infarct, aneurysms, and arteriovenous malfor-
mations occasionally may be detected by TCCS,117,135-141
although it is not the preferred imaging method for these
conditions. TCD or TCCS can be used to assess CA in the
ischemic hemisphere; it often is impaired even in minor
stroke.142 When CA is impaired, less favorable outcome is
observed after recombinant tissue-plasminogen activator
(rtPA) administration.143,144 Similarly, impaired CA ipsilateral
to an acute ischemic stroke is associated with larger
infarction.145
Chronic Hypoperfusion
Focal arterial stenosis disrupts the laminar flow pattern in
diastole. This is associated with flow instability and turbulence
with an increase in the range of flow velocities known as
spectral broadening.3 As the degree of stenosis progresses, flow
velocities increase at the point of maximum narrowing. In
very severe stenosis the reduced residual lumen causes flow
velocities to decrease and leads to flow cessation with com-
plete lumen obliteration. Doppler insonation proximal to an
occluded vessel shows a stump flow pattern on spectral wave-
form. The downstream cerebrovascular impedance can be
estimated with the pulsatility and resistance indices.28,146 In
patients with stenosis in proximal cerebral vessels and impaired
CA, even slight MAP fluctuations may result in ischemia
because of global or local hypoperfusion, especially when col-
lateral circulation is insufficient.147 Hence where there is slowly
developing stenosis, ischemia can result from chronic and
acute factors. Typically slowly progressive stenosis causes a
decrease in local perfusion, mostly at the regions farthest from
the main supplying arteries—the arterial border zones.147 The
perfusion decrease makes these areas susceptible to fluctua-
tions in systemic MAP, oxygen supply, and local vascular resis-
tance. When a CPP decrease is short-lived, transient neurologic
deficits may occur, but when the CPP reduction lasts longer,
infarction can occur. Thus it is important to know in advance
whether a patient who has arterial stenosis is at risk of brain
infarct caused by hypoperfusion alone.148 TCD studies suggest
that the risk of cerebral infarction in the territory of a severely
stenotic or occluded ICA or MCA is greater (relative risk = 8)
in subjects with a diminished CA response (Fig. 30.7). The
annual risk of ipsilateral stroke in these patients with impaired
CA is 23.7%.149 However, an association between impaired
CVR and recurrent ischemic events in asymptomatic carotid
stenosis patients is yet to be determined.150
Selection of Patients for Extracranial
to Intracranial Bypass Surgery
It was expected that some patients with severe stenosis or
occlusion in a major cerebral artery may benefit from
extracranial-to-intracranial (EC-IC) bypass. Although the
EC-IC bypass trial performed in the 1980s found no benefit,151
308 Section V—Cerebral Blood Flow
a b c
Fig. 30.7  Line graph that illustrates the vasomotor reactivity (VMRr)
determined with bilateral transcranial Doppler (TCD) blood flow velocity
(BFV) measurements in the right middle cerebral artery (MCA) (green
curve, VMRr = 91%) and left MCA (yellow curve, VMRr = 58%) in a
patient with occlusion of the left internal carotid artery and 70% to 80%
stenosis of the right internal carotid artery, at rest (a), after hyperventilation
(b), after breath holding (c). (White curve illustrates patient’s heart rate.)
First time measure 10:18:00 and last time measure 10:23:00. (Courtesy
Dr. G. Kozyra, the Department of Neurology, Medical University of Gdańsk, Poland).
it is very likely that this trial included many patients whose
strokes were thromboembolic and so did not have impaired
local perfusion. TCD has been proposed for the selection of
patients for bypass—patients with impaired CA and in whom
thromboembolic infarction is unlikely.152 Although in these
patients the CA response improved after surgery153,154 and
selected studies showed clinical benefits,155,156 most current
guidelines do not recommend the use of EC-IC arterial bypass
for the treatment of acute ischemic stroke157,158 or in primary159
or secondary160 stroke prevention.
TCD testing of CA in patients with moyamoya disease also
may be useful because these patients often have multiple
severe proximal intracranial stenosis and extensive collateral
development (Fig. 30.8). Surgical revascularization proce-
dures when performed are associated with an improved CA
response. In addition, it seems that new vessels develop better
from grafts to areas where the CA is impaired, independent of
local CBF, suggesting that impaired CA may promote a local
angiogenic drive.161,162
Assessment of Cerebral Autoregulation
to Guide Conservative Management
of Occlusive Cerebrovascular Disease
In the setting of impaired CA the primary goal is to maintain
adequate CPP, such as by optimizing cardiac function, limiting
orthostatic MAP fluctuations, and carefully using antihyper-
tensive medication. By contrast, when CA is intact, manage-
ment may be better centered on control of hypertension and
thromboembolism. The relative merits of surgical or medical
treatment for impaired CA still have to be elucidated. In addi-
tion, CA function can improve over time without surgical
revascularization163,164 because adequate medical therapy for
months to several years may allow sufficient collateral circula-
tion to form, thus reducing the risk of stroke from
P1 rest
M1
P1 P1 bh
P1 contr
A
P1 rest
PoCoA
P1 P1 bh
B
Fig. 30.8  Transcranial color-coded duplex sonography (TCCS) examina-
tion and transcranial Doppler (TCD) assessment of vasomotor reactivity
in the P1 segment of posterior cerebral artery in (A) a healthy volunteer:
mean blood flow velocity (BFV) at rest is 46 cm/second (P1 rest), mean
BFV during hypercarbia after 30 seconds of breath holding (BH) is 69 cm/
second (P1), and breath holding index (BHI) is 1.36; and (B) in a patient
with moyamoya disease: mean BFV at rest is 103 cm/second (P1 rest),
mean BFV during hypercarbia is 109 cm/second (P1 bh) and after 30
seconds of breath holding (BHI = 0.2). In this patient the middle cerebral
artery is occluded and there is collateral flow via posterior communicat-
ing artery (PoCoA) and compensatory high velocities in the P1. (Courtesy
Department of Neurology, Ernst Moritz Arndt University of Greifswald, Germany.)
hypoperfusion. However, these same studies also show that
CA function does not improve in many patients, so surgery
then may be the preferred treatment option.
Hyperperfusion Syndrome
After Carotid Endarterectomy
Hyperperfusion syndrome is a rare but serious complication
of carotid endarterectomy (CEA) that causes cerebral edema
and intracerebral hemorrhage.165,166 It may also occur after
angioplasty or stenting of the extracrainal and intracranial
arteries.167,168 TCD assessment of CA can help predict which
patients are at high risk for this phenomenon.165,169-171 For
example, Hosoda et al. observed postoperative hyperperfu-
sion after CEA only in patients following CEA with a less than
10% CBF response after acetazolamide.169 In these high-risk
patients, strict MAP control should be instituted in the early
postoperative period.167,170
Balloon Occlusion Test
Balloon occlusion test (BOT) of a major vessel (e.g., the ICA
or VA) is used to assess the adequacy of collateral circulation
before permanent vessel occlusion or to help guide decisions
about aneurysm surgery or need for a bypass. Patients who
tolerate up to 30 minutes of vessel occlusion without clinical
deficits are considered able to tolerate permanent occlusion.172
However, approximately 10% of these patients still may

Fig. 30.9  Computed tomography (CT) image (left) shows extensive hemispheric hypodensity, effacement of the sulci, midline shift, and hemorrhage
(arrows) in a traumatic brain injury (TBI) patient. Transcranial color-coded duplex sonography (TCCS) shows no blood flow in the major basal arteries
and oscillating waveform pattern in the terminal internal carotid artery, consistent with brain death. ICA, Internal carotid artery.
develop an infarct. Several techniques including pharmaco-
logically induced hypotension, stump pressure measurements,
and perfusion imaging are used to increase the sensitivity of
BOT, although there is no consensus on the optimal protocol.
TCD testing of CA with acetazolamide has been performed
during BOT with promising results.172,173
Increased Intracranial Pressure, Cerebral
Circulatory Arrest, and Brain Death
The cranium acts as a near rigid container of virtually incom-
pressible brain (80%), blood (12%), and cerebrospinal fluid
(CSF) (8%)174 that are in a state of dynamic equilibrium.
When CPP is normal (80-100 mm Hg), ICP is usually less
than 15 mm Hg. During every cardiac systole some CSF and
venous blood is pushed out of the cranium to prevent an ICP
increase. When this compensatory mechanism is reduced or
absent, ICP can increase logarithmically as pathologic volume
further increases.175 This in turn can reduce CBF, and it then
may be necessary to focus management to maintain CBF even
if this causes a further increase in ICP, because neural tissue
generally is more sensitive to ischemia than to increased ICP.176
However, a decrease in CPP causes vasodilation that may be
further exacerbated by increased CO2, potassium ions, and
adenosine associated with hypoxia from insufficient CBF. At
this point, significant vasodilation does not help improve CBF.
Increased ICP also constricts cerebral veins and sinuses, and
the pressure is transmitted through its thin, compliant walls
inside so the point of maximal resistance of the cerebral vas-
cular bed shifts from the arterial to the venous system.177 In
other words, CBF decreases because of lower and lower CPP,
and dilated arteries contain more blood and increased cerebral
blood volume further increases ICP. When the ICP equals
systemic MAP, CPP decreases to zero, leading to complete
CBF cessation. A sudden increase of Pulsatility Index (PI =
[peak-systolic velocity − end-diastolic velocity]/time-averaged
maximum mean velocity) or Resistivity Index (RI = [peak-
systolic velocity − end-diastolic velocity]/peak-systolic veloc-
ity) values may suggest a critical increase of ICP.
When there is cerebral circulatory arrest, brain death
follows. Although TCD is not a formal test to determine brain
death, it has very high specificity and sensitivity—100% and
96%, respectively in one study178—and can be used to help in
brain death assessment, for example, to shorten the process of
Section V—Cerebral Blood Flow 309
ICA R
brain death prognosis before organ donation.179,180 Transor-
bital imaging may be used when transtemporal imaging is
not feasible.181 Typical flow patterns are reduced or absent
diastolic flow, reverberant flow, and short systolic spikes
(Fig. 30.9). In addition, PI and RI are high because of mark-
edly reduced systolic flows. However, these patterns also may
be seen temporarily following bolus administration of seda-
tives. In these patients TCD can be used to help make deci-
sions about brain death when sedative drugs preclude formal
testing.182 The Doppler findings are perhaps best used to
exclude brain death rather than make the diagnosis, which still
has a clinical foundation (see Chapter 13). TCCS may be more
reliable than TCD to demonstrate cerebral circulatory arrest
because the suitability of a temporal acoustic window can be
easily determined. This can be important because lack of a
Doppler signal from arteries cannot be discriminated from
lack of a window when TCD is used.183 In cases of difficult
acoustic window, the use of contrast enhancement with TCCS
may help in detecting cerebral circulatory arrest.184
Traumatic Brain Injury
TCD has several applications after TBI: noninvasive evalua-
tion of ICP, assessment of need for invasive monitoring, blood
flow evaluation, assessment of autoregulation and vasoreac-
tivity, help in calculating ICP indices such as cerebrovascular
reactivity and pressure reactivity, and outcome predic-
tion.41,185,186 Three phases associated with changes in CBF have
been described during the first 2 weeks after TBI.187,188 Phase
1 occurs on the day of injury and is associated with a low CBF,
normal MCA BFV on TCD, and normal arteriovenous differ-
ence in oxygen (AVDO2). Phase 2 occurs 1 to 2 days postinjury,
when CBF and MCA BFV on TCD are increased, and AVDO2
is decreased (i.e., relative hyperemia). Phase 3 occurs 4 to 15
days postinjury, when CBF is decreased and MCA BFV on
TCD is increased. This suggests a vasospastic phase with the
peak BFVs generally measured at days 9 to 11. Elevated BFVs
and increased PI and RI are independent factors associated
with poor outcome from TBI (Figs. 30.10 and 30.11).135,189,190
TCD can be used to assess CA after TBI191 and noninvasively
assess cerebral perfusion pressure.192 For example, Czosnyka
et al. used semi-continuous monitoring of MCA velocity with
invasive intracranial and arterial pressure measurement to cal-
culate the ability of severe TBI patients to autoregulate in
310 Section V—Cerebral Blood Flow

Fig. 30.10  Traumatic brain injury (TBI) patients who have very high or very low impedance indices and in particular on side of injury have worse
outcomes than patients with indices within normal range (left). Graph that illustrates transcranial color-coded duplex sonography (TCCS) data of
Doppler impedance indices (right) in a patient who had surgery for an intracerebral hematoma. TCCS performed after the first surgery showed high
indices that prompted repeat surgery. The second surgery led to a decrease of the indices. In contrast, intracranial pressure (ICP) measurements were
not as sensitive to changes in the patient’s hemodynamics. RI, Resistivity index ([peak-systolic velocity − end-diastolic velocity]/peak-systolic velocity).

Ved - ipsilateral hyperventilation and mannitol effectively lower ICP only

Ved - contralateral when CA is intact. Minor head injuries also can disturb CA,
ICP4 [mm Hg]
RI - ipsilateral
RI - contralateral
although the effect on outcome is unclear.198
ICP3
indexes and mean velocities in MCA’s
Standardized values of impedance

300%
ICP5
ICP2
Liver Failure
Cerebral edema, increased ICP, and impaired CA may com-
200% plicate acute and chronic liver failure because of disturbed
flow-metabolism coupling.199-201 However, it can be difficult to
use invasive ICP monitors because of liver-induced coagula-
100%
tion abnormalities. Instead TCD and TCCS can be used to
MCA spasm - ipsilateral noninvasively assess intracranial hemodynamics in patients
ICP1 D1 <2.5 mm

Hyperemia in MCA
who develop hepatic encephalopathy. Changes in CA in ful-
MCA spasm
- ipsilateral
D1 >5 mm
- contralateral
D1 <2.5 mm
minant hepatic failure may develop similar to that seen in
0%
4 10 15 20 TBI.202 Changes in MCA blood flow velocity generally are
y
jur

ry

ry
ge
f in

ge

Days after injury

proportional to changes in CBF because the MCA diameter
ur

ur
yo

Is

II s
Da

does not change significantly. TCD findings can help predict

Fig. 30-11  Transcranial color-coded duplex sonography (TCCS)
brainstem death in fulminant hepatic failure203; however, it is
study that illustrates the utility of TCCS monitoring in a traumatic
brain injury (TBI) patient with an intracranial hematoma and
still to be elucidated whether TCD monitoring can help make
subarachnoid hemorrhage. The intracranial hematoma was operated a difference to outcome. Where TCD may be particularly
on twice. After the first surgery, impedance indices in both middle useful is to guide induced hypothermia for patients with
cerebral arteries (MCAs) remained, which suggested that local intracranial hepatic encephalopathy through measurement of the PI to
pressure (ICP) was still increased. A postoperative computed tomography assess the ICP response to therapy. The PI is considered a
(CT) scan showed that the intracranial hematoma was not completely measure of distal CVR, although mathematical models suggest
evacuated. After the second surgery, the impedance indices decreased a complex relationship between PI and multiple hemody-
to normal values. At 5 days, blood flow velocities increased (hyperemia), namic variables.204
because the diameters of both MCAs did not change. Ten days after
injury, blood flow velocities increased due to vasospasm. RI, Resistivity
index ([peak-systolic velocity − end-diastolic velocity]/peak-systolic
velocity).
Microembolic Signals and
Monitoring During Vascular
response to spontaneous fluctuations in arterial pressure
Procedures
based on the regression of CPP on mean and systolic BFVs.41 Microembolic signals (MESs) can be detected using TCD fre-
They found that loss of CA within 2 days of TBI was associated quency downstream from vascular atherothrombotic or car-
with poor outcome,193 although other studies did not confirm diothrombotic lesions. The nature of MESs is heterogeneous,
this observation.194 Systolic flow velocities may provide a and their prevalence is highly variable; however, MESs are
stronger association with outcome than mean flow veloci- thought to represent the cerebral embolic load. For this reason
ties.195 Disruption of CA also is associated with poor condition TCD monitoring for MESs can be useful during a variety of
at presentation.196 Moderate hyperventilation (PaCO2 3.7 kPa) vascular or cardiac procedures including CEA, carotid angio-
can help improve CA after TBI,194 although an effect on plasty and stenting (CAS), intracranial endovascular proce-
outcome is unclear.197 These same studies suggest that dures, or cardiac surgery; predict stroke risk in patients with

asymptomatic carotid stenosis205; and to guide antiplatelet or
anticoagulation therapy in select patients at risk for embolic
stroke (e.g., after carotid or vertebral artery dissection or
symptomatic carotid stenosis).206
Stroke is the most common major periprocedural compli-
cation of CEA or CAS (2%-11% of patients).207,208 TCD mon-
itoring can help guide management to reduce the perioperative
stroke rate.209,210 During CEA hemodynamic changes mea-
sured by TCD insonation of the MCA manifest as ipsilateral
reductions in BFV consistent with a flow volume decline.211 A
slight decrease first may be observed during administration of
anesthetic agents. At cross-clamping, a more severe reduction
(60%-90% of preclamp values) may be detected. When mean
BFV is less than 15 cm/second (10%-20% of cases),212,213 cere-
bral ischemia will develop if not immediately corrected212,214
(e.g., through placement of a shunt to restore blood flow back
to 50%-120% of preclamp values). A decrease in MCA BFV
of more than 70% at cross-clamping is considered by some as
threshold for a shunt,209,215 although others propose higher
cutoff values.210,212 High-dose thiopental sodium or propofol
anesthesia during this phase of the operation is another
option to attenuate the effects of brain ischemia.216 A moder-
ate increase in TCD velocities is not uncommon during the
days after the operation. However, a persistent and severe
increase is seen in patients who develop the hyperperfusion
syndrome and can be used to guide blood pressure
management.
During CEA both particulate microemboli and air micro-
bubbles occur (Fig. 30.12); these can be detected as MESs by
TCD. Microembolism is particularly prevalent at cross-clamp
release, but only MESs detected during dissection, wound
closure, shunting, and in the immediate postoperative phase
of surgery have been associated with brain infarction.210,217,218
In particular more than 10 MESs during initial exposure219
and more than 20 MESs per hour during the immediate post-
operative phase are associated with brain infarcts.217,218 MES
detection therefore can prompt the surgeon to change the
operative technique or to use antithrombotic agents.220
Carotid angioplasty and stenting is associated with a greater
risk of microemboli than CEA.221,222 Consequently antiem-
bolic devices are increasingly used to reduce stroke risk during
CAS. In these patients TCD monitoring can help assess the
efficacy of cerebral protection provided by these devices
during stenting, and through assessment of TCD spectral pat-
terns evaluate the hemodynamic responses to the proce-
dure.223,224 TCD-identified microemboli during poststent
dilation (>5 showers), particulate macroembolism, massive air
CTVII
50
0 occur. If bubbles collapse sufficiently violently to produce
–50
–100
Fig. 30.12  Microembolic signals (arrows) in the transcranial Doppler
(TCD) velocity waveforms from the middle cerebral artery. (Courtesy
Dr. G. Kozyra, the Department of Neurology, Medical University of Gdańsk, Poland.)
Section V—Cerebral Blood Flow 311
embolism, hypotension with a significant reduction of MCA
BFVs, and angioplasty-induced asystole are associated with
adverse outcomes.225,226 However, the frequency of procedure-
related silent cerebral lesions (i.e., radiographic and not clini-
cal) appears to be independent of the number of MESs during
the procedure.227 TCD monitoring for MESs also is useful
during and after transluminal angioplasty in the posterior
circulation to guide appropriate use of anticoagulant or anti-
platelet therapies.228 Thromboembolic events that result in
neurologic deficits occur in 2.4% to 5.2% of patients who
undergo endovascular aneurysm occlusion.229 TCD monitor-
ing during and immediately after coiling can help identify
patients at high risk of thromboembolic complications.230,231
Acute stroke or cognitive decline associated with cerebral
hypoperfusion or emboli can complicate cardiac surgery.232-
239
Both stroke and cognitive dysfunction are associated with
the number of intraoperative MESs240 Theses MESs are het-
erogeneous and may represent gaseous particles, atheroscle-
rotic plaque components, platelet-fibrin aggregates, and
lipid-laden or fatty particles.241,242 The introduction of mem-
brane rather than bubble oxygenators and in-line filtration
can help decrease the number of MESs during cardiopulmo-
nary bypass.243 Just as in other invasive procedures, MESs do
not appear at random during the course of cardiac surgery;
clamping and unclamping of the aorta are associated with
more than 60% of the total number of MESs.244-247 In partic-
ular, the release of an aortic clamp placed over an atheroscle-
rotic plaque is associated with an increased risk of cerebral
embolism.248,249 The number of MESs is especially high
during cardiac ejection after the release of aortic cross-clamps
and immediately after bypass. These signals correspond to
atheromatous debris arising from the aorta. Better selection
of the site of clamp placement by epiaortic US and the use of
filters distal to the clamp may help reduce perioperative
stroke.250
Therapeutic Use of Ultrasound
in Acute Stroke
Successful thrombolysis depends on the delivery of tissue plas-
minogen activator (tPA) to the thrombus through residual
blood flow around the arterial obstruction.251,252 Experimental
and limited clinical studies suggest that low-frequency253,254 or
diagnostic255 US also can have thrombolytic effects and so may
augment the effects of tPA.256-258 This has encouraged assess-
ments of US for thrombolysis even with standard US equip-
ment.259 In addition, US can increase the uptake of tPA into a
clot,252 increase the binding of tPA to fibrin,260 and reversibly
increase fluid permeation through fibrin.261 These effects may
occur through acoustic cavitations because degassing reduces
the US effect on flow through fibrin. If bubbles are induced
to grow by US to diameters larger than the pore size of the
fibrin lattice surrounding it, stretching of clot fibers may
broadband acoustic emissions, additional inertial cavitations
may produce localized stresses, hot spots, or microjets
that may further alter the structure of clot fibers.262 It also
is possible that the heat generated by US accelerates throm-
bolysis.263 US-accelerated thrombolysis may be further
enhanced by administration of US contrast MBs.264-270 Human
CLOTBUST studies271,272 involved MBs with dedicated
312 Section V—Cerebral Blood Flow
machines to make US delivery easier and more reliable for
thrombolysis, in which complete MCA recanalization with
dramatic clinical recovery occurred in 49% of patients receiv-
ing intravenous rtPA combined with transcranial ultrasound
versus 30% in patients receiving intravenous rtPA only.123
Transcranial US can be attenuated by bones of the cranium;
consequently, its use may not be possible in 10% to 15% of
patients.8 There are several strategies to overcome this problem:
(1) use of endovascular wires and transducers to deliver US
locally; (2) use of lower-frequency and subsequently higher-
power US for transcutaneous US-enhanced thrombolysis; or
(3) use of US contrast to induce and increase the number of
cavitations at the site where the US beam of a high mechanical
index is targeted.273-278 Frequencies used include 20 kHz, 40
kHz, 170 kHz, 300 kHz, 1 MHz, and 2 MHz, at intensities
from 0.25 to 10 W/cm2.264,279-283
Conclusion
Optimal management of patients in the NCCU depends in
part on understanding the physiology for the individual
patient. TCD ultrasonography first described in the 1980s is a
noninvasive test that can be used to help understand patient
pathophysiology and monitor for responses to management.
Technologic advances such as TCCS now make TCD more
reliable and accurate. This is a well-known test that records
the velocity (in cm/sec) and direction of blood flow in the
basal vessels of the brain and is most frequently used to evalu-
ate for vasospasm, particularly in the MCA, in SAH patients.
In addition, TCD and TCCS can be used to noninvasively
evaluate: 1) intracranial pressure, 2) other causes of vessel
stenosis, e.g., atherosclerosis, 3) TBI patients, 4) hemodynamic
alterations observed in intracranial and extracranial occlusive
disease, and 5) emboli among other disease states. However,
TCD does not measure flow rate (mL/min) or perfusion (mL
per 100/g/min). Instead, changes in TCD velocities reflect rela-
tive CBF changes provided that the recorded vessel diameter
remains constant. Therefore TCD also can be used to examine
CBF reactivity in response to CO2, blood pressure, acetazol-
amide, or head position and thus help assess cerebral auto-
regulation. While noninvasive, TCD is not a technique that is
easy to use for continuous monitoring, although continued
technologic improvements now make this possible. In addi-
tion, TCD ultrasound examination requires training and prac-
tice, and so the accuracy of the examination often depends on
the operator’s skill and experience.
References
1. Alexandrov AV, Sloan MA, Tegeler CH, et al. Practice standards for
transcranial Doppler (TCD) ultrasound. Part II. Clinical indications and
expected outcomes. J Neuroimaging 2010;Oct 26. Epub ahead of print.
2. Edmonds Jr HL, Isley MR, Sloan TB, et al. American Society of
Neurophysiologic Monitoring and American Society of Neuroimaging joint
guidelines for transcranial Doppler ultrasonic monitoring. J Neuroimaging
2011;21(2):177–83.
3. Nelson TR, Pretorius DH. The Doppler signal: where does it come from and
what does it mean? AJR Am J Roentgenol 1988;151(3):439–47.
4. Taylor KJW, Burns PN, Wells PNT. Clinical applications of Doppler
ultrasound. New York: Raven Press; 1995.
5. Hoeks AP, Reneman RS. Biophysical principles of vascular diagnosis. J Clin
Ultrasound 1995;23(2):71–9.
6. Krejza J, Mariak Z, Melhem ER, et al. A guide to the identification of major
cerebral arteries with transcranial color Doppler sonography. AJR Am J
Roentgenol 2000;174(5):1297–303.
7. Krejza J, Mariak Z, Babikian VL. Importance of angle correction in the
measurement of blood flow velocity with transcranial Doppler sonography.
AJNR Am J Neuroradiol 2001;22(9):1743–7.
8. Krejza J, Swiat M, Pawlak MA, et al. Suitability of temporal bone acoustic
window: conventional TCD versus transcranial color-coded duplex
sonography. J Neuroimaging 2007;17(4):311–14.
9. Nyborg WL. Biological effects of ultrasound: development of safety
guidelines. Part II. General review. Ultrasound Med Biol 2001;27(3):301–33.
10. Barnett SB, Ter Haar GR, Ziskin MC, et al. International recommendations
and guidelines for the safe use of diagnostic ultrasound in medicine.
Ultrasound Med Biol 2000;26(3):355–66.
11. Wu J, Winkler AJ, O’Neill TP. Effect of acoustic streaming on ultrasonic
heating. Ultrasound Med Biol 1994;20(2):195–201.
12. Daffertshofer M, Gass A, Ringleb P, et al. Transcranial low-frequency
ultrasound-mediated thrombolysis in brain ischemia: increased risk of
hemorrhage with combined ultrasound and tissue plasminogen activator:
results of a phase II clinical trial. Stroke 2005;36(7):1441–6.
13. Mariak Z, Krejza J, Swiercz M, et al. Human brain temperature in vivo: lack
of heating during color transcranial Doppler ultrasonography.
J Neuroimaging 2001;11(3):308–12.
14. Karagoz I, Kartal MK. A new safety parameter for diagnostic ultrasound
thermal bioeffects: safe use time. J Acoust Soc Am 2009;125(6):3601–10
15. Krasovitski B, Frenkel V, Shoham S, et al. Intramembrane cavitation as a
unifying mechanism for ultrasound-induced bioeffects. Proc Natl Acad Sci U
S A 2011;108(8):3258–63.
16. Kondo T, Kodaira T, Kano E. Free radical formation induced by ultrasound
and its effects on strand breaks in DNA of cultured FM3A cells. Free Radic
Res Commun 1993;19(Suppl 1):S193–200.
17. Hynynen K, McDannold N, Sheikov NA, et al. Local and reversible
blood-brain barrier disruption by noninvasive focused ultrasound at
frequencies suitable for trans-skull sonications. Neuroimage 2005;24(1):
12–20.
18. Zhong W, Sit WH, Wan JM, et al. Sonoporation induces apoptosis and cell
cycle arrest in human promyelocytic leukemia cells. Ultrasound Med Biol
2011;37(12):2149–59.
19. Sarvazyan AP, Rudenko OV, Nyborg WL. Biomedical applications of
radiation force of ultrasound: historical roots and physical basis. Ultrasound
Med Biol 2010;36(9):1379–94.
20. Lassen NA. Cerebral blood flow and oxygen consumption in man. Physiol
Rev 1959;39(2):183–238.
21. Strandgaard S, Olesen J, Skinhoj E, et al. Autoregulation of brain circulation
in severe arterial hypertension. BMJ 1973;1(5852):507–10.
22. Waldemar G, Schmidt JF, Andersen AR, et al. Angiotensin converting enzyme
inhibition and cerebral blood flow autoregulation in normotensive and
hypertensive man. J Hypertens 1989;7(3):229–35.
23. Rudzinski W, Swiat M, Tomaszewski M, et al. Cerebral hemodynamics and
investigations of cerebral blood flow regulation. Nucl Med Rev Cent East Eur
2007;10(1):29–42.
24. Wallis SJ, Firth J, Dunn WR. Pressure-induced myogenic responses in human
isolated cerebral resistance arteries. Stroke 1996;27(12):2287–90.
25. Garcia-Roldan JL, Bevan JA. Flow-induced constriction and dilation of
cerebral resistance arteries. Circ Res 1990;66(5):1445–8.
26. Thorin-Trescases N, Bevan JA. High levels of myogenic tone antagonize the
dilator response to flow of small rabbit cerebral arteries. Stroke
1998;29(6):1194–200.
27. McDonald DA. Arterial impedance. In: McDonald DA, editor. Blood flow in
arteries. Baltimore: Williams & Wilkins; 1977. p. 351–89.
28. Kontos HA, Raper AJ, Patterson JL. Analysis of vasoactivity of local pH,
PaCO2 and bicarbonate on pial vessels. Stroke 1977;8(3):358–60.
29. Baffi J, Gorcs T, Slowik F, et al. Neuropeptides in the human superior
cervical ganglion. Brain Res 1992;570(1-2):272–8.
30. Bleys RL, Cowen T. Innervation of cerebral blood vessels: morphology,
plasticity, age-related, and Alzheimer’s disease-related neurodegeneration.
Microsc Res Tech 2001;53(2):106–18.
31. Baumbach GL, Heistad DD. Effects of sympathetic stimulation and changes
in arterial pressure on segmental resistance of cerebral vessels in rabbits and
cats. Circ Res 1983;52(5):527–33.
32. Morita-Tsuzuki Y, Hardebo JE, Bouskela E. Inhibition of nitric oxide
synthase attenuates the cerebral blood flow response to stimulation of
postganglionic parasympathetic nerves in the rat. J Cereb Blood Flow Metab
1993;13(6):993–7.
33. Kalaria RN, Stockmeier CA, Harik SI. Brain microvessels are innervated by
locus ceruleus noradrenergic neurons. Neurosci Lett 1989;97(1-2):203–8.
34. Moreno MJ, Lopez DP, Conde MV, et al. Cat cerebral arteries are
functionally innervated by serotoninergic fibers from central and peripheral
origins. Stroke 1995;26(2):271–5.

35. Sato A, Sato Y, Uchida S. Activation of the intracerebral cholinergic nerve
fibers originating in the basal forebrain increases regional cerebral blood
flow in the rat’s cortex and hippocampus. Neurosci Lett 2004;361(1-3):90–3.
36. Cauli B, Tong XK, Rancillac A, et al. Cortical GABA interneurons in
neurovascular coupling: relays for subcortical vasoactive pathways.
J Neurosci 2004;24(41):8940–9.
37. Mulligan SJ, MacVicar BA. Calcium transients in astrocyte endfeet cause
cerebrovascular constrictions. Nature 2004;431(7005):195–9.
38. Aaslid R, Lash SR, Bardy GH, et al. Dynamic pressure–flow velocity
relationships in the human cerebral circulation. Stroke 2003;34(7):1645–9.
39. Paulson OB, Strandgaard S, Edvinsson L. Cerebral autoregulation.
Cerebrovasc Brain Metab Rev 1990;2(2):161–92.
40. Tiecks FP, Lam AM, Aaslid R, et al. Comparison of static and dynamic
cerebral autoregulation measurements. Stroke 1995;26(6):1014–19.
41. Czosnyka M, Brady K, Reinhard M, et al. Monitoring of cerebrovascular
autoregulation: facts, myths, and missing links. Neurocrit Care 2009;10(3):
373–86.
42. Grossmann WM, Koeberle B. The dose-response relationship of
acetazolamide on the cerebral blood flow in normal subjects. Cerebrovasc
Dis 2000;10(1):65–9.
43. Heuser D, Astrup J, Lassen NA, et al. Brain carbonic acid acidosis after
acetazolamide. Acta Physiol Scand 1975;93(3):385–90.
44. Vorstrup S, Henriksen L, Paulson OB. Effect of acetazolamide on cerebral
blood flow and cerebral metabolic rate for oxygen. J Clin Invest 1984;74(5):
1634–9.
Section V—Cerebral Blood Flow 313
45. Okazawa H, Yamauchi H, Sugimoto K, et al. Effects of acetazolamide on
cerebral blood flow, blood volume, and oxygen metabolism: a positron
emission tomography study with healthy volunteers. J Cereb Blood Flow
Metab 2001;21(12):1472–9.
46. Sorteberg W, Lindegaard KF, Rootwelt K, et al. Effect of acetazolamide on
cerebral artery blood velocity and regional cerebral blood flow in normal
subjects. Acta Neurochir (Wien) 1989;97(3-4):139–45.
47. Sullivan HG, Kingsbury TB, Morgan ME, et al. The rCBF response to
Diamox in normal subjects and cerebrovascular disease patients. J Neurosurg
1987;67(4):525–34.
48. Kuwabara Y, Ichiya Y, Sasaki M, et al. Time dependency of the acetazolamide
effect on cerebral hemodynamics in patients with chronic occlusive cerebral
arteries. Early steal phenomenon demonstrated by [15O]H2O positron
emission tomography. Stroke 1995;26(10):1825–9.
49. Yonas H, Pindzola RR. Physiological determination of cerebrovascular
reserves and its use in clinical management. Cerebrovasc Brain Metab Rev
1994;6(4):325–40.
50. Petrella JR, DeCarli C, Dagli M, et al. Age-related vasodilatory response to
acetazolamide challenge in healthy adults: a dynamic contrast-enhanced MR
study. AJNR Am J Neuroradiol 1998;19(1):39–44.
A complete list of references for this chapter can be found online at
www.expertconsult.com.

References
1. Alexandrov AV, Sloan MA, Tegeler CH, et al. Practice standards for
transcranial Doppler (TCD) ultrasound. Part II. Clinical indications and
expected outcomes. J Neuroimaging 2010;Oct 26. Epub ahead of print.
2. Edmonds Jr HL, Isley MR, Sloan TB, et al. American Society of
Neurophysiologic Monitoring and American Society of Neuroimaging joint
guidelines for transcranial Doppler ultrasonic monitoring. J Neuroimaging
2011;21(2):177–83.
3. Nelson TR, Pretorius DH. The Doppler signal: where does it come from
and what does it mean? AJR Am J Roentgenol 1988;151(3):439–47.
4. Taylor KJW, Burns PN, Wells PNT. Clinical applications of Doppler
ultrasound. New York: Raven Press; 1995.
5. Hoeks AP, Reneman RS. Biophysical principles of vascular diagnosis. J Clin
Ultrasound 1995;23(2):71–9.
6. Krejza J, Mariak Z, Melhem ER, et al. A guide to the identification of major
cerebral arteries with transcranial color Doppler sonography. AJR Am J
Roentgenol 2000;174(5):1297–303.
7. Krejza J, Mariak Z, Babikian VL. Importance of angle correction in the
measurement of blood flow velocity with transcranial Doppler sonography.
AJNR Am J Neuroradiol 2001;22(9):1743–7.
8. Krejza J, Swiat M, Pawlak MA, et al. Suitability of temporal bone acoustic
window: conventional TCD versus transcranial color-coded duplex
sonography. J Neuroimaging 2007;17(4):311–14.
9. Nyborg WL. Biological effects of ultrasound: development of safety
guidelines. Part II. General review. Ultrasound Med Biol 2001;27(3):
301–33.
10. Barnett SB, Ter Haar GR, Ziskin MC, et al. International recommendations
and guidelines for the safe use of diagnostic ultrasound in medicine.
Ultrasound Med Biol 2000;26(3):355–66.
11. Wu J, Winkler AJ, O’Neill TP. Effect of acoustic streaming on ultrasonic
heating. Ultrasound Med Biol 1994;20(2):195–201.
12. Daffertshofer M, Gass A, Ringleb P, et al. Transcranial low-frequency
ultrasound-mediated thrombolysis in brain ischemia: increased risk of
hemorrhage with combined ultrasound and tissue plasminogen activator:
results of a phase II clinical trial. Stroke 2005;36(7):1441–6.
13. Mariak Z, Krejza J, Swiercz M, et al. Human brain temperature in vivo: lack
of heating during color transcranial Doppler ultrasonography.
J Neuroimaging 2001;11(3):308–12.
14. Karagoz I, Kartal MK. A new safety parameter for diagnostic ultrasound
thermal bioeffects: safe use time. J Acoust Soc Am 2009;125(6):3601–10
15. Krasovitski B, Frenkel V, Shoham S, et al. Intramembrane cavitation as a
unifying mechanism for ultrasound-induced bioeffects. Proc Natl Acad Sci
U S A 2011;108(8):3258–63.
16. Kondo T, Kodaira T, Kano E. Free radical formation induced by ultrasound
and its effects on strand breaks in DNA of cultured FM3A cells. Free Radic
Res Commun 1993;19(Suppl 1):S193–200.
17. Hynynen K, McDannold N, Sheikov NA, et al. Local and reversible
blood-brain barrier disruption by noninvasive focused ultrasound at
frequencies suitable for trans-skull sonications. Neuroimage
2005;24(1):12–20.
18. Zhong W, Sit WH, Wan JM, et al. Sonoporation induces apoptosis and cell
cycle arrest in human promyelocytic leukemia cells. Ultrasound Med Biol
2011;37(12):2149–59.
19. Sarvazyan AP, Rudenko OV, Nyborg WL. Biomedical applications of
radiation force of ultrasound: historical roots and physical basis.
Ultrasound Med Biol 2010;36(9):1379–94.
20. Lassen NA. Cerebral blood flow and oxygen consumption in man. Physiol
Rev 1959;39(2):183–238.
21. Strandgaard S, Olesen J, Skinhoj E, et al. Autoregulation of brain circulation
in severe arterial hypertension. BMJ 1973;1(5852):507–10.
22. Waldemar G, Schmidt JF, Andersen AR, et al. Angiotensin converting
enzyme inhibition and cerebral blood flow autoregulation in normotensive
and hypertensive man. J Hypertens 1989;7(3):229–35.
23. Rudzinski W, Swiat M, Tomaszewski M, et al. Cerebral hemodynamics and
investigations of cerebral blood flow regulation. Nucl Med Rev Cent East
Eur 2007;10(1):29–42.
24. Wallis SJ, Firth J, Dunn WR. Pressure-induced myogenic responses in
human isolated cerebral resistance arteries. Stroke 1996;27(12):2287–90.
25. Garcia-Roldan JL, Bevan JA. Flow-induced constriction and dilation of
cerebral resistance arteries. Circ Res 1990;66(5):1445–8.
26. Thorin-Trescases N, Bevan JA. High levels of myogenic tone antagonize the
dilator response to flow of small rabbit cerebral arteries. Stroke 1998;29(6):
1194–200.
27. McDonald DA. Arterial impedance. In: McDonald DA, editor. Blood flow in
arteries. Baltimore: Williams & Wilkins; 1977. p. 351–89.
Section V—Cerebral Blood Flow 313.e1
28. Kontos HA, Raper AJ, Patterson JL. Analysis of vasoactivity of local pH,
PaCO2 and bicarbonate on pial vessels. Stroke 1977;8(3):358–60.
29. Baffi J, Gorcs T, Slowik F, et al. Neuropeptides in the human superior
cervical ganglion. Brain Res 1992;570(1-2):272–8.
30. Bleys RL, Cowen T. Innervation of cerebral blood vessels: morphology,
plasticity, age-related, and Alzheimer’s disease-related neurodegeneration.
Microsc Res Tech 2001;53(2):106–18.
31. Baumbach GL, Heistad DD. Effects of sympathetic stimulation and changes
in arterial pressure on segmental resistance of cerebral vessels in rabbits
and cats. Circ Res 1983;52(5):527–33.
32. Morita-Tsuzuki Y, Hardebo JE, Bouskela E. Inhibition of nitric oxide
synthase attenuates the cerebral blood flow response to stimulation of
postganglionic parasympathetic nerves in the rat. J Cereb Blood Flow
Metab 1993;13(6):993–7.
33. Kalaria RN, Stockmeier CA, Harik SI. Brain microvessels are innervated by
locus ceruleus noradrenergic neurons. Neurosci Lett 1989;97(1-2):203–8.
34. Moreno MJ, Lopez DP, Conde MV, et al. Cat cerebral arteries are
functionally innervated by serotoninergic fibers from central and peripheral
origins. Stroke 1995;26(2):271–5.
35. Sato A, Sato Y, Uchida S. Activation of the intracerebral cholinergic nerve
fibers originating in the basal forebrain increases regional cerebral blood
flow in the rat’s cortex and hippocampus. Neurosci Lett 2004;361(1-3):
90–3.
36. Cauli B, Tong XK, Rancillac A, et al. Cortical GABA interneurons in
neurovascular coupling: relays for subcortical vasoactive pathways.
J Neurosci 2004;24(41):8940–9.
37. Mulligan SJ, MacVicar BA. Calcium transients in astrocyte endfeet cause
cerebrovascular constrictions. Nature 2004;431(7005):195–9.
38. Aaslid R, Lash SR, Bardy GH, et al. Dynamic pressure–flow velocity
relationships in the human cerebral circulation. Stroke 2003;34(7):1645–9.
39. Paulson OB, Strandgaard S, Edvinsson L. Cerebral autoregulation.
Cerebrovasc Brain Metab Rev 1990;2(2):161–92.
40. Tiecks FP, Lam AM, Aaslid R, et al. Comparison of static and dynamic
cerebral autoregulation measurements. Stroke 1995;26(6):1014–19.
41. Czosnyka M, Brady K, Reinhard M, et al. Monitoring of cerebrovascular
autoregulation: facts, myths, and missing links. Neurocrit Care 2009;10(3):
373–86.
42. Grossmann WM, Koeberle B. The dose-response relationship of
acetazolamide on the cerebral blood flow in normal subjects. Cerebrovasc
Dis 2000;10(1):65–9.
43. Heuser D, Astrup J, Lassen NA, et al. Brain carbonic acid acidosis after
acetazolamide. Acta Physiol Scand 1975;93(3):385–90.
44. Vorstrup S, Henriksen L, Paulson OB. Effect of acetazolamide on cerebral
blood flow and cerebral metabolic rate for oxygen. J Clin Invest 1984;74(5):
1634–9.
45. Okazawa H, Yamauchi H, Sugimoto K, et al. Effects of acetazolamide on
cerebral blood flow, blood volume, and oxygen metabolism: a positron
emission tomography study with healthy volunteers. J Cereb Blood Flow
Metab 2001;21(12):1472–9.
46. Sorteberg W, Lindegaard KF, Rootwelt K, et al. Effect of acetazolamide on
cerebral artery blood velocity and regional cerebral blood flow in normal
subjects. Acta Neurochir (Wien) 1989;97(3-4):139–45.
47. Sullivan HG, Kingsbury TB, Morgan ME, et al. The rCBF response to
Diamox in normal subjects and cerebrovascular disease patients.
J Neurosurg 1987;67(4):525–34.
48. Kuwabara Y, Ichiya Y, Sasaki M, et al. Time dependency of the
acetazolamide effect on cerebral hemodynamics in patients with chronic
occlusive cerebral arteries. Early steal phenomenon demonstrated by [15O]
H2O positron emission tomography. Stroke 1995;26(10):1825–9.
49. Yonas H, Pindzola RR. Physiological determination of cerebrovascular
reserves and its use in clinical management. Cerebrovasc Brain Metab Rev
1994;6(4):325–40.
50. Petrella JR, DeCarli C, Dagli M, et al. Age-related vasodilatory response to
acetazolamide challenge in healthy adults: a dynamic contrast-enhanced
MR study. AJNR Am J Neuroradiol 1998;19(1):39–44.
51. Brian JE. Carbon dioxide and the cerebral circulation. Anesthesiology
1998;88(5):1365–86.
52. Kanno I, Uemura K, Higano S, et al. Oxygen extraction fraction at
maximally vasodilated tissue in the ischemic brain estimated from the
regional CO2 responsiveness measured by positron emission tomography.
J Cereb Blood Flow Metab 1988;8(2):227–35.
53. Kazumata K, Tanaka N, Ishikawa T, et al. Dissociation of vasoreactivity to
acetazolamide and hypercapnia. Comparative study in patients with chronic
occlusive major cerebral artery disease. Stroke 1996;27(11):2052–8.
54. Gambhir S, Inao S, Tadokoro M, et al. Comparison of vasodilatory effect of
carbon dioxide inhalation and intravenous acetazolamide on brain
313.e2 Section V—Cerebral Blood Flow
vasculature using positron emission tomography. Neurol Res 1997;19(2):
139–44.
55. Ringelstein EB, Van Eyck S, Mertens I. Evaluation of cerebral vasomotor
reactivity by various vasodilating stimuli: comparison of CO2 to
acetazolamide. J Cereb Blood Flow Metab 1992;12(1):162–8.
56. Lam JM, Hsiang JN, Poon WS. Monitoring of autoregulation using laser
Doppler flowmetry in patients with head injury. J Neurosurg
1997;86(3):438–45.
57. Tiecks FP, Lam AM, Aaslid R, et al. Comparison of static and dynamic
cerebral autoregulation measurements. Stroke 1995;26(6):1014–19.
58. Bouma GJ, Muizelaar JP. Relationship between cardiac output and cerebral
blood flow in patients with intact and with impaired autoregulation.
J Neurosurg 1990;73(3):368–74.
59. Aaslid R, Lindegaard KF, Sorteberg W, et al. Cerebral autoregulation
dynamics in humans. Stroke 1989;20(1):45–52.
60. Tiecks FP, Douville C, Byrd S, et al. Evaluation of impaired cerebral
autoregulation by the Valsalva maneuver. Stroke 1996;27(7):1177–82.
61. Kofke WA, Dong ML, Bloom M, et al. Transcranial Doppler
ultrasonography with induction of anesthesia for neurosurgery.
J Neurosurg Anesthesiol 1994;6(2):89–97.
62. Renou AM, Vernhiet J, Macrez P, et al. Cerebral blood flow and metabolism
during etomidate anaesthesia in man. Br J Anaesth 1978;50(10):1047–51.
63. Harrison JM, Girling KJ, Mahajan RP. Effects of target-controlled infusion
of propofol on the transient hyperaemic response and carbon dioxide
reactivity in the middle cerebral artery. Br J Anaesth 1999;83(6):839–44.
64. Matta BF, Lam AM. Nitrous oxide increases cerebral blood flow velocity
during pharmacologically induced EEG silence in humans. J Neurosurg
Anesthesiol 1995;7(2):89–93.
65. McCulloch TJ, Visco E, Lam AM. Graded hypercapnia and cerebral
autoregulation during sevoflurane or propofol anesthesia. Anesthesiology
2000;93(5):1205–9.
66. Mayberg TS, Lam AM, Matta BF, et al. Ketamine does not increase cerebral
blood flow velocity or intracranial pressure during isoflurane/nitrous oxide
anesthesia in patients undergoing craniotomy. Anesth Analg 1995;81(1):
84–9.
67. Kuroda Y, Murakami M, Tsuruta J, et al. Blood flow velocity of middle
cerebral artery during prolonged anesthesia with halothane, isoflurane, and
sevoflurane in humans. Anesthesiology 1997;87(3):527–32.
68. Matta BF, Heath KJ, Tipping K, et al. Direct cerebral vasodilatory effects of
sevoflurane and isoflurane. Anesthesiology 1999;91(3):677–80.
69. Mielck F, Stephan H, Buhre W, et al. Effects of 1 MAC desflurane on
cerebral metabolism, blood flow and carbon dioxide reactivity in humans.
Br J Anaesth 1998;81(2):155–60.
70. Gupta S, Heath K, Matta BF. Effect of incremental doses of sevoflurane on
cerebral pressure autoregulation in humans. Br J Anaesth
1997;79(4):469–72.
71. Heath KJ, Gupta S, Matta BF. The effects of sevoflurane on cerebral
hemodynamics during propofol anesthesia. Anesth Analg 1997;85(6):1284–7.
72. Matta BF, Lam AM, Mayberg TS, et al. Cerebrovascular response to carbon
dioxide during sodium nitroprusside- and isoflurane-induced hypotension.
Br J Anaesth 1995;74(3):296–300.
73. Strebel S, Lam AM, Matta B, et al. Dynamic and static cerebral
autoregulation during isoflurane, desflurane, and propofol anesthesia.
Anesthesiology 1995;83(1):66–76.
74. Summors AC, Gupta AK, Matta BF. Dynamic cerebral autoregulation
during sevoflurane anesthesia: a comparison with isoflurane. Anesth Analg
1999;88(2):341–5.
75. Girling KJ, Cavill G, Mahajan RP. The effects of nitrous oxide and oxygen
on transient hyperemic response in human volunteers. Anesth Analg 1999;
89(1):175–80.
76. Bedforth NM, Girling KJ, Harrison JM, et al. The effects of sevoflurane and
nitrous oxide on middle cerebral artery blood flow velocity and transient
hyperemic response. Anesth Analg 1999;89(1):170–4.
77. Matta BF, Lam AM. Nitrous oxide increases cerebral blood flow velocity
during pharmacologically induced EEG silence in humans. J Neurosurg
Anesthesiol 1995;7(2):89–93.
78. Kovarik WD, Mayberg TS, Lam AM, et al. Succinylcholine does not change
intracranial pressure, cerebral blood flow velocity, or the
electroencephalogram in patients with neurologic injury. Anesth Analg
1994;78(3):469–73.
79. Kofke WA, Shaheen N, McWhorter J, et al. Transcranial Doppler
ultrasonography with induction of anesthesia and neuromuscular blockade
in surgical patients. J Clin Anesth 2001;13(5):335–8.
80. Hanel F, Werner C, von Knobelsdorff G, et al. The effects of fentanyl and
sufentanil on cerebral hemodynamics. J Neurosurg Anesthesiol 1997;9(3):
223–7.
81. Paris A, Scholz J, von Knobelsdorff G, et al. The effect of remifentanil on
cerebral blood flow velocity. Anesth Analg 1998;87(3):569–73.
82. Moppett IK, Sherman RW, Wild MJ, et al. Effects of norepinephrine and
glyceryl trinitrate on cerebral haemodynamics: transcranial Doppler study
in healthy volunteers. Br J Anaesth 2008;100(2):240–4.
83. Moppett IK, Wild MJ, Sherman RW, et al. Effects of ephedrine, dobutamine
and dopexamine on cerebral haemodynamics: transcranial Doppler studies
in healthy volunteers. Br J Anaesth 2004;92(1):39–44.
84. Strebel SP, Kindler C, Bissonnette B, et al. The impact of systemic
vasoconstrictors on the cerebral circulation of anesthetized patients.
Anesthesiology 1998;89(1):67–72.
85. Endoh H, Honda T, Ohashi S, et al. The influence of nitroglycerin and
prostaglandin E1 on dynamic cerebral autoregulation in adult patients
during propofol and fentanyl anaesthesia. Anaesthesia 2001;56(10):947–52.
86. Brauer P, Kochs E, Werner C, et al. Correlation of transcranial Doppler
sonography mean flow velocity with cerebral blood flow in patients with
intracranial pathology. J Neurosurg Anesthesiol 1998;10(2):80–5.
87. Demolis P, Tran Dinh YR, Giudicelli JF. Relationships between cerebral
regional blood flow velocities and volumetric blood flows and their
respective reactivities to acetazolamide. Stroke 1996;27(10):1835–9.
88. van Gijn J, Rinkel GJE. Subarachnoid haemorrhage: diagnosis, causes and
management. Brain 2001;124:249–78.
89. Shimoda M, Oda S, Tsugane R, et al. Intracranial complications of
hypervolemic therapy in patients with a delayed ischemic deficit attributed
to vasospasm. J Neurosurg 1993;78(3):423–9.
90. Willinsky RA, Taylor SM, terBrugge K, et al. Neurologic complications of
cerebral angiography: prospective analysis of 2,899 procedures and review
of the literature. Radiology 2003;227(2):522–8.
91. Anderson GB, Ashforth R, Steinke DE, et al. CT angiography for the
detection of cerebral vasospasm in patients with acute subarachnoid
hemorrhage. AJNR Am J Neuroradiol 2000;21(6):1011–15.
92. Heiserman JE. MR angiography for the diagnosis of vasospasm after
subarachnoid hemorrhage. Is it accurate? Is it safe? AJNR Am J Neuroradiol
2000;21(9):1571–2.
93. Aaslid R, Huber P, Nornes H. Evaluation of cerebrovascular spasm with
transcranial Doppler ultrasound. J Neurosurg 1984;60(1):37–41.
94. Lysakowski C, Walder B, Costanza MC, et al. Transcranial Doppler versus
angiography in patients with vasospasm due to a ruptured cerebral
aneurysm: a systematic review. Stroke 2001;32(10):2292–8.
95. Mariak Z, Krejza J, Swiercz M, et al. Accuracy of transcranial color Doppler
ultrasonography in the diagnosis of middle cerebral artery spasm
determined by receiver operating characteristic analysis. J Neurosurg 2002;
96(2):323–30.
96. Krejza J, Mariak Z, Lewko J. Standardization of flow velocities with respect
to age and sex improves the accuracy of transcranial color Doppler
sonography of middle cerebral artery spasm. Am J Roentgenol 2003;181(1):
245–52.
97. Proust F, Callonec F, Clavier E, et al. Usefulness of transcranial color-coded
sonography in the diagnosis of cerebral vasospasm. Stroke 1999;30(5):
1091–8.
98. Krejza J, Kochanowicz J, Mariak Z, et al. Middle cerebral artery spasm after
subarachnoid hemorrhage: detection with transcranial color-coded duplex
US. Radiology 2005;236(2):621–9.
99. Swiat M, Weigele J, Hurst RW, et al. Middle cerebral artery vasospasm:
transcranial color-coded duplex sonography versus conventional
non-imaging TCD. Crit Care Med 2009;37(3):963–8.
100. Krejza J, Rudzinski W, Pawlak MA, et al. Angle-corrected imaging
transcranial Doppler sonography versus imaging and nonimaging
transcranial Doppler sonography in children with sickle cell disease. AJNR
Am J Neuroradiol 2007;28(8):1613–18.
101. Muizelaar JP, Zwienenberg M, Rudisill NA, et al. The prophylactic use of
transluminal balloon angioplasty in patients with Fisher grade 3
subarachnoid hemorrhage: a pilot study. J Neurosurg 1999;91(1):51–8.
102. Murayama Y, Malisch T, Guglielmi G, et al. Incidence of cerebral vasospasm
after endovascular treatment of acutely ruptured aneurysms: report on 69
cases. J Neurosurg 1997;87(6):830–5.
103. Wardlaw JM, Offin R, Teasdale GM, et al. Is routine transcranial Doppler
ultrasound monitoring useful in the management of subarachnoid
hemorrhage? J Neurosurg 1998;88(2):272–6.
104. Lindegaard KF, Nornes H, Bakke SJ, et al. Cerebral vasospasm diagnosis by
means of angiography and blood velocity-measurements. Acta
Neurochirurgica 1989;100(1-2):12–24.
105. Lang EW, Diehl RR, Mehdorn HM. Cerebral autoregulation testing after
aneurysmal subarachnoid hemorrhage: the phase relationship between
arterial blood pressure and cerebral blood flow velocity. Crit Care Med
2001;29(1):158–63.

106. Tseng MY, Czosnyka M, Richards H, et al. Effects of acute treatment with
pravastatin on cerebral vasospasm, autoregulation, and delayed ischemic
deficits after aneurysmal subarachnoid hemorrhage: a phase II randomized
placebo-controlled trial. Stroke 2005;36(8):1627–32.
107. Rajajee V, Fletcher JJ, Pandey AS, et al. Low pulsatility index on transcranial
doppler predicts symptomatic large vessel vasospasm after aneurysmal
sub-arachnoid hemorrhage. Neurosurgery 2011;Nov 14. Epub ahead of print.
108. Washington CW, Zipfel GJ. Participants in the International Multi-
disciplinary Consensus Conference on the critical care management of
subarachnoid hemorrhage. Detection and monitoring of vasospasm and
delayed cerebral ischemia: a review and assessment of the literature.
Neurocrit Care 2011;15(2):312–7.
109. Grosset DG, Straiton J, McDonald I, et al. Use of transcranial Doppler
sonography to predict development of a delayed ischemic deficit after
subarachnoid hemorrhage. J Neurosurg 1993;78(2):183–7.
110. Vora YY, Suarez-Almazor M, Steinke DE, et al. Role of transcranial Doppler
monitoring in the diagnosis of cerebral vasospasm after subarachnoid
hemorrhage. Neurosurgery 1999;44(6):1237–47.
111. Krejza J, Szydlik P, Liebeskind DS, et al. Age and sex variability and normal
reference values for the V(MCA)/V(ICA) index. AJNR Am J Neuroradiol
2005;26(4):730–5.
112. Swiercz M, Kochanowicz J, Weigele J, et al. Learning vector quantization
neural networks improve accuracy of transcranial color-coded duplex
sonography in detection of middle cerebral artery spasm-preliminary
report. Neuroinformatics 2008;6(4):279–90.
113. Soustiel JF, Bruk B, Shik B, et al. Transcranial Doppler in vertebrobasilar
vasospasm after subarachnoid hemorrhage. Neurosurgery 1998;43(2):
282–91.
114. Newell DW, Grady MS, Eskridge JM, et al. Distribution of angiographic
vasospasm after subarachnoid hemorrhage: implications for diagnosis by
transcranial Doppler ultrasonography. Neurosurgery 1990;27(4):574–7.
115. Sloan MA, Haley Jr EC, Kassell NF, et al. Sensitivity and specificity of
transcranial Doppler ultrasonography in the diagnosis of vasospasm
following subarachnoid hemorrhage. Neurology 1989;39(11):1514–18.
116. Allendoerfer J, Goertler M, von Reutern GM. Prognostic relevance of
ultra-early Doppler sonography in acute ischaemic stroke: a prospective
multicentre study. Lancet Neurol 2006;5(10):835–40.
117. Gerriets T, Goertler M, Stolz E, et al. Feasibility and validity of transcranial
duplex sonography in patients with acute stroke. J Neurol Neurosurg
Psychiatry 2002;73(1):17–20.
118. Malferrari G, Bertolino C, Casoni F, et al. The eligible study: ultrasound
assessment in acute ischemic stroke within 3 hours. Cerebrovasc Dis
2007;24(5):469–76.
119. Sekoranja L, Loulidi J, Yilmaz H, et al. Intravenous versus combined
(intravenous and intra-arterial) thrombolysis in acute ischemic stroke: a
transcranial color-coded duplex sonography–guided pilot study. Stroke
2006;37(7):1805–9.
120. Wong KS, Li H, Chan YL, et al. Use of transcranial Doppler ultrasound to
predict outcome in patients with intracranial large-artery occlusive disease.
Stroke 2000;31(11):2641–7.
121. Wunderlich MT, Stolz E, Seidel G, et al. Conservative medical treatment
and intravenous thrombolysis in acute stroke from carotid T occlusion.
Cerebrovasc Dis 2005;20(5):355–61.
122. Wunderlich MT, Goertler M, Postert T, et al. Recanalization after
intravenous thrombolysis: does a recanalization time window exist?
Neurology 2007;68(17):1364–8.
123. Alexandrov AV, Molina CA, Grotta JC, et al. Ultrasound-enhanced systemic
thrombolysis for acute ischemic stroke. N Engl J Med 2004;351(21):2170–8.
124. Demchuk AM, Burgin WS, Christou I, et al. Thrombolysis in brain
ischemia (TIBI) transcranial Doppler flow grades predict clinical severity,
early recovery, and mortality in patients treated with intravenous tissue
plasminogen activator. Stroke 2001;32(1):89–93.
125. Molina CA, Ribo M, Rubiera M, et al. Microbubble administration
accelerates clot lysis during continuous 2-MHz ultrasound monitoring in
stroke patients treated with intravenous tissue plasminogen activator.
Stroke 2006;37(2):425–9.
126. Hackworthy RA, Sorensen SG, Fitzpatrick PG, et al. Dependence of
assessment of coronary artery reperfusion during acute myocardial
infarction on angiographic criteria and interobserver variability. Am J
Cardiol 1988;62(9):538–42.
127. Gahn G, Gerber J, Hallmeyer S, et al. Contrast-enhanced transcranial
color-coded duplex sonography in stroke patients with limited bone
windows. AJNR Am J Neuroradiol 2000;21(3):509–14.
128. Gerriets T, Postert T, Goertler M, et al. DIAS I: duplex-sonographic
assessment of the cerebrovascular status in acute stroke—a useful tool for
future stroke trials. Stroke 2000;31(10):2342–5.
Section V—Cerebral Blood Flow 313.e3
129. Kenton AR, Martin PJ, Abbott RJ, et al. Comparison of transcranial
color-coded sonography and magnetic resonance angiography in acute
stroke. Stroke 1997;28(8):1601–6.
130. Baumgartner RW, Mattle HP, Schroth G. Assessment of >/=50% and <50%
intracranial stenoses by transcranial color-coded duplex sonography. Stroke
1999;30(1):87–92.
131. Gortler M, Niethammer R, Widder B. Differentiating subtotal carotid artery
stenoses from occlusions by colour-coded duplex sonography. J Neurol
1994;241(5):301–5.
132. Seidel G, Gerriets T, Kaps M, et al. Dislocation of the third ventricle due to
space-occupying stroke evaluated by transcranial duplex sonography.
J Neuroimaging 1996;6(4):227–30.
133. Stolz E, Gerriets T, Fiss I, et al. Comparison of transcranial color-coded
duplex sonography and cranial CT measurements for determining third
ventricle midline shift in space-occupying stroke. AJNR Am J Neuroradiol
1999;20(8):1567–71.
134. Gerriets T, Stolz E, Modrau B, et al. Sonographic monitoring of midline
shift in hemispheric infarctions. Neurology 1999;52(1):45–9.
135. Krejza J, Mariak Z, Bert RJ. Transcranial colour Doppler sonography in
emergency management of intracerebral haemorrhage caused by an
arteriovenous malformation: case report. Neuroradiology
2000;42(12):900–4.
136. Becker G, Winkler J, Hofmann E, et al. Differentiation between ischemic
and hemorrhagic stroke by transcranial color-coded real-time sonography.
J Neuroimaging 1993;3(1):41–7.
137. Seidel G, Kaps M, Dorndorf W. Transcranial color-coded duplex
sonography of intracerebral hematomas in adults. Stroke
1993;24(10):1519–27.
138. Maurer M, Shambal S, Berg D, et al. Differentiation between intracerebral
hemorrhage and ischemic stroke by transcranial color-coded duplex-
sonography. Stroke 1998;29(12):2563–7.
139. Baumgartner RW, Mattle HP, Schroth G. Transcranial colour-coded duplex
sonography of cerebral arteriovenous malformations. Neuroradiology 1996;
38(8):734–7.
140. Griewing B, Motsch L, Piek J, et al. Transcranial power mode Doppler
duplex sonography of intracranial aneurysms. J Neuroimaging 1998;8(3):
155–8.
141. Klotzsch C, Bozzato A, Lammers G, et al. Three-dimensional transcranial
color-coded sonography of cerebral aneurysms. Stroke 1999;30(11):
2285–90.
142. Aries MJ, Elting JW, De Keyser J, et al. Cerebral autoregulation in stroke: a
review of transcranial Doppler studies. Stroke 2010;41(11):2697–704. Epub
2010, Oct 7.
143. Eames PJ, Blake MJ, Dawson SL, et al. Dynamic cerebral autoregulation and
beat to beat blood pressure control are impaired in acute ischaemic stroke.
J Neurol Neurosurg Psychiatry 2002;72(4):467–72.
144. Reinhard M, Wihler C, Roth M, et al. Cerebral autoregulation dynamics in
acute ischemic stroke after rtPA thrombolysis. Cerebrovasc Dis 2008;26(2):
147–55.
145. Reinhard M, Rutsch S, Lambeck J, et al. Dynamic cerebral autoregulation
associates with infarct size and outcome after ischemic stroke. Acta Neurol
Scand 2012;125(3):156–62. Epub 2011, Apr 6.
146. Gosling RG, King DH. Arterial assessment by Doppler-shift ultrasound.
Proc R Soc Med 1974;67(6 Pt 1):447–9.
147. Derdeyn CP, Grubb Jr RL, Powers WJ. Cerebral hemodynamic impairment:
methods of measurement and association with stroke risk. Neurology 1999;
53(2):251–9.
148. Powers WJ. Cerebral hemodynamics in ischemic cerebrovascular disease.
Ann Neurol 1991;29(3):231–40.
149. Kuroda S, Houkin K, Kamiyama H, et al. Long-term prognosis of medically
treated patients with internal carotid or middle cerebral artery occlusion:
can acetazolamide test predict it? Stroke 2001;32(9):2110–16.
150. King A, Serena J, Bornstein NM, et al. Does impaired cerebrovascular
reactivity predict stroke risk in asymptomatic carotid stenosis? A
prospective substudy of the asymptomatic carotid emboli study. Stroke
2011;42(6):1550–5
151. Failure of extracranial-intracranial arterial bypass to reduce the risk of
ischemic stroke. Results of an international randomized trial. The EC/IC
Bypass Study Group. N Engl J Med 1985;313(19):1191–200.
152. Adams HP, Jr., Powers WJ, Grubb RL, Jr., et al. Preview of a new trial of
extracranial-to-intracranial arterial anastomosis: the carotid occlusion
surgery study. Neurosurg Clin N Am 2001;12(3):613–61x.
153. Karnik R, Valentin A, Ammerer HP, et al. Evaluation of vasomotor
reactivity by transcranial Doppler and acetazolamide test before and after
extracranial-intracranial bypass in patients with internal carotid artery
occlusion. Stroke 1992;23(6):812–17.
313.e4 Section V—Cerebral Blood Flow
154. Vorstrup S, Brun B, Lassen NA. Evaluation of the cerebral vasodilatory
capacity by the acetazolamide test before EC-IC bypass surgery in patients
with occlusion of the internal carotid artery. Stroke 1986;17(6):1291–8.
155. Kuroda S, Kamiyama H, Abe H, et al. Acetazolamide test in detecting
reduced cerebral perfusion reserve and predicting long-term prognosis in
patients with internal carotid artery occlusion. Neurosurgery 1993;32(6):
912–18.
156. Schmiedek P, Piepgras A, Leinsinger G, et al. Improvement of
cerebrovascular reserve capacity by EC-IC arterial bypass surgery in
patients with ICA occlusion and hemodynamic cerebral ischemia.
J Neurosurg 1994;81(2):236–44.
157. European Stroke Organisation (ESO) Executive Committee; ESO Writing
Committee. Guidelines for management of ischaemic stroke and transient
ischaemic attack 2008. Cerebrovasc Dis 2008;25(5):457–507.
158. Adams HP, del Zoppo G, Alberts MJ, et al. Guidelines for the early
management of adults with ischemic stroke: a guideline from the American
Heart Association/American Stroke Association Stroke Council, Clinical
Cardiology Council, Cardiovascular Radiology and Intervention Council,
and the Atherosclerotic Peripheral Vascular Disease and Quality of Care
Outcomes in Research Interdisciplinary Working Groups. Stroke 2007;
38(5):1655–711.
159. Goldstein LB, Bushnell CD, Adams RJ, et al. Guidelines for the primary
prevention of stroke: a guideline for healthcare professionals from the
American Heart Association/American Stroke Association. Stroke 2011;
42(2):517–84.
160. Furie KL, Kasner SE, Adams RJ, et al. Guidelines for the prevention of
stroke in patients with stroke or transient ischemic attack: a guideline for
healthcare professionals from the American heart association/American
stroke association. Stroke 2011;42(1):227–76.
161. Iwama T, Hashimoto N, Takagi Y, et al. Predictability of extracranial/
intracranial bypass function: a retrospective study of patients with occlusive
cerebrovascular disease. Neurosurgery 1997;40(1):53–9.
162. Nariai T, Suzuki R, Matsushima Y, et al. Surgically induced angiogenesis to
compensate for hemodynamic cerebral ischemia. Stroke 1994;25(5):1014–21.
163. Ogasawara K, Ogawa A, Terasaki K, et al. Use of cerebrovascular reactivity
in patients with symptomatic major cerebral artery occlusion to predict
5-year outcome: comparison of xenon-133 and iodine-123-IMP single-
photon emission computed tomography. J Cereb Blood Flow Metab 2002;
22(9):1142–8.
164. Widder B, Kleiser B, Krapf H. Course of cerebrovascular reactivity in
patients with carotid artery occlusions. Stroke 1994;25(10):1963–7.
165. Henderson RD, Phan TG, Piepgras DG, et al. Mechanisms of intracerebral
hemorrhage after carotid endarterectomy. J Neurosurg 2001;95(6):964–69.
166. Piepgras DG, Morgan MK, Sundt TM, Jr., et al. Intracerebral hemorrhage
after carotid endarterectomy. J Neurosurg 1988;68(4):532–6.
167. Zhang R, Zhou G, Xu G, et al. Posterior circulation hyperperfusion
syndrome after bilateral vertebral artery intracranial stenting. Ann Vasc
Surg 2009;23(5):686.e1–5.
168. Sfyroeras GS, Karkos CD, Arsos G, et al. Cerebral hyperperfusion after
carotid stenting: a transcranial Doppler and SPECT study. Vasc
Endovascular Surg 2009;43(2):150–6
169. Hosoda K, Kawaguchi T, Ishii K, et al. Prediction of hyperperfusion after
carotid endarterectomy by brain SPECT analysis with semiquantitative
statistical mapping method. Stroke 2003;34(5):1187–93.
170. Reigel MM, Hollier LH, Kazmier FJ, et al. Late survival in abdominal aortic
aneurysm patients: the role of selective myocardial revascularization on the
basis of clinical symptoms. J Vasc Surg 1987;5(2):222–7.
171. Pennekamp CW, Tromp SC, Ackerstaff RG, et al. Prediction of cerebral
hyperperfusion after carotid endarterectomy with transcranial Doppler. Eur
J Vasc Endovasc Surg 2012;Jan 18. Epub ahead of print.
172. Jain R, Hoeffner EG, Deveikis JP, et al. Carotid perfusion CT with balloon
occlusion and acetazolamide challenge test: feasibility. Radiology 2004;
231(3):906–13.
173. Okudaira Y, Arai H, Sato K. Cerebral blood flow alteration by
acetazolamide during carotid balloon occlusion: parameters reflecting
cerebral perfusion pressure in the acetazolamide test. Stroke 1996;27(4):
617–21.
174. Mokri B. The Monro-Kellie hypothesis: applications in CSF volume
depletion. Neurology 2001;56(12):1746–8.
175. Kingman TA, Mendelow AD, Graham DI, et al. Experimental intracerebral
mass: time-related effects on local cerebral blood flow. J Neurosurg 1987;
67(5):732–8.
176. Sahay KB, Mehrotra R, Sachdeva U, et al. Elastomechanical characterization
of brain tissues. J Biomech 1992;25(3):319–26.
177. Johnston IH, Rowan JO. Raised intracranial pressure and cerebral blood
flow. 3. Venous outflow tract pressures and vascular resistances in
experimental intracranial hypertension. J Neurol Neurosurg Psychiatry
1974;37(4):392–402.
178. Hadani M, Bruk B, Ram Z, et al. Application of transcranial doppler
ultrasonography for the diagnosis of brain death. Intensive Care Med 1999;
25(8):822–8.
179. Marinoni M, Alari F, Mastronardi V, et al. The relevance of early TCD
monitoring in the intensive care units for the confirming of brain death
diagnosis. Neurol Sci 2011;32(1):73–7. Epub 2010, Sep 25.
180. Sharma D, Souter MJ, Moore AE, et al. Clinical experience with transcranial
Doppler ultrasonography as a confirmatory test for brain death: a
retrospective analysis. Neurocrit Care 2011;14(3):370–6.
181. Soldatos T, Karakitsos D, Wachtel M, et al. The value of transcranial
Doppler sonography with a transorbital approach in the confirmation of
cerebral circulatory arrest. Transplant Proc 2010;42(5):1502–6.
182. Ducrocq X, Hassler W, Moritake K, et al. Consensus opinion on diagnosis
of cerebral circulatory arrest using Doppler-sonography: Task Force Group
on cerebral death of the Neurosonology Research Group of the World
Federation of Neurology. J Neurol Sci 1998;159(2):145–50.
183. Krejza J, Fryc J, Owlasiuk M, et al. Transcranial color Doppler sonography
in the emergency diagnosis of middle cerebral artery occlusion in a patient
after head injury. Clin Imaging 2001;25(2):90–4.
184. Llompart-Pou JA, Abadal JM, Velasco J, et al. Contrast-enhanced
transcranial color sonography in the diagnosis of cerebral circulatory arrest.
Transplant Proc 2009;41(5):1466–8
185. Sorrentino E, Budohoski KP, Kasprowicz M, et al. Critical thresholds for
transcranial Doppler indices of cerebral autoregulation in traumatic brain
injury. Neurocrit Care 2011;14(2):188–93.
186. Melo JR, Di Rocco F, Blanot S, et al. Transcranial Doppler can predict
intracranial hypertension in children with severe traumatic brain injuries.
Childs Nerv Syst 2011;27(6):979–84. Epub 2011, Jan 5.
187. Martin NA, Doberstein C, Zane C, et al. Posttraumatic cerebral arterial
spasm: transcranial Doppler ultrasound, cerebral blood flow, and
angiographic findings. J Neurosurg 1992;77(4):575–83.
188. Steiger HJ, Aaslid R, Stooss R, et al. Transcranial Doppler monitoring in
head injury: relations between type of injury, flow velocities, vasoreactivity,
and outcome. Neurosurgery 1994;34(1):79–85.
189. Zurynski YA, Dorsch NW, Fearnside MR. Incidence and effects of increased
cerebral blood flow velocity after severe head injury: a transcranial Doppler
ultrasound study. II. Effect of vasospasm and hyperemia on outcome.
J Neurol Sci 1995;134(1-2):41–6.
190. Kochanowicz J, Krejza J, Mariak Z, et al. Detection and monitoring of
cerebral hemodynamic disturbances with transcranial color-coded duplex
sonography in patients after head injury. Neuroradiology 2006;48(1):
31–6.
191. Len TK, Neary JP. Cerebrovascular pathophysiology following mild
traumatic brain injury. Clin Physiol Funct Imaging 2011;31(2):85–93.
192. Brandi G, Béchir M, Sailer S, et al. Transcranial color-coded duplex
sonography allows assessing cerebral perfusion pressure noninvasively
following severe traumatic brain injury. Acta Neurochir (Wien)
2010;152(6):965–72.
193. Czosnyka M, Smielewski P, Kirkpatrick P, et al. Continuous assessment of
the cerebral vasomotor reactivity in head injury. Neurosurgery
1997;41(1):11–7.
194. Newell DW, Weber JP, Watson R, et al. Effect of transient moderate
hyperventilation on dynamic cerebral autoregulation after severe head
injury. Neurosurgery 1996;39(1):35–43.
195. Budohoski KP, Reinhard M, Aries MJ, et al. Monitoring cerebral
autoregulation after head injury. Which component of transcranial Doppler
flow velocity is optimal? Neurocrit Care 2011;Jun 21. Epub ahead of print.
196. Smielewski P, Czosnyka M, Kirkpatrick P, et al. Evaluation of the transient
hyperemic response test in head-injured patients. J Neurosurg 1997;86(5):
773–8.
197. Muizelaar JP, Lutz HA, III, Becker DP. Effect of mannitol on ICP and CBF
and correlation with pressure autoregulation in severely head-injured
patients. J Neurosurg 1984;61(4):700–6.
198. Strebel S, Lam AM, Matta BF, et al. Impaired cerebral autoregulation after
mild brain injury. Surg Neurol 1997;47(2):128–31.
199. Larsen FS, Ejlersen E, Hansen BA, et al. Functional loss of cerebral blood
flow autoregulation in patients with fulminant hepatic failure. J Hepatol
1995;23(2):212–17.
200. Larsen FS, Olsen KS, Ejlersen E, et al. Cerebral blood flow autoregulation
and transcranial Doppler sonography in patients with cirrhosis. Hepatology
1995;22(3):730–6.
201. Strauss GI, Hansen BA, Herzog T, et al. Cerebral autoregulation in patients
with end-stage liver disease. Eur J Gastroenterol Hepatol
2000;12(7):767–71.

202. Larsen FS, Ejlersen E, Strauss G, et al. Cerebrovascular metabolic
autoregulation is impaired during liver transplantation. Transplantation
1999;68(10):1472–6.
203. Larsen FS, Pott F, Hansen BA, et al. Transcranial Doppler sonography may
predict brain death in patients with fulminant hepatic failure. Transplant
Proc 1995;27(6):3510–11.
204. de Riva N, Budohoski KP, Smielewski P, et al. Transcranial Doppler
pulsatility index: what it is and what it isn’t. Neurocrit Care 2012;Feb 4.
Epub ahead of print.
205. King A, Shipley M, Markus H, et al. Optimizing protocols for risk
prediction in asymptomatic carotid stenosis using embolic signal detection:
the Asymptomatic Carotid Emboli Study. Stroke 2011;42(10):2819–24.
Epub 2011, Aug 18.
206. King A, Bath PM, Markus HS. Clopidogrel versus dipyridamole in addition
to aspirin in reducing embolization detected with ambulatory transcranial
Doppler: a randomized trial. Stroke 2011 Mar;42(3):650–5. Epub 2011,
Jan 21.
207. Executive Committee for the Asymptomatic Carotid Atherosclerosis Study.
Endarterectomy for asymptomatic carotid artery stenosis. JAMA 1995;
273(18):1421–8.
208. Hartmann A, Hupp T, Koch HC, et al. Prospective study on the
complication rate of carotid surgery. Cerebrovasc Dis 1999;9(3):152–6.
209. Spencer MP. Transcranial Doppler monitoring and causes of stroke from
carotid endarterectomy. Stroke 1997;28(4):685–91.
210. Ackerstaff RG, Moons KG, van de Vlasakker CJ, et al. Association of
intraoperative transcranial doppler monitoring variables with stroke from
carotid endarterectomy. Stroke 2000;31(8):1817–23.
211. Kofke WA, Brauer P, Policare R, et al. Middle cerebral artery blood flow
velocity and stable xenon-enhanced computed tomographic blood flow
during balloon test occlusion of the internal carotid artery. Stroke 1995;
26(9):1603–6.
212. Halsey Jr JH. Risks and benefits of shunting in carotid endarterectomy. The
International Transcranial Doppler Collaborators. Stroke 1992;23(11):
1583–7.
213. Hayes PD, Vainas T, Hartley S, et al. The Pruitt-Inahara shunt
maintains mean middle cerebral artery velocities within 10% of
preoperative values during carotid endarterectomy. J Vasc Surg 2000;
32(2):299–306.
214. Babikian VL, Cantelmo NL, Wijman CAC. Neurovascular monitoring
during carotid endarterectomy. In: Babikian VL, Wechsler LR, editors.
Transcranial Doppler ultrasonography. 2nd ed. Boston: Butterworth-
Heinemann; 1999. p. 231–45.
215. Cao P, Giordano G, Zannetti S, et al. Transcranial Doppler monitoring
during carotid endarterectomy: is it appropriate for selecting patients in
need of a shunt? J Vasc Surg 1997;26(6):973–9.
216. Frawley JE, Hicks RG, Gray LJ, et al. Carotid endarterectomy without a
shunt for symptomatic lesions associated with contralateral severe stenosis
or occlusion. J Vasc Surg 1996;23(3):421–7.
217. Levi CR, O’Malley HM, Fell G, et al. Transcranial Doppler detected cerebral
microembolism following carotid endarterectomy. High microembolic
signal loads predict postoperative cerebral ischaemia. Brain 1997;
120 (Pt 4):621–9.
218. Cantelmo NL, Babikian VL, Samaraweera RN, et al. Cerebral
microembolism and ischemic changes associated with carotid
endarterectomy. J Vasc Surg 1998;27(6):1024–30.
219. Jansen C, Ramos LM, van Heesewijk JP, et al. Impact of microembolism
and hemodynamic changes in the brain during carotid endarterectomy.
Stroke 1994;25(5):992–7.
220. Lennard N, Smith J, Dumville J, et al. Prevention of postoperative
thrombotic stroke after carotid endarterectomy: the role of transcranial
Doppler ultrasound. J Vasc Surg 1997;26(4):579–84.
221. Jordan WD, Jr., Voellinger DC, Doblar DD, et al. Microemboli detected by
transcranial Doppler monitoring in patients during carotid angioplasty
versus carotid endarterectomy. Cardiovasc Surg 1999;7(1):33–8.
222. Gupta N, Corriere MA, Dodson TF, et al. The incidence of microemboli to
the brain is less with endarterectomy than with percutaneous
revascularization with distal filters or flow reversal. J Vasc Surg 2011;53(2):
316–22. Epub 2010, Dec 3.
223. Reimers B, Corvaja N, Moshiri S, et al. Cerebral protection with filter
devices during carotid artery stenting. Circulation 2001;104(1):12–5.
224. Jeong HS, Song HJ, Lee JH, et al. Interpretation of TCD spectral patterns
detected during carotid artery stent interventions. J Endovasc Ther 2011;
18(4):518–26.
225. Antonius Carotid Endaterectomy, Angioplasty, and Stenting Study Group.
Transcranial Doppler monitoring in angioplasty and stenting of the carotid
bifurcation. J Endovasc Ther 2003;10(4):702–10.
Section V—Cerebral Blood Flow 313.e5
226. Ackerstaff RG, Suttorp MJ, van den Berg JC, et al. Prediction of early
cerebral outcome by transcranial Doppler monitoring in carotid bifurcation
angioplasty and stenting. J Vasc Surg 2005;41(4):618–24.
227. Rosenkranz M, Fiehler J, Niesen W, et al. The amount of solid cerebral
microemboli during carotid stenting does not relate to the frequency of
silent ischemic lesions. AJNR Am J Neuroradiol 2006;27(1):157–61.
228. Sawada M, Hashimoto N, Nishi S, et al. Detection of embolic signals during
and after percutaneous transluminal angioplasty of subclavian and
vertebral arteries using transcranial Doppler ultrasonography. Neurosurgery
1997;41(3):535–40.
229. Pelz DM, Lownie SP, Fox AJ. Thromboembolic events associated with the
treatment of cerebral aneurysms with Guglielmi detachable coils. AJNR Am
J Neuroradiol 1998;19(8):1541–7.
230. Song JK, Niimi Y, Fernandez PM, et al. Thrombus formation during
intracranial aneurysm coil placement: treatment with intra-arterial
abciximab. AJNR Am J Neuroradiol 2004;25(7):1147–53.
231. Ries T, Buhk JH, Kucinski T, et al. Intravenous administration of
acetylsalicylic acid during endovascular treatment of cerebral aneurysms
reduces the rate of thromboembolic events. Stroke 2006;37(7):1816–21.
232. Gilman S. Cerebral disorders after open-heart operations. N Engl J Med
1965;272:489–98.
233. McKhann GM, Goldsborough MA, Borowicz Jr LM, et al. Cognitive
outcome after coronary artery bypass: a one-year prospective study. Ann
Thorac Surg 1997;63(2):510–15.
234. Blauth CI, Smith PL, Arnold JV, et al. Influence of oxygenator type on the
prevalence and extent of microembolic retinal ischemia during
cardiopulmonary bypass. Assessment by digital image analysis. J Thorac
Cardiovasc Surg 1990;99(1):61–9.
235. Gravlee GP, Hudspeth AS, Toole JF. Bilateral brachial paralysis from
watershed infarction after coronary artery bypass. A report of two cases and
review of the predisposing anatomic and physiological mechanisms.
J Thorac Cardiovasc Surg 1984;88(5 Pt 1):742–7.
236. Padayachee TS, Parsons S, Theobold R, et al. The detection of microemboli
in the middle cerebral artery during cardiopulmonary bypass: a
transcranial Doppler ultrasound investigation using membrane and bubble
oxygenators. Ann Thorac Surg 1987;44(3):298–302.
237. Pugsley W, Klinger L, Paschalis C, et al. The impact of microemboli during
cardiopulmonary bypass on neuropsychological functioning. Stroke 1994;
25(7):1393–9.
238. Vanninen R, Aikia M, Kononen M, et al. Subclinical cerebral complications
after coronary artery bypass grafting: prospective analysis with magnetic
resonance imaging, quantitative electroencephalography, and
neuropsychological assessment. Arch Neurol 1998;55(5):618–27.
239. Wityk RJ, Goldsborough MA, Hillis A, et al. Diffusion- and perfusion-
weighted brain magnetic resonance imaging in patients with neurologic
complications after cardiac surgery. Arch Neurol 2001;58(4):571–6.
240. Clark RE, Brillman J, Davis DA, et al. Microemboli during coronary artery
bypass grafting: genesis and effect on outcome. J Thorac Cardiovasc Surg
1995;109(2):249–57.
241. Challa VR, Moody DM, Bell MA. Small capillary and arteriolar dilatations
(SCADS) in brain after cardiopulmonary bypass. In: McKhan GM, Gibbs
CJ, editors. Neurological outcomes after coronary bypass surgery. New
York: Thieme; 1993. p. 330–4.
242. Moody DM, Bell MA, Challa VR, et al. Brain microemboli during cardiac
surgery or aortography. Ann Neurol 1990;28(4):477–86.
243. Deverall PB, Padayachee TS, Parsons S, et al. Ultrasound detection of
micro-emboli in the middle cerebral artery during cardiopulmonary bypass
surgery. Eur J Cardiothorac Surg 1988;2(4):256–60.
244. Barbut D, Hinton RB, Szatrowski TP, et al. Cerebral emboli detected during
bypass surgery are associated with clamp removal. Stroke
1994;25(12):2398–402.
245. Barbut D, Yao FS, Hager DN, et al. Comparison of transcranial Doppler
ultrasonography and transesophageal echocardiography to monitor emboli
during coronary artery bypass surgery. Stroke 1996;27(1):87–90.
246. Yao FS, Barbut D, Hager DN, et al. Detection of aortic emboli by
transesophageal echocardiography during coronary artery bypass surgery.
J Cardiothorac Vasc Anesth 1996;10(3):314–17.
247. Barbut D, Caplan LR. Brain complications of cardiac surgery. Curr Probl
Cardiol 1997;22(9):449–80.
248. Gardner TJ, Horneffer PJ, Manolio TA, et al. Stroke following coronary
artery bypass grafting: a ten-year study. Ann Thorac Surg
1985;40(6):574–81.
249. Roach GW, Kanchuger M, Mangano CM, et al. Adverse cerebral outcomes
after coronary bypass surgery. Multicenter Study of Perioperative Ischemia
Research Group and the Ischemia Research and Education Foundation
Investigators. N Engl J Med 1996;335(25):1857–63.
313.e6 Section V—Cerebral Blood Flow
250. Caplan LR. Protecting the brains of patients after heart surgery. Arch
Neurol 2001;58(4):549–50.
251. Blinc A, Kennedy SD, Bryant RG, et al. Flow through clots determines the
rate and pattern of fibrinolysis. Thromb Haemost 1994;71(2):230–5.
252. Francis CW, Blinc A, Lee S, et al. Ultrasound accelerates transport of
recombinant tissue plasminogen activator into clots. Ultrasound Med Biol
1995;21(3):419–24.
253. Behrens S, Daffertshofer M, Spiegel D, et al. Low-frequency, low-intensity
ultrasound accelerates thrombolysis through the skull. Ultrasound Med
Biol 1999;25(2):269–73.
254. Akiyama M, Ishibashi T, Yamada T, et al. Low-frequency ultrasound
penetrates the cranium and enhances thrombolysis in vitro. Neurosurgery
1998;43(4):828–32.
255. Spengos K, Behrens S, Daffertshofer M, et al. Acceleration of thrombolysis
with ultrasound through the cranium in a flow model. Ultrasound Med
Biol 2000;26(5):889–95.
256. Kramer C, Aguilar MI, Hoffman-Snyder C, et al. Safety and efficacy of
ultrasound-enhanced thrombolysis in the treatment of acute middle
cerebral artery infarction: a critically appraised topic. Neurologist 2011;
17(6):346–51.
257. Rubiera M, Alexandrov AV. Sonothrombolysis in the management of acute
ischemic stroke. Am J Cardiovasc Drugs 2010;10(1):5–10.
258. Tsivgoulis G, Eggers J, Ribo M, et al. Safety and efficacy of ultrasound-
enhanced thrombolysis: a comprehensive review and meta-analysis of
randomized and nonrandomized studies. Stroke 2010;41(2):280–7. Epub
2009, Dec 31.
259. Alexandrov AV, Demchuk AM, Felberg RA, et al. High rate of complete
recanalization and dramatic clinical recovery during tPA infusion when
continuously monitored with 2-MHz transcranial Doppler monitoring.
Stroke 2000;31(3):610–14.
260. Siddiqi F, Odrljin TM, Fay PJ, et al. Binding of tissue-plasminogen activator
to fibrin: effect of ultrasound. Blood 1998;91(6):2019–25.
261. Siddiqi F, Blinc A, Braaten J, et al. Ultrasound increases flow through fibrin
gels. Thromb Haemost 1995;73(3):495–8.
262. Polak JF. Ultrasound energy and the dissolution of thrombus. N Engl J Med
2004;351(21):2154–5.
263. Krejza J, Weigele JB, Alokaili R, et al. Sonothrombolysis in acute ischemic
stroke for patients ineligible for rt-PA. Neurology 2006;66(1):154–5.
264. Tachibana K, Tachibana S. Albumin microbubble echo-contrast material as
an enhancer for ultrasound accelerated thrombolysis. Circulation 1995;
92(5):1148–50.
265. Molina CA, Ribo M, Rubiera M, et al. Microbubble administration
accelerates clot lysis during continuous 2-MHz ultrasound monitoring in
stroke patients treated with intravenous tissue plasminogen activator.
Stroke 2006;37(2):425–9.
266. Xie F, Tsutsui JM, Lof J, et al. Effectiveness of lipid microbubbles and
ultrasound in declotting thrombosis. Ultrasound Med Biol 2005;31(7):
979–85.
267. Dijkmans PA, Knaapen P, Sieswerda GT, et al. Quantification of myocardial
perfusion using intravenous myocardial contrast echocardiography in
healthy volunteers: comparison with positron emission tomography. J Am
Soc Echocardiogr 2006;19(3):285–93.
268. Cintas P, Nguyen F, Boneu B, et al. Enhancement of enzymatic fibrinolysis
with 2-MHz ultrasound and microbubbles. J Thromb Haemost 2004;2(7):
1163–6.
269. Culp WC, Porter TR, Lowery J, et al. Intracranial clot lysis with intravenous
microbubbles and transcranial ultrasound in swine. Stroke 2004;35(10):
2407–11.
270. Mizushige K, Kondo I, Ohmori K, et al. Enhancement of ultrasound-
accelerated thrombolysis by echo contrast agents: dependence on
microbubble structure. Ultrasound Med Biol 1999;25(9):1431–7.
271. Saqqur M, Tsivgoulis G, Molina CA, et al. Design of a PROspective
multi-national CLOTBUST collaboration on reperfusion therapies for
stroke (CLOTBUST-PRO). Int J Stroke 2008;3(1):66–72.
272. Alexandrov AV, Demchuk AM, Burgin WS, et al. Ultrasound-enhanced
thrombolysis for acute ischemic stroke. Phase I. Findings of the
CLOTBUST trial. J Neuroimaging 2004;14(2):113–17.
273. Rosenschein U, Rozenszajn LA, Kraus L, et al. Ultrasonic angioplasty in
totally occluded peripheral arteries: initial clinical, histological, and
angiographic results. Circulation 1991;83(6):1976–86.
274. Siegel RJ, Cumberland DC, Crew JR. Ultrasound recanalization of diseased
arteries: from experimental studies to clinical application. Surg Clin North
Am 1992;72(4):879–97.
275. Luo H, Steffen W, Cercek B, et al. Enhancement of thrombolysis by external
ultrasound. Am Heart J 1993;125(6):1564–9.
276. Steffen W, Cumberland D, Gaines P, et al. Catheter-delivered high intensity,
low frequency ultrasound induces vasodilation in vivo. Eur Heart J 1994;
15(3):369–76.
277. Rosenschein U, Bernstein JJ, DiSegni E, et al. Experimental ultrasonic
angioplasty: disruption of atherosclerotic plaques and thrombi in vitro and
arterial recanalization in vivo. J Am Coll Cardiol 1990;15(3):711–17.
278. Hong AS, Chae JS, Dubin SB, et al. Ultrasonic clot disruption: an in vitro
study. Am Heart J 1990;120(2):418–22.
279. Sehgal CM, Leveen RF, Shlansky-Goldberg RD. Ultrasound-assisted
thrombolysis. Invest Radiol 1993;28(10):939–43.
280. Suchkova V, Siddiqi FN, Carstensen EL, et al. Enhancement of fibrinolysis
with 40-kHz ultrasound. Circulation 1998;98(10):1030–5.
281. Olsson SB, Johansson B, Nilsson AM, et al. Enhancement of thrombolysis
by ultrasound. Ultrasound Med Biol 1994;20(4):375–82.
282. Blinc A, Francis CW, Trudnowski JL, et al. Characterization of ultrasound-
potentiated fibrinolysis in vitro. Blood 1993;81(10):2636–43.
283. Lauer CG, Burge R, Tang DB, et al. Effect of ultrasound on tissue-type
plasminogen activator-induced thrombolysis. Circulation
1992;86(4):1257–64.
V
Chapter
31  
Laser Doppler Flowmetry,
Thermal Diffusion
Flowmetry, and Orthogonal
Polarizing Spectral Imaging
Frederik A. Pennings
Introduction
Brain functional and structural integrity depends on a con-
stant supply of oxygen and glucose through cerebral blood
flow (CBF) to the cerebral microcirculation. Accordingly,
adequate function of the cerebral microcirculation is a prereq-
uisite for sufficient oxygen (O2) delivery to brain cells. Under
normal conditions, the brain makes adjustments in local CBF
to support metabolic demands and prevent energy failure.
However, if an imbalance occurs between supply (CBF) and
consumption (cerebral metabolic rate of oxygen [CMRO2]),
cerebral ischemia may develop. If CBF falls below a critical
level and oxygen extraction has reached its limit, electrical and
certain metabolic functions are lost, and irreversible brain
damage (infarction) may occur.1 The development of tissue
infarction is related to both the degree and duration of CBF
reduction.2 This important concept implicates that certain
thresholds exist below which infarction occurs and that there
is a limited period of time that ischemia may still be reversible
(“therapeutic window”). Therefore, it may be helpful to have
monitoring instruments that can detect cerebral ischemia
before these thresholds are crossed and allow the clinician to
take measures in a timely manner to prevent the development
of brain infarction. In addition instruments are needed that
provide information on morphology and function of the
cerebral microcirculation to better understand the pathophys-
iology of cerebral ischemia. In particular, continuous assess-
ment of CBF helps to assess cerebral autoregulation and
pressure-reactivity. This is important because both can be
independent factors associated with outcome after acute brain
injury.3
Current treatment modalities to prevent or reduce ischemic
brain damage are mostly guided by parameters such as intra-
cranial pressure (ICP), cerebral perfusion pressure (CPP),
brain tissue oxygen tension (PbtO2), and cerebral metabolism
and not by direct CBF measurements. Direct monitoring of
CBF could provide the opportunity to diagnose and to correct
insufficient CBF before deficits in tissue oxygenation and
314
metabolism are recognized. Local microvascular CBF can be
measured continuously and invasively in the neurocritical
care unit (NCCU) with laser Doppler flowmetry (LDF) and
thermal diffusion flowmetry (TDF). Many years of clinical
experience have been obtained with these monitoring modali-
ties. This chapter gives an overview of LDF and TDF: history,
technology, validation, clinical applications, and develop-
ments are described. In addition, orthogonal polarizing spec-
tral (OPS) imaging and its successor sidestream dark field
(SDF) imaging are discussed as examples of optical techniques
that allow direct observation of the human microcirculatory
morphology and function.
Laser Doppler Flowmetry
Bedside evaluation of CBF has been a long-standing goal in
neurocritical care. Potentially, it could assess flow alterations,
measure the effect of therapy, and evaluate the integrity of
cerebral autoregulation. LDF was first introduced in 1977 to
measure cutaneous blood flow, and after commercialization
in the early 1980s, a stable and steadily growing clinical appli-
cation has occurred.4,5 LDF provides continuous, qualitative
measurements of microvascular perfusion intraoperatively or
in the intensive care unit as a bedside monitoring device; this
is attractive to clinicians working with critical neurologic dis-
eases in which cerebral ischemia is frequently encountered.
However, the routine clinical applicability of this optical tech-
nique remains logistically complex and is hindered by various
artifacts and is limited by the absence of quantitative regional
cerebral blow flow (rCBF) values. Consequently diagnostic
and prognostic values have not been reported and instead LDF
is best used as a trend monitor in individual patients.
In LDF, a fiber-optic laser probe with a diameter of 0.5 to
1 mm is applied to the surface of the brain or within the brain
to illuminate a tissue volume of approximately 1 mm3 with
monochromatic laser light of various wavelengths from 670
to 810 nm. As light impinges on the brain tissue, photons are
scattered and Doppler shifted in a random fashion via dynamic
© Copyright 2013 Elsevier Inc. All rights reserved.

Non-Doppler shifted light
Scattering
tissue target
Photodetector
Doppler shifted light
Laser diode
Fig. 31.1  Optical schematic of laser Doppler flowmetry.
red blood cells and stationary tissue cells. A fraction of scat-
tered shifted and nonshifted photons is detected by photore-
ceptors, which generates an electric signal. This electric signal
contains information about power and frequency propor-
tional to the volume or concentration (fraction of total
photons that is frequency shifted) and velocity of red blood
cells (signal frequency broadening) (Fig. 31.1). The commer-
cially available laser Doppler signal processors use Bonner and
Nossal’s algorithm to analyze the signal and to produce a flow
output. The product of volume and velocity scales linearly
with the perfusion (CBF) and is expressed in arbitrary units
(AUs).6,7
The principles of LDF measurements are implemented in
surface (cortical) probes and intraparenchymal (subcortical)
probes to obtain real-time continuous local microvascular
arbitrary flow values from capillaries at the cortical surface
and within the brain. The cortical probe consists of a thin
contoured disk with a long optical fiber that is connected
to a monitor. The probe is positioned subdurally and so can
be prone to motion artifacts caused by brain pulsations or
subsidence of brain swelling. To avoid disturbances of the
flow values, the probe should be placed in a relatively
normal brain region without large vessels. The intraparen-
chymal probe is introduced through a fixed bolt in the skull,
which eliminates motion artifacts, and advanced to the
white matter. The benefit is that measurements are obtained
from the same local brain region as other intracranial moni-
toring devices such as ICP and microdialysis, which may
improve interpretations of those parameters. The main
manufacturers of laser Doppler instruments are Perimed AB
(Stockholm, Sweden), Moor Instruments Ltd. (Axminster,
UK), Vasamedics Inc. (St. Paul, MN), Transonic Systems Inc.
(Ithaca, NY), Oxford Optronix Ltd (Oxford, UK) and LEA
Medizintechnik (Giessen, Germany).
LDF provides a reliable measurement of microvascular
(local) rCBF as demonstrated by multiple validation studies.
A high correlation is found between LDF and several other
blood flow monitoring techniques such as xenon (Xe) clear-
ance method, radioactive microspheres, hydrogen clearance
technique, iodoantipyrine method, and thermal diffusion.8-14
Furthermore, it has been successfully applied in the brain-
injured patient to assess autoregulation, carbon dioxide (CO2)
reactivity, and response to therapeutic interventions and to
detect ischemic insults.15-18 For example, Schubert et al. have
used LDF studies to detect hypoperfusion independent from
Section V—Cerebral Blood Flow 315
ICP in the early and the acute stages of subarachnoid hemor-
rhage (SAH).19
In the intensive care unit LDF can be used to help predict
outcome. For example, Lam et al. continually assessed auto-
regulation with LDF in 31 comatose head-injured patients. A
linear relationship between cortical LDF and CPP was used as
an indicator for loss of autoregulation. Intact autoregulation
correlated with a good outcome and persistent loss of auto-
regulation with a poor outcome.20 However, loss of contact
with the cerebral surface was responsible for unreliable data
in 16% of patients. Smielewski et al. assessed the transient
hyperemic response after traumatic brain injury using LDF
and found that patients with an unfavorable outcome had
significantly lower LDF readings than patients with a favorable
outcome.21 The concomitant use of LDF with transcranial
Doppler (TCD) to help detect delayed cerebral ischemia con-
firmed that high-flow velocities and ischemia are not often
related.22 LDF also can be used intraoperatively to assess
microvascular flow alterations in a region of interest during
aneurysm surgery, arteriovenous malformation (AVM) resec-
tions, and bypass procedures and to interpret the physiologic
changes associated with the surgery, understand the cortical
microcirculation, and help predict postoperative hyper- or
hypoperfusion syndromes.23-25 In addition LDF studies have
been used to assess blood flow in pericranial flaps during
reconstructive surgery.26
Despite LDF’s excellent temporal and dynamic resolution
with ultra short time responses to fluctuations in CBF there
are several limiting factors that may compromise data
quality.27 Different inter- and intraindividual LDF signals can
be generated by device-related factors such as calibration pro-
cedure, monochromatic wavelength, and probe configura-
tion. The penetration depth of LDF is approximately 1 mm,
depending on wavelength distance between transmitting and
receiving fibers (fiber separation). A shorter wavelength or a
smaller fiber separation will give lesser tissue penetration.
Artifacts can be induced by a large number of external
derangements such as the heterogeneity of microvascular
architecture, changes in hematocrit, tissue or probe motion,
room temperature, strong external light, and sound. LDF
only measures a very small area of brain tissue that may not
necessarily represent the global CBF due to the heterogeneity
of the cerebral microcirculation. Multichannel instruments
and laser Doppler perfusion imaging were introduced to
overcome potential limitations associated with temporal and
spatial resolution.28,29 These factors make inter- and intraindi-
vidual comparison of data difficult. Therefore a detailed stan-
dardized measurement protocol is a prerequisite to obtain
reproducible and comparable LDF data. This protocol
includes at least device-related technical data, anatomic site
selection, and frequent calibrations (every 15 minutes) to cir-
cumvent improper data collection due to changes in hemato-
crit, motion, room temperature, strong external light, and
sound. Finally, the inability of LDF to produce absolute CBF
values hinders the definition of a threshold value that indi-
cates cerebral ischemia; that is to say, LDF is a trend monitor.
Thermal Diffusion Flowmetry
TDF is based on thermal conductivity of brain tissue to
measure quantitative blood flow. In 1933, Gibbs demonstrated
that changes in CBF could be detected with a heated
316 Section V—Cerebral Blood Flow
1 mm
Passive
thermistor
Temperature
Measurement diameter = 8 mm
Active
thermistor 8 mm
A B
Fig. 31.2  A, Schematic of the Hemedex probe illustrating thermal diffusion technology. B, An annotated view of the Hemedex monitor shows 30
minutes of intraoperative cerebral blood flow (CBF) measurements made in the middle cerebral artery (MCA) vascular territory during an EC/IC bypass.
Note the hyperemia after the release of the bypass clamp and the subsequent establishment of a new and higher CBF level. (Courtesy Frank Bowman, PhD.)
thermocouple and Carter and Atkinson developed a mathe-
matical model to calculate absolute rCBF values.30,31 In 1973
TDF was introduced and validation studies showed that rCBF
values obtained by TDF are in good agreement with rCBF
values obtained in xenon-enhanced computed tomography
(Xe-CT) studies and the hydrogen clearance method.32,33
Currently two technical systems are commercially available
that use the temperature difference measured between a
thermistor and temperature sensor to calculate CBF. A small
Silastic probe(s) (Flowtronics, Inc., Phoenix, AZ) with two
gold disks can be placed subdurally on the cortex to measure
superficial rCBF. During placement large vessels and patho-
logic tissue need to be avoided to prevent artifacts. A confi-
dence factor can be calculated by turning off the heater to
determine proper contact with the cerebral cortex. A confi-
dence factor of 0.8 or higher is considered to represent reliable
data. When the correct position is confirmed the dura can be
closed. Recently, an intraparenchymal thermal diffusion
microprobe (Hemedex Inc., Cambridge, MA) has been intro-
duced and validated that eliminates artifacts caused by loss of
surface contact and obtains rCBF measurements from the
same vascular territory as other intraparenchymal monitoring
devices (i.e., ICP, PbtO2).33 This novel TDF microprobe consists
of a thermistor in the tip of the probe and a temperature
sensor located a few millimeters proximal that provides sensi-
tive and continuous absolute rCBF values even in the case of
minor flow changes (Fig. 31.2, A). It is inserted through a bolt
fixed in the skull with the tip approximately 25 mm below the
dura in normal brain tissue, which can be verified with a CT
scan. Automatic recalibration occurs approximately every 30
minutes; this results in a loss of data for 2 to 5 minutes. In
addition, significant baseline shifts of CBF that occur after
each recalibration independent of hemodynamic stability are
a concern.34 Both systems have a built-in safety mechanism
that prevents CBF measurements when temperature is greater
than 39o C. This safety mechanism thus interrupts monitoring
of the cerebral perfusion during periods of high fever that are
frequent in patients with severe closed head injury or sub-
arachnoid hemorrhage.
TDF provides in real time the dynamic changes in actual
cerebral perfusion, that is, quantitative data. Therefore this
Release of
cross clamp
Release of Hyperemia New CBF
bypass clamp
technique can be applied intraoperatively and at the bedside
to prevent, detect, or treat abnormalities in cerebral perfusion
and so alter clinical outcome. Superficial (cortical) rCBF
values between 40 to 70 mL/100 g/min are normal and values
less than 20 mL/100 g/min or greater than 70 mL/100 g/min
represent respectively ischemia and hyperemia.35 The TDF
microprobe measures subcortical white matter perfusion, and
a mean TDF value of 18 to 25 mL/100 g/min is considered
normal.36
In critically ill brain-injured patients, rCBF monitoring with
Silastic TDF sensors can help predict outcome. Specifically,
patients with a significant increase in rCBF from baseline
values, have a good outcome, whereas patients with initial very
low values and those without an increase in flow from baseline
have a poor outcome.35 In this study, however, reliable data
could only be obtained in 37 of 56 patients because the routine
application of this technique is not always easy. rCBF can
complement parameters such as CCP, ICP and PbtO2, and so
facilitate the interpretation of these parameters. In closed head
injury, monitoring of CPP and treatment of low CPP is con-
sidered necessary to prevent cerebral ischemia. However, with
a disrupted autoregulation, a CPP greater than 70 mm Hg may
still be associated with ischemic events, and therefore restora-
tion and maintenance of adequate perfusion could be per-
formed under guidance of continuous rCBF monitoring.37
Likewise use of a Hemedex probe can help evaluate cerebral
autoregulation and CO2 vasoreactivity by allowing for calcula-
tion of local cerebral vascular resistance (CPP/rCBF) in
response to a blood pressure or hyperventilation challenge.38
Alternatively continuous monitoring of TDF-CBF, CPP, and
brain oxygen allows calculation of flow- and oxygen-related
autoregulation indices.39 TDF provides absolute values (unlike
LDF) and therefore has the ability to distinguish between
elevations of ICP associated with ischemia or hyperemia. A
study of SAH and traumatic brain injury patients found a
correlation between CBF measured by the TDF microprobe
and PbtO2, indicating that tissue oxygen levels were determined
in part by regional CBF. This same study showed that loss of
data due to recalibrations or periods of fever is a concern. The
TDF microprobe allowed CBF measurement 64% of the time
when PbtO2 monitoring was possible.34

In SAH, TDF can be used to help detect vasospasm and
monitor for delayed cerebral ischemia.36,40 Vajkoczy et al.
found a threshold of 15 mL/100 g/min to diagnose symptom-
atic vasospasm had a sensitivity of 90% and specificity of
75%.36 In addition, the effectiveness of treatment can be deter-
mined with continuous rCBF monitoring. Therefore TDF
represents a useful and effective technique to help detect
delayed cerebral ischemia in SAH. Whether the use of this
technique in the NCCU can lead to improved clinical outcome
will need further study.
TDF can be used intraoperatively to study hemodynamic
changes that occur in the surrounding normal brain tissue
during aneurysm clipping, bypass surgery (Fig. 31.2, B), or
resections of AVM’s and brain tumors.41-44 Acute changes in
cortical CBF with vascular manipulations during aneurysm
clipping have been demonstrated with TDF. Furthermore,
increased perinidal cortical CBF values were found with
TDF and LDF after AVM resection. Detection of hyperemic
values may indicate pending normal perfusion pressure
breakthrough, and therapeutic interventions (i.e., barbiturate
coma) can be instituted promptly to avoid further brain tissue
damage. A decrease in peritumoral CBF was found before
tumor removal, which recovered after completion of the
resection, indicating that compression of the peritumoral
microvasculature was responsible for a decrease in blood flow.
During epilepsy surgery, it was demonstrated that normaliza-
tion of peri-ictal bihemispheric cortical CBF occurs in tem-
poral lobe epilepsy from a reduction of cortical CBF
interictally.45,46
Limitations associated with TDF include “nonmonitroing”
during recalibration and fever, sensor displacement or loss of
brain surface contact, and that only a small area of brain tissue
is monitored (20 to 30 mm3). Therefore global ischemia or
focal ischemia in other unmeasured flow regions will be
missed. This limitation may be overcome partially by the use
of multiple sensors on a strip or placement of multiple intra-
parenchymal microprobes. Complications associated with
TDF microprobes are very rare and similar to other implanted
intraparenchymal probes (i.e., an approximate 1% risk of sig-
nificant bleeding or infection).
*
A V
A 100 µm
Fig. 31.3  Images of the cerebral microcirculation before (A) and during (B) hyperventilation. Besides a large amount of subarachnoid blood (*),
arterioles (A) and venules (V) can be distinguished from each other. After hyperventilation a decrease in arteriolar diameter is observed with a bead
string–like constriction pattern (B). Note that the caliber of the venules is unaffected.
Section V—Cerebral Blood Flow 317
Orthogonal Polarizing
Spectral Imaging
Disturbances of microvascular function seem to play a role
in the development of cerebral ischemia. Various cerebral
pathologies such as SAH or traumatic brain injury may impair
microcirculatory function by dysregulation of the vascular
tone or intravascular microthrombosis.47,48 Furthermore, the
development of cerebral ischemia itself due to a mismatch in
supply and demand may alter microcirculatory hemodynam-
ics and intensify ischemia. These assumptions are based on
direct visualization of the cerebral microcirculation in animals
with a cranial window technique or conventional intravital
microscopy and on indirect monitoring techniques in
humans.49 Therefore the ability to observe the morphology
and function of the human cerebral microcirculation may aid
in the understanding of microvascular disturbances contrib-
uting to ischemia.
The authors introduced a novel optical technique in the late
1990s, OPS imaging, that allows for the direct visualization of
the microcirculation and qualitative evaluation of changes in
flow, vessel diameter, capillary density, and vasoreactivity.50-52
A thorough evaluation of the diagnostic value and limitations
of OPS imaging is useful to define its role in clinical research.
A first step in this direction was the validation of this optical
technique. Several animal studies demonstrated that OPS
imaging produced similar or better images of the micro­
circulation compared to conventional intravital microscopy,
which allows quantification of microvascular changes.53-55 It
was demonstrated that this also is true for the functional
assessment of the human cerebral microcirculation. Moreover,
with this technique changes in CO2 reactivity and autoregula-
tion of microvascular flow could be assessed in patients
undergoing aneurysm surgery and in a focal contusion model
(Fig. 31.3).51,56 This renders OPS imaging particularly useful
to assess changes in microvascular flow during physiologic
challenges in which perfusion is changed.
To further define the clinical value of OPS imaging it is
necessary to investigate whether this imaging technique can
provide new insights on underlying microcirculatory changes
*
A
B
318 Section V—Cerebral Blood Flow
that may play a role in the pathophysiology of cerebral isch-
emia. In addition, the question needs to be answered whether
OPS imaging data can be used to help improve outcome in
patients in whom microcirculatory disturbances contribute to
ischemia. Uhl et al. compared the cerebral microcirculation of
patients with incidental aneurysms (n = 3) to the cerebral
microcirculation of 10 SAH patients, before and after clip-
ping.57 They reported three major findings in patients with
SAH: the detection of microvasospasm in SAH patients in
combination with a reduced functional capillary density,
which increased significantly during the operation. A relation-
ship between intraoperative microvasospasm and clinical
outcome could not be observed and further evaluation is
needed. It has been observed that the human cerebral micro-
circulation is compromised in the course of neurovascular
diseases, which may contribute to cerebral ischemia.51,52,58
Clinical studies, however, also failed to demonstrate a correla-
tion between the severity of microcirculatory impairment and
clinical outcome. The small number of studied patients may
explain this. In patients with septic shock, sublingual micro-
circulatory function is severely altered, and these alterations
are more pronounced in nonsurvivors. In addition, improve-
ment in microcirculatory function after resuscitation can be a
better outcome predictor than changes in cardiac index, blood
pressure, or lactate.59,60 Thus OPS imaging may enable the
correlation of microvascular hemodynamic changes in the
human brain with metabolic and clinical parameters.
Several limitations of OPS imaging need to be addressed.
First, a craniotomy is required to obtain information on the
cerebral microcirculation. This limits its use only to the
intraoperative period. To overcome this limitation, extracra-
nial compartments that may represent the cerebral micro­
circulation could be studied. In this regard, the retina or
sublingual microcirculation may serve as potential targets.
Second, the superficial penetration depth of 500 µm and the
image quality limits the visualization of the framework of the
capillary network and density, which represents the meta-
bolic activity of the surrounding brain tissue. To resolve these
drawbacks SDF imaging (MicroVision Medical Inc., Amster-
dam, the Netherlands) has been introduced into the field of
neurosurgery.61 In this modality, a light guide imaging the
microcirculation is surrounded by light-emitting diodes of a
wavelength (530 nm) absorbed by the hemoglobin of eryth-
rocytes so that they can be clearly observed as flowing cells.
The concentrically placed light-emitting diodes at the tip of
the probe directly penetrate deep into the tissue that illumi-
nates the microcirculation. This way of observing the cere-
bral microcirculation provides clear images of the capillaries
and leukocytes without blurring. The improved resolution
allows for the determination of leukocyte-endothelium adhe-
sion and quantification of vascular leakage. Third, the image
processing software used in previous studies prohibited ac-
curate erythrocyte velocity measurement above 1 mm/sec.
Thus essential information on arteriolar velocity could not
be calculated. Therefore a semiquantitative method, the mi-
crocirculatory flow index (MFI), was introduced to describe
the changes in flow velocities and flow patterns.58 However,
the software has been improved to eliminate most of these
problems and allow velocity measurements of ve­nules and
arterioles as well as an MFI for the area of interest.62
In conclusion, OPS imaging and its successor
SDF can provide continuous information on important
microcirculatory hemodynamic parameters as well as changes
in mechanisms that regulate flow. These qualities may hold
promise to evaluate underlying microcirculatory mechanisms
responsible for ischemia.
References
1. Astrup J, Siesjo BK, Symon L. Thresholds in cerebral ischemia—the ischemic
penumbra. Stroke 1981;12:723–5.
2. Jones TH, Morawetz RB, Crowell RM, et al. Thresholds of focal cerebral
ischemia in awake monkeys. J Neurosurg 1981;54:773–82.
3. Czosnyka M, Brady K, Reinhard M, et al. Monitoring of cerebrovascular
autoregulation: facts, myths, and missing links. Neurocrit Care
2009;10(3):373–86. Epub 2009 Jan 6.
4. Bonner RF, Nossal R. Continuous measurement of tissue blood flow by
laser-Dopplerspectroscopy. Am J Physiol 1977;2:H441–8.
5. Rajan V, Varghese B, van Leeuwen TG, et al. Review of methodological
developments in laser Doppler flowmetry. Lasers Med Sci (online first)
2008;60.
6. Bonner RF, Nossal R. Principles of laser-Doppler flowmetry. In: Shepherd
AP, Oberg PA, editors. Laser Doppler flowmetry. Boston: Kluwer Academic;
1990. p.17–45.
7. Bolognese P, Miller JI, Heger IM, et al. Laser Doppler flowmetry in
neurosurgery. J Neurosurg Anesthesiol 1993;5:151–8.
8. Neufeld GR, Galante SR, Whang JM, et al. Skin blood flow from gas
transport: helium xenon and laser Doppler compared. Microvasc Res 1988;
35:143–52.
9. Olsson P. A comparison between the 133Xe washout and laser Doppler
techniques for estimation of nasal mucosal blood flow in humans. Acta
Otolaryngol 1986;102:106–12.
10. Eyre JA, Essex TJH, Flecknell PA, et al. A comparison of measurements of
cerebral blood flow in the rabbit using laser Doppler spectroscopy and
radionuclide labelled microspheres. Clin Phys Physiol Meas 1988;9:65–74.
11. Lindsberg PJ, O’Neill JT, Paakari IA, et al. Laser-Doppler flowmetry in
monitoring regulation of rapid microcirculatory changes in spinal cord. Am
J Physiol 1992;263:H285–92.
12. Fakuda O, Endo S, Kuwayama N, et al. The characteristics of laser-Doppler
flowmetry for the measurement of regional cerebral blood flow.
Neurosurgery 1995;36:358–64.
13. Rundquist I, Smith QR, Michel ME, et al. Sciatic nerve blood flow measured
by laser Doppler flowmetry and [14C]iodoantipyrine. Am J Physiol 1985;
248:H311–17.
14. Nitzan M, Fairs S, Roberts V. Simultaneous measurement of skin blood flow
by the transient thermal-clearance method and laser Doppler flowmetry.
Med Biol Eng Comput 1988;26:407–10.
15. Kirkpatrick PJ, Smielweski P, Piechnik S, et al. Early effects of mannitol in
patients with head injuries assessed using bedside multimodality monitoring.
Neurosurgery 1996;39:714–20.
16. Kirkpatrick PJ, Smielweski P, Czosnyka M, et al. Continuous monitoring of
cortical perfusion by laser Doppler flowmetry in ventilated patients with
head injury. J Neurol Neurosurg Psychiatry 1994;57:1382–8.
17. Le Bihan D, Turner R. The capillary network: a link between IVIM and
classical perfusion. Magn Reson Med 1992;27:171–8.
18. Zweifel C, Czosnyka M, Lavinio A, et al. A comparison study of cerebral
autoregulation assessed with transcranial Doppler and cortical laser Doppler
flowmetry. Neurol Res 2010;32(4):425–8.
19. Schubert GA, Seiz M, Hegewald AA, et al. Hypoperfusion in the acute phase
of subarachnoid hemorrhage. Acta Neurochir Suppl 2011;110(Pt 1):35–8.
20. Lam JMK, Hsiang JNK, Poon WS. Monitoring of autoregulation using laser
Doppler flowmetry in patients with head injury. J Neurosurg 1997;86:438–45.
21. Smielewski P, Czosnyka M, Kirkpatrick P, et al. Evaluation of the transient
hyperemic response test in head injured patients. J Neurosurg 1997;86:
773–8.
22. Johnson WD, Bolognese P, Miller JI, et al. Continuous postoperative lCBF
monitoring in aneurysmal SAH patients using a combined ICP-laser
Doppler fiberoptic probe. J Neurosurg Anesthesiol 1996;8:199–207.
23. Rosenblum BR, Bonner RF, Oldfield EH. Intraoperative measurement of
cortical blood flow adjacent to cerebral AVM using laser Doppler
velocimetry. J Neurosurg 1987;66:396–9.
24. Leahy MJ, de Mul FFM, Nilsson GE, et al. Principles and practice of the
laser-Doppler perfusion technique. Technol Health Care 1999;7:143–62.
25. Kawamata T, Kawashima A, Yamaguchi K, et al. Usefulness of intraoperative
laser Doppler flowmetry and thermography to predict a risk of postoperative
hyperperfusion after superficial temporal artery-middle cerebral artery

bypass for moyamoya disease. Neurosurg Rev 2011;34(3):355–62; discussion
362. Epub 2011 Jun 4.
26. Klein KU, Stadie A, Fukui K, et al. Measurement of cortical microcirculation
during intracranial aneurysm surgery by combined laser-Doppler flowmetry
and photospectrometry. Neurosurgery 2011;69(2):391–8.
27. Miles B, Davis S, Crandall C, et al. Laser-Doppler examination of the blood
supply in pericranial flaps. J Oral Maxillofac Surg 2010;68(8):1740–5. Epub
2010 May 20.
28. Hill SA, Pigott MI, Saunders MI, et al. Microregional blood flow in murine
and human tumours assessed using laser-Doppler microprobes. Br J Cancer
1996;S260–3.
29. Warden K, Jakobsson A, Nilsson GE. Laser-Doppler perfusion imaging by
dynamic light scattering. IEEE Trans Biomed Eng 1993;40:309–16.
30. Gibbs FA. A thermoelectric blood flow recorder in the form of a needle. In:
Proceedings of the Society for Experimental Biology and Medicine, San
Francisco, 1933;141–6.
31. Carter LP, Atkinson JR. Cortical blood flow in controlled hypotension as
measured by thermal diffusion. J Neurol Neurosurg Psychiatry 1973;36:
906–13.
32. Gaines C, Carter LP, Crowell RM. Comparison of local cerebral blood flow
determined by thermal and hydrogen clearance. Stroke 1983;14:66–9.
33. Vajkoczy P, Roth H, Horn P, et al. Continuous monitoring of regional
cerebral blood flow: experimental and clinical validation of a novel thermal
diffusion microprobe. J Neurosurg 2000;93:265–74.
34. Jaeger M, Soehle M, Schuhmann MU, et al. Correlation of continuously
monitored regional cerebral blood flow and brain tissue oxygen. Acta
Neurochir 2005;147:51–6.
35. Sioutos PJ, Orozco JA, Carter LP, et al. Continuous regional cerebral cortical
blood flow monitoring in head injured patients. Neurosurgery 1995;36:
943–50.
36. Vajkoczy P, Horn P, Thome C, et al. Regional cerebral blood flow monitoring
in the diagnosis of delayed ischemia following aneurismal subarachnoid
hemorrhage. J Neurosurg 2003;98:1227–34.
37. Miller JI, Chou MW, Capocelli A, et al. Continuous intracranial
multimodality monitoring comparing local cerebral blood flow, cerebral
perfusion pressure, and microvascular resistance. Acta Neurochir Suppl 1998;
71:82–4.
38. Rosenthal G, Sanchez-Mejia RO, Phan N, Hemphill 3rd JC, et al.
Incorporating a parenchymal thermal diffusion cerebral blood flow probe in
bedside assessment of cerebral autoregulation and vasoreactivity in patients
with severe traumatic brain injury. J Neurosurg 2011;114(1):62–70. Epub
2010 Aug 13.
Section V—Cerebral Blood Flow 319
39. Hecht N, Fiss I, Wolf S, et al. Modified flow- and oxygen-related
autoregulation indices for continuous monitoring of cerebral autoregulation.
J Neurosci Methods 2011;15:201(2):399–403. Epub 2011 Aug 17.
40. Hänggi D. Participants in the International Multi-Disciplinary Consensus
Conference on the Critical Care Management of Subarachnoid Hemorrhage.
Monitoring and detection of vasospasm. II: EEG and invasive monitoring.
Neurocrit Care 2011;15(2):318–23
41. Nagao S, Veta K, Mino S, et al. Monitoring of cortical blood flow during
excision of arteriovenous malformation by thermal diffusion method. Surg
Neurol 1989;32:137–43.
42. Ohmoto T, Nagao S, Mino S, et al. Monitoring of cortical blood flow during
temporary arterial occlusion in aneurysm surgery by the thermal diffusion
method. Neurosurgery 1991;28:49–54.
43. Thome C, Vajkoczy P, Horn P, et al. Continuous monitoring of regional
blood flow during temporary arterial occlusion in aneurysm surgery.
J Neurosurg 2001;95:402–11.
44. Carter LP, Weinand ME, Oommen KJ. Cerebral blood flow monitoring in
intensive care by thermal diffusion. Acta Neurochir Suppl 1993;59:43–6.
45. Gonzalez G, Rivero S, Ahern GL, et al. Normalization of periictal
bihemispheric cerebral perfusion in temporal lobe epilepsy. Pathophysiology
2004;11:31–4.
46. Weinand ME, Carter LP, Oommen KJ, et al. Response of human epileptic
temporal lobe cortical blood flow to hyperventilation. Epilepsy Res 1995;21:
221–6.
47. Cook DA. Mechanisms of cerebral vasospasm in subarachnoid haemorrhage.
Pharmacol Ther 1995;66:259–84.
48. Stein SC, Graham DI, Chen XH, et al. Association between intravascular
microthrombosis and cerebral ischemia in traumatic brain injury.
Neurosurgery 2004; 54:687–91.
49. Caro CG, Pedley TJ, Schroter RC, et al. The mechanics of the circulation.
Oxford, UK. Medical Publications, 2006.
50. Groner W, Winkelman JW, Harris AG, et al. Orthogonal polarization spectral
imaging: a new method for study of the microcirculation. Nat Med
1999;5:1209–12.
A complete list of references for this chapter can be found online at
www.expertconsult.com.

References
1. Astrup J, Siesjo BK, Symon L. Thresholds in cerebral ischemia—the ischemic
penumbra. Stroke 1981;12:723–5.
2. Jones TH, Morawetz RB, Crowell RM, et al. Thresholds of focal cerebral
ischemia in awake monkeys. J Neurosurg 1981;54:773–82.
3. Czosnyka M, Brady K, Reinhard M, et al. Monitoring of cerebrovascular
autoregulation: facts, myths, and missing links. Neurocrit Care 2009;10(3):
373–86. Epub 2009 Jan 6.
4. Bonner RF, Nossal R. Continuous measurement of tissue blood flow by
laser-Doppler spectroscopy. Am J Physiol 1977;2:H441–8.
5. Rajan V, Varghese B, van Leeuwen TG, et al. Review of methodological
developments in laser Doppler flowmetry. Lasers Med Sci (online first)
2008:60.
6. Bonner RF, Nossal R. Principles of laser-Doppler flowmetry. In: Shepherd
AP, Oberg PA, editors. Laser Doppler flowmetry. Boston: Kluwer Academic;
1990. p.17–45.
7. Bolognese P, Miller JI, Heger IM, et al. Laser Doppler flowmetry in
neurosurgery. J Neurosurg Anesthesiol 1993;5:151–8.
8. Neufeld GR, Galante SR, Whang JM, et al. Skin blood flow from gas
transport: helium xenon and laser Doppler compared. Microvasc Res 1988;
35:143–52.
9. Olsson P. A comparison between the 133Xe washout and laser Doppler
techniques for estimation of nasal mucosal blood flow in humans. Acta
Otolaryngol 1986;102:106–12.
10. Eyre JA, Essex TJH, Flecknell PA, et al. A comparison of measurements of
cerebral blood flow in the rabbit using laser Doppler spectroscopy and
radionuclide labelled microspheres. Clin Phys Physiol Meas 1988;9:65–74.
11. Lindsberg PJ, O’Neill JT, Paakari IA, et al. Laser-Doppler flowmetry in
monitoring regulation of rapid microcirculatory changes in spinal cord. Am
J Physiol 1992;263:H285–92.
12. Fakuda O, Endo S, Kuwayama N, et al. The characteristics of laser-Doppler
flowmetry for the measurement of regional cerebral blood flow.
Neurosurgery 1995;36:358–64.
13. Rundquist I, Smith QR, Michel ME, et al. Sciatic nerve blood flow measured
by laser Doppler flowmetry and [14C]iodoantipyrine. Am J Physiol 1985;
248:H311–17.
14. Nitzan M, Fairs S, Roberts V. Simultaneous measurement of skin blood flow
by the transient thermal-clearance method and laser Doppler flowmetry.
Med Biol Eng Comput 1988;26:407–10.
15. Kirkpatrick PJ, Smielweski P, Piechnik S, et al. Early effects of mannitol in
patients with head injuries assessed using bedside multimodality monitoring.
Neurosurgery 1996;39:714–20.
16. Kirkpatrick PJ, Smielweski P, Czosnyka M, et al. Continuous monitoring of
cortical perfusion by laser Doppler flowmetry in ventilated patients with
head injury. J Neurol Neurosurg Psychiatry 1994;57:1382–8.
17. Le Bihan D, Turner R. The capillary network: a link between IVIM and
classical perfusion. Magn Reson Med 1992;27:171–8.
18. Zweifel C, Czosnyka M, Lavinio A, et al. A comparison study of cerebral
autoregulation assessed with transcranial Doppler and cortical laser Doppler
flowmetry. Neurol Res 2010;32(4):425–8.
19. Schubert GA, Seiz M, Hegewald AA, et al. Hypoperfusion in the acute phase
of subarachnoid hemorrhage. Acta Neurochir Suppl 2011;110(Pt 1):35–8.
20. Lam JMK, Hsiang JNK, Poon WS. Monitoring of autoregulation using laser
Doppler flowmetry in patients with head injury. J Neurosurg 1997;86:
438–45.
21. 21. Smielewski P, Czosnyka M, Kirkpatrick P, et al. Evaluation of the
transient hyperemic response test in head injured patients. J Neurosurg
1997;86:773–8.
22. Johnson WD, Bolognese P, Miller JI, et al. Continuous postoperative lCBF
monitoring in aneurysmal SAH patients using a combined ICP-laser
Doppler fiberoptic probe. J Neurosurg Anesthesiol 1996;8:199–207.
23. Rosenblum BR, Bonner RF, Oldfield EH. Intraoperative measurement of
cortical blood flow adjacent to cerebral AVM using laser Doppler
velocimetry. J Neurosurg 1987;66:396–9.
24. Leahy MJ, de Mul FFM, Nilsson GE, et al. Principles and practice of the
laser-Doppler perfusion technique. Technol Health Care 1999;7:143–62.
25. Kawamata T, Kawashima A, Yamaguchi K, et al. Usefulness of intraoperative
laser Doppler flowmetry and thermography to predict a risk of postoperative
hyperperfusion after superficial temporal artery-middle cerebral artery
bypass for moyamoya disease. Neurosurg Rev 2011;34(3):355–62; discussion
362. Epub 2011 Jun 4.
26. Klein KU, Stadie A, Fukui K, et al. Measurement of cortical microcirculation
during intracranial aneurysm surgery by combined laser-Doppler flowmetry
and photospectrometry. Neurosurgery 2011;69(2):391–8.
Section V—Cerebral Blood Flow 319.e1
27. Miles B, Davis S, Crandall C, et al. Laser-Doppler examination of the blood
supply in pericranial flaps. J Oral Maxillofac Surg 2010;68(8):1740–5. Epub
2010 May 20.
28. Hill SA, Pigott MI, Saunders MI, et al. Microregional blood flow in murine
and human tumours assessed using laser-Doppler microprobes. Br J Cancer
1996;S260–3.
29. Warden K, Jakobsson A, Nilsson GE. Laser-Doppler perfusion imaging by
dynamic light scattering. IEEE Trans Biomed Eng 1993;40:309–16.
30. Gibbs FA. A thermoelectric blood flow recorder in the form of a needle. In:
Proceedings of the Society for Experimental Biology and Medicine, San
Francisco, 1933;141–6.
31. Carter LP, Atkinson JR. Cortical blood flow in controlled hypotension as
measured by thermal diffusion. J Neurol Neurosurg Psychiatry 1973;36:
906–13.
32. Gaines C, Carter LP, Crowell RM. Comparison of local cerebral blood flow
determined by thermal and hydrogen clearance. Stroke 1983;14:66–9.
33. Vajkoczy P, Roth H, Horn P, et al. Continuous monitoring of regional
cerebral blood flow: experimental and clinical validation of a novel thermal
diffusion microprobe. J Neurosurg 2000;93:265–74.
34. Jaeger M, Soehle M, Schuhmann MU, et al. Correlation of continuously
monitored regional cerebral blood flow and brain tissue oxygen. Acta
Neurochir 2005;147:51–6.
35. Sioutos PJ, Orozco JA, Carter LP, et al. Continuous regional cerebral cortical
blood flow monitoring in head injured patients. Neurosurgery 1995;36:
943–50.
36. Vajkoczy P, Horn P, Thome C, et al. Regional cerebral blood flow monitoring
in the diagnosis of delayed ischemia following aneurismal subarachnoid
hemorrhage. J Neurosurg 2003;98:1227–34.
37. Miller JI, Chou MW, Capocelli A, et al. Continuous intracranial
multimodality monitoring comparing local cerebral blood flow, cerebral
perfusion pressure, and microvascular resistance. Acta Neurochir Suppl 1998;
71:82–4.
38. Rosenthal G, Sanchez-Mejia RO, Phan N, Hemphill 3rd JC, et al.
Incorporating a parenchymal thermal diffusion cerebral blood flow probe in
bedside assessment of cerebral autoregulation and vasoreactivity in patients
with severe traumatic brain injury. J Neurosurg 2011;114(1):62–70. Epub
2010 Aug 13.
39. Hecht N, Fiss I, Wolf S, et al. Modified flow- and oxygen-related
autoregulation indices for continuous monitoring of cerebral autoregulation.
J Neurosci Methods 2011;15:201(2):399–403. Epub 2011 Aug 17.
40. Hänggi D. Participants in the International Multi-Disciplinary Consensus
Conference on the Critical Care Management of Subarachnoid Hemorrhage.
Monitoring and detection of vasospasm. II: EEG and invasive monitoring.
Neurocrit Care 2011;15(2):318–23
41. Nagao S, Veta K, Mino S, et al. Monitoring of cortical blood flow during
excision of arteriovenous malformation by thermal diffusion method. Surg
Neurol 1989;32:137–43.
42. Ohmoto T, Nagao S, Mino S, et al. Monitoring of cortical blood flow during
temporary arterial occlusion in aneurysm surgery by the thermal diffusion
method. Neurosurgery 1991;28:49–54.
43. Thome C, Vajkoczy P, Horn P, et al. Continuous monitoring of regional
blood flow during temporary arterial occlusion in aneurysm surgery.
J Neurosurg 2001;95:402–11.
44. Carter LP, Weinand ME, Oommen KJ. Cerebral blood flow monitoring
in intensive care by thermal diffusion. Acta Neurochir Suppl 1993;59:
43–6.
45. Gonzalez G, Rivero S, Ahern GL, et al. Normalization of periictal
bihemispheric cerebral perfusion in temporal lobe epilepsy. Pathophysiology
2004;11:31–4.
46. Weinand ME, Carter LP, Oommen KJ, et al. Response of human epileptic
temporal lobe cortical blood flow to hyperventilation. Epilepsy Res 1995;21:
221–6.
47. Cook DA. Mechanisms of cerebral vasospasm in subarachnoid haemorrhage.
Pharmacol Ther 1995;66:259–84.
48. Stein SC, Graham DI, Chen XH, et al. Association between intravascular
microthrombosis and cerebral ischemia in traumatic brain injury.
Neurosurgery 2004; 54:687–91.
49. Caro CG, Pedley TJ, Schroter RC, et al. The mechanics of the circulation.
Oxford, UK. Medical Publications 2006.
50. Groner W, Winkelman JW, Harris AG, et al. Orthogonal polarization spectral
imaging: a new method for study of the microcirculation. Nat Med
1999;5:1209–12.
51. Pennings FA, Bouma GJ, Ince C. Direct observation of the human
microcirculation during aneurysm surgery reveals increased arteriolar
contractility. Stroke 2004;35(6):1284–88.
319.e2 Section V—Cerebral Blood Flow
52. Pennings FA, Albrecht KW, Muizelaar JP, et al. Abnormal responses of the
human cerebral microcirculation to papaverin during aneurysm surgery.
Stroke 2009;40(1):317–20. Epub 2008 Oct 9.
53. Harris AG, Sinitsina I, Messmer K. The Cytoscan model E-II, a new
reflectance microscope for intravital microscopy: comparison with the
standard fluorescence method. J Vasc Res 2000;37:469–76.
54. Langer S, Harris AG, Biberthaler P, et al. Orthogonal polarization spectral
imaging as a tool for the assessment of hepatic microcirculation: a validation
study. Transplantation 2001;71:1249–56.
55. Mathura KR, Vollebregt KC, Boer K, et al. Comparison of OPS imaging and
conventional capillary microscopy to study the human microcirculation.
J Appl Physiol 2001;91:74–8.
56. Kroppenstedt SN, Thomale UW, Griebenow M, et al. Effects of early and late
intravenous norepinephrine infusion on cerebral perfusion, microcirculation,
brain-tissue oxygenation, and edema formation in brain-injured rats. Crit
Care Med 2003;31:2211–21.
57. Uhl E, Lehmberg J, Steiger HJ, et al. Intraoperative detection of early
microvasospasm in patients with subarachnoid hemorrhage by using
orthogonal polarization spectral imaging. Neurosurgery 2003;52:1307–15.
58. Pennings FA, Ince C, Bouma GJ. Continuous real-time visualization of the
human cerebral microcirculation during arteriovenous malformation
surgery using orthogonal polarization spectral imaging. Neurosurgery 2006;
59:167–71.
59. De BD, Creteur J, Preiser JC, et al. Microvascular blood flow is altered in
patients with sepsis. Am J Respir Crit Care Med 2002;166:98–104.
60. Sakr Y, Dubois MJ, De BD, et al. Persistent microcirculatory alterations are
associated with organ failure and death in patients with septic shock. Crit
Care Med 2004;32:1825–31.
61. Ince C. The microcirculation is the motor of sepsis. Crit Care Suppl 2005;4:
S13–9.
62. Dobbe JG, Streekstra GJ, Atasever B, et al. Measurement of functional
microcirculatory geometry and velocity distributions using automated image
analysis. Med Biol Eng Comput 2008;46:659–70.
V
Chapter
32  
Jugular Bulb Oximetry
Amit Prakash and Basil F. Matta
Introduction
Jugular bulb venous monitoring provides information about
global cerebral hemodynamics and metabolism and has
become an important tool in the clinical management of
neurologically injured patients. Other techniques to measure
cerebral oxygenation include near infrared spectroscopy
(NIRS), which noninvasively detects changes in blood hemo-
globin concentrations associated with neural activity and
multiparameter intraparenchymal probes that are used to
measure brain tissue and cerebrospinal fluid PO2, PaCO2, and
pH. This chapter focuses on jugular bulb venous oximetry.
Anatomy
The venous drainage from the brain forms six major cranial
sinuses (superior and inferior sagittal, occipital, right and left
transverse, and the straight sinus). These venous channels,
located between the dura mater and the periosteum lining the
cranium, are devoid of valves or muscle in their walls. The
final common pathway for the majority of blood returning
from the brain is through the right and left sigmoid sinus,
which curves down the posterior fossa, over the mastoid
portion of the temporal bone, and then runs forward to form
the right and left internal jugular veins in the posterior part
of the jugular foramen.
The internal jugular vein (IJV) begins as a small dilation
called the jugular bulb at the jugular foramen, and runs down
the neck in the carotid sheath to form the brachiocephalic vein
by joining the subclavian vein behind the medial end of the
clavicle. Within the carotid sheath, the IJV is lateral to the
vagus nerve and the carotid artery.
Posterior to the vein lie the transverse processes of the cervi-
cal vertebra, the cervical plexus, phrenic nerve, vertebral vein,
the first part of the subclavian artery, and the thoracic duct
toward the left. Anterolateral to the vein are superficial cervical
fascia, platysma, transverse cutaneous nerve, deep cervical
fascia, and the sternocleidomastoid muscle. The lower end of
IJV drains into the right atrium.
The jugular bulb is the final common pathway for venous
blood that drains from the cerebral hemispheres, cerebel-
lum, and brainstem. Therefore the jugular venous oxygen
saturation (SjvO2) reflects the balance between supply and
consumption of oxygen by the brain. Contribution of the
extracranial circulation is estimated to range from zero to
6.6%. This is exacerbated if the blood is contaminated with
blood from the facial vein that joins the IJV a few centime-
ters below the jugular bulb.
320
Physiologic Basis of
SjvO2 Interpretation
Jugular bulb oximetry has evolved from a combination of old
and new technologies. The first measurements taken from the
jugular bulb were used to calculate cerebral blood flow (CBF)
using the Fick principle.1
Cerebral metabolic rate of oxygen (CMRO2) is calculated
from the CBF and the difference between the arteriovenous
difference in oxygen (AVDO2) using the equation:
CMRO2 = CBF × (arterial oxygen content minus jugular
venous bulb oxygen content)
The oxygen content can be determined using co-oximetry,
or calculated using the equations:
Arterial oxygen content =
[(Hgb × 1.34 × SaO2 ) + (PaO2 × 0.0031)]
Jugular venous oxygen content =
[(Hgb × 1.34 × SjvO2 ) + (PvO2 × 0.0031)]
Where Hgb = hemoglobin in g/dL, SaO2 = arterial satura-
tion, SjvO2 = jugular venous oxygen saturation, PaO2 = oxygen
partial pressure in arterial blood in mm Hg, and PvO2 =
oxygen partial pressure in venous blood in mm Hg.
Because the hemoglobin concentration is the same in arte-
rial and venous blood, and the amount of dissolved oxygen is
minimal, CMRO2 can be estimated from CBF and the differ-
ence in the arteriojugular bulb oxygen saturation.
CMRO2 = CBF × (SaO2 − SjvO2 )
Therefore the arteriovenous oxygen content (or saturation)
difference can be used to relate for changes in metabolism
with alterations in CBF. The equation may be written as
follows:
AVDO2 = Hgb × 1.34 (SaO2 − SvO2 ) + 0.003 (PaO2 − PvO2 )
In the absence of infarction or necrosis, the cerebral AVDO2
and CBF appear to be inversely related.2 Hypoperfusion and
ischemia result in an increased AVDO2, whereas hyperemia
will lead to a reduced AVDO2. The same principle can be
applied to other metabolites such as glucose and lactate.
Insertion Technique
The use of ultrasound (US) in accordance with National Insti-
tute for Clinical Excellence (NICE) guidelines3 for anatomic
localization and IJV puncture has facilitated procedure
© Copyright 2013 Elsevier Inc. All rights reserved.

performance. Besides, SjvO2 monitoring is commonly used to
manage head injured patients. Because severe head injury fre-
quently has been associated with cervical injury, US-guided
insertion reduces the need for head manipulation, thus making
it a safer technique. Overall, SjvO2 catheters have a complica-
tion rate similar to that seen with the placement of antegrade
jugular venous lines.4 Jugular bulb catheters are best avoided
in those with coagulation abnormalities, sepsis, or local neck
trauma.
Optimal Side for Monitoring
Jugular Bulb Oxygenation
SjvO2 monitoring accurately reflects global and hemispheric
cerebral oxygenation, only when the dominant jugular bulb is
cannulated.5 However, selecting the optimal side for a jugular
bulb catheter remains controversial. Cortical blood from the
sagittal sinus flows into the right lateral sinus and subcortical
blood from the straight sinus usually goes to the left lateral
sinus. Overall, flow is greater to the right jugular venous bulb.6,7
Up to 15% difference between right and left SjvO2 has been
demonstrated in patients with head trauma.8 If an intracranial
pressure (ICP) monitor is in situ, it is possible to determine
the dominant side of venous drainage by sequential temporary
manual occlusion of the right and then left internal jugular
vein and observing the resultant ICP change.9 Compressing
the dominant jugular vein will lead to venous obstruction and
a greater increase in ICP, thus providing a clue for the best
side for SjvO2 monitoring. If the ICP increase is similar during
right and left IJV compression, the side with the greatest degree
of pathology as determined by computed tomography (CT)
scan is used to insert a jugular bulb catheter. In the case of
diffuse injury, the right side is used.10 In the authors’ unit, right-
sided jugular venous oximetry is used almost exclusively.
Site of Cannulation
US-guided visualization insertion reduces the need for scru-
pulous landmark identification. However, there are two main
sites for retrograde cannulation of the internal jugular vein:
(1) lateral to the carotid artery at the level of the inferior
border of the thyroid cartilage6 and (2) at the junction of the
sternal and clavicular heads of the sternocleidomastoid muscle
(Fig. 32.1). The cannulation can be approached either by the
Seldinger method or cannula over needle method. One
method is to approach the vessel at the junction of the sternal
and cla­vicular heads of the sternocleidomastoid muscle under
US guidance.
Once the site is selected, it is prepared with antiseptic solu-
tion, draped with sterile towels, and infiltrated with local anes-
thetic. The patient is then placed in the horizontal position,
with the neck in the neutral position or slightly turned to the
contralateral side. The IJV is then localized with an US probe,
between the junction of the sternal and clavicular heads of the
sternocleidomastoid. Under US guidance, the skin is punc-
tured with a 16-gauge 5.25-inch Angiocath catheter (Becton,
Dickinson, Franklin Lakes, NJ) mounted on a 5-mL syringe.
With continuous gentle aspiration the IJV is punctured at
almost 90 degrees under direct vision, which is confirmed by
the flashback in the syringe. Once the vein is entered, the
operating hand with the needle and the syringe is dropped
down to about 15 to 20 degrees in a cranial direction, while
Section V—Cerebral Blood Flow 321
IJV
Carotid A
SCM
Fig. 32.1  The authors’ technique for insertion of jugular bulb
catheters. The junction between the sternal and clavicular heads of the
sternocleidomastoid muscle (SCM) is localized and the skin is punctured
under ultrasound guidance with a 16-gauge, 5.25-inch Angiocath
catheter (Becton, Dickinson, Franklin Lakes, NJ) mounted with a 5-mL
syringe. During gentle aspiration the needle is passed in a cranial
direction for 1 to 2 cm, at an angle of 15 to 20 degrees in the sagittal
plane. Once the vein is entered, the catheter is advanced over the needle
until a slight elastic resistance is felt, or when the tip of the catheter is
estimated to be just behind the mastoid process. Carotid A, Carotid
artery; IJV, internal jugular vein.
continuously aspirating, gently to maintain intraluminal posi-
tion of the needle. The catheter then is advanced over the
needle until a slight elastic resistance is felt, or when the tip of
the catheter is estimated to be just behind the mastoid process.
If difficulty in localizing the vein is encountered, the patient
is tilted 15 degrees head down until the vein is located, after
which the head is elevated. A guidewire (Seldinger technique)
may be used in a similar manner.
If needed, central venous access also can be obtained as a
part of the same procedure. Once the wire for the SjvO2 cath-
eter is in situ, a guidewire is passed in the antegrade direction
through a separate puncture site, using the Seldinger tech-
nique under US guidance. After placement of the guidewires
for both lines, the retrograde catheter for SjvO2 followed by
the central venous catheter is threaded over the guidewire.
This sequence prevents the accidental shearing of the central
venous catheter when the SjvO2 catheter is inserted. The cath-
eter is connected to a slow continuous infusion of 0.9% saline,
to prevent blockage.
For continuous monitoring, a fiber-optic catheter is passed
through a 16-gauge cannula placed in a similar manner as
described previously. A pressurized flush system is used to
maintain catheter patency and reduce the incidence of wall
artefacts.11 A lateral cervical spine x-ray or an anteroposterior
chest x-ray that includes a view of the neck is obtained to
confirm the catheter position.12 The tip of the catheter should
lie at the level of and just medial to the mastoid bone above
the lower border of C1.
322 Section V—Cerebral Blood Flow
Measurement Obtained
from Jugular Bulb Catheters
SjvO2 can be measured by sampling blood intermittently or
continuously by using a fiber-optic oximetric catheter. Serial
or intermittent sampling is cheaper and allows calculation of
AVDO2, glucose, and lactate. The interpretation of jugular
oxygen saturation is based on several assumptions. However,
there is a good correlation between fiber-optic and catheter
SjvO2 and with direct brain oxygen measurements.13 Some of
the early problems encountered with the use of fiber-optic
oximetric catheters seem to have been reduced by the develop-
ment of new “stiffer” catheters less prone to kinking and
curling back, and by careful positioning and calibration. This
has been confirmed by Gunn et al.11 in intracranial surgery
and by Souter and Andrews,14 and Coplin et al.15 in an inten-
sive care setting. Nonetheless, suspected jugular bulb desatu-
ration should be verified by co-oximetry before therapeutic
interventions.
Two fiber-optic catheters are available for jugular bulb
oximetry: (1) the Oximetrix (Abbott Laboratories, North
Chicago) and 2) the Edslab II (Baxter Healthcare Corpora-
tion, Irvine, CA).16 Both are 4F-gauge double-lumen cathe-
ters. After insertion the distal lumen should be aspirated for
in vivo calibration. The other lumen contains two optical
fibers, one to transmit and the other to receive light trans-
mitted to and then returned from the jugular bulb. The
Oximetrix catheter has three light-emitting diodes; the
Edslab II has two. These send red and near infrared light at
1-ms intervals to blood passing through the jugular bulb,
where it is absorbed, reflected, and refracted. The reflected
light from hemoglobin is detected by a photoelectric sensor
and averaged for the previous 5 seconds to calculate hemo-
globin oxygen saturation that is then updated every second.
If the light source at the catheter tip abuts the vessel wall, a
low light intensity or signal quality alarm is displayed. The
Edslab II system requires the insertion of the patient’s hemo-
globin concentration for in vivo calibration; this may affect
its accuracy when the hemoglobin concentration is unstable
such as with rapid massive blood loss. However, presently
there is no clinical data that compare the two catheters in
such circumstances.
Continuous fiber-optic Oximetrix catheters can be cali-
brated before (in vitro), or after insertion (in vivo). In vivo
calibration is more accurate and subject to less drift.9,14,17 Drift
can be further reduced by recalibration at 12-hour intervals.
The sampling port of the catheter should be continuously
flushed with heparinized saline (1 IU/mL) at a rate of 2 to
4 mL/hr to maintain its patency. Contamination of jugular
bulb blood with extracranial blood depends on the speed of
blood withdrawal from the catheter, with up to 25% higher
SjvO2 values with faster rates of blood withdrawal (>2 mL/
min)18 (Fig. 32.2).
Complications on Insertion
The commonest complications are carotid artery puncture
and hematoma formation, which occur in about 1% to 4% of
insertions and are usually self-limiting. Pneumothorax, venous
air embolism, and venous thrombosis are infrequent compli-
cations, as is damage to adjacent structures such as the carotid
artery, vagus and phrenic nerves, and the thoracic duct. There
80
SjvO2 %
40
0
PaCO2 PaCO2
25 mm Hg 30 mm Hg
2 mL/min
4 mL/min
10 mL/min
Fig. 32.2  The speed of blood withdrawal from jugular bulb
catheters affects accuracy of reading. Jugular venous oxygen
saturation (SjvO2) values are higher with faster rates of blood withdrawal
because of contamination with extracranial blood. Optimal rate appears
to be 2 mL/min.
is an increased risk of local and systemic infection with long-
term placement.
Interpretation of the Measurements
Normal SjvO2 values range between 55% and 75%. In simple
terms, SjvO2 levels less than 55% suggest cerebral oxygen
demand exceeding supply, for example, as a consequence of
hypoperfusion (ischemia), whereas levels greater than 75%
indicate relative hyperemia (Tables 32.1 and 32.2). Microdi-
alysis studies in traumatic brain injury (TBI) patients demon-
strate that jugular oxygen desaturation is associated with
elevated glutamate.19 Evidence for cellular dysfunction
detected by microdialysis of metabolites such as glutamate,
glycerol, lactate, and pyruvate consistently occurs when SjvO2
is less than 45%.20 However, SjvO2 is a global hemispheric
measurement and hence regional ischemia cannot be detected.
For example, Chieregato et al.21 suggest that jugular bulb
oximetry, without ICP monitoring, is suboptimal to manage
a patient with subarachnoid haemorrhage (SAH) and raised
intracranial pressure.22 Therefore although a normal SjvO2
does not guarantee absence of regional ischemia, a low SjvO2
indicates either an increase in oxygen extraction or a reduction
in oxygen delivery, which may be an early warning of cerebral
ischemia.23 The volume of ischemic brain to cause a change in
SjvO2 is large, however. For example, Coles et al.24 examined
15 TBI patients with positron emission tomography (PET)
within 24 hours of head injury to map CBF, CMRO2, and
oxygen extraction fraction (OEF). The SjvO2 correlated with
the ischemic brain volume (measured by PET OEF) but SjvO2
values of 50% only occurred when ischemic brain volume was
170 plus or minus 63 mL (mean ± 95% coagulation index
[CI]), that is, 13 plus or minus 5% of the brain. Finally in some
patients, changes in SjvO2 associated with intracranial hyper-
tension may only be observed after herniation.
 Section V—Cerebral Blood Flow 323

outcome in TBI.19,30,31 For example, Robertson et al.19 observed

Table 32.1  A Summary of the Main that mortality was 21% in TBI patients with no desaturation,
Causes and Treatment of Low and High 37% with a single episode, and 69% after multiple episodes.
Jugular Saturations Although reduced SjvO2 is associated with a wide AVDO2,
High SjvO2 Abnormal Hyperemia outcome also may be poor with a reduced AVDO2 because
autoregulation cerebral metabolism is reduced once there is cerebral infarc-
Increased oxygen Polycythemia tion.31 Together these data suggest a potential benefit to detect
supply and treat episodes of jugular destauration.27
Deceased oxygen Hypothermia
consumption Sedative drugs
Many episodes of SjvO2 desaturation appear to be associ-
Anesthetic drugs ated with suboptimal cerebral perfusion pressure (CPP), with
Cerebral Infarction cerebral arterial vasospasm, or from hyperventilation.
Extracerebral blood However, abnormalities in jugular oxygenation may persist
Low SjvO2 Abnormal even when ICP and CPP are corrected. For example, Le Roux
autoregulation et al.32 studied 32 patients with severe TBI who had elevated
Deceased oxygen Hypoxia ICP and were treated with surgery or mannitol. Patients whose
supply Hypotension
Intracranial hypertension
AVDO2 failed to improve were more likely to develop a cere-
Hyperventilation bral infarct or have a poor outcome than when AVDO2
Low cardiac output improved (i.e., decrease >1 vol%) after correction of the ele-
Anemia vated ICP or decreased CPP. In addition, when very early
Increased oxygen Hyperthermia jugular bulb oximetry data from patients with severe TBI are
consumption Seizures
Sepsis
analyzed, there is evidence that altered and inadequate CPP is
common in the first few hours after injury. These findings
SjvO2, Jugular venous oxygen saturation. have important implications for the emergency management
of such patients.33
The role of lactate after TBI has been evaluated using ret-
rograde jugular catheters. Some studies suggest that an increase
Table 32.2  Arteriovenous Difference in CSF lactate is a useful marker of cerebral ischemia, and in
in Oxygen particular knowledge of cerebral lactate (or cerebral anaerobic
metabolism) can help define infarction.34 However Poca
AVDO2 = CMRO2 / CBF (calculation − 1.39 (SaO2 − 
SjvO2) × Hgb/100 et al.35 found that arteriojugular venous differences of lactate
Normal = 5-7.5 vol% (5.1 - 8.3 vol%) do not reliably reflect increased cerebral lactate production
Narrow AVDO2 − <5 vol% = hyperemia (CBF > CMRO2) when compared with microdialysis measurements of lactate.
Wide AVDO2 − >7.5 vol% = low flow (CBF < CMRO2) This difference, however, may be explained, one method
AVDO2, Arteriovenous difference in oxygen; CBF, cerebral blood flow; being a global measure and the other a local measure. Other
CMRO2, cerebral metabolic rate of oxygen; Hgb, hemoglobin; markers such as endothelin-1,36-38 cytokines,39 arteriovenous
SaO2, arterial saturation; SjvO2, jugular venous oxygen saturation. difference of PaCO2,21 markers of coagulation and inflamma-
tion thrombin-antithrombin complex, fibrin D-dimer, and
prothrombin fragment,40,41 markers of oxidative damage viz
Care must be taken when using a jugular bulb catheter malonyaldehyde,42 and type B natriuretic peptide43 have all
because incorrect placement can result in extracerebral con- been assessed using jugular bulb catheters (i.e., not with a
tamination or artifacts associated with catheter movement fiber-optic oximeter), in research studies. In addition after
may occur. A slow continuous infusion of 0.9% saline and injections of ice-cold indocyanine green, global CBF values
flush system to maintain patency can help reduce the inci- can be assessed using the mean transit time principle by exam-
dence of wall artifacts.11 Anemia also may narrow the AVDO2 ining temperature and dye concentration changes in the tho-
and affect measurements. Similarly the rate of sedative infu- racic aorta and the jugular bulb.44
sion (e.g., propofol) can affect readings.25 In addition, many
retrograde jugular catheters require constant recalibration
and unless done, accuracy suffers.13,26 Long-term use is associ- Intraoperative Management of Patients
ated with a small risk of infection and thrombosis. Undergoing Intracranial Aneurysm Surgery
Several studies have described the use of jugular bulb cathe-
ters during neurosurgical procedures and in particular intra-
Clinical Applications cranial aneurysm surgery. For example, Matta et al. observed
a 50% incidence of jugular venous desaturation in patients
Traumatic Brain Injury and during neurosurgical procedures.4 However, the incidence of
Subarachnoid Hemorrhage severe desaturation (defined as SjvO2 <45%) was 17%. The
Despite intensive care management, with invasive hemody- episodes of desaturation could not have been predicted by
namic and ICP monitoring, episodes of cerebral venous desat- the PaCO2 level alone—without SjvO2 monitoring many of
uration are common in comatose patients with TBI and the episodes would have gone unnoticed. Moss et al. studied
SAH.27-29 Patients with episodes of cerebral venous oxygen the effects of changing mean arterial pressure (MAP) on
desaturation (SjvO2 <50% for more than 15 minutes) have a SjvO2 and lactate oxygen index (LOI) in 26 patients during
higher mortality than those without. There is a well-described cerebral aneurysm surgery.45 An SjvO2 value less than 54%
association between episodes of SjvO2 desaturation and poor was considered to indicate cerebral hypoperfusion. They were
324 Section V—Cerebral Blood Flow
able to identify a critical MAP of between 80 and 110 mm Hg
in 9 patients. Patients with an LOI less than 0.08 at any time
during the procedure had a worse outcome immediately after
surgery. Although an increase in MAP resulted in a similar
increase in SjvO2 in 19 of the patients, this was not always
accompanied by an improved LOI. Together these studies
highlight the potential benefit of jugular bulb cannulation to
assess cerebral hypoperfusion during the intra- and postop-
erative management of SAH patients. In addition, significant
desaturation has been observed with intraoperative aneurysm
rupture.46
Carotid Endarterectomy
Detection of cerebral ischemia throughout cross clamping
during carotid endarterectomy (CEA) is critical. Several
studies have demonstrated the potential value of SjvO2 mon-
itoring to detect cerebral ischemia during carotid endarter-
ectomy. For example, Crossman et al.46 studied 37 patients
who underwent awake CEA to correlate neurologic deficits
with SjvO2 desaturation. A 25% change in SjvO2 was associ-
ated with the development of clinically apparent cerebral
ischemia. Moritz et al. examined 48 patients during awake
CEA. Twelve patients developed intraoperative neurologic
deficit, which correlated with SjvO2 desaturations. They
found that arteriojugular venous lactate content difference
(≥0.16 mmol/L) and lactate oxygen index was associated
with intraoperative ischemia.47
Diagnosis of Brain Death
Although the diagnosis of brain death remains a clinical one,
confirmatory or complementary tests are useful. Extreme
hyperoxia of the jugular venous blood has been suggested as
a possible sign of brain death.48 Díaz-Regañón et al.49 assessed
the usefulness of SjvO2 monitoring as a complementary test
to diagnose brain death. They studied 118 patients and mea-
sured the ratio of central venous oxygen (ScO2) to SjvO2 at the
time of diagnosis of brain death. A ratio of less than 1 was
found in 114 of the 118 patients.
Cardiac and Thoracic Surgery
Jugular bulb oxygen desaturation is associated with poor
outcome in patients undergoing cardiac surgery.50 For
example, Croughwell et al. used jugular bulb catheters to cal-
culate AVDO2 in 255 patients during cardiopulmonary
bypass.51 Desaturation (SjvO2 <50%) occurred in 23% of the
patients and was associated with worse postoperative cogni-
tive function. The desaturations observed were the result of
inadequate CBF in proportion to oxygen consumption, as
indicated by a decreased CBF to AVDO2 ratio. Other potential
causes include microembolic events and disturbances in cere-
bral autoregulation. The episodes of desaturation were
observed during both normothermic and hypothermic car-
diopulmonary bypass. In a follow-up study, the same group
highlighted the importance of continuously monitoring cere-
bral oxygenation during cardiac surgery as neither mixed
venous oximetry nor systemic pump venous saturation was
able to detect the adequacy of cerebral perfusion in these
patients.52 During off-pump coronary artery bypass grafting
surgery jugular desaturations are common but can be
prevented by achieving normal values of mixed SvO2, partial
pressure of PaCO2 and central venous pressure (CVP); this
provides insight to potential neurocritical care unit (NCCU)
management strategies. Others have also found a benefit to
PaCO2 control. For example, during one lung ventilation for
lung surgery, SjvO2 has been used to examine effects on cere-
bral oxygen. In these patients Iwata et al.53 observed that
hypercapnia, not hypertension, significantly improved cere-
bral oxygen balance.
Cardiac Arrest
The ability to diagnose evidence for brain injury after cardiac
arrest is important in some patients. Takasu et al. continuously
monitored systemic and jugular venous oxygenation during
the 24 hours after resuscitation in cardiac arrest in eight
patients.54 The three patients that survived had significantly
lower SjvO2 (67%) than nonsurvivors (80%), whereas mixed
venous saturations were higher in the survivors than in the
nonsurvivors (74% and 64%, respectively). The authors
suggest that damaged neurons are unable to utilize oxygen
adequately, and this resulted in the high SjvO2 value. However,
normal SjvO2 values vary between 55% and 85%, and so it is
unlikely that this information will be of benefit in the immedi-
ate management of these patients.
Hypothermia is indicated for cardiac arrest. Although
hypothermia improves outcome in patients after cardiac
arrest, there remains a question whether the cardiodepressive
effects of hypothermia may lead to secondary brain damage.
Studies by Bisschops et al.55 using a jugular bulb catheter
among other monitors suggest that jugular bulb oxygenation
gradually increases, whereas cerebral oxygen extraction
remains normal, and transcranial Doppler (TCD) measure-
ments of flow decrease, suggesting decreased cerebral meta-
bolic activity.
As a Guide to Clinical Management
in the Intensive Care Unit
In a survey of the intensive care management of severe TBI in
the United Kingdom, published in 1996, 12% of the units
questioned used a jugular bulb catheter for routine monitor-
ing, in more than 50% of their TBI patients.56 Since 2001 the
use of jugular bulb oximetry has progressed from being a
research tool to a common clinical tool used in many hospi-
tals. In some patients it is used alone to assess cerebral oxy-
genation but in many it is used with other measures such as
near infrared spectroscopy (NIRS) or direct brain oxygen.
These various tools can supplement and complement
each other.
Prolonged hyperventilation can aggravate neurologic
outcome. However, acute hyperventilation (HV) may be life-
saving in many patients provided the potential deleterious
consequences of HV are identified.57 Hypocapnia from HV
induces cerebral vasoconstriction, and the resultant decrease
in cerebral blood volume and ICP may improve CPP. However,
excessive cerebral vasoconstriction causes cerebral ischemia.
In addition, hyperventilation shifts the oxygen-hemoglobin
dissociation curve to the left, and reduces the amount of
oxygen released from haemoglobin to brain tissue. Therefore
HV should not be used to lower ICP without some measure
of cerebral oxygenation (CBF).58,59 Hyperoxia can be used as
 Section V—Cerebral Blood Flow 325

8 to changes in perfusion pressure. For example, Lewis et al.

Focal pathology demonstrated that a critically low SjvO2 is a late indicator of
None failed autoregulation, and when used with TCD may be useful
6 to determine the level of “optimal” CPP in the early resuscita-
∆PbO2 (mm Hg)

tion of TBI62 (Fig. 32.4).

3
Conclusion
By providing an indication of inadequate cerebral oxygen-
0 ation and by inference inadequate CBF, cerebral oximetry
may allow appropriate therapy to be given before neurologic
damage becomes permanent. There is no gold standard to
–3 measure cerebral oxygenation. Jugular venous bulb oximetry
22–26 26–30 30–34 34–38 is a global hemispheric measure with low sensitivity for
PaCO2 (mm Hg) regional ischemia and so it is best used with other monitors
of brain function. Regional brain oxygenation can be exam-
Fig. 32.3  Graph that illustrates the effect of hyperventilation on brain ined by using NIRS or direct brain oxygen monitors (see
tissue oxygenation in areas of focal pathology and with no focal
Chapters 33 and 35), and these techniques can be used in a
pathology. (Reproduced with permission from Gupta AK, Hutchinson PJ, Al-Rawi
P, et al. Measuring brain tissue oxygenation compared with jugular venous oxygen
complementary fashion in select patients. At present no one
saturation for monitoring cerebral oxygenation after traumatic brain injury. Anesth
method is sufficiently reliable to facilitate the clinical man-
Analg 1999;88:549–53.) agement of neurologically injured patient with absolute cer-
tainty when used alone. However, further research and
improvement in technology, and the ability to incorporate
SjvO2 <50% these methods into multimodal monitoring can permit
early interventions with a possible consequence of improved
No outcome.

SpO2 <50% Yes Correct hypoxemia References

1. Ketty SS, Schmidt CF. The nitrous oxide method for the quantitative
No determination of cerebral blood flow in man: theory, procedure and normal
values. J Clin Invest 1948;27:476–83.
2. Obrist W, Langfitt T, Jaggi J, et al. Cerebral blood flow and metabolism in
PaCO2 <4.0 kPa Yes Raise PaCO2 comatose patients with acute head injury: relationship to intracranial
hypertension. J Neurosurg 1984;61:241–53.
No 3. National Institute for Clinical Excellence. NICE technology appraisal
guidance No 49: guidance on the use of ultrasound locating devices for
placing central venous catheters. 2002. London NICE Available from
Hemoglobin <10g/dL Yes Blood transfusion www.nice.org.uk/pdf/ultrasound_49_GUIDANCE.pdf
4. Matta BF, Lam AM, Mayberg TS, et al. A critique of the intraoperative use of
jugular venous bulb catheters during neurosurgical procedures. Anesth Analg
No 1994;79:745–50.
5. Lam JM, Chan MS, Poon WS. Cerebral venous oxygen saturation
monitoring: is dominant jugular bulb cannulation good enough? Br J
Mean arterial Support blood Neurosurg 1996;10(4):357–64.
Yes
pressure flow pressure 6. Gibbs EL, Gibbs FA. The cross sectional areas of the vessels that form the
torcular and the manner in which blood is distributed to the right and to
No the left lateral sinus. Anat Rec 1934;54:419.
7. Gibbs EL, Lennox WG, Giggs FA. Bilateral internal jugular blood,
comparison of A-V differences, oxygen-dextrose ratios and respiratory
Mannitol, furosemide, quotients. Am J Psychiatry 1945;102:184–90.
Intracranial propofol, thiopentone,
hypertension Yes 8. Stocchetti N, Paparella A, Brindelli F, et al. Cerebral venous oxygen
hypothermia saturation studied with bilateral samples in the internal jugular veins.
Neurosurgery 1994;34:38–44.
Fig. 32.4  Suggested algorithm to treat jugular venous desaturation. 9. Andrews PJD, Dearden NM, Miller JD. Jugular bulb cannulation: description
g/dL, Grams per deciliter (conventional unit); kPa, kilo pascals (1 kpa = of a cannulation technique and a validation of a new continuous monitor.
7.5 mm Hg); PaCO2 , arterial carbon dioxide tension; SjvO2 , jugular Br J Anaesth 1991;67:553–8.
venous oxygen saturation; SpO2 , arterial saturation. 10. Matta BF, Lam AM, Mayberg TS, et al. A critique of the intraoperative use of
jugular venous bulb catheters during neurosurgical procedures. Anesth Analg
1994;79:745–50.
11. Gunn HC, Matta BF, Lam AM, et al. Accuracy of continuous jugular bulb
a temporary measure to improve cerebral oxygen delivery venous oximetry during intracranial surgery. J Neurosurg Anesthesiol
during hyperventilation.60,61 SjvO2 monitoring in such sce- 1995;7:174–7.
narios, allows fine-tuning of ventilation so that optimal brain 12. Bankier AA, Fleischmann D, Windiscch A, et al. Position of jugular oxygen
oxygenation is achieved (Fig. 32.3). saturation catheter in patients with head trauma: assessment by use of plain
films. Am J Radiol 1995;164:437–41.
Brain injury often is associated with impaired cerebral pres- 13. Gopinath SP, Valadka AB, Uzura M, et al. Comparison of jugular venous
sure autoregulation. The critical perfusion threshold can be oxygen saturation and brain tissue PO2 as monitors of cerebral ischemia
established by close examination of SjvO2 changes in response after head injury. Crit Care Med 27:2337–45, 1999
326 Section V—Cerebral Blood Flow
14. Souter MJ, Andrews PJD. Validation of the Edslab dual lumen oximetry
catheter for continuous monitoring of jugular bulb oxygen saturation after
severe head injury. Br J Anaesth 1996;76:744–6.
15. Coplin WM, O’Keefe GE, Grady MS, et al. Accuracy of continuous jugular
bulb oximetry in the intensive care unit. Neurosurgery 1998;42(3):533–9.
16. De Deyne C, Decruyenaere J, Colardyn F. How to interpret jugular bulb
oximetry? In: Vincent JL, editor. Yearbook of intensive care and emergency
medicine. Berlin: Springer-Verlag, 1996. p. 731–41.
17. Lewis SB, Myburgh JA, Reilly PL. Detection of cerebral venous desaturation
by continuous jugular bulb oximetry following acute neurotrauma. Anaesth
Intensive Care 1995;23:307–14.
18. Matta BF, Lam AM. The speed of blood withdrawal affects the accuracy of
jugular venous bulb oxygen saturation measurements. Anesthesiology 1996;
86:806–8.
19. Robertson CS, Gopinath SP, Goodman C, et al. SjvO2 monitoring in head
injured patients. J Neurotrauma 1995;12:891–6.
20. Chan MT, Ng SC, Lam JM, et al. Re-defining the ischemic threshold for
jugular venous oxygen saturation—a microdialysis study in patients with
severe head injury. Acta Neurochir 2005 Suppl;95:63–6.
21. Chieregato A, Marchi M, Fainardi E, et al. Cerebral arterio-venous PaCO2
difference, estimated respiratory quotient, and early posttraumatic outcome:
comparison with arterio-venous lactate and oxygen differences. J Neurosurg
Anesthesiol 2007;19(4):222–8.
22. Chieregato A, Targa L, Zatelli R. Limitations of jugular bulb oxyhemoglobin
saturation without intracranial pressure monitoring in subarachnoid
hemorrhage. J Neurosurg Anesthesiol 1996;8:21–5.
23. Gupta AK, Hutchinson PJ, Al-Rawi P, et al. Measuring brain tissue
oxygenation compared with jugular venous oxygen saturation for
monitoring cerebral oxygenation after traumatic brain injury. Anesth Analg
1999;88:549–53.
24. Coles JP, Fryer TD, Smielewski P, et al. Incidence and mechanisms of cerebral
ischemia in early clinical head injury. J Cereb Blood Flow Metab 2004;24(2):
202–11.
25. Haranishi Y, Kuroda Y, Matayoshi Y, et al. Infusion rate of propofol and
jugular venous oxygen saturation. J Anesth 2007;21(4):516–8. Epub 2007
Nov 1.
26. Goetting MG, Preston G. Jugular bulb catheterization: experience with 123
patients. Crit Care Med 1990;18(11):1220–3.
27. Schneider GH, Helden AV, Lanksch WR, et al. Continuous monitoring of
jugular bulb oxygen saturation in comatose patients—therapeutic
implications. Acta Neurochir (Wein) 1995;134:71–5.
28. Sheinberg GM, Kanter MJ, Robertson CS, et al. Continuous monitoring of
jugular venous oxygen saturation in head-injured patients. J Neurosurg
1992;76:212–7.
29. Gopinath SP, Rogertson CS, Constant CF, et al. Jugular venous desaturation
and outcome after head injury. J Neurol Neurosurg Psychiatry 1994;57:
717–23.
30. Robertson CS, Contant CF, Narayan RK, Grossman RG. Cerebral blood flow,
AVDO2, and neurologic outcome in head-injured patients. J Neurotrauma
1992;9(Suppl 1):S349–58.
31. Jaggi JL, Obrist WD, Gennarelli TA, et al. Relationship of early cerebral
blood flow and metabolism to outcome in acute head injury. J Neurosurg
1990;72:176–82.
32. Le Roux P, Lam AM, Newell DW, et al. Cerebral arteriovenous difference of
oxygen: a predictor of cerebral infarction and outcome in severe head injury.
J Neurosurg 1997;87:1–8.
33. De Deyne C, Decruyenaere J, Calle P, et al. Analysis of very early jugular bulb
oximetry data after severe head injury: implications for the emergency
management. Eur J Emerg Med 1996;3(2):69–72.
34. Robertson CS, Grossman RG, Goodman C, et al. The predictive value of
cerebral anerobic metabolism with infarction after head injury. J Neurosurg
1987;67:361–8.
35. Poca MA, Sahuquillo J, Vilalta A, et al. Lack of utility of arteriojugular
venous differences of lactate as a reliable indicator of increased brain
anaerobic metabolism in traumatic brain injury. J Neurosurg 2007;106(4):
530–7.
36. Menon DK, Day D, Kuc RE, et al. Arteriojugular endothelin-1 gradients in
aneurysmal subarachnoid haemorrhage. Clin Sci (Lond) 2002;103(Suppl
48):399S–403S.
37. Chatfield DA, Brahmbhatt DH, Sharp T, et al. Juguloarterial endothelin-1
gradients after severe traumatic brain injury. Neurocrit Care 2011;14(1):
55–60.
38. Helmy A, Carpenter KL, Menon DK, et al. The cytokine response to human
traumatic brain injury: temporal profiles and evidence for cerebral
parenchymal production. J Cereb Blood Flow Metab 2011;31(2):658–70.
Epub 2010 Aug 18.
39. Muroi C, Mink S, Seule M, et al. Monitoring of the inflammatory response
after aneurysmal subarachnoid haemorrhage in the clinical setting: review of
literature and report of preliminary clinical experience. Acta Neurochir
Suppl 2011;110(Pt 1):191–6.
40. Nekludov M, Antovic J, Bredbacka S, et al. Coagulation abnormalities
associated with severe isolated traumatic brain injury: cerebral arterio-
venous differences in coagulation and inflammatory markers. J Neurotrauma
2007;24(1):174–80.
41. Paolin A, Nardin L, Gaetani P, et al. Oxidative damage after severe head
injury and its relationship to neurological outcome. Neurosurgery 2002;
51(4):949–54; discussion 954–5.
42. Powner DJ, Hergenroeder GW, Awili M, et al. Hyponatremia and
comparison of NT-pro-BNP concentrations in blood samples from jugular
bulb and arterial sites after traumatic brain injury in adults: a pilot study.
Neurocrit Care 2007;7(2):119–23.
43. Jägersberg M, Schaller C, Boström J, et al. Simultaneous bedside
assessment of global cerebral blood flow and effective cerebral perfusion
pressure in patients with intracranial hypertension. Neurocrit Care 2010;
12(2):225–33.
44. Moss E, Dearden NM, Berridge JC. Effects of changes in mean arterial
pressure on SjvO2 during cerebral aneurysm surgery. Br J Anaesth 1995;75:
527–30.
45. Rozet I, Newell DW, Lam AM. Intraoperative jugular bulb desaturation
during acute aneurysmal rupture. Can J Anaesth 2006;53(1):97–100.
46. Crossman J, Banister K, Bythell V, et al. Predicting clinical ischaemia during
awake carotid endarterectomy: use of the SjvO2 probe as a guide to selective
shunting. Physiol Meas 2003;24:347–35.
47. Moritz S, Kasprzak P, Woertgen C, et al. The accuracy of jugular bulb
venous monitoring in detecting cerebral ischemia in awake patients
undergoing carotid endarterectomy. J Neurosurg Anesthesiol 2008;20:
8–14.
48. Ogawa M, Sugimoto T, Katsurada T. Hyperoxia of internal jugular venous
blood in brain death. J Neurosurg 1973;39:442–7.
49. Díaz-Regañón G, Miñambres E, Holanda M, et al. Usefulness of venous
oxygen saturation in the jugular bulb for the diagnosis of brain death: report
of 118 patients. Intensive Care Med 2002;28:1724–8.
50. Miura N, Yoshitani K, Kawaguchi M, et al. Jugular bulb desaturation during
off-pump coronary artery bypass surgery. J Anesth 2009;23(4):477–82. Epub
2009 Nov 18.
A complete list of references for this chapter can be found online at
www.expertconsult.com.

References
1. Ketty SS, Schmidt CF. The nitrous oxide method for the quantitative
determination of cerebral blood flow in man: theory, procedure and normal
values. J Clin Invest 1948;27:476–83.
2. Obrist W, Langfitt T, Jaggi J, et al. Cerebral blood flow and metabolism in
comatose patients with acute head injury: relationship to intracranial
hypertension. J Neurosurg 1984;61:241–53.
3. National Institute for Clinical Excellence. NICE technology appraisal
guidance No 49: guidance on the use of ultrasound locating devices for
placing central venous catheters. 2002. London NICE Available from
www.nice.org.uk/pdf/ultrasound_49_GUIDANCE.pdf
4. Matta BF, Lam AM, Mayberg TS, et al. A critique of the intraoperative use of
jugular venous bulb catheters during neurosurgical procedures. Anesth Analg
1994;79:745–50.
5. Lam JM, Chan MS, Poon WS. Cerebral venous oxygen saturation
monitoring: is dominant jugular bulb cannulation good enough? Br J
Neurosurg 1996;10(4):357–64.
6. Gibbs EL, Gibbs FA. The cross sectional areas of the vessels that form the
torcular and the manner in which blood is distributed to the right and to
the left lateral sinus. Anat Rec 1934;54:419.
7. Gibbs EL, Lennox WG, Giggs FA. Bilateral internal jugular blood,
comparison of A-V differences, oxygen-dextrose ratios and respiratory
quotients. Am J Psychiatry 1945;102:184–90.
8. Stocchetti N, Paparella A, Brindelli F, et al. Cerebral venous oxygen
saturation studied with bilateral samples in the internal jugular veins.
Neurosurgery 1994;34:38–44.
9. Andrews PJD, Dearden NM, Miller JD. Jugular bulb cannulation: description
of a cannulation technique and a validation of a new continuous monitor.
Br J Anaesth 1991;67:553–8.
10. Matta BF, Lam AM, Mayberg TS, et al. A critique of the intraoperative use of
jugular venous bulb catheters during neurosurgical procedures. Anesth Analg
1994;79:745–50.
11. Gunn HC, Matta BF, Lam AM, et al. Accuracy of continuous jugular bulb
venous oximetry during intracranial surgery. J Neurosurg Anesthesiol
1995;7:174–7.
12. Bankier AA, Fleischmann D, Windiscch A, et al. Position of jugular oxygen
saturation catheter in patients with head trauma: assessment by use of plain
films. Am J Radiol 1995;164:437–41.
13. Gopinath SP, Valadka AB, Uzura M, et al. Comparison of jugular venous
oxygen saturation and brain tissue PO2 as monitors of cerebral ischemia
after head injury. Crit Care Med 27:2337–45, 1999
14. Souter MJ, Andrews PJD. Validation of the Edslab dual lumen oximetry
catheter for continuous monitoring of jugular bulb oxygen saturation after
severe head injury. Br J Anaesth 1996;76:744–6.
15. Coplin WM, O’Keefe GE, Grady MS, et al. Accuracy of continuous jugular
bulb oximetry in the intensive care unit. Neurosurgery 1998;42(3):533–9.
16. De Deyne C, Decruyenaere J, Colardyn F. How to interpret jugular bulb
oximetry? In: Vincent JL, editor. Yearbook of intensive care and emergency
medicine. Berlin: Springer-Verlag; 1996. p. 731–41.
17. Lewis SB, Myburgh JA, Reilly PL. Detection of cerebral venous desaturation
by continuous jugular bulb oximetry following acute neurotrauma. Anaesth
Intensive Care 1995;23:307–14.
18. Matta BF, Lam AM. The speed of blood withdrawal affects the accuracy of
jugular venous bulb oxygen saturation measurements. Anesthesiology 1996;
86:806–8.
19. Robertson CS, Gopinath SP, Goodman C, et al. SjvO2 monitoring in head
injured patients. J Neurotrauma 1995;12:891–6.
20. Chan MT, Ng SC, Lam JM, et al. Re-defining the ischemic threshold for
jugular venous oxygen saturation—a microdialysis study in patients with
severe head injury. Acta Neurochir 2005 Suppl;95:63–6.
21. Chieregato A, Marchi M, Fainardi E, et al. Cerebral arterio-venous PaCO2
difference, estimated respiratory quotient, and early posttraumatic outcome:
comparison with arterio-venous lactate and oxygen differences. J Neurosurg
Anesthesiol 2007;19(4):222–8.
22. Chieregato A, Targa L, Zatelli R. Limitations of jugular bulb oxyhemoglobin
saturation without intracranial pressure monitoring in subarachnoid
hemorrhage. J Neurosurg Anesthesiol 1996;8:21–5.
23. Gupta AK, Hutchinson PJ, Al-Rawi P, et al. Measuring brain tissue
oxygenation compared with jugular venous oxygen saturation for
monitoring cerebral oxygenation after traumatic brain injury. Anesth Analg
1999;88:549–53.
24. Coles JP, Fryer TD, Smielewski P, et al. Incidence and mechanisms of cerebral
ischemia in early clinical head injury. J Cereb Blood Flow Metab 2004;24(2):
202–11.
Section V—Cerebral Blood Flow 326.e1
25. Haranishi Y, Kuroda Y, Matayoshi Y, et al. Infusion rate of propofol and
jugular venous oxygen saturation. J Anesth 2007;21(4):516–8. Epub 2007
Nov 1.
26. Goetting MG, Preston G. Jugular bulb catheterization: experience with 123
patients. Crit Care Med 1990;18(11):1220–3.
27. Schneider GH, Helden AV, Lanksch WR, et al. Continuous monitoring of
jugular bulb oxygen saturation in comatose patients—therapeutic
implications. Acta Neurochir (Wein) 1995;134:71–5.
28. Sheinberg GM, Kanter MJ, Robertson CS, et al. Continuous monitoring of
jugular venous oxygen saturation in head-injured patients. J Neurosurg
1992;76:212–7.
29. Gopinath SP, Rogertson CS, Constant CF, et al. Jugular venous desaturation
and outcome after head injury. J Neurol Neurosurg Psychiatry 1994;57:
717–23.
30. Robertson CS, Contant CF, Narayan RK, Grossman RG. Cerebral blood flow,
AVDO2, and neurologic outcome in head-injured patients. J Neurotrauma
1992;9(Suppl 1):S349–58.
31. Jaggi JL, Obrist WD, Gennarelli TA, et al. Relationship of early cerebral
blood flow and metabolism to outcome in acute head injury. J Neurosurg
1990;72:176–82.
32. Le Roux P, Lam AM, Newell DW, et al. Cerebral arteriovenous difference of
oxygen: a predictor of cerebral infarction and outcome in severe head injury.
J Neurosurg 1997;87:1–8.
33. De Deyne C, Decruyenaere J, Calle P, et al. Analysis of very early jugular bulb
oximetry data after severe head injury: implications for the emergency
management. Eur J Emerg Med 1996;3(2):69–72.
34. Robertson CS, Grossman RG, Goodman C, et al. The predictive value of
cerebral anerobic metabolism with infarction after head injury. J Neurosurg
1987;67:361–8.
35. Poca MA, Sahuquillo J, Vilalta A, et al. Lack of utility of arteriojugular
venous differences of lactate as a reliable indicator of increased brain
anaerobic metabolism in traumatic brain injury. J Neurosurg 2007;106(4):
530–7.
36. Menon DK, Day D, Kuc RE, et al. Arteriojugular endothelin-1 gradients in
aneurysmal subarachnoid haemorrhage. Clin Sci (Lond) 2002;103(Suppl
48):399S–403S.
37. Chatfield DA, Brahmbhatt DH, Sharp T, et al. Juguloarterial endothelin-1
gradients after severe traumatic brain injury. Neurocrit Care
2011;14(1):55–60.
38. Helmy A, Carpenter KL, Menon DK, et al. The cytokine response to human
traumatic brain injury: temporal profiles and evidence for cerebral
parenchymal production. J Cereb Blood Flow Metab 2011;31(2):658–70.
Epub 2010 Aug 18.
39. Muroi C, Mink S, Seule M, et al. Monitoring of the inflammatory response
after aneurysmal subarachnoid haemorrhage in the clinical setting: review of
literature and report of preliminary clinical experience. Acta Neurochir
Suppl 2011;110(Pt 1):191–6.
40. Nekludov M, Antovic J, Bredbacka S, et al. Coagulation abnormalities
associated with severe isolated traumatic brain injury: cerebral arterio-
venous differences in coagulation and inflammatory markers. J Neurotrauma
2007;24(1):174–80.
41. Paolin A, Nardin L, Gaetani P, et al. Oxidative damage after severe head
injury and its relationship to neurological outcome. Neurosurgery 2002;
51(4):949–54; discussion 954–5.
42. Powner DJ, Hergenroeder GW, Awili M, et al. Hyponatremia and
comparison of NT-pro-BNP concentrations in blood samples from jugular
bulb and arterial sites after traumatic brain injury in adults: a pilot study.
Neurocrit Care 2007;7(2):119–23.
43. Jägersberg M, Schaller C, Boström J, et al. Simultaneous bedside assessment
of global cerebral blood flow and effective cerebral perfusion pressure in
patients with intracranial hypertension. Neurocrit Care 2010;12(2):
225–33.
44. Moss E, Dearden NM, Berridge JC. Effects of changes in mean arterial
pressure on SjvO2 during cerebral aneurysm surgery. Br J Anaesth
1995;75:527–30.
45. Rozet I, Newell DW, Lam AM. Intraoperative jugular bulb desaturation
during acute aneurysmal rupture. Can J Anaesth 2006;53(1):97–100.
46. Crossman J, Banister K, Bythell V, et al. Predicting clinical ischaemia during
awake carotid endarterectomy: use of the SjvO2 probe as a guide to selective
shunting. Physiol. Meas 2003;24:347–35.
47. Moritz S, Kasprzak P, Woertgen C, et al. The accuracy of jugular bulb venous
monitoring in detecting cerebral ischemia in awake patients undergoing
carotid endarterectomy. J Neurosurg Anesthesiol 2008;20(1):8–14.
48. Ogawa M, Sugimoto T, Katsurada T. Hyperoxia of internal jugular venous
blood in brain death. J Neurosurg 1973;39:442–7.
326.e2 Section V—Cerebral Blood Flow
49. Díaz-Regañón G, Miñambres E, Holanda M, et al. Usefulness of venous
oxygen saturation in the jugular bulb for the diagnosis of brain death: report
of 118 patients. Intensive Care Med 2002;28:1724–8.
50. Miura N, Yoshitani K, Kawaguchi M, et al. Jugular bulb desaturation during
off-pump coronary artery bypass surgery. J Anesth 2009;23(4):477–82. Epub
2009 Nov 18.
51. Croughwell ND, Newman MF, Blumenthal JA, et al. Jugular bulb saturation
and cognitive dysfunction after cardiopulmonary bypass. Ann Thorac Surg
1994;58:1702–8.
52. Croughwell ND, White WD, Smith LR, et al. Jugular bulb saturation and
mixed venous saturation during cardiopulmonary bypass. J Card Surg 1995;
10:503–8.
53. Iwata M, Inoue S, Kawaguchi M, et al. The effect of hypercapnia and
hypertension on cerebral oxygen balance during one-lung ventilation for
lung surgery during propofol anesthesia. J Clin Anesth 2010;22(8):608–13.
54. Takasu A, Yagi K, Ishihara S, Okada Y. Combined continuous monitoring of
systemic and cerebral oxygen metabolism after cardiac arrest. Resuscitation
1995;29:189–94.
55. Bisschops LL, Hoedemaekers CW, Simons KS, et al. Preserved metabolic
coupling and cerebrovascular reactivity during mild hypothermia after
cardiac arrest. Crit Care Med 2010;38(7):1542–7.
56. Matta B, Menon D. Severe head injury in the United Kingdom and Ireland: a
survey of practice and implications for management. Crit Care Med 1996;
24:1743–8.
57. Muizelaar JP, Marmarou A, Ward JD, et al. Adverse effects of prolonged
hyperventilation in patients with severe head injury: a randomised clinical
trial. J Neurosurg 1991;75:731–9.
58. Schneider GH, Helden AV, Lanksch WR, et al. Continuous monitoring of
jugular bulb oxygen saturation in comatose patients—therapeutic
implications. Acta Neurochir 1995;134:71–5.
59. Sheinbrg M, Kanter MJ, Rogertson CS. Continuous monitoring of jugular
venous oxygen saturation in head-injured patients. J Neurosurg
1992;76:212–7.
60. Matta BF, Lam AM, Mayberg TS. The influence of arterial hyperoxygenation
on cerebral venous oxygen content during hyperventilation. Can J Anaesth
1994;41:1041–6.
61. Thiagarajan A, Goverdhan P, Chari P, et al. The effect of hyperventilation
and hyperoxia on cerebral venous oxygen saturation in patients with
traumatic brain injury. Anesth Analg 1998;87:850–3.
62. Lewis S, Wong M, Myburgh J, et al. Determining cerebral perfusion pressure
thresholds in severe head trauma. Acta Neurochir Suppl (Wien) 1998;
71:174–6.

Near Infrared Spectroscopy
Pippa G. Al-Rawi and Peter J. Kirkpatrick
Introduction
Frans Jöbsis1 first described the use of in vivo tissue near
infrared spectroscopy (NIRS) in the human brain in 1977.
The technique is based on the concept that light of wave-
lengths 680 to 1000 nm is able to penetrate human tissue
and is absorbed by the chromophores oxyhemoglobin (HbO2)
and deoxyhemoglobin (HHb) and the cytochrome oxidase.
Changes in the detected light levels therefore can represent
changes in concentrations of these chromophores.
The noninvasive nature of the technique, which uses high-
intensity light emitted from sources placed on the surface of
the scalp, led to its first clinical application in monitoring the
cerebral oxygenation status of premature infants. Initially,
near infrared spectroscopy NIRS was not quantified and only
provided a trend of increased or decreased oxygenation. The
early machines were large and cumbersome, and the detected
signals showed considerable drift and were prone to move-
ment artifact.
The clinical use of NIRS for monitoring the brain is well
established in neonates, in whom transillumination is possible
because of the thin skull and small dimensions.2 However, the
clinical application of NIRS to monitor the adult brain has
been hampered because it must be applied in reflectance
mode. This has resulted in concerns about quantification, the
volume and type of tissue being illuminated, and most signifi-
cantly the issue of signal contamination by the extracranial
tissue layers.3 A number of algorithms have been applied to
try to overcome these issues, and techniques such as time-
resolved, phase-resolved, and spatially resolved spectroscopy
have been developed.
The use of NIRS to evaluate cerebral hemodynamic changes
has been reported in a variety of medical and neurosurgical
conditions, including those associated with disturbed cerebral
circulation.4-12 The authors’ experience has shown that in
patients with head injury, changes in oxyhemoglobin correlate
well with changes in jugular venous oxygen saturation (SjvO2),
transcranial Doppler (TCD), and laser Doppler.9 The authors
also have seen that NIRS can provide a more sensitive indica-
tor of desaturation events than SjvO2, and a threshold for
cerebral ischemia has been defined.13
NIRS is an evolving technology, and there has been renewed
interest in NIRS as an easy-to-use, noninvasive technique to
© Copyright 2013 Elsevier Inc. All rights reserved.
Chapter
33  
V
measure tissue oxygenation in the adult brain. In addition,
NIRS use has expanded to examine oxygenation in other
tissues and to estimate the adequacy of systemic circulation14;
this can be particularly useful in resuscitation from shock or
sepsis.15-17 Technologic advances have led to the development
of compact, portable instruments that detect changes in
optical attenuation of several wavelengths of light, with the
potential to derive a tissue oxygen saturation from spontane-
ous HHb and HbO2 signal changes.13,18-20 NIRS shows promise
as a clinical tool in the context of bedside cerebral blood flow
measurements, and to image the brain.21
This chapter looks at the clinical application of NIRS but
stresses the importance of both the assumptions on which
NIRS is based, and the limitations of the technology, to inter-
pret the results correctly. It describes the authors’ use of this
technique and how it is integrated with other monitoring
modalities.
Basic Principles
The theory behind NIRS is described in detail elsewhere,22-24
but the basic principle is based on the fact that light in the
near infrared range (680-1000 nm) can pass through skin,
bone, and other tissues with relative ease. The near infrared
light is delivered via fiber-optic bundles, or optodes, placed
either on opposite sides of the head, or close together (3.5-6 cm
distance) on the same side (Fig. 33.1). Light penetrates the
scalp and brain tissue and is subjected to scatter and absorp-
tion within the various tissues. A fraction of the photons is
captured by the second optode and conveyed to a measuring
device.
The concentration of some light-absorbing compounds
such as melanin, bilirubin, and water remains virtually con-
stant over time. However, the concentrations of others, such
as HbO2, HHb, and cytochrome oxidase, vary according to
tissue oxygenation and metabolism. Cytochrome oxidase, the
terminal enzyme of the respiratory chain, also contributes to
the near infrared spectrum of cortical tissue. To calculate the
concentration of the chromophobes requires knowledge of
the absorption spectra for HbO2 and HHb, and assumes
constant scattering properties. The modified Beer-Lambert
law that includes a differential path length factor to account
for scattering within the tissue can be used to convert
327
328 Section V—Cerebral Blood Flow
Emitter
Fig. 33.1  Schematic representation of near infrared
spectroscopy on a patient’s head. CSF, Cerebrospinal
fluid.
measurements of light attenuation into concentrations of
these chromophores. The Beer-Lambert law is defined as:
∆Α = L ⋅ ∆µa
where A = light attenuation, L = differential path length, and
µa = absorption scattering coefficient.
This calculation is very reliable when applied to an infant’s
head because the skull is thin enough to allow transillumina-
tion of light from one side to the other. In an adult, however,
the relative thickness of the scalp, skull, and brain prevents
transmission spectroscopy, and NIRS must be used in reflec-
tance mode, with the emitting and receiving optodes placed
on the same side of the head. Reflectance mode oximetry
depends on a proportion of the light that passes through brain
tissue. The human head is comprised of multiple tissue layers,
all of which have different scattering properties and contain
varying concentrations of light-absorbing compounds.25 This
results in the introduction of unknown, nonlinear variables
for light absorption and scattering coefficients.26 Various algo-
rithms have been applied to attempt quantification, but doubts
are expressed over their accuracy.27-30
Extracranial Contamination
Extracranial contamination is a major issue during the use
of NIRS in the adult brain, although extracranial contribu-
tions to NIRS measurements initially were thought to be
insignificant. Bone was considered to be transparent to light,
whereas the skin contribution was estimated to be approxi-
mately 5%. Over time, however, numerous studies have
demonstrated that extracranial contamination is a significant
problem.29,31-33
In reflectance mode, the near infrared light emitter and
detectors are placed within a few centimeters of each other
on the surface of the scalp. The larger the optode distance
used, the greater the proportion of near infrared light that
should pass through the brain tissue. However, even when a
large inter-optode distance is used there is still a major con-
tribution from the extracranial tissue to the NIRS signal.34
Other methods are used to correct for the effects of the
extracranial tissue layers; one is to subtract measurements
from two receiving optodes placed in line with the transmit-
ting optode. The underlying assumption is that the path
length of light through the extracranial tissues is the same for
m Detector
6.0 c
4.5-
Skin
Bone
CSF
Brain
tissue
both of the receiving probes. The difference in absorption
coefficients between the two probes should then reflect
absorption and scatter within the cerebral tissue alone.
However, this is an oversimplification that has resulted in
measurement inaccuracies.35
The concepts about light distribution in the adult head are
complex and are still being investigated. Mathematical in vitro
modeling has attempted to predict the distribution of light
through the extracerebral layers.36 Multilayer studies have
highlighted the influence that superficial tissue layers can have
on light distribution in tissues and provide further evidence
that NIRS measurements may underestimate changes in con-
centration. Mathematical simulations using Monte Carlo
techniques and finite element modeling have demonstrated
that the cerebrospinal fluid (CSF) layer that surrounds the
brain can act as an optical short-cut for light as it passes
through the tissues. It has been suggested that this phenom-
enon can prevent the near infrared light from penetrating
any deeper than the cerebral grey matter, or even cause it to
bypass the cerebral compartment altogether.37,38
Modern near infrared spectrometers detect oxygenation
changes in cerebral tissues. However, many machines still
remain sensitive, by varying degrees, to changes in extracere-
bral oxygenation. These changes can influence estimates of
cerebral oxygenation. Conditions in which changes in extra-
cerebral oxygenation may seem unlikely should not be taken
for granted. For example, changes in extracerebral oxygen-
ation in the scalp have been detected during the performance
of functional activation tasks. Even in the absence of extrace-
rebral oxygenation changes, the extracerebral tissues may con-
found attempts to calibrate and quantify changes in cerebral
oxygenation. In any NIRS study it is essential to consider
whether the changes detected could be due, even in part, to
extracerebral contamination. To reliably measure cerebral
oxygenation with NIRS, extracerebral changes must be shown
to be insignificant, or their effect eliminated.
Instrumentation
There has been rapid growth and development of NIRS tech-
nology, and novel optical instruments continue to be devel-
oped and evaluated for diagnostic and functional imaging in
the clinical environment. Although the early machines
were useful only as trend monitors, current commercial

machines detect changes in optical attenuation of a number
of wavelengths of light, are compact, are portable, and enable
noninvasive measurements of cerebral oxygenation at the
bedside.13,18-20
The INVOS Cerebral Oximeter (Somanetics, Troy, MI) and
the Hamamatsu 100, 200, and 300 (Hamamatsu Photonics
KK, Hamamatsu, Japan) are among popular commercially
available devices for brain monitoring. In addition, multi-
probe devices (e.g., INVOS 5100) are used in newborns or
pediatric intensive care units (ICUs) to examine regional cere-
bral (rSO2C), splanchnic (rSO2S), and renal (rSO2R) tissue
oxygenation. Several research groups also have produced their
own specialized in-house NIRS equipment, which may be
available locally. When published values from different instru-
ments are compared, it should be remembered that different
instruments may provide different values for cerebral tissue
oxygen saturations because the data are derived from different
volumes of tissue.39 Among current techniques, spatially
resolved spectroscopy appears to be the most promising, and
reliably shows changes in chromophores in brain tissue.13,40,41
Spatially Resolved Spectroscopy
Spatially resolved spectroscopy is described in a number of
publications.8,42,43 Spatial resolution relies on the measure-
ment of the attenuation gradient as a function of source-
detector separation. Using a modified diffusion equation, a
product of the absorption and scattering coefficients is calcu-
lated. When the tissue is treated as homogeneous, the scatter-
ing coefficient can be assumed to be a constant (k) in the near
infrared wavelength. To increase the accuracy of the calcula-
tion, however, a wavelength dependency for the scattering
coefficient is derived in the form of:
k(1− hλ)
where λ = wavelength and h is the normalized slope of the
scattering coefficient along λ. From here, the relative absorp-
tion coefficients and thus the relative concentrations of HbO2
and HHb can be obtained.
Hamamatsu NIRO 300
The NIRO 300 is a noninvasive bedside monitor that provides
continuous online measurements of hemoglobin and cyto-
chrome oxidase concentrations, and a calculated tissue oxygen
index (TOI). The optodes are held in a black custom-made
holder that allows them to be set at either a 4.5- or 5-cm dis-
tance. By using two sets of optodes, two channel simultaneous
measurements are possible.
Four wavelengths of light (775, 810, 850, and 910 nm) are
delivered by four pulsed laser diodes, and the scattered light
is detected by three closely placed photodiodes. The concen-
tration changes of the chromophores HbO2, HHb, total hemo-
globin, and cytochrome oxidase are measured by conventional
differential spectroscopy using a modified Beer-Lambert
law,44,45 whereas the basic principle behind calculation of the
tissue oxygenation index is spatially resolved reflectance spec-
troscopy. The basic measurement made is the rate of increase
of light attenuation with respect to source or detector spacing,
and TOI is calculated from this using photon diffusion
theory.43 The hemoglobin and cytochrome concentration
Section V—Cerebral Blood Flow 329
changes are measured by the middle photodiode, whereas TOI
is measured by using all three. TOI is the ratio of oxygenated
to total tissue hemoglobin and can be expressed as:
HbO2
× 100
HbO2 + HHb
The NIRO 200 and 100 are the latest commercial updates to
the NIRO 300 and use only three wavelengths of delivered
light (775, 810, and 850 nm) for the measurement calcula-
tions. The NIRO 200 is a two-channel system that allows
bihemispheric data to be collected, whereas the NIRO 100 is
a single-channel machine.
Clinical Neuromonitoring
Clinical Application
NIRS has many applications in the ICU and the operating
room, and it can be used to detect real-time changes in
regional oxygen saturation of cerebral and somatic tissues.14
Its use is well described in cardiac surgery46 and neonatal and
pediatric critical care,2,47-49and most recently several studies
have described use of NIRS to evaluate tissue perfusion and
the dynamic response of the microcirculation to a stress test
in patients who are dehydrated, septic, or in shock.15-17,48,50
These studies in critically ill adults suggest NIRS can be used
to guide goal-directed therapy and that blood pressure is not
always a reliable surrogate for peripheral blood flow and
oxygen delivery. In addition, greater mortality is observed in
septic patients in whom forearm skeletal muscle rSO2 is less
than or equal to 60% throughout the first 24 hours of ICU
care and low tissue oxygen saturation (StO2) persists in sicker
patients (e.g., greater Acute Physiology and Chronic Health
Evaluation [APACHE] or Sequential Organ Failure Assess-
ment [SOFA] scores). This information about systemic
oxygenation can be very useful in many patients in the neuro-
critical care unit (NCCU) who may develop sepsis or systemic
inflammatory response syndrome (SIRS) among other sys-
temic disorders, because many shock states have been shown
to be associated with a decrease in perfused capillary density
and an increase in the heterogeneity of microcirculatory
perfusion.15
The ability to monitor cerebral oxygenation noninvasively
and at the bedside in patients at risk of cerebral ischemia is a
potential advantage of NIRS and its use to evaluate changes
in regional cerebral oxygenation, cerebral blood flow (CBF),
and oxygen utilization in the brain is reported in a variety of
medical and neurosurgical conditions,8,13,14,51-54 in particular
those associated with disturbed cerebral circulation such as
head injury and intracranial hemorrhage, and in patients
undergoing carotid endarterectomy.9,55-57 In head-injured
patients changes in HbO2 detected by NIRS appear to correlate
well with changes in SjvO2, TCD, and laser Doppler flowmetry
(LDF).9 The information for NIRS also appears to have some
prognostic information (e.g., cerebral oxygen desaturation is
associated with prolonged intensive care and total hospital
lengths of stay).58 Zweifel et al.4 compared an NIRS-based
index of cerebrovascular pressure reactivity (PRx), called total
hemoglobin reactivity (THx), against standard measurements
of PRx that requires invasive intracranial pressure (ICP) mon-
itoring. PRx and THx had a significant association across
330 Section V—Cerebral Blood Flow
averaged individual recordings and across patients. Optimal
cerebral perfusion pressure (CPP) could also be assessed with
THx in about half of patients. These data suggest that NIRS
may be useful in some traumatic brain injury (TBI) patients
to optimize and target therapy particularly those who do not
have an ICP monitor. Others have described the use of NIRS
to help guide therapy for vasospasm after subarachnoid hem-
orrhage (SAH)9 and observed that NIRS may distinguish dif-
ferent cerebral oxygenation patterns associated with different
seizure types.59
Intraoperative use of NIRS during cardiac surgery7,8 or
carotid endarterectomy (CEA) is well described.10,13,55,60,61
These studies have helped to define a potential NIRS-based
“threshold” for ischemia particularly when using spatially
resolved spectroscopy (e.g., NIRO 300, Hamamatsu Photon-
ics).13 However, it is difficult to determine how useful NIRS is
during cardiac surgery without fully understanding how
extracorporeal circulation affects the assumptions on which
NIRS technology is based. The information from CEA studies
may be more relevant to the NCCU. For example, the NIRO
500 can be used successfully to monitor patients who undergo
CEA under general anesthesia.10 In some studies once the
extracranial component is subtracted, a threshold for severe
critical ischemia can be defined.62,63 This threshold then can
be used to guide the selective use of an intraoperative carotid
artery shunt. Other studies using bilateral cerebral oximetry
in patients undergoing CEA under regional anesthesia62 or
when INVOS is compared with somatosensory evoked poten-
tials51; however, they have not found it possible to determine
the critical cerebral oxygen saturation or change in saturation
that requires shunt insertion because of marked individual
variability of rSO2 and the derived changes. In addition, cere-
bral oxygen saturation may decrease without neurologic
dysfunction.
Troubleshooting
Troubleshooting advice provided by the manufacturer comes
with all near infrared spectrometers and in the first instance
the user should consult the appropriate operating guide.
However, some issues may be considered common to all NIRS
devices and studies.
First, the light source and detector optodes should be spaced
according to the study requirements, although increasing the
optode distance may help improve the sensitivity of NIRS to
brain tissue changes.30 An optode distance greater than 3 to
4 cm should reduce the extracerebral contribution because as
the optode distance is increased, a greater proportion of light
passes through the brain. Germon et al.34 in a study in which
optode distance was increased during NIRS studies suggest
that a separation of 5.5 cm may increase sensitivity to cerebral
changes, whereas extracranial effects may plateau at 4.8 to
5.5 cm. By contrast when placed close together most of the
transmitted light likely passes through the scalp and superfi-
cial tissue. For example, in normal volunteers during verbal
fluency tasks, optodes placed 5 mm apart may only measure
skin changes rather than reflect any cerebral function.64 The
distance between optodes and emitter also may depend on
practical constraints such as other monitors, surgical sites,
dressings, and the signal-to-noise ratio. Measurement of both
arterial blood pressure and peripheral saturation can address
in part how extracranial contamination affects the signal.
Direct monitoring of skin flow is possible using LDF to assess
relative changes in scalp blood flow. If the LDF probe is placed
on the skin in the immediate vicinity of the NIRS probes, skin
flow changes that may affect NIRS measurements can be
crudely monitored.
Second, most NIRS systems are placed in the frontotempo-
ral region for adult brain tissue monitoring. For optimal per-
formance the skin surface should be clean and dry and the
optodes placed high on the forehead and in a position to avoid
the sagittal sinus, muscle (temporalis muscle), and midline.
Hair also should be avoided whenever possible because it can
reduce return of the near infrared signal because of light
absorption by the hair follicles. When hair is present it also is
difficult to adequately seal out light. Third, ambient light
should be eliminated. The use of a bandage to eliminate light
that covers the optode holder is beneficial. It also will help to
reduce scalp movement during the study. Care is needed to
avoid tissue ischemia and necrosis especially with long-term
monitoring. Fourth, it is useful to support the optical fiber
cables to prevent both damage and the weight of the fibers
dislodging the optodes. Finally, using NIRS as part of a mul-
timodal monitoring system helps to confirm true clinical
changes versus artifact. When interpreting NIRS values, signal
change or percentage change from baseline may be more
useful than absolute values. This is particularly true if com-
parisons between individual patients, clinical centers, or NIRS
machines are required.
Limitations
A number of limitations should be considered before NIRS
can contribute to clinical decision making. First, the major
limitation of NIRS is extracranial contamination. The hetero-
geneity of the scalp and skull affect the transmission of near
infrared light through the head and may cause significant
individual variability. Second, NIRS monitors cerebral oxygen
saturation in an uncertain mixture of arteries, capillaries, and
veins. The sampling volume of gray and white matter is
unknown. Third, current NIRS probes are not magnetic reso-
nance imaging (MRI) compatible; this limits its use in poten-
tial MRI validation studies. Fourth, NIRS does not resolve
focal abnormalities in CBF but may be useful as a monitor of
general changes that indicate secondary brain injury. Fifth,
some early studies suggest elevated intracranial pressure may
limit the reliability of NIRS.65 Finally, the quantitative accu-
racy of calculated cerebral oxygen saturation still needs to be
established, and a range of normal values defined.
Evaluation of the NIRO 300 as a Monitor
of Cerebral Ischemia
Patients who undergo CEA often may experience brief periods
of cerebral ischemia during cross-clamping of the internal
carotid artery (ICA). During CEA, segregation of the NIRS
signal between intra- and extracranial vascular territories is
possible by staged application of vascular clamps to the exter-
nal and internal carotid arteries. In this way, the authors have
previously identified NIRS thresholds for cerebral ischemia by
extracting the extracranial signal.62
At the authors’ institution routine multimodal monitoring
during CEA includes frontal cutaneous LDF, TCD mean flow
velocity (FV) measurements of the ipsilateral middle cerebral

LDF
NIRS
TCD
cfm
Fig. 33.2  Intraoperative multimodal monitoring setup, including cerebral function monitoring (cfm), transcranial Doppler (TCD), laser Doppler flow
(LDF), and near infrared spectroscopy (NIRS).
artery, and mean arterial blood pressure (ABP).10 The LDF
(Moor Instruments Ltd, Axminster, UK) is modified to use
light in the visible spectrum (wavelength 650 nm) to avoid any
interference with the NIRS signal. The authors incorporated
the NIRO 300 into this multimodal monitoring system to
detect cerebral hemodynamic changes during highly con-
trolled conditions and determine the anatomic source of the
calculated TOI (%) (Fig. 33.2).13
The authors studied 167 patients who had elective CEA.
Monitoring was applied after induction of anesthesia. NIRS
optodes were placed high on the ipsilateral forehead to avoid
the temporalis muscle and hair and sufficiently lateral from
the midline to avoid the superior sagittal sinus. Optode spacing
was 5 cm (see Fig. 33.2). Sequential clamping of the external
carotid artery (ECA) followed by the ICA was performed
intraoperatively with a sufficient time interval (≈ 2 minutes)
to assess the extracranial contribution to the NIRS signal.
Simultaneous measurements of hemoglobin concentration
and TOI changes (ΔTOI) were assessed by observing changes
in LDF (ΔLDF) and FV (ΔFV) to identify the extracerebral
and intracerebral contribution to the NIRS signal. The time
of intraoperative events and in particular the time of vascular
clamp applications and data signals from all monitored
parameters were digitized and collected on computers using
specialized software.66 Maximum physiologic stability gener-
ally is observed at the time of clamp application rather than
removal; therefore data from this period were used for analy-
sis. Multivariable linear regression was used to calculate
the correlation coefficient between: (1) ΔTOI and ΔFV and
(2) ΔTOI and ΔLDF. To account for baseline variation in TOI,
interrupted time series analysis was applied. From these data
a baseline threshold of 2% was derived, and therefore a ΔTOI
greater than 2% was considered significant.
Section V—Cerebral Blood Flow 331
Typical data obtained during CEA for an individual patient
are illustrated in Figure 33.3. The authors found that changes
in TOI correlated with changes in intracranial blood flow
(i.e., FV), but not with cutaneous skin flow (LDF). Notably,
ΔTOI was not seen without an associated ΔFV. These obser-
vations demonstrate that TOI is derived largely from the
intracranial (ICA) vascular bed. The sensitivity and specificity
of TOI change to intracranial changes (as detected by TCD)
was 87.5% and 100%, respectively.67 Most importantly, the
authors defined a %ΔTOI threshold of −13%, above which no
patients showed any evidence of ischemia (Fig. 33.4). This
threshold provided 100% sensitivity and 93.2% specificity to
identify patients who met preset criteria for cerebral ischemia
during CEA.13
It is important to realize that both relative changes in chro-
mophore concentration (using a modified Beer-Lambert Law)
and a quantified TOI (using spatially resolved spectroscopy)
are provided with the NIRO 300. Biphasic changes were seen
in HbO2 and HHb on sequential clamping, whereas ΔTOI was
only associated with ICA-related chromophore changes (see
Fig. 33.3). The HbO2 and HHb values derived by the NIRO
300 therefore cannot reliably be taken to represent the intra-
cranial compartment in the same way as TOI and should be
interpreted with caution.
Cerebral Blood Volume
and Cerebral Blood Flow
There is no ideal method to measure cerebral blood volume
(CBV) or CBF safely, rapidly, repeatedly, and noninvasively at
the bedside. Established methods to measure CBF directly
tend to be technically challenging, are time consuming, use
radioactive substances, or require patient transport to an
332 Section V—Cerebral Blood Flow

ECA on ICA on Shunt inserted imaging site away from the NCCU.68-71 Because NIRS can
measure HHb and HbO2, it follows that it also can measure
1000 total hemoglobin (THgb). Assuming the whole blood HHb
800 concentration and the large to small cerebral vessel-to-hema-
LDF (au)

600 tocrit ratio remains constant for the duration of the measure-
400 ment, changes in THgb may be assumed to imply a change in
200
CBV, according to the following equation:
0 ∆CBV = ∆[THgb] × (0.89 / Hgb)
80
60 Calculation of absolute quantified CBV and CBF also is pos-
FVm (cm/s)

sible, based on the generalized application of the Fick prin-

40
ciple by using HbO2 and HHb as endogenous intravascular
20 tracer molecules, and assuming no venous outflow of the
0 chromophores during the measurement, quantified CBF and
4 CBV can be obtained by using transient graded hypoxia.72-75
HHb (µmol/L)

2.75 Despite a good agreement between NIRS CBF and 133Xe

washout,73,76 this technique has not been applied clinically
1.5
because of concerns over the induced hypoxemia and ques-
0.25 tions over the accuracy of the calculated CBV. Studies have
–1 shown significantly different CBV readings and a large inter-
4 subject variability.77
HbO2 (µmol/L)

2.39 However, it also is possible to measure blood flow by using

0.76 NIRS and a light-absorbing tracer. Broadly speaking, the rate
of accumulation of a tracer in a given tissue is equal to the
–0.87 rate of inflow minus the rate of outflow. If a tracer is intro-
–2.5 duced rapidly and the rate of accumulation is measured over
70 time, blood flow can be measured as a ratio of the tracer
62.5 accumulated to the quantity of tracer introduced, over a given
TOI (%)

55 time.
47.5
40
Indocyanine Green
0 1 2 3 4 5 6 7 8 9 Indocyanine green (ICG) is used as a diagnostic tool in
Time several medical applications. It is considered safe, and serious
side effects are uncommon. ICG shows a strong absorption in
Fig. 33.3  Graphic display of data obtained from a patient during the near infrared range, and changes in the attenuation of
elective carotid endarterectomy. The vertical lines demonstrate time
near infrared light, and therefore the optical density of bio-
of application of vascular clamps. ECA, External carotid artery; FVm,
mean ipsilateral middle cerebral artery flow velocity; HbO2, oxyhemoglobin
logic tissue, have been recorded after the administration of
concentration; HHb, deoxyhemoglobin concentration; ICA, internal ICG.78-81 Using ICG as the intravascular tracer, the arterial
carotid artery; LDF, frontal cutaneous laser Doppler flow; TOI, tissue concentration can be measured either by repeated blood
oxygen index.
10

5
0
–87 –77 –67 –57 –47 –37 –27 –17 7 3 13 23 33 43 53
% TOI diff (baseline, ICA clamp on)

–5

–10

–15

–20

–25

–30
Fig. 33.4  Scatterplot that compares percentage
change in middle cerebral artery (MCA) flow –35
velocity (%ΔFV) with percentage change in tissue
oxygen index (%ΔTOI) from baseline to –40
postapplication when the internal carotid artery –45
(ICA) was clamped in 167 patients during carotid
endarterectomy. Patients with a sustained fall in –50
ipsilateral cerebral function monitoring are
–55
marked in red. Dotted line indicates a level of
−13% ΔTOI. % FV diff (baseline, ICA clamp on)

sampling or with a peripheral dye densitometer, whereas the
concentration of ICG in the brain is measured by NIRS. The
theory behind ICG use during NIRS is described in detail
elsewhere.82 After baseline measurements rapid bolus injec-
tion of ICG (0.3-0.5 mg/kg) is given through a central vein
and the arterial concentration measured either by blood sam-
pling or with a peripheral dye densitometer. The rate of
arrival of ICG in the brain is detected by NIRS, and ICG con-
centrations can be calculated from the changes in optical
density.83 Using a modified Fick principle,84 the amount of
ICG dye delivered to the brain can be calculated from the
area under the curve of the arterial ICG concentration
change. CBF, CBV, and mean transit time (MTT) then can be
calculated.82-88
ICG NIRS has high spatial and temporal resolution, is safe,
and can be performed easily at the bedside. Most studies have
concluded that it shows reproducibility and merits further
development as a clinical tool. In animal studies a good cor-
relation is found with 133Xe washout88,89 and ICG used with
NIRS to measure CBF. The technique has been validated in
infants when compared with other methods for resting CBF
measurements.87,90 However, attempts to quantify NIRS ICG
in the adult brain have given erroneously low values.85,91-94
Further validation and modification of this technique as a
bedside measurement of CBF in adults are needed.
Near Infrared Optical Imaging
NIR optical imaging is an emerging technology with great
potential as an affordable noninvasive functional imaging
modality for use in clinical practice. It provides the opportu-
nity to study brain function through hemodynamic, meta-
bolic, and neuronal measures. van Houten et al.95 described
the first imaging spectrometer capable of providing a two-
dimensional image of the neonatal brain. Since then, a variety
of imaging methods have been applied,96-99 and continuous
real-time optical imaging of the brain has been shown to be
both feasible and achievable.
The major requirement for bedside optical imaging is an
NIRS device that can accurately measure photon path length
from multiple detectors and process this information into
tomographic images. The use of near infrared light for imaging
presents a special challenge because of the multiple scattering
that takes place within the tissue, which complicates the image
reconstruction. This has necessitated development of highly
sophisticated image reconstruction algorithms. The detailed
principles of these techniques are beyond the scope of this
chapter, and can be found elsewhere.100-102 Novel optical
instruments are being developed and evaluated by several
groups for diagnostic and functional imaging in the clinical
environment.103-109
NIR optical imaging involves acquiring multiple reflectance
measurements using small source-detector spacings placed
over a large area of the head, either simultaneously or in rapid
succession. The theory is based on a diffusion approximation,
and two-dimensional images of hemodynamic and oxygen-
ation changes are reconstructed, based on the chromophore
concentrations in the tissues. Brain activity with responses as
fast as 100 ms can be studied and the resolution obtained is
in the order of 10 mm. However, only the cortical surface is
imaged.
The first reported observations of NIRS absorbance changes
attributable to repetitive total Hb changes in response to
Section V—Cerebral Blood Flow 333
stimulation in the human brain frontal region by a cognitive
process, were published in 1993.110 Since then, optical topog-
raphy using multichannel NIRS instruments has been success-
fully used to assess infant and adult brain function in a wide
variety of tasks,98,108,111-114 and strong correlation, particularly
HbO2, has been found between functional magnetic resonance
imaging (fMRI) changes and NIR optical measures.114,115 The
current resolution of available instruments is similar to posi-
tron emission tomography (PET); however, changes in skin
blood flow still need to be accounted for.64
NIR optical imaging, in which a transverse slice or three-
dimensional image is obtained, is technically more difficult.
Measurements are made using large source-detector spacings,
and consequently light must be integrated over time intervals
of several seconds or longer to obtain an adequate signal.
Although this prevents monitoring of fast hemodynamic
changes associated with functional activation, it does obtain
information on oxygenation deeper in the brain, and has been
used to reveal variations in blood volume and tissue oxygen-
ation in newborn babies suffering from hypoxic-ischemic
brain injury.100,102,106,107 At publication, no standardized
approach for NIR optical imaging data analysis and image
reconstruction has been established.
Future Developments
NIRS is an evolving technology that holds significant potential
for technical advancement. Improvements in the methodol-
ogy, quantitative accuracy, and regional specificity will increase
its clinical applicability. In particular, NIRS shows promise as
a low-cost noninvasive clinical tool for bedside CBF measure-
ments and as a cerebral imaging modality to map structure
and function. For these techniques to improve, more sophis-
ticated imaging algorithms must be developed to account for
the heterogeneity of the tissues and to enable image recon-
struction with higher spatial resolution and sensitivity. Inte-
gration of NIRS into multimodality monitoring and imaging
systems, such as computed tomography (CT), MRI, and ultra-
sound, will allow for co-registration.
There are considerable research efforts to further develop
NIRS instrumentation and image reconstruction algorithms.
Portable NIRS instruments combined with wireless telemetry
are being tested. Soon it is likely that more multichannel
imaging near infrared spectrometers will be developed to
enable detailed spatial information on cerebral oxygenation,
perfusion, and blood flow to be obtained at the bedside. In
addition, combined ICP and NIRS monitors have been devel-
oped, and with these devices new algorithms can be developed
to subtract the extracerebral contamination from the NIRS
signal.116
Conclusion
All neuromonitoring methods, including NIRS, require tech-
nical expertise and experience to efficiently interpret the
results. However, NIRS has several potential advantages over
other monitors. It allows for continuous noninvasive moni-
toring that is safe and relatively easy to use and can register
complex mechanisms of oxygen balance in the brain. Having
a continuous, real-time measure of cerebral oxygenation can
potentially identify critical ischemic events before they mani-
fest clinically and, in some cases, before other monitors may
document changes. The major limitations associated with
334 Section V—Cerebral Blood Flow
NIRS, however, are changes that occur in the extracerebral
tissues that can significantly degrade the results. Among
current commercially available NIRS instruments, the NIRO
300 appears to most reliably reflect changes in cerebral tissue
oxygenation to a high degree of sensitivity and specificity, and
a threshold for cerebral ischemia during carotid endarterec-
tomy has been defined.
NIRS technology has developed rapidly. A number of
instruments for CBF, regional imaging, and functional activa-
tion are being developed. NIRS could be of enormous clinical
value as a portable bedside monitoring tool that can be
coupled with other cerebral imaging modalities such as MRI
and PET. However, before routine use for monitoring the
adult brain, it is recommended that several aspects of NIRS
technology need refinement. Further validation studies to
investigate NIRS reliability in different clinical scenarios are
required. The accuracy of the regional cerebral oxygen satura-
tion as an absolute value may be deceiving because no reliable
data about normal or critical threshold values exist, and so the
clinician is unable to check the accuracy of the given values.
For this reason, only the trend information should be included
in the evaluation of cerebral oximetry.
References
1. Jöbsis FF. Non-invasive infrared monitoring of cerebral and myocardial
oxygen sufficiency and circulatory parameters. Science 1977;198:1264–7.
2. Toet MC, Lemmers PM. Brain monitoring in neonates. Early Hum Dev
2009;85(2):77–84. Epub 2009, Jan 17.
3. Schwarz G, Litscher G, Kleinert R, et al. Cerebral oximetry in dead subjects.
J Neurosurg Anesthesiol 1996;8(3):189–93.
4. Zweifel C, Castellani G, Czosnyka M, et al. Noninvasive monitoring of
cerebrovascular reactivity with near infrared spectroscopy in head-injured
patients. J Neurotrauma 2010;27(11):1951–8. Epub 2010, Oct 28.
5. Kim MN, Durduran T, Frangos S, et al. Noninvasive measurement of
cerebral blood flow and blood oxygenation using near-infrared and diffuse
correlation spectroscopies in critically brain-injured adults. Neurocrit Care
2010;12(2):173–80.
6. Mutoh T, Ishikawa T, Suzuki A, et al. Continuous cardiac output and
near-infrared spectroscopy monitoring to assist in management of
symptomatic cerebral vasospasm after subarachnoid hemorrhage.
Neurocrit Care 2010;13(3):331–8.
7. du Plessis AJ, Newburger J, Jonas RA, et al. Cerebral oxygenation supply and
utilisation during infact cardiac surgery. Ann Neurol 1995;37:488–97.
8. Al-Rawi PG, Smielewski P, Hobbiger H, et al. Assessment of spatially
resolved spectroscopy during cardiopulmonary bypass. J Biomed Opt
1999;4(2):208–16.
9. Kirkpatrick PJ, Smielewski P, Czosnyka M, et al. Near infrared spectroscopy
use in patients with head injury. J Neurosurg 1995;83:963–70.
10. Kirkpatrick PJ, Smielewski P, Whitfield P, et al. An observational study of
near infrared spectroscopy during carotid endarterectomy. J Neurosurg
1995;82:756–63.
11. Kurth CD, Steven JM, Nicholson SC. Cerebral oxygenation during pediatric
surgery using deep hypothermic circulatory arrest. Anaesthesiology
1995;82:74–82.
12. Liem KD, Hopman JCW, Oeseburg B, et al. Cerebral oxygenation and
haemodynamics during induction of extracorporeal membrane oxygenation
as investigated by near infrared spectroscopy. Paediatrics 1995;95:555–61.
13. Al-Rawi PG, Kirkpatrick PJ. Tissue oxygen index (TOI): thresholds for
cerebral ischaemia using near infrared spectroscopy. Stroke 2006;37:
2720–5.
14. Moerman A, Wouters P. Near-infrared spectroscopy (NIRS) monitoring in
contemporary anesthesia and critical care. Acta Anaesthesiol Belg 2010;61(4):
185–94.
15. De Backer D, Ospina-Tascon G, Salgado D, et al. Monitoring the
microcirculation in the critically ill patient: current methods and future
approaches. Intensive Care Med 2010;36(11):1813–25. Epub 2010, Aug 6.
16. Rodriguez A, Lisboa T, Martín-Loeches I, et al. Mortality and regional
oxygen saturation index in septic shock patients: a pilot study. J Trauma
2011;70(5):1145–52.
17. Nanas S, Gerovasili V, Renieris P, et al. Non-invasive assessment of the
microcirculation in critically ill patients. Anaesth Intensive Care 2009;37(5):
733–9.
18. Dullenkopf A, Frey B, Baenziger O, et al. Measurement of cerebral
oxygenation state in anaesthetized children using the INVOS 5100 cerebral
oximeter. Paediatr Anaesth 2003;13:384–91.
19. McKeating EG, Monjardino JR, Signorini DF, et al. A comparison of the
INVOS 3100 and the Critikon 2020 near-infrared spectrophotometer as
monitors of cerebral oxygenation. Anaesthesia 1997;52:136–40.
20. Yoshitani K, Kawaguchi M, Tatsumi K, et al. A comparison of the INVOS
4100 and the NIRO 300 near-infrared spectrometers. Anesth Analg 2002;94:
586–90.
21. Ferrari M, Mottola L, Quaresima V. Principles, techniques and limitations of
near infrared spectroscopy. Can J Appl Physiol 2004;29(4):463–87.
22. Cope M, Delpy DT. System for long-term measurement of cerebral blood
and tissue oxygenation on newborn infants by near infra-red
transillumination. Med Biol Eng Comp 1988;26:289–94.
23. Piantadosi CA. Near infrared spectroscopy: principles and application to
non-invasive assessment of tissue oxygenation. J Crit Care 1989;4:308–18.
24. Wyatt JS, Cope M, Delpy DT, et al. Measurement of optical path length for
cerebral near-infrared spectroscopy in newborn infants. Dev Neurosci 1990;
12:140–4.
25. Dehghani H, Delpy DT. Near infrared spectroscopy of the adult head: effect
of scattering and absorbing obstructions in the CSF layer on light
distribution in the tissue. Appl Opt 2000;39:4721–9.
26. Okada E, Firbank M, Schweiger M, et al. Theoretical and experimental
investigation of near-infrared light propagation in a model of the adult head.
Appl Opt 1997;36(1):21–31.
27. Harris DN, Bailey SM. Near infrared spectroscopy in adults: does the
INVOS 3100 really measure intracerebral oxygenation? Anaesthesia 1993;48:
694–6.
28. Komiyama T, Quaresima V, Shigematsu H, et al. Comparison of two spatially
resolved near-infrared photometers in the detection of tissue oxygen
saturation: poor reliability at very low oxygen saturation. Clin Sci 2001;101:
715–8.
29. Kytta J, Ohman J, Tanskanen P, et al. Extracranial contribution to cerebral
oximetry in brain dead patients: a report of six cases. J Neurosurg
Anaesthesiol 1999;11(4):252–4.
30. Litscher G, Schwarz G. Transcranial cerebral oximetry: is it clinically useless
at this moment to interpret absolute values obtained by the INVOS 3100
cerebral oximeter? Biomed Tech (Berl) 1997;42(4):74–7.
31. Young AER, Germon TJ, Barnett NJ, et al. Behaviour of near-infrared light
in the adult human head: implications for clinical near-infrared
spectroscopy. Br J Anaesth 2000;84(1):38–42.
32. Germon TJ, Evans P, Barnett N, et al. Optode separation determines
sensitivity of near infrared spectroscopy to intra- and extracranial
oxygenation changes. J Cereb Blood Flow Metab 1996;15:617.
33. Samra SK, Stanley JC, Zelenock GB, et al. An assessment of contributions
made by extracranial tissues during cerebral oximetry. J Neurosurg
Anaesthesiol 1999;11(1):1–5.
34. Germon TJ, Evans DH, Barnett N, et al. Cerebral near infrared spectroscopy:
emitter-detector separation must be increased. Br J Anaesth 1999;82(6):
831–7.
35. Yoshitani K, Kawaguchi M, Okuno T, et al. Measurement of optical path
length using phase-resolved spectroscopy in patients undergoing
cardiopulmonary bypass. Anesth Analg 2007;104:341–6.
36. Hiraoka M, Firbank M, Essenpries M. A Monte Carlo investigation of optical
path length in inhomogeneous tissue and its application to near infrared
spectroscopy. Phys Med Biol 1993;38:1859–76.
37. Firbank M, Schweiger M, Delpy DT. Investigation of light piping through
clear regions of scattering objects. SPIE 1995;2389:167–73.
38. Okada E, Firbank M, Schweiger M, et al. A theoretical and experimental
investigation of the effect of sulci on light propagation in brain tissue. SPIE
1995;2626:2–8.
39. Delpy DT, Cope M. Quantification in tissue near-infrared spectroscopy. Phil
Trans R Soc Lond B 1997;352:649–59.
40. Fantini S, Franceschini MA, Maier JS, et al. Frequency-domain multichannel
optical detector for non-invasive tissue spectroscopy and oximetry. Opt Eng
1995;34:32–42.
41. Hazeki O, Tamura M. Quantitative analysis of haemoglobin oxygenation
state of brain in situ by near-infrared spectrophotometry. J App Physiol
1988;64:796–802.
42. Matcher SJ, Kirkpatrick PJ, Nahid K, et al. Absolute quantification methods
in tissue near infrared spectroscopy. Proc SPIE 1993;2389:486–95.
43. Suzuki S, Takasaki S, Ozaki T, et al. A tissue oxygenation monitor using NIR
spatially resolved spectroscopy. Proc SPIE 1999;3597:582–92.

44. Delpy DT, Cope M, van der Zee P, et al. Estimation of optical pathlength
through tissue from direct time of flight measurement. Phys Med Biol 1988;
33(12):1433–42.
45. Patterson MS, Chance B, Wilson BC. Time resolved reflectance and
transmittance for the noninvasive measurement of tissue optical properties.
Appl Opt 1989;28:2331–6.
46. Brady K, Joshi B, Zweifel C, et al. Real-time continuous monitoring of
cerebral blood flow autoregulation using near-infrared spectroscopy
in patients undergoing cardiopulmonary bypass. Stroke 2010;41(9):
1951–6.
47. Dani C, Pratesi S, Fontanelli G, et al. Blood transfusions increase cerebral,
splanchnic, and renal oxygenation in anemic preterm infants. Transfusion
2010;50(6):1220–6. Epub 2010, Jan 22.
Section V—Cerebral Blood Flow 335
48. Hanson SJ, Berens RJ, Havens PL, et al. Effect of volume resuscitation
on regional perfusion in dehydrated pediatric patients as measured by
two-site near-infrared spectroscopy. Pediatr Emerg Care 2009;25(3):
150–3.
49. Petrova A, Mehta R. Near-infrared spectroscopy in the detection of regional
tissue oxygenation during hypoxic events in preterm infants undergoing
critical care. Pediatr Crit Care Med 2006;7(5):449–54.
50. Lima A, van Bommel J, Jansen TC, et al. Low tissue oxygen saturation at the
end of early goal-directed therapy is associated with worse outcome in
critically ill patients. Crit Care 2009;13(Suppl 5):S13. Epub 2009, Nov 30.
A complete list of references for this chapter can be found online at
www.expertconsult.com.

References
1. Jöbsis FF. Non-invasive infrared monitoring of cerebral and myocardial
oxygen sufficiency and circulatory parameters. Science 1977;198:1264–7.
2. Toet MC, Lemmers PM. Brain monitoring in neonates. Early Hum Dev
2009;85(2):77–84. Epub 2009, Jan 17.
3. Schwarz G, Litscher G, Kleinert R, et al. Cerebral oximetry in dead subjects.
J Neurosurg Anesthesiol 1996;8(3):189–93.
4. Zweifel C, Castellani G, Czosnyka M, et al. Noninvasive monitoring of
cerebrovascular reactivity with near infrared spectroscopy in head-injured
patients. J Neurotrauma 2010;27(11):1951–8. Epub 2010, Oct 28.
5. Kim MN, Durduran T, Frangos S, et al. Noninvasive measurement of
cerebral blood flow and blood oxygenation using near-infrared and diffuse
correlation spectroscopies in critically brain-injured adults. Neurocrit Care
2010;12(2):173–80.
6. Mutoh T, Ishikawa T, Suzuki A, et al. Continuous cardiac output and
near-infrared spectroscopy monitoring to assist in management of
symptomatic cerebral vasospasm after subarachnoid hemorrhage. Neurocrit
Care 2010;13(3):331–8.
7. du Plessis AJ, Newburger J, Jonas RA, et al. Cerebral oxygenation supply
and utilisation during infact cardiac surgery. Ann Neurol 1995;37:488–97.
8. Al-Rawi PG, Smielewski P, Hobbiger H, et al. Assessment of spatially
resolved spectroscopy during cardiopulmonary bypass. J Biomed Opt
1999;4(2):208–16.
9. Kirkpatrick PJ, Smielewski P, Czosnyka M, et al. Near infrared spectroscopy
use in patients with head injury. J Neurosurg 1995;83:963–70.
10. Kirkpatrick PJ, Smielewski P, Whitfield P, et al. An observational study of
near infrared spectroscopy during carotid endarterectomy. J Neurosurg
1995;82:756–63.
11. Kurth CD, Steven JM, Nicholson SC. Cerebral oxygenation during pediatric
surgery using deep hypothermic circulatory arrest. Anaesthesiology 1995;
82:74–82.
12. Liem KD, Hopman JCW, Oeseburg B, et al. Cerebral oxygenation and
haemodynamics during induction of extracorporeal membrane
oxygenation as investigated by near infrared spectroscopy. Paediatrics
1995;95:555–61.
13. Al-Rawi PG, Kirkpatrick PJ. Tissue oxygen index (TOI): thresholds for
cerebral ischaemia using near infrared spectroscopy. Stroke 2006;37:
2720–5.
14. Moerman A, Wouters P. Near-infrared spectroscopy (NIRS) monitoring in
contemporary anesthesia and critical care. Acta Anaesthesiol Belg 2010;
61(4):185–94.
15. De Backer D, Ospina-Tascon G, Salgado D, et al. Monitoring the
microcirculation in the critically ill patient: current methods and future
approaches. Intensive Care Med 2010;36(11):1813–25. Epub 2010, Aug 6.
16. Rodriguez A, Lisboa T, Martín-Loeches I, et al. Mortality and regional
oxygen saturation index in septic shock patients: a pilot study. J Trauma
2011;70(5):1145–52.
17. Nanas S, Gerovasili V, Renieris P, et al. Non-invasive assessment of the
microcirculation in critically ill patients. Anaesth Intensive Care 2009;37(5):
733–9.
18. Dullenkopf A, Frey B, Baenziger O, et al. Measurement of cerebral
oxygenation state in anaesthetized children using the INVOS 5100 cerebral
oximeter. Paediatr Anaesth 2003;13:384–91.
19. McKeating EG, Monjardino JR, Signorini DF, et al. A comparison of the
INVOS 3100 and the Critikon 2020 near-infrared spectrophotometer as
monitors of cerebral oxygenation. Anaesthesia 1997;52:136–40.
20. Yoshitani K, Kawaguchi M, Tatsumi K, et al. A comparison of the INVOS
4100 and the NIRO 300 near-infrared spectrometers. Anesth Analg 2002;
94:586–90.
21. Ferrari M, Mottola L, Quaresima V. Principles, techniques and limitations
of near infrared spectroscopy. Can J Appl Physiol 2004;29(4):463–87.
22. Cope M, Delpy DT. System for long-term measurement of cerebral blood
and tissue oxygenation on newborn infants by near infra-red
transillumination. Med Biol Eng Comp 1988;26:289–94.
23. Piantadosi CA. Near infrared spectroscopy: principles and application to
non-invasive assessment of tissue oxygenation. J Crit Care 1989;4:308–18.
24. Wyatt JS, Cope M, Delpy DT, et al. Measurement of optical path length for
cerebral near-infrared spectroscopy in newborn infants. Dev Neurosci
1990;12:140–4.
25. Dehghani H, Delpy DT. Near infrared spectroscopy of the adult head: effect
of scattering and absorbing obstructions in the CSF layer on light
distribution in the tissue. Appl Opt 2000;39:4721–9.
26. Okada E, Firbank M, Schweiger M, et al. Theoretical and experimental
investigation of near-infrared light propagation in a model of the adult
head. Appl Opt 1997;36(1):21–31.
Section V—Cerebral Blood Flow 335.e1
27. Harris DN, Bailey SM. Near infrared spectroscopy in adults: does the
INVOS 3100 really measure intracerebral oxygenation? Anaesthesia
1993;48:694–6.
28. Komiyama T, Quaresima V, Shigematsu H, et al. Comparison of two
spatially resolved near-infrared photometers in the detection of tissue
oxygen saturation: poor reliability at very low oxygen saturation. Clin Sci
2001;101:715–8.
29. Kytta J, Ohman J, Tanskanen P, et al. Extracranial contribution to cerebral
oximetry in brain dead patients: a report of six cases. J Neurosurg
Anaesthesiol 1999;11(4):252–4.
30. Litscher G, Schwarz G. Transcranial cerebral oximetry: is it clinically useless
at this moment to interpret absolute values obtained by the INVOS 3100
cerebral oximeter? Biomed Tech (Berl) 1997;42(4):74–7.
31. Young AER, Germon TJ, Barnett NJ, et al. Behaviour of near-infrared light
in the adult human head: implications for clinical near-infrared
spectroscopy. Br J Anaesth 2000;84(1):38–42.
32. Germon TJ, Evans P, Barnett N, et al. Optode separation determines
sensitivity of near infrared spectroscopy to intra- and extracranial
oxygenation changes. J Cereb Blood Flow Metab 1996;15:617.
33. Samra SK, Stanley JC, Zelenock GB, et al. An assessment of contributions
made by extracranial tissues during cerebral oximetry. J Neurosurg
Anaesthesiol 1999;11(1):1–5.
34. Germon TJ, Evans DH, Barnett N, et al. Cerebral near infrared
spectroscopy: emitter-detector separation must be increased. Br J Anaesth
1999;82(6):831–7.
35. Yoshitani K, Kawaguchi M, Okuno T, et al. Measurement of optical path
length using phase-resolved spectroscopy in patients undergoing
cardiopulmonary bypass. Anesth Analg 2007;104:341–6.
36. Hiraoka M, Firbank M, Essenpries M. A Monte Carlo investigation of
optical path length in inhomogeneous tissue and its application to near
infrared spectroscopy. Phys Med Biol 1993;38:1859–76.
37. Firbank M, Schweiger M, Delpy DT. Investigation of light piping through
clear regions of scattering objects. SPIE 1995;2389:167–73.
38. Okada E, Firbank M, Schweiger M, et al. A theoretical and experimental
investigation of the effect of sulci on light propagation in brain tissue. SPIE
1995;2626:2–8.
39. Delpy DT, Cope M. Quantification in tissue near-infrared spectroscopy.
Phil Trans R Soc Lond B 1997;352:649–59.
40. Fantini S, Franceschini MA, Maier JS, et al. Frequency-domain
multichannel optical detector for non-invasive tissue spectroscopy and
oximetry. Opt Eng 1995;34:32–42.
41. Hazeki O, Tamura M. Quantitative analysis of haemoglobin oxygenation
state of brain in situ by near-infrared spectrophotometry. J App Physiol
1988;64:796–802.
42. Matcher SJ, Kirkpatrick PJ, Nahid K, et al. Absolute quantification methods
in tissue near infrared spectroscopy. Proc SPIE 1993;2389:486–95.
43. Suzuki S, Takasaki S, Ozaki T, et al. A tissue oxygenation monitor using
NIR spatially resolved spectroscopy. Proc SPIE 1999;3597:582–92.
44. Delpy DT, Cope M, van der Zee P, et al. Estimation of optical pathlength
through tissue from direct time of flight measurement. Phys Med Biol
1988;33(12):1433–42.
45. Patterson MS, Chance B, Wilson BC. Time resolved reflectance and
transmittance for the noninvasive measurement of tissue optical properties.
Appl Opt 1989;28:2331–6.
46. Brady K, Joshi B, Zweifel C, et al. Real-time continuous monitoring of
cerebral blood flow autoregulation using near-infrared spectroscopy in
patients undergoing cardiopulmonary bypass. Stroke 2010;41(9):
1951–6.
47. Dani C, Pratesi S, Fontanelli G, et al. Blood transfusions increase cerebral,
splanchnic, and renal oxygenation in anemic preterm infants. Transfusion
2010;50(6):1220–6. Epub 2010, Jan 22.
48. Hanson SJ, Berens RJ, Havens PL, et al. Effect of volume resuscitation on
regional perfusion in dehydrated pediatric patients as measured by two-site
near-infrared spectroscopy. Pediatr Emerg Care 2009;25(3):150–3.
49. Petrova A, Mehta R. Near-infrared spectroscopy in the detection of regional
tissue oxygenation during hypoxic events in preterm infants undergoing
critical care. Pediatr Crit Care Med 2006;7(5):449–54.
50. Lima A, van Bommel J, Jansen TC, et al. Low tissue oxygen saturation at
the end of early goal-directed therapy is associated with worse outcome in
critically ill patients. Crit Care 2009;13(Suppl 5):S13. Epub 2009, Nov 30.
51. Kirkpatrick PJ, Smielewski P, Lam JMK, et al. Use of near infrared
spectroscopy for the clinical monitoring of adult brain. J Biomed Opt
1996;1(4):363–72.
52. McGown AD, Makker H, Elwell C, et al. Measurement of changes in
cytochrome oxidase redox state during obstructive sleep apnea using
near-infrared spectroscopy. Sleep 2003;26(6):710–6.
335.e2 Section V—Cerebral Blood Flow
53. Duncan A, Meek JH, Clemence M, et al. Optical pathlength measurement
on adult head, calf, forearm and the head of the newborn infant using
phase resolved optical spectroscopy. Physiol Med Biol 1995;40:295–304.
54. Nollert G, Jonas RA, Reichart B. Optimising cerebral oxygenation during
cardiac surgery: a review of experimental and clinical investigations with
near infrared spectrophotometry. Thorac Cardiovasc Surg 2000;48:247–53.
55. Beese U, Langer H, Lang W, et al. Comparison of near-infrared
spectroscopy and somatosensory evoked potentials for the detection of
cerebral ischaemia during carotid endarterectomy. Stroke 1998;29:2032–7.
56. Germon TJ, Kane NM, Manara AR, et al. Near-infrared spectroscopy in
adults: effects of extracranial ischaemia and intracranial hypoxia on
estimation of cerebral oxygenation. Br J Anaesth 1994;73:503–6.
57. Gopinath SP, Robertson CS, Contant CF, et al. Early detection of delayed
traumatic intracranial hematomas using near-infrared spectroscopy.
J Neurosurg 2001;83:438–44.
58. Yao F-S, Tseng C-C, Braverman JM, et al. Cerebral oxygen desaturation is
associated with prolonged lengths of stay in the intensive care unit and
hospital. Anesthesiology 1999;91:123.
59. Sokol DK, Markand ON, Daly EC, et al. Near infrared spectroscopy (NIRS)
distinguishes seizure types. Seizure 2000;9(5):323–7.
60. Williams IM, Mead G, Picton AJ, et al. The influence of contralateral
carotid stenosis and occlusion on cerebral oxygen saturation during carotid
artery surgery. Eur J Vasc Endovasc Surg 1995;10:198–206.
61. Samra SK, Dorje P, Zelenock GB, et al. Cerebral oximetry in patients
undergoing carotid endarterectomy under regional anaesthesia. Stroke
1996;27:49–55.
62. Kirkpatrick PJ, Lam JMK, Al-Rawi PG, et al. Defining thresholds for critical
ischaemia by using near-infrared spectroscopy in the adult brain.
J Neurosurg 1998;89(3):389–94.
63. Lam JMK, Smielewski P, Al-Rawi P, et al. Internal and external carotid
contributions to near-infrared spectroscopy during carotid endarterectomy.
Stroke 1997;28(5):906–11.
64. Takahashi T, Takikawa Y, Kawagoe R, et al. Influence of skin blood flow on
near-infrared spectroscopy signals measured on the forehead during a
verbal fluency task. Neuroimage 2011;May 10. Epub ahead of print.
65. Muellner T, Schramm W, Kwasny O, et al. Patients with increased
intracranial pressure cannot be monitored using near infrared spectroscopy.
Br J Neurosurg 1998;12(2):136–9.
66. Smielewski P, Czosnyka M, Steiner L, et al. ICM+: software for on-line
analysis of bedside monitoring data after severe head trauma. Acta
Neurochir Suppl 2005;95:43–9.
67. Al-Rawi PG, Smielewski P, Kirkpatrick PJ. Evaluation of a near infrared
spectrometer (NIRO 300) for the detection of intracranial oxygenation
changes in the adult head. Stroke 2001;32(11):2492–500.
68. Gallagher CN, Hutchinson PJ, Pickard JD. Neuroimaging in trauma. Curr
Opin Neurol 2007;20(4):403–9.
69. Obrist WD, Thompson HK, Wang HS, et al. Regional cerebral blood flow
estimated by 133-xenon inhalation. Stroke 1975;6:245–56.
70. Gideon P, Sperling B, Arlien-Soborg P, et al. Long-term follow-up of
cerebral infarction patients with proton magnetic resonance spectroscopy.
Stroke 1994;25:967–73.
71. Saunders DE, Howe FA, van den Boogaart A, et al. Continuing ischaemic
damage after acute middle cerebral artery infarction in humans
demonstrated by short-echo proton spectroscopy. Stroke 1995;26:
1007–13.
72. Edwards AD, Wyatt JS, Richardson CE, et al. Cotside measurements of
cerebral blood flow in ill newborn infants by near infrared spectroscopy.
Lancet 1988;ii:770–1.
73. Skov L, Pryds O, Griesen G. Estimating cerebral blood flow in newborn
infants: comparison of near infrared spectroscopy and 133Xe clearance.
Pediatr Res 1991;30:570–3.
74. Wyatt JS, Cope M, Delpy DT, et al. Quantification of cerebral blood volume
in newborn infants by near infrared spectroscopy. J Appl Physiol
1990;68:1086–91.
75. Elwell CE, Cope M, Edwards AD, et al. Quantification of adult cerebral
haemodynamics by near-infrared spectroscopy. App Physiol
1994;77:2753–60.
76. Bucher HU, Edwards AD, Lipp AE, et al. Comparison between near infrared
spectroscopy and 133xenon clearance for estimation of cerebral blood flow
in critically ill preterm infants. Pediatr Res 1992;33:56–60.
77. Wolf M, von Siebenthal K, Keel M, et al. Comparison of three methods to
measure absolute cerebral hemoglobin concentration in neonates by
near-infrared spectrophotometry. J Biomed Opt 2002;7:221–7.
78. Landsman MLJ, Kwant G, Mook GA, et al. Light-absorbing properties,
stability, and spectral stabilization of indocyanine green. J Appl Physiol
1976;40:575–83.
79. Hongo K, Kobayashi S, Okudera H, et al. Noninvasive cerebral optical
spectroscopy: depth resolved measurements of creebral haemodynamics
using indocyanine green. Neurol Res 1995;17:89–93.
80. Roberts I, Fallon P, Kirkham FJ, et al. Estimation of cerebral blood flow
with near infrared spectroscopy and indocyanine green. Lancet 1993;342:
1425.
81. Shinohara H, Tanaka A, Kital T, et al. Direct measurement of hepatic
indocyanine green clearance with near infrared spectroscopy: separate
evaluation of uptake and removal. Hepatology 1996;23:137–44.
82. Hopton P, Walsh TS, Lee A. Measurement of cerebral blood volume using
near-infrared spectroscopy and indocyanine green elimination. J Appl
Physiol 1999;87(5):1981–7.
83. Keller E, Nadler A, Alkhadi H, et al. Noninvasive measurement of regional
cerebral blood flow and regional cerebral blood volume by near infrared
spectroscopy and indocyanine green dye dilution. NeuroImage 2003;22(3):
828−39.
84. Springett R, Sakata Y, Delpy DT. Precise measurement of cerebral blood
flow in newborn piglets from the bolus passage of indocyanine green. Phys
Med Biol 2001;46:2209–25.
85. Gora F, Shinde S, Elwell CE, et al. Noninvasive measurement of cerebral
blood flow in adults using near-infrared spectroscopy and indocyanine
green: a pilot study. J Neurosurg Anaesthesiol 2002;14(3):218–22.
86. Boushel R, Langberg H, Olesen J, et al. Regional blood flow during exercise
in humans measured by near infrared spectroscopy and indocyanine green.
J Appl Physiol 2000;89:1868–78.
87. Patel J, Marks K, Roberts I, et al. Measurement of cerebral blood flow in
newborn infants using near infrared spectroscopy with indocyanine green.
Pediatr Res 1998;43:34–9.
88. Kuebler WM, Sckell A, Habler O, et al. Noninvasive measurement of
regional cerebral blood flow by near-infrared spectroscopy and indocyanine
green. J Cereb Blood Flow Metab 1998;18(4):445–56.
89. Colacino JM, Grubb B, Jöbsis FF. Infra-red technique for cerebral blood
flow: comparison with 133xenon clearance. Neurol Res 1981;3:17–31.
90. Roberts IG, Fallon P, Kirkham FJ, et al. Measurement of cerebral blood flow
during cardiopulmonary bypass with near infrared spectroscopy. J Thorac
Cardiovasc Surg 1998;115:94–102.
91. Keller E, Nadler A, Alkadhi H, et al. Noninvasive measurement of regional
cerebral blood flow and regional cerebral blood volume by near-infrared
spectroscopy and indocyanine green dye dilution. Neuroimage 2003;22(3):
828–39.
92. Rothoerl RD, Schebesch KM, Faltermeier R, et al. Lack of correlation
between Xenon133 and near infrared spectroscopy/indocyanine green rCBF
measurements. Neurol Res 2003;25(5):528–32.
93. Al-Rawi PG, Leung TS, Smielewski P, et al. Quantification of cerebral blood
flow using near infrared spectroscopy and indocyanine green in patients
with severe subarachnoid haemorrhage. Br J Neurosurg 2006;20(5):333–58.
94. Kohl-Bareis M, Obrig H, Steinbrink J, et al. Noninvasive monitoring of
cerebral blood flow by a dye bolus method: separation of brain from skin
and skull signals. J Biomed Opt 2002;7(3):464–70.
95. van Houten JP, Benaron DA, Spilman S, et al. Imaging brain injury using
time-resolved near infrared light scanning. Pediat Res 1996;39:470–6.
96. Wolf M, Ferrari M, Quaresima V. Progress of near-infrared spectroscopy
and topography for brain and muscle clinical applications. J Biomed Opt
2007;12(6):1–14.
97. Hoshi Y. Functional near-infrared spectroscopy: current status and future
prospects. J Biomed Opt 2007;12(6):062106-1–062106-9.
98. Obrig H, Villringer A. Beyond the visible; Imaging the human brain with
light. J Cereb Blood Flow Metab 2003;23(1):1–18.
99. Strangman G, Boas DA, Sutton JP. Non-invasive neuroimaging using
near-infrared light. Biol Psychiatry 2002;52:679–93.
100. Austin T, Gibson AP, Branco G, et al. Three dimensional optical imaging of
blood volume and oxygenation in the neonatal brain. NeuroImage
2006;31:1426–33.
101. Arridge SR, Schweiger M. Image reconstruction in optical tomography. Phil
Trans R Soc Lond B 1997;352:717–26.
102. Hebden JC. Advances in optical imaging of the newborn infant brain.
Psychophysiology 2003;40(4):501–10.
103. Benaron DA, Hintz SR, Villringer A, et al. Noninvasive functional
imaging of human brain using light. J Cereb Blood Flow Metab 2000;20:
469–77.
104. Danen RM, Wang Y, Li XD, et al. Regional imager for low resolution
functional imaging of the brain with diffusing near-infrared light.
Photochem Photobiol 1985;67:33–40.
105. Franceschini MA, Toronov V, Filiaci ME, et al. On line optical imaging of
the human brain with 160ms temporal resolution. Opt Express
2000;6:49–57.

106. Hebden JC, Gibson A, Austin T, et al. Imaging changes in blood volume
and oxygenation in the newborn infant brain using three-dimensional
optical tomography. Phys Med Biol 2004;49:1117–30.
107. Hintz SR, Cheong WF, van Houten JP, et al. Bedside imaging of intracranial
hemorrhage in the neonate using light: comparison with ultrasound,
computed tomography, and magnetic resonance imaging. Pediat Res
1999;45(1):54–9.
108. Kusaka T, Kawada K, Okubo K, et al. Noninvasive optical imaging in the
visual cortex in young infants. Hum Brain Mapp 2004;22:122–32.
109. Schroeter ML, Bucheler MM, Muller K, et al. Towards a standard analysis
for functional near-infrared imaging. Neuroimage 2004;21:283–90.
110. Chance B, Zhuang Z, UnAh C, et al. Cognition-activated low-frequency
modulation of light absorption in human brain. Proc Natl Acad Sci USA
1993;90:3370–4.
111. Villringer A, Planck J, Hock C, et al. Near infrared spectroscopy (NIRS):
a new tool to study hemodynamic changes during activation of brain
function in human adults. Neurosci Lett 1993;154:101–4.
Section V—Cerebral Blood Flow 335.e3
112. Hoshi Y, Tamura M. Dynamic multichannel near-infrared optical imaging
of human brain activity. J App Physiol 1993;75(4):1842–6.
113. Kato T, Kamei A, Takashima S, et al. Human visual cortical function during
photic stimulation monitoring by means of near-infrared spectroscopy.
J Cereb Blood Flow Metab 1993;13(3):516–20.
114. Mehagnoul-Schipper DJ, van der Kallen BF, Colier WN, et al.
Simultaneous measurements of cerebral oxygenation changes during brain
activation by near-infrared spectroscopy and functional magnetic resonance
imaging in healthy young and elderly subjects. Hum Brain Mapp 2002;16:
14–23.
115. Strangman G, Culver JP, Thompson JH, et al. A quantitative comparison of
simultaneous BOLD fMRI and NIRS recordings during functional brain
activation. Neuroimage 2002;17:719–31.
116. Keller E, Froehlich J, Muroi C, et al. Neuromonitoring in intensive care: a
new brain tissue probe for combined monitoring of intracranial pressure
(ICP) cerebral blood flow (CBF) and oxygenation. Acta Neurochir Suppl
2011;110(Pt 2):217–20.
VI
Chapter
34  
Intracranial Pressure
Randall M. Chesnut
This chapter is dedicated to professor Douglas Miller for his
personal influence on my understanding of brain injury and
his genius in originating the concept of targeted therapy, the
implications of which most of us are only now coming to
understand.
Introduction
An intracranial pressure (ICP) monitor is considered the gold
standard intracranial monitor. In particular, the most recent
guidelines for the management of severe traumatic brain
injury (TBI) contain level 2 recommendations (moderate
degree of clinical certainty) for the use of an ICP monitor in
comatose TBI patients who have an abnormal admission com-
puted tomography (CT) scan.1 Whether to use an ICP monitor
in comatose patients who have a normal CT scan after TBI or
in other disorders is less certain (level 3; clinical certainty not
established), although neurosurgeons often will consider an
ICP monitor in conditions as diverse as subarachnoid hemor-
rhage (SAH), meningitis, and liver failure.2,3 In addition, CT
scan findings may not always predict whether ICP is increased.4
Consequently, there is variation across centers on how fre-
quently an ICP monitor is used. There are very strong physi-
ologic arguments for ICP-based management for many acute
neurologic conditions that require admission to a neurocriti-
cal care unit (NCCU) and a well-described association between
elevated ICP and mortality. Having the ability to continuously
monitor ICP avoids “empirical” ICP treatment or blind pro-
phylactic treatment. This is important because many treat-
ments for ICP, although effective in lowering ICP, may have
other deleterious side effects.
The Monro-Kellie Doctrine
The basic concept of ICP is couched in the Monro-Kellie
doctrine.5-7 This doctrine is founded on the fixed volume
of the skull, such that the interior pressure is a function
of the “volumes” of the individual physiologic intracranial
compartments—brain, blood, and cerebrospinal fluid (CSF)—
plus mass lesions, when present (Fig. 34.1). Following trauma,
brain volume increases due to edema, either vasogenic or
cytotoxic, the latter being predominant.8 The vascular com-
partment is primarily composed of venous blood, the volume
of which can increase due to outflow obstruction. Such
obstruction can be extracranial (e.g., increased intrathoracic
pressure or compression of the jugular outflow system) or
internal, from kinking of vessels draining into the sagittal
338
sinus (Starling resistor),9 pressure-related venous compres-
sion, or thrombosis. The volume of the combined arteriove-
nous system can increase actively, due to increased flow
recruitment from elevated metabolic activity, or passively,
such as from hypercapnia or blood pressure elevation in the
face of abnormal pressure autoregulation.10 CSF production is
fixed, so volume increases within this compartment generally
increase as a result of outflow obstruction or abnormal
resorption.
Increases in the volume of one compartment (or the addi-
tion of a compartment such as a hematoma) may be buffered
by compensatory decreases in the volume of other compart-
ments. When the increase is sufficiently slow, these compensa-
tory changes can maintain a stable ICP up to the point of their
exhaustion (Fig. 34.2). At that point, the ICP rises rapidly with
any further increase in compartmental volume. If the rate of
initial volume increase is rapid, the compensatory mecha-
nisms are likely to fail earlier.
One consequence of these compensatory changes is that the
brain compliance changes as its volume buffering capacity is
exhausted. This means that for a given volume challenge, the
resultant ICP elevation will increase as the compliance
decreases. Therefore even when volume buffering keeps the
ICP relatively stable, knowing the compliance of the brain is
a very useful index of the status of that buffering system (vide
infra).
The Monro-Kellie Doctrine and
Intracranial Pressure Management
The Monro-Kellie doctrine helps explain ICP and is consis-
tent with the framework that supports current therapeutic
interventions, all of which aim to lower or prevent an
increase in the volume of one or more of the intracranial
compartments.
Mass Lesion
The simplest compartmental correction is for mass lesions.
Removal of a mass (e.g., a hematoma) is the oldest approach
to control intracranial hypertension and acts directly to
improve the intracranial volume balance. This is one of the
few areas in TBI or other acute neurologic disorders where a
pathophysiologic entity can be dealt with directly and defini-
tively and reinforces the value of a CT scan to avoid medical
treatment of a surgical lesion.
© Copyright 2013 Elsevier Inc. All rights reserved.
 Section VI—Intracranial Monitoring 339

MONRO KELLIE DOCTRINE - NORMAL STATE MONRO KELLIE DOCTRINE - COMPENSATED STATE

Pressure
Volume

Blood Blood
Blood compliance Blood compliance
CSF CSF
CSF compliance CSF compliance
Mass Mass
Brain Brain

A
Fig. 34.1  The Monro-Kellie doctrine. The fixed volume of the skull
contains brain, blood, and cerebrospinal fluid (CSF) compartments, plus
mass lesions when present. The combined volumes of these compart­
ments determine the intracranial pressure (ICP). The blood and CSF
MONRO KELLIE DOCTRINE - STRESSED COMPENSATION
compartments can shift some of their volume outside of the skull 
through natural means; this allows some degree of compensation for
increases in other compartments.

Pressure
Volume

Blood
The Cerebrospinal Fluid Compartment Blood compliance
CSF
The natural intracranial CSF buffering system attempts to CSF compliance
shunt a larger volume into the spinal subarachnoid space (i.e., Mass
it is subject to outflow tract patency). If interventricular pas- Brain
sages or egress from the fourth ventricle is compromised (e.g.,
a posterior fossa hematoma), the system will fail. An external
ventricular drain can be inserted in such instances, or when
caregivers want to facilitate CSF drainage by controlling it.
This can be useful if there is a sufficient CSF volume and if
the ventricular system does not collapse from drainage. An B
open ventriculostomy, however, renders ICP measurement
unreliable, so these drains should be clamped before recording
ICP.
MONRO KELLIE DOCTRINE - EXAUSTED COMPENSATION
Cerebral Blood Volume
Management of the cerebral blood volume (CBV) can address
Pressure

the venous or arterial compartments. Optimized drainage of Volume

intracranial venous blood by (1) elevating the head of the bed, Blood
(2) avoiding compression of the venous blood in the neck (e.g., Blood compliance
CSF
by a tight cervical orthosis or a tracheostomy tie), or (3) reduc- CSF compliance
ing intrathoracic pressures through careful ventilator manage- Mass
ment can help prevent elevation of the venous compartment Brain
or interference with its volume buffering capacity. CBV can
be decreased through vasoconstriction. Commonly this is
achieved through hyperventilation (HV) to induce hypocap-
nea. Because HV influences the resistance vessels that are on
the inflow side, it also decreases cerebral blood flow (CBF) and
so may cause ischemia.11,12 Prophylactic HV therefore has a C
limited if any role to play in ICP management because of this
Fig. 34.2  Volume compensation. An increase in the volume of one or
potential toxicity.12 If used, for example, when hyperemia con-
more intracranial compartments or the addition of a mass lesion initially
tributes to increased ICP, HV should be used carefully and an is compensated by shifting cerebrospinal fluid (CSF) into the spinal
additional monitor to look for ischemia should be considered subarachnoid space and venous blood out of the intracranial space (A).
if it is used aggressively. HV-induced vasoconstriction is a There is little effect on intracranial pressure (ICP). As volumes increase,
relatively transient effect, and is thought to last less than the ability to compensate decreases and ICP begins to rise (B). When
24 hours, with a risk of rebound vasodilation on discontinu- compensation is exhausted, any further volume challenge results in large
ation.13 Normobaric hyperoxia or hyperbaric oxygen therapy increases in ICP (C).
340 Section VI—Intracranial Monitoring
also may induce vasoconstriction and so alter ICP14-16 with
fewer deleterious effects on cerebral metabolism than HV.
However, the role of hyperbaric oxygen is still to be elucidated.
Finally, mannitol may induce transient vasoconstriction
through altered blood viscosity, but the effects on CBF and
volume and hence on ICP depend on autoregulation.17,18
The Brain
The natural volume buffering capacity of brain tissue is
limited and has a very long time constant. In general, the brain
volume increases after injury, due primarily to edema. Both
cytotoxic and vasogenic mechanisms occur in post-traumatic
edema. However, several lines of evidence support a cellular
origin that involves aquaporin-mediated transmembrane
water passage.8,19,20 The role of vasogenic mechanisms remains
unclear. This is important because current osmotic agents
(mannitol, hypertonic saline) work through the extracellular
compartment. The only medical treatment to decrease extra-
cellular edema in the brain compartment is hyperosmolar
therapy (osmotherapy). Alternatively, increased brain volume
can be reduced through the surgical removal of tissue, either
contused or pulped parenchyma or normal, “silent” tissue.21
The surgical removal of contusions remains controversial and
has largely been abandoned.
Decompressive Craniectomy
The final Monro-Kellie–related approach is to invalidate the
premise that underlies the doctrine through decompressive
craniectomy (DC); this renders the intracranial space no
longer a constant. A generous skull opening and a large durot-
omy or nonconstraining duraplasty compensates for increased
volumes of one or more intracranial compartments by increas-
ing the volume of containment and adding a compliance
buffer (because the bone-free area is not rigid). DC effectively
reduces increased ICP in conditions such as stroke, TBI, and
SAH.22-24 However, the exact timing of DC and its effect on
outcome particularly in TBI remains debated.25-28
Implications of Monro-Kellie–Driven
Therapy
The Monro-Kellie doctrine is a useful framework to organize
and understand medical and surgical approaches to the
injured brain. However, it is focused on the clinical manage-
ment of ICP, and the invoked therapeutic modalities are
phenomenologic and based on their perceived ability to lower
ICP. None of these treatments, however, specifically addresses
the pathophysiology of TBI and other acute brain injuries.
Osmotherapy probably works to decrease the water volume of
the least expanded compartment. Reduction of brain blood
volume is phenomenologic because primary venous or arte-
rial hypervolemia usually is not an underlying pathologic
event in adults, and induced vasoconstriction may decrease
oxygen delivery and be harmful. CSF drainage can address a
specific pathophysiologic abnormality if outflow obstruction
occurs but in the absence of hydrocephalus remains phenom-
enologic. Management such as sedation, analgesia, neuromus-
cular blockade, and second-tier interventions such a high-dose
barbiturates and DC also are focused on ICP and not the
underlying pathology. Therapeutic hypothermia, when deliv-
ered early and not as a treatment for intracranial hyperten-
sion, was an attempt to address some of the pathophysiologic
abnormalities initiated at the time of trauma, but unfortu-
nately this treatment does not improve outcome from adult
or pediatric TBI.29,30 By contrast, induced hypothermia is
useful in cardiac arrest31; however, in this condition it is not
used as specific ICP treatment.
The importance of this skepticism is to remind caregivers
that because treatment modalities are not targeted at specific
molecular or biochemical processes that must be reversed,
they need to avoid harming the patient in the choice of or
persistence with individual treatments. Obviously, ICP and
cerebral perfusion can be manipulated in many ways. When a
given approach appears ineffective or begins to exert systemic
toxicity, the therapeutic approach or tactics should immedi-
ately be changed. This presumes that an ICP monitor is in
place; in the modern NCCU ICP treatment should be guided
by such technology and not given empirically. There is no one
best way to treat increased ICP and ideally, a pragmatic
approach, guided by multimodality monitoring and coupled
with a view toward balanced critical care of the entire patient,
must form the guiding principles.
The Role of ICP Monitoring and
Treatment in TBI Management
The use of an ICP monitor and ICP treatment is best studied
in TBI and the techniques and expectations used to treat ele-
vated ICP in other conditions such as subarachnoid hemor-
rhage, intracerebral hemorrhage, ischemic stroke, encephalitis,
and metabolic disorders among others are based in large part
on the experience from TBI.32,33 The focus of this chapter
therefore is on ICP and TBI. ICP data can be used to predict
outcome and evolution of intracranial pathology, calculate
and manage cerebral perfusion pressure (CPP), direct man-
agement strategies, and limit the use of potentially deleterious
therapies. An ICP monitor is considered be a conditio sine qua
non of severe TBI care. Controversy about ICP monitoring,
however, is reflected in the variability of its clinical use.34-38 In
addition are various philosophies about how to integrate ICP
data into treatment—ICP-based therapy, CPP-based therapy,
Lund therapy, and optimized HV among others.39-41 Insertion
of an ICP monitor is the first step from “observational,” “clini-
cal,” or “empirical” management of severe TBI (or pathologies
in which increased ICP occurs) to monitor-based, physiologic
care. Despite experience dating back to the 1960s, considerable
controversy surrounds ICP-monitor–based care in TBI.
Is High Intracranial Pressure
Associated with Poor Outcome?
Since the seminal work of Lundberg and Guillaume and Janny
in the 1950s and 1960s, a large body of observational data
demonstrates that ICP elevation is associated with poor
outcome and that ICP cannot be estimated accurately or reli-
ably on clinical grounds.39,42-50 The association between intra-
cranial hypertension and poor outcome appears to be
independent from other confounding variables45,46 and pro-
portional to the amount of time that the ICP is greater than

20 to 25 mm Hg.45,51 In SAH, elevated ICP is associated with
poor outcome but on multivariate analysis is not an indepen-
dent factor.32 Outcome also may be poor because treatment
for high ICP can be harmful in some circumstances. For
example, mannitol administration can produce renal damage
or hypotension52 and HV may induce cerebral ischemia,12
whereas barbiturates and hypothermia both increase the risk
of infection and hypotension.29,30,53-55 Therapies to augment
CPP are associated with increased systemic complications,56
and even the most basic treatments, sedation and mechanical
ventilatory control, may increase ICU length of stay and expo-
sure to iatrogenic injuries such as ventilator-associated pneu-
monia or skin breakdown.57,58 In particular when the proper
ICP threshold is unclear (vide infra) or there is mild ICP
elevation without a perfusion deficit, a patient may be at more
risk of a treatment complication than of harm from ICP. In
these circumstances other monitors or information may help
guide ICP management.
Does Treating Intracranial
Hypertension Improve Outcome?
Whether treatment of intracranial hypertension is associated
with improved outcome is unclear. Certainly in individual
patients an effect can be seen, but there is no Class I evidence
that demonstrates ICP management is associated with better
outcome.59 Such a trial may be difficult to perform because
ICP monitoring and management have become standard of
practice. Early retrospective observational data not adjusted
for other variables suggest that ICP control may be beneficial.
For example, Marshall60 observed better outcome in patients
whose ICP was less than 15 mm Hg than those whose ICP was
greater than 40 mm Hg for more than 15 minutes. Saul and
Ducker61 also observed that patients with severe TBI whose
ICP was treated when it was greater than 15 mm Hg did better
than those treated when ICP was greater than 25 mm Hg.
Eisenberg et al.53 conducted a randomized clinical trial (RCT)
that examined high-dose barbiturate therapy for refractory
intracranial hypertension in severe TBI; ICP control rather
than clinical outcome was the primary objective. An associa-
tion between ICP control and improved outcome for all
patient groups (barbiturate group, control group, and control
group that crossed over to barbiturate treatment) was observed.
Whether these results apply to patients with mild elevated ICP
or ICP that is responsive to management is uncertain. In SAH,
Heuer et al.32 in a retrospective analysis of 433 patients
observed that among 21 patients whose raised ICP did not
respond to mannitol, all had a poor outcome. By contrast,
among 145 patients whose elevated ICP responded to man-
nitol, 66.9% had a favorable outcome. Most recently Stein
et al.62 performed an extensive literature review of TBI trials
and case series reported after 1970 and divided patients into
those with and without ICP monitoring and intensive therapy.
Meta-analysis showed that mortality was 12% lower and favor-
able outcome 6% greater in the aggressively treated group.
There also are several reports that use of an ICP monitor is
not associated with improved outcome after TBI or that “ICP
monitoring in accordance with the guidelines for the manage-
ment of severe traumatic brain injury63 is associated with
worsening of survival in TBI patients.”34,35 These studies have
several flaws that limit meaningful conclusions but they
Section VI—Intracranial Monitoring 341
emphasize two important issues about ICP monitors: (1) con-
tinued equipoise about the role of ICP monitors in TBI man-
agement, and (2) ICP monitoring exists only in the context of
an associated treatment philosophy, which confounds the
analyses of the value of ICP monitoring versus the efficacy of
a specific treatment strategy. In the two studies that reported
no benefit to an ICP monitor,34,35 all the patients were treated
for suspected intracranial hypertension, whether monitored
or not. Because insertion of an intraparenchymal ICP monitor
is of very low risk, it is unlikely that display of the ICP will
adversely influence outcome or improve recovery per se.
Implicit in any discussion about the role of an ICP monitor
in TBI management is that the interventions generated by the
ICP data improve outcome. Embedded within such a monitor-
driven therapy question are several issues, including (1) the
validity of the treatment threshold and whether that threshold
applies to all patients, (2) whether a given treatment regimen
is appropriate in all monitored patients, (3) the choice of
therapy, (4) the efficacy and risk-to-benefit ratio of various
treatments alone or in combination, and (5) an understanding
of the interaction between therapy and management interven-
tions made in a given patient for other reasons. Consequently,
although use of an ICP monitor may make sound physiologic
sense, the ability to demonstrate that its use makes a difference
to outcome may be confounded by several reasons: (1) wrong
treatment threshold, (2) wrong individual therapeutic agent,
(3) wrong therapeutic approach, (4) toxic effects of ICP treat-
ment, (5) misinterpretation of ICP data, (6) wrong patient
population, (7) lack of utility of ICP monitoring, or (8) a
combination of all. An example of how such confounding may
affect whether a monitor is useful are the RCTs that demon-
strated that routine use of right heart catheterization in the
ICU is not associated with better outcome64-66 (i.e., use of a
monitor does always mean better outcome unless effective
management strategies exist).
An Intracranial Pressure Threshold
Among several areas on how best to interpret and treat ICP is
the threshold value used for treatment and in particular
whether a single threshold applies to all patients or in one
patient at all times. Too high a value may allow unrecognized
neural injury, whereas too low a value may result in overtreat-
ment and iatrogenic complications. The most frequently used
threshold of 20 mm Hg was adopted from the accepted upper
limit of normal for lumbar CSF pressure measurements and
from early work by Lundberg.43 Schreiber et al. used multiple
regression modeling and found that ICP greater than or equal
to 15 mm Hg was associated independently with mortality
after severe TBI. However, they did not analyze treatment
thresholds per se.67 Marmarou et al.45 analyzed 428 monitored
patients from the Traumatic Coma Data Bank and found that
the proportion of total monitoring time that the ICP was
greater than 20 mm Hg was an independent predictor of
6-month outcome. However, ICP treatment thresholds of
20 or 25 mm Hg were used, which confounds separation
of the effects of ICP of greater than or equal to 20 mm Hg
from treatment toxicity. Nevertheless, an ICP threshold of
more than 20 mm Hg is now central to ICP management
based largely on post hoc analysis of studies in which 20 mm
Hg was the treatment trigger (i.e., there are no controlled
studies on of the “best” ICP treatment threshold). This is
342 Section VI—Intracranial Monitoring
important for several reasons. First, observational data for
which other monitors are used—jugular bulb catheters, brain
oxygen, or microdialysis—suggest that brain metabolism may
be abnormal even when ICP and CPP are normal.68,69 Second,
an ICP threshold of 20 mm Hg was established during a time
period when systemic hypertension was thought to be a risk
factor for intracranial hypertension, patients were routinely
kept “dry,” and CPP was not a monitored variable. Conse-
quently, many TBI patients in the intensive care unit (ICU)
had relatively low blood pressures and the only real limit was
to avoid systolic pressures less than 90 mm Hg. This represents
a mean arterial pressure of about 70 mm Hg. Because normal
autoregulation becomes impaired at approximately 50 mm
Hg, an ICP value of 20 mm Hg therefore may represent bor-
derline ischemia; that is, the threshold value may be a
by-product of the treatment regimens of that period. At the
time of this publication, the management focus is on not only
ICP but also CPP and in some centers brain oxygen or other
parameters. This combined with sedatives that lower the cere-
bral metabolic rate mean that the flow-to-perfusion conse-
quences of an ICP that is 20 mm Hg or just elevated may be
different today compared with the time period when this
threshold was decided upon.
Perhaps more important than a single ICP threshold may
be a trend over time, ICP waveform analysis, or whether the
ICP value is associated with other detrimental effects. The
concept that the ICP threshold may vary among patients and
within patients over time (i.e., targeted) is not new. In the
1970s and 1980s, professor Douglas Miller often used an ICP
threshold of 25 to 30 mm Hg in TBI patients with acceptable
indicators of cerebral oxygenation and compliance. This
concept of permissive intracranial hypertension is practiced
at several institutions including the author’s own but is best
guided by information from multimodality monitoring. For
example, a brain oxygen monitor may complement the infor-
mation from an ICP monitor. Some but not all observational
studies suggest that when both ICP and brain oxygen are
treated, the outcome may be better than if just ICP is treated
after TBI.70 This effect may occur in part because the deleteri-
ous consequences of ICP overtreatment may be avoided. Con-
sistent with this Chambers et al. calculated receiver-operating
characteristic curves for ICP and CPP in 227 patients71 and
found that the sensitivity of ICP for outcome increased at an
ICP of 10 mm Hg, but by 30 mm Hg it had only reached 61%,
with a cutoff value of 35 mm Hg that varied with the CT
classification.
The Value of Intracranial
Pressure Monitoring
Although understanding of ICP is incomplete and treatment
approaches still need to be refined, there are many other poten-
tial benefits from ICP monitoring. First, without an ICP
monitor, CPP is not known. Even transient episodes of isch-
emia can be devastating to the traumatized brain,72-74 making
it critical to accurately and continuously monitor CPP.75
Because insertion of intraparenchymal ICP monitors is safe,
the ability to monitor CPP per se is a supportable argument
for widespread ICP monitoring. Second, cerebral herniation
is a pressure issue and an ICP monitor may allow early detec-
tion. It is impossible to determine the pressure (pressure
gradient) empirically (i.e., the neurologic examination) when
brain herniation will occur although it is clear once it has
occurred.76 It is preferable to avoid herniation than to treat it
post hoc.77,78 Third, information from an ICP monitor may
provide useful acuity information and so guide patient care
and perhaps ICU resources. For example, a patient with a
worrisome-appearing CT scan79 who does not have intracra-
nial hypertension may not require the same degree of treat-
ment as a patient with a similar scan but elevated ICP.80
Similarly, a patient with elevated ICP that is refractory to
escalating management becomes an early candidate for
“second tier” treatments or if very high, even withdrawal of
care. Fourth, ICP trends can be an early warning of mass lesion
expansion, the appearance of new lesions, or evolution of
edema, ischemia, or hydrocephalus and allow these conditions
to be effectively managed before clinical findings change or
they are detected on periodic imaging studies.81 Finally, because
ICP values have prognostic value it can guide management
and prognosis discussions with the family.
Intracranial Pressure
Monitoring Technology
Technology evaluation is not an ideal topic for evidence-based
medicine. However, the various ICP monitor systems are well
described in the ICP monitoring technology section of the
guidelines for the management of severe traumatic brain
injury.75 This report concluded that at the time of its writing,
three technologies are sufficiently accurate to be interchange-
able in use: (1) a ventricular catheter connected to an external
strain gauge; (2) catheter tip strain gauge devices, or (3) cath-
eter tip fiber-optic technology. Each of these systems can be
placed into the ventricle. Catheter tip strain gauges or fiber-
optics also can be placed into the parenchyma. The most
commonly used devices include the Camino or the Ventrix
(Integra Neurosciences, Plainsboro, NJ), Codman microsen-
sor (Codman, Raynham, MA), the Spiegelberg ICP sensor and
compliance device (Spiegelberg KG, Hamburg, Germany), and
the Raumedic ICP sensor and multiparameter probe (Rau-
medic AG, Munchberg, Germany). The pneumatic Spiegel-
berg ICP monitor unlike the other parenchymal monitors also
allows in vivo calibration and intracranial compliance moni-
toring. The evidence report also concluded that fluid-coupled
or pneumatic devices placed in the subarachnoid, subdural,
and epidural compartments are less accurate.75 Noninvasive
technologies for ICP monitoring based on a variety of tech-
niques such as ultrasound, transcranial Doppler, acoustic
properties of cranial bones, tympanic membrane displace-
ment, heart rate variability and cardiac coupling, ophthalmo-
dynamometry, and measurement of optic nerve diameter are
under investigation. Although more than 30 noninvasive ICP
monitors have been patented since the 1990s, most are still too
cumbersome or not sufficiently accurate to be used in clinical
practice.
The ventricular catheter connected to a fluid-coupled exter-
nal strain gauge is considered the gold standard ICP monitor
in part because it can be used also to treat ICP through CSF
drainage. Control of elevated ICP is observed in about 50%
of patients in whom an external ventricular drain (EVD) is
inserted after other initial measures fail.82 Traditional ventricu-
lar catheter external transducers only permit intermittent ICP

monitoring when the drain is closed (i.e., not draining),
although some catheters have internal transducers that permit
CSF drainage at the same time ICP is monitored. An EVD may
miss episodes of increased ICP when it is draining.83 This gold
standard status, however, may take precedence because it was
the first monitoring system available. Furthermore, there are
no clinical outcome studies that show that one monitoring
technology is superior to others. Because a ventricular catheter
is placed into the ventricle, it is “logically” felt to optimally
represent the ICP. However, even though gradients have
been demonstrated between parenchymal and ventricular
systems,84-87 none of these has ever been found to be of clinical
significance. Parenchymal devices are easier to place, particu-
larly when altered ventricular anatomy may limit ventricular
catheter placement. However, intraparenchymal fiber-optic
and electronic strain gauge systems are more expensive and
cannot be recalibrated once in situ. Consequently, outside the
ability of ventricular-catheter monitors to drain CSF as a
potentially therapeutic maneuver, the choice of an ICP monitor
is probably best decided based on factors such as accuracy,
reliability, complication rates, ease of insertion, and cost.
Potential Complications of
Intracranial Pressure Monitors
Ventricular catheters and ICP bolts traditionally have been
placed by neurosurgeons. However, neurointensivists now are
inserting these monitors on a more frequent basis.88 The com-
plications rate appears to depend more on the technique used
and the device (ventricular catheter vs. parenchymal monitor)
than on who places the monitor once he or she has been
adequately trained.89,90
Infection
The definition of infection of an ICP monitor is somewhat
controversial. It is easier to detect bacterial involvement of
a ventricular catheter rather than a parenchymal monitor
because CSF sampling can be periodically performed. In the
absence of signs of ventriculitis, positive cultures may be
better termed colonization.63 When this definition (coloniza-
tion rather than infection) is used, positive cultures are
observed in 8% of ventriculostomy CSF cultures and 14% of
the cultures obtained from the tip of an intraparenchymal
device once it is removed. In clinical practice, however, routine
surveillance of intraparenchymal monitors is not performed.
By contrast CSF sampling is common with ventricular cath-
eters. Positive CSF cultures may prompt removal of the device
and antibiotics that may lengthen hospital stay, expose patients
to the potential hazards of parenteral antibiotics, and expose
them to the risk of catheter replacement. Therefore although
both types of monitors may be colonized, the consequences
are markedly different. This must be balanced against the
value of CSF drainage.
Iatrogenic catheter-related ventriculitis and meningitis
may occur in 5% to 20% of EVDs through direct catheter
contamination at insertion, through retrograde bacterial colo-
nization, or if introduced when the system is accessed or
flushed.91-95 The overall device infection rate is between 6
and 8 per 1000 drainage days.95 Risk factors for infection
include the presence of concurrent systemic infections, longer
Section VI—Intracranial Monitoring 343
duration of monitoring, presence of intraventricular hemor-
rhage or SAH, open skull fracture (with or without CSF leak),
trauma, flushing of the catheter, and CSF leakage at the inser-
tion site.93,96-100 Because most data are from case series, it
remains uncertain how best to prevent or manage infections.
However, strategies such as use of closed drainage systems, a
long subcutaneous tunnel, reduced duration of ventricular
cannulation (i.e., prompt removal when not needed), avoiding
catheter irrigation or flushing or if needed done using aseptic
techniques, and avoiding CSF leakage from any site are recom-
mended. There does not appear to be a role for continuous
antibiotic prophylaxis.75 Silver- or antibiotic-impregnated
catheters may decrease the incidence of catheter-related CSF
infections, although the exact role for these catheters is still
debated.101-103 Because CSF sampling may predispose to con-
tamination or infection, sampling should be indicated by spe-
cific clinical criteria rather than be a routine practice. Even a
reduction of CSF sampling from daily to every 3 days can
reduce the rate of ventriculitis from 10% to 3%.94 Use of stan-
dard management protocols particularly with a bundled
approach also can help reduce the infection rate.104,105 There
does not appear to be a role for routine replacement, because
it appears that the risk of replacement outweighs any potential
benefit.98,100,106-108 The use of parenchymal ICP monitors
should be considered when there is systemic infection or an
open skull fracture because there are no reports in the litera-
ture of infections associated with these devices in children or
adults.109-111 There is, however, one report of an abscess that
developed several weeks after an ICP monitor was removed
from a 35-week-old child who required both prolonged ICP
monitoring and steroids.112
Hemorrhage
The precise incidence of hemorrhagic complications follow-
ing ICP monitor insertion depends on what device is used,
insertion technique, and how hemorrhages were identified.
Ideally, imaging studies should be obtained before and after
ICP monitor insertion, but few studies include such imaging.
Retrospective case series suggest that the risk of intracranial
hemorrhage after placement of an EVD in adults is between
2% and 10%.113-116 The risk may be greater in children
(17.6%).117 Most hemorrhages are less than 15 mL in volume
and therefore “clinically insignificant.” The incidence of clini-
cally significant hemorrhages associated with ventriculostomy
placement is about 1%.116 After insertion of an intraparenchy-
mal ICP monitor in adults the incidence of intracranial hem-
orrhage is estimated to be between zero and 11%85,110,118,119 in
adults. The incidence is similar in children.111,117,118 Gelabert-
Gonzalez et al. described 1000 consecutive patients who had
intraparenchymal ICP monitors; 922 had postinsertion CT
imaging.110 The overall hemorrhage rate was 2.5%, and 0.66%
required surgery for an intraparenchymal or epidural hemor-
rhage. Two studies have examined intraparenchymal and
intraventricular devices concurrently. The hemorrhage inci-
dence is significantly greater for ventriculostomies.114,117
Technical Issues
Ventricular catheter displacement, accidental removal, or
blockage may occur. When blockage occurs either by debris
or blood that obliterates holes in the catheter or catheter
344 Section VI—Intracranial Monitoring

displacement into the parenchyma, CSF drainage can generate

a large pressure gradient between the catheter lumen and the Table 34.1  Advantages and Disadvantages
ventricle. In this circumstance ICP is underestimated when it of Ventricular and Parenchymal ICP Monitors
is monitored at the same time as CSF is drained and the CSF Ventriculostomy Intraparenchymal Monitor
waveform becomes flattened. When this occurs the drainage
Advantages Advantages
system should be closed to see if the waveform returns. If not, • Reference standard (by • Ease of placement
the system may need to be gently flushed with 1 to 2 mL of default) • Accurately reproduces the
normal saline. The role of thrombolytic agents to clear blood • Can be rezeroed CSF pulse waveform
clots with intraventricular hemorrhage is unclear. Technical • Cheap without mechanical
complications such as catheter breakage or dislodgement • Therapeutic drainage of damping
CSF
occur in about 4.5% of intraparenchymal devices. Most of
these occur during transport, nursing maneuvers, or patient Disadvantages Disadvantages
• Plugging • Can’t be rezeroed without
activities.110 None appear to influence the patient’s course or • Disconnection removal-potential for drift
outcome. Different parenchymal devices including fiber-optic, • Can’t measure ICP • More prone to technical
strain gauge, and pneumatic technologies are available. Only accurately when open to failure (catheter fracture,
the pneumatic Spiegelberg ICP monitor allows in vivo calibra- drainage pull-out, etc.)
tion. Bench testing shows that parenchymal monitors have • Ventriculitis • Incompatible with MRI
• Brain must be • More expensive
excellent accuracy; however, studies suggest suspicion that penetrated
zero-drift rates can be clinically important in strain gauge • Difficult to place with
monitors.120,121 Drift is very rare in fiber-optic catheters.122 In displaced, compressed,
addition, there are descriptions of disturbed baseline pressures or small ventricles
• Complications of
of the Codman and Raumedic ICP sensors by electrostatic overdrainage
discharges.123 • Requires rezeroing of
the system with any
change in the height of
the head of the bed
Optimal Placement of potential for error
Intraparenchymal Devices • Dampens the waveform
due to the mechanical
Intraparenchymal ICP monitors are placed into the brain properties of the tubing
parenchyma through a small burr hole and a cranial access and if air is present
device or bolt that may allow placement of one to three moni- CSF, Cerebrospinal fluid; ICP, intracranial pressure; MRI, magnetic resonance
tors. Usually these monitors are placed into the nondominant imaging.
frontal lobe when the injury or pathology is diffuse. However,
in focal injury these monitors may be better placed on the side
of the pathology or in pericontusional parenchyma because,
unlike ventricular catheters, intraparenchymal devices have or protocols to correct coagulopathies before an EVD or ICP
the potential to represent local, “compartmental” pressures, monitor is placed. Factor administration, however, may delay
should supratentorial gradients exist. In particular, intra­ the procedure. Studies suggest an international normalized
parenchymal devices placed in the hemisphere contralateral ratio (INR) less than or equal to 1.6 is acceptable to place a
to a mass lesion may underestimate ICP even with brain ventricular catheter following TBI.128,129 Many physicians
herniation.124 regard a platelet count of 100,000 necessary to safely place an
Some but not all clinical studies in TBI with older technol- ICP monitor; however, there is little literature on this topic.
ogy such as subarachnoid bolts (not recommended in the Davis et al. addressed the need to normalize coagulation
TBI evidence report75) often demonstrated interhemispheric parameters before inserting a fiberoptic intraparenchymal
gradients.125-127 Sahuquillo et al. using fiber-optic intraparen- ICP monitor.130 They retrospectively reviewed 157 patients
chymal monitors also found interhemispheric gradients in including laboratory values and postinsertion CT images.
patients with focal lesions that could be greater than 10 mm They included 10 patients with borderline INR values (1.3-
Hg.87 One quarter of such gradients produced a difference in 1.6) and 12 patients with INR values greater than or equal to
calculated CPP of greater than or equal to 5 mm Hg, which the 1.7 at the time of placement. Eleven of their patients had
authors felt had the potential to alter treatment. The authors platelet counts between 50,000 and 100,000 at the time of
therefore concluded that “to optimize ICP monitoring in insertion. Three patients (1.9%) had insertion-related pete-
patients with a mass lesion larger than 25 mL or midline shift, chial hemorrhages without clinical significance. They con-
the measuring device should preferably be implanted in both cluded that intraparenchymal monitor insertion with an INR
the brain parenchyma and the side of the mass” (Table 34.1). less than or equal to 1.6 is safe. They reported that the use of
fresh frozen plasma (FFP) to normalize the INR when less
than 1.6 delayed monitoring and was not necessary. These
Insertion of ICP Monitors data and other studies suggest that delaying ICP monitor
in Patients with Abnormal insertion may not be necessary in patients with mild coagu-
lopathy if an intraparenchymal device is used.122 In patients
Coagulation Studies with fulminant hepatic failure recombinant activated factor
Patients who require an ICP monitor may have abnormal VII (rFVIIa) has been recommended if an ICP monitor is
coagulation parameters. Many centers have routine practices required.131

How to Monitor Intracranial
Pressure and for How Long
Continuous digital recording is the best method to acquire
ICP data because this method can be linked to other monitors
to calculate derived indexes and should not miss transient
changes in ICP.132 This is not feasible in all ICUs. Instead
NCCU staff chart ICP values on a regular basis. End-hour
recording generally correlates well with continuous record-
ing133; however, to identify episodic events, recording every
10 minutes may be needed. Even episodes of increased ICP as
brief as 5 minutes can aggravate outcome134; use of alarms may
help identify this. Following TBI patients who develop
increased ICP generally do so in the first week135 and although
there are several patterns to the timing of ICP elevation, about
20% first show ICP elevations after 72 hours.136 The ICP
monitor should be kept in place while clinically indicated but
usually can be removed once the patient starts to follow com-
mands. If the patient remains comatose, removal of the
monitor can be considered if ICP is normal during the first
72 hours. If ICP is elevated, the monitor may be removed once
ICP is normal for at least 24 hours without any specific treat-
ment other than sedation for the ventilator. A follow-up CT
scan that demonstrates resolving mass effect is useful to guide
this decision, and other monitors such as brain oxygen or an
electroencephalogram (EEG) may help corroborate this.
Cerebral Compliance and
Compensatory Reserve
Cerebral compliance is the “stiffness” of the intracranial com-
partment and is physically represented by the ICP response to
a change in intracranial volume. In keeping with the Monro-
Kellie doctrine, a small increase in intracranial volume when
the compliance is good produces only a small increase in the
ICP (Fig. 34.3). When compliance is compromised, the ICP
response to the same volume change increases and becomes
Decreasing cerebral compliance
Pressure
∆P
∆V
∆P
∆V
Volume
Fig. 34.3  Cerebral compliance represents the ICP response (ΔP) to
a given volume challenge (ΔV). As compliance is progressively
compromised (see Fig. 34.1), the pressure response increases. When
compliance is mildly to moderately stressed, the intracranial pressure
(ICP) may remain acceptable despite marginal remaining buffering
capacity. When compliance is exhausted, a small volume challenge may
result in severe intracranial hypertension or acute herniation.
Section VI—Intracranial Monitoring 345
significant when compliance is very poor. The magnitude of
how a volume change influences ICP is illustrated by studies
on the amount of blood volume change necessary to produce
a 1 mm Hg ICP elevation. Studies on TBI patients suggest that
such volumes are as small as 0.42 to 0.5 cc, and less with
decreased compliance.137,138 Hypercapnea can increase the sen-
sitivity of the ICP to volume challenges.
Compensation for a volume change is time dependent.
Compensatory responses in the CSF or CBV compartments
can partially buffer volume increases (e.g., evolving cerebral
edema). By contrast, rapid volume challenges are less buffered
and have a greater influence on ICP. This may include naso-
tracheal suctioning, patient turning, or physiologic insults
such as hypercarbia (increased CBV) that all may increase ICP
when cerebral compliance is poor.
To quantify cerebral compliance requires the ability to
deliver a reproducible volume challenge and measure the ICP
change. The volume-pressure response (VPR) can be mea-
sured by rapidly injecting a small (1 cc) volume of fluid into
a patient’s ventricle through a ventriculostomy and measuring
the immediate ICP response (VPR = ICP change/volume
injected).139,140 This provides a snapshot of the compliance at
that moment but does not describe the shape of the volume-
pressure curve, which may vary among patients and within
patients over time.141 Assuming a mono-exponential relation-
ship, logarithmic conversion of this data provides the pressure-
volume index (PVI), which defines the volume change that
would produce a 10-fold increase in ICP (PVI = V/loge[Po/
Pm]).142 The PVI describes the shape of the compliance curve
but it does not represent the patient’s position along that
curve. This renders the PVR and PVI of complementary value.
The injection of fluid into the ventricular system, however,
requires “opening” the system. This appears to be associated
with an increased risk of infection.143 The ICP also may not
always readily return to baseline after fluid injection and
therefore cause intracranial hypertension. Although the same
calculations may be done with fluid withdrawal, this is more
difficult to perform and is rarely used. Hence measurement of
PVR and PVI (compliance) through a ventriculostomy is not
routine in clinical practice. Simple but very indirect methods
such as pressure on the patient’s abdomen and watching the
ICP response may provide qualitative estimates about the state
of compliance.
The Spiegelberg ICP sensor and compliance device allows
compliance testing. A ventriculostomy is inserted and at the
catheter tip is a small (0.1 cc) balloon that can be automati-
cally inflated and deflated in a cyclic fashion. Although the
response to such a small volume challenge can be difficult to
detect among the noise inherent in clinical ICP monitoring,
signal-triggered averaging is used to extract that signal and
analyze it to provide compliance data. Experimental and clini-
cal (in hydrocephalus) validation testing of this system against
VPR/PVI values provides support for its reliability.144,145
However, it is not in widespread clinical use.
The clinical evaluation of cerebral compliance is qualitative,
based on observation of the ICP waveform.146,147 The first three
wavelets prominent in the ICP pulse waveform are labeled
P1, P2, and P3 (Fig. 34.4). The first component (P1) is generated
by the systolic pulse wave. The second (P2—the “tidal wave”)
is a harmonic wave and reflects cerebral compliance. The
third (P3—the “dicrotic wave”) represents the reflection of the
dicrotic venous wave that results from closure of the aortic
346 Section VI—Intracranial Monitoring
P1
P2
P3
Fig. 34.4  A normal intracranial pressure (ICP) waveform, in which the
first peak (P1) is higher than subsequent peaks (P2 and P3).
A
B
Fig. 34.5  Intracranial pressure (ICP) waveform response to
treatment. ICP recording A represents abnormal compliance, with P2
and P3 being higher than P1. Treatment with osmotherapy and improving
the blood pressure resulted in a decrease of the ICP accompanied with
improvement in the waveform such that P1 is again greater than P2 and
P3 (recording B).
valve. Other, smaller wavelets may follow these three. Under
normal conditions (i.e., good compliance) P1 is the highest
component of the pulse waveform (Fig. 34.5, A). As compli-
ance worsens, often with intracranial hypertension, the ampli-
tude of P2 may increase until it equals or exceeds P1 (see Fig.
34.5, B), and comes to resemble the arterial pulse waveform.
These changes may be constant or may vary over time (i.e.,
with the respiratory cycle). The elevation of P2 over P1 is a
sensitive (99%) but not specific (1%-17%) predictor of subse-
quent ICP increases associated with nursing maneuvers.148
Patients with abnormal waveforms tend to have worse out-
comes. Waveform indices such as high-frequency centroid or
waveform transmission are associated with mortality. However,
quantitative ICP waveform analysis has had limited success in
clinical practice149-152 because the indices do not appear to be
independent of other parameters. Exactly how waveform anal-
ysis can help guide patient management or help predict intra-
cranial hypertension is still being elucidated.153,154
Advances in computerized monitoring and data processing
now permit online, real-time analysis of the interdependence
between the regulatory processes of ICP. This has led to the
development of derived indices of cerebrovascular pressure
reactivity (PRx) and cerebrospinal compensatory reserve
(RAP). These indices provide insight into the compensatory
reserve of the intracranial compartment and cerebrovascular
autoregulatory reserve.155,156 When the RAP index is near zero,
there is a good compensatory reserve, whereas a negative PRx
(–1 to 0) suggests there is preserved autoregulation. By con-
trast a high RAP index (+1), or a high PRx (>0.3) infers there
is little compensatory reserve in the intracranial compartment
and therefore even small changes in volume may lead to a
rapid ICP increase. These indices are primarily used in a
research environment and have not yet entered routine prac-
tice, although there is a described association between abnor-
mal RAP, PRx ,and poor outcome.157 Dynamic autoregulation
also may be assessed by analysis of the ICP response to induced
or physiologic challenges.80,155 This may help guide choice of
CPP thresholds.155 Finally, insight into the brain’s compensa-
tory reserve may be estimated from the Therapeutic Intensity
Level (TIL) score. This is a quantitative measure of what man-
agement is needed to manage ICP.158 A greater TIL implies that
multiple therapies or more complex therapies are required to
control ICP.
Conclusion
Elevated ICP is common after most forms of acute brain
injury, and its prevention and management are central to the
care of many patients admitted to the NCCU. There is well-
described association between significant ICP elevation and
poor outcome, and its avoidance or rapid treatment appears
to be associated with reduced morbidity and mortality when
part of protocol-driven therapy. There is, however, ongoing
debate about the precise treatment threshold, the optimal
approach to lower ICP, and whether or how mild elevation in
ICP should be managed, in both the adult and pediatric pop-
ulations.159 Other parameters such as CPP may be every bit
as important as ICP, and the interactions between these
variables and their management protocols require further
clarification. In addition with the advent of multimodality
monitoring, it is apparent that ICP values should be inter-
preted carefully and with other clinical and radiologic param-
eters. However, because ICP plays a fundamental role in
almost all aspects of TBI management, and in the care of
most patients admitted to the NCCU, knowing its precise
value is integral to direct therapy, allocate resources, estimate
prognosis, and prevent (rather than just treat) secondary
brain insults. Because there are many safe and effective ICP
monitoring technologies, there is little if any role for empiri-
cal ICP management. Furthermore, accurate monitoring of
the physiologic state and the response to treatment can help
optimize treatment protocols.
References
1. Bratton SL, Chesnut RM, Ghajar J, et al. Guidelines for the management of
severe traumatic brain injury. VI. Indications for intracranial pressure
monitoring. J Neurotrauma 2007;24(Suppl. 1):S37–44.
2. Odetola FO, Clark SJ, Lamarand KE, et al. Intracranial pressure monitoring
in childhood meningitis with coma: a national survey of neurosurgeons in
the United States. Pediatr Crit Care Med 2011;12(6):e350–6.
3. Kamat P, Kunde S, Vos M, et al. Invasive intracranial pressure monitoring is
a useful adjunct in the management of severe hepatic encephalopathy
associated with pediatric acute liver failure. Pediatr Crit Care Med 2011.
Epub ahead of print.
4. Kouvarellis AJ, Rohlwink UK, Sood V, et al. The relationship between basal
cisterns on CT and time-linked intracranial pressure in paediatric head
injury. Childs Nerv Syst 2011;27(7):1139–44.

5. Kellie G. An account of the appearances observed in the dissection of two
of the three individuals presumed to have perished in the storm of the 3rd,
and whose bodies were discovered in the vicinity of Leith on the morning of
the 4th November 1821 with some reflections on the pathology of the brain.
Transmissions of Medical and Chirurgical Science, Edinburgh 1824;1:84–169.
6. Monro A. Observations on the structure and function of the nervous system.
Edinburgh: Creech and Johnson; 1783.
7. Cushing H. The third circulation in studies in intracranial physiology and
surgery. London: Oxford University Press; 1926.
8. Marmarou A. A review of progress in understanding the pathophysiology
and treatment of brain edema. Neurosurg Focus 2007;22(5):E1.
9. Luce JM, Huseby JS, Kirk W, et al. Starling resistor regulates cerebral venous
outflow in dogs. J Appl Physiol 1982;53(6):1496–503.
10. Rosner MJ. Pathophysiology and management of increased intracranial
pressure. In: Andrews BT, editor. Neurosurgical intensive care. New York:
McGraw Hill; 1993. p. 57–112.
11. Gopinath SP, Robertson CS, Contant CF, et al. Jugular venous desaturation
and outcome after head injury. J Neurol Neurosurg Psychiatry 1994;57(6):
717–23.
12. Muizelaar JP, Marmarou A, Ward JD, et al. Adverse effects of prolonged
hyperventilation in patients with severe head injury: a randomized clinical
trial. J Neurosurg 1991;75(5):731–9.
13. Muizelaar JP, van der Poel HG, Li ZC, et al. Pial arteriolar vessel diameter
and CO2 reactivity during prolonged hyperventilation in the rabbit.
J Neurosurg 1988;69(6):923–7.
14. Rockswold SB, Rockswold GL, Vargo JM, et al. Effects of hyperbaric
oxygenation therapy on cerebral metabolism and intracranial pressure in
severely brain injured patients. J Neurosurg 2001;94(3):403–11.
15. Rockswold SB, Rockswold GL, Zaun DA, et al. A prospective, randomized
clinical trial to compare the effect of hyperbaric to normobaric hyperoxia on
cerebral metabolism, intracranial pressure, and oxygen toxicity in severe
traumatic brain injury. J Neurosurg 2010;112(5):1080–94.
16. Tolias CM, Reinert M, Seiler R, et al. Normobaric hyperoxia–induced
improvement in cerebral metabolism and reduction in intracranial pressure
in patients with severe head injury: a prospective historical cohort-matched
study. J Neurosurg 2004;101(3):435–44.
17. Muizelaar JP, Wei EP, Kontos HA, et al. Mannitol causes compensatory
cerebral vasoconstriction and vasodilation in response to blood viscosity
changes. J Neurosurg 1983;59(5):822–8.
18. Muizelaar JP, Lutz 3rd HA, Becker DP. Effect of mannitol on ICP and CBF
and correlation with pressure autoregulation in severely head-injured
patients. J Neurosurg 1984;61(4):700–6.
19. Stiefel MF, Tomita Y, Marmarou A. Secondary ischemia impairing the
restoration of ion homeostasis following traumatic brain injury. J Neurosurg
2005;103(4):707–14.
20. Marmarou A, Signoretti S, Fatouros PP, et al. Predominance of cellular
edema in traumatic brain swelling in patients with severe head injuries.
J Neurosurg 2006;104(5):720–30.
21. Bullock MR, Chesnut R, Ghajar J, et al. Guidelines for the surgical
management of traumatic brain injury. Neurosurgery 2006;58(Suppl. 3):Si–60.
22. Hutchinson P, Timofeev I, Kirkpatrick P. Surgery for brain edema.
Neurosurg Focus 2007;22(5):E14.
23. Timofeev I, Czosnyka M, Nortje J, et al. Effect of decompressive craniectomy
on intracranial pressure and cerebrospinal compensation following
traumatic brain injury. J Neurosurg 2008;108(1):66–73.
24. Li LM, Timofeev I, Czosnyka M, et al. Review article: the surgical approach
to the management of increased intracranial pressure after traumatic brain
injury. Anesth Analg 2010;111(3):736–48. Epub 2010, Aug 4.
25. Vahedi K, Hofmeijer J, Juettler E, et al. Early decompressive surgery in
malignant infarction of the middle cerebral artery: a pooled analysis of three
randomised controlled trials. Lancet Neurol 2007;6(3):215–22.
26. Hofmeijer J, Kappelle LJ, Algra A, et al. Surgical decompression for
space-occupying cerebral infarction (the Hemicraniectomy After Middle
cerebral artery infarction with Life-threatening Edema Trial [HAMLET]):
a multicenter, open, randomized trial. Lancet Neurol 2009;8(4):326–33.
27. Cooper DJ, Rosenfeld JV, Murray L, et al. Decompressive craniectomy in
diffuse traumatic brain injury. N Engl J Med 2011;364(16):1493–502. Epub
2011, Mar 25.
Section VI—Intracranial Monitoring 347
28. Hutchinson PJ, Timofeev I, Kolias AG, et al. Decompressive craniectomy for
traumatic brain injury: the jury is still out. Br J Neurosurg 2011;25(3):441–2.
29. Clifton GL, Miller ER, Choi SC, et al. Lack of effect of induction of
hypothermia after acute brain injury. N Engl J Med 2001;344(8):556–63.
30. Hutchison JS, Ward RE, Lacroix J, et al. Hypothermia therapy after traumatic
brain injury in children. N Engl J Med 2008;358(23):2447–56.
31. Bernard SA, Gray TW, Buist MD, et al. Treatment of comatose survivors of
out-of-hospital cardiac arrest with induced hypothermia. N Engl J Med
2002;346:557–63.
32. Heuer G, Smith MJ, Elliott JP, et al. The relationship between intracranial
pressure and other clinical variables in patients with aneurysmal
subarachnoid hemorrhage. J Neurosurg 2004;101:408–16.
33. Wendell LC, Khan A, Raser J, et al. Successful management of refractory
intracranial hypertension from acute hyperammonemic encephalopathy in a
woman with ornithine transcarbamylase deficiency. Neurocrit Care 2010;13;
113–7.
34. Cremer OL, van Dijk GW, van Wensen E, et al. Effect of intracranial pressure
monitoring and targeted intensive care on functional outcome after severe
head injury. Crit Care Med 2005;33(10):2207–13.
35. Shafi S, Diaz-Arrastia R, Madden C, et al. Intracranial pressure monitoring
in brain-injured patients is associated with worsening of survival. J Trauma
2008;64(2):335–40.
36. Bulger EM, Nathens AB, Rivara FP, et al. Management of severe head injury:
institutional variations in care and effect on outcome. Crit Care Med
2002;30(8):1870–6.
37. Ghajar J, Hariri RJ, Narayan RK, et al. Survey of critical care management of
comatose, head-injured patients in the United States. Crit Care Med
1995;23(3):560–7.
38. Hesdorffer DC, Ghajar J, Iacono L. Predictors of compliance with the
evidence-based guidelines for traumatic brain injury care: a survey of United
States trauma centers. J Trauma 2002;52(6):1202–9.
39. Becker DP, Miller JD, Ward JD, et al. The outcome from severe head injury
with early diagnosis and intensive management. J Neurosurg 1977;47(4):
491–502.
40. Rosner MJ, Rosner SD, Johnson AH. Cerebral perfusion pressure:
management protocol and clinical results. J Neurosurg 1995;83(6):949–62.
41. Grande PO. The “Lund concept” for the treatment of severe head trauma–
physiological principles and clinical application. Intensive Care Med
2006;32(10):1475–84. Epub 2006, Aug 2.
42. Guillaume J, Janny P. Monometrie intracranienne continué; Intérêt physio-
pathologique et clinique de la methode. La Presse Medicale 1951;59:953–5.
43. Lundberg N. Continuous recording and control of ventricular fluid pressure
in neurosurgical practice. Acta Psychiatr Scand Suppl 1960;149:1–193.
44. Haar FL, Sadhu VK, Sampson J, et al. Can CT scan findings predict ICP in
closed head injury patients? The Thirtieth Annual Meeting of the Congress
of Neurological Surgeons. Houston, Texas; 1980. p. 122–4.
45. Marmarou A, Anderson RL, Ward JD, et al. Impact of ICP instability and
hypotension on outcome in patients with severe head trauma. J Neurosurg
1991;75(Suppl):S159–66.
46. Marshall LF, Gautille T, Klauber MR, et al. The outcome of severe head
injury. J Neurosurg 1991;75(Suppl):S28–36.
47. Marshall LF, Smith RW, Shapiro HM. The outcome with aggressive
treatment in severe head injuries. Part I. The significance of intracranial
pressure monitoring. J Neurosurg 1979;50(1):20–5.
48. Miller JD, Becker DP, Ward JD, et al. Significance of intracranial
hypertension in severe head injury. J Neurosurg 1977;47(4):503–16.
49. Pitts E, Katkis J. Effects of megadose steroids on ICP in traumatic coma. In:
Shulman K, Marmarou A, Miller A, editors. Intracranial pressure IV. New
York: Springer-Verlag; 1972. p. 638–42.
50. Schoon P, Benito Mori L, Orlandi G, et al. Incidence of intracranial
hypertension related to jugular bulb oxygen saturation disturbances in severe
traumatic brain injury patients. Acta Neurochir Suppl 2002;81:285–7.
A complete list of references for this chapter can be found online at
www.expertconsult.com.

References
1. Bratton SL, Chesnut RM, Ghajar J, et al. Guidelines for the management of
severe traumatic brain injury. VI. Indications for intracranial pressure
monitoring. J Neurotrauma 2007;24(Suppl. 1):S37–44.
2. Odetola FO, Clark SJ, Lamarand KE, et al. Intracranial pressure monitoring
in childhood meningitis with coma: a national survey of neurosurgeons in
the United States. Pediatr Crit Care Med 2011;12(6):e350–6.
3. Kamat P, Kunde S, Vos M, et al. Invasive intracranial pressure monitoring is
a useful adjunct in the management of severe hepatic encephalopathy
associated with pediatric acute liver failure. Pediatr Crit Care Med 2011.
Epub ahead of print.
4. Kouvarellis AJ, Rohlwink UK, Sood V, et al. The relationship between basal
cisterns on CT and time-linked intracranial pressure in paediatric head
injury. Childs Nerv Syst 2011;27(7):1139–44.
5. Kellie G. An account of the appearances observed in the dissection of two of
the three individuals presumed to have perished in the storm of the 3rd, and
whose bodies were discovered in the vicinity of Leith on the morning of the
4th November 1821 with some reflections on the pathology of the brain.
Transmissions of Medical and Chirurgical Science, Edinburgh 1824;1:84–169.
6. Monro A. Observations on the structure and function of the nervous
system. Edinburgh: Creech and Johnson; 1783.
7. Cushing H. The third circulation in studies in intracranial physiology and
surgery. London: Oxford University Press; 1926.
8. Marmarou A. A review of progress in understanding the pathophysiology
and treatment of brain edema. Neurosurg Focus 2007;22(5):E1.
9. Luce JM, Huseby JS, Kirk W, et al. Starling resistor regulates cerebral venous
outflow in dogs. J Appl Physiol 1982;53(6):1496–503.
10. Rosner MJ. Pathophysiology and management of increased intracranial
pressure. In: Andrews BT, editor. Neurosurgical intensive care. New York:
McGraw Hill; 1993. p. 57–112.
11. Gopinath SP, Robertson CS, Contant CF, et al. Jugular venous desaturation
and outcome after head injury. J Neurol Neurosurg Psychiatry 1994;57(6):
717–23.
12. Muizelaar JP, Marmarou A, Ward JD, et al. Adverse effects of prolonged
hyperventilation in patients with severe head injury: a randomized clinical
trial. J Neurosurg 1991;75(5):731–9.
13. Muizelaar JP, van der Poel HG, Li ZC, et al. Pial arteriolar vessel diameter
and CO2 reactivity during prolonged hyperventilation in the rabbit.
J Neurosurg 1988;69(6):923–7.
14. Rockswold SB, Rockswold GL, Vargo JM, et al. Effects of hyperbaric
oxygenation therapy on cerebral metabolism and intracranial pressure in
severely brain injured patients. J Neurosurg 2001;94(3):403–11.
15. Rockswold SB, Rockswold GL, Zaun DA, et al. A prospective, randomized
clinical trial to compare the effect of hyperbaric to normobaric hyperoxia
on cerebral metabolism, intracranial pressure, and oxygen toxicity in severe
traumatic brain injury. J Neurosurg 2010;112(5):1080–94.
16. Tolias CM, Reinert M, Seiler R, et al. Normobaric hyperoxia–induced
improvement in cerebral metabolism and reduction in intracranial pressure
in patients with severe head injury: a prospective historical cohort-matched
study. J Neurosurg 2004;101(3):435–44.
17. Muizelaar JP, Wei EP, Kontos HA, et al. Mannitol causes compensatory
cerebral vasoconstriction and vasodilation in response to blood viscosity
changes. J Neurosurg 1983;59(5):822–8.
18. Muizelaar JP, Lutz HA, 3rd, Becker DP. Effect of mannitol on ICP and CBF
and correlation with pressure autoregulation in severely head-injured
patients. J Neurosurg 1984;61(4):700–6.
19. Stiefel MF, Tomita Y, Marmarou A. Secondary ischemia impairing the
restoration of ion homeostasis following traumatic brain injury.
J Neurosurg 2005;103(4):707–14.
20. Marmarou A, Signoretti S, Fatouros PP, et al. Predominance of cellular
edema in traumatic brain swelling in patients with severe head injuries.
J Neurosurg 2006;104(5):720–30.
21. Bullock MR, Chesnut R, Ghajar J, et al. Guidelines for the surgical
management of traumatic brain injury. Neurosurgery 2006;58(Suppl. 3):
Si–60.
22. Hutchinson P, Timofeev I, Kirkpatrick P. Surgery for brain edema.
Neurosurg Focus 2007;22(5):E14.
23. Timofeev I, Czosnyka M, Nortje J, et al. Effect of decompressive
craniectomy on intracranial pressure and cerebrospinal compensation
following traumatic brain injury. J Neurosurg 2008;108(1):66–73.
24. Li LM, Timofeev I, Czosnyka M, et al. Review article: the surgical approach
to the management of increased intracranial pressure after traumatic brain
injury. Anesth Analg 2010;111(3):736–48. Epub 2010, Aug 4.
25. Vahedi K, Hofmeijer J, Juettler E, et al. Early decompressive surgery in
malignant infarction of the middle cerebral artery: a pooled analysis of
three randomised controlled trials. Lancet Neurol 2007;6(3):215–22.
Section VI—Intracranial Monitoring 347.e1
26. Hofmeijer J, Kappelle LJ, Algra A, et al. Surgical decompression for
space-occupying cerebral infarction (the Hemicraniectomy After Middle
cerebral artery infarction with Life-threatening Edema Trial [HAMLET]):
a multicenter, open, randomized trial. Lancet Neurol 2009;8(4):326–33.
27. Cooper DJ, Rosenfeld JV, Murray L, et al. Decompressive craniectomy in
diffuse traumatic brain injury. N Engl J Med 2011;364(16):1493–502. Epub
2011, Mar 25.
28. Hutchinson PJ, Timofeev I, Kolias AG, et al. Decompressive craniectomy for
traumatic brain injury: the jury is still out. Br J Neurosurg 2011;25(3):
441–2.
29. Clifton GL, Miller ER, Choi SC, et al. Lack of effect of induction of
hypothermia after acute brain injury. N Engl J Med 2001;344(8):556–63.
30. Hutchison JS, Ward RE, Lacroix J, et al. Hypothermia therapy after
traumatic brain injury in children. N Engl J Med 2008;358(23):2447–56.
31. Bernard SA, Gray TW, Buist MD, et al. Treatment of comatose survivors of
out-of-hospital cardiac arrest with induced hypothermia. N Engl J Med
2002;346:557–63.
32. Heuer G, Smith MJ, Elliott JP, et al. The relationship between intracranial
pressure and other clinical variables in patients with aneurysmal
subarachnoid hemorrhage. J Neurosurg 2004;101:408–16.
33. Wendell LC, Khan A, Raser J, et al. Successful management of refractory
intracranial hypertension from acute hyperammonemic encephalopathy in
a woman with ornithine transcarbamylase deficiency. Neurocrit Care
2010;13;113–7.
34. Cremer OL, van Dijk GW, van Wensen E, et al. Effect of intracranial
pressure monitoring and targeted intensive care on functional outcome
after severe head injury. Crit Care Med 2005;33(10):2207–13.
35. Shafi S, Diaz-Arrastia R, Madden C, et al. Intracranial pressure monitoring
in brain-injured patients is associated with worsening of survival. J Trauma
2008;64(2):335–40.
36. Bulger EM, Nathens AB, Rivara FP, et al. Management of severe head
injury: institutional variations in care and effect on outcome. Crit Care
Med 2002;30(8):1870–6.
37. Ghajar J, Hariri RJ, Narayan RK, et al. Survey of critical care management
of comatose, head-injured patients in the United States. Crit Care Med
1995;23(3):560–7.
38. Hesdorffer DC, Ghajar J, Iacono L. Predictors of compliance with the
evidence-based guidelines for traumatic brain injury care: a survey of
United States trauma centers. J Trauma 2002;52(6):1202–9.
39. Becker DP, Miller JD, Ward JD, et al. The outcome from severe head injury
with early diagnosis and intensive management. J Neurosurg 1977;47(4):
491–502.
40. Rosner MJ, Rosner SD, Johnson AH. Cerebral perfusion pressure:
management protocol and clinical results. J Neurosurg 1995;83(6):
949–62.
41. Grande PO. The “Lund concept” for the treatment of severe head trauma–
physiological principles and clinical application. Intensive Care Med 2006;
32(10):1475–84. Epub 2006, Aug 2.
42. Guillaume J, Janny P. Monometrie intracranienne continué; Intérêt
physio-pathologique et clinique de la methode. La Presse Medicale 1951;59:
953–5.
43. Lundberg N. Continuous recording and control of ventricular fluid
pressure in neurosurgical practice. Acta Psychiatr Scand Suppl 1960;149:
1–193.
44. Haar FL, Sadhu VK, Sampson J, et al. Can CT scan findings predict ICP in
closed head injury patients? The Thirtieth Annual Meeting of the Congress
of Neurological Surgeons. Houston, Texas; 1980. p. 122–4.
45. Marmarou A, Anderson RL, Ward JD, et al. Impact of ICP instability and
hypotension on outcome in patients with severe head trauma. J Neurosurg
1991;75(Suppl):S159–66.
46. Marshall LF, Gautille T, Klauber MR, et al. The outcome of severe head
injury. J Neurosurg 1991;75(Suppl):S28–36.
47. Marshall LF, Smith RW, Shapiro HM. The outcome with aggressive
treatment in severe head injuries. Part I. The significance of intracranial
pressure monitoring. J Neurosurg 1979;50(1):20–5.
48. Miller JD, Becker DP, Ward JD, et al. Significance of intracranial
hypertension in severe head injury. J Neurosurg 1977;47(4):503–16.
49. Pitts E, Katkis J. Effects of megadose steroids on ICP in traumatic coma. In:
Shulman K, Marmarou A, Miller A, editors. Intracranial pressure IV. New
York: Springer-Verlag; 1972. p. 638–42.
50. Schoon P, Benito Mori L, Orlandi G, et al. Incidence of intracranial
hypertension related to jugular bulb oxygen saturation disturbances in
severe traumatic brain injury patients. Acta Neurochir Suppl 2002;81:
285–7.
51. Vik A, Nag T, Fredriksli OA, et al. Relationship of “dose” of intracranial
hypertension to outcome in severe traumatic brain injury.
J Neurosurg 2008;109(4):678–84.
347.e2 Section VI—Intracranial Monitoring
52. Chesnut RM, Gautille T, Blunt BA, et al. Neurogenic hypotension in
patients with severe head injuries. J Trauma 1998;44(6):958–63; discussion
63–4.
53. Eisenberg H, Frankowski R, Contant C, et al. High-dose barbiturate control
of elevated intracranial pressure in patients with severe head injury.
J Neurosurg 1988;69:15–23.
54. Schwartz ML, Tator CH, Rowed DW, et al. The University of Toronto head
injury treatment study: a prospective, randomized comparison of
pentobarbital and mannitol. Can J Neurological Sci 1984;11(4):434–40.
55. Ward JD, Becker DP, Miller JD, et al. Failure of prophylactic barbiturate
coma in the treatment of severe head injury. J Neurosurg 1985;62(3):383–8.
56. Robertson CS, Valadka AB, Hannay HJ, et al. Prevention of secondary
ischemic insults after severe head injury. Crit Care Med 1999;27(10):
2086–95.
57. Girard TD, Kress JP, Fuchs BD, et al. Efficacy and safety of a paired sedation
and ventilator weaning protocol for mechanically ventilated patients in
intensive care (Awakening and Breathing Controlled trial): a randomised
controlled trial. Lancet 2008;371(9607):126–34.
58. Kress JP, Pohlman AS, O’Connor MF, et al. Daily interruption of sedative
infusions in critically ill patients undergoing mechanical ventilation. N Engl
J Med 2000;342(20):1471–7.
59. Forsyth RJ, Wolny S, Rodrigues B. Routine intracranial pressure monitoring
in acute coma. Cochrane Database Syst Rev 2010; CD002043.
60. Marshall LF. Treatment of brain swelling and brain edema in man. Adv
Neurol 1980;28:459–69.
61. Saul TG, Ducker TB. Effect of intracranial pressure monitoring and
aggressive treatment on mortality in severe head injury. J Neurosurg
1982;56(4):498–503.
62. Stein SC, Georgoff P, Meghan S, et al. 150 years of treating severe traumatic
brain injury: a systematic review of progress in mortality. J Neurotrauma
2010;27(7):1343–53.
63. Bullock R, Chesnut R, Clifton G, et al. Guidelines for the management of
severe head injury. J Neurotrauma 1996;13(11):639–734.
64. Harvey S, Harrison DA, Singer M, et al. Assessment of the clinical
effectiveness of pulmonary artery catheters in management of patients in
intensive care (PAC-Man): a randomised controlled trial. Lancet 2005;
366(9484):472–7.
65. Richard C, Warszawski J, Anguel N, et al. Early use of the pulmonary artery
catheter and outcomes in patients with shock and acute respiratory distress
syndrome: a randomized controlled trial. JAMA 2003;290(20):2713–20.
66. Sandham JD, Hull RD, Brant RF, et al. A randomized, controlled trial of the
use of pulmonary-artery catheters in high-risk surgical patients. N Engl J
Med 2003;348(1):5–14.
67. Schreiber MA, Aoki N, Scott BG, et al. Determinants of mortality in
patients with severe blunt head injury. Arch Surg 2002;137(3):285–90.
68. Stiefel MF, Udoetek J, Spiotta A, et al. Conventional neurocritical care and
cerebral oxygenation after traumatic brain injury. J Neurosurg 2006;105:
568–75.
69. Le Roux P, Lam AM, Newell DW, et al. Cerebral arteriovenous difference of
oxygen: a predictor of cerebral infarction and outcome in severe head
injury. J Neurosurg 1997;87:1–8.
70. Nangunoori R, Maloney-Wilensky E, Stiefel MD, et al. Brain tissue oxygen
based therapy and outcome after severe traumatic brain injury: a systematic
literature review. Neurocritical Care 2012;17:131−8.
71. Chambers IR, Treadwell L, Mendelow AD. Determination of threshold
levels of cerebral perfusion pressure and intracranial pressure in severe
head injury by using receiver-operating characteristic curves: an
observational study in 291 patients. J Neurosurg 2001;94(3):412–6.
72. Chesnut RM, Marshall LF, Klauber MR, et al. The role of secondary brain
injury in determining outcome from severe head injury. J Trauma 1993;
34(2):216–22.
73. Fearnside MR, Cook RJ, McDougall P, et al. The Westmead Head Injury
Project outcome in severe head injury. A comparative analysis of
pre-hospital, clinical and CT variables. Br J Neurosurg 1993;7(3):267–79.
74. Jones PA, Andrews PJ, Midgley S, et al. Measuring the burden of secondary
insults in head-injured patients during intensive care. J Neurosurg
Anesthesiol 1994;6(1):4–14.
75. Bratton S, Bullock R, Carney N, et al. Guidelines for the management of
severe brain injury: 2007 Revision. J Neurotrauma 2007;24(Suppl .1):
S1–106.
76. Marshall LF, Barba D, Toole BM, et al. The oval pupil: clinical significance
and relationship to intracranial hypertension. J Neurosurg 1983;58:566–8.
77. Juul N, Morris GF, Marshall SB, et al. Intracranial hypertension and
cerebral perfusion pressure: influence on neurological deterioration and
outcome in severe head injury. The Executive Committee of the
International Selfotel Trial. J Neurosurg 2000;92(1):1–6.
78. Morris GF, Juul N, Marshall SB, et al. Neurological deterioration as a
potential alternative endpoint in human clinical trials of experimental
pharmacological agents for treatment of severe traumatic brain injuries.
Executive Committee of the International Selfotel Trial. Neurosurgery
1998;43(6):1369–72; discussion 72–4.
79. Marshall LF, Bowers-Marshall S, Klauber MR, et al. A new classification of
head injury based on computerized tomography. J Neurosurg 1991;
75(Suppl):S14–20.
80. Hiler M, Czosnyka M, Hutchinson P, et al. Predictive value of initial
computerized tomography scan, intracranial pressure, and state of
autoregulation in patients with traumatic brain injury. J Neurosurg 2006;
104(5):731–7.
81. Servadei F, Antonelli V, Giuliani G, et al. Evolving lesions in traumatic
subarachnoid hemorrhage: prospective study of 110 patients with emphasis
on the role of ICP monitoring. Acta Neurochir Suppl 2002;81:81–2.
82. Timofeev I, Dahyot-Fizelier C, Keong N, et al. Ventriculostomy for control
of raised ICP in acute traumatic brain injury. Acta Neurochir Suppl
2008;102:99–104.
83. Exo J, Kochanek PM, Adelson PD, et al. Intracranial pressure-monitoring
systems in children with traumatic brain injury: combining therapeutic and
diagnostic tools. Pediatr Crit Care Med 2010. Epub ahead of print.
84. Mollman HD, Rockswold GL, Ford SE. A clinical comparison of
subarachnoid catheters to ventriculostomy and subarachnoid bolts: a
prospective study. J Neurosurg 1988;68(5):737–41.
85. Ostrup RC, Luerssen TG, Marshall LF, et al. Continuous monitoring of
intracranial pressure with a miniaturized fiberoptic device. J Neurosurg
1987;67(2):206–9.
86. Piek J, Bock WJ. Continuous monitoring of cerebral tissue pressure in
neurosurgical practice–experiences with 100 patients. Intensive Care Med
1990;16(3):184–8.
87. Sahuquillo J, Poca MA, Arribas M, et al. Interhemispheric supratentorial
intracranial pressure gradients in head-injured patients: are they clinically
important? J Neurosurg 1999;90(1):16–26.
88. Ehtisham A, Taylor S, Bayless L, et al. Placement of external ventricular
drains and intracranial pressure monitors by neurointensivists. Neurocrit
Care 2009; 10:241–7.
89. Harris CH, Smith RS, Helmer SD, et al: Placement of intracranial pressure
monitors by non-neurosurgeons. Am Surg 2002;68(9):787–90.
90. Kaups KL, Parks SN, Morris CL: Intracranial pressure monitor placement
by midlevel practitioners. J Trauma 1998;45(5):884–6.
91. Beer R, Lackner P, Pfausler B, et al. Nosocomial ventriculitis and meningitis
in neurocritical care patients. J Neurol 2008;255:1617–24.
92. Lo CH, Spelman D, Bailey M, et al. External ventricular drain infections are
independent of drain duration: an argument against elective revision.
J Neurosurg 2007;106:378–83.
93. Camacho EF, Boszczowski I, Basso M, et al. Infection rate and risk factors
associated with infections related to external ventricular drain. Infection.
2011;39(1):47–51. Epub 2011, Jan 25.
94. Williams TA, Leslie GD, Dobb GJ, et al. Decrease in proven ventriculitis by
reducing the frequency of cerebrospinal fluid sampling from
extraventricular drains. J Neurosurg 2011. Epub ahead of print.
95. Scheithauer S, Bürgel U, Ryang YM, et al. Prospective surveillance of drain
associated meningitis/ventriculitis in a neurosurgery and neurological
intensive care unit. J Neurol Neurosurg Psychiatry 2009;80(12):1381–5.
Epub 2009, Mar 16.
96. Mayhall CG, Archer NH, Lamb VA, et al. Ventriculostomy-related
infections: a prospective epidemiologic study. N Engl J Med 1984;310(9):
553–9.
97. Aucoin P, Kotilainen H, Gantz N, et al. Intracranial pressure monitors:
epidemiologic study of risk factors and infections. Am J Med
1986;80:369–76.
98. Holloway KL, Barnes T, Choi S, et al. Ventriculostomy infections: the effect
of monitoring duration and catheter exchange in 584 patients. J Neurosurg
1996;85(3):419–24.
99. Lozier AP, Sciacca RR, Romagnoli MF, et al. Ventriculostomy-related
infections: a critical review of the literature. Neurosurgery 2002;51(1):170–
81; discussion 181–2.
100. Winfield JA, Rosenthal P, Kanter RK, et al. Duration of intracranial pressure
monitoring does not predict daily risk of infectious complications.
Neurosurgery 1993;33(3):424–30; discussion 430–1.
101. Zabramski JM, Whiting D, Darouiche RO, et al. Efficacy of antimicrobial-
impregnated external ventricular drain catheters: a prospective,
randomized, controlled trial. J Neurosurg 2003;98(4):725–30.
102. Fichtner J, Guresir E, Seifert V, et al. Efficacy of silver-bearing external
ventricular drainage catheters: a retrospective analysis. J Neurosurg
2010;112:840–6.

103. Muttaiyah S, Ritchie S, John S, et al. Efficacy of antibiotic-
impregnated external ventricular drain catheters. J Clin Neurosci 2010;17:
296–8.
104. Honda H, Jones JC, Craighead MC, et al. Reducing the incidence of
intraventricular catheter-related ventriculitis in the neurology-neurosurgical
intensive care unit at a tertiary care center in St Louis, Missouri: an 8-year
follow-up study. Infect Control Hosp Epidemiol 2010;31(10):1078–81.
105. Leverstein-van Hall MA, Hopmans TE, van der Sprenkel JW, et al. A bundle
approach to reduce the incidence of external ventricular and lumbar
drain-related infections. J Neurosurg. 2010;112(2):345–53.
106. Park P, Garton HJ, Kocan MJ, et al. Risk of infection with prolonged
ventricular catheterization. Neurosurgery 2004;55(3):594–9; discussion
599–601.
107. Wong GK, Poon WS, Wai S, et al. Failure of regular external ventricular
drain exchange to reduce cerebrospinal fluid infection: result of a
randomised controlled trial. J Neurol Neurosurg Psychiatry 2002;73(6):
759–61.
108. Paramore CG, Turner DA. Relative risks of ventriculostomy infection and
morbidity. Acta Neurochirurgica 1994;127(1–2):79–84.
109. Flibotte JJ, Lee KE, Koroshetz WJ, Rosand J, et al. Continuous antibiotic
prophylaxis and cerebral spinal fluid infection in patients with intracranial
pressure monitors. Neurocrit Care 2004;1(1):61–8.
110. Gelabert-Gonzalez M, Ginesta-Galan V, Sernamito-Garcia R, et al. The
Camino intracranial pressure device in clinical practice: assessment in 1000
cases. Acta Neurochir (Wien) 2006;148(4):435–41. Epub 2005, Dec 27.
111. Jensen RL, Hahn YS, Ciro E. Risk factors of intracranial pressure
monitoring in children with fiberoptic devices: a critical review. Surg
Neurol 1997;47(1):16–22.
112. Morton R, Lucas 2nd TH, Ko A, et al. Intracerebral abscess associated with
the Camino intracranial pressure monitor: case report and review of the
literature. Neurosurgery 2011. Epub ahead of print.
113. Narayan RK, Kishore PR, Becker DP, et al. Intracranial pressure: to monitor
or not to monitor? A review of our experience with severe head injury.
J Neurosurg 1982;56(5):650–9.
114. Guyot LL, Dowling C, Diaz FG, et al. Cerebral monitoring devices: analysis
of complications. Acta Neurochir Suppl 1998;71:47–9.
115. North B, Reilly P. Comparison among three methods of intracranial
pressure recording. Neurosurgery 1986;18(6):730–2.
116. Wiesmann M, Mayer TE. Intracranial bleeding rates associated with two
methods of external ventricular drainage. J Clin Neurosci 2001;8(2):126–8.
117. Anderson RC, Kan P, Klimo P, et al. Complications of intracranial pressure
monitoring in children with head trauma. J Neurosurg 2004;101(Suppl. 1):
53–8.
118. Blaha M, Lazar D. Traumatic brain injury and haemorrhagic complications
after intracranial pressure monitoring. J Neurol Neurosurg Psychiatry
2005;76(1):147.
119. Koskinen LO, Olivecrona M. Clinical experience with the intraparenchymal
intracranial pressure monitoring Codman MicroSensor system.
Neurosurgery 2005;56(4):693–8; discussion 698.
120. Citerio G, Piper I, Chambers IR, et al. Multicenter clinical assessment of the
Raumedic Neurovent-P intracranial pressure sensor: a report by the
BrainIT group. Neurosurgery 2008;63:1152–8.
121. Al-Tamimi YZ, Helmy A, Bavetta S, et al. Assessment of zero drift in the
Codman intracranial pressure monitor: a study from 2 neurointensive care
units. Neurosurgery 2009;64:94–8; discussion 98–9.
122. Martinez-Manas RM, Santamarta D, de Campos JM, et al. Camino
intracranial pressure monitor: prospective study of accuracy and
complications. J Neurol Neurosurg Psychiatry 2000;69:82–6.
123. Eide PK, Bakken A. The baseline pressure of intracranial pressure (ICP)
sensors can be altered by electrostatic discharges. Biomed Eng Online
2011;10(1):75. Epub ahead of print.
124. Dahlqvist MB, Andres RH, Raabe A, et al. Brain herniation in a patient
with apparently normal intracranial pressure: a case report. J Med Case Rep
2010;4:297.
125. Marshall LE, Zovickian J, Ostrup R, et al. Multiple simultaneous recordings
of ICP in patients with acute mass lesions. In: Miller JD, Teasdale GM,
Rowan JO, et al, editors. Intracranial pressure VI. Berlin: Springer-Verlag;
1986. p. 184–6.
126. Weaver DD, Winn HR, Jane JA. Differential intracranial pressure in patients
with unilateral mass lesions. J Neurosurg 1982;56(5):660–5.
127. Yano M, Ikeda Y, Kobayashi S, et al. Intracranial pressure in head-injured
patients with various intracranial lesions is identical throughout the
supratentorial intracranial compartment. Neurosurgery 1987;21(5):688–92.
128. Bauer DF, McGwin Jr G, Melton SM, et al. The relationship between INR
and development of hemorrhage with placement of ventriculostomy. J
Trauma 2010. Epub ahead of print.
Section VI—Intracranial Monitoring 347.e3
129. Matevosyan K, Madden C, Barnett SL, et al. Coagulation factor levels in
neurosurgical patients with mild prolongation of prothrombin time: effect
on plasma transfusion therapy. J Neurosurg 2010;114:3–7.
130. Davis JW, Davis IC, Bennink LD, et al. Placement of intracranial pressure
monitors: are “normal” coagulation parameters necessary? J Trauma
2004;57(6):1173–7.
131. Le TV, Rumbak MJ, Liu SS, et al. Insertion of intracranial pressure
monitors in fulminant hepatic failure patients: early experience using
recombinant factor VII. Neurosurgery 2010;66(3):455–8; discussion 458.
132. Smielewski P, Lavinio A, Timofeev I, et al. ICM? a flexible platform for
investigations of cerebrospinal dynamics in clinical practice. Acta Neurochir
Suppl 2008;102:145–51.
133. Venkatesh B, Garrett P, Fraenkel DJ, et al. Indices to quantify changes in
intracranial and cerebral perfusion pressure by assessing agreement
between hourly and semi-continuous recordings. Intensive Care Med 2004;
30:510–13.
134. Stein DM, Hu PF, Brenner M, et al. Brief episodes of intracranial
hypertension and cerebral hypoperfusion are associated with poor
functional outcome after severe traumatic brain injury. J Trauma 2011;
71(2):364–74.
135. Bremmer R, de Jong BM, Wagemakers M, et al. The course of intracranial
pressure in traumatic brain injury: relation with outcome and
CT-characteristics. Neurocrit Care 2010;12:362–8.
136. O’Phelan KH, Park D, Efird JT, et al. Patterns of increased intracranial
pressure after severe traumatic brain injury. Neurocrit Care 2009;10(3):
280–6.
137. Stocchetti N, Mattioli C, Paparella A, et al. Bedside assessment of CO2
reactivity in head injury: changes in CBF estimated by changes in ICP and
cerebral extraction of oxygen. J Neurotrauma 1993;10:187.
138. Yoshihara M, Bandoh K, Marmarou A. Cerebrovascular carbon dioxide
reactivity assessed by intracranial pressure dynamics in severely head
injured patients. J Neurosurg 1995;82(3):386–93.
139. Miller JD, Garibi J, Pickard JD. A clinical study of intracranial volume
pressure relationships. Br J Surg 1973;60(4):316.
140. Miller JD, Garibi J, Pickard JD. Induced changes of cerebrospinal fluid
volume: effects during continuous monitoring of ventricular fluid pressure.
Arch Neurol 1973;28(4):265–9.
141. Miller JD. Volume and pressure in the craniospinal axis. Clin Neurosurg
1975;22:76–105.
142. Maset AL, Marmarou A, Ward JD, et al. Pressure-volume index in head
injury. J Neurosurg 1987;67(6):832–40.
143. Luerssen TG, Chesnut RM, Van Berkum-Clark M, et al. Post traumatic
cerebrospinal fluid infections in the Traumatic Coma Data Bank: the
influence of type and management of ICP monitors. In: Avezaat C, van
Eijndhoven JHM, Maas AJR, et al, editors. Intracranial pressure VIII. Berlin:
Springer-Verlag; 1993. p. 157–63.
144. Piper I, Spiegelberg A, Whittle I, et al. A comparative study of the
Spiegelberg compliance device with a manual volume-injection method: a
clinical evaluation in patients with hydrocephalus. Br J Neurosurg 1999;
13(6):581–6.
145. Yau YH, Piper IR, Clutton RE, et al. An experimental evaluation of a new
intraparenchymal continuous compliance probe: preliminary studies. Acta
Neurochir 2002;81:181–2.
146. Adolph RJ, Fukusumi H, Fowler NO. Origin of cerebrospinal fluid
pulsations. Am J Physiol 1967;212(4):840–6.
147. Hamer J, Alberti E, Hoyer S, et al. Influence of systemic and cerebral
vascular factors on the cerebrospinal fluid pulse waves. J Neurosurg 1977;
46(1):36–45.
148. Mitchell P, Kirkness C, Burr R, et al. Intracranial pressure and transcranial
Doppler waveform analysis. In: Marmarou A, Bullock MR, Avezaat CJ,
editors. Intracranial pressure and neuromonitoring in brain injury. New
York: Springer; 1997. p. 420.
149. Contant Jr CF, Robertson CS, Crouch J, et al. Intracranial pressure
waveform indices in transient and refractory intracranial hypertension.
J Neurosci Methods 1995;57(1):15–25.
150. Panerai RB. Assessment of cerebral pressure autoregulation in humans:
a review of measurement methods. Physiol Meas 1998;19(3):305–38.
151. Robertson CS, Narayan RK, Contant CF, et al. Clinical experience with a
continuous monitor of intracranial compliance. J Neurosurg 1989;71
(5 Pt 1):673–80.
152. Wachi A, Marmarou A, Sato K. [Test of non-invasive PVI measure with a
fiber-optic pressure monitoring device in head injured patients]. No To
Shinkei 1990;42(6):561–8.
153. Hu X, Xu P, Asgari S, et al. Forecasting ICP elevation based on prescient
changes of intracranial pressure waveform morphology. IEEE Trans Biomed
Eng 2010;57(5):1070–8.
347.e4 Section VI—Intracranial Monitoring
154. Hamilton R, Xu P, Asgari S, et al. Forecasting intracranial pressure elevation
using pulse waveform morphology. Conf Proc IEEE Eng Med Biol Soc
2009;2009:4331–4.
155. Steiner LA, Czosnyka M, Piechnik SK, et al. Continuous monitoring of
cerebrovascular pressure reactivity allows determination of optimal cerebral
perfusion pressure in patients with traumatic brain injury. Crit Care Med
2002;30(4):733–8.
156. Czosnyka M, Guazzo E, Whitehouse M, et al. Significance of intracranial
pressure waveform analysis after head injury. Acta Neurochir 1996;138:531–
41; discussion 41–2.
157. Balestreri M, Czosnyka M, Steiner LA, et al. Association between outcome,
cerebral pressure reactivity and slow ICP waves following head injury. Acta
Neurochir Suppl 2005;95:25–8.
158. Maset AL, Marmarou A, Ward JD, et al. Pressure-volume index in head
injury. J Neurosurg 1987;67:832–40.
159. Padayachy LC, Figaji AA, Bullock MR. Intracranial pressure monitoring for
traumatic brain injury in the modern era. Childs Nerv Syst
2010;26(4):441–52.
VI
Chapter
35  
Brain Oxygen
Mauro Oddo and Peter D. le Roux
Introduction
Maintenance of an adequate tissue oxygenation is a funda­
mental objective in the field of critical care medicine in
general, and the assessment of tissue oxygenation is an essen­
tial part of the management of the critically ill patient. This
provides information about oxygen supply and use in specific
tissue beds.1 Hypoxia is defined as a reduction of tissue oxy­
genation to levels that are insufficient to maintain cellular
function and metabolism. The pathophysiology of tissue
hypoxia can be classified as follows:
Ischemic
Cytopathic
Anemic
Hypoxic
In the brain tissue, ischemic hypoxia may result from macro­
vascular ischemia (i.e., reduced or absent cerebral blood flow
[CBF], e.g., cerebral thrombosis, vasospasm, low arterial
carbon dioxide tension [PaCO2], alkalosis) or from micro­
vascular ischemia (e.g., blood-brain barrier disruption and
brain capillary endothelial dysfunction).2 Cytopathic hypoxia
in the brain includes a complex cascade of events after
the initial brain insult (e.g., glutamate excitotoxicity, intracel­
lular calcium influx, free radical and inflammatory cytokine
release) that lead to cell energy and mitochondrial dysfunc­
tion,3,4 diffusion limited oxygen delivery to the cell,5,6 and
reduced oxygen extraction.7 Anemia8,9 and impaired systemic
oxygenation10 are additional causes of brain tissue hypoxia
(Table 35.1).
Brain hypoxia can cause secondary brain damage. Thus
monitoring of brain tissue oxygen partial pressure to dis­
criminate between normal and critically impaired tissue
oxygenation has been integrated into the management of
patients with acute brain injury, and several lines of evidence
suggest that it is a clinically useful modality in neurocritical
care.11-22 In addition, several observational studies show that
cerebral perfusion pressure (CPP) and intracranial pressure
(ICP) are not surrogates for brain oxygen and that brain
oxygenation varies independently of cerebral hemodynamics
and ICP.19,23,24 Like other monitors, a brain oxygen monitor
is not intended to be used alone; instead it should be used
as a supplement or complement to other monitors, for
example, an ICP monitor (see Chapter 34) or microdialysis
(see Chapter 36).
348
Definition
Brain tissue oxygen is defined as the partial pressure of oxygen
in the interstitial space of the brain and reflects the availability
of oxygen for oxidative energy production. There has been
some debate as to whether the monitor systems measure tissue
oxygen pressure or tension, and, accordingly, several abbrevia­
tions have been used for brain tissue oxygen, including PbO2,
PbrO2, PbtO2, PtiO2, and BTO2. A consensus conference at the
13th International Symposium on Intracranial Pressure and
Brain Monitoring held July 2007 in San Francisco, California,
proposed that PbtO2 be used as the standard abbreviation.
Accordingly, this chapter uses the PbtO2 when referring to
brain tissue oxygen pressure. Consistent with this abbreviation
clinical studies suggest that PbtO2 may be best defined by the
equation:
PbtO2 = CBF × AVTO2
where AVTO2 is arterial oxygen tension (PaO2) minus oxygen
partial pressure in venous blood (PvO2)—that is, PbtO2
represents the interaction between plasma oxygen tension
and CBF.5
Technology
There are four methods to measure brain oxygen: jugular
venous bulb oximetry (Chapter 32), direct brain tissue oxygen
tension measurement, near infrared spectroscopy (Chapter
33), and oxygen-15 positron emission tomography (PET)
(Chapter 29). This chapter discusses direct brain tissue oxygen
tension measurement, which is now the most common tech­
nique used in the neurocritical care unit (NCCU) to assess
cerebral oxygenation. The technique involves the insertion of
a fine catheter (approximately 0.5 mm in diameter) into the
brain parenchyma, specifically the white matter, to continu­
ously measure PbtO2. The PbtO2 catheters or probes can be
inserted through a single- or multiple-lumen bolt secured in
a single burr hole or they may be tunneled under the scalp.
The procedure to insert the monitor may be performed in the
operating room or at the bedside in the NCCU. PbtO2 probes
generally are placed adjacent to an ICP monitor and through
the same triple-lumen bolt. Ideally the position and function
of the monitor should be confirmed with a head computed
tomography (CT) scan or oxygen challenge test particularly if
© Copyright 2013 Elsevier Inc. All rights reserved.
 Section VI—Intracranial Monitoring 349

e−
Table 35.1  Causes of Brain Tissue Hypoxia
Etiology Pathophysiology
Current measurement
Ischemic
Macrovascular ↓ O2 delivery, ↓ CBF (thrombosis,
vasospasm, low PaCO2) + −
Microvascular BBB disruption, vasogenic edema, 700 mV polarization
endothelial dysfunction
Cytopathic ↓ O2 extraction (cytotoxic edema,
diffusion limited O2 delivery,
mitochondrial dysfunction) 1
3
Anemic ↓ Hemoglobin concentration
Hypoxic ↓ PaO2, SaO2

BBB, Blood-brain barrier; CBF, cerebral blood flow; PaCO2, arterial carbon
dioxide tension; PaO2, arterial oxygen tension; SaO2, arterial oxygen 5
saturation. 2 4

O2

the initial PbtO2 reading after 30 to 60 minutes of stabilization

is abnormal. A
Two primary technologies underlie PbtO2 measurement:
one is based on the Clark principle, the other on an optical
technique. The Clark principle uses the electrochemical prop­
erties of noble metals to measure the oxygen content of tissue.1
The Clark electrode consists of a membrane that covers a layer
of electrolyte and two metallic electrodes. Oxygen diffuses
through the membrane and is electrochemically reduced at
the cathode. The greater the oxygen partial pressure, the more
oxygen diffuses through the membrane. The change in voltage
between the reference electrode and the measuring electrode
is proportional to the amount of oxygen molecules being
reduced on the cathode. The Licox PbtO2 probe (Integra Life­
sciences, Plainsboro, NJ) uses such a closed polarographic
Clark-type cell with reversible electrochemical electrodes (Fig.
35.1). This oxygen-consuming process is temperature depen­
dent, and so requires constant calibration of the system to
patient temperature. The Licox system includes a brain tem­
perature probe (thermocouple) that is inserted through a
triple-lumen bolt with the PbtO2 and ICP monitor (e.g.,
Camino, Integra Lifesciences). This temperature probe mea­
sures brain temperature and allows for automatic calibration.
The new Licox PMO probe can measure PbtO2 and brain tem­
perature using a single probe.25 The Neurovent-P Temp (Rau­ B
medic AG, Munchberg, Germany), which uses the same
polarographic technique as the Licox, also can measure PbtO2 Fig. 35.1  A, Schematic diagram of the Licox polarographic oxygenation
and brain temperature in one catheter. There are important probe. The numbered components of the diagram are (1) polyethylene
tube diffusion membrane; (2) polarographic gold cathode; (3) polaro-
differences in values obtained by the Neurovent and Licox
graphic silver anode; (4) cell filled with electrolyte; and (5) cerebral tissue.
devices and so the values obtained by these devices cannot be B, Schematic diagram of the Licox probe that illustrates placement
used interchangeably.26-28 Purpose-designed triple-lumen through a cranial bolt into the cerebral tissues. Placement is similar to
cranial access devices29 allow simultaneous PbtO2, ICP, and an intracranial pressure (ICP) monitor and is frequently used through 
cerebral microdialysis monitoring. However, if brain tempera­ the same bolt. (Adapted from Mulvey JM, Dorsch M, Mudaliar Y, et al. Multimo-
ture is not measured (e.g., a microdialysis catheter is inserted dality monitoring in severe traumatic brain injury: the role of brain tissue oxygen-
through the lumen instead of the temperature probe), the ation monitoring. Neurocrit Care 2004;1(3):391–402.)
monitor should be calibrated manually on a regular basis, for
example, every 30 minutes by using core body temperature.
The Licox system averages oxygen over a probe area of 14 to are used to measure concentrations of substances by photo­
18 mm2 and has excellent long-term stability, even after 7 days. chemical reactions that create changes in optical properties of
The Neurovent-P Temp has a greater surface area of 24 mm2. indicator compounds.1 OxyLabPO2 (Oxford Optronix Ltd.,
The second technique used to measure PbtO2 is based on a Oxford, UK) allows PbtO2 to be measured using optical fluo­
fluorescence quenching technique in which a marker changes rescence technology. Unlike the Licox, this process does not
color according to the ambient amount of gas. Optode sensors consume oxygen and does not affect the measured oxygen
350 Section VI—Intracranial Monitoring

level. However, the probe measures a smaller area. The
Neurotrend (Codman, Johnson & Johnson, Raynham, MA)
uses this technique, but it is no longer commercially available
for clinical use. The accuracy and clinical stability of the
Neurotrend sensor also appears to be less than the Licox
system.30 There are several other important differences
between techniques based on the Clark principle (e.g., the
Licox) and on an optical technique (e.g., Neurotrend). First,
Licox catheters are precalibrated and can be inserted without
any pre-use calibration. However, postinsertion stabilization
(about 1 hour) is required before readings are reliable. By
contrast, the Neurotrend monitor needs bedside calibration to
a defined oxygen concentration. Second, the catheters are of
different lengths; the Neurotrend is inserted at a greater depth
than the Licox catheter. Third, the critical PbtO2 threshold for
hypoxia is different, so it is difficult to compare studies that
use the different techniques. In addition, the Neurotrend
changed design in 1998, making it difficult to compare between
old and more recent studies that describe this technology.31
Catheter Placement
PbtO2 catheters are placed in the right frontal lobe white matter
in diffuse brain injury or on the side of maximal pathology
when there is focal injury. In subarachnoid hemorrhage (SAH)
patients, the authors place the PbtO2 monitor where maximal
vasospasm is likely to develop based on the distribution of
SAH and aneurysm location. Recommendations for the pre­
ferred brain area to be monitored with a PbtO2 probe are sum­
marized in Table 35.2.
In general, monitors are placed when the Glasgow Coma
Scale (GCS) score is equal to or less than 8 (i.e., consistent
with the indications for an ICP monitor) (see Chapter 34).
There are no specific guidelines on how long PbtO2 should be
monitored. In traumatic brain injury (TBI) patients the
authors leave the monitor in place until ICP is normal for 24
hours without any specific treatment other than sedation for
ventilation. On average, in patients with severe TBI this is 4 to
5 days. In SAH patients the monitor is kept in place during
the risk period for vasospasm. In addition, continuous PbtO2
can be monitored twice as long as jugular venous oxygen satu­
ration (SjvO2) and without needing recalibration. In studies
that compare PbtO2 and SjvO2, good quality data are acquired
95% of the time with PbtO2, but only 43% of the time for
SjvO2.32 Postinsertion noncontrast head CT confirmation of
probe position in the brain parenchyma (Fig. 35.2)
is important to interpret readings.33,34 In some patients a
CT-perfusion study also may be useful. For example, if PbtO2
readings are consistently low or poorly responsive to therapy,
it is useful to know the monitor’s proximity to a focal abnor­
mality or whether the region it is in is hypoperfused. In these
cases the PbtO2 threshold for treatment may be lower than
normally used. A “run-in” or equilibration time of up to 1
hour is required before readings are stable. Adjustment of
insertion depth (when used through an access device) is not
possible with the Licox system, and if the monitor is removed,
the insertion bolt then should be replaced if a new monitor is
inserted to avoid potential contamination.
Transiently increasing the inspiratory oxygen fraction
(FiO2) and observing the corresponding PbtO2 increase can
help confirm function or exclude the presence of surrounding
microhemorrhages or sensor damage at insertion. An oxygen
challenge is therefore performed when monitoring starts and,
in some institutions, daily thereafter as part of regular clinical
care to evaluate the function and responsiveness of the PbtO2
probe or to determine “oxygen reactivity.” The oxygen chal­
lenge involves increasing the FiO2 from baseline to 1.0 for
approximately 5 minutes. This typically leads to a several-fold
increase in the amount of oxygen dissolved in the plasma,
translating into a mean threefold increase in PbtO2.5,35 However,
the response to hyperoxia usually is less robust when the
monitor is in an underperfused (CBF is <20 mL/100 g/min)
region.36
Regional Versus Global
Measurement
PbtO2 probes sample approximately 15 mm2 of tissue around
the tip, and the PbtO2 value depends on O2 diffusion from the
vasculature to a small amount of tissue.5,6 Therefore it is a

Table 35.2  Recommended Areas for Brain

Oxygen Probe Insertion
Intracranial Pathology Catheter Location
Traumatic brain injury
Diffuse injury Right frontal lobe
Focal injury (subdural Pericontusional tissue
hematoma, contusion)
Subarachnoid hemorrhage Expected distribution area of the
parent artery of the aneurysm,
at highest risk for developing
symptomatic vasospasm and
delayed ischemia
Fig. 35.2  A computed tomography (CT) head scan that demonstrates
Cerebral infarction Area of lesion, at distance from
the infarcted tissue the position of a Licox probe in the white matter of the right frontal
cortex.

regional monitor with values that depend on probe location.
This has led to debate about probe location and whether the
value can be used to make decisions about global oxygenation.
In most patients PbtO2 is measured in “normal appearing”
frontal subcortical white matter; there is evidence that this
local measurement can be regarded as an indicator of global
oxygenation.31,32,37-39 Consistent with this, two clinical studies
have shown good correlation between PbtO2 and SjvO2, used
to assess global brain oxygenation, in areas without focal
pathology after TBI.32,37 In areas with focal pathology, this
correlation between PbtO2 and SjvO2 was absent, and the brain
tissue oximetry values reflected regional brain oxygenation
better than jugular bulb oximetry.37 Other studies also show
that when the probe is in tissue immediately adjacent to a
contusion or other pathology (e.g., a subdural hematoma),
values are often lower, even if CPP is higher.33,34,40 In addition,
when the monitor is placed adjacent to abnormal brain,
regional hypoxia lasts longer than in normal-appearing tissue33
and the relationship with outcome appears to be more robust.34
Values in Normal
and Pathologic Conditions
Normal Values
The Licox monitor provides a measure of PbtO2 in units
of tension (mm Hg). Usually oxygen content is expressed in
units of concentration (mL O2/100 mL) and oxygen delivery
to the brain and cerebral metabolic rate of oxygen (CMRO2)
are expressed in mL O2/100 g brain/minute. The conversion
factor: 1 mm Hg equals 0.003 mL O2/100 g brain can be used
to compare PbtO2 with other oxygen concentration measure­
ments. Normal PbtO2 measurements have been acquired
experimentally, but human measurements have been restricted
to “normal” values obtained during neurosurgical procedures
and in normal-appearing brain after TBI. In animals, normal
PbtO2 varies from 30 to 40 mm Hg.41,42 In humans without
severe brain injury, PbtO2 ranges from 37 to 48 mm Hg.39,43-45
Mean values of 23 mm Hg have been measured in awake
subjects undergoing functional neurosurgery.46 PbtO2 values in
patients with normal ICP are between 25 and 30 mm Hg in
TBI.32,47
Hypoxic Thresholds
The identification of hypoxic brain tissue permits interven­
tion potentially before irreversible injury occurs; that is, the
therapeutic window can be widened. In addition, threshold
values provide therapeutic endpoints. However, in managing
a patient, trends over time and how the PbtO2 value relates to
other variables such as ICP and CPP, also should be consid­
ered. In addition, threshold is not the only factor that is
important in terms of outcome; the duration spent below
threshold and the depth or severity of brain tissue hypoxia
also are important.48-50 Indeed the duration of PbtO2 less than
10 mm Hg is an independent factor associated with poor
outcome after severe TBI, and this association also is indepen­
dent of ICP.17 Threshold values can vary with probe type and
probe site, and so values obtained with different makes of
monitors (e.g., Licox vs. Neurovent vs. Neurotrend) are not
interchangeable.27,31,34 Various PbtO2 thresholds have been pro­
posed.6,33,48,49,51-56 A PbtO2 less than 20 mm Hg indicates
Section VI—Intracranial Monitoring 351
Table 35.3  Brain Oxygen Values in Humans*
Condition PbtO2 Values
Normal 25-50 mm Hg
Hypoxic thresholds
Moderate brain hypoxia 15-25 mm Hg
Critical brain hypoxia <15 mm Hg*
Severe brain hypoxia <10 mm Hg
Elevated >50 mm Hg
PbtO2, Brain tissue oxygen pressure.
*PbtO2 thresholds according to Brain Trauma Foundation guidelines. In a
National Institutes of Health–sponsored phase II trial that is expected to finish
patient recrurtment in 2013 to examine PbtO2, treatment is initiated when
PbtO2 is <20 mm Hg.
compromised brain tissue oxygen and often is taken as a
threshold at which to consider or initiate therapy to correct
brain oxygen.9,33 This value is the threshold at which therapy
is initiated in an National Institutes of Health (NIH)–
sponsored phase II trial that is expected to finish patient
recruitment in late 2013. The 2007 guidelines for severe TBI
recommend that PbtO2 values less than 15 mm Hg be consid­
ered as the critical threshold for “ischemia.”57 Microdialysis
studies show that at this level other markers of ischemia are
elevated.13 Normal mitochondria require an oxygen level of
about 1.5 mm Hg to function; this corresponds with a PbtO2
level between 15 and 20 mm Hg in normal white matter.31 A
PbtO2 less than 10 mm Hg is a marker of severe brain tissue
hypoxia and is an independent factor associated with both
mortality and unfavorable outcome.14,17,49 Values of 0 mm Hg
that persist longer than 30 minutes and show no response to
an oxygen challenge are consistent with brain death.58,59 Table
35.3 summarizes normal and hypoxic thresholds of PbtO2.
Observational Data
A large body of observational clinical data on PbtO2, primarily
in patients with severe TBI and less frequently SAH, has
accrued since the 1980s. These findings may be summarized
as follows. First, PbtO2 may be abnormal, despite a normal ICP
and CPP both in adults and children19,21,23,24,46,60 and when
SjvO2 is normal.37 This suggests that PbtO2 may be a novel
target for resuscitation following brain injury and can supple­
ment the information provided by an ICP monitor or
retrograde jugular catheter.46 Second, compromised PbtO2 is
common after acute brain injury even after adequate resusci­
tation, defined as normal ICP and CPP, and an episode of
reduced PbtO2 may complicate the ICU course of up to 70%
of patients, again often independently of ICP and CPP.19,23,24,61
Third, imaging studies obtained at admission after TBI do
not reliably predict whether brain hypoxia will develop.62
Fourth, reduced PbtO2 is associated with other markers of
cellular distress such as an elevated lactate-to-pyruvate
ratio on microdialysis that can be corrected with a variety
of therapies.12,63 Common therapies include head position,
venti­lator manipulation, CPP augmentation, and sedation
that overall are successful in correcting the abnormality in
about 70% of episodes of reduced PbtO2.10,54,61,64-71 Fifth, PbtO2
values may help guide therapeutic interventions (e.g., blood
352 Section VI—Intracranial Monitoring
transfusion),9,72-75 osmotherapy,76-79 and decompressive crani­
ectomy (DC),80-84 or identify deleterious effects of other treat­
ments such as vasoconstriction with hyperventilation or
shivering with induced hypothermia.69,70,85,86 In addition,
PbtO2 monitors may identify patients at risk for cerebral com­
promise associated with hospital transport.87 Sixth, PbtO2-
pressure reactivity may be used to assess cerebral autoregulation
and help target CPP in individual patients.88-91 Seventh, PbtO2
has been used for early detection of delayed cerebral ischemia
(DCI) in SAH patients and to evaluate the effects of various
therapies for DCI, temporary artery occlusion during surgery,
angiography, or pharmacologic angioplasty.76,92-96 In addition,
studies in SAH have shown that in some patients nimodipine
or intra-arterial papaverine can have unexpected adverse
effects on PbtO2.97,98 Finally, several observational studies dem­
onstrate an independent association between low PbtO2 and
poor outcome in adults as well as children with TBI and in
SAH.14,15,17,21,48,49,56,99-103 Together these various observations
suggest that care based on information provided by a PbtO2
monitor may be a reasonable strategy in severe acute brain
injury. Consistent with this several observational studies or
based on historical controls, but not all, suggest that the com­
bination of PbtO2- and ICP-based care is associated with better
outcome than just ICP-based care alone.18,51,104-107 In part this
potential outcome benefit may be associated with better-
targeted care in each patient, determination of an “optimal”
CPP, and avoidance of inappropriate ICP treatment and
potential side effects if ICP is slightly elevated (≈25 mm Hg)
but PbtO2 is normal, a concept known as permissive intracra-
nial hypertension. At the time this book was published a mul­
ticenter phase II trial was underway to further evaluate the
role of PbtO2-based care in severe TBI.
Interpretation of Measured Values
Information from a PbtO2 monitor, like other monitors, should
be interpreted with other data such as clinical examination,
CT scan findings, ICP, CPP, pulmonary status, and hemoglo­
bin among other variables. In addition, the PbtO2 value needs
to be temperature corrected. This is performed automatically
when a brain temperature probe is placed with a Licox; if a
temperature probe is not inserted, systemic temperature can
be manually entered into the device. It also is essential to know
exactly where the monitor is placed (see earlier text). In the
brain the sensor reads an integration of all small vessels in the
area of the probe, and so the PbtO2 value depends on the rela­
tive predominance of veins or arteries, and the number, diam­
eter, and spatial distribution of the local microvasculature and
so can be influenced by changes in cerebral capillary perfu­
sion.108 Because the venous fraction within cortical microvas­
culature is 70%, it has been suggested that PbtO2 predominantly
reflects venous PO2.109 However, exactly what PbtO2 measures
is still to be fully elucidated but it likely represents (1) the
balance between regional oxygen delivery and cellular oxygen
consumption, (2) oxygen diffusion rather than total oxygen
delivery or cerebral oxygen metabolism, or (3) oxygen that
accumulates in brain tissue because PET studies suggest it may
correlate inversely with oxygen extraction fraction
(OEF).5,54,109,110
Many factors can affect PbtO2. Among these the effect of
CPP and CBF has been the most studied.52,54,66 PbtO2 is associ­
ated with regional CBF.109,111-114 This relationship follows the
autoregulation curve that regulates CBF across a wide range
of mean arterial pressures (MAPs).115 CPP and MAP augmen­
tation can increase CBF and therefore PbtO2,76,54,66,114 whereas
the PbtO2 response to an oxygen challenge is reduced in areas
with low CBF.116 Together this suggests that PbtO2 can provide
information about CBF and impending cerebral ischemia in
certain conditions. However, PbtO2 is more than a marker
of ischemia and PET, and microdialysis studies suggest that
hypoxia in the brain can be independent of CPP and more
related to diffusion rather than perfusion abnormalities.4-6,117
The total amount of oxygen that diffuses across the blood-
brain barrier per unit time is defined as follows:
CBF × (CaO2 − CVO2 )[1] or CBF × [% SaO2 × (1.34) ×
hemoglobin + 0.003 × (PaO2 )] − [% SvO2 ×
(1.34) × hemoglobin + 0.003 × (PvO2 )] 5,118
The very close association between the product of CBF and
arteriovenous oxygen tension difference and PbtO2 implies a
relationship between the amount of dissolved plasma oxygen
that passes through a given volume of brain per unit time and
the steady-state oxygen concentration in brain tissue. Rosen­
thal et al., who used regional CBF and PbtO2 monitoring,
showed that PbtO2 may more appropriately reflect the product
of CBF and arteriovenous oxygen tension difference (CBF ×
AVTO2). This suggests that besides CBF, PaO2 is also an impor­
tant determinant of PbtO210,63,119 and that PbtO2 is not an isch­
emia monitor per se.
Unlike a SjvO2 monitor, which reflects the venous oxygen
content in blood that exits the brain and so indicates the
balance between oxygen delivery and oxygen use, PbtO2 is
more a measurement of the oxygen that accumulates in brain
tissue. Based on the formula PbtO2 = CBF × AVTO2, reduced
PbtO2 can occur when CBF is reduced (ischemic hypoxia) and
also when there is normal CBF, for example, with impaired O2
extraction due to increased gradients for oxygen diffusion,6
cell energy crisis,4,120 or mitochondrial dysfunction121 (cyto­
pathic hypoxia). Hence PbtO2 may be considered a marker of
“cellular function” rather than just simply an “ischemia
monitor,” which makes it an appropriate target for therapy in
NCCU patients.
Safety
The Licox system is safe. Overall the incidence of device-
associated contusion is less than 2%. Most of these are of no
clinical consequence and instead are only identified on
follow-up CT.48,49,56,122-128 This incidence is similar to that
observed with intraparenchymal ICP monitors and is less than
that associated with external ventricular drains. Catheter-
related infections have not been reported.48,49,56,123-127 Technical
complications such as device displacement or malfunction are
more common and may complicate up to 10% of device inser­
tions. In addition, device removal (all three monitors and
bolt) is suggested when a magnetic resonance imaging (MRI)
scan is obtained. Over time gliosis around the probe may
affect the reading. However, most studies suggest the measure­
ments remain accurate with little if any drift. Display errors
can occur, but the maximal probe display error is reported to
be 1.07 plus or minus 2.14%, tested at temperatures between
22° C and 37° C and at oxygen pressures of 0, 44, and 150 mm
Hg.122,125 This small error is unlikely to be relevant to daily
clinical practice.

Clinical Applications
Indications for Use
PbtO2 monitoring has been used most frequently in severe TBI
and poor-grade SAH. There are reports, however, of monitor­
ing use in a variety of other conditions including in brain
tumors, intracerebral hemorrhage stroke, cerebral edema
associated with metabolic abnormalities, and meningitis,
among others. Like ICP monitoring there is no Class I evi­
dence on which to base recommendation about use, and the
evidence is derived mainly from retrospective case-control
and prospective observational studies (Class III evidence).
Based on these studies, PbtO2 monitoring is recommended in
patients with TBI and a GCS score less than 9, abnormal head
CT scan, polytrauma, and hemodynamic instability (i.e., when
ICP monitoring is indicated). In patients with SAH, PbtO2
monitoring is recommended in those with a GCS score less
than 9 and in those at high risk for DCI (e.g., a patient with
thick SAH and intraventricular hemorrhage on admission
head CT scan). PbtO2 monitoring also may be considered in
patients with large middle cerebral artery stroke who are at
high risk for malignant brain edema. Table 35.4 summarizes
some of the important applications of PbtO2 monitoring in
neurocritical care.
Other Monitors
PbtO2 monitoring is best used with other monitors, in par­
ticular ICP and in some patients SjvO2 and a postinsertion
head CT scan. In addition, PbtO2 data supplement that derived
from cerebral microdialysis, CBF velocity measured by tran­
scranial Doppler ultrasound or regional cerebral blood flow
measurements, near infrared spectroscopy, and intracranial
temperature.31,50,126
Intensive Care Unit Management
Observational data suggest PbtO2 monitoring can provide
valuable insight into the intensive care unit (ICU) care of
Table 35.4  Some Clinical Applications of
PbtO2 Monitoring in Neurocritical Care
1. CPP management
a. MAP target
b. Cerebral autoregulation and individual CPP target
c. Induced hypertension and triple H therapy
2. ICP control
a. Choice of osmotic agent (mannitol vs. hypertonic saline)
b. Timing for decompressive craniectomy
3. Hemoglobin threshold for blood transfusion
a. Treatment of anemia in patients with impaired
cerebrovascular reserve (poor-grade SAH)
4. Management of mechanical ventilation
a. PaO2/FiO2 ratio, PEEP
b. Optimal PaCO2 target
CPP, Cerebral perfusion pressure; FiO2, inspiratory oxygen fraction;
ICP, intracranial pressure; MAP, mean arterial pressure; SAH, subarachnoid
hemorrhage; PaCO2, arterial carbon dioxide tension; PaO2, arterial oxygen
tension; PbtO2, brain tissue oxygen pressure; PEEP, positive end-expiratory
pressure.
Section VI—Intracranial Monitoring 353
patients with severe TBI, SAH, and stroke (e.g., autoregula­
tion)88-90 that may help identify an individual CPP target, and
potentially guide several therapies including (1) CPP54,67,68,92,96;
(2) induced hypertension66; (3) osmotherapy76-79,129; (4)
DC80-84; (5) hyperventilation69,70,86,115,130; (6) normobaric hyper­
oxia (although its role in resuscitation remains controver­
sial)5,36,63,71,119,131-134; (7) blood transfusion, particularly in
patients with impaired cerebrovascular reserve8,135-138 since the
transfusion threshold used in general critical care may not
apply to these patients,9,72-74 and so rather than use a hemo­
globin (Hgb) threshold for transfusion, blood is administered
when oxygen delivery is compromised139-142; (8) fluid balance143;
(9) titration of sedatives including use of propofol or barbi­
turates for burst suppression or ICP control144-146; and
(10) induced normothermia.147
Subarachnoid Hemorrhage
Continuous PbtO2 monitoring can help detect DCI in SAH.
This can supplement information obtained by once daily tran­
scranial Doppler.90,148 In addition PbtO2 data can be used to
guide management of DCI and evaluate autoregulation. In
particular, observational clinical studies suggest that induced
hypertension alone, but not hypervolemia and hemodilution
(which have the opposite effect) improves PbtO2,92,93 suggesting
there is a limited if any role for “triple H” prophylaxis after
SAH. PbtO2 monitoring also has been used during cerebral
angiography and pharmacologic angioplasty.98,124
Use During Surgery
Use of a PbtO2 monitor during aneurysm surgery is well
described.94,95,149-151 A correctly positioned PbtO2 monitor
allows the effects of temporary arterial occlusion to be exam­
ined; reduced PbtO2 and especially brain hypoxia indicate low
CBF and are associated with cerebral infarction.149,152 PbtO2
measurements also have been used during surgery to examine
oxygenation of cerebral tissue during arteriovenous malfor­
mation (AVM)150,153 or tumor surgery.154,155 Reduced PbtO2
before AVM resection suggests low perfusion and chronic
hypoxia, whereas a marked PbtO2 increase after AVM removal
indicates hyperperfusion. The effects of inhalational agents156,157
and propofol144,158 on cerebral autoregulation and oxygenation
during anesthesia have been examined using PbtO2 monitor­
ing. These studies show a dose-dependent loss of autoregula­
tion but a corresponding increase in PbtO2 provided CPP is
maintained with inhalational agents, but not with propofol.
PbtO2 Reactivity
The increase in PbtO2 relative to an increase in arterial PO2 is
termed brain tissue oxygen reactivity. It is believed that this
reactivity is controlled by an oxygen regulatory mechanism,
and that this mechanism may be disturbed after brain injury.
van Santbrink et al. examined the brain tissue oxygen response,
that is, the change in PbtO2 in response to changes in PaO2 and
showed that a greater response in the first 24-hours post injury
was independently associated with an unfavorable outcome.159
The effect of PaCO2 on this mechanism has been studied in
dogs160 and in pigs115; PbtO2 shows a linear correlation with
carbon dioxide (CO2) and MAP. A linear correlation with CBF
during CO2 reactivity testing also was found. This suggests
354 Section VI—Intracranial Monitoring
that PbtO2 is influenced by factors that regulate CBF, namely
CO2 and MAP.
Oxygen Reactivity Index
Soehle et al.161 introduced the concept of PbtO2 autoregulation,
defined as the ability of the brain to maintain PbtO2 despite
CPP changes. This may facilitate identification of appropriate
individual CPP targets. Additional studies by Lang et al.
showed a significant correlation between static cerebral auto­
regulation (determined using blood flow velocity in relation
to changing CPP) and cerebral tissue oxygen reactivity (the
rate of PbtO2 change relative to changing CPP).91,102 This sug­
gests a close link between regulation of CBF and oxygenation.
Impaired brain tissue oxygen pressure reactivity index is asso­
ciated with worse outcome after TBI,89 SAH,90 and stroke.88
These findings suggest that manipulation of PbtO2 by altering
PaO2 or CPP can be used to optimize CPP management. In
patients with acute ischemic stroke, impaired CPP-brain tissue
oxygen reactivity index predicts the development of malig­
nant edema after middle cerebral artery infarction.88
Outcome Prediction
The association between reduced PbtO2 or brain hypoxia,
including the duration and the depth or severity of brain
tissue hypoxia, and poor outcome in TBI has been well
described in several observational studies, although the
strength of this relationship may depend in part on probe
location relative to other pathology.* This includes an associa­
tion between low PbtO2 and long-term neuropsychological
performance.162 Importantly, episodes of brain tissue hypoxia
may occur when CPP and ICP are normal,23,50,60,103 thus
emphasizing the value of using both monitors. Reduced PbtO2
also is associated with outcome in SAH, although the relation­
ship is less robust than in TBI.148,163 This association between
outcome and PbtO2 has led to the concept of PbtO2-based care.
Some but not all observational series suggest that the addition
of PbtO2-based care to conventional ICP- and CPP-based care
is associated with improved outcome after severe TBI.18,51,104-
107,164,165
At publication, this question was being evaluated in a
multicenter phase II clinical trial.
Conclusion
Monitoring of brain tissue oxygen partial pressure is a safe and
reliable technique that allows bedside evaluation of cellular
function in patients with severe brain injury. PbtO2 depends
on CBF, arteriovenous oxygen tension difference, Hgb concen­
tration, and systemic oxygenation among other variables. Low
PbtO2 is frequently observed after acute brain injury and can
result from several pathologic mechanisms including isch­
emia, impaired cell oxygen extraction, anemia, and reduced
systemic oxygenation, among others. PbtO2 monitoring helps
optimize CPP, PaCO2, PaO2, and hemoglobin targets, and
guide management of elevated ICP. This can help avoid delete­
rious side effects of treatments and reduce the extent of brain
tissue hypoxia, a variable that is associated with poor outcome.
As such, PbtO2 monitoring can complement the use of existing
intracranial monitors and help target and individualize
*References 14, 17, 21, 34, 48, 49, 56, 99, 101, 103, 130.
therapy for NCCU patients. Future studies are needed to
examine whether PbtO2-targeted therapy may improve the
outcome of patients with severe brain injury.
References
1. Siegemund M, van Bommel J, Ince C. Assessment of regional tissue
oxygenation. Intensive Care Med 1999;25:1044–60.
2. Coles JP. Regional ischemia after head injury. Curr Opin Crit Care 2004;10:
120–5.
3. Marmarou A, Signoretti S, Fatouros P, et al. Mitochondrial injury measured
by proton magnetic resonance spectroscopy in severe head trauma patients.
Acta Neurochir Suppl 2005;95:149–51.
4. Vespa P, Bergsneider M, Hattori N, et al. Metabolic crisis without brain
ischemia is common after traumatic brain injury: a combined microdialysis
and positron emission tomography study. J Cereb Blood Flow Metab 2005;
25:763–74.
5. Rosenthal G, Hemphill JC, 3rd, Sorani M, et al. Brain tissue oxygen tension
is more indicative of oxygen diffusion than oxygen delivery and metabolism
in patients with traumatic brain injury. Crit Care Med 2008;36:1917–24.
6. Menon DK, Coles JP, Gupta AK, et al. Diffusion limited oxygen delivery
following head injury. Crit Care Med 2004;32:1384–90.
7. Coles JP, Cunningham AS, Salvador R, et al. Early metabolic characteristics
of lesion and nonlesion tissue after head injury. J Cereb Blood Flow Metab
2009;29(5):965–75.
8. Hare GM. Anaemia and the brain. Curr Opin Anaesthesiol 2004;17:363–9.
9. Oddo M, Milby A, Chen I, et al. Hemoglobin concentration and cerebral
metabolism in patients with aneurysmal subarachnoid hemorrhage. Stroke
2009;40(4):1275–81.
10. Rosenthal G, Hemphill JC, Sorani M, et al. The role of lung function in
brain tissue oxygenation following traumatic brain injury. J Neurosurg 2008;
108:59–65.
11. Hoffman WE, Charbel FT, Edelman G. Brain tissue oxygen, carbon dioxide,
and pH in neurosurgical patients at risk for ischemia. Anesth Analg 1996;
82(3):582–6.
12. Hlatky R, Valadka AB, Goodman JC, et al. Patterns of energy substrates
during ischemia measured in the brain by microdialysis. J Neurotrauma
2004;21(7):894–906.
13. Hlatky R, Valadka AB, Goodman JC, et al. Evolution of brain tissue injury
after evacuation of acute traumatic subdural hematomas. Neurosurgery
2007;61(Suppl 1):249–54; discussion 254–5.
14. Maloney-Wilensky E, Gracias V, Itkin A, et al. Brain tissue oxygen and
outcome after severe traumatic brain injury: a systematic review. Crit Care
Med 2009;37(6):2057–63. Review.
15. Chang JJ, Youn TS, Benson D, et al. Physiologic and functional outcome
correlates of brain tissue hypoxia in traumatic brain injury. Crit Care Med
2009;37(1):283–90.
16. Friess SH, Ralston J, Eucker SA, et al. Neurocritical care monitoring
correlates with neuropathology in a swine model of pediatric traumatic
brain injury. Neurosurgery 2011;69(5):1139–47; discussion 1147.
17. Oddo M, Levine JM, Mackenzie L, et al. Brain hypoxia is associated with
short-term outcome after severe traumatic brain injury independently of
intracranial hypertension and low cerebral perfusion pressure. Neurosurgery
2011;69(5):1037–45; discussion 1045.
18. Nangunoori R, Maloney-Wilensky E, Stiefel MD, et al. Brain tissue oxygen
based therapy and outcome after severe traumatic brain injury: a systematic
literature review. Neurocrit Care 2011;Aug 16. Epub ahead of print.
19. Eriksson EA, Barletta JF, Figueroa BE, et al. Cerebral perfusion pressure and
intracranial pressure are not surrogates for brain tissue oxygenation in
traumatic brain injury. Clin Neurophysiol 2011;Nov 18. Epub ahead of print.
20. Feng JF, Zhao X, Gurkoff GG, et al. Post-traumatic hypoxia exacerbates
neuronal cell death in the hippocampus. J Neurotrauma 2012;Jan 30. Epub
ahead of print.
21. Rohlwink UK, Zwane E, Graham Fieggen A, et al. The relationship between
intracranial pressure and brain oxygenation in children with severe
traumatic brain injury. Neurosurgery 2012;70(5):1220–31.
22. Bhatia A, Gupta AK. Neuromonitoring in the intensive care unit. II. Cerebral
oxygenation monitoring and microdialysis. Intensive Care Med 2007;33:
1322–8.
23. Stiefel MF, Udoetuk JD, Spiotta AM, et al. Conventional neurocritical care
and cerebral oxygenation after traumatic brain injury. J Neurosurg 2006;105:
568–75.
24. Gracias VH, Guillamondegui OD, Stiefel, MF, et al. Cerebral cortical
oxygenation: a pilot study. J Trauma 2004;56: 469–74.

25. Stewart C, Haitsma I, Zador Z, et al. The new Licox combined brain tissue
oxygen and brain temperature monitor: assessment of in vitro accuracy and
clinical experience in severe traumatic brain injury. Neurosurgery 2008;63:
1159–64; discussion 1164–5.
26. Orakcioglu B, Sakowitz OW, Neumann JO, et al. Evaluation of a novel brain
tissue oxygenation probe in an experimental swine model. Neurosurgery
2010;67(6):1716–22; discussion 1722–3.
27. Dengler J, Frenzel C, Vajkoczy P, et al. Cerebral tissue oxygenation measured
by two different probes: challenges and interpretation. Intensive Care Med
2011;37:1809–15.
28. Dengler J, Frenzel C, Vajkoczy P, et al. The oxygen reactivity index and its
relation to sensor technology in patients with severe brain lesions. Neurocrit
Care 2012;Mar 7. Epub ahead of print.
29. Hutchinson PJ, Hutchinson DB, Barr RH, et al. A new cranial access device
for cerebral monitoring. Br J Neurosurg 2000;14:46–8.
30. Jaeger M, Soehle M, Meixensberger J. Brain tissue oxygen (PtiO2): a clinical
comparison of two monitoring devices. Acta Neurochir Suppl 2005;95:
79–81.
31. Maloney-Wilensky E, Le Roux P. The physiology behind direct brain oxygen
monitors and practical aspects of their use. Childs Nerv Syst
2010;26(4):419–30.
32. Kiening KL, Unterberg AW, Bardt TF, et al. Monitoring of cerebral
oxygenation in patients with severe head injuries: brain tissue PO2 versus
jugular vein oxygen saturation. J Neurosurg 1996;85:751–7.
33. Longhi L, Pagan F, Valeriani V, et al. Monitoring brain tissue oxygen tension
in brain-injured patients reveals hypoxic episodes in normal-appearing and
in peri-focal tissue. Intensive Care Med 2007;33:2136–42.
34. Ponce LL, Pillai S, Cruz J, et al. Position of probe determines prognostic
information of brain tissue pO2 in severe traumatic brain injury.
Neurosurgery 2012;Jan 27. Epub ahead of print.
35. Scheufler KM. Tissue oxygenation and capacity to deliver O2: do the two go
together? Transfus Apher Sci 2004;31:45–54.
36. Hlatky R, Valadka AB, Gopinath SP, et al. Brain tissue oxygen tension
response to induced hyperoxia reduced in hypoperfused brain. J Neurosurg
2008;108(1):53–8.
37. Gupta AK, Hutchinson PJ, Al-Rawi P, et al. Measuring brain tissue
oxygenation compared with jugular venous oxygen saturation for
monitoring cerebral oxygenation after traumatic brain injury. Anesth Analg
1999;88:549–53.
Section VI—Intracranial Monitoring 355
38. Gopinath SP, Valadka A, Contant CF, et al. Relationship between global and
cortical cerebral blood flow in patients with head injuries. Neurosurgery
1999;44:1273–8; discussion 1278–9.
39. Hoffman WE, Charbel FT, Edelman G, et al. Brain tissue oxygen pressure,
carbon dioxide pressure and pH during ischemia. Neurol Res 1996;18:
54–6.
40. Sarrafzadeh AS, Kiening KL, Bardt TF, et al. Cerebral oxygenation in
contusioned vs. nonlesioned brain tissue: monitoring of PtiO2 with Licox
and Paratrend. Acta Neurochir Suppl 1998;71:186–9.
41. Critchley GR, Bell BA. Acute cerebral tissue oxygenation changes following
experimental subarachnoid hemorrhage. Neurol Res 2003;25:451–6.
42. Zauner A, Bullock R, Di X, et al. Brain oxygen, CO2, pH, and temperature
monitoring: evaluation in the feline brain. Neurosurgery 1995;37:1168–76;
discussion 1176–7.
43. Hoffman WE, Charbel FT, Edelman G. Brain tissue oxygen, carbon dioxide,
and pH in neurosurgical patients at risk for ischemia. Anesth Analg
1996;82:582–6.
44. Hoffman WE, Charbel FT, Edelman G, et al. Brain tissue oxygen pressure,
carbon dioxide pressure, and pH during hypothermic circulatory arrest. Surg
Neurol 1996;46:75–9.
45. Meixensberger J, Dings J, Kuhnigk H, et al. Studies of tissue PO2 in normal
and pathological human brain cortex. Acta Neurochir Suppl (Wien) 1993;59:
58–63.
46. Pennings FA, Schuurman PR, van den Munckhof P, et al. Brain tissue oxygen
pressure monitoring in awake patients during functional neurosurgery: the
assessment of normal values. J Neurotrauma 2008;25:1173–7.
47. Dings J, Meixensberger J, Amschler J, et al. Brain tissue pO2 in relation to
cerebral perfusion pressure, TCD findings and TCD-CO2-reactivity after
severe head injury. Acta Neurochir (Wien) 1996;138:425–34.
48. Valadka AB, Gopinath SP, Contant CF, et al. Relationship of brain tissue PO2
to outcome after severe head injury. Crit Care Med 1998;26:1576–81.
49. van den Brink WA, van Santbrink H, Steyerberg EW, et al. Brain oxygen
tension in severe head injury. Neurosurgery 2000;46:868–76; discussion
876–8.
50. Oddo M, Levine JM, Frangos S, et al. Brain lactate metabolism in humans
with subarachnoid hemorrhage. Stroke 2012;43(5):1418–21.
A complete list of references for this chapter can be found online at
www.expertconsult.com.

References
1. Siegemund M, van Bommel J, Ince C. Assessment of regional tissue
oxygenation. Intensive Care Med 1999;25:1044–60.
2. Coles JP. Regional ischemia after head injury. Curr Opin Crit Care 2004;10:
120–5.
3. Marmarou A, Signoretti S, Fatouros P, et al. Mitochondrial injury measured
by proton magnetic resonance spectroscopy in severe head trauma patients.
Acta Neurochir Suppl 2005;95:149–51.
4. Vespa P, Bergsneider M, Hattori N, et al. Metabolic crisis without brain
ischemia is common after traumatic brain injury: a combined microdialysis
and positron emission tomography study. J Cereb Blood Flow Metab 2005;
25:763–74.
5. Rosenthal G, Hemphill JC, 3rd, Sorani M, et al. Brain tissue oxygen
tension is more indicative of oxygen diffusion than oxygen delivery and
metabolism in patients with traumatic brain injury. Crit Care Med 2008;36:
1917–24.
6. Menon DK, Coles JP, Gupta AK, et al. Diffusion limited oxygen delivery
following head injury. Crit Care Med 2004;32:1384–90.
7. Coles JP, Cunningham AS, Salvador R, et al. Early metabolic characteristics
of lesion and nonlesion tissue after head injury. J Cereb Blood Flow Metab
2009;29(5):965–75.
8. Hare GM. Anaemia and the brain. Curr Opin Anaesthesiol 2004;17:363–9.
9. Oddo M, Milby A, Chen I, et al. Hemoglobin concentration and cerebral
metabolism in patients with aneurysmal subarachnoid hemorrhage. Stroke
2009;40(4):1275–81.
10. Rosenthal G, Hemphill JC, Sorani M, et al. The role of lung function in
brain tissue oxygenation following traumatic brain injury. J Neurosurg
2008;108:59–65.
11. Hoffman WE, Charbel FT, Edelman G. Brain tissue oxygen, carbon dioxide,
and pH in neurosurgical patients at risk for ischemia. Anesth Analg 1996;
82(3):582–6.
12. Hlatky R, Valadka AB, Goodman JC, et al. Patterns of energy substrates
during ischemia measured in the brain by microdialysis. J Neurotrauma
2004;21(7):894–906.
13. Hlatky R, Valadka AB, Goodman JC, et al. Evolution of brain tissue injury
after evacuation of acute traumatic subdural hematomas. Neurosurgery
2007;61(Suppl 1):249–54; discussion 254–5.
14. Maloney-Wilensky E, Gracias V, Itkin A, et al. Brain tissue oxygen and
outcome after severe traumatic brain injury: a systematic review. Crit Care
Med 2009;37(6):2057–63. Review.
15. Chang JJ, Youn TS, Benson D, et al. Physiologic and functional outcome
correlates of brain tissue hypoxia in traumatic brain injury. Crit Care Med
2009;37(1):283–90.
16. Friess SH, Ralston J, Eucker SA, et al. Neurocritical care monitoring
correlates with neuropathology in a swine model of pediatric traumatic
brain injury. Neurosurgery 2011;69(5):1139–47; discussion 1147.
17. Oddo M, Levine JM, Mackenzie L, et al. Brain hypoxia is associated with
short-term outcome after severe traumatic brain injury independently of
intracranial hypertension and low cerebral perfusion pressure.
Neurosurgery 2011;69(5):1037–45; discussion 1045.
18. Nangunoori R, Maloney-Wilensky E, Stiefel MD, et al. Brain tissue oxygen
based therapy and outcome after severe traumatic brain injury: a systematic
literature review. Neurocrit Care 2011;Aug 16. Epub ahead of print.
19. Eriksson EA, Barletta JF, Figueroa BE, et al. Cerebral perfusion pressure and
intracranial pressure are not surrogates for brain tissue oxygenation in
traumatic brain injury. Clin Neurophysiol 2011;Nov 18. Epub ahead of
print.
20. Feng JF, Zhao X, Gurkoff GG, et al. Post-traumatic hypoxia exacerbates
neuronal cell death in the hippocampus. J Neurotrauma 2012;Jan 30. Epub
ahead of print.
21. Rohlwink UK, Zwane E, Graham Fieggen A, et al. The relationship between
intracranial pressure and brain oxygenation in children with severe
traumatic brain injury. Neurosurgery 2012;70(5):1220–31.
22. Bhatia A, Gupta AK. Neuromonitoring in the intensive care unit. II.
Cerebral oxygenation monitoring and microdialysis. Intensive Care Med
2007;33:1322–8.
23. Stiefel MF, Udoetuk JD, Spiotta AM, et al. Conventional neurocritical care
and cerebral oxygenation after traumatic brain injury. J Neurosurg 2006;
105:568–75.
24. Gracias VH, Guillamondegui OD, Stiefel, MF, et al. Cerebral cortical
oxygenation: a pilot study. J Trauma 2004;56: 469–74.
25. Stewart C, Haitsma I, Zador Z, et al. The new Licox combined brain tissue
oxygen and brain temperature monitor: assessment of in vitro accuracy and
clinical experience in severe traumatic brain injury. Neurosurgery
2008;63:1159–64; discussion 1164–5.
Section VI—Intracranial Monitoring 355.e1
26. Orakcioglu B, Sakowitz OW, Neumann JO, et al. Evaluation of a novel
brain tissue oxygenation probe in an experimental swine model.
Neurosurgery 2010;67(6):1716–22; discussion 1722–3.
27. Dengler J, Frenzel C, Vajkoczy P, et al. Cerebral tissue oxygenation
measured by two different probes: challenges and interpretation. Intensive
Care Med 2011;37:1809–15.
28. Dengler J, Frenzel C, Vajkoczy P, et al. The oxygen reactivity index and its
relation to sensor technology in patients with severe brain lesions.
Neurocrit Care 2012;Mar 7. Epub ahead of print.
29. Hutchinson PJ, Hutchinson DB, Barr RH, et al. A new cranial access device
for cerebral monitoring. Br J Neurosurg 2000;14:46–8.
30. Jaeger M, Soehle M, Meixensberger J. Brain tissue oxygen (PtiO2): a clinical
comparison of two monitoring devices. Acta Neurochir Suppl 2005;95:
79–81.
31. Maloney-Wilensky E, Le Roux P. The physiology behind direct brain
oxygen monitors and practical aspects of their use. Childs Nerv Syst 2010;
26(4):419–30.
32. Kiening KL, Unterberg AW, Bardt TF, et al. Monitoring of cerebral
oxygenation in patients with severe head injuries: brain tissue PO2 versus
jugular vein oxygen saturation. J Neurosurg 1996;85:751–7.
33. Longhi L, Pagan F, Valeriani V, et al. Monitoring brain tissue oxygen tension
in brain-injured patients reveals hypoxic episodes in normal-appearing and
in peri-focal tissue. Intensive Care Med 2007;33:2136–42.
34. Ponce LL, Pillai S, Cruz J, et al. Position of probe determines prognostic
information of brain tissue pO2 in severe traumatic brain injury.
Neurosurgery 2012;Jan 27. Epub ahead of print.
35. Scheufler KM. Tissue oxygenation and capacity to deliver O2: do the two go
together? Transfus Apher Sci 2004;31:45–54.
36. Hlatky R, Valadka AB, Gopinath SP, et al. Brain tissue oxygen tension
response to induced hyperoxia reduced in hypoperfused brain. J Neurosurg
2008;108(1):53–8.
37. Gupta AK, Hutchinson PJ, Al-Rawi P, et al. Measuring brain tissue
oxygenation compared with jugular venous oxygen saturation for
monitoring cerebral oxygenation after traumatic brain injury. Anesth Analg
1999;88:549–53.
38. Gopinath SP, Valadka A, Contant CF, et al. Relationship between global and
cortical cerebral blood flow in patients with head injuries. Neurosurgery
1999;44:1273–8; discussion 1278–9.
39. Hoffman WE, Charbel FT, Edelman G, et al. Brain tissue oxygen pressure,
carbon dioxide pressure and pH during ischemia. Neurol Res 1996;18:
54–6.
40. Sarrafzadeh AS, Kiening KL, Bardt TF, et al. Cerebral oxygenation in
contusioned vs. nonlesioned brain tissue: monitoring of PtiO2 with Licox
and Paratrend. Acta Neurochir Suppl 1998;71:186–9.
41. Critchley GR, Bell BA. Acute cerebral tissue oxygenation changes
following experimental subarachnoid hemorrhage. Neurol Res 2003;25:
451–6.
42. Zauner A, Bullock R, Di X, et al. Brain oxygen, CO2, pH, and temperature
monitoring: evaluation in the feline brain. Neurosurgery 1995;37:1168–76;
discussion 1176–7.
43. Hoffman WE, Charbel FT, Edelman G. Brain tissue oxygen, carbon dioxide,
and pH in neurosurgical patients at risk for ischemia. Anesth Analg
1996;82:582–6.
44. Hoffman WE, Charbel FT, Edelman G, et al. Brain tissue oxygen pressure,
carbon dioxide pressure, and pH during hypothermic circulatory arrest.
Surg Neurol 1996;46:75–9.
45. Meixensberger J, Dings J, Kuhnigk H, et al. Studies of tissue PO2 in normal
and pathological human brain cortex. Acta Neurochir Suppl (Wien) 1993;
59:58–63.
46. Pennings FA, Schuurman PR, van den Munckhof P, et al. Brain tissue
oxygen pressure monitoring in awake patients during functional
neurosurgery: the assessment of normal values. J Neurotrauma 2008;25:
1173–7.
47. Dings J, Meixensberger J, Amschler J, et al. Brain tissue pO2 in relation to
cerebral perfusion pressure, TCD findings and TCD-CO2-reactivity after
severe head injury. Acta Neurochir (Wien) 1996;138:425–34.
48. Valadka AB, Gopinath SP, Contant CF, et al. Relationship of brain tissue
PO2 to outcome after severe head injury. Crit Care Med 1998;26:1576–81.
49. van den Brink WA, van Santbrink H, Steyerberg EW, et al. Brain oxygen
tension in severe head injury. Neurosurgery 2000;46:868–76; discussion
876–8.
50. Oddo M, Levine JM, Frangos S, et al. Brain lactate metabolism in humans
with subarachnoid hemorrhage. Stroke 2012;43(5):1418–21.
51. Stiefel MF, Spiotta A, Gracias VH, et al. Reduced mortality rate in patients
with severe traumatic brain injury treated with brain tissue oxygen
monitoring. J Neurosurg 2005;103:805–11.
355.e2 Section VI—Intracranial Monitoring
52. Doppenberg EM, Zauner A, Bullock R, et al. Correlations between brain
tissue oxygen tension, carbon dioxide tension, pH, and cerebral blood
flow: a better way of monitoring the severely injured brain? Surg Neurol
1998;49:650–4.
53. Doppenberg EM, Zauner A, Watson JC, et al. Determination of the
ischemic threshold for brain oxygen tension. Acta Neurochir Suppl
1998;71:166–9.
54. Johnston AJ, Steiner LA, Coles JP, et al. Effect of cerebral perfusion pressure
augmentation on regional oxygenation and metabolism after head injury.
Crit Care Med 2005;33:189–95; discussion 255–7.
55. Zauner A, Doppenberg EM, Woodward JJ, et al. Continuous monitoring of
cerebral substrate delivery and clearance: initial experience in 24 patients
with severe acute brain injuries. Neurosurgery 1997;41:1082–91; discussion
1091–3.
56. van Santbrink H, Maas AI, Avezaat CJ. Continuous monitoring of partial
pressure of brain tissue oxygen in patients with severe head injury.
Neurosurgery 1996;38:21–31.
57. Bratton SL, Chestnut RM, Ghajar J, et al. Guidelines for the management of
severe traumatic brain injury. X. Brain oxygen monitoring and thresholds.
J Neurotrauma 2007;24(Suppl 1):S65–70.
58. Smith ML, Counelis GJ, Maloney-Wilensky E, et al. Brain tissue oxygen
tension in clinical brain death: a case series. Neurol Res 2007;29:755–9.
59. Figaji AA, Kent SJ. Brain tissue oxygenation in children diagnosed with
brain death. Neurocrit Care 2010;12(1):56–61.
60. Chen HI, Stiefel MF, Oddo M, et al. Detection of cerebral compromise with
multimodality monitoring in patients with subarachnoid hemorrhage.
Neurosurgery 2011;69(1):53–63; discussion 63.
61. Bohman LE, Heuer GG, Macyszyn L, et al. Medical management of
compromised brain oxygen in patients with severe traumatic brain injury.
Neurocrit Care 2011;14(3):361–9.
62. Katsnelson M, Mackenzie L, Frangos S, et al. Are initial radiographic and
clinical scales associated with subsequent intracranial pressure and brain
oxygen levels after severe traumatic brain injury? Neurosurgery 2012;70(5):
1095–105.
63. Tolias CM, Reinert M, Seiler R, et al. Normobaric hyperoxia—induced
improvement in cerebral metabolism and reduction in intracranial pressure
in patients with severe head injury: a prospective historical cohort-matched
study. J Neurosurg 2004;101:435–44.
64. Ledwith MB, Bloom S, Maloney-Wilensky E, et al. Effect of body position
on cerebral oxygenation and physiologic parameters in patients with acute
neurological conditions. J Neurosci Nurs 2010;42(5):280–7.
65. Pascual JL, Georgoff P, Maloney-Wilensky E, et al. Reduced brain tissue
oxygen in traumatic brain injury: are most commonly used interventions
successful? J Trauma 2011;70(3):535–46.
66. Johnston AJ, Steiner LA, Chatfield DA, et al. Effect of cerebral perfusion
pressure augmentation with dopamine and norepinephrine on global and
focal brain oxygenation after traumatic brain injury. Intensive Care Med
2004;30:791–7.
67. Marin-Caballos AJ, Murillo-Cabezas F, Cayuela-Dominguez A, et al.
Cerebral perfusion pressure and risk of brain hypoxia in severe head injury:
a prospective observational study. Crit Care 2005;9:R670–6.
68. Stocchetti N, Chieregato A, De Marchi M, et al. High cerebral perfusion
pressure improves low values of local brain tissue O2 tension (PtiO2) in
focal lesions. Acta Neurochir Suppl 1998;71:162–5.
69. Clausen T, Scharf A, Menzel M, et al. Influence of moderate and profound
hyperventilation on cerebral blood flow, oxygenation and metabolism.
Brain Res 2004;1019:113–23.
70. Carmona Suazo JA, Maas AI, van den Brink WA, et al. CO2 reactivity and
brain oxygen pressure monitoring in severe head injury. Crit Care Med
2000;28:3268–74.
71. Menzel M, Doppenberg EM, Zauner A, et al. Increased inspired oxygen
concentration as a factor in improved brain tissue oxygenation and
tissue lactate levels after severe human head injury. J Neurosurg 1999;91:
1–10.
72. Leal-Noval SR, Rincon-Ferrari MD, Marin-Niebla A, et al. Transfusion of
erythrocyte concentrates produces a variable increment on cerebral
oxygenation in patients with severe traumatic brain injury: a preliminary
study. Intensive Care Med 2006;32:1733–40.
73. Smith MJ, Stiefel MF, Magge S, et al. Packed red blood cell transfusion
increases local cerebral oxygenation. Crit Care Med 2005;33:1104–8.
74. Zygun DA, Nortje J, Hutchinson PJ, et al. The effect of red blood cell
transfusion on cerebral oxygenation and metabolism after severe traumatic
brain injury. Crit Care Med 2009;37:1074–8.
75. Figaji AA, Zwane E, Kogels M, et al. The effect of blood transfusion on
brain oxygenation in children with severe traumatic brain injury. Pediatr
Crit Care Med 2010;11(3):325–31.
76. Al-Rawi PG, Zygun D, Tseng MY, et al. Cerebral blood flow augmentation
in patients with severe subarachnoid haemorrhage. Acta Neurochir Suppl
2005;95:123–7.
77. Francony G, Fauvage B, Falcon D, et al. Equimolar doses of mannitol and
hypertonic saline in the treatment of increased intracranial pressure. Crit
Care Med 2008;36(3):795–800.
78. Oddo M, Levine JM, Frangos S, et al. Effect of mannitol and hypertonic
saline on cerebral oxygenation in patients with severe traumatic brain
injury and refractory intracranial hypertension. J Neurol Neurosurg
Psychiatry 2009;80(8):916–20.
79. Sakowitz OW, Stover JF, Sarrafzadeh AS, et al. Effects of mannitol bolus
administration on intracranial pressure, cerebral extracellular metabolites,
and tissue oxygenation in severely head-injured patients. J Trauma 2007;
62:292–8.
80. Weiner GM, Lacey MR, Mackenzie L, et al. Decompressive craniectomy for
elevated intracranial pressure and its effect on the cumulative ischemic
burden and therapeutic intensity levels after severe traumatic brain injury.
Neurosurgery 2010;66:1111–19.
81. Strege RJ, Lang EW, Stark AM, et al. Cerebral edema leading to
decompressive craniectomy: an assessment of the preceding clinical and
neuromonitoring trends. Neurol Res 2003;25:510–5.
82. Stiefel MF, Heuer GG, Smith MJ, et al. Cerebral oxygenation following
decompressive hemicraniectomy for the treatment of refractory intracranial
hypertension. J Neurosurg 2004;101:241–7.
83. Jaeger M, Soehle M, Meixensberger J. Effects of decompressive craniectomy
on brain tissue oxygen in patients with intracranial hypertension. J Neurol
Neurosurg Psychiatry 2003;74:513–5.
84. Ho CL, Wang CM, Lee KK, et al. Cerebral oxygenation, vascular reactivity,
and neurochemistry following decompressive craniectomy for severe
traumatic brain injury. J Neurosurg 2008;108:943–9.
85. Oddo M, Frangos S, Maloney-Wilensky E, et al. Effect of shivering on brain
tissue oxygenation during induced normothermia in patients with severe
brain injury. Neurocrit Care 2010;12(1):10–6.
86. Coles JP, Minhas PS, Fryer TD, et al. Effect of hyperventilation on cerebral
blood flow in traumatic head injury: clinical relevance and monitoring
correlates. Crit Care Med 2002;30:1950–9.
87. Swanson E, Mascitelli J, Stiefel M, et al. The effect of patient
transport on brain oxygen in comatose patients. Neurosurgery 66:
925–32, 2010.
88. Dohmen C, Bosche B, Graf R, et al. Identification and clinical impact of
impaired cerebrovascular autoregulation in patients with malignant middle
cerebral artery infarction. Stroke 2007;38:56–61.
89. Jaeger M, Schuhmann MU, Soehle M, et al. Continuous assessment of
cerebrovascular autoregulation after traumatic brain injury using brain
tissue oxygen pressure reactivity. Crit Care Med 2006;34:1783–8.
90. Jaeger M, Schuhmann MU, Soehle M, et al. Continuous monitoring of
cerebrovascular autoregulation after subarachnoid hemorrhage by brain
tissue oxygen pressure reactivity and its relation to delayed cerebral
infarction. Stroke 2007;38:981–6.
91. Lang EW, Czosnyka M, Mehdorn HM. Tissue oxygen reactivity and cerebral
autoregulation after severe traumatic brain injury. Crit Care Med 2003;31:
267–71.
92. Muench E, Horn P, Bauhuf C, et al. Effects of hypervolemia and
hypertension on regional cerebral blood flow, intracranial pressure, and
brain tissue oxygenation after subarachnoid hemorrhage. Crit Care Med
2007;35:1844–51; quiz 52.
93. Ekelund A, Reinstrup P, Ryding E, et al. Effects of iso- and hypervolemic
hemodilution on regional cerebral blood flow and oxygen delivery for
patients with vasospasm after aneurysmal subarachnoid hemorrhage. Acta
Neurochir (Wien) 2002;144:703–12; discussion 712–3.
94. Kett-White R, Hutchinson PJ, Al-Rawi PG, et al. Cerebral oxygen and
microdialysis monitoring during aneurysm surgery: effects of blood
pressure, cerebrospinal fluid drainage, and temporary clipping on
infarction. J Neurosurg 2002;96:1013–9.
95. Kett-White R, Hutchinson PJ, Czosnyka M, et al. Effects of variation in
cerebral haemodynamics during aneurysm surgery on brain tissue oxygen
and metabolism. Acta Neurochir Suppl 2002;81:327–9.
96. Dudkiewicz M, Proctor KG. Tissue oxygenation during management of
cerebral perfusion pressure with phenylephrine or vasopressin. Crit Care
Med 2008;36:2641–50.
97. Stiefel MF, Heuer GG, Abrahams JM, et al. The effect of nimodipine on
cerebral oxygenation in patients with poor-grade subarachnoid
hemorrhage. J Neurosurg 2004;101:594–9.
98. Stiefel MF, Spiotta A, Udoetek J, et al. Intra-arterial papaverine used to
treat cerebral vasospasm reduces brain oxygen. Neurocritical Care 2006;4:
113–8.

99. Figaji AA, Zwane E, Thompson C, et al. Brain tissue oxygen tension
monitoring in pediatric severe traumatic brain injury. part 1: relationship
with outcome. Childs Nerv Syst 2009;25(10):1325–33.
100. Ramakrishna R, Stiefel M, Udoteuk J, et al. Brain oxygen and outcome in
patients with aneurysmal subarachnoid hemorrhage. J Neurosurgery
2008;109(6):1075–82.
101. Dings J, Jager A, Meixensberger J, et al. Brain tissue pO2 and outcome after
severe head injury. Neurol Res 1998;20(Suppl 1):S71–5.
102. Ang BT, Wong J, Lee KK, et al. Temporal changes in cerebral tissue
oxygenation with cerebrovascular pressure reactivity in severe traumatic
brain injury. J Neurol Neurosurg Psychiatry 2007;78:298–302.
103. Chang JJ, Youn TS, Benson D, et al. Physiologic and functional outcome
correlates of brain tissue hypoxia in traumatic brain injury. Crit Care Med
2009;37:283–90.
104. Spiotta AM, Stiefel MF, Gracias VH, et al. Brain tissue oxygen-directed
management and outcome in patients with severe traumatic brain injury.
J Neurosurg 2010;113(3):571–80.
105. Narotam PK, Morrison JF, Nathoo N. Brain tissue oxygen monitoring in
traumatic brain injury and major trauma: outcome analysis of a brain
tissue oxygen-directed therapy. J Neurosurg 2009;111(4):672–82.
106. Martini RP, Deem S, Yanez ND, et al. Management guided by brain tissue
oxygen monitoring and outcome following severe traumatic brain injury.
J Neurosurg 2009;111(4):644–9.
107. Meixensberger J, Jaeger M, Vath A, et al. Brain tissue oxygen guided
treatment supplementing ICP/CPP therapy after traumatic brain injury.
J Neurol Neurosurg Psychiatry 2003;74:760–4.
108. Haitsma IK, Maas AI. Advanced monitoring in the intensive care unit:
brain tissue oxygen tension. Curr Opin Crit Care 2002;8:115-20.
109. Scheufler KM, Rohrborn HJ, Zentner J. Does tissue oxygen-tension reliably
reflect cerebral oxygen delivery and consumption? Anesth Analg 2002;95:
1042–8.
110. Scheufler K-M, Lehnert A, Rohrborn H-J, et al: Individual values of brain
tissue oxygen pressure, microvascular oxygen saturation, cytochrome redox
level and energy metabolites in detecting critically reduced cerebral energy
state during acute changes in global cerebral perfusion. J Neurosurg
Anesthesiol 2004; 16:210–19.
111. Hemphill JC, 3rd, Smith WS, Sonne DC, et al. Relationship between brain
tissue oxygen tension and CT perfusion: feasibility and initial results. AJNR
Am J Neuroradiol 2005;26:1095–100.
112. Jaeger M, Soehle M, Schuhmann MU, et al. Correlation of continuously
monitored regional cerebral blood flow and brain tissue oxygen. Acta
Neurochir (Wien) 2005;147:51–6; discussion 56.
113. Menzel M, Doppenberg EM, Zauner A, et al. Cerebral oxygenation in
patients after severe head injury: monitoring and effects of arterial
hyperoxia on cerebral blood flow, metabolism and intracranial pressure.
J Neurosurg Anesthesiol 1999;11:240–51.
114. Reinert M, Barth A, Rothen HU, et al. Effects of cerebral perfusion pressure
and increased fraction of inspired oxygen on brain tissue oxygen, lactate
and glucose in patients with severe head injury. Acta Neurochir (Wien)
2003;145:341–9; discussion 349–50.
115. Hemphill JC, 3rd, Knudson MM, Derugin N, et al. Carbon dioxide
reactivity and pressure autoregulation of brain tissue oxygen. Neurosurgery
2001;48:377–83; discussion 383–4.
116. Hlatky R, Valadka AB, Gopinath SP, et al. Brain tissue oxygen tension
response to induced hyperoxia reduced in hypoperfused brain. J Neurosurg
2008;108:53–8.
117. Oddo M, Le Roux P. Brain tissue oxygen monitors: more than an ischemia
monitor. Crit Care Med 2008;36:1984–5.
118. Kety SS, Schmidt CF. The nitrous oxide method for the quantitative
determination of cerebral blood flow in man: theory, procedure and
normal values. J Clin Invest 1948;27:476–83.
119. Nortje J, Coles JP, Timofeev I, et al. Effect of hyperoxia on regional
oxygenation and metabolism after severe traumatic brain injury:
preliminary findings. Crit Care Med 2008;36:273–81.
120. Abate MG, Trivedi M, Fryer TD, et al. Early derangements in oxygen and
glucose metabolism following head injury: the ischemic penumbra and
pathophysiological heterogeneity. Neurocrit Care 2008;9:319–25.
121. Merenda A, Bullock R. Clinical treatments for mitochondrial dysfunctions
after brain injury. Curr Opin Crit Care 2006;12:90–6.
122. Dings J, Meixensberger J, Jager A, et al. Clinical experience with 118 brain
tissue oxygen partial pressure catheter probes. Neurosurgery 1998;43:
1082–95.
123. Brawanski A, Faltermeier R, Rothoerl RD, et al. Comparison of near-
infrared spectroscopy and tissue p(O2) time series in patients after severe
head injury and aneurysmal subarachnoid hemorrhage. J Cereb Blood Flow
Metab 2002;22:605–11.
Section VI—Intracranial Monitoring 355.e3
124. Carvi y Nievas M, Toktamis S, Hollerhage HG, et al. Hyperacute
measurement of brain-tissue oxygen, carbon dioxide, pH, and intracranial
pressure before, during, and after cerebral angiography in patients with
aneurysmatic subarachnoid hemorrhage in poor condition. Surg Neurol
2005;64:362–7; discussion 367.
125. Dings J, Meixensberger J, Roosen K. Brain tissue pO2-monitoring: catheter
stability and complications. Neurol Res 1997;19:241–5.
126. Lang EW, Mulvey JM, Mudaliar Y, et al. Direct cerebral oxygenation
monitoring—a systematic review of recent publications. Neurosurg Rev
2007;30:99–106; discussion 107.
127. Nortje J, Gupta AK. The role of tissue oxygen monitoring in patients with
acute brain injury. Br J Anaesth 2006;97:95–106.
128. Bailey RL, Quattrone F, Curtain C, et al. The safety of multimodal
monitoring in severe brain injury. Montreal, Canada, Neurocritical Care
Society Meeting, 2011.
129. Hartl R, Bardt TF, Kiening KL, et al. Mannitol decreases ICP but does not
improve brain-tissue pO2 in severely head-injured patients with
intracranial hypertension. Acta Neurochir Suppl 1997;70:40–2.
130. Sarrafzadeh AS, Kiening KL, Callsen TA, et al. Metabolic changes during
impending and manifest cerebral hypoxia in traumatic brain injury. Br J
Neurosurg 2003;17:340–6.
131. Bullock MR. Hyperoxia: good or bad? J Neurosurg 2003;98:943–4;
discussion 944.
132. Diringer MN. Hyperoxia: good or bad for the injured brain? Curr Opin
Crit Care 2008;14:167–71.
133. Diringer MN, Aiyagari V, Zazulia AR, et al. Effect of hyperoxia on cerebral
metabolic rate for oxygen measured using positron emission tomography
in patients with acute severe head injury. J Neurosurg 2007;106:526–9.
134. Longhi L, Stocchetti N. Hyperoxia in head injury: therapeutic tool? Curr
Opin Crit Care 2004;10:105–9.
135. Borgstrom L, Johannsson H, Siesjo BK. The influence of acute
normovolemic anemia on cerebral blood flow and oxygen consumption of
anesthetized rats. Acta Physiol Scand 1975;93:505–14.
136. Dexter F, Hindman BJ. Effect of haemoglobin concentration on brain
oxygenation in focal stroke: a mathematical modelling study. Br J Anaesth
1997;79:346–51.
137. Hare GM, Mazer CD, Hutchison JS, et al. Severe hemodilutional anemia
increases cerebral tissue injury following acute neurotrauma. J Appl Physiol
2007;103:1021–9.
138. Hare GM, Mazer CD, Mak W, et al. Hemodilutional anemia is associated
with increased cerebral neuronal nitric oxide synthase gene expression.
J Appl Physiol 2003;94:2058–67.
139. Kramer A, Le Roux P. Red blood cell transfusion and transfusion
alternatives in traumatic brain injury. Curr Treat Options Neurol 2012;14:
150–63.
140. Le Roux PD. Participants in the International Multi-disciplinary Consensus
Conference on the Critical Care Management of Subarachnoid
Hemorrhage. Anemia and transfusion after subarachnoid hemorrhage.
Neurocrit Care 2011;15(2):342–53.
141. Oddo M, Levine JM, Kumar M, et al Anemia and brain oxygen after severe
traumatic brain injury. Intensive Care Med 2012;38(9):1497–504.
142. Naidech AM, Jovanovic B, Wartenberg KE, et al. Higher hemoglobin is
associated with improved outcome after subarachnoid hemorrhage. Crit
Care Med 2007;35:2383–9.
143. Fletcher JJ, Bergman K, Blostein PA, et al. Fluid balance, complications, and
brain tissue oxygen tension monitoring following severe traumatic brain
injury. Neurocrit Care 2010;13(1):47–56.
144. Johnston AJ, Steiner LA, Chatfield DA, et al. Effects of propofol on cerebral
oxygenation and metabolism after head injury. Br J Anaesth 2003;91:781–6.
145. Chen HI, Malhotra NR, Oddo M, et al. Barbiturate infusion for intractable
intracranial hypertension and its effect on brain oxygenation. Neurosurgery
2008;63:880–6; discussion 886–7.
146. Thorat JD, Wang EC, Lee KK, et al. Barbiturate therapy for patients with
refractory intracranial hypertension following severe traumatic brain
injury: its effects on tissue oxygenation, brain temperature and
autoregulation. J Clin Neurosci 2008;15:143–8.
147. Oddo M, Frangos S, Milby A, et al. Induced normothermia attenuates
cerebral metabolic distress in patients with aneurysmal subarachnoid
hemorrhage and refractory fever. Stroke 2009;40(5):1913–6. Epub 2009,
Feb 26.
148. Kett-White R, Hutchinson PJ, Al-Rawi PG, et al. Adverse cerebral events
detected after subarachnoid hemorrhage using brain oxygen and
microdialysis probes. Neurosurgery 2002;50:1213–21; discussion 21–2.
149. Gelabert-Gonzalez M, Fernandez-Villa JM, Ginesta-Galan V. Intra-operative
monitoring of brain tissue O2 (PtiO2) during aneurysm surgery. Acta
Neurochir (Wien) 2002;144:863–6; discussion 866–7.
355.e4 Section VI—Intracranial Monitoring
150. Hoffman WE, Charbel FT, Edelman G, et al. Brain tissue oxygenation in
patients with cerebral occlusive disease and arteriovenous malformations.
Br J Anaesth 1997;78:169–71.
151. Hoffman WE, Wheeler P, Edelman G, et al. Hypoxic brain tissue following
subarachnoid hemorrhage. Anesthesiology 2000;92:442–6.
152. Szelenyi A, Jung CS, Schon H, Seifert V. Brain tissue oxygenation
monitoring supplementary to somatosensory evoked potential monitoring
for aneurysm surgery. Initial clinical experience. Neurol Res 2002;24:555–62.
153. Hoffman WE, Charbel FT, Edelman G, et al. Brain tissue gases and pH
during arteriovenous malformation resection. Neurosurgery 1997;40:294–
300; discussion 301.
154. Pennings FA, Bouma GJ, Kedaria M, et al. Intraoperative monitoring of
brain tissue oxygen and carbon dioxide pressures reveals low oxygenation
in peritumoral brain edema. J Neurosurg Anesthesiol 2003;15:1–5.
155. Tijero T, Ingelmo I, Garcia-Trapero J, et al. Usefulness of monitoring brain
tissue oxygen pressure during awake craniotomy for tumor resection: a case
report. J Neurosurg Anesthesiol 2002;14:149–52.
156. Hoffman WE, Edelman G. Isoflurane increases brain oxygen reactivity in
dogs. Anesth Analg 2000;91:637–41.
157. Hoffman WE, Edelman G. Enhancement of brain tissue oxygenation during
high dose isoflurane anesthesia in the dog. J Neurosurg Anesthesiol
2000;12:95–8.
158. Steiner LA, Johnston AJ, Chatfield DA, et al. The effects of large-dose
propofol on cerebrovascular pressure autoregulation in head-injured
patients. Anesth Analg 2003;97:572–6.
159. van Santbrink H, van den Brink WA, Steyerberg EW, et al. Brain tissue
oxygen response in severe traumatic brain injury. Acta Neurochir (Wien)
2003;145:429–38; discussion 438.
160. Hoffman WE, Edelman G, Wheeler P. Cerebral oxygen reactivity in the dog.
Neurol Res 2000;22:620–2.
161. Soehle M, Jaeger M, Meixensberger J. Online assessment of brain tissue
oxygen autoregulation in traumatic brain injury and subarachnoid
hemorrhage. Neurol Res 2003;25:411–7.
162. Meixensberger J, Renner C, Simanowski R, et al. Influence of cerebral
oxygenation following severe head injury on neuropsychological testing.
Neurol Res 2004;26:414–7.
163. Meixensberger J, Vath A, Jaeger M, et al. Monitoring of brain tissue
oxygenation following severe subarachnoid hemorrhage. Neurol Res
2003;25:445–50.
164. Adamides AA, Cooper DJ, Rosenfeldt FL, et al. Focal cerebral oxygenation
and neurological outcome with or without brain tissue oxygen-guided
therapy in patients with traumatic brain injury. Acta Neurochir (Wien)
2009;151(11):1399–409.
165. Martini RP, Deem S, Yanez ND, et al. Management guided by brain tissue
oxygen monitoring and outcome following severe traumatic brain injury.
J Neurosurg 2009;111(4):644–9.
VI
Chapter
36  
Cerebral Microdialysis
Martin Smith
Introduction
Cerebral microdialysis (MD) is a well-established laboratory
tool that is being increasingly used as a bedside monitor to
provide online analysis of brain tissue biochemistry during
neurointensive care. MD samples substances present in brain
extracellular fluid (ECF) and has opened new avenues to
monitor the injured brain. Because MD measures changes at
the cellular level, it is an attractive technique to detect and
monitor cerebral hypoxia, ischemia, and other causes of cel-
lular dysfunction. To date MD has been predominantly applied
to patients with traumatic brain injury (TBI) and subarach-
noid hemorrhage (SAH) and has increased understanding
of the pathophysiology of these conditions. Studies also have
addressed its potential role as a clinical monitoring technique
that can guide individualized and targeted therapy after brain
injury.1,2 Future studies must identify the effect of therapeutic
maneuvers on brain tissue biochemistry and establish the rela-
tionship between MD variables and outcome.
History
The first use of the MD technique was in 1966 when Bito
inserted membrane-lined sacks containing 6% dextran into
the cerebral hemispheres of dogs and analyzed the fluid that
they contained for amino acid levels when they were removed
10 weeks later.3 During the 1970s Delgado developed the
“dialytrode” device,4 which was refined later and simplified by
Tossman and Ungerstedt into the “microdialysis” method.5
This initially was used to study neurochemical processes in
animals, but in 1990, the first application in the human brain
was described.6 The use of cerebral MD was popularized
following the introduction of commercially available micro-
dialysis catheters and a bedside analyzer (CMA 600, CMA
Microdialysis, Solna, Sweden) and was first used as a neuro-
chemical monitoring tool in neurocritical care in 1992.7
Devices for microdialysis in the human brain were CE-marked
according to the European Medical Device Directive in 1995
and approved by the U.S. Food and Drug Administration
in 2002.
Today MD is a well-established research tool but it has not
yet become widely established as a routine clinical monitor.8
However, some centers that have extensive experience of its
use as a research tool are beginning to incorporate MD moni-
toring into clinical practice.2 When considering the clinical
role of MD, it is important to appreciate that it contributes a
single, albeit vital, component of the whole picture of the
356
injured brain. MD should therefore be seen as a tool to explore
the neurochemical features of secondary brain injury pro-
cesses with other monitors as part of a multimodal monitor-
ing technique.9
Principles of Microdialysis
The principles of the MD technique have been reviewed in
detail elsewhere.1,2,10-14 In brief, a MD catheter consists of a fine
double-lumen probe, lined at its tip with a semipermeable
dialysis membrane. The probe tip is placed into biologic
tissue, in this case the brain, and perfused via an inlet tube
with fluid isotonic to the tissue interstitium. The perfusate
passes along the membrane before exiting through outlet
tubing into a microvial designed to collect minute amounts of
fluid (Figs. 36.1 and 36.2). The microvials are removed regu-
larly, usually every hour, and placed in a bedside analyzer,
which measures and records the changing brain tissue chem-
istry. Subsequently the samples can be analyzed off-line for
other substances.
MD is based on the principle of diffusion of water-soluble
substances across the semipermeable MD membrane driven
by their concentration gradient. Molecules at high concentra-
tion in the brain ECF pass into the perfusate with minimum
passage of water, and because the perfusate flows along the
membrane and is removed at a constant rate, the concentra-
tion gradient across the membrane is maintained. In this way
the MD catheter acts as an artificial blood capillary with the
perfusate gradually equilibrating to the composition of ECF.
The concentration of substrate in the collected fluid (the
microdialysate) depends in part on the balance between
substrate delivery to, and uptake or excretion from, the ECF
(Fig. 36.3). This simple concept provides a powerful technique
in which any molecule small enough to pass across the
membrane can be sampled.
Unless there is total equilibration across the dialysis mem-
brane, the concentration of a given molecule in the microdi-
alysate will be lower than its concentration in the brain ECF.
The proportion of the true ECF concentration collected in the
microdialysate is termed the relative recovery and is dependent
on many factors, including dialysis membrane pore size,
membrane area, perfusate flow rate, and diffusion properties
of the substance. Because of the many variables that affect the
relative recovery, it is essential to consider the sampling
methods and materials used when comparing measured MD
values. In clinical practice the most commonly used system
comprises a catheter that is 10 mm in length with a 20-kDa
© Copyright 2013 Elsevier Inc. All rights reserved.
 Section VI—Intracranial Monitoring 357

(CMA 70, CMA Microdialysis, Solna, Sweden) or 100-kDa
(CMA 71, CMA Microdialysis, Solna, Sweden) molecular
weight cutoff, perfused with commercially available physio-
logic perfusion fluid (Perfusion Fluid CNS, CMA Microdialy-
sis, Solna, Sweden) at a rate of 0.3 µL/min. Commonly
measured bedside MD variables are reliably captured using a
2
1 0.3 µL/min and a 10-cm catheter achieves a recovery rate in
5
3 clinical purposes, continuous sampling would be advanta-
6 4
7
Fig. 36.1  Photograph of the components of a clinical microdialysis
catheter (CMA Microdialysis, Solna, Sweden). (1), Pump connector;
(2), inlet tube; (3), microdialysis catheter; (4), microdialysis membrane;
(5), outlet tube; (6), microvial holder; (7), microvial for collection of
microdialysate. (Reproduced with permission from Tisdall MM, Smith M. Cerebral
microdialysis: research technique or clinical tool. Br J Anaesth 2006;97(1):18–25.)
20-kDa molecular weight cutoff catheter, whereas 100-kDa
cutoff catheters allow investigation of higher molecular weight
biomarkers. For the more clinically relevant MD variables,
100-kDa catheters have equivalent recovery to 20-kDa cutoff
catheters,15 and many centers now use the higher cutoff cath-
eters as a routine.16 A perfusate flow rate of 0.3 µL/min
allows hourly sampling at the bedside, striking a balance
between adequate microdialysate volume and acceptable
recovery rates. Higher flow rates permit more frequent
sampling but at the cost of lower concentration of measured
substances in the microdialysate. A standard flow rate of
excess of 70% for routinely measured metabolites.17 By using
lower perfusate flow rates it is possible to approach 100%
recovery and measurement of the absolute concentration of a
substance in brain ECF.18
Although hourly MD sampling rates are adequate for most
geous when rapid and transient changes in brain metabolite
levels need to be determined. A continuous MD method has
been described for research purposes but it is not yet
sufficiently developed for clinical application.19 This system
incorporates continuous flow of dialysate into a glucose and
lactate analyzer that uses flow-injection dual-assay enzyme-
based biosensors able to determine values of metabolites
every 30 seconds. The analyzer can detect changes 9 minutes
after an event has occurred, and the temporal resolution is
limited only by probe-to-sensor tubing length and perfusate
flow rate.

Microdialysate collected Isotonic fluid perfused

via outlet tube via inlet tube

Microdialysis
catheter
Molecules in ECF
equilibrate across MD
membrane

Semi-permeable
Brain tissue membrane
interstitium

Fig. 36.2  Schematic representation of a microdialysis catheter in brain tissue. Fluid isotonic to the brain extracellular fluid is pumped through
the microdialysis catheter at a rate of 0.3 µL/min. Molecules at high concentration in the brain extracellular fluid equilibrate across the semipermeable
microdialysis membrane and can be analyzed in the collected perfusate (the microdialysate). ECF, Extracellular fluid; MD, microdialysis. (Modified with
permission from Tisdall MM, Smith M. Cerebral microdialysis: research technique or clinical tool. Br J Anaesth 2006;97(1):18–25.)
358 Section VI—Intracranial Monitoring
Decreased
glucose and oxygen
Ischemia
Capillary
Degeneration
Fig. 36.3  Schematic representation of the relationship between blood capillary and microdialysis catheter in brain tissue. The microdialysis
catheter acts as an artificial blood capillary and the concentration of substrate in the collected fluid (microdialysate) is related to the balance between
substrate delivery to, and uptake or excretion from, the extracellular fluid. (Reproduced with permission from Tisdall MM, Smith M. Cerebral microdialysis:
research technique or clinical tool. Br J Anaesth 2006;97(1):18–25.)
Clinical Relevance of Monitoring
Tissue Biochemistry in the
Injured Brain
Acute brain injury frequently is exacerbated by secondary
events that lead to secondary brain injury. This occurs follow-
ing activation by the primary injury of an autodestructive
cascade of metabolic, immunologic, biochemical, and inflam-
matory changes that render the brain more susceptible to
systemic physiologic insults that can result in irreversible cell
damage or death.20 Although these pathologic processes are
poorly understood, they include calcium overload, increased
production of free radicals, release of neurotoxic levels of
excitatory amino acids (EAAs), and failing cellular metabo-
lism. Ultimately these changes can cause cellular swelling,
increase in intracranial pressure (ICP), further neuronal loss,
and, if unchecked, can result in increased mortality and wors-
ened outcome in survivors.
Secondary brain injury is a potentially modifiable cause of
mortality and morbidity, and the primary aim of neurocritical
care treatment of brain injury is to prevent or reduce the
burden of secondary injury.21 Conventional intracranial mon-
itoring techniques, such as measurement of ICP, are often
“reactive” and may indicate changes only when irreversible
tissue damage has already occurred. Monitoring brain tissue
biochemistry through microdialysis has the potential to guide
individualized therapy after brain injury,2 to identify impend-
ing or early onset secondary injury in some cases before there
is a change in ICP,22 to detect cerebral compromise when ICP
or cerebral perfusion pressure (CPP) is normal,23 and allow
timely implementation of neuroprotective strategies.24,25
Glucose
lactate:
markers virgule
pyruvate
ratio
marker Glycerol
Microdialysis catheter
Microdialysis Markers
Many substances can be measured using MD, but the key
variables that are most relevant to neurocritical care can be
categorized as follows11:
• Energy-related metabolites, for example, glucose, lactate,
and pyruvate
• Neurotransmitters, for example, glutamate, aspartate,
gamma-aminobutyric acid (GABA)
• Markers of tissue damage and inflammation, for example,
glycerol, potassium, cytokines
• Exogenous substances, for example, drugs
The different types of markers that can be recovered by MD
highlights the utility of the technique to provide surrogate
biochemical outcomes that reflect underlying pathologic
mechanisms of injury.26 These highlighted markers represent
a small fraction of those that have been recovered by MD, and
a comprehensive review lists those that are less well evalu-
ated.10 Several studies have addressed the recovery of protein
biomarkers such as S100β, tau, beta amyloid proteins, and
neurofilament.27-29 What these various biomarkers mean is still
being elucidated, and the role of biomarkers in neurocritical
care is discussed further in Chapter 18.
Commercially available assays for bedside use are those for
glucose, lactate, pyruvate, glycerol, and glutamate, and tenta-
tive normal values for these variables in adults are described
(Table 36.1). These normal values were derived from a study
by Reinstrup et al., who inserted MD catheters into the frontal
cortex of patients undergoing surgery for benign posterior
fossa lesions and collected microdialysate samples in the
postoperative course to demonstrate baseline metabolite
 Section VI—Intracranial Monitoring 359

Table 36.1  Suggested Normal Concentrations Table 36.2  Biochemical Markers of

of Commonly Measured Biochemical Markers Secondary Brain Injury
in Microdialysate Samples from the
Microdialysis
Uninjured Human Brain Variable Biomarker for: Comments
Normal Value Normal Value Low glucose • Hypoxia/ • Should be interpreted
Microdialysate ± SD ± SD ischemia in association with
Concentration Reinstrup et al.30 Schulz et al.31 • Reduced serum glucose
cerebral concentration
Glucose (mmol/L) 1.7 ± 0.9 2.1 ± 0.2
glucose supply
Lactate (mmol/L) 2.9 ± 0.9 3.1 ± 0.3 • Cerebral
hyperglycolysis
Pyruvate (µmol/L) 166 ± 47 151 ± 12
Increased LPR • Hypoxia/ • Most reliable
Lactate/pyruvate ratio 23 ± 4 19 ± 2
ischemia biomarker of ischemia
Glycerol (µmol/L) 82 ± 44 82 ± 12 • Cellular redox • Independent of
state catheter recovery
Glutamate (µmol/L) 16 ± 16 14 ± 3.3 • Reduced • Tissue hypoxic
From Reinstrup P, Stahl N, Mellergard P, et al. Intracerebral microdialysis in
cerebral threshold for raised
clinical practice: baseline values for chemical markers during wakefulness, glucose supply LPR not established
anesthesia, and neurosurgery. Neurosurgery 2000;47(3):701–9; and • Impairment of
Schulz MK, Wang LP, Tange M, et al. Cerebral microdialysis monitoring: glycolytic
determination of normal and ischemic cerebral metabolisms in patients with pathway
aneurysmal subarachnoid hemorrhage. J Neurosurg 2000;93(5):808–14.
Increased • Hypoxia/ • Increased glycerol
glycerol ischemia may also occur due to
• Cell spillover from
membrane systemic glycerol or
degradation from the formation of
concentrations from the uninjured human brain.30 Concen- glycerol from glucose
trations of metabolites in microdialysate samples from patients
Increased • Hypoxia/ • Wide variability in
with SAH but no clinical or radiologic evidence of cerebral glutamate ischemia glutamate levels
ischemia also have been considered to represent “normal” • Excitotoxicty within and between
values.31 A summary of the pathophysiologic changes moni- patients
tored by cerebral MD biomarkers is shown in Table 36.2.
LPR, Lactate/pyruvate ratio.
There have been far fewer MD studies in children; these
suggest there may be differences in MD findings in children
compared with adults.32 Brain ischemia and hypoxia underlie
many of the secondary cerebral insults that can cause brain
damage; this results in a typical pattern of change in MD cell integrity. Microdialysate glucose levels are reduced in
variables including an increase in glycerol and glutamate con- patients following TBI, and a concentration consistently less
centrations, a reduction in glucose, and an increase in the than 0.66 mmol/L in the first 50 hours post injury is associ-
lactate/pyruvate ratio (LPR) and lactate/glucose ratio (LGR). ated with poor outcome.35 The etiology of this low glucose
Although an elevated LPR frequently is interpreted as a sign concentration is likely to be multifactorial. In the acute period
of cerebral hypoxia or ischemia there also are many other after TBI there is typically a reduction in oxidative metabo-
causes of an elevated LPR that are not related to hypoxia or lism36 and increase in glucose metabolism.37 Extremely low
ischemia.33 Consistent with this Nelson et al.34 observed that brain ECF glucose is observed during periods of severe hypoxia
altered local biochemistry detected with MD in severe TBI or ischemia after TBI2,38 and SAH,39 and associated with a
patients represented long-term metabolic patterns and was brain tissue oxygen (PbtO2) of less than 1.3 kPa (10 mm Hg).40,41
weakly correlated with ICP and CPP; that is, metabolic changes However, there is a poor correlation between positron emis-
other than those related to pressure or flow can influence MD sion tomography (PET)–defined ischemia and low MD
findings. glucose concentration42 suggesting that in some cases at least,
low MD glucose concentration may be associated with hyper-
glycolysis rather than decreased supply of glucose and oxygen
Markers of Glucose Metabolism because of reduced cerebral perfusion.
Tissue hypoxia often is the final common pathway for cellular Glucose is taken up into neurons and glia and initially
damage after acute brain injury, and the most commonly metabolized to pyruvate by glycolysis. When there is adequate
investigated substances assayed by MD relate to the aerobic oxygen delivery and tissue oxygenation, pyruvate enters the
and anaerobic metabolism of glucose. The determinants of citric acid cycle and ultimately is metabolized to carbon
cerebral extracellular glucose concentration are complex and dioxide and water with the generation of adenosine triphos-
depend on peripheral blood glucose concentration, local cap- phate (ATP). During ischemia, pyruvate is diverted into the
illary blood flow, and brain cell uptake of glucose. A particular anaerobic pathway and is metabolized to lactate. The mea-
advantage of cerebral MD monitoring after brain injury is its surement of ECF lactate and pyruvate concentrations there-
ability to assess not only the cerebral delivery of glucose but fore provides information about the extent of anaerobic
also its use. glycolysis. However, the absolute value of lactate within the
The energy demands of the brain are met by the metabo- brain ECF is not in itself indicative of the degree of anaerobic
lism of glucose, and a continued supply is vital to maintain metabolism for several reasons. The production of lactate
360 Section VI—Intracranial Monitoring
depends on a continued supply of glucose, which may fail
during complete ischemia, and the increased levels of gluta-
mate and potassium that are associated with ischemia drive
astrocyte lactate production and hyperglycolysis.37 An increase
in ECF glutamine may reflect this accelerated astrocyte
metabolism and be found in vulnerable rather than already
ischemic tissue.43 In addition, lactate can act as a metabolic
substrate to sustain increased energy needs through the
astrocyte-neuron lactate shuttle.44-46 To correct for the vari-
able sources of ECF lactate and the dynamic variation in
glucose delivery, the LPR47-49 and LGR31,38 have been used as
more accurate markers of anaerobic metabolism. Because
lactate and pyruvate have very similar molecular weights, the
LPR is independent of catheter recovery in vivo.50 The LPR is
therefore a robust and reliable marker of cell energy dysfunc-
tion and is the most widely monitored MD variable after
brain injury.
In the human brain, worsening hypoxia, ischemia, or edema
can lead to an increase in the LPR,51-53 which is associated with
severe reductions in PbtO254 and increases in PET-measured
oxygen extraction fraction.55 An increase in LPR above estab-
lished thresholds (20-25) is associated with poor outcome
after TBI56-58 and SAH,49,59,60 and traditionally is assumed to
indicate tissue hypoxia or ischemia. However, it has proved
difficult to establish the tissue hypoxic threshold for a raised
LPR,61 and it is increasingly apparent that anaerobic glycolysis
may occur from ineffective utilization of delivered oxygen
because of mitochondrial failure, among other causes.33,42 For
example, Vespa et al. compared MD variables with PET-
derived measurement of glucose and oxygen metabolism in
19 patients with severe TBI.42 There was a 25% incidence of
metabolic crisis, defined in this study as LPR greater than 40,
but only a 2.4% incidence of PET-defined ischemia. It was
concluded that these data indicate that the LPR is a reliable
marker of cellular dysfunction and of inadequate substrate
delivery. An elevated LPR therefore may be classified as type I
(ischemic), in which the pyruvate is decreased and there is a
marked increase in lactate, or type II (nonischemic hypergly-
colysis), in which the pyruvate is normal or increased. In a
type I LPR elevation there is oxygen and glucose lack, whereas
in type II LPR elevation there is mitochondrial failure or
failure of effective use of delivered oxygen and glucose.42 In
animal models of TBI, average LPR has a strong correlation
with total injury volume on neuropathologic examination,62
confirming the value of LPR monitoring in the neurocritical
care unit (NCCU).
Markers of Tissue Damage
Failure of cellular metabolism results in disruption of cell
membrane function that then leads to intracellular influx of
calcium, induction of phospholipase, and ultimately to deg-
radation of cell membranes. This results in release of phos-
pholipids and, after enzymatic breakdown, free fatty acids and
glycerol into the brain ECF. Glycerol is reliably recovered from
the ECF and is therefore a useful MD marker of tissue hypoxia
and cell damage.63,64 The degree of the MD glycerol elevation
associated with pathology (e.g., hypoxia or ischemia) may be
dramatic, with four- or eightfold increases recorded in severe
or complete ischemia respectively.31 High levels of interstitial
glycerol are associated with an unfavorable outcome after TBI
and indicate the severity of parenchymal damage.65 Cerebral
MD glycerol concentration typically is elevated in the first 24
hours after severe TBI, presumably as a result of the primary
injury, and then exponentially declines during the ensuring 3
days. Clausen et al. observed that PbtO2 less than 1.3 kPa
(10 mm Hg) and CPP less than 70 mm Hg were associated
with elevated mean cerebral MD glycerol levels, although indi-
vidual episodes below the same thresholds also occurred
without an increase in MD glycerol concentration.63 In this
study mean MD glycerol concentration was similar in patients
with favorable and unfavorable outcomes. The interpretation
of elevated glycerol levels after TBI therefore requires further
validation. Glycerol also is able to leak through a damaged
blood brain barrier (BBB) and cause spuriously high brain
ECF glycerol concentration because of the high levels of
plasma glycerol that occur due to stress-induced triglyceride
degradation or the administration of exogenous glycerol.66 To
help distinguish a compromised BBB effect from a true intra-
cerebral event, systemic glycerol concentration can be simul-
taneously measured via an MD catheter placed in the
abdominal subcutaneous adipose tissue.64
Excitotoxicity
Ischemia, TBI, SAH, and other pathologies can lead to cell
depolarization and release of EAAs such as glutamate and
aspartate. Excitotoxicity is one of several mechanisms impli-
cated in neuronal injury, and there was early enthusiasm
about using MD to measure glutamate because animal studies
demonstrated raised glutamate concentration in global cere-
bral ischaemia67 and TBI.68 Subsequent studies in humans
also demonstrated an association between increased MD glu-
tamate concentration and poor outcome after TBI18 and
SAH.49,59 For example, Staub et al. observed 30-fold increases
in ECF glutamate, aspartate, and GABA in SAH patients with
unfavorable outcome at 3 months compared with those with
favorable outcome.69 The increased concentration of EAA in
patients with poor outcome followed a biphasic course, with
maximal concentrations on the first to second, and seventh,
days after the insult. It is likely that the early increase repre-
sents the degree of initial injury and the later increase the
development of vasospasm-related ischemic injury. Prolonged
elevation of MD glutamate also is observed after severe TBI
and is associated with poor outcome.70 High MD-glutamate
levels have been correlated with clinical events involving
hypoxia, ischemia, reduced PbtO2,40 and low CPP.71 The role of
glutamate in excitotoxicity after TBI has been challenged,72
and there are multiple causes of increased ECF glutamate
concentration after brain injury.10 Nevertheless, recent studies
suggest glutamate levels may provide useful information in
some patients. For example, Timofeev et al. found that MD
glutamate was associated weakly with mortality in 223 severe
TBI patients but that this association disappeared after mul-
tivariable analysis.56 Chamoun et al. observed two patterns of
glutamate elevation after TBI: (1) glutamate levels tended to
normalize over the monitoring period (120 hours); and (2)
glutamate levels tended to increase with time or remain
abnormally elevated.73 Pattern 1 patients had a lower mortal-
ity rate (17.1% versus 39.6%) and a better 6-month func-
tional outcome among survivors (41.2% versus 20.7%)
compared to pattern 2. In addition glutamate levels greater
than 20 mmol/L were associated with a nearly twofold
increase in mortality.

Variability of Measured Metabolites
After brain injury there are wide variations in MD variables
over time between different subjects and within individu-
als.7,24,49,70 These are likely to represent the changing levels of
metabolic activity within the injured brain55 but make it dif-
ficult to interpret single MD measurements or measurement
obtained in isolation. Although “normal” and threshold MD
values are published, cerebral MD must be seen as a trend
monitor, and the information provided by MD should be
interpreted with other measured variables, clinical informa-
tion, or radiologic findings.2
Catheter Placement
MD monitors local tissue biochemistry and reflects meta-
bolic disturbances and neurochemical changes only in the
region of the brain where the catheter is located. Differences
between MD measured variables are observed in areas close
to, and far away from, focal traumatic lesions. For example,
Engstrom et al. found in 22 severe TBI patients that MD
markers of ischemia were significantly different in tissue
adjacent to a parenchymal lesion when compared with
normal tissue in the contralateral hemisphere.74 Vespa et al.
also observed regional differences in brain tissue chemistry
after TBI and confirmed the importance of catheter loca-
tion.75 In this study, there was persistent metabolic dysfunc-
tion in pericontusional tissue, both in baseline values and
trends across time, compared with normal tissue. However,
despite biochemical differences between “perilesional” and
less-injured brain, Timofeev et al.76 observed that both areas
showed a relationship between other physiologic variables
(e.g., ICP, CPP, and PbtO2) and biochemistry. In particular,
reduced CPP or brain oxygen was associated with worse neu-
rochemistry (e.g., increased LPR), although these effects were
greater in perilesional tissue and when cerebrovascular reac-
tivity was abnormal.
There are consensus recommendations on where to place
MD catheters to monitor patients after TBI and SAH.77 The
MD catheter should be placed in “at-risk” tissue (e.g., the area
surrounding a mass lesion after TBI), or the vascular territory
most likely to be affected by vasospasm after SAH. These loca-
tions should allow biochemical changes to be measured in the
area of brain most vulnerable to secondary injury. In diffuse
axonal injury catheter placement in the nondominant frontal
lobe is recommended. However, others suggest that the cath-
eter should always be placed in a normal brain region so that
it can be used to indicate global cerebral metabolism.78
Whether white or gray matter should be monitored is another
variable to consider to interpret MD results. White matter has
a lower metabolic demand than gray matter and receives a
smaller proportion of the cerebral blood flow. Most studies
allude to white matter placement. Commercial MD catheters
have a gold tip so that the catheter position can be confirmed
by computed tomography scan.77
The MD catheter is a parenchymal probe that usually is
inserted though a cranial access device, including single-
lumen bolts designed only for MD catheters or ones that
contain multiple-angled lumens to transmit other parenchy-
mal probes such as PbtO2 or ICP sensors.79 The catheter also
can be placed through a burr hole or under direct vision at
the time of craniotomy. These techniques require the catheter
Section VI—Intracranial Monitoring 361
to be tunneled under the skin and fixed with sutures. Although
the catheter can be inserted to varying depths at any orienta-
tion, it is susceptible to migration or displacement depending
on the security of the fixation.
There always is potential for tissue trauma at the time of
catheter insertion. Animal and human studies demonstrate
that artificially high levels of substrate that result from trauma
or inflammation at the time of insertion usually abate within
the first hour of placement.77 A “run-in” period of at least
1 hour should therefore be allowed before using MD data.
Clinical Applications of
Microdialysis Monitoring
Several lines of clinical evidence suggest that MD may assist
clinical decision making, such as management of cerebral per-
fusion pressure,80 hyperventilation,81 and the appropriateness
of surgical procedures.82 This includes deciding when none-
mergent surgery may be performed after severe TBI. For
example, Yokobori et al.83 examined 25 patients with MD and
several other monitors. They observed that the extracellular
glucose concentration improved, and the value of LPR
decreased 4 days after injury, whereas the average PRx
decreased daily and became negative on the fifth day after
injury. These results suggested that the injured brain may
become “less vulnerable” to secondary insults 4 to 5 days after
injury. Microdialysis also may be used to guide therapies such
as hyperoxia,84 induced hypothermia,85 or normothermia;86,87
help establish optimal hemoglobin88,89, and glucose90,91 levels
for transfusion and glycemic control; and provide information
to help determine prognosis. Finally, MD findings may provide
early warning of bacterial meningitis in conditions such as
poor grade SAH, in which clinical signs may be masked by the
effect of the hemorrhage.92 However, to offer added value over
and above that provided by other intracranial monitors, cere-
bral MD must not only guide treatment, but also do so in a
way that reduces the burden of secondary brain injury and
thereby offers the potential to improve functional outcome in
survivors.
Monitoring Evolution of Brain Injury
Monitoring of biochemical changes in at-risk brain tissue
using MD may provide clinically useful indications of evolv-
ing brain injury at the bedside. Hlatky et al. compared focal
measurements of MD and PbtO2 from brain tissue under an
evacuated subdural hematoma to global measures of ICP, CPP,
and jugular venous oxygen saturation (SjvO2) in patients with
severe TBI.93 Seventeen of the 33 (52%) patients developed
delayed focal secondary injury (contusion or infarction) that
was associated with reduced PbtO2 and LPR elevation in the
absence of global changes in ICP, CPP, or SjvO2. There were
no changes in MD variables and PbtO2 in the 11 patients with
an uneventful clinical course. The remaining 5 patients devel-
oped fatal refractory intracranial hypertension and demon-
strated markedly elevated LPR and glutamate, and reduced
glucose concentration. Other studiers have suggested that
changes in MD markers may precede changes in ICP. For
example, Adamides et al., who studied 14 patients with severe
TBI, observed that elevations in brain lactate, LPR and glyc-
erol often preceded an episode of intracranial hypertension by
362 Section VI—Intracranial Monitoring
more than 2 hours.22 In particular the absolute elevation of
brain lactate was the earliest and strongest predictor of ele-
vated ICP.
Prognostication
Information obtained from MD may help determine progno-
sis in the unconscious patient when clinical examination is not
possible. Several studies in both TBI and SAH have demon-
strated an association with disturbed MD values and worse
clinical condition and outcome. In SAH, disturbed cerebral
metabolism is related to the severity of SAH. Sarrafzadeh et al.
studied 149 SAH patients who underwent surgery; cerebral
metabolism was deranged in poor (IV and V) compared to
good (I-III) grade patients.60 Noteworthy in particular were
significant elevations of LPR, LGR, and glycerol in poor grade
patients. Elevated LPR and glutamate concentration also were
markers of poor outcome at 12 months. After severe TBI,
metabolic changes and alterations in CBF may be more
complex, but consistently low MD glucose levels early after
injury have been associated with poor outcome35 as have
increases in LPR.57,58 These extracellular metabolic markers
may be independently associated with outcome after severe
TBI. For example, Timofeev et al. collected prospective obser-
vational neuromonitoring data from 223 severe TBI patients.56
In a multivariate logistic regression model that used data aver-
aged over the whole monitoring period (median duration 4
days), significant independent factors associated with mortal-
ity included brain glucose, LPR, ICP, cerebrovascular pressure
reactivity index, and age, whereas pyruvate was a significant
independent negative predictor of mortality. Levels of gluta-
mate and glycerol were associated with mortality in univariate
but not multivariate analysis.
Prediction of Secondary Injury
Because cerebral MD measures change at the cellular level, it
is a useful tool to describe molecular events triggered by brain
injury. Brain ECF glutamate and glycerol concentrations cor-
relate with regional cerebral blood flow after SAH, and LPR
shows high sensitivity and specificity for brain energy dys-
function that often is associated with symptoms of ischemia.94
In addition, abnormal MD findings may precede changes in
ICP.22 MD therefore has the potential to detect secondary
brain insults including hypoxia or ischemia before changes
can be identified in the patient’s neurologic status or by more
conventional monitoring techniques such as ICP measure-
ment. For example, Skjoth-Rasmussen et al. observed an isch-
emic MD pattern, defined as an increase in LPR and LGR
followed by an increase in glycerol concentration in 17 of 18
patients who experienced a delayed ischemic neurologic
deficit (DIND) after SAH but in only 3 of 24 who did not.25
The mean delay from the peak of the LPR and LGR to the
occurrence of the DIND was 23 hours (range 4 to 50 hours)
and the mean delay from the onset of a full ischemic pattern
was 11 hours (range –11 hours to + 48 hours). The ability of
MD to predict the occurrence of DIND with high sensitivity
and specificity warrants further investigation. In a recent pro-
spective observational study Belli et al. analyzed the correla-
tion between MD markers metabolic impairment and changes
in ICP after TBI.24 An LPR greater than 25 and glycerol greater
than 100 µmol/L were associated with a significantly greater
risk of imminent intracranial hypertension, with odds ratios
of 9.8 (95% CI 5.8-16.1) and 2.2 (95% CI 1.6-3.8), respec-
tively. An abnormal LPR was predictive of an increase in ICP
above normal levels in 89% of cases. Not all studies have found
evidence for the predictive value of MD. For example, Peerde-
man et al. observed no association between MD glycerol
concentration and secondary events, such as intracranial
hypertension and low CPP, after TBI.65 Glutamate appears to
be a sensitive indicator of impending cerebral ischemia after
SAH95 but appears to have less predictive value after TBI.24
It remains to be seen whether the early warnings provided
by MD measured biochemical changes after TBI and SAH in
some studies can be exploited to expand the window for thera-
peutic intervention and whether this will improve outcome.
Monitoring the Effects of Treatment
on the Injured Brain
Clinical use of MD in the NCCU usually is applied to TBI and
SAH patients, and this has improved understanding of the
many pathophysiologic processes that occur in the injured
brain, particularly when used with other monitors or imaging
techniques (e.g., PET96). In addition, cerebral MD has increased
understanding of how current and potential neurointensive
care treatment strategies affect the injured brain. This can be
of great importance since each therapy can have both benefits
and deleterious side effects and most are best used in a tar-
geted fashion. For example, Oddo et al. have observed that
induced normothermia can improve brain metabolism after
acute brain injury but that when shivering occurs, tissue
hypoxia may occur.86,87 MD therefore is being applied increas-
ingly as a surrogate endpoint to evaluate therapeutic interven-
tional strategies or as a means of choosing an appropriate
management.
Assessment of Adequacy of Cerebral Perfusion
There is a wide range recommended for the “optimal” CPP
after TBI and other brain injuries. Rather than there being a
single CPP level that applies to all patients, the optimal CPP
varies among patients and over time in individual patients;
MD can help identify this level. For example, Nordstrom et al.
studied 50 severe TBI patients using an MD catheter placed in
the at-risk tissue surrounding a focal lesion and a second in
the contralateral frontal lobe (normal tissue).80 The lactate
concentration was higher in the at-risk compared with normal
brain and LPR increased in the at-risk tissue when CPP fell
below 50 mm Hg (see also the example at Fig. 36.4). These
authors concluded that cerebral MD can be used to assess the
safe lower limit of CPP and that CPP management might be
individualized by MD rather than delivered to generic target
values. However, these conclusions have been challenged by
data from a later study.75 Although frequent and sustained
increases in LPR and glutamate were observed in pericontu-
sional but not normal brain tissue of TBI patients, these
abnormalities were not related to CPP alterations. In particu-
lar CPP that decreased to less than 60 mm Hg was not associ-
ated with signs of increased metabolic distress, and it also was
not possible to identify an optimal CPP range using MD.
However, MD markers of ischemia were improved by removal
of mass lesions, suggesting that although dynamic changes in
MD values can occur with changes in CPP, there are persistent

40 Low CPP
30
LPR
20
10
Increased blood pressure
0
00:00
Fig. 36.4  Graph illustrating the changes in the lactate/pyruvate ratio (LPR) in “at-risk” (A) and normal (B) brain during a period of low and normal
cerebral perfusion pressure. The normal range for the LPR is shown by the shaded area. Note the rise in LPR in the at-risk tissue during a period of
cerebral hypoperfusion with normal values measured by the catheter in normal brain tissue. CPP, cerebral perfusion pressure. (Modified with permission
from Tisdall MM, Smith M. Cerebral microdialysis: research technique or clinical tool. Br J Anaesth 2006;97(1):18–25.)
differences in regional brain chemistry that far outweigh the
impact of global measures of perfusion. Although many
studies confirm the presence of a biochemical “penumbral”
zone surrounding focal traumatic brain lesions,75,93 it remains
to be proven whether therapeutic intervention, such as CPP-
directed therapy, can protect this penumbral zone from further
damage.74
Glycemic Control
The association between hyperglycemia and outcome after
acute brain injury is well described. However, the optimal
management of blood glucose and its relationship to brain
glucose is still to be fully elucidated. Manipulation of blood
glucose concentration can alter ECF glucose in both normal97
and injured98 brain. Poor outcome in the acute phase after
severe TBI is associated with high systemic glucose concentra-
tions99 but also with low cerebral MD glucose concentra-
tions.35 Vespa et al. observed that tight systemic glycemic
control (defined in this study as blood glucose concentration
of 5.0-6.7 mmol/L [90-120 mg/dL]) did not improve func-
tional outcome after TBI and was associated with an increased
incidence of MD markers of cellular distress.98 Similar find-
ings are observed in SAH patients.90 Schlenk et al. observed
that insulin infusion was associated with stable blood glucose
concentration but also with a significant decrease in cerebral
glucose.100 The brain glucose decrease is likely associated with
high brain glucose utilization because the LPR and glutamate
did not increase, effectively excluding ischemia as the cause of
the low brain glucose concentration. Other investigators101
have observed that systemic hyperglycemia (>7.8 mmol/L
Section VI—Intracranial Monitoring 363
B
12:00 00:00 12:00
Time (hours)
[>140 mg/dL]) occurs more frequently in patients with DIND
than in asymptomatic patients but is unrelated to cerebral
glucose levels. In this study, both cerebral low-glucose epi-
sodes (<0.6 mmol/L [<10.8 mg/dL]) and high-glucose epi-
sodes (>2.6 mmol/L [>47 mg/dL]) occurred independently of
blood glucose concentration. However, low cerebral glucose
occurred more frequently in symptomatic patients and was
associated with other signs of cellular distress such as increase
in LPR, glutamate, and glycerol concentrations. The combina-
tion of low cerebral glucose and systemic hyperglycemia also
was associated with unfavorable outcome. These findings
merit further investigation and suggest that established targets
for glycemic control after brain injury might not be univer-
sally applicable and that cerebral glucose concentration may
present a target for individualized glucose management.
Further studies are required to identify which patients might
benefit from such intervention. Studies also suggest that more
“moderate” systemic glucose control may be reasonable
because brain glucose increases once systemic glucose is
7.8 mmol/L (140 mg/dL), whereas “tight” glucose control may
increase the risk of cell energy dysfunction in the brain.90,91
Hyperventilation
Hyperventilation (HV) can be used to control increased ICP
after TBI, but to do so requires an additional monitor to
ensure that it does not contribute to brain ischemia. Although
there appears to be no role for prophylactic HV, brief periods
of HV therapy do not appear to worsen brain ischemia or
neurologic outcome, although this may depend on how or
when HV is used. Marion et al. examined the effect of 30
364 Section VI—Intracranial Monitoring
minutes of HV (mean PaCO2 24.6 mm Hg) on the extracel-
lular metabolites associated with ischemia after severe TBI.81
HV trials were performed 24 to 36 hours after injury and again
at 3 to 4 days. MD lactate, pyruvate, and glutamate were mea-
sured before and for 4 hours after HV. At 24 to 36 hours, HV
resulted in 13.7% to 395% increases in MD glutamate in 14
of 20 patients, and increases in LPR of 10.8% to 227% in all
20 patients. Interestingly, these changes were not associated
with a significant decrease in local cerebral blood flow (CBF).
At 3 to 4 days after injury, 10 of 13 patients had an increase
in glutamate, whereas only 3 of 13 had an increase in LPR.
These data indicate that even brief periods of HV during the
first 4 days after TBI can increase markers of cellular ischemia
in selectively vulnerable regions of brain and that HV-induced
changes are much more common early after injury. It remains
to be seen whether these findings can translate into effective
titration of HV guided by changes in MD variables.
Testing New Treatment Interventions
MD also has been used to assess novel therapeutic interven-
tions. There is interest in normobaric hyperoxia (NBH) or
hyperbaric oxygen as a potential treatment strategy after
severe TBI. Some investigators102 but not all103 have observed
that NBH is associated with improved markers of cerebral
metabolism after severe TBI. Two studies have used MD with
other monitoring modalities to investigate the underlying
mechanisms of action of NBH. Nortje et al. compared MD
and PbtO2 to 15O-PET variables within a region of interest
around MD catheters placed in at-risk tissue in 11 patients
with severe TBI.84 MD lactate and pyruvate were unchanged
during NBH and, although the LPR showed a statistically
significant reduction, the magnitude of the reduction was
small. Global 15O-PET variables were unchanged during NBH,
but a universal increase in cerebral metabolic rate of oxygen
was seen in at-risk tissue. These data suggest that NBH results
in significant enhancement of oxidative metabolism in tissue
that is physiologically at risk of nonsurvival. In another study,
MD variables were combined with near infrared spectroscopy
measurement of changes in oxidized cytochrome c oxidase
concentration.104 Cytochrome c oxidase is the terminal elec-
tron acceptor of the mitochondrial transfer chain and there-
fore plays a crucial role in the dynamics of cellular oxygen
utilization and energy production.105 Changes in cytochrome
c oxidase concentration have been validated as a marker of
cellular energy status.106 During short periods of hyperoxia
(FiO2 = 1), MD lactate concentration and LPR were signifi-
cantly decreased, and this was associated with an increase in
the oxidation state of cytochrome c oxidase. This study con-
firms oxidation in brain cellular and mitochondrial redox
state after TBI. The findings of both these studies are consis-
tent with increased aerobic metabolism and imply a preferen-
tial metabolic benefit from hyperoxia. The clinical significance
of these findings needs to be confirmed in future studies.
Research Applications
of Microdialysis
Novel Biomarkers
Any molecule present in the brain ECF that is small enough
to cross the semipermeable dialysis membrane will be
collected in the microdialysate. This opens the door to inves-
tigate novel biomarkers of brain injury; the possible applica-
tions are immense and as yet unexplored.
Brain ECF S100β has been successfully measured in vivo
using MD, and increases in ECF S100β are related to second-
ary events, including intracranial hypertension after TBI and
vasospasm after SAH.107 The 100-kDa MD cutoff catheter
allows for the improved recovery of S100β.28 MD also has been
used to measure brain extracellular N-acetyl aspartate (eNAA)
after severe TBI.108 Levels of eNAA were 34% less in nonsur-
vivors compared with survivors, with a nonrecoverable
decrease in eNAA concentration from day 4 after injury
onward. This was associated with a rise in LPR and glycerol.
Because NAA is synthesized in the mitochondria of neurons,
these results confirm the importance of mitochondrial dys-
function as a contributor to poor outcome following TBI. The
measurement of eNAA also may have potential as a prognostic
marker after TBI and could be used to determine the efficacy
of therapeutic interventions aimed at mitochondrial function.
Severe TBI often is associated with axonal injury and loss.
Neurofilament heavy chains (NF-H) are cytoskeletal proteins
found in axons, and studies show that with a 100-kDa MD
cutoff membrane that two proteolytic breakdown products of
NF-H, NfH(476-986) and NfH(476-1026), can be measured
using cerebral MD with a relative recovery of 20%. In severe
TBI, Petzold et al. observed that the MD NF-H levels were
associated with the mechanism of injury and other physio-
logic parameters and were of prognostic value.27 Other studies
suggest that MD monitoring of brain ECF total tau and
β-amyloid 1-42 proteins that are important biomarkers for
axonal injury and for Alzheimer’s disease may help evaluate
axonal injury after moderate to severe TBI.29
The temporal profile of nitric oxide (NO) metabolite con-
centrations (i.e., nitrite and nitrate ([NOx]), also has been
investigated using MD.109 NOx concentrations demonstrated
a typical temporal profile after SAH of an initial peak followed
by an exponential decay. The decrease in NO metabolites over
time was associated with altered MD-derived energy- or
damage-related compounds and could be related to reduced
NO availability, potentially leading to an imbalance of vaso-
dilatory and vasoconstrictive factors. In preliminary TBI
studies, higher concentrations of NO measured with cerebral
MD are associated with more favorable metabolism.110 Because
NO is a vasodilator, this effect may be associated with increased
blood flow and delivery of oxygen and glucose.
MD catheters with high cutoff membranes allow macro-
molecules to be sampled off-line.16 Changes in extracellular
concentrations of cytokines, chemokines, and neurotrophic
factors have been described after brain injury.111 Multiple
compounds, including interleukin (IL)-1β, IL-6, and IL-8;
macrophage inflammatory protein-1β; vascular endothelial
growth factor; and fibroblast growth factor-2, have been mea-
sured. This illustrates the opportunity to monitor many bio-
chemical events simultaneously in the human brain and may
provide insight into events ranging from aneurysm size to
delayed ischemia,112 although the prognostic significance of
MD-measured inflammatory markers has still to be eluci-
dated.113 In addition MD allows multiple body fluids to be
examined (e.g., plasma, cerebrospinal fluid [CSF], and brain
ECF). For example, Sarrafzadeh et al. observed that in SAH,
cerebral, but not plasma IL-6, levels were associated with
delayed cerebral ischemia (DCI) after SAH.114 Catheters with

high cutoff membranes have the potential to expand MD to
the study of protein chemistry as a routine bedside method
during neurointensive care.115
Drug Delivery
The pharmacokinetics of drugs in plasma has been extensively
studied; however, drug pharmacokinetics and pharmacody-
namics in the human brain has received much less attention
because of the obvious difficulty in sampling. MD can be used
to investigate drug penetration in the human brain and
measure concentration at, or as close as possible to, the site of
action. Many drugs have been studied using MD, and these
are reviewed in detail elsewhere.26 In one study of particular
relevance to neurointensive care, Tisdall et al. investigated the
relationship between brain ECF free phenytoin concentration
measured with MD and plasma free phenytoin concentra-
tion.116 There was no relationship between microdialysate and
plasma free phenytoin concentration, suggesting that in the
context of brain injury, measurement of plasma free phenyt-
oin concentration may not provide an accurate reflection of
regional brain ECF drug concentration. This has implications
for dosing regimens that rely on measurement of plasma
levels. MD also may be used to measure muscle levels of drugs
such as muscle relaxants to help examine the drug use and
effects in patients in the NCCU.117 Finally, it is possible to
engineer the delivery of substances to the brain using MD, and
this has many potential applications in the NCCU and in the
treatment of other disorders such as brain tumors.118-121
Proteomics
A potentially exciting new field is the application of pro-
teomics to MD. Using a proteomic approach based on two-
dimensional gel-electrophoresis and subsequent mass
spectroscopy, Maurer et al. created a protein map for the
global protein expression pattern of the microdialysate in
three patients with thromboembolic stroke.122 Ten proteins
were found exclusively in the microdialysate (but not in CSF)
and indicate the possibility of monitoring disease progression.
In another study, protein expression profiles in microdialysate
samples were measured in SAH patients.123 Differences in
protein expression were detected in those with vasospasm
compared with those without, and the changes in protein
concentrations were detected 3.8 days before the onset of
symptoms. It therefore may become possible to use MD pro-
teomic techniques to identify early markers for vasospasm and
enable selective early therapeutic intervention in this high-risk
group of patients.
Conclusion
Cerebral MD presently is the only method available to measure
brain tissue biochemistry at the bedside, and the introduction
of a commercially available analyzer has made online cerebral
MD monitoring a reality in the NCCU. Its use has provided
insight into the pathophysiology after acute brain injury, and
MD abnormalities are associated with worse clinical condition
and outcome. There also is accumulating evidence that MD
may have the potential to provide early warning of impending
neurologic deterioration, thus allowing timely implementa-
tion of neuroprotective strategies. However, MD reflects only
Section VI—Intracranial Monitoring 365
local tissue biochemistry, and accurate placement of the cath-
eter is crucial. Furthermore, because there are wide variations
in measured variables, trend data are more important than
absolute values. MD is used routinely in a few centers but it
has not yet been introduced into widespread clinical practice.
Although clinical experience is rapidly increasing, carefully
designed prospective studies are required to determine the
value of cerebral MD in the management of brain-injured
patients. However, because of its unique ability to contribute
important information about the processes of secondary
brain injury, MD has the potential to become established as a
key component of multimodal monitoring during neuroin-
tensive care. In addition, MD values may be used as surrogate
endpoints in the early phase clinical trials.
References
1. Hutchinson PJ. Microdialysis in traumatic brain injury: methodology and
pathophysiology. Acta Neurochir Suppl 2005;95:441–5.
2. Tisdall MM, Smith M. Cerebral microdialysis: research technique or clinical
tool. Br J Anaesth 2006;97(1):18–25.
3. Bito L, Davson H, Levin E, et al. The concentrations of free amino acids and
other electrolytes in cerebrospinal fluid, in vivo dialysate of brain, and blood
plasma of the dog. J Neurochem 1966;13(11):1057–67.
4. Delgado JM, DeFeudis FV, Roth RH, et al. Dialytrode for long term
intracerebral perfusion in awake monkeys. Arch Int Pharmacodyn Ther
1972;198(1):9–21.
5. Tossman U, Ungerstedt U. Microdialysis in the study of extracellular levels of
amino acids in the rat brain. Acta Physiol Scand 1986;128(1):9–14.
6. Hillered L, Persson L, Ponten U, et al. Neurometabolic monitoring of the
ischaemic human brain using microdialysis. Acta Neurochir (Wien) 1990;
102(3-4):91–7.
7. Persson L, Hillered L. Chemical monitoring of neurosurgical intensive
care patients using intracerebral microdialysis. J Neurosurg 1992;76(1):
72–80.
8. Goodman JC, Robertson CS. Microdialysis: is it ready for prime time? Curr
Opin Crit Care 2009;15(2):110–17.
9. Tisdall MM, Smith M. Multimodal monitoring in traumatic brain injury:
current status and future directions. Br J Anaesth 2007;99(1):61–7.
10. Hillered L, Vespa PM, Hovda DA. Translational neurochemical research in
acute human brain injury: the current status and potential future for
cerebral microdialysis. J Neurotrauma 2005;22(1):3–41.
11. Johnston AJ, Gupta AK. Advanced monitoring in the neurology intensive
care unit: microdialysis. Curr Opin Crit Care 2002;8(2):121–7.
12. Landolt H, Langemann H. Cerebral microdialysis as a diagnostic tool in
acute brain injury. Eur J Anaesthesiol 1996;13(3):269–78.
13. Lee GJ, Park JH, Park HK. Microdialysis applications in neuroscience. Neurol
Res 2008;30(7):661–8.
14. Peerdeman SM, van Tulder MW, Vandertop WP. Cerebral microdialysis as a
monitoring method in subarachnoid hemorrhage patients, and correlation
with clinical events: a systematic review. J Neurol 2003;250(7):797–805.
15. Hutchinson PJ, O’Connell MT, Nortje J, et al. Cerebral microdialysis
methodology: evaluation of 20 kDa and 100 kDa catheters. Physiol Meas
2005;26(4):423–8.
16. Hillman J, Aneman O, Anderson C, et al. A microdialysis technique for
routine measurement of macromolecules in the injured human brain.
Neurosurgery 2005;56(6):1264–8.
17. Hutchinson PJ, O’Connell MT, Al-Rawi PG, et al. Clinical cerebral
microdialysis: a methodological study. J Neurosurg 2000;93(1):37–43.
18. Langemann H, Alessandri B, Mendelowitsch A, et al. Extracellular levels of
glucose and lactate measured by quantitative microdialysis in the human
brain. Neurol Res 2001;23(5):531–6.
19. Bhatia R, Hashemi P, Razzaq A, et al. Application of rapid-sampling, online
microdialysis to the monitoring of brain metabolism during aneurysm
surgery. Neurosurgery 2006;58(4 Suppl 2):ONS–20.
20. Teasdale GM, Graham DI. Craniocerebral trauma: protection and retrieval
of the neuronal population after injury. Neurosurgery 1998;43(4):723–37.
21. Smith MI. Neurocritical care: has it come of age? Br J Anaesth
2004;93(6):753–5.
22. Adamides AA, Rosenfeldt FL, Winter CD, et al. Brain tissue lactate elevations
predict episodes of intracranial hypertension in patients with traumatic
brain injury. J Am Coll Surg 2009;209(4):531–9.
366 Section VI—Intracranial Monitoring
23. Chen HI, Stiefel MF, Oddo M, et al. Detection of cerebral compromise with
multimodality monitoring in patients with subarachnoid hemorrhage.
Neurosurgery 2011;69:53–63.
24. Belli A, Sen J, Petzold A, Russo S, et al. Metabolic failure precedes
intracranial pressure rises in traumatic brain injury: a microdialysis study.
Acta Neurochir (Wien) 2008;150(5):461–9.
25. Skjoth-Rasmussen J, Schulz M, Kristensen SR, et al. Delayed neurological
deficits detected by an ischemic pattern in the extracellular cerebral
metabolites in patients with aneurysmal subarachnoid hemorrhage.
J Neurosurg 2004;100(1):8–15.
26. Helmy A, Carpenter KL, Hutchinson PJ. Microdialysis in the human brain
and its potential role in the development and clinical assessment of drugs.
Curr Med Chem 2007;14(14):1525–37.
27. Petzold A, Tisdall MM, Girbes AR, et al. In vivo monitoring of neuronal loss
in traumatic brain injury: a microdialysis study. Brain 2011;134(Pt 2):
464–83.
28. Afinowi R, Tisdall M, Keir G, et al. Improving the recovery of S100β protein
in cerebral microdialysis: implications for multimodal monitoring in
neurocritical care. J Neurosci Methods 2009;181(1):95–9.
29. Marklund N, Blennow K, Zetterberg H, et al. Monitoring of brain interstitial
total tau and beta amyloid proteins by microdialysis in patients with
traumatic brain injury. J Neurosurg 2009;110(6):1227–37.
30. Reinstrup P, Stahl N, Mellergard P, et al. Intracerebral microdialysis in
clinical practice: baseline values for chemical markers during wakefulness,
anesthesia, and neurosurgery. Neurosurgery 2000;47(3):701–9.
31. Schulz MK, Wang LP, Tange M, et al. Cerebral microdialysis monitoring:
determination of normal and ischemic cerebral metabolisms in patients with
aneurysmal subarachnoid hemorrhage. J Neurosurg 2000;93(5):808–14.
32. Charalambides C, Sgouros S, Sakas D. Intracerebral microdialysis in children.
Childs Nerv Syst 2010;26(2):215–20.
33 Larach DB, Kofke WA, Le Roux P. Potential non-hypoxic/ischemic causes
of increased cerebral interstitial fluid lactate/pyruvate ratio: a review of
available literature. Neurocrit Care 2011;Feb 19. Epub ahead of print.
34. Nelson DW, Thornquist B, MacCallum RM, et al. Analyses of cerebral
microdialysis in patients with traumatic brain injury: relations to
intracranial pressure, cerebral perfusion pressure and catheter placement.
BMC Med 2011 2;9:21.
35. Vespa PM, McArthur D, O’Phelan K, et al. Persistently low extracellular
glucose correlates with poor outcome 6 months after human traumatic brain
injury despite a lack of increased lactate: a microdialysis study. J Cereb Blood
Flow Metab 2003;23(7):865–77.
36. Hovda DA, Yoshino A, Kawamata T, et al. Diffuse prolonged depression of
cerebral oxidative metabolism following concussive brain injury in the rat:
a cytochrome oxidase histochemistry study. Brain Res 1991;567(1):1–10.
37. Bergsneider M, Hovda DA, Shalmon E, et al. Cerebral hyperglycolysis
following severe traumatic brain injury in humans: a positron emission
tomography study. J Neurosurg 1997;86(2):241–51.
38. Goodman JC, Valadka AB, Gopinath SP, et al. Extracellular lactate and
glucose alterations in the brain after head injury measured by microdialysis.
Crit Care Med 1999;27(9):1965–73.
39. Unterberg AW, Sakowitz OW, Sarrafzadeh AS, et al. Role of bedside
microdialysis in the diagnosis of cerebral vasospasm following aneurysmal
subarachnoid hemorrhage. J Neurosurg 2001;94(5):740–9.
40. Valadka AB, Goodman JC, Gopinath SP, et al. Comparison of brain tissue
oxygen tension to microdialysis-based measures of cerebral ischemia in
fatally head-injured humans. J Neurotrauma 1998;15(7):509–19.
41. Robertson CS, Gopinath SP, Uzura M, et al. Metabolic changes in the brain
during transient ischemia measured with microdialysis. Neurol Res 1998;20
Suppl 1:S91–4.
42. Vespa P, Bergsneider M, Hattori N, et al. Metabolic crisis without brain
ischemia is common after traumatic brain injury: a combined microdialysis
and positron emission tomography study. J Cereb Blood Flow Metab
2005;25(6):763–74.
43. Samuelsson C, Hillered L, Enblad P, et al. Microdialysis patterns in
subarachnoid hemorrhage patients with focus on ischemic events and brain
interstitial glutamine levels. Acta Neurochir (Wien) 2009;151(5):437–46;
discussion 446. Epub 2009, Mar 19.
44. Boumezbeur F, Petersen KF, Cline GW, et al. The contribution of blood
lactate to brain energy metabolism in humans measured by dynamic 13C
nuclear magnetic resonance spectroscopy. J Neurosci 2010;30(42):13983–91.
45. Pellerin L, Magistretti PJ. Food for thought: challenging the dogmas. J Cereb
Blood Flow Metab 2003;23(11):1282–6.
46. Gallagher CN, Carpenter KL, Grice P, et al. The human brain utilizes lactate
via the tricarboxylic acid cycle: a 13C-labelled microdialysis and high-
resolution nuclear magnetic resonance study. Brain 2009;132(10):2839–49.
47. Hutchinson PJ, Al-Rawi PG, O’Connell MT, et al. Monitoring of brain
metabolism during aneurysm surgery using microdialysis and brain
multiparameter sensors. Neurol Res 1999;21(4):352–8.
48. Nilsson OG, Polito A, Saveland H, et al. Are primary supratentorial
intracerebral hemorrhages surrounded by a biochemical penumbra? A
microdialysis study. Neurosurgery 2006;59(3):521–8.
49. Persson L, Valtysson J, Enblad P, et al. Neurochemical monitoring using
intracerebral microdialysis in patients with subarachnoid hemorrhage.
J Neurosurg 1996;84(4):606–16.
50. Persson L, Hillered L. Intracerebal microdialysis. J Neurosurg 2008;85:984–5.
A complete list of references for this chapter can be found online at
www.expertconsult.com.

References
1. Hutchinson PJ. Microdialysis in traumatic brain injury: methodology and
pathophysiology. Acta Neurochir Suppl 2005;95:441–5.
2. Tisdall MM, Smith M. Cerebral microdialysis: research technique or clinical
tool. Br J Anaesth 2006;97(1):18–25.
3. Bito L, Davson H, Levin E, et al. The concentrations of free amino acids
and other electrolytes in cerebrospinal fluid, in vivo dialysate of brain, and
blood plasma of the dog. J Neurochem 1966;13(11):1057–67.
4. Delgado JM, DeFeudis FV, Roth RH, et al. Dialytrode for long term
intracerebral perfusion in awake monkeys. Arch Int Pharmacodyn Ther
1972;198(1):9–21.
5. Tossman U, Ungerstedt U. Microdialysis in the study of extracellular levels
of amino acids in the rat brain. Acta Physiol Scand 1986;128(1):9–14.
6. Hillered L, Persson L, Ponten U, et al. Neurometabolic monitoring of the
ischaemic human brain using microdialysis. Acta Neurochir (Wien)
1990;102(3-4):91–7.
7. Persson L, Hillered L. Chemical monitoring of neurosurgical intensive
care patients using intracerebral microdialysis. J Neurosurg 1992;76(1):
72–80.
8. Goodman JC, Robertson CS. Microdialysis: is it ready for prime time? Curr
Opin Crit Care 2009;15(2):110–17.
9. Tisdall MM, Smith M. Multimodal monitoring in traumatic brain injury:
current status and future directions. Br J Anaesth 2007;99(1):61–7.
10. Hillered L, Vespa PM, Hovda DA. Translational neurochemical research in
acute human brain injury: the current status and potential future for
cerebral microdialysis. J Neurotrauma 2005;22(1):3–41.
11. Johnston AJ, Gupta AK. Advanced monitoring in the neurology intensive
care unit: microdialysis. Curr Opin Crit Care 2002;8(2):121–7.
12. Landolt H, Langemann H. Cerebral microdialysis as a diagnostic tool in
acute brain injury. Eur J Anaesthesiol 1996;13(3):269–78.
13. Lee GJ, Park JH, Park HK. Microdialysis applications in neuroscience.
Neurol Res 2008;30(7):661–8.
14. Peerdeman SM, van Tulder MW, Vandertop WP. Cerebral microdialysis as a
monitoring method in subarachnoid hemorrhage patients, and correlation
with clinical events: a systematic review. J Neurol 2003;250(7):797–805.
15. Hutchinson PJ, O’Connell MT, Nortje J, et al. Cerebral microdialysis
methodology: evaluation of 20 kDa and 100 kDa catheters. Physiol Meas
2005;26(4):423–8.
16. Hillman J, Aneman O, Anderson C, et al. A microdialysis technique for
routine measurement of macromolecules in the injured human brain.
Neurosurgery 2005;56(6):1264–8.
17. Hutchinson PJ, O’Connell MT, Al-Rawi PG, et al. Clinical cerebral
microdialysis: a methodological study. J Neurosurg 2000;93(1):37–43.
18. Langemann H, Alessandri B, Mendelowitsch A, et al. Extracellular levels of
glucose and lactate measured by quantitative microdialysis in the human
brain. Neurol Res 2001;23(5):531–6.
19. Bhatia R, Hashemi P, Razzaq A, et al. Application of rapid-sampling, online
microdialysis to the monitoring of brain metabolism during aneurysm
surgery. Neurosurgery 2006;58(4 Suppl 2):ONS–20.
20. Teasdale GM, Graham DI. Craniocerebral trauma: protection and retrieval
of the neuronal population after injury. Neurosurgery 1998;43(4):723–37.
21. Smith MI. Neurocritical care: has it come of age? Br J Anaesth 2004;93(6):
753–5.
22. Adamides AA, Rosenfeldt FL, Winter CD, et al. Brain tissue lactate
elevations predict episodes of intracranial hypertension in patients with
traumatic brain injury. J Am Coll Surg 2009;209(4):531–9.
23. Chen HI, Stiefel MF, Oddo M, et al. Detection of cerebral compromise with
multimodality monitoring in patients with subarachnoid hemorrhage.
Neurosurgery 2011;69:53–63.
24. Belli A, Sen J, Petzold A, Russo S, et al. Metabolic failure precedes
intracranial pressure rises in traumatic brain injury: a microdialysis study.
Acta Neurochir (Wien) 2008;150(5):461–9.
25. Skjoth-Rasmussen J, Schulz M, Kristensen SR, et al. Delayed neurological
deficits detected by an ischemic pattern in the extracellular cerebral
metabolites in patients with aneurysmal subarachnoid hemorrhage.
J Neurosurg 2004;100(1):8–15.
26. Helmy A, Carpenter KL, Hutchinson PJ. Microdialysis in the human brain
and its potential role in the development and clinical assessment of drugs.
Curr Med Chem 2007;14(14):1525–37.
27. Petzold A, Tisdall MM, Girbes AR, et al. In vivo monitoring of neuronal
loss in traumatic brain injury: a microdialysis study. Brain 2011;
134(Pt 2):464–83.
28. Afinowi R, Tisdall M, Keir G, et al. Improving the recovery of S100β
protein in cerebral microdialysis: implications for multimodal monitoring
in neurocritical care. J Neurosci Methods 2009;181(1):95–9.
Section VI—Intracranial Monitoring 366.e1
29. Marklund N, Blennow K, Zetterberg H, et al. Monitoring of brain
interstitial total tau and beta amyloid proteins by microdialysis in patients
with traumatic brain injury. J Neurosurg 2009;110(6):1227–37.
30. Reinstrup P, Stahl N, Mellergard P, et al. Intracerebral microdialysis in
clinical practice: baseline values for chemical markers during wakefulness,
anesthesia, and neurosurgery. Neurosurgery 2000;47(3):701–9.
31. Schulz MK, Wang LP, Tange M, et al. Cerebral microdialysis monitoring:
determination of normal and ischemic cerebral metabolisms in patients with
aneurysmal subarachnoid hemorrhage. J Neurosurg 2000;93(5):808–14.
32. Charalambides C, Sgouros S, Sakas D. Intracerebral microdialysis in
children. Childs Nerv Syst 2010;26(2):215–20.
33. Larach DB, Kofke WA, Le Roux P. Potential non-hypoxic/ischemic causes
of increased cerebral interstitial fluid lactate/pyruvate ratio: a review of
available literature. Neurocrit Care 2011;Feb 19. Epub ahead of print.
34. Nelson DW, Thornquist B, MacCallum RM, et al. Analyses of cerebral
microdialysis in patients with traumatic brain injury: relations to
intracranial pressure, cerebral perfusion pressure and catheter placement.
BMC Med 2011 2;9:21.
35. Vespa PM, McArthur D, O’Phelan K, et al. Persistently low extracellular
glucose correlates with poor outcome 6 months after human traumatic
brain injury despite a lack of increased lactate: a microdialysis study. J
Cereb Blood Flow Metab 2003;23(7):865–77.
36. Hovda DA, Yoshino A, Kawamata T, et al. Diffuse prolonged depression of
cerebral oxidative metabolism following concussive brain injury in the rat:
a cytochrome oxidase histochemistry study. Brain Res 1991;567(1):1–10.
37. Bergsneider M, Hovda DA, Shalmon E, et al. Cerebral hyperglycolysis
following severe traumatic brain injury in humans: a positron emission
tomography study. J Neurosurg 1997;86(2):241–51.
38. Goodman JC, Valadka AB, Gopinath SP, et al. Extracellular lactate and
glucose alterations in the brain after head injury measured by microdialysis.
Crit Care Med 1999;27(9):1965–73.
39. Unterberg AW, Sakowitz OW, Sarrafzadeh AS, et al. Role of bedside
microdialysis in the diagnosis of cerebral vasospasm following aneurysmal
subarachnoid hemorrhage. J Neurosurg 2001;94(5):740–9.
40. Valadka AB, Goodman JC, Gopinath SP, et al. Comparison of brain tissue
oxygen tension to microdialysis-based measures of cerebral ischemia in
fatally head-injured humans. J Neurotrauma 1998;15(7):509–19.
41. Robertson CS, Gopinath SP, Uzura M, et al. Metabolic changes in the brain
during transient ischemia measured with microdialysis. Neurol Res 1998;
20(Suppl 1):S91–4.
42. Vespa P, Bergsneider M, Hattori N, et al. Metabolic crisis without brain
ischemia is common after traumatic brain injury: a combined microdialysis
and positron emission tomography study. J Cereb Blood Flow Metab 2005;
25(6):763–74.
43. Samuelsson C, Hillered L, Enblad P, et al. Microdialysis patterns in
subarachnoid hemorrhage patients with focus on ischemic events and brain
interstitial glutamine levels. Acta Neurochir (Wien) 2009;151(5):437–46;
discussion 446. Epub 2009, Mar 19.
44. Boumezbeur F, Petersen KF, Cline GW, et al. The contribution of blood
lactate to brain energy metabolism in humans measured by dynamic 13C
nuclear magnetic resonance spectroscopy. J Neurosci
2010;30(42):13983–91.
45. Pellerin L, Magistretti PJ. Food for thought: challenging the dogmas. J
Cereb Blood Flow Metab 2003;23(11):1282–6.
46. Gallagher CN, Carpenter KL, Grice P, et al. The human brain utilizes lactate
via the tricarboxylic acid cycle: a 13C-labelled microdialysis and high-
resolution nuclear magnetic resonance study. Brain 2009;132(10):2839–49.
47. Hutchinson PJ, Al-Rawi PG, O’Connell MT, et al. Monitoring of brain
metabolism during aneurysm surgery using microdialysis and brain
multiparameter sensors. Neurol Res 1999;21(4):352–8.
48. Nilsson OG, Polito A, Saveland H, et al. Are primary supratentorial
intracerebral hemorrhages surrounded by a biochemical penumbra? A
microdialysis study. Neurosurgery 2006;59(3):521–8.
49. Persson L, Valtysson J, Enblad P, et al. Neurochemical monitoring using
intracerebral microdialysis in patients with subarachnoid hemorrhage. J
Neurosurg 1996;84(4):606–16.
50. Persson L, Hillered L. Intracerebal microdialysis. J Neurosurg
2008;85:984–5.
51. Enblad P, Valtysson J, Andersson J, et al. Simultaneous intracerebral
microdialysis and positron emission tomography in the detection of
ischemia in patients with subarachnoid hemorrhage. J Cereb Blood Flow
Metab 1996;6(4):637–44.
52. Stahl N, Mellergard P, Hallstrom A, et al. Intracerebral microdialysis and
bedside biochemical analysis in patients with fatal traumatic brain lesions.
Acta Anaesthesiol Scand 2001;45(8):977–85.
366.e2 Section VI—Intracranial Monitoring
53. Zetterling M, Hallberg L, Hillered L, et al. Brain energy metabolism in
patients with spontaneous subarachnoid hemorrhage and global cerebral
edema. Neurosurgery 2010;66(6):1102–10.
54. Benveniste H. Brain microdialysis. J Neurochem 1989;52(6):1667–79.
55. Hutchinson PJ, Gupta AK, Fryer TF, et al. Correlation between cerebral
blood flow, substrate delivery, and metabolism in head injury: a combined
microdialysis and triple oxygen positron emission tomography study. J
Cereb Blood Flow Metab 2002;22(6):735–45.
56. Timofeev I, Carpenter KL, Nortje J, et al. Cerebral extracellular chemistry
and outcome following traumatic brain injury: a microdialysis study of 223
patients. Brain 2011;134(Pt 2):484–94.
57. Hutchinson PJ, Al-Rawi PG, O’Connell MT, et al. Head injury monitoring
using cerebral microdialysis and Paratrend multiparameter sensors.
Zentralbl Neurochir 2000;61(2):88–94.
58. Zauner A, Doppenberg EM, Woodward JJ, et al. Continuous monitoring of
cerebral substrate delivery and clearance: initial experience in 24 patients
with severe acute brain injuries. Neurosurgery 1997;41(5):1082–91.
59. Kett-White R, Hutchinson PJ, Al-Rawi PG, et al. Adverse cerebral events
detected after subarachnoid hemorrhage using brain oxygen and
microdialysis probes. Neurosurgery 2002;50(6):1213–21.
60. Sarrafzadeh A, Haux D, Kuchler I, et al. Poor-grade aneurysmal
subarachnoid hemorrhage: relationship of cerebral metabolism to outcome.
J Neurosurg 2004;100(3):400–6.
61. Johnston AJ, Steiner LA, Coles JP, et al. Effect of cerebral perfusion pressure
augmentation on regional oxygenation and metabolism after head injury.
Crit Care Med 2005;33(1):189–95.
62. Friess SH, Ralston J, Eucker SA, et al. Neurocritical care monitoring
correlates with neuropathology in a swine model of pediatric traumatic
brain injury. Neurosurgery 2011;Jun 10. Epub ahead of print.
63. Clausen T, Alves OL, Reinert M, et al. Association between elevated brain
tissue glycerol levels and poor outcome following severe traumatic brain
injury. J Neurosurg 2005;103(2):233–8.
64. Hillered L, Valtysson J, Enblad P, et al. Interstitial glycerol as a marker for
membrane phospholipid degradation in the acutely injured human brain.
J Neurol Neurosurg Psychiatry 1998;64(4):486–91.
65. Peerdeman SM, Girbes AR, Polderman KH, et al. Changes in cerebral
interstitial glycerol concentration in head-injured patients: correlation with
secondary events. Intensive Care Med 2003;29(10):1825–8.
66. Konig K, Rickels E, Heissler HE, et al. Artificial elevation of brain tissue
glycerol by administration of a glycerol-containing agent: case report.
J Neurosurg 2001;94(4):621–3.
67. Benveniste H, Drejer J, Schousboe A, et al. Elevation of the extracellular
concentrations of glutamate and aspartate in rat hippocampus during
transient cerebral ischemia monitored by intracerebral microdialysis.
J Neurochem 1984;43(5):1369–74.
68. Nilsson P, Hillered L, Ponten U, et al. Changes in cortical extracellular levels
of energy-related metabolites and amino acids following concussive brain
injury in rats. J Cereb Blood Flow Metab 1990;10(5):631–7.
69. Staub F, Graf R, Gabel P, et al. Multiple interstitial substances measured by
microdialysis in patients with subarachnoid hemorrhage. Neurosurgery
2000;47(5):1106–15.
70. Hutchinson PJ, Al-Rawi PG, O’Connell MT, et al. On-line monitoring of
substrate delivery and brain metabolism in head injury. Acta Neurochir
Suppl 2000;76:431–5.
71. Vespa P, Prins M, Ronne-Engstrom E, et al. Increase in extracellular
glutamate caused by reduced cerebral perfusion pressure and seizures after
human traumatic brain injury: a microdialysis study. J Neurosurg 1998;
89(6):971–82.
72. Obrenovitch TP, Urenjak J. Is high extracellular glutamate the key to
excitotoxicity in traumatic brain injury? J Neurotrauma 1997;14(10):677–98.
73. Chamoun R, Suki D, Gopinath SP, et al. Role of extracellular glutamate
measured by cerebral microdialysis in severe traumatic brain injury. J
Neurosurg 2010;113(3):564–70.
74. Engstrom M, Polito A, Reinstrup P, et al. Intracerebral microdialysis in
severe brain trauma: the importance of catheter location. J Neurosurg 2005;
102(3):460–9.
75. Vespa PM, O’Phelan K, McArthur D, et al. Pericontusional brain tissue
exhibits persistent elevation of lactate/pyruvate ratio independent of
cerebral perfusion pressure. Crit Care Med 2007;35(4):1153–60.
76. Timofeev I, Czosnyka M, Carpenter K, et al. Interaction between brain
chemistry and physiology after traumatic brain injury: impact of
autoregulation and microdialysis catheter location. J Neurotrauma
2011;Apr 13. Epub ahead of print.
77. Bellander BM, Cantais E, Enblad P, et al. Consensus meeting on
microdialysis in neurointensive care. Intensive Care Med
2004;30(12):2166–69.
78. Vespa PM. The implications of cerebral ischemia and metabolic
dysfunction for treatment strategies in neurointensive care. Curr Opin Crit
Care 2006;12(2):119–23.
79. Hutchinson PJ, Hutchinson DB, Barr RH, et al. A new cranial access device
for cerebral monitoring. Br J Neurosurg 2000;14(1):46–8.
80. Nordstrom CH, Reinstrup P, Xu W, et al. Assessment of the lower
limit for cerebral perfusion pressure in severe head injuries by bedside
monitoring of regional energy metabolism. Anesthesiology 2003;98(4):
809–14.
81. Marion DW, Puccio A, Wisniewski SR, et al. Effect of hyperventilation on
extracellular concentrations of glutamate, lactate, pyruvate, and local
cerebral blood flow in patients with severe traumatic brain injury. Crit Care
Med 2002;30(12):2619–25.
82. Boret H, Fesselet J, Meaudre E, et al. Cerebral microdialysis and P(ti)O2 for
neuro-monitoring before decompressive craniectomy. Acta Anaesthesiol
Scand 2006;50(2):252–4.
83. Yokobori S, Watanabe A, Matsumoto G, et al. Time course of recovery from
cerebral vulnerability after severe traumatic brain injury: a microdialysis
study. J Trauma 2011;Apr 15. Epub ahead of print.
84. Nortje J, Coles JP, Timofeev I, et al. Effect of hyperoxia on regional
oxygenation and metabolism after severe traumatic brain injury:
preliminary findings. Crit Care Med 2008;36(1):273–81.
85. Schneweis S, Grond M, Staub F, et al. Predictive value of neurochemical
monitoring in large middle cerebral artery infarction. Stroke 2001;32:
1863–7.
86. Oddo M, Frangos S, Milby A, et al. Induced normothermia attenuates
cerebral metabolic distress in patients with aneurysmal subarachnoid
hemorrhage and refractory fever. Stroke 2009;40(5):1913–6.
87. Oddo M, Frangos S, Maloney-Wilensky E, et al. Effect of shivering on brain
tissue oxygenation during induced normothermia in patients with severe
brain injury. Neurocrit Care 2010;12(1):10–6.
88. Oddo M, Milby A, Chen I, et al. Hemoglobin concentration and cerebral
metabolism in patients with aneurysmal subarachnoid hemorrhage: a
microdialysis study. Stroke 2009;40(4):1275–81.
89. Kurtz P, Schmidt JM, Claassen J, et al. Anemia is associated with metabolic
distress and brain tissue hypoxia after subarachnoid hemorrhage. Neurocrit
Care 2010;13:10–6.
90. Oddo M, Schmidt JM, Carrera C, et al. Impact of tight glycemic control on
cerebral glucose metabolism after severe brain injury: a microdialysis study.
Crit Care Med 2008;36:3233–8.
91. Schlenk F, Vajkoczy P, Sarrafzadeh A. Inpatient hyperglycemia following
aneurysmal subarachnoid hemorrhage: relation to cerebral metabolism and
outcome. Neurocrit Care 2009;11(1):56–63. Epub 2009, May 6.
92. Schlenk F, Frieler K, Nagel A, et al. Cerebral microdialysis for detection of
bacterial meningitis in aneurysmal subarachnoid hemorrhage patients: a
cohort study. Crit Care 2009;13(1):R2. Epub 2009, Jan 20.
93. Hlatky R, Valadka AB, Goodman JC, et al. Evolution of brain tissue injury
after evacuation of acute traumatic subdural hematomas. Neurosurgery
2004;55(6):1318–23.
94. Sarrafzadeh AS, Haux D, Ludemann L, et al. Cerebral ischemia in
aneurysmal subarachnoid hemorrhage: a correlative microdialysis-PET
study. Stroke 2004;35(3):638–43.
95. Nilsson OG, Brandt L, Ungerstedt U, et al. Bedside detection of brain
ischemia using intracerebral microdialysis: subarachnoid hemorrhage and
delayed ischemic deterioration. Neurosurgery 1999;45(5):1176–84.
96. Gallagher CN, Carpenter KL, Grice P, et al. The human brain utilizes lactate
via the tricarboxylic acid cycle: a 13C-labelled microdialysis and high-
resolution nuclear magnetic resonance study. Brain 2009;132(Pt 10):
2839–49.
97. Abi-Saab WM, Maggs DG, Jones T, et al. Striking differences in glucose and
lactate levels between brain extracellular fluid and plasma in conscious
human subjects: effects of hyperglycemia and hypoglycemia. J Cereb Blood
Flow Metab 2002;22(3):271–9.
98. Vespa P, Boonyaputthikul R, McArthur DL, et al. Intensive insulin therapy
reduces microdialysis glucose values without altering glucose utilization or
improving the lactate/pyruvate ratio after traumatic brain injury. Crit Care
Med 2006;34(3):850–6.
99. Jeremitsky E, Omert LA, Dunham CM, et al. The impact of hyperglycemia
on patients with severe brain injury. J Trauma 2005;58(1):47–50.
100. Schlenk F, Graetz D, Nagel A, et al. Insulin-related decrease in cerebral
glucose despite normoglycemia in aneurysmal subarachnoid hemorrhage.
Crit Care 2008;12(1):R9.
101. Schlenk F, Nagel A, Graetz D, et al Hyperglycemia and cerebral glucose in
aneurysmal subarachnoid hemorrhage. Intensive Care Med
2008;34(7):1200–7.

102. Tolias CM, Reinert M, Seiler R, et al. Normobaric hyperoxia–induced
improvement in cerebral metabolism and reduction in intracranial pressure
in patients with severe head injury: a prospective historical cohort-matched
study. J Neurosurg 2004;101(3):435–44.
103. Magnoni S, Ghisoni L, Locatelli M, et al. Lack of improvement in cerebral
metabolism after hyperoxia in severe head injury: a microdialysis study. J
Neurosurg 2003;98(5):952–8.
104. Tisdall MM, Tachtsidis I, Leung TS, et al. Increase in cerebral aerobic
metabolism by normobaric hyperoxia after traumatic brain injury. J
Neurosurg 2008;109:424–32.
105. Richter OM, Ludwig B. Cytochrome c oxidase—structure, function, and
physiology of a redox-driven molecular machine. Rev Physiol Biochem
Pharmacol 2003;147:47–74.
106. Springett R, Wylezinska M, Cady EB, et al. Oxygen dependency of cerebral
oxidative phosphorylation in newborn piglets. J Cereb Blood Flow Metab
2000;20(2):280–9.
107. Sen J, Belli A, Petzold A, Russo S, et al. Extracellular fluid S100β in the
injured brain: a future surrogate marker of acute brain injury? Acta
Neurochir (Wien) 2005;147(8):897–900.
108. Belli A, Sen J, Petzold A, et al. Extracellular N-acetylaspartate depletion in
traumatic brain injury. J Neurochem 2006;96(3):861–9.
109. Sakowitz OW, Wolfrum S, Sarrafzadeh AS, et al. Temporal profiles of
extracellular nitric oxide metabolites following aneurysmal subarachnoid
hemorrhage. Acta Neurochir Suppl 2002;81:351–4.
110. Carpenter KL, Timofeev I, Al-Rawi PG, et al. Nitric oxide in acute brain
injury: a pilot study of NO(x) concentrations in human brain
microdialysates and their relationship with energy metabolism. Acta
Neurochir Suppl 2008;102:207–13.
111. Mellergard P, Aneman O, Sjogren F, et al. Changes in extracellular
concentrations of some cytokines, chemokines, and neurotrophic factors
after insertion of intracerebral microdialysis catheters in neurosurgical
patients. Neurosurgery 2008;62(1):151–7.
Section VI—Intracranial Monitoring 366.e3
112. Hanafy KA, Grobelny B, Fernandez L, et al. Brain interstitial fluid
TNF-alpha after subarachnoid hemorrhage. J Neurol Sci
2010;291(1-2):69–73.
113. Cederberg D, Siesjö P. What has inflammation to do with traumatic brain
injury? Childs Nerv Syst 2010;26(2):221–6.
114. Sarrafzadeh A, Schlenk F, Gericke C, et al. Relevance of cerebral
interleukin-6 after aneurysmal subarachnoid hemorrhage. Neurocrit Care
2010;13(3):339–46.
115. Hillman J, Aneman O, Persson M, et al. Variations in the response of
interleukins in neurosurgical intensive care patients monitored using
intracerebral microdialysis. J Neurosurg 2007;106(5):820–5.
116. Tisdall M, Russo S, Sen J, et al. Free phenytoin concentration measurement
in brain extracellular fluid: a pilot study. Br J Neurosurg 2006;20(5):285–9.
117. Fernández-Candil J, Gambús PL, Trocóniz IF, et al. Feasibility of
microdialysis to determine interstitial rocuronium concentration in the
muscle tissue of anesthetized neurosurgical patients. J Neurosurg
Anesthesiol 2010;22(2):163–9.
118. Redgate ES, Boggs S, Grudziak A, et al. Polyamines in brain tumor therapy.
J Neurooncol 1995;25(2):167–79.
119. Ronquist G, Hugosson R, Sjolander U, et al. Treatment of malignant glioma
by a new therapeutic principle. Acta Neurochir (Wien) 1992;114(1-2):8–11.
120. Marcus HJ, Carpenter KL, Price SJ, et al. In vivo assessment of high-grade
glioma biochemistry using microdialysis: a study of energy-related
molecules, growth factors and cytokines. J Neurooncol 2010;97(1):11–23.
121. Blakeley JO, Olson J, Grossman SA, et al. New Approaches to Brain Tumor
Therapy (NABTT) Consortium. Effect of blood brain barrier permeability
in recurrent high grade gliomas on the intratumoral pharmacokinetics of
methotrexate: a microdialysis study. J Neurooncol 2009;91(1):51–8. Epub
2008, Sep 12.
122. Maurer MH, Berger C, Wolf M, et al. The proteome of human brain
microdialysate. Proteome Sci 2003;1(1):7.
123. Maurer MH, Haux D, Sakowitz OW, et al. Identification of early markers
for symptomatic vasospasm in human cerebral microdialysate after

Brain Temperature
Nino Stocchetti and Elisa R. Zanier
Introduction
Mammals have evolved sophisticated thermoregulatory sys­
tems to maintain body temperature within an optimal range
for cellular biochemical reactions. Under normal conditions,
excluding hibernating mammals, core temperature is main­
tained around 37° C. However, body temperature fluctuates
over the day with an early morning nadir and peak in the
evening and with activity. Temperature also is lower in older
individuals and males and will vary according to the site at
which it is measured. Mean rectal temperature is between
36.7° C and 37.5° C, whereas mean axillary temperatures are
between 35.5° C and 37° C. Minor fluctuations around the
set-point of 37.5° C are well tolerated, but variation that
exceeds a few degrees may be deleterious. Body temperature
results from a balance between mechanisms for heat produc­
tion and mechanisms for heat dispersion. These mechanisms
are coordinated by the central nervous system, specifically, by
the preoptic and anterior hypothalamic nuclei.1 The brain
itself is particularly sensitive to temperature. In general,
hyperthermia is associated with exacerbation of brain injury,
whereas induced hypothermia, or even prevention of fever
may be neuroprotective.2 Brain temperature (BT) may there­
fore be an important therapeutic target; however, in clinical
practice, core body temperature is commonly used as a
surrogate.
This chapter reviews the physiology of brain temperature
regulation, literature regarding the clinical impact of tempera­
ture abnormalities on brain injury, and methods to monitor
brain temperature. In addition, methods to measure body
temperature are briefly reviewed.
Brain Temperature Under
Physiologic Conditions
The Brain as a Heater
All metabolic processes produce heat. For example, glucose
and oxygen are converted via the Krebs cycle into adenosine
triphosphate (ATP), water, and heat. The brain consumes 25%
of total body glucose and 20% of total body oxygen for ATP
production. About 43% of the energy contained in glucose is
converted to ATP, and 67% is dissipated as heat. Given its high
metabolic activity, the brain generates a considerable amount
of heat.
© Copyright 2013 Elsevier Inc. All rights reserved.
Chapter
37  
VI
Maintaining a constant brain temperature depends on a
“tight” balance between heat production and heat dissipa­
tion. The brain dissipates heat primarily to the systemic cir­
culation. Blood entering the brain is at core temperature,
which is lower than brain temperature. Heat is transferred
from the brain to the circulating blood. Increased cerebral
blood flow (CBF) increases heat transfer and therefore brain
cooling. As a result of this heat transfer, the temperature of
venous blood leaving the brain in the internal jugular vein is
greater than core blood temperature.3 On average, the gradi­
ent between brain and body core temperature is modest,
ranging from 0.3° C to 1.1° C,4-8 and usually is higher in the
brain. There is a reasonable correlation between BT and sys­
temic measures of temperature; however, this relationship
changes at the extremes of the physiologic temperature range
such that the brain-body gradient may be reversed with
brain hypothermia or hyperthermia.9 Some consider a rever­
sal of this gradient a poor prognostic sign in traumatic brain
injury (TBI).10
In some species, anatomic variations in brain vasculature
improve cooling efficiency. In dogs, goats, and sheep the inter­
nal carotid artery may be small or absent, and blood circulates
to the skull base via the external carotid artery. Before entering
the circle of Willis, the external carotid artery divides into a
series of small arteries, “the carotid rete,” or “rete mirabilis,”
which flow through the cavernous sinus. During panting,
evaporation in the oral and nasal cavities lowers venous blood
temperature in the cavernous sinus. Heat then is transferred
from arterial blood entering the brain in the carotid rete to
venous blood in the cavernous sinus, thus facilitating heat
removal from the brain.
Heat also may be dispersed outward by direct conduction
through the calvaria,11 but the skull acts as an insulator,
impeding efficient heat exchange. However, when the skull
is opened, for example, after decompressive craniectomy, a
temperature gradient forms between exposed (noninsulated)
brain and ambient temperature; this may result in a BT that
is lower than core temperature.4,12,13
Brain Temperature Is Heterogeneous
There is spatial variation in brain temperature. Structures that
are more metabolically active or deeper have higher tempera­
tures.3 A thermal gradient exists within the brain that allows
heat transfer from deep gray structures to white matter to
367
368 Section VI—Intracranial Monitoring
cortex. This has been confirmed in human volunteers by
proton magnetic resonance spectroscopy.14 Furthermore,
when temperature probes are withdrawn from the lateral ven­
tricle to the brain surface, slight (nonstatistically significant)
temperature differences are detected.15 In patients with hydro­
cephalus, BT has been measured at various depths beneath
the pial surface. Temperature increases gradually as a function
of depth, and the highest temperature is measured in the
ventricle.16
Brain Temperature Is Dynamic
Because the determinants of brain temperature are dynamic,
brain temperature is dynamic. Heat production is determined
by cerebral metabolic rate, and heat removal depends on
regional CBF and arterial blood temperature. Each of these
variables fluctuates. The first BT measurements were obtained
in animals in which large BT fluctuations—up to 2° C to
3° C —were observed on exposure to various environmental
challenges and upon performance of different behaviors and
tasks.3
When systemic temperature increases abruptly, e.g., in the
heat shock, high fever, or strenuous physical exercise, systemic
arterial temperature may increase to a point where the normal
temperature gradient between blood and brain disappears or
is even reversed. Conversely, BT may decrease below core tem­
perature when brain metabolism is inhibited, for example,
profound general anesthesia.3 Another “special case” is the
intentional induction of systemic hypothermia, for example,
after cardiac arrest. External or internal cooling reduces core
and blood temperature, leading to a sharp gradient between
the hot brain and the cooled blood. Specific studies docu­
menting the effect of systemic hypothermia on BT, such as
after cardiac arrest, are limited only to a few cases.17 Variations
in metabolic rate or in CBF in the single patient can change
or reverse the gradient between BT and core temperature. This
may in part explain variable findings on the gradients between
core and brain temperature reported in the literature.1,7,8
Clinical Relevance of
Brain Temperature
Brain Temperature and
Intracranial Pressure
Because the cellular machinery for energy production is sensi­
tive to temperature, changes in BT may have profound conse­
quences. In particular BT, brain metabolic rate, oxygen, and
glucose consumption are linked. Furthermore temperature
elevations have been reported to increase inflammatory pro­
cesses, neuronal excitotoxicity, neurotransmitter release, free
radical production, and intracellular glutamate concentra­
tions, and potentiate the sensitivity of neurons to excitotoxic
injury in the brain.18-20 These may influence cerebral metabolic
rate of oxygen (CMRO2) that is physiologically coupled to
CBF through modulation of vessel diameter. Therefore an
increase CMRO2 results in vasodilation and increased CBF,
which in turn, leads to increased cerebral blood volume. This
increase in blood volume is compensated by egress of cerebro­
spinal fluid from the cranial vault and collapse of low-pressure
cerebral veins. Once these compensatory mechanisms are
exhausted, intracranial pressure (ICP) increases. Clinically, an
increase in ICP may be observed with fever in patients with
acute brain damage.4,21 This ICP increase may be a surrogate
for increased fuel delivery to the brain in the setting of
increased metabolic demand. It has been demonstrated that
during fever there may be preservation of a normal cerebral
lactate/pyruvate ratio (measured by microdialysis) indicating
adequate substrate delivery.21 However, once maximal cerebral
vasodilation occurs, further increases in CMRO2 may not be
compensated by further increases in CBF and there may be
cellular energy crisis. On the other hand microdialysis studies
show that induced normothermia can attenuate the fever-
associated metabolic distress.22
The relationship between BT and ICP is not simple. A clear
relationship does not exist when cumulative BT and ICP data
are pooled.8,23 In a population, ICP depends on several factors
and cannot be predicted solely by BT. However, in certain
individual cases an association between ICP and BT may be
detected.21
Hyperthermia Is Deleterious in
Experimental Models of Brain Injury
In the experimental setting it has been demonstrated that
hyperthermia aggravates neuronal injury after traumatic24 or
ischemic insults to the brain.25-28 Even mild hyperthermia
may exacerbate outcome.29 In animal models of focal cerebral
ischemia, infarct volume varies with temperature30,31 and any
deleterious effect is proportional to the degree of hyperther­
mia. The impact of hyperthermia (and also hypothermia)
on intracerebral hemorrage (ICH) is less clear.32 In an ICH
model in rats, mild to moderate temperature elevations do
not worsen outcome,33 although induced hypothermia can
improve outcome.34 Hyperthermia may damage uninjured
brain cells; cell membranes and mitochondria are selectively
vulnerable.3 Damage is not limited to neurons but also
involves glia and endothelial cells. Consequences of hyper­
thermia include increased glutamate release, disruption of
the blood-brain barrier, increased proinflammatory cytokines
and exacerbation of inflammatory cascades, up-regulation
of various enzymes, and expression of heat shock proteins,
among others.3,25
Fever Is Associated with Worse Outcomes
After Severe Brain Injury
Fever is common after severe brain injury. However, the
precise mechanism by which neurologic injury results in
fever is still to be fully elucidated.35 This may include direct
injury to the thermoregulatory centers in the hypothalamus.
In addition, fever is common when there is intracranial hem­
orrhage and, in particular, intraventricular hemorrhage.36
Other risk factors include the severity of injury, depressed
consciousness, and use of antiepileptic drugs, especially phe­
nytoin.37 Although a causal role has not been established in
humans, an association between hyperthermia and worse
outcome has been demonstrated in patients admitted to
neurocritical care for various pathologies.38-40 For example,
observational clinical studies demonstrate that elevated body
temperature is associated in a dose-dependent way with
longer intensive care unit (ICU) and hospital length of stay,
higher mortality, and worse hospital disposition.38,39 An asso­
ciation between fever burden and worse outcome has been

found in subarachnoid hemorrhage,41 and in acute ischemic
stroke there is a strong relationship between admission tem­
perature and mortality.42 Higher temperature also is associ­
ated with an increased risk of hemorrhagic transformation
of an ischemic stroke.43 In patients with ICH who survive the
first 72 hours after hospital admission, duration of fever is
associated with poor outcome, and fever seems to be an
independent prognostic factor in these patients.44 Fever also
is common after TBI, especially in the more severe cases.45,46
In a meta-analysis of 39 studies that included more than
14,000 patients Greer et al. found a consistent association
between higher body temperature and worse outcomes
across multiple outcome measures.40
Some recreational drugs cause metabolic neural activation
and often increase oxygen consumption and heat production.
This is typical of cocaine, heroin, amphetamine-like sub­
stances, and ketamine. Brain temperature has been studied in
rats during cocaine administration, which confirms a dose-
dependent temperature increase.3 Heat stroke causes hyper­
thermia without an elevation in the hypothalamic set-point.
It is characterized by a core body temperature greater than
40° C due to exposure and is accompanied by hot and dry skin,
nervous system failure (including delirium, convulsions, and
coma), and a systemic inflammatory response. It may culmi­
nate in multiorgan system failure and death.47
Therapeutic Hypothermia
The neuroprotective effects of hypothermia have been dem­
onstrated in the experimental setting in models of ischemia
and trauma. In the clinical setting therapeutic hypothermia
is associated with decreased mortality and improved neuro­
logic outcome after cardiac arrest.48-51 There also may be a
benefit to prehospital hypothermia for cardiac arrest.52 Con­
sequently the European Resuscitation Council and American
Heart Association guidelines recommend the use of hypo­
thermia in patients after cardiac arrest. Hypothermia also is
thought to be associated with better neurologic outcome
after neonatal birth asphyxia. Prophylactic hypothermia as
a method for neuroprotection after TBI in humans has
received considerable interest. Multiple single-center studies
have suggested benefit; however, two large multicenter trials
failed to show efficacy.53,53a A meta-analysis suggested that
prophylactic hypothermia was associated with a nonsignifi­
cant reduction in mortality and a significant improvement
in functional outcome.54,55 Virtually every single center study
has suggested that hypothermia can help control intracranial
hypertension after TBI. There is a consensus, however, that
hypothermia is not necessary if ICP is controlled with other
therapies.56 Two multicenter randomized trials to examine
hypothermia in TBI are in progress in Europe (Eurotherm;
www.eurotherm3235trial.eu/home/index.phtml) and in Aus­
tralia and New Zealand (POLAR; www.anzicrc.monash.org/
polar-rct.html). In addition, it does not appear that routine
pharmacologic or physical cooling provides a benefit in
stroke patients.57 One promising new area for induced hypo­
thermia (core temperature of 32° C to 34° C) is for patients
with acute liver failure and encephalopathy.58
Most studies of therapeutic hypothermia have measured
core body temperature, and data on BT during hypothermia
are scarce. Nordmark et al. recorded brain temperature from
four patients undergoing therapeutic hypothermia (32° C to
Section VI—Intracranial Monitoring 369
34° C) after cardiac arrest and observed brain temperatures as
low as 32° C. There was a decrease in the lactate/pyruvate ratio
that paralleled induction of hypothermia.17 Yoo et al. com­
pared brain and axillary temperature in 54 patients submitted
to normothermia and 9 to hypothermia: brain temperature
was lowered below normal but remained consistently higher
than axillary temperature by approximately 1° C.59
Fever Control and Induced Normothermia
Fever in the neurocritical care unit (NCCU) is very common
and its adverse effect on brain injury is well described. Hence
fever control rather than induced hypothermia may be a rea­
sonable treatment option in the NCCU. Consistent with this
premise it appears that hypothermia does not provide any
added benefit to cerebral protection compared with normo­
thermia among patients undergoing brain surgery.60 Clinical
studies show that prophylactic normothermia using intravas­
cular cooling can reduce significant increases in BT in patients
with severe TBI61 and in a 2010 case control study, Badjatia
et al.62 observed that fever control may be associated with
improved outcome after subarachnoid hemorrhage (SAH).
Several recent advances in temperature management mean
that fever control and maintaining normothermia now are
feasible on a consistent basis in the NCCU. However, there are
still many unanswered questions about its use, such as in
which patients, indications, and when to start fever control.
Brain Temperature Monitoring
Systemic and Core Temperature
Systemic or core temperature is routinely measured in every
patient in the NCCU. From this, brain temperature is inferred.
However, several lines of evidence show that brain tempera­
ture differs from core temperature and cannot be estimated
reliably by monitors placed outside the brain, including the
tympanic membrane.1
It is rare that temperature is monitored using a thermom­
eter. Instead, thermocouple or thermistor devices, infrared
sensors, or liquid crystal devices are used. Thermocouples are
based on the phenomenon that a conductor subject to a tem­
perature gradient will generate a voltage that can be measured.
A thermistor is a semiconductor that varies its resistance
according to temperature. Infrared sensors (e.g., tympanic
membrane thermometers) detect heat energy given off by
radiation; that is, they do not need to be in direct contact with
the object measured. However, this may affect accuracy
because the device picks up temperature from the external ear
rather than the tympanic membrane. Newer tympanic mem­
brane thermometer models are more accurate.63 Thermo­
tropic liquid crystals usually are incorporated into disposable
plastic strips and have a range between 34° C and 40° C. They
are not as accurate as other monitors.
There are several sites from which temperature can be mon­
itored. Broadly speaking the sites can be divided into: (1) core
or central sites that represent the temperature of highly per­
fused organs, for example, esophageal, pulmonary artery,
nasopharynx, or tympanic membrane, the last because of its
proximity to the internal carotid artery; (2) intermediate
sites (e.g., oral [sublingual], rectal, or bladder); or (3) periph­
eral (e.g., forehead skin or axilla). Core or central sites are
370 Section VI—Intracranial Monitoring
preferable because they are less influenced by vasomotor or
thermoregulatory mechanisms. Intermediate sites tend to lag
behind central monitors in a response to temperature changes
and may be influenced by urine flow (bladder) or bacteria in
rectum (rectal), whereas peripheral sensors are affected by
many environmental influences.
Direct Brain Temperature
Monitoring Methods
Brain temperature can be measured directly using probes
placed into cerebral tissue or the lateral ventricles. The first
thermal recording from the brain of an awake animal dates
back to 1860, when the relationship between temperature and
brain activation was noted.3 Intraventricular monitors provide
a global (average) measure of brain temperature, whereas
intraparenchymal temperature monitors (which are often
coupled with brain tissue oxygen sensors) measure regional
temperature. Various commercially available probes exist for
clinical use (http://raumedic.com; www.integra-ls.com). An
earlier version of the Licox (Integra Lifesciences, New Bruns­
wick, NJ) system included separated sensors for brain oxygen
and brain temperature. Newer Licox probes incorporate both
sensors in a single probe. In vitro analysis of the new Licox
probe for simultaneous reading of brain oxygen tension and
BT reveals a tendency to underestimate BT, measuring a mean
of 0.67 plus or minus 0.22° C below the reference experimen­
tal temperature.64 There is ongoing work to be able to incor­
porate BT into a single probe that can monitor multiple
parameters.65
Brain temperature also has been studied noninvasively
using magnetic resonance spectroscopy (MRS)66 and diffusion
magnetic resonance imaging (MRI).67 Focal readings, in
volumes as small as 4 cm3, have been performed with limited
accuracy (approximately 1° C).11 In addition, BT is higher
than that of the extracranial structures with wide variations
observed depending on the regions studied. In contrast, Childs
et al. found that the average BT was 0.5° C lower than tem­
perature measured by MRS at three extracerebral sites, includ­
ing oral cavity, tympanum, and temporal artery.14 In MR-based
studies the mean brain-body (rectal) temperature difference
is about 1.3 plus or minus 0.4° C.68
Brain Temperature
Monitoring Complications
Complications associated with the Licox monitor including
the BT probe are very rare and generally are associated with
insertion of the bolt to hold the probes rather than the monitor
itself.69 Similar to other invasive intracranial monitors, the
main risks include cerebral hemorrhage and infection.
Brain Temperature and
Brain Oxygen Tension
Temperature has an effect on brain tissue oxygen tension.
Given a constant tissue oxygen content, pressure is propor­
tional to temperature, according to the fundamental gas laws.
Therefore oxygen tension increases during hyperthermia and
decreases during hypothermia because of this biophysical
property. Hypoxia itself may induce hypothermia.70 For this
reason the Licox probe used to measure brain oxygen includes
a temperature correction tool in its monitor, whereas the new
Licox probe incorporates BT in its measurements.64 Oxygen
tension may vary with changes of BT for several reasons, such
as CBF changes, shift of the oxyhemoglobin curve, and pos­
sible changes in metabolic rate, among others. Interpretation
of brain oxygen tension during hyperthermia or hypothermia
therefore requires consideration of these factors.8 Similar
changes in oxygen associated with temperature are observed
in normal individuals, for example, during exercise.71
Brain Temperature and Local Cerebral
Blood Flow Measurement
BT can be used to determine local CBF (see Chapter 31). This
CBF monitor (Hemedex) is based on the principle that
thermal conductivity of cerebral cortical tissues varies propor­
tionally with CBF. Therefore measurement of thermal diffu­
sion at the cortical surface may be used to determine CBF. This
monitoring system consists of two small metal plates (therm­
istors) that are inserted into brain tissue. One of the plates is
heated, thereby establishing a temperature gradient between
the two thermistors; this allows CBF to be calculated from the
temperature difference between the plates.72 However, the
system is not able to measure CBF when temperature is greater
than 39° C.
Selective Brain Cooling
How to cool is beyond the scope of this chapter, but a brief
mention is made of selective brain cooling because systemic
hypothermia, the method used in major clinical trials, is
limited by the time to target temperature, the depth of hypo­
thermia, and complications. Selective brain cooling may solve
these issues. A variety of devices, for example, intraventricular
cooling catheters,73 neck pads,74 or cool packs applied to the
dura during a craniotomy75 among many others, have been
used to selectively cool the brain and generally they demon­
strate localized cooling of the brain while maintaining relative
systemic normothermia. The most promise may lie in the use
of the paranasal sinuses76 or be derived from an understanding
of how seals selectively cool their brains to reduce oxygen
consumption in the brain during a dive.77 In addition, implant­
able cooling devices in the field of neuromodulation may
become clinically feasible in the near future.78
Conclusion
Brain temperature is an important and dynamic variable in
patients with acute brain damage. However, BT cannot be
reliably estimated from extracerebral measurements, and it
varies according to brain region. The main determinants of
BT include cerebral metabolic rate and CBF. It is well under­
stood that fever is deleterious to the brain in a variety of
disease states. Induced hypothermia confers neuroprotection
after cardiac arrest, but its role is less certain for other disor­
ders such as TBI, SAH, or stroke. Instead, fever control and
normothermia may be reasonable alternatives.
Brain temperature may be measured invasively, usually in
combination with other intracranial monitoring devices such
as ICP and brain tissue oxygen probes. Brain temperature
monitoring may be of particular value when using therapeutic
temperature modulation in neurocritical care, such as during

controlled normothermia or induced hypothermia, or to
guide the interpretation of other important intracranial
parameters, such as brain tissue oxygenation.
References
1. Childs C. Human brain temperature: regulation, measurement and
relationship with cerebral trauma. Part 1. Br J Neurosurg 2008;4:486–96.
2. Badjatia N. Hyperthermia and fever control in brain injury. Crit Care Med
2009;37(7 Suppl):S250–7.
3. Kiyatkin EA. Brain hyperthermia as physiological and pathological
phenomena. Brain Res Rev 2005;50(1):27–56.
4. Rossi S, Roncati Zanier E, et al. Brain temperature, body core temperature,
and intracranial pressure in acute cerebral damage. J Neurol Neurosurg
Psychiatry 2001;71(4):448–54.
5. Mellergård P. Intracerebral temperature in neurosurgical patients:
intracerebral temperature gradients and relationships to consciousness level.
Surg Neurol 1995;43(1):91–5.
6. Soukup J, Zauner A, Doppenberg EM, et al. The importance of brain
temperature in patients after severe head injury: relationship to intracranial
pressure, cerebral perfusion pressure, cerebral blood flow, and outcome.
J Neurotrauma 2002;19(5):559–71.
7. Rumana CS, Gopinath SP, Uzura M, et al. Brain temperature exceeds
systemic temperature in head-injured patients. Crit Care Med 1988;23:
562–7.
8. Spiotta AM, Stiefel MF, Heuer GG, et al. Brain hyperthermia after traumatic
brain injury does not reduce brain oxygen. Neurosurgery 2008;62(4):864–72.
9. Smith CM, Adelson PD, Chang YF, et al. Brain-systemic temperature
gradient is temperature-dependent in children with severe traumatic brain
injury. Pediatr Crit Care Med 2011;12(4):449–54.
10. Kuo JR, Lo CJ, Wang CC, et al. Measuring brain temperature while
maintaining brain normothermia in patients with severe traumatic brain
injury. J Clin Neurosci 2011;18(8):1059–63. Epub 2011, Jul 1.
11. Corbett R, Laptook A, Weatherall P. Noninvasive measurements of human
brain temperature using volume-localized proton magnetic resonance
spectroscopy. J Cereb Blood Flow Metab 1997;17(4):363–9.
12. Nakagawa K, Hills NK, Kamel H, et al. The effect of decompressive
hemicraniectomy on brain temperature after severe brain injury. Neurocrit
Care 2011;15(1):101–6.
13. Suehiro E, Fujisawa H, Koizumi H, et al. Significance of differences between
brain temperature and core temperature (delta T) during mild hypothermia
in patients with diffuse axonal injury. Neurol Med Chir (Tokyo) 2011;51(8):
551–5.
14. Childs C, Hiltunen Y, Vidyasagar R, et al. Determination of regional brain
temperature using proton magnetic resonance spectroscopy to assess
brain-body temperature differences in healthy human subjects. Magn Reson
Med 2007;57(1):59–66.
15. Fountas KN, Kapsalaki EZ, Feltes CH, et al. Intracranial temperature: is it
different throughout the brain? Neurocrit Care 2004;1(2):195–9.
16. Hirashima Y, Takaba M, Endo S, et al. Intracerebral temperature in patients
with hydrocephalus of varying aetiology. J Neurol Neurosurg Psychiatry
1998;64(6):792–4.
17. Nordmark J, Rubertsson S, Mörtberg E, et al. Intracerebral monitoring in
comatose patients treated with hypothermia after a cardiac arrest. Acta
Anaesthesiol Scand 2009;53(3):289–98.
18. Suehiro E, Fujisawa H, Ito H, et al. Brain temperature modifies glutamate
neurotoxicity in vivo. J Neurotrauma 1999;16:285–97.
19. Dietrich WD, Chatzipanteli K, Vitarbo E, et al. The role of inflammatory
processes in the pathophysiology and treatment of brain and spinal cord
trauma. Acta Neurochir Suppl 2004; 89:69–74.
20. Takagi K, Ginsberg MD, Globus MY, et al. Effect of hyperthermia on
glutamate release in ischemic penumbra after middle cerebral artery
occlusion in rats. Am J Physiol 1994;267:H1770–6.
21. Stocchetti N, Protti A, Lattuada M, et al. Impact of pyrexia on
neurochemistry and cerebral oxygenation after acute brain injury. J Neurol
Neurosurg Psychiatry 2005;76(8):1135–9.
22. Oddo M, Frangos S, Milby A, et al. Induced normothermia attenuates
cerebral metabolic distress in patients with aneurysmal subarachnoid
hemorrhage and refractory fever. Stroke 2009;40(5):1913–6.
23. Huschak G, Hoell T, Wiegel M, et al. Does brain temperature correlate with
intracranial pressure? J Neurosurg Anesthesiol 2008;20(2):105–9.
24. Dietrich WD, Alonso O, Halley M, et al. Delayed posttraumatic brain
hyperthermia worsens outcome after fluid percussion brain injury: a light
and electron microscopic study in rats. Neurosurgery 1996;38(3):533–41.
Section VI—Intracranial Monitoring 371
25. Takagi K, Ginsberg MD, Globus MY, et al. Effect of hyperthermia on
glutamate release in ischemic penumbra after middle cerebral artery
occlusion in rats. Am J Physiol 1994;267(5 Pt 2): H1770–6.
26. Busto R, Dietrich WD, Globus MY, et al. The importance of brain
temperature in cerebral ischemic injury. Stroke 1989;20(8):1113–4.
27. Corbett D, Thornhill J. Temperature modulation (hypothermic and
hyperthermic conditions) and its influence on histological and behavioral
outcomes following cerebral ischemia. Brain Pathol 2000;10(1):145–52.
28. Reglodi D, Somogyvari-Vigh A, Maderdrut JL, et al. Postischemic
spontaneous hyperthermia and its effects in middle cerebral artery occlusion
in the rat. Exper Neurol 2000;163(2):399–407.
29. Sakurai A, Atkins CM, Alonso OF, et al. Mild hyperthermia worsens the
neuropathological damage associated with mild traumatic brain injury in
rats. J Neurotrauma 2011;25. Epub ahead of print.
30. Kim Y, Busto R, Dietrich WD, et al. Delayed postischemic hyperthermia in
awake rats worsens the histopathological outcome of transient focal cerebral
ischemia. Stroke 1996;27(12):2274–81.
31. Noor R, Wang CX, Shuaib A. Effects of hyperthermia on infarct volume
in focal embolic model of cerebral ischemia in rats. Neurosci Lett 2003;
349(2):130–2.
32. MacLellan CL, Clark DL, Silasi G, et al. Use of prolonged hypothermia to
treat ischemic and hemorrhagic stroke. J Neurotrauma 2009;26(3):313–23.
33. MacLellan CL, Colbourne F. Mild to moderate hyperthermia does not
worsen outcome after severe intracerebral hemorrhage in rats. J Cereb Blood
Flow Metab 2005;25:1020–9.
34. MacLellan CL, Davies LM, Fingas MS, et al. The influence of hypothermia
on outcome after intracerebral hemorrhage in rats. Stroke 2006;37(5):1266–
70. Epub 2006, Mar 30.
35. Thompson HJ. Elevated body temperature in the neuroscience intensive care
unit. Crit Care Med 2005;33:1672.
36. Commichau C, Scarmeas N, Mayer SA. Risk factors for fever in the
neurologic intensive care unit. Neurology 2003;60:837–41.
37. Rabinstein AA, Sandhu K. Non-infectious fever in the neurological intensive
care unit: incidence, causes and predictors. J Neurol Neurosurg Psychiatry
2007;78:1278–80.
38. Kilpatrick MM, Lowry DW, Firlik AD, et al. Hyperthermia in the
neurosurgical intensive care unit. Neurosurgery 2000;47(4):850–6.
39. Diringer MN, Reaven NL, Funk SE, et al. Elevated body temperature
independently contributes to increased length of stay in neurologic intensive
care unit patients. Crit Care Med 2004;32(7):1489–95.
40. Greer DM, Funk SE, Reaven NL, et al. Impact of fever on outcome in
patients with stroke and neurologic injury: a comprehensive meta-analysis.
Stroke 2008;39(11):3029–35.
41. Naidech AM, Bendok BR, Bernstein RA, et al. Fever burden and functional
recovery after subarachnoid hemorrhage. Neurosurgery 2008;63(2):212–8.
42. Kammersgaard LP, Jorgensen HS, Rungby JA, et al. Admission body
temperature predicts long-term mortality after acute stroke: the Copenhagen
Stroke Study. Stroke 2002;33(7):1759–62.
43. Leira R, Sobrino T, Blanco M, et al. A higher body temperature is associated
with haemorrhagic transformation in patients with acute stroke untreated
with recombinant tissue-type plasminogen activator (rtPA). Clin Sci (Lond)
2012;122(3):13–9.
44. Schwarz S, Hafner K, Aschoff A, et al. Incidence and prognostic significance
of fever following intracerebral hemorrhage. Neurology 2001;54(2):354–61.
45. Soukup J, Zauner A, Doppenberg EM, et al. The importance of brain
temperature in patients after severe head injury: relationship to intracranial
pressure, cerebral perfusion pressure, cerebral blood flow, and outcome.
J Neurotrauma 2002;19(5):559–71.
46. Stocchetti N, Rossi S, Roncati Zanier E, et al. Pyrexia in head-injured
patients admitted to intensive care. Intensive Care Med 2002;28(11):1555–62.
47. Bouchama A, Knochel JP. Heat stroke. N Engl J Med 2002;346:1978–88.
48. Hypothermia after Cardiac Arrest Study Group. Mild therapeutic
hypothermia to improve the neurologic outcome after cardiac arrest. N Engl
J Med 2002;346(8): 49–56.
49. Bernard SA, Gray TW, Buist MD, et al. Treatment of comatose survivors of
out-of-hospital cardiac arrest with induced hypothermia. N Engl J Med
2002;346(8):557–63.
50. Holzer M, Bernard S, Hachimi-Idrissi S, et al. Hypothermia for
neuroprotection after cardiac arrest: systematic review and individual patient
data meta-analysis. Crit Care Med 2005;33(2):414–8.
A complete list of references for this chapter can be found online at
www.expertconsult.com.

References
1. Childs C. Human brain temperature: regulation, measurement and
relationship with cerebral trauma. Part 1. Br J Neurosurg 2008;4:486–96.
2. Badjatia N. Hyperthermia and fever control in brain injury. Crit Care Med
2009;37(7 Suppl):S250–7.
3. Kiyatkin EA. Brain hyperthermia as physiological and pathological
phenomena. Brain Res Rev 2005;50(1):27–56.
4. Rossi S, Roncati Zanier E, et al. Brain temperature, body core temperature,
and intracranial pressure in acute cerebral damage. J Neurol Neurosurg
Psychiatry 2001;71(4):448–54.
5. Mellergård P. Intracerebral temperature in neurosurgical patients:
intracerebral temperature gradients and relationships to consciousness level.
Surg Neurol 1995;43(1):91–5.
6. Soukup J, Zauner A, Doppenberg EM, et al. The importance of brain
temperature in patients after severe head injury: relationship to intracranial
pressure, cerebral perfusion pressure, cerebral blood flow, and outcome.
J Neurotrauma 2002;19(5):559–71.
7. Rumana CS, Gopinath SP, Uzura M, et al. Brain temperature exceeds
systemic temperature in head-injured patients. Crit Care Med 1988;23:
562–7.
8. Spiotta AM, Stiefel MF, Heuer GG, et al. Brain hyperthermia after traumatic
brain injury does not reduce brain oxygen. Neurosurgery 2008;62(4):
864–72.
9. Smith CM, Adelson PD, Chang YF, et al. Brain-systemic temperature
gradient is temperature-dependent in children with severe traumatic brain
injury. Pediatr Crit Care Med 2011;12(4):449–54.
10. Kuo JR, Lo CJ, Wang CC, et al. Measuring brain temperature while
maintaining brain normothermia in patients with severe traumatic brain
injury. J Clin Neurosci 2011;18(8):1059–63. Epub 2011, Jul 1.
11. Corbett R, Laptook A, Weatherall P. Noninvasive measurements of human
brain temperature using volume-localized proton magnetic resonance
spectroscopy. J Cereb Blood Flow Metab 1997;17(4):363–9.
12. Nakagawa K, Hills NK, Kamel H, et al. The effect of decompressive
hemicraniectomy on brain temperature after severe brain injury. Neurocrit
Care 2011;15(1):101–6.
13. Suehiro E, Fujisawa H, Koizumi H, et al. Significance of differences between
brain temperature and core temperature (delta T) during mild hypothermia
in patients with diffuse axonal injury. Neurol Med Chir (Tokyo) 2011;51(8):
551–5.
14. Childs C, Hiltunen Y, Vidyasagar R, et al. Determination of regional brain
temperature using proton magnetic resonance spectroscopy to assess
brain-body temperature differences in healthy human subjects. Magn Reson
Med 2007;57(1):59–66.
15. Fountas KN, Kapsalaki EZ, Feltes CH, et al. Intracranial temperature: is it
different throughout the brain? Neurocrit Care 2004;1(2):195–9.
16. Hirashima Y, Takaba M, Endo S, et al. Intracerebral temperature in patients
with hydrocephalus of varying aetiology. J Neurol Neurosurg Psychiatry
1998;64(6):792–4.
17. Nordmark J, Rubertsson S, Mörtberg E, et al. Intracerebral monitoring in
comatose patients treated with hypothermia after a cardiac arrest. Acta
Anaesthesiol Scand 2009;53(3):289–98.
18. Suehiro E, Fujisawa H, Ito H, et al. Brain temperature modifies glutamate
neurotoxicity in vivo. J Neurotrauma 1999;16:285–97.
19. Dietrich WD, Chatzipanteli K, Vitarbo E, et al. The role of inflammatory
processes in the pathophysiology and treatment of brain and spinal cord
trauma. Acta Neurochir Suppl 2004; 89:69–74.
20. Takagi K, Ginsberg MD, Globus MY, et al. Effect of hyperthermia on
glutamate release in ischemic penumbra after middle cerebral artery
occlusion in rats. Am J Physiol 1994;267:H1770–6.
21. Stocchetti N, Protti A, Lattuada M, et al. Impact of pyrexia on
neurochemistry and cerebral oxygenation after acute brain injury. J Neurol
Neurosurg Psychiatry 2005;76(8):1135–9.
22. Oddo M, Frangos S, Milby A, et al. Induced normothermia attenuates
cerebral metabolic distress in patients with aneurysmal subarachnoid
hemorrhage and refractory fever. Stroke 2009;40(5):1913–6.
23. Huschak G, Hoell T, Wiegel M, et al. Does brain temperature correlate with
intracranial pressure? J Neurosurg Anesthesiol 2008;20(2):105–9.
24. Dietrich WD, Alonso O, Halley M, et al. Delayed posttraumatic brain
hyperthermia worsens outcome after fluid percussion brain injury: a light
and electron microscopic study in rats. Neurosurgery 1996;38(3):533–41.
25. Takagi K, Ginsberg MD, Globus MY, et al. Effect of hyperthermia on
glutamate release in ischemic penumbra after middle cerebral artery
occlusion in rats. Am J Physiol 1994;267(5 Pt 2): H1770–6.
26. Busto R, Dietrich WD, Globus MY, et al. The importance of brain
temperature in cerebral ischemic injury. Stroke 1989;20(8):1113–4.
Section VI—Intracranial Monitoring 371.e1
27. Corbett D, Thornhill J. Temperature modulation (hypothermic and
hyperthermic conditions) and its influence on histological and behavioral
outcomes following cerebral ischemia. Brain Pathol 2000;10(1):145–52.
28. Reglodi D, Somogyvari-Vigh A, Maderdrut JL, et al. Postischemic
spontaneous hyperthermia and its effects in middle cerebral artery occlusion
in the rat. Exper Neurol 2000;163(2):399–407.
29. Sakurai A, Atkins CM, Alonso OF, et al. Mild hyperthermia worsens the
neuropathological damage associated with mild traumatic brain injury in
rats. J Neurotrauma 2011;25. Epub ahead of print.
30. Kim Y, Busto R, Dietrich WD, et al. Delayed postischemic hyperthermia in
awake rats worsens the histopathological outcome of transient focal cerebral
ischemia. Stroke 1996;27(12):2274–81.
31. Noor R, Wang CX, Shuaib A. Effects of hyperthermia on infarct volume in
focal embolic model of cerebral ischemia in rats. Neurosci Lett 2003;349(2):
130–2.
32. MacLellan CL, Clark DL, Silasi G, et al. Use of prolonged hypothermia to
treat ischemic and hemorrhagic stroke. J Neurotrauma 2009;26(3):313–23.
33. MacLellan CL, Colbourne F. Mild to moderate hyperthermia does not
worsen outcome after severe intracerebral hemorrhage in rats. J Cereb Blood
Flow Metab 2005;25:1020–9.
34. MacLellan CL, Davies LM, Fingas MS, et al. The influence of hypothermia
on outcome after intracerebral hemorrhage in rats. Stroke 2006;37(5):1266–
70. Epub 2006, Mar 30.
35. Thompson HJ. Elevated body temperature in the neuroscience intensive care
unit. Crit Care Med 2005;33:1672.
36. Commichau C, Scarmeas N, Mayer SA. Risk factors for fever in the
neurologic intensive care unit. Neurology 2003;60:837–41.
37. Rabinstein AA, Sandhu K. Non-infectious fever in the neurological intensive
care unit: incidence, causes and predictors. J Neurol Neurosurg Psychiatry
2007;78:1278–80.
38. Kilpatrick MM, Lowry DW, Firlik AD, et al. Hyperthermia in the
neurosurgical intensive care unit. Neurosurgery 2000;47(4):850–6.
39. Diringer MN, Reaven NL, Funk SE, et al. Elevated body temperature
independently contributes to increased length of stay in neurologic intensive
care unit patients. Crit Care Med 2004;32(7):1489–95.
40. Greer DM, Funk SE, Reaven NL, et al. Impact of fever on outcome in
patients with stroke and neurologic injury: a comprehensive meta-analysis.
Stroke 2008;39(11):3029–35.
41. Naidech AM, Bendok BR, Bernstein RA, et al. Fever burden and functional
recovery after subarachnoid hemorrhage. Neurosurgery 2008;63(2):212–8.
42. Kammersgaard LP, Jorgensen HS, Rungby JA, et al. Admission body
temperature predicts long-term mortality after acute stroke: the Copenhagen
Stroke Study. Stroke 2002;33(7):1759–62.
43. Leira R, Sobrino T, Blanco M, et al. A higher body temperature is associated
with haemorrhagic transformation in patients with acute stroke untreated
with recombinant tissue-type plasminogen activator (rtPA). Clin Sci (Lond)
2012;122(3):13–9.
44. Schwarz S, Hafner K, Aschoff A, et al. Incidence and prognostic significance
of fever following intracerebral hemorrhage. Neurology 2001;54(2):354–61.
45. Soukup J, Zauner A, Doppenberg EM, et al. The importance of brain
temperature in patients after severe head injury: relationship to intracranial
pressure, cerebral perfusion pressure, cerebral blood flow, and outcome.
J Neurotrauma 2002;19(5):559–71.
46. Stocchetti N, Rossi S, Roncati Zanier E, et al. Pyrexia in head-injured
patients admitted to intensive care. Intensive Care Med 2002;28(11):1555–62.
47. Bouchama A, Knochel JP. Heat stroke. N Engl J Med 2002;346:1978–88.
48. Hypothermia after Cardiac Arrest Study Group. Mild therapeutic
hypothermia to improve the neurologic outcome after cardiac arrest. N Engl
J Med 2002;346(8): 49–56.
49. Bernard SA, Gray TW, Buist MD, et al. Treatment of comatose survivors of
out-of-hospital cardiac arrest with induced hypothermia. N Engl J Med
2002;346(8):557–63.
50. Holzer M, Bernard S, Hachimi-Idrissi S, et al. Hypothermia for
neuroprotection after cardiac arrest: systematic review and individual patient
data meta-analysis. Crit Care Med 2005;33(2):414–8.
51. Sagalyn E, Band RA, Gaieski DF, et al. Therapeutic hypothermia after cardiac
arrest in clinical practice: review and compilation of recent experiences. Crit
Care Med 2009;37(7 Suppl):S223–6.
52. Cullen D, Augenstine D, Kaper L, et al. Therapeutic hypothermia initiated in
the pre-hospital setting: a meta-analysis. Adv Emerg Nurs J 2011;33(4):
314–21.
53. Clifton GL, Miller ER, Choi SC, et al. Lack of effect of induction of
hypothermia after acute brain injury. N Engl J Med 2001;344(8):556–63.
53a.  Clifton GL, Valadka A, Zygun D, et al. Very early hypothermia induction in
patients with severe brain injury (the National Acute Brain Injury Study:
Hypothermia II): a randomised trial. Lancet Neurol 2011;10(2):131–9.
371.e2 Section VI—Intracranial Monitoring
54. Mcilvoy LH. The effect of hypothermia and hyperthermia on acute brain
injury. AACN Clin Issues 2005;16(4): 488–500.
55. Polderman KH. Induced hypothermia and fever control for prevention and
treatment of neurological injuries. Lancet 2008;371(9628):1955–69.
56. Childs C, Wieloch T, Lecky F, et al. Report of a consensus meeting on human
brain temperature after severe traumatic brain injury: its measurement and
management during pyrexia. Front Neurol 2010;1:146.
57. Den Hertog HM, van der Worp HB, Tseng MC, et al. Cooling therapy for
acute stroke. Cochrane Database Syst Rev 2009;(1):CD001247.
58. Rabinstein AA. Treatment of brain edema in acute liver failure. Curr Treat
Options Neurol 2010;12(2):129–41.
59. Yoo DS, Kim DS, Park CK, et al. Significance of temperature difference
between cerebral cortex and axilla in patients under hypothermia
management. Acta Neurochir Suppl 2002;81:85–7.
60. Milani WR, Antibas PL, Prado GF. Cooling for cerebral protection during
brain surgery. Cochrane Database Syst Rev 2011;(10):CD006638.
61. Fischer M, Lackner P, Beer R, et al. Keep the brain cool—endovascular
cooling in patients with severe traumatic brain injury: a case series study.
Neurosurgery 2011;68(4):867–73; discussion 873.
62. Badjatia N, Fernandez L, Schmidt JM, et al. Impact of induced
normothermia on outcome after subarachnoid hemorrhage: a case-control
study. Neurosurgery 2010;66(4):696–700; discussion 700–1.
63. Bock M, Hohlfeld U, von Engeln K, et al. The accuracy of a new infrared ear
thermometer in patients undergoing cardiac surgery. Can J Anaesth 2005;
52(10):1083–7.
64. Stewart C, Haitsma I, Zador Z, et al. The new Licox combined brain tissue
oxygen and brain temperature monitor: assessment of in vitro accuracy and
clinical experience in severe traumatic brain injury. Neurosurgery 2008;
63(6):1159–65.
65. Li C, Wu PM, Wu Z, et al. Brain temperature measurement: a study of in
vitro accuracy and stability of smart catheter temperature sensors. Biomed
Microdevices 2011;Sep 28. Epub ahead of print.
66. Murakami T, Ogasawara K, Yoshioka Y, et al. Brain temperature measured by
using proton MR spectroscopy predicts cerebral hyperperfusion after carotid
endarterectomy. Radiology 2010;256(3):924–31.
67. Kozak LR, Bango M, Szabo M, et al. Using diffusion MRI for measuring the
temperature of cerebrospinal fluid within the lateral ventricles. Acta Paediatr
2010;99(2):237–43. Epub 2009, Oct 20.
68. Covaciu L, Rubertsson S, Ortiz-Nieto F, et al. Human brain MR spectroscopy
thermometry using metabolite aqueous-solution calibrations. J Magn Reson
Imaging 2010;31(4):807–14.
69. Maloney-Wilensky E, Gracias V, Itkin A, et al. Brain tissue oxygen and
outcome after severe traumatic brain injury: a systematic review. Crit Care
Med 2009;37(6):2057–63.
70. Osaka T. Hypoxia-induced hypothermia mediated by noradrenaline and
nitric oxide in the rostromedial preoptic area. Neuroscience 2011;179:170–8.
Epub 2011, Jan 28.
71. Rasmussen P, Nybo L, Volianitis S, et al. Cerebral oxygenation is reduced
during hyperthermic exercise in humans. Acta Physiol (Oxf)
2010;199(1):63–70. Epub 2010, Jan 25.
72. Bhatia A, Gupta AK. Neuromonitoring in the intensive care unit. I.
Intracranial pressure and cerebral blood flow monitoring. Intensive Care
Med 2007;33(7):1263–71.
73. Moomiaie RM, Gould G, Solomon D, et al. Novel intracranial brain cooling
catheter to mitigate brain injuries. J Neurointerv Surg 2011;Jun 14. Epub
ahead of print.
74. Mink S, Schwarz U, Mudra R, et al. Treatment of resistant fever: new method
of local cerebral cooling. Neurocrit Care 2011;15(1):107–12.
75. King C, Robinson T, Dixon CE, et al. Brain temperature profiles during
epidural cooling with the ChillerPad in a monkey model of traumatic brain
injury. J Neurotrauma 2010;27(10):1895–903. Epub 2010, Sep 16.
76. Gallup AC, Hack GD. Human paranasal sinuses and selective brain cooling:
a ventilation system activated by yawning? Med Hypotheses 2011;77(6):970–
3. Epub 2011, Sep 8.
77. Blix AS, Walløe L, Messelt EB, et al. Selective brain cooling and its vascular
basis in diving seals. J Exp Biol 2010;213(Pt 15):2610–6.
78. Fujii M, Fujioka H, Oku T, et al. Application of focal cerebral cooling for the
treatment of intractable epilepsy. Neurol Med Chir (Tokyo)
2010;50(9):839–44.
VII
Chapter
38  
Device Development
Gerald P. Roston and Brandon von Tobel
Introduction
In times past, tribal medicine men jealously guarded their
secret knowledge, typically passing it from father to son,
thereby protecting the family’s status and importance. Over
the centuries mankind has come to realize that broadly sharing
this critical knowledge benefits society. As a result, in today’s
society more than 1 million people worldwide study medicine.
Interestingly, the learning process is not radically different
today than in years past: students apprentice under those with
greater experience and learn by reading and practicing the
taught skills.
Although this process is quite adequate to teach skills, it falls
short with the ability to produce devices or medicines. Con-
sider the carnage that may ensue if people were provided a
recipe to make aspirin, or a pacemaker, or a monitor rather
than being able to buy one whose efficacy, or functionality, is
well understood.1 The key word is “buy,” which implies the
existence of an entity willing to sell, possibly on a large scale.
This is the focus of this chapter, namely, how to translate an
idea for a medical product from a notion to a salable product.
Because this textbook is about monitoring, the focus of this
chapter is on medical devices rather than pharmaceutical
compounds. Although much of the material presented applies
equally to both, the time and expense associated with drug
development are typically considerably greater than that for
medical devices. In addition, the business models may be dif-
ferent. It is not uncommon for a pharmaceutical start-up to
be acquired before market introduction but after certain mile-
stones are achieved, whereas medical device companies typi-
cally are not acquired until the product has successfully
launched.
Much of the material in this chapter discusses the business
and financial aspects of bringing a product to market, because
most medical practitioners do not have a business back-
ground. By contrast, many medical practitioners have a scien-
tific or engineering background and so only a general overview
of the product development life cycle and some key develop-
ment activities are discussed. However, the medical device
development process has become more complex over the years
in large part because of regulatory requirements, business
decisions (reimbursement), and technology advances. This
development process has recently been detailed2 (Table 38.1)
and practical ways to address specific challenges have been
detailed in workshops of academia, government, and industry
representatives.3 Several universities around the world also
have developed dedicated educational programs in life science
technology innovation.4 In addition, most books that address
374
medical device development focus on the technical aspects of
the process.5-8 The more business-oriented focus of this
chapter complements these other sources.
Developing Medical Products
For new or established companies, the decision to pursue the
development of a new product is never taken lightly because
a significant commitment of the company’s resources is
required. To bring a new drug to market, for example, may
require the better part of a decade and cost more than $500
million. Therefore one of the most important aspects of the
product development process is to determine if there is a need
for the product. The need assessment is revisited frequently
during the development process, whose steps are (1) idea gen-
eration, (2) conceptual design, (3) detail design, (4) clinical
trials, and (5) manufacturing. This section outlines the activi-
ties that occur during each step of this process.9
Idea Generation
All products available for purchase start as an idea in some-
one’s head. The original idea may have sourced from an engi-
neer who realized that some technology could be of value to
a medical practitioner or the idea may have sourced from a
clinician who identified a need during the course of practice.
These two camps are referred to as “technology push” and
“market pull.” Technology push typically requires more time
and effort to fully develop, but often may lead to a broad range
of products, often more than imagined by the original inves-
tor. Market pull ideas typically are accepted more quickly into
the marketplace, but usually do not lead to a broad suite of
follow-on products.
Once a product idea has been formed, the concept must be
tested. For medical devices the first tests performed are risk
analyses—does the use of the product pose greater risk to the
user than the problem it is intended to address? Assuming that
the product does not pose unreasonable risk, the next task is
to determine if the product meets customer needs. There are
several methods to perform this assessment.10
The easiest method, but the one that provides the least valu-
able information, is to perform secondary research, that is, to
analyze published documents to understand the market.
Although secondary research is important, it cannot truly
answer the question, “If I make it, will they buy it?” To learn
the answer to this question, the best approach is to ask poten-
tial customers, in particular, those with purchase authority. To
© Copyright 2013 Elsevier Inc. All rights reserved.
 Section VII—Computers, Engineering, and the Future 375
Table 38.1  The Five Major Phases
and Decision Gates for Medical
Device Development
This process is applicable to a broad range of medical
technologies and may be used to develop sophisticated
premarket approval and premarket notification (510[k])
devices and less sophisticated devices that may be exempt
from most regulatory requirements.
• Phase I/gate 1: Initiation, opportunity, and risk analysis
• Phase II/gate 2: Formulation, concept, and feasibility
• Phase III/gate 3: Design, development, verification, and
validation
• Phase IV/gate 4: Final validation and product launch
preparation
• Phase V: Product launch and post-launch assessment
From Pietzsch JB, Shluzas LA, Pate-Cornell ME, et al. Stage-gate process for
the development of medical devices. J Med Dev 2009;3:021004–15.
maximize the value of the information obtained from poten-
tial customers, a structured approach, such as quality func-
tional deployment (QFD)11 is recommended. The reason that
this process is so powerful is that it is a well-defined process
to translate customer needs to product features and keep them
evident throughout the design process.
The final step in this phase of the design process is to thor-
oughly document the product requirements.12 Though there
is some disagreement, requirements should state only what the
product should do and not how it does it. This approach
provides the broadest palette for conceptual designs.
Conceptual Design
Conceptual design is the process of structuring a solution to
a design problem. During this stage the process is mostly
qualitative, not quantitative. For instance, if one were design-
ing a robotic surgical assistant, a choice between electric and
hydraulic actuators would be decided during this phase. This
decision may require some quantitative analysis. For example,
“Can a motor fit within the space allotted and still provide
sufficient motive force without requiring the user to specify
part numbers, bolt patterns, and so on.”13
For engineers, the conceptual design phase is usually the
most enjoyable. Conceptual design requires great creativity, as
a number of widely disparate approaches should be developed
to ensure that the best ones are considered for further devel-
opment. Methods used to generate ideas include (1) literature
searches, (2) biomimicry, (3) reverse engineering of related
products, (4) study of analogous systems, and (5) brainstorm-
ing, among others. Once a number of concepts have been
developed, they are assessed against the requirements previ-
ously developed to determine those that best meet the speci-
fied needs.
At this juncture it is important to note several issues that
are relevant to design of medical devices rather than other
product types. In the United States, the Food and Drug
Administration (FDA) regulates medical devices. Although
entire books are written on the subject of complying with FDA
requirements,14-16 the following two (and potentially three)
issues greatly affect the development process.17
Record Keeping
The Safe Medical Devices Act of 1990 added design valida-
tion requirements to the Good Manufacturing Practices
(GMP) requirements.18 Fundamentally, the entity that designs
a medical device must document the procedures used to
ensure that the specified requirements are met. Procedures
should be generic rather than detailed. That is, they should
include major activities and assignments for subsystems, as
opposed to detailing responsibility for component selection.
It also is understood that design procedures change, thus the
standard requires that an entity’s procedure be regularly
reviewed and updated.
Device Class
All medical devices are categorized into one of three device
classes:
 Class I: These are non–life-sustaining devices whose failure
poses no risk to life. Examples of class I devices include
bandages, stethoscopes, and examination gloves. Most
class I devices are exempt from the premarket notification
or GMP regulation.
 Class II: These devices are non–life sustaining; however,
the controls in place for class I devices are insufficient to
ensure safety and effectiveness. Examples of class II devices
include catheters, powered wheelchairs, and many clinical
chemistry test systems. Although some class II devices are
exempt from premarket notification, others may require
clinical studies as rigorous as an investigational device
exemption (IDE) for premarket approval (PMA).
 Class III: These devices are typically those that support or
sustain human life and whose failure is life threatening.
Examples of class III devices include heart valves, implanted
electrical devices, and joint implants. The vast majority of
class III devices require PMA, a process that includes labo-
ratory testing, animal testing, failure mode analysis, manu-
facturing standards, and safety and efficacy testing as
determined in a human clinical trial. Needless to say, to
obtain PMA can be a long, costly process.
When setting out to develop a medical device, the design team
should determine the device class, as early as possible so that
the planning process can incorporate the appropriate activities
into the overall project plan. For new devices for which prec-
edent is unclear, the design team should submit a Request for
Determination to the FDA.
Extent of Business Activities
In the grossest sense, the activities needed to bring a device to
market are conception, design, production, distribution, and
support. A key decision for a new business entity is to deter-
mine which of these activities are to be performed in-house
and which are outsourced. This decision has two major
impacts: first, the amount of investment needed, and second,
the FDA regulations that must be addressed. For many com-
panies, outsourcing manufacturing offers a good blend of
activities, allowing the company to minimize capital costs,
earn a reasonable rate of return, and reduce FDA regulatory
issues (though the manufacturing partner must comply with
FDA requirements).
376 Section VII—Computers, Engineering, and the Future
The conceptual design phase ends with a design review that
presents the requirements, the concepts developed, and the
concept(s) selected for further development.
Detail Design
Most novel designs includes one, and sometimes more, ele-
ments that have not been previously realized. For example,
lab-on-chip devices replace manual fluid introduction and
mixing with microfluidic components. Before the design for
the final product is completed, the development team typically
performs a series of experiments to validate the functionality
of the novel component.8 Staying with the example, do the
microfluidic devices deliver sufficient head and flow to provide
sufficient agitation?
Once all of the subsystems are known to work, it is common
to couple all of them together in a bench-top prototype to
assess overall system functionality. This prototype may bear
little physical resemblance to the final device. However, the
manner in which it operates is analogous. Depending on the
nature of the product, the bench-top prototype may be used
to perform in vitro or in vivo testing, again providing further
evidence of the device’s safety and efficacy.
Once this testing is completed, the team should have suf-
ficient information to design the final device. This detailed
design consists of all of the information necessary to reliably
and repeatedly produce the device. Often, methods such as
Design for Six Sigma19,20 are used to help ensure that changes
in performance due to manufacturing variations do not result
in performance falling outside the specified bounds. This
design phase ends with a review that presents the results of
the experiments performed, the detail design, the bill of mate-
rials, and how these perform together to meet the design
specifications.
Clinical Trials
Certain classes of medical device must undergo clinical inves-
tigation21 before the FDA provides marketing approval. Animal
model testing often is performed before human trials, espe-
cially in those cases for which safety and efficacy are unknown.
Animal use to test a device is a privilege and numerous con-
trols exist to ensure that no more than the minimum number
of animals are used, that they are properly housed and cared
for, and that upon the conclusion of the experiment, they are
humanely euthanized.
The institutional review board of the facility that oversees
the trial must approve all devices that undergo human clinical
investigation. In addition, the FDA must approve trials for
significant risk devices through submission of an IDE. The
IDE exempts the manufacturer from certain portions of the
Food, Drug, and Cosmetic (FD&C) Act during the trial phase.
However, the manufacturer is still required to comply with all
design controls and all necessary manufacturing controls to
ensure patient safety and compliance with the manufacturer’s
quality claims.
The goal of a clinical trial is to determine the safety of a
device and its effectiveness in terms of the intended claims.22
This requires concise and specific definition of the study’s
objectives. The objectives must consider factors such as
comparison to approved modalities of treatment or whether
it treats a symptom or an underlying cause. Once these
factors are known, details of the trail design, such as study
size, candidate pool, and controls, among others can be
formulated. In addition, trials must be carefully designed
to minimize errors and biases. From a business perspective,
trial design is critical because the claims that the business
wishes to make about its device must be supported by clini-
cal outcomes.
Manufacturing
Once the design is completed, a pilot run is performed to
validate manufacturing processes. This is the first build of the
device in its final form using the manufacturing documenta-
tion. This part of the process also includes assembly workforce
training, validation of supplier plans, and checking that man-
ufacturing objectives have been achieved.23
A critical aspect of this phase is product packaging and
labeling. Packaging must be designed to ensure that the
product reaches the end-user in an appropriate condition
(which typically means that the product maintains sterility)
and that the package can be easily opened. Labeling includes
all printed or graphic material that accompanies a product
and advertising. Depending on the type of product, labeling
requirements may include the manufacturer’s contact infor-
mation, intended uses, and directions for use, to name a few.
Once all of this information is available, it is submitted to
the FDA, either as a 510(k) premarket notification submission
or a PMA submission. Approval from the FDA is required to
legally offer the product for sale in the United States. Other
regulations apply in other countries. However, FDA approval
is not necessarily sufficient for the product to gain acceptance.
In the United States, much of the cost for medical products
and procedures is paid for by third-party insurers. Underlying
these payments is a list of codes that source from the Centers
for Medicare & Medicaid Services (CMS). Because codes are
defined in terms of services provided, and not specific devices,
new products may be covered by existing codes. However,
novel devices may require the development of new CMS
codes, and without these codes, many practitioners may be
hesitant to adopt a new device.
Financing Product Development
A corollary to Murphy’s law states, “The first 90% of the task
takes 90% of the money, and the last 10% takes the other
90%.” Based on the previous discussion about medical device
design and development, this statement may, in fact, be overly
optimistic. Depending on the type of device, the development
process can require years and tens of millions of dollars.
Because most medical device inventors are not independently
wealthy, outside investment often is required to bring a new
medical device to market.
The world of outside investment can be broadly divided
into two categories: those who invest primarily to obtain a
(hopefully very high) return on their investment and all
others. The former are typified by angel investors and venture
capitalists; the latter by universities, federal and state govern-
ments, family, and others. Angel investors and venture capital-
ists play such an important role in medical device development
that the entire next section is devoted to them. This section
starts with some general guidance about business practices
and then discusses non-equity investment opportunities.
 Section VII—Computers, Engineering, and the Future 377
The founder of a business focused on developing a new
medical device is convinced of the value and importance of
the invention. However, Ralph Waldo Emerson’s adage that if
one “build[s] a better mousetrap and the world will beat a
path to your door” is not always true. It takes considerable
effort to convince others that the potential product will both
address an important medical need and produce a profit.24
Simply providing experimental data is insufficient because the
inventor will be asking potential investors to trust him or her
with millions of dollars.
To access outside funding of most any sort, a company must
have a business plan.25,26 There are innumerable books about
writing business plans27,28 (see further suggestions later in this
chapter). Two cautionary notes: first, most business plan soft-
ware programs should be avoided, because they typically do
not add much value. Second, the writing of the plan should
not be outsourced because the entrepreneur must know the
entire plan, the sources used to develop it, and so on. Con-
tained within the plan is a financial model,29 arguably the
single most important aspect of the plan. Attention should be
focused on building a rational model, because potential inves-
tors will be more interested in the assumptions that underlie
the model rather than the actual number presented.
As a final cautionary note, inventors who intend to seek
outside investment should engage accountants and attorneys.
Although it is true that forming a business requires little more
than completing a simple form, the business’s rules of opera-
tion (referred to as an operating agreement for limited liability
companies [LLCs] and bylaws for corporations) dictate how
the business is run. If these rules of operation are not profes-
sionally written, they may impede potential investments. Sim-
ilarly, an accounting system that does not adhere to generally
accepted accounting principles (GAAPs) will make it very dif-
ficult for a proper financial analysis to be performed, which
again may impede potential investments. As a technology-
focused business, the most valuable assets are the inventor’s
intellectual property. Although it is certainly possible to apply
for patents and trademarks independently, engaging an intel-
lectual property (IP) attorney is strongly recommended to
ensure that the protection sought is as broad as possible.
Finally, an attorney should review all dealings that can materi-
ally affect the business, such as grants of corporate ownership,
licensing agreements, and partnering deals, among others,
before being executed.
University Research
If the developer is a university employee when the medical
device idea is first invented, the first stop is the school’s office
of technology transfer. Since the passage of the University and
Small Business Patent Procedures Act (also known as the Bayh-
Dole Act) in 1980, most universities have created offices that
deal exclusively with licensing and IP. For a university inventor,
working with the office of technology transfer provides several
important benefits. First, as dictated by the Bayh-Dole Act, if
the research was funded by the federal government, the univer-
sity has the right to retain ownership of the IP created. As such,
should the technology be commercialized, the inventor will
typically receive financial remuneration as a percentage of the
licensing revenue earned by the university. Second, the inven-
tor of the IP is often granted first right-of-refusal with regard
to licensing the technology for commercial purposes. Often,
for faculty-founded start-ups, the university will provide
favorable terms, such as defraying patent prosecution costs
until the company has revenue. Third, the office of technology
transfer may have specialists whose role is to help with
company formation. These individuals can provide guidance
with business plan writing, recommendations for local service
providers, and identification of funding opportunities.
The passage of the Bayh-Dole Act also has motivated uni-
versities to commercialize technology. Universities with active
technology transfer offices can generate tens of millions of
dollars annually in license fees, royalty, and equity. To capital-
ize on this opportunity, a number of states and universities
have created funding sources for university inventors to assist
them to bring their inventions to market. Examples of such
programs include the following:
 Boston University Ignition Award and Launch Award30:
The former helps bridge the gap between basic science and
the product development and the latter is designed to help
faculty members start new companies based on technolo-
gies that they invented at the university.
 Michigan Universities Commercialization Initiative31: A
collaboration designed to complement and enhance the
technology transfer at Michigan academic and research
institutions by supporting commercialization of IP.
 University of Utah Technology Commercialization Pro­
ject32: Focused on further developing novel techno­
logies that are near commercialization in all areas of
technology.
For the university inventor, such initiatives offer great value.
First, the competition for funds is limited to a small pool of
applicants, thus the odds of funding are relatively high. Second,
the peer-review process typically used provides some valida-
tion of the commercial potential of the technology. Third, the
funds often can be used for non–technology-focused activi-
ties, such as engaging business and marketing professionals
from outside the university, which helps to accelerate the
growth of the business. The only downside to these programs
is that the funding available is typically limited.
Government Research Funds
Growing a technology-focused start-up from inception to
profitability is a very high-risk proposition. Though exact
figures are not available, it is estimated that less than one
quarter of all businesses are still operational after 10 years—
for technology-focused start-ups, the number is smaller. On
the other hand, new products and companies account for a
disproportionate fraction of a country’s gross domestic
product (GDP).33 As such, all levels of government are incen-
tivized to support early-stage companies because of the poten-
tial return on investment. In the United States, one federally
mandated program that supports small businesses dominates
all others: the Small Business Innovation Research (SBIR)
program (and the very closely related Small Business Technol-
ogy Transfer [STTR] program).34,35
Originally authorized by Congress in 1982, key objectives
of the programs are to stimulate U.S. technologic innovation,
create new opportunities for small businesses to participate in
federally sponsored research and development, and increase
private-sector commercialization of innovations derived from
federal research and development (R&D). As legislated, all
378 Section VII—Computers, Engineering, and the Future
SBIR STTR
Phase I 6 months 12 months
Feasibility $100k $100k
Phase II 2 years 2 years
Prototype $750k $750k
Phase III As long as it takes
Commercialization non-SBIR/STTR funding
Fig. 38.1  General time and funding outline for Small Business Innovation
Research (SBIR) and Small Business Technology Transfer (STTR) programs.
federal agencies whose extramural research and development
budgets exceed $100 million have a mandated set-aside of
2.5% of their budget to fund SBIR projects. In addition, all
federal agencies whose extramural research and development
budgets exceed $1 billion have a mandated set aside of 0.3%
of their budget to fund STTR projects. In fiscal year (FY) 2007,
12 agencies participated in the SBIR/STTR program, with
total funding of $2.315 billion.
Of the 12 participating agencies, the Department of Defense
(DOD) accounted for 54.9%, the National Institutes of Health
(NIH) accounted for 28.1%, and the National Aeronautics and
Space Administration (NASA), Department of Energy (DOE),
and National Science Foundation (NSF) combined accounted
for 14.2% of the of the total funding, respectively. Developers
of medical devices would typically focus on applications to the
NIH; however, the DOD and NSF occasionally seek medical
device applications as well.
SBIR/STTR programs are three-phased programs, as out-
lined in Figure 38.1. The durations and dollar figures are
statutory guidelines—each agency sets its own rules.
The key requirements to allow access to the source of
funding are that the firm be a U.S. for-profit business with 500
or fewer employees and that work be performed in the United
States. Other requirements about corporate ownership, prin-
cipal investigator employment, and so on can be found online.
Major changes to the SBIR program are being proposed in
Congress during the production of this book, and so the
information provided herein is likely to change. (The reader
may refer to information about these programs online at
www.sbir.gov.)
Participating in the program is straightforward. A small
business must first obtain a data universal numbering system
(DUNS) number (fedgov.dnb.com), federal tax identification
number (www.irs.gov), and a bank account that accepts elec-
tronic funds transfers, and then register with the central con-
tractor registration (www.ccr.gov). Next, the small business
reads the solicitations published by the 12 agencies and applies
for phase I projects for which it is qualified. The phase I appli-
cation takes the form of a proposal along with a project
budget, brief resumes of key personnel, and letters of support
from partners or customers. Small businesses that successfully
complete a phase I project will almost always be invited to
submit a phase II proposal. The phase II proposal is very much
like the phase I, but with two additions: a section that describes
the outcome of the phase I project, and a commercialization
plan that describes how the technology being developed will
be brought to market.
There are some important differences between the SBIR/
STTR programs offered by the agencies. The DOD and several
others are contracting agencies, whereas the NIH and NSF and
others are granting agencies. With contracting agencies the
agency establishes the needs (which are typically highly
focused), there are more fiscal requirements, and the project
initiator (who works for the agency) is the primary proposal
reviewer. With granting agencies, the investigator identifies
the problem and specifies the approach (as long as it falls
within the agency’s purview), there is more fiscal flexibility,
and proposals are peer reviewed. From a business perspective,
winning SBIR contracts from contracting agencies is benefi-
cial because on successful completion of phase II, the agency
may wish to purchase the product that has just been devel-
oped. Winning SBIR contracts from granting agencies is ben-
eficial because this indicates that the technology was deemed
meritorious to a panel of experts, and so provides great cred-
ibility to the company.
As valuable as the SBIR/STTR programs are, all potential
applicants for SBIR/STTR funding should be aware of several
facts:
 The programs are highly competitive. On average, fewer
than one in eight phase I proposals is funded and typically
half of all phase II proposals are funded.
 Federal funding has restrictions. Project money may not
be used for certain necessary business activities such as
patent prosecution, thus other sources of financing are
required.
 The SBIR/STTR cycle is slow. It is not atypical for a project
(defined as the first writing of a phase I proposal to the
end of the phase II project) to exceed 3.5 years. For certain
types of products, where time-to-market is important, this
long time frame may not be acceptable.
 The federal government has rights to IP developed.
Quoting a recent DOD SBIR solicitation: “The govern-
ment receives a royalty-free license for its use, reserves the
right to require the patent holder to license others in
certain limited circumstances, and requires that anyone
exclusively licensed to sell the invention in the United
States must normally manufacture it domestically.”
State Commercialization Funds
State governments realize that having a vibrant start-up com-
munity is valuable for the state. Companies that become suc-
cessful will hire more people, spend more money, and pay
more taxes. To help build such communities, many states
have created programs that support technology-focused
start-up businesses. Such programs span the gamut from
business plan competitions that provide ten of thousands of
dollars to SBIR matching programs that offer hundreds of
thousands of dollars to other competitive programs that
provide millions of dollars. The following list highlights
several such programs.
 Texas Emerging Technology Fund (ETF): Established in
2005, the Texas ETF was intended to “expedite innovation
and commercialization of research, promote a substantial
increase in high-quality jobs, and increase higher educa-
tion applied technology research capabilities.”36 Compa-
nies with disruptive technologies were encouraged to
apply for funding up to $5 million. The decision process
included a thorough vetting process (technology, IP, and
 Section VII—Computers, Engineering, and the Future 379
commercial potential), which added significant credibility
when seeking future investments. The funding mecha-
nisms were very favorable from the companies’ perspective
and typically took the form of equity. (It is no longer
available.)
 Kentucky SBIR/STTR Matching Funds Award37: The pur­
pose of the program is to foster job creation and economic
development in Kentucky by increasing the competitive
position of small businesses to attract SBIR/STTR funding.
This is accomplished by providing matching funds to com-
panies that have been granted a federal SBIR/STTR
program, phase I or phase II, in one of the state’s identified
focus areas. Companies can receive up to 100% of the
amount of the SBIR/STTR award (with some limitation)
and may receive up to a total of five such awards. The
application is not subject to vetting (other than ensuring
that certain criteria are met), and the funds are provided
in the form of a grant.
Although these programs vary widely, they typically have
several attributes in common:
 The majority of the funding provided is to be used within
the state that makes the award. In fact, companies that
move out of the awarding state may be obligated to repay
the monies upon so doing.

The programs typically focus on commercialization, not
basic research. The states are investing in the company
with the hope that jobs, and taxes, will be created in the
relatively near future.

The funding may be provided as a grant, loan, or convert-
ible debt. The entities that run these programs typically
structure the deals so that they do not hinder or preclude
future investments.

Funds often are provided on a first-come, first-served basis,
with a fixed amount of funding allocated to the program.
As such, meritorious applications to these programs can
go unfunded if the program exhausts its prescribed dollar
allocation (e.g., toward the end of the fiscal year).

There is one potential problem with state-run programs.
In certain states, the legislation for the program may not
provide sufficient privacy protection for documents sub-
mitted for consideration. Before applying for one of these
programs, and before submitting any written reports, a
device developer must fully understand what access to
documents is available to interested parties using the
Freedom of Information Act (FOIA).
Other than potential FOIA issues, these state-run programs
are valuable. Companies need unrestricted money to find cus-
tomers, market their products, and seek potential investors;
other sources of funding, such as SBIR/STTR, are limited in
their ability to support such activities. Thus for companies
that seek to commercialize a product that results from R&D
activities, these funds can be valuable. However, the award
amounts are typically insufficient to bring a product, espe-
cially a medical device, to market on their own.
Other Sources of Funds
There are several other possible sources of funding for start-up
businesses. Although these sources typically do not have the
financial resources to fund a company from inception to first
revenue, they can provide easy access to cash and other useful
expertise.
Friends, Family, and Fools
Entrepreneurs often turn to their friends, families, other
people they know, and credit cards to help fund the early
stages of a business. These individuals typically invest because
they personally know the entrepreneur and trust that he or
she will provide a return on their investment. Unlike angel
investors, these individuals are not required to be high–net
worth individuals. However, it is critically important that such
transactions be managed properly, with appropriate legal
documentation and signatures.
Foundations, Not-for-Profit Organizations
Organizations whose mission is aligned with the company’s
product development efforts can potentially provide assis-
tance. In some cases, the assistance may be other than cash,
because some not-for-profits are restricted in their ability to
finance for-profit businesses. However, these organizations
typically have extensive networks of like-minded people
with whom they can share information about the entrepre-
neur’s efforts, which can lead to investments and other
opportunities.
Commercial Banks
Commercial banks rarely provide loans to start-up businesses
because collateral is required to secure the loan. Unless the
entrepreneur is willing to collateralize his or her own assets,
such as a house, there is rarely anything of bank-accepted
value within the business to offer as collateral.
Figure 38.2 overlies a product’s life cycle with the typically
available sources of funding for the particular stage of
development. The dollar values shown provide guidance
about the typical order of magnitude of the funding. The light
blue shaded section, that period of time during which cumu-
lative profits are negative, is often referred to as the “valley
of death.”
Angel and Venture Funding
Angel investors and venture capital (VC) firms are potential
sources of capital available to entrepreneurs trying to start a
company. Angel investors (also known simply as angels) are
high–net worth individuals who invest their own money into
companies. Often angels have domain expertise or a personal
interest in the technology being developed. Although the mag-
nitude of angel investing in the United States is large, esti-
mated to exceed $23.3 billion in 2010, individual angels, or
angel groups, typically do not invest more than several
hundred thousand dollars into a single firm. Due to the high
costs associated with bringing a medical product to market,
angels are not typically well equipped to provide substantial
assistance to entrepreneurs in this field. Thus the remainder
of this section focuses on venture capital.
A venture fund is a private equity investment entity that
invests capital in companies on behalf of third-party investors.
Usually structured as an LLC or a general partnership, VC
firms receive investment capital from limited partners, be they
380 Section VII—Computers, Engineering, and the Future

Investigation Feasibility Development Introduction Growth Maturity

Equity markets,
banks ($10M)

Profit Venture capital ($1M)

Pre-seed funds, angels ($100k)

SBIR/STTR, other gov.

support ($100k)

Friends, family,
founders ($10k)

Time
Fig. 38.2  The life cycle of a start-up company, showing stages of development and typically available funding sources. SBIR, Small Business Innovation
Research; STTR, Small Business Technology Transfer.

Pre-Deal Capitalization Structure Class B Preferred Units

Common Common Pro forma Pro forma Preferred Pro forma Pro forma
shares options units ownership issuance units ownership
Entrepreneur 10,000,000 – 10,000,000 100.0% – 10,000,000 66.7%
Venture capitalist – – – 0.0% 5,000,000 5,000,000 33.3%
Total 10,000,000 – 10,000,000 100.0% 5,000,000 15,000,000 100.0%

Fig. 38.3  Simplified capitalization table showing the change in percentage ownership upon accepting an investment.

high–net worth individuals or large financial institutions or
some combination of the two. VC firms typically make invest-
ments in exchange for company shares, that is, equity in the
company is exchanged for capital financing from the venture
capitalist.
The exact number of shares that the VC firm receives for
the cash investment is determined by what is called the pre-
money valuation of the given company. The pre-money valu-
ation is a negotiated value of the company’s intrinsic value
before taking money from the VC firm. If the VC firm and the
company agree that the company is worth $10 million before
the investment (i.e., the pre-money valuation is $10 million),
and the amount of invested capital is $5 million, then the VC
firm is effectively buying 33% of the company. By putting $5
million into a company that is worth $10 million, the post-
money valuation of the company is $15 million. More simply
put, $5 million added to $10 million equals $15 million; $5
million divided by $15 million equals 33% of the company
(Fig. 38.3).38
Accompanying the capital that entrepreneurs can receive
from VC firms is the experience that venture capitalists can
offer to start-up companies. For immature companies, VC can
provide the added benefit of the managerial, technical, and
industry-based expertise of the venture capitalists. For medical
device companies specifically, the venture capitalist may
help the entrepreneur navigate through the complex regula-
tory approval process, set up clinical trials, understand
the reimbursement coding system, establish a manufacturing
relationship, ensure timely and complete filing of IP protect-
ing patents, negotiate distribution contracts, or develop a
go-to-market strategy to name just a few of the “bonuses” that
the ideal VC firm can offer to the entrepreneur.39
The timeless VC adage in the medical device space goes
something like this: every entrepreneur has impenetrable IP
and every venture capitalist promises to aid the entrepreneur
to surmount every hurdle during the process of taking a
product to market. The truth of the matter usually lies some-
where in between the two.
What Equity Investors Are Seeking
One of the first hindrances to receiving VC is getting the atten-
tion of a venture capitalist. The best means to find a venture
capitalist is through personal contacts who can then recom-
mend the inventor and thus personalize the introduction. A
trusted contact can break down even the strongest of barriers.
Another means to meet a venture capitalist is at a conference.
The last method is the “cold call” or “cold email,” in which the
venture capitalist is contacted without a formal introduction.
The venture capitalist is thereby introduced to the company
or idea for a company in an impromptu and extemporaneous
way. (A list of venture capitalists that focus on medical devices
can be located at the following website: www.devicelink.com.)
David Lawee, the founder of Mosaic Venture Partners offers
these tips for getting in the door:
 It dramatically improves your chances to come in through
a trusted reference.
 You have to be savvy about who those people [are]…not
to listen to everybody and say, “Oh, that guy’s a trusted
 Section VII—Computers, Engineering, and the Future 381
referral.” If you associate with someone who is not consid-
ered to be credible, then you get tainted by that.
 There is so much available on the Internet. Go out, read
up, and learn about the market. Figure out how you’re
going to get to the influencers. We do that all the time for
our companies. You need to do that…it’s just a normal
business skill.40
During the initial contact phase, it is customary to provide the
venture capitalist with an executive summary or a “one-pager”
that summarizes the company. The one-pager is ideal because
it forces the entrepreneur to include only the most pertinent
information that relates to the device in a “one page only”
format. After sending the one-pager and arranging a confer-
ence call or an in-person meeting, it is imperative to have a
solid investor PowerPoint presentation. This presentation not
only helps guide the company during the story-telling process,
but also shows the VC firm how detail-oriented and stream-
lined the company truly is.
Most venture capitalists target medical device technology
companies at a specific stage, be it early stage (i.e., pre-FDA
approval) or late stage and post-FDA approval (i.e., market
launch). Some venture capitalists are stage agnostic and will
entertain entrepreneurs at any stage in the development cycle.
The goal of VC investing is to make money on a calculated
risk investment. The goal, as novel as it may sound, is to invest
as little as possible and to generate as high a rate of return as
possible. Buy low, sell high. All venture capitalists have their
own investment thesis, but certain investment metrics do
exist. Following is a general list of the basic investment criteria
of a venture capitalist:
 Unmet clinical need
 Large uncrowded market

Strong intellectual property

Clinical and regulatory process

System and physician economics

Established reimbursement protocol

Distribution and sales strategies

Experienced management team

Multiple on invested capital

Overall cash return on investment
This list, if thoughtfully addressed by the medical device
company in question, will help guide the venture capitalist’s
decision-making process. The goal of the start-up company is
to put together a clear presentation that shows that the device
is novel and meets an unmet need for a specific patient popu-
lation. With that in mind, the entrepreneur must show how
this device is a major advancement over existing technology
and that it fulfills a truly unmet clinical need in a market that
will generate significant returns.41
When the market is considered, the saying is “the bigger,
market, the better.” Unfortunately orphan diseases often are
excluded after a market-based analysis. Some big markets
include the most common disease pathologies such as diabe-
tes, congestive heart failure, obesity, or sleep apnea.
Strong intellectual property is the sine qua non of VC invest-
ing. Patents protect an investment and in essence are the true
assets of a start-up company. Because patent assessment
requires real dollars and cents in terms of costs for legal review,
this part of the due diligence process is often the last step after
a letter of intent, or financial term sheet, is extended to the
company.42
The clinical and regulatory process is one of the many
hurdles that entrepreneurs must leap over on the path to
taking a medical device to market. This process, and indeed it
is a process, requires a deliberate and well-executed strategy
when dealing with the FDA. Whether the company is filing a
510(k) or a PMA, the clinical trials that will support the FDA
filing require thoughtful planning, careful selection of sites
and principal investigators, and clearly defined endpoints and
protocols. The choice of the PMA or 10(k) mechanism can be
important because most recalls are of medical devices origi-
nally cleared through the 510(k) process or were considered
of low risk and so exempt from review.43
When determining the economics of a medical device, the
venture capitalist considers the micro- and macro-effects that
the product will have on the system. Devices often are used
directly by doctors or other medical practitioners. When ana-
lyzing a device, the cost to the doctor, namely the physician
economics or micro-effect, must be considered in the grand
economic picture. With respect to the macro-effect, the
venture capitalist considers the costs to the system as a whole,
for example, does the device increase office throughput or
reduce operating room time? Because of the American third-
party payer system, venture capitalists must consider the reim-
bursement rate of a medical device. Devices are paid for as
part of a Diagnosis-Related Group (DRG) hospital code,
through a Current Procedural Terminology (CPT) code, or by
the patient him- or herself (an out-of-pocket expense). The
process to establish a new CPT code is a lengthy (greater than
1 year) and exacting process that relies on careful filings and
ultimately on sales generated from the device. The venture
capitalist will consider the true “purchaser” of the device when
evaluating a device.
When it comes to distributing and selling a device, the
venture capitalist will consider the marketplace and the big
players in the given clinical space. Because of established dis-
tribution channels, large medical device companies can easily
add a new product to their sales representatives’ “bags” and
effectively blanket the country with the device in question.
Unfortunately, a fledgling company does not have the where-
withal or the resources to emulate this national or global
approach. The venture capitalist will consider the competitive
environment during his or her analysis and weigh the different
sales strategies.
How intimately venture capitalists work with their portfolio
companies will determine how important the vetting of the
management team’s biographies is to them. In the end, the
shepherding of portfolio companies through the entire process
will require intense interaction. The better the communica-
tion process is, the more successful the experience will prove
to be. Hardworking and assiduous people often generate suc-
cessful outcomes. The returns that a venture capitalist looks
for in its investment process are based largely on the method-
ology of the specific VC firm. In the end, a patent-protected
device that addresses a disease state with a large market and
that can be produced with high gross margins and low operat-
ing costs is the venture capitalist’s four-leaf clover.
Pros and Cons of an Equity Investment
The pros and cons of taking on an equity investment lie in the
locus of control. Control comes in two forms: financial and
administrative. Before taking money from a venture capitalist,
the entrepreneur has 100% control over his or her company.
382 Section VII—Computers, Engineering, and the Future
The decision making is not diluted by a board of directors and
is solely in the hands of the founders. Once the company
decides to take on capital from a venture capitalist, the power
structure of the company changes significantly. In exchange
for the risk of their investment, the venture capitalists gain
significant control over company decisions, in addition to a
significant portion of the company’s ownership (and conse-
quently value).
The key to ensuring a safe journey lies in shrewd negotia-
tion of the investment documents. Every line in every one of
the investment documents, from the purchase agreement
to the shareholders’ agreement, is written for a reason.44 To
ensure that no surprises arise, seasoned legal representatives
should scour the documents and negotiate the most favorable
terms possible. The list of problems that can arise from poorly
negotiated documents is limitless, so the enterpreneur must
be keenly aware of all the pitfalls and windfalls that exist in
the legal contracts.
The pros of taking on VC money depend upon the venture
capitalist in question. Before taking money from a given
venture capitalist, the venture capitalist should be asked for
permission to contact some of their portfolio company chief
executive officers (CEOs) or general managers (GMs). A
transparent venture capitalist will not find anything amiss
with this request and should facilitate this process. A good
firm has nothing to hide. If red flags emerge from these con-
versations, the answer to the question of whether or not this
venture capitalist will add value on top of his or her equity
investment should be apparent.
If the venture capitalist receives positive reviews, the
ideal venture capitalist will buttress the management team
with expertise in any number of medical device-related
areas. Suffice it to say, a seasoned VC firm will offer its experi-
ence in the field of medical devices, its contact lists, and
the overall expertise of its team, and so ensure that capital
will be coupled not with a loss of control but a gain of a
valued resource.
The Equity Investment Process
After a phone call or an in-person investor PowerPoint pre-
sentation, the venture capitalist will perform due diligence on
the company based on the interest level generated by these
interactions. The key to getting the attention of the venture
capitalist is a clear presentation that hits on the key parameters
that venture capitalists consider important to make an equity
investment. Initial due diligence can either precede or come
after another conference call or another in-office meeting. The
number of the calls and meetings varies from firm to firm, but
the average inclusive number probably exceeds three (calls and
meetings). The venture capitalist also may supplement infor-
mation requests and in-person demonstrations by the
company with calls to physicians and other trusted experts in
the given field. A typical due diligence binder that the venture
capitalist will request may include any of the following
documents:
1. Financial plan and top line assumptions for the plan
2. Milestone chart reflecting key milestones of the plan
3. Term sheet proposal
4. Investor presentation
5. Regulatory process
6. Management biographies
7. Organization chart
8. Patent family tree
9. Summary of preclinical studies
10. Summary of clinical trial protocols
11. Clinical trial contacts
12. Explanation or overview of device
13. Capitalization table
14. Market model
15. Indications for gold standard competitor
16. Competitive grid
17. Clinical literature review
18. Clinical literature articles
At this stage in the investment algorithm, a venture capitalist
will decide whether to extend a letter of intent, often referred
to as a term sheet. This letter of intent will clearly define the
investment parameters of the venture capitalist, namely, the
pre-money valuation, the amount of invested capital, whether
or not warrants are to be a part of the transaction, the tranch-
ing of the investment, and any milestones that the company
must meet to receive the different investment tranches.
(“Tranche” is French for a slice, so if a venture capitalist
decides to invest $6 million, it may provide the investment in
three $2 million tranches.) The letter of intent also usually
defines the capitalization structure of the company and the
terms of the stock investment—conversion, voting rights,
redemption rights, antidilution protection, liquidation prefer-
ence, and first-offer rights. The rest of the letter of intent
defines the closing conditions, the transactions fees, and the
legal and confidential nature of the agreement.
After the so-called “doctor calls” and the negotiation of the
letter of intent, the legal evaluation of the company’s IP is
usually the last step in a VC firm’s due diligence process
because of the high cost of IP attorneys’ fees.
If he or she decides to pursue an investment, the venture
capitalist will send a letter of intent to the company with a
monetary valuation of the company. From this starting point,
the negotiation between the company and the venture capital-
ist begins. Ideally, the terms on either end meet somewhere in
the middle about the valuation and the acceptable terms for
the transaction to occur.
Conclusion
To bring a novel medical device to market is a challenging
proposition. The range of activities that need to be successfully
managed range from understanding the clinical issues to
developing an engineered device to raising capital to comply-
ing with federal regulations to eventually selling the product.
Despite the challenges, this is a process undertaken by people
on a daily basis because of the potential rewards. The single
most important fact to remember is that others who have
done this before are available, sometimes at little to no cost,
to assist new inventors in their endeavors. Therefore, one of
the most important activities in which a new entrepreneur
should participate is networking. Networking provides the
opportunity to learn about resources available and lets others
learn about the new project. Through networking it is possible
to obtain recommendations for service providers, find em­
ployees, and identify people who can serve on a board of
advisors.
 Section VII—Computers, Engineering, and the Future 383
Regardless of where an enterpreneur lives, there are local
resources that can provide assistance. Most every city has a
chamber of commerce,45 an organization whose focus is on
the local business community. Chambers typically have regu-
larly scheduled meetings, which are usually great places for
networking. They also offer a range of services, such as provid-
ing group insurance, that may be of value to small businesses.
In addition, many chambers have SCORE offices,46 business
counseling and mentoring organizations, another invaluable
resource for startups.
Many cities, or regions, have economic development
corporations, typically public-private partnerships focused on
growing businesses. Services offered vary widely, but some
offer business incubator space, accelerator programs, and/or
consulting services for entrepreneurs. All states have state-
wide economic development corporations that establish
programs (such as those discussed previously), provide
tax incentives, and a broad range of programs to assist
businesses.
All states also have small business (and technology) devel-
opment centers.47 These organizations provide counseling,
training, research, and advocacy for small businesses. The
program, run jointly with the federal Small Business Admin-
istration, typically provides everything from classroom learn-
ing to one-on-one coaching.
The bottom line is this: starting a business and working to
make it successful is an extremely rewarding process. In so
doing, the entrepreneur enriches himself or herself, his or her
community, and in the case of medical devices, society as a
whole. Although challenging, the resources available to entre-
preneurs are greater than ever, and make now the best time to
start a business.
References
1. Harman RJ. Development and control of medicines and medical devices.
London: Pharmaceutical Press; 2004.
2. Pietzsch JB, Shluzas LA, Pate-Cornell ME, et al. Stage-gate process for the
development of medical devices. J Med Dev 2009;3:021004–15.
3. Linehan JH, Chaney A. Academic/industry challenges for medical device
development. Sci Transl Med 2010;2:63mr6.
4. Yock PG, Brinton TJ, Zenios SA. Teaching biomedical technology innovation
as a discipline. Sci Transl Med 2011;3:92cm18.
5. Kutz M. Biomedical engineering & design handbook. Vols. I and II. New
York: McGraw-Hill Professional; 2009.
6. Fries RC. Reliable design of medical devices. 2nd ed. Boca Raton, FL: CRC
Press; 2005.
7. Prutchi D, Norris M. Design and development of medical electronic
instrumentation: a practical perspective of the design, construction, and test
of medical devices. Hoboken, NJ: Wiley-Interscience; 2004).
8. Kucklick TR. The medical device R&D handbook. 2nd ed. Boca Raton, FL:
CRC Press; 2005.
9. Pahl G. Engineering design: a systematic approach. London: Springer; 1996.
10. Pugh S. Creating innovative products using total design: the living legacy of
Stuart Pugh. Reading, MA: Addison-Wesley; 1996.
11. Cohen L. Quality function deployment: how to make QFD work for you.
Reading, MA: Addison-Wesley; 1995.
12. Grady JO. System requirements analysis. New York: McGraw-Hill; 1993.
13. Blanchard BS. Systems engineering and analysis. Upper Saddle River, NJ:
Pearson Prentice Hall; 2006.
14. Fries RC. Medical device quality assurance and regulatory compliance. New
York: Marcel Dekker; 1998.
15. Cheng M, World HO. Medical device regulations: global overview and
guiding principles. Geneva: World Health Organization; 2003.
16. U.S. Food and Drug Administration—Center for Devices and Radiological
Health. at http://www.fda.gov/AboutFDA/CentersOfficesMedicalProductsand
Tobacco/CDRH/default.htm. Accessed Sept 4, 2012.
17. Fries RC. Handbook of medical device design. New York: Marcel Dekker;
2001.
18. US FDA/CDRH: Good manufacturing practice/quality systems. at
http://www.fda.gov/medicaldevices/deviceregulationandguidance
/postmarketrequirements/qualitysystemsregulations/default.htm.
Accessed Sept 4, 2012.
19. El-Haik B. Medical device design for six sigma: a road map for safety and
effectiveness. Hoboken, NJ: Wiley-Interscience; 2008.
20. Justiniano JM. Six sigma for medical device design. Boca Raton, FL: CRC
Press; 2005.
21. Becker KM, Whyte JJ. Clinical evaluation of medical devices: principles and
case studies. Totowa, NJ: Humana Press; 2006.
22. Chiacchlerini RP. Medical device clinical trial design, conduct, and analysis.
Health care technology policy I: the role of technology in the cost of health
care; 1994. p. 278–85.
23. DeVor RE. Statistical quality design and control: contemporary concepts and
methods. Upper Saddle River, NJ: Pearson/Prentice-Hall; 2007.
24. Dorf RC. Technology ventures: from idea to enterprise. New York:
McGraw-Hill; 2007.
25. Mullins J. The new business road test: what entrepreneurs and executives
should do before writing a business plan. 2nd ed. Upper Saddle River, NJ: FT
Press; 2008.
26. Kawasaki G. The art of the start: the time-tested, battle-hardened guide for
anyone starting anything. New York: Portfolio Hardcover; 2004.
27. Rule RC. The rule book of business plans for startups. Irvine, CA:
Entrepreneur Press; 2000.
28. DeThomas A, Derammelaere S. Writing a convincing business plan.
Hauppage, NY: Barron’s Educational Series; 2008.
29. Benninga S. Financial modeling. 3rd ed. Cambridge, MA: MIT Press; 2008.
30. BU—Office of Technology Development—awards/funding overview.
Available at http: www.bu.edu/otd/awards. Accessed Sept 4, 2012.
31. Michigan Universities Commercialization Initiative (MUCI) Home Page.
Available at http://www.muci.org. Accessed Sept 4, 2012.
32. The University of Utah. Available at http://www.research.utah.edu/funding/
tcp.html. Accessed Sept 4, 2012.
33. Lord Robert McCredie, M.O.A.K.F.F.F. Innovation: from new knowledge to
new products. (2002).Available at http://www.atse.org.au/
index.php?sectionid=344. Accessed Sept 4, 2012.
34. SBIR—Small business innovation research. Available at www.sbir.gov.
Accessed Sept 4, 2012.
35. SBIR/STTR Gateway—news and search engine. Available at www.zyn.com/
sbir. Accessed Sept 4, 2012.
36. Texas Life Science Center. Available at http://www.texaslifesciences.com/.
Accessed Sept 4, 2012.
37. Technology & innovation—funding| The Kentucky Cabinet for Economic
Development. Available at www.thinkkentucky.com/DCI/SBIR. Accessed
Sept 4, 2012.
38. Wilmerding A. Deal terms. Boston: Aspatore; 2003.
39. Marks, K.H. 0.14.2.b: The handbook of financing growth: equity
investment documentation—encyclopedia—library—VC experts.
Available at http://vcexperts.com/vce/encyclopedia-of-private-equity.
Accessed Sept 4, 2012.
40. Knight J. Equity financing: understanding venture capital. Available
atwww.canadaone.com/ezine/aug00/venture_capital.html. Accessed Sept 4,
2012.
41. Knight HJ. Patent strategy for researchers and research managers. Chichester,
UK: John Wiley & Sons; 2001.
42. Elka N. Medical device firms can boost their chances of success. Available
atwww.altassets.com/casefor/sectors/2007/nz11478.php. Accessed Sept 4,
2012.
43. Zuckerman DM, Brown P, Nissen SE. Medical device recalls and the FDA
approval process. Arch Intern Med 2011;171:1006–11.
44. Feld B, Mendelson J. Venture deals: be smarter than your lawyer and venture
capitalist. Hoboken, NJ: John Wiley & Sons; 2011.
45. Chamber of Commerce, USA Directory, Visitor Bureaus United States
Directory. Available at www.2chambers.com. Accessed Sept 4, 2012.
46. SCORE Counselors to America’s small business. Available at www.score.org/
index.html. Accessed Sept 4, 2012.
47. America’s Small Business Development Center Network.
Available at www.asbdc-us.org. Accessed Sept 4, 2012.
VII
Chapter
39  
Engineering Issues
Brett Trimble and Jens Bracht
Introduction
History and Background
By most accounts the modern era of neuromonitoring began
in the 1950s with reports of continuous intracranial pressure
(ICP) measurements in humans by Janny et al.1,2 ICP was
measured by connecting a strain gauge pressure transducer to
a fluid-filled tube connected to a catheter with its distal end
implanted in the patient’s cerebral ventricle. The use of a
strain gauge, as opposed to a manometer, to measure pressure
made ICP monitoring safe, easy, and accurate. The frequency
response of the sensor and associated electronics was suffi-
cient to allow visualization of the pulsatile ICP waveform. In
addition periodic measurements could be plotted over time,
leading to the discovery of the well-known and clinically
useful Lundberg waves. Continuous ICP monitoring has
become the cornerstone of critical care monitoring for patients
admitted to neurocritical care units (NCCUs).
From an engineering point of view, it is interesting that the
application of a simple electrical sensor and analog electronics
resulted in a significant improvement in clinical practice. This
is a pattern that has repeated itself in the years since Lund-
berg’s original work in the 1960s. The remarkable develop-
ments in electronics and sensing technologies driven by the
computer, telecommunications, and aerospace and defense
industries have been applied to many aspects of medicine
including monitoring devices used in neurocritical care.
Clinical Background
An examination of engineering issues associated with neuro-
monitoring devices must be made in the context of the goals
and challenges of neurocritical care. Neurocritical care patients
commonly suffer from traumatic brain injury (TBI), from
neurovascular diseases such as subarachnoid hemorrhage, or
have undergone a neurosurgical procedure such as resection
of a brain tumor. In addition to the original injury or disease,
patients are at risk of “secondary” injury because of the unique
nature of the head and brain. One of the primary goals of
neurocritical care is the prevention of these secondary inju-
ries. The purpose of monitoring equipment used in neuro-
critical care is to enable the clinician to identify signs and
symptoms of the primary disease, to warn of impending sec-
ondary insults, and to help judge the efficacy of treatment.
Because the brain is encased in the rigid skull, there is
limited room for the brain to swell when injured. Left
unchecked, swelling (edema) can lead to serious morbidity or
384
death. For this reason ICP monitoring is important in neuro-
critical care. Nutrients such as oxygen and glucose are sup-
plied and waste products removed from the brain through the
blood circulation. Although the brain is a relatively small
organ at 2% of body weight, it receives 20% of the body’s
blood flow and accounts for 20% of oxygen and 25% of
glucose consumption.3 Significantly, the brain does not store
oxygen. Alterations in blood flow are common in neurocritical
care patients, and there is much interest in monitoring blood
flow in both the large vessels leading to and from the brain
and the microvasculature within the brain parenchyma. Sys-
temic blood pressure is commonly monitored because it is the
driving force for cerebral blood flow (CBF). The concentra-
tion of carbon dioxide (CO2), which is a powerful vasodilator
and affects CBF is indirectly monitored in the expired breath.
CO2 also has been monitored directly in the brain. Oxygen
tension can be measured in both the circulation and in brain
tissue. The oxygen saturation of systemic blood is routinely
monitored, and estimates of regional saturation of cerebral
blood also are possible.
It is common for alterations in cerebral metabolism to
occur in neurocritical care patients. The brain normally pro-
duces energy in the form of adenosine triphosphate (ATP)
through the Krebs cycle; oxygen and glucose are necessary for
this ATP production. When cerebral metabolism is disturbed,
for example, by a lack of oxygen availability, the brain relies
more on anaerobic glycolysis that produces less ATP. Under
these conditions, brain cells may not function normally and
may even lose structural integrity.4 This condition is some-
times called hyperglycolysis. This condition as well as other
metabolic disturbances can be inferred by various chemicals
released by the brain. A portable analyzer can test samples
collected from the interstitial fluid by dialysis probes implanted
directly in the brain to determine the concentration of the
chemical in question. The clinical utility of these measure-
ments is the subject of significant research.
Neuromonitoring Systems:
An Engineering Prospective
The Ideal Monitor
The ideal monitoring system would be noninvasive, provide
continuous information, interrogate the entire brain, present
the information in a way that is easily understood, be compat-
ible with other monitoring and imaging systems including
magnetic resonance imaging (MRI), take up little or no space,
© Copyright 2013 Elsevier Inc. All rights reserved.
 Section VII—Computers, Engineering, and the Future 385
meet all regulatory requirements both medical and technical,
and cost very little. Few if any currently marketed devices meet
all of these requirements. The following sections describe the
basic components of most common existing monitoring
systems, give an overview of the types of sensing techniques
that are or have been used, and assess the pros and cons of
each technique.
Most monitoring systems used in neurocritical care consist
of a sensor in contact with the scalp or directly implanted into
the brain, some kind of fixation device that keeps the sensor
in place during the monitoring period, connecting cables, and
a stand-alone electronic monitor to operate the sensor and
display data or connect to a bedside monitor.
Fixation Techniques
Fixation is an extremely important aspect of monitoring
systems. Fixation systems must ensure consistent alignment or
contact between the sensor and the patient. Two primary fixa-
tion techniques are used for indwelling sensors: bolt fixation
and tunneling. Most indwelling sensors take the overall physi-
cal shape of a long slender cylinder usually referred to as a
catheter or probe with the sensor located at the tip of the probe
(Fig. 39.1).
Bolt Fixation
The bolt device is composed of a short metal cylinder with
self-tapping threads at one end and some kind of compression
fitting at the other end. The bolt is usually threaded to a depth
of approximately the thickness of an adult skull of approxi-
mately 0.7 cm. The thread pitch is usually sized such that at
least three or four fully formed threads will be in contact with
the skull with the bolt in place. If the thread pitch is too large,
not enough contact is made by the thread crests to ensure a
Fig. 39.1  Bolt fixation device. Licox oxygen sensing catheter and
fixation bolt. (Used with permission of Integra LifeSciences Corporation.)
secure and leakproof fit. The lead-in threads are usually
tapered in a manner similar to pipe threads with the crest
height of the initial one to three pitches smaller than the
diameter of the twist drill used to cut the insertion hole in the
skull. The compression fitting must resist axial movement of
the probe and provide a leakproof seal around the outer
surface of the probe. The maximum compressive force that
can be exerted must be limited to an amount that will not
damage the probe. Providing strain relief at the junction
between the top of the bolt and the probe is a good practice.
The ultimate strength of both the bolt and probe in bending
and tension must be carefully considered because the probe
and bolt often are subjected to significant loading when the
probe or a connecting cable is inadvertently pulled when it
catches on a fixed object during patient transport, the patient
falls out of bed with the cable wrapped on the bed frame or
other stationary object, or the patient pulls on the probe in a
semiconscious state. It is far better for the bolt or probe
(outside the patient) to break than the bolt to break in the
skull (Fig. 39.2).
Tunneling
Perhaps the oldest fixation method is tunneling. Tunneling
refers to the surgical technique of routing an elongated cath-
eter under the scalp toward the insertion site. This method has
the advantage of good fixation to the patient using stitches in
the scalp and better infection control than simply routing the
catheter out of the scalp directly over the insertion site. This
technique is commonly used with fluid-filled ventriculostomy
ICP monitoring. It also has become common for transducer-
tipped probes to be affixed to the patient via tunneling. It is
important to recognize that the probe must be capable of
making a right-angle turn into the skull into the insertion site
under the scalp without being damaged. Because the connec-
tors on the proximal end of most probes are too large to be
tunneled under the scalp, many of these devices are not
Fig. 39.2  Bolt fixation device. Camino intracranial pressure bolt in situ.
(Used with permission of Integra LifeSciences Corporation.)
386 Section VII—Computers, Engineering, and the Future
Fig. 39.3  Tunneling fixation. Ventrix intracranial pressure catheter
tunneled under scalp. The inset shows the sensor in the tip of the
ventricular catheter. Note the shape of the probe tip (bullet-like) that
limits tissue injury during insertion. (Used with permission of Integra
LifeSciences Corporation.)
tunneled per se but are routed through a plastic sleeve that
has itself been tunneled under the scalp. This “tunneling
sleeve” usually takes the form of a plastic tube connected at
one end to a solid trocar. After the sleeve has been tunneled
under the scalp, the trocar is cut from the sleeve and the cath-
eter is passed through the sleeve toward the insertion site. The
tip of the catheter is then implanted and the catheter and/or
sleeve are sutured to the scalp via loops (Fig. 39.3).
Other Fixation Methods
Noninvasive monitoring devices such as electroencephalo-
graph (EEG) electrodes, near infrared (NIR) oximeters, and
transcranial Doppler (TCD) probes are affixed by hand, by
mechanical means, or by the use of some kind of adhesive
tape–like substance. Heavy devices that are not usually used
to make continuous measurements such as TCD transducers
are simply held to the head by the technician or are sometimes
fitted to a device that looks like an eyeglass frame. Lighter
transducers like EEG electrodes or oximeter leads usually are
held in place by adhesive patches. The adhesives used must
not irritate the skin during monitoring durations of up to
several days, must not cause hair to be removed when the
patch is removed, and must securely hold the transducer in
place over the monitoring period.
The Bedside Monitor
Most if not all neurocritical care patients are connected to
bedside monitors (BSMs). The BSM collects signals from
external monitors and displays them on a single screen. Gen-
erally each parameter is connected to the BSM via detachable
“modules” that are part of the BSM and accept analog or serial
input from an external monitor. Each module is tailored to a
Fig. 39.4  Hewlett-Packard bedside monitor connecting cables and
Camino monitor. (Used with permission of Integra LifeSciences Corporation.)
specific parameter type such as pressure or temperature.
Commonly displayed parameters include electrocardiogram
(ECG), heart rate, temperature, blood oxygen saturation,
cardiac output, and mean arterial blood pressure among other
options. Most physiologic pressures such as arterial pressure,
pulmonary artery pressure, and pulmonary capillary wedge
pressure are monitored using strain gauges. BSMs are there-
fore designed to provide electrical excitation of the strain
gauge and to interpret pressure from the electrical signals
returned form the strain gauge. Neurologic-related parame-
ters, other than ICP signals measured by fluid filled systems,
are generally not included in the current generation of BSMs.
Devices that measure parameters that are not included as stan-
dard features of the BSM commonly connect to it via analog
strain gauge emulation or by serial means when available.
Cable connectors must accommodate the various makes and
models of BSMs (Fig. 39.4).
Primary Sensors
The injured brain can be a difficult sensing environment. In
particular intracranial hemorrhage of one form or another
results in free and clotted blood both of which can damage or
confound many types of indwelling senor elements. Blood or
large proteins in the cerebrospinal fluid (CSF) also can deposit
on indwelling sensors and so damage or confound them.
Monitoring durations may continue for several days and
increase the chances that blood, blood clots or proteins may
deposit or form on implanted sensors.
Invasive catheters must be stiff enough to be pushed past
an opening in the dura mater and into parenchyma anywhere
from a few millimeters to several centimeters, such as with a
ventricular catheter. However, once implanted, catheters
should be flexible enough to impart the least amount of force
on the tissue if overloaded. The hair, scalp, and skull of varying
thickness are all factors to account for when using noninvasive
monitors.
 Section VII—Computers, Engineering, and the Future 387
Many primary sensor technologies have been used in neu-
rocritical care monitors. Electrical, optical, chemical, and
ultrasound are some of the broad categories.
Electrical Sensors
Strain Gauge
Doped silicon piezoresistive strain gauge sensors are used
commonly to transduce ICP. Miniaturized versions can be
placed in the distal tip of catheters, whereas larger versions
are placed outside the head in fluid communication with the
ventricle or less commonly the subarachnoid space. Both full-
and half-bridge devices, with and without temperature com-
pensation, are common. These devices are accurate and
reliable and in the larger sizes low cost, because these larger
devices are manufactured in large quantities for use in periph-
eral arterial line blood pressure monitoring. The use of the
strain gauge to sense pressure has the added advantage that
BSMs are designed to operate them directly; this eliminates
the need for complicated electronics between the sensor and
the BSM. The wire bond junctions must be protected from
corrosive environments, often with low modulus adhesive
coatings. The chip must be insulated from mechanical strain
other than that imparted by what is to be measured, such as
ICP. This is a difficult design requirement, and much intel-
lectual property and many trade secrets revolve around this
issue (Fig. 39.5).
Polarographic Electrodes
Electrochemical “polarographic” electrodes are used to mea­
sure oxygen tension in the brain parenchyma. A metal cathode
and anode and electrolyte solution are contained in an oxygen-
permeable plastic tube. Oxygen molecules that have diffused
through the probe wall react with water at the surface of the
cathode to produce hydroxide ions. Free electrons then are
liberated, causing current flow between the cathode and
anode. This current is sensed by the monitor and is propor-
tional to the partial pressure of oxygen in the tissues in contact
with the probe. This method has proven to be accurate and
reliable and does not require complex optical or electronic
devices for operation (Fig. 39.6).
Fig. 39.5  Strain gauge pressure transducer within NeuroSensor ICP/CBF
probe. ICP/CBF, Intracranial pressure/cerebral blood flow. (Used with
permission of Integra LifeSciences Corporation.)
Electroencephalogram
EEG sensors are simple electrodes typically connected by a
single wire to one side of a differential amplifier with the other
side connected to a common reference. Arrays of up to 256
leads are common and can measure the electrical activity of
the brain through the scalp, the dura, pia mater, or directly
in the cortex. EEG monitoring is a very mature technology,
but there have been several recent improvements including
noise suppression and signal processing techniques. In addi-
tion, improvements in analog electronics and the advent of
digital signal processing (DSP) have made EEG monitoring
more reliable and easier to understand for the clinician. For
example, useful time and frequency domain transformations
are now easy and inexpensive to implement with commer-
cially available DSP chips and software. Because of these tech-
nologic advances, continuous EEG is used more frequently in
NCCUs, and its use has provided insight into physiologic
effects of brain injury and how seizures themselves may cause
secondary injury.5
Optical Sensors
Intensity Modulation
Many optical methods have been used to sense neurologic
parameters such as pressure, oxygen tension, and blood oxygen
saturation. Most indwelling optical sensors rely on a fiber-
optic connection to optical components outside the head.
Optical ICP sensors use a primary mechanical sensor and an
optical proximity detector to transduce pressure. One com-
monly used device makes use of a miniaturized thin-walled
bellows closed at one end as the primary transducer. The
intensity of light reflected by the closed end of the bellows
from one optical fiber to another is used to sense the bellows’
position. The geometry of the reflections modulates the inten-
sity of the returned light signal, which in turn is proportional
e−
Current measurement
+ −
700 mV polarization
1
3
5
2 4
O2
Fig. 39.6  Licox oxygen sensor that uses polarograph electrodes.
1, Oxygen-permeable plastic tube; 2, cathode; 3, anode; 4, electrolyte
solution; 5, tissues in contact with probe. (Used with permission of Integra
LifeSciences Corporation.)
388 Section VII—Computers, Engineering, and the Future
to pressure. The only optical components used are the optical
fibers, a light-emitting diode (LED), and two photodiodes, all
of which are contained in the probe.
Interferometer Proximity Detection
Another optical pressure sensor device used to measure ICP
uses an interferometer to measure the depth of an optical
cavity formed by a silicon diaphragm bonded over a cavity
etched in a small cylinder of borosilicate glass. The cavity
is vented to atmosphere on its proximal side; therefore,
changes in cavity depth caused by displacement of the silicon
diaphragm are proportional to ICP. This sensor has the advan-
tage of being formed of all glass materials, which reduces the
effects of differential thermal expansion or water take-up
causing apparent pressure changes not due to a change in the
patient’s ICP.
Fluorescence Quenching
Optical sensors have been used to measure oxygen and other
parameters in the brain using the technique of fluorescence
quenching. Typically silicone room temperature vulcanizing
(RTV) adhesive is applied to the end of an optical fiber. The
RTV is doped with dye that fluoresces at a specific wavelength.
The intensity or duration of the light signal generated by the
fluorescence also is altered by the presence of a target molecule
such as oxygen. The fluorescent dye also can be immobilized
by applying it to porous glass microbeads. The rate of fluores-
cent decay when the lamp or LED used for excitation is turned
off is used to measure temperature in some devices. These
systems can be reasonably accurate and cost effective.
Spectroscopy
Optical techniques also may be used to measure concentration
of hemoglobin and the total oxygen saturation of hemoglobin
in blood. The absorption of light at convenient wavelengths
in near-infrared (NIR) near the isobestic point for hemoglo-
bin are measured and used to determine the absolute con­
centrations of oxyhemoglobin, deoxyhemoglobin, and total
saturation in blood. The same principle is used in noninvasive
cerebral oximeters to estimate regional blood oxygen satura-
tion in brain tissues. The forehead is illuminated by two emit-
ter-receiver pairs with NIR light at two wavelengths. The
attenuation of the returned light signals is used to estimate
regional blood oxygen saturation.6 These systems appear to
function reasonably well in uninjured brain but can be con-
founded by factors such as intracranial bleeding. They also
suffer from an inability to depth resolve the return signals; this
may lead to inaccuracies due to signals from scalp blood. More
sophisticated techniques can be used to overcome many of
these obstacles; however, these techniques are not in commer-
cially available systems. Chapter 33 contains a more detailed
description of these methods.
Laser Doppler Flowmetry
Optical methods are used to estimate blood flow in brain
parenchyma. The well-known laser Doppler flowmetery tech-
nique has been used in many products over the past 15 to 20
years. The tissue is illuminated with collimated light from a
laser source in the near infrared. Light is scattered by station-
ary structures in the tissues as well as the moving red blood
cells. Light reflected by the moving red blood cells undergoes
a Doppler phase shift dependent on the blood cells’ velocity.
This in turn causes the spectra of the returned light to be
broadened by some 20 Hz to 20 kHz. This spectral broadening
is used to estimate the velocity of the moving red blood cells.
The intensity of the returned signal is used to estimate their
concentration, and the product of the two terms, commonly
referred to as flux, is proportional to blood flow.7 This mea-
surement technique is attractive because it is continuous,
allows for the visualization of the blood flow pulse waves, and
can be used in febrile patients. The technique, however, is not
quantitative.
Thermal Diffusion
Thermal diffusion techniques are used to measure blood flow
in the parenchyma. Two thermistors are located near the distal
tip of a probe. Current is passed through one thermistor to
raise its temperature a predetermined amount above the tissue
temperature. Heat from this thermistor is transferred by both
conduction and convection. The convective heat transfer is
due to blood flowing past the probe. Knowledge of thermal
properties of the parenchyma, the temperature rise of the
thermistor, the power required to maintain the thermistor at
temperature, and other factors can be used to estimate the rate
of convection and therefore the blood flow near the probe tip.8
Other Sensors
Microdialysis
Microdialysis probes are used to collect samples of chemical
substances from the interstitial fluid. The catheter is composed
of a small-diameter tube with a membrane at the distal end
filled with a perfusion fluid. The proximal end of the catheter
ends in a collection vile. Chemicals diffuse across the dialysis
membrane into a perfusion fluid inside the catheter. The
probe membrane can be constructed such that chemicals of
different molecular weight can be collected. The chemical of
interest is collected in a sample vial connected to the catheter.
The sample is then analyzed in a bedside chemical analyzer.
The list of potential substances that can be measured is long,
but in the clinical environment glucose, lactate, and pyruvate
are the most frequently analyzed substances. Microdialysis is
a excellent research tool, but its use in routine clinical practice
has been limited by the intermittent measurements and need
to transfer samples from the patient to the analyzer (Fig. 39.7).
Doppler
Ultrasound Doppler techniques are used to noninvasively
measure the blood flow velocity in the large conductive vessels
in the circle of Willis. The transducer head, including trans-
ducer and receiver, is positioned where the skull is thin (tem-
poral region) and the ultrasound energy, typically using
frequencies of around 2 MHz or a multiple thereof, is focused
on the vessel of interest. The phase shifted return signal is used
to calculate velocity. Although accurate blood velocity mea-
surements can be made, it is harder to use as a continuous
monitor because it can be difficult to maintain contact and
alignment of the transducer head.
 Section VII—Computers, Engineering, and the Future 389
Fig. 39.7  Microdialysis system showing microdialysis catheter, pump and
analyzer. (Courtesy CMA Microdialysis.)
Safety
Safety is a paramount consideration in the design, manufac-
ture, and use of monitoring devices in neurocritical care.
Safety hazards generally relate to the probe and sensor, ancil-
lary components such as fixation devices, connecting cables,
and the electronic monitor used with the sensor. Manufactur-
ers of medical device products use formal risk analysis proce-
dures to identify hazards and mitigation methods and to
determine whether the residual risks after mitigation are
acceptable. The list of potential safety concerns is exhaustive;
the more important issues posed by neurocritical care moni-
toring devices are discussed in the following text.
Probe and Ancillary Components
The shape of the distal tip of implanted probes should be
given careful consideration. An ogive or bullet-nose shape is
probably the best and safest design. A completely blunt shape
with a sharp transition between the axial tip and the outer
diameter of the catheter tip is probably the worst because it
tends to tear tissues during implantation. The behavior under
unanticipated loading of components that are implanted in
the patient, the connection to components implanted in the
patient, or which secure implanted parts should be considered
to limit transmission of these loads to the patient as much as
possible. The methods of sealing implanted probes and fixa-
tion devices against leaks must be given careful thought to
minimize the chances of infection. Materials that contact the
patient must be chosen to minimize irritation and must be
tested for biocompatibility, for example, per International
Standards Organization (ISO)10993 Biological Evaluation of
Medical Devices.
Electronic Components
Electronic monitors and cables pose their own set of safety
concerns. Malfunctioning electrical devices can display erro-
neous data, cause burns and electrical shock, and interfere
with the proper functioning of other electronic devices. Elec-
trical safety issues such as patient electrical isolation, leakage
currents, radiated emissions, susceptibility to radiated emis-
sions, flammability, and a host of other potential safety issues
are covered in the electrical safety standard International
Electrotechnical Commission (IEC) 60601 Medical Electrical
Equipment—Part 1: General Requirements for Safety and
related documents. Medical devices sold in the United States,
the European Union, and many other parts of the world must
meet this standard as tested by an independent test laboratory
such as Underwriters Laboratories (UL), Technical Inspection
Association (TUV), British Standards Institution (BSI), and
the like.
Magnetic Resonance Imaging Safety
An area of increasing concern is the safety of medical devices
in the MRI environment. Devices made of or containing metal
can translate in the scanner due to the strong magnetic field
and can rotate to align with the direction of the magnetic field,
imparting torsional forces. In addition, lead wires or other
elongated components capable of carrying current can couple
with the radiofrequency (RF) field in the scanner causing the
device to heat, particularly at the distal tip. There are reports
of patients and others sustaining injuries due to each of these
safety issues. The American Society for Testing and Materials
(ASTM) International in close cooperation with the U.S. Food
and Drug Administration (FDA) has developed test standards
to measure magnetically induced displacement force, mag-
netically induced torque, and RF-induced heating of passive
implants along with a guide for MRI safety labeling.9-11 Of the
three test methods RF heating is by far the most difficult and
complex. The understanding of factors that influence RF
heating in the MRI environment is rapidly advancing and as
such, testing requirements are in flux.12 As of this writing the
revision of the standard for measurement of RF-induced
heating on or near passive implants is ASTM 2182 11a. In
addition, a joint working group of the IEC and ISO has devel-
oped a test standard (TS) for MRI safety testing of active
implantable medical devices (AIMDs) that is a precursor for
an International Standards document. RF heating is influ-
enced by the strength of the electromagnetic fields present in
the scanner. These fields vary spatially depending on such
factors as location within the RF coil and interactions between
the electromagnetic fields and the device and patient. Current
best practice appears to be to measure the heating of the
device in a phantom within an RF coil or MRI system while
controlling or measuring the electromagnetic fields to obtain
a baseline understanding of the response of the device to a
well-understood set of conditions. This information is then
used in conjunction with computer modeling of the patient,
device, and the scanner to predict heating in actual use.
Regardless of the form the test method ultimately takes, it is
important for devices to be tested for MRI safety and for clini-
cians to closely follow the MRI safety instructions provided
by manufacturers (Figs. 39.8 and 39.9).
The Regulatory Environment
No discussion of engineering aspects of medical devices would
be complete without mention of the regulations that govern
the design, manufacture, and distribution of these devices.
Like the practice of medicine the business of medical devices
is highly regulated. In the United States the primary governing
laws are contained in the Code of Federal Regulations titled
Quality System Regulation.13 This set of regulations is enforced
by the FDA. The Council of European Communities Medical
Device Directives of 1993 that was last amended in 200714
390 Section VII—Computers, Engineering, and the Future
Electric Field with ASTM Phantom in RF Coil
200
–0.4
–0.2
150
0
Z (m)
100
0.2
0.4
50
0.6
0 This estimate can be compared with competing projects or
–0.2 0 0.2
X (m)
Fig. 39.8  Finite-difference time domain (FDTD) plot of electrical fields in
and around American Society for Testing and Materials (ASTM) phantom
within magnetic resonance imaging radiofrequency (MRI RF) coil. (Used
with permission of Integra LifeSciences Corporation 2008.)
Fig. 39.9  American Society for Testing and Materials (ASTM) phantom
in 1.5T magnetic resonance imaging (MRI) scanner during device MRI
safety test. (Used with permission of Integra LifeSciences Corporation 2008.)
governs the products made or sold in the European Union
(EU) and the manufacturers thereof. It is critical for engineers
to understand these and many other regulations as they
impact most technical decisions. Furthermore, it is important
to consider these regulatory issues early in design and devel-
opment to ensure that any new idea is feasible in the clinical
environment before expending time and labor. At present the
FDA divides medical devices into three classes according
to their level of risk to a patient. There are two alternative
regulatory standards: (1) 510(k) that requires the device be
similar to an already marketed device (i.e., it should be low
risk), or (2) premarket approval (PMA) that requires clinical
testing and inspection. However, device classification errors
can be associated with patient safety and recalls15 and so the
FDA asked the Institute of Medicine to review the classifica-
tion and review process in 2011; these recommendations are
pending.
Business Considerations
The majority of medical devices are sold by for-profit compa-
nies; therefore the decision to develop and market a medical
device is in large part an economic one. Most companies apply
a disciplined and analytical approach that uses one or more
forms of return on investment analysis. The cost of develop-
ing the product and the costs of making and selling it are
weighed against future revenue streams from sales of the
product. The time value of money is always a factor in these
calculations. One easy-to-understand approach is to deter-
mine the net present value of future revenues less all expenses.
simply the interest earned by placing the same amount of
money in a savings account. It is important for engineers and
clinicians to understand the basic economic principals that
govern how companies decide what products to develop and
when to develop them. Obviously products that address a
large market, are not costly to produce, command a high
price, and are inexpensive and quick to develop are favored.
References
1. Gullaume J, Janny P. Continuous intracranial manometry; physiopathologic
and clinical significance of the method. Presse Med 1951;59(45):953–5.
2. Lundberg N. Continuous recording and control of ventricular fluid pressure
in neurosurgical practice. Acta Psych Neurol Scand 1960;36(Suppl. 149):
1–193.
3. Magistretti PJ, Pellerin L, Martin J. Brain energy metabolism. From the
American College of Neuropsychopharmacology (ACNP) website
publication Psychopharmacology—The Forth Generation of Progress, 2000.
http://www.acnp.org/g4/GN401000064/Default.htm
4. Duke T. Dysoxia and lactate. Arch Dis Child 1999;81:343–50.
5. Vespa PM, Miller C, McArthur D, et al. Nonconvulsive electrographic
seizures after traumatic brain injury result in a delayed, prolonged increase
in intracranial pressure and metabolic crisis. Crit Care Med 2007;35(12):
2830–6.
6. Chance B, Cope M, Gratton E, et al. Phase measurement of light absorption
and scatter in human tissue. Am Inst Physics 1998;69(10):3457–81.
7. Shepherd AP, Oberg PA. Laser-Doppler blood flowmetry. Boston: Kluwer
Academic; 1990.
8. Martin GT, Bowman HF. Validation of real-time continuous perfusion
measurement. Med Bio Eng Comput 2000;38(3):319–26.
9. ASTM F 2052–00, Standard test method for measurement of magnetically
induced displacement force on passive implants in the magnetic resonance
environment.
10. ASTM F 2213-04, Standard test method for measurement of magnetically
induced torque on medical devices in the magnetic resonance
environment.
11. ASTM F 2182-02, Standard test method for measurement of radio
frequency induced heating near passive implants during magnetic
resonance imaging.
12. Kainz W. MR heating tests of MR critical implants. J Magn Reson Imaging
2007;26:1182–85.
13. 21 CFR Part 820—Quality system regulation, Revisions as of 31 March,
2006.
14. Council Directive 93/42/EEC of 14 June 1993 concerning medical devices,
Amended 11/10/2007.
15. Zuckerman DM, Brown P, Nissen SE. Medical device recalls and the FDA
approval process. Arch Intern Med 2011;171(11):1006–11. Epub 2011 Feb 14.

Multimodality Monitoring
and Artificial Intelligence
Richard S. Moberg and J. Michael Schmidt
Introduction
Multimodality monitoring can be defined as the simultaneous
collection of data from multiple diverse sources pertaining to
a single patient coupled with the ability to view the data in an
integrated and time-synchronized manner. Alternatively, mul-
timodality monitoring may be considered the use of more
than one complementary method to monitor a single organ,
when no one single method can provide complete informa-
tion. It is the first step to create a bedside environment where
advanced knowledge tools can be applied to the data to aid
the clinician in patient management. However, few, if any,
neurocritical care units (NCCUs) have such an environment
due to a variety of barriers (see Chapter 4). Assuming these
problems will be solved, this chapter reviews the creation of a
“knowledge infrastructure” in neurocritical care. The ratio-
nale and benefits are presented along with a historical perspec-
tive, current progress, and future prospects.
Rationale
An accepted goal of neuromonitoring in intensive care is to
enable the detection of harmful physiologic events before they
cause irreversible damage to the brain.1 Ideally monitoring
should be used to predict “events” in the intensive care unit
(ICU).2 However, the brain is a complex system with inter-
related hemodynamic, metabolic, and electrical subsystems.
From simply a logical perspective, the most effective monitor-
ing paradigm would be one that monitors multiple sites in this
multifaceted system. But to date, no formal studies have been
conducted that pair multimodal monitoring with a treatment
protocol to demonstrate better outcome, although one such
trial, the Brain Oxygen and Outcome Study in Traumatic
Brain Injury (BOOST), a phase 2 trial to examine physiologic
efficacy of multimodal monitoring, is likely to complete
enrollment in 2013.3 For the moment, however, a common-
sense rationale must suffice.
The need for a knowledge-rich infrastructure made possible
by multimodal data extends beyond the NCCU to all of high-
acuity critical care. First, in intensive care, clinical information
systems acquire and store physiologic variables and device
parameters online at least every minute.4 Physiologic signals
may contain useful data in frequencies of ~0.2 kHz. These data
can be lost when data less than 0.5 to 1 kHz is sampled over
a short time (1 to 2 ms).5 In addition, high-resolution
© Copyright 2013 Elsevier Inc. All rights reserved.
Chapter
40  
VII
physiologic data can be lost if there is no means to store or
archive it. In most NCCUs today, physicians and nurses can
only view continuous physiologic data by looking at the
bedside monitor. This information, once it has scrolled across
the screen, is lost, and if the staff is not at the bedside only
intermittently recorded data are available; that is, data collec-
tion and storage is essential to NCCU informatics. Second, in
the severely brain-injured patient, use of the clinical examina-
tion to detect changes in patient state is severely limited. Thus
physicians and nurses rely in large part on changes in physi-
ologic metrics and on results from diagnostic tests to make
patient management decisions. During rounds of critically ill
patients each morning, a physician may be confronted with
more than 200 variables.6 The high data volume and the time-
critical situations physicians are confronted with create a
problem.7 This is accentuated because human beings are not
able to judge the degree of relatedness between more than two
variables. Furthermore, constant information overload is one
contributing cause for preventable medical errors.8 A knowl-
edge infrastructure can provide a closer coupling between the
clinician and the data with the promise to decrease errors and
improve outcomes. Finally, as the aging population fills the
health care system, it is predicted there will be a significant
shortage of caregivers, particularly in the ICU.9 Because of
higher efficiencies, workflow software driven by multimodal
patient data has the potential for minimizing the number of
staff needed to accomplish the same amount of work.
The current federal (and international) emphasis on
medical economics provides further justification for multi-
modal neuromonitoring. The rising costs of health care in the
United States has led the government to conclude that better
information in a variety of areas could eventually alter the way
in which medicine is practiced and yield lower health care
spending without having adverse effects on health.10 Other
countries have come to similar conclusions, and such thinking
has prompted a rise in comparative effectiveness research that
evaluates the impact (primarily costs and benefits) of different
treatment options.11 This emphasis on informatics under-
scores the need for a more comprehensive and coordinated
approach to data collection, something that multimodal mon-
itoring can provide.
There also are benefits to patient safety in creating a knowl-
edge infrastructure. There are a few reports of accidental pa­
tient deaths that could have been prevented if medical devices
were able to intercommunicate. In one example, a ventilator
391
392 Section VII—Computers, Engineering, and the Future
was momentarily turned off to eliminate blurring of x-ray
images that were being obtained. The film plate jammed
under the table and the anesthesiologist went to help. In so
doing, the anesthesiologist forgot to restart the ventilator and
the patient died.12 Had there been a method for the x-ray
machine and the ventilator to communicate, the patient death
may have been prevented.
Neurocritical care has evolved as a distinct speciality dur­
ing the past decade. In large part patient management
depends on understanding patient physiology, so techniques
to monitor the brain also have evolved. Increasingly, NCCUs
use more than one monitor of brain function, and an increas-
ing number of publications and reviews allude to the poten-
tial benefits of multimodal monitoring in various aspects of
neurocritical care, including traumatic brain injury (TBI),13-
15,16
subarachnoid hemorrhage,17 management of cerebral
hemodynamics,18-20 and determination of brain death,21 and
in understanding patient physiology in general.22-25,24,26-30 Fur-
thermore when multimodality monitoring is used, it has
become apparent that single monitors may miss episodes of
cerebral compromise.31
Requirements of Multimodal
Neuromonitoring Systems
Recording multiple simultaneous measurements from a
variety of diverse sources brings about problems not
encountered in conventional monitoring. These problems
can be broadly classified as dealing with the following
requirements of monitoring: continuous, comprehensive,
and communicative.
Continuous Monitoring
Continuous monitoring is required in critical care to not miss
clinically significant events. More specifically, the frequency of
monitoring needs to be higher than the duration of the events
to be detected. Hemphill et al.32 compared medical records
(with approximately hourly entries) to automatic continuous
data recordings (approximately 1-second intervals) recorded
simultaneously in 16 patients and found that the medical
record data tended to underestimate the number of secondary
brain insults detected. They concluded that higher frequency
data (i.e., higher than is normally acquired in the medical
record) may be necessary for more precise evaluation of sec-
ondary brain injury in neurocritical care. A notable example
of the value of continuous monitoring is with electroencepha-
logram (EEG) recordings. In 1993, Jordan monitored EEG
continuously rather than intermittently and showed the
number patients with seizures and the frequency of seizures
present in critical care patients was much greater than what
was conventionally thought.33 He also showed that continuous
monitoring of EEG made an impact on clinical decision
making in 51% of patients (n = 124). Subsequently, Claassen
and Mayer34 showed that continuous EEG influenced patient
management on 50% of monitored days.
The problem inherent in continuous monitoring is the
amount of data collected when monitoring from multiple
sources. Storage of trend data and a few waveforms is insig-
nificant with current storage capacities. However, EEG is
generally recorded with a higher sampling rate and usually
with 16 or more channels. Assuming a 256-Hz sampling
rate, 16 channels, and 16-bit precision, storage of 1 day of
EEG requires more than 700 megabytes of space. Adding
simultaneous video of the patient can bring the storage rate
from 250 megabytes to more than 1 gigabyte per hour,
depending on the frame rate and type of video compression
used. Consequently, most institutions do not store all the
video data but rather just the segments of interest, a process
that requires manual intervention. Similarly manual inter-
vention may be necessary to exclude potential artifacts
or during periods when patients are disconnected from a
monitor.
Comprehensive Monitoring
Multimodal neuromonitoring must be comprehensive in that
all of the necessary data for particular monitoring goals
need to be collected simultaneously, time synchronized, and
displayed in an integrated fashion. This is a challenge in
the NCCU because most monitors stand alone and are self-
contained, provided by a commercial vendor. However, this
challenge is likely to diminish since the American Society for
Testing and Materials adopted the concept of an “integrated
clinical environment” in 2009.35 Time synchronization of data
received from multiple devices in a variety of formats presents
a significant problem to developers of multimodal data col-
lection systems. The data sampling rates may be widely differ-
ent and the data may be sent to the collection device
continuously or in packets with varying time intervals. The
accuracy of the time synchronization task, however, depends
on the use of the data. For slow varying trends where only a
visual correlation is to be made, a high degree of synchroniza-
tion may not be required.
There are some applications where very accurate time syn-
chronization of data is required. An example is combining
data from different sources to improve the quality of data. For
example, EEG data contaminated with electrocardiogram
(ECG) or pulse artifact can falsely trigger automated seizure
detection algorithms. Other physiologic signals can be used to
help remove or reduce such artifacts in the EEG using various
analysis methods such as adaptive filtering36 or independent
component analysis.37 For example, ECG can be used to help
remove pulse artifact and a respiration signal can be used to
aid in removing high-frequency ventilator artifact from the
EEG. Figure 40.1 shows EEG data (left panel) contaminated
with pulse artifact. Using simultaneously collected ECG, the
EEG can be adaptively filtered to remove the artifact (right
panel). The notation on the first channel of each panel shows
the difference in an amplitude envelope function caused by
the artifact (8.2 µV p-p with the artifact versus 5.4 µV p-p
without).
Communicative Monitoring
Once collected, the information in multiple data streams must
be extracted and communicated to the clinician. Perhaps the
simplest and most effective way to extract information is to
observe relationships in the data by plotting time-synchronized
trends on a single display. Such displays are standard on con-
ventional vital-signs monitors because they facilitate the visual
detection of correlations between measurements. Figure 40.2
shows an example that plots trends of ICP, arterial blood
 Section VII—Computers, Engineering, and the Future 393

EEG with ECG Artifact (shown at 50uV p-p) EEG after Adaptive Filtering with ECG Channel (shown at 50uV p-p)

T3-C3 T3-C3

90% PDF amp = 8.2 uV p-p 90% PDF amp = 5.4 uV p-p

C3-Cz C3-Cz

Cz-C4 Cz-C4

C4-T4 C4-T4

ECG ECG

180 181 182 183 184 185 186 187 188 189 190 180 181 182 183 184 185 186 187 188 189 190
Time (sec) Time (sec)

Fig. 40.1  Electroencephalographic data contaminated with electrocardiogram (ECG) artifacts (left panel) and after adaptive filtering with the ECG
(right panel) to remove the artifact.

50
ICP 40
[mm Hg] 30
20
10
200

ABP 150
[mm Hg] 100
50
200
150
FV 100
[cm/s] 50
0
90
85
TOI 80
[%] 75
70
22
20
PbtO2 18
[mm Hg] 16
22:10 22:15 22:20 22:25 22:30 22:35 22:40 22:45 22:50 22:55 23:00 23:05
Time [h:m]
Fig. 40.2  Simultaneous trends of 1 hour monitoring of intracranial pressure (ICP), arterial blood pressure (ABP), flow velocity (FV), tissue
oxygenation index (TOI), and brain tissue oxygenation (PbtO2). This provides a synchronized view of physiologic data. (Courtesy Dr. Marek Czosnyka.)

pressure (ABP), flow velocity, tissue oxygenation index (TOI),
and brain tissue oxygenation. Episodes of ICP increases are
associated with hyperemia due to unstable arterial pressure.
Figure 40.3 shows a more comprehensive visual display, a
neurocritical care dashboard, which includes data represented
by waveforms, images, and other formats in addition to trends.
Going further, simple statistical analyses on the recorded
data can be revealing. For example, Figure 40.4 shows
the change in brain oxygen versus cerebral perfusion pressure
at two points in time in the same patient using scatter
plots. It illustrates the dynamic nature of autoregulation in
neurocritical care. These plots can be calculated with most
spreadsheet software using data exported from a neuromoni-
toring data collection system or with more specialized statis-
tics software. These capabilities also exist in some of the
neuromonitoring systems described in the next section.
In summary, multimodality monitoring creates new prob-
lems with data collection, but it is anticipated that most will
be solved with near-term technology. For data analysis, there
are many ways to process multiparametric data to extract and
display information useful to the clinician. This is an active
area of critical care research.
394 Section VII—Computers, Engineering, and the Future

Fig. 40.3  A neurocritical care dashboard that includes information from multiple domains. (Courtesy Dr. Val Nenov.)

Cerebral Perfusion Pressure by Brain Oxygen Cerebral Perfusion Pressure by Brain Oxygen
Patient 105, day 1 11:00 am − 1:00 pm Patient 105, day 1 11:00 am − 1:00 pm
r2 = 0.6199 r2 = 0.0.2586
34 34

32
32
30

28 30
Brain oxygen

Brain oxygen

26
28
24

22 26

20
24
18

16 22
45 50 55 60 65 70 75 80 85 90 95 100 54 56 58 60 62 64 66 68 70 72 74 76 78 80
Cerebral perfusion pressure Cerebral perfusion pressure
Fig. 40.4  Scatter plots of brain oxygen vs. cerebral perfusion pressure on subsequent days made using the Statistica Data Mining software.
(Courtesy Dr. Micheal Schmidt.)

Creating a Bedside Knowledge
Environment
The Role of Artificial Intelligence
Once continuous multimodal data can be recorded, it will
be possible to create a bedside knowledge environment that
couples decision support to patient management. These
methods are largely in the domain of artificial intelligence
(AI). AI is defined as the science and engineering of making
intelligent machines and was coined by computer scientist
John McCarthy in 1956.38 It is a field that attempts to trans-
form data into reasoning and uses a variety of tools such as
artificial neural networks (ANN) and statistical classifiers.
Neural networks are composed of “neuron-like” software ele-
ments that, in the aggregate, can exhibit simple learning
behaviors. They have a particular use in the analysis of neuro-
critical care data in that they can be trained to find patterns
in data both in a generalized way and within a single patient.
Although the potential of AI is significant, publications
about using AI to analyze data specifically in neurocritical care
 Section VII—Computers, Engineering, and the Future 395
are sparse, most likely due to the challenges to collecting mul-
tiparametric data. Some of the work is reviewed here and then
the promise AI has to offer in creating a knowledge environ-
ment at the bedside is discussed. A number of references can
provide a more in-depth explanation of AI concepts.39,40
An active area of AI research in critical care is to attempt to
improve the accuracy of alarms and the detection of events.
Most clinicians are familiar with the problem of alarms in a
multimonitor environment, such as critical care or surgery.
Imhoff 41 recently reviewed the literature and showed that up
to 90% of all alarms in critical care monitoring are false
alarms, which leads to a dangerous desensitization of the staff
toward true alarms.42 Several studies have shown encouraging
results with the application of AI techniques to alarm manage-
ment,43-45 but these techniques have not made their way to
routine clinical use, most likely because of liability concerns
of medical device vendors.
AI techniques can be used to assess the context of the
patient, taking into consideration the variability among
patients, to provide more individualized monitoring. For
example, Apiletti and colleagues used real-time classification
techniques to detect different severity levels of a monitored
patient’s clinical situation,46 whereas Zhang and Szolovits,
using neural networks, showed patient-specific modeling in
real time to be feasible and effective to generate bedside
alerts.47 An analysis of the dynamics of the critical care work
environment as a complex cognitive system was performed by
Patel et al.; this work has implications for the design of deci-
sion support tools.48
Specific to neurocritical care, Nelson et al.49,50 analyzed
patterns of cerebral microdialysis in patients with TBI and,
with a neural network methodology, investigated pattern
Fig. 40.5  A display of automated seizure detection (red colored trace) that uses a combination of statistical techniques and an expert system approach
(in a neonatal electroencephalogram).
relationships to intracranial pressure and cerebral perfusion
pressure. Minhas et al.51 used decision tree analysis to iden-
tify which patients would be colonized with antibiotic-
resistant bacteria on admission to the NCCU. Several groups
have studied the prognostic and predictive abilities of AI
techniques. For example, Swiercz et al.52 used neural
network techniques with a wavelet algorithm incorporated
to predict ICP values, whereas Nikiforidis et al.53 constructed
a Bayesian belief network (BBN) to predict short-term prog-
nosis of TBI patients in the intensive care unit, and Väth
et al. used ANN to analyze the accuracy of outcome predic-
tion after TBI.54
Discriminate function analysis is used to determine which
combination of variables discriminate between two or more
naturally occurring states. The resulting discriminate function
can be used in monitoring to alert the clinician to a change in
the state of the patient. The bispectral index (BIS) is an
example that integrates several disparate descriptors of the
EEG55 and has gained wide use in surgery to assess anesthetic
effects on the brain and has been used as a measure of sedation
in critical care (see Chapters 9 and 25).
Another AI approach is used when “human expertise” is
needed to solve problems. Human expertise is formalized as a
knowledge base, a series of rules, or other means (or a com-
bination of these) so the computer can use this information
to process the data. Expert systems, as these are called, often
are used to augment statistical analyses when they are not
robust enough to provide the desired results. Figure 40.5
shows the results of an automated seizure detection algorithm
developed using an expert systems approach. The first pass of
the analysis calculates metrics for a variety of features such as
rhythmicity and amplitude. The second pass, using expert
396 Section VII—Computers, Engineering, and the Future

systems tools, adds the knowledge of an expert epileptologist
and follows a rule-based approach to evaluate the metrics for
conditions such as their stability or fluctuation over time and
symmetry of activity in adjacent channels. This combination
of techniques significantly increases the accuracy of the seizure
detection and allows a more precise determination of seizure
start and stop on a per-channel basis.
Historical Notes and Review
of Existing Systems
There are very few commercial systems that integrate neuro-
logic data. Consequently a few university-based NCCUs have
developed their own systems. These various systems and range
of capabilities are reviewed briefly.
University-Developed Systems
Some of the earliest publications in multimodality neuro-
monitoring came from the University of Graz, Austria, in the
late 1980s.56,57 Beginning mostly with the simultaneous collec-
tion and integration of electrophysiologic signals (EEG and
evoked potentials), the researchers have added a full com­
plement of physiologic signals in their data collection system
and have used the system to investigate a variety of procedures
in surgery and critical care.58 Figure 40.6 illustrates the
integrated display of signals in a patient in the NCCU whose
condition deteriorated to brain death.
Starting in the early 1990s, researchers in the Department
of Neurosurgery at the University of California Los Angeles
(UCLA) began collecting multichannel EEG signals continu-
ously on every patient in neurocritical care. EEG is one of the
more challenging signals to collect continuously and to review
because of the volumes of data collected. The UCLA group
therefore developed the percent alpha variability (PAV), which
condensed certain features of the EEG into a variable that
could be trended along with other physiologic parameters in
a multimodal fashion.59 The PAV is associated with outcome
in TBI and among other diagnoses provides insight about
ischemia.60,61 The data collection capabilities of the UCLA
system also include other physiologic parameters, patient
demographics, and images. This integrated “dashboard” is
illustrated in Figure 40.3.
Investigators interested in ICP in the NCCU at Adden-
brooke’s Hospital in Cambridge, England, developed a system
that collects data from multiple sources so it could be corre-
lated to ICP.22 This multimodality monitoring system is based
on a standard personal computer equipped with a digital-to-
analog converter and serial data interfaces, and has evolved
from the1980s into a system, ICM+, for multimodal NCCU
monitoring and ICP waveform analysis.22,62
At San Francisco General Hospital, the University of Cali-
fornia San Francisco (UCSF) Brain and Spinal Injury Center

EEG BAEP I-V IPL SSEP CCT HR HRV ICP BP Temp.

Zervikal Kortikal Sys Dis
16:51

<2%
15:40

N13 Hirntod
15:15

>2%

14:10

12:29

8:15

N20
6:50

ICP
0.5
0.1 µV 0.5
µV µV 6:20
0 24 0 24 0 10 3 6 5 20 15 30 3 12 20 120 0 20 0 50 60 200 60 140 36 39
Hz Hz msec msec msec msec msec 1/min % mm Hg °C

Fig. 40.6  Multimodal data recording from a 57-year-old woman with an intracerebral hemorrhage in the left hemisphere that shows a
progression to brain death following an intracranial pressure (ICP) increase. The recorded modalities from left to right are electroencephalo
graph (EEG) compressed spectral array (two channels), brainstem auditory evoked potentials (BAEP), the interpeak latency (IPL) of peak I to IV of the
BAEP, somatosensory evoked potentials (SSEP) with the N14 and N20 marked and the central conduction time (CCT) calculated, heart rate (HR), heart
rate variability (HRV), ICP, blood pressure ([BP], systolic and diastolic), and temperature. Zervical and Kortikal refer to the cervical and cortical responses
from the SSEP. Hirntod indicates brain death. (Courtesy Dr. Gerhard Litscher.)
 Section VII—Computers, Engineering, and the Future 397

Fig. 40.7  ICM+ software developed at the Cambridge Neurosurgical Unit.

has an integrated data repository that attempts to link bedside

critical care physiology, critical care charting, and the hospital
patient electronic medical records. These represent three dis-
tinct systems: the locally developed Aristein system (bedside
continuous physiology data collected at least once each
minute from each ICU bed), Philips’ Carevue critical care
bedside nurse charting system, and Siemens’ Invision elec-
tronic medical record. Selected data from each system is
brought together using a distinct patient identifier into a
common server that is located behind the hospital firewall.
These data are then coded daily and Health Insurance Porta-
bility and Accountability Act identifiers removed before the
data are passed outside the firewall into a metadata ware-
house that can be used for advanced informatics analysis.
The ultimate goal is to make this metadata warehouse avail-
able for pooling of coded anonymous data from multiple
institutions.63

Commercial Systems
Several of the available commercial multimodal neuromoni-
toring systems originated from those developed in university
and research settings. Global Care Quest (Aliso Viejo, CA)
was formed to commercialize the data integration work at
UCLA (see Fig. 40.3). ICM+, a Windows program for multi-
modal neuromonitoring, grew out of the system used at
Cambridge University (Fig. 40.7). The author’s [R.M.] early
work in data integration (Fig. 40.8) led to a research system Fig. 40.8  An early multimodality neuromonitoring system deve­
(CNS Technology, Ambler, PA) used for data collection in loped by the author [RM] in 1996. It collected data from a vital signs
multicenter clinical trials funded by the National Institutes of monitor, transcranial Doppler, brain oxygen (RunMan from Dr. Britton
Health. It is now a commercial multimodal data collection Chance), laser Doppler flowmetry, ScVO2, and EEG. The system was used
and display system called the Component Neuromonitoring to investigate software architectures for multimodal data collection and
System (CNS) Monitor. in its current iteration is a single device operated through a touch screen.
398 Section VII—Computers, Engineering, and the Future
Fig. 40.9  ICUpilot screen shot.
Several companies that manufacture a monitor for a spe-
cific modality have developed research systems that record
additional data, and some of these systems are commercially
available. GMS (Kiel-Mielkendorf, Germany), which initially
developed the Licox brain tissue oxygen monitor, developed
a product called the MMM (Multi Modality Monitor) that
collected data from a variety of related monitors. GMS
became part of Integra (Plainsboro, NJ) that has the
Moebius system for collection of multimodality data. CMA
Microdialysis (Solna, Sweden), the clinical arm of which is
now known as Dipylon Medical, developed the ICUpilot
software that is widely used for analysis of data. The soft-
ware collects data from a variety of neuromonitoring devices
and other physiologic monitors, although it is integrated
into the CMA analyzer, and stores the data in an SQL data-
base. The user then can review the data in a variety of
formats and process the data with built-in statistical tools
(Fig. 40.9).
Manufacturers of continuous EEG monitors have added the
capability to record and trend other parameters along with the
EEG trends. Most notably, the Axon Systems (Hauppauge,
NY) Eclipse Workstation, although focused on intraoperative
monitoring, can record EEG, evoked potentials, physiologic
trends, transcranial Doppler, and a variety of other signals
(Fig. 40.10).
Capabilities of vital sign monitors are constantly expand-
ing to incorporate new measurements and most can trend
ICP, cerebral perfusion pressure, brain oxygen, and a limited
number of EEG channels. The expectation is that multimodal
neuromonitoring will be integrated into conventional vital
signs monitoring in the future. But, at present, the comput-
ing power in most vital sign monitors is not adequate to
provide the advanced processing (e.g., multichannel spectral
analysis of EEG, seizure detection) that is required in neuro-
critical care.
At the NCCU at Columbia University College of Physicians
and Surgeons, efforts have been made to get multiple com-
mercial products to work together. Capsule Technologies
(Paris, France) is one of several companies that provide
general purpose data integrating systems. Capsule Technolo-
gies provide a wide variety of device interfaces and hardware
to collect trends of numerical data on a central server.
Although targeted at workflow and patient record applica-
tions, the technology has been used in the Columbia
University NCCU as one component of their multimodal
neuromonitoring system. Excel Medical Electronics (Jupiter,
FL) produces the BedMaster software, which, like the Capsule
Technologies product, integrates data from multiple sources
and provides an integrated display of that data. An advantage
of the BedMaster software is that it is able to emulate a
monitor display and it allows remote access to patient data
information through a Citrix server (Fig. 40.11). To visualize
multimodality data, ICUpilot is used to query the BedMaster
database. A separate system is used to collect EEG data from
each bedside.
In summary, current products or combinations of products
can provide multimodal data collection. Each system has
potential advantages and disadvantages but in large part each
is a means to display multiple data sources and at best inte-
grate them. However the ideal information environment in
neurocritical care has still to be developed and in particular
interpretative systems or decision support does not yet exist
in the NCCU. This future is the focus of the next section.
The Future
While the future of neuromonitoring may rely on nanotech-
nology or molecular solutions,64 monitoring devices are here
to stay for the foreseeable future. There are several trends that
will influence the course of multimodal monitoring.
 Section VII—Computers, Engineering, and the Future 399
Sensors
Sensors, both invasive and noninvasive, are moving toward
multimodal data collection to reduce the “real estate” on and
in the head. For example, collecting tissue oxygen, ICP, and
temperature is now possible in the same catheter (Raumedic,
Münchberg, Germany). The recent announcement of a “lab-
on-a-tube” incorporated on the outside of a ventricular
Fig. 40.10  Display from Axon Systems
Eclipse Neurological Workstation.
Fig. 40.11  Display from the BedMaster
XA software.
drainage catheter is an exciting concept (Fig. 40.12).65,66 Pres-
sure, temperature, oxygen, and glucose microsensors to
measure these parameters in real time are rolled in a spiral and
attached to the catheter.
Many sensors may truly be noninvasive in the future;
this likely will increase the number of patients who are moni-
tored for ICP and even cerebrovascular reactivity.67-69 It is
400 Section VII—Computers, Engineering, and the Future
conceivable that a noninvasive sensor may combine EEG,
evoked potentials, near-infrared spectroscopy, and transcra-
nial Doppler (TCD) in an easy-to-apply headband. An inter-
esting helmet for continuous monitoring of TCD, EEG,
somatosensory evoked potentials, brainstem auditory evoked
potentials, and brain oxygen (via near-infrared spectroscopy
[NIRS]) was constructed by Dr. Gerhard Litscher in Graz,
Austria, in 1998.70 Although somewhat unwieldy for routine
use, the system demonstrated the feasibility of this concept
and the ability to set up and start collecting these parameters
in as little as 10 minutes.
Monitoring Modalities
New monitoring modalities or improved existing methods
will enter routine use. For example, NIRS was originally
Fig. 40.12  The Lab-on-a-Tube consisting of a spiral-wound strip of
sensors on a ventricular drainage catheter. (Printed with permission from the
Mayfield Clinic.)
Data
manager
Patient record and
remote review
External database
Standard protocol
Adapters
Patient-specific
Monitors database
Fig. 40.13  A unified information architecture for neurocritical care.
introduced as a relative measure of trends of brain oxygen,
and now systems are available that provide absolute readings
(see Chapter 33). Recent work shows that NIRS can be used
to track the kinetics of the tracer dye indocyanine green to
calculate an index of cerebral blood flow.71 Similarly, diffuse
correlation spectroscopy (DCS) uses near-infrared light to
assess changes in cerebral blood flow, but does not require a
tracer. Although it is a relatively new modality, blood flow
calculations using DCS have been validated both in animal
and human studies,72-74 which show it to be a promising non-
invasive measurement.
Unified Information Architecture
Neurocritical care monitoring, as with much of critical care
medicine, will benefit greatly from a unified information
architecture. The architecture would consist of a bedside envi-
ronment where a seamless data pathway exists from the
patient, to processing methods, to displays, to the patient
record, to workflow software and decision support methods,
to a data warehouse for later use. The architecture would
support medical devices that would plug in and be automati-
cally recognized, as well as software-based knowledge modules
that would “plug in” to provide new calculations, new displays,
practice guidelines, decision support methods, and other
bedside-usable information. The system would understand
the context of the patient and customize the monitoring to
specific problems.
Although somewhat futuristic, work is currently underway
to design components of such an architecture, and to develop
standards and address regulatory issues. Figure 40.13 illus-
trates the concept and is based on the Integrated Clinical
Environment (ICE).75,76 ICE is a proposed standard for infor-
mation flow at the bedside that includes medical device “plug-
and-play,” the creation of a patient-centric database, and the
use of the data by knowledge tools such as workflows. The
concept is being developed by a consortium of industry
Knowledge
manager
Patient
context
Bedside
Guideline
engine
P
Decision G G
support
Knowledge
plug-ins
Events
manager D D
Learning D L
manager
 Section VII—Computers, Engineering, and the Future 401
members, academic organizations, professional societies, and
regulatory agencies and is moving through the standards
development process.75
In ICE, data are collected from devices that use a standard
protocol to ensure consistency in how data are received,
labeled, and stored. Because there is a wide variety of medical
device communications protocols and data naming conven-
tions, adapters are proposed that transport the medical device
data via a standard protocol that provides information about
the device as well as the data from the device. In this manner,
the receiving software does not have to know about every new
device that is developed.
A key benefit of the information architecture is the ability
to uncouple the knowledge components from basic data col-
lection devices and the ability to distribute them as separate
“knowledge plug-ins.” At present, at least in the United States,
a medical device manufacturer must “re-clear” a device with
the Food and Drug Administration (FDA) if a new metric is
added to it that provides an additional use (e.g., a measure of
autoregulation or a seizure detection method). Device manu-
facturers often are not willing to spend the time or money to
go through this FDA process. This hurdle means that many
innovative techniques for data analysis and decision support
remain on the shelves of the inventor. Uncoupling these
knowledge components from the device manufacturer would
then shift the safety and quality requirements to the develop-
ers of the components, and a process to certify them would
need to be developed.
Annotated Database
A wider adoption of multimodal neuromonitoring will facili-
tate the creation of a comprehensive database of neurophysi-
ologic signals (and related data) that can be used for
retrospective review, research, and teaching. It is proposed
that a standardized method to annotate these multimodal data
sets be developed. Such a database could be used for validation
and conformance testing of “knowledge plug-ins,” such as
event detectors, decision support modules, and protocol strat-
egies. It would expedite their development and commercializa-
tion similar to the way the annotated ECG database (developed
at Massachusetts Institute of Technology more than 30 years
ago) ushered in modern day arrhythmia analysis.77
Conclusion
The concept of multimodality monitoring in neurocritical
care is not new; it has been discussed for more than 30 years.
However, even today there are only a few institutions that
provide something that approaches multimodality monitor-
ing in a comprehensive manner. The primary reason for this
is the lack of a medical device communications standard. It
is surprising that today it is possible to buy any number of
devices for a home computer such as cameras, printers, and
scanners, and they all connect to the computer “out of the
box.” Yet with medical devices this is not the case. Without a
database of physiologic signals collected in a consistent
manner, it is difficult to make progress in creating a truly
integrated knowledge environment.
Future evolution of NCCU informatics is likely to accelerate
for several reasons. First, the problems in the information
infrastructure encountered in the NCCU are not unique to
this field. The recent efforts to solve much broader problems
in hospital informatics will help in the NCCU. Second, indus-
try is starting to focus on automatic data collection, including
a host of smaller companies that will challenge the single-
vendor monopolies and their proprietary data pathways.
Third, a renewed effort to achieve a device communications
standard coupled to a bedside information environment has
made significant progress. Fourth the ASTM has adopted the
concept of an integrated clinical environment.35 Finally,
various professional societies have begun to document the
increase in patient safety and worker efficiencies that device
interoperability and intelligent software can yield. Together
these factors mean that working in a knowledge environment
at the bedside is within sight.
Disclaimer
The author (R.M.) is a founder of one of the companies (CNS
Technology) whose product is mentioned in this chapter.
References
1. Gupta AK. Monitoring the injured brain in the intensive care unit.
J Postgrad Med 2002;48:218–25.
2. Cohen MJ, Grossman AD, Morabito D. Identification of complex metabolic
states in critically injured patients using bioinformatic cluster analysis. Crit
Care 2010;14:R10.
3. Sahuquillo J. Does multimodality monitoring make a difference in
neurocritical care? Eur J Anaesthesiol 2008;42(Suppl):83–6.
4. Imhoff M, Fried R, Gather U, et al. Dimension reduction for physiological
variables using graphical modeling. AMIA Annu Symp Proc 2003:313–17.
5. Burykin A, Peck T, Krejci V, et al. Toward optimal display of physiologic
status in critical care. I. Recreating bedside displays from archived
physiologic data. J Crit Care 2010;26:105.e1–e9.
6. Morris G, Gardner R. Computer applications. In: Hall J, Schmidt G, Wood L,
editors. Principles of critical care. New York: McGraw-Hill; 1992. p. 500–14.
7. Imhoff M. Detecting relationships between physiological variables using
graphical modeling. In: Proc AMIA Symp 2002; 340–44, AMIA (Bethesda),
2002.
8. Jennings D, Amabile T, Ross L. Informal assessments: data-based versus
theory-based judgments. In: Kahnemann D, Slovic P, Tversky A, editors.
Judgments under uncertainty: heuristics and biases. Cambridge: Cambridge
University Press; 1982. p. 211–30.
9. Angus DC, Kelley MA, Schmitz RJ, et al. Caring for the critically ill patient.
Current and projected workforce requirements for care of the critically ill
and patients with pulmonary disease: can we meet the requirements of an
aging population? JAMA 2000;284:2762–70.
10. Congressional Budget Office. Research on the comparative effectiveness of
medical treatments: issues and options for an expanded federal role.
Washington; 2007.
11. Maas AI, Menon DK, Lingsma HF, et al. Re-orientation of clinical research
in traumatic brain injury: report of an international workshop on
comparative effectiveness research. J Neurotrauma 2011;29 [Epub ahead of
print].
12. Turn your alarms on! Anesthesia Patient Safety Foundation Newsletter
2004:43. Available at www.apsf.org.
13. Hlatky R, Robertson CS. Multimodality monitoring in severe head injury.
Curr Opin Anaesthesiol 2002;15:489–93.
14. Kett-White R, Hutchinson PJ, Czosnyka M, et al. Multi-modal monitoring of
acute brain injury. Adv Tech Stand Neurosurg 2002;27:87–134.
15. Tisdall MM, Smith M. Multimodal monitoring in traumatic brain injury:
current status and future directions. Br J Anaesth 2007;99:61–7.
16. Nangunoori R, Maloney-Wilensky E, Stiefel M, et al. Brain tissue oxygen-
based therapy and outcome after severe traumatic brain injury: a systematic
literature review. Neurocrit Care 2011;16 [Epub ahead of print].
17. Levine JM. Critical care management of subarachnoid hemorrhage. Curr
Neurol Neurosci Rep 2008;8:518–25.
18. Czosnyka M, Kirkpatrick PJ, Pickard JD. Multimodal monitoring and
assessment of cerebral haemodynamic reserve after severe head injury.
Cerebrovasc Brain Metab Rev 1996;8:273–95.
402 Section VII—Computers, Engineering, and the Future
19. Chan KH, Dearden NM, Miller JD, et al. Multimodality monitoring as a
guide to treatment of intracranial hypertension after severe brain injury.
Neurosurgery 1993;32:547–53.
20. Stiefel MF, Verdiramo AM, Schuster JM, et al. Multimodality monitoring
in the management of refractory intracranial hypertension: a case report.
J Trauma 2005;59:757–61.
21. Shiogai T, Takeuchi K. Multimodal neuromonitoring in impending brain
death. Minerva Anesthesiol 1994;60:583–8.
22. Czosnyka M, Whitehouse H, Smielewski P, et al. Computer supported
multimodal bed-side monitoring for neuro intensive care. Int J Clin Monit
Comput 1994;11:223–32.
23. De Georgia MA, Deogaonkar A. Multimodal monitoring in the neurological
intensive care unit. Neurologist 2005;11:45–54.
24. Kirkpatrick PJ, Czosnyka M, Pickard JD. Multimodal monitoring in
neurointensive care. J Neurol Neurosurg Psychiatry 1996;60:131–9.
25. Sorani MD, Hemphill 3rd JC, Morabito D, et al. New approaches to
physiological informatics in neurocritical care. Neurocrit Care 2007;7:
45–52.
26. Miller JD, Piper IR, Jones PA. Integrated multimodality monitoring in the
neurosurgical intensive care unit. Neurosurg Clin N Am 1994;5:661–70.
27. Mulvey JM, Dorsch NW, Mudaliar Y, et al. Multimodality monitoring in
severe traumatic brain injury: the role of brain tissue oxygenation
monitoring. Neurocrit Care 2004;1:391–402.
28. Vespa PM. Multimodality monitoring and telemonitoring in neurocritical
care: from microdialysis to robotic telepresence. Curr Opin Crit Care
2005;11:133–8.
29. Wartenberg KE, Mayer SA. Multimodal brain monitoring in the neurological
intensive care unit: where does continuous EEG fit in? J Clin Neurophysiol
2005;22:124–7.
30. Wartenberg KE, Schmidt JM, Mayer SA. Multimodality monitoring in
neurocritical care. Crit Care Clin 2007;23:507–38.
31. Chen HI, Stiefel MF, Oddo M, et al. Detection of cerebral compromise with
multimodality monitoring in patients with subarachnoid hemorrhage.
Neurosurgery 2011;69(1):53–63.
32. Hemphill 3rd JC, Barton CW, Morabito D, et al. Influence of data resolution
and interpolation method on assessment of secondary brain insults in
neurocritical care. Physiol Meas 2005;26:373–86.
33. Jordan KG. Continuous EEG and evoked potential monitoring in the
neuroscience intensive care unit. J Clin Neurophysiol 1993;10:445–75.
34. Claassen J, Mayer SA. Continuous electroencephalographic monitoring in
neurocritical care. Curr Neurol Neurosci Rep 2002;2:534–40.
35. ASTM subcommittee F29.21. ASTM Standard F2761–09 Medical devices and
medical systems: essential safety requirements for equipment comprising the
patient-centric integrated clinical environment (ICE). Part 1. General
requirements and conceptual model. ASTM International http://
www.astm.org/Standards/F2761.htm;2009.
36. He P, Wilson G, Russell C, et al. Removal of ocular artifacts from the EEG: a
comparison between time-domain regression method and adaptive filtering
method using simulated data. Med Biol Eng Comput 2007;45:495–503.
37. Devuyst S, Dutoit T, Stenuit P, et al. Removal of ECG artifacts from EEG
using a modified independent component analysis approach. Conf Proc
IEEE Eng Med Biol Soc 2008;2008:5204–7.
38. McCarthy J, Minsky M, Rochester N, et al. A proposal for the Dartmouth
Summer Research Project on Artificial Intelligence; 1965. Available at http://
www-formal.standard.edu/jmc/history/dartmouth.html.
39. Arbib MA. The handbook of brain theory and neural networks. 2nd ed.
Cambridge, Mass: MIT Press; 2003.
40. Haykin SS. Neural networks and learning machines. 3rd ed. New York:
Prentice Hall; 2009.
41. Imhoff M, Kuhls S. Alarm algorithms in critical care monitoring. Anesth
Analg 2006;102:1525–37.
42. Meredith C, Edworthy J. Are there too many alarms in the intensive care
unit? An overview of the problems. J Adv Nurs 1995;21:15–20.
43. Kalogeropoulos D, Carson ER, Collinson PO. Clinical-HINTS: integrated
intelligent ICU patient monitoring and information management system.
Stud Health Technol Inform 1997;43(Pt B):906–10.
44. Aboukhalil A, Nielsen L, Saeed M, et al. Reducing false alarm rates for
critical arrhythmias using the arterial blood pressure waveform. J Biomed
Inform 2008;41:442–51.
45. Blum JM, Kruger GH, Sanders KL, et al. Specificity improvement for
network distributed physiologic alarms based on a simple deterministic
reactive intelligent agent in the critical care environment. J Clin Monit
Comput 2009;23:21–30.
46. Apiletti D, Baralis E, Bruno G, et al. Real-time analysis of physiological data
to support medical applications. IEEE Trans Inf Technol Biomed
2009;13:313–21.
47. Zhang Y, Szolovits P. Patient-specific learning in real time for adaptive
monitoring in critical care. J Biomed Inform 2008;41:452–60.
48. Patel VL, Zhang J, Yoskowitz NA, Green R, Sayan OR. Translational
cognition for decision support in critical care environments: a review.
J Biomed Inform 2008;41:413–31.
49. Nelson DW, Bellander BM, Maccallum RM, et al. Cerebral microdialysis of
patients with severe traumatic brain injury exhibits highly individualistic
patterns as visualized by cluster analysis with self-organizing maps. Crit Care
Med 2004;32:2428–36.
50. Nelson DW, Thornquist B, MacCallum RM, et al. Analyses of cerebral
microdialysis in patients with traumatic brain injury: relations to
intracranial pressure, cerebral perfusion pressure and catheter placement.
BMC Med 2011;9:21.
A complete list of references for this chapter can be found online at
www.expertconsult.com.
 Section VII—Computers, Engineering, and the Future 402.e1
References
1. Gupta AK. Monitoring the injured brain in the intensive care unit.
J Postgrad Med 2002;48:218–25.
2. Cohen MJ, Grossman AD, Morabito D. Identification of complex metabolic
states in critically injured patients using bioinformatic cluster analysis. Crit
Care 2010;14:R10.
3. Sahuquillo J. Does multimodality monitoring make a difference in
neurocritical care? Eur J Anaesthesiol 2008;42(Suppl):83–6.
4. Imhoff M, Fried R, Gather U, et al. Dimension reduction for physiological
variables using graphical modeling. AMIA Annu Symp Proc 2003:313–17.
5. Burykin A, Peck T, Krejci V, et al. Toward optimal display of physiologic
status in critical care. I. Recreating bedside displays from archived
physiologic data. J Crit Care 2010;26:105.e1–e9.
6. Morris G, Gardner R. Computer applications. In: Hall J, Schmidt G, Wood L,
editors. Principles of critical care. New York: McGraw-Hill; 1992. p. 500–14.
7. Imhoff M. Detecting relationships between physiological variables using
graphical modeling. In: Proc AMIA Symp 2002; 340–44, AMIA (Bethesda),
2002.
8. Jennings D, Amabile T, Ross L. Informal assessments: data-based versus
theory-based judgments. In: Kahnemann D, Slovic P, Tversky A, editors.
Judgments under uncertainty: heuristics and biases. Cambridge: Cambridge
University Press; 1982. p. 211–30.
9. Angus DC, Kelley MA, Schmitz RJ, et al. Caring for the critically ill patient.
Current and projected workforce requirements for care of the critically ill
and patients with pulmonary disease: can we meet the requirements of an
aging population? JAMA 2000;284:2762–70.
10. Congressional Budget Office. Research on the comparative effectiveness of
medical treatments: issues and options for an expanded federal role.
Washington; 2007.
11. Maas AI, Menon DK, Lingsma HF, et al. Re-orientation of clinical research
in traumatic brain injury: report of an international workshop on
comparative effectiveness research. J Neurotrauma 2011;29 [Epub ahead of
print].
12. Turn your alarms on! Anesthesia Patient Safety Foundation Newsletter
2004:43. Available at www.apsf.org.
13. Hlatky R, Robertson CS. Multimodality monitoring in severe head injury.
Curr Opin Anaesthesiol 2002;15:489–93.
14. Kett-White R, Hutchinson PJ, Czosnyka M, et al. Multi-modal monitoring of
acute brain injury. Adv Tech Stand Neurosurg 2002;27:87–134.
15. Tisdall MM, Smith M. Multimodal monitoring in traumatic brain injury:
current status and future directions. Br J Anaesth 2007;99:61–7.
16. Nangunoori R, Maloney-Wilensky E, Stiefel M, et al. Brain tissue oxygen-
based therapy and outcome after severe traumatic brain injury: a systematic
literature review. Neurocrit Care 2011;16 [Epub ahead of print].
17. Levine JM. Critical care management of subarachnoid hemorrhage. Curr
Neurol Neurosci Rep 2008;8:518–25.
18. Czosnyka M, Kirkpatrick PJ, Pickard JD. Multimodal monitoring and
assessment of cerebral haemodynamic reserve after severe head injury.
Cerebrovasc Brain Metab Rev 1996;8:273–95.
19. Chan KH, Dearden NM, Miller JD, et al. Multimodality monitoring as a
guide to treatment of intracranial hypertension after severe brain injury.
Neurosurgery 1993;32:547–53.
20. Stiefel MF, Verdiramo AM, Schuster JM, et al. Multimodality monitoring in
the management of refractory intracranial hypertension: a case report.
J Trauma 2005;59:757–61.
21. Shiogai T, Takeuchi K. Multimodal neuromonitoring in impending brain
death. Minerva Anesthesiol 1994;60:583–8.
22. Czosnyka M, Whitehouse H, Smielewski P, et al. Computer supported
multimodal bed-side monitoring for neuro intensive care. Int J Clin Monit
Comput 1994;11:223–32.
23. De Georgia MA, Deogaonkar A. Multimodal monitoring in the neurological
intensive care unit. Neurologist 2005;11:45–54.
24. Kirkpatrick PJ, Czosnyka M, Pickard JD. Multimodal monitoring in
neurointensive care. J Neurol Neurosurg Psychiatry 1996;60:131–9.
25. Sorani MD, Hemphill 3rd JC, Morabito D, et al. New approaches to
physiological informatics in neurocritical care. Neurocrit Care 2007;7:
45–52.
26. Miller JD, Piper IR, Jones PA. Integrated multimodality monitoring in the
neurosurgical intensive care unit. Neurosurg Clin N Am 1994;5:661–70.
27. Mulvey JM, Dorsch NW, Mudaliar Y, et al. Multimodality monitoring in
severe traumatic brain injury: the role of brain tissue oxygenation
monitoring. Neurocrit Care 2004;1:391–402.
28. Vespa PM. Multimodality monitoring and telemonitoring in neurocritical
care: from microdialysis to robotic telepresence. Curr Opin Crit Care
2005;11:133–8.
29. Wartenberg KE, Mayer SA. Multimodal brain monitoring in the neurological
intensive care unit: where does continuous EEG fit in? J Clin Neurophysiol
2005;22:124–7.
30. Wartenberg KE, Schmidt JM, Mayer SA. Multimodality monitoring in
neurocritical care. Crit Care Clin 2007;23:507–38.
31. Chen HI, Stiefel MF, Oddo M, et al. Detection of cerebral compromise with
multimodality monitoring in patients with subarachnoid hemorrhage.
Neurosurgery 2011;69(1):53–63.
32. Hemphill 3rd JC, Barton CW, Morabito D, et al. Influence of data resolution
and interpolation method on assessment of secondary brain insults in
neurocritical care. Physiol Meas 2005;26:373–86.
33. Jordan KG. Continuous EEG and evoked potential monitoring in the
neuroscience intensive care unit. J Clin Neurophysiol 1993;10:445–75.
34. Claassen J, Mayer SA. Continuous electroencephalographic monitoring in
neurocritical care. Curr Neurol Neurosci Rep 2002;2:534–40.
35. ASTM subcommittee F29.21. ASTM Standard F2761–09 Medical devices and
medical systems: essential safety requirements for equipment comprising the
patient-centric integrated clinical environment (ICE). Part 1. General
requirements and conceptual model. ASTM International http://
www.astm.org/Standards/F2761.htm;2009.
36. He P, Wilson G, Russell C, et al. Removal of ocular artifacts from the EEG: a
comparison between time-domain regression method and adaptive filtering
method using simulated data. Med Biol Eng Comput 2007;45:495–503.
37. Devuyst S, Dutoit T, Stenuit P, et al. Removal of ECG artifacts from EEG
using a modified independent component analysis approach. Conf Proc
IEEE Eng Med Biol Soc 2008;2008:5204–7.
38. McCarthy J, Minsky M, Rochester N, et al. A proposal for the Dartmouth
Summer Research Project on Artificial Intelligence; 1965. Available at http://
www-formal.standard.edu/jmc/history/dartmouth.html.
39. Arbib MA. The handbook of brain theory and neural networks. 2nd ed.
Cambridge, Mass: MIT Press; 2003.
40. Haykin SS. Neural networks and learning machines. 3rd ed. New York:
Prentice Hall; 2009.
41. Imhoff M, Kuhls S. Alarm algorithms in critical care monitoring. Anesth
Analg 2006;102:1525–37.
42. Meredith C, Edworthy J. Are there too many alarms in the intensive care
unit? An overview of the problems. J Adv Nurs 1995;21:15–20.
43. Kalogeropoulos D, Carson ER, Collinson PO. Clinical-HINTS: integrated
intelligent ICU patient monitoring and information management system.
Stud Health Technol Inform 1997;43(Pt B):906–10.
44. Aboukhalil A, Nielsen L, Saeed M, et al. Reducing false alarm rates for
critical arrhythmias using the arterial blood pressure waveform. J Biomed
Inform 2008;41:442–51.
45. Blum JM, Kruger GH, Sanders KL, et al. Specificity improvement for
network distributed physiologic alarms based on a simple deterministic
reactive intelligent agent in the critical care environment. J Clin Monit
Comput 2009;23:21–30.
46. Apiletti D, Baralis E, Bruno G, et al. Real-time analysis of physiological data
to support medical applications. IEEE Trans Inf Technol Biomed
2009;13:313–21.
47. Zhang Y, Szolovits P. Patient-specific learning in real time for adaptive
monitoring in critical care. J Biomed Inform 2008;41:452–60.
48. Patel VL, Zhang J, Yoskowitz NA, Green R, Sayan OR. Translational
cognition for decision support in critical care environments: a review.
J Biomed Inform 2008;41:413–31.
49. Nelson DW, Bellander BM, Maccallum RM, et al. Cerebral microdialysis of
patients with severe traumatic brain injury exhibits highly individualistic
patterns as visualized by cluster analysis with self-organizing maps. Crit Care
Med 2004;32:2428–36.
50. Nelson DW, Thornquist B, MacCallum RM, et al. Analyses of cerebral
microdialysis in patients with traumatic brain injury: relations to
intracranial pressure, cerebral perfusion pressure and catheter placement.
BMC Med 2011;9:21.
51. Minhas P, Perl TM, Carroll KC, et al. Risk factors for positive admission
surveillance cultures for methicillin-resistant Staphylococcus aureus and
vancomycin-resistant enterococci in a neurocritical care unit. Crit Care Med
2011;39(10):2322–9.
52. Swiercz M, Mariak Z, Krejza J, et al. Intracranial pressure processing with
artificial neural networks: prediction of ICP trends. Acta Neurochir (Wien)
2000;142:401–6.
53. Nikiforidis GC, Sakellaropoulos GC. Expert system support using Bayesian
belief networks in the prognosis of head-injured patients of the ICU. Med
Inform (Lond) 1998;23:1–18.
54. Vath A, Meixensberger J, Dings J, et al. Prognostic significance of advanced
neuromonitoring after traumatic brain injury using neural networks.
Zentralbl Neurochir 2000;61:2–6.
402.e2 Section VII—Computers, Engineering, and the Future
55. Rampil IJ. A primer for EEG signal processing in anesthesia. Anesthesiology
1998;89:980–1002.
56. Pfurtscheller G, Druschky K, Kamp HD, et al. [Multimodal evoked potentials
and heart rate variability in comatose patients. 1. Method of measuring and
normal values]. EEG EMG Z Elektroenzephalogr Elektromyogr Verwandte
Geb 1987;18:108–14.
57. Pfurtscheller G, Schwarz G, Schroettner O, et al. Continuous and
simultaneous monitoring of EEG spectra and brainstem auditory and
somatosensory evoked potentials in the intensive care unit and the operating
room. J Clin Neurophysiol 1987;4:389–96.
58. Litscher G, Schwarz G. Noninvasive bioelectrical neuromonitoring in
anaesthesia and critical care. Eur J Anaesthesiol 2001;18:785–8.
59. Vespa PM, Nuwer MR, Juhasz C, et al. Early detection of vasospasm after
acute subarachnoid hemorrhage using continuous EEG ICU monitoring.
Electroencephalogr Clin Neurophysiol 1997;103:607–15.
60. Vespa PM, Boscardin WJ, Hovda DA, et al. Early and persistent impaired
percent alpha variability on continuous electroencephalography monitoring
as predictive of poor outcome after traumatic brain injury. J Neurosurg
2002;97:84–92.
61. Hebb MO, McArthur DL, Alger J, et al. Impaired percent alpha variability on
continuous electroencephalography is associated with thalamic injury and
predicts poor long-term outcome after human traumatic brain injury.
J Neurotrauma 2007;24:579–90.
62. Smielewski P, Czosnyka M, Zabolotny W, et al. A computing system for the
clinical and experimental investigation of cerebrovascular reactivity. Int J
Clin Monit Comput 1997;14:185–98.
63. Hemphill C. Personal communication Oct. 5, 2009.
64. Kohli-Seth R, Oropello JM. The future of bedside monitoring. Crit Care Clin
2000;16:557–78, vii-viii.
65. Li C, Wu P, Narayan RK, et al. A novel lab-on-a-tube for multimodal monitoring
of patients with traumatic brain injury. Lab Chip 2009;9(14):1988–90.
66. Li C, Wu P-M, Jung W, et al. A novel lab-on-a-tube for multimodal
monitoring of patients with traumatic brain injury. Proceedings from the
IEEE Transducers 2009 symposium. p. 1449-52, Denver; 2009.
67. Li C, Shutter LA, Wu PM, et al. Potential of a simple lab-on-a-tube for
point-of-care measurements of multiple analytes. Lab Chip
2010;10(11):1476–9.
68. Chacón M, Pardo C, Puppo C, et al. Non-invasive intracranial pressure
estimation using support vector machine. Conf Proc IEEE Eng Med Biol Soc
2010;2010:996–9.
69. Rajajee V, Vanaman M, Fletcher JJ, et al. Optic nerve ultrasound for the
detection of raised intracranial pressure. Neurocrit Care 2011;Jul 19 [Epub
ahead of print].
70. Kasprowicz M, Schmidt E, Kim DJ, et al. Evaluation of the cerebrovascular
pressure reactivity index using non-invasive Finapres arterial blood pressure.
Physiol Meas 2010;31(9):1217–28.
71. Litscher G. A multifunctional helmet for noninvasive neuromonitoring.
J Neurosurg Anesthesiol 1998;10:116–19.
72. Bein B, Meybohm P, Cavus E, et al. A comparison of transcranial Doppler
with near infrared spectroscopy and indocyanine green during hemorrhagic
shock: a prospective experimental study. Crit Care 2006;10:R18.
73. Zhou C, Eucker SA, Durduran T, et al. Diffuse optical monitoring of
hemodynamic changes in piglet brain with closed head injury. J Biomed Opt
2009;14:034015.
74. Buckley EM, Cook NM, Durduran T, et al. Cerebral hemodynamics in
preterm infants during positional intervention measured with diffuse
correlation spectroscopy and transcranial Doppler ultrasound. Opt Express
2009;17:12571–81.
75. Medical Device Plug and Play. <http://mdpnp.org/ICE.html>.
76. Essential safety requirements for equipment comprising the patient-centric
integrated clinical environment (ICE). Part I. General requirements and
conceptual model. In: F2761: ASTM; 2009.
77. Moody GB, Mark RG. The impact of the MIT-BIH arrhythmia database.
IEEE Eng Med Biol Mag 2001;20:45–50.

Simulation in the ICU
W. Andrew Kofke
Introduction
Simulation: a situation or environment created to allow
persons to experience a representation of a real event for
the purpose of practice, learning, evaluation, testing, or to
gain understanding of systems or human actions.
—definition taught at the Center for Medical
Simulation, Cambridge, Mass.
In the 1960s, Red Cross first aid classes taught a new cardio-
pulmonary resuscitation (CPR) method, having discarded the
chest pressure-arm lift method, the primary resuscitation
method of the 1950s and 1960s and perhaps before. So how
was CPR practiced? On Resusci-Anne—a relatively new
Laerdal product that was used everywhere to teach mouth-to-
mouth and chest compressions following the contemporary
recommendations of the American Heart Association. By
today’s standards this is a relatively primitive simulator, one
that is still in use.
However, perhaps the most sophisticated simulator of all
had already been demonstrated in the late 1960s by Steven
Abrahamson, the Dean of Education at University of South-
ern California.1 Abahamson et al. developed a high-fidelity
simulator—SIM-ONE—that was more sophisticated then
than some simulators that are currently available; it even fas-
ciculated after succinylcholine injection. However, SIM-ONE
required a room full of computers, each with an attendant
technician in the manner of a NASA space launch. Dr. Abra-
hamson went on sabbatical, his simulator was dismantled, so
the story goes, and it took advances in computer technology
and the vision of subsequent simulation pioneers before
resurrection of this technology in the 1990s. Simulation is
now an exponentially growing field, driven by the patient-
safety movement, changes in the educational opportunities
available to trainees, and evolving computer capabilities.
Medical-simulation centers that numbered in the 10s in the
early 1990s now number in the 1000s worldwide. Further-
more there is accumulating evidence that simulation can be
used in training of physicians and health care providers to
help reduce medical errors, improve technical skills for some
invasive procedures, and enhance communication skills.2,3
Simulation in terms of a taxonomy can be characterized as
follows:
 Task training. This entails use of simulation devices meant
to teach specific psychomotor tasks with complexity that
© Copyright 2013 Elsevier Inc. All rights reserved.
Chapter
41  
VII
can range from simple insertion of an intravenous cannula
into a simulated arm to a computerized virtual reality
world that allows an operator to learn highly complex
procedures such as transesophageal echocardiography,
laparoscopy, or endoscopy procedures.4 One example of a
new virtual reality simulator is that developed by Banerjee
et al.5,6 to teach ventriculostomy placement (Fig. 41.1).
 PC-based simulation. This entails use of PC-based pro-
grams that are installed on laptop computers. They can be
set up to mimic real life scenarios such that the user has
to run through all of the cognitive decision algorithms that
are required in clinical practice. Moreover, in accordance
with the apparent preferences of the current younger
generation, the PC-based simulation also can be set up as
a game.
 Standardized patients. What can be better than working on
a real human? This notion has resulted in the development
of standardized patient programs in many schools and
a standardized patient society (www.aspeducators.org).
Standardized patients are real people, individuals who are
trained to display the appropriate behaviors of patients
with given diseases in given situations or individuals with
the relevant physical findings. Both types of standardized
patients can be used to teach novices how to deal with a
variety of problems.
 High-fidelity simulation. This uses simulators that bring
together all of the physiology and pharmacologic responses
of a human in a manikin, which breathes, may produce or
consume a variety of different gases, and talks in an inter-
active manner with the students. After appropriate suspen-
sion of disbelief, this manikin becomes for all intents and
purposes a real human being with a real problem that
students have to diagnose and treat appropriately.7
All of these elements of simulation can be further integrated
to create scenarios that encompass any or all of these methods.
Students can perform procedures, make decisions, and inter-
act with a patient (manikin or standardized patient who may
be attached to a task trainer). At an even higher level, students
may interact with other health care providers on a team. This
had led increasingly to the use of such simulation tools to train
health care providers in the principles of crew/crisis resource
management (CRM) (teamwork), another lesson from the
aviation industry. And finally simulation can be used as an
essential tool for microsystem or macrosystem simulations.
403
404 Section VII—Computers, Engineering, and the Future
Fig. 41.1  ImmersiveTouch. Ventriculostomy simulation is depicted,
showing the observer’s view and the operator’s view. (From Issenberg SB,
Chung HS, Devine LA. Patient safety training simulations based on competency
criteria of the accreditation council for graduate medical education. Mt Sinai J Med
2011;78(6):842–53, with permission.)
Such simulations encompass handling multiple patients with
management of medical, interpersonal, and system issues.
In the context of critical care, CRM is an important new
vista that can be used to improve team performance.
Crisis Resource Management
In 2000, the American Heart Association established
the National Registry of Cardiopulmonary Resuscitation
(NRCPR) to assist participating hospitals with systematic data
collection on resuscitative efforts. In so doing, the AHA pro-
vided a prospective, multisite, observational mechanism to
determine the epidemiology and outcomes from in-hospital
resuscitation. Initial results from the registry were reported
by Peberdy et al.8 in 2003 from 14,720 cardiac arrests in adults
at 207 hospitals, of which 86% had an organized emergency
response team continuously available. The researchers
reported that 44% of adult in-hospital cardiac arrest victims
had a return of spontaneous circulation (ROSC), but only
17% survived to hospital discharge. Neurologic outcome was
reported as good in 86% of discharged survivors. This 17%
discharge survival rate is consistent with reports of others.9
Various models have been proposed to prevent cardiac
arrests and improve survival following cardiac arrest, such as
using emergency medicine residents as responders10 and creat-
ing emergency response teams,11 but these recommendations
have yet to be translated into improved survival after inpatient
cardiac arrest. Several reasons can be attributed to the rela-
tively poor results after inpatient cardiac arrest. These cer-
tainly include patients’ premorbid severity of illness such that
any practice or process changes in resuscitation would not be
expected to have a meaningful impact. Nonetheless, several
studies emphasize the important role that various practices
have on patient outcomes after resuscitation. For example,
time to CPR was identified as an important factor in pre­
dicting outcome for both inpatient9 and out-of-hospital
cardiac arrest,12 and organized early medical emergency
response team involvement appears to positively affect
outcomes.13-15 Trauma surgeons have been leaders in this
aspect of resuscitation, and have demonstrated the advantage
of using a team model to resuscitate victims of severe trauma.16
Providing further support to this concept, there is now com-
pelling data on the advantages of CRM team training on error
rates and practice process variables and, moreover, the value
of attending intensivist contribution to critical care outcome.
All of these observations and those of others clearly indicate
process change as an area for improvement that could improve
survival after in-hospital resuscitation. The notable disparity
in ROSC and discharge survival rates suggest that improve-
ments in processes of practice can reasonably be targeted as
aspects of patient care that could increase the rate of meaning-
ful survival after cardiac arrest or near arrest.
The aviation and nuclear energy industries, as well as the
military, all described as high reliability organizations, have a
rich and successful history of using crisis simulation to train
personnel to deal with high-risk situations in complex
environments.17-19 More recently, the pioneering work of anes-
thesiologists, critical care specialists, and others has extended
the knowledge developed in these industries to acute care
medicine and the operating room.20-23 These providers have
collaborated with industry to develop realistic computerized
manikin systems, which simulate medical crisis situations
typically encountered in the operating room and intensive
care unit (ICU) and that may be directed at physicians, nurses,
advanced practitioners, residents, or students.24-27 The systems
are being further developed and adapted to medical crisis situ-
ations outside of the operating room. For example, these units
can provide both inexperienced and experienced personnel in
a number of fields with exposure to common and uncommon
medical crises such as myocardial ischemia, hypothermia,
acute respiratory distress syndrome, septic shock, anaphylaxis,
complex arrhythmias, Advanced Cardiac/Trauma Life Support
protocols, pneumothorax, difficult airway situations, and
many others. Moreover, the realistic setting of the simulation
“theater” can be used to create realistic situations that teach
important lessons about optimal team dynamics and can help
improve leadership, teamwork, and self-confidence skills to
manage medical emergencies. It should be remembered that
effective instructor training is needed for successful imple-
mentation of simulation based CRM learning.28
What Is Crisis Resource
Management?
Risser et al.29 in their analysis of medical teamwork failures
describe a core team as “a set of 3 to 10 clinically skilled care-
givers who work together during a shift and have been trained
to use specific teamwork behaviors to tightly coordinate and
manage their clinical actions.” Risser et al. adapted knowledge
from aviation CRM to the emergency department acute care
setting and derived five team dimensions:
1. Maintenance of team structure and climate. This refers ini-
tially to:
 The establishment of the team leader and organization
of the team
 Cultivation of team climate
 Constructive resolution of conflicts
Success in the subsequent four dimensions depends
on this one being successfully implemented. Examples
of common team failures in this area include not
 Section VII—Computers, Engineering, and the Future 405
assigning roles properly and not holding team members
accountable.
Dimensions 2, 3, and 4, described next, are the opera-
tional core of a high-quality team and are functionally
intertwined, essentially occurring simultaneously. The
caregivers may have to balance competing demands and
must maintain a high level of situational awareness.30
2. Application of problem-solving strategies. This describes:

Conducting situational planning

Application of decision-making methods
 Engaging in error correction actions
This dimension relies on adequate communication. This
means that all team members are a part of decision-
making, that the leader asks for and is given relevant infor-
mation, and that relevant protocols are implemented.
This dimension also has caregivers being aware of the
three basic types of errors: slips, lapses, and mistakes,31 and
being assertive enough to identify and correct them.
Examples of common team failures in this area include not
identifying relevant established protocols, not engaging
other team members in decision making, not alerting the
team to potential bias and error, and not asserting the need
for corrective actions.
3. Communication with the team. This refers to:
 The use of local standards for effective communication
 Provision of supporting information
 Requesting supporting information
The emphasis of this dimension is the assurance
within the team of a common understanding of patient
and operational issues that affect the performance of
the team and patient outcome. There is timely informa-
tion transfer and maintenance of team situational aware-
ness. Examples of common team failures in this area
include not using a check-back process to verify com-
munication, not offering information that supports a
decision, not seeking needed information, and not com-
municating plans to the team members.
4. Execution of plans and management of workload. This
refers to:
 Plan implementation
 Primary and secondary triage
 Task prioritization
 Management of team workload and resources
 Cross-monitoring team member actions
 Maintaining situational awareness
The emphasis of this dimension is to encourage willing-
ness of team members to ask for assistance and to assist
others to eliminate work overload and poor integration of
everyone’s tasks. Examples of common team failures in
this area include not executing the protocol or plan, not
integrating team member assessments of patient needs,
poor task prioritization, and inadequate cross-monitoring
of team members and the team’s plan execution.
5. Improvement of team skill. This refers to engaging in infor-
mal and formal team improvement strategies. This entails
informal discussion at the time of the event and later
formal review and discussion.
Baker et al.,32 in a recent review of this topic, outline knowl-
edge, skills, and attitudes that members of a high-quality team
should possess that facilitate coordinated, adaptive perfor-
mance and provide support of one’s teammates, objectives,
and mission. Baker et al., in general agreement with Risser
et al., indicates that each team member is able to do the fol-
lowing to identify when errors occur and how to recover and
correct for these errors:
 Anticipate the needs of others
 Adjust to each other’s actions and to the changing
environment

Have a shared understanding of how a procedure should
happen
Hunt et al.,33 in a review of teamwork training, indentifies
several factors listed in Table 41.1 as elements of high per-
forming teams.
Simulation as a Strategy for CRM
Training in Critical Care
Recent advances in simulation technology suggest that simu-
lation can be used as an important tool to implement a strat-
egy to improve critical care outcomes by using it as an essential
tool to teach CRM.
Building on the seminal work of Abrahamson34 in the
1960s to develop the first medical simulator; high-fidelity
patient simulation has become widely available with an expo-
nential jump in progress and acceptance over the past 15
years. The number of simulation centers worldwide num-
bered about 10 in 1994, 200 in 2001, and 452 in 2005.35 Pres-
ently, 126 simulation centers are listed in the directory of the
Society for Simulation in Healthcare.36 Ongoing technical
improvements have been evident with a substantial amount
of study to ascertain efficacy of human acute-care simulation,
as was first suggested by Abrahamson.34 Gaba et al. and others
have described the use of a simulator suite to recreate the
operating room for research and clinical training.37-40 Investi-
gators have subsequently demonstrated that simulators can
help reduce morbidity due to critical incidents,20,37,41-45 to
provide CRM training,37,46 and to more effectively provide
resident trainees with a safe and effective milieu to learn
appropriate procedures in acute care situations.47,48 Recent
research has demonstrated that simulation can be applied to
effectively measure some quality measures in cardiac arrest
response teams49 and that it can be used in surgical proce-
dures to increase efficiency and safety.50 Simulation is now an
accepted tool in the improvement of patient safety in many
disparate situations.
From a psychological perspective, Rose et al.51 demon-
strated the feasibility of the notion that prior training of a
nonmedical task in a simulated environment can be expected
to result in transfer of skills to real life situations. For medical
procedural simulations, several investigators have demon-
strated that prior procedure simulation results in better sub-
sequent performance in a simulated environment.52-54 Tuggy
et al.55 reported improved sigmoidoscopy performance on
human volunteers with prior simulation. Grantcharov et al.56
and Seymour et al.50 using a prospective randomized design
demonstrated the capability of prior simulator training to
significantly decrease errors and improve efficiency in the
operating room. This presumably translates into better patient
surgical outcome, but that element was not studied. Ahlberg
et al.57 showed that previous training on a simulator resulted
in improved colonoscopy performance. Conversely, in another
406 Section VII—Computers, Engineering, and the Future

Table 41.1  Characteristics of High Performing Teams

Situation Awareness (SA)
Leadership
Followership
Closed loop communication
Critical language and
standardized practices
Assertive communication
Adaptive behaviors
Workload management
Debriefing
Team performance is improved when team members continually assess their environment and update
each other in a process call “shared cognition” so that they are making decisions based on current
information and can have a “shared mental model” of the current state of affairs and an updated
plan of action with contingencies. SA allows team to maintain “big picture” view of situation.
Effective military and aviation teams have higher SA than low-performing teams.
An effective team leader can both command the team and values input from team members.
Flattening the hierarchy improves safety as information can flow both directions, while leaders who
maintain an authoritarian type of leadership “reinforce large authority gradients, creating
unnecessary risk.” A leader should try not to perform procedures unless the procedure is essential
and no one else is capable of doing it. Stepping back and keeping a “bird’s eye view” allows the
leader to take in and process more information and contributes to situational awareness.
The nonleader members of the team are called followers. Good “followership” is just as important
for good team functioning as good leadership. Followers need to know their individual role on the
team but also contribute to overall team functionality. They must contribute to situational
awareness by verbalizing observations about changes in the environment, ideas about diagnosis, to
decrease the leaders’ workload if necessary and finally to help the leader avoid mistakes; for
example, “the team leader might focus on an incorrect diagnosis and apply the wrong rule
(treatment) owing to a fixation error, or be incapacitated, hence everyone in the team should
always be alert.” Finally, followers must not assume that the leader knows everything and should
feel obligated to share observations that might impact outcome.
Closed looped communication is used to ensure that the message that was sent is heeded and
understood, and involves: “(1) the sender initiating a message, (2) the receiver receiving the
message, interpreting it, and acknowledging its receipt, and (3) the sender following up to ensure
the intended message was received.”
Critical language refers to the use of a catch phrase that means something to every member of an
organization and requires specific action (standardized practices). United Airlines developed the
CUS program for “I’m concerned, I’m uncomfortable, this is unsafe, or I’m scared,” which is
adopted within the culture as meaning, “We have a serious problem; stop and listen to me.”
Another example of a standardized approach to improve the effectiveness of communication is
SBAR (situation, background, assessment, recommendation). This is a tool that gives an outline of
how “awareness and education regarding the fact that nurses, physicians, and other clinicians are
taught to communicate in very different styles.”
Safe patient care may depend on the ability for a team member to speak up and get the attention
of other team members when they believe something might be going wrong. This is more likely to
happen if a team member believes speaking up will not be held against him or her. A recurrent
phrase is that people can still show deference to expertise, but speak up in a “nonthreatening and
respectful manner.” The idea is that all team members may have valuable input, “regardless of
rank.” The “hint and hope” model has been described as a common and dangerous way of trying
to indirectly communicate with other team members.
Teams whose members are flexible and perform as needed to optimize team functioning and
demonstrate “adaptive behaviors” are those that can truly benefit from the synergy of an effective
team. Examples of adaptive behavior that optimize team functionality include the following: “(1)
team members ask for help when overloaded, (2) team members monitor each other’s performance
to notice any performance decreases (mutual performance monitoring), or (3) team members take
an active role in assisting other team members who are in need of help (backup behavior). An
essential component to the above actions happening is trust among team members.”
Workload management depends on team members demonstrating adaptive behaviors. This principle
requires: (1) “proper allocation of tasks to individuals, (2) avoidance of work overloads in self and
in others, (3) prioritization of tasks during periods of high workload, and (4) preventing
nonessential factors from distracting attention from adherence to protocols, particularly those
relating to critical tasks.”
Debriefing is the process of reviewing a simulation or real event after it is complete to optimize any
lessons that can be learned. When using simulation as a teaching tool, simulation with no
debriefing and feedback does not result in effective learning. In terms of real events, teams that
debrief themselves afterwards have been shown to be higher performing.

From Hunt EA, Shikofsk, NA, Stavroudis TA, et al. Simulation: translation to improved team performance. Anesthesiol Clin 2007;25(2):301-19.

study of technical skill on patients, Gerson58 compared resi-
dents trained in endoscopy on patients with those trained on
a simulator and concluded that traditional methods were
superior, although they did not account for possible morbidity
in the patients who were subjects of the traditional training.
Notably lacking are studies that demonstrate that such prior
procedural training has a positive impact on measures of per-
formance and quality care.
Gaba and others have pioneered the application of simula-
tion training to the medical community, following many of
 Section VII—Computers, Engineering, and the Future 407
the tenants of aviation training. His group initially used simu-
lation technology to evaluate human performance during
anesthesia crisis situations; this resulted in the development
of Anesthesia Crisis Resource Management.59,60 The principles
from the early work are being adapted to critical care.
CRM training applied to medicine is similar in aviation
training, including the three essential elements of CRM train-
ing: knowledge, practice, and recurrence.61 Simulated scenar-
ios are created and applied to evaluate team performance, in
addition to preventing small management errors that can lead
to lethal results, such as in cardiopulmonary arrest or critical
care transport. Debriefing is then used as an essential element
of CRM training. Simulation to teach CRM has been shown
to be ineffective if there is no debriefing. For example, Savodelli
et al.62 demonstrated the importance of debriefing in a study
showing that those trainees who received audio or visual feed-
back following a simulated scenario performed significantly
better on subsequent scenarios than their counterparts who
had not received debriefing.
Neurocritical care units (NCCUs) are comprised of dispa-
rately composed dynamic teams that must act correctly and
fast when unexpected critical events arise. NCCU patients
typically require complex physiologic monitoring with new
technologies with which all members may not be familiar.
Simulation is helpful in such circumstances to replicate the
NCCU environment in advance of the real crisis. Complex
scenarios that require successful team interactions can be
simulated, such as cardiac arrest, cerebral herniation, exsan-
guination, arrhythmias, transport, and team/family behav-
ioral problems. Specific human patient simulators, such as the
Human Patient Simulator (HPS) produced by Medical Educa-
tion Technologies, have been used to enhance neurocritical
care education.63
Procedures performed in the NCCU, such as intubation and
central venous pressure (CVP) catheter placement, also can be
included in such simulations, including hand washing and
sterile procedures. For example, an improper sterile technique
could be used by an actor in a simulation with the teaching
goal to reinforce assertive tactful communication by other
team members.
Leadership failure has been quoted as a common error in
simulated ICU crisis situations.64 The use of simulation and
associated debriefing sessions to create the need for leadership
has been favorably received as a valuable teaching tool at
Stanford.64
Use of standardized patients (SPs) is another means to
augment team communication skills. For example, SPs can be
used to create problems in team dynamic during a crisis or to
simulate problems in dealing with families, such as breaking
bad news, doing difficult consents, or simulating discussions
regarding end-of-life care.
Equipment problems can be an obstacle to successful team
function. Simulation has a dual role here. The first is to test
equipment before purchase for any issues in its use in the local
environment. The second is to ensure adequate familiarity of
team members with the equipment that is available to them.
Hunt et al.65 have demonstrated the utility of team training in
a simulated environment to ensure that team members make
correct crucial decisions regarding pacemaker use.
Reports indicate that mishaps with transport related to un-
familiarity with equipment and with potential obstacles that
may arise are frequent and important problems.66 Simulation
scenarios can be conducted to practice difficult transports. For
example, a sick intubated patient with multiple brain and car-
diovascular monitors could be taken to radiology in a simula-
tion scenario. Ensuring an adequate oxygen supply, having
airway equipment on hand, and accidental infusion discon-
tinuation can be simulated. Application of teamwork princi-
ples to such situations appears to decrease the incidence of
transport mishaps.67,68 Similarly simulation-based training can
be used to enhance “hand-off ” skills and communication.69
Does CRM Work in Medicine?
Many institutions presently are grappling with whether and
how to invest in simulation and CRM training, but without
scientific evidence it is difficult to determine who should
participate, the level of training, and the impact on patient
care. While simulation-based learning has greatly enhanced
the technical skill of physicians and other health care profes-
sionals in performing selected interventions, the benefits of
simulation-based learning coupled with CRM have not been
adequately investigated in “real life” patient care situations.
Furthermore, the perceived advantages of using human
simulators in clinical laboratories as a teaching method have
not been translated into measurable benefits for patients
and institutions. The benefits of simulation team training
also may require repeated simulation exposure rather than a
single exposure.70 Cost-constrained hospitals are therefore
reluctant to commit the necessary manpower and equipment
resources to sustain these programs, although programs can
be shaped to meet specific institutional needs and budget
limitations.
Nonetheless, from this technology and background has
sprung a growing interest among health care professionals
across disciplines and over the past decade to develop and
train CRM principles. It is clear that simulation and medical
education are now at a historic patient-safety interface. This
has been largely driven by the Institute of Medicine report71
“To Err is Human.” This report recommended development
of patient-safety programs that “establish interdisciplinary
team training programs, such as simulation, that incorporate
proven methods for team management.” Their conclusions
were derived from studies such as those of Risser et al.,21 which
retrospectively demonstrated that lack of team behaviors
likely contributed to adverse events in the emergency depart-
ment. Moreover, Moorthy et al.72 demonstrated, in the absence
of CRM training, many teamwork deficiencies in a simulated
surgery environment. More recently Morey et al.,22 in a multi-
institutional project, demonstrated significant reduction in
errors when an intensive CRM team training, but without use
of simulation, was implemented.
Several other studies have evaluated the impact of
CRM. The outcome measures from these studies have reported
satisfaction by students with the training,20,73-76 improved
team performance dynamics in clinical situations18 or in a
simulated environment,37,75,77-79 decreased incidence of bedside
errors,18 and improved performance in a simulation environ-
ment.37,75,80,81 As a general rule, better teamwork translates into
better performance.82 Most recently DeVita et al.80 have pro-
vided some important data about the efficacy of training
teams in a simulated environment. They found that prior
CRM training for medical emergency response teams results
in better later performance of the same teams in the same
408 Section VII—Computers, Engineering, and the Future
simulated environment. The potential benefits of simulation
may extend beyond CRM. For example, simulation studies or
simulation modeling may be used to inform infection control
methods,83 identify features of team efficiency, such as task
allocation and closed-loop communication, that can translate
into improved outcome,84 explain variance across patient
cohorts,85 enhance drug dosing,86 evaluate performance of
common ICU monitors (e.g., glucometer),87 evaluate the
effects of fatigue on performance,88 or maximize design,
enrollment, and execution of clinical trials,89 among others.
Conclusion
In summary, there has been significant progress in patient
safety but there remains a need for practitioners to use CRM
methods that have been found to be helpful in high reliability
organizations such as the airline industry or nuclear energy
industries. Based on these organizations’ excellent safety
records, it becomes logical to suggest that using simulation of
rare events for the experienced or simulation of common
events for the inexperienced should be the next vista in patient
safety and in critical care taught largely in the context of CRM
training with simulation as an important tool.
It is clear that many advances are emerging in this field and
that the medical culture of the twenty-first century is going to
be extraordinarily different from that of the twentieth century.
How health care providers are educated also is changing;
alternative media resources such as Twitter,90 web-based edu-
cation, computer-assisted learning,91 and simulation-based
medical education92 that apply to both students and estab-
lished multidisciplinary teams93 now are commonplace. We
are going to transition from “see one, do one, teach one,” to a
culture of “practice it until proficient and then, and only then,
… do one.” Indeed, it should not be surprising to find the
twenty-first century routine to include the use of high-stakes
simulation for certification and hospital credentialing. As
such these advances represent an important cultural shift in
medicine, one that is still evolving and likely to affect the
practice of neurocritical care, including how various monitors
are used.
References
1. Abrahamson S. An interview of Professor Stephen Abrahamson. Educ Health
(Abingdon) 2003;16(3):378–84.
2. Smith CC, Huang GC, Newman LR, et al. Simulation training and its effect
on long-term resident performance in central venous catheterization. Simul
Healthc 2010;5(3):146–51.
3. Issenberg SB, Chung HS, Devine LA. Patient safety training simulations
based on competency criteria of the accreditation council for graduate
medical education. Mt Sinai J Med 2011;78(6):842–53.
4. Bose RR, Matyal R, Warraich HJ, et al. Utility of a transesophageal
echocardiographic simulator as a teaching tool. J Cardiothorac Vasc Anesth
2011;25(2):212–15.
5. Luciano CJ, Banerjee PP, Lemole Jr GM, et al. Second generation haptic
ventriculostomy simulator using the ImmersiveTouch system. Stud Health
Technol Inform 2006;119:343–8.
6. Banerjee PP, Luciano CJ, Rizzi S. Virtual reality simulations. Anesthesiol Clin
2007;25(2):337–8.
7. Kofke WA, Rosen KA, Barbaccia J, et al. The value of acute care simulation.
W V Med J 2000;96(2):396–402.
8. Peberdy MA, Kaye W, Ornato JP, et al. Cardiopulmonary resuscitation of
adults in the hospital: a report of 14720 cardiac arrests from the National
Registry of Cardiopulmonary Resuscitation. Resuscitation 2003;58:297–308.
9. Herlitz J, Bång A, Alsén B, et al. Characteristics and outcome among patients
suffering from in hospital cardiac arrest in relation to the interval between
collapse and start of CPR. Resuscitation 2002;53(1):21–7.
10. Adams BD, Zeiler KK, Jackson WO, et al. Emergency medicine residents
effectively direct inhospital cardiac arrest teams. Am J Emerg Med 2005;
23(3):304–10.
11. Foraida MI, Devita MA, Braithwaite RS, et al. Improving the utilization of
medical crisis teams (Condition C) at an urban tertiary care hospital. J Crit
Care 2003;18:87–94.
12. Holmberg M, Holmberg S, Herlitz J, et al. Factors modifying the effect of
bystander cardiopulmonary resuscitation on survival in out-of-hospital
cardiac arrest patients in Sweden. Eur Heart J 2001;22(6):511–19.
13. Bellomo R, Goldsmith D, Uchino S, et al. A prospective before-and-after trial
of a medical emergency team. Med J Aust 2003;179:283–7.
14. Goldhill DR, Worthington LM, Mulcahy AJ, et al. The patient-at-risk team:
identifying and managing seriously ill ward patients. Anaesthesia 1999;54:
853–60.
15. Buist MD, Moore GE, Bernard SA, et al. Effects of a medical emergency team
on reduction of incidence of and mortality from unexpected cardiac arrests
in hospital: preliminary study. Br Med J 2002;324(7334):387–90.
16. Petrie D, Lane P, Stewart T. An evaluation of patient outcomes comparing
trauma team activated versus trauma team not activated using TRISS
analysis. Trauma and Injury Severity Score. J Trauma Inj Infect Crit Care
1996;41(5):870–3.
17. Wiener EL, Helmreich RL. Cockpit resource management. San Diego, CA:
Academic Press; 1993.
18. Orlady J. Airline pilot training today and tomorrow, cockpit resource
management. In: Weiner EL, Kanki BG, Helmreich RL, editors. Cockpit
resource management. San Diego: Academic Press; 1995. p. 447–78.
19. Eisen LA, Savel RH. What went right: lessons for the intensivist from the
crew of U.S. Airways Flight 1549. Chest 2009;136(3):910–17.
20. Schroeder T, Schierbeck J, Howardy P, et al. Effect of labetalol on CBF and
middle cerebral arterial flow velocity in healthy volunteers. Neurol Res 1991;
13:10.
21. Risser DT, Rice MM, Salisbury ML, et al. The potential for improved
teamwork to reduce medical errors in the emergency department. The
MedTeams Research Consortium. Ann Emerg Med 1999;34(3):373–83.
22. Morey JC, Simon R, Jay GD, et al. Error reduction and performance
improvement in the emergency department through formal teamwork
training: evaluation results of the MedTeams project. Health Serv Res 2002;
37(6):1553–81.
23. Ziewacz JE, Arriaga AF, Bader AM, et al. Crisis checklists for the operating
room: development and pilot testing. J Am Coll Surg 2011;213(2):212–17.
24. Fernandez GL, Lee PC, Page DW, et al. Implementation of full patient
simulation training in surgical residency. J Surg Educ 2010;67(6):393–9.
25. Lambden S, Martin B. The use of computers for perioperative simulation in
anesthesia, critical care, and pain medicine. Anesthesiol Clin 2011;29(3):
521–31.
26. Mould J, White H, Gallagher R. Evaluation of a critical care simulation series
for undergraduate nursing students. Contemp Nurse 2011;38(1-2):180–90.
27. Pascual JL, Holena DN, Vella MA, et al. Short simulation training improves
objective skills in established advanced practitioners managing emergencies
on the ward and surgical intensive care unit. J Trauma 2011;71(2):330–8.
28. Cheng A, Donoghue A, Gilfoyle E, et al. Simulation-based crisis resource
management training for pediatric critical care medicine: a review for
instructors. Pediatr Crit Care Med 2011;14 [Epub ahead of print].
29. Risser DT, Rice MM, Salisbury ML, et al. The potential for improved
teamwork to reduce medical errors in the emergency department. Ann
Emerg Med 1999;34(3):373–83.
30. Wright MC, Taekman JM, Endsley MR. Objective measures of situation
awareness in a simulated medical environment. Qual Saf Health Care 2004;
13(Suppl 1):165–71.
31. Reason J. The contribution of latent human failures to the breakdown of
complex systems. Philos Trans R Soc Lond B Biol Sci 1990;327(1241):
475–84.
32. Baker DP, Salas E, King H, et al. The role of teamwork in the professional
education of physicians: current status and assessment recommendations.
Joint Commission J Quality Patient Safety 2005;31(4):185–202.
33. Hunt EA, Shilkofski NA, Stavroudis TA, et al. Simulation: translation to
improved team performance. Anesthesiol Clin 2007;25(2):301–19.
34. Abrahamson S, Denson JS, Wolf RM. Effectiveness of a simulator in training
anesthesiology residents. J Med Educ 1969;44:515–19.
35. Berman S. Individual lifetime achievement: Jeffrey B. Cooper, PhD. Joint
Commission J Quality Patient Safety 2003;29(12):625–33.
36. Directory of Sim Centers. Available at http://ssih.org/directory-of-sim-
centers-2. The Society for Simulation in Healthcare. Accessed June 28, 2012.
37. Howard SK, Gaba DM, Fish KJ, et al. Anesthesia crisis resource management
training: teaching anesthesiologists to handle critical incidents. Aviat Space
Environ Med 1992;63(9):763–70.
 Section VII—Computers, Engineering, and the Future 409
38. Good ML, Gravenstein JS. Anesthesia simulators and training devices. Int
Anesthesiol Clin 1989;27:161–8.
39. Chopra V, Engbers FH, Geerts MJ, et al. The Leiden anesthesia simulator. Br
J Anaesth 1994;73(3):287–92.
40. Devitt JH, Kurrek M, Cohen M, et al. Testing internal consistency and
construct validity during evaluation of performance in a patient simulator.
Anesth Analg 1998;86(6):1160–4.
41. DeAnda A, Gaba DM. Unplanned incidents during comprehensive
anesthesia simulation. Anesth Analg 1990;71(1):77–82.
42. DeAnda A, Gaba DM. Role of experience in the response to simulated
critical incidents. Anesth Analg 1991;72(3):308–15.
43. Schwid H, O’Donnell D. Anesthesiologists’ management of simulated critical
incidents. Anesthesiology 1991;76:495–501.
44. Botney R, Gaba DM, Howard SK, et al. The role of fixation error in
preventing the detection and correction of a simulated volatile anesthetic
overdose (abstract). Anesthesiology 1993;79:A1115.
45. Botney R, Gaba DM, Howard SK. Anesthesiologists performance during
simulated loss of pipeline oxygen (abstract). Anesthesiology 1993;79(A1118).
46. Holzman RS, Cooper JB, Gaba DM, et al. Anesthesia crisis resource
management: real-life simulation training in operating room crises. J Clin
Anesth 1995;7(8):675–87.
47. Good ML, Gravenstein JS, Mahla ME, et al. Can simulation accelerate the
learning of basic anesthesia skills by beginning anesthesia residents.
Anesthesiology 1992;77:A1133.
48. Chopra V, Gesink B, DeJong J, et al. Does training on an anesthesia
simulator lead to improvement in performance? Br J Anaesth 1994;73:293–7.
49. Hunt EA, Nelson KL, Shilkofski NA, et al. Performance of crucial
resuscitation maneuvers during simulated pediatric cardiopulmonary
arrests: the effect of an educational intervention. Simul Healthc 2006;
1(2):139 (abstract).
50. Seymour NE, Gallagher AG, Roman SA, et al. Virtual reality training
improves operating room performance: results of a randomized, double-
blinded study. Ann Surg 2002;236(4):458–63.
A complete list of references for this chapter can be found online at
www.expertconsult.com.
 Section VII—Computers, Engineering, and the Future 409.e1
References
1. Abrahamson S. An interview of Professor Stephen Abrahamson. Educ Health
(Abingdon) 2003;16(3):378–84.
2. Smith CC, Huang GC, Newman LR, et al. Simulation training and its effect
on long-term resident performance in central venous catheterization. Simul
Healthc 2010;5(3):146–51.
3. Issenberg SB, Chung HS, Devine LA. Patient safety training simulations
based on competency criteria of the accreditation council for graduate
medical education. Mt Sinai J Med 2011;78(6):842–53.
4. Bose RR, Matyal R, Warraich HJ, et al. Utility of a transesophageal
echocardiographic simulator as a teaching tool. J Cardiothorac Vasc Anesth
2011;25(2):212–15.
5. Luciano CJ, Banerjee PP, Lemole Jr GM, et al. Second generation haptic
ventriculostomy simulator using the ImmersiveTouch system. Stud Health
Technol Inform 2006;119:343–8.
6. Banerjee PP, Luciano CJ, Rizzi S. Virtual reality simulations. Anesthesiol Clin
2007;25(2):337–8.
7. Kofke WA, Rosen KA, Barbaccia J, et al. The value of acute care simulation.
W V Med J 2000;96(2):396–402.
8. Peberdy MA, Kaye W, Ornato JP, et al. Cardiopulmonary resuscitation of
adults in the hospital: a report of 14720 cardiac arrests from the National
Registry of Cardiopulmonary Resuscitation. Resuscitation 2003;58:297–308.
9. Herlitz J, Bång A, Alsén B, et al. Characteristics and outcome among patients
suffering from in hospital cardiac arrest in relation to the interval between
collapse and start of CPR. Resuscitation 2002;53(1):21–7.
10. Adams BD, Zeiler KK, Jackson WO, et al. Emergency medicine residents
effectively direct inhospital cardiac arrest teams. Am J Emerg Med
2005;23(3):304–10.
11. Foraida MI, Devita MA, Braithwaite RS, et al. Improving the utilization of
medical crisis teams (Condition C) at an urban tertiary care hospital. J Crit
Care 2003;18:87–94.
12. Holmberg M, Holmberg S, Herlitz J, et al. Factors modifying the effect of
bystander cardiopulmonary resuscitation on survival in out-of-hospital
cardiac arrest patients in Sweden. Eur Heart J 2001;22(6):511–19.
13. Bellomo R, Goldsmith D, Uchino S, et al. A prospective before-and-after trial
of a medical emergency team. Med J Aust 2003;179:283–7.
14. Goldhill DR, Worthington LM, Mulcahy AJ, et al. The patient-at-risk team:
identifying and managing seriously ill ward patients. Anaesthesia 1999;54:
853–60.
15. Buist MD, Moore GE, Bernard SA, et al. Effects of a medical emergency team
on reduction of incidence of and mortality from unexpected cardiac arrests
in hospital: preliminary study. Br Med J 2002;324(7334):387–90.
16. Petrie D, Lane P, Stewart T. An evaluation of patient outcomes comparing
trauma team activated versus trauma team not activated using TRISS
analysis. Trauma and Injury Severity Score. J Trauma Inj Infect Crit Care
1996;41(5):870–3.
17. Wiener EL, Helmreich RL. Cockpit resource management. San Diego, CA:
Academic Press; 1993.
18. Orlady J. Airline pilot training today and tomorrow, cockpit resource
management. In: Weiner EL, Kanki BG, Helmreich RL, editors. Cockpit
resource management. San Diego: Academic Press; 1995. p. 447–78.
19. Eisen LA, Savel RH. What went right: lessons for the intensivist from the
crew of US Airways Flight 1549. Chest 2009;136(3):910–17.
20. Schroeder T, Schierbeck J, Howardy P, et al. Effect of labetalol on CBF and
middle cerebral arterial flow velocity in healthy volunteers. Neurol Res 1991;
13:10.
21. Risser DT, Rice MM, Salisbury ML, et al. The potential for improved
teamwork to reduce medical errors in the emergency department. The
MedTeams Research Consortium. Ann Emerg Med 1999;34(3):373–83.
22. Morey JC, Simon R, Jay GD, et al. Error reduction and performance
improvement in the emergency department through formal teamwork
training: evaluation results of the MedTeams project. Health Serv Res 2002;
37(6):1553–81.
23. Ziewacz JE, Arriaga AF, Bader AM, et al. Crisis checklists for the operating
room: development and pilot testing. J Am Coll Surg 2011;213(2):212–17.
24. Fernandez GL, Lee PC, Page DW, et al. Implementation of full patient
simulation training in surgical residency. J Surg Educ 2010;67(6):393–9.
25. Lambden S, Martin B. The use of computers for perioperative simulation in
anesthesia, critical care, and pain medicine. Anesthesiol Clin 2011;29(3):
521–31.
26. Mould J, White H, Gallagher R. Evaluation of a critical care simulation series
for undergraduate nursing students. Contemp Nurse 2011;38(1-2):180–90.
27. Pascual JL, Holena DN, Vella MA, et al. Short simulation training improves
objective skills in established advanced practitioners managing emergencies
on the ward and surgical intensive care unit. J Trauma 2011;71(2):330–8.
28. Cheng A, Donoghue A, Gilfoyle E, et al. Simulation-based crisis resource
management training for pediatric critical care medicine: a review for
instructors. Pediatr Crit Care Med 2011;14 [Epub ahead of print].
29. Risser DT, Rice MM, Salisbury ML, et al. The potential for improved
teamwork to reduce medical errors in the emergency department. Ann
Emerg Med 1999;34(3):373–83.
30. Wright MC, Taekman JM, Endsley MR. Objective measures of situation
awareness in a simulated medical environment. Qual Saf Health Care
2004;13(Suppl 1):165–71.
31. Reason J. The contribution of latent human failures to the breakdown of
complex systems. Philos Trans R Soc Lond B Biol Sci 1990;327(1241):
475–84.
32. Baker DP, Salas E, King H, et al. The role of teamwork in the professional
education of physicians: current status and assessment recommendations.
Joint Commission J Quality Patient Safety 2005;31(4):185–202.
33. Hunt EA, Shilkofski NA, Stavroudis TA, et al. Simulation: translation to
improved team performance. Anesthesiol Clin 2007;25(2):301–19.
34. Abrahamson S, Denson JS, Wolf RM. Effectiveness of a simulator in training
anesthesiology residents. J Med Educ 1969;44:515–19.
35. Berman S. Individual lifetime achievement: Jeffrey B. Cooper, PhD. Joint
Commission J Quality Patient Safety 2003;29(12):625–33.
36. Directory of Sim Centers. Available at http://ssih.org/directory-of-sim-
centers-2. The Society for Simulation in Healthcare. Accessed June 28, 2012.
37. Howard SK, Gaba DM, Fish KJ, et al. Anesthesia crisis resource management
training: teaching anesthesiologists to handle critical incidents. Aviat Space
Environ Med 1992;63(9):763–70.
38. Good ML, Gravenstein JS. Anesthesia simulators and training devices. Int
Anesthesiol Clin 1989;27:161–8.
39. Chopra V, Engbers FH, Geerts MJ, et al. The Leiden anesthesia simulator.
Br J Anaesth 1994;73(3):287–92.
40. Devitt JH, Kurrek M, Cohen M, et al. Testing internal consistency and
construct validity during evaluation of performance in a patient simulator.
Anesth Analg 1998;86(6):1160–4.
41. DeAnda A, Gaba DM. Unplanned incidents during comprehensive
anesthesia simulation. Anesth Analg 1990;71(1):77–82.
42. DeAnda A, Gaba DM. Role of experience in the response to simulated
critical incidents. Anesth Analg 1991;72(3):308–15.
43. Schwid H, O’Donnell D. Anesthesiologists’ management of simulated critical
incidents. Anesthesiology 1991;76:495–501.
44. Botney R, Gaba DM, Howard SK, et al. The role of fixation error in
preventing the detection and correction of a simulated volatile anesthetic
overdose (abstract). Anesthesiology 1993;79:A1115.
45. Botney R, Gaba DM, Howard SK. Anesthesiologists performance during
simulated loss of pipeline oxygen (abstract). Anesthesiology 1993;79(A1118).
46. Holzman RS, Cooper JB, Gaba DM, et al. Anesthesia crisis resource
management: real-life simulation training in operating room crises. J Clin
Anesth 1995;7(8):675–87.
47. Good ML, Gravenstein JS, Mahla ME, et al. Can simulation accelerate the
learning of basic anesthesia skills by beginning anesthesia residents.
Anesthesiology 1992;77:A1133.
48. Chopra V, Gesink B, DeJong J, et al. Does training on an anesthesia
simulator lead to improvement in performance? Br J Anaesth 1994;73:
293–7.
49. Hunt EA, Nelson KL, Shilkofski NA, et al. Performance of crucial
resuscitation maneuvers during simulated pediatric cardiopulmonary
arrests: the effect of an educational intervention. Simul Healthc 2006;
1(2):139 (abstract).
50. Seymour NE, Gallagher AG, Roman SA, et al. Virtual reality training
improves operating room performance: results of a randomized, double-
blinded study. Ann Surg 2002;236(4):458–63.
51. Rose FD, Attree EA, Brooks BM, et al. Training in virtual environments:
transfer to real world tasks and equivalence to real task training. Ergonomics
2000;43(4):494–511.
52. Eversbusch A, Grantcharov TP. Learning curves and impact of psychomotor
training on performance in simulated colonoscopy: a randomized trial using
a virtual reality endoscopy trainer. Surg Endosc 2004;18(10):1514–18.
53. Wilhelm DM, Ogan K, Roehrborn CG, et al. Assessment of basic endoscopic
performance using a virtual reality simulator. J Am Coll Surg 2002;195(5):
675–81.
54. Uchal M, Raftopoulos Y, Tjugum J, et al. Validation of a six-task simulation
model in minimally invasive surgery. Surg Endosc 2005;19(1):109–16.
55. Tuggy ML. Virtual reality flexible sigmoidoscopy simulator training: impact
on resident performance. J Am Board Fam Pract 1998;11(6):426–33.
56. Grantcharov TP, Kristiansen VB, Bendix J, et al. Randomized clinical trial of
virtual reality simulation for laparoscopic skills training. Br J Surg 2004;
91(2):146–50.
409.e2 Section VII—Computers, Engineering, and the Future
57. Ahlberg G, Hultcrantz R, Jaramillo E, et al. Virtual reality colonoscopy
simulation: a compulsory practice for the future colonoscopist? Endoscopy
2005;37(12):1198–204.
58. Gerson LB, Van Dam J. A prospective randomized trial comparing a virtual
reality simulator to bedside teaching for training in sigmoidoscopy.
Endoscopy 2003;35(7):569-75; erratum appears in Endoscopy 2004;36(2):185.
59. Howard SK, Gaba DM, Fish KJ, et al. Anesthesia crisis resource management
training: teaching anesthesiologists to handle critical incidents. Aviat Space
Environ Med 1992;63(9):763–70.
60. Cooper JB. Are simulation and didactic crisis resource management (CRM)
training synergistic? Qual Saf Health Care 2004;13(6):413–14.
61. Baker DP, Gustafson S, Beaubien JM, et al. Medical teamwork and patient
safety: the evidence-based relation. Literature review. AHRQ Publication
No. 05-0053; 2005. Available from: http://www.ahrq.gov/qual/medteam.
62. Savoldelli GL, Naik VN, Park J, et al. Value of debriefing during simulated
crisis management: oral versus video-assisted oral feedback. Anesthesiology
2006;105(2):279–85.
63. Musacchio Jr MJ, Smith AP, McNeal CA, et al. Neuro-critical care skills
training using a human patient simulator. Neurocrit Care 2010;13(2):
169–75.
64. Lighthall GK, Barr J, Howard SK, et al. Use of a fully simulated intensive care
unit environment for critical event management training for internal
medicine residents. Crit Care Med 2003;31(10):2437–43.
65. Hunt EA, Nelson KL, Shilkofski NA. Simulation in medicine: addressing
patient safety and improving the interface between healthcare providers and
medical technology. Biomed Instrum Technol 2006;40(5):399–404.
66. Vos GD, Nissen AC, Nieman FH, et al. Comparison of interhospital pediatric
intensive care transport accompanied by a referring specialist or a specialist
retrieval team. Intensive Care Med 2004;30(2):302–8.
67. Flabouris A, Runciman WB, Levings B. Incidents during out-of-hospital
patient transportation. Anaesth Intensive Care 2006;34(2):228–36.
68. Lupton BA, Pendray MR. Regionalized neonatal emergency transport. Semin
Neonatol 2004;9(2):125–33.
69. Chen JG, Mistry KP, Wright MC, et al. Postoperative handoff
communication: a simulation-based training method. Simul Healthc 2010;
5(4):242–7.
70. Stocker M, Allen M, Pool N, et al. Impact of an embedded simulation team
training programme in a paediatric intensive care unit: a prospective,
single-centre, longitudinal study. Intensive Care Med 2011 Oct 1 [Epub
ahead of print].
71. Kohn L, Donaldson M. To err is human: building a safer health system.
Washington, DC: National Academy Press, 1999.
72. Moorthy K, Munz Y, Adams S, et al. Human factors analysis of technical and
team skills among surgical trainees during procedural simulations in a
simulated operating theatre. Ann Surg 2005;242(5):631–9.
73. France DJ, Stiles R, Gaffney EA, et al. Crew resource management
training: clinicians’ reactions and attitudes. AORN J 2005:82(2):212–24.
74. Lighthall GK, Barr J, Howard SK, et al. Use of a fully simulated intensive care
unit environment for critical event management training for internal
medicine residents. Crit Care Med 2003;31(10):2437–43.
75. Blum RH, Raemer DB, Carroll JS, et al. A method for measuring the
effectiveness of simulation-based team training for improving
communication skills. Anesth Analg 2005;100(5):1375–80.
76. Ali J, Dunn J, Eason M, et al. Comparing the standardized live trauma
patient and the mechanical simulator models in the ATLS initial assessment
station. J Surg Res 2010;162(1):7–10.
77. Shapiro MJ, Morey JC, Small SD, et al. Simulation based teamwork training
for emergency department staff: does it improve clinical team performance
when added to an existing didactic teamwork curriculum? Qual Safet Health
Care 2004;13(6):417–21.
78. Holcomb JB, Dumire RD, Crommett JW, et al. Evaluation of trauma team
performance using an advanced human patient simulator for resuscitation
training. J Trauma 2002;52(6):1078–85.
79. Holcomb JB, Dumire RD, Crommett JW, et al. Evaluation of trauma team
performance using an advanced human patient simulator for resuscitation
training. J Trauma 2002;52(6):1078–85.
80. DeVita MA, Schaefer J, Lutz J, et al. Improving medical emergency team
(MET) performance using a novel curriculum and a computerized human
patient simulator. Qual Safet Health Care 2005;14(5):326–31.
81. Aggarwal R, Undre S, Moorthy K, et al. The simulated operating theatre:
comprehensive training for surgical teams. Qual Saf Health Care 2004;
13(Suppl 1):27–32.
82. Siassakos D, Fox R, Crofts JF, et al. The management of a simulated
emergency: better teamwork, better performance. Resuscitation 2011;82(2):
203–6.
83. Karkada UH, Adamic LA, Kahn JM, et al. Limiting the spread of highly
resistant hospital-acquired microorganisms via critical care transfers: a
simulation study. J Intensive Care Med 2011;37(10):1633–40.
84. Siassakos D, Bristowe K, Draycott TJ, et al. Clinical efficiency in a simulated
emergency and relationship to team behaviours: a multisite cross-sectional
study. Br J Obstet Gyenaecol 2011;118(5):596–607.
85. Chase JG, Le Compte AJ, Suhaimi F, et al. Tight glycemic control in critical
care—the leading role of insulin sensitivity and patient variability: a review
and model-based analysis. Comput Methods Programs Biomed 2011;
102(2):156–71.
86. Roberts JA, Kirkpatrick CM, Lipman J. Monte Carlo simulations:
maximizing antibiotic pharmacokinetic data to optimize clinical practice for
critically ill patients. J Antimicrob Chemother 2011;66(2):227–31.
87. Karon BS, Boyd JC, Klee GG. Glucose meter performance criteria for tight
glycemic control estimated by simulation modeling. Clin Chem 2010;56(7):
1091–7.
88. Sharpe R, Koval V, Ronco JJ, et al. The impact of prolonged continuous
wakefulness on resident clinical performance in the intensive care unit: a
patient simulator study. Crit Care Med 2010;38(3):766–70.
89. Brindley PG, Dunn WF. Simulation for clinical research trials: a theoretical
outline. J Crit Care 2009;24(2):164–7.
90. Mistry V. Critical care training: using Twitter as a teaching tool. Br J Nurs
2011;20(20):1292–6.
91. Kleinpell R, Ely EW, Williams G, et al. Web-based resources for critical care
education. Crit Care Med 2011;39(3):541–53.
92. McGaghie WC, Issenberg SB, Petrusa ER, et al. A critical review of
simulation-based medical education research: 2003-2009. Med Educ 2010;
44(1):50–63.
93. Frengley RW, Weller JM, Torrie J, et al. The effect of a simulation-based
training intervention on the performance of established critical care unit
teams. Crit Care Med 2011;39(12):2605–11.
VII
Chapter
42  
Robotic Telepresence in
Neurocritical Care: The Next
Paradigm in Critical Care
Paul M. Vespa
Introduction
Timely assessment and treatment of intensive care unit (ICU)
patients is a major goal in intensive care.1,2 This requirement
can be difficult given the shortage of available intensive care
physicians and surgeons1 and ICU beds. Delays in both patient
assessment and physician response can result in lost opportu-
nities to improve patient outcome and can result in increased
morbidity and length of stay.3 These problems particularly
affect neurocritical care patients, because the dichotomy
between the need for rapid intervention and the shortage of
highly trained experts is most acute.4 Most neurocritical care
patients are at high risk for further secondary brain injury
such as brain ischemia or elevated intracranial pressure (ICP),5
and the time window for intervention is quite short. Hence
there is a great need to quickly identify and rapidly respond
to secondary brain insults once they occur. However, there is
a nationwide shortage of neurointensivists. The use of infor-
mation technology may hold the answer to facilitate this goal
of quick diagnosis and rapid response despite the workforce
shortage.6
This chapter outlines the role of information technology to
change the paradigm for patient care using robotic telepres-
ence (RTP). Robots have been used increasingly in a variety
of surgical procedures, in large part to augment surgical visu-
alization, dexterity, and precision, although the effect on
patient outcome and cost efficacy rather than on surrogate
endpoints is still debated.7-15 In recent years there also has been
a rapid expansion of ICU telemedicine that combines audio-
visual technology with the electronic medical record to provide
critical care from a remote location.16 This includes continu-
ous offsite monitoring for the entire ICU or all ICUs in a
hospital,17 or as is the focus of this chapter, targeted individual
patient evaluation. This chapter outlines how RTP works, how
it is used, how it impacts patient care, and how developers
anticipate it will be used in the future in the neurointensive
care unit. There are essentially two components to the RTP
system: the robot and the computerized information system.
These two elements have synergy and facilitate high-quality,
timely clinical care. RTP may represent the next clinical care
paradigm in the NCCU. This approach, perhaps through
implementation of evidence-based practice in the ICU applied
410
by an intensivist who can conduct remote rounds has the
potential to improve patient outcomes.
Physician Response Paradigms
Rapid response and implementation of goal-directed treat-
ment improve outcomes for critically ill patients with sepsis.18
Physician attention to patients is divided into prospective
and responsive attention.19 In most neurointensive care units
(NICU), prospective physician attention is conducted on
morning bedside rounds. Morning bedside rounds are made
to evaluate the patient’s clinical condition (the physical exam-
ination), and to evaluate the laboratory studies, brain images,
and other data that accumulate over time. Morning rounds in
the NICU are a mixture of retrospective analysis of trends in
data and prospective planning of treatment for the day. These
rounds are usually based on scheduled periods of time and
result in a well-formulated “plan of attack” for the day, which
likely will change several times during the course of the day
based on intercurrent changes in patient condition. For a
typical brain injury patient with ICP monitoring, it has been
calculated that the intensivist evaluates more than 650 data
points during each episode of bedside rounds.2 These data
points consist of hourly vital signs, waveform morphology
values, brain monitoring values such as brain tissue oxygen
and microdialysis, radiologic studies, clinical examination
findings, laboratory data, verbal story-line information, and
visual-cue information from body language and appearance.
These data points involve integration of graphic information,
longitudinal trends of data, and cross-sectional review of data
at fixed points in time. This analysis is done prospectively
during morning rounds to make a reasoned clinical treat-
ment plan.
Physicians also respond to changes in the clinical condition
of the patient by being paged by the nurse or other health care
professionals. This can be called an urgent or emergent response,
and occurs throughout the day and night. During such an
urgent response, the physician often needs to consider the
same set of data points and to determine the evolution
of the patient over time. However, the physician may not be
at the bedside when paged and may not have all these data
points available to him or her; yet he or she still is required to
© Copyright 2013 Elsevier Inc. All rights reserved.
 Section VII—Computers, Engineering, and the Future 411
make a clinical decision.1,4 The response is by telephone and,
at best, a delayed face-to-face evaluation of the patient, often
hours later. This means at the time of the initial call, the physi-
cian may not have access to visual, graphic, or waveform infor-
mation, and perhaps the electronic medical records or
electronic charting data. Thus the physician is at a disadvan-
tage in processing the clinical information and reaching a
clinical decision during the time of urgent patient need. This
disadvantage may result in poor clinical judgment or medical
errors. Medical errors can occur when incomplete informa-
tion is presented and/or verbal information alone is used.3
One way to avoid medical errors is to increase intensivist
coverage in the ICU.1 Because there is a shortage of intensiv-
ists, it is difficult to increase the number of intensivists, per se.
However, using RTP technology, one may be able to increase
the percentage of time in which the intensivist is telepresent
in the ICU. As discussed later in this chapter, robotic telepres-
ence can help reduce or avoid medical errors.
Robotic Telepresence Technology
InTouch Technologies Inc. (InTouch Health), a privately held
company based in Santa Barbara, CA, manufactures the
robotic telecommunication system currently in use in the
author’s ICU.2 The company has pioneered Remote Presence
technology for hospital providers. Through its RP-7 Remote
Presence Robotic System, a proprietary mobile robotic and
communications platform, health care professionals are able
to consult with hospital-based patients and staff more easily
and frequently. The InTouch Health solution leverages the
time and expertise of health care professionals across multiple
care facilities to improve the efficiency and effectiveness of
care delivery. The RP-7 Robotic System enables “remote pres-
ence” that allows individuals to project themselves from one
geographic location to another, via the wireless mobile robot,
such that they can interact from that remote location.
There are two main components to the Remote Presence
System: the RP-7 Robot (Fig. 42.1) and control station (Fig.
42.2). The technology is dependent on the use of broadband
internet connectivity and existing wireless network within the
hospital. Most hospitals already have the necessary network
infrastructure in place. The system incorporates two-way
audio and video that permits observation and face-to-face
interaction between health care professionals and patients.
The robot operates in the hospital environment on an 802.11
Wi-Fi network. It is controlled remotely by a health care pro-
fessional using a laptop or desktop computer, called a control
station, which can be linked to the Internet through secure
broadband connection. Security protocols enable Health
Insurance Portability and Accountability Act (HIPAA) com-
pliant audiovisual communication across the public internet,
via a hardwire or cellular connection.
From the control station, the health care professional can
drive the robot throughout the facility and communicate in
real time with patients, their families, and medical staff. The
RP-7 is 5.4 feet tall, weighs 220 lb, and travels up to 2 miles
per hour. The anthropomorphic design consists of a “head,”
which contains a flat screen monitor, microphone, and digital
camera with pan, tilt, and 10× zoom features. The central body
houses the speaker, robot-side volume control, and expansion
bay that can include a handset for private conversation, a
printer for physician’s notes, a digital stethoscope, and other
video inputs. The control station has a share-and-display
feature that permits images on the control station to be dis-
played remotely on the head/face of the robot, visible to the
audience in the ICU (see Fig. 42.2). The robot rides on a
A B
Fig. 42.1  Telepresence robot (Intouch Technologies, Santa Barbara,
CA). A, the frontal view of the robot, with head and neck consisting of
a flatscreen TV, and mobile neck device enabling 340-degree head
movement, and attached camera system for wide view and focused
views. Sensors along the base are designed to override manual controls
and enable collision avoidance. B, Rear projection of the robot showing
portals for digital and video diagnostic equipment, such as digital
stethoscope.
Fig. 42.2  Desktop physician control station. The
control station contains software, camera, microphone,
and digital Internet connection system, as well as a control
stick device. The physician drives the robot like playing a
video game. The double screen facilitates simultaneous
display of radiology and laboratory images and the intensive
care unit environment.
412 Section VII—Computers, Engineering, and the Future
A B
Fig. 42.3  Sample case display. These are picture files taken using the desktop control station of a new patient admission to the ICU. A, A
neurointensive care fellow is presenting the patient to the attending who is running the robot at 21:00. B, The appearance of a hard-copy computed
tomography scan, brought with the patient from an outside hospital, showing the Fisher grade 3 subarachnoid hemorrhage. C, The bloody output
from the ventriculostomy catheter 2 hours later, indicating release of blood via the ventriculostomy catheter. Note the intravenous pump display to
the left, indicating infusion of 3% sodium chloride.
patented holonomic platform about 20 inches wide, made up
of three balls that allow it to move in any direction, thus
greatly increasing maneuverability in tight spaces. An array of
infrared proximity sensors placed around the lower section of
the robot prevents it from colliding with people or objects.
The robot is powered by a rechargeable battery with a life of
5 to 6 hours depending on usage. The robot moves from
bedside to bedside under the control of the physician, is
capable of traveling on elevators, and has been safely used in
operating and procedure rooms. Direct feeds from clinical
information systems, such as digital radiology systems and lab
systems, are routinely done using the doctor’s workstation. At
UCLA, caregivers use the Global Care Quest (GCQ, Mission
Viejo, CA) clinical information system to view all types of
electronic medical record data. In this fashion, the physician
can discuss cases with colleagues, evaluate radiology, and eval-
uate the bedside environment (Fig. 42.3).
HIPAA compliance and consent for telemedicine is required
for any system.6,20 There are formal consent forms for all ICU
procedures, including RTP. Information technology and com-
pliance officer review of RTP is done routinely at centers that
implement the technology; acceptance of RTP technology has
been uniform across various regions of the country. Tele-
medicine has enjoyed a robust experience with medical-legal
circles. Increased physician participation and enhanced data
review actually makes medical-legal positions better com-
pared with telephonic communication. Multiple systems can
be used to provide medical documentation and computer
order entry, depending on the particular hospital. A separate
internal documentation system, called Stroke-Respond, is cur-
rently available, and can integrate into a hospital-based elec-
tronic medical record system for transparency. In July 2008,
Medicare initiated special ICU billing codes for telemedicine,
called 0188T and 0189T. These codes are currently under
study by Medicare.
Changing the Physician
Response Paradigm
The response to an urgent condition using RTP makes use of
visual information and real-time audiovisual communication
with the bedside nurse. In a recent study, the importance of
C
visual information was highlighted.20 The main types of criti-
cal data acquired from the telemedicine interaction were
judged by the attending physician to be important in the deci-
sion pathway. The distribution of these critical data is as
follows: 67% visual and 33% verbal information. Of the visual
information, the breakdown was as follows: physical examina-
tion (40.6%), physiologic monitor graphics (11.5%), printed
information from the medical record (5.8%), body language
of the nurse (5.7%), information about a tube or catheter
(3.8%). In a randomized controlled trial of telemedicine for
acute ischemic stroke, telemedicine provided important visual
and verbal information, improved diagnostic accuracy, and
enabled more complete data assessment as compared with the
telephonic paradigm.21 Similar reliance on visual and graphic
information has been reported using the electronic-ICU
system (Visicu, Baltimore, MD).22 These data speak to the
need for visual information to be part of the physician response
paradigm.
Uses of Robotic Telepresence
Rapid Response to Unstable Intensive Care
Unit Patient Condition
The RTP system enables the clinician to make a rapid response
to a sudden deterioration in patient condition. In a recent
study, Vespa et al20 demonstrated a marked reduction in physi-
cian response time for a face-to-face evaluation of the patient.
Overall there was a marked reduction in the attending physi-
cian response latency using RTP. The mean response time was
significantly shorter for patients with brain ischemia (7.8 ± 2.8
vs. 152 ± 85 minutes) and elevated intracranial pressure (11 ±
14 vs. 108 ± 55 minutes) (P < .001). Univariate analysis dem-
onstrated a significant reduction in the response latency for
each category and for the composite overall (P < .001). Figure
42.4 illustrates a comparison of face-to-face response times
using the RTP system by specific categories.
The reduction in response time to elevated ICP and brain
ischemia is clinically meaningful in that the time window to
intervention is quite short. For example, customary time for
an attending physician to visually see a patient with elevated
ICP averaged 108 minutes. With the use of RTP, the mean
time to a face-to-face contact was about 11 minutes. This
 Section VII—Computers, Engineering, and the Future 413

reduction in latency is substantial and clinically meaningful
because the duration of untreated, elevated ICP can be a
major determinant factor for morbidity and mortality after
traumatic brain injury. There was a similar reduction in
latency for other paroxysmal events that could contribute
irreversible brain damage, such as brain ischemia, seizures,
hypotension, and hypoxemia. This reduction was mirrored by
a reduction in the length of stay in those patients who are
most at risk for brain ischemia, namely those with subarach-
noid hemorrhage, stroke, and brain trauma. The impact of
rapid response on patient outcome remains untested. However,
the concept of rapid response to patient instability has been
found to reduce mortality overall in general critical care,22-25
suggesting that special methods to facilitate rapid response
Comparison of Response Time
350
300
250
Minutes
would have similar overall benefits. Other telemedicine
systems have reported improved outcomes for critically ill
patients.5
Prospective Compliance Monitoring
and Mentoring
A fundamental principal of safety in the ICU is compliance
and close mentoring of less experienced practitioners by their
senior colleagues.26 This requires presence in the ICU by the
senior practitioner. The RTP system enables the clinician to
make rounds during night hours and during times when tra-
ditionally the physician is not in the ICU, and facilitates face-
to-face discussions with simultaneous evaluation of imaging
and laboratory values. This enables physicians to supervise
directly the ongoing care of the patient and to enable mentor-
ing of the bedside nurse. In a recent study, such compliance
monitoring and mentoring accounted for 12% of all encoun-
ters, and together with avoidance of medical errors, accounted
for 20% of all encounters with the robot. Figure 42.5 illus-
trates a simultaneous evaluation of an ischemic limb with
Doppler ultrasound of the dorsalis pedis artery. The mutual
simultaneous evaluation and treatment of patients, with both

200 nurse and remote doctor acting together, facilitates patient

150 care. In addition, there are a number of groups that are using
100
the robot to enhance mentoring of inexperienced nurses by a
senior nurse educator. This provides the bedside nurse with
50 an approachable person with whom he or she can discuss
0 ongoing nursing issues that would otherwise not be discussed
with the physician. The senior nurse educator has the poten-
ia

P
s

ia
on

es
ia

um
ia

IC
em

tial to staff numerous hospitals and ICUs from a central loca-

em
ur

ur
si

hm

iri
lig
en

ed

iz
ox

ch

el
yt

Se

tion. Similar telementoring, teleproctoring, and telestration

O
ot

D
at
yp

is
rh
yp

ev

n
H

Ar

systems also have been used to enable trainees and mentors to

ai
H

El

Br

remotely observe surgical cases in the operating room or

Actual response time
Customary response time
Fig. 42.4  Face-to-face response time using the robot (solid bars)
compared with the customary face-to-face response time using
the traditional in-person paradigm. The responses for a number of
neurologic emergencies are individually shown, with mean and standard
deviation data between 5 and 20 cases per category. ICP, Intracranial
pressure.
provide intraoperative consultation by other surgeons at
remote sites.27-29 These systems thus can enhance education
and mentoring in the ICU.
Improving Efficiency of Care
Given the universal shortage of hospital and intensive care
unit beds, a significant focus on improving efficiency of care

A B
Fig. 42.5  Nurse mentoring examples: A, An example of physician-operated robot discussion with a nurse at the patient’s bedside. Review of laboratory
and imaging data are being simultaneously done, and discussion of results and formulation of treatment plans are occurring. B, An example of
mentoring of a nurse while examining an ischemic limb using physical exam (note dusky appearance of dorsal leg) and Doppler ultrasound of the
dorsalis pedis. Clinical decision making and confirmation of the nurse’s diagnosis of limb ischemia were made.
414 Section VII—Computers, Engineering, and the Future

and bed allotment is underway. The robot may be a mecha-

nism to facilitate improved efficiency of care in the same way
that it is used to improve compliance with treatment proto-
cols. This concept has been studied in the postoperative
setting, in which clinical pathways have been established to
facilitate quality of care and expedite efficiency of the hospital
stay. Gandsas and colleagues30 recently presented results from
a retrospective trial of RTP versus traditional rounding on
routine postoperative patients in a gastric bypass surgery
practice. In this study, patients were very satisfied with the
telerounding concept, and postoperative morbidity was not
increased. The length of stay was reduced with the use of
robotic telerounding. The attending surgeon was able to
increase the time spent doing office and operating room work A
because of better efficiency in the rounding process. This
enabled an overall improvement in physician workflow and
resulted in overall hospital and physician profit.
In the UCLA study,20 a similar improvement in efficiency
was found for an academic neurointensive care unit. In this
study, a comparison of the length of stay and costs associated
with the primary diagnoses in a neurointensive care unit were
examined 1 year before implementing the robot and com-
pared with the initial year of routine robot use. The diagnoses
were subarachnoid hemorrhage, brain trauma, intracerebral
hemorrhage, brain tumors, and ischemic stroke. The use of
the robot was associated with a substantial reduction in mean
length of stay for most diagnoses except brain tumors, and a
net cost savings of more than $1.1 million during that year.
B
Supervision and Mentoring for Invasive
Fig. 42.6  A, Mentoring of a house staff member in the performance of
Procedures in the Intensive Care Unit an external ventriculostomy catheter insertion using robotic telepresence.
Because there is a shortage of neurosurgeons and intensivists, The robotic physician was able to determine and guide the trajectory
invasive procedures may be delayed or not done. Robotics angle for drilling the hole and for catheter insertion. B, Mentoring of a
provides an opportunity for telementoring and teleproctor- nurse practitioner inserting a radial arterial line. The angle of needle
ing, although there is very limited study of this concept in the approach, as well as needle depth, was discussed in real time and the
NCCU.31 In a pilot study, the ability to supervise invasive mentoring suggestions were provided in order to get the ideal arterial
procedures using RTP has been demonstrated. This experi- blood jet flow, before attempting to place the guidewire and catheter.
ence mirrors that of general surgeons who routinely supervise
laparoscopic surgery of junior colleagues. In the author’s
NICU, there have been 40 invasive procedures using RTP to
mentor junior house staff in a pilot protocol. Under this para- design remains to be defined.32,35-38 In the near future, robots
digm, the attending specialist is able to watch the procedure, may help establish a reproducible method to quantitate pro-
provide real-time feedback to the junior physician, and coach cedural performance and so help assess surgical or critical care
the entire procedure. There is no delay in image acquisition trainees. Robots also may be adapted to a hybrid operating
or audiovisual communication. Using this paradigm, 97% of room (radiology suite) to augment neuroendovascular proce-
procedures were successfully completed, while one procedure dures and integrate various imaging modalities.39 This may
was aborted because of inability to gain venous access using become of particular relevance to the NCCU and performance
the technique. Importantly, 60% of these procedures were of thrombolytic procedures for stroke.
judged to be emergent and provided stabilizing or life-saving
care to the patient. An example case of ventriculostomy place-
ment in a patient with traumatic brain injury and elevated Emergency Department Triage
ICP is shown in Figure 42.6. Arterial access with visualization and Consultation
of pulsatile blood flow is appreciated through this system. There is a growing experience with telemedicine in the emer-
More recently studies have addressed the prospect of semi- gency department setting for neurocritical care consulta-
autonomous surgery and specifically whether a robotic system tions.21,40-43 Telemedicine enables rapid and more accurate
is capable of performing a burr hole procedure in specific assessment of stroke patients, enables early decisions about
locations and along predetermined trajectories to a precise administration of tissue plasminogen activator, and facilitates
depth.32,33 The robot, however, does not replace a health care early goal-directed treatment that enhances cerebral blood
provider and cannot be performed without a human being flow and avoids hemorrhagic complications. Use of telemedi-
present.34 In addition, whether robot-assisted surgery in neu- cine for acute stroke can result in improved outcomes for large
rosurgery improves upon surgical outcome and optimal populations, both in rural and urban settings. A number of
 Section VII—Computers, Engineering, and the Future 415
academic and private organizations have been formed to
enable delivery of acute stroke care to underserved regions.
This technology enables accurate triage of patients and enables
appropriate transfers of complex patients to tertiary care
centers in a hub and spoke pattern.44,45
A similar experience has been demonstrated in emergency
neurosurgical evaluation and treatment.46 In this study, the
use of a videoconferencing technology was associated with a
reduction in mortality using a combination of teleradiology
and videoconferencing, compared with telephone-only con-
sultation (25% vs. 35%, P < 0.025). This study did not use the
robotic device, but used the Polycom view station (Polycom
Inc, San Jose, CA). However, there was a high failure rate, 30%
of all cases, for this videoconferencing device, during this
study. Furthermore, these authors found that the teleradiology
component had the largest impact on mortality. Thus atten-
tion to the specific videoconferencing technology appears to
be warranted. In contrast, in a 4-year experience with the
Telepresence Robot, the failure rate has been less than 1%.
In the Neurocritical Care Robotic Telepresence Network,
there is connectivity to multiple community-based hospitals
and UCLA-affiliated hospitals. Triage in the emergency depart-
ment, with continuity of care in the ICU has become a very
useful model. The most recent development of the robotic
telepresence system is the feature called Multi-Presence. Multi-
Presence enables multiple physicians and nurses to be engaged
simultaneously through the robot with the patient, and to
communicate with each other at the same time. This enables
three- and four-way discussion groups and enables collective
problem solving (Fig. 42.7).
Conclusion
RTP is a powerful tool to facilitate patient care. The health
care and ICU world is changing, with a 24-7-365 mentality
for access to information and discussion. Seemingly, the cur­
rent traditional paradigm for patient care, with intermittent
rounds and telephone-based care, is being challenged by the
demands to be omnipresent and always available. RTP has
Fig. 42.7  Multipresence. Multiple users are able to see the patient via
the robot as well as see each other and have a group conversation
simultaneously. Two consultant physicians are shown to the far left of
the image, and the attending is shown in the bottom screen evaluating
the patient, who is shown in the large center screen. All participants can
communicate with each other in real time.
the potential to achieve a new paradigm in patient care and at
the same time make the physician more mobile and less
hospital-bound. The concept of changing the paradigm of
delivering critical care is penetrating the collective conscious-
ness,47 although barriers such as the perceived ease of use
and complexity, perceived usefulness, perceived behavioral
control, and ethical and legal dilemmas may affect adaptation
of robotic telepresence in the ICU.48,49 With robotic telepres-
ence, there is the potential to deliver the right care, right away,
to neurocritical care patients.
References
1. Angus DC, Shorr AF, White A, et al, Committee on Manpower for
Pulmonary and Critical Care Societies (COMPACCS). Critical care delivery
in the United States: distribution of services and compliance with Leapfrog
recommendations. Crit Care Med 2006;34:1016–24.
2. Vespa P. Multimodality monitoring and telemonitoring in neurocritical care:
from microdialysis to robotic telepresence. Curr Opin Crit Care 2005;11:
133–8.
3. Heffner JE, Ellis R, Zeno B. Safety in training and learning in the intensive
care unit. Crit Care Clin 2005;21:129–48.
4. Pronovost PJ, Thompson DA, Holzmueller CG, et al. The organization of
intensive care unit physician services. Crit Care Med 2007;35:2256–61.
5. Vespa PM. The implications of cerebral ischemia and metabolic dysfunction
for treatment strategies in neurointensive care. Curr Opin Crit Care 2006;12:
119–23.
6. Cummings J, Krsek C, Vermoch K, et al, University HealthSystem
Consortium ICU Telemedicine Task Force. Intensive care unit telemedicine:
review and consensus recommendations. Am J Med Qual 2007;22:239–50.
7. Barbash GI, Glied SA. New technology and health care costs: the case of
robot-assisted surgery. N Engl J Med 2010;363:701–4.
8. Maeso S, Reza M, Mayol JA, et al. Efficacy of the Da Vinci Surgical System in
abdominal surgery compared with that of laparoscopy: a systematic review
and meta-analysis. Ann Surg 2010;252:254–62.
9. Barnett JC, Judd JP, Wu JM, et al. Cost comparison among robotic,
laparoscopic, and open hysterectomy for endometrial cancer. Obstet Gynecol
2010;116:685–93.
10. Maan ZN, Gibbins N, Al-Jabri T, et al. The use of robotics in
otolaryngology-head and neck surgery: a systematic review. Am J
Otolaryngol 2011; Jun 7 [Epub ahead of print].
11. Kowalczyk KJ, Yu HY, Ulmer W, et al. Outcomes assessment in men
undergoing open retropubic radical prostatectomy, laparoscopic radical
prostatectomy, and robotic-assisted radical prostatectomy. World J Urol
2011; Mar 2 [Epub ahead of print].
12. Atluri P, Woo YJ. Minimally invasive robotic mitral valve surgery. Expert Rev
Med Devices 2011;8(1):115–20.
13. Antoniou GA, Riga CV, Mayer EK, et al. Clinical applications of robotic
technology in vascular and endovascular surgery. J Vasc Surg 2011;53(2):
493–9.
14. Khatlani K, Sharma S, Mendoza PJ, et al. The current state of robot assisted
radical prostatectomy. Minerva Urol Nefrol 2010;62(2):193–201.
15. Hellenthal NJ, Hussain A, Andrews PE, et al. Surgical margin status after
robot assisted radical cystectomy: results from the International Robotic
Cystectomy Consortium. J Urol 2010;184(1):87–91.
16. New England Healthcare Institute. Tele-ICUs: remote management in
intensive care units. Cambridge, Mass: Massachusetts Technology
Collaborative and Health Technology Center; 2007.
17. Thomas EJ, Lucke JF, Wueste L, et al. Association of telemedicine for remote
monitoring of intensive care patients with mortality, complications, and
length of stay. JAMA 2009;302(24):2671–8.
18. Rivers E, Nguyen B, Havstad S, et al, Early Goal-Directed Therapy
Collaborative Group. Early goal-directed therapy in the treatment of severe
sepsis and septic shock. N Engl J Med 2001;345:1368–77.
19. Pronovost PJ, Angus DC, Dorman T, et al. Physician staffing patterns and
clinical outcomes in critically ill patients: a systematic review. JAMA 2002;
288:2151–62.
20. Vespa, P, Miller, CM, Hu, X, et al. Intensive care unit robotic telepresence
facilitates rapid physician response to unstable patients and decreased cost in
neurointensive care. Surg Neurol 2007;67:331–7.
21. Meyer BC, Raman R, Hemmen T, et al. Efficacy of site-independent
telemedicine in the STRokE DOC trial: a randomised, blinded, prospective
study. Lancet Neurol 2008;7:787–95.
416 Section VII—Computers, Engineering, and the Future
22. Breslow MJ, Rosenfeld BA, Doerfler M, et al. Effect of a multiple-site
intensive care unit telemedicine program on clinical and economic
outcomes: an alternative paradigm for intensivist staffing. Crit Care Med
2004;32:31–8.
23. Ranji SR, Auerbach AD, Hurd CJ, et al. Effects of rapid response systems on
clinical outcomes: systematic review and meta-analysis. J Hosp Med 2007;2:
422–32.
24. Rosenfeld BA, Dorman T, Breslow MJ, et al. Intensive care unit telemedicine:
alternate paradigm for providing continuous intensivist care. Crit Care Med
2000;28:3925–31.
25. Winters BD, Pham JC, Hunt EA, et al. Rapid response systems: a systematic
review. Crit Care Med 2007;35:1238–4.
26. Hussain, K. Medication safety and transfusion errors in the ICU and beyond.
Crit Care Clin 2005;21:91–110.
27. Smith CD, Skandalakis JE. Remote presence proctoring by using a
wireless remote-control videoconferencing system. Surg Innov 2005;12(2):
139–43.
28. Agarwal R, Levinson AW, Allaf M, et al. The RoboConsultant: telementoring
and remote presence in the operating room during minimally invasive
urologic surgeries using a novel mobile robotic interface. Urology 2007;
70(5):970–4.
29. Lai F. Robotic telepresence for medical training and education. Stud Health
Technol Inform 2009;142:151–2.
30. Gandsas A, Parekh M, Bleech MM, et al. Robotic telepresence: profit analysis
in reducing length of stay after laparoscopic gastric bypass. J Am Coll Surg
2007;205(1):72–7.
31. Ereso AQ, Garcia P, Tseng E, et al. Usability of robotic platforms for remote
surgical teleproctoring. Telemed J E Health 2009;15(5):445–53.
32. Louw DF, Fielding T, McBeth PB, et al. Surgical robotics: a review and
neurosurgical prototype development. Neurosurgery 2004;54(3):525–37.
33. Brodie J, Eljamel S. Evaluation of a neurosurgical robotic system to make
accurate burr holes. Int J Med Robot 2011;7(1):101–6.
34. Stüer C, Ringel F, Stoffel M, et al. Robotic technology in spine surgery:
current applications and future developments. Acta Neurochir Suppl 2011;
109:241–5.
35. Le Roux P, Das H, Esquenazi S, et al. Robot-assisted microsurgery: feasibility
in a rat microsurgical model. Neurosurgery 2001;48:584–9.
36. Arata J, Tada Y, Kozuka H, et al. Neurosurgical robotic system for brain
tumor removal. Int J Comput Assist Radiol Surg 2011;6(3):375–85.
37. Alric M, Chapelle F, Lemaire JJ, et al. Potential applications of medical and
non-medical robots for neurosurgical applications. Minim Invasive Ther
Allied Technol 2009;18(4):193–216.
38. Pandya S, Motkoski JW, Serrano-Almeida C, et al. Advancing neurosurgery
with image-guided robotics. J Neurosurg 2009;111(6):1141–9.
39. Murayama Y, Irie K, Saguchi T, et al. Robotic digital subtraction angiography
systems within the hybrid operating room. Neurosurgery 2011 Jan 26 [Epub
ahead of print].
40. Audebert HJ, Schenkel J, Heuschmann PU, et al, Telemedic Pilot Project for
Integrative Stroke Care Group. Effects of the implementation of a
telemedical stroke network: the Telemedic Pilot Project for Integrative Stroke
Care (TEMPiS) in Bavaria, Germany. Lancet Neurol 2006;5:742–8.
41. Choi JY, Porche NA, Albright KC, et al. Using telemedicine to facilitate
thrombolytic therapy for patients with acute stroke. Jt Comm J Qual Patient
Saf 2006;32:199–205.
42. Schwab S, Vatankhah B, Kukla C, et al, TEMPiS Group. Long-term
outcome after thrombolysis in telemedical stroke care. Neurology 2007;69:
898–903.
43. Hess DC, Wang S, Gross H, et al. Telestroke: extending stroke expertise into
underserved areas. Lancet Neurol 2006;5:275–8.
44. Lai F. Stroke networks based on robotic telepresence. J Telemed Telecare
2009;15(3):135–6.
45. Jabbour P, Gonzalez LF, Tjoumakaris S, et al. Stroke in the robotic era. World
Neurosurg 2010;73(6):603–4
46. Wong HT, Poon WS, Jacobs P, et al. The comparative impact of video
consultation on emergency neurosurgical referrals. Neurosurgery 2006;59:
607–13.
47. Groves Jr RH, Holcomb Jr BW, Smith ML. Intensive care telemedicine:
evaluating a model for proactive remote monitoring and intervention in the
critical care setting. Stud Health Technol Inform 2008;131:131–46.
48. Benmessaoud C, Kharrazi H, MacDorman KF. Facilitators and barriers to
adopting robotic-assisted surgery: contextualizing the unified theory of
acceptance and use of technology. PLoS One 2011;6(1):e16395.
49. Mavroforou A, Michalodimitrakis E, Hatzitheo-Filou C, et al. Legal and
ethical issues in robotic surgery. Int Angiol 2010;29(1):75–9.

Information Processing, Data
Acquisition, and Storage
Per Enblad, Ian Piper, and Richard O. Sinnott
Introduction
Neurocritical care monitoring generates enormous amounts
of various kinds of data from multiple sources. The informa-
tion obtained serves the purpose of detecting secondary
brain insults and guiding the immediate management of the
individual patient. Stored data also may be used for quality
assurance and to capture the intracenter and intercenter
management variation that can be fundamental to being able
to show positive effects in phase 3 trials. Furthermore, the
collection and storage of monitoring data into databases
fosters research, for example, defining critical threshold levels
for managing raised intracranial pressure (ICP). Such re-
search may increase the understanding of secondary brain
injury mechanisms and define the proper use and value of
new monitoring techniques. Ultimately, this may lead both
to better detection of dangerous secondary brain insults and
to new treatment approaches. If the future expectations of
neurocritical care and multimodality monitoring are to be
realized, it is essential to improve information processing and
to standardize data acquisition storage and analysis of data.
In these aspirations, information technology (IT) plays a
crucial role.
Information Technology
IT is defined by the Information Technology Association of
America (ITAA) as “the study, design, development, imple-
mentation, support or management of computer-based infor-
mation systems, particularly software applications and
computer hardware.” It deals with the use of electronic com-
puters and computer software to convert, store, protect,
process, transmit, and securely retrieve information. Infor-
matics in the neurosciences was first applied to basic science
research but there now are several large databases, such as the
Alzheimer’s Disease Neuroimaging Database (ADNI) or the
Glioma Molecular Diagnostic Initiative (GMDI), that provide
researchers in any location open access to data that can drive
research forward.1 It is clear that the computer science expert
also must be included in the neurointensive care team today
and in the future. The application and development of IT in
the neurocritical care unit (NCCU) is the challenge faced to
further develop neurocritical care. In particular, informatics
in the NCCU can be leveraged to help provide insight into the
© Copyright 2013 Elsevier Inc. All rights reserved.
Chapter
43  
VII
disorders that are treated there and so help design new treat-
ment strategies.2
There is accumulating evidence that demonstrates that IT
can improve patient health care. Bates and Gawande describe
trials in which technology was associated with a reduction in
medication errors and errors related to transitions of care, and
in which technology aided in earlier detection of adverse
events. For example, Kupermann et al.,4 in a randomized con-
trolled trial of technology for early detection of adverse events,
showed a reduction in time to treatment of 11% and a 29%
reduction in the duration of dangerous conditions to patients.
Rosenfield5 conducted a study of IT-based remote monitoring
of a 10-bed intensive care unit (ICU) and reported a reduction
in mortality of greater than 40% and a reduced length of stay
of 30% compared with historical controls.
Critics of this type of research point out that better resolu-
tion of events is of no value unless their direct management
influences clinical outcome. Single-center studies with small
numbers of patients are not suited to answer such questions,
and multicenter randomized controlled trials to validate
IT-driven management are not readily funded nor easily justi-
fied as a research priority. Paradoxically, the patient popula-
tions that may benefit most from better IT-based event
detection, such as the brain injury population, are the most
challenging in which to conduct a controlled management
trial because of continuing intercenter management varia-
tion6,7 fostered in large part by a lack of evidence for any type
of effective therapy.8
Nevertheless, indirect evidence is available and continues to
become available. There are increasing numbers of reports
that indicate the importance of providing specialist neuro-
critical care in the management of traumatic brain injury
(TBI) patients.9-11 For example, the report by Patel9 of 2300
patients treated in nonneurosurgical hospitals showed a 2.15-
fold increase in the odds of death compared with those treated
in a neurosurgical center. These studies do not indicate which
aspects of critical care management are key, but management
(whether surgically or medically focused) aimed at earlier
detection and treatment of adverse events must in part be
responsible. In support of this and despite criticism from
enthusiasts of evidence-based medicine (EMB), neuroin­
tensive care centers with a track record in the aggressive man-
agement of secondary insults continue to report improvements
417
418 Section VII—Computers, Engineering, and the Future
in outcome statistics of patients compared with historical
controls.12
The Brain Monitoring with Information Technology
(BrainIT) group has been a strong proponent for adopting IT
methods for the early detection and management of second-
ary insults in patients with brain injury (www.brainit.org).
Analyses in progress from this group on the minute-by-
minute physiologic data of 200 head-injured patients obtained
from 22 neurointensive care centers across Europe also may
indicate that more complex summary measures, such as the
Pressure-Time Index,13 better relate to clinical outcome than
do simple measures such as the mean, median, or total dura-
tion of insult burden.
Researchers continue to hope for definitive controlled trial
evidence that IT-led management yields improved patient
outcome, but experience to date of funding and conducting
such studies is limited. Perhaps research time and funds are
better spent on conducting trials of new forms of manage-
ment without the need to trial new health care support systems
as well? However, to demonstrate benefit of a therapy requires
that all other aspects of management be homogenous. In
conditions as complex as TBI and subarachnoid hemorrhage,
there remain many opinions as to what constitutes “standard”
or “optimal” management, and without information about
this, even effective therapies may appear to be ineffective in
clinical practice because of variation in care. The authors are
inclined to agree with the sentiments of Socrates as portrayed
in the “letter of dissent”14 when arguing against “Enthusiasti-
cus” that perhaps the loudest supporters of EBM are the hos-
pital accountants keen to keep health care costs down. There
is no question that better monitoring and event detection
technology for health care is needed and that more research
to optimize that technology is also needed, but should their
adoption depend upon large-scale clinical trials? Perhaps now
the questions to focus upon are no longer if but when and no
longer why but how.14a
Data Acquisition
Data acquisition is the sampling of the real world to generate
data that can be manipulated by a computer. Collection of
minute-by-minute physiologic monitoring data is routine in
many NCCUs. However, there are no agreed-upon standards
to define the collection of data. Cerebral perfusion pressure
(CPP = Blood pressure[BP] − ICP) monitoring is a good
example. When CPP values are reported, how does one correct
for policies that differ between centers on the level to which
the BP transducer is zeroed? Tracking changes in bed tilt
(which affects the size of hydrostatic pressure gradients
between the head and heart) is also a significant technical issue
without defined standards. As a result, it is often difficult to
compare even the most common forms of monitoring when
data are presented. Development of standards and guidelines
in this area is an important step for the future design
and conduct of trials of new intensive care monitoring and
treatment methods. In an attempt to standardize how to
approach standardizing correction of hydrostatic pressure
gradients, the BrainIT group has published a BrainIT web
standard (http://www.brain-it.eu/).
With this background in mind, certain individuals working
within the field of TBI research met to discuss the founda-
tion and development of a network to create an IT-based
infrastructure aimed at improving the standards for multi-
center studies of monitoring and managing TBI patients
during their acute stay in intensive care. (These same princi-
ples can be adapted to patients with other disorders admitted
to the NCCU.) It was agreed that a different approach was
needed, one that focused on using IT-based methods to
increase the resolution of data capture and the quality and
validation of data captured. The pervasive nature of the
Internet and extensive use by clinicians of email systems fos-
tered the creation of such a network as an Internet-based
e-Infrastructure: the BrainIT group.15
The BrainIT group works collaboratively to develop stan-
dards to collect and analyze of data from brain-injured patients
toward providing a more efficient infrastructure to assess new
health technology. In several international meetings, the group
has defined a core dataset designed to be collected using
PC-based tools and to provide a common minimal dataset for
all studies, regardless of the underlying research question. This
data definition period was funded as part of a European Com-
munity study (QLGT-2000-00454).16 The meetings brought
together clinical and scientific experts from the domain of TBI
basic research and of multicenter clinical trials, such as the
European Brain Injury Consortium,17 as well as representa-
tives from the medical device and pharmaceutical industries.
A series of meetings and workshops conducted over 1 year
enabled the group to define a minimum set of data that could
be collected from all patients with TBI, which would be useful
in most research projects conducted in this population of
patients. To facilitate discussion, the core dataset was subdi-
vided into four logical groups: (1) demographic and clinical
information, (2) minute-by-minute monitoring information,
(3) intensive care management information, and (4) second-
ary insult treatment information.
From these meetings a consensus dataset was formed, which
includes nine categories:
1. Demographic and one-off clinical data (e.g., pre-
neurosurgical hospital data, first and worst computed
tomography scan data)
2. Daily management data (e.g., daily summary measures of
the use of sedatives, analgesics, vasopressors, fluid input/
output balance)
3. Laboratory data (e.g., blood gas, haematology, biochemis-
try data)
4. Event data (e.g., nursing maneuvers, physiotherapy, medical
procedures such as line insertion, calibrations)
5. Surgical procedures
6. Monitoring data summary (e.g., type and placement loca-
tion of ICP sensors, BP lines)
7. Neuroevent summary (e.g., Glasgow Coma Scale [GCS],
pupil size and reactivity)
8. Targeted therapies (e.g., mannitol given for raised ICP,
pressor given for arterial hypotension)
9. Vital monitoring data (e.g., minute-by-minute BP, ICP,
systemic arterial oxygen saturation [SaO2] collected from
the bedside monitoring)
The full details of the core dataset definition and the collabo-
ration structure of the group can be found in the BrainIT
publication: BrainIT Group: Core Concept and Data Defini-
tion.16 This same approach has been undertaken in the United
States, sponsored by several federal agencies including the
 Section VII—Computers, Engineering, and the Future 419

Ventilation change:
Normal
Mild hyperventilation (4-4.5 kPa)
Severe hyperventilation (<4.0 kPa)
Other (specify)
Sedation/Analgesia
Y/N
Volume expansion
Crystalloids
Colloids (including plasma)
Hypertonic saline
Erythrocytes
Other (specify)
Inotropes
Catecholamines
Other (specify)
Anti-hypertensive therapy ICP
Beta-blockers CCP
Alpha2 antagonists Neurologic deterioration
Other (specify) (eg: change in GCS, pupils)
Anti-pyretic therapy
Fever
Paracetamol
Cooling Hyponatremia
Other (specify) Hypoxia
Cerebral vasoconstriction Hypercapnia
Dihydroergotamine Hypotension
Indomethacin
SjvO2 desaturation
Other (specify)
Osmotic therapy Hyperglycemia
Mannitol Seizures
Glycerol Other (specify)
Other (specify)
Barbiturates
Dose/day
CSF drainage
Closed
Intermittent drainage
Open drainage
Steroids
Y/N
Hypothermia
Y/N
Head elevation
Y/N
Fig. 43.1  Tracking therapies and targets. CCP, Cerebral
Other treatments
Insulin perfusion pressure; CSF, cerebrospinal fluid; GCS,
Other (specify) Glasgow Coma Scale; LCP, intracranial pressure; SjvO2,
jugular venous oxygen saturation.

National Institutes of Health to develop the “Common Data
Elements” (http://www.commondataelements.ninds.nih.gov/
and references 18 and 19). Unique to other dataset definitions,
the BrainIT core dataset defined a special approach to quantify
secondary insult management. This is medical management
therapy given to patients specifically to treat secondary insults
that occur despite the patient’s baseline intensive care medical
management. To distinguish therapy given to patients to treat
secondary insults from those of baseline intensive care, the
authors have devised a coding system that allows specific cat-
egories of therapy to be assigned a “therapy target.” For
example, vasopressors may be given to treat systemic hypoten-
sion or to treat reduced CPP secondary to raised ICP. Choos-
ing an appropriate target for each secondary insult therapy
will enhance the usefulness of the database on medical therapy.
Each therapy must be assigned a target chosen from a drop-
down list. If drugs are given, then one can indicate continuous
infusion if drugs are delivered by a continuous infusion pump
or one can indicate boluses if it is delivered noncontinuously.
Figure 43.1 summarizes the minimum choice of therapy cat-
egories and associated targets for the BrainIT core dataset.
This therapy-tracking model is designed to be implemented
easily in software.
It is one thing to define on paper a dataset and another to
actually collect the data. Although a paper-based feasibility
exercise established some baseline information, the acid test
was still to develop a series of IT-based tools to collect the core
dataset and to prospectively trial the collection of core data
from a number of centers with NCCUs.
A 3-year follow-up EC-funded study (QLGC-2002-00160)
enabled the group to develop IT methods to collect the core
dataset and to assess the feasibility and accuracy to collect this
core dataset from 22 neurointensive care centers. The main
data collection instrument for the episodic nonmonitoring
data was a personal digital assistant (PDA)-based data collec-
tion tool.
Bedside nursing staff entered clinical data on hand held
PDAs that supported the BrainIT core dataset definition
420 Section VII—Computers, Engineering, and the Future
Center
Data collection tool
Data converter tool
Regional
data validater
+ 20% sample
Data validation tool for validation
Fig. 43.2  BrainIT data validation flow of information (black lines, Incoming raw data flow; blue lines, validation requests/data flow).
through a Java Struts-based tool. This allowed the core dataset
to be entered by roaming research nurses using a set of PDA
documents accessed via a series of buttons and tabs. With this
system, indicators were present that showed data documents
that were fully complete, partially complete, or totally incom-
plete. When convenient, the PDA was connected to a docking
station and a client program allowed viewing and saving of
patient data collected. An anonymization routine removed
patient identification elements from the collected data and
labeled the patient data file with a unique BrainIT study code
generated from the BrainIT website. A local database held in
each center linked the anonymized data to local center patient
ID information that was needed during the data checking
stage of the study. The multicenter ethics approval precluded
connection of the PC client system holding the data to any
computer connected to the Internet. Local research nurses
downloaded the anonymized data from the PDA system client
PC onto a memory stick or CD and transferred the data to an
Internet-connected PC for upload of the data to the BrainIT
database via the website data upload page.
Because this was a multicenter study collecting data from
several countries with different languages, a multilanguage
implementation was needed to foster ease of use by local
nursing staff. A training course was held for the data valida-
tion nursing staff in Glasgow on the use of this data collection
instrument, which also included using their medical term and
language expertise to translate all PDA labels and text output
into six European languages (English, French, German,
Spanish, Flemish, and Italian). Data could be entered in the
local language, exported in an XML file format where a table
lookup driven by an XSL transformation converted the data
into a standard English language version.
Data validation research nurse staff were hired on a coun-
try-by-country basis to check samples of the collected data
against gold standard clinical record sources to quantify the
accuracy for collection of the BrainIT core-dataset using the
group IT-based data collection methods. Anonymized data
Glasgow
Common data format
Validation request tool
was uploaded via the BrainIT web-upload services where a
server-side data-converter tool converted data from center-
based formats into BrainIT data format to generate data cat-
egory files that were imported into the BrainIT database (SQL
Server 2005). A validation request tool sampled 20% of the
data sent for each data category and generated a validation
request file listing the timestamps and data items to be
checked by local data validators. Emails were generated to the
data validation staff that contained the validation data
requested documents listing the data items to be checked.
Data validators entered into a data validation tool the
requested data items for checking from source documenta-
tion held in each local center. Validation data then were
uploaded to the BrainIT data coordinating center via the
website and using data validation checking software tools, the
validated data were checked against the data items originally
sent, from which percentage accuracy data was calculated.
Figure 43.2 shows the flow of data between a remote center
and the BrainIT database with the validation procedure
selecting a random sample of 20% of data items uploaded per
data type, which are sent to data validation nurses. The data
validation nurses enter requested data into a PC-based valida-
tion data tool and upload the data to the data manager, who
can then check the accuracy of data for each data category
and estimate an overall error rate.
As part of this validation process, in addition to the cate-
gorical and numeric clinical data being checked for accuracy,
the BrainIT system also assessed the minute-by-minute moni-
toring data. Random samples of monitoring data channels
uploaded (e.g., ICP, SaO2) were selected and validation staff
asked to manually enter the hourly recorded values from the
nurse’s chart (or local gold standard data source) for the first
and last 24-hour periods of bedside monitoring for a given
patient for a given channel. These validation values could then
be compared with a range of summary measures (e.g., mean,
median) from the computer-based monitoring data acquired
from the patient.
 Section VII—Computers, Engineering, and the Future 421
Table 43.1  Percentage Error Rate by Data
Type Class with Common Error Types
Data Class Error Rate (%) Common Errors
Laboratory 2 PaCO2, FiO2 value
Demographic 4 Monitoring on arrival
at neurosurgery,
intubation on arrival
at neurosurgery
Neuro 5 Pupil Size, GCS versus
observations (code 1 versus
unknown)
Monitoring 5 ICP type, ICP location
summary
Daily 5 Infusion type (bolus vs
management infusion or both),
summary drug number (1, >1)
Targeted 6 Nonstandard target,
therapy no target specified
Surgeries 34 ICP monitor; surgery
for skull fracture or
mass lesion
FiO2, Fraction of inspired oxygen; GCS, Glasgow Coma Scale; ICP, intracranial
pressure; pCO2, partial pressure of carbon dioxide.
By late 2008, 384 TBI patients’ core data had been collected
from 22 European neurointensive centres. The first 200
patients, data were cleaned and validation analyses conducted.
In total, 19,461 comparisons were made between collected
data elements and source documentation data.20 The number
of comparisons made per data category was in proportion to
the size of the data received for that category with the largest
number checked in laboratory data (5667) and the least in the
surgery data (567). Table 43.1 summarizes error rates by data
class. Error rates were generally less than or equal to 6%, with
the exception being the surgery data class where a high error
rate of 34% was found.
The proportion of surgery errors was due primarily to the
classification system used to simplify and thereby reduce the
burden of data entry. For example, through discussions with
local nursing and data validation staff it was found that there
was particular confusion over when to record ICP sensor
placement and a craniectomy or craniotomy when the proce-
dures occurred together as the primary surgical procedure. As
such, confusion over coding these two procedures by both the
local research nurse and the data validation nurse accounted
for the majority of errors in this data category.
This study conducted by the BrainIT group is one of only
a few projects to attempt to prospectively assess the data
capture error rate within an academic environment. It was
shown that it is feasible to collect the BrainIT dataset from
multiple centers in an international setting with human-
intensive (research nurse) IT-based methods and the accuracy
of the data collected is greater than or equal to 94%, with the
exception of the surgery data type implementation. One of the
challenges to implementing the system over multiple sites is
data security. In the short term this is overcome in part by
allowing voluntary entry of data by investigators at each site.21
It was also shown that computer-collected minute-by-minute
vital signs data, summarized as end-hour averages, correlate
well with nursing chart end-hour recordings. This allows the
end-hour averaged computer records to be used in database
analyses assessing nurses’ chart-recorded detection of events
with computer-based sampling. These validation results cal-
culated on a subset of patients provides an estimate of the data
quality for future analyses on the full patient cohort of 350
patients collected as part of the EC-funded study, which was
conducted over the same time period by the same staff using
the same data methods. Clearly, though, future data collection
projects will generate datasets under differing data collection
conditions and will require a separate validation stage if con-
fidence is to be maintained in the level of data accuracy.
However, the costs of maintaining such a data validation
network are prohibitively high. To maintain a full-time data
validation nurse within each participating country costs in
excess of 1 million Euros per year. Such large running costs
for an academic network are not sustainable in the long term
and a more cost-effective solution for data validation must
be found.
Information Processing
Practical information processing can be described as a cycle,
wherein data that may have no inherent meaning to the
observer is converted into information that does have meaning
for the observer. In neurointensive care, improved informa-
tion processing is now a necessity. Furthermore, with advanced
bioinformatics that use analytic techniques, the complex
interrelationship between physiologic variables can be
explored. In turn, this may predict future events in the NCCU
rather than just outcome; that is, it can facilitate targeted
management.22 With increasing use of computer technology
that generates large quantities of high-resolution datasets,
interpretation of the data is now the limiting step. Three
examples of research information processing approaches that
may address some of the problems with interpretation of large
amounts of neurocritical care data follow.
Mathematical Models
Cerebral autoregulation is the process by which cerebral blood
flow (CBF) is maintained constant over a specific CPP range.
Cerebral autoregulation can be an important factor in clinical
decisions about treatment of the patient’s injuries; if it is mis-
diagnosed, undesirable effects, such as ischemia or hypoxia,
could occur from inappropriate treatment. Autoregulation is
typically classed as either static or dynamic, depending on the
time frame over which it is assessed.23
Tests of static autoregulation tend to be based on the
manipulation of BP by pharmacologic agents with measures
of volume CBF at two different BP levels with measurements
occurring many minutes apart. The current gold standard
methods to assess static autoregulation involve the use of
ultrasound or invasive isotope infusion, or require the patient,
who is critically ill, to be taken out of the NCCU to a posi-
tron emission tomography/single photon emission computed
tomography (PET/SPECT) scanner for blood-flow measure-
ments. For patients in the ICU, this approach can be used at
most only once or twice a day.
Dynamic autoregulation assesses the short-term response
of the cerebrovascular bed over 10 to 20 seconds to perturba-
tions in BP caused by leg cuff deflation or brief unilateral
422 Section VII—Computers, Engineering, and the Future
carotid compression. There is no accepted standard method
for dynamic autoregulation testing, although there is a con-
sensus that dynamic testing of autoregulation is more clini-
cally practical than static autoregulation testing.23
Ideally, what is needed is a method that can assess the status
of autoregulation on a continuous basis with minimal distur-
bance to the patient’s normal clinical management or moni-
toring. This is often difficult to achieve with techniques that
measure CBF or velocity, because they typically require spe-
cialized monitoring and expert measurement. However, most
patients have minute-by-minute monitoring of BP and ICP as
standard, and a number of groups have been studying the
time-dependent variation in BP and ICP (cerebrovascular
pressure transmission) with a view to revealing relationships
between them and highlighting patterns that may indicate
abnormal autoregulation.
Mathematical modeling of the autoregulatory process is
one approach that has been studied. Several models have
already been produced using different analysis techniques
including linear regression, spectral analysis, transfer function
analysis, cross correlation function, or impulse response anal-
ysis.24 Some of these approaches have had their performance
assessed in a comparative study, but these methods are based
upon single or pulse train analysis that requires specialized
monitoring with high data-sampling rates.25 To be practical in
the clinical environment, models should be developed and
assessed from data with minute-by-minute BP and ICP sam-
pling rates that are a more common data capture rate in many
ICUs.
There are several models of cerebral autoregulation; the
Ursino Lodi model is an example of a mathematical modeling
approach.26 Further work by the BrainIT group20 has reworked
the Ursino Lodi autoregulation model to derive an index of
autoregulation (G) that allows comparison of autoregulatory
models and provides a standard measure of autoregulation
obtained from animal models. This modified Ursino index
also correlates with the Bouma index27 of static autoregula-
tion.21 The application of this type of autoregulatory modeling
in TBI datasets is illustrated in Figure 43.3, which shows a
Outcome Group Comparison
350
300
* (p = 0.03)
250
Variance of G
200
150
100
50
25
Bad Good
Fig. 43.3  Box plot representation of variance of G that compares
outcome groups.
good correlation of the reworked Ursino Lodi model output
G with clinical outcome over the later stages of patient ICU
stay. Results from a pilot study in 12 patients showed that there
was a significantly greater mean variance in G in patients with
bad outcome than good outcome.21
Such modeling approaches based upon mathematical
models rather than data driven models can be inherently
more informative, because their underlying mechanism can
be studied and adapted as knowledge of the cerebrovascular
system develops. Modeling approaches that rely solely on the
relationship between two or more physiologic parameters may
not be so easily extended.
Fractal Analysis
During the past two decades, analysis techniques have moved
from classic stochastic process analysis using basic statistics to
more nonlinear systems or chaos theory–based approaches
that look at patterns in the variability of the time series. Physi-
ologic time series exhibit complex multifractal properties,28 so
changes and alterations in the underlying physiologic pro-
cesses can be detected when a classification and analysis based
on this nonlinear and chaotic nature of the time series is
designed. The main idea used in the initial analysis and fractal
characterization of the waveform leverages the relationship
between the mathematical properties of a wavelet transforma-
tion and a signal’s localized fractal nature.
Wavelets29 can be thought of as a time-frequency analysis
technique analogous to a Fourier transformation but without
the inherent frequency-related problems of Fourier analysis.
As an example of this type of analysis, ICP waveform signals
were randomly sampled and cleaned from the BrainIT dataset,
and a moving window approach was created to allow the
repeated application of the mean Holder function (which is a
measure of the fractal nature of a signal) along the time course
of the original signal (Fig. 43.4). Although only a sample
analysis, a clear downward trend in the Holder component
that indicates a decrease in the chaotic (fractal) nature of the
signal is observed before increases in ICP instability occur.21
It is likely that further study of the fractal nature of physiologic
time series generated from patients in the NCCU may provide
new approaches to data interpretation.
Neural Networks
Artificial Neural Network
An artificial neural network (ANN) is a digital simulation of
a brain model, and can be considered a pattern classifier, that
relates input patterns to output patterns. This is a very robust
tool for classification of multiple factors and for multivariate
analysis that may facilitate recognition of patterns or clusters
of data, including in ICU studies that may otherwise be dif-
ficult to recognise.30 In an ANN, individual “neurons” are
interconnected from an input layer (the inputs to which are
the patient parameters) through one or more hidden layers to
an output layer (Fig. 43.5). Each neuron within the ANN
performs some numerical operation on its input connections
to generate the values for its output parameters. This is typi-
cally accomplished by assigning (initially random) weights to
all of the inputs and outputs. All but the simplest ANNs
require some amount of training for the output patterns to
become relevant to the input patterns. This training usually
 Section VII—Computers, Engineering, and the Future 423

ICP Wave Form

200

150
ICP (mm Hg)

100

50

0
0 2000 4000 6000 8000 10000 12000 14000
A Time (min)

Trend of Mean Holder Exponent

1.5
Mean holder exponent

1.0

0.5
Fig. 43.4  Time series plots of (A) raw intracranial
0.0 pressure (ICP) data and (B) the mean Holder
exponent (as a measure of the fractal nature of
–0.5 the signal). As ICP becomes more unstable, the
Holder exponent drops and shows a clear trend
0 2000 4000 6000 8000 10000 12000 14000
many hours before the occurrence of the ICP
B Time (min) instability.

Input layer
Hidden layer(s)
Fig. 43.5  Schematic representation of an artificial neural network.
requires that the weights on the network be adjusted after each
training event to optimize the output and so depends on the
existence of a (typically large) dataset that matches the inputs
to the required outputs. When training, great care must be
taken to ensure that the network is not overtrained, which
typically results in the network being highly accurate for only
a small subset of the input pattern set, yet it must be ensured
that the network is sufficiently trained to offer realistic results
for previously unseen input patterns.
Bayesian Approach to Neural
Network Training
The power of neural networks as general nonlinear models
can lead to problems. Overly complex models may be derived
that fit the limited training set perfectly but that are highly
inaccurate when presented with new data. This problem is
Output layer referred to as “overfitting.” Also, because of the complexity of
the functions computed, the techniques used to compute con-
fidence intervals for simpler models (e.g., linear or polynomial
models) cannot be used for neural networks. In many domains,
and clearly in any domain that includes medical treatment
(such as is presented here), the confidence assigned to a pre-
diction is of critical importance.
An appropriate approach to resolve these problems applies
Bayesian principles to the problem of neural network training.
Rather than generating a single network to model the training
set, many models (typically hundreds) are generated in a
Monte Carlo Markov Chain (MCMC). The series of models
generated by the MCMC has the property that it conforms to
the Bayesian posterior distribution on the space of possible
models conditioned on the training data. Thus to generate a
prediction, the whole set of models derived during training is
used to generate a corresponding distribution of outputs.
Some summary measure of the Bayesian Artificial Neural
Network (BANN) output such as the median or the mode of
this distribution can be used as the prediction, and the spread
of the output values can be used to assign confidence intervals
for this prediction, derived on the basis of the Bayesian prin-
ciples of probability. The problem of overfitting is addressed,
because in the process of combining predictions from hun-
dreds of models, the divergent predictions of the models that
overfit the training data average out.
The confidence intervals generated in such a model are
superior to those normally used for simpler models such as
linear and polynomial models. As an example, when a linear
model is fit, confidence intervals are given based on the
assumption that the data are in fact generated by a linear
424 Section VII—Computers, Engineering, and the Future

Edinburgh Uppsala

1.0 1.0
0.9 0.9

Probability of favorable outcome

Probability of favorable outcome

0.8 0.8
0.7 0.7
0.6 0.6
0.5 0.5
0.4 0.4
0.3 0.3
0.2 0.2
0.1 0.1
0.0 0.0

0 20 40 60 80 100 0 20 40 60 80 100
% Monitoring time with CPP <60 % Monitoring time with CPP <60
Fig. 43.6  Bayesian Artificial Neural Network–generated probability distribution plots for the mean (±95% confidence limits) for the
likelihood of a favorable clinical outcome with increasing time spent with cerebral perfusion pressure (CPP) less than 60 mm Hg. Data
are presented from two centers. Note the difference in predicted outcome with CPP management; this suggests that these two patient populations
have different underlying pathophysiologic mechanisms or clinical characteristics.

process. However, this assumption particularly in complex data sets often means that research data entry is duplicated
medical datasets is not justified, and as a result leads to unre- and often requires nursing staff to enter the same data on
alistically optimistic confidence intervals being assigned to more than one data-entry system. If one could directly access
model predictions. The confidence intervals derived by the hospital-based data remotely (and securely) for research pur-
Bayesian neural network approach realistically reflect uncer- poses, then considerable savings could be made to implement
tainty about both the form and the parameter settings of the and finance many clinical research projects.
model, as well as the variance of the training data. Grid technology provides software methods (middleware)
Bayesian networks have been applied to critical care that can provide distributed access and collection (federation)
research.31 This type of BANN approach also has found some of data to occur. However the medical domain is especially
success in neurocritical care applications. For example Howells demanding in its adherence to ethics and information gover-
et al.32 applied a BANN model to show how patient outcome nance, which in turn requires fine-grained security. Many
probabilities change as a function of physiologic variables, IT-based data storage solutions already exist that cross primary
such as the percentage of monitoring time that CPP is less care, secondary care, and a range of specialized resources such
than 60 mm Hg. Figure 43.6 shows graphically the output of as disease registries in the clinical domain. Many of these
such a model with the probability of a favorable outcome solutions have been developed in isolation and have different
(Glasgow Outcome Scale [GOS] 4 or 5) as a function of the data security policies, or in many cases, no security policies
percentage of monitoring time that a patient’s CPP was less other than protection at the hospital firewall level. This
than 60 mm Hg. In the patients in Edinburgh, longer dura- deficiency in security policies is magnified when crossing
tions of CPP insult were associated with death. In contrast, in national boundaries. Dealing with such heterogeneity from
those in Uppsala, longer durations of CPP insult suggested the data perspective has been one of the drivers behind grid
favorable outcome. Using this same approach, there are several technologies.
commercially available systems that offer prediction of forth- Given the sensitivity of data, the establishment of security
coming adverse events, such as the BioSign device (Fig. 43.7), policies and their enforcement is essential in the clinical
which derives a BioSign index value from five parameters domain. A cornerstone of these solutions is to ensure fine-
(heart rate, respiration rate, body temperature, oxygen satura- grained security. However, it is an unavoidable fact that the
tion, and BP), from which subsequent cardiovascular instabil- specific privileges required in a particular trial or study will
ity can be predicted.33 not be known when the system is first created. Similarly a
The BrainIT network approach to collaborative neurocriti- doctor in one hospital may have privileges to access various
cal care research used the accepted model of upload of local systems in that hospital, but these do not transfer directly
center data to a central data repository or database. To do so when this doctor attends a different hospital. Therefore
required hiring of research nurse staff employed specifically systems capable of adding and removing resources or privi-
to validate uploaded data. The costs of maintaining such a leges “on the fly” are necessary, wherein the corresponding
data validation network are expensive and not sustainable in allocation of privileges can be added or removed depending
the long term and so a more cost-effective solution for data on the needs of different trials or health care systems. In grid
validation must be found.21 Much of the data collected for parlance, the framework by which such rules and regulations
research purposes already is collected as part of the routine on the resources and the users that may access them and
clinical management. However, in many centers the lack of under what conditions is conceptually called a virtual organi-
transparent mechanism that allows access to hospital-based zation (VO).
 Section VII—Computers, Engineering, and the Future 425
Fig. 43.7  BioSign technology to help predict cardiovascular instability.
The basic models suggested for the majority of grid-based
security systems can be broadly broken down into the “AAA”
categories:
 Authentication. Establish the identity of the person who
requests access to a resource.

Authorization. Having established identity, establish and
enforce what that person is allowed to do on a given
resource.

Accountability. Establish the activities, and time of activi-
ties, of a particular person on that resource (or resources)
so that they cannot subsequently deny potential misuse
later on (nonrepudiation).
The following example illustrates the interplay of role-based
access to federated resources. A particular BrainIT trial is
undertaken with the Southern General Hospital in Glasgow.
When users attempt to access the BrainIT portal, they are
redirected to their home institutions when they are asked to
authenticate. After authenticating, the attributes needed to
access the BrainIT portal for authorized access to the federated
clinical data are accessible. The portal then uses these attri-
butes to configure the contents of the portal; that is, the users
are restricted to see and do what their roles dictate. Figure 43.8
shows two different users have logged in to the system, one
with the brainIT-investigator role (right side) and one with the
brainIT-nurse role (left side). The investigator role has more
privileges in this particular trial and hence is allowed access to
a wider range of information within the context of that trial
(e.g., GCS, ICP, glucose levels, and patient identification). The
nurse role on the other hand is restricted to a subset of non–
patient-identifying information.
The systems described show the proof of concept upon
which clinical data can be accessed and used within a secure
framework. A more sophisticated grid infrastructure (AVERT-
IT project; http://www.avert-it.org) is under design.34 In this
virtual environment near real-time information is captured
from six NCCUs across Europe (located in Barcelona, Glasgow,
Heidelberg, Monza, Uppsala, and Vilnius) to explore the
potential prediction of hypotensive events. To support this
work, researchers are exploring how hypotensive events might
be predicted using a novel BANN, to be trained against an
existing BrainIT dataset before a prospective clinical trial is
undertaken to demonstrate the effectiveness of the AVERT-IT
project concept (AVERT-IT portal). Figure 43.9 is a block
diagram that illustrates the virtual organization structure.
With the AVERT-IT systems, models are proposed based
upon pushed data transfer only; that is, the clinical data pro-
vider firewalls will reject all incoming data requests, but will
push data out of the firewalls into demilitarized zones set up
as a buffer between the grid re++search domain and the
domain of live health care provision. One significant challenge
that remains to be solved in the AVERT-IT system and within
the wider e-research community is with real-time or near real-
time data. Monitoring data that needs to be streamed to
support the Bayesian adverse event prediction algorithms
offers new challenges that have hitherto not been addressed.
Conclusion
The future of information processing, data acquisition, and
storage within the neurocritical care environment must be a
collaborative approach. Gone are the days when neurosurgeons
or intensivists were able to follow and implement technical
developments outside their own specific area of expertise. Col-
laborations between clinicians and information technologists
are key, but often too large a knowledge gap must be bridged
426 Section VII—Computers, Engineering, and the Future

Fig. 43.8  Role-based access to portal and restrictions to data access.

AVERT IT Existing
data
HypoPredict Bedside Bedside Bedside
collection
web app Local monitor monitor monitor
HypoPredict
engine
Patient
Web Pages
WebServer for
(for both HypoPredict HypoPredict
and GRID system)
Web Pages Physiologic
for
GRID GRID data
secure
access
services
Event/treatment
Local data
HypoPredict
database

GRID Demographic
role DB data
Existing hospital IT
GRID
Secure internet
access to/from This software runs on a infrastrucure
secure
clinical trail
access computer supplied by
management
Web app services the Neurosurgery center
to an AVERT-IT spec
Neurosurgery ICU
Key: = C3Amulet, = GRID

Fig. 43.9  Block diagram that shows the virtual organization design where hospital datasets are transferred to a local grid repository in
each neurocritical care center. Secure grid access services support role-based access to the repository with monitoring of all six centers’ data remotely
by a data manager via grid-based web services.
 Section VII—Computers, Engineering, and the Future 427
to make such collaborations functional. Many hospitals now
employ clinical physicists or scientists who are seconded into
areas of specialization, including neurosurgery and intensive
care. It may well be that such staff could help bridge this gap.
Clinical scientists are the natural interface between clinicians
and their data. Often they are responsible for assisting clinical
staff to develop and maintain data collection and analysis
tools and, as such, they understand better than many the details
and limitations of the data and the methods to collect and
condition that data. Clinical scientists are not IT experts, but
given their numerate and technical background, are better
able to act as a communication channel between the informat-
ics experts and clinicians. Whatever the formula is, the modern
NCCU is best based upon functional collaborations between
domain experts who all pull in the same direction. And that
direction is to provide a sound evidence base to help improve
patient monitoring and management.
References
1. Riddick G, Fine HA. Integration and analysis of genome-scale data from
gliomas. Nat Rev Neurol 2011;7(8):439–50.
2. Hemphill JC, Andrews P, De Georgia M. Medscape. Multimodal monitoring
and neurocritical care bioinformatics. Nat Rev Neurol 2011;7(8):451–60.
3. Bates D, Gawande A. Improving safety with information technology. N Engl
J Med 2003;348:2526–34.
4. Kuperman GJ, Teich JM, Tanasijevi MJ, et al. Improving response to critical
laboratory values with automation: results of a randomised controlled trial.
J Am Med Inform Assoc 1999;6:512–22.
5. Rosenfield BA, Dorman T, Breslow MJ, et al. Intensive care unit
telemedicine; alternative paradigm for providing continuous intensive care.
Crit Care Med 2000;28:3925–31.
6. Stochetti N, Penny K, Dearden M, et al. Intensive care management of
head-injured patients in Europe: a survey from the European Brain Injury
Consortium. Intensiv Care Med 2001;27:400–6.
7. Bulger E, Nathens A, Rivara F, et al. Management of severe head injury:
institutional variations in care and effect on outcome. Crit Care Med
2002;30:1870–6.
8. Narayan R, Michel M, Clinical Trials in Head Injury Study Group. Clinical
trials in head injury. J Neurotrauma 2002;19:503–57.
9. Patel H, Bouamra O, Woodford M, et al. Trends in head injury outcome
from 1989 to 2003 and the effect of neurosurgical care: an observational
study. Lancet 2005;366:1538–44.
10. Varelas PM, Eastwood D, Yun H, et al. Impact of a neurointensivist on
outcomes in patients with head trauma treated in a neurosciences intensive
care unit. J Neurosurg 2006;104:713–9.
11. Clayton T, Nielson R, Manara A. Reduction in mortality from severe head
injury following introduction of a protocol for intensive care management.
Br J Anaesth 2004;93:761–7.
12. Elf K, Nilsson P, Enblad P. Outcome after traumatic brain injury improved
by an organised secondary insult program and standardised care. Crit Care
Med 2002;30:2129–34.
13. Chambers I, Jones P, Milly Lo T, et al. Critical thresholds of intracranial
pressure and cerebral perfusion pressure related to age in paediatric head
injury. J Neurol Neurosurg Psychiatry 2006;77:234–40.
14. Smith D. Evidence based medicine: Socratic dissent. Br Med J
1995;310:1126–7.
14a.  Piper I. Not if but when; no longer why but how. Critical Care
2007:11(1):117.
15. Brain Monitoring with Information Technology (BrainIT) Research Group:
http://www.brainit.org.
16. Piper I, Citerio C, Chambers I, et al. The BrainIT Group: concept and core
dataset definition. Acta Neurochir 2003;145:615–29.
17. European Brain Injury Consortium – EBIC: http:www.ebic.nl.
18. Maas AI, Harrison-Felix CL, Menon D, et al. Common data elements for
traumatic brain injury: recommendations from the interagency working
group on demographics and clinical assessment. Arch Phys Med Rehabil
2010;91(11):1641–9.
19. Manley GT, Diaz-Arrastia R, Brophy M, et al. Common data elements for
traumatic brain injury: recommendations from the biospecimens and
biomarkers working group. Arch Phys Med Rehabil 2010;91(11):1667–72.
20. Shaw M, Piper I, Chambers I, et al. on behalf of the BrainIT Group. The
Brain Monitoring with Information Technology (BrainIT) collaborative
network: data validation results. Acta Neurochir Suppl 2008;102:217–21.
21. Piper I, Chambers I, Citerio G, et al. The brain monitoring with Information
Technology (BrainIT) collaborative network: EC feasibility study results and
future direction. Acta Neurochir (Wien) 2010;152:1859–71.
22. Buchman TG. The digital patient: predicting physiologic dynamics with
mathematical models. Crit Care Med 2009;37:1167–8.
23. Tieks F, Lam AM, Aaslid R, et al. Comparison of static and dynamic cerebral
autoregulation measurements. Stroke 1995;26:1014–9.
24. Panerai RB. Assessment of cerebral pressure autoregulation in humans—
a review of measurement methods. Physiol Meas 1998;19:305–38.
25. Panerai RB. Neural network modelling of dynamic cerebral autoregulation:
assessment and comparison with established methods. Med Eng Phys
2004;26(1):43–52.
26. Ursino M, Lodi CA. A simple mathematical model of the interaction
between intracranial pressure and cerebral hemodynamics. J Appl Physiol
1997;82(4):1256–69.
27. Bouma G, Muizelaar P, Bandow K, et al. Blood pressure and intracranial
pressure-volume dynamics in severe head injury: relationship with cerebral
blood flow. J Neurosurg 1992;77:15–9.
28. Bassigthwaighte JB, Liebowitch LS, West BJ. Fractal physiology. Oxford:
OxfordUniversity Press; 1994.
29. Muzy JF, Bacry E, Arneodo A. Wavelets and multifractal formalism for
singular signals: application to turbulence data. Phys Rev Lett
1991;67(25):3515–8.
30. Cohen MJ, Grossman AD, Morabido D, et al. Identification of complex
metabolic states in critically injured patients using bioinformatic cluster
analysis. Crit Care 2010;14:R10.
31. Peelen L, de Keizer NF, Jonge E, et al. Using hierarchical dynamic Bayesian
networks to investigate dynamics of organ failure in patients in the intensive
care unit.
J Biomed Inform 2010;43:273–86.
32. Howells T, Elf K, Jones P, et al. Pressure reactivity as a guide in the treatment
of cerebral perfusion pressure in patients with brain trauma. J Neurosurg
2005;102:311–7.
33. BioSign Technology. http://www.isis-innovation.com/documents/Biosign.pdf.
34. Stell A, Sinnott R, Jiang J, et al. Federating distributed clinical data for the
prediction of adverse hypotensive events. Philos Transact A Math Phys Eng
Sci 2009;367(1898):2679–90.
VII
Chapter
44  
VISICU and the eICU Program
Pamela J. Amelung and Martin E. Doerfler
Introduction
Neurocritical care originated from the principles of respira-
tory intensive care established during the poliomyelitis epi-
demic of the mid-twentieth century, evolving to a discipline
that provides comprehensive medical and specialized neuro-
logic support for patients with life-threatening neurologic
diseases.1 The early neurologic critical care unit (NCCU) ini-
tially focused on the management of postoperative neurosur-
gical patients. Today neurocritical care has expanded to
include the management of patients with intracranial hemor-
rhage, traumatic brain injury (TBI), stroke, complex seizures,
elevated intracranial pressure (ICP), and the systemic compli-
cations of brain injury. Rapid advances in neurologic and
neurosurgical interventions and interventional neuroradiol-
ogy in the last two decades, and the appreciation that a brain
affected by a primary injury is influenced by systemic altera-
tions that affect its function (secondary injury), have led to
the development of specialized NCCUs and specialized train-
ing for physicians and nurses, all focused on treating life-
threatening neurologic diseases.2
Today, neurocritical care is one of the newest and fastest
growing specialties in medicine, and an increasing number of
institutions throughout the world now have dedicated neuro-
critical care units. Additionally, there are a number of hospi-
tals without dedicated neurocritical care units but with
neurointensivists acting as consultants.3
Development of VISICU and
the eICU Program
The Leapfrog Initiative
As neurocritical care has come to the forefront in the care of
seriously ill or injured neurologic patients, so has the concept
of specialty critical care medicine in general become more
widely accepted over the past two decades. In its 1999 report,
“To Err Is Human: Building a Safer Health System,” the
National Academy of Sciences’ Institute of Medicine (IOM)
estimated at least 44,000 and as many as 98,000 people die
each year in hospitals from preventable medical errors.4 In a
subsequent report, “Crossing the Quality Chasm: A New
Health System for the 21st Century,” published in 2001, the
IOM called for a redesign of the U.S. health care system to
help achieve the goals stated in the first report and stated: “You
can’t get there from here.”5 This means that the problem of
quality improvement will not be solved by doing more of what
428
has been done in the past, or what is being done currently, but
rather by using new tools to help solve core problems. The
report also called for use of information technology to help
drive the quality improvements. In response to the IOM
reports, The Leapfrog Group, an organization of Fortune 500
companies and large private and public health care purchas-
ers, identified four hospital quality and safety practices that
could help prevent errors and save lives, including intensive
care units (ICUs) staffed by intensivists 24 hours a day and all
week long. Estimates suggest that 50,000 lives and $4.3 billion
may be saved each year in the United States through 24-hour
intensivist staffing.6,7
Too few ICU patients are cared for by critical care special-
ists; not every hospital has the patient volume or the financial
resources to hire dedicated intensivists. In addition there is a
shortage of trained intensivists.8,9 About 6000 intensivists are
estimated to be in practice in the United States today, and less
than 15% of all ICU beds have dedicated intensivists. There
are approximately 6000 ICUs in the United States, which rep-
resents 5% to 10% of hospital beds. On a daily basis, these
ICUs care for approximately 55,000 patients. Only one third
of these patients are treated by an intensivist, either as the
primary physician or as a consultant.10 In addition, there is a
current nursing shortage and aging of the most experienced
critical care nurses. The first of the “baby boomers” are reach-
ing retirement age at a time when technology provides for
unprecedented life expectancy and recovery from illness and
injury. However the relative lack of intensivists and critical
care nurses is not likely to improve.
ICU Outcomes with an Intensivist Model
Many variables determine outcomes in critically ill patients,
including the primary diagnosis that leads to admission,
patient characteristics, the evolution of secondary disorders,
and the care that these patients receive. There is a wealth of
data to suggest that intensivists, having been trained specifi-
cally in the care of the critically ill patient, are better at man-
agement of comorbidities and avoidance of complications,
and so can improve outcomes in ICU patients. This has stimu-
lated considerable interest in ICU performance. Patients
managed by a dedicated critical care service in a single hospital
surgical ICU have a shorter length of stay (LOS), fewer com-
plications, and lower hospital charges than patients cared for
by general surgeons and house staff.11 In 1999, a study by
Pronovost et al12 found that daily rounds by an intensivist
were associated with improved outcomes in patients who
© Copyright 2013 Elsevier Inc. All rights reserved.
 Section VII—Computers, Engineering, and the Future 429
underwent abdominal aortic surgery. Others have demon-
strated reduction in mortality,13 including a systematic litera-
ture review in which 16 of 17 studies showed a decrease in
hospital mortality with high-intensity ICU staffing (the inten-
sivist was the primary attending or there was a mandatory
critical care consultation for all ICU patients).10 However,
Levy et al14 have debated the benefit of the intensivist model
and suggest that intensive care provided by critical care spe-
cialists does not improve outcomes and may be detrimental
to patients. In this study, hospital mortality was higher for
patients managed by critical care physicians, even after adjust-
ment for severity of illness and other factors that may have
influenced the probability of being cared for by a critical care
physician. However, this study is observational in nature,
making cause-and-effect conclusions difficult. Furthermore,
heterogeneity beyond that accounted for by applying adjust-
ments may limit result interpretation.
Most studies identify ICU physician staffing as an opportu-
nity to reduce in-hospital mortality. Assuming 3.5 million
patients are admitted each year to ICUs in the United States
and a 40% reduction in mortality with dedicated intensivists,
it is estimated that 54,000 lives could be saved with a full
intensivist model as described by the Leapfrog Group.
However, only 10% to 15% of U.S. hospitals have such a
program in place, in large part because of a shortage of inten-
sivists. This shortage of intensivists is expected to worsen in
coming years. The technology tools of the eICU program,
including virtual presence through audiovisual technology,
provides an alternative means to bring intensivists expertise
to hospitals that lack sufficient numbers of intensivists.
Telemedicine to Leverage Scarce Resources
Grundy and colleagues15,16 first introduced telemedicine as a
tool for critical care in the late 1970s. Interactive television was
used to provide consultation with university-based critical
care physicians for patients in the ICU of a 100-bed hospital
that lacked physicians with critical care expertise. During an
18-month period, 1548 telemedicine “visits” were made to 395
patients. The process was thought to be a reliable method to
extend the availability of specialist expertise and to be benefi-
cial to patient care. However it was another 15 years before the
concept of telemedicine in critical care reemerged, albeit in a
very different form. In 2000, Rosenfeld17 reported that intro-
duction of around-the-clock intensivist oversight through
telemedicine produced reductions in mortality and LOS
similar to those reported by implementation of an intensivist
program.17 Subsequently, Breslow and Rosenfeld founded
VISICU, a start-up company that provided the first con­
tinuous, remote, centralized, computer-assisted intensivist
program for critical care. Sentara Healthcare in southeast Vir-
ginia was the first to implement this program in 2000. After
the first year, hospital mortality decreased by 27%, ICU LOS
by 17%, and hospital LOS by 13% compared with the previous
year.18 Variable costs decreased and hospital revenue increased
because of increased patient throughput. These improvements
have been sustained over subsequent years.
What is telemedicine? Because onsite intensivist staffing is
associated with reduced mortality,13,19,20 telemedicine in the
ICU has evolved in part to address limited clinician supply
through a technologic solution. ICU telemedicine can come
in several forms that range from consultations regarding
individual patients21 to continuous offsite monitoring for the
whole ICU,22 but all combine audiovisual technology (video-
conferencing technology and telemetry) and electronic
medical records to provide critical care from a remote site. It
is estimated now that about 10% of the total ICU beds in the
United States use this technology.23 The telemedicine coverage
may come in several forms, including (1) type of coverage
(proactive, reactive, or mixed), (2) structure (a single consult-
ing “hub” that covers multiple ICU “spokes,” a “parent” hos-
pital that covers another hospital in its system, or a network
of clinicians who monitor from multiple sites), and (3) hours
of operation (24 hours or night only). In part, the telemedi-
cine setup depends on physical ICU staffing (open or closed),
intensivist availability at physical ICUs, number and type of
ICUs, and hospital resources or affiliation with a tele-ICU
vendor.
Telemedicine and General ICU Outcome
Several factors may influence how a telemedicine program
affects clinical outcome: (1) integration of the information
systems of the tele-ICU and the monitored units, (2) physician
acceptance, and (3) how the telemedicine system is used. In
addition, other delivery systems and work-process innova-
tions rather than technology alone can impact outcome.24
Several studies have examined how telemedicine in general
ICU care may influence outcome.19,22,25-30 Some suggest a
benefit in the sickest patients,22 because these may be the
patients most likely to have unexpected events. In 2011, Young
et al30 published a systematic review and metaanalysis that
described the clinical impact of ICU telemedicine. They iden-
tified 13 studies, only seven of which were published in peer-
reviewed literature that examined the association between
ICU telemedicine and clinical outcomes. Each study used a
before-and-after design, but severity of illness and case-mix
was not always adjusted for. On metaanalysis there was an
association between reduced ICU mortality and LOS and a
telemedicine ICU. However, there was no association between
admission to an ICU under a telemedicine model and
in-hospital mortality or hospital LOS. The confidence inter-
vals were wide, and so a clinically significant effect could not
be excluded. In addition, most of the studies included in the
analysis used telemedicine at night to enhance patient safety
in ICUs that already had high-quality daytime staffing. In
some of the studies, authority was delegated to the telemedi-
cine physicians only in life-threatening events. The value of
telemedicine may not be so much in improved safety but
rather in the ability to provide evidence-based care on a
24-hour basis in a proactive fashion.31-33 Furthermore the
benefit may be greatest in smaller ICUs with limited staffing
where the telemedicine intensivists can interact with bedside
staff to optimize proactive evidence-based management rather
than reactive care.26
Evolution of VISICU
Since the original Sentara project, VISICU has implemented
more than 35 programs in 26 states, bringing the eICU
program to more than 200 hospitals and more than 8% of the
adult ICU beds in the United States. More than 300,000
patients now have their ICU care supplemented by a remote
care team annually (Fig. 44.1). These eICU programs can
430 Section VII—Computers, Engineering, and the Future
Fig. 44.1  eICU programs are used to monitor critically ill patients in
intensive care units, emergency departments, and postanesthesia care
units, as well as other areas in university hospitals, community hospitals,
and critical access hospitals.
Table 44.1  Realized Benefits of the eICU
Program
CLINICAL BENEFITS
Decreased severity-adjusted mortality
Decreased adverse outcomes in low-risk patients
Decreased frequency of complications
Increased compliance with best practices
FINANCIAL BENEFITS
Decreased severity-adjusted ICU LOS
Decreased severity-adjusted floor LOS
Decreased cost-per-day for remaining days in the ICU
Decreased nursing staff turnover
Avoided capital for new ICU beds
Improved throughput
Increased hospital revenue
Increased physician pro-fee billing
ICU, Intensive care unit; LOS, length of stay.
contribute to potential clinical and financial benefits including
at large academic medical centers and level I trauma centers
(Table 44.1). For example, in the 30-bed surgical trauma unit
at the Hospital of the University of Pennsylvania, Kohl et al
compared care in the year before implementation of the eICU
program with care in the 2 1 2 years following. In an ICU that
already was staffed with dedicated intensivists, fellows, and
residents, significant reductions in ICU (63%) and hospital
(45%) mortality were observed; that is, 336 lives were saved.34
Hospital and ICU LOS also were reduced with an estimated
$5.8 million savings in direct costs. Studies have also demon-
strated an eICU program to be useful in rural hospitals.35 For
example, in the Avera-McKennan hospital of the Avera Health-
care System in South Dakota, significant reductions in stan-
dardized (APACHE III) ICU and hospital mortality ratios
(0.63 and 0.67 to 0.25 and 0.53, respectively) and ICU and
hospital LOS were observed with use of an eICU program.35
Setup of an eICU
A variety of clinical process changes accompany the introduc-
tion of an eICU. The eICU program is not a single technology
or a single product. It is a set of tools and an innovation stream
that results in a technology-supported team approach to the
care of ICU patients. It can be transformational because
it reorganizes work processes and care models. The eICU
program does not replace bedside doctors and nurses or sig-
nificantly alter their roles and responsibilities, but instead adds
an additional layer of support for the care of the most critically
ill patients whose conditions can, and often do, change rapidly.
Critically ill patients require frequent assessment throughout
the day and night, which is probably an important mechanism
by which intensivists improve care by providing regular avail-
ability and focus on the constant titration of critical care. This
allows for more rapid interventions such as fine-tuning thera-
pies and prompt detection of emerging problems, which if not
recognized early can lead to life-threatening complications.
These tasks (assessment/titration/intervention) represent a
core activity of the remote care team and allow health systems
to leverage existing intensivists to provide 24/7 coverage to all
critically ill patients.
The eICU program is designed to ensure intensivist involve-
ment in care for every ICU patient, and to provide the inten-
sivists with tools that improve efficiency. Cardiologists
frequently say “time is heart.” Stroke specialists now say “time
is brain.” In trauma there is the concept of “the golden hour”
to prevent major systemic damage. In the ICU, there fre-
quently are many important things going on such that only
emergent situations truly generate rapid responses and many
pathophysiologic processes progress further than is ideal
before they are readdressed. Often the traditional systems that
are in place to take care of patients become overwhelmed. In
this setting, improvements in efficiency lead to improved care
as patient-related events are addressed at an earlier stage in
their evolution.
eICU Technology
An essential part of the eICU system that leads to improved
efficiency is the technology infrastructure that manages the
flow of clinical information from the hospitals to the eICU
center. This includes the real-time data from bedside moni-
tors, and all the supporting data (i.e., lab results, medications,
patient registration and administration data, and medical
images such as x-rays and magnetic resonance images), and
audio and video streams for remote observation of patients
(Fig. 44-2). Using evidence-based algorithms and clinician-set
parameters, the software parses patient data in real time to
identify incipient clinical events and alerts clinicians before
 Section VII—Computers, Engineering, and the Future 431

many significant adverse events develop. Smart Alert prompts
evaluate patients’ electronic monitoring data in real time and
flag potentially worrisome trends and readings that exceed
specified thresholds (Fig. 44-3). This focuses the intensivist’s
attention and can promote timely intervention to avoid crises.
The eCareManager System provides a “dashboard-like” inter-
face that helps the remote intensivists coordinate patient care
with doctors and nurses at the bedside. An online decision-
support tool called The Source draws on the latest evidence-
based care guidelines. In addition, the eCareManager system
is built upon a relational database, which allows data organiza-
tion to drive best practices such as simplified multidisciplinary
rounds and the tracking of performance according to evidence-
based best practices.
Clinical Transformation Process
The real significance of the eICU program lies in the way it
changes care delivery. The technology is only enabling; the key
is a clinical process transformation that accompanies the
technology. The hospital admitting or managing physician
remains the attending physician of record and is responsible
for the patient care plan and determining which on-site con-
sultants assist with care delivery. This concept is particularly
important in understanding how an eICU program staffed
with intensivists can be of great assistance in the care of sub-
specialty patients such as the neurocritical care patient. It is
not the role or responsibility of the eICU team to determine
the specific needs of an ICU patient or to define the primary
course of treatments; those evaluations and decisions are best
done at the bedside by the physicians and surgeons who know
the patient. The role of the eICU team is to closely monitor
the patient’s data for evidence that care needs are changing or
that expected data are available, and then to alter or initiate
care in accordance with the previously expressed plan and
direction of the bedside health care providers. If there is a
question whether the original plan is still best considering the
changes or new data, the eICU intensivist seeks clarification
from the managing bedside team and then resumes giving
support accordingly.

eICU Team Responsibilities

Fig. 44.2  eICU clinicians monitor patient data from a remote site, using
Smart Alerts, real-time monitor data, laboratory results, and other
ancillary test results to intervene on the patients’ behalf.
The eICU team can regularly fine-tune the acute care of ICU
patients through observation of trends. Consider a patient
who presents with an intracerebral hemorrhage. The plan of
care following initial evaluation by a neurologist and neuro-
surgeon may include admission to an ICU with an ICP
monitor, frequent follow-up neurologic evaluation, blood
pressure control, and intubation for airway protection. The
remote eICU physician monitoring the patient follows the
plan of the bedside team and monitors ICP and cerebral per-
fusion pressure (CPP) in the patient, and regularly confers
with the bedside nurse via camera to determine stability in the
patient’s neurologic findings. Several hours into the patient’s
ICU course, the eICU staff is alerted via Smart Alerts (see
Fig. 44-3) that, although the patient’s vital signs are not out

Reload Patient Status Font Size: 16-point Select Units Log out

Selected Patient -Additional Alert Details- Selected Units:

<No Selection> Please select an active alert to see the additional information There are 10 unit(s) being
monitored in 6 hospital(s).

Time Site Bed Patient HR/MAP MAP Limit MAP Trend HR Limit HR Trend 02 Sat/RR

11:59 M-MICU 3 GEHRIG, L HR 175, MAP 44

11:55 C-SICU 8 COBB, T H [101] 124

11:47 M-CICU 12 WAGNER, H H [130] 132

11:45 G-CICU 4 CLEMENTE, R H [100] 104

11:30 G-MICU 5 MAYS, W H [120] 126 02 [92] 90, RR [26] 30

11:28 M-SICU 10 STARGELL, W L [70] 65

11:20 M-MICU 3 MUSIAL, S L [92] 77

Fig. 44.3  Smart Alerts software continuously track vital signs and trends, and notify eICU clinicians of aberrancies, enabling quick action and thereby
potentially preventing problems from escalating. H, High; HR, heart rate; L, low; MAP, mean arterial pressure; O2 Sat, oxygen saturation; RR, respiratory
rate.
432 Section VII—Computers, Engineering, and the Future
of the set range of acceptable values, the trend has changed
such that the heart rate has dropped 25 beats per minute and
the mean arterial pressure has risen 30 mm Hg. Inspection of
the ICP trend reveals that it has increased. Concerned about
a potential Cushing response, the eICU physician contacts the
bedside nurse, orders an emergent repeat head computed
tomography (CT) scan, and alerts the attending physician of
record, consulting neurologist, and neurosurgeon. While
awaiting formal input from the neurologic experts, the eICU
physician begins simple measures to treat elevated ICP, such
as sedation and elevation of the head of bed. When the CT
scan is completed and demonstrates extension of the hemor-
rhage with worsening edema, the eICU intensivist confers
with the neurologist or neurosurgeon immediately to discuss
further patient treatment.
The eICU team also can intervene in emergencies and
manage best practices (e.g., glucose control, deep vein throm-
bosis (DVT) and stress ulcer prophylaxis, low tidal volume
ventilation). In many eICU programs, the attending physician
defines in advance the requirements for communication by
the eICU center before nonemergent actions are taken for
their patients. This may range from specific review of all non-
emergent decisions to allowing the eICU intensivist to act as
a surrogate decision maker, calling the bedside team as indi-
cated by specific conditions and degree of expertise in those
decisions. Examination of communication methods suggests
better outcomes when the remote care team is allowed more
autonomy to address new problems and institute best prac-
tices.36 Independent of communication preferences, the eICU
physician reviews all patient data at regular intervals (from
hourly for the sickest patients to every 4 to 6 hours for the
most stable). Each eICU caregiver monitors and delivers care
from a workstation comprised of several desktop monitors
that provide access to the eCareManager application suite,
including audiovisual tools, the hospital information system,
picture archiving and communication system, and remote-
view applications for the bedside monitors (see Fig. 44.2).
Caregivers consist of physicians, nurses, clerical staff, and,
in some programs, a pharmacologist or pharmacist. Staff
numbers vary based on the number of beds in the network.
With smaller networks (less than 70 beds being monitored),
the remote care team is comprised of one intensivist and one
critical care nurse. As the number of monitored beds increases,
the number of care providers increases, with additional nurses
typically added after 60 to 70 beds are reached and a second
physician (not necessarily a second intensivist) added after
100 to 120 beds. There can be variability in how tasks are
allocated among the various care providers, however, several
core features must be in place to achieve quality goals, such as
a comprehensive daily care plan that addresses key clinical
issues (e.g., ventilator weaning, nutrition, risk assessment,
social issues).37 The care plan is developed by the on-site team
and optimally encompasses the input of all relevant care pro-
viders. The eICU team monitors the care plan and assists with
implementing the various components, and can take primary
responsibility for processes such as compliance with best prac-
tices. High compliance rates can be achieved by centralizing
these essential activities and assigning them to dedicated per-
sonnel. Because the remote team has no other clinical respon-
sibilities, it is efficient and continuously attentive. The
responsibilities of the remote team include (1) continuous
monitoring of the progress of each patient; (2) titration
of therapies, as necessary, to achieve care plan objectives;
(3) identification of emerging problems and initiation of
appropriate countermeasures; (4) facilitation of communica-
tion among members of the care team; and (5) responsibility
for best practice compliance.38
Several lines of evidence suggest that allowing the remote
team to fulfill these responsibilities is associated with improved
ICU practices in both urban and rural hospitals, including
many evidence-based practices, such as DVT prophylaxis
compliance, glycemic control,37 and 6- and 24-hour care
bundles in patients with sepsis, often resulting in reduced
mortality.39 In addition to best practice improvements,
improvements in other aspects of patient care have been
observed with an eICU program. For example, Shafer et al
observed a reduction in the number of cardiopulmonary
arrests in their ICUs and the number of patients who died of
cardiopulmonary arrest once an eICU program was instituted
at their hospital.40
The Impact Beyond the ICU
There are several potential advantages to a tele-ICU that
extend beyond the ICU. For example, tele-ICU care may con-
tribute to more judicious decisions about which patients are
to benefit from ICU admission and which patients should
receive palliative care, that is, better resource use.41 Further-
more, any telemedicine program means that there is a large
volume of electronic data that is collected in real time. Used
correctly, this can drive clinical practice improvement through
“instantaneous” quality metrics. Alternatively, the aggregated
data can be used to answer important research questions. For
this to be a reality, however, informatics infrastructure and
protection of patient privacy must be addressed.42,43
A remote eICU monitoring program can help increase
attending physician oversight and availability. This may be
particularly helpful in hospitals with residency programs and
enhance education.44 House officers have an immediate point
of contact to discuss complex cases or even ask simple ques-
tions. Often a new house officer may be hesitant to call an
attending physician in the middle of the night, especially if it
is not clear to the house officer that the question is urgent.
The eICU physician can serve as a buffer for the house officer,
even if it is simply to advise the house officer whether an
issue for the bedside attending physician should be addressed
immediately or can wait until morning. Intensivists in the
eICU center have time to educate house officers and can
quickly email or fax relevant information including publica-
tions, using The Source for topical information or as a
conduit to the medical literature and papers of interest. Staff
in the eICU center should be regarded as a resource for house
staff, but should not usurp house staff autonomy. In that
regard, eICU physicians are trained to include house officers
in any decision making that is of educational value. In addi-
tion, the presence of the eICU clinicians can help reduce the
stress that critical care nurses often experience by providing
them a constant safety net. Together these resources for nurse
and house staff can provide better outcomes for ICU patients
by eliminating the wait time often needed to reach a physi-
cian. An eICU program allows an “ICU without walls”; it is
already available in some emergency departments (EDs) and
postanesthesia care units (PACUs) and is being deployed
through wireless mobile units to assist with rapid response in
 Section VII—Computers, Engineering, and the Future 433
general acute care settings. Several eICU programs also are
adding stroke assessment and intervention programs using
their eICU technology and operational backbones. Office-
based or at-home stroke specialists then can work collabora-
tively with the existing eICU teams and so more rapidly assess
and intervene in acute stroke cases from the ED through the
ICU care stages.
Neurocritical Care Today
In Support of Dedicated Neurointensivists
In the same way patients with acute cardiac dysfunction are
best cared for in coronary care units and by cardiologists,
patients with acute neurologic disorders should be cared for
in an NCCU with dedicated staff. Neurointensivists combine
knowledge of neurologic diseases with the techniques of
intensive care and defragment care by focusing on the inter-
play between the brain and other systems. Specific areas of
expertise of neurocritical care specialists include ICP manage-
ment, hemodynamic augmentation to improve cerebral blood
flow, therapeutic hypothermia, and advanced neuromonitor-
ing (e.g., continuous electroencephalography, brain-tissue
oxygen, and microdialysis).45 Patients with neurologic disor-
ders such as TBI, intracerebral hemorrhage, and stroke who
are admitted to NCCUs appear to have reduced mortality and
better functional outcomes than those admitted to general
ICUs.46-48 In addition, there is reduced LOS and lower overall
costs when patients with acute neurologic disorders are cared
for by neurointensivists in a dedicated NCCU.49-52 These out-
comes likely are associated with developments in several areas,
including improved understanding of disease pathophysiol-
ogy, neuromonitoring, new and effective treatments for
complex neurologic conditions, and staff with specialized
neurocritical care skills. Multimodality neuromonitoring may
play an important role because the importance of early iden-
tification and treatment of secondary insults after acute brain
injury is well established.53 More important than the monitor
is the interpretation of the data and what is done with the
information. This is where the neurointensivist becomes
essential; monitoring enables the neurointensivist to antici-
pate, prevent, and treat potentially dangerous secondary
insults. However, there still is a national shortage of neuroin-
tensivists and many hospitals do not have dedicated NCCUs;
that is, few centers have the expertise and technology to
provide this level of care. Thus for the majority of neurocriti-
cal care patients there is a gap in actual care compared to what
is now known to be the best care; constant surveillance by
specialized nurses, immediate availability of neuroimaging,
specially trained physicians, and specialized interventions that
together can provide constant observation and immediacy of
action to detect and treat neurologic deterioration or patient-
related factors that may aggravate the initial brain injury. In
addition, the technology can be leveraged to enhance safety,
reduce errors, or avoid adverse drug reactions.54,55
Possibilities for the Practice of
Neurocritical Care with an eICU Program
Innovations can help solve important problems. For example,
several stroke networks based on telepresence have evolved so
that stroke expertise is immediately available in remote or
smaller community hospitals.56 Those issues tackled by the
eICU program are fundamental to medical care today:
 The shortage of appropriate specialists
 The shortage and high burn-out rate of specialized nurses
 The need to handle large amounts of data and filter
through the noise and find the important data that needs
attention
 The need for decision support systems to synthesize evolv-
ing medical knowledge and supply it at the point of care
 The demand for detailed documentation of care, and the
cost and complexity of increasingly sick patients in the
hospital
Most NCCUs are equipped with advanced treatments that
may improve prognosis in patients with severe neurologic
conditions that can threaten both survival and brain function.
The advancement of the specialty of neurocritical care, the
creation of specialized NCCUs, and the continued expansion
in number and size of eICU programs are all occurring in
response to the growing needs for evidence-based, timely care
in an efficient, effective manner to produce the best possible
outcomes for patients, health care providers, and health
systems. This creates an opportunity for synergy between
these evolving concepts.
The future likely has the most complex neurocritical care
patients cared for in specialized units by specialized clinicians
using specialized monitoring equipment. It also likely contin-
ues to have some highly complex patients in less complex set-
tings cared for by less specialized caregivers. Less complex but
quite ill or injured patients may also remain in nonspecialized
settings. In short, critically ill patients will continue to exist in
many setting from complex specialized ICUs to general med-
surg units to PACUs to waiting for beds in EDs or neuroradiol-
ogy suites. These patients may be monitored by and their care
supported by dedicated eICU teams led by intensivists. For
example, again consider a patient who presents with an intra-
cerebral hemorrhage to a small critical access hospital. Such a
facility may have a general intensivist on staff but is not a neu-
rointensivist. However, if this critical-access facility is part of
an eICU program, one centered at a large community or aca-
demic hospital, the ED staff can contact the eICU intensivist
for advice when the patient presents to the ED. The intensivist
is able to quickly assess the patient, determine the need for and
arrange immediate transfer to a tertiary care setting, and at the
same time can arrange neuroimaging upon arrival and confer
with the on-call neurologist or neurosurgeon at the referral
center; that is, the eICU intensivist can coordinate patient care
and communication between health care providers.
When the patient is admitted to the ICU, the eICU clini-
cians can assist the bedside neurologic team in providing
round-the-clock clinical surveillance and interpretation of
monitoring technology, and guide specialized treatments to
reduce any immediate or delayed brain damage. Monitoring
the injured brain has become an integral part of the manage-
ment of severely brain-injured patients in intensive care but
has evolved in recent years such that a variety of techniques
now are available (Table 44.2). These monitoring techniques
can provide the neurointensivist with crucial information
about brain physiology and metabolism. Combined, these
techniques (“multimodal monitoring”) can produce an even
more accurate overall picture. This approach, however, is labor
434 Section VII—Computers, Engineering, and the Future
Table 44.2  Potential Neurologic Procedures/
Monitoring/Interventions by an eICU
Program
Neuro-ICU Monitoring Techniques
INTRACRANIAL PRESSURE MONITORING
Intraventricular catheters
Fiberoptic monitors
Subarachnoid bolts
Epidural monitors
Continuous brain tissue monitoring and cerebral blood flow
probes
Jugular venous oxygen saturation monitoring
Invasive hemodynamic monitoring
Continuous electroencephalogram (EEG) monitoring
PROCEDURES MONITORED
Ventricular drainage
Thrombolysis
Plasmaphersis
Hypothermia
INTERVENTIONS
Administration of blood products for urgent indications
Code supervision
Airway management
Fluid bolus for hypotension
Ventilator management
Management of seizure activity
Intensive insulin therapy
Acute management of delirium, psychosis, or severe agitation
Management of severe hypertension
Culture-directed or empiric antibiotics
Pain management
ROUTINE FLUID AND ELECTROLYTE MANAGEMENT
Fluid bolus for low urine output
Routine electrolyte replacement
Interpretation of electrocardiogram
Use of diuretics
intensive and often requires specialized knowledge to best
interpret the information. This is something eICU staff can
help with, and through communication with the bedside
team, alert them of detrimental trends and variations, and
intervene, if needed, to elicit corrections as directed by the
patient’s care plan developed by the bedside team.
eICU clinicians can evaluate and intervene in response to
compromise of any organ system (see Table 44.2). For example,
suppose the patient with intracerebral bleed develops sympa-
thetic nervous system overdrive that may occur with intracra-
nial hemorrhage. When this occurs it may be necessary to
stress the heart and avoid lower CPP to resuscitate the brain
and allow the heart function to recover later. eICU program
technology allows careful monitoring to ensure the blood
pressure remains in a range that would not jeopardize CPP.
An early warning system known as Smart Alert Prompts con-
tinuously analyzes each patient’s vital signs and tracks changes
and trends to identify problems that require rapid attention.
These alerts allow the eICU team to respond immediately to
the slightest changes in the patient’s condition and quickly
identify the most appropriate treatment. The criteria for
prompts can be individualized for each patient so that eICU
clinicians can maintain blood pressure or other parameters
within a range specified by the bedside team.
Outcome Analysis
Data suggest that neurology-based ICUs may benefit from an
eICU program. Saint Lukes Health System in Kansas City
implemented an eICU program in 2006 which monitored,
among others, a neurosurgical ICU. APACHE III scores did
not change; however, compared with pre-eICU, in the year and
a half following the start of eICU, risk-adjusted hospital mor-
tality for neurosurgical ICU patients (standardized mortality
ratio, or observed to predicted mortality) decreased from 0.86
to 0.67 (leading to 15% fewer hospital deaths in that popula-
tion) and ICU LOS decreased from 1.52 to 1.11 (22% fewer
ICU days).57
Conclusion
Treatment of TBI, stroke, and cerebral hemorrhage are just
some of the several pursuits of NCCUs and neurointensivists
today. Future directions include the clinical implementation
of brain resuscitation and brain-sparing therapies, sophisti-
cated monitoring of electrophysiologic and intracranial physi-
ologic indices, and further understanding of the dysfunction
of other organs that follows brain and nerve failure. The intro-
duction of neurointensivists and NCCUs is associated with
reduced hospital mortality and resource use. However, there
is no change overall in readmission rates or long-term mortal-
ity. This suggests that, although neurocritical care has evolved
recently, there is further room for improvement. Implementa-
tion of an eICU program potentially can help expand the
reach of neurointensivists and bring about needed improve-
ments so that all neurocritical care patients can benefit from
the expertise of specially trained physicians.
References
1. Menon D. Neurocritical care: turf label, organizational construct, or clinical
asset? Curr Opin Crit Care 2004;10:91–3.
2. Suarez J. Outcome in neurocritical care: advances in monitoring and
treatment and effect of a specialized neurocritical care team. Crit Care Med
2006;34:S232–8.
3. Wright W. Putting neurocritical care units on the map. 2007. Neurocritical
Care Society: http://www.neurocriticalcare.org/files/public/
advocacyarticle4.pdf.
4. Kohn L, Corrigan J, Donaldson M. To err is human: building a safer health
system. In: IOM Committee on Healthcare in America, editor. Washington,
DC: The National Academies Press; 1999.
5. Committee on Quality of Health Care in America, IOM. Crossing the quality
chasm: a new health system for the 21st century. Washington, DC: The
National Academies Press; 2001.
6. Lwin AK, Shepard DS. Estimating lives and dollars saved from universal
adoption of the Leapfrog safety and quality standards: 2008 update.
Washington, DC: Leapfrog Group; 2008.
7. Milstein A, Galvin RS, Delbanco SF, et al. Improving the safety of health
care: the Leapfrog initiative. Eff Clin Pract 2000;3(6):313–6.
8. Gutsche JT, Kohl BA. Who should care for intensive care unit patients? Crit
Care Med 2007;35(2):S18–S23.
9. Kahn JM, Matthews FA, Angus DC, et al. Barriers to implementing the
Leapfrog Group recommendations for intensivist physician staffing: a survey
of intensive care unit directors. J Crit Care 2007;22(2):97–103.
10. Pronovost PJ, Angus DC, Dorman T, et al. Physician staffing patterns and
clinical outcomes in critically ill patients: a systematic review. JAMA 2002;
288:2151–62.
11. Hanson C, Deutschman C, Anderson HL, et al. Effects of an organized
critical care service on outcomes and resource utilization: a cohort study.
Crit Care Med 1999;27:270–4.
12. Pronovost PJ, Jenckes MW, Dorman T, et al. Organizational characteristics of
intensive care units related to outcomes of abdominal aortic surgery. JAMA
1999;281:1310–17.
 Section VII—Computers, Engineering, and the Future 435
13. Young MB, Birkmeyer JD. Potential reduction in mortality rates using an
intensivist model to manage intensive care units. Eff Clin Pract 2000;3:284–9.
14. Levy MN, Rapoport J, Lemeshow S, et al. Association between critical care
physician management and patient mortality in the intensive care unit. Ann
Intern Med 2008;148:801–9.
15. Grundy BL, Crawford P, Jones PK, et al. Telemedicine in critical care: an
experiment in health care delivery. JACEP 1977;6:439–44.
16. Grundy BL, Jones PK, Lovitt A. Telemedicine in critical care: problems in
design, implementation, and assessment. Crit Care Med 1982;10:471–5.
17. Rosenfeld BA, Dorman T, Breslow MJ, et al. Intensive care unit telemedicine:
alternate paradigm for providing continuous intensivist care. Crit Care Med
2000;28:3925–31.
18. Breslow MJ, Rosenfeld BA, Doerfler M, et al. Effect of a multiple-site
intensive care unit telemedicine program on clinical and economic
outcomes: an alternative paradigm for intensivist staffing. Crit Care Med
2004;32:31–8.
19. Angus DC, Kelley M, Schmitz R, et al. Committee on Manpower for
Pulmonary and Critical Care Societies (COMPACCS). Caring for the
critically ill patient: current and projected workforce requirements for care
of the critically ill and patients with pulmonary disease: can we meet the
requirements of an aging population? JAMA 2000;284(21):2762–70.
20. Gajic O, Afessa B, Hanson AC, et al. Effect of 24-hour mandatory versus
on-demand critical care specialist presence on quality of care and family and
provider satisfaction in the intensive care unit of a teaching hospital. Crit
Care Med 2008;36 (1):36–44.
21. Vespa PM, Miller C, Hu X, et al. Intensive care unit robotic telepresence
facilitates rapid physician response to unstable patients and decreased cost in
neurointensive care. Surg Neurol 2007;67(4):331–7.
22. Thomas EJ, Lucke JF, Wueste L, et al. Association of telemedicine for remote
monitoring of intensive care patients with mortality, complications, and
length of stay. JAMA 2009;302(24):2671–78.
23. New England Healthcare Institute. Tele-ICUs: remote management in
intensive care units. Cambridge, MA: Massachusetts Technology
Collaborative and Health Technology Center; 2007.
24. Berenson RA, Grossman JM, November EA. Does telemonitoring of
patients—the eICU—improve intensive care? Health Aff (Millwood) 2009;
28(5):w937–47.
25. Gracias V. Outcomes of SICU patients after implementation of an electronic
ICU (“eICU”) system and off-site intensivist. Paper presented at 42nd World
Congress of Surgery, Montreal, Quebec; August 2007.
26. Zawada ET Jr, Herr P, Larson D, et al. Impact of an intensive care unit
telemedicine program on a rural health care system. Postgrad Med 2009;
121(3):160–70.
27. Van der Kloot T, Riker R, Goran S, et al. Improved hospital survival after
implementing a remote ICU telemedicine program. Paper presented at 14th
Annual Meeting and Exposition of the American Telemedicine Association,
Las Vegas, NV; April 27, 2009.
28. McCambridge M, Jones K, Paxton H, et al. Association of health information
technology and teleintensivist coverage with decreased mortality and
ventilator use in critically ill patients. Arch Intern Med 2010;170(7):648–53.
29. Morrison JL, Cai Q, Davis N, et al. Clinical and economic outcomes of the
electronic intensive care unit: results from two community hospitals. Crit
Care Med 2010;38(1):2–8.
30. Young LB, Chan PS, Lu X, et al. Impact of telemedicine intensive care unit
coverage on patient outcomes: a systematic review and meta-analysis. Arch
Intern Med 2011;171(6):498–506.
31. Luyt CE, Combes A, Aegerter P, et al. Mortality among patients admitted to
intensive care units during weekday day shifts compared with “off ” hours.
Crit Care Med 2007;35(1):3–11.
32. Girard TD, Kress JP, Fuchs BD, et al. Efficacy and safety of a paired sedation
and ventilator weaning protocol for mechanically ventilated patients in
intensive care (Awakening and Breathing controlled trial): a randomised
controlled trial. Lancet 2008;371(9607):126–34.
33. Sairanen T, Soinila S, Nikkanen M, et al. Finnish Telestroke Task Force. Two
years of Finnish Telestroke: thrombolysis at spokes equal to that at the hub.
Neurology 2011;76(13):1145–52.
34. Kohl BA, Gutsche JT, Kim P, et al. Effect of telemedicine on mortality and
length of stay in a university ICU. Crit Care Med 2007;35:A22.
35. Zawada ET, Herr P, Erickson D, et al. Financial benefit of a teleintensivist
program to a rural health system. Chest 2007;132:444.
36. Cowboy EN, Rajamani S, Jamil MG, et al. Impact of remote ICU
management on ventilator days. Crit Care Med 2005;33:A1.
37. Breslow M. Remote ICU care programs: current status. J Crit Care 2007;22:
66–76.
38. Aaronson ML, Zawada ET, Herr P. Role of a telemedicine intensive care unit
program (TISP) on glycemic control (GC) in seriously ill patients in a rural
health system. Chest 2006;130:226S.
39. Rincon T, Bourke G, Ikeda D. Centralized, remote care improves sepsis
identification, bundle compliance and outcomes. Chest 2007;132:557b.
40. Shaffer JP, Breslow MJ, Johnson JW, et al. Remote ICU management
improves outcomes in patients with cardiopulmonary arrest. Crit Care Med
2005;33:A5.
41. Ramsay A, Rabinowitz T. Developing a rural telemedicine program to
provide palliative care consultation. Paper presented at 14th Annual Meeting
and Exposition of the American Telemedicine Association, Las Vegas, NV;
April 27, 2009.
42. Gamble KH. Critical care network. ICU telemedicine can help ease the
burden of caring for critically ill patients—provided all the right pieces are
in place. Healthcare Inform 2009;26(12):26–30.
43. McShea M, Holl R, Badawi O, et al. The eICU research institute—a
collaboration between industry, health-care providers, and academia. IEEE
Eng Med Biol Mag 2010;29(2):18–25.
44. Conde JG, De S, Hall RW, et al. Telehealth innovations in health education
and training. Telemed J E Health 2010;16(1):103–6.
45. Rincon F, Mayar SA. Neurocritical care: a distinct discipline? Curr Opin Crit
Care 2007;13:115–21.
46. Diringer MN, Edwards DF. Admission to a neurologic/neurosurgical
intensive care unit is associated with reduced mortality rate after
intracerebral hemorrhage. Crit Care Med 2001;29:635–40.
47. Mirski MA, Chang CW, Cowan R. Impact of a neuroscience intensive care
unit on neurosurgical patient outcomes and cost of care: evidence-based
support for an intensivist-directed specialty ICU model of care. J Neurosurg
Anesthesiol 2001;13:83–92.
48. Elf K NP, Enblad P. Outcome after traumatic brain injury improved by an
organized secondary insult program and standardized neurointensive care.
Crit Care Med 2002;30:2129–34.
49. Mirski MA, Chang CW, Cowan R. Impact of a neuroscience intensive care
unit on neurosurgical patient outcomes and cost of care: evidence-based
support for an intensivist-directed specialty ICU model of care. J Neurosurg
Anesthesiol 2001;13:83–92.
50. Suarez JI. Outcome in neurocritical care: advances in monitoring and
treatment and effect of a specialized neurocritical care team. Crit Care Med
2006;34:S232–8.
A complete list of references for this chapter can be found online at
www.expertconsult.com.
 Section VII—Computers, Engineering, and the Future 435.e1
References
1. Menon D. Neurocritical care: turf label, organizational construct, or clinical
asset? Curr Opin Crit Care 2004;10:91–3.
2. Suarez J. Outcome in neurocritical care: advances in monitoring and
treatment and effect of a specialized neurocritical care team. Crit Care Med
2006;34:S232–8.
3. Wright W. Putting neurocritical care units on the map. 2007. Neurocritical
Care Society: http://www.neurocriticalcare.org/files/public/
advocacyarticle4.pdf.
4. Kohn L, Corrigan J, Donaldson M. To err is human: building a safer health
system. In: IOM Committee on Healthcare in America, editor. Washington,
DC: The National Academies Press; 1999.
5. Committee on Quality of Health Care in America, IOM. Crossing the quality
chasm: a new health system for the 21st century. Washington, DC: The
National Academies Press; 2001.
6. Lwin AK, Shepard DS. Estimating lives and dollars saved from universal
adoption of the Leapfrog safety and quality standards: 2008 update.
Washington, DC: Leapfrog Group; 2008.
7. Milstein A, Galvin RS, Delbanco SF, et al. Improving the safety of health
care: the Leapfrog initiative. Eff Clin Pract 2000;3(6):313–6.
8. Gutsche JT, Kohl BA. Who should care for intensive care unit patients? Crit
Care Med 2007;35(2):S18–S23.
9. Kahn JM, Matthews FA, Angus DC, et al. Barriers to implementing the
Leapfrog Group recommendations for intensivist physician staffing: a survey
of intensive care unit directors. J Crit Care 2007;22(2):97–103.
10. Pronovost PJ, Angus DC, Dorman T, et al. Physician staffing patterns and
clinical outcomes in critically ill patients: a systematic review. JAMA
2002;288:2151–62.
11. Hanson C, Deutschman C, Anderson HL, et al. Effects of an organized
critical care service on outcomes and resource utilization: a cohort study.
Crit Care Med 1999;27:270–4.
12. Pronovost PJ, Jenckes MW, Dorman T, et al. Organizational characteristics of
intensive care units related to outcomes of abdominal aortic surgery. JAMA
1999;281:1310–17.
13. Young MB, Birkmeyer JD. Potential reduction in mortality rates using an
intensivist model to manage intensive care units. Eff Clin Pract 2000;3:
284–9.
14. Levy MN, Rapoport J, Lemeshow S, et al. Association between critical care
physician management and patient mortality in the intensive care unit. Ann
Intern Med 2008;148:801–9.
15. Grundy BL, Crawford P, Jones PK, et al. Telemedicine in critical care: an
experiment in health care delivery. JACEP 1977;6:439–44.
16. Grundy BL, Jones PK, Lovitt A. Telemedicine in critical care: problems in
design, implementation, and assessment. Crit Care Med 1982;10:471–5.
17. Rosenfeld BA, Dorman T, Breslow MJ, et al. Intensive care unit telemedicine:
alternate paradigm for providing continuous intensivist care. Crit Care Med
2000;28:3925–31.
18. Breslow MJ, Rosenfeld BA, Doerfler M, et al. Effect of a multiple-site
intensive care unit telemedicine program on clinical and economic
outcomes: an alternative paradigm for intensivist staffing. Crit Care Med
2004;32:31–8.
19. Angus DC, Kelley M, Schmitz R, et al. Committee on Manpower for
Pulmonary and Critical Care Societies (COMPACCS). Caring for the
critically ill patient: current and projected workforce requirements for care
of the critically ill and patients with pulmonary disease: can we meet the
requirements of an aging population? JAMA 2000;284(21):2762–70.
20. Gajic O, Afessa B, Hanson AC, et al. Effect of 24-hour mandatory versus
on-demand critical care specialist presence on quality of care and family and
provider satisfaction in the intensive care unit of a teaching hospital. Crit
Care Med 2008;36(1):36–44.
21. Vespa PM, Miller C, Hu X, et al. Intensive care unit robotic telepresence
facilitates rapid physician response to unstable patients and decreased cost in
neurointensive care. Surg Neurol 2007;67(4):331–7.
22. Thomas EJ, Lucke JF, Wueste L, et al. Association of telemedicine for remote
monitoring of intensive care patients with mortality, complications, and
length of stay. JAMA 2009;302(24):2671–78.
23. New England Healthcare Institute. Tele-ICUs: remote management in
intensive care units. Cambridge, MA: Massachusetts Technology
Collaborative and Health Technology Center; 2007.
24. Berenson RA, Grossman JM, November EA. Does telemonitoring of
patients—the eICU—improve intensive care? Health Aff (Millwood)
2009;28(5):w937–47.
25. Gracias V. Outcomes of SICU patients after implementation of an electronic
ICU (“eICU”) system and off-site intensivist. Paper presented at 42nd World
Congress of Surgery, Montreal, Quebec; August 2007.
26. Zawada ET Jr, Herr P, Larson D, et al. Impact of an intensive care unit
telemedicine program on a rural health care system. Postgrad Med
2009;121(3):160–70.
27. Van der Kloot T, Riker R, Goran S, et al. Improved hospital survival after
implementing a remote ICU telemedicine program. Paper presented at 14th
Annual Meeting and Exposition of the American Telemedicine Association,
Las Vegas, NV; April 27, 2009.
28. McCambridge M, Jones K, Paxton H, et al. Association of health information
technology and teleintensivist coverage with decreased mortality and
ventilator use in critically ill patients. Arch Intern Med 2010;170(7):648–53.
29. Morrison JL, Cai Q, Davis N, et al. Clinical and economic outcomes of the
electronic intensive care unit: results from two community hospitals. Crit
Care Med 2010;38(1):2–8.
30. Young LB, Chan PS, Lu X, et al. Impact of telemedicine intensive care unit
coverage on patient outcomes: a systematic review and meta-analysis. Arch
Intern Med 2011;171(6):498–506.
31. Luyt CE, Combes A, Aegerter P, et al. Mortality among patients admitted to
intensive care units during weekday day shifts compared with “off ” hours.
Crit Care Med 2007;35(1):3–11.
32. Girard TD, Kress JP, Fuchs BD, et al. Efficacy and safety of a paired sedation
and ventilator weaning protocol for mechanically ventilated patients in
intensive care (Awakening and Breathing controlled trial): a randomised
controlled trial. Lancet 2008;371(9607):126–34.
33. Sairanen T, Soinila S, Nikkanen M, et al. Finnish Telestroke Task Force. Two
years of Finnish Telestroke: thrombolysis at spokes equal to that at the hub.
Neurology 2011;76(13):1145–52.
34. Kohl BA, Gutsche JT, Kim P, et al. Effect of telemedicine on mortality and
length of stay in a university ICU. Crit Care Med 2007;35:A22.
35. Zawada ET, Herr P, Erickson D, et al. Financial benefit of a teleintensivist
program to a rural health system. Chest 2007;132:444.
36. Cowboy EN, Rajamani S, Jamil MG, et al. Impact of remote ICU
management on ventilator days. Crit Care Med 2005;33:A1.
37. Breslow M. Remote ICU care programs: current status. J Crit Care
2007;22:66–76.
38. Aaronson ML, Zawada ET, Herr P. Role of a telemedicine intensive care unit
program (TISP) on glycemic control (GC) in seriously ill patients in a rural
health system. Chest 2006;130:226S.
39. Rincon T, Bourke G, Ikeda D. Centralized, remote care improves sepsis
identification, bundle compliance and outcomes. Chest 2007;132:557b.
40. Shaffer JP, Breslow MJ, Johnson JW, et al. Remote ICU management
improves outcomes in patients with cardiopulmonary arrest. Crit Care Med
2005;33:A5.
41. Ramsay A, Rabinowitz T. Developing a rural telemedicine program to
provide palliative care consultation. Paper presented at 14th Annual Meeting
and Exposition of the American Telemedicine Association, Las Vegas, NV;
April 27, 2009.
42. Gamble KH. Critical care network. ICU telemedicine can help ease the
burden of caring for critically ill patients—provided all the right pieces are
in place. Healthcare INFORM 2009;26(12):26–30.
43. McShea M, Holl R, Badawi O, et al. The eICU research institute—a
collaboration between industry, health-care providers, and academia. IEEE
Eng Med Biol Mag 2010;29(2):18–25.
44. Conde JG, De S, Hall RW, et al. Telehealth innovations in health education
and training. Telemed J E Health 2010;16(1):103–6.
45. Rincon F, Mayar SA. Neurocritical care: a distinct discipline? Curr Opin Crit
Care 2007;13:115–21.
46. Diringer MN, Edwards DF. Admission to a neurologic/neurosurgical
intensive care unit is associated with reduced mortality rate after
intracerebral hemorrhage. Crit Care Med 2001;29:635–40.
47. Mirski MA, Chang CW, Cowan R. Impact of a neuroscience intensive care
unit on neurosurgical patient outcomes and cost of care: evidence-based
support for an intensivist-directed specialty ICU model of care. J Neurosurg
Anesthesiol 2001;13:83–92.
48. Elf K NP, Enblad P. Outcome after traumatic brain injury improved by an
organized secondary insult program and standardized neurointensive care.
Crit Care Med 2002;30:2129–34.
49. Mirski MA, Chang CW, Cowan R. Impact of a neuroscience intensive care
unit on neurosurgical patient outcomes and cost of care: evidence-based
support for an intensivist-directed specialty ICU model of care. J Neurosurg
Anesthesiol 2001;13:83–92.
50. Suarez JI. Outcome in neurocritical care: advances in monitoring and
treatment and effect of a specialized neurocritical care team. Crit Care Med
2006;34:S232–8.
51. Suarez JI, Zaidat OO, Suri MF, et al. Length of stay and mortality in
neurocritically ill patients: impact of a specialized neurocritical care team.
Crit Care Med 2004;32:2311–7.
435.e2 Section VII—Computers, Engineering, and the Future
52. Varelas PN, Conti MM, Spanaki MV, et al. The impact of a neurointensivist-
led team on a semiclosed neurosciences intensive care unit. Crit Care Med
2004;32:2191–8.
53. DeGeorgia MA, Krieger DW, Abou-Chebl DA, et al. Cooling for acute
ischemic brain damage, a feasibility trial of endovascular cooling. Neurology
2004;63:312–7.
54. Meidl TM, Woller TW, Iglar AM, et al. Implementation of pharmacy services in
a telemedicine intensive care unit. Am J Health Syst Pharm 2008;65(15):1464–9.
55. Hassan E, Badawi O, Weber RJ, et al. Using technology to prevent adverse drug
events in the intensive care unit. Crit Care Med 2010;38(6 Suppl):S97–S105.
56. Lai F. Stroke networks based on robotic telepresence. J Telemed Telecare
2009;15(3):135–6.
57. Lem V, eICU medical director, St. Luke’s Health System, Kansas City.
Personal communication.
VII
Chapter
45  
Medical Informatics
J. Claude Hemphill III, Marco D. Sorani, Stuart Russell, and Geoffrey T. Manley
Introduction
The term medical informatics frequently is used to describe
systems applied to hospital and physician office record
keeping, operations, and regulatory documentation, such as
electronic medical records (EMR), computerized order entry
(CPOE), and databases for billing or review by regulatory
bodies such as The Joint Commission. However, these systems
generally provide an electronic recapitulation of written
records rather than new insight into disease treatments. In
the context of monitoring in neurocritical care, medical infor-
matics is more appropriately defined as the acquisition,
storage, and analysis of neuromonitoring data to (1) facilitate
knowledge and discovery about pathogenesis and treatment
of disease, and (2) improve methods to integrate the com­
plex array of clinical, physiologic, and genetic data into
practical bedside care.1 Unfortunately, although monitoring—
specifically neuromonitoring—has advanced tremendously in
the four decades since the origins of critical care, for the
purposes of clinical care, data are considered in a way similar
to what was used a generation ago. However, it is clear that
the complexity of data in the neurocritical care unit (NCCU)
requires a fresh and novel approach to medical informatics
to complement the development of new neuromonitoring
paradigms.2,3 This chapter provides insight into this clinical
problem and describes some initial efforts toward this end.
The Data Environment
The NCCU is an extremely data-intense environment. The art
of neurocritical care is to use these data in real-time to make
decisions about patient care.2 Patients routinely are monitored
frequently or continuously using 30 or more parameters
including blood pressure (BP), heart rate (HR), systemic
arterial oxygen saturation (SaO2), respiratory rate (RR), tidal
volume, peak inspiratory pressures, intracranial pressure
(ICP), and body temperature (T), to name a few. Add to these
the variety of new advanced neuromonitoring tools such as
brain tissue oxygen tension (PbtO2), cerebral blood flow (CBF),
and microdialysis parameters and several obvious questions
emerge:
1. What do we do with all these data?
2. Are all of these data useful and important?
3. How are artifacts “cleaned?”
4. When and how often are the data collected?
5. Which data are discarded and when?
436
Every NCCU already uses some form of informatics technol­
ogy for these purposes. However, in most circumstances, this
“technology” consists of monitors that display but do not
capture and analyze the data. Often raw data are simply visu­
ally extracted from the monitor and recorded on a paper chart
at defined intervals. Bedside nurses and physicians then review
this paper record of the data and rely on their expertise and
intuition, without the aid of additional analytic tools, to apply
it to clinical care. At a simple level, this leads to data cleaning
because obvious artifacts, for example, a monitor that was
turned off during patient transport, are accounted for. The
presence of large amounts of data that require complex inte­
gration and robust analytic tools is now a routine problem
for the corporate world, such as the financial, airline, and
computer industries. These industries have made significant
advances in information technology over the past 20 years.
Remarkably, only recently have attempts been initiated to
push critical care informatics beyond the paper chart and
clinician intuition to address the questions posed earlier.
Figure 45.1 provides an example of the complexity and
volume of even a short duration of critical care data.
What Is Not Known
The fundamental purpose of neurocritical care (for central
nervous system problems) is the identification, prevention,
and treatment of secondary brain injury. Physiology matters
and much of the focus in the NCCU is on management of
secondary brain insults (SBIs), which may include hypoten­
sion, hypoxia, fever, elevated ICP, hyperglycemia, and seizures.
In fact, much of the moment-to-moment treatment provided
in neurocritical care involves ensuring that these physiologic
parameters, which can be seen as surrogate or intermediate
targets with presumed impact on survival and long-term
functional outcome, are within specified ranges. This often
leads to a lengthy series of bedside orders that delineate accept­
able ranges of one parameter at a time (e.g., “Call medical
doctor if systolic BP <90 mm Hg”; “Call medical doctor if
SaO2 <94%”; “Call medical doctor if ICP >20 mm Hg”). The
bedside nurse then calls the physician and abnormalities in a
single parameter are treated in a reactive fashion (e.g., acet­
aminophen after a temperature elevation, mannitol after con­
tinued ICP elevation) with little attention to how multiple
parameters may interact. Yet, while patients are treated in a
univariate manner, they live in a multivariate world in which
many factors interact to determine injury and outcome. Fur­
thermore, the clinical response is to a certain threshold rather
© Copyright 2013 Elsevier Inc. All rights reserved.
 Section VII—Computers, Engineering, and the Future 437

Fig. 45.1  Critical care data are voluminous and complex. These graphs depict two days of continuous neurocritical care physiologic data from
a single patient represented on a time-series plot. The top panel shows blood pressure, central venous pressure, heart rate, and intracranial pressure.
The middle panel shows brain tissue oxygen, brain temperature, and jugular venous oxygen saturation. The bottom panel shows measures obtained
from the mechanical ventilator. These data are too noisy and compressed for visual inspection to be informative to identify single events, trends, or
relationships between parameters, thus necessitating summary measures, indices, or informatics-based analyses.

issues in a more comprehensive and sophisticated manner,

Table 45.1  Several Existing Neurocritical Care
Informatics Questions
Do secondary brain insults (e.g., fever, elevated intracranial
pressure [ICP], hypotension) have a dose-response
relationship with outcome?
Are there multivariate interactions and relationships among
various physiologic parameters that influence outcome
beyond individual parameters?
Are there event signatures that would predict the future
occurrence of a deleterious problem (e.g., elevated ICP or
organ failure) that would allow proactive measures to
prevent its occurrence?
How do we integrate new measures (e.g., PbtO2)?
How often do we need to collect physiologic data to optimize
patient care?
How can we classify patients to understand their
heterogeneity?
How can we integrate physiologic data with data from
genomic, proteomic, imaging, and other types of data?
How can physiologic data be presented in real-time and in
a user-friendly manner to provide decision-enabling
information to clinicians?
To what extent is prior medical history (from before
hospitalization or earlier in hospital course) important in
interpreting current physiologic values and events?
than to trends over time and with little insight into the
complex interactions between various physiologic factors.
Table 45.1 enumerates several questions about secondary
brain injury in neurocritical care that remain elusive because
of limitations in existing approaches to informatics. It should
be emphasized that a fundamental limitation that has ham­
pered attempts to address these issues has been inadequacies
in data acquisition and storage. To begin to address these
data acquisition tools well beyond the paper chart are required.
This is simply not a recapitulation of a paper record in an
electronic form as occurs in many commercial EMRs4 but
rather one in which all data, including patient records, labora­
tory analysis, imaging, and continuous physiologic data, are
not only acquired but also archived, integrated, and synchro­
nized.5,6 Many monitors used in the NCCU are stand-alone
and self-contained; that is, they are not designed to operate
together. This has been an obstacle to NCCU informatics but
should become less of a barrier since the American Society for
Testing and Materials adopted the concept of an integrated
clinical environment in 2009.7 Other chapters in this text focus
on the various monitors and the challenges to integration
methods, and they should be seen as both complementary and
required aspects of neurocritical care informatics.
Are More Data Better? The Example
of Defining “Dose” of Secondary
Brain Insults
Many studies have identified SBIs, such as hypotension,
hypoxia, elevated ICP, and fever, as being associated with poor
outcome after acute neurologic injury or illness.8-13 However,
the manner in which these events have been defined often
varies widely across studies. Some studies have used any
occurrence of an SBI (e.g., systolic BP <90 mm Hg), while
others have used the number of times an event has occurred,
and others have used the duration below a specific threshold.
Many times, the choice of a specific definition depends on the
limitation imposed by the resolution, or frequency, with which
physiologic vital signs were collected in the medical record or
438 Section VII—Computers, Engineering, and the Future
research database. For example, in a study that identified fever
as a predictor of hospital length of stay, only the maximum
daily temperature was recorded in the database used for study
analysis.10 In another study of the impact of hypotension in
the emergency department (ED) on outcome after TBI, the
number of times the systolic BP was less than 90 mm Hg was
assessed, regardless of depth or duration of the episodes.14 All
of these various methods are attempts to define a “dose” of the
deleterious event, a concept familiar in pharmacology but
often not used in the NCCU.
In pharmacokinetics, dose or “exposure” is usually defined
as area under the curve of concentration versus time. Intui­
tively this makes sense for SBIs; a longer duration of a lower
BP should be more injurious than a brief event that just barely
passed the threshold definition. However, appreciating this
dose is not readily apparent from current paper or electronic
bedside charting. Even this seemingly simple assessment is an
example of informatics technology; the resolution at which
data are collected must be defined and analysis must be per­
formed on raw collected data to derive a summary dose
term—SBI burden. Figure 45.2 demonstrates the example of
fever burden15 as area under the curve above a defined thresh­
old. Importantly, initial studies suggest that it makes a differ­
ence both how frequently data are collected and how “dose” is
defined. It is not surprising that more frequent data collection
identifies more SBIs. The extension of this is the conclusion
that current intensive care unit (ICU) charting practices
(usually writing down a value hourly) are missing many events
of potential clinical significance. Furthermore, integrating
these data as “area under the curve” seems to be more infor­
mative than just identifying presence or number of events.
Hypotension burden is more associated with outcome than
just number of events of hypotension.16 Fever burden is a
broader method of assessing normothermia control than just
time over a threshold.15,17 Thus just the assessment of SBIs as
“dose” is a straightforward and seemingly simple example of
ICU informatics that is still not clinically routine because of
shortcomings in data acquisition, charting, and bedside data
analysis tools.
39
Body temperature (°C)
38
37
36
35
0 24 48 72 96 120 144
Hours from hospital admission
Fig. 45.2  Fever curve with cutpoint at 38.0° C (dashed line). Depending
on the measurement method, this patient could be considered: 
(1) positive for any fever during hospitalization, (2) had 15 febrile
episodes, (3) had a total duration of fever of 78.2 hours, or (4) had a
total fever dose (area under the curve [AUC]) of 28.1 degree-hours. AUC
is calculated as the sum of all the solid regions over the 38.0° C cutpoint.
New Measures Derived from Old—
The Example of PRx
A compelling reason to pursue critical care informatics is the
possibility that measurements derived from analysis of single
or multiple parameters might be more informative than the
raw data of a single parameter alone. Although a very simpli­
fied example, the most common measurement routinely used
in this fashion is cerebral perfusion pressure (CPP). CPP is
not measured directly but rather is a derived measure calcu­
lated as the difference between the mean arterial pressure
(MAP) and the ICP. CPP is routinely calculated by monitors,
charting systems, or by hand and is used in real time at the
bedside in neurocritical care. However, current systems and
bedside charting methods are poorly equipped to derive mea­
sures that involve anything more complex than simple math­
ematics. Even so, some early neurocritical care informatics
systems have provided the ability to assess some new param­
eters derived from relationships between existing parameters.
PRx, the pressure reactivity index, is one example.
Cerebral autoregulation is a critical component to ensure
adequate blood flow to normal and injured brain tissue.
Impaired autoregulation is associated with worsened outcome
after TBI and management of CPP based on an individual
patient’s autoregulation status may improve outcome. How­
ever, autoregulation status is difficult to assess from usual
monitored parameters. PRx is a moving correlation coefficient
between ICP and MAP and is one measure of whether auto­
regulation is intact or impaired.18,19 PRx ranges between −1
and 1, with a positive number indicating some correlation
between the two parameters and a negative number indicating
an absence of correlation. If ICP and MAP are correlated, this
means that the ICP increases as the MAP increases; this occurs
when the brain’s autoregulation status is impaired or the MAP
is shifted off the normal autoregulation curve. If there is no
correlation, then ICP and MAP are independent and auto­
regulation is presumably intact. A PRx of less than 0.3 is often
considered as a threshold to assess intact autoregulation. PRx
has been used in studies of both TBI and nontraumatic intra­
cerebral hemorrhage to define optimal CPP, the CPP at which
PRx is brought within the noncorrelating range.20 Impor­
tantly, PRx is determined, not as a simple arithmetic function
between single current values of existing parameters, but
rather as a statistically analyzed relationship between time-
series data from existing parameters (ICP and MAP); that is,
it is a form of real-time multivariable modeling. Although a
seemingly straightforward example of a single derived para­
meter, PRx represents a leap forward in applying real-time
analytics to raw data, which then can be used to drive clinical
care. It does, however, require time-synchronization and inte­
gration of digital data acquisition to allow real-time analysis.
Such systems now exist for that purpose (e.g., ICM+).21
168 Advanced Critical Care
Informatics I: Data-Driven
Methods for Prediction
The previous examples represent fairly straightforward appli­
cations of medical informatics that become possible only
when data are collected electronically in a high-resolution
format. The methods of analysis (e.g., area under the curve,
 Section VII—Computers, Engineering, and the Future 439
correlation coefficients) are conventional in statistics and
provide targeted summaries of measurements collected over
an extended period for a single patient. Far more is possible,
however. This section goes beyond summarization of mea­
surements into the realm of data-driven methods that inter­
pret measurements and predict outcomes, or better yet events
that may occur in the ICU22 based on experience represented
by repositories of data collected from many patients. The
next section considers model-based methods that combine
repository data with expert knowledge of physiology to reason
about underlying patient states and their anticipated evolu­
tion over time.
It is has been said that critical care is “the art of managing
extreme complexity.”23 Informatics should thus be seen as a
potential tool to help manage this complexity. Both data-
driven and model-based methods do this by analyzing raw
data and generating outputs that are much closer to actionable
decisions. Of course, expert bedside clinicians do this already.
Through accumulated experience, they have learned that
certain patterns in the data—often, patterns across time and
across multiple parameters and monitoring tools (heretofore
described as sensors)—may indicate future problems that can
be avoided with preemptive intervention. Advanced informat­
ics techniques can support this process through data visualiza­
tion and by extracting features not readily available from raw
data and clinical charting and, to some extent, they can auto­
mate this process, at least in part.
The fields of statistics and machine learning have produced
a huge range of methods to analyze data and make predic­
tions. The methods are typically trained on previously col­
lected data and can be tested on unseen data to evaluate the
accuracy of their predictions. Most of the methods used in
medical informatics are supervised learning methods, which
means that the training data contain ground-truth or expert-
supplied answers for each of the instances. (Unsupervised
methods such as clustering algorithms are useful for explor­
atory analysis—“making sense” of a large data collection or
discovering unsuspected regularities through “data mining.”)
Within the class of supervised learning methods, many differ­
ent prediction methods are used. Three are considered here:
linear models (which include most indices), neural networks,
and decision trees.
Linear Models and Indices
The simplest way to build a predictor from multivariate ICU
data is to combine them linearly—that is, by taking a weighted
sum. Linear regression sets the weights to give the best linear
fit to a continuous output variable such as length of ICU stay,
whereas logistic regression passes the sum through a soft
threshold function to give an estimated probability for a
binary output variable such as mortality. The Glasgow Coma
Score (GCS) can be thought of as a linear model that sums
three terms: eye, verbal, and motor responses, with weights 4,
5, and 6, respectively. Although there is no explicit outcome
predictor, thresholds often are applied to the GCS to predict
outcome. As the GCS example illustrates, selecting or design­
ing highly predictive indices that aggregate or summarize the
raw data is a crucial contribution.
Recent techniques have included the development of indices,
such as total fever burden, elevated ICP dose, or burden of
brain hypoxia, based on area-under-the-curve calculations.
For example, Van Santbrink et al.24 investigated the prognostic
value of a derived index known as brain tissue oxygen response
(TOR). Patients with an unfavorable outcome had a higher
TOR during the first 24 hours. Logistic regression analysis
supported the independent predictive value of TOR for
outcome. Oddo et al.25 examined the burden of brain hypoxia
and intracranial hypertension after TBI and found that brain
hypoxia was associated with poor short-term outcome after
severe TBI independently of elevated ICP, low CPP, and injury
severity. Amin and Kulkarni26 used the opinions of clinical
experts to develop a “fuzzy” GCS to predict cognitive recovery
and found that it improved the information content for pre­
diction. Chambers et al.27 developed pressure-time indices and
identified age-related thresholds for ICP and CPP. These
indices also were predictive of score on the Glasgow Outcome
Scale (GOS) and could be compared independently of age.
Hornero et al.28 studied changes in ICP complexity, estimated
by the approximate entropy of the ICP signal, as subjects
progressed from normal to elevated ICP. They concluded that
decreased complexity coincides with episodes of hypertension,
suggesting that regulatory mechanisms that govern ICP are
disrupted during acute rises. These different examples all dem­
onstrate early efforts to develop informatics methods to aid in
outcome prediction. Regression analysis, however, assumes
that all data are of value and usually assumes a linear relation­
ship between the parameters and outcome. In addition it is
difficult to include time-series data in this form of analysis,
and so other tools may be better suited to bioinformatics in
the NCCU.
Neural Networks
Real-world systems rarely behave in a simple, linear fashion
throughout their entire dynamic range, and physiologic
systems are no exception. For example, if MAP is used as an
input variable in an outcome predictor, poor outcomes would
be expected for very low and very high MAP with better out­
comes in between. A linear model cannot generate this behav­
ior; instead, a nonlinear predictor such as a neural network is
needed. A neural network is essentially a nonlinear input-
output function with many tunable weights. (The “network”
simply displays the internal structure of the nonlinear func­
tion, with weights on the links and intermediate values com­
puted at the internal nodes.) The training process iteratively
modifies the weights to improve the degree of fit between the
network’s predictions and the actual training data.
Several recent studies have described the use of neural net­
works in critical care patients29 and in particular those with
acute neurologic insults. For example, Vath et al.30 used neural
networks to predict outcome after TBI for different combina­
tions of clinical and neuromonitoring parameters. Different
network models were composed using clinical data plus addi­
tional inputs for the number of events related to ICP and PbtO2
levels.30 Lang et al.31 attempted to determine whether neural
network modeling would improve outcome prediction com­
pared with logistic regression analysis. They confirmed the
importance of age, GCS, and hypotension in outcome predic­
tion. Model performance was similar. Becalick and Coats32
developed a neural network using anatomic and physiologic
predictors. Head injury, age, and chest injury were the most
important predictors. Li et al.33 compared three different
models to predict a particular TBI decision about open-skull
440 Section VII—Computers, Engineering, and the Future

GCS

,5 6,
7,
3,4 8

Pupillary Glucose level

response on the 2nd day

Unilaterally or
Bilaterally
bilaterally
normal
absent ≤169 mg/dL
>169 mg/dL
n = 119
Age
Unfav 96.6% SAH
1
>48 yrs Intracranial
≤48 yrs
Absent Present diagnosis
n = 12
Unfav 91.7% n = 67 n = 49
Glucose level Epidural hematoma Acute subdural hematoma
Unfav 1.5% Unfav 28.6%
on admission Diffuse brain injury Intracerebral hematoma
2
9 6

WBC count n = 28
≤179 mg/dL >179 mg/dL Unfav 89.3%
on admission
3
n = 10 n = 17 ≤15700 mm3 >15700 mm3
Unfav 10% Unfav 70.6%
8 4 n = 29 n = 14
Unfav 15.6% Unfav 42.8%
7 5
Fig. 45.3  Decision tree to predict outcome based on sequential partitioning of patients based on specific characteristics. Structure of the
tree is based on classification and regression analysis that creates a hierarchical set of binary decisions (in this case) based on optimizing outcome
prediction rather than separating patients based on clinically driven characteristics or cutpoints. SAH, Subarachnoid hemorrhage; Unfav, unfavorable
outcome; WBC, white blood cell. (Reprinted with permission from Rovlias A, Kotsou S. Classification and regression tree for prediction of outcome after severe head
injury using simple clinical and laboratory variables. J Neurotrauma 2004;21:886–93. Copyright Mary Ann Liebert, Inc.)

surgery: logistic regression, a multilayer perceptron neural
network, and a radial basis function neural network. Sensitiv­
ity and specificity for the regression model were lower than
for the network models to predict a neurosurgeon’s decision,
so confirming the need for a nonlinear predictor. Recently
Nelson et al.34 used mixed effects models and nonlinear (arti­
ficial neural networks) analysis to examine microdialysis data
in TBI. This analysis showed multiple perturbations in metab­
olism and that the long-term metabolic patterns were weakly
correlated to ICP and CPP; that is, factors other than pressure
and/or flow likely influence the microdialysis findings.
Decision Trees
Whereas linear models and neural networks generate smoothly
varying functions of the input variables, decision trees use
hard thresholds (or discrete inputs, such as gender or injury
type) to split the input space into distinct regions, each of
which can be treated separately from the others. The training
algorithm starts with a “stump” and adds branches to create
subsets of the training data that are homogeneous with respect
to the output variable. Decision trees are essentially flowcharts
that can be used to categorize items such as patients, as shown
in Figure 45.3. For example, a set of patients in a study might
all be grouped together at the “top” of a tree. At each branch,
the set of patients can be divided into two (e.g., “Is a patient
female or male?”) or more (e.g., “Did a patient receive 5 mg,
10 mg, or 25 mg of drug?”) subsets. These subsets can further
be subdivided. The nature and sequence of the rules can be
optimized according to various algorithms. Ultimately, the
final subsets are evaluated based on an endpoint (e.g., females
who received 5 mg of drug showed 70% improvement, females
who received 10 mg showed 80% improvement). This type of
evaluation might suggest a decision such as increasing drug
dosage for poor responders.
Decision trees are useful methods to analyze nonlinear,
multivariate data, particularly with discrete inputs, and have
been implemented with growing frequency for prognosis.
They often incorporate demographic as well as physiologic
parameters. Although not necessarily more accurate than
neural networks, trees are much easier to understand—the
patterns that the learning algorithms (e.g., demographic risk
factors or critical physiologic thresholds) find are very clear
when one looks at the main branches of the tree. Furthermore,
the prediction for any given patient is easily explained based
on the corresponding path through the tree.
 Section VII—Computers, Engineering, and the Future 441
Decision tree analysis has been used to predict TBI outcome
or verify other methods of outcome prediction. For example,
Rovlias and Kotsou 35 developed a tree model to predict neu­
rologic outcome after TBI; their results indicated that GCS
was the best predictor. The overall cross-validated predictive
accuracy was 87%. Andrews et al.36 compared decision tree
analysis and logistic regression and used decision tree analysis
to identify patient subgroups and critical values in assessed
variables after TBI. The best predictors of mortality were
duration of hypotensive, pyrexic, and hypoxemic insults.
Decision tree analysis confirmed some of the results of logistic
regression and challenged others. Temkin et al.37 developed
prediction trees of verbal IQ, Halstead’s Impairment Index,
and work status at 1 year after injury. They found that the
trees were simple enough to be used in a clinical setting, and
predictions were accurate enough for clinical utility. Other
studies have used decision tree analysis to predict memory
recovery, disability, or productive activity after TBI based on
measures of initial injury severity, tests of attention, demo­
graphic characteristics,38 or the TBI Model Systems database.39
Choi et al.40 employed a tree to predict outcomes among the
five Glasgow Outcome Scale categories. Analysis revealed that
the important prognostic factors differed among patient sub­
groups. The overall accuracy of the technique was 78%. Deci­
sion tree analysis has also been used to predict which patients
admitted to the NCCU are likely to have colonization with
methicillin-resistant Staphylococcus aureus or vancomycin-
resistant Enterococcus.41
Advanced Critical Care
Informatics II: Model-Based
Methods
Although data-driven prediction systems have obvious uses
and reflect some aspects of expert understanding, they typi­
cally lack any notion of underlying mechanisms. To give a
trivial example, the GCS, taken literally, gives incorrect pre­
dictions for a deaf-blind patient who may make no verbal
response to spoken commands. A physician obviously under­
stands how the patient’s condition affects the ability to respond
to verbal stimuli, and if the patient’s history is unknown, the
physician might well suspect the presence of such a condition
if other responses appear normal. This section describes two
methods to represent the causal processes that underlie mea­
sured data. To the extent that specific causal processes are
understood, this knowledge can be applied directly to inter­
pret data and make predictions, possibly augmented by
machine learning methods and patient-specific adaptation.
Dynamic System Models
Dynamic system models are sets of algebraic and differential
equations whose variables represent physiologic quantities.
The models specify exactly how a system evolves over time,
given the initial conditions. They are among the most intuitive
mathematical tools to develop because they are based on ordi­
nary physical understanding, for example, of pressure-volume-
flow relationships in cerebrospinal fluid dynamics, or on
experimentally derived relationships such as autoregulation.
Several examples of dynamic system models relevant to neu­
rocritical care exist. In the late 1990s, Ursino et al.42,43
produced a series of quantitatively rigorous studies on cere­
bral vascular reserve, cerebral hemodynamics during arterial
and CO2 pressure changes, cerebral vasospasm, ICP dynamics,
and relationships between CPP and autoregulation. Zhu and
Diao44 developed a model to examine brain temperature gra­
dients during selective cooling of the brain surface after head
injury. Thoman et al.45 developed the ability to visualize intra­
cranial dynamics during simulated clinical scenarios. Previous
studies had isolated physiologic variables and shown their
effects on brain dynamics, but no studies had shown the com­
bined effects of these variables on intracranial dynamics.
Finally, Zhang et al.46 investigated differences in brain response
due to impacts based on a finite element model with anatomic
features of a human head. The model predicted higher posi­
tive pressures accompanied by a large localized skull deforma­
tion at the impact site from a lateral impact compared to a
frontal impact. Overall, dynamic systems modeling has yielded
a great deal of qualitative understanding of human physiology
in general and intracranial dynamics in particular.
The drawback of classic dynamic systems models is the
need for exact and complete knowledge of the underlying
system. Whereas it is possible to estimate the numerical coef­
ficients of differential equation models by laboratory experi­
ment or by fitting the model to data from patient populations,
a generic model is seldom correct for any particular patient.
Sometimes this can be overcome by further training to fit
coefficients to patient-specific data. For example, Ursino’s
group incorporated estimation of intracranial compliance
into its predictive modeling protocol.42,43 A more serious
problem is the assumption of determinism in the model; for
example, no reasonable model can predict the exact time and
characteristics of a myocardial infarction, rupture of an aortic
aneurysm, or occurrence of a cerebral hemorrhage, because
the amount of detail and exactness of initial conditions
required are simply too great. Put another way, deterministic
models have no need for observation of the patient because
they already “know” what is going to happen.
Dynamic Bayesian Networks
A dynamic Bayesian network, or DBN, can be thought of as a
generalization of dynamic system models that allows for
uncertainty about how the underlying system state—in this
case, the patient state—evolves over time (the “transition
model”) and how that system state is manifested in sensor
measurements (the “sensor model”). This may predict “transi­
tions” between states in the ICU.47 The network structure of
the DBN reflects dependencies among the variables. In addi­
tion, each node is annotated with the conditional probability
distribution for that variable given its parents in the network.
Both network structure and conditional distributions may be
learned from data or specified based on expert knowledge.
Similar to dynamic system models, DBNs provide a natural
and intuitive format to express knowledge of causal processes.
Figure 45.4 shows a simple example to predict heart rate (HR)
from multiple sensors.
Given values for the measurement variables in a DBN
(ECG-HR, Pleth-HR [heart rate from the oxygen saturation
monitor], and SpO2 in the example), posterior probability
distributions can be inferred (by standard algorithms) for all
the unobserved variables in the model. This includes the
current state of the patient (True-HR, patient-movement), the
442 Section VII—Computers, Engineering, and the Future
Time t
Resting
HR
True-HR
Patient
movement
ECG Pleth ECG
detached detached detached
ECG-HR SpO2 Pleth-HR ECG-HR
Fig. 45.4  Qualitative structure of a dynamic Bayesian network (DBN) that models a patient’s true heart rate changing over time, its
measurement by electrocardiogram and Pleth (heart rate from the oxygen saturation monitor), and three possible causes of artifacts
in those measurements. The network also shows measured oxygen saturation (SpO2) and two causes of artifact in that measurement. It also shows
the influence of an underlying, constant patient parameter (resting heart rate) on the actual, moment-to-moment heart rate. ECG, Electrocardiograph;
HR, heart rate.
current state of the sensors (ECG-detached, Pleth-detached)
that might affect the readings, and underlying patient param­
eters (Resting-HR). Notice the important distinction between
measured and true values of physiologic variables; a DBN
typically models sensor noise, drift, miscalibration, and failure,
whereas deterministic dynamic system models typically make
no distinction between true state and measured state.
Detailed models of physiology need to be coupled to
detailed sensor models if real, artifact-ridden data are to be
understood properly. For example, careful modeling of the
process by which arterial BP is measured, including artifactual
sensor states causesd by rezeroing and blood draws, can yield
almost perfect detection and discounting of artifactual sensor
values and allows correct inference of the true BP even during
extended interruptions.48 In another study, Celi et al.49 used
this type of modeling to predict fluid requirements for criti­
cally ill patients.
Current Challenges in Physiologic
Data Analysis
There are at least three significant challenges in translating
analytic methods for clinical utility. These are the multivariate,
highly variable, and vast nature of the data generated by neu­
rocritical care monitoring. Within these challenges are issues
related to artifact detection, integration of event data (e.g.,
medical administration), and noise or variability in sensors.
Neurocritical data is multivariate (e.g., ICP, MAP), multi­
modal (e.g., numerical, text, image), and multidomain (e.g.,
physiologic, biochemical, clinical). Decisions usually cannot
be made on the basis of just one parameter. For example,
treatment of arterial pressure may change intracranial blood
volume and result in a change of ICP, and the CPP also may
be affected. Furthermore, parameters may be directly, inversely
or even sporadically correlated. Information that is multi­
modal also must be quantified and integrated when making
patient decisions. Computed tomography (CT) scan images,
while complex, can be scored, if not evaluated pixel by pixel.
However, this type of analysis may not capture the qualitative
overall impression of injury state assessed by an experienced
clinician or radiologist. Thus analysis of these data likely
requires multiple domains of expertise.
Time t+1 Time t+2...
True-HR True-HR
Patient Patient
movement movement
Pleth ECG Pleth
detached detached detached
SpO2 Pleth-HR ECG-HR SpO2 Pleth-HR
Neurocritical data can be highly variable. Such data are not
necessarily “bad,” but they can be difficult to interpret. Vari­
ance can be introduced by factors such as faulty instru­
mentation, clinical interventions, or physiologic variability.
Approaches to process missing data include interpolation/
extrapolation, random filling, removal of the record, or
inference from DBNs. Filters, simple yet robust statistical
parameters such as medians, change-point algorithms, and
information theory may be applied to simplify highly variable
data.50,51 Thus fundamental goals of physiologic informatics
must be to ensure reliability of the data and establish relation­
ships between physiologic data and clinical events.
Vast amounts of neurocritical data are collected. For
example, if HR is monitored every minute for a day, 1440 data
points are generated. If 20 parameters are continuously cap­
tured from 20 patients for a week, then 4,032,000 data points
are generated. Data rapidly reaches “genomic” orders of mag­
nitude. Ironically, even with large data volumes, neurocritical
care data can be sparse. For example, even if physiologic
parameters are recorded every minute, laboratory values may
only be measured every day, and images may only be captured
once. Furthermore, events of interest such as rapid ICP
increases are sporadic. Thus development of medical infor­
matics goes hand-in-hand with the development of data
acquisition methods and repositories of large amounts of
critical care data. Just as efforts to develop specimen and data
repositories for genetic data have had to address evolving
issues of privacy, heterogeneity, scale, and standardization of
terminology, emerging approaches to physiologic critical care
data registries for the purposes of knowledge discovery likely
will be confronted with the same regulatory issues as well.
Overcoming these challenges (Table 45.2), along with issues
of user-friendly data visualization that could be used at the
bedside for real-time clinical decision making, lies at the heart
of current and future approaches to data acquisition and
analysis in critical care.
Medical Informatics and Decision
Support—Enhancing Expertise
Medical informatics is a broad, multidisciplinary field con­
cerned with the processing of data and knowledge in medicine
 Section VII—Computers, Engineering, and the Future 443
Table 45.2  Challenges to the Development
of the Field of Neurocritical Care Informatics
Data acquisition electronically as a standard in intensive care
units
Interoperability of different monitoring devices to acquire data
Standardization of global controlled vocabularies of
terminology
Volume and resolution of data
Regulatory aspects of data collection (e.g., Health Insurance
Portability and Accountability Act and privacy boards)
Automated artifact detection and cleaning of acquired data
before analysis
Missing data (e.g., due to artifact, disconnection, different
time domains)
Varying types of data (e.g., numeric, imaging, text,
continuous, ordinal)
Integration with electronic medical records and other 1. Sorani MD, Hemphill JC 3rd, Morabito D, et al. New approaches to
patient data
Whether to use all data or discard selectively or empirically
Real-time analysis and feedback to the bedside
Clinician acceptance and use of analyzed data
and health care. A primary goal of medical informatics is to
provide decision support for clinicians.52 The field of medical
informatics can be characterized in many ways, but three large
subdomains are information systems, knowledge representa­
tion, and data analysis.53 Recently, the American Medical
Informatics Association augmented its activities to encompass
“translational bioinformatics,” including “analytic and inter­
pretive methods to optimize the transformation of increas­
ingly voluminous biomedical data into proactive, predictive,
preventive, and participatory health.”54 This effort lies at the
intersection of information systems, knowledge representa­
tion, and data analysis.
Here, the focus is on the analysis of physiologic data cap­
tured from neuromonitoring, but soon, increased efforts will
be made to integrate such data with computer-based deci­
sion support systems, EMRs, and other technologies. In
addition to standard methods for systems to communicate
(HL7) and submit claims (ICD-10), other standards could
improve the access and exchange of patient information.
Advances in the Unified Medical Language System (UMLS)
of the National Library of Medicine have made medical
record information standards (Systemized Nomenclature of
Medicine, [SNOMED]) more accessible.55
Conclusion
Neurocritical care is complex, and informatics methods hold
some promise to manage this complexity. Initial approaches
using comparatively simple analytical advances have already
demonstrated the early potential of new informatics tech­
niques. However, a recent systematic review concluded that
while prognostic models are frequently published, they are
“developed from small samples of patients, their methodolog­
ical quality is poor and they are rarely validated on external
populations. Furthermore, they are not clinically practical as
they are not presented to physicians in a user-friendly way.”56
Similarly, commercial products suffer from lack of flexibility,
problems with interoperability, complexity, and high cost.
Along with advances in neuromonitoring devices (e.g., brain-
tissue oxygen monitors, microdialysis), it is likely necessary to
develop user-friendly methods of advanced data visualization
and analysis to understand how to use these data for maximum
patient benefit. It is hoped that as medical informatics em­
braces translational bioinformatics, public access “genomic-
style” repositories of physiologic data and software could do
much to support global clinical and research efforts. As moni­
toring and neurocritical care informatics evolve, it will be
important to constantly reevaluate how they can aid decision
making, standardize care, serve as a training resource, or
provide better information access. The potential benefits are
alluring, but the complexity and regulatory, ethical, and ana­
lytic challenges should not be underestimated (Table 45.2).
References
physiological informatics in neurocritical care. Neurocrit Care 2007;7:45–52.
2. Hemphill JC, Andrews P, De Georgia M. Multimodal monitoring and
neurocritical care bioinformatics. Nat Rev Neurol 2011;7(8):451–60.
3. Diedler J, Czosnyka M. Merits and pitfalls of multimodality brain
monitoring. Neurocrit Care 2010;12(3):313–6.
4. Kumar S, Aldrich K. Overcoming barriers to electronic medical record
(EMR) implementation in the US healthcare system: a comparative study.
Health Informatics J 2010;16:306–18.
5. Gomez H, Camacho J, Yelicich B, et al. Development of a multimodal
monitoring platform for medical research. Conf Proc IEEE Eng Med Biol
Soc 2010;1:2358–61.
6. Saeed M, Villarroel M, Reisner AT, et al. Multiparameter Intelligent
Monitoring in Intensive Care II (MIMIC-II): a public-access intensive care
unit database. Crit Care Med 2011;39:952–60.
7. ASTM subcommittee F29.21. ASTM Standard F2761-09 Medical devices and
medical systems—essential safety requirements for equipment comprising
the patient-centric integrated clinical environment (ICE)—Part 1: general
requirements and conceptual model. ASTM International [online]; 2009.
http://www.astm.org/Standards/F2761.htm.
8. Chesnut RM. Avoidance of hypotension: conditio sine qua non of successful
severe head-injury management. J Trauma 1997;42:S4–9.
9. Chesnut RM, Marshall LF, Klauber MR, et al. The role of secondary brain
injury in determining outcome from severe head injury. J Trauma 1993;34:
216–22.
10. Diringer MN, Reaven NL, Funk SE, et al. Elevated body temperature
independently contributes to increased length of stay in neurologic intensive
care unit patients. Crit Care Med 2004;32:1489–95.
11. Kilpatrick MM, Lowry DW, Firlik AD, et al. Hyperthermia in the
neurosurgical intensive care unit. Neurosurgery 2000;47:850–56.
12. Robertson CS, Valadka AB, Hannay HJ, et al. Prevention of secondary
ischemic insults after severe head injury. Crit Care Med 1999;27:2086–95.
13. Struchen MA, Hannay HJ, Contant CF, et al. The relation between acute
physiological variables and outcome on the glasgow outcome scale and
disability rating scale following severe traumatic brain injury. J Neurotrauma
2001;18:115–25.
14. Manley G, Knudson MM, Morabito D, et al. Hypotension, hypoxia, and head
injury: frequency, duration, and consequences. Arch Surg 2001;136:1118–23.
15. Diringer MN. Treatment of fever in the neurologic intensive care unit with a
catheter-based heat exchange system. Crit Care Med 2004;32:559–64.
16. Barton CW, Hemphill JC, Morabito D, et al. A novel method of evaluating
the impact of secondary brain insults on functional outcomes in traumatic
brain-injured patients. Acad Emerg Med 2005;12:1–6.
17. Schmutzhard E, Engelhardt K, Beer R, et al. Safety and efficacy of a novel
intravascular cooling device to control body temperature in neurologic
intensive care patients: a prospective pilot study. Crit Care Med
2002;30:2481–88.
18. Steiner LA, Czosnyka M, Piechnik SK, et al. Continuous monitoring of
cerebrovascular pressure reactivity allows determination of optimal cerebral
perfusion pressure in patients with traumatic brain injury. Crit Care Med
2002;30:733–38.
19. Czosnyka M, Smielewski P, Kirkpatrick P, et al. Continuous monitoring of
cerebrovascular pressure-reactivity in head injury. Acta Neurochir Suppl
1998;71:74–7.
20. Diedler J, Sykora M, Rupp A, et al. Impaired cerebral vasomotor activity in
spontaneous intracerebral hemorrhage. Stroke 2009;40:815–9.
21. Smielewski, P, Czosnyka M, Steiner LA, et al. ICM+: software for on-line
analysis of bedside monitoring data after severe head trauma. Acta
Neurochir Suppl 2005;95:43–9.
444 Section VII—Computers, Engineering, and the Future
22. Buchman TG. The digital patient: predicting physiologic dynamics with
mathematical models. Crit Care Med 2009;37:1167–8.
23. Gawande A. The checklist. The New Yorker. December 10, 2007.
24. van Santbrink H, vd Brink WA, Steyerberg EW, et al. Brain tissue oxygen
response in severe traumatic brain injury. Acta Neurochir (Wien) 2003;145:
429–38.
25. Oddo M, Levine JM, Mackenzie L, et al. Brain hypoxia is associated with
short-term outcome after severe traumatic brain injury independently of
intracranial hypertension and low cerebral perfusion pressure. Neurosurgery
2011;69(5):1037–45.
26. Amin AP, Kulkarni HR. Improvement in the information content of the
Glasgow Coma Scale for the prediction of full cognitive recovery after head
injury using fuzzy logic. Surgery 2000;127:245–53.
27. Chambers IR, Jones PA, Lo TY, et al. Critical thresholds of intracranial
pressure and cerebral perfusion pressure related to age in paediatric head
injury. J Neurol Neurosurg Psychiatry 2006;77:234–40.
28. Hornero R, Aboy M, Abasolo D, et al. Interpretation of approximate entropy:
analysis of intracranial pressure approximate entropy during acute
intracranial hypertension. IEEE Trans Biomed Eng 2005;52:1671–80.
29. Cohen MJ, Grossman AD, Morabito D, et al. Identification of complex
metabolic states in critically injured patients using bioinformatic cluster
analysis. Crit Care 2010;14:R10.
30. Vath A, Meixensberger J, Dings J, et al. Prognostic significance of advanced
neuromonitoring after traumatic brain injury using neural networks.
Zentralbl Neurochir 2000;61:2–6.
31. Lang EW, Pitts LH, Damron SL, et al. Outcome after severe head injury: an
analysis of prediction based upon comparison of neural network versus
logistic regression analysis. Neurol Res 1997;19:274–80.
32. Becalick DC, Coats TJ. Comparison of artificial intelligence techniques with
UKTISS for estimating probability of survival after trauma. UK Trauma and
Injury Severity Score. J Trauma 2001;51:123–33.
33. Li YC, Liu L, Chiu WT, et al. Neural network modeling for surgical decisions
on traumatic brain injury patients. Int J Med Inform 2000;57:1–9.
34. Nelson DW, Thornquist B, MacCallum RM, et al. Analyses of cerebral
microdialysis in patients with traumatic brain injury: relations to
intracranial pressure, cerebral perfusion pressure and catheter placement.
BMC Med 2011;2(9):21.
35. Rovlias A, Kotsou S. Classification and regression tree for prediction of
outcome after severe head injury using simple clinical and laboratory
variables. J Neurotrauma 2004;21:886–93.
36. Andrews PJ, Sleeman DH, Statham PF, et al. Predicting recovery in patients
suffering from traumatic brain injury by using admission variables and
physiological data: a comparison between decision tree analysis and logistic
regression. J Neurosurg 2002;97:326–36.
37. Temkin NR, Holubkov R, Machamer JE, et al. Classification and regression
trees (cart) for prediction of function at 1 year following head trauma.
J Neurosurg 1995;82:764–71.
38. Stuss DT, Binns MA, Carruth FG, et al. Prediction of recovery of continuous
memory after traumatic brain injury. Neurology 2000;54:1337–44.
39. Brown AW, Malec JF, McClelland RL, et al. Clinical elements that predict
outcome after traumatic brain injury: a prospective multicenter recursive
partitioning (decision-tree) analysis. J Neurotrauma 2005;22:1040–51.
40. Choi SC, Muizelaar JP, Barnes TY, et al. Prediction tree for severely
head-injured patients. J Neurosurg 1991;75:251–5.
41. Minhas P, Perl TM, Carroll KC, et al. Risk factors for positive admission
surveillance cultures for methicillin-resistant Staphylococcus aureus and
vancomycin-resistant enterococci in a neurocritical care unit. Crit Care Med
2011;39(10):2322–9.
42. Ursino M, Giulioni M, Lodi CA. Relationships among cerebral perfusion
pressure, autoregulation, and transcranial Doppler waveform: a modeling
study. J Neurosurg 1998;89:255–66.
43. Ursino M, Lodi CA, Rossi S, et al. Estimation of the main factors affecting
icp dynamics by mathematical analysis of PVI tests. Acta Neurochir Suppl
1998;71:306–9.
44. Zhu L, Diao C. Theoretical simulation of temperature distribution in the
brain during mild hypothermia treatment for brain injury. Med Biol Eng
Comput 2001;39:681–7.
45. Thoman WJ, Lampotang S, Gravenstein D, et al. A computer model of
intracranial dynamics integrated to a full-scale patient simulator. Comput
Biomed Res 1998;31:32–46.
46. Zhang L, Yang KH, King AI. Comparison of brain responses between frontal
and lateral impacts by finite element modeling. J Neurotrauma 2001;
18:21–30.
47. Peelen L, de Keizer NF, de Jonge E, et al. Using hierarchical dynamic
Bayesian networks to investigate dynamics of organ failure in patients in the
intensive care unit. J Biomed Inform 2010;43:273–86.
48. Aleks N, Russell S, Madden MG, et al. Probabilistic detection of short events,
with application to critical care monitoring. In: Advances in neural
information processing systems 21. Cambridge, MA: MIT Press; 2009.
49. Celi LA, Hinske LC, Alterovitz G, et al. An artificial intelligence tool to
predict fluid requirement in the intensive care unit: a proof-of-concept
study. Crit Care 2008;12:R151.
50. Aboy M, McNames J, Thong T, et al. An automatic beat detection algorithm
for pressure signals. IEEE Trans Biomed Eng 2005;52:1662–70.
A complete list of references for this chapter can be found online at
www.expertconsult.com.
 Section VII—Computers, Engineering, and the Future 444.e1
References
1. Sorani MD, Hemphill JC 3rd, Morabito D, et al. New approaches to
physiological informatics in neurocritical care. Neurocrit Care 2007;7:45–52.
2. Hemphill JC, Andrews P, De Georgia M. Multimodal monitoring and
neurocritical care bioinformatics. Nat Rev Neurol 2011;7(8):451–60.
3. Diedler J, Czosnyka M. Merits and pitfalls of multimodality brain
monitoring. Neurocrit Care 2010;12(3):313–6.
4. Kumar S, Aldrich K. Overcoming barriers to electronic medical record
(EMR) implementation in the US healthcare system: a comparative study.
Health Informatics J 2010;16:306–18.
5. Gomez H, Camacho J, Yelicich B, et al. Development of a multimodal
monitoring platform for medical research. Conf Proc IEEE Eng Med Biol
Soc 2010;1:2358–61.
6. Saeed M, Villarroel M, Reisner AT, et al. Multiparameter Intelligent
Monitoring in Intensive Care II (MIMIC-II): a public-access intensive care
unit database. Crit Care Med 2011;39:952–60.
7. ASTM subcommittee F29.21. ASTM Standard F2761-09 Medical devices and
medical systems—essential safety requirements for equipment comprising
the patient-centric integrated clinical environment (ICE)—Part 1: general
requirements and conceptual model. ASTM International [online]; 2009.
http://www.astm.org/Standards/F2761.htm.
8. Chesnut RM. Avoidance of hypotension: conditio sine qua non of successful
severe head-injury management. J Trauma 1997;42:S4–9.
9. Chesnut RM, Marshall LF, Klauber MR, et al. The role of secondary brain
injury in determining outcome from severe head injury. J Trauma 1993;34:
216–22.
10. Diringer MN, Reaven NL, Funk SE, et al. Elevated body temperature
independently contributes to increased length of stay in neurologic intensive
care unit patients. Crit Care Med 2004;32:1489–95.
11. Kilpatrick MM, Lowry DW, Firlik AD, et al. Hyperthermia in the
neurosurgical intensive care unit. Neurosurgery 2000;47:850–56.
12. Robertson CS, Valadka AB, Hannay HJ, et al. Prevention of secondary
ischemic insults after severe head injury. Crit Care Med 1999;27:2086–95.
13. Struchen MA, Hannay HJ, Contant CF, et al. The relation between acute
physiological variables and outcome on the glasgow outcome scale and
disability rating scale following severe traumatic brain injury. J Neurotrauma
2001;18:115–25.
14. Manley G, Knudson MM, Morabito D, et al. Hypotension, hypoxia, and
head injury: frequency, duration, and consequences. Arch Surg 2001;136:
1118–23.
15. Diringer MN. Treatment of fever in the neurologic intensive care unit with a
catheter-based heat exchange system. Crit Care Med 2004;32:559–64.
16. Barton CW, Hemphill JC, Morabito D, et al. A novel method of evaluating
the impact of secondary brain insults on functional outcomes in traumatic
brain-injured patients. Acad Emerg Med 2005;12:1–6.
17. Schmutzhard E, Engelhardt K, Beer R, et al. Safety and efficacy of a novel
intravascular cooling device to control body temperature in neurologic
intensive care patients: a prospective pilot study. Crit Care Med 2002;30:
2481–88.
18. Steiner LA, Czosnyka M, Piechnik SK, et al. Continuous monitoring of
cerebrovascular pressure reactivity allows determination of optimal cerebral
perfusion pressure in patients with traumatic brain injury. Crit Care Med
2002;30:733–38.
19. Czosnyka M, Smielewski P, Kirkpatrick P, et al. Continuous monitoring of
cerebrovascular pressure-reactivity in head injury. Acta Neurochir Suppl
1998;71:74–7.
20. Diedler J, Sykora M, Rupp A, et al. Impaired cerebral vasomotor activity in
spontaneous intracerebral hemorrhage. Stroke 2009;40:815–9.
21. Smielewski, P, Czosnyka M, Steiner LA, et al. ICM+: software for on-line
analysis of bedside monitoring data after severe head trauma. Acta
Neurochir Suppl 2005;95:43–9.
22. Buchman TG. The digital patient: predicting physiologic dynamics with
mathematical models. Crit Care Med 2009;37:1167–8.
23. Gawande A. The checklist. The New Yorker. December 10, 2007.
24. van Santbrink H, vd Brink WA, Steyerberg EW, et al. Brain tissue oxygen
response in severe traumatic brain injury. Acta Neurochir (Wien)
2003;145:429–38.
25. Oddo M, Levine JM, Mackenzie L, et al. Brain hypoxia is associated with
short-term outcome after severe traumatic brain injury independently of
intracranial hypertension and low cerebral perfusion pressure. Neurosurgery
2011;69(5):1037–45.
26. Amin AP, Kulkarni HR. Improvement in the information content of the
Glasgow Coma Scale for the prediction of full cognitive recovery after head
injury using fuzzy logic. Surgery 2000;127:245–53.
27. Chambers IR, Jones PA, Lo TY, et al. Critical thresholds of intracranial
pressure and cerebral perfusion pressure related to age in paediatric head
injury. J Neurol Neurosurg Psychiatry 2006;77:234–40.
28. Hornero R, Aboy M, Abasolo D, et al. Interpretation of approximate entropy:
analysis of intracranial pressure approximate entropy during acute
intracranial hypertension. IEEE Trans Biomed Eng 2005;52:1671–80.
29. Cohen MJ, Grossman AD, Morabito D, et al. Identification of complex
metabolic states in critically injured patients using bioinformatic cluster
analysis. Crit Care 2010;14:R10.
30. Vath A, Meixensberger J, Dings J, et al. Prognostic significance of advanced
neuromonitoring after traumatic brain injury using neural networks.
Zentralbl Neurochir 2000;61:2–6.
31. Lang EW, Pitts LH, Damron SL, et al. Outcome after severe head injury: an
analysis of prediction based upon comparison of neural network versus
logistic regression analysis. Neurol Res 1997;19:274–80.
32. Becalick DC, Coats TJ. Comparison of artificial intelligence techniques with
UKTISS for estimating probability of survival after trauma. UK Trauma and
Injury Severity Score. J Trauma 2001;51:123–33.
33. Li YC, Liu L, Chiu WT, et al. Neural network modeling for surgical decisions
on traumatic brain injury patients. Int J Med Inform 2000;57:1–9.
34. Nelson DW, Thornquist B, MacCallum RM, et al. Analyses of cerebral
microdialysis in patients with traumatic brain injury: relations to
intracranial pressure, cerebral perfusion pressure and catheter placement.
BMC Med 2011;2(9):21.
35. Rovlias A, Kotsou S. Classification and regression tree for prediction of
outcome after severe head injury using simple clinical and laboratory
variables. J Neurotrauma 2004;21:886–93.
36. Andrews PJ, Sleeman DH, Statham PF, et al. Predicting recovery in patients
suffering from traumatic brain injury by using admission variables and
physiological data: a comparison between decision tree analysis and logistic
regression. J Neurosurg 2002;97:326–36.
37. Temkin NR, Holubkov R, Machamer JE, et al. Classification and regression
trees (cart) for prediction of function at 1 year following head trauma. J
Neurosurg 1995;82:764–71.
38. Stuss DT, Binns MA, Carruth FG, et al. Prediction of recovery of continuous
memory after traumatic brain injury. Neurology 2000;54:1337–44.
39. Brown AW, Malec JF, McClelland RL, et al. Clinical elements that predict
outcome after traumatic brain injury: a prospective multicenter recursive
partitioning (decision-tree) analysis. J Neurotrauma 2005;22:1040–51.
40. Choi SC, Muizelaar JP, Barnes TY, et al. Prediction tree for severely
head-injured patients. J Neurosurg 1991;75:251–5.
41. Minhas P, Perl TM, Carroll KC, et al. Risk factors for positive admission
surveillance cultures for methicillin-resistant Staphylococcus aureus and
vancomycin-resistant enterococci in a neurocritical care unit. Crit Care Med
2011;39(10):2322–9.
42. Ursino M, Giulioni M, Lodi CA. Relationships among cerebral perfusion
pressure, autoregulation, and transcranial Doppler waveform: a modeling
study. J Neurosurg 1998;89:255–66.
43. Ursino M, Lodi CA, Rossi S, et al. Estimation of the main factors affecting
icp dynamics by mathematical analysis of PVI tests. Acta Neurochir Suppl
1998;71:306–9.
44. Zhu L, Diao C. Theoretical simulation of temperature distribution in the
brain during mild hypothermia treatment for brain injury. Med Biol Eng
Comput 2001;39:681–7.
45. Thoman WJ, Lampotang S, Gravenstein D, et al. A computer model of
intracranial dynamics integrated to a full-scale patient simulator. Comput
Biomed Res 1998;31:32–46.
46. Zhang L, Yang KH, King AI. Comparison of brain responses between frontal
and lateral impacts by finite element modeling. J Neurotrauma
2001;18:21–30.
47. Peelen L, de Keizer NF, de Jonge E, et al. Using hierarchical dynamic
Bayesian networks to investigate dynamics of organ failure in patients in the
intensive care unit. J Biomed Inform 2010;43:273–86.
48. Aleks N, Russell S, Madden MG, et al. Probabilistic detection of short events,
with application to critical care monitoring. In: Advances in neural
information processing systems 21. Cambridge, MA: MIT Press; 2009.
49. Celi LA, Hinske LC, Alterovitz G, et al. An artificial intelligence tool to
predict fluid requirement in the intensive care unit: a proof-of-concept
study. Crit Care 2008;12:R151.
50. Aboy M, McNames J, Thong T, et al. An automatic beat detection algorithm
for pressure signals. IEEE Trans Biomed Eng 2005;52:1662–70.
51. Eide PK. Assessment of quality of continuous intracranial pressure
recordings in children. Pediatr Neurosurg 2006;42:28–34.
52. Ubeyli ED. Medical informatics: a model developed for diabetes education
via telemedicine. J Med Syst 2009;33:113–9.
444.e2 Section VII—Computers, Engineering, and the Future
53. Schuemie MJ, Talmon JL, Moorman PW, et al. Mapping the domain of
medical informatics. Methods Inf Med 2009;48:76–83.
54. Butte AJ. Translational bioinformatics: coming of age. J Am Med Inform
Assoc 2008;15:709–14.
55. Coonan KM. Medical informatics standards applicable to emergency
department information systems: making sense of the jumble. Acad Emerg
Med 2004;11:1198–205.
56. Perel P, Edwards P, Wentz R, et al. Systematic review of prognostic models in
traumatic brain injury. BMC Med Inform Decis Mak 2006;6:38.

Noninvasive Brain Monitoring
Marek Czosnyka, Bernhard Schmidt, Eric Albert Schmidt, Rohan Ramakrishna,
Pierre D. Mourad, and Michel Kliot
Introduction
During the first decade of the 21st century, many major
advances have revolutionized neurocritical care. These changes
have been achieved mainly because neurocritical care now is
regarded as a distinct specialty in its own right, necessitating
specialists familiar with patients with neurologic injury. Appli-
cation of new technologies toward the central nervous system
(CNS) in an effort to understand and manage the physiology
of the brain has been central to this mission. Such technolo-
gies include intracranial pressure monitoring, brain oxygen
and cerebral blood flow (CBF) monitoring, microdialysis
techniques, and jugular venous blood saturation monitoring,
to name a few. Each of these technologies used in neurocritical
care, however, have important limitations. First, their use is
generally limited to major medical centers in part because of
the personnel required to install and manage these devices, the
training required, and the cost of implementation. Second,
some of devices are frequently criticized as being too global
and therefore nonspecific in their measurements, or con-
versely too focal and then not representative of the entire
neurologic system. Third and perhaps most important, most
of the devices and in particular those to measure the gold
standard parameter, intracranial pressure (ICP), are invasive
and so there is a reluctance to insert these devices. It seems
that technologies able to take neurocritical care to the next
level will need to be noninvasive to garner widespread use, as
well as multimodal, because clinical experience has indicated
that the neurologic system cannot be optimally managed
without an integrated approach that factors the multiple
important determinants of brain pathophysiology.
Because of its anatomy, the brain is not an easy organ on
which to perform noninvasive monitoring. The surrounding
skull represents an important barrier to ultrasonic insonation,
for example. Thanks to technologic refinements, various
direct sensors used in neurocritical care have become safer and
less invasive. They have been miniaturized, sometimes made
wireless, and the infection risk has decreased as these monitors
are excluded from intracranial contents such as the cerebro-
spinal fluid (CSF). In the majority of acute neurologic injuries
the small risk of sensor placement is outweighed by the clinical
benefits of having better insight into assessment of the timing
and extent of secondary insults such as ischemia. Nevertheless,
the benefits of noninvasive brain monitoring are obvious, par-
ticularly in cases with known contraindications (e.g., hepatic
© Copyright 2013 Elsevier Inc. All rights reserved.
Chapter
46  
VII
encephalopathy in which an underlying coagulopathy increases
the risk of hemorrhage from insertion of an invasive monitor-
ing device) or when the potential benefits of invasive monitor-
ing are unclear (e.g., moderate stroke, moderate and mild
head injury, or eclampsia, among others).
This chapter reviews some of the noninvasive technologies
available now as well as on the horizon that may enhance
neurocritical care.
Noninvasive Intracranial
Pressure and Cerebral
Perfusion Pressure Monitoring
ICP can be estimated noninvasively using various methods:
(1) tympanic membrane displacement,1 (2) fontanometry,2
(3) analysis of transcranial Doppler (TCD) pulse waveform,3,4
or (4) ocular nerve sonography.5,6 However, none of these
noninvasive methods can be considered reliable or precise
enough to be used routinely in the clinical management of
intracranial hypertension. Nor is there a validated noninvasive
system that can replace invasive ICP monitors for quantifica-
tion of ICP.7 Instead current noninvasive monitors provide
supporting qualitative data that may be helpful in making
decisions about use of more invasive direct measurement
devices.
Various methods of noninvasive ICP measurement are
described in the contemporary literature8:
 Prediction of ICP from noninvasive transocular venous
and arterial hemodynamic measurements (reported 95%
confidence interval for prediction ±12 mm Hg)9
 Change in a skull diameter10 (no consistent 95% percent
limit)
 Change in blood flow velocity in the straight sinus11 (esti-
mated 95% confidence limit is low, ±8.8 mm Hg, but this
is a difficult measurement to make in adults)
 Methods based on electroencephalography (reported
lowest 95% limits for prediction ±8 mm Hg)12
 Methods based on magnetic resonance imaging (MRI)
analysis of CSF volume, as well as arterial blood inflow and
venous blood outflow from the cranium13 (there is no
consistent 95% confidence limit reported)
 MRI or ultrasound-based measurement of the optic nerve
sheath diameter14 (95% confidence limit ±20 mm Hg)
445
446 Section VII—Computers, Engineering, and the Future
Intracranial Pressure Monitoring Based
on Transcranial Doppler
Methods using TCD ultrasonography are discussed in more
detail because this technology is commonly used to noninva-
sively assess ICP. Aaslid’s design of TCD sonography in 198215
permitted bedside monitoring of CBF velocity (the instanta-
neous speed of flowing blood, typically in units of cm/second),
which may be used assuming vasospasm is absent as an index
of the CBF (the instantaneous volume of flowing blood, typi-
cally in units of mL/second; see also Chapter 30). The measure-
ment may be conducted noninvasively, repeatedly, and even
continuously. This is a “big tube technique,” which measures
arterial blood flow velocity in the major branches of the circle
of Willis, most commonly the middle cerebral artery (MCA).
A compliant branch of the MCA may be interpreted as an
implanted pressure transducer, in that the pattern of blood
flow within the MCA is modulated by transmural pressure
(i.e., cerebral perfusion pressure [CPP]) and the distal vascular
resistance (also modulated by CPP), and therefore analysis of
MCA blood flow should produce a measure of the net pressure
acting on the “transducer” (again, the MCA). But the calibra-
tion factor for the MCA, the nonlinear distortion, and other
factors (in particular, their stability in time) are unknown.
There is a reasonable correlation between the pulsatility
index (PI) of the MCA blood flow velocity and CPP after head
injury, but absolute measurements of CPP cannot be extrapo-
lated.16 Others suggest that critical closing pressure (CCP),
derived from flow velocity and arterial pressure waveform,
approximates ICP.17 The accuracy of this method has, however,
never been satisfactorily demonstrated.18 Aaslid et al. pro-
posed19 that an index of CPP could be derived from the ratio
of the amplitudes of the first harmonics of the arterial blood
pressure and the MCA velocity (detected by TCD sonography)
multiplied by the mean flow velocity in the MCA:
nCPP = A1/F1 × FVm
where F1 and A1 are first harmonic components of flow veloc-
ity and arterial pressure pulse waveforms, and FVm is time-
averaged flow velocity. However, the 95% confidence limit for
predicted values is as wide as plus or minus 20 mm Hg.
CPP affects the shape of the blood flow velocity waveform.
However, arterial pulse waveform, heart rate, tension of arte-
rial carbon dioxide (CO2), distal vascular resistance, and even
age, all affect the flow velocity (FV) waveform as well. Some
simple formulas to assess CPP noninvasively from arterial
blood pressure (ABP) and FV waveforms have been proposed.
One particular calculation has achieved satisfactory accuracy
(error less than ±10 mm Hg in more than 80% measure-
ments20; including in traumatic brain injury [TBI] patients,
shown in work by Mourad, Kliot et al.):
nCPP = MAP × FVd/FVm + 14
where FVd is diastolic flow velocity; FVm is mean flow veloc-
ity, and 14 mm Hg is “calibrating constant” for TBI patients.
nCPP is useful both to estimate absolute CPP and to
monitor changes in CPP in time (Fig. 46.1). The 95% confi-
dence limit for estimation of CPP is 12 mm Hg (prospective
trial)20 and 18 mm Hg in retrospective analysis of these
authors’ own database. Although this seems to be satisfactory
for CPP, such precision is not good enough to estimate ICP.
The pulsatility index (measured by TCD) increases as ICP
increases.3 The prediction of absolute ICP using PI may not
to be accurate enough because many other factors can
influence PI such as arterial pulse, heart rate, PaCO2, vascu-
lar tone, proximal stenosis, and spasm, among others.
However, data from Bellner et al.21 suggest a very narrow
95% confidence limit for prediction in head-injured pa-
tients: ±4.2 mm Hg. The same prediction limit derived from
data obtained at Cambridge and the University of Washing-
ton is plus or minus 20 mm Hg. Such a huge discrepancy
ought to be explained.
A moving-average model of transmission between ABP and
ICP, modified by the relationship between ABP and FV gives
a mean absolute error of 6 mm Hg and 95% confidence limit
for predictor of plus or minus 16 mm Hg.22 The method is
based on analysis of a large database of patients with homo-
geneous pathology who received ICP, ABP, and FV direct
monitoring and is most probably pathology dependent. The
method is useful for monitoring ICP changes over time (Fig.
46.2). However, a change in PaCO2, vessel spasm and proximal
stenosis can confound the results.
TCD methods (with easy access to bilateral Doppler
machines) theoretically allow the assessment of inter­
hemispheric gradients of ICP or CPP. As long as CSF com-
municates freely between different fluid cavities within the
brain, there should not be any substantial differences in
regionally measured ICP. Direct measurements of pressures in
two CSF compartments are performed rarely. In head injury,
intrahemispheric pressure gradients are reported23 with inter-
hemispheric ICP differences greater than 20 mm Hg mea-
sured in one severely injured patient.24 Other studies of
patients with focal injuries or intracranial masses show a
median interhemispheric difference of 5.5 mm Hg after cra-
niotomy but before parenchymal resection, ranging from 3 to
16 mm Hg.25 There are intrahemispheric pressure gradients in
critical closing pressure and noninvasive CPP associated with
midline shift, side of contusion (assessed using computed
tomography [CT]), or side of craniectomy.26 Both nCPP and
CCP measurements indicate that cerebral perfusion or level
of cerebrovascular dilation is greater on the side of a contusion
or more swollen brain in the case of midline shift. This may
support the hypothesis that vascular expansion and not a
brain tissue volume increase is associated with side-to-side
differences seen on CT scan. This hypothesis may be further
supported by the observation that cerebral autoregulation is
worse on the side of contusion or brain expansion.27
Intracranial Pressure Monitors Based
on the Optic Nerve
The optic nerve sheath (ONS) is an extension of the dura mater,
and the space within the sheath is continuous with the intra-
cranial contents. Therefore pressure in the ONS increases when
ICP increases.28 This can be detected by physical examination
(i.e., funduscopy) to detect papilledema. The Frisen scale can
be used to grade the severity of papilledema (0, normal, through
5, severe). Although papilledema indicates increased ICP, fun-
duscopy is not suited to monitoring patients in the NCCU.
First, it is a qualitative “point-in-time” assessment. Second, the
sensitivity and specificity to detect papilledema vary among
medical specialists and may depend on pupillary dilation.
Third, in acute brain injury, papilledema may take several
hours to develop after ICP increases. Fourth, the absence of
papilledema does not preclude increased ICP.29 In addition,
spontaneous venous pulsations, the absence of which are
 120
100
MAP 80
60
[mm Hg] 40
20
0
120
FVI 100
FVr
[cm/s] 50
0

100
80
nCPPr 60
[mm Hg] 40
20
0

100
80
nCPPI 60
[mm Hg] 40
20
0
10:00 11:00 00 13:00 14:00 15:00 16:00
Dissection Ahepatic Reperfusion
Fig. 46.1  An example of noninvasive cerebral perfusion pressure (nCPP) monitoring based on transcranial Doppler (TCD) obtained during
a liver transplant. nCPP fluctuated quite considerably during dissection and ahepatic and reperfusion phases, with an overall tendency to fall during
the operation. This was most probably caused by a gradual fall in mean arterial pressure (MAP). The difference between arterial blood pressure (ABP)
and nCPP was stable over the whole data collection period and did not exceed 20 mm Hg. Blood flow velocity on left and right was symmetrical,
with a specific increase during the reperfusion phase (vasodilation). nCPP was symmetrical with a 20-minute period of discrepancy (about 10:40 on
x-axis), when short-term irregularity of ABP was noticed and at the beginning of the reperfusion period, with absolute values of difference between
left and right never greater than 10 mm Hg. From a technical standpoint it was very difficult to keep the TCD probes in place during the 6-hour-long
operation. FVl, Flow velocity, left; FVr, flow velocity right; nCCPr, noninvasive cerebral perfusion pressure right; nCPPl, noninvasive cerebral perfusion
pressure left.

160
120
FV
80
[cm/s]
40
0
200 ABP
160
ABP 120
[mm Hg] 80
40
0
100 ICP
80
ICP 60
[mm Hg] 40
20
0
750 LDF
60
LDF 450
[a.u.] 300
150
0
nICP
100
80
nICP 60
[mm Hg] 40
20
0
0 2814 5629 8443 11258 TIME [sec]
Fig. 46.2  Traces showing noninvasive intracranial pressure (nICP) assessment in a traumatic brain injury (TBI) patient. Middle cerebral
artery (MCA) blood flow velocity (FV), arterial blood pressure (ABP), intraparenchymal intracranial pressure (ICP), and cortical blood flow using laser-
Doppler flowmetry (LDF) were recorded. Retrospectively, simulated nICP using TCD and ABP recordings indicated good agreement between real and
simulated elevations of ICP during plateau waves.
448 Section VII—Computers, Engineering, and the Future
consistent with increased ICP, are absent in about 10% of
normal subjects. Fifth, the optic nerve does not swell in atro-
phic areas30 (i.e., papilledema is not reliable in patients who
have had previous optic nerve injury). To overcome the limita-
tions of funduscopy, several techniques based on evaluation of
the ONS have been used to evaluate ICP, including (1) ophthal-
modynamometry, (2) scanning laser tomography (SLT), (3)
optical coherence tomography, and (4) ONS ultrasound. In
addition, color Doppler ultrasonography to insonate the oph-
thalmic artery and the central retinal artery and vein can be
used.31,32 Although these techniques can be used to quantify
ICP, none is suitable for ICP monitoring in the ICU at present.
Ophthalmodynamometry
This technique is an attempt to quantify ICP. First, the baseline
intraocular pressure (IOP) is measured. Then a suction cup is
placed on the eye, and IOP increased until the central retinal
vein collapses. This pressure added to the baseline IOP indi-
cates the central retinal vein venous outflow pressure that has
a linear relationship with ICP.33,34
Scanning Laser Tomography
SLT uses a laser to scan the retinal surface. It is primarily used
in patients with glaucoma but has been used to quantify pap-
illedema in patients with idiopathic intracranial hypertension
in which optic nerve height and volume measurements are
associated with opening pressure on lumbar puncture.37,38
Optical Coherence Tomography
Infrared light to measure total retinal thickness, retinal nerve
fiber layer thickness, and optic nerve head morphology is used
to quantify papilledema. This technique can be used to distin-
guish normal nerves from mild papilledema but at present is
not suited to NCCU monitoring.35,36
Optic Nerve Sheath Ultrasound
Measurement of optic nerve diameter using ultrasound, typi-
cally using a frequency between 5 and 10.5 MHz, is the most
studied noninvasive method of ICP assessment based on the
ONS.6,39-41 The technique is relatively easy to learn and interob-
server variation is low.42,43 The ONS begins to expand when
the ICP is greater than 15 mm Hg, and beyond this level is a
linear correlation between ONS enlargement and ICP
increases. An ONS diameter on ultrasound between 4.8 and
5.9 mm measured 3 mm posterior to the globe is considered
the threshold value to identify ICP greater than 20 mm Hg.14,40,41
Although this technique has promise, it has not been used to
monitor ICP in the ICU because eye injury or swelling may
limit its use, off-axis measurement of the ONS diameter pro-
vides an incorrect value, and artifacts may limit quality.44
Continuous Monitoring
of Cerebral Autoregulation
Cerebral autoregulation is an essential intrinsic, self-protecting
brain mechanism in many disorders that require NCCU care
(e.g., TBI, subarachnoid hemorrhage [SAH], and stroke). On
FVr 150 B
[cm/s] A
120
90
60
30
0
0 5 10 15 20
Time [sec.]
Fig. 46.3  The transient hyperemic response test is an example of
a noninvasive test of autoregulation. Only a transcranial Doppler
(TCD) machine is needed. The hyperemic phase (B) after short-term
compression of the ipsilateral common carotid artery (CCA) is compared
to a baseline (A) that in this case represents evidence of preserved
autoregulation. The method is not suitable for continuous monitoring
and is contraindicated when there is significant atherosclerotic carotid
disease. FVr, Flow velocity right obtained from the middle cerebral artery.
the other hand some therapies for severe TBI, in particular the
Lund concept, are based on the premise that autoregulation is
impaired.45 Methods for noninvasive assessment of autoregu-
lation through study of CBF can be divided into (1) tests in
which an external stimulus to challenge regulation mecha-
nisms is applied or (2) methods suitable for continuous moni-
toring, that is, without an external stimulus. Methods to
examine autoregulation include (1) static rate of autoregula-
tion, (2) lower body negative pressure, (3) tilt test, (4) CO2
reactivity, (5) leg-cuff test, and (6) transient hyperemic
response (Fig. 46.3). These tests provide information at a point
in time, but in many conditions, cerebral autoregulation may
fluctuate quite rapidly, particularly in NCCU patients. In such
cases, continuous monitoring of autoregulation is clinically
useful. These methods rely on analysis of spontaneous con-
tinuous fluctuations of MAP or CPP. Thanks to refinement of
computer-assisted methodology, fluctuations of MAP greater
than 5 mm Hg are suitable to retrieve information about auto-
regulation. In some cases information is less precise than in
tests using an external stimulus, but the ability to time-average
continuously obtained autoregulation indices can improve the
signal-to-noise ratio of measures of autoregulation.46
System Analysis: Phase Shift and Transfer
Function Between Arterial Pressure and
Flow Velocity Waves
In this methodology it is presumed that autoregulation
mechanisms can be described by a linear dynamic system
with arterial blood pressure as input and blood flow velocity
the output. The fast pulsatile component associated with the
cardiac cycle is probably too fast to help assess autoregula-
tion, in part because cerebral autoregulation acts primarily
through altering the diameter of the secondary cerebral arte-
rioles on time scales of a few seconds in healthy individuals,
and longer for injured patients. However, respiratory waves
(from 0.4-0.16 Hz) and slow waves (0.05-0.005 Hz) are
slower and therefore they can carry information about cere-
bral autoregulation. The first step in processing FV and ABP
time-series is therefore filtering out pulsatile components,
typically with a low pass filter of cutoff frequency of 0.5 Hz.
A noninvasive method to derive autoregulatory status from
 Section VII—Computers, Engineering, and the Future 449

110
ABP 100
[mm Hg] 90
80

60
FV
[cm/s] 55
50
10:36 10:38 10:40 10:42 10:44 10:46 10:48
Time [h:m]
0.8
Coherence 0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 1 2 3 4 5 6 7 8 9 10
Frequency [b/min]
180
160
140
120
Phase 100
[deg] 80
60
40
20
0
0 1 2 3 4 5 6 7 8 9 10
Frequency [b/min]
Fig. 46.4  With working autoregulation, slow waves of flow velocity (FV) are leading changes in arterial pressure (ABP) (upper graph), with pulse and
respiratory waveforms filtered out. Coherence shows strong association (around 0.8) between waves in FV and ABP at a frequency around 2 cycles
per minute. Therefore in system analysis, calculated ABP to FV phase shift (bottom panel) is positive (in this case little bit above 60 degrees). b/min,
Beats per minute.

natural fluctuations in MCA flow velocity and ABP involves
the assessment of phase shift between the superimposed
respiratory or slow ABP and blood FV waves during sponta-
neous recordings or slow breathing. Coherence function,
which is a Fourier transform of normalized cross-correlation
between ABP and FV series is calculated. The modulus of
coherence indicates how strong components of given fre-
quency are associated with each other. Another transforma-
tion method is a transfer function between ABP and FV. It is
a complex function, having its gain and phase shift. Phase
indicates which variable changes before another. A zero-
degree phase shift (e.g., both ABP and FV change at the same
time, without delay) indicates absent autoregulation, whereas
a positive phase shift (>30 degrees, i.e., FV changes before
ABP) indicates intact autoregulation47,48 (Fig. 46.4). Higher
FV (output) and ABP (input) transfer function also indicate
worse autoregulation.49 In both methods, the absolute value
of coherence between FV and MAP should be more than 0.5
in the analyzed frequency bandwidth. Both methods presume
a linear relationship between changes in FV and MAP within
a studied range of fluctuations.
spontaneous ABP fluctuations, values of ARI are obtained by
fitting the second-order linear model proposed by Tiecks
et al.50 that describes the FV response to a step-change in ABP.
Transfer function analysis is used to quantify the dynamic
relationship between mean ABP (input) and mean FV
(output). The inverse Fourier transformation then is per-
formed to obtain the FV impulse response in the time domain.
Impulse response is in turn integrated to yield an estimate of
the FV response to a hypothetical step-change in ABP. Each of
the 10 models,50 corresponding to ARI values from 0 (absence
of autoregulation) to 9 (best autoregulation) is fitted to the
first 10 seconds of the FV step response and the best fit, as
selected by the minimum squared error, is considered the
representative value of ARI for that segment of data. A similar
technique using repetitive thigh cuff tests is able to generate
rapid increases or decreases of MAP of plus or minus
15 mm Hg and so examine the symmetry of cerebral auto-
regulation (i.e., is there an equal compensatory response to
abrupt increases or decreases in ABP).51 Alternatively, inter-
mittent application of thigh cuffs using square wave sequences
can be used to modulate ABP and so assess CA.52

Calculation of Autoregulation Index Time Correlation Method (Mx Index)

Autoregulation index (ARI) is an “extension” of system analy- This method is based on assumptions that the predominant
sis using modeling of the step-response of the system generat- source of FV fluctuations over a short period of time is MAP
ing changes of flow velocity from changes in arterial pressure. fluctuations. By continuous monitoring over consecutive
To assess dynamic cerebral autoregulation (CA) based on time-averaged samples of FV and MAP (30 to 60, 5- to
450 Section VII—Computers, Engineering, and the Future

100
ABP 90
[mm Hg] 80
70
60
ICP 40
[mm Hg]
20

60
CPP 50
[mm Hg] 40
30
80
FV
[cm/s] 60
40
1
Mx 0.5
0
11:10 11:20 11:30 11:40 11:50 12:00
A Time [h:m]

120
ABP 100
[mm Hg] 80
60
25
ICP 20
[mm Hg] 15

25
CPP 20
[mm Hg] 15
10
120
FV 100
[cm/s] 80
60
40
1
Mx 0.5
0
B 13:56 14:00 14:04 14:08 Time [h:m]
Fig. 46.5  Autoregulation is a dynamic process and can be disturbed over relatively short time periods, such as during plateau waves of intracranial
pressure (ICP) (A). The Mx index increases during the plateau wave and then recovers to baseline values around 0. A similar observation occurs during
hypotension (B). These examples illustrate why, in neurocritical care, autoregulation should be monitored continuously rather than occasionally tested.
ABP, Arterial blood pressure; CCP, cerebral perfusion pressure; FV, flow velocity; Mx, autoregulation.

10-second averages are usually taken), a correlation coefficient
between mean MAP and mean FV is calculated. This coeffi-
cient termed the mean index (Mxa; Mx used in head injury53
is a correlation between CPP and FV. Mxa can be calculated
noninvasively, providing a noninvasive measure of blood pres-
sure is available). A positive coefficient signifies a positive
association between FV and MAP (i.e., disturbed autoregula-
tion). A zero or negative correlation coefficient (i.e., absent or
negative association) implies intact autoregulation. The calcu-
lation may be repeated with a moving time window in which
analysis occurs; therefore Mxa may form new variable and
indicate changes of cerebral autoregulation with time. This
index seems to be ideal to monitor transient changes
in autoregulation in response to secondary insults such as
arterial hypotension or intracranial hypertension (Fig. 46.5).
Both Mxa and ARI methods are associated with TBI
outcome.53,54 Some studies, however, suggest that systolic flow
indices (Sx and Sxa) demonstrate a stronger association with
outcome after TBI than the mean flow indices (Mx and Mxa).55
Wavelet Analysis
When the time series are either nonstationary or nonlinearly
related to one another, the justification for transfer function
theory is lost. To address this, a procedure has been introduced
that overlaps with transfer function theory in the linear
 Section VII—Computers, Engineering, and the Future 451
90
FV
[cm/s]
80
70
62
TOI 60
[%]
58
12:45 12:50 12:55
Fig. 46.6  Contemporary near infrared machines now are faster, are better able to reduce the influence of extracranial circulation, and
have an improved signal-to-noise ratio. A slow fluctuation of blood flow velocity (FV) recorded by transcranial Doppler (0.05-0.005 Hz), from
which information for continuous monitoring of cerebral autoregulation is generated, is replicated, without visible phase shift, by the tissue oxygen
index (TOI) monitored by near infrared spectroscopy (NIRS). It suggests that TOI can be used as a noninvasive index of autoregulation.
domain, but in this method complex continuous wavelet
transforms are used to study the phase difference between
fluctuations in ABP and MCA FV. Its application does not
require the assumption that the relation between the two time
series is or is not linear. Three parameters are used to charac-
terize recorded signals: (1) variability of the signal, (2) syn-
chronization that is analogous of coherence between two
variables, and (3) wavelet “gain,” which characterizes amplifi-
cation of the output signal, flow velocity compared to input
(i.e., arterial pressure).56 Increased gain (with good coherence)
may be interpreted as worse cerebral autoregulation.
Multimodal Decomposition
To further overcome problems related to nonstationary and
nonlinearity, a novel computational method called multi-
modal pressure-flow (MMPF) analysis was developed to study
the ABP-FV relationship during the Valsalva maneuver. The
MMPF method enables evaluation of autoregulatory dynam-
ics based on instantaneous phase analysis of blood pressure
(BP) and blood FV oscillations induced by the intervention.
While applying this technique, a characteristic phase lag
between blood flow volume and BP oscillations induced by
the Valsalva maneuver was found in healthy subjects, and this
phase lag was reduced in patients with hypertension and
stroke.57 These findings suggested that ABP-FV phase lag
could serve as an index of CA. The same method can be used
for continuous monitoring of CA, analyzing spontaneous
fluctuation of ABP.
Possible Use of Near Infrared Spectroscopy
Transcranial Doppler is operator dependent, and in some
patients it is difficult to find and insonate the MCA. Further-
more, once found the ultrasound probe may shift in position
over time scales of a few minutes to hours. Therefore other
noninvasive modalities proportional to changes in CBF have
been used instead of blood flow velocity. For example, near
infrared spectroscopy (NIRS)–derived oxygenated hemoglo-
bin (HbO2), tissue oxygen index (TOI), and cerebral oxygen-
ation index (CO) are associated with arterial oxygen saturation,
13:00 13:05 13:10 13:15
Time [h:m]
CBF, cerebral oxygen diffusivity, and CMRO2. Over a short
period of time (a few minutes), CBF probably is the major
determinant of fluctuations of detected TOI (Fig. 46.6). NIRS
optodes do not require precise focus on the MCA or continu-
ous correction for shifts. Preliminary studies based on slow
respiration in patients with carotid artery stenosis,58 transfer
functions between slow waves in neonates,59 and a moving
linear correlation coefficient between CPP and INVOS cere-
bral oximeter waveforms in a pig model suggest that artifact-
free continuous monitoring of CA using NIRS can become
clinically feasible.60 For example, Brady et al.60 defined a cere-
bral oximetry index (COx) in pigs that had 92% sensitivity
(73%-99%) and 63% specificity (48%-76%) to detect loss of
autoregulation associated with hypotension when COx was
above a threshold of 0.36. Further studies in a swine model of
hypoxic-asphyxic cardiac arrest suggest that use of NIRS to
calculate the lower limit of autoregulation (LLA) is compa-
rable to LLA calculated using laser Doppler flow measure-
ments.61 In the clinical environment NIRS-based studies in
adults who undergo cardiac surgery suggest there is a wide
range of MAP at the LLA62 thus emphasizing a role for moni-
toring rather than an empiric choice of MAP.
NIRS also has been used to assess cerebrovascular reactivity
noninvasively using the total hemoglobin reactivity index
(THx) that is analogous to the pressure reactivity index (PRx).
This NIRS-based cerebrovascular reactivity index THx can be
used as a noninvasive substitute for PRx, but only during
phases with sufficient slow wave power in the input signal.63
Brain Oxygenation
NIRS is a noninvasive method to study brain oxygen satura-
tion (see Chapter 33). In this method an NIRS sensor is
applied to the forehead and its probe emits and detects wave-
lengths of light millimeters from its tip. As the near infrared
absorption characteristics of hemoglobin are known, the
system calculates brain tissue oxygen saturation by measuring
differences in intensity of light as it passes through the brain.
The sensor then converts these wavelengths to a measurement
that represents regional oxygen saturation. The saturations
range from 15% to 95%. In the brain the major source of
452 Section VII—Computers, Engineering, and the Future

80

ABP 60
[mm Hg] 40
20
0
65
60
FV 55
[cm/s] 50
45
40
35
30
71
70
TOI 69
[%] 68
67
66

5.5
5
EtCO2 4.5
[kPa] 4
3.5
3
17:36 17:39 17:42 17:45 17:48 17:51
Time [h:m]
Fig. 46.7  Line graphs illustrating carbon dioxide (CO2) reactivity testing with a near infrared spectroscopy (NIRS) machine. The increase
in blood flow velocity (FV) in the hypercapnic phase and tissue oxygen index (TOI) are almost correlated. If calibration of TOI is stable across a population
of patients, NIRS may become a viable test cerebrovascular reactivity. ABP, Arterial blood pressure; etCO2, end-tidal carbon dioxide.

tissue oxygen content is the blood, which is influenced by
brain oxygen delivery, consumption, and the arteriovenous
blood volume ratio in the microcirculation.
There is a larger experience with this technology in cardiac
surgery patients and in the pediatric population than in
NCCU use. Such studies indicate that optimal saturations are
about 75% and that levels less than 50% or a 20% reduction
from baseline are associated with worse neurologic outcome.64,65
These values are consistent with animal data showing that
baseline values are about 70%, cerebral lactate increases when
NIRS brain oxygen saturation is 45%, the electroencephalo-
gram (EEG) changes when saturation is 42%, and adenosine
triphosphate (ATP) levels are reduced when saturation is
33%.66 In patients with severe TBI, noninvasive transcranial
oxygen saturation (StcO2) values greater than 70% are associ-
ated with better outcome and fewer interventions to treat ICP.
However, the difference in StcO2 between those who died or
survived was only 10% (70% vs. 60%).67 NIRS also can be used
to detect peri-infarct depolarizations (PIDs) in ischemic
stroke. This has significant implications; if penumbral isch-
emia can be predicted by PIDs and is associated with NIRS
recordings, then peri-infarct tissue may be salvaged by therapy
to improve cerebral perfusion.54
In adults there are several limitations to NIR-derived cere-
bral oxygen levels: (1) high scattering and attenuation of light
in tissue, especially the cranium, (2) the difficulty of estimat-
ing the path length of light through tissue, (3) the inability
to distinguish between blood contained in arteries and veins,
and (4) contamination of NIRS readings by extracranial
tissues. Indeed, Kirkpatrick et al. in a study of patients under-
going carotid endarterectomy, showed that the specificity or
sensitivity of the NIRS technique was very limited unless the
extracranial tissues were removed from the measurement, in
this study made possible by clamping the external carotid
artery (ECA).68 When the ECA was clamped, there was better
concordance between ischemia, defined by TCD and EEG
and NIRS readings. Newer NIRS designs apply an algorithm
called spatially resolved spectroscopy, to calculate the tissue-
oxygen index. This index appears independent of the extra-
cranial circulation and is associated with a CBF decrease
during carotid artery clamping.69 Although this makes this
technique useful during carotid surgery, the application of
NIRS as a noninvasive modality in NCCU patients is still
being elucidated but includes detection of secondary cerebral
ischemia, assessment of cerebrovascular reactivity to PaCO2
changes (Fig. 46.7), or assessment of cerebral pressure
autoregulation.70,71
Imaging
Imaging modalities offer the advantage of giving both quan-
titative and qualitative measurements of cerebral physiology.
The disadvantage of any imaging modality is that only a snap-
shot in time is examined—they cannot provide continuous
data. Nonetheless, different imaging modalities can aid in
noninvasive monitoring of patients in the NCCU.
Magnetic Resonance Imaging
The role of MRI in neurocritical care is discussed in Chapter
28. In particular, advanced MRI techniques can help in iden-
tifying patients with severe diffuse axonal injury (DAI) and in
 Section VII—Computers, Engineering, and the Future 453
outcome prediction. For example, advanced techniques such
as diffusion tensor imaging (DTI) are more sensitive than
conventional MRI sequences to detect white matter injury.
These techniques can also be used to examine the extent of
structural integrity after hypoxic-ischemic injury and thus
inform care.72 There are a variety of imaging sequences but
one, MRI spectroscopy (MRS), lends itself to noninvasive
monitoring of cellular function. In MRS intracellular metabo-
lites such as N-acetyl aspartate (NAA), choline, creatine (Cr),
and lactate are measured. These metabolites provide a reliable
assessment of cellular function: (1) NAA indicates the propor-
tion of dead or living neuronal cells; (2) choline levels are
associated with cell membrane activity; (3) Cr provides insight
into global metabolic level; and (4) increased lactate suggests
cellular ischemia. In TBI, the NAA/Cr ratio is decreased by as
much as 20% in areas of DAI.73 The combination of MRI and
MRS also can aid in TBI outcome prediction, particularly
when the MRI is near normal.74
With the advent of functional MRI (fMRI), it has become
possible to assess the integrity of vascular coupling and neural
function at risk in areas of brain injury and stroke. Thus fMRI
can help assess function of tissues at risk noninvasively. fMRI
can be difficult to perform in sedated patients but spontane-
ous blood oxygen level–dependent (BOLD) fluctuations can
be observed without patient cooperation (i.e., resting state
fMRI) The BOLD contrast originates from intravoxel mag-
netic field inhomogeneity induced by paramagnetic deoxyhe-
moglobin in erythrocytes. This imaging is based on the
concept that neural activity is intricately coupled to regional
CBF. When a task is performed, regional CBF increases dis-
proportionately, which can be measured by techniques such
as arterial spin labeling (ASL). The CBF increase overcompen-
sates for the stimulus-evoked increase in oxygen consumption
needed to fuel the elevated neural activity relative to basal
conditions. In animal studies of cerebral ischemia, neural
activity can be lost in primary somatosensory cortex but vas-
cular coupling remains intact before changes in apparent dif-
fusion coefficient on MRI occur,75 that is, fMRI could be a
predictor of at-risk brain. In TBI, intact task-correlated
sensory and cognitive BOLD hemodynamic responses to
stimuli have been observed in a comatose patient in whom
electrophysiologic studies indicated absent thalamocortical
processing, suggesting that fMRI may help predict outcome in
nonresponsive TBI patients.76
Other Noninvasive Methods of Metabolic
and Flow Imaging
Positron emission tomography (PET) scanning is an imaging
modality that uses cellular consumption of substances such as
glucose (18 F [fluorodeoxyglucose]) as its imaging substrate
(see Chapter 29). This allows identification of glucose use in
the brain. Alternatively, oxygen metabolism can be studied
using PET by injection of 15-O–labeled water. Regional CBV
can be assessed by inhalation of 15-O–labeled carbon mon-
oxide and CMRO2 measured using 15-O–labeled oxygen.77
PET can then be combined with CT or MRI for dual imaging
that gives more accurate anatomic localization. Single photon
emission computed tomography (SPECT) is another “meta-
bolic imaging technique” in which technetium-99-m-
hexamethyl propyleneamine is intravenously injected into
the patient’s bloodstream (see Chapter 29). Whereas PET is
quantitative, SPECT is qualitative but can be used to examine
for hypoperfusion and so is suited to examination for condi-
tions such as vasospasm after SAH, particularly when per-
formed with TCD evaluation. PET and SPECT have been used
in a variety of neurologic diseases including Alzheimer’s
disease, Huntington’s chorea, cerebral microangiopathy, and
TBI.78-82 In TBI, brain hypometabolism is frequent and the
severity of trauma correlates well with degree of hypometabo-
lism.83,84 This can be detected using PET or SPECT, although
the results may always be concordant such as when edema
compromises blood flow without causing ischemia. In this
case PET would be normal but SPECT abnormal.
In CT perfusion imaging, iodinated contrast is adminis-
tered and then regions of interest are referenced to known
blood vessels such as the pericallosal arteries or superior sagit-
tal sinus. From these data, values such as CBV, CBF, and mean
transit time (MTT) can be generated.85,86 If the perfusion scan
is obtained at two different blood pressures, the perfusion
characteristics can be examined for loss of autoregulation. In
normal autoregulation, reduced CBF is thought to trigger
vasodilation with a subsequent increase of CBV elevation to
maintain adequate tissue perfusion. Reduced CBV in the pres-
ence of ischemia in turn indicates that autoregulation is
impaired.
New Directions: Optoacoustic Techniques
The optoacoustic (or photoacoustic) technique uses short
near-infrared laser pulses to generate ultrasonic waves in
tissue due to fast thermoelastic expansion of heated volumes.
Using advanced signal processing, Petrov et al.87 demon-
strated a good correlation with known oxygenation values in
the internal jugular vein and those obtained by their opto-
acoustic probe in normal volunteers. Further development of
this technique could provide an alternative to invasive
methods of brain oxygenation monitoring and provide a
noninvasive insight into edema or hemodynamic changes in
the brain.88-90
Electrophysiology
in Neurocritical Care
Evaluation of brain function using electrophysiologic tech-
niques such as the EEG, bispectral index (BIS), evoked poten-
tial monitoring, and the rheoencephalogram is noninvasive
and these techniques are useful adjuncts to patient care in the
NCCU.
EEG and BIS
Electrical activity continuously emanates from the superficial
cerebral cortex; this can be examined using the EEG. Several
lines of evidence provide a compelling argument for bedside
continuous EEG (cEEG) in the NCCU to detect seizures,
nonconvulsive seizures, ICP changes, and cerebral ischemia,
outcome prediction, and to titrate therapies such as barbitu-
rates in a variety of conditions such as TBI, intracerebral
hemorrhage, stroke, SAH, CNS infections, and after cardiac
arrest91-98 (see also Chapter 25). cEEG provides good tempo-
ral and spatial resolution but can be technically difficult to
implement and is susceptible to drug effects and artifacts
454 Section VII—Computers, Engineering, and the Future
such as eye movements, patient movement, or electrical
interference.99 New software algorithms make it easier for
non-electrophysiologists and non-epileptologists to identify
relevant information (e.g., compressed spectral array for sei-
zures and relative alpha variability for ischemia). However,
vast amounts of data are still generated and successful use of
cEEG requires efficient computer networks and information
technology (IT) support and is resource dependent. This may
limit cEEG use to larger centers.100 However, a decreased
length of stay may offset this increased cost.101
BIS represents a summary of EEG activity in which 0 rep-
resents a flat EEG and 100 represents normal wakefulness. To
perform BIS, a sensor is placed over the forehead from which
data are gathered. Although not as sensitive as cEEG, BIS is
simple and values greater than 60 are associated with better
outcome and better control of ICP.102
Evoked Potentials
Evoked potential monitoring includes modalities such as
somatosensory evoked potentials (SSEPs), visual evoked
potentials (VEPs), auditory evoked potentials (AEPs), and
brainstem auditory evoked potentials (BAEPs; see Chapter
24). Evoked potentials (EPs) differ from EEG in that they
represent the brain’s response to repetitive stimuli along a
specific nerve pathway. In addition EPs are an order of mag-
nitude smaller in voltage than the voltages seen on EEG. SSEPs
are the most frequently used of these methods in the intensive
care unit (ICU). In this technique, responses to median nerve
(or tibial nerve) stimulation are recorded at the cortical level.
A variety of parameters are measured and useful in the NCCU:
cervical (C7, C2) and cortical evoked potentials; the interpeak
latencies C2/N20, also referred to as the central conduction
time (CCT); and the interrelation of the amplitudes N20/N13.
SSEPs may be abnormal even when there are demonstrable
lesions on imaging studies. EP monitoring is primarily used
to predict outcome rather than as a guide to therapy. In
general, a significant prolongation of CCT, reduced amplitude
of cortical responses, a low amplitude quotient N20/N13, and
bilateral loss of the cortical response are associated with poor
outcome.102
Rheoencephalography
Disrupted cerebral vasomotor autoregulation is associated
with poor outcome. CT perfusion (discussed earlier) or elec-
trical impedance studies, namely rheoencephalography (REG)
can be used to examine this phenomenon. REG works by
applying electrodes on the scalp in a bipolar arrangement and
is based on monitoring pulse-synchronous variations in
cranial electrical impedance over time to derive CBF measure-
ments. Animal studies suggest REG can be used to noninva-
sively assess CBF in intracerebral hemorrhage and to assess
compliance of the cerebral arterial bed.103 Scalp thickness,
however, can affect the measurement quite significantly,
despite rigorous efforts to mathematically exclude it104 and so
REG is infrequently used in the ICU.
Conclusion
A variety of invasive intracranial monitors are used in the
NCCU. At present invasive ICP measurement is the gold
standard monitor, and in TBI, guidelines recommend that ICP
should be monitored invasively in patients with severe injury
(i.e., Glasgow Coma Scale score ≤8). Similar indications are
used for patients in coma associated with pathologies such as
SAH or stroke. For other patients—those with moderate
TBI—the role for invasive ICP monitoring is unclear. In these
patients noninvasive estimation of ICP may help select patients
for invasive monitoring. In addition, noninvasive monitoring
may be useful when the risk of invasive monitoring is high
such as in hepatic encephalopathy.
There is a relatively long history and several approaches—
imaging (CT and MRI), TCD, NIRS, evoked potentials, EEG—
and techniques to evaluate the optic nerve sheath among
others in noninvasive monitoring of intracranial hemody-
namics. These techniques remain largely exploratory in large
part because none of methods are accurate enough and easy
to use at same time. For example, ophthalmodynamometry,
although precise, is unsuitable for continuous monitoring and
cumbersome for repeated measurement. In addition, the mea-
sured characteristic often is not only a function of ICP but
also other factors such as ABP (TCD, ultrasound) or func-
tional integrity of a specific brain region (EEG). At present
ultrasound measurements of the ONS diameter and Doppler
flow are the most promising noninvasive techniques. These
methods may be useful in select patients, but future improve-
ments will depend on improvement in accuracy. Ideally, in the
future noninvasive monitoring could replace invasive intra-
cranial monitors.
References
1. Samuel M, Burge DM, Marchbanks RJ. Tympanic membrane displacement
testing in regular assessment of intracranial pressure in eight children with
shunted hydrocephalus. J Neurosurg 1998;88(6):983–95.
2. Plandsoen WCG, de Jong DA, Maas AIR, et al. Fontanelle pressure
monitoring in infants with the Rotterdam teletransducer: a reliable
technique. Med Prog Technol 1987;13:21–271.
3. Klingelhofer J, Conrad B, Benecke R, et al. Evaluation of intracranial
pressure from transcranial Doppler studies in cerebral disease. J Neurol
1988;235:159–62.
4. Prunet B, Asencio Y, Lacroix G, et al. Noninvasive detection of elevated
intracranial pressure using a portable ultrasound system. Am J Emerg Med
2011; Jun 10. Epub ahead of print.
5. Soldatos T, Chatzimichail K, Papathanasiou M, et al. Optic nerve
sonography: a new window for the non-invasive evaluation of intracranial
pressure in brain injury. Emerg Med J 2009;26(9):630–4. Review.
6. Moretti R, Pizzi B. Ultrasonography of the optic nerve in neurocritically ill
patients. Acta Anaesthesiol Scand 2011;55(6):644–52.
7. Behrens A, Lenfeld N, Ambarki K, et al. Trans-cranial Doppler pulsatility
index: not an accurate method to assess intracranial pressure. Neurosurgery
2010;66(6):1050–7.
8. Rosenberg JB, Shiloh AL, Savel RH, et al. Non-invasive methods of
estimating intracranial pressure. Neurocrit Care 2011;15(3):599–608.
9. Ragauskas A, Daubaris G, Dziugys A, et al. Innovative non-invasive method
for absolute intracranial pressure measurement without calibration. Acta
Neurochir Suppl 2005;95:351–61.
10. Ueno T, Ballard RE, Shuer LM, et al. Intracranial pressure dynamics during
simulated microgravity using a new noninvasive ultrasonic technique.
J Gravit Physiol 1998;5(1):P39–40.
11. Schoser BG, Riemenschneider N, Hansen HC. The impact of raised
intracranial pressure on cerebral venous hemodynamics: a prospective venous
transcranial Doppler ultrasonography study. J Neurosurg 1999;91(5):744–9.
12. Zhao YL, Zhou JY, Zhu GH. Clinical experience with the non-invasive ICP
monitoring system. Acta Neurochir Suppl 2005;95:351–3.
13. Raksin PB, Alperin N, Sivaramakrishnan A, et al. Noninvasive intracranial
compliance and pressure based on dynamic magnetic resonance imaging of
blood flow and cerebrospinal fluid flow: review of principles,
implementation, and other noninvasive approaches. Neurosurg Focus
2003;14(4):e4.
 Section VII—Computers, Engineering, and the Future 455
14. Geeraerts T, Merceron S, Benhamou D, et al. Non-invasive assessment of
intracranial pressure using ocular sonography in neurocritical care patients.
Intensive Care Med 2008;34(11):2062–7.
15. Aaslid R, Markwalder TM, Nornes H. Noninvasive transcranial Doppler
ultrasound recording of flow velocity in basal cerebral arteries. J Neurosurg
1982;57(6):769–74.
16. Chan KH, Miller DJ, Dearden M, et al. The effect of changes in cerebral
perfusion pressure upon middle cerebral artery blood flow velocity and
jugular bulb venous oxygen saturation after severe brain trauma.
J Neurosurg 1992;77:55–61.
17. Thees C, Scholz M, Schaller MDC, et al. Relationship between intracranial
pressure and critical closing pressure in patients with neurotrauma.
Anesthesiology 2002;96(3):595–9.
18. Czosnyka M, Smielewski P, Piechnik S, et al. Critical closing pressure in
cerebrovascular circulation. J Neurol Neurosurg Psychiatry
1999;66(5):606–11.
19. Aaslid R, Lundar T, Lindegaard K-F, et al. Estimation of cerebral perfusion
pressure from arterial blood pressure and transcranial Doppler recordings.
In: Miller JD, Teasdale GM, Rowan JO, et al, editors. Intracranial pressure VI.
Berlin/Heidelberg: Springer Verlag; 1986. p. 229–31.
20. Schmidt EA, Czosnyka M, Gooskens I, et al. Preliminary experience
of the estimation of cerebral perfusion pressure using transcranial
Doppler ultrasonography. J Neurol Neurosurg Psychiatry 2001;70(2):
198–204.
21. Bellner J, Romner B, Reinstrup P, et al. Transcranial Doppler sonography
pulsatility index (PI) reflects intracranial pressure (ICP). Surg Neurol
2004;62(1):45–51; discussion 51.
22. Schmidt B, Klingelhofer J, Schwarze JJ, et al. Noninvasive prediction of
intracranial pressure curves using transcranial Doppler ultrasonography and
blood pressure curves. Stroke 1997;28(12):2465–72.
23. Reulen HJ, Graham R, Spatz M, et al. Role of pressure gradient and
bulk flow in dynamics of vasogenic brain edema. J Neurosurg 1977;46:
24–35.
24. Carhuapoma JR, Qureshi AI, Bhardwaj A, et al. Interhemispheric intracranial
pressure gradients in massive cerebral infarction. J Neurosurg Anesthesiol
2002;14(4):299–303.
25. Bundgaard H, Cold GE. Studies of regional subdural pressure gradients
during craniotomy. Br J Neurosurg 2000;14(3):229–34.
26. Kumar A, Schmidt EA, Hiler M, et al. Asymmetry of critical closing
pressure following head injury. J Neurol Neurosurg Psychiatry 2005;76(11):
1570–3.
27. Schmidt EA, Czosnyka M, Steiner LA, et al. Asymmetry of pressure
autoregulation after traumatic brain injury. J Neurosurg 2003;99(6):
991–8.
28. Hansen HC, Helmke K, Kunze K. Optic nerve sheath enlargement in acute
intracranial hypertension. Neuroophthalmology 1994;14:345–54.
29. Steffen H, Eifert B, Aschoff A, et al. The diagnostic value of optic disc
evaluation in acute elevated intracranial pressure. Ophthalmology 1996;103:
1229–32.
30. Hayreh SS. Optic disc edema in raised intracranial pressure. V. Pathogenesis.
Arch Ophthalmol 1977;95:1553–65.
31. Querfurth HW, Lagrèze WD, Hedges TR, et al. Flow velocity and pulsatility
of the ocular circulation in chronic intracranial hypertension. Acta Neurol
Scand 2002;105:431–40.
32. Miller MM, Chang T, Keating R, et al. Blood flow velocities are reduced in
the optic nerve of children with elevated intracranial pressure. J Child
Neurol 2009;24:30–5.
33. Querfurth HW, Arms SW, Lichy CM, et al. Prediction of intracranial
pressure from noninvasive transocular venous and arterial hemodynamic
measurements: a pilot study. Neurocrit Care 2004;1:183–94.
34. Querfurth HW, Lieberman P, Arms S, et al. Ophthalmodynamometry for
ICP prediction and pilot test on Mt. Everest. BMC Neurol 2010 1;10:106.
35. Scott CJ, Kardon RH, Lee AG, et al. Diagnosis and grading of papilledema in
patients with raised intracranial pressure using optical coherence
tomography vs clinical expert assessment using a clinical staging scale. Arch
Ophthalmol 2010;128:705–11.
36. Rebolleda G, Munoz-Negrete FJ. Follow-up of mild papilledema in
idiopathic intracranial hypertension with optical coherence tomography.
Invest Ophthalmol Vis Sci 2009;50:5197–200.
37. Trick GL, Vesti E, Tawansy K, et al. Quantitative evaluation of papilledema in
pseudotumor cerebri. Invest Ophthalmol Vis Sci 1998;39:1964–71.
38. Heckmann JG, Weber M, Junemann AG, et al. Laser scanning tomography of
the optic nerve vs CSF opening pressure in idiopathic intracranial
hypertension. Neurology 2004;62:1221–3.
39. Tayal VS, Neulander M, Norton HJ, et al. Emergency department
sonographic measurement of optic nerve sheath diameter to detect findings
of increased intracranial pressure in adult head injury patients. Ann Emerg
Med 2007;49:508–14.
40. Moretti R, Pizzi B, Cassini F, et al. Reliability of optic nerve ultrasound for
the evaluation of patients with spontaneous intracranial hemorrhage.
Neurocrit Care 2009;11:406–10.
41. Rajajee V, Vanaman M, Fletcher JJ, et al. Optic nerve ultrasound for the
detection of raised intracranial pressure. Neurocrit Care 2011;Jul 19. Epub
ahead of print.
42. Potgieter DW, Kippin A, Ngu F, et al. Can accurate ultrasonographic
measurement of the optic nerve sheath diameter (a non-invasive measure of
intracranial pressure) be taught to novice operators in a single training
session? Anaesth Intensive Care 2011;39:95–100.
43. Ballantyne SA, O’Neill G, Hamilton R, et al. Observer variation in the
sonographic measurement of optic nerve sheath diameter in normal adults.
Eur J Ultrasound 2002;15:145–9.
44. Copetti R, Cattarossi L. Optic nerve ultrasound: artifacts and real images.
Intensive Care Med 2009;35:1488–9.
45. Sharma D, Vavilala MS. Lund concept for the management of traumatic
brain injury: a physiological principle awaiting stronger evidence.
J Neurosurg Anesthesiol 2011;23(4):363–7.
46. Aoi MC, Matzuka BJ, Olufsen MS. Toward online, noninvasive, nonlinear
assessment of cerebral autoregulation. Conf Proc IEEE Eng Med Biol Soc
2011;2011:2410–3.
47. Birch AA, Dirnhuber MJ, Hartley-Davies R, et al. Assessment of
autoregulation by means of periodic changes in blood pressure. Stroke
1995;26(5):834–7.
48. Reinhard M, Müller T, Guschlbauer B, et al. Transfer function analysis for
clinical evaluation of dynamic cerebral autoregulation: a comparison
between spontaneous and respiratory-induced oscillations. Physiol Meas
2003;24(1):27–43.
49. Zhang R, Zuckerman JH, Giller CA, et al. Transfer function analysis of
dynamic cerebral autoregulation in humans. Am J Physiol 1998;274
(1 Pt 2):H233–41.
50. Tiecks FP, Lam AM, Aaslid R, et al. Comparison of static and dynamic
cerebral autoregulation measurements. Stroke 1995;26(6):1014–9.
A complete list of references for this chapter can be found online at
www.expertconsult.com.
 Section VII—Computers, Engineering, and the Future 455.e1
References
1. Samuel M, Burge DM, Marchbanks RJ. Tympanic membrane displacement
testing in regular assessment of intracranial pressure in eight children with
shunted hydrocephalus. J Neurosurg 1998;88(6):983–95.
2. Plandsoen WCG, de Jong DA, Maas AIR, et al. Fontanelle pressure
monitoring in infants with the Rotterdam teletransducer: a reliable
technique. Med Prog Technol 1987;13:21–271.
3. Klingelhofer J, Conrad B, Benecke R, et al. Evaluation of intracranial
pressure from transcranial Doppler studies in cerebral disease. J Neurol
1988;235:159–62.
4. Prunet B, Asencio Y, Lacroix G, et al. Noninvasive detection of elevated
intracranial pressure using a portable ultrasound system. Am J Emerg Med
2011; Jun 10. Epub ahead of print.
5. Soldatos T, Chatzimichail K, Papathanasiou M, et al. Optic nerve
sonography: a new window for the non-invasive evaluation of intracranial
pressure in brain injury. Emerg Med J 2009;26(9):630–4. Review.
6. Moretti R, Pizzi B. Ultrasonography of the optic nerve in neurocritically ill
patients. Acta Anaesthesiol Scand 2011;55(6):644–52.
7. Behrens A, Lenfeld N, Ambarki K, et al. Trans-cranial Doppler pulsatility
index: not an accurate method to assess intracranial pressure. Neurosurgery
2010;66(6):1050–7.
8. Rosenberg JB, Shiloh AL, Savel RH, et al. Non-invasive methods of
estimating intracranial pressure. Neurocrit Care 2011;15(3):599–608.
9. Ragauskas A, Daubaris G, Dziugys A, et al. Innovative non-invasive method
for absolute intracranial pressure measurement without calibration. Acta
Neurochir Suppl 2005;95:351–61.
10. Ueno T, Ballard RE, Shuer LM, et al. Intracranial pressure dynamics during
simulated microgravity using a new noninvasive ultrasonic technique.
J Gravit Physiol 1998;5(1):P39–40.
11. Schoser BG, Riemenschneider N, Hansen HC. The impact of raised
intracranial pressure on cerebral venous hemodynamics: a prospective
venous transcranial Doppler ultrasonography study. J Neurosurg
1999;91(5):744–9.
12. Zhao YL, Zhou JY, Zhu GH. Clinical experience with the non-invasive ICP
monitoring system. Acta Neurochir Suppl 2005;95:351–3.
13. Raksin PB, Alperin N, Sivaramakrishnan A, et al. Noninvasive intracranial
compliance and pressure based on dynamic magnetic resonance imaging of
blood flow and cerebrospinal fluid flow: review of principles,
implementation, and other noninvasive approaches. Neurosurg Focus
2003;14(4):e4.
14. Geeraerts T, Merceron S, Benhamou D, et al. Non-invasive assessment of
intracranial pressure using ocular sonography in neurocritical care patients.
Intensive Care Med 2008;34(11):2062–7.
15. Aaslid R, Markwalder TM, Nornes H. Noninvasive transcranial Doppler
ultrasound recording of flow velocity in basal cerebral arteries. J Neurosurg
1982;57(6):769–74.
16. Chan KH, Miller DJ, Dearden M, et al. The effect of changes in cerebral
perfusion pressure upon middle cerebral artery blood flow velocity and
jugular bulb venous oxygen saturation after severe brain trauma.
J Neurosurg 1992;77:55–61.
17. Thees C, Scholz M, Schaller MDC, et al. Relationship between intracranial
pressure and critical closing pressure in patients with neurotrauma.
Anesthesiology 2002;96(3):595–9.
18. Czosnyka M, Smielewski P, Piechnik S, et al. Critical closing pressure in
cerebrovascular circulation. J Neurol Neurosurg Psychiatry
1999;66(5):606–11.
19. Aaslid R, Lundar T, Lindegaard K-F, et al. Estimation of cerebral perfusion
pressure from arterial blood pressure and transcranial Doppler recordings.
In: Miller JD, Teasdale GM, Rowan JO, et al, editors. Intracranial pressure
VI. Berlin/Heidelberg: Springer Verlag; 1986. p. 229–31.
20. Schmidt EA, Czosnyka M, Gooskens I, et al. Preliminary experience of the
estimation of cerebral perfusion pressure using transcranial Doppler
ultrasonography. J Neurol Neurosurg Psychiatry 2001;70(2):198–204.
21. Bellner J, Romner B, Reinstrup P, et al. Transcranial Doppler sonography
pulsatility index (PI) reflects intracranial pressure (ICP). Surg Neurol
2004;62(1):45–51; discussion 51.
22. Schmidt B, Klingelhofer J, Schwarze JJ, et al. Noninvasive prediction of
intracranial pressure curves using transcranial Doppler ultrasonography
and blood pressure curves. Stroke 1997;28(12):2465–72.
23. Reulen HJ, Graham R, Spatz M, et al. Role of pressure gradient and bulk
flow in dynamics of vasogenic brain edema. J.Neurosurg 1977;46:24–35.
24. Carhuapoma JR, Qureshi AI, Bhardwaj A, et al. Interhemispheric
intracranial pressure gradients in massive cerebral infarction. J Neurosurg
Anesthesiol 2002;14(4):299–303.
25. Bundgaard H, Cold GE. Studies of regional subdural pressure gradients
during craniotomy. Br J Neurosurg 2000;14(3):229–34.
26. Kumar A, Schmidt EA, Hiler M, et al. Asymmetry of critical closing
pressure following head injury. J Neurol Neurosurg Psychiatry
2005;76(11):1570–3.
27. Schmidt EA, Czosnyka M, Steiner LA, et al. Asymmetry of pressure
autoregulation after traumatic brain injury. J Neurosurg 2003;99(6):991–8.
28. Hansen HC, Helmke K, Kunze K. Optic nerve sheath enlargement in acute
intracranial hypertension. Neuroophthalmology 1994;14:345–54.
29. Steffen H, Eifert B, Aschoff A, et al. The diagnostic value of optic disc
evaluation in acute elevated intracranial pressure. Ophthalmology
1996;103:1229–32.
30. Hayreh SS. Optic disc edema in raised intracranial pressure. V.
Pathogenesis. Arch Ophthalmol 1977;95:1553–65.
31. Querfurth HW, Lagre’ze WD, Hedges TR, et al. Flow velocity and pulsatility
of the ocular circulation in chronic intracranial hypertension. Acta Neurol
Scand 2002;105:431–40.
32. Miller MM, Chang T, Keating R, et al. Blood flow velocities are reduced in
the optic nerve of children with elevated intracranial pressure. J Child
Neurol 2009;24:30–5.
33. Querfurth HW, Arms SW, Lichy CM, et al. Prediction of intracranial
pressure from noninvasive transocular venous and arterial hemodynamic
measurements: a pilot study. Neurocrit Care 2004;1:183–94.
34. Querfurth HW, Lieberman P, Arms S, et al. Ophthalmodynamometry for
ICP prediction and pilot test on Mt. Everest. BMC Neurol 2010 1;10:106.
35. Scott CJ, Kardon RH, Lee AG, et al. Diagnosis and grading of papilledema
in patients with raised intracranial pressure using optical coherence
tomography vs clinical expert assessment using a clinical staging scale. Arch
Ophthalmol 2010;128:705–11.
36. Rebolleda G, Munoz-Negrete FJ. Follow-up of mild papilledema in
idiopathic intracranial hypertension with optical coherence tomography.
Invest Ophthalmol Vis Sci 2009;50:5197–200.
37. Trick GL, Vesti E, Tawansy K, et al. Quantitative evaluation of papilledema
in pseudotumor cerebri. Invest Ophthalmol Vis Sci 1998;39:1964–71.
38. Heckmann JG, Weber M, Junemann AG, et al. Laser scanning tomography
of the optic nerve vs CSF opening pressure in idiopathic intracranial
hypertension. Neurology 2004;62:1221–3.
39. Tayal VS, Neulander M, Norton HJ, et al. Emergency department
sonographic measurement of optic nerve sheath diameter to detect findings
of increased intracranial pressure in adult head injury patients. Ann Emerg
Med 2007;49:508–14.
40. Moretti R, Pizzi B, Cassini F, et al. Reliability of optic nerve ultrasound for
the evaluation of patients with spontaneous intracranial hemorrhage.
Neurocrit Care 2009;11:406–10.
41. Rajajee V, Vanaman M, Fletcher JJ, et al. Optic nerve ultrasound for the
detection of raised intracranial pressure. Neurocrit Care 2011;Jul 19. Epub
ahead of print.
42. Potgieter DW, Kippin A, Ngu F, et al. Can accurate ultrasonographic
measurement of the optic nerve sheath diameter (a non-invasive measure
of intracranial pressure) be taught to novice operators in a single training
session? Anaesth Intensive Care 2011;39:95–100.
43. Ballantyne SA, O’Neill G, Hamilton R, et al. Observer variation in the
sonographic measurement of optic nerve sheath diameter in normal adults.
Eur J Ultrasound 2002;15:145–9.
44. Copetti R, Cattarossi L. Optic nerve ultrasound: artifacts and real images.
Intensive Care Med 2009;35:1488–9.
45. Sharma D, Vavilala MS. Lund concept for the management of traumatic
brain injury: a physiological principle awaiting stronger evidence.
J Neurosurg Anesthesiol 2011;23(4):363–7.
46. Aoi MC, Matzuka BJ, Olufsen MS. Toward online, noninvasive, nonlinear
assessment of cerebral autoregulation. Conf Proc IEEE Eng Med Biol Soc
2011;2011:2410–3.
47. Birch AA, Dirnhuber MJ, Hartley-Davies R, et al. Assessment of
autoregulation by means of periodic changes in blood pressure. Stroke
1995;26(5):834–7.
48. Reinhard M, Müller T, Guschlbauer B, et al. Transfer function analysis for
clinical evaluation of dynamic cerebral autoregulation: a comparison
between spontaneous and respiratory-induced oscillations. Physiol Meas
2003;24(1):27–43.
49. Zhang R, Zuckerman JH, Giller CA, et al. Transfer function analysis of
dynamic cerebral autoregulation in humans. Am J Physiol 1998;274(1 Pt
2):H233–41.
50. Tiecks FP, Lam AM, Aaslid R, et al. Comparison of static and dynamic
cerebral autoregulation measurements. Stroke 1995;26(6):1014–9.
51. Aaslid R, Blaha M, Sviri G, et al. Asymmetric dynamic cerebral
autoregulatory response to cyclic stimuli. Stroke 2007;38(5):1465–9.
52. Katsogridakis E, Bush G, Fan L, et al. Random perturbations of arterial
blood pressure for the assessment of dynamic cerebral autoregulation.
Physiol Meas 2012;33(2):103–16.
455.e2 Section VII—Computers, Engineering, and the Future
53. Czosnyka M, Smielewski P, Piechnik S, et al. Cerebral autoregulation
following head injury. J Neurosurg 2001;95(5):756–63.
54. Panerai RB, Kerins V, Fan L, et al. Association between dynamic cerebral
autoregulation and mortality in severe head injury. Br J Neurosurg
2004;18(5):471–9.
55. Budohoski KP, Reinhard M, Aries MJ, et al. Monitoring cerebral
autoregulation after head injury: which component of transcranial
Doppler flow velocity is optimal? Neurocrit Care 2012;17(2):211–8.
56. Latka M, Turalska M, Glaubic-Latka M, et al. Phase dynamics in cerebral
autoregulation. Am J Physiol Heart Circ Physiol 2005;289(5):H2272–9.
57. Novak V, Yang AC, Lepicovsky L, et al. Multimodal pressure-flow method to
assess dynamics of cerebral autoregulation in stroke and hypertension.
Biomed Eng Online 2004 5;3(1):39.
58. Reinhard M, Wehrle-Wieland E, Grabiak D, et al. Oscillatory cerebral
hemodynamics–the macro- vs. microvascular level. J Neurol Sci
2006;250(1-2):103–9.
59. Wong FY, Leung TS, Austin T, et al. Impaired autoregulation in preterm
infants identified by using spatially resolved spectroscopy. Pediatrics
2008;121(3):e604–11.
60. Brady KM, Lee JK, Kibler KK, et al. Continuous time-domain analysis of
cerebrovascular autoregulation using near-infrared spectroscopy. Stroke
2007;38(10):2818–25.
61. Lee JK, Yang ZJ, Wang B, et al. Noninvasive autoregulation monitoring in a
Swine model of pediatric cardiac arrest. Anesth Analg 2012;114(4):825–36.
Epub 2012, Feb 7.
62. Joshi B, Ono M, Brown C, et al. Predicting the limits of cerebral
autoregulation during cardiopulmonary bypass. Anesth Analg
2012;114(3):503–10. Epub 2011m Nov 21.
63. Diedler J, Zweifel C, Budohoski KP, et al. The limitations of near-infrared
spectroscopy to assess cerebrovascular reactivity: the role of slow
frequency oscillations. Anesth Analg 2011;113(4):849–57. Epub 2011,
Aug 4.
64. Murkin J Adams S, Schaefer B, et al. Monitoring cerebral oxygen saturation
significantly decreases stroke rate in CABG patients: a randomized blinded
study. Presented at the Outcomes 2004: The Key West Meeting, Fla., May
19–24, 2004. The Heart Surgery Forum 7(6), 2004[abstract].
65. Papadimos T, Marcos P. Cerebral oximetry and an unanticipated circulatory
arrest. Anaesthesia 2004;59:309–10.
66. Sakamoto T, Zurakowski D, Duebner LF, et al. Cerebral Ischemia caused by
obstructed SVC cannula is detected by NIRS. J Thorac Cardiovasc Surg
2004;128:220–32.
67. Dunham C, Ransom K, McAuley C, et al. Severe brain injury ICU
outcomes are associated with cranial-arterial pressure index and
noninvasive bispectral index and transcranial oxygen saturation: a
prospective, preliminary study. Crit Care 2006;10(6):1–9.
68. Kirkpatrick J, Lam K, Al-Rawi P, et al. Defining thresholds for critical
ischemia by using near-infrared spectroscopy in adult brain. J Neurosurg
1998;89(3):389–94.
69. Al-Rawi PG, Kirkpatrick PJ. Tissue oxygen index: thresholds for cerebral
ischemia using near-infrared spectroscopy. Stroke 2006;37(11):2720–5.
70. Zweifel C, Castellani G, Czosnyka M, et al. Continuous assessment of
cerebral autoregulation with near-infrared spectroscopy in adults after
subarachnoid hemorrhage. Stroke 2010;41(9):1963–8. Epub 2010, Jul 22.
71. Yokose N, Sakatani K, Murata Y, et al. Bedside monitoring of cerebral blood
oxygenation and hemodynamics after aneurysmal subarachnoid
hemorrhage by quantitative time-resolved near-infrared spectroscopy.
World Neurosurg 2010;73(5):508–13.
72. Wang S, Wu E, Tam C, et al. Characterization of white matter injury in a
hypoxic ischemic neonatal rat model by diffusion tensor MRI. Stroke
2008;38:2348–53.
73. Cecil KM, Hills EC, Sandel ME, et al. Proton magnetic resonance
spectroscopy for detection of axonal injury in the splenium of the corpus
callosum of brain-injured patients. J Neurosurg 1998;88(5):795–801
74. Carpentier A, Galanaud D, Puybasset L, et al. Early morphologic and
spectroscopic magnetic resonance in severe traumatic brain injuries can
detect “invisible brain stem damage” and predict “vegetative states.”
J Neurotrauma 2006;23(5):674–85.
75. Shen Q, Ren H, Cheng H, et al. Functional perfusion and diffusion MRI of
acute focal ischemic brain injury. J Cerebr Blood Flow Metab
2005;25:1265–79.
76. Moritz C, Rowley H, Haughton V, et al. Functional MR Imaging assessment
of a non-responsive brain injured patient. Magn Reson Imaging
2001;19:1129–32.
77. Driniger M, Aiyagari V, Zazulia A, et al. Effect of hyperoxia on cerebral
metabolic rate for oxygen measured using positron emission tomography
in patients with acute severe head injury. J Neurosurg 2007;106:526–9.
78. Boecker H, Kuwert T, Langen K. SPECT with HMPAO compared to PET
with FDG in Huntingdon disease. J Comput Assist Tomogr 1994;19:
542–8.
79. Davis P, Micra S, Alazraki N. The brain in older persons with and without
dementia: findings on MRI, PET, and SPECT. Am J Roentgenol
1994;162:1267–78.
80. Gross H, Kling A, Henry G, et al. Local cerebral glucose metabolism in
patients with long term behavioral and cognitive deficits following mild
TBI. J Neuropsych Clin Neurosci 1996;8:324–34.
81. Mielke R, Petrzyk U, Jacobs A. HMPAO SPECT and FDG PET in
Alzheimer’s and vascular dementia: comparison of perfusion and metabolic
pattern. Eur J Nuc Med 1994;10:1052–60.
82. Yamaki T Imahori Y, Ohomori Y. Cerebral hemodynamics and metabolism
of severe diffuse brain injury measured by PET. J Nuc Med 1996;37:
1166–70.
83. Alavi A. Functional and anatomical studies of head injury.
J Neuropsychiatry Clin Neurosci 1989;1(Suppl):S45–50.
84. Alavi A, Fazekas T, Alves W. PET in the evaluation of head injury.
J Cerebral Blood Flow Metab 1987;(Suppl):646.
85. Furukawa M, Kashiwagi S, Matsunaga N, et al. Evaluation of cerebral
perfusion parameters measured by perfusion CT in chronic cerebral
ischemia: comparison with xenon CT. J Comput Assist Tomogr 2002;26:
272–8.
86. Soustiel J, Mahamid E, Goldsher D, et al. Perfusion CT for early assessment
of traumatic cerebral contusions. Neuroradiology 2008;50:189–96.
87. Petrov YY, Petrova IY, Patrikeev IA, et al. Multiwavelength optoacoustic
system for noninvasive monitoring of cerebral venous oxygenation: a pilot
clinical test in the internal jugular vein. Opt Lett 2006;31(12):1827–9.
88. Jo J, Yang X. Functional photoacoustic imaging to observe regional brain
activation induced by cocaine hydrochloride. J Biomed Opt
2011;16(9):090506.
89. Zhang Y, Hong H, Cai W. Photoacoustic imaging. Cold Spring Harb Protoc
2011;2011(9):pii.
90. Xu Z, Zhu Q, Wang LV. In vivo photoacoustic tomography of mouse
cerebral edema induced by cold injury. J Biomed Opt 2011;16(6):066020.
91. Ronne-Engstrom E, Winkler T. Continuous EEG monitoring in patients
with traumatic brain injury reveals a high incidence of epileptiform
activity. Acta Neurol Scand 2006;114:47–53.
92. Vespa PM, Nuwer MR, Nenov V. Increased incidence and impact of
nonconvulsive and convulsive seizures after traumatic brain injury as
detected by continuous EEG monitoring. J Neurosurg 1999;91:750–60.
93. Young GB, Doig GS. Continuous EEG monitoring in comatose intensive
care patients: epileptiform activity in etiologically distinct groups.
Neurocrit Care 2005;2:5–10.
94. Hirsch LJ. Continuous EEG monitoring in the intensive care unit: an
overview. J Clin Neurophysiology 2004;21:323–40.
95. Carrera E, Claassen J, Oddo M, et al. Continuous electroencephalographic
monitoring in critically ill patients with central nervous system infections.
Arch Neurol 2008;65(12):1612–8.
96. Oddo M, Carrera E, Claassen J, et al. Continuous electroencephalography
in the medical intensive care unit. Crit Care Med 2009;37(6):2051–6.
97. Rossetti AO, Urbano LA, Delodder F, et al. Prognostic value of continuous
EEG monitoring during therapeutic hypothermia after cardiac arrest. Crit
Care 2010;14(5):R173. Epub 2010, Sep 29.
98. Mani R, Schmitt SE, Mazer M, et al. The frequency and timing of
epileptiform activity on continuous electroencephalogram in comatose
post-cardiac arrest syndrome patients treated with therapeutic
hypothermia. Resuscitation 2012;Feb 23. Epub ahead of print.
99. Young GB. Continuous EEG monitoring in the ICU: challenges and
opportunities. Can J Neurol Sci 2009;36(Suppl 2):S89–91.
100. Young GB. Continuous EEG monitoring in the ICU. Acta Neurol Scand
2006;114:67–8.
101. Vespa PM, Nenov V, Nuwer MR. Continuous EEG monitoring in the
intensive care unit: early findings and clinical efficacy. J Clin
Neurophysiology 1999;16:1–13
102. Freye E. Cerebral monitoring in the operating room and the intensive care
unit: an introductory for the clinician and a guide for the novice wanting
to open a window to the brain. J Clin Monit Comput 2005;19:77–168,
2005.
103. Traczewski W, Moskala M, Kruk D, et al. The role of computerized
rheoencephalography in the assessment of normal pressure hydrocephalus.
J Neurotrauma 2005;22(7):836–43.
104. Pérez JJ Guijarro E, Barcia JA. Influence of the scalp thickness on the
intracranial contribution to rheoencephalography. Phys Med Bio
2004;49(18):4383–94.
VII
Chapter
47  
Auditory Signals for
Noninvasive Monitors
Richard P. Dutton and John M. Sewell
Introduction
Noninvasive monitoring of cerebral blood flow (CBF) among
other aspects of neurologic function will add important tools
to the armamentarium of the clinician. A noninvasive monitor
could be applied to a much wider range of patients than exist-
ing intracranial pressure (ICP) monitors, which require that
the patient be in an intensive care unit (ICU). It is now feasible
to measure ICP with optic nerve ultrasound or with venous
ophthalmodynamometry.1-4 Investigators have also combined
ICP monitors with other noninvasive devices or used near-
infrared spectroscopy to measure cerebrovascular pressure
reactivity.5,6 The role of transcranial Doppler or color-coded
duplex sonography that also can be used to noninvasively
assess ICP and cerebral perfusion pressure7 is described in
Chapter 30. It also is possible that a noninvasive monitor
could provide more physiologically relevant information
about intracranial physiology (e.g., blood flow rather than
blood pressure). This chapter describes one approach to this
challenge, based on detection and processing of sound waves
generated by the heart and transmitted through arterial vessels
to the brain. Discussion of this brain acoustic monitor
(BAM)8,9 begins with a description of the pathophysiology of
brain injury and the physics of the cranium that make acoustic
monitoring possible. Following is a discussion of how the
BAM evolved to its present state, with evidence from a number
of studies of its use in traumatic brain injury (TBI) and stroke.
Illustrations of the output of the BAM as well as interpretation
steps used to determine the existence of CBF anomalies are
provided. A discussion of challenges and future directions of
this technology concludes the chapter.
Pathophysiology of Detectable
Brain Injury
The brain has the most highly regulated vasculature in the
body. Neuronal function depends on aerobic metabolism,
which in turn requires continuous delivery of oxygenated
blood. Local autoregulatory processes act to maintain even
blood flow and pressure within the brain across a wide range
of systemic pressures. This autoregulation is exquisitely sensi-
tive to disruption by TBI and other pathologies,10 but very
456
difficult to study given the logistic constraints involved in use
of the imaging tools, such as positron emission tomography
(PET) scanning, in trauma or stroke patients in the emergency
setting. Acoustics, however, offers an alternative approach to
cerebrovascular monitoring that is less expensive and easier to
apply early after injury.
Acoustics, the science of sound wave transmission, is com-
monly used in medicine to characterize the flow of blood and
air through the body. The stethoscope, the very symbol of the
physician, is one of the oldest and most useful of all medical
devices. A stethoscope is a simple machine that amplifies natu-
rally produced sound waves to an amplitude at which they can
be detected and discriminated by the human ear. In this way
physicians can examine blood flow through the heart and
great vessels and airflow through the lungs. Although normal,
laminar, flow is relatively inaudible, turbulence produced by
obstructions or narrowing is clearly heard, and characterized
variously as bruits, rales, wheezing, or the like. This is the
principle that underlies the function of the BAM. Blood flow
in the cranium is not audible to an external stethoscope, even
with electronic amplification. The cranium does flex or vibrate
in time to the arterial pulse, however, on a microscopic level.
A highly sensitive contact sensor can be used to detect these
vibrations and display them digitally. The BAM is thus an
electronic stethoscope that “listens” to the blood moving
through the brain and detects alterations in flow. Studies at
the R Adams Cowley Shock Trauma Center of the University
of Maryland have found that the brain acoustic signal the
BAM measures is identical in morphology and component
frequencies to the arterial pulse signal captured from any
other artery in the body in normal patients, but substantially
different following any trauma to the brain.
The observation of what the BAM measured provided the
principal impetus for the study of this technology and evolved
into assessing the signal itself and its characteristics as they
differed in “normal” volunteers and patients with TBI. The
process involved signal analysis in both time and frequency
domains as is shown in the following examples, in which the
brain-emitted signals are compared with those from a refer-
ence artery, such as the radial or digital artery. In the latest
embodiment of the BAM system the metrics that indicate
critical differences between reference and brain are coded into
© Copyright 2013 Elsevier Inc. All rights reserved.
 Section VII—Computers, Engineering, and the Future 457
software that will identify anomalies in signals immediately.
To broaden the basis of comparison, the BAM findings have
been compared with computed tomography (CT) scan results,
and with a group of indicators such as evidence of head insult,
alteration of consciousness, Glasgow Coma Scale (GCS) score
of less than 14, postconcussive symptoms, and general presen-
tation. When these indicators are included, there is a high
incidence of abnormal BAM scores in patients with evidence
of TBI, compared with a very low incidence of abnormalities
in normal volunteers.
Although in some ways this approach is novel, it is actually
an extension of general auscultation with a visual display used
instead of detection of audible sound differences. It is possible
to amplify a BAM signal into the audible range, but the human
ear is unable to detect the subtle differences in amplitude and
frequency associated with TBI that become evident when a
visual display is used to characterize the wave. Quantification
of digitized data also makes it possible to track the progress
of an individual patient over the course of treatment and
recovery.
The diagnosis and subsequent clinical management of any
patient with an acute brain injury depends on clinicians
who integrate information from the patient’s history, from
physical examination, from imaging studies, from laboratory
tests, and from changes that occur over time. No one device
can possibly provide an absolute black or white diagnosis in
every patient. Instead the BAM is seen as an adjunctive tech-
nology that is simple, rapid, logistically compatible with the
early care of trauma patients, and able to provide objective
data on the quality of blood flow in the potentially injured
brain.
Evolution of the Acoustic
Monitoring Approach and Current
Embodiment
Some notes on the somewhat serendipitous evolution of this
approach will shed light on the mechanisms involved in brain
acoustics. The study began with a visit from a group of U.S.
Army Rangers to acoustic engineers now working at Active
Signal Technologies. Their objectives were to determine if the
science of exploring the ocean with sound could be used to
explore the brain and to develop an ICP monitor the size of a
cigarette pack—the space the rangers had for such equipment.
The initial efforts following these discussions indicated that
when small exciters at the temple interrogated the brain and
the results were measured at the forehead (a system based on
active sonar), variations in ICP would cause a difference in
acoustic response. The approach encountered the challenge of
calibrating this difference against absolute levels of ICP across
different patients due to individual differences. During the
course of this effort, a physician noted that the passive sensing
itself generated waveforms on the data-gathering analyzer that
resembled those of the arterial line displays on the patient
monitors. Furthermore, there was a difference in the mor-
phology of the signals between healthy patients (those admit-
ted with decreased level of consciousness due to alcohol, but
with no brain injury) and those with documented TBI and
elevated ICP. Unlike the ICP values themselves, these differ-
ences appeared to be constant over a range of patients, at least
at the basic level of those with and without TBI. Following this
proof of principle study, further studies of in-hospital, pre-
hospital and stroke applications of passive acoustic monitor-
ing applied to cerebral and systemic blood flow have been
conducted.
The Acoustic Science
of Brain Monitoring
There is a limited literature about passive brain monitoring;
the approach described here relies on the first principles
of acoustics.11 Fundamentally, stethoscopes are air-coupled
devices designed primarily to detect sounds generated in air-
filled structures (the lungs). To better detect variations in
brain sound emissions, sensors were built to “match” the
brain, which is close in density to water. The sensor systems
relied on “hydrophone” technology,12 instead of diaphragm-
based air-coupled stethoscope technology. This approach was
used because the cranial sensors were not immersed in the
fluid of the brain, but rather “coupled” to the outside of the
skull at a point anterior to the upper portions of the forehead
above the sinus cavity. A similar approach was used to capture
systemic arterial signals directly over the radial or digital
artery.
In the application of these principles to monitoring, the
first finding was that the resulting system did not detect
audible signals from the brain, although in some cases the
resolution of the system could be increased to detect anoma-
lous responses (refer to Moretti and Pizzi’s discussion of the
basic technology as applied to active and passive stethoscopy
in high-noise environments13). Rather, the critical and subtle
features of the signal that contained most of the information
had to rely on a high-resolution digital display.
The second finding was that the brain signal was out of
phase by nearly 180 degrees with signals that came from a
location over the temporal artery, and from the reference
sensor. This is a useful feature of this technology because it
enables the monitoring staff to readily distinguish the brain
signal from one that might come from the superficial vascu-
lature over the forehead. This distinction is shown in Figure
47.1. When there is uniform pressure from the arterial pulse
wave entering the cranium, the side of the head follows the
phase of the pulse, and the front of the head where the sensors
are placed is receding. This is the general rule, although there
are some exceptions based on anatomic variations among
patients.
Initial Studies
The initial BAM evaluation took place in a study of 30 patients
who were admitted to the ICU with severe TBI and invasive
ICP monitors. The results of this study were based on the
time-variant waveforms or “time-domain” signals collected.
The original results proved promising14 and the analysis tech-
niques moved to conventional time and frequency domain
signal analysis. Further studies were performed in stroke
patients15 and other environments including the admitting
area of a trauma center to determine early response to brain
insult and the prehospital environment. In addition studies
have been performed to examine how BAM relates to conven-
tional TBI metrics, and how the BAM tracked TBI effects over
a period of time following the original incident.
458 Section VII—Computers, Engineering, and the Future

0.60 Plot 0

Plot 1

Plot 2

0.40

0.20

0.00

–0.20

–0.40
B 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 3.25 3.50 3.75 4.00

Fig. 47.1  A, Outward flexion of the cranium in response to the arterial pulse. The top of the figure represents the anterior skull, which “recedes”
from a forehead sensor with each pulse. P, Pulse. B, A representative brain acoustic monitor signal from a healthy patient. The red trace is from the
radial artery, the green is from the temporal artery, and the black from the brain. The true brain signal is out of phase with the more superficial
signals.

Current Status
The BAM system continues to evolve (an example of a BAM
used between 2005 and 2008 is shown in Fig. 47.2). The radial
and finger sensor locations are shown, as well as the contact
points of the head sensors. All sensors are held in place with
bands except for the finger sensor, which is attached in much
the same way as a pulse oximeter probe. To operate the system
the researcher:
1. Places the sensors on the patient
2. Enters patient data in the acquiring laptop (or personal
digital assistant [PDA])
3. Activates the acquisition system, which records the sounds
from all three locations, two head and one reference
4. Checks the integrity of the signals
5. Repeats if necessary until signals are representative
6. Saves the data file and removes sensors and bands
This process normally takes approximately 5 minutes and can
be performed without disrupting indicated clinical care (e.g.,
the study may be performed after initial physical assessment,
while the patient is waiting for a CT scan). The BAM monitor-
ing procedure also includes observing the autograding results
of the system.
 Section VII—Computers, Engineering, and the Future 459
Fig. 47.2  The brain acoustic monitor in place on a volunteer. A, Laptop to acquire data. The radial and finger sensors also are visible.
B, Radial sensor. C, Head sensor on forehead.
BAM Measurements in Normal Volunteers
and Traumatic Brain Injury Patients
The BAM measures both the time variant amplitude of pulse
waveforms and the frequency response of those waveforms.
Examples are shown in the following text to illustrate the
acoustic properties of normal subjects (controls) compared
with those with TBI, beginning with data from a normal and
a mild TBI patient shown in Figure 47.3.
In Figure 47.3 A and D show the “time domain” signal in
volts, representing the amplitude and direction of skull deflec-
tion in real time from two points and a reference. The green
trace is from a sensor on the right forehead, the black trace is
from a sensor on the left forehead, and the red trace is from
the reference sensor over either the radial artery or the finger.
In a non-TBI patient (see Fig. 47.3, A), these signals are analo-
gous, and morphologically similar, to the conventional trace
produced by an arterial pressure catheter.
C
The system then performs a fast-Fourier transformation
(FFT) on these signals that determines the component fre-
quencies in the raw signal in a band from less than 1 Hz to
80 Hz—the frequency responses displayed in the middle of
each panel (see Fig. 47.3, B and E). In the absence of severe
pathology, the highest peak in the frequency response should
be the first one (the fundamental), which is determined by
the arterial pulse rate. In a normal individual this should be
around 72 beats per minute, or just greater than 1 Hz. The
frequency responses of the 3 signals from the normal volun-
teer (see Fig. 47.3, B) essentially overlap starting at the
approximately 1 Hz fundamental and descending to the
approximately 90 dB noise floor of the instrument, whereas
the TBI patient’s brain responses (see Fig. 47.3, E) diverge
from the arterial reference greater than 20 Hz.
The current BAM system has a signal-processing algorithm
that subtracts the frequency response of the artery from the
brain after averaging, and highlights the divergence by
460 Section VII—Computers, Engineering, and the Future

0.20 LB (0)

ref (1)

RB (2)
0.15

0.10

0.05

0.00

–0.05

–0.10
A 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 3.25 3.50 3.75 4.00

–20.00

–40.00

–60.00

–80.00

–100.00

–110.00
B 0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00

Fig. 47.3  Representative normal and abnormal signals from the brain acoustic monitor (BAM). A, Normal raw or “time domain” signal from
the BAM. The red trace is the radial artery signal; the black and green traces are from the left and right sides of the brain. B, Normal “frequency
domain” signal from the BAM, showing the superposition of frequencies in the right and left brain and the radial artery.
 Section VII—Computers, Engineering, and the Future 461

25.00

20.00

15.00

10.00

5.00

0.00

–5.00

–10.00

–15.00
5.00 30.00 55.00 80.00
C
0.60 LB (0)

ref (1)
0.50
RB (2)

0.40

0.30

0.20

0.10

0.00

–0.10

–0.20

–0.30
D 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 3.25 3.50 3.75 4.00

Fig. 47.3 cont’d  C, Normal “subtraction display” signal from the BAM, showing that brain and systemic (radial artery) signals have equivalent
frequency distribution. D, Abnormal raw or time domain signal from the BAM, in a patient with traumatic brain injury (TBI). The red trace is the radial
artery signal; the black and green traces are from the left and right sides of the brain.
Continued
462 Section VII—Computers, Engineering, and the Future

–20.00

–40.00

–60.00

–80.00

–100.00

–110.00
E 0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00

25.00

20.00

15.00

10.00

5.00

0.00

–5.00

–10.00

–15.00
5.00 25.00 50.00 80.00
F
Fig. 47.3 cont’d  E, Abnormal frequency domain signal from the BAM, in a patient with TBI, showing the deviation of frequency response between
the right and left brain signals (black and green traces) and the systemic signal (red). F, Abnormal subtraction display signal from the BAM, in a patient
with TBI, quantifying the deviation in frequency response between the brain and systemic signals shown in (E). LB, Left brain; RB, right brain.

calculating mean values of these relative frequency responses.
The relative frequency responses of the left and right forehead
signals are shown in panels of Figure 47.3 from 10 to 80 Hz—
the band determined on the basis of correlation of critical
frequencies within the arterial spectrum from both trauma
and stroke patients. In normal individuals the morphology of
the arterial pressure trace should be transmitted cleanly
throughout the entire arterial tree, and so the relative fre-
quency response of the brain (see Fig. 47.3, C) should be a
nearly flat line close to 0 dB. Indeed the normal relative fre-
quency response curves in Figure 47.3, C are scattered around
0 dB and their mean values are close to 0 dB: left brain
 Section VII—Computers, Engineering, and the Future 463
approximately minus 3 dB, right brain approximately minus
2 dB. Turbulence in the brain signal resulting from disrupted
autoregulation will cause a discrepancy between the brain and
systemic arterial frequency response, yielding a relative fre-
quency response (see Fig. 47.3, F) that rises to greater than
20 Hz.
Analysis Approach of
Acoustic Response
Of the three principal metrics used for numerical assessment
of the BAM signal, the first two (amplitude and ratio) are
measured directly from the time domain signal (the raw data)
and the third (divergence) is measured from the relative fre-
quency response, which uses a systemic arterial control. These
metrics are defined in the following text.
1. Amplitude: An acceptable and representative amplitude is
sometimes difficult to obtain for the brain because it is
based on skull excursion for each arterial pulse and
is only about 0.001 inch per heartbeat. The solid state
sensors of the BAM can consistently measure a signal
equal to about 0.1 V from positive to negative peak in a
healthy normal subject and so amplitude less than 0.1 V
is considered a positive indication for TBI. The systemic
arterial reference signal is much stronger and conse-
quently the system attenuates it in hardware or normal-
izes it in software so that it is consistently comparable to
the brain signal.
2. Ratio: The second metric is an initial crude assessment of
signal morphology that calculates the peak excursion
above the mean line divided by the peak excursion below
the mean. Positive excursion should be at least twice as
great as negative excursion and a ratio less than 2 is con-
sidered a positive indication for TBI. This is similar to the
normal calculation of mean arterial pressure (MAP) in
which the MAP is equal to (diastolic plus 1/3 [systolic
minus diastolic]). Interestingly, this is a telling indicator.
There can be some slight deviations, but the ratio cannot
go below 2 positive to negative deflection to be safely con-
sidered normal.
3. Divergence: The final and most important metric is
derived from the frequency analysis. In a normal subject
noise levels as indicated by the frequency-response slope
should drop off rapidly at higher frequencies, whereas in
the presence of brain trauma there will be higher ampli-
tudes persisting into higher frequencies in the band. The
divergence is defined as the maximum deviation of the
brain signal’s relative frequency response above 0 dB.
When the divergence is 10 dB or more, the patient is con-
sidered to have an anomalous brain noise compared to
reference. In a mundane comparison, it is as if there is a
quiet or laminar flow in the arteries when there is no dis-
turbance, and an increase caused by some alteration of
arterial tone causing disturbed flow (severely disturbed
flow would be turbulent flow similar to that caused by a
hose constriction).16
All three metrics on both forehead sensors must register a
“pass” to indicate a normal BAM; control and TBI patient
BAM outputs are shown in Figure 47.3. Here both the
normal and mild TBI patient pass the amplitude
criteria—both have amplitudes of at least 0.1 V in their brain
signals bilaterally. However, the ratios of the normal are
greater than 2 (see Fig. 47.3, A) with both left and right brain
signals exhibiting good pulsatile shape, whereas the brain-
injured patient has ratios less than 2 and the brain signals are
uneven—an unevenness not caused by artifact. The diver-
gences of the normal brain signals in Figure 47.3, C are 7 and
8 dB for the right and left brain, respectively, indicating BAM
negative for TBI, whereas the TBI signals have a maximum
divergence of 20 dB (see Fig. 47.3, F, left brain), indicating
BAM positive for TBI.
This analysis now has been automated, so that the results
of the amplitude, ratio and divergence calculations appear on
a pop-up panel with a color coding of yellow or green that
indicate whether the person either is out of the bounds of the
metrics established for controls or, in the case of green, that
the person passes all criteria. In the event of a yellow, the
screen shows the sensor(s) and metric(s) that are out of
bounds and the procedure is to adjust the sensor position(s),
band tension, and so on to accommodate for individual dif-
ferences and retest. When the results appear to repeat (at least
three times) they are considered representative, whether
yellow or green, and the data are saved. In general, it is difficult
to obtain a green reading when there is any anomaly in the
patient’s blood flow, but yellow readings can often be cor-
rected through adjustments to sensor placement. An example
of these displays is shown in Figure 47.4, in which the yellow
reading is checked and the left brain sensor adjusted to yield
a green reading with all metrics being within acceptable limits.
The system now has evolved from initial time domain read-
ings early in the study, to the later time and frequency domain
data collection, comparison with a systemic arterial reference,
and analysis with the autograding now in a clearly readable
format.
Example of the Acoustic Response
of a Stroke Patient
Passive acoustics may have place in the field of stroke. Acoustic
technology could be especially useful if it allowed for a rapid
and early discrimination of ischemic versus hemorrhagic
stroke, and could thus indicate or facilitate rapid treatment
with clot-busting drugs, where efficacy is time dependent.17
Currently, the diagnosis of ischemic stroke is made by a com-
bination of physical examination and CT or magnetic reso-
nance imaging (MRI) scan, the imaging often being needed to
exclude cerebral hemorrhage that would contraindicate
thrombolytics. BAM technology has been used in an attempt
to discriminate between these two forms of stroke.16 The
results are best appreciated using two data examples followed
by a summary of the study results about the applicability of
passive acoustics to stroke.
Figure 47.5 illustrates a patient with an ischemic stroke who
was monitored with the BAM within 3 hours of symptom
onset and who received tissue-plasminogen activator (rt-PA).
The monitoring took place both before and after administra-
tion of the drug and the data examples are interesting for
several reasons. Only the divergence measurements (see Anal-
ysis Approach of Acoustic Response) are shown here (see Fig.
47.5) as they were the most telling in the stroke studies. As can
be seen in Figure 47.5, A, the divergence rises in the middle of
464 Section VII—Computers, Engineering, and the Future
A B
Fig. 47.4  A, Automated interpretive display of a brain acoustic monitor session. Yellow indicates a deviation from normal in the signal, which may
be due to pathology or to misapplication of a sensor. B, Automated interpretive display of a brain acoustic monitor session. Green indicates a normal
signal.
the band to a level greater than 10 dB and this in several fre-
quency points. After rt-PA (see Fig. 47.5, B), the divergence is
uniformly reduced and resembles a normal response. Similar
results were observed in other patients, although in some cases
in which the divergence was reduced, the patient did not
recover clinically. These results suggest that brain acoustics
can be used to identify ischemic insults.
Hemorrhagic Stroke
Patients with hemorrhagic stroke were examined using BAM.
The primary focus of the study was to examine whether isch-
emic or hemorrhagic stroke could be differentiated before 3
hours, that is, determine eligibility for rt-PA. There were two
categories of BAM measurements of hemorrhagic stroke: less
than 3 hours of symptom onset or greater than 3 hours.
Studies within 3 hours had a normal divergence or a less noisy
signal than the ischemic patients. When examined more than
3 hours after symptom onset, patients often had higher than
normal divergence. This may represent “mass effect” contrib-
uting to ischemia rather than active bleeding. The difference
between the ischemic (noisy or high in divergence) and the
hemorrhagic (quiet, low, or normal divergence) acoustic
response within 3 hours suggests that BAM may be used to
differentiate stroke subtype. Further studies are needed to
validate this.
Future Potential Uses of Brain
Acoustic Monitor Technology
The Challenge of Prehospital Triage
More effective triage tools can help prehospital emergency
medicine. For example, any patient who may have suffered a
TBI must be transported to a trauma center, where a head CT
scan is obtained. However, greater than 90% of admission CT
scans to rule out brain trauma are normal. In addition, many
of the patients with normal CT scans may still have brain
injury such as a concussion. The diagnosis of concussion is a
clinical one, and there currently is not an objective physiologic
test; this has important implications in sports medicine and
in the military. A simple noninvasive test that could facilitate
triage of TBI patients or inform the diagnosis of mild TBI may
be feasible using acoustic technology.
Initial Issues and Challenges
There are several challenges and questions when BAM is used
as a “triage tool.” First, can passive acoustics detect an anomaly
very early after injury, or does it take time for the disruption
of blood flow to become apparent? Second, can a BAM system
be configured for prehospital, sports field, or far-forward mili-
tary application? This would require a modified sensor suite
 Section VII—Computers, Engineering, and the Future 465

20.00

10.00

0.00

–10.00

–20.00
0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00
A

20.00

10.00

0.00

–10.00

–20.00
0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00 90.00 100.00
B
Fig. 47.5  A, Abnormal brain acoustic monitor (BAM) subtraction display from a patient with an evolving ischemic stroke. Note the deviation between
brain and systemic signals in the midfrequency range (arrow). B, Improved BAM signal in the same patient, following thrombolytic therapy.

and a small and rugged electronics package. Third, can the

BAM system operate well in a noisy outdoor environment, or
in a medevac unit (ground or helicopter) where vibration and
noise may limit BAM signal? To address these issues 150
patients were evaluated during transport to a trauma center
(prehospital) and compared with data from patients exam-
ined in-hospital.
Delay of Initial Acoustic Response?
Tissue injury appears to cause almost instantaneous acousti-
cally detectable changes in the brain. Brain acoustic monitor-
ing often occurred less than 15 minutes after brain injury, and
in all the injury was sensitive to acoustic detection. However,
some patients without injury had an abnormal BAM response,
466 Section VII—Computers, Engineering, and the Future
suggesting that early BAM information could not be used
reliably for a triage decision, but only as a what-if guide.
Could a Device Be Made Small Enough
for Triage Purposes?
The electronics and associated software for the BAM are
simple and in theory adaptable to small platforms, especially
with recent advances in technology. In early studies the
Compaq iPAQ was used in the configuration shown in Figure
47.6. This particular device did not prove rugged enough over
a long period of time, but showed that there was no inherent
reason that the BAM could not be reduced to a size compatible
with prehospital triage or handheld use.
Could the Brain Acoustic Monitor Work
in Noisy or “Hostile” Environments?
BAM has been studied in a variety of ground ambulances,
helicopters, and fixed-wing transport aircraft. Better record-
ings are found when the environment was quiet and the
vehicle was not moving; however, the technologic challenges
presented do not seem insurmountable. They fall into three
categories:
Short-term medevac or ambulance transport: In these

instances, medics indicate that the most important
Fig. 47.6  The brain acoustic monitor configured to run on a
personal digital assistant, as deployed with the U.S. Army. This
can be used for prehospital monitoring.
Table 47.1  Sensitivity and Specificity of BAM Abnormalities in Patients with Mild TBI and
Postconcussive Symptoms 7 to 10 Days after the Injury Compared with Other Potential
Indicators of TBI
BAM CT Scan
Sensitivity 87% 6%
Specificity 19% 99%
Total subjects 284 286
BAM, Brain acoustic monitor; CT, computed tomography; GCS, Glasgow Coma Scale; LOC, level of consciousness.
decision is the triage decision that is made before loading
the patient, thus there is no concern for vibration, although
noise is present. The system has been tested to less than
110 dBA noise environment with no adverse effects.
 Longer-term medevac or ambulance transport: Whether
military or civilian, the platform excitation levels tend to
overwhelm the BAM signal. When this occurs it is not
possible to obtain representative measurements. Noise
cancellation approaches have been explored, but when the
signal to noise is greater than 20 in amplitude as is often
the case in these instances, even noise cancellation tech-
niques are limited.
 Longer-term fixed-wing transport: Here there are more
constant “lines” of excitation and these can be canceled,
even in the band of the BAM. Projects funded by the U.S.
Air Force long-distance critical care aeromedical transpor-
tation teams continue to explore this question. It appears,
however, that reliable BAM signals can be obtained after
noise cancellation in fixed-wing transports.
Thus the BAM is versatile, but in very high amplitude vibra-
tion excitation environments that have few periodic compo-
nents, it will not operate well. The information must be
gathered before or after the transport.
Acoustic Technology for
In-Hospital Monitoring
BAM technology is unlikely to replace CT or MRI once a
patient reaches the hospital, but it does have several potential
uses including evaluation of evolving pathophysiology over
time to predict problems, use as a continuous monitoring
system, or use during surgical procedures, in which neurologic
monitoring is indicated but invasive intracranial monitoring
is not justified.
The Brain Acoustic Monitor
as a Leading Indicator
BAM changes appear to have a predictive nature. Table 47.1
shows the sensitivity and specificity of BAM abnormalities
(measured at the time of admission in patients with mild TBI)
associated with the presence of three or more postconcussive
symptoms 7 to 10 days after the injury, and compares it with
other potential indicators: CT scan; GCS score less than 15;
evidence of head trauma; and documented loss of conscious-
ness (LOC). BAM changes have high sensitivity (most patients
with symptoms had an abnormal BAM) although low speci-
ficity (some patients with abnormal BAM did not develop
GCS <15 Evidence of Head Trauma Documented LOC
7% 38% 43%
92% 60% 58%
288 288 282
 Section VII—Computers, Engineering, and the Future 467
symptoms). The patients studied all had mild TBI and so CT
and GCS had low sensitivity. None of the indications used
alone appeared to predict outcome. Future studies need to
explore whether the use of BAM together with other param-
eters may create a more accurate predictive tool.
The Brain Acoustic Monitor
as a Continuous Monitor
Acoustic technology may allow for continuous brain monitor-
ing of patients with the potential for deterioration of neuro-
logic status, but who are in good enough condition that an
invasive intracranial monitor is not justified. Because BAM
abnormalities occur in real time relative to changes in cerebral
perfusion, the BAM may be useful. Anecdotally observation
has shown rapid deterioration of BAM signals in patients with
ICP spikes associated with suctioning, pain, or hypoxia, and
rapid improvement of signals in severe TBI patients treated by
head elevation, sedation, or osmotherapy. However, in some
cases in which the ICP changed, the BAM did not. The signifi-
cance of this finding is not clear because changes in intra­
cranial pressure may not be the appropriate comparator
for changes in intracranial blood flow. Preliminary studies in
severe TBI patients suggest that the BAM may predict outcome
in some patients better than ICP,18 but these were patients who
received treatment for ICP. A limitation to use of the BAM for
continuous monitoring is that close coupling of the BAM
sensor to the forehead is required, because a certain amount
of pressure is required to achieve good signals. This may be
feasible for a 15-second “snapshot” BAM recording even when
repeated, but if maintained for longer periods likely will
produce skin breakdown or patient discomfort.
The Brain Acoustic Monitor
in the Operating Room
Surgical procedures (e.g., orthopedic) may be delayed in poly-
trauma patients with moderate or severe TBI because the
anesthetic and surgical stress may produce secondary brain
injury. In addition, the ability to monitor neurologic function
and detect deterioration through clinical examination is lost
in an anesthetized patient. Short-term BAM monitoring
during anesthesia may permit earlier surgery or procedures
such as angiography. The hypothesis requires further testing
but it appears that anesthesia per se—as seen in patients with
intoxication but not brain injury—may not cause abnormal
BAM readings.
Special Cases: Blast and Concussion
Two clinical questions outside the ICU may lend themselves
to BAM use: (1) improvised explosive device (IED)–induced
blast injury in which a triage decision to move the patient to
the next echelon of care or return to duty is critical; and (2)
sports-related concussion that involves a decision whether the
athlete can return to competition. Both injuries can generate
changes in cerebral autoregulation, and thus blood flow
anomalies that change the acoustic signature of the brain,
similar to those noted in general TBI. Detailed studies are
needed to confirm this. An exploration of BAM as a diagnostic
tool in mild TBI was conducted in 2008 at the Shock Trauma
Center, and suggested that BAM could help discriminate those
patients with mild TBI most likely to have postconcussive
symptoms.19
Concussion
A concussion sub-study was performed in a prehospital trial
of the BAM. Data were analyzed post hoc. Concussion was
considered if a patient with a closed head injury and a GCS
of 14 or 15 had documented LOC (brief). In those with no
documented LOC, a GCS of 14 or 15, and negative CT scan
48% had anomalies in their acoustic response similar to the
overall negative CT rate population. Those with a documented
LOC were more likely to show acoustic anomalies particularly
if the GCS was 14 (60%).
Blast
Whether blast injury causes a change in brain acoustics is
more difficult to answer. Blast should generate physical distur-
bances in the brain20; however, BAM assessment of these
injuries in the battlefield has not been possible, and there are
few blast patients seen in civilian trauma centers. There is a
great need in the military for TBI diagnostics. BAM signals
improve relatively quickly after injury (blood flow returning
to normal even as neuronal and glial damage persists). Delayed
BAM evaluation after soldiers have returned from war zones
requires study to determine whether there is any association
between changes in BAM signals over time and neuropsycho-
logical evaluation.
Challenges and Next Steps
in System Development
Multicenter Testing
Acoustic brain monitoring is in its infancy, but its develop-
ment can be driven in a number of different directions by both
clinical need and technologic possibility. At the time of pub-
lication, BAM has only been tested in a single center, the
R Adams Cowley Shock Trauma Center, albeit on more than
1000 patients and in one of the largest trauma centers in the
world. The BAM requires multicenter testing before U.S. Food
and Drug Administration (FDA) clearance.
Sensitivity and Specificity Verification
To date, the BAM has been tested on patients who have had a
head CT due to suspected TBI; BAM sensitivity is more than
90% with a specificity between 15% and 40%. It is conceivable
that BAM detects “physiologic” anomalies that are not so
severe that an anatomic defect can be seen on CT, but this
hypothesis still needs study. In 2011, Dutton et al published a
prospective study on 369 patients with a mechanism of injury
consistent with TBI who were studied with CT, BAM, and
clinical evaluation.19 In addition, BAM data was obtained
from 50 normal volunteers and 49 trauma control patients
thought not to have TBI. BAM discriminated between patients
with mild TBI versus without TBI. An abnormal BAM signal
was suggestive of a new abnormality on CT. The sensitivity of
BAM abnormality for head CT abnormality was 100%, with
a specificity of 30.14%. The results suggested that BAM screen-
ing may be a useful diagnostic adjunct in patients with mild
468 Section VII—Computers, Engineering, and the Future
TBI. In particular when there is loss of consciousness, an
abnormal BAM reading adds significance to LOC as a predic-
tor of a new abnormality on head CT.
What the Brain Acoustic Monitor Measures
The BAM measures only injuries that create a global acoustic
effect on the brain and likely measures the effects of an inter-
ference in the normal perfusion patterns of the brain. Studies
that can verify exactly what is measured should help define a
role for passive brain acoustics in brain injury and the neuro-
critical care unit. This may require PET imaging or be feasible
with CT perfusion scanning.
Emergency Medical Services Applications
Because existing BAM technology is handheld, it most likely
will find a home in the emergency medical services (EMS)
arena.
Will Future BAM Configurations Be
Compatible with the EMS Environment?
The latest embodiment of the BAM for EMS is in a rugged
PDA. An early prototype is shown in Figure 47.6. The Air
Force’s Critical Care Air Transport Team is funding the devel-
opment of a plug-in type module, whereby the BAM can be
used with other types of vital sign assessment modalities. This
would allow a medic to quickly assess for intracranial pathol-
ogy and then track a patient’s course over a period of extended
transport.
Is the Brain Acoustic Monitor
Rugged Enough?
This is a constant question of the EMS groups. In its original
EMS embodiment, the iPAQ, the system was prone to failure
due to connector, battery, and other issues. Newer plug-in
versions appear to survive the rigors of field use better than
early prototypes.
What About Cost?
The BAM system and the technology that underlies it are
simple in nature (i.e., an electronic stethoscope with display
and simple interpretation algorithms). Cost is an important
question to EMS groups, many of which are volunteer services
with limited budgets. Indications are that the BAM will be
relatively inexpensive, particularly when compared with other
diagnostic tools for brain injury now in use.
Stroke Applications
BAM technology may have applications in stroke diagnosis.
Several observations support this concept although more
study is required to verify this.
1. Stroke appears to offer a “purer” case for passive acoustics,
unlike trauma, which can involve diffuse, focal, or a com-
bination of injuries. Stroke generally is of two types—
ischemic or hemorrhagic, and at least within 3 hours after
the onset of symptoms there is a different BAM signature.
This may be why the “stroke” signatures are easier to dis-
tinguish than in trauma, e.g. subarachnoid hemorrhage
(SAH) in a trauma patient will look different and often
“noisier” than SAH in a stroke patient, possibly because in
TBI there is more than one cause of brain tissue
disturbance.
2. Recent pooled analysis of the ATLANTIS,21, 22 ECASS,23,24
and NINDS25 rt-PA trials shows that earlier treatment is
better with the greatest benefit in patients whose treatment
started within 90 minutes of symptom onset.26 During this
time period and in patients who also have a very high
stroke scale (i.e., >20), there is a higher divergence by a
factor of 10 dB more than the normal threshold for
“typical” ischemic stroke. Thus passive acoustics may help
determine a role for rt-PA.
3. There also is a higher incidence of arterial “noise” in
patients prone to stroke, such as plaque in cerebral arteries.
This may be an adjunct to an evaluation of stroke risk.
Conclusion
Passive brain acoustics in its current embodiment in the BAM
has revealed a new and potentially fruitful embodiment of a
very traditional tool: the stethoscope. Many of the develop-
ments that have contributed to this technology come from the
arena of underwater acoustics and face the same issues that
many such transition technologies face. Studies to date,
however, demonstrate enough potential to suggest that the
BAM system is a low-risk method to examine how trauma and
stroke affect the brain.
References
1. Geeraerts T, Merceron S, Benhamou D, et al. Non-invasive assessment of
intracranial pressure using ocular sonography in neurocritical care patients.
Intensive Care Med 2008;34(11):2062–7. Epub 2008 May 29.
2. Rajajee V, Vanaman M, Fletcher JJ, et al. Optic nerve ultrasound for the
detection of raised intracranial pressure. Neurocrit Care 2011;Jul 19. Epub
ahead of print.
3. Moretti R, Pizzi B. Ultrasonography of the optic nerve in neurocritically ill
patients. Acta Anaesthesiol Scand 2011;55(6):644–52. Epub 2011 Apr 4.
4. Querfurth HW, Lieberman P, Arms S, et al. Ophthalmodynamometry for
ICP prediction and pilot test on Mt. Everest. BMC Neurol 2010;10:106.
5. Kasprowicz M, Schmidt E, Kim DJ, et al. Evaluation of the cerebrovascular
pressure reactivity index using non-invasive finapres arterial blood pressure.
Physiol Meas 2010;31(9):1217–28. Epub 2010 Jul 28.
6. Zweifel C, Castellani G, Czosnyka M, et al. Noninvasive monitoring of
cerebrovascular reactivity with near infrared spectroscopy in head-injured
patients. J Neurotrauma 2010; 27:1951–8.
7. Brandi G, Bechir M, Sailer S, et al. Transcranial color-coded duplex
sonography allows to assess cerebral perfusion pressure noninvasively
following severe traumatic brain injury. Acta Neurochir (Wien) 2010;152:
965–72.
8. Apparatus and method for measurement of intracranial pressure with lower
frequencies of acoustic signal, US Patent Number 5,919,144, issued July 6,
1999.
9. Brain assessment monitor, US Patent Number 6,887,199, issued May3, 2005.
10. Goodman JC. Pathophysiology. In: Marion D, editor. Traumatic brain injury.
New York: Thieme; 1999. p. 147.
11. Kinsler LE, Frey AR. Fundamentals of acoustics. 2nd ed. New York: John
Wiley & Sons; 1962. p. 108. ff.
12. Burdic WS. Underwater acoustic system analysis. Englewood Cliffs, NJ:
Prentice-Hall; 1984. p. 64.ff.
13. Houtsma AJ, Sewell JM. A dual mode noise-immune stethoscopes for use in
noisy vehicles. Army Sciences Conference. Orlando: Sept. 2006.
14. Dutton R, Sewell J, Aarabi B, et al. preliminary trial of a noninvasive brain
acoustic monitor in trauma patients with severe closed head injury.
J Trauma 2002;53(5):857–63.
 Section VII—Computers, Engineering, and the Future 469
15. LaMonte MP, Sewell JM. A non-invasive portable acoustic diagnostic system
to differentiate ischemic from hemorrhagic stroke. J Neuroimaging 2005;15:
57–63.
16. Hademenenos GJ, Massoud TF. The physics of cardiovascular disease. New
York: Springer; 1998. p. 104.
17. LaMonte MP, Medlin T, Gaasch W. Intravenous thrombolytic therapy for
stroke: a review of recent studies and controversies. Ann Emerg Med 1999;
34(2):244–55.
18. Dutton RP, Sewell J, Aarabi B, et al. preliminary trial of a noninvasive brain
acoustic monitor in trauma patients with severe closed head injury.
J Trauma 2002;53:857–63.
19. Dutton RP, Prior K, Cohen R, et al. Diagnosing mild traumatic brain injury:
where are we now? J Trauma 2011;70(3):554–9.
20. Bayly P. MRI measurement of acceleration-induced brain deformation,
neurological effects of blast injury conference. Bethesda, MD, April 2008.
21. Clark WM, Albers GW, Madden KP, et al. Recombinant tissue-type
plasminogen activator (alteplase) for ischemic stroke 3 to 5 hours after
symptom onset: the ATLANTIS study—a randomized controlled trial. JAMA
1999;282:2019–26.
22. Clark WM, Wissman S, Albers GW, et al. The rt-PA (alteplase) 0- to 6-hour
acute stroke trial. Part A (A0267g): results of a double-blind, placebo-
controlled multicenter study. Stroke 2000;31(4):811-6.
23. Hacke W, Kaste M, Fieschi C, et al. For the ECASS Study Group. Intravenous
thrombolysis with recombinant tissue plasminogen activator for acute
hemispheric stroke: the European Cooperative Acute Stroke Study (ECASS).
JAMA 1995;274:1017–25.
24. Hacke W, Kaste M, Fieschi C, et al. Randomized double-blind, placebo-
controlled trial of thrombolytic therapy with intravenous alteplase in acute
ischaemic stroke (ECASS II). Lancet 1998;352;1245–51.
25. The National Institute of Neurological Disorders and Stroke (NINDS) rt-PA
Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke.
N Engl J Med 1995;333(24): 1581–7.
26. Association of outcome with early stroke treatment: pooled analysis of
ATLANTIS, ECASS, and NINDS rt-PA stroke trials. Lancet 2004;768–74.
VII
Chapter
48  
Past, Present, and Future
Developments of Intracranial
Monitoring
David M. Benglis, Jr., Brett Trimble, and M. Ross Bullock
“Monitors themselves do not save lives, but the insightful
interpretation of the information derived from them can.”
—Saul and Ducker
Introduction
Traumatic brain injury (TBI) in the United States affects more
than 1.5 million individuals per year costing more than $60
billion in both direct medical and indirect costs such as loss
of productivity.1-4 Following the initial insult, a series of
complex pathophysiologic cascades are set in place, which
synergistically interact to cause delayed, secondary brain
damage, over hours, weeks, and even up to a year after the TBI.
Physicians have the capability to monitor and reverse these
secondary damage mechanisms. A host of modalities, includ-
ing cerebral blood flow (CBF), brain oxygen, intracranial pres-
sure (ICP), biomarkers, and electrical cortical activity, among
others, are available in state-of-the-art neurocritical care
units (NCCU).
Despite this, no single brain monitoring method has yet
emerged to supersede serial neurologic observations by a well-
trained neuroscience critical care nurse. Many promising
brain-monitoring technologies have been evaluated, but after
initial enthusiastic use, have not continued to be used in
patient care. Examples include jugular bulb oximetry and the
“Neurotrend” system. Each of these represents ingenious tech-
nology able to generate meaningful data, yet they have failed
to affect patient care in the long term. Nevertheless, technolo-
gies to monitor the brain continue to evolve and be developed,
and much of what is done in the NCCU is to recognize changes
in brain physiology and function and how the brain and the
rest of the body interact. Ideally this is done through multi-
modality monitoring with a bioinformatics support platform,
two concepts that will drive the future of monitoring in
neurocritical care.5-7
Aims of This Chapter
In an ideal system, information derived from monitoring
should be specific, online or real-time, and filtered so that the
pertinent information is presented, changes from the norm
470
are identified, and measures to correct these abnormalities
can be instituted. Ideally these methods should also include
understanding the mechanisms associated with brain damage
at the cellular level. This chapter briefly discusses the current
neuromonitoring devices and techniques available, as well as
new technologies projected to be available in the upcoming
years. In addition it covers the role of industry in the
advancement of new ideas and nanotechnology, and envis-
ages the intensive care unit (ICU) of the future. Key issues
relevant to modern ICUs are whether improved brain moni-
toring techniques can ultimately affect individuals’ quality of
life and outcome.
The Basics: Glasgow Coma Scale
and Physical Examination
The brain is 2% of total body mass yet receives 15% to 20%
of the cardiac output.8 Because this tissue is extremely sensitive
to hypoxia and ischemia, maneuvers to prevent critical reduc-
tions in arterial blood pressure (invasive arterial monitor or
noninvasive cuff), peripheral arterial oxygen content (oxygen
saturation monitor), intrinsic cardiac function (electrocardio-
gram), and pulmonary function (end-tidal CO2 or arterial
blood gas analysis) output assessment are important in critical
care. A standardized neurologic exam (Glasgow Coma Scale
[GCS]) introduced by Teasdale and Jennett in 1974 can be
used for a basic assessment of brain function. It has prevailed
as a simple yet critical bedside neuromonitoring technique
because of its sensitivity to detect meaningful changes and its
ability to be reproduced with a little variation.9
When possible, new technologies should be compared to
established gold standard techniques for specific measure-
ment of changes in the injured brain. For example, a decrease
in brain oxygen partial pressure reported by an intraparenchy-
mal probe, should be confirmed (e.g., by xenon-enhanced
computed tomography [Xe-CT] CBF) that those changes are
actually accurate and relevant. With the development of por-
table computed tomography (CT) scanners that also allow
evaluation of perfusion and flow, this may become a com-
monly accessible modality in many NCCUs.10
© Copyright 2013 Elsevier Inc. All rights reserved.
 Section VII—Computers, Engineering, and the Future 471
A more than a number and that use of a single threshold at
D mm Hg
–100
–80
–60
–40
–20
–0
120 105 90 75 60 45 30 15 0 initiation of monitoring.24 A mismatch or uncoupling between
Minutes
B D = Tapping of fluid (8 mL)
Fig. 48.1  A, Ventricular monitoring device used by Nils Lundberg
(Swedish neurosurgeon).163 B, Ventricular pressure tracing before and
after drainage of cerebrospinal fluid. (Courtesy Wilkins RH, Wilkins G, et al.
Neurosurgical classics II. American Association of Neurological Surgeons, Park
Ridge, IL, 2000.)
Monitors of Brain Substrate
Delivery
ICP Monitoring
Although ICP monitoring has been in use for greater than 50
years (Fig. 48.1, A and B), its utility is still questioned by
some, who allege its use has not been demonstrated to save
lives or improve outcome, in controlled trials. Such a trial
sponsored by the National Institutes of Health (NIH) in Latin
America to address this criticism was completed in late 2012.
Nevertheless, these issues have been extensively reviewed in
the latest revision of the severe TBI management guidelines,
and level II evidence exists to supports ICP treatment in indi-
viduals with sustained ICPs greater than 20 mm Hg.11 With
ICP and mean arterial blood pressure (MAP) measurements,
cerebral perfusion pressure (CPP) can then be derived. The
CPP value necessary to prevent ischemia varies according to
the individual, but in general is between 50 and 70 mm
Hg.11,12 Requirements for ICP monitors should be that the
device is low cost, accurate, and reliable. Fluid-coupled exter-
nal strain gauge intraventricular catheters traditionally cost
less, but may experience significant measurement drifts over
time and have a greater incidence of malfunction. Transducer-
tipped devices (micro strain gauge [Codman, Raynham, MA]
versus fiber-optic (Integra NeuroSciences, Plainsboro, NJ)
cost slightly more, may be placed in either brain parenchyma
or the ventricle, and may have greater accuracy over time
(e.g., microstrain catheter), but cannot be recalibrated once
placed.11-13 Wet/dry systems are now available, such as the
Spiegelberg (Hamburg, Germany) transducer, which keep the
compartments to measure ICP and drain CSF separate. These
devices also have automatic zeroing functions. In conclusion,
ICP monitoring currently plays a major role in monitoring
brain status in the ICU. It should be considered in all patients
where ICP is a question until alternative methods prove to be
better. Developments will see an evolution of noninvasive
devices, although at present none are able to replace invasive
ICP monitors.14-17 In addition, it is well recognized that ICP is
which to treat may be an oversimplification. Hence further
evolution of ICP monitoring will include use of trend analysis
and indices of autoregulation, pressure reactivity, and compli-
ance to name a few.18-22
SjvO2, PbtO2, CBF, NIRS
Because brain ischemia and hypoxia are common following
TBI, the maintenance of tissue oxygenation above critical
values is essential in modern NCCUs.23 This reduction in
blood flow and oxygen delivery can occur within minutes after
TBI; this stresses the importance of early intervention and
substrate demand and substrate supply also may be present
because of the loss of cerebral autoregulation, and these
changes can be present on a regional or more global scale.25
These tissues are thus at risk for permanent damage from
critical reductions in the continuous delivery of glucose,
oxygen, and other essential substrates.19 Aerobic metabolism
becomes superseded by anaerobic mechanisms, ionic homeo-
stasis is compromised, and calcium influx into cells along
with potassium efflux causes widespread neuronal dysfunc-
tion and edema.
SjvO2
Jugular venous oxygen saturation (SjvO2) monitoring was
implemented in the early 1990s as a promising method to
detect post-traumatic changes in global cerebral oxygenation
(see Chapter 32).26 The average SjvO2 in normal subjects is
62%; a decrease less than 50% is defined as critical ischemia
and is associated with a worse outcome.27 Mainstream use of
this monitoring device has been limited, however, because of
problems encountered with catheter migration or poor posi-
tioning, calibration errors (e.g., light absorbance), limited
“time of good data quality,” and the inability to detect all
causes of decreased brain oxygenation.26,28-30 Thus in some
studies, more than 50% of measurements were erroneous,
leading to far less use of the method in most centers. SjvO2
also requires intensive supervision by specialized personnel.31
Future use of retrograde jugular catheters may also include
the sampling of biomarkers, cytokines, or other factors such
as endothelin that may influence CBF.32,33
PbtO2
A more direct approach to measure focal fluctuations in brain
oxygen tension (PbtO2) can be accomplished with Licox (GMS-
Integra-Kiel, Mielkendorf, Germany) or Neurotrend (Codman,
472 Section VII—Computers, Engineering, and the Future
Raynham, MA) sensors (see Chapter 35).34 Optical sensors
contained within the Neurotrend probe detect dye color
changes resulting from differences in gas concentrations and
pH of the surrounding tissues. One advantage of this particu-
lar device is the capability to measure oxygen concentrations
and other modalities including carbon dioxide (PbtCO2), pH,
and temperature (T).34 The Licox sensor consists of a minia-
turized Clark electrode with a polyethylene casing enclosing a
solution of buffered potassium phosphates. Oxygen concen-
tration is correlated with the electrical current created across
a silver anode and gold-coated silver cathode.8 Sampling area
is approximately 14 mm2, which is seven times the sampling
area of the Neurotrend probe.34
Critical thresholds of oxygen concentration have been
determined for this technique. The oxygen value picked up by
the sensor is a function of diffusion of oxygen from the eryth-
rocyte or plasma to the extracellular space and it integrates all
of the venous and arterial vessel oxygen tensions in the vicin-
ity.23 Following placement, an increase of inspiratory oxygen
fraction (FiO2) to 100% can confirm the sensitivity of the
device.35 With Neurotrend, sustained values of PbtO2 that are
less than 20 to 25 mm Hg are associated with poor outcome.10
Using Licox technology, PbtO2 concentrations between 25 and
30 mm Hg are found in TBI patients with CPP and ICP in the
normal range, but values less than 10 to 15 mm Hg are con-
sidered a sign of tissue at risk, and values less than 5 to 10 mm
Hg indicate ongoing critical values, or, rarely, sensor malfunc-
tion. Furthermore, brain hypoxia (<10 mm Hg) appears to be
an independent factor associated with poor outcome.36,37 Mea-
sures to increase PbtO2 then need to be instituted.11,30,38,39 When
PO2 is maintained at normal values, it is associated with better
outcomes in TBI patients.40 Persistent decreases in PbtO2 also
correlate with increases in brain lactate, and can result from
with sustained ICP elevations, respiratory distress, and hypo-
tension. In these studies, however, it was not possible to cor-
relate brain PaCO2 and pH with cerebral blood flow (CBF)
and PbtO2.10,25 For example, van den Brink et al. in a study of
101 patients with severe nonpenetrating TBI (GCS ≤8), the
magnitude and length of time of low PbtO2, measured by Licox
were an independent factor associated with an unfavorable
outcome and death. In addition, they observed that the devices
demonstrated little zero drift and values were reliable over the
time course of monitoring.41
It remains unclear what is the optimal brain territory in
which to place the probe.42 When placed in contused brain,
increasing the FiO2 does not result in increases in PbtO2 to the
same degree it does in noncontused brain, thus suggesting
that CBF is too low to provide adequate substrate delivery
in these contused areas.43 Therefore to recognize critical
decreases of PbtO2 some groups recommend placement in
nonlesioned white matter in the frontal lobe for diffuse cere-
bral injury or on the affected side for a unilateral injury.34,43
Because of this heterogeneity of measured values between dif-
ferent areas of brain and the potential influence of local
factors (e.g., microhemorrhages) that may contribute to the
measurement, CT analysis of probe position is necessary.41,44,45
Most authors agree that PbtO2 monitoring should not be used
as a “stand-alone” monitoring technique and is best com-
bined with ICP monitoring.23
The Neurotrend sensors are no longer available because of
commercial reasons. This emphasizes that although neuro-
monitoring technologies may be accurate and useful in
clinical practice, small companies may be unable to sustain
production costs for sophisticated microsensor systems
without sufficient volumes of sales. The intrinsic conservatism
of clinicians and hospital purchasing systems compound the
problem. The award of Small Business Innovation Research
(SBIR) grants may serve as a mechanism to help defray
production costs, until uptake of new technologies achieves
critical levels for commercial viability. In late 2012, an NIH-
sponsored phase II randomized clinical trial to compare
guided management of severe TBI using ICP alone to ICP and
PbtO2 (based on a Licox monitor) was halfway through patient
recruitment. This trial hopefully will help clarify the utility
and cost effectiveness of PbtO2 monitoring.
Cerebral Blood Flow (LDF, TDF Probes)
Laser Doppler flowmetry (LDF) consists of an implanted sub-
cortical fiber-optic laser probe (0.5-1 mm in diameter) that
measures the shift of reflected laser light induced by move-
ment of red blood cells (RBCs). Data are presented in real
time, and variations in blood flow are monitored (see Chapter
31). The information is not quantitative, however, and repre-
sents only relative changes. Absolute values may be obtained
when LDF is calibrated against other quantitative measures of
CBF such as xenon-enhanced CT (see Chapter 27).12,46 Thermal
dilution flow (TDF) methods (e.g., Bowman or Carter probes)
are based on conductance of small amounts of heat between
two small gold electrodes, one equipped with a sensitive ther-
mocouple sensor (see Chapter 31). CBF is proportional to heat
conductance, and the values are thus absolute, but the probes
are usually configured for placement in deep white matter
rather than gray matter. CBF changes therefore are less (e.g.,
normal CBF in white matter is ∼22 mL/100 g/min versus
80 mL/100 g/min in gray cortex). These CBF sensors usually
are placed through a “bolt” system and can be placed at cra-
niotomy (e.g., for aneurysm clipping). Clinical acceptance of
both these monitors of CBF is limited in part because relative
CBF values are reported, because of questionable reliability
and validity, and because of questions on how to use the
information.47 In particular it is difficult to make conclusions
about CBF unless the state of metabolism is known. Neverthe-
less, use of these devices can supplement other monitors and
provide information about autoregulation.48,49 Noninvasive
methods to assess regional and global blood flow and oxygen-
ation are under deve­lopment and may one day replace current
LDF and TDF monitors.
Noninvasive Monitoring of Brain
Oximetry: NIRS
Near-infrared spectroscopy (NIRS) technology was initially
pioneered by Jobsis and Chance, and is an emerging method
to monitor cerebral oxygenation, blood flow, blood volume,
cerebrovascular reactivity, and perhaps even edema in both
adults and children (see Chapter 33).31,50-56 The basis of NIRS
theory rests on the unique properties of light in the near
infrared range to be scattered and absorbed when passing
through biologic tissues by factors found within those tissues
(e.g., chromophores such as water, lipids, cytochrome c
oxidase, and various hemoglobin compounds). In basic terms
brain NIRS involves transmission of light from an emitting
source to a sensor placed nearby. Although transmission NIRS

works well in neonates because of their semitransparent skull
and scalp, NIRS monitoring in adults is degraded by the
increase in skull and scalp tissue density.57,58 To overcome this
limitation, reflectance mode NIRS is used, where an emitter
and detector are separated by a specific distance on the scalp
with the premise that a fixed amount of transmitted, reflected,
and scattered light follows an elliptical path whose depth of
penetration is proportional to the distance of separation
between the emitter and the detector.59
Devices Based on Absorption
Measurements
The absorption spectra of deoxyhemoglobin (HHb) and oxy-
hemoglobin (HbO2) in the infrared spectra between 700 and
900 nm has long been used to measure blood oxygen satura-
tion.54,59 The attenuation of optical signals at convenient wave-
lengths in the near infrared range around the isosbestic point
for hemoglobin (800 nm), are measured and used to deter-
mine the molar concentrations of HHb and HbO2 and oxygen
saturation in blood.53,60 This is possible because water and
most other chromophores are functionally transparent or do
not dramatically vary in absorption characteristics within the
bandwidth interrogated61 (Fig. 48.2).
The INVOS Cerebral Oximeter (Somanetics Inc., Troy, MI)
uses a similar technique to estimate blood oxygen saturation
in the brain. Optical transceiver pairs are placed on the left
and right forehead over the prefrontal cortex. Typically, the
attenuation of light signals at two wavelengths is used to esti-
mate regional blood oxygen saturation. The separation
between emitter-detector pairs is optimized to give as much
signal as possible from tissues that underlie the skull.
These systems are used often during cardiac surgery to
detect cerebral blood desaturation episodes and show promise
in cerebral hematoma detection. However, there is limited
enthusiasm for their use in NCCU monitoring, in part because
of several limitations. First, attenuation due to scattering
0.30
0.25
Water
Absorption factor (cm–1)
0.20
0.15
0.10 HHb
0.05
OxyHb
0
600 700 800 900
Wavelength (nm)
Fig. 48.2  Absorption spectra of hemoglobin, deoxyhemoglobin (HHb),
and water. OxyHb, Oxyhemoglobin. (Reused with permission from Chance B,
Cope M, Gratton E, et al. Rev Sci Instrum 1998;69:3457–81.)
Section VII—Computers, Engineering, and the Future 473
cannot be identified and subtracted. Second, signals from
blood in tissues overlying the brain such as the scalp cannot
be reliably identified and subtracted.62 Third, hair on the scalp
also complicates measurements, and current systems only can
be used in areas of the head not covered by hair such as the
forehead. A few small clinical series, however, describe the
potential application in both the operating room and in
the ICU. For example, Brawanski et al. described a correlation
between intraparenchymal oxygen probes and NIRS data in
nine TBI patients.63 Kirkpatrick et al. found that NIRS regis-
tered changes in cerebral oxygenation that recovered once
blood flow was restored during vessel cross-clamping in
carotid endarterectomy.64 In 12 TBI patients the same investi-
gators reported an increased frequency of correlative changes
(e.g., decrease in peripheral saturation, intracranial hyperten-
sion, hyperemia) in NIRS versus SjvO2.31
Several of the drawbacks of the first generation of cerebral
oximeters have been addressed using techniques and equip-
ment borrowed from radar and communications applica-
tions.59,65-68 This includes time of flight or time delay methods
and multifrequency, multiwavelength phase modulation
methods similar to the well-known time division multiplex
(TDM) and wavelength division multiplex (WDM) methods
used extensively in telecommunications applications. In short,
a radiofrequency carrier wave is applied to the transmitted
optical signal. Phase shifts in the returned optical signals
coupled with the attenuations can be used to deconvolute scat-
tering and absorption independently. In addition, the phase
shift information can be used to spatially resolve the measure-
ments and construct an image.60 In the future it is conceivable
that this technique could be applied to the whole brain and
provide a continuous image of brain oxygen saturation or a
related parameter. However, no commercial product based on
these techniques is available despite the significant body of
research (Fig. 48.3).
Devices Based on Light
Scattering Measurements
The scattering of light in tissues is caused by refractive index
mismatch between adjoining materials and structures within
the tissue. In the near infrared bandwidth of 700 to 900 nm,
MRI
B
4
A
cm
1000 1100
cm
C
9
Fig. 48.3  Functional magnetic resonance imaging (fMRI) (A) versus near
infrared phased array device (B, C) images of hemoglobin increase. (From
Chance B, Nioka S, Zhao Z. A wearable brain imager. IEEE Eng Med Biol Mag
2007;26(4):30–7.)
474 Section VII—Computers, Engineering, and the Future
scattering is the dominant effect thought to outstrip absorp-
tion by two orders of magnitude.59 In vivo scattering mea­
surements show promise in the ability of NIRS to detect
experimental brain edema in rodents.69,70 In these experiments
the optical density of brain tissues was directly interrogated
using NIRS signals and compared with the specific gravity of
tissue samples harvested during the experiment and to ICP
measurements. Water intoxication by direct parenchymal
injection was used to induce brain edema and intracranial
hypertension. The NIRS-measured optical density correlated
with the specific gravity of the tissue and decreased with tissue
swelling.69 In follow-up experiments, transgenic aquaporin-4
knockout mice and wild-type control animals were compared
under the same conditions. The knockout mice had less
decrease in optical density, suggesting that water uptake by
cells may be responsible for the increased scattering and the
subsequent decline in optical density of the tissue.71
Further development of this device holds the promise of a
continuous brain edema detector. If optical density is not
affected by blood volume or indicates an increase in optical
density, the technique may be able to differentiate between
edema from cellular swelling or from vascular engorgement.
The potential shortcomings associated with this technique
include: (1) the effect of solutes such as mannitol and (2) that
the measurement by its nature must be made in situ. Produc-
tion of a “piggyback” device added to current intraparenchy-
mal monitors may soon be feasible. In addition, prototype
devices based on NIRS technology have been used for intra-
cranial hematoma detection, and although at present these
devices cannot replace CT they can in the future play an
important role in the NCCU if the need for patient transport
is avoided.72
Microdialysis
Microdialysis (MD) was introduced to the clinical environ-
ment in the 1980s.73 It consists of a fine probe (0.62 mm in
diameter) placed directly into the brain and may be combined
with ICP and PbtO2 probes.74 The probe is perfused with a
sterile “mock extracellular fluid [ECF]” solution at a slow rate
that allows continuous sampling of the parenchymal extracel-
lular environment with the transfer of substances less than
20 kDa across a dialysis membrane.8 Assays are based on
high-performance liquid chromatography (HPLC) techniques
in small aliquots of dialysate fluid (60 µL). This allows iden-
tification and measurement of various molecules including
glucose, lactate, potassium, pyruvate, nitric oxide, glutamate,
and glycerol.75-77 The use of MD probes with much higher
molecular weight cutoff (e.g., 70 or 100 kDa) allows protein
fragments, and small peptides, such as cytokines (e.g., inter-
leukin [IL]-1, IL-6, trophic factors, and amyloid precursor
protein fragments) to be measured.78 Microdialysis has several
limitations. First, measurements are representative of relative
concentrations of the extracellular molecules and do not rep-
resent actual concentrations. This can limit data comparison
between different machines and centers.79 Second, the infor-
mation is not truly online, although newer devices, such as
the CMA600 (CMA Microdialysis, Stockholm, Sweden) have
shortened times of bedside substrate analysis to approxi-
mately 1 to 2 hours.79,80 Third, result specificity may be affected
by cell edema, and changes in the ECF space around the
catheter.12,29,77,81
TBI, subarachnoid hemorrhage (SAH), and stroke among
other disorders treated in the NCCU are associated with a
cascade of events: excitatory amino acid (EAA) release, the
opening of various ion channels that results in calcium influx
and potassium efflux, failure of the sodium-potassium ade-
nosine triphosphatase (Na+/K+-ATPase) transport, mitochon-
drial derangement, cytoskeletal breakdown, and a shift from
aerobic to anaerobic metabolism that evolves over time82 (see
Fig. 48.1). These changes have been evaluated in humans using
microdialysis and show that adverse clinical events (e.g., ele-
vated ICP, hypotension, or hypoxia) are associated with dialy-
sate concentration increases (e.g., lactate, potassium, EAAs,
or decreases such as glucose.75,83-86 There is a suggestion that
the changes—elevated lactate-to-pyruvate (LP) ratios and
glycerol—may precede an increase in ICP.87 In addition, a well-
described and independent association exists between extra-
cellular metabolic markers and outcome after TBI.88 However,
it remains unclear whether treatment guided by microdialysis
findings can alter outcome, although its use can be used to
identify episodes of cerebral compromise not detected by con-
ventional techniques (ICP and CPP)89 and to better stratify or
target therapies based on metabolic disturbances.90
Microdialysis in the NCCU is largely a research tool, and
more research and technology development are required
before it is adapted into routine clinical use.91 Development
likely will include the ability to obtain both regional and non-
regional information; the construction of analyzers (e.g.,
direct electrochemical detectors) that can give real-time read-
outs of an analyte such as lactate or glutamate; and evolution
of dialysis membranes. However, the “best” analyte to be mea-
sured remains unclear.
Functional Monitoring: Glasgow
Coma Scale, Electroencephalogram,
Electrocorticogram, Evoked
Potentials, Pupillometer
Glasgow Coma Scale and
Neurologic Assessment
Outcome following acute brain injury including TBI, SAH,
and stroke is associated with the initial neurologic condition.
Hence assessment of consciousness, pupils, and limb move-
ments is a foundation of neuromonitoring (see Chapter 10).
This neurologic assessment is a continual process and can
provide reliable and dynamic information about the injured
brain. The examination, however, may be limited following
intubation, sedation, and administration of chemical paraly-
sis, and so standardized and objective scales are valuable. Since
Teasdale and Jennett’s initial description of the GCS in 1974,
it has evolved into the gold standard neuromonitoring tech-
nique because it is simple, reproducible, objective, and has
limited intra- and interobserver variability. Other monitors
therefore should be used with the GCS.8
Brainstem Auditory Evoked Potentials
Brainstem auditory evoked potentials (BAEPs) assess the
integrity of the auditory efferent pathways. Normally there are
five waveforms: waves I through V represent signals generated
at the vestibulocochlear nerve, cochlear nuclei, superior
 Section VII—Computers, Engineering, and the Future 475
olivary complex, lateral lemniscus, and inferior colliculi,
respectively. BAEPs are useful in posterior fossa surgery but
have a limited role in assessment of patients in the NCCU,
although they may help confirm the diagnosis of brain death
(e.g., absence of all waveforms or presence of wave I).8 On the
other hand, somatosensory evoked potentials (SSEPs) may
play a role in prediction of outcome.92 This is further reviewed
in Chapter 24, and guidelines on the use of clinical neuro-
physiology in the ICU are available.93
Electroencephalogram, Electrocorticogram,
Cerebral Function Monitoring: Seizures
and Spreading Depression
Subclincal, nonmotor, or partial seizures may occur in about
20% of patients in the NCCU if looked for, and these noncon-
vulsive seizures can adversely influence outcome.94 Overt
motor seizures, however, may be masked by chemical paraly-
sis, so monitoring for epileptiform activity can be an impor-
tant adjunct to ICP, brain oximetry, or CBF monitors.12,95,96
Electroencephalogram (EEG) data also are associated with
outcome and can be used to guide therapeutic interventions
before irreversible damage ensues.97-99 The presence of epilep-
tiform activity also is linked to episodic increases in ICP and
the LP ratios.100 Finally, EEG data can be used as a surrogate
endpoint to test the efficacy of new drugs without having to
wait for functional outcome measures (typically 6 months to
1 year).
EEG monitoring is noninvasive, and machines used in the
NCCU today are small and portable, contain data reduction
software (e.g., microprocessor technology that splits informa-
tion into spectra dependent on frequency, amplitude, and
location), provide continuous data, and eliminate the need for
a full-time technical assistant at the bedside.8,101 The simplest
form of EEG monitoring includes a noninvasive strip of 2 to
4 electrodes (e.g., bispectral index [BIS]) that analyzes the
depth of sedation and whether seizure activity is present.
These findings correlate with the GCS. However, BIS cannot
regionalize information. Larger EEG units including 8- to
14-channel 10- to 12-electrode montages provide better spatial
resolution and can collect specific information such as alpha,
delta, beta, and theta proportions of the raw EEG.97,98,100
Sophisticated units display complex information in more
simple formats that are easy to read and interpret in the NCCU
and can present the information in color spectra similar to the
appearance of a positron emission tomography (PET) scan.102
Following craniotomy, invasive EEG monitoring is feasible
with placement of an electrocorticogram (ECoG) recording
strip or grid on the cortex near the injury site.103,104 The use of
a single linear strip facilitates removal at the bedside once the
monitoring period is complete. Collected data can be stored
and analyzed offsite by specialized personnel. Use of ECoG
provides insight into cortical spreading depression (CSD) and
peri-infarct depolarization (PID). CSD consists of a wave of
depressed cortical activity or amplitude that spreads slowly
across cerebral gray matter.105 The depression is accompanied
by a transient loss of ion homeostasis, changes in glucose and
lactate, and an increase in CBF and oxygen followed by olige-
mia. ECoG changes are similar to those recorded during PIDs,
which occur in the periphery of infarct regions.103,106-109
However, PID and CSD are different, in that PID may contrib-
ute to irreversible ischemic damage. Whether CSD harbors
long-term effects in patients and what they are is still to be
fully elucidated.104,109-111
Animal studies dating back 60 years show that several
mechanisms including potassium and glutamate in the ECF
can influence CSD. However, the presence of CSD in humans
after TBI has only recently been confirmed and depends in
part on the ability to place ECoG monitors in perilesional
zones rather than standard frontal sites that are often distant
from a lesion.103,104,112,113 CSD may be observed in a third of
these patients and is more likely to occur in younger patients.
In addition, CSD may occur with or precede a seizure, although
this observation is inconsistent and further study of this rela-
tionship is required. Both CSD and PID events can be attenu-
ated with N-methyl-D-aspartic acid receptor blockade in
animals, and so monitoring for these changes in humans may
provide a selection tool for those patients who could benefit
from these therapies in the future.112 Implantable fiber-based
optical probes also have been used to detect preictal changes
in the rodent hippocampus and determine whether scattering
coefficients can predict electrographic seizure onset in vivo.70
Pupillometer
Pupillary assessment is important when defining prognosis
after brain injury and to identify herniation; this component
of the neurologic assessment has the potential to become
“automated” and more quantitative and sensitive through use
of a pupillometer.114-119 The neurologic pupillary index (NPi)
is an objective algorithm based on the pupillary indices
obtained from healthy subjects that uses a handheld digital
camera pupillometer that automatically tracks and analyzes
pupil dynamics (NeurOptics Inc., Irvine, CA)120 (Fig. 48.4).
This technology has the potential to limit subjective descrip-
tions of pupillary change and reduce interobserver variabil-
ity.121 The results of pupil assessment are stored in the device’s
memory and projected onto a liquid crystal display (LCD)
screen. Use of the pupillometer may provide an early warning
of ICP changes that cannot be appreciated by visual inspec-
tion by the health care provider alone (i.e., clinical measures
Fig. 48.4  NeurOptics ForSite quantitative dynamic pupillometer.
Liquid crystal display screen during data targeting. Data acquisition
phase. (Permission granted from NeurOptics, Irvine, Calif.)
476 Section VII—Computers, Engineering, and the Future
could be instituted early, to prevent ICP peaks and CPP
reduction).
New Technologies That May Change
Neuromonitoring: Biomarkers,
Nanotechnology, Microelectronics,
Wireless Systems, “Lab on a Chip”
Biomarkers: The Search for a Troponin
Equivalent in the Brain
The term biomarker has different meanings to different users,
but in the NCCU it implies a substance or parameter that
represents either organ outcome or function. Ideally bio-
markers also need to respond to treatment. Two examples are
the reduction in viral load levels and increase in CD4 cells in
human immunodeficiency virus (HIV) with proper antiret-
roviral treatment and the reduction in cardiac specific tropo-
nin levels following reestablishment of blood flow to the
heart. The ideal “brain biomarker” is still being sought (see
Chapter 18) but has the potential to improve understanding
of the mechanisms associated with cellular, organ, and sys-
temic damage to facilitate future drug development and to
alert clinicians when changes occur to guide manage-
ment.122,123 Similar to how cardiac troponins are specific to
cardiac muscle injury, the ideal neurologic biomarker should
be accessible, specific to the nervous system, identify progres-
sion of secondary damage, and predict outcome.124,125 Ideally
it should appear in blood and CSF immediately after an
injury.124 Serum biomarkers are advantageous because of
sampling ease.126
Potential biomarkers include cytokines (e.g., IL-6, IL-1,
tumor necrosis factor [TNF], and transforming growth
factor-β [TGF-β]) that increase in blood 48 hours post injury,77
brain-specific creatine kinase (CK), glial fibrillary acidic
protein (GFAP), lactate dehydrogenase (LDH), myelin basic
protein (MBP), neuron-specific enolase (NSE), S-100β
(protein expressed constitutively by brain astrocytes), and
spectrin breakdown products (SBP 120), among others, or a
combination of these markers.124 NSE, a glycolytic enzyme,
appears rapidly in serum after brain injury and is specific to
brain tissue124,127 but results are conflicting and in some studies
has failed to separate persons with mild TBI from normal
controls.124,128,129 S-100β also has undergone extensive study
and its levels are associated with the GCS, imaging, and
outcome. However, it also is found in tissue outside the brain,
adipose tissue, and chondrocytes.127,130-132 GFAP is released fol-
lowing TBI but not after multiple trauma without TBI; the
GFAP levels correspond to TBI severity and outcome.133,134
Alpha II-spectrin is a cytoskeletal protein found in axons
and presynaptic membranes that is cleaved by caspase-3
and calpain, two enzymes that are activated following TBI and
can be measured in both blood and CSF. This protein is ele-
vated in patients following TBI compared with controls,124,135
and secondary increases in CSF are observed with clinical
deterioration.
Despite much research there is no ideal biomarker for prac-
tical clinical use in TBI.124,136 It also is conceivable that rather
than a single biomarker, a panel of several biomarkers may be
important.137,138 Proteomic analysis may provide a vehicle for
new molecular discovery.139 A large multicenter trial is
underway to examine multiple biomarkers in TBI (Banyan
Biomarkers, Inc. Alachua, FL). Once an ideal biomarker is
defined, an enzyme-linked immunosorbent assay (ELISA) or
electrochemical detection system could be developed, and a
“chip” or kit devised to allow cheap, rapid, serial measure-
ments at the bedside. The U.S. military has invested heavily in
research designed to produce such a biomarker detection
device that is the size of a cell phone and can be deployed with
medics in combat zones.
Brain Temperature Monitoring
Hyperthermia is harmful after TBI, stroke, SAH, and other
disorders treated in the NCCU, and so monitoring of brain
and systemic temperature is important (see Chapter 37).140
Miniaturized thermocouples incorporated with brain oxygen
sensors are available for parenchymal monitoring because
cerebral temperatures may differ from core temperatures.82 A
large body of evidence in animals and humans suggests that
induction of mild hypothermia (32° C to 35° C) early after
injury may arrest or slow secondary mechanisms and in
humans can control elevated ICP.141-143 However, the results of
randomized clinical trials using hypothermia other than in
cardiac arrest have not demonstrated a consistent benefit of
therapeutic hypothermia after brain injury.144 In part, this
failure of clinical trials may be associated with patient
selection—only specific patients will benefit—when and how
hypothermia is induced, time to target temperature, duration
of hypothermia, and speed of rewarming.142 Use of a brain
temperature monitor, perhaps as part of a micromachined
smart catheter that measures many parameters may facilitate
answers to these questions or simply better target fever control
in the ICU.145-147
ELISA for the Brain?
Electronic and microsensor technology has always been the
limiting factor for neuromonitoring methods development.
Huge advances in computer memory integration, micropro-
cessor speeds, microelectronics, and etched microcircuitry
and immunologic linking have been made. These advances
have led to a “quantum leap” in microanalytic technologies,
such as ELISA techniques (Fig. 48.5) and protein microarray
technologies. This has revolutionized the concept of “point
of care” testing (POCT), which may, in years to come,
render the practice of daily blood drawing in the ICU obso-
lete. These technologies have yet to be incorporated into
neuromonitoring, but may offer the prospect of CSF or
microdialysis based continuous “flow-through” ELISA, for a
cluster of biomarkers, which could detect incipient brain
damage and thus offer new ways to treat or prevent such
damage.
Nanotechnology, Microprocessing,
and Long-Term Projections
“Nanotechnology is concerned with materials and systems
whose structure and components exhibit novel and signifi-
cantly improved physical, chemical, and biologic properties,
 Section VII—Computers, Engineering, and the Future 477
Absorbent pad
Plastic backing
Control reagent
Test reagent
Membrane
with/without
blocking reagent
Conjugate pad
with reagent
Sample pad
with/without reagent
Wicking pad with reagent Sample flow
Fig. 48.5  Enzyme-linked immunosorbent assay (ELISA)–based
flow-through system. The membrane is placed at the bottom of a
sample reservoir and on top of an absorbent pad. Reagents added to
the sample reservoir flow through the membrane into the pad, allowing
discrete steps without having to use bags or other containers.
phenomena, and processes because of their small size.”148
Dimensions of these devices are in the range of 10 to 100 nm.149
This infers control of the orientation of matter at the molecu-
lar or atomic level. At this scale quantum mechanical effects
sometimes cause these structures to have properties different
from bulk materials. Nanostructures such as nano­wires and
quantum dots hold promise in electronics, medicine, and
other applications. In medicine, nanostructures might be used
to interact with individual cells, molecules, proteins, or deoxy-
ribonucleic acid (DNA). In electronics, nanostructures may
allow molecular level electronics and computing that would
enable integrated circuits to operate near the theoretical limit
for power consumption per bit of processing, resulting in the
smallest, fastest electronic devices possible.
Nanowires are structures with a diameter on the order of
tens of nanometers and an unconstrained length. For this
reason nanowires are sometimes said to be one dimensional.
Many materials are used in their construction including
metals (e.g., nickel and gold), insulators (e.g., titanium dioxide
and silicon dioxide), and inorganic or organic molecules (e.g.,
DNA). Electrons in nanowires are constrained laterally; there-
fore certain structures such as carbon nanotubes exhibit dis-
crete values of conductance. The future use of nanowires may
be to connect nanodevices in the construction of complex
circuits for computing, sensing, or wireless data transfer. Sig-
nificant research is being conducted into the use of nanowires
as electronic sensors of genes and proteins. One operating
principle is that a change in chemical potential occurring
during DNA hybridization can act as a field effect gate on a
doped, single-crystal nanowire. The effect is similar to a silicon
field effect transistor (FET). An advantage of such sensors is
that the limitation on the number of sensor elements rests
only in the ability to connect to the individual nanowires.
Theoretically, dense sensor arrays could be fabricated on very
small devices, enabling measurement of large numbers of
genes or proteins from small tissue samples or even cells.150
Fig. 48.6  Silicone nanowire sensors for real-time, electronically
transduced, label-free biomolecular detection of electrolytes.
Nanowires are 14 nm each. (From Cheng M, Cuda G, Bunimovich Y, et al.
Nanotechnologies for biomolecular detection and medical diagnostics. Curr Opin
Chem Biol 2006;10[1]:11–9.)
Nanowires have been used to construct a dopamine sensor
for in vivo use in an implantable deep brain stimulator
(DBS) electrode.151 It is not hard to imagine miniaturized
implantable sensor systems based on nanotechnology capable
of an array of sensing functions that communicate wirelessly
to monitors outside the body. Production relevant to the
NCCU may also come in the form of biomolecular detection
devices, whereby 40-nm fiber-optic probes that use fluores-
cence are able to monitor intra- and extracellular biochemi-
cal events. When the probes are removed, cellular events may
continue to be monitored due to deposition of these fluores-
cent molecules.152 Delays in sampling and analysis of micro-
dialysate may become things of the past with the introduction
of “lab on a chip” technology that will give online answers
about the parenchymal microenvironment or specific bio-
markers released following brain injury.153 Other forms of
biochemical detection include micro and nano electrome-
chanical systems (MEMS/NEMS), which consist of carbon
nanotubes that transduce biochemical events into electrical
signals. These signals could then be processed, analyzed, and
presented in a format to allow for protocolized decision
making in the NCCU.153,154 The tools needed to create even
the basic structures in the sizes and aspect ratios needed are
in their infancy. For example, even such exotic methods as
electron beam lithography cannot yet produce the small
width, high-aspect ratio semiconductor materials needed for
nano­wire biochemical sensor arrays. New methods such as
superlattice nanowire pattern (SNAP) transfer are being
developed that use novel processing techniques, in this case a
fluidics approach (Fig. 48.6). Given the tremendous aca-
demic interest in this area, it is hoped that materials and
techniques soon will be developed and applied to neurocriti-
cal care and other aspects of neurology and neurosurgery.
Applications such as continuous EEG mapping, in situ
protein, and DNA or other chemical analysis are tantalizing
but not yet clinically feasible.
478 Section VII—Computers, Engineering, and the Future
Design of the Future Neuroscience
Intensive Care Unit
“Superdocs,” Central Consoles, Data
Routing, and the Challenges in Integrating
These Methods
Advances in microchip technology are occurring rapidly.
These improvements will enable more refined means of data
processing, storage, and acquisition in the NCCU. The ICU of
the future may be a console operated system in which a central
physician is the “air traffic controller” who has access to all
data continuously and can thus make decisions and institute
therapies at remote locations with wireless systems, possibly
in a different hospital or even city (see also Chapters 4, 42, 43,
and 44).155 In central California, such systems have allowed
smaller level II hospitals to continue to function with remote
intensive care specialists linked in by telemedicine. Early
research suggests this telemedicine approach may help improve
patient outcome, in part because it can answer staffing
needs.156,157 Linked drug-device combinations or “smart
systems” may automate certain functions such as administer-
ing benzodiazepines during the presence of seizure activity,
venting an ICP device, or administering mannitol during ICP
elevations.158 However, there are many hurdles, including
technologic and legal to overcome before such systems become
commonplace.159
The Role of Industry
Industry is a major driving force in whether certain concepts
receive the financial support to be marketed and developed.
Furthermore, crossover technologic advances, such as those
made with microprocessors in the computer industry, trans-
late into new developments downstream for medical devices.
Products are in a constant flux, undergoing development and
replacement. The big neuroscience companies involved in
neurocritical care (e.g., Integra Neuroscience-Plainsboro, NJ;
Medtronic, Goleta, CA; Codman, Raynham, MA) serve as a
“choke point” for the integration and commercialization of
new ideas in NCCU. Development of devices sometimes
depends on the potential market and return on investment
rather than the engineering and clinical expertise, and there
have been visionary, futuristic ideas that have not yet received
the financial support required for widespread production. For
example, one idea that is not yet used in NCCUs is the concept
of monitoring oxidized and reduced states of nicotinamide
adenine dinucleotide (NADH).160 Alternate methods for the
financial jumpstarting of ideas include SBIR grants funded by
the National Institutes of Health (NIH). The SBIR program
supports small businesses to stimulate technologic innovation
in biomedical research (see Chapter 38).161
Innovation is the key to progress in neuromonitoring. This
depends on the interaction between physicians, surgeons and
scientists, information technologists, or engineers from many
disciplines. However, legislation, at least in the United States
has created intense scrutiny of the relationship between health
care providers and industry. This scrutiny comes from both
the public and from Congress (e.g., the Grassley-Kohl Physi-
cian Payments Sunshine Act of 2007). However, it is important
to recognize there is a difference between industry-provided
consultation fees and royalty revenue. Critics argue
that consultation fees are often rewards for “loyal” users of a
particular product, drug, or device. By contrast, royalty
revenue is distributed to those who develop a particular
product, that is to say, innovation is rewarded. On the other
hand, the Bayh-Dole Act for universities of 1980 has catalyzed
an interest by academic institutions to promote innovation
because it allows universities to patent and exclusively license
federally funded inventions. In turn this facilitates the transla-
tion of ideas generated on the bench to patients at the bedside
that requires close collaboration between scientists, entrepre-
neurs, venture capitalists, and industry.162
Conclusion
The field of neuromonitoring is continually expanding with
new developments. This chapter has covered past, present, and
future techniques used to monitor patients in the NCCU.
Much of the research and applications have been in TBI.
Industry plays a major role in which concepts receive financial
backing and ultimately end up in mainstream use. Monitors
should present pertinent, online, and specific information.
Furthermore, the information gathered may also facilitate an
understanding of the mechanisms involved in brain injury,
including TBI, stroke, SAH, and cerebral edema. This may lead
to the engineering of better medicines, the production of
more refined clinical trials, and the improvement of patient
outcomes. Promising methods that may be used increasingly
in the future include biomarkers, seizure and CSD sensing
equipment, and global blood oxygen saturation analysis
through NIRS, and further in the future, nanotechnology and
LOCs.
Given the enormously responsive research and develop-
ment environment in the modern world, the challenge to all
who work in NCCU is clear; there is a need to devise new ideas
and partner with industry, to advance the field of brain moni-
toring and improve outcome for patients. In this manner
patient management can be moved from empiric to targeted
stratified care and ultimately to individualized care based on
genomic and mechanistic analysis.
References
1. Selassie AW, Zaloshnja E, Langlois JA, et al. Incidence of long-term disability
following traumatic brain injury hospitalization, United States, 2003. J Head
Trauma Rehabil 2008;23(2):123–31.
2. Rutland-Brown W, Langlois JA, Thomas KE, et al. Incidence of traumatic
brain injury in the United States, 2003. J Head Trauma Rehabil 2006;21(6):
544–8.
3. Langlois JA, Rutland-Brown W, Wald MM. The epidemiology and impact of
traumatic brain injury: a brief overview. J Head Trauma Rehabil 2006;21(5):
375–8.
4. Finkelstein E, Corso PS, Miller TR. The incidence and economic burden of
injuries in the United States. New York: Oxford University Press; 2006.
5. Wijman CA, Smirnakis SM, Vespa P, et al. Research and technology in
neurocritical care. Neurocrit Care 2012;16(1):42–54.
6. Hemphill JC, Andrews P, De Georgia M. Multimodal monitoring and
neurocritical care bioinformatics. Nat Rev Neurol 2011;7(8):451–60.
doi:10.1038/nrneurol.2011.101.
7. Diedler J, Czosnyka M. Merits and pitfalls of multimodality brain
monitoring. Neurocrit Care 2010;12(3):313–6.
8. Reilly P, Bullock R. Head injury: pathophysiology and management. 2nd ed.
London/New York: Hodder Arnold; 2005.
9. Teasdale G, Jennett B. Assessment of coma and impaired consciousness: a
practical scale. Lancet 1974;2(7872):81–4.
10. Zauner A, Doppenberg E, Soukup J, et al. Extended neuromonitoring:
new therapeutic opportunities? Neurol Res 1998;20(Suppl 1):S85–90.
 Section VII—Computers, Engineering, and the Future 479
11. Guidelines for the management of severe traumatic brain injury.
J Neurotrauma 2007;24 Suppl 1:S1–106.
12. Gupta AK, Bullock MR. Monitoring the injured brain in intensive care:
present and future. Hosp Med 1998;59(9):704–13.
13. Czosnyka M, Czosnyka Z, Pickard JD. Laboratory testing of three
intracranial pressure microtransducers: technical report. Neurosurgery
1996;38(1):219–24.
14. Behrens A, Lenfeld N, Ambarki K, et al. Transcranial Doppler pulsatility
index: not an accurate method to assess intracranial pressure. Neurosurgery
2010;66(6):1050–7.
15. Querfurth HW, Lieberman P, Arms S, et al. Ophthalmodynamometry for
ICP prediction and pilot test on Mt. Everest. BMC Neurol 2010;10:106.
16. Moretti R, Pizzi B. Ultrasonography of the optic nerve in neurocritically ill
patients. Acta Anaesthesiol Scand 2011;55(6):644–52.
17. Rosenberg JB, Shiloh AL, Savel RH, et al. Non-invasive methods of
estimating intracranial pressure. Neurocrit Care 2011;15(3):599–608.
18. Aoi MC, Matzuka BJ, Olufsen MS. Toward online, noninvasive, nonlinear
assessment of cerebral autoregulation. Conf Proc IEEE Eng Med Biol Soc
2011;2011:2410–3.
19. Aries MJ, Czosnyka M, Budohoski KP, et al. Continuous monitoring of
cerebrovascular reactivity using pulse waveform of intracranial pressure.
Neurocrit Care 2012;Apr 3. Epub ahead of print.
20. Sorrentino E, Diedler J, Kasprowicz M, et al. Critical thresholds for
cerebrovascular reactivity after traumatic brain injury. Neurocrit Care
2012;16(2):258–66.
21. Radolovich DK, Aries MJ, Castellani G, et al. Pulsatile intracranial pressure
and cerebral autoregulation after traumatic brain injury. Neurocrit Care
2011;15(3):379–86.
22. Lavinio A, Rasulo FA, De Peri E, et al. The relationship between the
intracranial pressure-volume index and cerebral autoregulation. Intensive
Care Med 2009;35(3):546–9. Epub 2008, Oct 11.
23. Mazzeo AT, Bullock R. Monitoring brain tissue oxymetry: will it change
management of critically ill neurologic patients? J Neurol Sci 2007;261(1-2):
1–9.
24. Bouma GJ, Muizelaar JP, Bandoh K, et al. Blood pressure and intracranial
pressure-volume dynamics in severe head injury: relationship with cerebral
blood flow. J Neurosurg 1992;77(1):15–9.
25. Doppenberg EM, Zauner A, Bullock R, et al. Correlations between brain
tissue oxygen tension, carbon dioxide tension, pH, and cerebral blood flow–
a better way of monitoring the severely injured brain? Surg Neurol 1998;
49(6):650–4.
26. Sheinberg M, Kanter MJ, Robertson CS, et al. Continuous monitoring of
jugular venous oxygen saturation in head-injured patients. J Neurosurg
1992;76(2):212–7.
27. Robertson C. Desaturation episodes after severe head injury: influence on
outcome. Acta Neurochir Suppl (Wien) 1993;59:98–101.
28. Siggaard-Andersen O, Fogh-Andersen N, Gothgen IH, et al. Oxygen status of
arterial and mixed venous blood. Crit Care Med 1995;23(7):1284–93.
29. Sarrafzadeh AS, Kiening KL, Unterberg AW. Neuromonitoring: brain
oxygenation and microdialysis. Cur Neurol Neurosci Rep 2003;3(6):517–23.
30. Kiening KL, Unterberg AW, Bardt TF, et al. Monitoring of cerebral
oxygenation in patients with severe head injuries: brain tissue PO2 versus
jugular vein oxygen saturation. J Neurosurg 1996;85(5):751–7.
31. Kirkpatrick PJ, Smielewski P, Czosnyka M, et al. Near-infrared spectroscopy
use in patients with head injury. J Neurosurg 1995;83(6):963–70.
32. Helmy A, Carpenter KL, Menon DK, et al. The cytokine response to human
traumatic brain injury: temporal profiles and evidence for cerebral
parenchymal production. J Cereb Blood Flow Metab 2011;31(2):658–70.
33. Chatfield DA, Brahmbhatt DH, Sharp T, et al. Juguloarterial endothelin-1
gradients after severe traumatic brain injury. Neurocrit Care 2011;14(1):
55–60.
34. Lang EW, Mulvey JM, Mudaliar Y, et al. Direct cerebral oxygenation
monitoring: a systematic review of recent publications. Neurosurg Rev
2007;30(2):99–106; discussion 107.
35. Scheufler K. Tissue oxygenation and capacity to deliver O2 do the two go
together? Transfus Apheresis Sci 2004;31:45–54.
36. Maloney-Wilensky E, Gracias V, Itkin A, et al. Brain tissue oxygen and
outcome after severe traumatic brain injury: a systematic review. Crit Care
Med 2009;37(6):2057–63.
37. Oddo M, Levine JM, MacKenzie L, et al. Brain hypoxia is a predictor of
short-term outcome after severe traumatic brain injury independent of
intracranial hypertension and low cerebral perfusion pressure. Neurosurgery
2001;69:1037–45.
38. Pascual JL, Georgoff P, Horan A, et al. Brain tissue hypoxia in traumatic
brain injury: are most commonly used interventions successful at correcting
brain tissue oxygen deficits? J Trauma 2011;70(3):535–46.
39. Bohman LE, Heuer GG, Macyszyn L, et al. Medical management of
compromised brain oxygen in patients with severe traumatic brain injury.
Neurocrit Care 2011;14:361–9.
40. Nangunoori R, Maloney-Wilensky E, Stiefel MD, et al. Brain tissue oxygen
based therapy and outcome after severe traumatic brain injury: a systematic
literature review. Neurocrit Care 2011;Aug 16. Epub ahead of print.
41. van den Brink WA, van Santbrink H, Steyerberg EW, et al. Brain oxygen
tension in severe head injury. Neurosurgery 2000;46(4):868–76; discussion
876–8.
42. Ponce LL, Pillai S, Cruz J, et al. Position of probe determines prognostic
information of brain tissue PO2 in severe traumatic brain injury.
Neurosurgery 2012;Jan 27. Epub ahead of print.
43. Sarrafzadeh AS, Kiening KL, Bardt TF, et al. Cerebral oxygenation in
contusioned vs. nonlesioned brain tissue: monitoring of PtiO2 with Licox
and Paratrend. Acta Neurochir 1998;71:186–9.
44. van den Brink WA, Haitsma IK, Avezaat CJ, et al. Brain parenchyma/pO2
catheter interface: a histopathological study in the rat. J Neurotrauma 1998;
15(10):813–24.
45. Lubbers DW, Baumgartl H, Zimelka W. Heterogeneity and stability of local
PO2 distribution within the brain tissue. Adv Exper Med Biol 1994;345:567–74.
46. Carter LP, Weinand ME, Oommen KJ. Cerebral blood flow (CBF)
monitoring in intensive care by thermal diffusion. Acta Neurochir Suppl
(Wien) 1993;59:43–6.
47. Vajkoczy P, Roth H, Horn P, et al. Continuous monitoring of regional
cerebral blood flow: experimental and clinical validation of a novel thermal
diffusion microprobe. J Neurosurg 2000;93(2):265–74.
48. Rosenthal G, Sanchez-Mejia RO, Phan N, et al. Incorporating a parenchymal
thermal diffusion cerebral blood flow probe in bedside assessment of
cerebral autoregulation and vasoreactivity in patients with severe traumatic
brain injury. J Neurosurg 2011;114(1):62–70. Epub 2010, Aug 13.
49. Hänggi D, Participants in the International Multi-Disciplinary Consensus
Conference on the Critical Care Management of Subarachnoid Hemorrhage.
Monitoring and detection of vasospasm. II: EEG and invasive monitoring.
Neurocrit Care 2011;15(2):318–23
50. McCormick PW, Stewart M, Goetting MG, et al. Noninvasive cerebral optical
spectroscopy for monitoring cerebral oxygen delivery and hemodynamics.
Crit Care Med 1991;19(1):89–97.
A complete list of references for this chapter can be found online at
www.expertconsult.com.
 Section VII—Computers, Engineering, and the Future 479.e1
References
1. Selassie AW, Zaloshnja E, Langlois JA, et al. Incidence of long-term
disability following traumatic brain injury hospitalization, United States,
2003. J Head Trauma Rehabil 2008;23(2):123–31.
2. Rutland-Brown W, Langlois JA, Thomas KE, et al. Incidence of traumatic
brain injury in the United States, 2003. J Head Trauma Rehabil 2006;21(6):
544–8.
3. Langlois JA, Rutland-Brown W, Wald MM. The epidemiology and impact
of traumatic brain injury: a brief overview. J Head Trauma Rehabil 2006;
21(5):375–8.
4. Finkelstein E, Corso PS, Miller TR. The incidence and economic burden of
injuries in the United States. New York: Oxford University Press; 2006.
5. Wijman CA, Smirnakis SM, Vespa P, et al. Research and technology in
neurocritical care. Neurocrit Care 2012;16(1):42–54.
6. Hemphill JC, Andrews P, De Georgia M. Multimodal monitoring and
neurocritical care bioinformatics. Nat Rev Neurol 2011;7(8):451–60.
doi:10.1038/nrneurol.2011.101.
7. Diedler J, Czosnyka M. Merits and pitfalls of multimodality brain
monitoring. Neurocrit Care 2010;12(3):313–6.
8. Reilly P, Bullock R. Head injury: pathophysiology and management. 2nd ed.
London/New York: Hodder Arnold; 2005.
9. Teasdale G, Jennett B. Assessment of coma and impaired consciousness:
a practical scale. Lancet 1974;2(7872):81–4.
10. Zauner A, Doppenberg E, Soukup J, et al. Extended neuromonitoring: new
therapeutic opportunities? Neurol Res 1998;20(Suppl 1):S85–90.
11. Guidelines for the management of severe traumatic brain injury.
J Neurotrauma 2007;24(Suppl 1):S1–106.
12. Gupta AK, Bullock MR. Monitoring the injured brain in intensive care:
present and future. Hosp Med 1998;59(9):704–13.
13. Czosnyka M, Czosnyka Z, Pickard JD. Laboratory testing of three
intracranial pressure microtransducers: technical report. Neurosurgery
1996;38(1):219–24.
14. Behrens A, Lenfeld N, Ambarki K, et al. Transcranial Doppler pulsatility
index: not an accurate method to assess intracranial pressure. Neurosurgery
2010;66(6):1050–7.
15. Querfurth HW, Lieberman P, Arms S, et al. Ophthalmodynamometry for
ICP prediction and pilot test on Mt. Everest. BMC Neurol 2010;10:106.
16. Moretti R, Pizzi B. Ultrasonography of the optic nerve in neurocritically ill
patients. Acta Anaesthesiol Scand 2011;55(6):644–52.
17. Rosenberg JB, Shiloh AL, Savel RH, et al. Non-invasive methods of
estimating intracranial pressure. Neurocrit Care 2011;15(3):599–608.
18. Aoi MC, Matzuka BJ, Olufsen MS. Toward online, noninvasive, nonlinear
assessment of cerebral autoregulation. Conf Proc IEEE Eng Med Biol Soc
2011;2011:2410–3.
19. Aries MJ, Czosnyka M, Budohoski KP, et al. Continuous monitoring of
cerebrovascular reactivity using pulse waveform of intracranial pressure.
Neurocrit Care 2012;Apr 3. Epub ahead of print.
20. Sorrentino E, Diedler J, Kasprowicz M, et al. Critical thresholds for
cerebrovascular reactivity after traumatic brain injury. Neurocrit Care
2012;16(2):258–66.
21. Radolovich DK, Aries MJ, Castellani G, et al. Pulsatile intracranial pressure
and cerebral autoregulation after traumatic brain injury. Neurocrit Care
2011;15(3):379–86.
22. Lavinio A, Rasulo FA, De Peri E, et al. The relationship between the
intracranial pressure-volume index and cerebral autoregulation. Intensive
Care Med 2009;35(3):546–9. Epub 2008, Oct 11.
23. Mazzeo AT, Bullock R. Monitoring brain tissue oxymetry: will it change
management of critically ill neurologic patients? J Neurol Sci
2007;261(1-2):1–9.
24. Bouma GJ, Muizelaar JP, Bandoh K, et al. Blood pressure and intracranial
pressure-volume dynamics in severe head injury: relationship with cerebral
blood flow. J Neurosurg 1992;77(1):15–9.
25. Doppenberg EM, Zauner A, Bullock R, et al. Correlations between brain
tissue oxygen tension, carbon dioxide tension, pH, and cerebral blood
flow–a better way of monitoring the severely injured brain? Surg Neurol
1998;49(6):650–4.
26. Sheinberg M, Kanter MJ, Robertson CS, et al. Continuous monitoring of
jugular venous oxygen saturation in head-injured patients. J Neurosurg
1992;76(2):212–7.
27. Robertson C. Desaturation episodes after severe head injury: influence on
outcome. Acta Neurochir Suppl (Wien) 1993;59:98–101.
28. Siggaard-Andersen O, Fogh-Andersen N, Gothgen IH, et al. Oxygen
status of arterial and mixed venous blood. Crit Care Med 1995;23(7):
1284–93.
29. Sarrafzadeh AS, Kiening KL, Unterberg AW. Neuromonitoring: brain
oxygenation and microdialysis. Cur Neurol Neurosci Rep 2003;3(6):
517–23.
30. Kiening KL, Unterberg AW, Bardt TF, et al. Monitoring of cerebral
oxygenation in patients with severe head injuries: brain tissue PO2 versus
jugular vein oxygen saturation. J Neurosurg 1996;85(5):751–7.
31. Kirkpatrick PJ, Smielewski P, Czosnyka M, et al. Near-infrared spectroscopy
use in patients with head injury. J Neurosurg 1995;83(6):963–70.
32. Helmy A, Carpenter KL, Menon DK, et al. The cytokine response to
human traumatic brain injury: temporal profiles and evidence for
cerebral parenchymal production. J Cereb Blood Flow Metab 2011;31(2):
658–70.
33. Chatfield DA, Brahmbhatt DH, Sharp T, et al. Juguloarterial endothelin-1
gradients after severe traumatic brain injury. Neurocrit Care
2011;14(1):55–60.
34. Lang EW, Mulvey JM, Mudaliar Y, et al. Direct cerebral oxygenation
monitoring: a systematic review of recent publications. Neurosurg Rev
2007;30(2):99–106; discussion 107.
35. Scheufler K. Tissue oxygenation and capacity to deliver O2 do the two go
together? Transfus Apheresis Sci 2004;31:45–54.
36. Maloney-Wilensky E, Gracias V, Itkin A, et al. Brain tissue oxygen and
outcome after severe traumatic brain injury: a systematic review. Crit Care
Med 2009;37(6):2057–63.
37. Oddo M, Levine JM, MacKenzie L, et al. Brain hypoxia is a predictor of
short-term outcome after severe traumatic brain injury independent of
intracranial hypertension and low cerebral perfusion pressure.
Neurosurgery 2001;69:1037–45.
38. Pascual JL, Georgoff P, Horan A, et al. Brain tissue hypoxia in traumatic
brain injury: are most commonly used interventions successful at
correcting brain tissue oxygen deficits? J Trauma 2011;70(3):535–46.
39. Bohman LE, Heuer GG, Macyszyn L, et al. Medical management of
compromised brain oxygen in patients with severe traumatic brain injury.
Neurocrit Care 2011;14:361–9.
40. Nangunoori R, Maloney-Wilensky E, Stiefel MD, et al. Brain tissue
oxygen based therapy and outcome after severe traumatic brain injury:
a systematic literature review. Neurocrit Care 2011;Aug 16. Epub ahead of
print.
41. van den Brink WA, van Santbrink H, Steyerberg EW, et al. Brain oxygen
tension in severe head injury. Neurosurgery 2000;46(4):868–76; discussion
876–8.
42. Ponce LL, Pillai S, Cruz J, et al. Position of probe determines prognostic
information of brain tissue PO2 in severe traumatic brain injury.
Neurosurgery 2012;Jan 27. Epub ahead of print.
43. Sarrafzadeh AS, Kiening KL, Bardt TF, et al. Cerebral oxygenation in
contusioned vs. nonlesioned brain tissue: monitoring of PtiO2 with Licox
and Paratrend. Acta Neurochir 1998;71:186–9.
44. van den Brink WA, Haitsma IK, Avezaat CJ, et al. Brain parenchyma/pO2
catheter interface: a histopathological study in the rat. J Neurotrauma 1998;
15(10):813–24.
45. Lubbers DW, Baumgartl H, Zimelka W. Heterogeneity and stability of local
PO2 distribution within the brain tissue. Adv Exper Med Biol 1994;345:
567–74.
46. Carter LP, Weinand ME, Oommen KJ. Cerebral blood flow (CBF)
monitoring in intensive care by thermal diffusion. Acta Neurochir Suppl
(Wien) 1993;59:43–6.
47. Vajkoczy P, Roth H, Horn P, et al. Continuous monitoring of regional
cerebral blood flow: experimental and clinical validation of a novel thermal
diffusion microprobe. J Neurosurg 2000;93(2):265–74.
48. Rosenthal G, Sanchez-Mejia RO, Phan N, et al. Incorporating a
parenchymal thermal diffusion cerebral blood flow probe in bedside
assessment of cerebral autoregulation and vasoreactivity in patients with
severe traumatic brain injury. J Neurosurg 2011;114(1):62–70. Epub 2010,
Aug 13.
49. Hänggi D, Participants in the International Multi-Disciplinary Consensus
Conference on the Critical Care Management of Subarachnoid
Hemorrhage. Monitoring and detection of vasospasm. II: EEG and invasive
monitoring. Neurocrit Care 2011;15(2):318–23
50. McCormick PW, Stewart M, Goetting MG, et al. Noninvasive cerebral
optical spectroscopy for monitoring cerebral oxygen delivery and
hemodynamics. Crit Care Med 1991;19(1):89–97.
51. Brazy JE. Cerebral oxygen monitoring with near infrared spectroscopy:
clinical application to neonates. J Clin Monit 1991;7(4):325–34.
52. Jobsis FF. Noninvasive, infrared monitoring of cerebral and myocardial
oxygen sufficiency and circulatory parameters. Science 1977;198(4323):
1264–7.
479.e2 Section VII—Computers, Engineering, and the Future
53. McCormick PW, Stewart M, Goetting MG, et al. Regional cerebrovascular
oxygen saturation measured by optical spectroscopy in humans. Stroke
1991;22(5):596–602.
54. Lee JK, Kibler KK, Benni PB, et al. Cerebrovascular reactivity measured by
near-infrared spectroscopy. Stroke 2009;40(5):1820–6. Epub 2009, Mar 12.
55. Zweifel C, Castellani G, Czosnyka M, et al. Noninvasive monitoring of
cerebrovascular reactivity with near infrared spectroscopy in head-injured
patients. J Neurotrauma 2010(11):1951–8. Epub 2010, Oct 28.
56. Brady K, Joshi B, Zweifel C, et al. Real-time continuous monitoring of
cerebral blood flow autoregulation using near-infrared spectroscopy in
patients undergoing cardiopulmonary bypass. Stroke 2010;41(9):1951–6.
Epub 2010, Jul 22.
57. Brown JI, Moulton RJ, Konasiewicz SJ, et al. Cerebral oxidative metabolism
and evoked potential deterioration after severe brain injury: new evidence
of early posttraumatic ischemia. Neurosurgery 1998;42(5):1057–63;
discussion 1063–4.
58. Germon TJ, Evans PD, Manara AR, et al. Sensitivity of near infrared
spectroscopy to cerebral and extra-cerebral oxygenation changes is
determined by emitter-detector separation. J Clin Monitor Comput
1998;14(5):353–60.
59. Chance B, Cope M, Gratton E, et al. Phase measurement of light absorption
and scatter in human tissues. Rev Sci Instrum 1998;69(10):3457–81.
60. Chance B, Nioka S, Zhao Z. A wearable brain imager. IEEE Eng Med Biol
Mag 2007;26(4):30–7.
61. Izzetoglu M, Bunce SC, Izzetoglu K, et al. Functional brain imaging using
near-infrared technology. IEEE Eng Med Biol Mag 2007;26(4):38–46.
62. Kytta J, Ohman J, Tanskanen P, et al. Extracranial contribution to cerebral
oximetry in brain dead patients: a report of six cases. J Neurosurg
Anesthesiol 1999;11(4):252–4.
63. Brawanski A, Faltermeier R, Rothoerl RD, et al. Comparison of near-
infrared spectroscopy and tissue p(O2) time series in patients after severe
head injury and aneurysmal subarachnoid hemorrhage. J Cereb Blood Flow
Metab 2002;22(5):605–11.
64. Kirkpatrick PJ, Smielewski P, Whitfield PC, et al. An observational study of
near-infrared spectroscopy during carotid endarterectomy. J Neurosurg
1995;82(5):756–63.
65. Hueber DM, Franceschini MA, Ma HY, et al. Non-invasive and quantitative
near-infrared haemoglobin spectrometry in the piglet brain during hypoxic
stress, using a frequency-domain multidistance instrument. Phys Med Biol
2001;46(1):41–62.
66. Tu T, Chen Y, Zhang J, et al. Analysis on performance and optimization of
frequency-domain near-infrared instruments. J Biomed Opt
2002;7(4):643–9.
67. Chen Y, Tailor DR, Intes X, et al. Correlation between near-infrared
spectroscopy and magnetic resonance imaging of rat brain oxygenation
modulation. Phys Med Biol 2003;48(4):417–27.
68. Chance B, Ma HY, Nioka S. Quantitative brain tissue oximetry, phase
spectroscopy and imaging the range of homeostasis in piglet brain. Adv
Exp Med Biol 2003;530:13–8.
69. Owen C, Gill A, Binder D. Detection of cerebral edema through an
implanted optical probe. Seattle, WA: American Epilepsy Society; 2008.
70. Owen C, Mathews M, Abokasis D, et al. Pre-ictal optical changes in rodent
hippocampus detected with an implanted optical probe. American Epilepsy
Society December 5-9 Seattle, 2008.
71. Binder D. Early detection of cerebral edema through an implanted optical
probe. In: Proceedings of the AANS National Meeting 2008; Chicago, 2008
(abstract).
72. Robertson CS, Zager EL, Narayan RK, et al. Clinical evaluation of a
portable near-infrared device for detection of traumatic intracranial
hematomas. J Neurotrauma 2010;27(9):1597–604.
73. Meyerson BA, Linderoth B, Karlsson H, et al. Microdialysis in the human
brain: extracellular measurements in the thalamus of parkinsonian patients.
Life Sci 1990;46(4):301–8.
74. Sarrafzadeh AS, Sakowitz OW, Callsen TA, et al. Bedside microdialysis for
early detection of cerebral hypoxia in traumatic brain injury. Neurosurg
Focus 2000;9(5):e2.
75. Reinert M, Khaldi A, Zauner A, et al. High level of extracellular potassium
and its correlates after severe head injury: relationship to high intracranial
pressure. J Neurosurg 2000;93(5):800–7.
76. Khaldi A, Zauner A, Reinert M, et al. Measurement of nitric oxide and
brain tissue oxygen tension in patients after severe subarachnoid
hemorrhage. Neurosurgery 2001;49(1):33–8; discussion 38–40.
77. Bullock MR, Doppenberg E, Zauner A, et al. New techniques for
multi-modality monitoring of the injured brain: opportunities for therapy?
Jap J Int Crit Care Med 1997;9(11):1295–300.
78. Hillman J, Aneman O, Anderson C, et al. A microdialysis technique for
routine measurement of macromolecules in the injured human brain.
Neurosurgery 2005;56(6):1264–8; discussion 1268–70.
79. Johnston AJ, Gupta AK. Advanced monitoring in the neurology intensive
care unit: microdialysis. Curr Opin Crit Care 2002;8(2):121–7.
80. Zauner A, Doppenberg E, Woodward JJ, et al. Multiparametric continuous
monitoring of brain metabolism and substrate delivery in neurosurgical
patients. Neurol Res 1997;19(3):265–73.
81. Meixensberger J, Kunze E, Barcsay E, et al. Clinical cerebral microdialysis:
brain metabolism and brain tissue oxygenation after acute brain injury.
Neurol Res 2001;23(8):801–6.
82. Mathew P, Gentleman D, Bullock MR. Brain monitoring in severe head
injury: a practical guide. Trauma 1999;1:105–14.
83. Bullock R, Zauner A, Woodward JJ, et al. Factors affecting excitatory amino
acid release following severe human head injury. J Neurosurg
1998;89(4):507–18.
84. Bullock R, Zauner A, Myseros JS, et al. Evidence for prolonged release of
excitatory amino acids in severe human head trauma: relationship to
clinical events. Ann N Y Acad Sci 1995;765:290–7; discussion 298.
85. Persson L, Hillered L. Chemical monitoring of neurosurgical intensive care
patients using intracerebral microdialysis. J Neurosurg 1992;76(1):72–80.
86. Hutchinson PJ, al-Rawi PG, O’Connell MT, et al. On-line monitoring of
substrate delivery and brain metabolism in head injury. Acta Neurochir
2000;76:431–5.
87. Belli A, Sen J, Petzold A, et al. Metabolic failure precedes intracranial
pressure rises in traumatic brain injury: a microdialysis study. Acta
Neurochir (Wien) 2008;150(5):461–9; discussion 470.
88. Timofeev I, Carpenter KL, Nortje J, et al. Cerebral extracellular chemistry
and outcome following traumatic brain injury: a microdialysis study of 223
patients. Brain 2011;134(Pt 2):484–94. Epub 2011, Jan 18.
89. Chen HI, Stiefel MF, Oddo M, et al. Detection of cerebral compromise with
multimodality monitoring in patients with subarachnoid hemorrhage.
Neurosurgery 2011;69:53–63.
90. Oddo M, Levine J, Frangos S, et al. Brain lactate metabolism in humans
with subarachnoid haemorrhage. Stroke 2012;Feb 16. Epub ahead of print.
91. Goodman JC, Robertson CS. Microdialysis: is it ready for prime time? Curr
Opin Crit Care 2009;15(2):110–7.
92. Houlden DA, Taylor AB, Feinstein A, et al. Early somatosensory evoked
potential grades in comatose traumatic brain injury patients predict
cognitive and functional outcome. Crit Care Med 2010;38:167–74.
93. Guerit JM, Amantini A, Amodio P, et al. Consensus on the use of
neurophysiological tests in the intensive care unit (ICU):
electroencephalogram (EEG), evoked potentials (EPs), and
electroneuromyography (ENMG). Neurophysiol Clin 2009;39:71–83.
94. Vespa PM, McArthur DL, Xu Y, et al. Nonconvulsive seizures after
traumatic brain injury are associated with hippocampal atrophy. Neurology
2010;75:792–8.
95. Vespa P. Continuous EEG monitoring for the detection of seizures in
traumatic brain injury, infarction, and intracerebral hemorrhage: “to detect
and protect.” J Clin Neurophysiol 2005;22(2):99–106.
96. Vespa PM, Nuwer MR, Nenov V, et al. Increased incidence and impact of
nonconvulsive and convulsive seizures after traumatic brain injury as
detected by continuous electroencephalographic monitoring. J Neurosurg
1999;91(5):750–60.
97. Hebb MO, McArthur DL, Alger J, et al. Impaired percent alpha variability
on continuous electroencephalography is associated with thalamic injury
and predicts poor long-term outcome after human traumatic brain injury.
J Neurotrauma 2007;24(4):579–90.
98. Vespa PM, Boscardin WJ, Hovda DA, et al. Early and persistent impaired
percent alpha variability on continuous electroencephalography monitoring
as predictive of poor outcome after traumatic brain injury. J Neurosurg
2002;97(1):84–92.
99. Rossetti AO, Oddo M, Logroscino G, et al. Prognostication after cardiac
arrest and hypothermia: a prospective study. Ann Neurol 2010;67:301–7.
100. Vespa PM, Miller C, McArthur D, et al. Nonconvulsive electrographic seizures
after traumatic brain injury result in a delayed, prolonged increase in
intracranial pressure and metabolic crisis. Crit Care Med 2007;35(12):2830–6.
101. Friedman D, Claassen J, Hirsch LJ. Continuous electroencephalogram
monitoring in the intensive care unit. Anesth Analg 2009;109:506–23.
102. Wallace BE, Wagner AK, Wagner EP, et al. A history and review of
quantitative electroencephalography in traumatic brain injury. J Head
Trauma Rehabil 2001;16(2):165–90.
103. Strong AJ, Fabricius M, Boutelle MG, et al. Spreading and synchronous
depressions of cortical activity in acutely injured human brain. Stroke
2002;33(12):2738–43.
 Section VII—Computers, Engineering, and the Future 479.e3
104. Fabricius M, Fuhr S, Willumsen L, et al. Co-occurrence of spreading
depression and seizures after acute brain injury. Clin Neurophysiol 2008;
119(9):1973–84.
105. Leao A. Spreading depression of activity in cerebral cortex. J Neurophysiol
1944;7:359–90.
106. Hopwood SE, Parkin MC, Bezzina EL, et al. Transient changes in cortical
glucose and lactate levels associated with peri-infarct depolarisations,
studied with rapid-sampling microdialysis. J Cereb Blood Flow Metab
2005;25(3):391–401.
107. Lauritzen M, Jorgensen MB, Diemer NH, et al. Persistent oligemia of rat
cerebral cortex in the wake of spreading depression. Ann Neurol 1982;
12(5):469–74.
108. Hossman K. Periinfarct depolarizations. Cerebrovasc Brain Metab Rev
1996;8:195–208.
109. Kraig RP, Nicholson C. Extracellular ionic variations during spreading
depression. Neuroscience 1978;3(11):1045–59.
110. Busch E, Gyngell M, Elis M, et al. Potassium-induced cortical spreading
depressions during focal cerebral ischemia in rats: contribution to lesion
growth assessed by diffusion-weighted NMR and biochemical imaging.
J Cereb Blood Flow Metab 1996;16:1090–9.
111. Dreier JP. The role of spreading depression, spreading depolarization and
spreading ischemia in neurological disease. Nat Med 2011;17:1–9.
112. Fabricius M, Fuhr S, Bhatia R, et al. Cortical spreading depression and
peri-infarct depolarization in acutely injured human cerebral cortex. Brain
2006;129(Pt 3):778–90.
113. Mayevsky A, Doron A, Manor T, et al. Cortical spreading depression
recorded from the human brain using a multiparametric monitoring
system. Brain Res 1996;740(1-2):268–74.
114. Chesnut RM, Gautille T, Blunt BA, et al. The localizing value of asymmetry
in pupillary size in severe head injury: relation to lesion type and location.
Neurosurgery 1994;34(5):840–5; discussion 845–6.
115. Andrews BT, Pitts LH. Functional recovery after traumatic transtentorial
herniation. Neurosurgery 1991;29(2):227–31.
116. Braakman R, Gelpke GJ, Habbema JD, et al. Systematic selection of
prognostic features in patients with severe head injury. Neurosurgery 1980;
6(4):362–70.
117. Sakas DE, Bullock MR, Teasdale GM. One-year outcome following
craniotomy for traumatic hematoma in patients with fixed dilated pupils.
J Neurosurg 1995;82(6):961–5.
118. Chen JW, Gombart ZJ, Rogers S, et al. Pupillary reactivity as an early
indicator of increased intracranial pressure: the introduction of the
Neurological Pupil index. Surg Neurol Int 2011;2:82. Epub 2011, Jun 21.
119. Fountas KN, Kapsalaki EZ, Machinis TG, et al. Clinical implications of
quantitative infrared pupillometry in neurosurgical patients. Neurocrit Care
2006;5(1):55–60.
120. Hults KN, Knowlton SL, Oliver JW, et al. A study of pupillary assessment in
outpatient neurosurgical clinics. J Neurosci Nurs 2006;38(6):447–52.
121. Litvan I, Saposnik G, Maurino J, et al. Pupillary diameter assessment: need
for a graded scale. Neurology 2000;54(2):530–1.
122. Ingebrigtsen T, Romner B. Biochemical serum markers of traumatic brain
injury. J Trauma 2002;52(4):798–808.
123. Narayan RK, Michel ME, Ansell B, et al. Clinical trials in head injury.
J Neurotrauma 2002;19(5):503–57.
124. Pineda JA, Wang KK, Hayes RL. Biomarkers of proteolytic damage
following traumatic brain injury. Brain Pathol 2004;14(2):202–9.
125. Antman EM, Tanasijevic MJ, Thompson B, et al. Cardiac-specific troponin
I levels to predict the risk of mortality in patients with acute coronary
syndromes. N Engl J Med 1996;335(18):1342–9.
126. Hergenroeder G, Redell JB, Moore AN, et al. Identification of serum
biomarkers in brain-injured adults: potential for predicting elevated
intracranial pressure. J Neurotrauma 2008;25(2):79–93.
127. Johnsson P. Markers of cerebral ischemia after cardiac surgery.
J Cardiothorac Vasc Anesth 1996;10(1):120–6.
128. Ross SA, Cunningham RT, Johnston CF, et al. Neuron-specific enolase as
an aid to outcome prediction in head injury. Br J Neurosurg 1996;10(5):
471–6.
129. Yamazaki Y, Yada K, Morii S, et al. Diagnostic significance of serum
neuron-specific enolase and myelin basic protein assay in patients with
acute head injury. Surg Neurol 1995;43(3):267–70; discussion 270–1.
130. Woertgen C, Rothoerl RD, Metz C, et al.. Comparison of clinical,
radiologic, and serum marker as prognostic factors after severe head injury.
J Trauma 1999;47(6):1126–30.
131. Raabe A, Grolms C, Seifert V. Serum markers of brain damage and
outcome prediction in patients after severe head injury. Br J Neurosurg
1999;13(1):56–9.
132. Raabe A, Menon DK, Gupta S, et al. Jugular venous and arterial
concentrations of serum S-100B protein in patients with severe head injury:
a pilot study. J Neurol Neurosurg Psychiatry 1998;65(6):930–2.
133. Pelinka LE, Kroepfl A, Schmidhammer R, et al. Glial fibrillary acidic protein
in serum after traumatic brain injury and multiple trauma. J Trauma 2004;
57(5):1006–12.
134. Papa L, Lewis LM, Falk JL, et al. Elevated levels of serum glial fibrillary
acidic protein breakdown products in mild and moderate traumatic brain
injury are associated with intracranial lesions and neurosurgical
intervention. Ann Emerg Med 2011;Nov 8. Epub ahead of print.
135. Cardali S, Maugeri R. Detection of alphaII-spectrin and breakdown
products in humans after severe traumatic brain injury. J Neurosurg Sci
2006;50(2):25–31.
136. Laskowitz DT, Grocott H, Hsia A, et al. Serum markers of cerebral
ischemia. J Stroke Cerebrovasc Dis 1998;7(4):234–41.
137. Siman R, Toraskar N, Dang A, et al. A panel of neuron-enriched proteins as
markers for traumatic brain injury in humans. J Neurotrauma
2009;26(11):1867–77
138. Siman R, Giovannone N, Toraskar N, et al. Evidence that a panel of
neurodegeneration biomarkers predicts vasospasm, infarction, and outcome
in aneurysmal subarachnoid hemorrhage. PLoS One 2011;6(12):e28938.
Epub 2011, Dec 9.
139. Prieto DA, Ye X, Veenstra TD. Proteomic analysis of traumatic brain injury:
the search for biomarkers. Expert Rev Proteomics 2008;5(2):283–91.
140. Metz C, Holzschuh M, Bein T, et al. Moderate hypothermia in patients with
severe head injury: cerebral and extracerebral effects. J Neurosurg
1996;85(4):533–41.
141. Busto R, Globus MY, Dietrich WD, et al. Effect of mild hypothermia on
ischemia-induced release of neurotransmitters and free fatty acids in rat
brain. Stroke 1989;20(7):904–10.
142. Polderman KH. Induced hypothermia and fever control for prevention and
treatment of neurological injuries. Lancet 2008;371(9628):1955–69.
143. Sadaka F, Veremakis C. Therapeutic hypothermia for the management of
intracranial hypertension in severe traumatic brain injury: a systematic
review. Brain Inj 2012;Mar 26. Epub ahead of print.
144. Hutchison JS, Ward RE, Lacroix J, et al. Hypothermia therapy after
traumatic brain injury in children. N Engl J Med 2008;358(23):2447–56.
145. Badjatia N. Hyperthermia and fever control in brain injury. Crit Care Med
2009;37(7 Suppl):S250–7.
146. Childs C, Wieloch T, Lecky F, et al. Report of a consensus meeting on
human brain temperature after severe traumatic brain injury: its
measurement and management during pyrexia. Front Neurol 2010;1:146.
147. Li C, Wu PM, Wu Z, et al. Brain temperature measurement: a study of in
vitro accuracy and stability of smart catheter temperature sensors. Biomed
Microdevices 2011;Sep 28. Epub ahead of print.
148. Leary SP, Liu CY, Yu C, et al. Toward the emergence of nanoneurosurgery.
Part I: progress in nanoscience, nanotechnology, and the comprehension of
events in the mesoscale realm. Neurosurgery 2005;57(4):606–34; discussion
634.
149. Lakhtakia A. Nanometer structures: theory, modeling, and simulation.
Bellingham, WA: SPIE Press; 2004.
150. Cheng M, Cuda G, Bunimovich Y, et al. Nanotechnologies for biomolecular
detection and medical diagnostics. Curr Opin Chem Biol 2006;10(1):11–9.
151. Varadan VK, Society of Photo-optical Instrumentation Engineers, et al.
Nanosensors, microsensors, and biosensors and systems 2007. 21-22 March,
2007, San Diego.
152. Vo-Dinh T, Kasili P. Fiber-optic nanosensors for single-cell monitoring.
Anal Bioanal Chem 2005;382(4):918–25.
153. Elder JB, Liu CY, Apuzzo ML. Neurosurgery in the realm of 10(-9). Part 1:
stardust and nanotechnology in neuroscience. Neurosurgery
2008;62(1):1–20.
154. Patolsky F, Zheng G, Lieber CM. Nanowire-based biosensors. Anal Chem
2006;78(13):4260–9.
155. Elder JB, Liu CY, Apuzzo ML. Neurosurgery in the realm of 10(-9). Part 2:
applications of nanotechnology to neurosurgery: present and future.
Neurosurgery 2008;62(2):269–84; discussion 284–5.
156. Zawada ET Jr, Herr P, Larson D, et al. Impact of an intensive care unit
telemedicine program on a rural health care system. Postgrad Med
2009;121(3):160–70.
157. Young LB, Chan PS, Lu X, et al. Impact of telemedicine intensive care unit
coverage on patient outcomes: a systematic review and meta-analysis. Arch
Intern Med 2011;171(6):498–506
158. Gomez EJ, Hernando Perez ME, Vering T, et al. The INCA system: a further
step towards a telemedical artificial pancreas. IEEE Trans Inf Technol
Biomed 2008;12(4):470–9.
479.e4 Section VII—Computers, Engineering, and the Future

159. Gamble KH. Critical care network. ICU telemedicine can help ease the
burden of caring for critically ill patients–provided all the right pieces are
in place. Healthcare Inform 2009;26(12):26–30.
160. Mayevsky A, Chance B. Oxidation-reduction states of NADH in vivo: from
animals to clinical use. Mitochondrion 2007;7(5):330–9.
161. Small business innovation research (SBIR) and small business technology
transfer (STTR) programs. Available at http://grants.nih.gov/grants/
funding/sbirs. Accessed 2008.
162. Thursby JG, Thursby MC. Research and development: where is the new
science in corporate R&D? Science 2006;8(5805):1547–8.
163. Lundberg N. Continuous recording and control of ventricular fluid
pressure in neurosurgical practice. Acta Psychiatr Scand Suppl 1960;36:
1–193.