Foreword by the late

Dr. Leon Goldman

LASERS IN MEDICINE
Edited by

Ronald W. Waynant

CRC PR E S S
Boca Raton London New York Washington, D.C.

Library of Congress Cataloging-in-Publication Data

Lasers in Medicine / edited by Ronald W. Waynant. p. ; cm. Includes bibliographical references and index. ISBN 0-8493-1146-2 (alk. paper) 1. Lasers in medicine. 2. Lasers in surgery. I. Waynant, Ronald W. [DNLM: 1. Lasers. 2. Keratomileusis, Laser In Situ. 3. Laser Coagulation. 4. Laser Surgery. 5. Microscopy, Confocal. WB 117 L34341 2001[ R857.L37L382 2001 610.28--dc21 2001043530 CIP

This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish reliable data and information, but the authors and the publisher cannot assume responsibility for the validity of all materials or for the consequences of their use. Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, microlming, and recording, or by any information storage or retrieval system, without prior permission in writing from the publisher. All rights reserved. Authorization to photocopy items for internal or personal use, or the personal or internal use of specic clients, may be granted by CRC Press LLC, provided that $1.50 per page photocopied is paid directly to Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923 USA The fee code for users of the Transactional Reporting Service is ISBN 0-8493-1146-2/02/$0.00+$1.50. The fee is subject to change without notice. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged. The consent of CRC Press LLC does not extend to copying for general distribution, for promotion, for creating new works, or for resale. Specic permission must be obtained in writing from CRC Press LLC for such copying. Direct all inquiries to CRC Press LLC, 2000 N.W. Corporate Blvd., Boca Raton, Florida 33431. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identication and explanation, without intent to infringe.

Visit the CRC Press Web site at www.crcpress.com

2002 by CRC Press LLC No claim to original U.S. Government works International Standard Book Number 0-8493-1146-2 Library of Congress Card Number 2001043530 Printed in the United States of America 1 2 3 4 5 6 7 8 9 0 Printed on acid-free paper

Dedication

To my mother, father, wife and children for their patience with this project. Never has a book required so much.

Foreword

The Early Development

It is curious that the early interest in this darling of the physicists called a laser focused on concerns about safety. With memories of the hazards of x-ray development, it was not known what this powerful light could do. In fact, the early worries regarding laser applications in ophthalmology were that there would be x-ray changes in the eyes that could be dangerous. Focusing on the hazards to the eye, our interest was with the dangers to the workmen who were building lasers. With that idea, we established the rst comprehensive laser laboratory for medicine at the University of Cincinnati in 1961. The development of a safety program for the use of lasers in medicine was the rst goal. It was fortunate that the laser in the visible light range, the ruby, was the rst laser we used. The ruby laser furnished the operating surgeon with a special scalpel that was related to the color of the target area. The interest in these studies was tremendous, so funding was given by the John A. Hartford Foundation for the development of the extensive laser research laboratory at the Childrens Hospital of the Medical Center of the University of Cincinnati. This permitted a number of disciplines in medicine and biology to do basic work, rst with the ruby laser, followed by the CO2 laser and the argon laser, and, in 1962, the development of the rst Q-switched ruby laser. After several years, the Laser Research Laboratory at the Childrens Hospital was moved to the Department of Dermatology, College of Medicine, University of Cincinnati. This laboratory continued to be used for basic biomedical and then clinical applications in many areas of surgery. The Laser Research Laboratory also developed practical laser surgery teaching courses for hands-on learning with an emphasis on controls. These courses continue today with more or fewer controls, relatively few with hands-on learning some even of value for the expenses required. In 1993, Puliato introduced a teaching course with diagnostics and design engineering in addition to laser surgery. Throughout the years, the programs of research and development continued and new applications were discovered, which led to the development of laser biomedical institutes and the laser biomedical industry. The laser biomedical industry even today is called a protless prot organization. This is because laser biomedical instrumentation is considered a form of technology development rather than of market development. The hazards that the laser biomedical industry might be inuenced by vivid market developments have been shown recently in two areas: one, the development of laparoscopic cholecystectomy and, second, the renaissance of laser dentistry, which we rst established in the 1960s. For laparoscopic cholecystectomy, the marketing inuence initially recommended the use of expensive laser systems, discouraging the general surgeons, who had been the last group of conservative medical disciplines attracted to the laser. As a result, untrained surgeons ventured into this high surgical technology for laparoscopic cholecystectomy at the time, a new ambulatory technology that had been treated conservatively over a period of years. The application of this new surgical technology proceeded without the continued development of controls, basic training and safety courses, which, in turn, led to

disappointments, concerns, and reactions. However, the result was an over-enthusiasm for the short morbidity associated with gallbladder operations. The belated application of control electrosurgery led to some disenchantment with laparoscopic laser surgery among general surgeons and even the public. However, a small sampling of surgeons continued to use laparoscopic laser surgery and, eventually, its use expanded to the appendectomy, hiatus hernia and herniorrhaphy. Again, the need for controls and even the actual need for such surgery were the driving forces behind the developing technology. A similar vivid market episode led to the revival of laser dentistry, which had been neglected because of disinterest among dentists, the expense of instrumentation for general practice and the lack of continued research in that eld after the 1960s. The result of this market inuence was the increased sales of Nd:YAG lasers; expensive and not necessary for common soft tissue dental surgery. This led to the purchase of unwanted expensive instruments that the practicing dentist could not afford. The current revival of basic research and continued interest in CO2 and holmium lasers for operative dentistry are gradually rebuilding the condence of dental surgeons in particular, and the continued research on inexpensive laser diagnostics is inuencing even regular dental practitioners.

The Current Laser Theme

The rapid production of laser biomedical instrumentation, the recognition of the importance of laser research in the laboratory and beyond, as well as the inuence of laser biomedical engineers are all characteristics of the modern scene. Efforts to develop the new comprehensive laser engineering programs in a formal style have not been possible. So-called optical engineering is provided mostly in short courses in postgraduate engineering programs few real programs in optical engineering are available. The original proposal of laser engineering included, rst, two years of basic optical-, electrical- and mechanical-engineering courses, then cooperative laser engineering for the subsequent three years. Cooperative engineering, rst established by Herman Schneider, Dean of the College of Engineering at the University of Cincinnati, means one semester at the College of Engineering alternating with the next semester in industry. For laser engineering, this industry meant factory laser production, laser applications in the military or the laser biomedical industry. An alternate worker held the students job position while he or she returned to college. Cooperative engineering also permitted more students to attend engineering college through their jobs in industry. Briey, the contact with the actual laser industry during the engineering training period made for a practical insight into what needed to appear on the oor of the factory. The problem here, of course, is nding faculty for the third, fourth and fth years at the College of Engineering. It is the obligation of the laser industry, the military and the biomedical groups to help supply such faculties at both undergraduate and graduate levels. Leroy Hood, M.D., Gates Professor and Chairman of the Department of Biotechnology at the University of Washington School of Medicine, believes that, The future problems of biology will primarily have to do with the analyses of complex systems and networks and, accordingly, we must develop tools to handle these complex systems. The development of these tools will require the joining of applied science, engineering and other disciplines to biology to develop the complex tools of the future. We see a special role in developing new computational tools in both hardware and software areas. My feeling is that lasers would play a very important role in the development of new biological and medical technologies. There is a need for the use of the consortium in laser biomedical technology; the need for cooperative groups to work together to solve difcult problems is important. As indicated, this has not been possible

in the United States, whereas it is possible, with government help, in Japan. The only consortium presently available in the United States now seems to be the automobile consortium for the electric automobile. The secrecy and competition of the laser biomedical industry has made it difcult to develop a consortium.

The Future
Our usual biomedical lasers are slowly becoming smaller, more exible, and somewhat less expensive. Technology is producing new lasers with more Q-switched systems, which brings about special safety problems such as more viable plume fragments. In general, the new lasers are being developed for specic applications. Our interest in the future of laser biomedical instrumentation is to consider development of multiwave systems. This is to attempt to have more applications in single or compact unit systems. We survey the initial development of the following multi-wave systems: R. Rox Anderson suggests the economical junction diode systems alone or pumping solid state lasers. Multiple heavy metal vapor systems, which we have just started. Multi-wavelength systems of the parametric oscillator types such as the MOPO of Spectra Physics from the third harmonic of Nd:YAG with 200 nm up to 4500 nm; Continuum with the second harmonic and a smaller number of wavelengths. Coherent Laser Co. is also considering a parametric oscillator system; the type is not known. Our program at the Naval Medical Center is to develop a seven-wavelength system for our needs, such as for blood vessel disorders, for pigmentations, for tattoo removal, for cancer, and for localized refractory dermatologic conditions. For that, we are working through J.W. Steger, CAPT MC USN, with the copper vapor laser pumping a ruby laser without Q-switching, 513, 578, 513 + 578, and 694 in the nanosecond range. This combination gives us four wavelengths. Then, with another heavy metal vapor laser, barium, we are able to have 1500, 1300, and a third wavelength of a mixture of the 1500 and 1300. Also, the copper vapor laser can pump gold and pump lead, giving us additional wavelengths. A third area of multiple laser systems is the MOPO 700 series, as indicated, of Spectra Physics, which can provide wavelengths from the 200 nm to the 4500 nm range with millijoule output. With the biomedical engineering developments for instrumentation of this latter group, we see the future of laser biomedical instrumentation. Research and development is just starting. The complex problem of laser safety for such a system is now with development by Rockwell. It is difcult in this current era, at least in the United States, with concerns for the effects of new programs on health costs, for continued basic and applied laser research.

Conclusions
There is proof that laser surgery reduces the cost of in-patient surgery, a signicant advantage in modern medical care. One can also establish that decreased morbidity is available through diagnostics and treatments by laser systems. So, will there be approval to use these lasers for more laser medicine and surgery in this current cost conscious era? Can we now afford all this laser research and development? A consortium in laser biomedical manufacturing, basic controlled research, and a comprehensive biomedical engineering program for instrumentation for what the patient needs can we pay for all of

this? Hopefully, the laser market of the future will continue to be developed and to enable the American market to compete with the Japanese and European markets. Another economic facet is the increased, often practical value in the training of physician assistants. For example, this is done in the naval medical program for training of dermatology corpsmen as laser technicians. This makes available a very effective dermatologists assistant. This training is offered mostly in California at present. Current biomedical lasers are becoming more exible, smaller and perhaps not as expensive. There are now more laser ofce and ambulatory surgical practices than ever before. Diagnostic laser facilities will be more available in the coming years. As technology expands, the biomedical laser of the future will include more multi-wave systems with diodes, heavy metal vapors and optical parametric oscillator types. Adequate funding, because of the need for such lasers in laser biomedical engineering, will be necessary, as well as the development of complex laser safety programs. If the laser can continue to prove that it will reduce the cost of medical care and be used by trained critical practitioners, in a country where laser medicine and surgery began, laser medicine and surgery can continue as major parts of the practice of both medicine and surgery. Leon Goldman, M.D.

Preface

The purpose of assembling this book, with chapters from a group of experts whose backgrounds come mainly from the experimental side of the physical sciences, was to try to set down the the basics of laser interaction with tissue and describe how these basics have been applied in some of the medical specialties. The reasoning behind this is that 1) there are many areas in medical science where lasers or modern optics might have application if only some physical scientists knew about them and could transfer their knowledge to the medical area, and 2) clinicians need to know some basic principles from physical science to understand the opportunities and limitations of lasers and optical science as they try to apply these devices in medicine. Lasers have made many advances in medicine, especially in ophthalmology, dermatology and cardiology. A wave of enthusiasm follows the use of lasers that causes patients to request its use, even when it may not be warranted. Hopefully, this book will be able to convey to physicians enough basic information to better assess a lasers usefulness for a specic purpose, so they will not purchase or try to use a laser when it is not the best solution. The converse of enabling physicians to understand the limitations of a laser is to encourage them to try using a laser when it may do a better job than the conventional practice. In this regard, I am pleased and fortunate to have a foreword by the late Leon Goldman, known widely as the father of laser medicine. Leon was a pioneering soul whose enthusiasm for laser use was instrumental in much of the early progress in laser medicine. He was largely responsible for the founding of the American Society for Laser Surgery and Medicine, which provides a forum for training and education and discussions among laser medical professionals. I feel fortunate to have known Leon, and treasure his foreword to this book. I am grateful to all the contributing authors for their hard work with their chapters and for their encouragement. I am also grateful to George Pettit for his assistance in the early stages of the book, and to Marcia Patchan for her help in keeping things organized at the nish. Ronald W. Waynant Clarksville, MD Editor-in-Chief

Editor

Ronald W. Waynant received his B.E.S. in electrical engineering from the Johns Hopkins University in 1962 and M.S.E.E. and Ph.D. from Catholic University in Washington, D.C. in 1966 and 1971, respectively. From 1962 to 1966, Dr. Waynant worked in development and applications of solid state lasers at Westinghouse in Baltimore and continued part time in electrooptics and low light level television until 1969, when he joined the Naval Research Laboratory, where he carried out gas laser research that led to the rst vacuum ultraviolet laser the hydrogen laser. This work in the vacuum ultraviolet provided the spark that began serious x-ray laser research. Dr. Waynant also worked on the kinetics of rare gas halide excimer lasers, waveguide excimer lasers, microwave excited excimer lasers, and in excimer laser lithography. In the course of this work, he has published nearly 80 scientic papers and has given more than 90 contributed and invited talks on his work. In 1986, Dr. Waynant joined the Food and Drug Administration to assist in the development of a laser surgery research program that includes studies of basic laser interaction with tissue, ber optic delivery of laser energy to target tissue, and concern for long-term effects that might follow laser surgery. Dr. Waynant is a member of the American Physical Society and the Society of Photooptical Instrumentation Engineers. He is a Fellow of the Institute of Electrical and Electronics Engineers, the Optical Society of America and the American Institute of Medical and Biological Engineers.

Contributors

Thomas P. Coohill
Siena College Loudonville, New York

George H. Pettit
Summit Autonomous Orlando, Florida

Brian C. Wilson
Ontario Cancer Institute Princess Margaret Hospital and University of Toronto Toronto, Ontario Canada

Craig Gardner
Rio Grande Medical Technologies, Inc. Albuquerque, New Mexico

Qiushi S. Ren
Bascom Palmer Eye Institute Biophysics Lab Miami, Florida

Jordan D. Haller
School of Public Health Columbia University New York, New York

Jonah G. Sinowitz
Department of History Stanford University Stanford, Connecticut

Mark H. Wholey
Department of Radiology Shadyside Hospital Pittsburgh and University of Pittsburgh School of Medicine Pittsburgh, Pennsylvania

Tiina Karu
Laser Technology Research Center Russian Academy of Science Moscow Region, Russian Federation

Sune Svanberg
Department of Physics Lund Institute of Technology Lund, Sweden

Douglas R. Wyman
Department of Medical Physics Hamilton Regional Cancer Center and McMaster University Hamilton, Ontario Canada

Stuart L. Marcus
Dusa Pharmaceuticals, Inc. Valhalla, New York

Keith P. Thompson
Department of Ophthalmology Emory University School of Medicine Atlanta, Georgia

Glenn N. Merberg
Integration Services Group, Inc. Bethesda, MD

Ronald W. Waynant
Food and Drug Administration Center for Devices and Radiological Health Rockville, Maryland

Osvaldo Yano
College of Physicians and Surgeons Columbia University New York, New York

Mehmet Oz
College of Physicians and Surgeons Columbia University New York, New York

Jean-Marie Parel
Bascom Palmer Eye Institute Biophysics Lab Miami, Florida

A.J. Welch
Department of Biomedical Engineering University of Texas Austin, Texas

Table of Contents

Basics of Lasers Ronald W. Waynant and Glenn N. Merberg

1.1 Laser Principles ........................................................................................................................... 1 1.2 Laser Materials ............................................................................................................................ 1 1.3 Pump Sources.............................................................................................................................. 3 1.4 Resonators ................................................................................................................................... 6 1.5 Major Types of Lasers ................................................................................................................. 6 1.6 Medical Lasers ........................................................................................................................... 14 1.7 Measuring Laser Power............................................................................................................. 15 1.8 Focusing Laser Energy .............................................................................................................. 15 1.9 Basics of Fiber Optics................................................................................................................ 16 1.10 Optical Materials....................................................................................................................... 20 1.11 The Future of Medical Lasers and Fiber Optics ...................................................................... 24 References ............................................................................................................................................ 25

Optical and Thermal Response of Tissue to Laser Radiation A.J. Welch and Craig Gardner
2.1 Introduction .............................................................................................................................. 27 2.2 The Optical Response Of Tissue .............................................................................................. 28 2.3 Thermal Response..................................................................................................................... 41 2.4 Summary ................................................................................................................................... 45 References ............................................................................................................................................ 45

Dosimetry and Thermal Monitoring Brian C. Wilson and Douglas R. Wyman

3.1 Introduction .............................................................................................................................. 47 3.2 Optical Dosimetry..................................................................................................................... 47 3.3 Thermal Dosimetry................................................................................................................... 68 3.4 Radiologic Imaging Methods ................................................................................................... 79 3.5 Summary ................................................................................................................................... 79 References ............................................................................................................................................ 80

Uses and Effects of Ultraviolet Radiation on Cells and Tissues Thomas P. Coohill
4.1 Introduction to Ultraviolet Radiation ..................................................................................... 86

4.2. A Division of the Ultraviolet for Photobiological Studies ...................................................... 86 4.3 UV Sources ................................................................................................................................ 88 4.4 Absorption of Ultraviolet ......................................................................................................... 90 4.5 Direct vs. Indirect Effects of UV............................................................................................... 96 4.6 Action Spectroscopy: Effect as a Function of . ...................................................................... 96 4.7 Effects of UV on Cells ............................................................................................................. 100 4.8 Complex Responses of Tissues............................................................................................... 101 4.9 Repair....................................................................................................................................... 103 4.10 General Effects......................................................................................................................... 104 4.11 A Perfect Laser for UV Photobiological Studies.................................................................... 104 4.12 Useful Review Texts ................................................................................................................ 105 References .......................................................................................................................................... 105

The Physics of Ultraviolet Laser Ablation George H. Pettit

5.1 Introduction ............................................................................................................................ 109 5.2 Deposition of Ultraviolet Radiation in Organic Materials ................................................... 113 5.2 Target Decomposition ............................................................................................................ 120 5.3 The Ablation Plume ................................................................................................................ 122 5.4 Repetitive Irradiation.............................................................................................................. 128 References .......................................................................................................................................... 129

Tissue Diagnostics Using Lasers Sune Svanberg

6.1 Introduction ............................................................................................................................ 135 6.2 Light Interaction with Tissue ................................................................................................. 136 6.3 Spectroscopic Diagnostics of Malignant Tumors ................................................................. 141 6.4 Spectroscopic Diagnostics of Atherosclerotic Plaque ........................................................... 151 6.5 Light Scattering and Tissue Transillumination ..................................................................... 157 6.6 Outlook.................................................................................................................................... 163 Acknowledgments ............................................................................................................................. 164 References .......................................................................................................................................... 165

Low-Power Laser Effects Tiina Karu

7.1 7.2 Introduction ............................................................................................................................ 171 Primary and Secondary Mechanisms of the Action of Monochromatic Visible and Near Infrared Radiation on Cells.................................................................................... 172 7.3 Explanation of Controversies and Limitations of LowPower Laser Effects on Cellular Level ..................................................................................................................... 187 7.4 Clinical Applications of Low-Power Laser Effects ................................................................ 196 7.5 Summary ................................................................................................................................. 200 References .......................................................................................................................................... 201

Therapeutic and Diagnostic Application of Lasers in Ophthalmology Keith P. Thompson, Qiushi S. Ren, and Jean-Marie Parel,
8.1 Introduction ............................................................................................................................ 211 8.2 Basic Ocular Anatomy and Physiology.................................................................................. 212 8.3 Transmission and Absorptive Properties of Ocular Tissues................................................. 214 8.4 Photothermal Laser Applications........................................................................................... 217 8.5 Photodisruptive Laser Applications....................................................................................... 231 8.6 Photochemical Laser Applications: Photoablation and Photodynamic Therapy................ 233 8.7 Diagnostic Laser Applications ................................................................................................ 239 8.8 Summary ................................................................................................................................. 240 Acknowledgments ............................................................................................................................. 241 References .......................................................................................................................................... 241

Cardiovascular Applications of Lasers Jordan D. Haller

Introduction ............................................................................................................................ 247 History (with Jonah G. Sinowitz) ........................................................................................... 248 Anatomy and Pathology ......................................................................................................... 249 Physics...................................................................................................................................... 251 Angioplasty (with Mark H. Wholey) ...................................................................................... 262 Transmyocardial Laser Revascularization (TMLR) .............................................................. 268 Other Surgical Applications of Lasers in Cardiology (with Osvaldo J. Yano and Mehmet C. Oz)............................................................................. 271 9.8 Aids to Welding: Tissue Solders ............................................................................................. 272 References .......................................................................................................................................... 278 9.1 9.2 9.3 9.4 9.5 9.6 9.7

10

Lasers in Photodynamic Therapy Stuart L. Marcus

10.1 Introduction ............................................................................................................................ 287 10.2 Tissue Optical Properties, Photobleaching, and Light Delivery Systems............................. 289 10.3 PHOTOFRIN PDT for Superficial Bladder Cancer .............................................................. 291 10.4 PHOTOFRIN PDT in the Treatment of Endobronchial Lung Cancer................................ 296 10.5 PHOTOFRIN PDT in Gynecologic Malignancies ................................................................ 300 10.6. PHOTOFRIN PDT in Head and Neck Cancer...................................................................... 302 10.7 PHOTOFRIN PDT in Gastrointestinal Cancer..................................................................... 304 10.8 PHOTOFRIN PDT for Intraoperative Abdominal or Thoracic PDT.................................. 306 10.9 PHOTOFRIN for Intraoperative PDT in the Treatment of Intracranial Tumors .............. 307 10.10 PHOTOFRIN PDT in the Treatment of Ocular Cancer....................................................... 310 10.11 PHOTOFRIN PDT in the Treatment of Cutaneous and Subcutaneous Tumors............... 311 10.12 New Photosensitizers for PDT ............................................................................................... 313 10.13 Conclusions ............................................................................................................................. 317 References .......................................................................................................................................... 317 Index...................................................................................................................................................325

1
Basics of Lasers
1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 Laser Principles....................................................................... 1 Laser Materials........................................................................ 1 Pump Sources......................................................................... 3 Resonators............................................................................... 6 Major Types of Lasers ............................................................ 6 Medical Lasers ...................................................................... 14 Measuring Laser Power........................................................ 15 Focusing Laser Energy ......................................................... 15 Basics of Fiber Optics .......................................................... 16
General Considerations

Ronald W. Waynant
Food and Drug Administration Center for Devices and Radiological Health

Glenn N. Merberg
Integration Services Group, Inc.

1.10 Optical Materials .................................................................. 20 1.11 The Future of Medical Lasers and Fiber Optics ................ 24 References ........................................................................................ 25

1.1 Laser Principles

To understand the operation of the laser requires a knowledge of the energy levels associated with atoms, ions and molecules. In thermal equilibrium, the energy levels are populated according to the Boltzmann distribution, which forbids the conditions in which an upper level might have a greater population than a lower level. Because, for lasing, an upper level must be more highly populated than the lower level, lasing will not take place. Lasing can take place only when a material is not in thermal equilibrium. This non-equilibrium is created by an excitation source sometimes called a pump source. Just as thousands of atoms, ions or molecules can be laser materials, numerous pump sources can excite the materials. In many cases, the gain produced by the pumped laser material is low. To make a device, it is necessary to use an optical resonator to repeatedly reect the signal through the amplifying material to add to the intensity. Therefore, the basic elements of a laser are: (1) a laser material, (2) a pump source and (3) a resonant cavity, as shown in Figure 1.1. Figure 1.2 shows the concept of an excited material in which the higher energy level is more populated than the lower energy level. This is clearly a non-equilibrium system. Without the resonator, the excited system is capable of amplifying a signal only once, as shown in Figure 1.3. The addition of a resonator (two mirrors aligned to precisely reect the energy back and forth) allows a signal to be amplied to high intensity. Figure 1.4 shows the laser oscillator that results.

1.2 Laser Materials

Laser materials have been the focus of 30 years of research. (A list of generic laser materials is contained in Table 1.1.) The research began with solid state lasers such as ruby and neodymium doped glass lasers. In the early days of lasers, all the easy-to-grow solids were grown with a wide collection of dopants. Most of these devices were inefciently pumped by ashlamps and many of the materials

0-8493-1146-2/02/$0.00+$1.50 2002 by CRC Press LLC

Lasers in Medicine

mirror

feedback and oscillation

atmos (laser medium)

mirror

laser output beam

R=100%

R=80%

pumping process

FIGURE 1.1

Basic laser components.1

energy

upper level

laser action

population inversion

lower level

population
FIGURE 1.2 Population inversion.1

Inverted laser medium

Input light beam

amplified output light beam

FIGURE 1.3

Laser Amplier.1

Basics of Lasers

reflected light waves

Inverted laser medium

100% reflecting mirror

partially transmitting mirror

FIGURE 1.4

Laser oscillator.1 TABLE 1.1 Laser Materials2

Laser Material 1. Solid a. Doped Crystal Host b. Semiconductor Liquid a. Dyes in Solvent Gases a. Atomic b. Molecular c. Excimer - Diatomic - Triatomic - Ionic Comments Ruby, garnets, etc., usually optically pumped Electrically pumped, high efciency Tunable, optically pumped Rare gases, metal vapor Infrared, ultraviolet Ultraviolet Visible Ultraviolet, vacuum ultraviolet, x-ray

2. 3.

were easily damaged by the radiation they generated. These problems fueled the development of gas lasers, which were rf- or dc-discharge pumped in the beginning. Semiconductor lasers were discovered but made little impact because of their need for cryogenic cooling. The desire for more energy and more laser types led to owing gas lasers and to high energy electron beams for laser excitation. Electron beams worked just ne for laser research, but they were impractical for most applications. Low inductance discharge excitation systems were developed for excimer lasers. In the meantime, the development of the heterostructure semiconductor enabled room temperature operation of diode lasers, which have high energy conversion efciency. Combinations of diode arrays made suitable excitation sources for solid materials, which led to a renaissance of solid state laser research and a desire to retry many of the old materials with the new pump. The combination has led to smaller, more practical lasers capable of portable uses. From this more than 30 years of laser research has come a handful of lasers that have begun to be developed for medical applications. We will expand on these lasers below. Here, it is sufcient to say that much must be done to make these devices better and more appropriate for surgery and other medical applications.

1.3 Pump Sources

The rst laser was pumped by a ashlamp. Flashlamps of the sort appropriate for laser pumping were developed by Edgerton for ash photography. They consist of a quartz envelop, two tungsten electrodes, a low pressure gas usually xenon and an energy storage system suitable for single or repetitive pulsing of the lamp. A collection of some of the common ashlamps that have been used is shown in Figure 1.5. The lamp emits a blackbody energy distribution of temperature near 5500 K. The dopant in the laser material absorbs a small fraction of this energy, hence the poor efciency from ashlamp pumping.

Lasers in Medicine

FIGURE 1.5

Typical linear and helical ashlamps used for laser pumping.2

Beam

Output mirror

Total reflector

FIGURE 1.6

DC discharge system.

RF supply Metal-rf electrodes

Waveguide Ceramic insulator

FIGURE 1.7

RF discharge excitation.

Discharge pumping allows the energy source to drive the exciting electrons directly within the laser material. There is no conversion to optical energy and back to excited state as in ashlamp pumping, but discharge pumping has its problems too. Greater care must be taken of electron temperature to optimize excitation of the correct level. Gases can be added to the discharge to improve energy transfer to the excited state or to remove energy from the lower laser level, but the effects of these gases on the discharge conditions must be a concern. Figure 1.6 shows a typical longitudinal dc discharge, while

Basics of Lasers

MIRROR OPTICAL WINDOW CATHODE (SCREEN) ELECTRON BEAM -HV (ELECTRON BEAM POWER SUPPLY) + HV (DISCHARGE POWER SUPPLY)

ANODE (SOLID)

DISCHARGE REGION

FIGURE 1.8

E-beam system for laser excitation.

Figure 1.7 is an rf discharge. Both types of discharge have been used frequently for excitation of low pressure gas lasers. Electron beams are capable of exciting nearly any material. Solids and semiconductors have been studied by e-beam excitation as well as gases. High pressure gases can be excited by e-beams quite easily. E-beams are excellent for quick studies, but are impractical for lasers that are intended for applications. The maintenance required by electron beam machines is too costly for e-beam pumped lasers to be useful. A typical electron beam pumped laser system is shown schematically in Figure 1.8. Electric current from rather low voltage sources can be used to pump semiconductor lasers. Since this type of pumping is at least 30% efcient (compared with a few percent at best from most other sources), it is easy to see the widespread popularity of diode lasers and the interest in using them to pump other lasers. Chemical excitation is, in some ways, an attractive and efcient process. By inducing a chemical reaction that leaves a reactant in an excited state, it is possible to build a population inversion capable of laser action. Very little energy is needed from the outside to ignite the chemical reaction. Since the energy is limited only by the volume of the population, these devices are scalable to high power. (A list of numerous laser-producing reactions is given in Table 1.2.) However, many of the gases are toxic or highly reactive and handling a large volume of them is an unpleasant, unsafe and costly task. Nuclear pumping has some attraction because the emission from nuclear reactors can excite laser materials, but this method of pumping is impractical for lasers that must be used for applications remote from the reactors.
TABLE 1.2 Major Chemical Lasers2
Typical Reaction O2* + I O2 + I* Same as HF F + H2 HF* + H H + F2 HF* + F H + Cl2 HCl* + Cl F + D2 DF* + D D + F2 DF* + D H + Br2 HBr* + Br CS + O CO* + S DF* + CO2 CO2* + DF Wavelength (nm) l.3 1.31.4 2.63.5 3.54.1 3.54.1 4.04.7 4.95.8 1011

Laser I HF (overtone) HF HCl DF HBr CO CO2

Lasers in Medicine

1.4 Resonators
Two types of resonators can be important for laser applications: stable and unstable. Each type has some important principles. Stable resonators are the most common type. Examples of numerous stable resonators, which have been given considerable research, are shown in Figure 1.9. The most popular are characterized by a repeated and stable radiation pattern within the resonator. Stable resonators are characterized by somewhat wider beam divergence and somewhat poorer coupling to the excited gas volume inside the tube. These resonators are needed to achieve oscillation in low gain systems. Unstable resonators are practical only with high gain laser materials, but they produce just the opposite effects to stable resonators. Examples are given in Figure 1.10. These lasers do not direct a ray back and forth over the same path, but actually utilize a mild walk-off path that couples much of the excited volume into an output that can be made nearly parallel. This non-divergent beam can be focused into a smaller focal spot than the stable resonator output.

1.5 Major Types of Lasers

Far Infrared lasers 1. CO2 (10.6 m): The carbon dioxide laser sparked a recovery of lasers for surgery after early difculties with pulsed laser treatment of tumors. Figure 1.11 shows the energy levels of the CO2 laser. Carbon dioxide lasers are usually discharge excited. High CW powers are possible with either dc or rf excitation of mixtures of gases that include CO2, N2, He and H2 in approximately a 1:1:8:trace ratio. Examples of these excitation systems that could be used to excite CO2 are shown in Figures 1.6 and 1.7. Powers of 50100 W can easily be obtained from a modest 14 m tube. Mid Infrared 1. Er:YAG (2.94 m): This is a vibronic solid state laser with an emission line exactly matching the region of highest absorption of water. The crystal is not as durable as some of the other materials. It can be run at a 510 Hz rate with average power to 1015 W. The system can be q-switched with some difculty. Ordinary stochastic pulsing can be obtained for periods of a few hundred microseconds. The total energy per pulse can be as high as 0.5 to 1.0 joules per pulse, and it can operate at a rate of 16 pps. 2. Ho:YAG (2.1 m): Holmium is another vibronic laser with emission near 2 microns. It also can be used with the garnet hosts and, in some cases, thulium is used as a sensitizer (i.e., the thulium absorption bands extract energy from the pump, but transfer it to holmium to improve its efciency. Thulium lasers as well at 2.9 m. Holmium suffers from room temperature population of the lower laser level, which will make it difcult to operate without cryogenic cooling. 3. Nd:YAG (1.06 m): One of the early lasers and still a most important one, it encounters little attenuation and penetrates tissue rather well from 15 mm or so. Figure 1.12 gives the energy levels of the Nd:YAG laser. The laser can operate cw or pulsed with pulses from nanoseconds to picoseconds and pulse compression to femtoseconds is possible. The laser can be frequency doubled, tripled, quadrupled routinely and even higher harmonics can be generated. Numerous garnets and other hosts can be used with neodymium to make slightly different wavelengths. The device can be diode pumped or ashlamp pumped. A wide variety of pumping congurations have been used. Typical output powers for CW and pulsed operation are given in the Table 1.3. Near Infrared 1. Alexandrite (700826 nm): This material is Cr:BeAl2O4 and its energy levels are not very different from those of ruby, as shown in Figure 1.13. It was the rst solid state tunable laser and is one of

Basics of Lasers

R1 (Mirror radius) =

(a) Plane-parallel

R2 =

L R2 >
(b) Large-radius mirrors >> L

R1 = L

(c) Confocal

R2 = L

a/ 2

R1 = L / 2

(d) Spherical

R2 = L / 2

R1 > L

(e) Concave-convex

R2 = - (R1 - L )

R1 = L

(f) Hemispherical

R2 =

FIGURE 1.9 Examples of stable resonators.

negative branch confocal

positive branch confocal

FIGURE 1.10 Examples of unstable resonators.

Lasers in Medicine

CO2 C O C O O C O O C O N

N2 N

He

He

Symmetric stretch 3000

Bending

Asymmetric stretch
J(odd) 55 53

J(even)

Energy (cm-1)

2000
2

58 56

10

m .4

J 54 52 2 0

00 1 m 9.4

3 1

v=1

Next quantum state in helium is 67.7 times the v = 0 to v = 1 spacing in nitrogen

10 0 1000

02 0

4.23 m 01'0 00 0
v1 v2 v3
0

v=0

1'S

FIGURE 1.11 Energy levels of the CO2 molecule.3

20

Pump bands

18

16

14

F3/2

11502 cm-1 R2 11414 R1

Laser transition

Energy ( x 10 cm )

-1

12

F3/2

10
Laser transition

I 15/2

~ 6000 cm-1

8
4 4

I 13/2

~ 4000 cm-1

6
4

I 15/2
2526 2473
4

4
4

I 13/2

I 11/12

2
4

I 11/2

2146 2111 2029 2001 848 311 197 0 134

I 9/2

Ground level

I 9/2

FIGURE 1.12 Energy levels of the Nd: YAG laser.3

Basics of Lasers

TABLE 1.3

Neodymium Laser Performance2

Pump Flashlamp Flashlamp Flashlamp Diode Flashlamp Diode Diode Flashlamp Diode Arc lamp Diode Diode Arc lamp Flashlamp Wavelength (nm) 263266 266 355 523 523527 532 532 532 1047 1053 1061 1064 1064 1064 Power (W) 0.04 0.0012 0.00120 0.00002-0.01 13 0.0010.08 0.0000010.0004 0.0150 0.0350.25 20 0.1 0.0023 0.51800 1500

Type Pulsed quad Nd glass Pulsed quad Nd:YAG Pulsed trip Nd:YAG Pulsed doub Nd:YLF Pulsed doub Nd:YLF CW doub Nd:YAG Pulsed doub Nd:YAG Pulsed doub Nd:YAG CW Nd-YLF CW Nd-YLF CW ND, Cr-GSGG CW Nd-YAG CW Nd-YAG Pulsed Nd-YAG

25

T1 T2 T2

T1 T2 T2 T1 E

T2
4

20
Energy, 103 cm-1

Vibronic band Fixed wavelength

T1

15

T1 2 E

2 2

Tunable wavelength

1 3

E T1

A2

10
0.6943 m

0.6804

0.701 -0.818 m

T2

T2

1.61- 1.74 m
4

1.63- 2.11 m
A2
4

0
Cr (Al2 O3) Ruby
3+

A2
3+

A2

T1

Cr (BeAl2O4) Alexandrite

Ni (MgF2)

2+

Co (MgF2)

2+

FIGURE 1.13 Energy levels of several vibronic lasers.2

a number of vibronic solid state lasers listed in Table 1.4. While the laser suffers from a lower stimulated emission cross section than neodymium, the cross section improves as the crystal temperature rises. Visible 1. Ti:Al2O3 (6601180 nm): Titanium doped sapphire has a large cross section comparable to neodymium, but its excited state lifetime is only 3.2 s, too short to store and build up a substantial upper laser level population. Laser pumping is usually required. 2. Cr: Al2O3 (ruby694.3 nm): The rst laser produced, its high threshold makes it less desirable that many of the other lasers. Still, it nds a few applications where its wavelength and pulsed characteristics are desirable, such as in tattoo removal. 3. Semiconductor Diode Lasers (6351550 nm): Semiconductor lasers were rst produced in the early 1960s. Initially GaAs devices, they were quite weak in output and required cryogenic cooling.

10

Lasers in Medicine

TABLE 1.4
Type

Vibronic Lasers
Pump source Arc lamp Flashlamp KrF Excimer 1320 nm Nd:YAG Laser or lamp Laser or lamp Laser Laser Laser Usually laser Operation CW Pulsed Pulsed Pulsed Pulsed or cw Pulsed or cw Pulsed or cw Pulsed or cw CW Pulsed or cw Wavelength (nm) 730810 701858 309325 17502500 720840 760920 720842 11671345 18702160 6601180

Alexandrite Alexandrite Ce-YLF Co-MgF2 Cr-LiCaAlF6 Cr-LiSrAlF6 Emerald (Cr doped) Fosterite (Cr doped) Thulium-YAG Ti-Sapphire

10 m

250

p n 11 Active region Facet

FIGURE 1.14 Diode laser structures.

A schematic of the tiny structure of these diode lasers is shown in Figure 1.14. Improvements over the years have developed new heterostructure junctions that allow room temperature operation. Output power of arrays are now powerful enough to be used as pumps for other materials. These devices have always been efcient emitters, and thus make a convenient, low voltage pump especially useful with solid state laser materials. These devices are usable in their own right as easily modulated sources for ber optic communication sources. The new technology has allowed these devices to move to both shorter and longer wavelengths as well. The wavelengths available and the materials that emit them are given in Table 1.5. Table 1.5 shows the power available from diode lasers at the various wavelengths. Pumping of lasers can be accomplished by the arrays that can be constructed at 810 nm. This wavelength is absorbed strongly by some solid state materials. For communications purposes, a diode composed of In0.73Ga0.27As0.58P0.42 will produce a diode emission of 1.33 m, and a material composed of In0.58Ga0.42As0.9P0.1 will emit at 1.55m. These wavelengths are preferred for communications because of the improved transmission of ber optics. 4. Copper/Gold Vapor (511, 578/628 nm): Metal vapor lasers such as those operating on copper and gold can achieve high average power by running at high pulse repetition rates. Unable to run

Basics of Lasers

11

TABLE 1.5

Semiconductor Laser Performance

Material InGaAsP InGaAsP Ga0.5In0.5P GaAlAs GaAlAs GaAlAs GaAlAs GaAlAs GaAlAs GaAlAs GaAlAs GaAlAs GaAs InGaAs InGaAs Power 3 mW 3 mW 10 mW 8 mW 35 mW 100 mW single 10 W linear array 60W quasi-cw (pulsed)array 1500W quasi-cw stacked array 150 mW 100 mW Pulsed only Pulsed only Pulsed only 50 mW

Wavelength (nm) 635 660 670 750 780 810 810 810 810 830 850 880 or 895 905 910 980

continuously because of lower level bottlenecking, these lasers can operate at a pulse rate of several kilohertz and can reach an average power of more than 100W. Gold lasers are of considerable medical interest because the wavelength is appropriate for photodynamic therapy. Other metal vapor lasers are shown in Table 1.6. A second problem of metal vapor lasers is that no good source of blue radiation has been found to allow red, green, or blue color projection. Several candidates have been worked on, but are not able to compete with the strength of the lasers shown in Table 1.6. 5. Dye Lasers (broadly tunable): Dye lasers are perhaps the least understood, but some of the most widely applied lasers. Originally made from some of the wide variety of dyes used by the clothing industry, little is known about the energy structure of these large molecules. Their absorption and emission spectra usually show broad features. Typical generic energy levels are shown in Figure 1.15. The upper state lifetimes are short and require fast pumping pulses (nanoseconds. Pumping is usually accomplished by ashlamp or by laser excitation. The broad emission enables tunable laser emission to be generated, which allows spectroscopic studies. In addition, the wide bandwidth also allows ultrashort pulse generation. Dyes are mixed with a wide variety of solvents at concentration levels of millimoles or lower. For continuous or rapidly pulsed operation, and to prevent deterioration by the short wavelength pump, the dyes are rapidly circulated to lower temperature. A selection of dyes, solvents and pumping sources that allow the generation of laser wavelengths from the ultraviolet to the infrared and with power up to a few watts average has been produced. Several schemes have been invented to allow tunable operation of dye lasers. A rather simple method is shown in Figure 1.16. Figure 1.17 shows a more sophisticated ring dye laser. 6. Argon/Krypton (515, 488/647): These ion lasers typically emit tens of watts and have been built to emit hundreds of watts of cw laser power. They are gas lasers operating at about a torr of pressure of pure argon or krypton. However, their construction has been a monument to gas tube design. Figure 1.18 shows the energy levels involved in the argon ion laser. Since energy injected into the gas must rst ionize the gas and then excite and invert the upper ion level, a high current is necessary. The lasers are very inefcient and operate at high power only when pumped with high discharge current. The high current discharge places such thermal stress on the tube that only a few materials can survive even when a magnetic eld is used to keep the discharge away from the walls, and strong water cooling is used to remove heat. The high power continuous emission of these devices offers blue or green emission capable of driving some chemical reactions. When mode locked to generate a few tens of picosecond pulses,

12

Lasers in Medicine

TABLE 1.6
Element Copper Gold Barium Lead Manganese Calcium

Metal Vapor Lasers

Wavelength (nm) 511, 578 628 312 1130 1500 722.9 534 1290 852.4 866.2 Power (relative to Cu) 1 0.1 to 0.3 Low Low 0.3 to 0.5 0.2 to 0.3 0.2 to 0.3 Comments

Secondary laser line Ba liquid a problem Ba liquid a problem 1000-1100 C temperature Mn vapor a problem Mn vapor a problem

Triplet absorption (can soak up laser emission) Excited state (S1) Radiationless decay Excitation

Laser emission

Alternative path through triplet states (not a laser transition)

Ground state (S0)

Radiationless decay

Singlet states
FIGURE 1.15 Typical dye laser energy levels.2

Triplet states

the laser can be used as a synchronous pump for dye lasers. When operated at lower pressure, the generation of higher stages of ionization can produce shorter cw wavelengths some reaching the mid-ultraviolet nearly to the vacuum ultraviolet. Ultraviolet 1. Excimer Lasers: The name excimer comes from two words, excited and dimer. A dimer is a molecule made up of two similar atoms. Some dimers form only when one of the atoms is excited. An example is the rare gases that form an excited dimer or excimer such as Xe2, which lives as an excimer for less than 200 ns. While numerous rare gas excimers can be formed at high pressure

Basics of Lasers

13

Cavity focusing mirror Pump beam Prism (fixed) Focusing lens Dye cell or jet

Output mirror Rear cavity mirror (moved to tune wavelength)

Output beam

FIGURE 1.16 Tunable dye laser.2

Ion laser pump beam Pump mirror Dye jet Mirror Astigmatism compensator Mirror

Variable focus auxiliary beam waist Double galvoplate Output mirror Output beam

Mirror
Unidirectional device Thin etalon Scanning Birefringent etalon filter Collimated arm

FIGURE 1.17 Ring dye laser.

and pumped with electron beams, it has been found that rare gases form molecules with halogen atoms much more easily. These rare gas halogen excimers can be formed at only an atmosphere or two and can be pumped with low inductance discharge systems. Because the upper laser level has such a short lifetime, excimer laser pulses are usually 1020 ns in duration. Typical pulse energies range from 100 to 500mJ from modest excimer lasers. Repetition rates of 50 to 100 Hz or greater also are typical. As this presents a high power to bers used to transmit the radiation, efforts have been made to lengthen the pulse to 100200 ns. Although silica windows transmit to short wavelength, no silica ber has been satisfactory for delivery of ArF (193nm) radiation. In addition, the use of halogen gases poses a safety problem during the relling and gas changing cycles. Table 1.7 shows the gas combinations and the wavelengths they generate. Considerable interest has been generated by improvements in the F2 laser at 157 nm brought on by the use of He in the discharge. This system, technically not an excimer, is now capable of more than 100 mJ per pulse and can be used as a pump for other solids or gases. 2. Free Electron Lasers (Tunable): Free Electron Lasers (FELs) are discussed here only because of the interest in these devices for medical applications. The FEL operates on the principle of perturbing a beam made up of free electrons by use of a series of magnetic pole pieces arranged in a section of a storage ring. This magnetic pole assembly is called a wiggler. FELs are large devices that

14

Lasers in Medicine

4p 2D0

3/2 5/2

1/2

3/2 4p 2p0 5/2 7/2 3/2 1/2 4p 4D0

4965 150,000 4880 A

o

4658

4545 4765

5145 5287

4889 140,000
Energy (cm-1) 3/2 1/2 4s 2p

130,000
+

0 + 1.P

Ar

3p5 2p

15.75 V 127.109.9 cm-1

> 50 laser transitions in Arl, 0.7 30 m

3p6 'S

Ar

FIGURE 1.18 Energy levels of argon ion laser.

TABLE 1.7

Rare Gas Excimer Lasers (Bold indicates strongest)

F Cl 308 222 175 Br 282 206 I 253

Xe Kr Ar Ne

351, 353 249 193 108

currently exist as facilities at a few locations. As facilities they offer tunable, directional laser emission having many of the properties of the more conventional lasers. Currently operating systems emit mostly in the mid infrared from about 210 microns and have macropulse (12 microsecond pulses) energy of about 1200 mJ at a rate of 1030 Hz. Within the macropulse is a series of pulses 2 ps wide separated by 350 ps. Currently, these devices at Duke University, the University of California at Santa Barbara (200400 m wavelength), Stanford University and Vanderbilt University are being used to conduct exploratory research on numerous medical problems. A new storage ring FEL designed to work at shorter wavelengths (from the visible through the ultraviolet, vacuum ultraviolet and perhaps into the x-ray region) has been built at Duke University.

1.6 Medical Lasers

All of the above lasers can and have been used for medical purposes. If other wavelengths are needed, they can likely be found among lasers already discovered. In a sense, however, the problem of lasers in medicine lies not in discovering a magical wavelength, but in learning how to best deliver the right wavelength, the right pulse width, the right pulse shape, the right beam prole and the proper amount of energy to create the best medical outcome possible. The exact shape of the box(es) that contains the equipment will not be discussed in detail. That is a matter of choice for the medical device industry, which, nonetheless, must produce a safe, effective and competitive laser system.

Basics of Lasers

15

1.7 Measuring Laser Power

Measurement of laser power is, in many cases, simply a matter of placing a power meter in the beam path. Numerous companies make excellent meters capable of making accurate measurements of laser power and energy. Of somewhat greater concern, however, is the measurement of the laser pulse width and phenomena associated with time dependent change. This is no problem for pulses of nanosecond width or greater because excellent electronic systems exist that can make accurate measurements. Below a nanosecond, time dependent measurements become more difcult. If the pulse wavelength is in the visible, streak tubes and image intensiers are able to record the pulse. Electronic devices as well have response down to 50 ps or so. However, if the time variant pulse is in the infrared, the problem is somewhat different. Photon energies are not capable of being seen by traditional photocathodes. Nonlinear correlation can give an approximation of the pulse shape, but these techniques do not allow real time, pulse-to-pulse visualization.

1.8 Focusing Laser Energy

Focusing laser energy produces one of the most difcult to measure parameters focal spot size. This is difcult to measure because it usually results in an intensity that evokes a nonlinear or damaging response on any material that may be placed at the spot. Exact methods of measuring spot size have not been developed. There are some guidelines that can be used, however. First, a more or less uniform phase output beam of wavelength, , emitted through an aperture, d, will have an angular divergence, , such that

= -d

(1.1)

When a lens of focal length, f, is placed in the above uniform beam, then the beam can be focused to a spot size, do, where

do = ( F # ) ( )
and

(1.2)

f F # = -d

(1.3)

and is known as the f-number. One further approximation, as the depth of eld is also f-number dependent, an estimate of the volume in the focal region, Fv, is that

Fv = ( F # ) ( )

(1.4)

These approximations can be of help when focusing a beam onto a target such as tissue or a ber surface. Since beams may not be perfectly uniform or diffraction limited, and lenses may have aberrations as well, it is useful to devise methods to get better information on focal spot size for each specic application.

16

Lasers in Medicine

1.9 Basics of Fiber Optics

1.9.1 General Considerations
Fiber optics can be thought of as conduits for light. In fact, before the advent of modern ber optics, hollow tubes with reective inner surfaces were used to guide light energy. Todays ber optics are solid-core exible strands of highly transparent material. Light is guided through the bers by a process called total internal reection (TIR). High quality optical bers can carry light for miles with negligible attenuation. TIR is a consequence of the refraction of light by a refractive index gradient. Refraction can be dened by considering the interface between two optical materials with indices of refraction nl and n2. A ray of light crossing the interface from nl to n2 is deected by the index mismatch according to Snells law:

n1 sin 1= n2 sin 2

(1.5)

where 1 and 2 are the angles the ray of light makes with the normal on either side of the interface. In the case where n1 > n2 the light will be deected away from the normal upon crossing the interface. Figure 1.19 illustrates refraction for light rays with various angles of incidence on an interface with n1 > n2. Note that, for many incident rays, there are two resultant rays. These rays can be thought of as the reected and transmitted rays. A reected ray always leaves the interface at an angle, r , equal to the angle of incidence, i. Its intensity is a function of the index mismatch between the two materials and the angle of incidence. The transmitted ray has a path dened by Snells law, as described above. Some incident rays, which are said to be totally internally reected, produce only reected rays. Light rays with angles of incidence greater than the critical angle, cr , experience TIR. The critical angle, cr , is dened by

sin cr = n1/n 2

(1.6)

Fiber optics guide light by TIR. The ber typically consists of concentric layers, as shown in Figure 1.20. The core material is surrounded by a cladding material of slightly lower index of refraction. Light is contained in the ber by TIR at the core-cladding interface and is guided along the ber. Numerical Aperture To be guided by the ber optic, light must be launched so that it satises the criteria for TIR. Light entering the ber is refracted at the end face of the ber, and must strike the cladding at an angle greater

transmitted ray
2 1 r cr

n2 n1

interface

reflected ray

FIGURE 1.19 Refraction and total internal reection.

Basics of Lasers

17

Buffer coating

Core

Cladding

FIGURE 1.20 Geometry of a ber optic.

acceptance cone

ac

FIGURE 1.21 Numerical aperture and acceptance cone.

than the critical angle. Using simple geometry, we can dene a cone that includes all the rays that will be guided by the ber, as is shown in Figure 1.21. This cone is called the acceptance cone, and the sine of its half-angle, ac, is dened as the Numerical Aperture (NA) of the ber. The NA is related to the indices of refraction of the core and cladding of the ber by the equation

NA = sin ac =

n core n clad

(1.7)

Fresnel Reection Light crossing a refractive index gradient will be partially reected, as was discussed previously. Similarly, light launched into a ber will be partially reected at each of the end faces. The equation for reection of light striking an interface at normal incidence is

(n n ) + (k k ) R = ----------------------------------------------------2 1 2 2 1 2 (n + n ) + (n + n )

1 2

1 2

(1.8)

where R is the fraction of the light that is reected, n and k are the real and imaginary components of the indices of refraction, respectively, and the subscripts 1 and 2 denote the media on either side of the interface. The quantity k is often referred to as the extinction coefcient, and can be considered negligible over spectral regions where media 1 and 2 are highly transparent. A simple approximation of the total light transmitted through a ber after reections from both end-faces are considered is given by

T = 2n/n2 + 1

(1.9)

18

Lasers in Medicine

where T is the fraction transmitted and n is the index of refraction of the core of the ber. Conventional ber optics have indices of refraction of about 1.5. Therefore, about 92% of the light launched into the ber can be transmitted. Sometimes, anti-reective coatings are applied to the ends of bers, but these coatings are wavelength specic and need to be reapplied every time the ber is cleaved. Bending Fiber optics can be bent gently without inducing substantial optical attenuation. Only at severe bends does the optical transmission begin to fall. The severity of bending of a ber is described in terms of bend radius, which can be thought of as the radius of the circle the ber would form if it were bent into a closed loop. Figure 1.22 shows the effects of bending on the transmission of bers with various NAs.5 In addition to optical considerations, bers have mechanical minimum bend radii. Bending bers to smaller radii results in catastrophic failure. Conventional bers, which are made of synthetic silica, are extremely strong and, thus, very exible. The recommended minimum bend radius for silica bers is about 50100 times the ber diameter.5 For long-term bends, such as in a rigid device or for packaging, the suggested minimum bend radius is about twice this value. Often, the exibility of a ber will be specied in terms of strain-to-failure. Strain is a unit of deformation, and is dened as change in length per unit length, or

= l/l

(1.10)

where is strain, l is sample length and l is the elongation of the sample under some load. Strain-tofailure denes the strain point at which the sample will fracture. The quantity, f , is related to the minimum bend radius by the relationship

f = r/R
where r is the ber radius and R is the minimum bend radius.
100

(1.11)

90

Transmission (%)

80

70

60

TECS (NA = 0.39) PCS (NA = 0.40) GLPC (NA = 0.18)

3 4 Bend Radius (cm)

FIGURE 1.22 Effect of bending on ber transmission.

Basics of Lasers

19

Laser Damage High energy lasers are capable of causing damage in the ber optics used for their delivery. The sources of this damage are widespread.6 For cw, or long pulsewidth lasers (>50 s), the damage is most frequently associated with gradual heating of the ber. Often, this is a result of misalignment of the laser. Typically, the cladding material or epoxy in the connector heats and causes failure of the ber. For short pulse lasers, however, there is the added issue of high peak powers. Short pulses with high peak powers may cause photoacoustic effects that damage the ber. In addition, these lasers may cause a plasma to develop at the launch face of the ber. The plasma is witnessed as a blue spark at the end of the ber accompanied by an audible snapping sound. Most often, laser damage occurs on either of the end faces, or at the point of rst focus inside the ber. This is the point at which the light has its highest intensity within the ber. Laser damage thresholds are lower at the surfaces than in a materials bulk. The situation is compounded if the surfaces are not properly prepared. A well polished surface will have a surface roughness of one tenth the wavelength of the laser light. In addition, care should be taken not to embed polishing materials in the surface. Certain techniques for cleaving, or breaking the ends of bers produce excellent quality faces. Under these conditions, the surface may have a damage threshold that approaches that of the bulk. Alternatively, the ber can be polished. Several commercial ber polishers are available. Connectors Often, bers are terminated with connectors. Connectors are ttings epoxied onto the bers to allow them to be quickly connected to an optical system. These precision devices allow the ber to be removed and replaced repeatedly without changing the alignment of the optical system. Connectors can also be used to connect two lengths of bers together. Fiber Bundles There are really two types of ber bundle, coherent and incoherent. Coherent bundles have consistent relative positions of all of the constituent bers along their length, and are used in imaging endoscopes. The resolution of the image is determined by the number of bers in the array. Often, some of the bers are used for illumination and the remainder are used to carry an image back to a viewnder or camera. Incoherent bundles are far less expensive and are ideal for the delivery of light or laser energy. They are sometimes used because a bundle of small bers is much more exible than a large ber with the same total core diameter. Laser power delivery bundles are used in applications such as angioplasty. Contact Tips for Silica Fibers Fibers are often used in conjunction with various tips that may act to shape the laser beam or cut tissue.7 Contact tips concentrate the laser energy at the distal end of the ber and do not transmit a free beam. The tip is placed in direct contact with the tissue, and cuts it as the laser energy is absorbed. Very little laser energy falls on surrounding tissue. Sapphire tips can be attached to silica bers to provide a wide range of cutting tools. Spherical, tapered, absorbing and transmitting tips are all used for various cutting applications with different results. Some attachable tips are actually made of silica rather than sapphire. Some silica bers are available with molded tips. These bers have different shapes machined into their distal ends as an alternative to attachable contact tips. Launching Laser Energy into a Fiber Typically, the laser beam is focused with one or more lenses to a spot size smaller than the ber core diameter. If the spatial mode of the beam is poor, it may become necessary to pass the beam through a spatial lter, which consists of a series of lenses and apertures that expand the beam and remove its highest order modes. After passing through this type of device, the beam can usually be focused much more easily. Most experts agree that the nal focus of the beam should be in front of the ber. This avoids having the beam focus inside the ber. If the beam were to focus in the ber, sufciently high intensities could develop that would damage the ber. At the launch face of the ber, the beam waist should be about 7080% of the core diameter. In addition, it is recommended that the NA of the ber be underlled by 2030%.

20

Lasers in Medicine

1.10 Optical Materials

Various materials are used to transmit light over different spectral regions. Silica ber optics, which are excellent for the transmission of visible light, are highly absorbing at mid-infrared wavelengths. Similarly, materials exist that transmit infrared laser energy, but are opaque to visible light. The main factors that determine a materials transparency window are scattering and absorption. Absorption is the conversion of light energy into heat. Scattering is the redistribution of light energy over different paths. A materials intrinsic transparency is dened by the absorption and scattering losses caused by the molecular structure of the material. These losses are graphically represented with a V-curve. Figure 1.23 shows V-curves for oxide, chalcogenide and halide glasses. These curves were generated from data presented by Shibata, et al.8 The minima in the V-curves are 0.16 dB/km at 1.6 m, 0.01 dB/km at 4.54 m and 0.001 dB/km at 3.44 m for oxide, chalcogenide and halide glasses, respectively.9 Due to extrinsic factors such as bubbles, impurity absorption bands, and inclusions, the real losses in these materials may be substantially higher, as is shown in Figure 1.24. The intrinsic losses shown Figure 1.23 derive from three main mechanisms. The short wavelength limit to a materials transparency is dened by the band gap of the material. High energy photons are
101

Transmission Loss (dB/km)

100

multiphonon edge

Oxide
10-1 electronic transitions 10-2 Rayleigh scatter 1 2 3 Wavelength (m)

Chalcogenide

10

-3

Halide
4 5

FIGURE 1.23 Theoretical loss curves for oxide, uoride, and chalcogenide glasses.

10
Attenuation Coefficient (dB/m)

Sapphire Chalcogenide glass Hollow sapphire waveguide

1
Silica Silver halide Fluoride glass

.1

Laser (m) (cm-1) HF Er:YAG CO CO2 2.8 2.9 5.2 10.6 1000 3000 200 600

.01 1 2 3 4 5 6 7 8 9 10 11 12
Wavelength (m)

FIGURE 1.24 Attenuation coefcients of various infrared transmitting ber optics.

Basics of Lasers

21

sufciently energetic to excite electronic processes such as valence band to conduction band transitions. The UV-edge of the V-curve represents optical absorption due to electronic transitions. The line in Figure 1.23 labeled multiphonon edge describes attenuation due to atomic and molecular vibrations excited by the incident energy. The position of this line is dictated by the size of the atoms and strength of the bonding in the material. Generally, larger atoms and weaker bonds correspond to better IR transmission. Any light at longer wavelength than the multiphonon edge is coupled into exciting vibrational modes, and is not transmitted through the material. The line in Figure 1.23 labeled Rayleigh scatter describes attenuation due to microuctuations in the index of refraction of the material. The scattering has an A.-4 dependence that is associated with scattering centers much smaller than the wavelength of the light. Note that the intersection of the Rayleigh scatter and the multiphonon edge dene the minimum attenuation for the optical material. In addition to the intrinsic factors described by the v-curve, there are many extrinsic sources of optical attenuation. Impurities may absorb or scatter light. Absorption caused by impurities might be electronic transitions or vibrational in nature. Scattering from impurities usually arises from differences in the indices of refraction between the impurity and the optical material. Bubbles in optical materials are said to be optically thick because their indices are very different from those of the optical materials themselves. Optically thick scattering centers are a strong source of scattering losses. UV-Vis-NIR Fiber Optics Silica-core ber optics are used for delivery of medical laser energy in the spectral region from 200 to 2400 nm. These bers are well developed because they have applications in telecommunications. As a result, the optical and mechanical properties of silica ber optics approach theoretical limits. For medical applications, there are three varieties of silica-core bers. While the claddings on the three varieties differ, the cores are all SiO2, or silica. Silica-silica, or glass-clad ber optics, are those bers with a silica glass cladding around a silica glass core. The cladding glass is typically doped with uorine or boron to lower its refractive index relative to the core glass. Frequently, these bers will be referred to as all-glass bers. There are several advantages to using silica-silica bers rather than plastic or silicon-clad bers.5 All-glass bers have the lowest optical loss coefcients of any optical bers. In the UV and IR spectral regions, where ber losses may be appreciable, all-glass bers are required for adequate transmission. These bers also offer higher laser damage thresholds. If the laser beam should become temporarily misaligned, or if the core is overlled with laser energy, a glass cladding is less likely to damage than a polymer cladding because the glass can withstand temperatures up to 1800C, whereas the plastic claddings can be used up to only a few hundred degrees. This high temperature capability may also be important for hot-tip laser procedures. All-glass bers are also preferred for accurate beam delivery. The low NA of silica-silica bers (NA = 0.22) provides less beam divergence at the distal end of the ber than high NA plastic-clad bers (NA = 0.37). In addition, the NAs of all-glass bers are more stable with varying temperature than the NAs of plasticclad bers because of the low expansion of silica. For less demanding laser delivery applications, there are two less expensive types of silica-core ber optics. Polymer-clad silica (PCS) bers have a soft, low index silicone cladding around a silica-glass core. Typically a Tefzel or Nylon buffer coating is applied for extra protection. The cladding may be removed by mechanical means, and this becomes necessary for contact tip applications. Unfortunately, the cores of PCS bers are free to move relative to the cladding. This phenomenon is known as pistoning, and makes connection difcult. Hard uoropolymer-clad silica (TECS) bers have a very thin, hard plastic cladding. The claddings are typically about 15 m thick and are protected by a buffer coating. The hard uoropolymer actually bonds to the silica core, so no pistoning occurs. An important parameter to consider when selecting a silica ber for a specic application is the bers OH content, which will affect both the spectral transmission and the cost of the ber. Spectral transmission curves of high and low OH silica bers are provided in Figure 1.25.5 Generally, high OH bers are used for the transmission of excimer laser energy, while low OH bers are necessary for visible and NIR transmission.

22

Lasers in Medicine

250 200 dB/km 150 100 50 0 350 Transmission/Meter 97.9% 90.0% 99.8% 550 750 950 1150 1350 Wavelength, nm 1550 1750 1950 High OH Low OH

FIGURE 1.25 Effect of OH- content on silica ber transmission.

Infrared Transmitting Fiber Optics In the past, much of the research extolling the merits of infrared laser tissue ablation has been presented with the disclaimer, If only a suitable delivery system was available. In fact, there are optical bers available for the delivery of energy from many of the infrared lasers that are of interest to the medical community. While it is true that many of these bers require more careful handling than conventional silica ber optics, they do offer the ability to deliver infrared laser energy for a variety of minimally invasive procedures. Lasers such as the Er:YAG, HF, CO and CO2 operate in the infrared spectral region, where the absorption coefcient of tissue is many times higher than at visible wavelengths. The efciency of laser tissue ablation is strongly dependent on the absorption coefcient of the target tissue. In addition, strongly absorbed laser energy creates less thermal damage in surrounding tissue. Therefore, infrared laser surgery represents a relatively precise method of tissue removal. The development of infrared beroptic catheter technology may result in a multitude of new, noninvasive therapeutic procedures. Figure 1.24 shows the attenuation spectra of some of the bers currently available for delivery of infrared laser energy. Also shown in this gure are the emission wavelengths of several clinically useful lasers and the approximate absorption coefcients of tissue at these wavelengths.10 The attenuation spectra shown account only for the propagation loss in the bers, reported in units of attenuation per length. Fresnel losses caused by reections from the launch and delivery faces of the ber are neglected. Generally, an attenuation coefcient of 1 dB/m or less is considered acceptable for delivery of energy over a distance of several meters. The following sections describe each of the types of ber depicted in Figure 1.24. Table 1.7 summarizes the properties of the bers. Fluoride Glass Fiber Optics Heavy-metal uoride glass (HMFG) optical bers are similar to conventional silica ber optics in many ways. Like conventional bers, they possess a glass-core/glass-cladding structure, have very low optical losses and are quite exible. In addition, uoride glasses offer transparency through the visible portion of the spectrum out to about 4.5 m wavelength.11 These bers have a demonstrated ability to deliver high energy pulses from the Er:YAG laser. Fibers with core diameters of 250 m have been used to deliver 450 mJ pulses (900 J/cm2) of laser energy at 2.94 m wavelength for short periods of time.1214 Unfortunately, the extended infrared transparency offered by uoride glasses is derived from the materials relatively weak molecular bonding.15 Thus, the same glass structure that is blessed with a wide transparency window is, at the same time, cursed with a low maximum use temperature (150 C), limited mechanical strength and some degree of chemical reactivity. The past 10 years have seen an enormous push to further develop uoride glasses. Much of this research was fueled by the extremely low intrinsic optical attenuation coefcients of the material, and directed at fullling the objective of a repeaterless transoceanic beroptic telecommunications link. While this goal has yet to be realized, many advances have been made in the eld. Fluoride bers with minimum

Basics of Lasers

23

attenuation coefcients of several dB/km are readily available. For medical applications involving only a few meters of material, the bers are essentially lossless. One of the major frustrations in working with uoride glasses has been the chemical reactivity of the bers. The industry standard uorozirconate, or ZBLAN, bers are somewhat soluble in water and so degrade rapidly in an aqueous environment. In addition, laser damage to the ber end-faces during the delivery of high uences of Er:YAG laser energy has been associated with a hydrolysis reaction. Compositions of uoride glass with improved resistance to attack by water are traditionally more difcult to manufacture in ber form.16,17 Recently, however, low-loss bers with enhanced chemical durability have been produced.18 Fibers fabricated from these new compositions have substantially higher laser damage thresholds for the Er:YAG laser14 and are available from at least one commercial source. Single-Crystal Sapphire Fiber Optics Single crystals of Al203, or sapphire, which are grown into optical bers, offer many advantages for the delivery of Er:YAG laser energy. Sapphire is extremely hard, has a melting temperature greater than 2000C, is biologically inert and does not degrade during laser delivery. The bers can be used in direct contact with tissue. Unlike silica and uoride glasses, however, sapphire bers are grown without an optical cladding. Core-only bers are susceptible to excessive loss of laser energy to the ber environs, unless they are properly jacketed. Teon and other polymers have been applied to sapphire bers for this purpose,19 and further research on optical cladding materials is ongoing. Sapphire can be grown into single-crystal bers by one of two methods. The edge-dened, lm-fed growth process (EFG) has been used to grow sapphire bers for structural reinforcement applications for nearly 20 years.20 Unfortunately, structural sapphire ber typically has high optical losses and is unsuitable for laser power delivery. Efforts to produce optical quality EFG sapphire ber are currently under way, and attenuation coefcients of less than 2 dB/m have recently been measured with the Er:YAG laser [J. Fitzgibbon, Saphikon, Inc., Milford, NH. Private communication 1992]. EFG sapphire ber is offered commercially, but optical quality ber is not yet available. Laser-heated pedestal growth (LHPG) of single-crystal sapphire has been shown to be a viable process for producing optical quality bers.19,21 Fibers with optical attenuation coefcients of less than 1 dB/m at the Er:YAG laser wavelength are readily produced. LHPG sapphire bers with core diameters of 340 m have been used to deliver 600 mJ pulses of energy from the Er:YAG laser.19 Unfortunately, there are, as yet, no commercial sources for LHPG sapphire. Chalcogenide Glass Fiber Optics Chalcogenide glasses contain the group VI elements S, Se or Te as the network-forming anion.22 These glasses are generally opaque to visible radiation, but transmit well in the near- and mid-infrared spectral regions. Optical bers fabricated from chalcogenide glass compositions have been used to deliver several hundred watts of CO laser radiation at wavelengths between 5.2 and 5.9 m.23 The bers can also be used to transmit radiation from the CO2 laser, but laser damage from a phenomenon known as thermal lensing limits the deliverable power to about 10 W. Traditionally, chalcogenide glass bers have been extremely brittle and difcult to work with. However, recent advances in cladding and coating technology have revolutionized this eld.24 New chalcogenide bers are available that offer good exibility and strength. Like all chalcogenide glasses, the new bers have a low use temperature and should not be exposed to temperatures much above 100C. Chalcogenide glasses have good chemical durability at room temperature and are not subject to attack by water. Chalcogenide glass bers are available commercially from a variety of sources. Unfortunately, the properties of bers from different vendors vary substantially. A range of chalcogenide ber properties is included in Table 1.1. Polycrystalline Fiber Optics Polycrystalline (PC) ber optics have been fabricated from the halides of silver, thallium and potassium. The materials are extruded through polished dies at elevated temperature and pressure. Silver halides

24

Lasers in Medicine

(AgCl-AgBr) are currently the best PC bers for the delivery of CO and CO2 laser energy. Clad and unclad silver halide bers are readily produced with attenuation coefcients of 3 and <1 dB/m at 10.6 m wavelength, respectively.25,26 Clad bers deliver energy with better beam quality and are less sensitive to hostile delivery environments. Unfortunately, irregularities at the core-cladding interface cause optical scattering and increase the total attenuation in these bers. Severe aging problems have also been associated with polycrystalline bers. The optical properties of the bers degrade over time. Packaging of silver halide bers is critical for several reasons. The bers are corrosive to many materials, and are quite sensitive to ultraviolet radiation. Exposure to uorescent room light causes a dramatic increase in the attenuation coefcients of the bers. In addition, the bers will permanently deform if bent beyond a certain point. Fortunately, delivery catheters have been developed to protect the bers from such damage. Most often, the bers are found in hand pieces at the end of articulated arms used with surgical CO2 lasers. Laser damage constraints generally limit these bers to delivering about 10 W of CO2 laser energy.26 Hollow Waveguides Hollow waveguides fall into two general categories, leaky and guided-mode propagating waveguides.27 Both types transmit laser energy through the hollow bore of a tubular ber. Leaky waveguides are made from metals or dielectric coated metals, and transmit energy along a highly reective inner surface. In some cases, both metallic and dielectric coatings are applied to a substrate to create a leaky guide.28 Guided-mode hollow waveguides operate on the principle of total internal reection, much like conventional ber optics. Sapphire,29 polycrystalline alumina,30,31 and several oxide glasses32 have been used as guided-mode waveguides. Both guided-mode and leaky waveguides are effective for delivery of CO2 laser power. Because the laser power propagates in air rather than in a solid, laser damage thresholds are high, and Fresnel reection losses are nonexistent. Guided-mode waveguides can be thought of as optical bers with air as the core. At certain wavelengths, the refractive index of specic materials may dip below unity because of a phenomenon known as anomalous dispersion. At these wavelengths, tubes of the material can act as waveguides with an air core (nair = 1). Sapphire tubes have been used as waveguides for CO2 lasers with attenuation coefcients of less than 0.5 dB/m.29 Interestingly, sapphire hollow waveguides are single-mode waveguides, which allows their output energy to be refocused easily. Doped silica tubes have also been used as guided-mode hollow waveguides, but their attenuation coefcients have been high.33,34 Dielectric coated metal leaky guides have demonstrated the ability to deliver CO2 laser radiation with attenuation coefcients of less than 0.1 dB/m.35 These waveguides have been fabricated with both circular and rectangular cross sections. A water-cooled version has been used to deliver almost 3000 W of power from a CO2 laser.36 Multilayer coated plastic guides are susceptible to laser damage at powers greater than 50 W. Hollow waveguides generally have very limited exibility. Most often, they are used in hand pieces and connected to articulated arms from surgical CO2 lasers. One commercial hollow waveguide boasts a 50 cm minimum bend radius. Some xed-curve sections of hollow waveguides are available for specic applications.

1.11 The Future of Medical Lasers and Fiber Optics

As accurate as the current coverage is in describing the current state of products commonly available for medical applications, a new wave of laser and ber optic devices is beginning to make its way from the research labs. This new technology will soon begin to replace everything currently described. In a few years, medicine and biology will be ooded with new technology, briey outlined in the next several pages. In a nutshell, here is what will happen. Solid state devices will replace all gas lasers wherever possible. Semiconductor diodes will either be used directly for applications within their power range or will be used as pumping sources for solid state laser materials. Solid state diodes are the material of choice because of their efcient conversion from electricity to light. New medical devices will consist of the following:

Basics of Lasers

25

Femtosecond lasers Ti:sapphire as well as other materials will enable exceedingly high power lasers, low energy pulsed systems that will allow surgeons to perform precise surgery without heating tissue as much. The high powers will allow table top accelerators that will generate x-rays, protons, and new nuclear isotopes, and enable technologies for medical diagnostics. New ber lasers Fiber lasers can also take advantage of solid state diode pumping. Fiber lasers, or lasers in which the ber core becomes the lasing material and the accompanying ber-cladding becomes the element that contains losses and guides the laser build-up without the need for a conventional resonant cavity, makes simple devices possible. Smaller, less expensive lasers and higher power from a reduced footprint system will make it easier for the laser to get into the operating room. All solid state, complete coverage of the visible and near infrared region Solid state pumped solid or ber lasers will have greater power and will cover all wavelengths. The high peak power from femtosecond laser will allow more use of non-linear optics and the generation of nearly every wavelength needed in the visible and infrared for those surgical and photochemical (photodynamic and therapeutic) applications. Mid-infrared semiconductor lasers These devices are also emerging from the research lab and will, if cooling problems can be solved, make excellent devices for gas and thin tissue sample analysis. The mid-infrared has the capability of determining exact chemical specicity and could possibly enable optical disease characterization. In other words, the next 10 years will see nearly all solid state lasers giving higher peak powers; higher average powers; more efcient operation covering the visible, the near IR, the ultraviolet and large parts of the vacuum ultraviolet, x-ray and the mid- and far-infrared as well. Most of the gas lasers will be replaced, as will dye lasers, chemical lasers and other less convenient lasers. Fiber optics could also take a great leap forward, if claims have merit. Photonic crystal type hollow waveguides are being touted in the popular press as having made a leap to low transmission loss approaching 0.01 db loss per kilometer. This would perhaps greatly improve both communications and medical applications, but one must be cautious until the results appear in professional journals and have been proven true. If these predictions hold true, the medical eld will likely soon have vastly new types of sources and delivery systems.

References
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. Siegman, A.E., Lasers, University Science Books, Mill Valley, CA, 1986. Hecht, J., The Laser Guidebook, McGraw-Hill, New York, 1992. Verdeyen, J.T., Laser Electronics, Prentice Hall, 1981. Special Section of the Proc. IEEE on Lasers in Med., T.F. Deutsch and R.W. Waynant, Eds., vol. 80, pp. 831930, 1992. McCann, B.P. and Powell, R.C., Optical Fibers in Medicine, Fiber Optic News, No. 11, (1991. De Hart, T.G., Where and why optical bers fail and how to prevent it, Photonics Spectra, November 1992, pp. 107-110. Advances in Nd:YAG Laser Surgery, S. Joffe and Y. Oguro, Eds., Springer-Verlag, New York, 1988. Shibata, S. et al., Prediction of loss minima in infrared optical bres, Electron. Lett 17, pp. 775777, 1981. Tran, D.C., Sigel, G. H., Jr. and Bendow, B., Heavy metal uoride glasses and bers: a review, J. Lightwave Technol., LT-2 (5) pp. 566586, 1984. Esterowitz, L. et al., Angioplasty with a laser and ber optics at 2.94m, Proc. S.P.I.E. 605, pp 3236, 1986. Drexhage, M.G., Infrared glass bers, Proc. S.P.I.E. 1228, pp. 211, 1990. Whitehurst, C. et al., Transmission of 2.94 m radiation by zirconium uoride optical bers, Proc. S.P.I.E. 1048, pp. 141144, 1989.

26

Lasers in Medicine

13. Kwark, B. et al., An analysis of Er:YAG laser for angioplasty: delivery system, ablation capability, and the effect of water content, Lasers in the Life Sciences 5, pp. 113128, 1992. 14. Merberg, G.N. et al., Evaluation of crystalline and chemically durable glass bers for Erbium-YAG laser delivery systems, Proc. S.P.I.E. 1228, pp. 216223, 1990. 15. Drexhage, M.G. and Moynihan, C.T., Infrared optical bers, Scientic American 259, pp. 110115, 1988. 16. Kanamori, T. et al., BaF2-CaF2-YF3-AlF3 based glass systems for infrared transmission, Japan. J.A.P. 20, pp. 326328, 1981. 17. Shahriari, M.R. et al., Fabrication of AlF3-based glass bers, Proc. Opt. Fiber Mat. and Processing Symp. 172, pp. 163168, 1990. 18. Iqbal, T. et al., AlF3-based glass bers with enhanced optical transmission, Electronics Lett. 27, pp. 2-3, 1991. 19. Merberg, G.N. and Harrington, J.A., Optical and mechanical properties of single-crystal sapphire optical bers, Appl. Opt. 32, pp. 32013209, 1993. 20. LaBelle, H.E., Jr., EFG, the invention and application to sapphire growth, J. Crystal Growth 50, pp. 817, 1980. 21. Jundt, D.H., Fejer, M.M. and Byer, R. L, Characterization of single-crystal sapphire bers for optical power delivery systems, Appl Phys. Lett. 55, pp. 21702172, 1989. 22. Nishii, J. et al., Recent advances and trends in chalcogenide glass ber technology: a review, J. NonCrystalline Solids 140 199208,1992. 23. Sato, S. et al., Multihundred watt CO laser power delivery through chalcogenide glass bers, Appl. Phys. Lett. 62 669-671, 1993. 24. Hilton, A.R., Sr., Chalcogenide glass optical bers, Proc. Soc.Photo-Optical Instr. Engrg. 1591, 3242, 1992. 25. Artjushenko, V. et al., Infrared cables and catheters for medical applications, Proc. Soc. Photo-optical Instr. Engrg. 1420, 157168, 1991. 26. Paisss, I., Moser, F. and Katzir, A., Core-clad silver halide bers for CO2 laser power delivery, Proc. Soc. Photo-Optical Instr. Engrg., 1420, 141148, 1991. 27. Miyagi, M., Consideration on realization of low-loss total reection-type hollow-core ber at midinfrared, Proc. Soc. Photo-Optical Instr. Engrg. 843, 7679, 1987. 28. Dror, J., et al., Hollow plastic waveguides internally coated with metal and dielectric lms, Proc. Soc. Photo-Optical Instr. Engrg. 843, 88-93, 1987. 29. Harrington, J.A. and Gregory, C.C., Hollow sapphire bers for delivery of CO2 laser energy, Optics Lett. 15, 541543, 1990. 30. Harrington, J.A., Gregory, C.C. and Nubling, R., Hollow waveguides for CO2 laser delivery systems, Proc. Soc. Photo-Optical Instr. Engrg., 1048, 117121, 1989. 31. Gregory, C.C. et al., Hollow curved Al2O3 waveguides for CO2 laser surgery, Proc. Soc. Photo-Optical Instrumentation Engrg. 1420, 169175, 1991. 32. Worrell, C.A., Infrared optical properties of glasses and ceramics for hollow waveguides operating at 10.6 m wavelength, Proc. Soc. Photo-Optical Instr. Engrg. 843, 8087, 1987. 33. Harrington, J.A. and Katzir, A., Specialty bers tackle tough problems, Laser Focus World 6, 139, 1991. 34. Saggese, S.J., Harrington, J.A. and Sigel, G.H., Jr., Attenuation of incoherent infrared radiation in hollow sapphire and silica waveguides, Optics Lett. 16, 2729, 1991. 35. Miyagi, M., Hongom A., and Matsuura, Y., Hollow IR waveguides, Proc. Soc. Photo-Optical Instr. Engrg. 1228, 2635, 1990. 36. Hongo, A. et al., Transmission of kilowatt-class CO2 laser light through dielectric-coated metallic hollow waveguides for material processing, Appl. Optics 51145120, 1992.

2
Optical and Thermal Response of Tissue to Laser Radiation
2.1 2.2 Introduction ......................................................................... 27 The Optical Response of Tissue.......................................... 28
Scattered Light Reection Reected Power Semi-Innite Slab Low Albedo High Albedo Rate of Heat Generation Selection of Wavelength Time Resolved Propagation

A.J. Welch
University of Texas

Craig Gardner
Rio Grand Medical Technologies, Inc.

2.3

Thermal Response................................................................ 41
Error Analysis

2.4 Summary............................................................................... 45 References ........................................................................................ 45

2.1 Introduction
A unied theory for the interaction of laser light with tissue is not possible at this time. However, it is possible to describe some basic interactions that provide some small insight into the effect of laser light on tissue. When an overall problem overwhelms analysis, engineers are taught to subdivide it into components that can be analyzed individually. Typically, lasertissue interactions have been divided into the optical response and the resulting thermal response. Actually, considerable progress has been made on many basic aspects of lasertissue interaction, and a few real problems have been analyzed. However, it has been difcult to string basic concepts together to describe nonlinear interactions such as tissue ablation, or situations where the physical properties of tissue do not remain constant during laser radiation. In this chapter, we will examine basic optical and thermal interactions. The applications and limitations of this analysis will be explored in terms of published experimental data. Typically, absorbed light is converted to heat; however, light absorbed by uorophores or photosensitizers can produce uorescence or photochemical reactions, respectively. A wide range of tissue reactions can be achieved by varying the laser pulse duration, energy per pulse, wavelength and spot size. In each case, the amount of energy absorbed in a specied volume is critical. The rate of heat generation S(r) at a position r(r = x,y,z) is equal to

S(r) = a(r) (r)

(2.1)

where (r) is the local uence rate [W/m2] and a is the local absorption coefcient [1/m] or [m1]. The total uence rate includes both the uence rate of scattered light and the uence rate of the attenuated collimated laser beam at r.

0-8493-1146-2/02/$0.00+$1.50 2002 by CRC Press LLC

27

28

Lasers in Medicine

2.2 The Optical Response of Tissue

If tissue is irradiated with a laser beam, the uence rate [W/m2] of the beam inside the tissue decreases exponentially (Beers law). The attenuation of the laser beam is due to absorption and possibly scattering of the collimated light according to the relation

E ( z ) = ( 1 rs ) E0 e
where E(z) E0 t rs

t z

(2.2)

is the attenuated beam within the tissue[W/m] is the irradiance [W/m2] is the attenuation coefcient [1/m] is the Fresnel specular reection for non-polarized light which is about 2% for light normal to an airtissue interface.

Lasertissue interaction is viewed as the propagation of photons through a volume of randomly distributed absorption and scattering centers. In the framework of this representation, the probability that a photon will be absorbed or scattered over a distance S is given by tS. Further, the probability that the photon is absorbed over the distance S is aS, where a is the absorption coefcient with units [1/m]. Likewise, it follows that the probability of a scattering event over S is s S, where s is the scattering coefcient with units [1/m]. Based on these denitions and assuming that absorption and scattering are disjoint events

t = a + s

(2.3)

Although Equation (2.2) assumes a continuous laser beam, Beers law holds for pulsed irradiation.

H ( z ) = ( 1 rs ) H0 e

t z

(2.4)

where H0 is the radiant exposure [J/m2] and H(z) is the energy per unit area of the attenuated laser beam [J/m2] at a distance z within the tissue.

2.2.1 Scattered Light

The radiance of scattered light, L(r, s), [W/(m2 sr)] at r in unit direction s is reduced by absorption and by scattering of light from direction s to other directions s as the light travels a distance ds. However, light is gained by scattering from other directions s' into direction s. The scattering from s to s is described by the single scattering phase function p(s, s ), which has units [1/sr]. For homogeneous tissue, we assume that the phase function can be described by the angle between s and s and does not depend on the initial direction s . That is,

p(s, s) = p(s s) = p(cos )

The expected (probability weighted) cosine of the scattering angle is given by

(2.5)

( cos ) p ( cos ) d( cos ) g = -----------------------------------------------------------p ( cos ) d( cos )

(2.6)

Light in tissue is forward scattered, and typically the value of g is between 0.85 and 0.99.

Optical and Thermal Response of Tissue to Laser Radiation

29

In concept, light propagation in a scattering medium can be described by the transport equation that describes the transfer of energy1

dL ( r, s ) 3 ------------------ = t L ( r, s ) + s p ( s, s ) L ( r, s ) d [ W m sr ] ds

(2.7)

Solutions of the transport equation include the exact adding-doubling method for one-dimensional problems,2 or the not-so-exact diffusion approximation for three-dimensional problems.1,3 However, most attempts to model light propagation in tissue rely on Monte Carlo simulations.4,5 The uence rate of scattered light at r is obtained by summing all of the light (radiance) through the point r. That is,

s ( r ) =

L ( r, s ) d [ W m ]
2

(2.8)

The importance of the contribution of scattered light to the spatial distribution of the rate of heat generation depends not only on the absorption coefcient at location of r but the relative ratio of the scattering coefcient to the attenuation coefcient. This ratio is called the albedo, a, where

s a = ----s = --------------t a + s

(2.9)

If a is greater than about 0.8, then the diffusion approximation provides a reasonable estimate of the distribution of scattered photons. When the albedo is less than 0.6, photons scattered from the collimated beam tend to be absorbed within or near the beam path, especially for beam diameters larger than the penetration depth of the beam, , where

1 = ---t

[m]

(2.10)

2.2.2 Reection
The reection of light at an interface can be described by considering a ray of light at angle 1, with respect to the surface normal as depicted in Figure 2.1. When the index of refractions of the two media are not matched, then the angle of reection is 1 and the angle of transmitted light 2 is given by Snells law

n 1 sin 1 = n 2 sin 2

(2.11)

1
air, n1

specular reflection

tissue, n2

transmitted into tissue

FIGURE 2.1 Reectance and transmittance of a nonpolarized ray of light incident upon an interface with index of refraction n1 for the top layer and n2 for the bottom layer. With permission from Kluwer Academic.17

30

Lasers in Medicine

At a boundary, the reection of a ray of non-polarized light at angle 1 with respect to the surface normal is given by the Fresnel law as

1 sin ( 1 2 ) tan ( 1 2 ) - ------------------------------ + ------------------------------r f ( ) = -2 sin 2( 1 + 2 ) tan 2( 1 + 2 )

(2.12)

2.2.3 Reected Power

The incremental power dP(r, s) owing in direction s with innitesimal solid angle d through an innitesimal area dA located at r is

dP ( r, s ) = L ( r, s ) cos dAd

(2.13)

where is the angle between s and a normal to the surface of dA. If the diffuse light is isotropic, then

L ( r, s ) = L 0

(2.14)

The reected power per unit area for diffuse (isotropic) light incident between 1 and 1 + d1 on the interface of two layers is

2 r f ( 1 ) L 0 cos 1 sin 1 d 1

(2.15)

where rf(1) is the Fresnel reection coefcient at 1. The specular reection of diffuse light incident on an airtissue interface is given by

power of reflected light r sd = ------------------------------------------------------power of incident light

For irradiance with diffuse light
2 2

(2.16)

r f ( 1 ) L 0 cos 1 sin 1 d 1

L0

r ( ) 2 sin cos d
f 1 1 1 0

r sd = --------------------------------------------------------------------- = --------------------------------------------------------------------2 L0 2 L 0 cos 1 sin 1 d 1

0 1

(2.17)

r ( ) sin ( 2 ) d
f 1 0

When light travels from layer 2 (tissue) to layer 1 (air), all rays at an angle 2 greater than a critical angle c are totally reected. That is,

r f ( 2 ) = 1 for 2 c

(2.18)

The fraction of light internally reected, rid, is given also by Equation (2.16). Unknown is the form of the radiance at the interface owing to the backscattering of light in tissue. If we assume that it is isotropic, then radiance can be represented by a constant. Thus the equation for internal reection, rid is

Optical and Thermal Response of Tissue to Laser Radiation

c

31

2 r f ( 2 ) L 0 cos 2 sin 2 d 2 + 2
0

r id = ------------------------------------------------------------------------------------------------------------------------------- L0
c

L cos sin d
0 2 2

(2.19)

r ( ) sin ( 2 ) d
f 2 2 0

1 - [ cos ( 2 c ) + 1 ] + -2

The critical angle c is given by Snells law for 1 = 90 as

n 1 c = arc sin --- n 2

(2.20)

The critical angle for total internal reection in tissue at an air boundary is about 45 for ntissue = 1.4. * An approximation r id for r id can be obtained by assuming rf(2) = 0 for 2 < c. Using Equation (2.19),

1 * - [ cos ( 2 c ) + 1 ] r id = -2 1 2 = cos c = 1 ---2 n

*

(2.21)

where n = n2/n1. r id is a lower bound on the fraction of diffuse (isotropic) light that is internally reected at the boundary. The importance of index refraction upon reectance is illustrated in Figure 2.2 for three situations: (1) reectance of collimated light normal to the interface; (2) diffuse light incident upon tissue rsd; (3) internal reection of diffuse light within tissue rid. When the media have similar boundaries, such as a watertissue interface, and internal reection is not signicant, the backscattered light leaves the tissue. This can produce a subsurface peak in uence rate just below the surface when the albedo is large and the spot size is much larger than the penetration depth.

2.2.4 Semi-Innite Slab

Many of the features of light propagation in tissue can be illustrated by considering a slab geometry where the laser beam is normally incident on the surface of the slab. Although the penetration of the light in the slab is primarily determined by the optical properties of the slab, the radius of the beam inuences penetration depth. These effects are illustrated in a series of gures where Monte Carlo simulations have been used to estimate the total light distributions (collimated plus scattered light) in an innite slab with an airtissue interface at the surface of the slab.

2.2.5 Low Albedo

An interesting and perhaps misunderstood situation occurs when the albedo is low, say 0.3 a 0.6. In this range, neither absorption nor scattering dominates. Light scattered from the laser beam is usually absorbed in the vicinity of the beam. Assuming a uniform irradiation of the slab simplies light propagation to a one-dimensional problem, and it is reasonable to expect the light to approximately decrease exponentially with depth. That is,

( z ) = k1 e

k2 z

(2.22)

32

Lasers in Medicine

Tissue Reflection Coefficient Collimated and Diffuse Light 0.8 0.7 0.6
Reflectance r id
surface collimated, rsc surface diffuse, rsd internal (approx) internal (integral),rid

r id (approximation)

0.5 0.4 0.3 0.2 0.1

r sc r sd

0.0 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0
Tissue Index of Refraction (relative to air interface)
FIGURE 2.2 Boundary reectance for (1) collimated light normal to surface of boundary, (2) diffuse light incident upon surface from region of low index of refraction, rsd, and (3) internal reection of isotropic diffuse light from region of high index of refraction, rid. The rid approximation is based on Equation (2.21). With permission from Kluwer Academic.17

Certainly k1 is not going to be much larger than E0(1 rs) where E0 is the irradiance and rs is the specular reectance at the surface. Suppose rs = 0 and k1 = E0 and the total uence follows Equation (2.22). The value of k2 follows from a simple energy balance. The total power absorbed per unit area is
k2 z a 0

S ( z ) dz = E e
0 0

a E0 dz = ---------- = E0 k2

(2.23)

For the assumed conditions, k2 must equal a. Thus, the distribution of total uence rate (collimated plus scattered light) is

( z ) = E0 e

a z

(2.24)

Equation (2.24) may be somewhat unexpected since the collimated light decreases according to Equation (2.1). For this one-dimensional example:

E ( z ) = E0 e

( a + s ) z

(2.25)

When the albedo is between 0.3 and 0.6, k1 is slightly larger (about 5%) than E0, the attenuation is not exactly exponential, and any exponential curve t has a coefcient k2 slightly larger (also about 5%) than a . For example, see Figure 2.3 for a = 0.5, g = 0.9 and an in nite beam radius. The z ( 1 r 2 ) E 0 e a curve of Figure 2.3 is not a bad approximation for the one-dimensional case, and satises the energy balance requirement.

Optical and Thermal Response of Tissue to Laser Radiation

33

a
1 r = 0.01 cm r = 0.03 cm r = 0.1 cm r=
8

0.8 Fluence rate on axis, (z) [W cm-2]

0.6

(1 -rs)exp(-az) (1 -rs)exp(-tz)

0.4

0.2

a = 0.5 a = 10 cm-1 g = 0.9 0 0.05 0.1 Depth, z [cm] 0.15 0.2

b
1 r = 0.01 cm r = 0.03 cm r = 0.1 cm r =
8

0.8 Fluence rate on axis, (z) [W cm-2]

0.6

(1 -rs)exp(-az) (1 -rs)exp(-tz)

0.4

0.2

a = 0.33 a = 10 cm-1 g = 0.9 0 0.05 0.1 Depth, z [cm] 0.15 0.2

FIGURE 2.3 Monte Carlo simulation of total uence rate at r = 0 as a function of depth for various beam sizes in a semi-innite medium with airtissue interface. Computations were made for uniformly distributed collimated laser z z beam. Tissue optical properties were a = 10 cm1 and g = 0.9. Curves of ( 1 r s ) E 0 e a and ( 1 r s ) E 0 e t provide bounds for the simulated uence rates. Computed for albedos of (a) 0.5, (b) 0.33 and (c) 0.6. (continued)

Equation (2.22) is still a reasonable approximation for total uence rate along the center line of the beam path in tissue, even when the spot size is small. Note that the total uence, , is approximately a z ( a + s ) z bounded from above by e and below by e as beam radius goes from to zero. The inuence of albedo is seen in Figure 2.3b for a = 0.33 and in Figure 2.3c for a = 0.6. In all Monte Carlo simulations, a = 10 cm1 and g = 0.9.

2.2.6 High Albedo

The chromophores that primarily affect wavelength-dependent absorption of light in the visible and IR spectrum of tissue are melanin, blood and water. However, between 600 nm and 1.2 m, the absorption coefcient of these chromophores is usually much, much less than the scattering coefcient of a tissue. Values for the albedo are typically from 0.75 to 0.99. Although a collimated beam is attenuated in a few

34

Lasers in Medicine

c
1 r = 0.01 cm r = 0.03 cm r = 0.1 cm r =
8

Fluence rate on axis, (z) [W cm-2]

0.8

0.6

(1 -rs)exp(-az) (1 -rs)exp(-tz)

0.4 0.2 a = 0.6 a = 10 cm-1 g = 0.9 0 0.05 0.1

0 0.15 0.2 Depth, z [cm]

FIGURE 2.3 (CONTINUED) Monte Carlo simulation of total uence rate at r = 0 as a function of depth for various beam sizes in a semi-innite medium with airtissue interface. Computations were made for uniformly distributed z collimated laser beam. Tissue optical properties were a = 10 cm1 and g = 0.9. Curves of ( 1 r s ) E 0 e a and t z provide bounds for the simulated uence rates. Computed for albedos of (a) 0.5, (b) 0.33 and (c) 0.6. ( 1 rs ) E0 e

optical depths, the resulting scattered light may propagate through several centimeters of tissue. Away from the collimated beam, the diffusion approximation can be used to describe the light distribution. The time varying diffusion approximation is

1 ( r, t ) 2 - = D ( r, t ) a ( r ) ( r, t ) + S 0 ( r, t ) -- ----------------c t
where c is the speed of light in tissue, S0 is the source term and D is the diffusion constant.

(2.26)

1 D = ---------------------------------------3 [ a + s ( 1 g ) ]

(2.27)

For steady state optics, ( ) ( t ) = 0 . A complete discussion of the diffusion approximation can be found in Refs. 1,3. The one-dimensional steady state solution (uniform irradiance E0 over the slab) for scattered uence rate s(z) for a semi-innite slab has the form
z z s ( z ) ----------- = Ae eff + Be t E0

(2.28)

where A and B are constants that depend on the boundary conditions and the effective attenuation coefcient, eff,

eff =

3 a [ a + s ( 1 g ) ]

(2.29)

The total uence rate (z) that includes the collimated light is
z s (z) -(z) + e t --------- = ---E0 E0

(2.30)

Optical and Thermal Response of Tissue to Laser Radiation

35

4 3 Fluence rate on axis, (z)/E [-] 2 1 0 -1 -2 -3 0 0.005 0.01 0.015 0.02 0.025 0.03 Depth, z [cm] Aexp(-tz) Bexp (-effz) (z)/Eo = s(z)/Eo + exp(-tz) s/Eo

exp(-tz)

FIGURE 2.4 Form of one-dimension diffusion approximation solution for the uence rate of the scattered light, z z z s ( z ) = Ae t + Be eff , and attenuation of the collimated beam, e t . Solution is for collimated irradiance of a semi-innite medium with an airtissue interface.

When the albedo is large,

eff < t
and, for many boundary conditions including index matched boundaries,

(2.31a)

A>0 B<0 A+B>0

eff z

(2.31b) (2.31c) (2.31d)

As z becomes large, the total uence rate is approximated by Ae . In fact, the basic denition of eff is not Equation (2.29), but rather the actual exponential attenuation coefcient of diffuse light with depth. The one-dimensional solution for the conditions of Equation (2.30) has the form depicted in Figure 2.4 for E0 = 1 and rs = 0.025. The curves were determined for the conditions of Figure 2.5; mismatched boundary a = 10 cm1, a = 0.95 and g = 0.9. eff z As expected, the uence rate is attenuated by Ae as z becomes large. Perhaps not expected is (1) the subsurface peak and (2) the uence rate below the surface are larger than the input irradiance. In the limit as a 0, the uence rate just below the surface (0+) 3 for matched boundary conditions.3

2.2.7 Rate of Heat Generation

Coagulation and ablation are governed by the rate of heat generation, S(r). As stated in Equation (2.4), the rate of heat generation is proportional to the product of the local absorption coefcient and local uence rate. An interesting question is the effect of scattering on S(r). In the following set of Monte Carlo simulations, the surface irradiance is 1 W/cm2, the absorption coefcient of the semi-innite, thick slab is 10 cm1, and the anisotropic factor, g, is 0.9. There is an index of refraction mismatch at the top surface to simulate an air (nair = 1)tissue (ntissue = 1.38) interface. Parameters are beam radius (0.01, 0.05 and 2.0 cm) and albedo (0.5, 0.8 and 0.95). The irradiance is assumed to be uniform within the beam radius.

36

Lasers in Medicine

0 0.05 0.1 Depth [cm] 0.15

a = 0.5 r = 0.2 cm
10

1.0 0.2 0.25 0.3 0.01 0.35 0.4 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 Radius [cm] 0.1

0 0.05 0.1 Depth [cm] 0.15 0.2

a = 0.85 r = 0.2 cm
10

1.0

0.1 0.25 0.3 0.35 0.4 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 Radius [cm] 0.01

0 0.05 0.1 Depth [cm] 0.15 0.2 0.25 0.3 0.35 0.4 0 0.05 0.1 0.15 0.2 0.25 0.01 10

a = 0.95 r = 0.2 cm

1.0 0.1

c
0.001

0.3

0.35

0.4

Radius [cm]

FIGURE 2.5 Monte Carlo simulation of the rate of heat generation for a beam radius of 0.2 cm. Irradiation parameters: Uniform irradiance E0 = 1 W/cm2, a = 10 cm1, g = 0.9 and mismatched boundary (n2/n1 = 1.38). The rate of heat generation is computed for albedos of 0.5, 0.85 and 0.95.

In the absence of scattering, the rate of heat generation just below the surface is 10 W/cm3 (neglecting specular reection for collimated light normal to the surface). The rate of heat generation decreases according to Beers law and the 1/e penetration depth is 0.1 cm. S(z) decreases to 1 W/cm3 at a depth of 0.23 cm. When scattering is zero, these values hold for any spot size as long as the irradiance is 1.0 W/cm2.

Optical and Thermal Response of Tissue to Laser Radiation

37

0 0.05 0.1 Depth [cm] 0.15 0.2 0.25 0.3 0.35 0.4 0

a = 0.5 r = 0.05 cm
10 1.0 0.1

a
0.01 0.001

0.05

0.1 0.15

0.2 0.25

0.3

0.35

0.4

Radius [cm]

0 0.05 0.1

a = 0.8 r = 0.05 cm 10 1.0 0.1

Depth [cm]

0.15 0.2 0.25

0.01 0.001

0.3 0.35 0.4 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 Radius [cm]

a = 0.95 r = 0.05 cm 10

0.05 0.1 Depth [cm] 0.15 0.2 0.25 0.3 0.35 0.4 0

1.0 0.1 0.01 0.001

0.05

0.1 0.15

0.2 0.25

0.3

0.35

0.4

Radius [cm]

FIGURE 2.6 Monte Carlo simulation of the rate of heat generation for a beam radius of 0.05 cm. Irradiation parameters: Uniform irradiance E0 = 1 W/cm2, a = 10 cm1, g = 0.9 and mismatched boundary (n2/n1 = 1.38). The rate of heat generation is computed for albedos of 0.5, 0.85 and 0.95.

For the Monte Carlo results presented in Figures 2.5, 2.6, and 2.7, the albedo is varied by changing the scattering coefcient and holding the absorption coefcient constant; the beam radius is 0.2, 0.05 and 0.01 cm respectively in these gures. For a large spot size (Figure 2.5), the rate of heat generation at the center of the beam ts a one-dimensional solution. Note that the depth of the 10 W/cm3 contour increases with increased albedo. When the albedo is equal to 0.95, all of the region above 0.035 cm has

38

Lasers in Medicine

0 0.05 10 0.1 1.0 Depth [cm] 0.15 0.2 0.25 0.3 0.35 0.4 0 0.05 0.1 0.15 0.2 0.25 0.001 0.1 0.01

a = 0.5 r = 0.01 cm

0.3

0.35

0.4

Radius [cm]

a = 0.8 r = 0.01 cm
0 0.05 10 0.1 Depth [cm] 0.15 0.2 0.001 0.25 0.3 0.35 0.4 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 Radius [cm] 0.1 1.0 0.01

0 0.05 0.1 0.1 Depth [cm] 0.15 0.2 0.25 0.3 0.35 0.4 0 0.05 0.1 0.15 0.2 0.25 0.01 0.001 10 1.0

a = 0.95 r = 0.01 cm

0.3

0.35

0.4

Radius [cm]

FIGURE 2.7 Monte Carlo simulation of the rate of heat generation for a beam radius of 0.01 cm. Irradiation parameters: Uniform irradiance E0 = 1 W/cm2, a = 10 cm1, g = 0.9 and mismatched boundary (n2/n1 = 1.38). The rate of heat generation is computed for albedos of 0.5, 0.85 and 0.95.

a rate of heat generation of at least 10 W/cm3. Without scattering, the rate of heat generation at 0.035 cm for a = 10 cm1 is 7.0 W/cm3. This is not a trick energy is conserved in both cases. As scattering increases, the curve depicting the decrease in the rate of heat generation is more quickly attenuated with depth (see Figure 2.8). As spot size is decreased (see Figures 2.6 and 2.7), radial scattering becomes increasingly important. As a result, the region of high rate of heat generation for a beam radius of 0.01 cm

Optical and Thermal Response of Tissue to Laser Radiation

39

1.6 Fluence rate, (z) [W cm-2] 1.4 1.2 1 0.8 0.6 0.4 0.2 0 0 0.05 0.1 0.15 Depth, z [cm] 0.2 0.25 0.3 a = 0.95 a = 0.85 a = 0.5 (1-rs)exp(-az)

FIGURE 2.8 Fluence rate as a function of depth at r = 0 (for the data of Figure 2.5). In addition, the uence rate for a = 0(s = 0) is plotted for a = 10 cm1.

is a depth of 0.01 cm. The penetration depths for 1 W/cm2 are about 0.15, 0.093, and 0.05 cm for albedos of 0.5, 0.85 and 0.95, respectively. From this analysis, it appears that scattering (as a single parameter) should not hinder coagulation or ablation. In fact, scattering increases the maximum value of the rate of heat generation when irradiance and the absorption coefcient are kept constant. The inability to coagulate or ablate tissue with a Nd:YAG laser ( = 1060 nm) is not because of the high albedo, but is due, rather, to the low values of absorption. Often, the tissue absorption coefcient at = 1060 nm is less than 0.5 cm1. In fact, over the entire spectrum from about 650 nm to 1.25 m, a is less than 1.0 cm1 for whitish-appearing tissue. To ablate whitish tissue it is necessary to increase surface absorption with blood (especially for 1060 nm radiation) or apply an exogenous dye. For example, Indocyanine Green has a broad absorption peak at 780 nm. Even a thin layer of ICG (5 g/L) with an absorption coefcient of 2200 cm1 can produce dramatic results during diode laser radiation. Once ablation is initiated, spots of carbonization form that provide the necessary absorption for continued ablation.

2.2.8 Selection of Wavelength

The selection of a laser should be based on the requirements for a particular task. In general, UV wavelengths transmit through water but are absorbed by proteins and amino acids. However, the primary absorber for wavelengths above 1.3 m is water. Between these wavelengths, tissue chromophores such as melanin and blood absorb the light. The inuence of these absorbers as a function of wavelength is illustrated in Figure 2.9. The least amount of absorption of light in tissue occurs in the 600 nm to 1000 nm band of wavelengths. The effect can be visualized by directing white light to one side of your hand. A red glow can be seen on the other side, indicating the red (630 nm) light has propagated through the tissue while shorter wavelengths have been absorbed. What is seen is light that has been scattered hundreds of times. Very few, if any, non-scattered photons are transmitted through the tissue of 1.0 cm (~ 50 optical depths). No nonscattered photons are transmitted through several cm thickness. Nevertheless, considerable research has been directed at using wavelengths in the 600 nm to 1000 nm window to image objects within tissue.

2.2.9 Time Resolved Propagation

Most of the imaging techniques exploit the increase in pathlength of scattered photons. Thus, for an impulse of light incident upon tissue, the light transmitted to a point r consists of non-scattered photons

40

Lasers in Medicine

absorption coefficient [mm-1]

104 103 102 101 100 10-1 10-2 10-3 10-4 10-5 0.1 excimer argon holmium neodymium aortic tissue hemoglobin water

10

wavelength [m]
FIGURE 2.9 Absorption spectra of aorta (interrupted line), blood (dotted line) and water (solid line).

that arrive rst (if there are any); this group is followed by photons scattered in a highly forward direction such that their pathlength is not much longer than the direct path of non-scattered photons (say, 10%). Other photons have longer pathlengths and arrive at longer times, as suggested by Figure 2.10. The diffusion approximations suggest photon propagation to an instantaneous point source in an innite medium can be represented by

( r, t ) = c ( 4 Dct )

32

r exp ----------- a ct 4 Dct

(2.32)

where c is the speed of light in the medium, D is the diffusion constant D = 1/3(a + s(1 g)) and r is the distance from the point source. In addition to imaging objects in turbid media, time-resolved measurements of light propagation provide information on the absorption coefcient, a, and the reduced scattering coefcient, s, where

s = s ( 1 g )
1.2

(2.33)

0.8

multiple scattered photons

intensity [a.u.] 0.6

0.4

0.2

light path < _1.1r

0 0 1 2 output time [a.u.] 3 4

FIGURE 2.10 Time-resolved propagation for an impulse of light incident at r = 0 with light measurement at position r.

Optical and Thermal Response of Tissue to Laser Radiation

41

The peak of Figure 2.10 is related to s and the attenuation within tissue is given by a.6 When a ray of light is delivered to a semi-innite slab, the reectance as a function of distance from the center of the ray and time is6

R ( r, t ) = ( 4 Dc )

3 2

( s ) t

1 5 2

r + ( s ) exp ------------------------- exp ( a ct ) 4 Dct

(2.34)

For total reectance, Equation (2.19) reduces to

( s ) - exp [ a ct ] exp -------------4 Dct R ( t ) = ---------------------------------------------------------12 32 ( 4 Dc ) s t

(2.35)

As ct becomes large, the measured attenuation provides a value of absorption. To avoid the difculties of time resolved measurements, it is possible to transform the diffusion approximation to the frequency domain. Images or optical properties are evaluated by examining the amplitude and phase resulting from irradiation with a modulated continuous wave laser.7

2.3 Thermal Response

The thermal response of tissue to laser radiation is governed by the local rate of heat generation (see Equation (2.4)). During the initial heating phase, the temperature response prior to the diffusion of heat from the region of heat generation is

a ( r ) t T ( r, t ) = ----------------c

(2.36)

Equation (2.4) is especially useful for estimating temperature of the end of a short pulse duration of t0.

a (r) T ( r, t 0 ) = ----c

(2.37)

where (r) is the uence [J/m2] distribution of the absorbed energy. When light scattering and specular reection can be neglected, the temperature at t0 at the surface of tissue is

a - W(r) T ( z = 0, r, t 0 ) = ---c

(2.38)

where W[J/m2] is the radial prole of the radiant exposure. In the absence of any phase changes, the transient temperature eld is described by appropriate boundary conditions and the heat conduction equation.

T ( r, t ) c ------------------ = k T ( r, t ) + S ( r, t ) t

(2.39)

where r is density [kg/m3], c is specic heat [J/kg K] and k is thermal conductivity [W/mK]. For an instantaneous point source in an innite medium, the transient temperature eld is given by

Q 1 r - exp --------T ( r, t ) = ----- ---------------------- 4t c 8 ( t ) 3 2

(2.40)

42

Lasers in Medicine

where = k/c is the thermal diffusivity [m2/s] and Q is the amount of heat instantaneously generated at r = 0. Temperature distributions for various simple source geometries are described in Carslaw and Jaeger8 for innite and semi-innite solids. Although analytic solutions exist for Beers law absorption of light and a Gaussian beam prole, most solutions of Equation (2.24) rely on either nite difference9,10 or nite element11 simulations. Excellent agreement between measured and computed temperatures have been reported for the eye.12 However, in situations involving an airtissue interface, Equation (2.39) has not adequately described the thermal process even for temperatures below 100C. Most models assure a convective boundary condition for heat ow f [W/m2] of

f = h ( T tissue T air )

(2.41)

where h is the convective heat transfer coefcient [W/m2 K]. Often, predicted and measured temperatures do not match, as illustrated in the following example of argon laser irradiation of in vitro human aorta. Example Torres et al.13,14 modeled and measured the temperature response of aorta to argon laser radiation. The model included a nite difference solution of the diffusion approximation for light propagation. Heat conduction was modeled with a nite difference algorithm based on Equations (2.24) and (2.26); the rate of heat generation was based on a diffusion approximation solution for uence rate produced by the argon laser radiation. Optical properties were determined from spectrophotometer measurements of in vitro tissue samples. Thermal properties for aorta were obtained from Valvano and Chitsabesan.15 Temperatures were measured with a thermal camera. Measured and computed temperatures at the center of the laser spot are presented in Figure 2.11. There are so many differences between the two curves that it is difcult to know why the curves did not match. Once predicted temperatures reached 100C (78C temperature rise), the temperature remained constant during a phase of water vaporization. Whenever experimental and predicted data do not match, many possibilities as to the source or sources of error can be found. This example exhibits a frightening number of possibilities. The task of predicting temperature is illustrated in Figure 2.12. Given a set of laser parameters (spot size, beam prole, irradiation time, power and wavelength), and a model for light propagation, it is possible to estimate the uence rate in the tissue. Computing the light distribution requires knowledge of the tissue optical properties, such as the absorption coefcient, a, and the reduced scattering coefcient, s = s(1 g).

80 70 Temperature Rise (0C) 60 50 40 30 20 10 0 0.0 0.5 1.0 Time (sec) 1.5 2.0 measured predicted

FIGURE 2.11 Predicted versus measured temperature during argon laser irradiation of normal aorta with 2W beam focused to a spot of 1.8 mm diameter. Optical properties used in the model were a = 3.0 cm1, s = 30 cm1 and g = 0.9.

Optical and Thermal Response of Tissue to Laser Radiation

43

Optical Properties

Absorption Coefficient

Thermal Properties

Laser Parameters

Light Propagation

Fluence rate, (r, t ) [W/m ]

2

Rate of Heat Generation

Heat Source, S (r, t ) [W/m ]

3

Heat Conduction

Temperature, T (r, t ) [oC]

FIGURE 2.12 Block diagram of steps required in the modeling of temperature resulting from laser radiation of tissue.

Once the uence rate has been computed, the rate of heat generation is determined using Equation (2.4). Then Equations (2.39) and (2.41) are used to compute the transient temperature eld. Sources of prediction errors are: Optical or thermal governing equations do not represent the physical problem. The computer program does not correctly represent the governing equation (Torres et al. used a two-dimensional, r, z, nite difference code). Values of the optical and thermal properties are not sufciently accurate for the required predictions. We have found that most users of models have complete faith in their programs at times unreasonable faith that assumes measurement differences are due to another botched experiment. Certainly, a number of possible sources of experimental errors exist, such as: Inability to correctly measure a transient temperature response (in this example, a 35 m Inframetrics model 600L thermal camera provided a two-dimensional image of surface temperature with a new image every 33.3 ms). Inability to correctly measure optical or thermal properties of aorta. (Optical properties were measured with a spectrophotometer and diffusion approximation algorithm. At 500 nm, a = 3 to 5 cm1 and s = 30 cm1; thermal properties of human aorta were obtained from Valvano and Chitsabesan.15 At 35C, k = 4.78 mW/cm C, = 1.27 103 cm2/s.) Irradiation parameters incorrectly measured. To determine the sources of error, it is necessary to examine Figure 2.11 and describe the major differences. Obviously, the measured steady state temperature is far below the computed temperature, and the initial slope of the temperature transient is about one third of the computed slope for the tissue sample with a measured absorption coefcient of 3 cm1. There is also a troublesome slope change at 60C in the experimental results.

2.3.1 Error Analysis

When large differences exist between measured and predicted results, the experimental design should be modied to simplify measurements and eliminate possible sources of measurement error. At the same time, it may be possible to reduce the dimensionality of a problem so measurements can be compared with an analytic solution. In this example, Torres expanded the laser beam, so temperatures at the center of the beam would be similar to those produced by collimated irradiation over the entire surface of a slab. The large beam minimized the effects of radial scattering of light and radial heat conduction. At least at the center of the beam, the optical and thermal variations were reduced to one dimension. The heat diffusion time given by

= -----4

(2.42)

where the characteristic length was assumed to be twice the effective penetration depth of the argon light (588 nm, 514.5 nm) in the aorta.16

44

Lasers in Medicine

2 = ----- eff

(2.43)

For these conditions, is approximately 740 milliseconds, and signicant heat conduction does not take place for the rst hundred milliseconds. On the other hand, a spot size of 1.0 mm produces a radial diffusion time of 250 ms and signicant heat conduction would take place in about 30 ms.16 During the period of minimal heat conduction, the surface temperature at the center of the laser spot is given by Equation (2.21). Irradiation with a large spot produced a 5:1 difference between predicted and measured temperature immediately after the onset of irradiation in Figure 2.13; the error must be due either to the value of the absorption coefcient a or the uence rate (0). Because aorta is approximately 80% water, it is difcult to believe that the thermal properties would produce such a discrepancy. It appears the product a(0) used in the model must have been too large. Independent Monte Carlo simulations of the uence rate agreed with the diffusion approximation results and both simulations yielded values of (0) that were about 20% larger than the irradiance. Thus, it appeared that the absorption coefcient a had been overestimated by a factor of three.13 By reducing the absorption coefcient (a = 1.6 cm1), the initial phase of the predicted and measured temperature matched. This did not solve the difference in the steady state temperature. Predicted temperatures below 100C were still 25% larger than measured values. Although steady state temperature is inversely proportional to thermal conductivity, there was no evidence that the true thermal conductivity would be a factor of 3:2 higher than the thermal conductivity measured by Valvano and Chitsabesan.15 Through additional experiments, Torres et al. determined that predicted and measured temperatures matched for non-biological materials with known optical properties. Thus, the only source of error had to be the governing equation or boundary conditions for heat transfer. The primary problem was surface cooling due to vaporization of water at the surface of the tissue. This temperature-enhanced heat loss was signicant for temperatures above 60C. A secondary heat loss was the phase change in tissue collagen that also occurred at about 60C. Torres et al. achieved excellent agreement between measured and predicted temperature when the absorption coefcient measured with a spectrophotometer was reduced and the boundary condition for an airtissue boundary was augmented with a heat loss due to vaporization of water. The reader is encouraged to examine the two papers by Torres et al.13,14 that illustrate the opticalthermal interaction of laser irradiated tissue. The single most important tissue property in lasertissue interaction is the absorption coefcient. Extreme care must be taken to accurately determine this parameter if the thermal response to laser light is modeled.
40 35 Temperature Rise (0C) 30 25 20 15 10 5 0 0.0 0.5 1.0 Time (sec) 1.5 2.0 measured predicted

FIGURE 2.13 Predicted versus measured temperature during laser irradiation of normal aorta with 7W focused to a 1 cm spot diameter. The absorption coefcient measured with a spectrophotometer for these samples was 5 cm1. The effect of heat conduction can be seen in the modeled temperatures but not in the measured temperatures, suggesting the actual characteristic length given in Equation (2.43) is less than the value obtained from the measured optical properties. That is, the penetration depth of light was deeper than expected, owing to the overestimation of the absorption coefcient.

Optical and Thermal Response of Tissue to Laser Radiation

45

2.4 Summary
Tissue response to laser radiation requires an evaluation of (1) the propagation of light in the tissue, (2) the rate of heat generation, and (3) the resulting temperature response. However, the optical-thermal interaction does not describe the effect on the tissue. Temperature-time-dependent rate processes describing thermal denaturation of tissue12,17 or ablation18,19 of tissue are usually the medical goals of laser treatment.

References
1. Ishimaru, I., Wave Propagation and Scattering in Random Media, Academic, New York, 1978, Vol. 1. 2. Prahl, S.A., M.J.C. van Gemert and A.J. Welch, Determining the optical properties of turbid media by using the adding-doubling method, Appl. Optics, 32(4):559-568, 1993. 3. Star, W., Diffusion theory of light transport, in Optical and Thermal Response of Laser Irradiated Tissue, A.J. Welch and M.J.C. van Gemert (Eds.), Plenum, 1995, pp 131200. 4. Keijzer, M., S.L. Jacques, S.A. Prahl, and A.J. Welch, Light distribution in artery tissue: Monte Carlo simulations for nite-diameter laser beams, Lasers Surg. Med., 9:148-154, 1989. 5. Jacques, S.L. and L.Wang, Monte Carlo modeling of light transport in tissues, in Optical and Thermal Response of Laser Irradiated Tissue, A.J. Welch and M.J.C. van Gemert (Eds.), Plenum Pub. Corp., New York, 1995, pp 305-332. 6. Patterson, M.S., B. Chance and B.C. Wilson, Time resolved reectance and transmittance for the noninvasive measurement of tissue optical properties, Appl. Optics, 28:2331-2336, 1989. 7. Dilworth, D.S., E.M. Leith, and J.L. Lopez, Three dimensional confocal imaging of objects embedded within thick diffusing media, Appl. Optics, 30:1796-1802, 1991. 8. Carslaw, H.S. and J.C. Jaeger, Conduction of Heat in Solids, Oxford University Press, Oxford, 1959. 9. Mainster, M.A. et al., Transient thermal behavior in biological systems, Bull. Math. Biophys.; 32:303314, 1970. 10. Incropera, F.P. and D.P. DeWitt, Fundamentals of Heat and Mass Transfer, Wiley, New York, 1985, Chap. 6. 11. Rastegar, S. et al., A theoretical study of the effect of optical properties in laser ablation of tissue, IEEE Trans. Biomed. Eng., 36(12):1180-1187, 1989. 12. Welch, A.J., Laser irradiation in tissue, in Heat Transfer in Medicine and Biology, A. Schitzer and R.C. Eberhardt, Eds., Plenum, New York, 1985, pp 135-183. 13. Torres, J.H. et al., Tissue optical property measurements: overestimation of absorption coefcient with spectrophotometric techniques, Lasers Surg. Med., 14:249-257, 1994. 14. Torres, J.H. et al., Experimental evaluation of mathematical models for predicting the thermal response of tissue to laser irradiation, Appl. Optics, 32(4):597-606, 1993. 15. Valvano, J.W. and B. Chitsabesan, Thermal conductivity and diffusivity of arterial wall and atherosclerotic plaque, Lasers Life Sci., 1:219-229, 1987. 16. van Gemert, M.J.C. and A.J. Welch, Time constants in thermal laser medicine, Lasers Surg. Med. 9:405-421, 1989. 17. Welch, A.J. et al., Denitions and overview of tissue optics, in Optical Thermal Response of LaserIrradiated Tissue, A.J. Welch and M.J.C. van Gemert, Eds., Plenum Press, New York, 1995, pp 15-46.

3
Optical and Thermal Dosimetry
3.1 3.2 Introduction ......................................................................... 47 Optical Dosimetry................................................................ 47
Terms and Denitions Light Detectors Source Light Measurements On-Line Output Monitoring In Situ Detectors Light Fluence Distributions in Tissue

Brian C. Wilson, Ph.D.

Ontario Cancer Institute and University of Toronto

3.3

Thermal Dosimetry.............................................................. 69
Introduction Thermography Electrical Probes Optical Fiber Temperature Probes Radiologic Imaging Methods

Douglas R. Wyman, Ph.D.

Hamilton Regional Cancer Center and McMaster University

3.4 Summary............................................................................... 80 References ........................................................................................ 80

3.1 Introduction
In many biomedical and clinical applications of lasers the accurate determination of treatment parameter such as wavelength or delivered optical power or energy, and of the spatial and temporal distribution of energy within laser-irradiated tissue, is essential to achieve the desired photobiological or clinical effect. For some purposes, it is possible to rely on the built-in dosimetry systems provided by the laser equipment manufacturer, where the user simply has to select the required treatment factors. However, as both equipment and applications become more complex and sophisticated, it is increasingly necessary either to check the laser system performance or to perform specialized measurements for particular treatment techniques, for example, the local energy uence rate on or in target tissues. The rst purpose of this chapter is to provide practical guidance in light dosimetry techniques and instrumentation. The presentation will be general, rather than specic to particular clinical applications, but the principles and approaches illustrated should enable specic measurements to be made with condence. The second purpose is to describe corresponding principles, techniques and instruments for determining the temperature distributions in tissue resulting from laser irradiation, because a wide range of clinical laser-based treatments are based on thermal effects. This will include both noninvasive, or remote sensing methods, based on the surface infrared emission from heated tissue, and the use of direct point temperature measurements with microthermocouples. Thermal dosimetry will be illustrated with reference to various specic clinical and preclinical applications.

3.2 Optical Dosimetry

In this section, after rst establishing the various terms and denitions relating to optical dosimetry, we will describe the performance characteristics of the different types of light detectors that are used

0-8493-1146-2/02/$0.00+$1.50 2002 by CRC Press LLC

47

48

Lasers in Medicine

commonly in dosimetry systems. The principles and methods for measuring laser beam parameters, with and without optical ber light delivery, will follow. Then, after an introduction to the characteristics of light uence distributions in tissue, the design and performance of optical ber-based probes will be described, leading to the general topic of performing interstitial measurements of light distributions in tissue. This section will conclude with a brief discussion of recently developed, noninvasive techniques for in situ optical dosimetry.

3.2.1 Terms and Denitions

Figure 3.1 illustrates the various laser treatment parameters for which measurement techniques will be discussed in later sections. These include both those parameters relating to the light as delivered by the laser or optical ber system and those involved in the spatial or temporal distribution of light within irradiated tissue. The corresponding terms and units are listed in Table 3.1, for both continuous wave (CW) and pulsed irradiation. For the laser output, delivered either as a direct beam or via optical bers, the optical energy, E0, and power, P0, are related simply by E0 = P0.t, where t is the treatment or exposure time. If the laser is operating with multiple output wavelengths, the power may be given either in terms of the power at each discrete wavelength or as the total power in all lines. The power density or intensity (and corresponding energy density) is a measure of how the laser output or light delivered to the tissue is spatially distributed. In the case of a uniform beam of area A, the average power density is simply equal to P0/A. However, in practice, there is often a considerable power variation across the beam. For direct optical delivery (e.g., via lenses or mirrors), the mode structure of the laser1 is a primary determinant of the beam shape (see Figure 3.2a). The separate modes are usually scrambled by optical ber delivery, but the ber output is not necessarily uniform and

Delivery System Laser

Tissue

. . . . .
Output Power Wavelength Pulse Parameters

Radiance Fluence (rate)

. . . .

Coupling Efficiency

Fiber Loss Bending Mode Mixing

. . .

Delivered Power Spot Area Source Distribution

FIGURE 3.1 Elements of laser treatment. The various dosimetric parameters at each point between the laser source and the irradiated tissue are indicated. TABLE 3.1
Term Radiance Fluence Rate Fluence

Radiometric Terms and Units

Symbol Denition radiant energy ux per unit solid angle in a given direction and per unit projected area perpendicular to that direction L(,t)d integral of the radiance over all solid angles at a point (t)dt integral of uence rate over the exposure time Units W.m2.sr1 W.m2 J.m2

Each radiometric term can also be expressed as a spectral quantity by dening it per unit wavelength.

Optical and Thermal Dosimetry

49

(a)

TEM 00

TEM

01

TEM 02

Laser

P i ii

(b)

Laser

ii x P i ii x

FIGURE 3.2 (a) Examples of low-order modes of laser operation. The fundamental transverse mode (TEM00) has a single Gaussian beam prole and is particularly suitable for direct tissue irradiation or efcient coupling into optical bers. (b) Beam proles from cleaved or (micro) lens-tipped optical bers.

depends both on how the laser beam is coupled into the ber and whether an optical element (e.g., a lens or diffuser) is used at the distal end (see Figure 3.2b). Note also that, in the case of a focused or defocused beam, the power density varies with distance even if there is no loss in total power. With a pulsed laser source (see Figure 3.3) the energy per pulse, Ep, can be determined from the average power and number of pulses per second (pulse repetition rate, Rp) by Ep = P0/Rp. The peak

Output Power

P o

CW

P p
(c)

Ep

time (a) Pulsed

P p P p
time

P p

/ e
(d)

Ep

(b)

FIGURE 3.3 Illustrations of the output power time dependence for (a) continuous wave (CW) and (b) pulsed sources, and the detailed pulsed shapes for (c) square-wave pulses and (d) Gaussian pulses.

50

Lasers in Medicine

power in the pulse, Pp depends on the pulse shape and, if present, substructure. In the simplest case of a square pulse of length (duration) , such as produced by a mechanical chopper or acousto-optic modulator, Pp = Ep/. For the more common (approximately) Gaussian pulse of width (full width at 1/e or 37% of the peak power), Pp = 0.886 Ep/. If there is substructure, for example, multiple micropulses within each main pulse (macropulse), then its characteristics must be known in detail and, depending on the biophysical effects induced in the tissue, the relevant dosimetric quantity may be either the peak power and energy in each micropulse or in the whole macropulse. The power and energy densities for pulsed irradiation are determined from the geometric characteristics of the beam in the same way as for CW sources. With the exception of the transparent elements of the eye (cornea, lens, vitreous), light is strongly scattered upon entering tissue, due to microscopic variations in the tissue refractive index.2,3 Generally, this does not result in energy loss, but the energy is spatially (and also, with short pulses in the subnanosecond range, temporally) redistributed. Before or after multiple scattering, the light may be absorbed by tissue chromophores (endogenous or exogenous), resulting in the photobiological effect, or be lost to the tissue volume as reected or transmitted light. Secondary photons (of longer wavelength) may be produced following absorption, which may, in turn, be scattered, reabsorbed or escape from the tissue.4 These absorption and scattering processes vary strongly with wavelength and tissue type. As illustrated in Figure 3.4, usually the energy ow at any point, x, within the, tissue is not equal in all directions, so that the angular distribution of the local power (energy) density is described by the radiance, L(x, ), which is the power density per unit solid angle, . The total energy uence rate, , at this point, is equal to the radiance integrated over all solid angles, and represents the total energy ow per second per unit cross-sectional area through the point. (The local energy absorption rate and consequent photobiological effects are generally insensitive to differences in the radiance distributions, but the latter do affect light dosimetry measurements: see Section 3.2.4.) At low uence rates, the local rate of energy absorption equals the product of the tissue absorption coefcient (probability of absorption per unit path length of photon travel) and the local energy uence rate. However, at high incident power densities, such as with short pulse, high-peak-power lasers ( 106 W.cm2), non-linear effects such as multi-photon absorption may occur, where the energy absorption rate depends on some power function of the local

Incident Light

x ~

L(x ~,)

Tissue

Local Fluence Rate (x ~)= L(x ~,)

Local Energy Absorption Rate . ~) =a(x ~) (x

FIGURE 3.4 Illustration of a non-isotropic radiance pattern of light at a point within tissue. The contribution to the local uence (rate) is not equal from all directions, due to the combination of the primary (unscattered) light and the scattered ux.

Optical and Thermal Dosimetry

51

uence rate (e.g. 2 for 2-photon absorption). This chapter will not be concerned with the details of these absorption processes and subsequent biological effects,5 except insofar as they inuence the types of physical dosimetry measurements required to characterize the light irradiation or resulting temperature distributions.

3.2.2 Light Detectors

All the dosimetric techniques described below rely on suitable detectors to convert the optical energy into electrical signals, which can be measured by standard electronic instruments.6,7 Alternative dosimeters, for example, actinometers based on the measurement of photochemical changes,8 will not be considered bacause they are not in common use. The selection of a particular detector for a specic application depends on matching the detector response (spectral range, sensitivity, response time, etc.) to the expected optical conditions. However, there are also signicant differences in factors such as ease of use and cost, and one must also consider whether a particular instrument is for a single use or for a range of potential applications in different light elds. 3.2.2.1 Photothermal Detectors Figure 3.5 shows a detector system based on the absorption of incident light by a blackbody surface (i.e., close to 100% absorption) and the measurement of the resulting temperature rise of the surface by conversion to an electrical signal. The main advantages of such detectors is that they have an almost at spectral response over a very wide wavelength range, do not depend on the pulse characteristics of the incident light, and can be used up to high power levels. Conversely, they have a slow time response (which limits their use if the light varies over a comparable time scale) and relatively poor sensitivity. They are, therefore, most often used for measurements of the average power output of lasers. The sensitive element may be of different size and shape for different applications, and the detector heads are often provided on a cable remote from the read-out instrument, which gives exibility in positioning the detector. 3.2.2.2 Photodiode Detectors Photodiodes are solid-state semiconductor devices in which the absorption of incident photons causes the production of free electron-hole pairs, which are subsequently detected as an induced electrical current. The efciency of this process is strongly wavelength dependent and depends also on the photodiode material, so that any one type of photodiode has a limited spectral range, and even within this range, the sensitivity is wavelength dependent, as illustrated in Figure 3.6a. However, within their ranges, the sensitivity of photodiodes is high and they have a good time response for many laser types and applications. The performance of photodiodes does, however, vary with the quality (and hence cost) of the device and associated electronics. For clinical dosimetry, high sensitivity, general-purpose radiometers using photodiodes are available (see Figure 3.7), with a range of detachable detectors of different spectral ranges, sensitivities and geometric forms, and with both power and energy registration. These may be bench-top or hand-held conguration. They cost from a few hundred to a few thousand dollars and are generally rugged and reliable. Depending on the manufacturer, they are either available with a built-in wavelength calibration or this is supplied in tabular or graphic form. Auto ranging is often provided for ease of use over a wide range of detected powers. If absolute light dosimetry is important, then it is advisable to ensure that the calibration can be traced back to a national standards laboratory. The upper power level is limited, both by potential damage to the photodiode itself and, more usually, by the builtin range of the electronics. 3.2.2.3 Photomultiplier Detectors Photomultipliers are vacuum-tube devices in which electrons released from a light-sensitive surface (photocathode) are electrostatically accelerated by a series of electrodes at increasing voltage, causing the release of further electrons, resulting in a highly amplied electrical pulse. The spectral range is also limited for these devices (see Figure 3.6b). They are relatively expensive and fragile and require special power supplies and detection electronics. Their advantages are very fast response time and extremely

52

Lasers in Medicine

FIGURE 3.5 Example of a photothermal power meter, based on light absorption by a blackbody surface and measurement of the consequent temperature rise. Such instruments typically have a range of interchangeable detector heads with either surface or volume detector elements. Different power ranges and laser types may be selected. Electronic time integration for energy measurements may also be provided.
1.0

Relative Response

0.8 0.6 Si

In Ga As

Bialkali Sb-Cs

0.4 Multialkali 0.2 0.0

400

800

1200

1600

2000

400

800

1200

1600

2000

Wavelength (nm) (a)

Wavelength (nm) (b)

FIGURE 3.6 (a) Typical spectral response of silicon and indium gallium arsenide detectors, suitable for measurements at visible and near-infrared wavelengths. (b) Typical spectral responses of photomultiplier tubes of different materials.

high sensitivity (low noise equivalent power) within their operating wavelength range, the latter resulting from the high signal gain within the tube itself. They can be operated in current mode, where the pulses from many incident photons are integrated as an electrical current, or in single photon counting mode, where individual pulses are detected digitally. In this mode, PMTs have typically about a 10% quantum efciency (i.e., one photon in ten is detected), and the sensitivity is limited only by the spontaneous emission of electrons from the photocathode. This can be minimized by cooling the tube (either electrically or with liquid N2) and, in a practical biomedical setting, a dark level of about 510 photons per minute can be achieved, which is equivalent to a CW power of about 1019 W. In this case, the upper limit on incident power is set by the electronics, typically ~107 counts per second (~1012 W). Thus, the price paid for extremely high sensitivity is a restricted power range. The balance can be shifted toward

Optical and Thermal Dosimetry

53

FIGURE 3.7 Example of a general-purpose, hand-held photodiode radiometer. These are available with a variety of detector head sizes and geometries for different applications.

higher powers by operating in current mode. Recently, solid state avalanche photodiodes have become available that also have intrinsic signal amplication, giving sensitivities on the order of those of some photomultiplier tubes. These have the added advantages of small size and ruggedness. 3.2.2.4 Accessory Devices Several items of equipment are useful accessories to the main detectors for practical dosimetry. Three of these are: 1. Filters and monochromators A wide range of spectral lters is available to select light only within a dened wavelength range.10 As illustrated in Figure 3.8, these may be band-pass, cut-on (high pass) or cut-off (low pass) depending, respectively, on whether light only within a dened band, above a given range, or below a given range is required. If scanning of a wide wavelength range is needed, the light can be passed through a monochromator placed in front of the detector. If very strong rejection of unwanted wavelengths is required (e.g., in measuring tissue uorescence on a background of ambient or scattered light), lters or monochromators can be stacked, e.g., two band-pass lters, each with 50% transmission at the peak and 10-4 transmission out of the band will, when used together, give 25% peak transmission and a 108 rejection factor for unwanted light. Note that there is always some loss at the selected wavelength, so that this must be taken into account for absolute dosimetry measurements. The loss through a monochromator depends on the design, slit width used and divergence or convergence of the incident light, so that it is not simple to determine these factors. Neutral density lters, which attenuate light over a very wide wavelength range, are useful for reducing the light intensity onto the detector to stay within the correct operating range. In most wavelength ranges, ND lters have a rather at transmission spectrum. The attenuation is usually cited in optical density units: OD = log10(Io/It), where Io = incident intensity and It = transmitted intensity. Thus, for example, lters of OD = 1, 2, 3 have attenuation factors of 10, 100, and 1000, respectively. For two or more lters used in combination, the ODs are added (attenuation factors multiplied).

54

Lasers in Medicine

T() Cut Off 1 Band-Pass Cut On

lo()

T() l() T()= l() lo()

lo()

T1 T2 l()

l()=lo().T1().T 2() T12()=T1().T 2()

FIGURE 3.8 Spectral transmissivities of different types of wavelength lters. Note that the maximum transmission is always less than 100%, and that there is some residual leakage at wavelengths outside the intended transmission range. For stacked lters, as illustrated on the right, the spectral transmission at each wavelength is the product of the individual transmissions.

2. Integrating spheres Most photodetectors have a at sensitive surface. This is suitable for measuring direct light beams where all, or a known fraction, of the beam area can be made normal to the detector surface. However, in many cases, particularly when optical bers are used, either for light delivery or collection, the light is distributed in many directions and not necessarily uniformly. As illustrated in Figure 3.9, an integrating sphere can be used to collect all the light, regardless of direction, so that, after being randomized by multiple scattering from the inner diffusely reecting surface, a known fraction of the light is detected. The larger the sphere, the simpler it becomes to couple the light into it, but the lower its sensitivity, because the fraction of light ultimately falling on the detector depends on the ratio of detector area to surface area of the sphere. For a 25 cm diameter sphere with a 1 cm diameter detector port, the efciency (light detected or light input) is typically about 1%. A bafe is often used to shield the detector from direct light from the source, thus reducing further the dependence of the signal on source position and geometry. 3. Beam choppers It is often convenient to chop a CW beam to provide (square wave) pulses of light. The magnitude of the change in detected signal can then be separated from any constant background light (see Figure 3.10). This increases the detection sensitivity, eliminates effects due to drift and, most importantly, permits measurements to be made under ambient room lighting. Simple and inexpensive mechanical choppers, operating up to a few kHz, are available. (Faster chopping can be obtained by using an acousto-optic modulator, but these are more expensive and complex to operate.) The detector output signal can be measured with an oscilloscope or, more accurately, using a lock-in amplier synchronized with the chopper.

3.2.3 Source Light Measurements

This section examines issues in the measurement of laser light sources, either using the direct laser beam or where optical bers are used for light delivery to the patient. It will be assumed that the appropriate detector system has been selected, based on matching the sensitivity, spectral range and temporal response to the source operating conditions.

Optical and Thermal Dosimetry

FIGURE 3.9 Illustration of the principle of an integrating sphere. The detected signal is approximately independent of the source geometry, since all photons are multiply scattered by the inner surface before reaching the detector. The bafe prevents the detector from seeing the source directly.

55

56

Lasers in Medicine

P o S B t

P o

Detector
t Scope

Driver

Lock-In Amplifier

Chopper

FIGURE 3.10 Principle of the use of an optical chopper to measure light signals in an ambient light background. The chopped signal is shown either measured with an oscilloscope or via a lock-in amplier (LIA) in which the constant baseline offset is subtracted.

3.2.3.1 Direct Beam Sources Figure 3.11 shows the two main conditions for direct beam measurements (with or without optical elements such as lenses, mirrors or lters between the laser and the detector). In the rst case, the beam is smaller than the detector-sensitive area. Thus, the optical power can be determined directly if the detector calibration is known at the source wavelength. In the second case, where the beam size at the detector is greater than the detector area, the signal at any position in the beam (after response calibration) is equal to the average power, P, over the detector area, a, i.e., to the local beam intensity (P/a). If the intensity is uniform, then the total power, Po = P . (A/a), where A is the cross-sectional area of the beam. If it is not uniform, as, for example, in the case of the output from a at-cut optical ber, then the total power can be obtained either by using an integrating sphere, as indicated above, or by integrating the local intensity measurements over all points in the beam. In most cases, the beam will be radially symmetrical, so that a single prole across a beam diameter sufces. 3.2.3.2 Optical Fiber Sources For a at-cut ber source, the beam prole can be obtained as described above. For bers having a light diffusing tip,11 it is easiest to use an integrating sphere (as illustrated in Figure 3.9) to measure the total output power, making sure the sphere is large enough to encompass the complete diffusing tip. If it is required to determine the spatial distribution of the ber output, for example, the isotropy of a point diffuser or the prole along the length of a line diffuser, then a small at-face detector can be scanned around or along the tip, ensuring that the detector is always oriented normally to the ber and at a constant distance. The total power can be determined in this way, by integrating over the surface area of, for example, a sphere traced by the detector at a constant distance. However, this procedure is time consuming and, unless a precision scanning mount is used for the detector and the complete surface is scanned, it is generally more accurate and convenient to use a calibrated integrating sphere.

Optical and Thermal Dosimetry

57

P o

(a)

(b)

FIGURE 3.11 Illustration of total beam power measurement for the cases of (a) beam size less than detector, (b) large beam, showing the measured signal prole from which the total power may be obtained by area integration.

A qualitative estimate of the output distribution from a ber source can be obtained by placing the source within a light-scattering medium (e.g., dilute milk). The output can then be visualized directly. A uorescent liquid, excited at the source wavelength, can also be used for this purpose, imaging the emitted light. Some trial and error with dilutions and placement of the source are usually needed to obtain a satisfactory picture in this way. It is possible to make this technique semi-quantitative, for example, by using a video or CCD camera to image the source.12 3.2.3.3 Spectral Measurements The output spectrum of a light source can be determined using either a monochromator or set of passband lters. The former gives a continuous wavelength scan with a spectral resolution determined by the bandwidth of the monochromator (typically a few nm for a simple laboratory instrument), while the latter samples the source output spectrum only at discrete wavelengths. More sophisticated and rapid systems (optical multichannel analyzer) use a diffraction grating spectrograph to spread the spectrum onto either a photodiode array or charged-coupled device (CCD). This can be particularly useful for measuring, for example, uorescence spectra in tissue. In cases where the source spectrum is broad band, it may be necessary to correct the measurements for the spectral response of the detector at each wavelength. If the measured power (density) at wavelength, , is Pm() and the corresponding detector responsivity is R(), then the true source power is

P t() = Pm()/R()
The total output power, integrating over all wavelengths is then

(3.1a)

Po = P t() d = Pm()/R() d

(3.1b)

3.2.3.4 On-Line Output Monitoring

It is often useful during a laser treatment to monitor the power delivered to the tissue in order, for example, to detect output variation. Also, many in vitro or in vivo spectroscopic measurements require

58

Lasers in Medicine

To Monitor Reflected Light Detector Fiber Lens Glass Slide To Tissue/ Delivery System To Monitor

Treatment Beam

Laser

Detector

(a)

(b)

FIGURE 3.12 Techniques for monitoring laser source output power, (a) in a direct beam, by detecting the specular reection from a glass slide and, (b) with ber coupling, by measuring the leakage component.

the delivered power to be known throughout the measurements. With clinical laser equipment, this online monitoring is often built into the system and used actively to control the output. When an open beam is used, the simplest method, as shown in Figure 3.12a, is to pass the beam through a thin glass slide and detect the small percentage of light reected. The slight loss in primary beam power can easily be determined by measuring the transmitted beam with and without the slide. One potential problem with this method is that the fraction reected depends on the polarization of the beam, which, in some lasers, may drift with time. With optical ber delivery, where the coupling efciency into the ber may be variable and it is not adequate simply to check the delivered power at the start and end of treatment, measuring the leakage from the ber can be a useful way to check at least that the output is not varying during treatment (see Figure 3.12b). In this case, because the leakage depends on the bending radius of the ber, it is necessary to hold the ber straight and at a xed position with respect to the detector by, for example, using a grooved clamp around the ber, to which the detector attaches. Even then, the method is not easily amenable to making absolute, reproducible measurements.

3.2.4 In Situ Detectors

For measurements of surface, intracavitary or interstitial uence (rates), as discussed in Section 3.2.5, specically designed detectors are usually required. Factors to be considered include safety, sterilizability, size, spatial and spectral response, degree of invasiveness and suitability for intraoperative, endoscopic or interstitial use. In some cases, small photodiodes held or sutured onto the tissue surface have been used directly.13 Electrical safety and restricted options for sterilization are limitations, and generally, the size of these detectors precludes their being used endoscopically or interstitially. On the other hand, calibration for absolute uence rate measurements is relatively straightforward. For interstitial and endoscopic measurements, optical ber detectors offer many advantages, at least in the visible and near-infrared wavelength ranges where a variety of types of ber with very low transmission losses and very small diameter are available. These can be easily coupled to appropriate photodetectors, depending on the required sensitivity, spectral and temporal response. If the uence to be measured is approximately isotropic, as in deep within a scattering tissue, then the distal end of the ber may simply be cleaved at, as illustrated in Figure 3.13a. The orientation of the ber tip does not alter the reading in this case, and the numerical aperture of the ber in the tissue determines what fraction of the light incident on the cleaved face will enter the ber and be detected. The detection sensitivity is proportional to the cross-sectional area of the ber core. In an anisotropic light eld, for example near an irradiated tissue surface or implanted (optical ber) source, the response of a at-cut detector ber depends on its orientation. It is possible to determine the uence rate at a point approximately by measuring at several orientations, as shown in Figure 3.13b, and

Optical and Thermal Dosimetry

59

Tissue

Tissue

(a)

(b)

FIGURE 3.13 Interstitial measurement of the local uence (rate) using cleaved optical ber detectors: (a)(approximately) isotropic eld using a single detector at an arbitrary orientation to the light beam, (b) anisotropic eld with measurements made simultaneously or sequentially at several specic orientations.

summing up the readings after appropriate weighting for the solid angle acceptance in each orientation.14 However, this is tedious and highly invasive, especially if the uence is required at multiple points in the tissue. Two solutions aimed at making detector ber tips with good isotropic response so that the orientation does not matter, even in an anisotropic eld, have been developed. The rst15 is to make the tip spherical and highly light scattering so that photons entering it from any direction are randomized and have the same probability of being transmitted to the photodetector (Figure 3.14a). The second type of isotropic detector16 is based on making the tip uorescent (Figure 3.14b). The emitted uorescent light is intrinsically isotropic and the tip geometry can be adjusted to make the response independent of the direction of the excitation light. The advantages of scattering-tip detectors are their relatively high sensitivity (typically ~105 104 cm2)16 and the fact that the spectral response is essentially set only by the photodetector used. Their principal limitation is that they cannot be made arbitrarily small, because the isotropy is lost if the optical pathlength in the tip is inadequate to give multiple scattering; commercially available isotropic probes have tip diameters of about 800 m. The isotropy of uorescent-tipped detectors is independent of size, provided the tip shape is maintained. However, any given detector has a limited wavelength range, and an additional pass-band or cut-on lter is required to discriminate against the excitation light (the direct detection of which is not isotropic). For a given tip diameter, the sensitivity is typically a thousand-fold less than the equivalent scattering tip, but this can be compensated for, at least in the visible range, by using more sensitive detection schemes, such as single photon counting. Fluorescent tips with good isotropy ( 15% except in the 20 backward angle of the ber stem) down to diameters of 170 m (which will pass down a 26 G needle) have been reported,16 although these are not currently available commercially. Both types of detector ber can be sterilized for clinical use. The useful life of the uorescent tips is limited by uorophore bleaching in some cases. The use of uorescent probes with pulsed irradiation can also be limited by the uorescence decay time of the tip (~ns). The response time using scattering tips is that of the photodetector, unless sub-picosecond pulses are involved. The calibration of optical ber detectors for in situ measurements in tissue involves numerous factors, particularly if absolute values of the local uence (rate) are required.1719 For relative measurements with a single detector, for example, monitoring with time during a treatment or at different positions in the radiation eld, calibration is unnecessary. For relative measurements between several detectors, it is sufcient to measure the response of each in the same (arbitrary) eld. It is, however, important to have the correct refractive index match (or mismatch) as in the intended treatment, because the detector

60

Lasers in Medicine

ex To Detector Tissue Tissue

em

Cut-On Filter

em (a) (b) ex

FIGURE 3.14 Isotropic uence detectors based on (a) light scattering in a spherical diffusing tip and (b) uorescence excitation in the tip, with a lter to remove the scattered excitation light that enters the ber.

isotropy and response depend on this and the effect is not, a priori, the same for each detector; for interstitial applications, it is convenient to use a broad (uniform) beam and a water tank for the relative calibration. If multiple bers are to be used sequentially with a single photodetector, it is also important to ensure the reproducibility of coupling between the ber and detector. The most accurate method to perform an absolute response calibration of an optical ber detector is to measure the response at depth in a medium of known light-scattering and -absorbing properties and refractive index, which approximate those of tissue.19 The medium, which should be large in volume so as to avoid boundary effects, is illuminated by a broad beam of known incident intensity. (Alternatively, an optical ber source of known output power can be used and the detector placed at a known distance from this.) A suitable model of light propagation in tissue (see Section 3.2.5) is then used to calculate the absolute uence rate, , at the position of the detector tip, so that the detector responsivity (at given wavelength) is R = S/, where S is the measured signal. In this form, the responsivity includes the photodetector response as well as the optical ber efciency, but the former can be removed if the photodetector is itself calibrated and the coupling efciency to the detector ber is known.

3.2.5 Light Fluence Distributions in Tissue

In this section, the general characteristics of light uence distributions in tissue and their dependence on tissue optical properties are discussed, together with practical methods to estimate or measure these distributions in vivo, including the use of the optical ber probes presented above. 3.2.5.1 Homogeneous Tissues The simplest case is the uence distribution in tissues that are optically homogenous, i.e., have approximately the same optical properties throughout the irradiated volume. For an external beam irradiance of incident uence rate 0 [W . m-2] the primary uence from the source is attenuated due to the combined scatter and absorption processes. Thus, at depth d in tissue, the primary uence rate, p, is given by

p = 0 . exp( -t.d)

(3.2)

where the total attenuation coefcient t = a + s and a, s are the absorption and scattering coefcients, respectively. Note that these coefcients, which describe the probabilities of each type of interaction, depend on the tissue and on the wavelength25 being the result of the chromophore content of the tissue

Optical and Thermal Dosimetry

61

(absorption) and the microscopic variations in refractive index (scattering). In cases where absorption dominates (e.g., at UV and mid- to far-infrared wavelengths), the total uence is simply equal to the primary uence, which falls with depth as exp( a. d), being a maximum (0) at the surface (Figure 3.15a). If the scattering is signicant (as in the case of visible and near-infrared wavelengths in soft tissues, with the exception of the transparent elements of the eye), then the uence rate within the tissue has a contribution from secondary, or scattered, photons. Thus, the total uence rate is

(d) = p(d) + s(d)

(3.3a)

For very high scatter-to-absorption ratio, this can be expressed analytically to a good approximation using diffusion theory20 as

(d) = A.0.exp( eff.d) B 0.exp( 't.d)

(3.3b)

The transport attenuation coefcient, t = a + 's. = a + s (lg), where g is the scattering anisotropy. (g = 0 for isotropic scatter, 1 for forward-peaked scatter.) For large incident beam size (5 10/'s) the effective attenuation coefcient can be written in terms of the absorption and transport scattering as

eff [3 a ( a + 's)]1/2

(3.4)

In Equation 3.3b, A and B are positive constants that depend on a, 's and the relative refractive index at the tissue surface. At large depth, d >>1/ eff, can be written as

(d) = ks.o.exp( eff.d)

where the backscatter factor, ks (A), increases with increasing tissue albedo ( s/ t). The consequences of high light scattering are seen in Figure 3.15b:

(3.5)

(a) The uence rate within the tissue near the surface is greater than the incident uence rate due to the backscattered photons. (b) There is sideways spreading of the beam prole. (c) The fall-off with depth depends on the scattering and absorption properties but is less rapid than that of the primary uence alone. (d) Backscattered light is lost through the irradiated surface as diffuse reectance, Rd. (e) The shape of the sub-surface distribution depends on the refractive index match or mismatch at the surface (e.g., water or air). Table 3.2 gives values of optical properties for some typical soft tissues and wavelengths, and also the 1/e depth, i.e., the depth over which the uence rate falls to 37% of its value ( = 1/ eff ). For long wavelengths (1500 nm) the attenuation of soft tissues is dominated by water absorption. In the near infrared (~7001300 nm), the absorption of other chromophores and of water is low and the tissue is scatter-dominated, giving the maximum depth of penetration of the light (optical window or therapeutic window). In the visible and UV, the wavelength dependence of absorption is very complex, due to the contribution of several types of endogenous chromophores, each with a distinct absorption spectrum (hemoglobin, melanin, cytochromes in the visible; nucleic acids, proteins and water in the UV). The scattering falls gradually with increasing wavelength and, in the visible range, a mixed scattering and absorption situation prevails. The attenuation is very sensitive to differences in tissue pigmentation, blood content and oxygenation in the visible and near infrared. In the UV, absorption again dominates, and differences among tissues are smaller.

62

Lasers in Medicine

Rd

d
Increasing Radiance Isotropy

o In - d e a

k s o In o e d (a)

- d e eff -td d (b)

r Tissue Tissue

In(r )

-ar e

In(r)

-effr e

r (c)

r (d)

FIGURE 3.15 Fluence (rate) distributions for external (laser) beam irradiation (a,b) or interstitial irradiation (c,d) for the cases of: (a,c) purely absorbing tissue, (c,d) scattering and absorbing tissue.

Optical and Thermal Dosimetry

63

TABLE 3.2
Tissue Any soft

Typical Optical Properties of Soft Tissues

<300 nm (excimer) 488/514 nm (argon) 633 nm (HeNe, dye) 1064 nm (Nd:YAG) ~2 m (Er:YAG:Ho) 2.94 m [H2O max] 10.6 m (CO2) a (cm1) 102104 ~1 ~10 ~10 <1 <10 ~5 0.55 ~102 ~3.103 ~103 s (cm1) ~103 102103 ~102 ~103 ~10
2

eff (cm1)

< 10 m

Rd (%)

Lightly pigmented Highly pigmented Blood Lightly pigmented Highly pigmented Blood Any soft

0.95 0.95 >0.97 0.9

510 >10 110 ~10 15 ~102 ~3.103 ~103

1 mm 1 mm 110 mm ~1 mm 210 mm ~100 m ~3 m ~10 m

1020 <10 2050 1020 2050

<50

Any soft

Note: The values should be considered as representing typical ranges in each case for general catagories of mammalian soft tissues [5, 10, 21, 22 and refs. therein]. The values for blood are for whole, oxygenated blood at 40% hematocrit. For absorption-dominated cases, the 1/e penetration depth is taken as d = 1/a; for high scattering cases (s >> a) as d~1/eff.

With interstitial irradiation, e.g., from an implanted optical ber source, the uence distribution around the ber is dependent on both the absorption and scattering properties of the tissue at the irradiation wavelength and the geometry of the source ber tip (cut-end, isotropic diffuser, cylindrical diffuser).11,23 In the simplest case of an isotropic (point) source of tip radius delivering an optical power po, the uence rate at radial distance, r, is given by:

absorption dominated: scatter dominated:

(r) = P0 . [exp ( a . (r-a))]/4r2 for r>a

(3.6a) (3.6b)

(r) = P0 . ([2 eff./ a)exp ( eff . (r-a))]/4r2 for r>a

Note that, when the tissue is primarily absorbing, the geometric factor describing the spherical spreading of the light varies as 1/r2, whereas, with high scattering, this has a weaker l/r dependence due to light scattered back toward the source (see Figure 3.15c and d). In both cases, the exponential attenuation (a. or eff ) dominates over the geometric spreading at large distances. The biological and resulting clinical consequences of the form of the uence distribution in tissue depend very strongly on the tissue type, wavelength and irradiation geometry. This will be clear in the other chapters dealing with specic applications of different laser types. However, to indicate the extremes, the low scattering and very high absorption at UV and mid- and far-IR wavelengths means that tissues may be cut or ablated with little spreading of the light beyond the deposition zone (depth ~ ). Conversely, with near-IR irradiation, the scattering causes energy absorption throughout a large volume of tissue, going beyond the geometric edge of the beam. The behavior in the visible range is intermediate and very strongly wavelength dependent; for example, there is a signicant difference in blood absorption between, say, an argon ion laser at 488/514 nm and a dye laser operating at 575/585 nm, the latter corresponding to an oxyhemoglobin absorption peak. This causes a profound difference in the connement of laserinduced heat damage to blood vessels, as in treatment of port-wine stain. 3.2.5.2 Inhomogeneous Tissue At wavelengths where absorption (due to water for mid-IR or cellular biomolecules for UV) dominates, tissues can generally be considered optically homogenous unless, for example, there are signicant variations in hydration state. However, in the near-IR and, especially, visible ranges, local variations in

64

Lasers in Medicine

sc e d
(a)

(i) (ii)

b d e d
(i)

b d Depth

(b) (ii) Depth

FIGURE 3.16 Idealized representation of the uencedepth distribution (i) and local rate of energy absorption (ii) in: (a) a skin layer model (sc: stratum corneum, e: epidermis, b: blood plexus, d: dermis) at wavelengths with high blood absorption. ea > da << ba, (b) a hot center model with discretely localized high absorbers.

chromophore content can markedly alter the distribution of uence and of absorbed energy deposition within the tissue. Two classes of tissue distribution have been considered; layered tissues and localized absorbing centers. The former is particularly relevant to skin irradiation,24 and the latter has been applied to the spatial connement of energy in melanin granules.25 An example of the uence rate, and resultant energy deposition rate, of visible light in a simplied model of the skin is shown in Figure 3.16a. The blood layer, in particular, affects the uence distribution, reducing the uence rate to deeper lying tissue and, at the same time, absorbing a signicant proportion of the light energy incident on it. In the case of localized absorbing centers, if the volume concentration of these is small, the effect on the light distribution may be slight (and approximated by assuming that the same total absorption is uniformly distributed); however, the rate of energy absorption within the centers themselves may be very high compared with that of the surrounding tissue structure (see Figure 3.16b), resulting in high local temperatures. In each case, the temperature reached by the highly absorbing elements depends on the local uence rate, their absorption coefcient and volume and the rate of heat dissipation. This will vary markedly with laser wavelength and pulse characteristics, with shorter pulses helping to conne the heat to the local absorber.26 3.2.5.3 In Situ Determination of Fluence Rate Distributions Two general approaches to the determination of light uence (rate) distributions in laser irradiated target tissues are direct interstitial measurement and calculation based on knowledge of the tissue optical properties and irradiation geometry. 1. Direct Measurements Direct in situ measurement of the uence rate at discrete points within tissue can be made using the probes described in Section 3.2.4.27,28 Isotropic probes placed interstitially will measure the total uence rates (Figure 3.17a). The use of such probes on a tissue

Optical and Thermal Dosimetry

65

Tissue (a) (b)

Tissue

Tissue (c) (d)

Tissue

FIGURE 3.17 Direct in situ measurement of local uence (rate) in tissue using isotropic optical ber detectors or photodiodes: (a) interstitial measurement of total uence rate, (b) measurement of surface uence (primary + backscattered), (c,d) measurement of primary incident uence, using (c) a small opaque shield to block the backscattered light or (d) a photodiode with the sensitive surface toward the beam. The backscatter can be monitored by switching the probe and shield positions in (c) or placing the sensitive photodiode face on the tissue (reverse of (d)).

surface in air (Figure 3.17b) may be uncertain by up to 30 or 40%, due to the refractive index mismatch between the probe and the tissueair interface. If the primary incident uence from surface irradiation is to be measured, then the sensitive tip should be shielded from backscattered light (Figure 3.17c). This is usually intrinsically the case if a at photodiode is placed face up on the tissue. The detector ( shield) should be small compared with the incident beam diameter so as not to distort the uence distribution signicantly. The diffuse reectance, Rd, can be measured using an integrating sphere on the tissue surface29 with narrow beam irradiation (Figure 3.18a), or by using a lens or optical ber (bundle) to transfer or image30 a fraction of the light onto a detector (Figure 3.18b). In either case, Rd should be measured with reference to a calibrated diffuse reectance standard, which comprises a at surface of a highly reecting material such as barium sulfate. The accuracy of direct uence measurement using interstitial probes is, of course, limited by the size and positional accuracy of the sensitive tip. This can be a particular problem in heterogeneous tissues, such as the skin, where the depth prole of the uence depends strongly on the specic thicknesses and optical attenuation of the different layers. To date, such direct microdosimetry studies have been very limited.16 A further problem with interstitial measurements, at least in vivo, is the possible error caused by bleeding around the tip due to the mechanical trauma of introducing the probe. This can signicantly reduce the detected signal, particularly at visible wavelengths where hemoglobin absorption is high.

66

Lasers in Medicine

P R

Tissue (a) (b)

Tissue

FIGURE 3.18 Measurement of total diffuse reectance, Rd, using: (a) an integrating sphere in contact with the tissue surface, (b) a non-contact camera with a collector lens at a xed distance from the surface and incident source position. In each case, a xed fraction of the diffusely reected light is detected, determined from equivalent measurements on a calibrated reectance standard.

2. Calculation If the optical properties (a, s, g, refractive index) of the tissue are known at the treatment wavelength, then the complete uence distribution can be calculated for a given irradiation geometry.3,11,20 For absorption-dominated cases, this is simple, as seen above. In the scattering-dominated case, at least for simple irradiation geometry and homogenous tissue volumes, the uence distribution can be calculated analytically using diffusion theory.31 In situations where this breaks down (e.g., if the scattering is not high compared with the absorption, or near sources and boundaries) or is too cumbersome to apply (e.g., for a heterogeneous tissue), numerical models such as Monte Carlo simulation can be used.32 The details of such models are beyond the scope of this chapter, but it should be emphasized, rst, that each model works accurately only for a dened and restricted set of circumstances and second, that the optical properties must generally be known to good accuracy for example, in the scatter-dominated case, an error of 10% in a and 's results in a 10% error in eff and hence to a 3040% error in the estimated uence rate at a depth of three penetrations depths and to correspondingly greater error at larger depths. Thus, even if the general distribution of uence rate is to be based on calculation, it is helpful to use interstitial detector measurements at a few points as a check. Particularly in the visible and near-IR, where there is considerable tissue-to-tissue and even subject-to-subject variation in absorption and scattering, such calculations of the uence beg the question of how to measure the optical properties. At other wavelengths, it is generally adequate to use generic values for each tissue, often determined from measurements made ex vivo. For homogeneous tissues, the absorption and scattering values can be determined either invasively or noninvasively, as illustrated in Figure 3.19. Invasive measurements The values of a and 's can be obtained by using two or more isotropic detectors interstitially to measure the uence rate, , at different known radial distances, r, from an isotropic source (or at different depths, d, during surface irradiation). For example, if 's >> a, then diffusion theory can be used and the values of a and 's calculated18 using Equation 3.6b. It is an important aspect of this method that the absolute uence must be measured. Only dependent variables can be derived from a relative measurement such as i/ j:

Optical and Thermal Dosimetry

67

(d1) (d2) d1 d2 Tissue (a) r1 r2

(r1) (r2)

Tissue

R() 1 2 3

Tissue (b)
FIGURE 3.19 Determination of tissue optical properties by (a) interstitial measurement of the local uence at multiple points with external or interstitial irradiation, (b) surface measurements of the local diffuse reectance as a function of the radial distance from the source on the tissue surface.

e.g., for an interstitial source and interstitial measurements, eff = {ln(rii/rjj)}/(rj ri) where the subscripts refer to different radial positions in the diffusion region. The accuracy of these calculations is increased if more than two measurement points are used, in which case, a best t to the diffusion equations can be performed.19 If only relative measurements can be made, yielding eff, then, because the total diffuse reectance, Rd, is uniquely dependent on the transport albedo a' = 's/(a + 's), also, measuring this allows both a and 's to be determined (using Equation( 3.4)) from

a = eff [(l a')/3]1/2 's = eff a'/[3(1-a')]1/2

(3.7a) (3.7b)

If the tissue is not scattering enough or the homogenous volume is too small (e.g., in the case of a solid tumor) to permit the use of diffusion theory, then multiple-point uence measurements (with or without measurement) still may permit the optical interaction parameters to be estimated, but more complex modeling is required. Note, however, that in this case, it may be necessary to also determine the scattering anisotropy, g, because the uence distribution may not depend simply on a and 's. This requires techniques that have not yet come into routine practice. Noninvasive measurements As shown in Figure 3.19b, a and 's can be determined by making multiple measurements of the radial dependence from a point source of the local diffuse reectance, R(). Again, if the diffusion conditions apply, R() can be expressed uniquely in terms of a and 's, so that the reectance curve can be tted to derive these coefcients,33 or a neural network that has been trained on model or experimental data can be used.34 It has been shown that absolute reectance measurements are not required if the range of values is chosen appropriately (typically /2<<5), because a and 's can be derived simply from the

68

Lasers in Medicine

t t t

t t

Tissue (a)

Tissue (b)

FIGURE 3.20 Time-of-f1ight measurements in tissue: (a) direct time domain, showing the temporal spreading (ns time scale) of the signal following a short (ps) input pulse, (b) frequency domain, showing the phase shift and demodulation of the detected signal with reference to the high frequency ( >~ 100 MHz) modulated input light. The tissue absorption and (transport) scattering coefcients can be derived from the detected signal, knowing the source-to-detector separation (which can be zero, allowing endoscopic or intravascular single ber probes).

shape of the reectance curve. Optical ber-based instruments have been developed, the most recent utilizing a broad-band source and spectrograph coupled to a 2-D CCD array to permit measurements of R() simultaneously at each value and for a range of wavelengths.35 This has been used for noninvasive absorption spectroscopy in vivo, to measure a(), from which specic chromophore features can be identied and their tissue concentration estimated. Recent modications to the local diffuse reectance method have been introduced, based on the photon time-of-ight in tissue4,36 as illustrated in Figure 3.20a. The principle is that photons undergoing multiple scattering between the source and a detector on the surface are delayed due to their variable pathlengths, l, through the tissue: time delay = l/ c', where c' is the speed of light in tissue (=co/n, with co = 3 1010 cm.s1 and n = tissue refractive index). Hence, if a short (~picosecond) pulse is incident on the tissue, then the detected pulse is broadened by the variable time delay (different pathlengths), and the shape of this time curve depends on a and 's (or also on s and g at very short times). An equivalent method, illustrated in Figure 3.20b, uses intensity-modulated incident light (typically at a frequency ~100500 MHz) and measures the phase shift and demodulation of the detected signal with respect to the input.4,36 Again, a and 's can be determined. A signicant advantage of these time-dependent measurements is that the reectance needs to be measured only at a single point on the tissue surface, and this can be close to the input source. Thus, endoscopic techniques become feasible. Their principal disadvantage at present is the technological complexity and cost compared with the steady-state, spatially resolved method. However, clinical systems based on light emitting diodes (LED) or diode laser sources are becoming available. The application of these noninvasive methods for cases of heterogeneous tissues is not clear. The local reectance signals can certainly be affected by heterogeneity, as in the case, for example, of different tissue layers.37 However, it is not known if this will be sensitive or specic enough to determine the distribution of optical properties with the accuracy required for clinical uence-rate calculations, for example, in skin. Finally, for some cases of layered tissues, and possibly also for localized highly absorbing centers within an otherwise homogenous volume, the technique of pulsed photothermal radiometry (PPTR) may play a role, at least for wavelengths where the absorption is signicant.38 As shown in Figure 3.21, infrared radiation (black body) is emitted from the tissue surface as the heat generated within the tissue by the absorption of light energy diffuses to the surface, following a short (~s) input pulse at the wavelength of interest. As the initial heat distribution depends on the distribution of the light uence and of the optical absorbers, the PPTR signal can be analyzed to estimate the optical absorption and scattering

Optical and Thermal Dosimetry

69

Input Pulse I.R. Signal t(s) Laser I.R. Detector t(ms) I.R. Lens

Tissue
FIGURE 3.21 Illustration of the principle of pulsed photothermal radiometry, for the case where the incident light pulse (~ s) at the wavelength of interest is delivered to the tissue surface by an optical ber and the emitted infrared radiation is collected by a lens and focused onto an IR sensitive photo detector (e.g., HgCdTe photodiode). The tissue optical properties at the irradiation wavelength are derived from the output signal magnitude and time dependence.

properties of the tissue itself39 or to measure absorbing exogenous dyes.40 It might also be possible to determine some information on the spatial distribution of localized absorbers such as a blood layer.41

3.3 Thermal Dosimetry

In this section, the measurement of temperature distributions in tissue resulting from laser irradiation will be considered.

3.3.1 Introduction
Many methods are available for measuring temperature changes in tissues resulting from the absorption of laser energy. Most of the methods currently in use were developed originally for industrial applications and then adapted to biomedical applications such as vascular imaging, tumor imaging and hyperthermia monitoring.42,43 The principal forms of tissue damage employed in surgical laser applications are photocoagulation, occurring at temperatures greater than approximately 60 C44 and vaporization at temperatures greater than 100 C. Although both of these are thermal effects, thermal monitoring is rarely performed during clinical laser surgery. In many established clinical applications, thermal tissue damage is well localized and predictable, rendering unnecessary the acquisition of temperature information. This is fortunate, because the task of measuring tissue temperatures with good spatial resolution and reasonable accuracy is formidable in the large local thermal gradients found in laser surgery sites. Thermal monitoring of laser-heated tissues has, in most cases, been restricted to laboratory investigations of effectiveness, optimization and safety that involve thermal dosimetry. Examples are temperature measurements during laser angioplasty;4551 laryngeal surgery;52 laser photocoagulation of the bladder wall,5356 abdominal wall,57 retina,58,59 brain,6062 stomach,63,64 port-wine stain;6568 and pulpal damage during dental surgery.69,70

70

Lasers in Medicine

There is, however, a medical laser technique in which small solid tumors (smaller than approximately 10 cm3) are destroyed thermally by laser energy directed through implanted optical bers, and for which thermal monitoring may become an integral component of the treatment. This minimally invasive technique, known as interstitial laser photocoagulation (ILP) or interstitial thermotherapy (ITT), involves delivering continuous laser energy at relatively low power (approximately 2 W per ber) with a long exposure duration (~1000s). ILP is currently being investigated for the destruction of primary tumors or isolated metastases in the liver or pancreas,7173 brain,7476 and head and neck.77,78 Still lower delivered powers (less than 1 W per ber) can be used to induce hyperthermia, a technique referred to as interstitial laser hyperthermia (ILH),79 possibly for use in combination with other modalities such as photodynamic therapy. Four general methods of thermal monitoring can be identied: thermography, electrical probes, optical ber probes and radiologic imaging. As shown in Table 3.3, these methods can be distinguished according to their invasiveness, portability and dimensionality (i.e., the number of spatial dimensions over which temperature data can be collected).
TABLE 3.3 Practical Characteristics of Methods of Thermal Monitoring in Vivo
Invasive yes yes no no Portable yes yes yes no Dimensionality 0 (point) or 1 (line) 0 or 1 2 (surface) 2 (cross-sectional) or 3 (volume)

Method Electrical Probes Optical Fiber Probes Thermography Radiologic Imaging

Generally, thermography and radiologic imaging methods offer the advantages of noninvasiveness and multidimensionality, but at the expense of accuracy and spatial resolution. In biomedical laser applications, most temperature measurements are made using electrical probes or thermography, although increasing use of optical ber probes might be anticipated. Most of the following discussion will focus on thermography and electrical temperature probes. Only brief summaries of thermal monitoring using optical ber probes and radiologic imaging will be given.

3.3.2 Thermography
Thermography is a noninvasive technique in which temperatures are monitored, recorded and displayed in a two-dimensional image, thereby allowing visualization of both thermal equilibrium and transient heating patterns.42 There are three types of thermography: liquid crystal, infrared and microwave thermography. Liquid crystal and infrared thermography are used to map surface temperatures, whereas microwave thermography can map subcutaneous temperatures. Of these three, only infrared thermography has been used appreciably in the biomedical laser eld. The discussion here will focus, therefore, on this technique, although brief discussions of liquid crystal and microwave thermography will also be provided, since these could nd biomedical laser application, for example, in monitoring ILP or ILH. 3.3.2.1 Infrared Thermography Basic Principles Infrared thermography involves the detection, by an infrared camera (pyrometer), of the electromagnetic radiation emitted from a surface at infrared wavelengths. The spectrum and total power emitted are governed by the blackbody radiation laws: Plancks Radiation law (spectral distribution), the Wien Displacement law (spectral peak) and the Stefan-Boltzmann law (total emitted power). A black body is an object that absorbs all radiation incident upon it, reecting and transmitting none. Most objects can be described as gray bodies, with radiative emission relative to a blackbody described by the emissivity, (0 1) and with total emitted power, E (W . m2), at absolute temperature T given by the modied Stefan-Boltzmann law:

E = T 4

(3.8)

Optical and Thermal Dosimetry

71

Camera Optical Path Patient Optics Video Display Detector Field of View

Microprocessor
FIGURE 3.22 Principle of infrared thermography, showing temperature mapping by a scanned detector eld of view.

where is the Stefan-Boltzmann constant, 5.672 108 W . m2 . oK4. Human skin, for example, is nearly a blackbody at infrared wavelengths ( 0.98).79 The spectral bands 35 m and 812 m are used for infrared thermography, because this minimizes absorption in air. The thermal detector is usually a crystal of mercury-cadmium-telluride or, for 35m radiation, indium-antimonide,48 in which the electrical conductivity varies with the total radiant power received, and which, in turn, is a function of surface temperature and emissivity. In practice, a measured voltage is translated through a calibration table to gray level intensity or a color band for display on a video monitor.48 (Conversion to a calibrated temperature map requires knowledge of the tissue emissivity.) An infrared camera is shown schematically in Figure 3.22. Temperature mapping is performed by optically scanning the narrow eld of view of the detector, using either rotating prisms or oscillating mirrors. Commercial cameras typically have spatial resolutions of approximately 1 mm and temperature resolutions of better than 0.2 C. 3.3.2.2 Infrared Thermography Biomedical Laser Applications In general medicine, infrared thermography has been applied primarily in the diagnosis of circulatory disorders and vascular disease, breast cancer and joint inammation.42,43 In the biomedical laser eld,48 infrared thermography applications are essentially limited to research studies. Typically, infrared cameras are used in experimental congurations to map tissue surface temperature distributions per se, or for radiometric inammation.42,43 Examples of surface temperature mapping are shown in Figure 3.23. The rst illustrates measurement of the heat accumulation and thermal damage following non-ablative CO2 1aser pulses.52 Note that the CO21aser wavelength, 10.6 m, falls within the detection band of cameras with mercury-cadmiumtelluride detectors, so that temperature measurements made during an irradiation may be erroneous, due to detection of reected radiation. Figure 23b illustrates a study in which a tissue phantom was used to model retinal photocoagulation during argon laser irradiation.59 Both the safety and efcacy of laser photocoagulation can be assessed using infrared thermography, for example, to estimate bladder wall temperatures during transurethral laser coagulation,54 to measure skin surface temperature during argon or tunable dye laser photocoagulation of port-wine stains,68 or to evaluate port-wine stain blood perfusion before and after argon laser therapy.65 Infrared cameras are also used to perform radiometry of laser wavelengths lying outside the two infrared detection bands. This involves wavelength conversion through absorption of the laser wavelength

72

Lasers in Medicine

CO2 Laser

Argon Laser

Infrared Camera Water Handpiece

Coagulated Egg White

Plastic Wrap Absorbing Layer

Biological Material

Infrared Camera

(a)

(b)

FIGURE 3.23 Examples of the application of infrared thermography: (a) to investigate heat accumulation and thermal damage to tissue following non-ablative CO2 laser pulses,50 (b) to model retinal photocoagulation, measuring the temperature changes induced in an absorbing layer.57
Black Body Nd : YAG Laser Beam (1064 nm) Infrared Camera 2 1 Darkened Surface Glass Hemisphere Tissue Sample Heat Gun Infrared Camera Sapphire Probe

1= 1.064 m 2= 2 - 5.8 m (a)

(b)

FIGURE 3.24 Radiometric applications of infrared cameras: (a) the use of a darkened external surface of a glass hemisphere to convert Nd:YAG laser energy (1.064 pm) reected from a tissue sample into thermal energy,61 (b) measurement of the emissivity of a sapphire ber tip probe.48

and re-emission of a thermal spectrum. Figure 3.24a illustrates a radiometric study in which the Nd:YAG power transmitted and reected from a tissue sample was converted on black-painted glass hemispheres and then measured with an infrared camera.63 A further radiometric application of infrared cameras is the determination of surface emissivity. This might be either an endpoint in itself or an intermediate step for temperature calibration. Emissivity can be determined by aiming the camera on the object placed in thermal equilibrium with a heated black body, and then adjusting the emissivity setting until there is no apparent temperature difference between

Optical and Thermal Dosimetry

73

Transparent Plastic Frame

Patient

Liquid Crystal Material In Inflated Balloon

Air Pump
FIGURE 3.25 Direct observation of thermal color map using liquid crystal thermography.

the tissue and the reference. Figure 3.24b illustrates one such experiment, to determine the emissivity of sapphire probes mounted as optical ber tips for use in laser angioplasty.50 3.3.2.3 Liquid Crystal Thermography Liquid crystals utilized for thermography are organic compounds, such as cholesterol derivatives, that selectively reect visible light in a narrow, temperature-dependent range of wavelengths. When incorporated into a contact applicator, such as the inatable balloon applicator depicted in Figure 3.25, they can be used to generate color maps that enable direct observation of a cutaneous thermal pattern. The color changes with progressive temperature elevation from longer to shorter wavelengths (reddish brown to dark blue). Liquid crystal thermography systems are simple, portable, commercially available and inexpensive. Biomedical applications include breast cancer detection81 and imaging of spinal root compression syndromes.82 Medical laser applications of liquid crystal thermography are limited by the contact nature of the technique. 3.3.2.4 Microwave Thermography Microwave thermography involves the detection of electromagnetic radiation emitted from the body at microwave frequencies (approximately 170 GHz). Unlike infrared and liquid crystal thermography, this allows reception of the thermal signals arising from subsurface tissues. The maximum sampling depth is determined by frequency. Microwave penetration in water decreases from approximately 3.3cm at 2.45GHz (a commonly used frequency) to 0.3cm at 9GHz.43 Conversely, spatial resolution increases with frequency. In microwave thermography, the thermal signal (microwave noise) emitted from the body is measured using a radiometer, consisting of an antenna and receiver. Both contact radiometers (110 GHz) and offset, or spaced, radiometers with focused antennas (1-7OGHz) are used, as shown in Figure 3.26. In the approximation that the probetissue coupling is optimal (black body), the power, P, collected by the receiver from a tissue volume at uniform temperature T is

P = kTf

(3.9)

where k = Boltzmann constant (1 . 38 1023 J . Kl) and f = receiver bandwidth (Hz). In practice, suboptimal coupling and tissue attenuation reduce the signal. Scanning is performed, as in infrared thermography, and the microwave signal is digitized, processed and displayed as a color image of temperature distribution. Biomedical applications of microwave thermography include cancer detection (breast, brain, thyroid) and noninvasive control of hyperthermia.83 Medical laser applications are likely to be conned to temperature imaging of internal (interstitial or endoscopic) laser irradiation, and may be limited by spatial resolution.

74

Lasers in Medicine

3.3.3 Electrical Probes

Unlike thermography, electrical probes for measuring tissue temperatures are implanted rather than applied externally. The four most common electrical temperature transducers are thermocouples, thermistors, resistive-temperature detectors (RTOs) and integrated circuit sensors. Practical descriptions of the basic operating principles of these devices can be found in commercial temperature measurement catalogs, such as Reference 84. RTOs are the most accurate and stable, but are expensive and seldom used clinically. The discussion here will be conned to thermocouples and thermistors, which are the only temperature probes commonly used in biomedical (laser) applications.
Offset Receiver 1m Reflector Video Display Contact Receiver Patient

Microprocessor

FIGURE 3.26 Microwave thermography using a contact or offset (focused ellipsoidal) receiver.

3.3.3.1 Thermocouples Basic Principles When two wires composed of dissimilar metals are joined at one end, a voltage is generated across the open ends of the wires, as indicated in Figure 3.27a. This thermoelectric (Seebeck) voltage, V, is a function of the sensing junction temperature and the composition of the two metals. For small changes in temperature, V increases linearly with the sensing junction temperature, T, through the Seebeck coefcient, = dV/dT. The most common bimetallic combinations are summarized in Table 3.4. Each of these has a nonlinear temperature vs. voltage characteristic (see Figure 3.28) that is usually tted by a high-order polynomial over a wide temperature range or a low-order polynomial over a narrow temperature range. At least one other junction in the measuring circuit must generate a thermoelectric voltage in series with V. In practice, circuits can be constructed with one such junction maintained at a known reference temperature, Tref, such as in an ice bath, and with all other junctions generating opposing voltages (see Figure 3.27b). Alternatively, both voltmeter leads can be extended to reference junctions located on an electrically insulating isothermal block, in which case, Tref is measured by a thermistor or RTO (Figure 3.27c). In both cases, Tref corresponds to a reference junction voltage, Vref, that can be subtracted from the measured voltage to give V. Thermocouples are the most versatile but least sensitive of electrical temperature probes; a typical thermocouple generates only 40 V/ C. They are, however, rugged, inexpensive, usable over a wide temperature range and simple, although requiring a reference temperature. Thermocouple temperature data can be acquired using a voltmeter or a commercially available data acquisition system with capability for multi-sensor input and computer interfacing. 3.3.3.2. Thermistors Basic Principles Increasing the temperature of a semiconductor decreases its resistivity by increasing the number of electrons excited thermally from the valence to the conduction band. The resistance (R) vs. absolute temperature (T) can be closely approximated by the Steinhart-Hart equation.84

Optical and Thermal Dosimetry

75

TABLE 3.4
Type +Metal Metal

Common Thermocouple Metal Combinations

E Chromel Constantan J Iron Constantan K Chromel Alumel R Platinum 13% Rhodium Platinum S Platinum 10% Rhodium Platinum T Copper Constantan

Voltmeter
+ Cu Cu V A B J

Voltmeter
+ Cu Cu A V A B Ice Bath Tref = 0oC J

Voltmeter
+ Cu Cu

Cu

Isothermal Block A J V

Cu R

(a)

(b)

(c)

FIGURE 3.27 Principle of the thermocouple showing dissimilar metals, A and B, generating a thermoelectric voltage, V, at their temperature sensing junction, J. At least one other junction in the voltage measuring circuit also consists of dissimilar metals (a) and contributes a separate voltage determined using a reference junction such as an ice bath (b) or isothermal block (c).

80
E

Voltage (mV)

60
K

40 20

R T S

500

1000 1500
o

2000

Temperature ( C)
FIGURE 3.28 Temperature vs. voltage characteristic for the six common thermocouple combinations listed in Table 3.5.82

1/T = Cl + C2lnR + C3(lnR)3

(3.10)

where Cl, C2 and C3 are material-dependent constants. Thus, thermistors are highly nonlinear devices. A typical R vs. T characteristic is shown in Figure 3.29.85 Linearized thermistors have been developed, although the use of computerized data acquisition generally renders this unnecessary. Thermistors are manufactured from oxides of nickel, manganese, iron, magnesium, copper, cobalt, titanium and other metals. They are the most fragile but also the most sensitive of electrical temperature probes, with temperature coefcients of resistivity typically around 4% per C. A disadvantage is their

76

Lasers in Medicine

106 104 102 1 10-2 0 2 4 6 (oK-1) 8 1000/Temperature

FIGURE 3.29 Resistivity vs. temperature characteristic of an Fe3O4 . MgCr2O4 thermistor.83

limited temperature range. The recommended maximum operating temperature is typically 100C. Higher temperature operation is possible, although at the risk of permanent loss of calibration.84 3.3.3.3 Thermocouples and Thermistors Biomedical Laser Applications Thermocouples and thermistors are useful for measuring temperatures in biomedical applications because they can be constructed with very small dimensions. So-called microthermocouples consist of two interwound wires with a combined outer diameter <<1 mm. These implantable temperature probes are available commercially, either bare or with a Teon coating to provide electrical insulation and waterproong. The smallest commercially available thermistors are also manufactured in tubular geometry, with diameters as small as 1 mm. Another feature useful in biomedical applications is that both devices have short response times, in the order of 1 s for thermistors and 0.1 s for microthermocouples. In the biomedical laser eld, electrical temperature probes are most often used for real-time temperature monitoring during interstitial laser heating of tumors (ILP and ILH), as illustrated in Figure 3.30. Most of this work to date has been in animal models.79,8690 Recently, ILP and ILH have been performed clinically in the brain with stereotactic implantation of microthermocouples or thermistors to monitor temperature,86,90 in the liver with implanted microthermocouples,73 and in head and neck cancer with a single implanted microthermocouple.91 Electrical temperature probes have been used in surgical laser studies to measure temperatures in normal tissues adjacent to the surgical target, for example, by implanting miniature thermistor probes beyond the edge of a CO2 laser lesion crater to investigate tissue damage and healing in the treatment of cervical intraepithelial neoplasia.57 Microthermocouples have also been applied to studies of the thermal effects of Nd:YAG and CO2 laser irradiation using animal models.79,80,8690 Recently, ILP has been used clinically in brain with stereotactic implantation of electrical temperature probes.76 The safety and efcacy of laser angioplasty have been assessed using microthermocouples by measuring the temperatures at or near the laser probe tip.49,51 The accuracy of these measurements may be limited by possible conduction errors. Other studies have used K-type thermocouples attached to the outer walls of harvested human femoral and popliteal arteries to help assess wall damage during laser angioplasty,51 as shown in Figure 3.31. A power meter for laser angioplasty has been developed based on temperature increase in water measured using a miniature thermistor.45 With the increasing use of the CO2 laser for soft tissue surgery in dentistry there is greater potential for thermal injury to teeth, which can result in irreversible pulpal damage. This problem has been

Resistivity (.cm)

Optical and Thermal Dosimetry

77

Optical Fiber Stereotactic Apparatus Temperature Probe Brain Tumor

(a)

Optical Fibers Temperature Probes Ultrasound Monitoring

Liver

Tumor

(b)

FIGURE 3.30 Interstitial laser photocoagulation (ILP) used to destroy inoperable tumors in (a) brain and (b) liver, with temperature monitoring using implanted microthermocouples/thermistors. Computed Tomography or Magnetic Resonance Imaging can be used at the same time to monitor tissue response, while ultrasound can serve the same function in the liver.70,73

Microthermocouples Plaque Laser Probe

Stenotic Artery

FIGURE 3.31 Microthermocouples attached to the outer wall and laser probe during laser angioplasty of a harvested human stenotic artery (modied from Ref. 49).

investigated by implanting miniature thermistors in the pulp chamber of freshly extracted human molars to assess thermal transfer to the pulp during laser irradiation of the external tooth surface.69 Electrical temperature probes can be made with intrinsic accuracy and precision of 0.1 C or better. In practice, however, they are generally subject to three main types of error,9295 two of which are particularly relevant to laser applications. The rst, self-heating, occurs when the probe directly absorbs the laser energy more rapidly than the surrounding tissues, and so records too high a temperature. The problem can be confounded by the probe, then heating the surrounding tissues. Self-heating errors arise only when the temperature probe is located within approximately two effective optical penetration depths (including geometric divergence) of the optical source, which is not usually the case. The second error, conduction, is introduced by thermal conduction within the temperature probe, resulting in poor spatial resolution. Conduction errors are greatest for metallic probes in high temperature gradients. They often arise in biomedical laser measurements, but their effects have not been well documented.

78

Lasers in Medicine

3.3.4 Optical Fiber Temperature Probes

Many concepts for ber-optic temperature probes have been developed in the last two decades.96,97 Most of these involve transmitting an optical input signal into one or more optical bers, and then deducing temperature from the optical signal returned by a sensor at the ber tip. The sensors can be either extrinsic, i.e., an added component located at the ber tip, or intrinsic, in which sensing takes place in the ber itself (core, cladding or jacket). The three main types of optical ber sensors are reective, uorescent and interferometric. Although commercial products are available, optical ber sensors are more expensive than electrical probes of comparable performance, and have not been widely used to date for biomedical temperature measurements. Their principal advantage is that they are unaffected by electrical and magnetic elds and so are well suited to applications such as temperature monitoring during magnetic resonance imaging, or during heating from radiofrequency electromagnetic radiation or radiofrequency currents (e.g., microwave hyperthermia).95 Because lasers have been used for distributed heating of small solid tumors7179 and also in conjunction with magnetic resonance imaging,75 it is possible that optical bers could be used in the future for both delivery of laser energy and temperature monitoring in the same application. 3.3.4.1 Optical Fiber Probes Basic Principles Five examples of reective (extrinsic) optical ber temperature sensors are illustrated in Figure 3.32.97 The rst is a bimetallic reector in which the differential thermal expansion of two metals in contact in the sensor generates a displacement that alters the amount of incident light reected. The second is also a bimetallic sensor in which each metal forms a reecting target, the relative displacement of which creates temperature-dependent interference fringes. In Figure 3.32c, the intensity of reected light is determined by a crystal with temperature-dependent birefringence, while, in Figure 3.32d, the sensor is a semiconductor crystal with a temperature-dependent absorption of infrared light. The sensor shown in Figure 3.32e is a mirror that reects light whose spectrum varies with temperature. The temperature is then determined from the intensity ratio at two discrete wavelengths in the optical output signal. In uorescent sensors, incident ultraviolet light is absorbed and re-emitted at visible wavelengths. Typically, the uorescing element is a phosphor coating placed on the tip of a single encapsulated, ultraviolet-transmitting ber, as shown in Figure 3.33.98 Temperature measurement is based on the temperature dependence of the intensity ratio at two discrete output wavelengths. Fluorescent sensors are the most commonly used optical ber temperature sensors in biomedical applications and have been available commercially since the 1980s. Interferometric temperature sensors are very sensitive to temperature-dependent changes in length (as evidenced, for example, in Figure 3.32b) and refractive index.96,97 Interferometric probes require a stable optical source, typically a gas laser. They are, therefore, relatively expensive and unlikely to be used widely for biomedical applications in the near future. Other optical ber sensors have been developed in addition to the three main types discussed above. Microbending of an optical ber can be used to measure temperature;99 infrared transmitting optical bers can be used to detect infrared radiation emitted by heated tissues directly, avoiding completely the requirement of an optical input source and an extrinsic sensing element; and infrared ber radiometry has been used to measure temperature transients following CO2 laser irradiation in tissue phantoms.100

3.3.5 Radiologic Imaging Methods

Certain radiologic imaging modalities can be used to measure temperatures in heated tissues. Unlike all other methods of thermal monitoring, radiologic imaging potentially can provide full cross-sectional or even three-dimensional temperature maps, although with poor spatial resolution compared with discrete interstitial probes. Radiologic imaging methods are, therefore, better suited to temperature

Optical and Thermal Dosimetry

79

Light source Aluminum Mirror stalk Transmit leg Thermal-contact plate Receiving leg Detector Silica substrate Input optical fiber From laser To Single-mode monitoring output optical fiber equipment

Bimetallic element

Silica Mirror stalk

(a) BIMETALLIC DISPLACEMENT

(b) BIMETALLIC INTERFERENCE

Protective jacket Sensing crystal Glass tube Fiber

Polarizer Transmit leg LED Detector Receiving leg

Mirror

Birefringent crystal

(c) BIREFRINGENT CRYSTAL

(d) SEMICONDUCTOR CRYSTAL

LED Lens

Connector Sensor

Photodiodes

Fiber

(e) SPECTRAL MODULATION

FIGURE 3.32 Various reective ber optic temperature probes. (Reprinted with permission from Fiber Optic Sensors, 2nd ed., Instrument Society of North America, 1992.)

Phosphor Layer (0.25mm thick)

Silica Core (0.4mm dia)

Silicone Cladding PFA (0.25mm thick)

PFA Sheath (0.7mm O.D.) PFA Encapsulation (0.8mm max. O.D.5mm long)

FIGURE 3.33 Fluorescent optical ber temperature sensor. (Reprinted with permission from Biomedical Thermology. Alan R. Liss, Inc. 1982.)

monitoring during cancer hyperthermia, which involves spatially distributed heating, than to measuring localized temperature increases during laser irradiation. Perhaps the most promising imaging modality for thermal monitoring is diffusion-sensitive magnetic resonance (MR) imaging.101 This involves calculating the molecular diffusion coefcient, D, from a ratio

80

Lasers in Medicine

of signal intensities in two MR images differently sensitized to diffusion, and then calculating temperature, T, based on the Stokes-Einstein relation

D exp( Ea/(kT))

(3.11)

where Ea is the activation energy for translational molecular diffusion. Thermometry based on ultrasound velocimetry has also been described.102 In this approach, a temperature map is calculated from a map of ultrasound velocity, based on prior knowledge of the coefcient of velocity variation with temperature.

3.4 Summary
Determination of light uence distributions in tissue prior to or during clinical laser irradiation is becoming practical due to the development of both invasive and noninvasive optical ber-based techniques. These techniques have resulted from improved understanding of tissue optical properties and light propagation in tissue. As laser-based therapeutic techniques become more sophisticated, there is an increasing reliance on accurate light dosimetry to ensure optimal therapy, and both general-purpose and applicationspecic instruments are expected to become an integral part of medical laser systems in the future. The principal forms of tissue damage exploited in many current clinical applications of surgical lasers are thermally induced, and a wide variety of thermal monitoring tools are available. Yet, to date, such monitoring of laser-irradiated tissues has been restricted primarily to pre-clinical laboratory investigations. A notable exception is interstitial photocoagulation of solid internal tumors, for which some form of in situ treatment monitoring is usually considered essential. Given that thermal monitoring enables important quantication of tissue damage during laser therapy, increased clinical utilization of the techniques described above can be anticipated.

References
1. Siegman, A.E., Lasers, University Science Books, Mill Valley (1986). 2. Wilson, B.C., Optical properties of tissues, in: Encyclopedia of Human Biology, (R. Dulbecco, Ed.), Vol.5, pp 578-597, Academic Press, New York (1991). 3. Patterson, M.S., B.C. Wilson and D.R. Wyman, The propagation of optical radiation in tissue II. Optical properties of tissues and resulting uence distributions, Lasers Med. Sci., 6, 379-390 (1991). 4. Wilson, B.C. and S.L. Jacques, Optical reectance and transmittance of tissues: principles and applications, IEEE J. Quant. Electr., 26, 2186-2199 (1990). 5. Bulnois, J.L., Photophysical processes in recent medical laser developments: a review, Lasers Med. Sci., 1,147-66 (1986). 6. Jacobs, S.F., Nonimaging Detectors, in: Handbook of Optics, (W.G. Driscoll and W. Vaughan, Eds.), Ch. 4, McGraw-Hill Co., New York (1978). 7. Sze, S.M., Physics of Semiconductor Devices, pp 743-789, J. Wiley & Sons Inc., New York, (1981). 8. Lilge, L. et al., Photoactivatable uorophores for the measurement of uence in turbid media, Photochem. Photobiol., 58, 37-44 (1993). 9. Demtrodes,W. Laser Spectroscopy, pp 196-227, Springer-Verlag, Berlin (1982). 10. Dobrowolski, J.A., Coating and Filters, in: Handbook of Optics, (W.G. Driscoll and W. Vaughan, Eds.), Ch. 8, McGraw-Hill Co., New York (1978). 11. Star, W.M., B.C. Wilson and M.S. Patterson, Light delivery and optical dosimetry in photodynamic therapy of solid tumors, in: Photodynamic Therapy, (B.W. Henderson and T.J. Dougherty, Eds.), pp 335-368, Marcel Dekker Inc, New York (1992). 12. Straight, R.C. et al., Application of charge-coupled device technology for measurement of laser light and uorescence distribution in tumors for photodynamic therapy, J. Photochem. Photobiol., 53 87-796 (1991).

Optical and Thermal Dosimetry

81

13. Sindelar, W.F. et al., Technique of photodynamic therapy for disseminated intraperitoneal malignant neoplasms: Phase I study, Arch. Surg. 126, 318-324 (1991) 14. Doiron, D.R., L.O. Svaasand and A.E. Proo, Light dosimetry in tissue: application to photodynamic therapy, in: Porphyrin Photosensitization, (D. Kessel and T.J. Dougherty, Eds.), pp 63-75, Plenum Press, New York (1983). 15. Star, W.M., J.P.A. Marijnissen and M.J.C. van Gemert, New trends in photobiology light dosimetry: status and prospects, J. Photochem. Photobiol., B1,149-187 (1987). 16. Lilge, L., T..Haw and B.C. Wilson, Miniature isotropic optical ber probes for quantitative light dosimetry in tissue, Phys. Med. Biol., 38, 215-230 (1993). 17. Marijnissen, J.P.A and W.P. Star, Quantitative light dosimetry in vitro and in vivo, Lasers Med. Sci., 2, 235-242 (1987). 18. Driver, I., C.P. Lowdell and D. V. Ash, In vivo measurement of the optical interaction coefcients of human tumors at 630 nm, Phys. Med. Biol., 36, 805-813 (1991). 19. Lilge, L. and B.C. Wilson, The accuracy of interstitial measurements of absolute light uence rate in the determination of tissue optical properties, Proc. Soc. Photo-Optical Instr. Eng., 1882, 291304 (1993). 20. Patterson, M.S., B.C. Wilson and D.R. Wyman, The propagation of optical radiation in tissue. I. Optical models of radiation transport and their application, Lasers Med. Sci., 6, 155-168 (1991). 21. Cheong, W.F., S.A. Prahl and A.J. Welch, A review of the optical properties of biological tissues, IEEE J. Ouant. Electr., 26, 2166-2185 (1990). 22. Svaasand, L.O. and C.J. Gomer, Optics of Tissues, in Dosimetry of Laser Radiation in Medicine and Biology, (G.J. Muller and D. H. Sliney, Eds), IS5, pp 114-132, SPIE Optical Eng. Press, Bellingham (1989). 23. McKenzie, A.L., How may external and interstitial illumination be compared in laser photodynamic therapy?, Phys. Med. Biol. 30, 455-460 (1985). 24. van Gemert, M.J.C. et al., Skin optics, IEEE Trans. Biomed. Eng., 36, 1146-1154 (1989). 25. Vitkin, I.A. et al., Optical and thermal characterization of natural (sepia afcinalis) melanin, J. Photochem. Photobiol., 59, 455-462 (1994) 26. Anderson, R.R. and J.A. Parrish, Selective photothermolysis: precise microsurgery by selective absorption of pulsed radiation, Science, 220, 524-527 (1983). 27. Star, W.M. et al., Light dosimetry for photodynamic therapy by whole bladder wall irradiation, J. Photochem. Photobiol., 46,619-624 (1987). 28. Chen, Q. et al., Effects of light beam size on uence distribution and depth of necrosis in supercially applied photodynamic therapy of normal rat brain, J. Photochem. Photobiol.,56, 379-384 (1992). 29. Anderson, R.R., J. Hu and J .A. Parrish, Optical radiation transfer in the human skin and applications in in vivo remittance spectroscopy, in: Bioengineering and the Skin ( R. Marks and P.A. Payne, Eds.), pp 253-265, MTP Press, Lancaster (1981). 30. Jacques, S.L. et al., Video reectometry to specify optical properties of tissue in Vivo, in: Medical Optical Tomography, (G. Muller et al., Eds.), IS11, pp 211-227, SPIE Optical Eng. Press, Bellingham (1993). 31. Jacques S.L. and S.A. Prahl, Modeling optical and thermal distributions in tissue during laser irradiation, Lasers Surg. Med., .6, 494-503(1987). 32. Flock, S.T. et al., Monte Carlo modeling of light propagation in highly scattering tissues-l: Model predictions and comparison with diffusion theory, IEEE Trans. Biomed. Eng., 36, 1162-1168 (1989). 33. Farrell, T.J., M.S. Patterson and B.C. Wilson, A diffusion theory model of spatially resolved, steadystate diffuse reectance for the noninvasive determination of tissue optical properties in vivo, Med. Phys., 19, 879-888 (1992). 34. Farrell, T.J., B.C. Wilson and M.S. Patterson, The use of a neural network to determine tissue optical properties from spatially resolved diffuse reectance measurements, Phys. Med. Biol., 37, 2281-2286 (1992).

82

Lasers in Medicine

35. Wilson, B.C., T.J. Farrell and M.S. Patterson, An optical ber-based diffuse reectance spectrometer for noninvasive investigation of photodynamic sensitizers in vivo, Proc. Soc. Photo-Optical Instr. Eng., IS6, 219-232 (1990). 36. Wilson, B.C. et al., Time-dependent optical spectroscopy and imaging for biomedical applications, Proc. IEEE, 80, 918-930 (1992). 37. Nossal, R.J., Kiefer, G.H. Weiss et al., Photon migration in layered media, Appl. Opt., 27, 33823391 (1988). 37. Anderson, R.R. et al., Pulsed photothermal radiometry in turbid media: internal reection of backscattered radiation stongly inuences optical dosimetry, Appl. Opt., 28, 2256-2262 (1989). 39. Prahl, S.A. et al., Determination of optical properties of turbid media using pulsed photothermal radiometry, Phys. Med. Biol., 37, 1203-1217 (1992). 40. Vitkin, I.A. et al., The feasibility of monitoring exogenous dye uptake in tissue in vivo using pulsed photothermal radiometry, J. Photochem. Photobiol., B16, 235-239 (1992). 41. Long, F.H., R.R. Anderson and T.F. Deutsch, Pulsed photothermal radiometry for depth proling of layered media, Appl. Phys. Lett., 51, 2076-2078 (1987). 42. Yang, W.J. and P.P.T. Yang, Literature survey on biomedical applications of thermography, BioMed. Mat. and Eng., 2, 7-18 (1992). 43. Biomedical Thermology, M. Guthrie and E. Albert, Eds., Alan R. Liss, Inc., New York (1982). 44. Thomsen, S., Pathologic analysis of photothermal and photomechanical effects of laser-tissue interactions, Photochem. Photobiol. 53, 825-835 (1992). 45. Tayler D. and D. Cumberland, A power meter for laser thermal angioplasty, Lasers Med. Sci., 3, 129-131 (1988). 46. Verdaasdonk, R.M.et al., Laser angioplasty with a metal laser probe (hot tip): probe temperature in blood, Lasers Med. Sci, 2, 153-158 (1987). 47. Welch, A.J., A.B. Bradley and J.H. Torres, Laser probe ablation of normal and atherosclerotic human aorta in vitro: a rst thermographic and histologic analysis, Circulation,76, 1353-1356 (1987). 48. Torres, J.H. et al., Limitations of a thermal camera in measuring surface temperature of laserirradiated tissues, Lasers Surg. Med., 10, 510-523 (1990). 49. Vincent, G.M. et al., Thermal laser probe angioplasty: inuence of constant tip temperature, plaque composition, and probe/vessel diameter ratio, Lasers Surg. Med.,10, 420-426 ( 1990). 50. Ashley, S. et al., Thermal characteristics of sapphire contact probe delivery systems for laser angioplasty, Lasers Surg. Med., 10, 234-244 (1990). 51. Silverman, S.H. et al., Effects of blood ow on laser probe temperature in human arteries, Lasers Surg. Med., 8: 555-561 (1988). 52. Brugmans, M.J.P. et al., Temperature response of biological materials to pulsed non-ablative CO2 laser irradiation, Lasers Surg. Med., 11, 587-594 (1991). 53. Staehler, G. et al., Dosimetry for Neodymium:YAG laser applications in urology, Lasers Surg. Med., 1, 191-197 (1980). 54. Rothenberger, K. J. et al., Transurethral laser coagulation for treatment of urinary bladder tumors, Lasers Surg. Med., 2, 255-260 (1983). 55. Hofstetter, A. and F. Frank, Laser use in urology, in Surgical Application of Lasers, (J.A. Dixon, Ed.), pp147, Yearbook Medical Publishers, Chicago (1983). 56. Pensel, J. et al., Temporal and spatial temperature prole of the bladder serosa in intravesical Neodymium:YAG laser irradiation, Eur. Urol., 7, 298 (1981).

57. Benina, J.H. and Y.J. Seto, Pathological and physical investigations into CO2 lasertissue interactions with specic emphasis on cervical intraepithelial neoplasm, Lasers Surg.Med., 1, 47-69 (1980).
58. Birngruber, R., Choroidal circulation and heat convection at the fundus of the eye: implications for laser coagulation and the stabilization of retinal temperature, in Laser Applications in Medicine and Biology, Vol. 5, (M. Wolbarsht, Ed.) Plenum Press, New York (1991).

Optical and Thermal Dosimetry

83

59. Jerath, M.R. et al., Dynamic optical property changes, implications for reectance feedback control of photocoagulation, J. Photochem. Photobiol., B16, 113-126 (1992). 60. Burke, L. et al., Thermal effects of the Nd:YAG and carbon dioxide lasers on the central nervous system, Lasers Surg. Med., 15, 67-71 (1985). 61. Wharen, R.E. Jr. et al., The Nd:YAG laser in neurosurgery: part 1. laboratory investigations: doserelated biological response of neural tissue, J. Neurosurg., 60, 531-539 (1984). 62. Gong-bai, C., Laser vaporization on intracranial tumors, Lasers Surg. Med., 1, 235-240 (1981). 63. Handrsson, T. W. et al., Theoretical and experimental investigations prove Nd:Y AG laser treatment to be safe, Lasers Surg. Med., 1, 253-262 (1981). 64. Marchesini, R. et al., Temperature rise in biological tissue during Nd:YAG laser irradiation, Lasers Surg. Med.,5, 75-82 (1985). 65. Troilius, A.K. et al., Evaluation of port-wine stain perfusion by laser doppler imaging and thermography before and after argon laser treatment, Acta. Derm. Venereol., 72, 6-10 (1992). 66. Patrice, T. et al., Thermography as a predictive tool for laser treatment of port-wine stain, Plast. Reconstr. Surg., 76, 554-557 (1985). 67. Mordon, S. et al., Relation between skin surface temperature and minimal blanching during argon, Nd:YAG 532, and CW dye 585 laser therapy of port-wine stains, Lasers Surg. Med., 13, 124-125 (1993). 68. Shakespeare, P.G., J. Hambleton and J.A.S. Carruth, Skin surface temperature during argon and tunable dye laser therapy of port-wine stains, Lasers Med. Sci., 6, 29-34 (1991). 69. Miserendino, L.J. et al., Evaluation of thermal cooling mechanisms for laser application to teeth, Lasers Surg. Med., 13, 83-88 (1993). 70. Sagi, A. et al., Heating of biological tissue by laser irradiation, temperature distribution during laser ablation, Opt. Eng., 31, 1425-1431 (1992). 71. Matthewson, K. et al., Biological effects of intrahepatic Neodymium:Yttrium-Aluminum-Garnet laser photocoagulation in rats, Gastroenterology, 93, 550-557 (1987). 72. Steger, A.C. et al., Interstitial laser hyperthermia, a new approach to local destruction of tumours, Brit. Med. J., 299, 362-365 (1989). 73. Schrder, T.M. et al., Fatal air embolism as a complication of laser-induced hyperthermia, Lasers Surg. Med., 9, 183-185 (1989). 74. Schrottner, O., P.W. Ascher and F. Ebner, Interstitial laser thermotherapy of brain tumours under MRI control, Abstract C-24, Fifth Int. Cong. Eur. Laser Assoc., Graz, Austria (1990). 75. Tracz, R.A. et al., Magnetic resonance imaging of interstitial laser photocogulation in brain, Lasers Surg. Med., 12, 165-173 (1992). 76. Roux, F.X. et al., Laser interstitial thermotherapy in stereotactical neurosurgery, Lasers Med.Sci., 7, 121-126 (1992). 77. Ohyama, M. et al., Laserthermia on head and neck malignancies experimental and clinical studies, Acta Otolaryngol., suppl. 458, 7-12 (1988). 78. Castro, D.J. et al., Interstitial laser phototherapy assisted by magnetic resonance imaging: a new technique for monitoring of lasertissue interaction, Lasers Surg. Med. suppl. 1. p.3 (abstr.) (1990). 79. Waldow, S.M., G.E. Russell and P.E. Wallner, Microprocessor-controlled Nd:YAG laser for hyperthermia induction in the RIF-l tumor, Lasers Surg. Med.12, 417-424 (1992). 80. van Hillegersberg, R. et al., Interstitial Nd:YAG Laser coagulation with a cylindrical diffusing ber tip in experimental liver metastases, Lasers Surg. Med.,14, 124-138 (1994). 81. Gautherie, M. et al., Long-term assessment of breast cancer risk by liquid-crystal thermal imaging, in: Biomedical Thermology. pp. 279-301, Alan R. Liss, Inc. New York (1982). 82. Pochaczevsky, R. et al., Thermographic study of extremity dermatomes in the diagnosis of spinal root compression syndromes, in: Biomedical Thermology, pp.339-360, Alan R. Liss, Inc. New York, (1982). 83. Leroy, Y., Microwave Radiometry and Thermography, Present and Prospective, in: Biomedical Thermology. pp.485-499, Alan R. Liss, Inc., New York (1982).

84

Lasers in Medicine

84. Omega Complete Temperature Measurement Handbook and Encyclopedia, 26, Omega Engineering Inc., USA (1988). 85. Askeland, D.A., The Science and Engineering of Materials. PWS-Kent, Boston (1989). 86. Godlewski, G. et al., Deep localized Neodymium (Nd)-YAG laser photocoagulation in liver using a new water cooled and echoguided handpiece, Lasers Surg. Med., 8, 510-509 (1988). 87. Panjehpour, M. et al., Nd:YAG laser-induced interstitial hyperthermia using a long frosted contact probe, Lasers Surg. Med.10, 16-24 (1990). 88. Wyman, D.R.et al., A control method for a nonlinear multivariable system: application to interstitial laser hyperthermia, IEEE Trans. Biomed. Eng. 38, 891-898 (1991). 89. Daikuzono, N. et al., Laserthermia, a new computer-controlled contact Nd:YAG system for intersititial local hyperthermia, Lasers Surg. Med., 8, 254-258 (1988). 90. Grossweiner, L.I. et al., Modeling of tissue heating with a pulsed Nd:YAG laser, Lasers Surg. Med.,10, 295-302 (1990). 91. Panjehpour, M. et al., Nd:YAG laser-induced hyperthermia treatment of spontaneously occurring veterinary head and neck tumors, Lasers Surg. Med., 11, 351-355 (1991). 92. Performance Evaluation of Hyperthermia Equipment, AAPM Report. 26, American Institute of Physics (1989). 93. Lyons, B.E., T.V. Samulski and R.H. Britt, Temperature measurements in high thermal gradients: II. analysis of conduction effects, Int. J. Radiat. Oncol. BioI. Phys., 11, 963-971 (1985). 94. Dickinson, R.J. Thermal conduction errors of Manganin-Constantan thermocouple arrays, Phys. Med. Biol., 30, 445-453 (1985). 95. Lyons, B.E., T.V. Samulski and R.H. Britt, Temperature measurements in high thermal gradients: I. The effects of conduction, Int. J. Radiat. Oncol. Biol. Phy., 11, 951-962 (1985). 96. Grattan, K.T.V., Fiber Optic Chemical Sensors and Biosensors. Vol. 11, O.S. Wolfbeis, Ed., CRC Press, Boca Raton, (1991). 97. D.A. Krohn, Fiber Optic Sensors 2nd Ed., Instr. Soc. A., North Carolina, (1992). 98. Wickersheim, K.A. and RV Alves, Fluoroptic Thermometry: A New RF-Immune Technology, in: Biomedical Thermology. pp.547-554, Alan R. Liss, Inc., New York (1992). 99. Gottlieb M. and G.B. Brandt, Measurement of temperature with optical bers using transmission intensity effects, Proc. Electro-Optics Conf., Anaheim (1979). 100. Zur, A. and A. Katzir, Use of infrared bers for low-temperature radiometric measurements, Appl. Phys. Lett., 48, 499-500 (1986). 101. LeBihan, D., J. Delannoy and R.L. Levin, Temperature mapping with MR imaging of molecular diffusion: application to hyperthermia, Radiology, 171, 853-857 (1989). 102. Robert, J. et al., Ultrasound Velocimetry for Hyperthermia Control, in: Biomedical Thermology, Alan R. Liss, Inc., New York (1982).

4
Uses and Effects of Ultraviolet Radiation on Cells and Tissues
List of Abbreviations Used in this Chapter .................................. 85 4.1 Introduction to Ultraviolet Radiation ................................ 86 4.2. A Division of the Ultraviolet for Photobiological Studies...86
Vacuum UV (10 nm190 nm) UV-C (190 nm290 nm) UVB (290 nm320 nm) UV-A (320 nm380 nm) Terrestrial Solar UV (290 nm380 nm)

4.3 4.4 4.5 4.6 4.7 4.8

UV Sources ........................................................................... 88 Absorption of Ultraviolet .................................................... 90

Division Between Ionizing and Non-Ionizing UV Molecular Absorption Absorption by Cells Absorption by Tissue

Direct vs. Indirect Effects of UV......................................... 96 Action Spectroscopy: Effect as a Function of ................. 96
UV-C (190 nm290 nm) Action Spectra UV-A (320 nm380 nm) Action Spectra

Effects of UV on Cells ....................................................... 100

Cell Death Cell Mutation Membrane effects

Complex Responses of Tissues.......................................... 101

Immune Effects Carcinogenesis

Thomas P. Coohill
Siena College

4.9 Repair .................................................................................. 103 4.10 General Effects.................................................................... 104 4.11 A Perfect Laser for UV Photobiological Studies.............. 104 4.12 Useful Review Texts............................................................ 105 References ...................................................................................... 105

List of Abbreviations used in this Chapter

AS eV UV UV-A UV-B UV-C VUV action spectra electron volt = 1.6 1019 J wavelength(s) 10 nm 380 nm 320 nm 380 nm 290 nm 320 nm 190 nm 290 nm 10 nm 190 nm

0-8493-1146-2/02/$0.00+$1.50 2002 by CRC Press LLC

85

86

Lasers in Medicine

4.1 Introduction to Ultraviolet Radiation

The known electromagnetic spectrum is open-ended and continuous, having no inherent upper and lower bounds. It extends from the longest measured wavelength (about 104m) radio waves to the shortest measured wavelength (about 10-16 nm) -rays.1 Of most interest to biologists and hence life on earth is the terrestrial solar region (290 nm to longer than 2000 nm).1 This region contains those wavelengths emitted by the sun that penetrate the earths atmosphere to reach the surface, although technically any wavelength reaching the biosphere (maximum height about 6 km) can affect living processes. Solar radiation is responsible for almost all of the energy added to the biosphere and drives such important bioreactions as photosynthesis and vision. Studies of the effects of non-ionizing electromagnetic radiation on living systems is known as photobiology. In addition, some photobiological research is conducted using ultraviolet (UV) wavelengths that are shorter than those contained in terrestrial solar radiation (i.e., shorter than 290 nm) This chapter limits itself to a discussion of the uses and effects of UV on cells and tissues. Ultraviolet radiation is that portion of the electromagnetic spectrum that extends from the lower wavelength limit of human vision (usually dened as 380 nm or 400 nm) to wavelengths as short as about 10 nm, where it overlaps the x-ray region. In the natural environment, the shortest wavelength of sunlight that can be routinely measured at the earths surface is about 290 nm, largely due to the absorption properties of ozone and other atmospheric gases. So, the only environmentally relevant UV region is from 290 nm to 380 nm. However, articial UV sources such as certain uorescent lamps, mercury and xenon arcs, and lasers, are readily available and extend the possibility of human exposure to UV down to wavelength limits of about 190 nm. Below 190 nm, air (oxygen) and water begin to absorb UV heavily making it difcult2 to expose biological samples except under extreme conditions (e.g., in a vacuum).

4.2. A Division of the Ultraviolet for Photobiological Studies

It is always difcult, and for some, unreasonable, to fractionate a continuum into convenient usable regions. For example, red is dened as very approximately, the wavelength region 630 nm to 700 nm.3 Such imprecise denitions sufce for normal conversation, but more rigorous guidelines should prevail for scientic presentation. This is especially important when the inclusion (or exclusion) of certain wavelengths can markedly alter the biological response of the effect being studied (as is often the case for the various denitions of the UV-B). The naming of some of the regions of the electromagnetic spectrum is anthropocentric. The visible region (380 nm780 nm) denes the approximate region of normal human vision and excludes, for example, cat vision in the infrared and insect vision in the UV. In fact, the term ultraviolet is itself vague and replaced a hodgepodge of earlier terms such as abiotic rays, chemical rays, and actinic rays.4 The wavelength bounds for the UV region are as approximate as those for any other region, but the term has proven useful. In a similar manner, the driving force for fractionating the biologically active UV was the desire of those working with human skin to bracket regions based on erythemal (sunburn) response. Although useful to dermatologists, these regions dened as (UV-A, UV-B, and UV-C) were for many years ignored by photobiologists. Thus, authors routinely used the terms near UV (near the visible) and far UV (far from the visible), or various other denitions, to attempt to identify the region of UV they were utilizing. The conventions listed in Table 4.1 are not unique to the author. Rather, they are common in the photobiological literature. They begin at the short wavelength () end of the UV.

In some cases, the photon is dened by the method of production, not merely its , e.g., -rays and x-rays.

Uses and Effects of Ultraviolet Radiation on Cells and Tissues

87

TABLE 4.1
Name

Common Fractionation of the Ultraviolet Region

Bounds in nm 290380 320380 290320 190290 10190 Comments all approximate Human vision occurs above 380 nm. 290 nm is shortest solar measurable at sea level. Ozone relativity transparent above 320 nm. Dominated by ozone absorption of solar UV. Oxygen (air) and water transparent above 190 nm. These are absent in terrestrial solar UV. Mostly ionizing photons. Absorbed heavily by water and air.

Terrestrial Solar UV UV-A UV-B UV-C Vacuum UV

Note: The denitions of the terms UV-A, UV-B, and UV-C vary with author. Those given above are commonly used by photobiologists. Other denitions vary by no more than about 10 nm for any given bound. See text for details.

4.2.1 Vacuum UV (10 nm190 nm)

There is no general agreement as to where the UV ends and x-rays begin. Most textbooks show the two regions overlapping. Hence, authors have cited as short as 1 nm as UV. Some photobiologists limit the UV to above 100 nm. A reasonable compromise is 10 nm. Photons in the vacuum UV (VUV) are heavily absorbed by water and oxygen (in air), both of which become essentially transparent (more than 50% transmission for a one cm path length)5 to UV at above 190 nm. Because of this limited penetration, VUV damage to cells is usually conned to a narrow region (a few microns) near the cell surface. The energies of single photons in this region are above 6.5 eV, sufcient to ionize many biomolecules.6 Because the biological effects due to ionizing photons are different from those due to non-ionizing photons, the vacuum UV often causes different types of cellular and molecular damage. Thus, vacuum UV effects can be qualitatively different from those of other UV regions. Frequently it is convenient to generally regard below 190 nm as ionizing and above 190 nm as non-ionizing, although the energy needed to reach the ionization band of molecules varies widely (see below).

4.2.2 UV-C (190 nm290 nm)

The shorter end of the UV-C (190 nm) is the region, where air and water become transparent. The longer limit (290 nm) is the shortest solar UV easily measured at the earths surface. Thus, by this convention, all of the UV-C is environmentally irrelevant. However, research in the UV-C range was central in elucidating many important features of cell functioning. DNA, the genetic material, has a peak absorption near 260 nm, which falls by a factor of six by 290 nm. This fact, combined with the readily available 254 nm germicidal UV uorescent source, allowed for simple experimental molecular manipulation of DNA, and UV-C photobiology was central in establishing the then-new eld of molecular biology. Accordingly, a considerable body of work that unraveled cellular function was accomplished in this environmentally irrelevant region.7

4.2.3 UV-B (290 nm320 nm)

The short limit (290 nm) of the UV-B can be dened as the shortest UV routinely measurable at the earths surface, where it is about one million-fold less prevalent than 320 nm radiation. The long limit (320 nm) is where ozone and other atmospheric components begin to appreciably (more than 50%) attenuate sunlight. This absorption prevents much of the signicant DNA damage that would result if no ozone layer existed. It is thought by some that the limited absorption of DNA in the UV-B (compared with the UV-C) is, in part, due to the evolution of life under the protective ozone screen. DNA absorption rapidly decreases toward the longer end of this region.8 Nevertheless, absorption of UV-B photons by DNA contributes to a wide variety of bioeffects. Stratospheric ozone depletion will cause an increase in the amount of UV reaching the biosphere and this increase will be fully contained within the UV-B. The UV-B is a transition region from the heavily damaging UV-C to the less damaging UV-A. The product

88

Lasers in Medicine

of solar UV exposure, and, for example, DNA absorption, peaks at about 300 nm.9 Therefore, the UVB is responsible for most of the damage inicted on organisms by sunlight. This UV-B denition also closely brackets the effective solar UV absorption properties of the ozone layer.

4.2.4 UV-A (320 nm380 nm)

The shorter limit of the UV-A (320 nm) is dened here as that at which ozone becomes transparent. The longer wavelength limit (380 nm) is where human vision begins and the UV ends. The UV-A is also biologically effective and causes cellular death, mutation, and DNA damage. However, the primary chromophores for these effects may be non-DNA sensitizers that are chemically matched to the photon energy and act as intermediates in relaying the absorbed energy to DNA.10 It is also the region where some photoreactivation can occur and is heavily absorbed by certain biopigments. Thus, the UV-A, which is still not clearly understood, is proving to be a very complex area of photobiology. Some UV-A effects, such as cellular mutation (which should ultimately be caused by effects in DNA) occur at UV-A exposures far less than those thought to affect DNA. The absorption spectrum of DNA in the UV-A, though low, is not clearly dened.8 Some UV-A effects seem to mimic those due to ionizing radiation although UVA photons do not ionize molecules. UV-A exposures from sunlight are considerable, constituting about 8% of the total sunlight spectrum (compared with less that 0.3% for the UV-B).

4.2.5 Terrestrial Solar UV (290 nm380 nm)

Terrestrial solar UV is easily dened as the bracket that is the limit of human vision (380 nm) at the long end, and the effective limit of the solar UV reaching the earths surface at the short (290 nm) end. This is the environmentally relevant UV region and is, of course, as dened here, just the UV-B plus the UV-A. No clear fractionation of the biologically effective UV is ever as precise, or as desirable, as a detailed description of the UV source, its output (both in distribution and intensity), and its measurement, which should be clearly stated by experimenters. For example, there are no pure UV-B sources (unless one considers monochromatic sources), so it is meaningless to describe a uence of 10 jm2 of UV-B. But, although the term UV-B cannot replace a detailed description, it is useful in dening the region of UV being utilized.

4.3 UV Sources
A variety of UV sources exist for use in biomedical applications. As shown in Table 4.2, the selection of an appropriate source for UV photobiological studies is often determined by availability and cost. Simple polychromatic sources are useful for studying bioeffects; monochromators for action spectroscopy; and synchrotrons and lasers for more sophisticated molecular effects and time-resolved spectroscopy. Polychromatic sources include the sun, solar simulators, and uorescent lamps.
TABLE 4.2 Comparison of UV Sources
Spectral Purity High Medium High Highest Intensity Medium Medium Medium Medium Medium* Highest Cost Low Low Medium Medium Highest High

Source Germicidal lamp Polychromatic uorescent Solar simulator Monochromator Synchrontron Laser

* Medium intensity in the range above 190 nm; relatively high below 190 nm.

Uses and Effects of Ultraviolet Radiation on Cells and Tissues

89

The advantage of solar radiation is that it is a normal component of the ambient radiation beneath which humans evolved. The curative effects of solar exposure are ancient in use and include the taking of plant (Ammi majus) extracts followed by solar exposure for the treatment of vitiligo. Detrimental effects of solar UV will be discussed in a following section. However, the sun is an awkward source for use in laboratory or therapeutic regimes. Its intensity and wavelength distribution vary in relation to position on the earths surface, time of day, season, atmospheric conditions, etc. Of more direct use is a solar simulator that lters the output from a xenon arc so that it more closely matches the terrestrial solar output. It is often used in biomedical research. The simplest polychromatic sources are uorescent bulbs. They contain Hg gas atoms which, when ionized by accelerated electrons, in turn excite other Hg atoms that emit photons according to the energy level differences of the Hg atom, especially the strong resonance output at 254 nm. A special phosphor coating on the inside of the bulb envelope converts these UV-C photons to UV-B, UV-A, and visible radiation. Some bulbs are specic for each of these regions. The glass envelope attenuates most of the UV-C. Certain uorescent sunlight bulbs are also manufactured, but a careful check of their spectral output is necessary to determine their properties. Even small amounts of short UV, less than 300 nm, can markedly affect the biological effects of these bulbs. Mylar ltration is advised. Incandescent halogen lamps produce large amounts of UV. Monochromatic sources also vary in intensity, spectral purity, wavelength range, cost, and ease of operation. A simple germicidal lamp acts the same as a uorescent bulb. It has no phosphor, but has a UV transparent envelope. It emits 86% of its radiation at 254 nm.5 It ts into simple uorescent sockets, such as a desk lamp, and is used in sterilization. The nearness of the 254 nm to the peak absorption of DNA (260 nm) makes this an ideal source for many photobiological experiments. More-intense sources can be fractionated into monochromatic regions, each with sufcient intensity to cause a biological effect. Mercury and xenon arcs are commonly used, often with the Hg-Xe mixed in combination, and have proven useful for the wavelength region 220 nm380 nm. Synchrotrons can produce high intensity UV radiation that is also tunable, but one has to travel to the few available sites to use them. But, for the widest potential application, nothing compares with the laser. Its spectral purity, collimation, power, and reasonable availability make it an ideal source for many uses. If only a single wavelength is needed (e.g., in matching the incident radiation to the peak absorption of a molecule) then a simple laser is sufcient. If a variety of are required, a tunable laser is necessary. The limited availability of inexpensive, tunable sources in the UV-C and UV-B has hampered the widespread use of lasers in UV photobiology. Some action spectra (see below) have already have been reported using tunable lasers. Recent advances in the area of UV and VUV lasers are covered in this book. It is especially important that one choose the correct UV source for the needed purpose, is aware of the absorbing and scattering effects of the materials (including gases and solutions) between the source and the sample, and measures the intensity and distribution as close to the target sample as possible. Table 4.3 lists the at which 50% and 90% of an incident beam of UV will be absorbed by some common
TABLE 4.3 Ultraviolet Absorption by Common Materials
for absorption of Thickness Ozone (0 C) Air (O2 at 0 C) Water Quartz (UV fused) Window glass Pyrex glass 1 cm 1 cm 1 cm 1 mm 1 cm 1 cm 50% 330 181 189 164 340 325 90% 314 174 186 160 332 306

Note: There are several different types of quartz and glass; some aqueous buffers will have higher UV absorption due to dissolved components. Data from a variety of sources, especially Jagger, 1967.

90

Lasers in Medicine

TABLE 4.4

Approximate Photon Flux (Photons cm-2 sec-1) for Various Radiation Sources*
100 nm 4 1014 7 1018 200 nm 2 10 4 1014 3 1019 1015 at 254 nm
13

Source Double monochromator Tunable dye laser Synchrotron RF-Linac Free electron laser Germicidal lamp Polychromatic uorescent Solar simulator Sunlight (noon)

300 nm 2 10 2 1016 4 1014 5 1019 2 1014 from 300400nm 2 1016 from 290400nm 1016 from 290400nm
15

400 nm 1 1015 1.5 1016 2 1014 1 1020

* Sources vary greatly in bandwidth and spectral purity. For example, the spectral band width for the synchrotron is 1%, for both lasers less than 0.1%, and for the monochromator about 2%. Wattage varies also. These values are for a beam passed through a monochromator. A storage ring undulator may increase the synchrotron readings by two orders of magnitude.

substances. Note the protective absorption of ozone beginning at about 330nm, which shields life forms from detrimental UV exposure. Air and water do not absorb heavily at above about 190 nm. Quartz is useful for photobiological studies because it transmits well at above 170 nm. Window glass begins to absorb appreciably below 340 nm. All of these values change with thickness. No listings are given for various plastics, which can have different absorption properties, even from batch to batch. In addition, UV ages plastic rapidly, changing its absorption properties. It is especially important that one be aware of everything in the path between sample and UV source. It is also crucial that the source be carefully described in any publication as to type, output (Table 4.4), measured parameters, and experimental conguration. If possible, each laboratory should carefully measure the output of any source using a calibrated spectroradiometer.

4.4 Absorption of Ultraviolet

To a rst approximation, the absorption of non-ionizing radiation follows the Beer-Lambert law:

I nsx --- = e Io
where Io = incident intensity, I = nal (sometimes transmitted) intensity, x = absorber thickness, n = number of absorbers, and s = absorption cross section. Thus, absorption is, to the rst order, an exponential process. Hence, we also dene:

Io log --- = optical density or absorbance I

The rst law of photochemistry states that a photon must be absorbed to produce an effect. The quantum yield () is then dened as:

number of molecules photochemically altered = ------------------------------------------------------------------------------------------------------------number of molecules that absorb a photon

The range of in photochemistry is usually from about 106 up to the maximum value of 1.5

Uses and Effects of Ultraviolet Radiation on Cells and Tissues

91

4.4.1 Division Between Ionizing and Non-Ionizing UV

There is an important division in the UV for radiation effects. In general, photons of wavelength shorter than about 190 nm have enough energy to ionize atoms and molecules. But the ionization potential of atoms and molecules is too high for wavelengths longer than 190 nm to have this effect. This distinction between ionizing and non-ionizing UV is crucial, because the mechanisms of damage by each type of radiation are different. Ionizing radiation cuts an energy deposition path through cells and tissues that can cause a variety of indirect as well as direct effects, often by the formation of free radicals. Nonionizing radiation has to be absorbed to effect a change. Hence, a non-ionizing UV photon can traverse a cell, not be absorbed, and the cell will have no history of exposure to it. Because almost all biomedical uses of UV employ sources at wavelengths greater than 190 nm, only the non-ionizing portion of the UV need concern us here. This is not to imply that ionizing UV research (in the vacuum UV) is unimportant, or will not have biomedical applications in the future.2

4.4.2 Molecular Absorption

As mentioned above, the rst rule of photochemistry is that a photon has to be absorbed to produce an effect. Molecules absorb non-ionizing radiation by three fundamental mechanisms. As shown in Figure 4.1, electrons, normally in the ground electronic state, can absorb photons directly and go to an excited state that has an energy difference equal to the energy of the absorbed photon (E1 in Figure 4.1). In addition, there are energy levels contained within the electronic levels that allow for changes in the vibrational and rotational modes of molecules. Table 4.5 shows the energy requirements for photons to excite the various energy levels present in molecules. In general, electrons can be excited into the ionization band by in the VUV. The remaining UV and visible regions can excite electrons from the ground state to various electronic excited states. Infrared radiation can excite molecules into higher

FIGURE 4.1 An energy level diagram for a diatomic molecule. Solid curved lines are the envelopes for the electronic ground state and the rst excited electronic state. Solid horizontal lines represent vibrational energy levels superimposed on the electronic ones. Not shown are rotational energy levels that would be superimposed on the vibrational energy levels. Vertical lines; solid line represents the absorption by the molecule of a photon of energy E1 , dashed lines the emission of a photon by the molecule, wavy lines the internal loss of energy by collision. For details see text.

92

Lasers in Medicine

TABLE 4.5

Absorption Region for Excitation of Biomolecules

Photon Energy to Excite in eV 6 1.0 0.1 0.01 Approximate Spectral Region Vacuum UV UV-A-B-C, visible Infrared Microwave

Energy Level Ionization band Electronic Vibrational Rotational

Note: Real values depend on the molecular structure and vary accordingly. 1eV = 1.6 1019 J.

vibrational states; microwave radiation excites molecular rotational energy levels. As stated, these values are rough approximations, as the structure of molecules determines their individual absorption properties. Figure 4.1 shows a typical energy level diagram for molecular absorption. As shown, an incident photon of energy El, is absorbed by the molecule, raising a ground state electron to the rst excited state. Here it has a variety of ways to return to the ground state. It can do so directly by emitting a photon of energy equal to E1 ; it can lose some energy of vibration by collision with other molecules and then emit a lower energy photon E2; or it can emit an even lower energy photon E3; returning to a higher vibrational level in the ground state, eventually going to the lowest vibrational energy level by collision. The last two mechanisms mean that a photon of longer than was absorbed is emitted. In general, the length of the conjugated double bond region of a molecule determines the long limit of absorption.

4.4.3 Absorption by Cells

Beginning at the cellular level, we can see from Table 4.6 that 200 nm UV-C is absorbed only slightly by viruses (about 30%), more so by bacterial cells (about 70%), and entirely by mammalian cells (more than 99%). This absorption is mainly due to the presence of endogenous molecules that absorb heavily in the UV-C. Figure 4.2 compares the absorption of two important biomolecules, DNA, the genetic material, and protein. Both molecules begin to absorb appreciably in the UV-B and substantially in the UV-C. On a weight-by-weight comparison, DNA absorbs about 20-fold as much 260 nm radiation as does protein. Therefore, the absorption of even homogenates of mammalian cells (Figure 4.3) is similar in the UV-C, in wavelength dependence, to that of DNA. The absorbance (a logarithmic function) of such homogenates is seven times higher at 240 nm than at 300 nm.11 As the UV wavelength decreases below 220 nm, the large number of peptide bonds present in proteins begin to contribute greatly to the total cellular absorption. At wavelengths below 190 nm, water and oxygen absorption prevails. In addition to these macromolecules, some cells contain certain pigments that can alter cellular absorption signicantly, even at the longest UV wavelengths (Figure 4.4). Also, cellular particles and organelles can absorb
TABLE 4.6 Estimate of the Percent Transmission to the Center of Selected Cells and Viruses in the Ultraviolet
Wavelength in nm Biological sample Bacteriophage (T2) Herpes simplex virus Bacterial cell Yeast cell Mammalian cell (umbonate) Mammalian cell (spherical) Mammalian tissue (100 thick) Diameter () 0.1 0.15 1 5 4.6 20 200 74 66 33 1.6 102 107 250 86 80 78 69 50 20 105 300 100 100 98 97 95 91 39 350 100 100 100 100 99 96 66

Note: All values are approximate; values at above 300 nm can vary widely due to the presence of endogenous chromophores; umbonate is the attened geometry of mammalian cells when attached to surfaces. See Coohill (1986) and Coohill and Sutherland (1989) for more complete discussion.

Uses and Effects of Ultraviolet Radiation on Cells and Tissues

93

FIGURE 4.2 The ultraviolet absorption of DNA and a typical protein. At short , the absorption of the numerous peptide bonds in the protein predominate.

FIGURE 4.3

Absorption spectra for a homogenate of monkey kidney cells (CV-1) and human HeLa cells.

94

Lasers in Medicine

FIGURE 4.4 Absorption spectra for four important molecules in human skin. Data is averaged from several authors, including Parrish et al.12 and Everett.48

and scatter UV radiation and, at some wavelengths, shield the center of the cell from a signicant portion of the incident beam. Such cytoplasmic screening can alter measured bioeffects to a large degree and must be considered when cellular exposure is attempted.11 Due to the nature of living material, it is often difcult, or impossible, to place the UV dosimeter at the site selected to be irradiated (e.g., the nucleus of a cell). In such cases, it is important to measure the intensity of the incident beam at the surface of the sample and on the exit side of the sample. Only then can a reasonable estimate of the exposure of the target area to the beam be derived. In the case of living tissue, it is often impossible to place a dosimeter on the exit side, so estimates must be made from separate measurements of the optical properties of the tissue being exposed. Such concerns are important because biological cells and tissues will absorb radiation in a wavelength-dependent manner and alter the exposure of the target accordingly. Figure 4.5 shows the penetration depth of various UV-C into mammalian cells. The horizontal lines show the level at which approximately 50% of a beam at that will be absorbed. Cells in a spherical conguration (e.g., in solution) absorb more heavily because the beam has to traverse more protoplasm. Cells in the

FIGURE 4.5 The approximate penetration depth into a single mammalian cell for 50% absorption of an ultraviolet beam at different . Spherical cell (often the shape in solution), and attened umbonate cell (often the shape in tissue or cell culture). For example, 50% of a beam of 280 nm radiation makes it to the center of a spherical cell, while more than 50% at that traverses an umbonate cell.

Uses and Effects of Ultraviolet Radiation on Cells and Tissues

95

attened (umbonate) conguration (e.g., in some tissues or in cell culture) allow more than 50% of a beam of 280 nm to completely traverse the cell. The absorption properties of cells are not merely a summation of the individual absorption of their component molecules. Scattering of incident radiation by cellular structures (granules, mitochondria, membranes, etc.) and the shielding of possible target chromophores by pigments and other molecules and structures add to the complexity of predicting the absorption properties of any cell. However, reasonable estimates can be made based largely on experimental measurement of cellular absorption (when available) and the known absorption properties of cellular constituents.

4.4.4 Absorption by Tissue

It is surprising how many people think that tissue (e.g., skin) is opaque to light. Any children who have gone into a dark closet, placed a lit ashlight into their mouths, and looked into a mirror, have seen the red end of the visible spectrum emerging from their cheeks. Of more concern is the statement by some scientists that UV does not penetrate into the human body. Of course UV, in a very wavelengthdependent manner, does penetrate into certain organs such as the skin and eye. This penetration is limited but, in turn, limits the uses of UV sources for biomedical purposes. In fact, tissue optics is a reasonably well developed eld, and long wavelength UV can penetrate at least 1 mm into the skin, i.e., traversing the stratum corneum and epidermis, and going well into the dermis. Although this penetration is highly dependent on skin type and pigmentation, about 40% of UV-A reaches the dermis in white skin, and about 10% in black skin.12 This means that a signicant amount of UV-A reaches blood vessels where it can be absorbed by blood and other moieties. Cells circulating in the blood system can also be affected.12 Much of the direct absorption of UV by the skin can be accounted for by the presence of endogenous pigments, especially melanin, hemoglobin, carotenes, and keratin (Figure 4.4), and exogenous pigments and drugs. Because skin is such a heterogeneous material, UV can be reected, absorbed, scattered, and re-scattered. This means that the direct component of the UV beam is augmented by additions from photons scattered and reected back into the beam pathway. Hence, at any tissue depth, the sum of the total UV exposure is just the direct plus the diffuse. Table 4.7 is a very approximate estimate of the percentage of UV at various wavelengths that will traverse the human epidermis. These values will vary widely among various skin types, and can change as an individual adapts to UV exposure. Dry skin will allow more penetration than wet, because water will reect a potion of the incident beam away from the skin. As shown, penetration of greater than 10% of the incident beam occurs for above 300 nm. Below 300 nm, absorption rapidly increases, making the skin essentially opaque to below about 290 nm. An excellent series of charts showing the penetration of different of UV into human skin can be found in Parrish, pp. 7475.12
TABLE 4.7 Rough Estimates of the Percent of UV Transmitted through Intact Human Epidermis
380 350 325 300 290 275 270240 % Transmission 58 41 36 11 2 0.1 <1.0

Note: Can be highly variable; data for normal Caucasian skin; averaged from a variety of sources, expecially Anderson and Parrish (1981) and Everett et al. (1966).

96

Lasers in Medicine

4.5 Direct vs. Indirect Effects of UV

Direct effects of UV exposure imply that the chromophore absorbing the radiation is changed so as to cause a bioresponse. This is often the case for UV exposure at between 190 nm and about 320 nm. Here, the effects of photon absorption by important molecules like DNA are largely known and include such photochemical changes as the formation of pyrimidine dimers (Figure 4.6), 64 photoproducts, DNA-protein cross links, and lesions that can lead to single and double strand breaks.1315 If left unrepaired, these lesions may lead to impairment, mutation, and even cell death. The biochemical and physiological consequences of UV exposure of some biological systems are partially characterized and, to some extent, understood16,17 when compared with other insults, e.g. ionizing radiation.18 However, in the UV-A, it is not uncommon to observe events such as sensitization due to photodynamic effects involving reactive oxygen species such as singlet oxygen and superoxide anion. There is also evidence for the generation of H2O2 by UV-A and a consequent production of OH radicals. In some ways, these UVA effects mimic the results of exposure to ionizing radiation.

FIGURE 4.6 One example of a common cellular photoproduct after UV radiation, the pyrimidine, here thymine, dimer. These dimers are formed by UV, and can be reversed by shorter UV, longer UV in the presence of the photoreactivation enzyme (PHR), or by excision repair (ER) in the absence of light. See text for details.

4.6 Action Spectroscopy: Effect as a Function of

4.6.1 UV-C (190 nm 290 nm) Action Spectra
Because some cellular molecules absorb heavily in the UV, certain bioresponses are highly dependent on the to which the cell is exposed. Indeed, it is sometimes possible to determine the type of molecule responsible for a given effect by comparing the absorption spectrum of the molecule to a spectrum derived by measuring the cellular response as a function of . The latter is called an action spectrum (AS), and was the rst method of analysis to point to chlorophyll as the major chromophore in photosynthesis. Action spectroscopy has played a central role in the elucidation of the mechanisms of cellular responses to UV radiation, especially the UV-C,5 and, as such, has contributed to our understanding of the molecular biology of cells. In 1930, Gates (Figure 4.7) reported19 that the AS for bacterial cell death closely followed the absorption of nucleic acid, not protein, which was then widely believed to be the genetic material. In retrospect, this was the rst clear evidence that DNA was the genetic material. Although AS utilizing small cells is somewhat simple, it is difcult to extend these studies to larger (e.g., mammalian cells) because of the substantial absorption of UV by large cells and tissues (Table 4.6). Figure 4.7 also shows two other AS in the UV-C for cell death, one for mammalian tissue20 and one for individual mammalian cells.21 Early attempts by Mayer and Schreiber (1934)20 failed to produce AS with the ne structure of those of Gates (1930)19 because they employed hanging-drop mammalian tissue samples that were essentially opaque to radiation near the peak of DNA absorption (260 nm). The target molecule (DNA) was shielded to an extent such that the shape of the AS did not parallel the absorption

Uses and Effects of Ultraviolet Radiation on Cells and Tissues

97

FIGURE 4.7 The absorption spectrum for DNA, and action spectra for: bacterial cell death,19 cultured mammalian cell death,21 and cell death in mammalian tissue.20 See text for details.

spectrum of the target. The advent of single cell mammalian culture techniques and the unique attened geometry that mammalian cells assume when in monolayer culture,11 allowed these studies to begin. Thus, AS for killing of cultured mammalian cells21 reported data similar to those of Gates19 with bacteria, but the peak was shifted to about 270 nm. This discrepancy can be accounted for if one looks at the absorption properties of single mammalian cells and considers that, as is the case in bacteria, the likely target molecule for cell killing is also DNA, which resides in the nucleus. This means that the UV beam has to traverse, on average, half of the cell to strike its target. In bacteria, this distance is small enough to allow one to neglect absorption effects; in mammalian cells the absorption is substantial and dependent (Figure 4.3). However, because the target and primary photoproduct for this effect was already known from work with bacteria, an AS for the production of pyrimidine dimers caused by UV exposure of that target was measured. Thus, the effects of cytoplasmic shielding were accounted for in the experimental results themselves. Accordingly, measurements of the AS for pyrimidine dimer production matched the action spectrum for cell death in mammalian cells (Figure 4.8). Therefore, DNA alone was responsible for mammalian cell lethality by UV-C.22

4.6.2. UV-A (320 nm 380 nm) Action Spectra

It was widely thought that experimental work in the UV-C and UV-B ( range 190 nm320 nm) could be extrapolated to predict photobiological responses in the UV-A (320 nm380 nm). This is not the case, and UV-A ASs are much more complex than had been predicted.10,17,23 Here, even events such as cell mutation, which surely involves the genetic material, can be affected at uences that appear to be below those necessary to affect DNA in these regions. Figure 4.9 details the divergence of several ASs for human cell photoresponses from the DNA absorption spectrum as the curves shift to longer in the UV-A. In the UV-C, all the measured effects, cell killing, mutagenesis, DNA breaks and cross-links, follow the absorption

98

Lasers in Medicine

FIGURE 4.8 The absorption spectrum for DNA and the averaged survival action spectrum for a large variety of mammalian cell lines. The points are the wavelength dependence of pyrimidine dimer formation by exposure of whole cells to UV. Note that these points closely follow the cell survival curve.7

spectrum of DNA. In the UV-B, breaks and cross-links are made at a rate above one that can be explained by DNA absorption alone. In the UV-A, all of the biological parameters are affected at levels that are not fully explainable by the absorption properties of DNA. The absolute absorption of moieties in the DNA molecule itself is difcult to measure at above 320 nm because of scattering and contamination by extraneous chromophores.8 It is thought that, if the absorption is not intrinsic to DNA, then an intermediate molecule may be involved that absorbs the incident UV-A photon and transfers the effect to DNA.10 The motivation for focusing on studies of the effects of UV-A and UV-B, the solar UV wavelength region, is provided by three observations: 1. Although many studies concerning the UV-C between 220 and 290 nm have been valuable in elucidating some of the mechanism of cellular function, these wavelengths are environmentally irrelevant. 2. Research has suggested that the nature of the primary and secondary chromophores, photoproducts and mechanisms for cellular response to solar UV appear, in some cases, to be very different from those elucidated for wavelengths shorter than about 300 nm. For example, in contrast to their marked sensitivity to UV-C, certain photosensitive human cell lines, such as Xeroderma pigmentosum cells (XP), exhibit the same sensitivity to UV-A as normal cells.2327 3. Solar UV impacts a large number of important bioresponses, such as skin cancer, plant growth, cellular survival and mutagenesis, etc. It is also the region where many UV laser sources are currently available. It is hoped that research in this area will ultimately reveal the nature of biological responses to a portion of the radiation present in a normal environmental setting. In addition, certain factors that may be involved in cell killing by UV-A, such as the disruption of the cellular cytoplasmic microtubule complex observed by Zamansky and Chou,28 are not evident at shorter .

Uses and Effects of Ultraviolet Radiation on Cells and Tissues

99

FIGURE 4.9 The UV-A, UV-B, UV-C action spectra for human cell killing, mutagenicity, DNA strand breaks, and DNA-protein cross-links. These are mostly from Ref.30, with the exception of the AS for mutagenicity, which is averaged from several reports. Note that all of these spectra diverge from the absorption spectrum of DNA in the UV-A.

Also, glutathione seems to protect cells from the deleterious effects of UV-A, suggesting that at least some UV-A damage is mediated via radical formation.29 The apparent lack of a close correlation between action spectra for biological end points and UV-A-induced DNA damages (SSB, DSB, and DNA-to protein cross-links) in human cells (Figure 4.9) can be explained in part by repair (see below) of these lesions.3034 Two studies looked into the repair of SB induced by sunlamp irradiation. Roza et al.27 reported the complete repair of all SSB within 60 mm of repair time (similar to the repair kinetics of x-ray-induced SSB) in primary human broblasts and concluded that the lesions are irrelevant for cell lethality. Holmberg et al.35 found biphasic SSB repair in primary human lymphocytes exposed to UV-A a fast (1 hour) and a slow (several hours) component, not observed by Roza et al.27 What is best known about the effect of solar UV on living systems is that the nature of the responses is complex. At present, it is not possible to document accurately the events that occur between photon absorption and biological response. It appears that a large and varied number of chromophores are involved, some of which may act only as intermediates. The nature of the nal photoproducts is also unclear, as is their role in such important events as cell death and mutation. How solar photodamage is processed or repaired is almost unknown. That both UV-B and UV-A can cause cell death and mutation is well established, but the mechanisms involved, especially in the UV-A region, are still obscure. Initial results that possibly link certain endogenous chromophores with some cellular responses are promising. Little is known about the synergistic or antagonistic effects of polychromatic exposure, especially mixtures of UV-B and UVA, which have such different properties, and essentially nothing is sure about such interactions for the full spectrum of natural solar radiation. There is strong evidence that the mechanism of killing is not the same for these different regions and that the initial damage may reside in a chromophore other than DNA. That is, the lack of agreement between those bioeffects in the UV-A and DNA absorption also seems to show that the primary chromophore(s) for mutagenesis is a non-DNA sensitizer, with

100

Lasers in Medicine

energy levels that match the incident photon energy, which acts as an intermediate in relaying the energy to DNA. The identity of such chromophores has eluded all efforts to nd them thus far.

4.7 Effects of UV on Cells

A considerable body of work that measures some of the bioresponses of cells to UV has been accomplished. Several major areas of effect will be noted.

4.7.1 Cell Death

The most frequently measured cellular response to UV is cell death, usually assayed by measuring the ability of the cell to form a multicellular colony. A large number of studies have shown that cell killing is mainly a response to DNA damage. For example, pyrimidine molecules can absorb UV photons, become excited, and interact to form dimers (Figure 4.6). These dimers are not recognized by the reproductive apparatus of the cell and can prevent cell proliferation. Other DNA photoproducts are also formed, including pyrimidine (6-4) pyrimidone, especially in the UV-A. It is also possible to produce breaks in the DNA molecule backbone, either by direct or indirect effects sometimes involving the repair of the photoproduct (see below). DNA can also link co-valently to protein molecules. These cross-links are very stable and detrimental to the cell. Cell survival after exposure to UV depends on a wide variety of conditions. Rapidly dividing cells are usually more sensitive than quiescent cells. Pigments shield the DNA of some cells to UV exposure. Cell geometry is important and can change during the cell cycle and under different irradiation conditions (e.g., cells adhered to plastic petri dishes or grown in solution).11 A possible factor in cell killing by solar UV is the disruption of the cellular cytoplasmic microtubule complex (cytoskeleton), which is important for normal cell growth.28 Micro-tubule dissembly was observed after treatment with UV-A (but not UVB); this process could be the causative factor or one of many contributing factors in cell death. Microenvironmental conditions, including effects on non-UV-transparent buffers used in experiments and the complex array of intercellular moieties in living tissue, must also be considered. A comparison of the actual exposures required to kill various cell lines shows marked variations. For example, normal mammalian cell lines exposed to broadband UV-A can differ in sensitivity by an order of magnitude.2630 The most probable cause of these differences, if rigorous spectral purity and comparable dosimetry are adhered to, could be that each cell line may harbor different kinds or amounts of endogenous photosensitizers. Cells lacking efcient repair systems can be more sensitive than normal cells. Depletion of endogenous glutathione markedly sensitizes human skin broblasts to killing by both UV-B and UV-A, but not UV-C.29 The degree of glutathione protection approaches that derived from excision repair (below). Glutathione protection conrms the indirect nature of UV-A lethality, suggesting that a major fraction of UV-A damage, and also (surprisingly) UV-B damage, may be caused by radical effects in photosensitized reactions of photodynamic action.

4.7.2 Cell Mutation

Anytime the genetic apparatus, especially the DNA molecule itself, is damaged, there is the possibility of a mutations occurring. All UV wavelengths are mutagenic if a high enough exposure at that wavelength reaches the DNA or if other chromophores absorb the photon and transmit the damage to the DNA molecule. As is the case in cell killing, substantially higher exposures are required in the UV-B, and even higher in the UV-A, to affect cells to a degree comparable to the effects derived from exposure to UV-C (Figure 4.9). As is the case with cell death, different human cell lines may require different exposures to any given UV wavelength to sustain similar levels of mutation This is especially true for human cells derived from patients with certain photosensitive diseases such as XP, which results from a deciency in excision repair (below). These different responses do not necessarily extend over the whole UV range. For example, Keyse et al.23 showed that a line of XP cells hypersensitive to UV-C was no more sensitive

Uses and Effects of Ultraviolet Radiation on Cells and Tissues

101

to UV-A than were normal human cells. It is also speculative to extrapolate to human cells the mutation rates determined with other cell types, e.g., in general, rodent cells are more sensitive to mutation by UV, possibly because many rodents are nocturnal and are exposed to little, if any, UV. Human cells are more procient than frog cells in repairing solar UV lesions. Marsupial cells seem to have a sensitivity between that of rodent and human cells. There is also one report that the wavelength 334 nm was nonmutagenic in human epithelial cells.30 These authors carefully raised the exposure as high as 80 KJm2, an exposure that was mutagenic at 365 nm, and still could detect no level of mutation. Other cell lines are mutagenic at 334 nm. These and other results with rodent and human cell lines show that it is not possible to accurately predict the degree of mutation for any wavelength in the UV region. Lack of correspondence with the spectrum of DNA in the UV-A (Figure 4.9) again suggests that the primary chromophore(s) for mutagenesis in that region is a nonDNA sensitizer. Spectral studies show that the yield of mutagenesis, for any given uence, varies with wavelength but not in a manner consistent with the involvement of any apparent specic primary chromophore. Further, different cell lines may harbor different primary chromophores. Once the critical damaging chromophores are identied, it might be easier to determine which photoproducts eventually lead to mutation at long UV wavelengths. Broadband polychromatic sources in the UV-B and UV-A regions are also capable of causing mutation, although it is difcult to know if such mutation induction is due to a small moiety of highly efcient short wavelength or a large moiety of longer wavelength radiation. In the study of Hitchins et al.,36 mutation was observed even though no radiation of less than 340 nm was present.

4.7.3 Membrane Effects

Membranes, especially if they contain molecules that act as dyes and absorb UV, are also sensitive to UV. Ion channels are affected, as is the respiratory system. In some cases, leakage of internal biomolecules has been reported. Of course, it is imperative that the integrity of the membrane be preserved. How much any of the above damages contribute to cell death is not presently known, but certain transport problems could, if not kill, at least hinder a cell from functioning properly.17 UV-A radiation, at equitoxic exposures, seems to be more effective in damaging membranes. For example, murine lymphoma cells lysed, and more than half of the remaining cells showed membrane damage if exposed to UV-A levels that killed 90% of the cells.37 Similar results were obtained with sheep red blood cells. Visible and infrared radiation had no lytic effects. However, human lymphoma cells showed membrane damage but no lysis at similar exposures. Whether any of this relates to UV-A cytoskeleton damage28 is not yet clear. But affecting membrane function and continuity by UV exposure can serve both research and therapeutic purposes. As shown in Figure 4.5, short UV wavelengths (e.g., the ArF laser output at 193 nm) have such limited penetration into a single cell that only the membrane and areas close to it are exposed. Cells damaged by membrane effects in this region should not be susceptible to mutation, as the genetic material is not exposed. Cells damaged near the peak of DNA absorption (260 nm) would be vulnerable to genetic change. At longer wavelengths, the situation is more cloudy. All UV wavelengths are mutagenic (even 193 nm if naked DNA is exposed). The UV-B and UVA regions are useful experimental tools because they penetrate deeper into tissue. How a mutagenic risk is estimated, however, depends on a large number of parameters. Low power level laser radiation, i.e., levels below which thermal damage predominates, may be useful therapeutically but could be mutagenic to the surviving cells or disrupt cellular membranes.

4.8

Complex Responses of Tissues

4.8.1 Immune Effects

Perhaps the most comprehensive effect of UV on animals (including humans) are on the immune response. Exposure to UV can compromise certain functions of the immune system, lowering an animals ability to combat invasion by foreign substances and leading to an increase in infectivity, the spreading

102

Lasers in Medicine

of disease, and the efcacy of vaccination against disease. Currently, effects on the immune system have been shown to be mainly due to exposure to UV-B. Whether UV-A exposure alone can elicit immune suppression is still somewhat in doubt. UV damage to the immune system may explain why certain viral infections, such as Herpes simplex, can recur in an individual who already has circulating antibodies against the virus. Solar UV exposure can cause herpes recurrences in individuals previously infected with the virus. The infectivity of the virus in individuals who do not harbor it in a latent form may also be increased by UV exposure. Somewhat similar results have been obtained with HIV viruses. The target chromophores for these immune effects are still questionable. However, one likely candidate is urocanic acid,38 which, after absorption of UV, may initiate a chain of molecular events that ultimately damage DNA. Figure 4.10 shows an action spectrum in mice for the immune response known as delayed contact hypersensitivity.39 This spectrum has been shown to follow the absorption spectrum of urocanic acid, which has a much higher extinction coefcient in the UV-B than does DNA. More recent results further implicate this compound as at least one of the chromophores in the immune systems response to UV. Those are detailed in this book. A recent study of human T-lymphocytes found them to be 20-fold more sensitive to UV-B radiation than are human broblast cells.40 More alarming, naked T-lymphocytes cells in culture could be killed by just a 1-minute exposure to sunlight. Even though little UV-B penetrates to capillaries in the skin, extracapillary lymphocytes could receive lethal exposures. How this relates to the role of UV in immune system suppression is still unknown. Sunlight can affect immune responses before it causes sunburn. It has also been shown that UV radiation can accelerate the death of mice challenged by a lethal infection of C. albicans.41 Here, the timing of the UV exposure was crucial, but the mechanism for the effect was unknown. It seems clear that these and other effects on the immune system are widespread and that the eld of photoimmunology42 is evolving rapidly. These effects may turn out to be the most damaging component of the human response to UV. Conversely, it may be possible in the future to control some adverse immune reactions in humans by careful exposure to certain of UV. Table 4.8 lists the predominant types of damage to cells by exposure to UV from different regions of the spectrum. It is important to remember that these damages reect the degree of involvement of each process in cellular response. For example, a UV-C exposure that will kill a cell may also cause some membrane damage, but the latter is a minor effect because the cell is killed by a different mechanism. Note that VUV and UV-A both damage membranes, the former due to the limited penetration of VUV into the cell (Figure 4.5), the latter to absorption of UV-A by components of the cell membrane. It

FIGURE 4.10 An action spectrum for photocarcinogenesis in mice, 43 and an AS for delayed contact hypersensitivity (an immune response) in mice.39

Uses and Effects of Ultraviolet Radiation on Cells and Tissues

103

TABLE 4.8
UV Region

Predominant Types of Damage to Cells and Tissues by UV Region

Major Damage Surface effects; membranes; ionizations. DNA altered; cell death; mutation; proteins; respiration; erthyma; carcinogenesis; immune effects. Transition region between UV-C and UV-A effects; solar UV-B causes cataract, erythema, carcinogenesis, immune supression. Various chromophores; indirect damage to DNA; repair systems; membranes; photoaging.

Vacuum UV, 10190 nm UV-C, 190290 nm UV-B, 290320 nm UV-A, 320380 nm

Note: All UV regions can cause cell death; the UV-A-B-C can also induce mutation, inhibition of respiration, and immune responses, the exposures required are highly dependent on .

should also be reafrmed that UV-A, UV-B, and UV-C are all mutagenic if enough exposure is given. Because of the absorption properties of pigments, lasers can be aimed at specic pigmented targets, e.g., melanin in melanocytes, hemoglobin in red blood cells, and exogenous drugs in photodynamic treatment. For example, by laser exposure, it is possible to disrupt melanosomes within a melanocyte without disrupting the cell.

4.8.2 Carcinogenesis
All known mutagens are potential carcinogens. The role of solar UV in human skin cancer has long been established.4 Squamous and basal cell carcinomas are believed to be the result of chronic UV exposure. The expression of these cancers is highly dependent on skin type and individual predisposition. Figure 4.10 also shows the AS for carcinogenesis and contact sensitivity in mice. The carcinogenic data is mainly from the analysis of De Gruijl and van der Leun43 and incorporates 12 separate studies. This AS shows major effects in the UV-C and UV-B, and falls off rapidly in the UV-A, with a minor peak at about 380 nm. However, due to the large preponderance of UV-A compared with UV-B in solar radiation (about a factor of 35), the contribution of UV-A to carcinogenesis is signicant. As shown, however, the major cause of UV-induced cancer is the UV-B component of sunlight. The extent of the role of UV in melanoma is still debatable. These skin cancers are often lethal if not detected early after onset, and appear to be, at least partially, the result of acute exposures to sunlight, especially in younger people. These sun-burning episodes may lead to the onset of melanoma several decades after the exposure. It needs to be pointed out that melanomas often appear on areas of the skin that are not usually exposed to UV. How, or even whether, these melanomas are associated with UV is unknown, but one possible mechanism may be damage to the immune system. The long delay in the response is not explained. Work on a UV AS for the production of melanoma in sh shows that UV-A contributes a considerable amount of damage.

4.9 Repair
Complicating any analysis of UV effects on cells and tissues is the fact that cells have evolved efcient molecular mechanisms to repair at least a portion of the damage caused by UV radiation.44 Usually, but not in all cases (e.g., direct splitting of dimers by short UV photons), cells damaged by UV-C can be repaired by UV-A (or even visible) radiation if an enzyme called the photoreactivating enzyme is present. In the presence of UV-A, this enzyme splits the dimers caused by UV-C and UV-B radiation back into monomers (Figure 4.6). A current controversy is whether this occurs to any real extent in human skin.45,46 High UV-A exposures can damage the repair system of the cell. In addition to these photon-mediated repairs, a process known as excision repair can occur after UV exposure in the absence of any light. Again, the integrity of the DNA molecule may be restored by a complicated removal of the damage and resynthesis of an exact copy of the predamaged molecule. Other repair processes exist and the extent and efciency of each is still somewhat undetermined in human tissue. For a good recent review see Sage.47

104

Lasers in Medicine

4.10 General Effects

The situation for UV exposure of cells and tissues is complex. A variety of active, quiescent, and dead cells present themselves for irradiation. These cells may have a varying amount of pigmentation and hence be differentially shielded from UV. Certain cells, somewhat dependent on their activity, may be able to repair some of the damage caused by UV, in some cases using photons of other UV wavelengths. Cells from certain diseased patients (e.g., XP) lack sufcient quantities of these enzymes for effective repair. If the radiation reaches the blood, then blood-borne components, including cells, could absorb enough UV and also be damaged. The nal result of all this interplay may not be clinically observable over the short term or ever. It is not unreasonable to expect a variety of uses in the future for targeting certain cells or tissues to monochromatic laser exposure. For example, an oral or a local dose of a proper photosensitizer that absorbs long wavelength UV can be given. The laser is then positioned (either within or outside the body) to irradiate a well dened area of tissue (or even circulating cells), with the laser chosen matched to the peak absorption of the photosensitizer. In this manner, only the targeted tissue is exposed. Such photosensitizers can kill cells. If a drug being used for other purposes is also a photosensitizer, then a warning advising patients to limit their solar exposure is usually listed in the descriptive folder packaged with the drug sample. It is important to estimate the degree of penetration of the UV from the source into the targeted tissue. Table 4.7 shows a rough estimate of the extent of penetration of UV through human epidermis. Any such measurement is highly dependent on skin type and pigmentation and will even vary among individuals of the same skin type. The table shows data averaged from two sources.12,48 Table 4.9 is an abbreviated list of some of the uses and effects of UV in biomedicine. Erythema is a common skin response mainly due to UV-B exposure, whereas photoaging is mainly a UV-A phenomenon. The treatment of various skin disorders, such as psoriasis, with drugs and UV is now common worldwide. But the surgical applications of UV rely mainly on the availability of lasers. These applications range from the sculpting of deformed corneas by microablation with 193 nm laser radiation to the ablation of arterial plaque by 308 nm laser radiation. The benets and risks of each of these procedures are being studied, but these and other treatments hold promise for an even wider use of lasers in medical treatment.
TABLE 4.9 Some Examples of the Uses and Effects of UV in Biomedicine
Surgical use Glaucoma Angioplasty Corned sculpting Endoscopy Keratogomy Other Erythema Vit. D synthesis Photoaging Photoimmunology Photodynamic therapy

Treatment for Acne vulgaris Psoriasis Mycosis fungoides Eczema Vitiligo

4.11 A Perfect Laser for UV Photobiological Studies

The wishes and requirements for the perfect UV radiation source may well be different for the majority of the photobiological community, when compared with the needs of physicists and physicians. As listed in Table 4.2, in many cases a photobiologist would prefer a high total energy source and would not be satised with a source with high peak power but low total output power. Of course, those scientists studying molecular effects such as time-resolved spectroscopy, and those physicians interested in ablating tissue, would require high peak power. In fact, some uses of lasers in therapy and diagnosis will require pulsed sources in which the exposure can be more easily controlled in stepwise increments. It is important to observe whether such high power lasers cause unwanted multiphoton effects, which may be of use in the study of certain molecular phenomena, but the natural exposure of cells and tissues usually occurs

Uses and Effects of Ultraviolet Radiation on Cells and Tissues

105

in a continuous manner and depends on the availability of certain chromophores for absorption. Except for isolated studies, it would also be preferable to have a tunable source for action spectra generation, and for exposure of different cells and tissues, especially in combination with different photosensitizers. Although it is wishful thinking, the breadth of tunability should be as wide as possible over the UV region, say from 200380 nm. Laser spectral purity is often more stringent than needed for the irradiation of biomolecules and cells, and already exceeds the requirements for many photobiological uses, as biomolecules have broad absorption spectra and cannot resolve photons to more than a few nm. The laser should also be simple to operate. Ideally, such a photon box should easily plug into an outlet, have an onoff switch, a selection dial, and an intensity dial. It should require few skills to operate and never break down. Athough the availability of such a source is still a fantasy, probably only with lasers can it even be approached.

4.12 Useful Review Texts

Several good texts to introduce scientists to the various areas of photobiology are available. Two books by Jagger5,17 provide both an introduction to UV photobiology and an analysis of solar UV effects. Smiths Photobiology49 is a general introduction that includes a fuller spectral range. Douglas et al.50 is also a good general review. Three texts conne themselves mainly to UV-A effects,12,51,52 all incorporating many biochemical applications. The rapidly emerging eld of photoimmunology is summarized by Parrish et al.42 Molecular photodamage and repair of DNA is detailed in Hanawalt and Setlow44 and Simic et al.31 Because of renewed interest, partially due to the depletion of stratospheric ozone and the consequent expected rise in UV-B reaching the biosphere, and the increased use of UV in diagnostic and therapeutic procedures, many shorter monographs that address more limited areas of photobiology are appearing.53 Recent books and review articles on photobiology are cited in.54,55

References
1. Green, A.E.S., T. Sawada, and E.P. Shettle, The middle ultraviolet reaching the ground, J. Photochem. Photobiol. 19, 251-259 (1974). 2. Coohill, T.P., Virus-cell interactions as probes for vacuum-ultraviolet radiation damage and repair, Photochem. Photobiol. 44, 359-363 (1986). 3. F.W. Sears, M.W. Zemansky, H.D. Young, University Physics, Addison-Wesley, Reading, MA (1987). 4. Blum, H F., Carcinogenesis by Ultraviolet Light, Princeton University Press, Princeton (1959). 5. Jagger, J., Jntroduction to Research in Ultraviolet Photobiology, Prentice-Hall, Englewood Cliffs, NJ (1967). 6. Setlow, R.B., Ultraviolet wavelength-dependent effects on proteins and nucleic acids, Radiat. Res. (Sup. 2), 276-289 (1960). 7. Coohill,T.P., Action spectra again?, Photochem. Photobiol. 54, 859-870 (1990). 8. Sutherland, J.C. and K.P. Grifn, Absorption spectrum of DNA for wavelengths greater an 300 nm, Radiat. Res. 41, 399-409 (1981). 9. Coohill,T.P and J. C. Sutherland, Free-electron lasers in ultraviolet photobiology, Photochem. Photobiol. 6, 1079-1082 (1989). 10. Coohill,T.P., M.J. Peak and J.G. Peak, The effects of the ultraviolet wavelengths present in sunlight on human cells in vitro, Photochem. Photobiol. 46, 1043-1050 (1987). 11. Coohill,T.P., D.J. Knauer, and D.G. Fry, The wavelength dependence of changes in cell geometry on the sensitivity to ultraviolet radiation of mammalian cellular capacity, Photochem. Photobiol. 30, 565-572 (1979). 12. Parrish, J.A. et al., UV-A, Biological Effects of Ultraviolet Radiation with Human Responses to Longwave Ultraviolet, Plenum Press, New York (1978).

106

Lasers in Medicine

13. Peak, M.J. and J.G. Peak, DNA-to-protein crosslinks and backbone breaks caused by far- and nearultraviolet, and visible radiations in mammalian cells, in Mechanisms of DNA Damage and Repair: Implications for Carcinogenesis and Risk Assessment, M.G. Simic, L. Grossman, and A.C. Upton, Eds., Plenum Press, New York, 193-202 (1968). 14. Rosenstein, B.S. and D.L. Mitchell, Action spectra for the induction of pyrimidine photoproducts and cyclobutane pyrimidine dimers in normal human skin broblasts, Photochem. Photobiol. 45, 775-780 (1987). 15. Peak, M.J., J.G. Peak and B.A. Carnes, Induction of direct and indirect single-strand breaks in human DNA by far- and near-ultraviolet radiations: Action spectrum and mechanisms. Photochem. Photobiol. 45, 381-397 (1987). 16. Urbach, F., Man and ultraviolet radiation, in Human Exposure to Ultraviolet Radiation: Risks and Regulations, W.F. Passchier and B.F.M. Bosnajkovic, Eds., Excerpta Medica, Amsterdam (1987). 17. Jagger, J. Solar-UV Actions of Living Cells, Praeger, New York (1985). 18. Elkind, M.M. and G.F. Whitmore, Radiobiology of Cultured Mammalian Cells, Gordon and Breach, New York (1967). 19. Gates, F.L., A study of the bactericidal action of ultraviolet light, III. The asborption of ultraviolet light by bacteria, J. Gen. Physiol. 14, 31-42 (1930). 20. Mayer, E. and H. Schreiber, Die Wellenangabhangigkeit der ultraviolettwirkung auf gewebekulturen (Reinkulturen), Protoplasma 21, 34-61 (1934). 21. Todd, P., T.P. Coohill, and J. A. Mahoney, Responses of cultured Chinese hamster cells to ultraviolet light of different wavelengths, Radiat. Res. 35, 390-400 (1968). 22. Coohill, T.P., Action spectra for mammalian cells in vitro, in Topics in Photomedicine, Ed. K.C. Smith, Plenum, New York, 1-37 (1984). 23. Keyse, S.M., S.H. Moss, and K.J.G. Davies, Action spectra for inactivation of normal and Xeroderma pigmentosum human skin broblasts by ultraviolet radiations, Photochem. Photobiol. 37, 307-312 (1983). 24. Kantor, G.J., Effects of sunlight on mammalian cells, Photochem. Photobiol. 41, 741-746 (1985). 25. Gill, R.F. and T.P. Coohill, A comparison of mammalian cell sensitivity to either 254 nm or articial, solar-similated radiation, Photochem. Photobiol. 45, 264-271 (1987). 26. Zamansky, G.B., Varying sensitivity of human skin broblasts to polychromatic ultraviolet light, Mutat. Res. 160, 55-60 (1986). 27. Roza, L., G.P. van der Schans, and P.H.M. Lohman, The induction and repair of DNA damage and its inuence on cell death in primary human broblasts exposed to UV-A or UV-C radiation. Mutat. Res. 146, 89-98 (1985). 28. Zamansky, G.B. and I. N. Chou, Environmental wavelength of ultraviolet light-induced cytoplasmic damage, J. Invest. Dermatol. 89, 603-606 (1987). 29. Tyrell, R.M. and M. Pidoux, Endogenous glutathione protects human skin broblasts against the cytotoxic action of UV-B, UV-A and near visible radiations, Photochem. Photobiol. 44, 561-564 (1986). 30. Jones, C.A., E. Huberman, M.L. Cunningham, and M.J. Peak, Mutagenesis and cytotoxicity in human epithelial cells by far and near ultraviolet radiations: Action spectra, Radiat. Res. 110, 244254 (1987). 31. Simic, G., L. Grossman, and A.C. Upton, Eds., Mechanisms of DNA Damage and Repair: Implications for Carcinogenesis and Risk Assessment, Plenum, New York (1986). 32. Peak, J.G. and M.J. Peak, Ultraviolet light induces double strand breaks in DNA of cultured human P3 cells as measured by neutral lter elution, Photochem. Photobiol. 53, 387-393 (1990). 33. Krell, K. and E.D. Jacobson, Sunlight-induced mutagenesis and toxicity in L51178Y mouse cells: determination and comparison with other light sources, Envirn. Mutagen. 2, 389-394 (1980). 34. Churchill, M.E., J.G. Peak and M.J. Peak, Repair of near visible and blue light induced DNA single strand breaks by the CHO lines AA8 and EM9, Photochem. Photobiol. 54, 639-644 (1991).

Uses and Effects of Ultraviolet Radiation on Cells and Tissues

107

35. Holmberg, M. et al., The repair of strand breaks in human lymphocytes exposed to near UV radiation (UV-A) and far UV radiation (UV-C), Photochem. Photobiol. 41, 437-444 (1985). 36. Hitchins, V.M. et al., The cytotoxic and mutagenic effects of UV-A radiation on L5178Y mouse lymphoma cells, Photochem. Photobiol. 44, 53-57 (1986). 37. Godar, D.E. and J.Z. Beer, UV-A1 induced anuclear damage in mammalian cells, in Biological Responses to UV-A Radiation, Ed. F. Urbach, Valdenmar Publishing, Overland Park, Kansas (1992). 38. DeFabo, E.C. and F.P. Noonan, Urocanic acid: on its role in the regulation of UV-B-induced systemic immune suppression, in Effects of Changes in Stratospheric Ozone and Global Climate, Ed E.J.G. Titus, US EPA, Washington D.C. (1986). 39. DeFabo, E.C., D.C. Reilly, and F.P. Noonan, Mechanisms of UV-A effects on immune function: preliminary studies, in Biological Responses to UV-A Radiation, Ed. F. Urbach. Valdenmar Publishing, Overland Park, Kansas (1992). 40. Arlett, C.F. et al., Hypersensitivity of human lymphocytes to UV-B and solar radiation, Cancer Res. 531 609-614 (1993). 41. Denkins, Y.M. and M.L. Kripke, Effect of UV irradiation on lethal infection of mice with Candida albicans, Photochem. Photobiol. 57, 266-271 (1993). 42. Parrish, J.A. M.L. Kripke, and W.L. Morison, in Photoimmunology, Plenum Press, N.Y. (1983). 43. de Gruijl, F.R. and J. van der Leun, Action spectra for photocarcinogenesis, in Biological Responses to UV-A Radiation, Ed. F. Urbach, Valdenmar Publishing, Overland Park Kansas (1992). 44. Hanawalt, P.C. and R.B. Setlow, Molecular Mechanisms for Repair of DNA, Parts A and B, Plenum, New York (1974). 45. Ley, R., Photoreactivation in tissues, in Biological Responses to UV-A Radiation, Ed. F. Urbach, Valdenmar Publishing, Overland Park, KS (1992). 46. Sutherland, B.M. et al., Pyrimidine dimer formation by UV-A radiation: implications for photoreactivation, in Biological Responses to UV-A Radiation, Ed. F. Urbach, Valdenmar Publishing, Overland Park, KS (1992). 47. Sage, E., Distribution and repair of photolesions in DNA: genetic consequences and the role of sequence content, Photochem. Photobiol. 57, 163-174 (1993). 48. Everett, M.A. et al., Penetration of epidermis by ultraviolet rays, Photochem. Photobiol. 5, 533-542 (1966). 49. Smith, K.C., The Science of Photobiology, Plenum, New York (1977). 50. Douglas R.H., J. Moan, and G. Ronto, Light in Biology and Medicine, Plenum, New York (1991). 51. Urbach, F. and R.W. Gange, The Biological Effects of UV-A Radiation, Praeger, New York (1986). 52. Urbach, F., Biological Responses to UV-A Radiation, Valdenmar Publishing, Overland Park, KS (1992). 53. Wilson, B.C., Ed., Lasers in medicine, Photochem. Photobiol. 53, special issue (1991). 54. Urbach, F., 25th Anniversary Symposium, Landmarks in Photobiology, Photochem. Photobiology. 655, 1055-1515. 55. Coohill, T.P., Photobiology for the 21st Century, Valdenmar Publishing, Overland Park, KS (2001).

5
The Physics of Ultraviolet Laser Ablation
5.1 5.2 Introduction ....................................................................... 109
Laser Characteristics Material Characteristics

Deposition of Ultraviolet Radiation in Organic Materials.....................................................................113

More Advanced Considerations

George H. Pettit
Summit Autonomous

5.3 Target Decomposition........................................................ 120 5.4 The Ablation Plume ........................................................... 122 5.5 Repetitive Irradiation ......................................................... 128 References ...................................................................................... 129

5.1 Introduction
In 1982, two articles describing the ablation of organic material by excimer lasers appeared in the scientic literature.1,2 Those initial studies demonstrated that intense pulses of ultraviolet radiation precisely etched submicron layers from synthetic polymer surfaces and caused negligible thermal damage to the remaining target. Such microscopic accuracy generated much interest in the materials processing community, as well as in several medical disciplines. Within 1 year, argon uoride (ArF, = 193 nm) excimer laser ablation was being explored as a means of reshaping the surface of the cornea to correct for vision defects.3 Soon thereafter, xenon chloride excimer laser radiation (XeCl, = 308 nm), which could be transmitted by a exible beroptic catheter, was studied as a tool for clearing atherosclerotic blood vessel obstructions in a noninvasive manner.4 Over the ensuing years, clinical interest in pulsed UV laser ablation of tissue has continued to grow, as discussed elsewhere in this book. In this chapter, the physical mechanisms underlying the ablation process are described. Tissue ablation is the discrete vaporization of a microscopic quantity of biological material by a pulse of laser radiation. It is a complex event encompassing multiple phenomena occurring on a short (nanosecond to microsecond) timescale. To provide an organizational framework, the material is partitioned into six categories, which are presented in roughly chronological order over the course of an ablation occurrence. Within several of these six areas, the material is further subdivided into two sections. The rst part deals with the subject in a general qualitative sense, while the second explores it on a more quantitative level. While this outline facilitates the discussion, it is an essentially arbitrary structure. Several of the delineated events, such as the absorption of laser radiation and the decomposition of the target material, occur simultaneously, and undoubtedly impact on each other. These interrelationships are important aspects of the ablation process and are discussed in some detail.

0-8493-1146-2/02/$0.00+$1.50 2002 by CRC Press LLC

109

110

Lasers in Medicine

Although the theme of the text is the medical application of laser radiation, UV laser ablation studies involving nonbiologic as well as biologic organic material are discussed here. A wealth of data exists in the scientic literature on the topic of synthetic polymer ablation, due to numerous important applications in the microelectronics industry and elsewhere (e.g., calibration of medical laser systems). In addition, the chemical and physical properties that facilitate analysis of the lasertarget interaction are better established for man-made plastics than for tissues. Because many aspects of the process are similar for both polymers and tissue, much useful information can be gleaned from these investigations. Where signicant differences are known to exist between biologic and nonbiologic substrate ablation, these are expressly stated in the text.

5.1.1 Laser Characteristics

Although the specic lasers mentioned in the introduction are both excimer systems, the ablation mechanisms described in this chapter apply generally to any device fullling three criteria: 1. The laser must produce radiation at an ultraviolet wavelength. The relevant part of the electromagnetic spectrum extends from about 180 nm to 380 nm. 2. The laser output should be in the form of submicrosecond pulses of radiation. 3. When the laser is delivered to an ablation site, the energy per pulse must be sufcient to produce a uence (energy per unit area) above some threshold. The exact value of this ablation threshold uence is both material- and laser wavelength-dependent. The importance of each of these characteristics is now considered in turn. Because the energy associated with an individual photon increases with decreasing wavelength, photon energies at short ultraviolet wavelengths are substantial. In early analysis of the photoablation process, it was hypothesized that these energetic UV photons caused direct bond-breaking of the molecular chains in the organic target, and that the photochemical fragmentation vaporized material with negligible heating (vibrational excitation).5 This mechanism may, in fact, play a signicant role in laser ablation at wavelengths below ~200 nm, where photon energies are signicantly greater than the typical molecular bond strengths in organic polymers. However, similar ablation effects are observed for longer ultraviolet laser wavelengths where the photon energy is insufcient to break organic bonds.4,6,7 There now appears to be general agreement among researchers that the photoablation phenomenon at most ultraviolet wavelengths is primarily due to multiple-photon processes, including localized heating.69 The absolute importance of single-photon bond-breaking to the event is still debated. A separate feature of ultraviolet radiation universally considered to be important for laser ablation is that it is very strongly absorbed in organic targets (i.e., tissue or synthetic polymers). Reported values for the 1/e penetration depth of 193 nm ArF excimer laser radiation in corneal tissue range from 0.3 to 4 m,10,11 and only slightly higher in the plastic polymethylmethacrylate (PMMA).12 This means that, when a laser pulse strikes an organic target, the radiation is deposited in a thin surface layer of the substrate, resulting in a localized region of very high deposited energy density. Whatever effect the laser pulse has on the material, i.e., either bond-breaking or vibrational heating, occurs in that narrow supercial volume. Laser alteration of the material is also conned by the submicrosecond pulse duration. Simply stated, laser energy is absorbed by the target volume much faster than it can diffuse as heat into the adjacent bulk. The competition between laser absorption and heat diffusion is shown schematically in Figure 5.1. One measure of the rate of heat diffusion out of a target region is the thermal relaxation time t, dened as the time required for the excess temperature in the target to fall by half. If the laser penetration depth is on the micron scale, then for virtually any biomedical application, the lateral dimensions of the absorbing volume will be several orders of magnitude larger than the depth. Heat diffusion out of the heated region is virtually unidirectional, into the deeper target. Under these conditions, the thermal relaxation time can be estimated by:13

The Physics of Ultraviolet Laser Ablation

111

incident laser pulse

absorbing region

target surface

heat diffusion

FIGURE 5.1 The fundamental lasertissue interaction, which is central to the ablation process. An incident laser pulse is absorbed in a surface layer of the target tissue. After absorption, the energy originally contained in the pulse can diffuse as heat into the surrounding tissue. By delivering the energy as fast as possible, i.e., in a very short pulse, and choosing a laser wavelength where the tissue is highly absorbing, the laser energy can be initially conned to a small region of the tissue and cause vaporization.

1 , t = -----------2 4

(5.1)

where is the absorption coefcient of the target at the laser wavelength and is the thermal diffusivity of the material. For aqueous biological tissues, values for in the ultraviolet typically lie between 102104 cm-1 (UV absorption in synthetic polymers can exceed 105 cm-1), while is on the order of 10-3 cm2 s1.14 Therefore, the relaxation time for a target tissue volume will be ~100 s or greater. A laser pulse more than two orders of magnitude shorter than this relaxation time will be completely absorbed before any signicant thermal diffusion occurs. If the uence of the incident laser pulse exceeds some threshold, the third essential criterion, material in the absorbing region, will decompose into smaller molecular weight products that are then ejected in the ablation event. As indicated above, the precise value of the ablation threshold uence is extremely case specic. For example, the measured ablation threshold for ArF laser ablation of the cornea is slightly less than 50 mJ/cm2 per pulse,11 while the threshold uence for XeCl laser ablation of calcied arterial wall tissue is approximately 3.0 J/cm2 per pulse.15,16 Practical considerations as to beam delivery and useful target spot size dictate that clinical laser systems for these applications produce tens to hundreds of millijoules per pulse. The earliest pulsed UV lasers generally available for ablation work were excimer devices. An excimer laser works by passing a controlled intense electrical discharge through a gas-lled cavity. The gas mixture in the cavity has three components: a noble gas (argon, krypton, or xenon), a halogen (either uorine or chlorine), and an inert buffer gas (helium and/or neon). The electric discharge leads to the transient formation of noble gas or halogen molecules, and the subsequent energetic dissociation of these unstable molecules releases the ultraviolet photons that make up the laser output. Excimer lasers typically produce pulses lasting 1030 ns, although special versions of the XeCl laser have been built in which the pulse is stretched to several hundred nanoseconds duration. This lengthening of the pulse facilitates the efcient coupling of the 308 nm radiation into optical bers,15 an essential task for laser angioplasty. Such an increase in laser pulse length has been shown to not substantially affect the tissue ablation achieved for constant-uence pulses16,18 consistent with the orders of magnitude longer thermal relaxation time. Despite the excellent tissue ablation characteristics of excimer lasers, the fairly large size of these devices and their requirement for pressurized halogen gases make them less than ideally suited for clinical use. These limitations have prompted development of alternative pulsed ultraviolet sources. Currently, the most advanced of these is the Q-switched nd:YAG laser. This solid-state laser produces intense (up to 1 joule or more) nanosecond pulses at a wavelength of 1.064 m. By passing this intense IR radiation through nonlinear crystals, it is possible to obtain ultraviolet pulses at 3 wavelengths: 355 nm, 266 nm,

112

Lasers in Medicine

and 213 nm. Laser radiation at 355 nm has been employed in experimental angioplasty work,19 while 213 nm radiation is being studied for corneal sculpting.20

5.1.2 Material Characteristics

What characteristics of the target material are important in the ablation process? Obviously, to be affected by an incident laser pulse, a target must absorb some fraction of the pulse energy. Absorption in organic material is due to specic molecular or submolecular features called chromophores. In biological systems, the most common molecule is water, composing typically ~70% of the total tissue mass. While absorption by water is stronger in the ultraviolet than in the visible, over the range of wavelengths considered here, the absorption coefcient for water at physiological temperatures does not exceed 10 cm-1.21 Therefore, water is not usually considered a signicant chromophore for UV laser pulses, although it has recently been hypothesized to play an important absorptive role at elevated target temperatures.22 The next most prevalent component in tissue is the organic fraction, composing ~25% of the total wet weight. This grouping includes the proteins, lipids, nucleic acids, etc., that make up the various cellular and extracellular structures. Such biochemical entities are typically rich in ultraviolet chromophores. Consequently, most of the energy in an incident UV laser pulse will be absorbed by the organic tissue component. Protein, the dominant organic entity in tissue, consists of a chain of amino acids connected by peptide bonds, as shown in Figure 5.2. Two ultraviolet chromophore types exist in the macromolecule the amino acid side chains and the peptide bond itself. The identity and concentration of the different side chain absorbers vary widely between proteins, while the peptide bond is always present in abundance. The absorptive strength of these different entities is conventionally reported in the biochemical literature as the molar extinction coefcient or molar absorptivity, e, reported in units of cm1 M1. Of the 20 common amino acids, three have side chains incorporating a 6-carbon aromatic ring: tryptophan, tyrosine, and phenylalanine. Benzene itself is a well-known ultraviolet chromophore, with a strong absorption peak at 198 nm ( = 8000 M1 cm1), and a second weaker absorption band centered around 255 nm (e = 230 M1 cm-1).23 When the aromatic ring structure is incorporated into an amino acid, these characteristics are essentially retained. The molar extinction coefcients for the aromatic amino acids at the 193 nm ArF laser wavelength lie between 25 104 M1 cm1.24 Tryptophan, the largest, most highly conjugated amino acid, also exhibits a signicant absorption band over a broad wavelength region centered around 275 nm.25 Tyrosine absorbs at these wavelengths as well, although at only one fourth the magnitude of tryptophan, due to a less complex side chain.25 Phenylalanine, the simplest aromatic amino acid, exhibits only very weak absorption in the vicinity of 255 nm.25 Therefore, a high concentration of aromatic amino acids confers a strong far ultraviolet absorption capacity to a protein molecule, as well as moderate mid-UV absorption (dependent primarily on tryptophan). Proteins composed of nonaromatic building blocks still absorb signicantly in the far ultraviolet, due to the peptide bond. This molecular structure exhibits an absorption peak at approximately 185 nm, which falls with increasing wavelength to essentially zero by ~240 nm.26 The molar extinction coefcient for the bond at the 193 nm ArF excimer wavelength is 5.5 x 103 cm-1 M-1,24 roughly an order of magnitude less than that for the aromatic amino acids. However, as this feature occurs once for every amino acid along the protein chain, the number of peptide bonds in the macromolecule is always very large. Therefore, far ultraviolet absorption by peptide bonds is important in all proteins, even those rich in aromatic side chains. Other organic biologic molecules, less prevalent than protein in most tissues, play a secondary role in the absorption of a UV laser pulse. All exhibit some absorption in the far ultraviolet. The nucleic acids (DNA and RNA) are also absorbing at mid-UV wavelengths between ~250 nm and 290 nm.24,26 This is not the case with carbohydrates, which are essentially transparent to radiation above about 230 nm.26 Lipids, which primarily contain simple hydrocarbon bonds, are also essentially nonabsorbing throughout most of the ultraviolet.

The Physics of Ultraviolet Laser Ablation

113

side chain X N H C H

peptide bond O C N H Y C H

side chain O C

amino acid

amino acid

FIGURE 5.2 The essential molecular structure of protein, an important tissue absorber of ultraviolet laser radiation. Proteins are biological polymers composed of multiple amino acid subunits linked by peptide bonds. The uniqueness of each amino acid is entirely due to the side chain. In turn, the structure and function of the protein macromolecule is dependent on the identity and ordering of the constituent amino acids.

The light-scattering potential of a target also affects its interaction with laser radiation. Tissue is an inherently scattering medium at ultraviolet wavelengths due to the high density of refractive index uctuations in biologic structures. Because scattering in a material typically increases as wavelength decreases, the phenomenon would be expected to be most pronounced at the shortest ultraviolet wavelengths. However, since the absorption coefcient in tissue increases even more dramatically with decreasing wavelength, scattering has the largest effect on laser pulse propagation in the near ultraviolet, at wavelengths longer than ~280 nm. For example, the scattering coefcient for 308 nm radiation in normal human aorta is 77 cm1, while the absorption coefcient in the same tissue is only 33 cm1.27 Thus, most of the attenuation of a forward directed XeCl laser pulse in aortic tissue is due to scattering rather than absorption.

5.2 Deposition of Ultraviolet Radiation in Organic Materials

The rst challenge in understanding the interaction of an ultraviolet laser pulse with an organic target is to describe how the pulse energy is deposited in the tissue. A common starting point for such analysis is Beers law, also referred to as Beer-Lambert law, which states that the incident laser pulse uence F0 will be attenuated exponentially versus depth x in the material: F ( x ) = F0 e
x

(5.2)

where F(x) is the residual pulse uence crossing x, and is the measured absorption coefcient of the target. (The absorption coefcient is also often noted by a). A clear distinction should be made between the absorption coefcient , commonly used in the physical literature, and the molar extinction coefcient e introduced above. The molar extinction coefcient is dened as:23 F ( x ) log --------- F0 10 = --------------------------Cx

(5.3)

where C is the molar concentration of the chromophore in question. Thus, for a single absorbing species, = e C log10(e). Equation 5.2 is only partially applicable to the case of UV laser tissue ablation. The effects of light scattering, which may be signicant at long ultraviolet wavelengths, are not included in the equation. In addition, is assumed to be a constant equal to the measured low intensity value. Under intense irradiation, the effective attenuation coefcient of the target may differ signicantly from the small-

114

Lasers in Medicine

signal value and may change during the laser pulse.28,29 Such effects are due to multiple-photon absorption and other nonlinear processes. Despite these shortcomings, Equation 5.2 can still be used to derive a qualitative description of the laser pulse deposition that yields useful insight into the ablation process. To understand the effect of a laser on a target, one should know the amount of pulse energy deposited per unit volume in the material. This quantity will also vary with x, and is dened here as E(x). As a laser pulse penetrates into an absorbing medium, if scattering may be neglected, the uence decrease per unit depth is equal to the energy absorbed per unit volume: dF x - = F0 e . E ( x ) = ----dx

(5.4)

Thus, like the pulse uence, the deposited energy density also decays exponentially with depth. However, note that E(x) is considerably more dependent on the target absorption coefcient , than F(x) is. Maximum energy deposition occurs at the surface (x = 0), where E(0) is equal to F0. On the basis of Equation 5.4, one can infer that the threshold uence for ablation of a substance, Fth, corresponds directly with a threshold deposited energy density, Eth = Fth. Experimental evidence suggests that Eth is an intrinsic material property, relatively independent of the ultraviolet laser wavelength.7,30 It is logical to assume that, as the incident uence F0 is increased above Fth material will be ablated to a nite depth, d, where E(x) falls to Eth. A mathematical expression relating incident uence to ablation depth can be derived using these relations. By starting with: E ( d ) = E th , and, substituting on both sides: F0 e we can arrive at the simple expression: 1 F0 - ln ------ . d = - F th
d

(5.5)

= F th ,

(5.6)

(5.7)

Equation 5.7 predicts that the etch depth per pulse will increase logarithmically with uence above threshold. (Experimental laser etching results typically exhibit signicant deviations from Equation 5.7, for reasons discussed below.) This theoretical framework can be used to explain several of the wavelength-dependent effects observed in experimental ablation studies. For example, consider a tissue being irradiated by one of two pulsed lasers operating at distinct wavelengths, 1 and 2, with absorption in the target being much higher at the rst wavelength ((1) (2)). According to Equation 5.4, for an incident pulse of uence F0, the deposited energy density at the target surface will be much greater for the laser operating at the 1 wavelength. Furthermore, the energy density will fall off much more rapidly with depth at 1 than at 2. These effects are shown graphically in Figure 5.3. In Figure 5.3a, the incident laser pulse is of relatively low uence. However, since the deposited energy density at the target surface is the product of uence and absorption coefcient, the ablation threshold is exceeded for the more strongly absorbed 1 laser radiation, and material is removed to depth d1. Therefore, the threshold uence for ablation is lower at the more strongly absorbed laser wavelength. This has been found to be true in virtually all ablation studies.31,32 In Figure 5.3b, the incident laser pulse is much more intense. The deposited energy density at the material surface exceeds Eth for either laser wavelength, therefore, ablation occurs in both cases. Note

The Physics of Ultraviolet Laser Ablation

115

Deposited Energy Density

a: low fluence laser pulse

E1(x) E2(x) Eth

d1 Depth in Material b: high fluence laser pulse

Deposited Energy Density

E1(x) E2(x)

Eth d 1 d2 Depth in Material

FIGURE 5.3 Deposition of two identical-uence laser pulses at different wavelengths, 1 and 2, in a material which absorbs much more strongly at 1. Ablation will occur in the material to a depth where E(x) falls to some threshold value, Eth. (a) Absorption of a low uence laser pulse. Ablation only occurs for the 1 laser wavelength. (b) Absorption of a high uence pulse. Laser radiation at either wavelength causes ablation, with the 2 pulse removing material to a greater depth.

that the etch depth achieved at 2, the weakly absorbed laser wavelength, exceeds the depth produced by the 1 laser pulse (d2 > d1). As the incident uence is increased still further, d2 will continue to grow much more rapidly than d1. These trends are also borne out in experimental work; at laser uences above the threshold regime, a smaller target absorption coefcient results in a greater dependence of etch rate on uence, and, consequently, for most practical uences, a signicantly deeper ablation depth per pulse.7,31,32 One other aspect of the lasertarget interaction can be inferred from Figure 5.3b. Since the uence in the laser pulse is the same at both wavelengths, the integral of E(x) over all depths, i.e., the area under either curve, is also the same. Note that, for the pulse at 1, the vast majority of the laser energy is absorbed in the ablation volume from the surface to d1. In the 2 case, the average value of E(x) in the ablation volume is much less, and a larger fraction of the pulse energy is deposited below d2, in the intact target. The excess energy in the ablated region under high absorption conditions results in decomposition of the original organic material into smaller molecular fragments. The ablation products for ArF excimer laser irradiation of PMMA are, on average, signicantly smaller in molecular weight than those for krypton uoride (KrF, = 248 nm) excimer laser irradiation, in part due to the 35-times higher absorption coefcient at the 193 nm wavelength (193 = 2.0 103 cm-1 versus 248 = 5.7 102 cm-1).31 The greater residual energy left in the target under low absorption conditions is converted to heat,30,33 which may contribute to damage of the site. The concepts presented in this section are exemplied particularly well by comparing ArF and KrF excimer laser ablation of corneal tissue. The cornea of the eye is composed primarily of collagen, which accounts for ~70% of the dry weight of the tissue.34 Collagen is a somewhat unusual protein, in that it is almost devoid of aromatic amino acids and absorbs ultraviolet radiation only through the peptide

116

Lasers in Medicine

10 Corneal Ablation Depth per Pulse (m)

0.1

0.01 193 nm 248 nm 0.001 10 100 1000

2

104

Laser Fluence per Pulse (mJ/cm )

FIGURE 5.4 The reported ablation depth per pulse in cornea as a function of incident laser uence at 193 nm and 248 nm, based on literature sources as cited in the text. The distinct ablation behavior observed at the two wavelengths is largely attributable to the much larger tissue absorption coefcient at the shorter ultraviolet wavelength.

bond chromophores. Because of this, the absorption coefcient of the cornea is much higher at 193 nm (between 2700 and 40,000 cm1)10,32 than at 248 nm (210 cm).32 The ablation behavior of the tissue is shown in Figure 5.4, based on published measurements of corneal etching by either ArF 32,33,3541 or KrF32,35 excimer lasers. ArF laser ablation of the tissue results in a lower threshold and a slower increase in etch depth versus tissue than does KrF laser ablation. In addition, KrF laser irradiation of the cornea is associated with more damage to the residual tissue. Furthermore, an examination of the ablation plumes produced at the two excimer wavelengths indicates much smaller ablation particles at 193 nm.42 All these ndings are consistent with the absorption analysis discussed above. 5.2.1 More Advanced Considerations One limitation of the simple theory presented above is that it assumes that is constant over the course of the laser pulse. At ablative laser intensities, the optical properties of the target can change substantially during nanosecond-timescale irradiation. For example, in the case of ArF laser ablation of cornea, a detectable decrease in laser transmission has been observed for irradiation of microtome thin tissue slices.29 Both transmission decreases and increases have been measured during UV-laser ablation of various synthetic polymers.4345 In this subsection, the theoretical description of the ablation process is extended to analyze this dynamic (i.e., radiation-dependent) target absorption. As an example, we will rst consider the well-studied case of 248 nm KrF laser ablation of polyimide. The measured relationship between polyimide etch rate and KrF laser pulse uence is shown in Figure 5.5. The discrete data points in the gure are taken from two experimental studies, one using a thin polyimide lm technique very accurate at low pulse uences (9) and the other based on perforation of thicker polyimide sheets by repetitive irradiation (o6). (The discrepancies between the two sets of data are most likely due to unavoidable inaccuracies in the repetitive irradiation method at laser uences just above the ablation threshold.) The continuous line in the gure is the blow-off theoretical relationship, based on reported values for the ablation threshold uence (28 mJ/cm2) and absorption coefcient (25 m-1).7 Clearly, material is being removed to a signicantly greater depth than predicted by the theory, particularly at higher laser pulse uences. The transmission characteristics of polyimide have also been studied over this same range of KrF laser pulse uences, as shown in Figure 5.6.43 The vertical axis in the gure indicates the fraction of incident KrF laser pulse energy transmitted through a 0.1 m-thick polyimide lm (1 = total pulse energy transmission). The continuous horizontal line in the gure indicates the predicted transmission behavior

The Physics of Ultraviolet Laser Ablation

117

0.8 0.7 Ablation Depth per Pulse (m) 0.6 0.5 0.4 0.3 0.2 0.1 0 10

100

1000

10000

Laser Fluence per Pulse (mJ/cm2)

FIGURE 5.5 The reported ablation depth per pulse in the synthetic polymer polyimide as a function of incident laser uence at 248 nm, indicated by discrete symbols, based on two literature sources as cited in the text. The continuous curve indicates the ablation behavior predicted by simple blow-off theory (Equation 5.7). The dashed curve is based on a slightly more complex model that takes into account radiation-induced changes in the material absorption coefcient (Equation 5.19).
1

Laser Energy Transmission

0.8

0.6

0.4

0.2

10

100

1000

10000

Laser Fluence per Pulse (mJ/cm2)

FIGURE 5.6 Measured transmission of a 0.1 m-thick lm of polyimide as a function of KrF laser pulse uence (see text). For ablative laser uences, the transmission of the lm increases substantially, due to radiation-induced decreases in the target absorption coefcient. Curves in the gure are theoretical models of the absorption behavior, as described in the text.

of the lm, i.e., a constant ~ 8%, based on Beers law and the measured absorption coefcient of the material. With increasing laser intensity, a progressively greater fraction of the incident 248 nm radiation passes through the polyimide lm. To understand these two interrelated ndings in Figures 5.5 and 5.6, we must analyze laser absorption by the target in a slightly more detailed way.43,46 We do this using the simple two-level absorption scheme shown in Figure 5.7. Consider an idealized organic target with a single population of chromophores, all with identical absorption characteristics and present in the material at a density . Initially, all chromophores are in the ground state (0). Absorption of photons from an incident laser pulse will promote some chromophores to the excited state (1). The densities of chromophores in the ground and excited states at depth x in the material and at time t during the irradiation are noted as 0(x,t) and 1(x,t). The

118

Lasers in Medicine

1(x,t) 1

l(x,t)

0(x,t)

FIGURE 5.7 Theoretical scheme for the single-photon absorption of ultraviolet laser radiation by an organic substrate. Material chromophores, initially in the ground state (0), are promoted to the excited state (1) by absorbing single photons from the incident laser pulse. The excited chromophore state may have substantially different absorption properties from the ground state.

afnity of a single ground-state chromophore for a photon at the laser wavelength is expressed as the absorption cross-section, 1, which has the units of area. (The cross-section in cm2 for an individual chromophore can be directly calculated from the molar extinction coefcient, when reported in M-1 cm1, simply by multiplying by 1.66 10-21.) The product of the initial ground-state chromophore density (0(x,0) = ) and the absorption cross-section must be equal to the material absorption coefcient: 1 =

(5.8)

Under conventional, i.e., low-intensity irradiation, the density of photons absorbed per volume of material is much less than and absorption in the material is accurately described by a constant . However, at high laser intensity, this is not necessarily the case. For polyimide, the density of building-block imide monomers in the intact material is about 2.3 x 1021 cm-3. Each monomer is rich in double bonds and other features known to be strongly absorbing in the ultraviolet,6 so that the number of 248 nm chromophores per monomer may be as high as 20. Therefore, in this situation can be estimated to be 45 1022 cm3. Using the above stated values for and Fth in Equation 5.4, we can estimate Eth to be about 7,000 J/cm3, or 9 1021 absorbed 248 nm laser photons per cm3 (based on the energy associated with each photon being 8 10-19 J). Therefore, just at the ablation threshold, the density of laser photons deposited in the surface region is almost one fourth as large as the chromophore density. This fraction becomes progressively more substantial as the incident pulse uence is increased above threshold. We therefore must consider excited state chromophores in analyzing the radiation transport process. Many plausible suppositions can be made about the rst excited state in Figure 5.7. For example, the chromophore might quickly (relative to the pulse duration) return to the ground state through either radiative or nonradiative decay. That mechanism would allow for some laser energy deposition in the material without appreciably altering the absorption, not the effect we are presently trying to describe. In this particular instance, we will assume that each chromophore promoted to the excited state is permanently bleached by absorption, as would occur with rapid dissociation of the excited state entity into new nonabsorbing species. If the laser ux (photon per area per time) crossing depth x in the material is given as I(x, t), then the density of ground and excited chromophore states in the material will evolve over time as: 0 ( x, t ) 1 ( x, t ) -------------------- = -------------------- = 1 0 ( x, t ) I ( x, t ) . t t

(5.9)

I(0,t) of course, is the incident photon ux, determined by the temporal characteristics of the laser. In a similar fashion, the spatial attenuation of the incident photon ux will be: I ( x, t ) ----------------- = 1 0 ( x, t ) I ( x, t ) x

(5.10)

The Physics of Ultraviolet Laser Ablation

119

Equations 5.9 and 5.10 can easily be solved. Integration of 5.9 yields: 1 ( x, t ) = 1 exp 1 I ( x, t ) dt
t

(5.11)

Substituting Equation 5.11 into 5.10 gives: I ---- = 1 I ( x, t ) exp 1 I ( x, t ) dt . x

t

(5.12)

Because the duration of excimer laser pulses is typically several orders of magnitude less than the interval between pulses, Equation 5.12 can be integrated from time zero to innity to obtain the attenuation for the total pulse photon density S(x), where:

S(x) =

I ( x, t ) dt .
0

(5.13)

S(x) times the photon energy is the uence crossing depth x in the material. Integration of Equation 5.12 then yields: dS ----- = 1 I ( x, t ) exp 1 I ( x, t ) dt dt dx
t

(5.14)

which results in: dS ----- = ( 1 exp ( 1 S ) ) dx

(5.15)

What does Equation 5.15 imply about the laser deposition in such a target? At low pulse uences, where 1 S 1, the expression reduces to: dS ----- = 1 S , dx

(5.16)

which, in light of Equation 5.8, is simply Beers law. At high uences, where 1 S 1, Equation 5.16 simplies to: dS ----- = . dx

(5.17)

Equation 5.17 indicates that the absorption becomes saturated at high intensity. The maximum attenuation rate of the incident photon ux per unit depth is simply equal to the density of chromophores in the material. Equation 5.15 can be integrated from the surface to an arbitrary depth x to nd the residual number of laser photons per area crossing x:

120

Lasers in Medicine

exp ( 1 S 0 ) 1 1 - ln 1 + --------------------------------S ( x ) = ---1 exp ( 1 x )

(5.18)

Returning to the blow-off model, laser etching of the target should occur to a depth where S(x) falls to the ablation threshold value, Sth, which again is simply the threshold uence divided by the energy per photon. This depth, d1, can be found from Equation 5.18: 1 exp ( 1 S 0 ) ( S 0 S th ) 1 - ln -------------------------------------- . - + -d 1 = -------------------- 1 exp ( 1 S th )

(5.19)

This expression is obviously quite different from Equation 5.7. It predicts the ablation behavior for a material with saturable absorption at the laser wavelength. We can test this theoretical description against the experimental results in Figures 5.5 and 5.6. If we assume that the density of 248 nm chromophores in the polyimide target = 4.5 1022 cm-3, then from Equation 5.8, 1 5.5 10-18 cm2. Plugging these values into Equation 5.19 gives the predicted etch depthuence relationship indicated by the dashed curve in Figure 5.5. Using the same Equation 5.5 constants in Equation 5.18 produces the calculated transmission behavior shown by the dashed curve in Figure 5.6. In both cases, model agreement with experiment is quite good for laser pulse uences below ~2 J/cm2. Above this uence, other nonlinear processes, such as plasma formation, that are not accounted for in this model, may play an important role.28 More important than the particulars of this mathematical model is the simple fact that extending the blow-off ablation theory to account for changes in target absorption during the laser pulse greatly improves the theorys accuracy in describing the etchdepth uence relationship. How important is dynamic target absorption in tissue ablation? In the best-studied case, ArF laser ablation of cornea, transmission of 193 nm laser radiation through corneal tissue has been observed to decrease at high ArF laser pulse uences. This transmission decrease is not due to increased reection or scattering of laser radiation from the ablation site,47,48 and hence represents enhanced laser absorption in the tissue. The dominant chromophore in the cornea, the peptide bond in the structural collagen molecules, is present in the tissue at a density of about 1021 cm-3.10 Using a value of 10,000 cm-1 for the tissue absorption coefcient (roughly the geometric mean of reported values) and 50 mJ/cm2 for the threshold laser uence, the ablation threshold absorbed energy density is ~5000 J/cm3, or 5 x 1020 193 nm laser photons per cm3. Thus, as was the case for polyimide, at ablative laser uences a signicant fraction of the target chromophores participate in the absorption process, and dynamic chromophore changes are a plausible explanation for the observed change in tissue optics. Laser-induced increases in absorption can be most readily accommodated in the model of Figure 5.7 by assuming that the excited chromophore state has a nite lifetime (relative to the laser pulse) and an absorption cross-section larger than s1.43 A detailed numerical analysis here is not useful, because there are substantial uncertainties in both the true value of for this interaction and in the corneal etch depth versus ArF laser uence relationship.10 Sufce it to say that nonlinear absorption processes, e.g., chromophore modication and plasma formation, undoubtedly do occur during tissue irradiation and do impact on the ablation behavior.

5.3 Target Decomposition

Having described how the laser energy is deposited in the tissue, we must now consider the consequences of that absorption. Figure 5.8 gives a simple schematic of the possible steps leading from radiation absorption to ablation. The ultraviolet laser photons initially promote target chromophores to excited electronic states. Some fraction of the absorbed energy may be reradiated, as occurs, for example, in uorescence. However, the vast majority of the pulse energy follows two pathways after the initial absorption. An electronically excited chromophore may directly dissociate into two new species (bond-

The Physics of Ultraviolet Laser Ablation

121

Absorption of UV Radiation

Electronic Excitation

Direct ''BondBreaking''

Organic Tissue Component

Vibrational Relaxation

Denaturation/ Decomposition

Molecular Collisions

Tissue Ablation

Aqueous Tissue Component

Heating

Expansion/ Vaporization

FIGURE 5.8 Simplied schematic of the energy pathways leading from laser irradiation to ejection of tissue decomposition fragments (ablation). Laser radiation is primarily absorbed by the structural organic tissue component. A combination of the weakening of this tissue scaffolding and generation of vapor phase decomposition molecules drives the ablation event.

breaking).5 This pathway may be particularly important at shorter ultraviolet wavelengths, where the photon energy exceeds the bond strength of most organic bonds. Alternatively, the energy in the electronically excited chromophore can be redistributed to the rest of the macromolecule in the form of vibrations (vibrational relaxation).49 This pathway is predominant for longer UV wavelengths, and likely plays some role at shorter wavelengths as well.69 Large organic polymers such as proteins will contain numerous chromophores within each molecule. Transfer of energy from multiple chromophores to the bulk macromolecule will drive it into pronounced vibratory oscillations. If severe enough, these vibrations, in turn, cause denaturation, i.e., change in the spatial folding of the biopolymer, and decomposition into smaller molecules. At the same time, because the oscillating organic molecules in tissue are bathed in an aqueous environment, some thermal energy transfer from the initial energy absorbers to surrounding water molecules also occurs. (Note that here we are talking about heating on a nanoscopic scale, not thermal diffusion out of the laser absorbing volume as described by Equation 5.1.) All of this occurs on a timescale faster than the nanosecond duration of the laser pulse.49,50 We know that the target undergoes heating because we observe in it a pressure transient characteristic of rapid thermal expansion (Grneisen stress, as discussed below),50,51 as well as a bulk temperature rise after the laser pulse.5255 Less certain is the degree of heating that occurs during the laser pulse and the contribution of such heating to the ablation event. Heating of the aqueous tissue component could theoretically be both of large magnitude and very rapid, in part due to temperature related changes in the ultraviolet absorption properties of water.22 Once initiated by heat transfer from organic molecules early in the laser pulse, the warming aqueous component could then directly absorb a signicant fraction of the laser energy. If the tissue water within the laser-absorbing volume could be heated very quickly, i.e., faster than it could physically expand, then the enormous resulting pressure (due to the relative incompressibility of water) could drive the explosive ejection of material.56 Less rapid aqueous heating could still contribute to the ablation event not through laser absorption, but through steam formation. These mechanisms are currently the subject of numerous laboratory investigations. Denaturation and fragmentation of the organic tissue component greatly reduces the structural integrity of the tissue.50,57 This structural weakening, combined with the internal pressure generated by gas formation and thermal stresses, leads to the ejection of ablated material. There is good evidence that this

122

Lasers in Medicine

ejection begins during the nanosecond timescale UV pulse. Target stress wave measurements conducted during photoablation of a variety of tissues50,51,58,59 as well as polyimide60 indicate the onset of material removal during the laser pulse. The photoacoustic measurements in these studies indicated that, as the incident uence was increased, material ejection occurred progressively earlier during the laser pulse, and that the process occurred via a rapid surface vaporization rather than a bulk explosion. In a separate examination of XeCl laser polyimide ablation using an ultrafast imaging apparatus to look directly at the target surface,61 etching of the polymer clearly occurred layer by layer over the course of the irradiation. This moving ablation front was observable late in the irradiation at low ablative pulse uences, and early at high laser intensity, in general agreement with the photoacoustic studies. Another measurable phenomenon that occurs during the laser irradiation is a marked fall in target reectivity. UV laser pulses specularly reected from a host of synthetic and biologic materials under ablation conditions exhibit a truncation in the latter portion of the temporal pulse proles.6264 In every case studied to date, this reected pulse clipping is rst observable at incident laser uences near the ablation threshold and becomes increasingly more pronounced as the laser intensity rises. The reectivity decrease is transient,65 although it may persist for up to several hundred microseconds after the ultraviolet laser pulse.66 Although the precise cause(s) of this phenomenon has not been completely established, both debris plume attenuation effects and target index of refraction changes have been advanced as possible mechanisms.62,65 Recent experimental measurements have, in fact, demonstrated refractive index changes in polyimide during UV laser irradiation.67,68 Whatever the cause, the effect has been utilized empirically to examine various aspects of the ablation process,62,69 and has been investigated as a diagnostic feedback monitor for clinical photoablation procedures.64,70 If the heating occurs much faster than the material can move, signicant Grneisen stress builds up in the laser-absorbing volume.71 To assess the magnitude of this phenomenon in a given situation, the heating period, i.e., the laser pulse duration, must be compared with the expansion time of the heated volume. In an fashion analogous to Equation 5.1, which dened a thermal relaxation time, an acoustic relaxation time, a, can be found by dividing the 1/e laser penetration depth by the speed of sound in the material.72 The speed of sound in most tissues will be approximately the same as in pure water, ~ 1500 m/s, so that: 1 a = ------------------------------------ . ( 1500m s ) ( )

(5.20)

If the laser pulse is short compared with a, the absorbing material is being heated much faster than it can possibly expand, and the Grneisen stress becomes important. It is worth pointing out that, in Equation 5.20, the acoustic relaxation time is inversely proportional to the absorption coefcient. Therefore, for 193 nm laser tissue ablation, where the absorption coefcient is at least several thousand inverse centimeters, that is, on the order of a few nanoseconds. Because ArF excimer laser pulses are typically 1020 ns long, the Grneisen stresses are probably not substantial in these cases. For the 308 nm XeCl laser, tissue absorption coefcients may be more than two orders of magnitude smaller, while the laser pulse duration usually does not exceed ~100 ns. Under those conditions, the Grneisen stress may become very important.52,72

5.4 The Ablation Plume

The most dramatic aspect of the ablation process is undoubtedly the visible explosion of debris from the tissue surface, accompanied by an audible sharp snap. Principal features of this event are illustrated in Figure 5.9. The ejection of ablated material causes generation of a recoil pressure wave that propagates into the target,59,73,74 and which possibly could cause ancillary damage to the tissue. In addition, a shock wave is typically formed in the ambient environment that moves away from the target ahead of the actual ablation plume.73,75,76 The dynamics of the explosion are highly dependent on the environment. Conned

The Physics of Ultraviolet Laser Ablation

123

Ablation in Air

Ablation in Liquid

incident laser pulse

incident laser pulse

shock wave ablated material

shock wave

ablated material tissue surface tissue surface

recoil momentum recoil momentum

(a)

(b)

FIGURE 5.9 Dynamics of the plume expansion following laser ablation. In air, the rush of material into the ambient atmosphere creates a shock wave, as well as a recoil force pressing downward on the tissue target. Similar effects are observed for ablation in a liquid environment, although the plume expansion, i.e., bubble formation, is constrained by the uid. This tamping leads to much greater recoil forces acting on the residual target.

plume expansion, e.g., expansion in a liquid such as blood or saline, is considerably retarded versus expansion in air, and is accompanied by much greater recoil stresses in the remaining tissue.74,77 What are the products of ablative photodecomposition? The dominant constituent in a tissue ablation plume is submicron diameter water droplets78 not surprising, given the target composition. Mass spectroscopy has been used to analyze the gaseous tissue ablation products. Corneal tissue ablation by either 193 nm or 248 nm excimer laser produces a variety of organic gases, e.g., C2H2 and CH2NH2, in addition to water vapor.79 Heavier molecular weight material, presumably macromolecular fragments, are also observed. Except for the water, similar species are also observed in ablation plume from synthetic polymers. Infrared spectroscopy and gas chromatography, as well as mass spectroscopy, have been used to identify the principle gaseous products of polyimide photoablation: CO2, CO, H2O, and HCN, as well as various light hydrocarbons (up to 4 carbon atoms in size).7,80 Larger molecules all the way up to visible soot particles have also been identied.6,7,81 A consistent nding in these studies is that smaller molecular weight products are more prevalent as higher laser pulse uences are used. This trend can be explained by referring back to Figure 5.3. For two laser pulses of different energy but the same wavelength, the average deposited energy in the ablated volume is greater for the higher energy pulse. This extra energy fragments the plume debris into progressively smaller molecules and accelerates them to higher velocities. There are also short-lived species that form early in the ablation event but do not persist long enough to be detected by the above methods. For example, laser-induced-uorescence measurements indicate that the radicals C2 and CN form transiently during polyimide photodecomposition.6 Because of their bioreactivity and potential deleterious impact on tissue, free radicals are particularly interesting transient ablation products. Electron paramagnetic resonance spectroscopy indicates that short-lived free radicals are formed during both 193 nm irradiation of cornea,82 and during 308 nm irradiation of cardiovascular tissue.83 In the former case, there is in vivo evidence that radical bioeffects do impact the corneal healing response.84 One of the most straightforward and useful methods for studying the dynamics of material ejection is laser-ash imaging.42 This technique is illustrated in Figure 5.10. A pulsed visible laser is used essentially as a stroboscopic light source to generate an image of the ablation site on a camera or other imaging device. By varying the delay between the ring of the ablating laser and the imaging pulse, it is possible to capture clear images of the target explosion at discrete time intervals, and hence to assess ejection

124

Lasers in Medicine

variable delay generator

pulsed UV ablation laser

target pulsed visible laser

imaging system

FIGURE 5.10 Ultrafast imaging technique used to study ablation plume expansion dynamics. By varying the delay between the UV ablation laser and the visible imaging laser (essentially a stroboscopic source), the time course of plume development can be explored. Motion artifacts due to the moving plume are reduced as the duration of the imaging laser pulse is shortened.

velocities and particle sizes. The resolution of the image will improve as the visible laser pulse is shortened because the blurring effect of motion is reduced. Successful imaging experiments of the photoablation event have typically used visible imaging pulses of a few nanoseconds or less duration. Typically, the optical axis of the imaging system is oriented parallel to the target surface to examine the orthogonal development of the plume. This technique was rst used to study excimer laser ablation of corneal tissue in air at ArF laser pulse uences of 100-900 mJ/cm2 and KrF laser pulse uences of 500 mJ/cm2.42 Using either excimer device, the plume was rst detectable 500 ns after onset of the 15 ns laser pulse. Images acquired at slightly different delay times indicated that the cloud of particulate debris was initially traveling normal to the tissue surface at transonic to supersonic velocities ( 340 m/s), but quickly slowed as it interacted with the viscous air to roughly 100 m/s by 10 s postirradiation. The plume velocity increased slightly with increasing ArF laser pulse uence, although the fastest plume expansion was observed during KrF laser irradiation of the tissue. This was attributed to the greater ablation mass and larger particle size at the longer ultraviolet laser wavelength (as discussed above) causing the plume to be less retarded by air viscosity. The major limitation of the imaging method is that it, of course, detects only solid particles. For ultraviolet laser ablation, a signicant fraction of the ablated material is in the form of low molecular weight gases (as discussed below), which are essentially invisible as they expand into air. Other techniques such as time-of-ight mass spectroscopy85,86 or probe-beam laser-induced uorescence6,31 have been used to examine the velocity of these ablation products. Results using these methods during ablation of synthetic polymers show gas-phase products move away from the target surface at substantially supersonic velocities (103-104 m/s). In contrast, in an aqueous environment, the gaseous plume becomes a bubble, which can be examined by ultrafast imaging. Such studies of bubble formation have been conducted during XeCl laser ablation of arterial tissue in saline,87 as well as an absorbing hemoglobin solution.88 During the rst ~100 s after a laser pulse, such bubbles expand with average velocities on the order of 20 m/s. As a bubble grows, the temperature of the initially hot gas inside falls quickly, causing a rapid decrease in the driving pressure. Eventually, after a few hundred microseconds, the compressive force of the surrounding uid causes the bubble to collapse. Interestingly, this collapse can be so violent in nature as to cause transient formation of a second bubble (sometimes called rebounding).88 The duration of the material ejection in air has also been well studied using the imaging technique of Figure 5.10. Images acquired during corneal ablation by the ArF excimer laser show debris leaving the tissue surface more than 5 s after the laser pulse.42 On a timescale of 10s to 100s of microseconds, the ejected plume material is shown to disperse in the air above the target. Essentially similar results have been reported for XeCl excimer laser irradiation of arterial tissue,89 and for KrF laser ablation of PMMA.76 As the ablated material begins to rush out of the target, it creates a blast or shock wave (a narrow layer of extremely high pressure gradients moving at supersonic velocity), which then propagates into the

The Physics of Ultraviolet Laser Ablation

125

ambient environment ahead of the ablation plume. Such shock waves are also visible using the imaging technique of Figure 5.10. These waves have been qualitatively observed to accompany ArF laser corneal tissue ablation,75 and have been quantitatively studied during ablation of synthetic polymers.76,90 For KrF laser ablation of polyimide, an initial blast wave velocity of 1200 m/s has been measured.90 As with the plume, the shock wave velocity rapidly falls off with expansion into the air. The recoil forces exerted on the remaining target by the ejected plume have been directly studied by mounting thin (30100 m) slices of tissue on a piezoelectric transducer.50,59 Measurements made during ArF laser corneal ablation indicate a transient pressure rise in the tissue of as much as 100 ATM due to the stress wave. Target pressure wave formation has also been studied using an acousto-optical technique and a synthetic model of corneal tissue: a thin sheet of polyimide oating on water.77 In that work, irradiating the surface polymer with ablative XeCl excimer laser pulses was found to produce true shock waves in the underlying water, with peak pressures calculated to exceed ~1000 ATM. The shock wave amplitude was even larger (by a factor of 510) at any given uence if the plume expansion was conned by a second overlying water layer. Notable in all such studies was that the stress or shock wave amplitude increased markedly with increasing laser pulse uence. These recoil forces in the target are not without biologic consequences. In one study of ArF laser, bone ablation osteocyte destruction was observed more than 1 mm beneath the surface of the etch crater,91 in all likelihood due to photoacoustic injury. Conning the ablation plume expansion, as occurs, for example, during laser angioplasty in a uid-lled blood vessel, has been demonstrated in several other experiments to lead to greatly increased tissue damage surrounding the target area.88,89,92,93 As the magnitude of the pressure wave in the target is a direct function of laser uence, these ndings suggest that the most efcacious laser therapy will be achieved at moderate pulse uences and, if possible, under conditions where the ablation plume can expand freely. It is often desirable to assess the recoil imparted to a tissue during ablation in a quantitative way. A clear analysis of this topic has been presented previously by Dingus.75 The essential features of his discourse are as follows. This discussion rst considers the case of expansion of the ablation plume into free space, then addresses the more complex situation of constrained or tamped vaporization. The law of conservation of momentum indicates that the total momentum of the ejecta in the ablation plume will be equal to the momentum delivered to the remaining target. If the ablation is assumed to be uniform over an irradiated target of area A then:

Mi vi ---------- = A

p ( t ) dt ,
0

(5.21)

where Mi is the total mass of the plume debris traveling orthogonally from the target at velocity vi, and p(t) is the pressure transient induced in the residual tissue. The integral of pressure over time is dened as the specic impulse, noted here as I. Note that this expression treats the process as one-dimensional, i.e., it assumes all ablated material is ejected perpendicular to the target surface. It is worth noting that even completely isotropic ejection into 2 solid angle would reduce the imparted momentum by only half. Furthermore, under most practical conditions, A is large enough so that the ejection is fairly anisotropic and Equation 5.21 closely approximates the experimental result. All the material is initially assumed to have approximately the same ejection velocity. If m0 and k are dened respectively as the ablated mass per area and the kinetic energy per area (k = m0 v2 /2), then the specic impulse is simply: I = m0 v = 2m 0 k

(5.22)

The kinetic energy per area in the ablation plume obviously cannot exceed the uence F in the incident laser pulse. This fact places an absolute upper limit on I, namely I max = 2m 0 F . However, in reality, I is far less than Imax. The ratio of the specic impulse generated in the target to the incident laser pulse

126

Lasers in Medicine

uence (I/F) is dened as the impulse coupling coefcient. Typical values of the impulse coupling coefcient from experimental ablation measurements are 10-5 to 10-3 s m-1. It is illustrative to apply these ideas to the experimental photoacoustic study of ArF excimer laser cornea ablation mentioned earlier.59 In that work, laser-induced pressure waves were detected by a piezoelectric transducer on the rear surface of 30 m-thick corneal tissue slices. Irradiation of the sample with ablative 16 ns laser pulses produced pressure pulses of approximately 30 ns duration (full width at half maximum). At the highest experimental laser uence, 570 mJ/cm2, the amplitude of the transducer signal corresponded to peak pressure of about 100 ATM in the stress wave. In light of Equations 5.21 and 5.22, we can use these values to estimate the specic impulse generated in the cornea to be roughly 0.3 N s m2 (since 1 ATM _ 105 n m-2). The impulse coupling coefcient in this case is therefore ~5 x 105 N s J1. For comparison, the theoretical Imax under these conditions is an order of magnitude larger, about 3.4 n s m-2, based on an ablation depth of ~1 m @ 570 mJ/cm2 (see Figure 5.4) and assuming the density of cornea to be around 1 gm/cm3. I is always much less than Imax for at least three reasons (aside from any angular distribution in plume particle velocities). First, not all the energy originally in the laser pulse ends up in the ablation plume. Referring back to Figure 5.3, a nite part of the incident pulse is deposited deep in the tissue, beneath the ablated region. (An additional small fraction of the pulse is reected from the target surface.) Second, not all the laser energy absorbed in the ablation volume is converted into the kinetic energy of the debris. The energy density Eth is required just to photodecompose the target material. Third, the kinetic energy in the plume is not uniformly distributed over the entire mass, as would be the case for maximal momentum generation. Based on these considerations, Dingus has shown that the maximum value of the impulse coupling coefcient is 0.5 to 0.6 times (Eth/)1/2, where is the material density. In addition, this optimal impulse coupling typically occurs when the laser pulse uence is six to seven times the ablation threshold. Evaluation of these issues becomes much more complicated under tamped conditions (e.g., ablation of a tissue in liquid).94 Analysis of conservation of momentum is no longer straightforward because there are not two quickly separating masses (the plume and the residual target). The conning ambient medium keeps the vaporized material in contact with the remaining bulk, and much of the energy originally absorbed in the ablation region can be transferred back into the solid. In one study of metal ablation under tamped versus untamped conditions, the kinetic energy coupled into the target mass was 1000 times greater with tamping than without.94 Tissue ablation in a uid, as occurs during laser angioplasty, is a practical example of at least partially tamped. In general, such enhanced coupling is lessened if the tamping medium has a high thermal conductivity or the ablation plume contains condensable vapors that are likely to precipitate out of the plume on striking the cold tamping medium. Both these qualities facilitate rapid transfer of energy from the plume into the ambient environment rather than into the tissue. The shock wave generation accompanying UV-laser photoablation is a complex nonlinear process. Nevertheless, it is important to have some understanding of such waves because of their potential to cause damage in delicate tissues. A simple yet instructive conceptual picture of this phenomenon has been presented in Ref. 95, and the essential features of that description are shown in Figure 5.11. Consider a laser pulse of cross-sectional area A striking a target at time t = 0 and generating an ablation plume that then rushes out into the uid ambient environment at a velocity u. The environment initially has a uniform density, 0, as indicated in the lower left hand of the gure. As the plume pushes outward, a leading-edge bow wave of swept up uid is created. If the plume velocity is high, i.e., at least on the order of the ambient sound velocity, then the stress in the bow wave becomes large. Important to note here is that, at high compressive stress, the wave propagation velocity is no longer constant (the characteristic sound speed), and generally increases with increasing pressure. Therefore, as the amplitude of the bow wave increases, the region of peak pressure begins to move faster than the rest of the wave. This leads to a marked sharpening of the leading edge of the wave into a very thin layer with a nearly discontinuous jump in pressure (hence the damage potential), and moving at supersonic velocities. The developed shock wave is shown in the right half of Figure 5.11 at time t after the laser pulse. For simplicity, the shock front is modeled as a discontinuous layer of uniform

The Physics of Ultraviolet Laser Ablation

127

time = 0 target ambient medium

time = t

plume area = A laser pulse u U shock front

density

density

0 distance

?
ut Ut

FIGURE 5.11 Theoretical model of shockwave propagation into an ambient environment on density 0 following laser target ablation. The shock front is modeled as a thin layer of density 1 propagating at a velocity U. The expanding ablation plume, of unknown density, propagates at an average velocity u behind the shock front (see text).

density 1. The leading edge of the discontinuity propagates into the undisturbed ambient region at a velocity U. Because of conservation of mass, the total material swept ahead of the plume must be contained in the shock region: 1 A ( Ut ut ) = 0 AUt ,

(5.23)

meaning that the lightly shaded regions in the two graphs of Figure 5.11 have the same area. This mass acquires a momentum of 0 A Utu. Using Equation 5.21, we can relate the momentum per area in the compressed layer to the specic impulse associated with the driving pressure. If p0 and p1 denote respectively the pressures in the undisturbed ambient uid and the shock region, then the net outward pressure acting on the shock front is p1p2, and: 0 Utu = ( p 1 p 0 ) t .

(5.24)

Because p1 p0 under shock wave conditions, the ambient pressure is inconsequential,96 and Equation 5.24 simplies to: 0 Uu = p 1 .

(5.25)

Equation 5.25 is known as a jump condition, and is a very useful tool for calculating shock wave peak pressure based on velocity measurements.73,96 Note, however, that two velocities, those of the shock front and the driving particles, are required to perform the calculation. While the former is relatively straightforward to measure, the latter can be extremely difcult to obtain. Fortunately, the two speeds are related by the equation of state. To a rst-order approximation: U = a + bu ,

(5.26)

where a and b are empirically determined constants.96 For water, a = 1483 m/s (the speed of sound) and b = 2.07.97

128

Lasers in Medicine

Shortly after the laser pulse, the shock wave separates from the ablation plume, and, from that instant onward, the momentum in the shock region remains constant. This does not mean, however, that the shock wave persists indenitely. As soon as it is no longer being driven by the plume, both the pressure and velocity of the shock fall precipitously. This is, in part, due to a geometric effect: as the shock moves away from the ablation site, it evolves from a nearly plane wave into a spherical wave,76,90 and the intensity of such a wave must have an inverse dependence on radius. In addition, shock waves heat the medium they are propagating through,95 and this heating takes energy out of the wave. Such processes eventually reduce shock waves to conventional pressure waves.

5.5 Repetitive Irradiation

Medical ablation procedures employ a train of many laser pulses to achieve appreciable tissue removal. The quality of an ablation crater formed by repetitive irradiation, i.e., the depth, topography, and extent of tissue damage, depends on specic laser parameters, tissue characteristics, and ambient conditions, as described in the sections above. In general, it is assumed that each laser irradiation in a train of identicaluence pulses removes a consistent amount of tissue, so that the crater depth increases linearly with the number of irradiations. For certain synthetic polymers, such as PMMA, the relationship between crater depth and number of laser pulses is complicated by a phenomenon termed incubation, particularly so for pulse uences just above the ablation threshold. Incubation is a process whereby multiple irradiations are required to initiate etching at modest laser intensities. This effect has been extensively studied44,98100 and has been shown to be due to progressive photochemical alterations within the irradiated material (i.e., fragmentation of large polymer molecules, formation of new UV absorbers, and accumulation of trapped gases). As for tissue targets, incubation has specically been shown not to occur during 193 nm corneal ablation,69 but may play some role in vascular tissue irradiation at 308 nm, where irradiation-induced long-lived changes in target optics have been observed.55 Another process, not yet documented for ultraviolet laser ablation of tissue but known to occur with synthetic polymers, is radiation hardening, where material removal occurs only for initial pulses in a train of irradiations. This phenomenon has been extensively observed for polyimide ablation, and, like incubation, is most pronounced for pulse uences in the vicinity of the ablation threshold. There are at least two causes for radiation hardening. First, weakly ablative polyimide irradiation ejects low molecular weight gases and leaves behind a surface layer of carbon crystallites.101 This carbon layer is much more resistant to ablation than the original polymer. Second, low uence ablation of polyimide leads to a progressive roughening of the crater oor through the formation of randomly scattered conical features,102104 as well as periodic ripple-like surface waves.105 Roughening of the surface increases the irradiation area, and hence decreases the effective uence striking the target. For tissue ablation, one established important consideration is the maximum repetition rate that can safely be used in a clinical procedure. Referring once again to Figures 5.1 and 5.3, some fraction of the incident laser energy for each pulse is deposited in the intact tissue below the ablation depth. This energy is primarily converted to heat. If the subsequent laser pulse arrives before sufcient thermal diffusion into the surrounding bulk tissue occurs (Equation 5.1), a pronounced localized temperature rise, and hence thermal damage, can occur. In one study of ArF laser corneal ablation using a microthermocouple technique,53 a temperature rise of 20 C was obtained during excised tissue irradiation at 30 Hz with 360 mJ/cm2 laser pulses. Not surprisingly, the heating was found to increase with either increasing laser pulse frequency or pulse uence. A major conclusion of that study was that the pulse repetition rate should be kept below 63 Hz for corneal sculpting procedures for conventional ArF laser uences (below ~200 mJ/cm2 per pulse).

The Physics of Ultraviolet Laser Ablation

129

References
1. Kawamura, Y., K. Toyoda, and S. Namba, Effective deep ultraviolet photoetching of polymethyl methacrylate by an excimer laser, Appl. Phys. Lett. 40, 374-375 (1982). 2. Srinivasan, R. and V. Mayne-Banton, Self-developing photoetching of poly(ethylene terephthalate) lms by far-ultraviolet excimer laser radiation, Appl. Phys. Lett. 41, 576-578 (1982). 3. Trokel, S.L., R. Srinivasan, and B. Braren, Excimer laser surgery of the cornea, Am. J. Ophthalmol. 96, 710-715 (1983). 4. Grundfest, W.S. et al., Laser ablation of human atherosclerotic plaque without adjacent tissue injury, J. Am. Coll. Cardiol. 5, 929-933 (1985). 5. Srinivasan, R., Ablation of polymers and biological tissue by ultraviolet lasers, Science 234, 559565 (1986). 6. Srinivasan, R., B. Braren, and R.W. Dreyfus, Ultraviolet laser ablation of polyimide lms, J. Appl. Phys. 61, 372-376 (1987). 7. Brannon, J.H. et al., Excimer laser etching of polyimide, J. Appl. Phys. 58, 2036-2043 (1985). 8. Andrew, J.E. et al., Direct etching of polymeric materials using a XeCl laser, Appl. Phys. Lett. 43, 717 (1983). 9. Kper S., and J. Brannon, Excimer laser ablation of polyimide: threshold behavior studied by single-shot-rate measurements, Conference on Lasers and Electro-Optics (CLEO), May 12-17, 1991, Baltimore, MD, Technical Digest Series 10, 506 (1991). 10. Pettit, G.H. and M.N. Ediger, Corneal tissue absorption coefcients for 193 nm and 213 nm ultraviolet radiation, Appl. Opt. 35, 3386-3391 (1996). 11. Puliato, C.A. et al., Excimer laser ablation of the cornea and lens, Ophthalmology 92, 741-748 (1985). 12. Cole, H.S., Y.S. Liu, and H.R. Philipp, Dependence of photoetching rates of polymers at 193 nm on optical absorption depth, Appl. Phys. Lett. 48, 76-77 (1986). 13. Furzikov, N.P., Different lasers for angioplasty, thermooptical comparison, IEEE J. Quantum. Electron. QE-23, 1751-1755 (1987). 14. McKenzie, A.L., Physics of thermal processes in lasertissue interaction, Phys. Med. Biol. 35, 11751209 (1990). 15. Laufer, G. et al., Characteristics of 308 nm excimer laser activated arterial tissue under ablative and non-ablative conditions. Lasers Surg. Med. 9, 556-571 (1989). 16. Taylor, R.S., A.J. Higginson, and K.E. Leopold, Dependence of the XeCl laser cut rate of plaque on the degree of calcication, laser uence, and optical pulse duration, Lasers Surg. Med. 10, 414-419 (1990). 17. Taylor, R.S. et al., Damage measurements of fused silica bers using long optical pulse XeCl lasers, Opt. Commun. 63, 26-31 (1987). 18. Taylor, R.S., D.L. Singleton, and G. Paraskevopoulos, Effect of optical pulse duration on the XeCl laser ablation of polymers and biological tissue, Appl. Phys. Lett. 50, 1779-1781 (1987). 19. Litvack, F. et al., Pulsed laser angioplasty: Wavelength power and energy dependencies relevant to clinical application, Lasers Surg. Med. 8, 60-65 (1988). 20. Gailitis, R.P. et al., Solid state ultraviolet laser (213 nm) ablation of the cornea and synthetic collagen lenticules, Lasers Surg. Med. 11, 556-562 (1991). 21. Hale, G.M. and M.R. Querry, Optical constants of water in the 200-nm to 200-m wavelength region, Appl. Opt. 12, 555-563 (1973). 22. Staveteig, P.T., R.K. Shori, and J.T. Walsh, Potential Role of Dynamic Water Absorption During 193-nm ArF Excimer Laser Ablation of Tissue Phantoms, in Conference on Lasers and Electro-Optics, Vol. 15, 1995 OSA Technical Digest Series (Optical Society of America, Washington, D.C., 1995), pp. 417-418.

130

Lasers in Medicine

23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51.

Solomons, T.W.G., Organic Chemistry, John Wiley & Sons, New York (1980). Kochevar, I.E., Biological effects of excimer laser radiation. Proc. IEEE, 80: 833-837 (1992). Lehninger, A.L., Biochemistry, Worth Publishers, New York (1975). Smith, K.C. and P.C. Hanawalt, Molecular Photobiology, Academic Press, New York (1969). Oraevsky, A.A. et al., XeCl laser ablation of atherosclerotic aorta: optical properties and energy pathways, Lasers Surg. Med. 12, 585-597 (1992). Pettit, G.H. and R. Sauerbrey, Fluence-dependent transmission of polyimide at 248 nm under laser ablation conditions, Appl. Phys. Lett. 58, 793-795 (1991). Ediger, M.N. et al., Transmission of corneal collagen during ArF excimer laser ablation, Lasers Surg. Med. 13, 204-210 (1992). Gorodetsky, G. et al., Calorimetric and acoustic study of ultraviolet laser ablation of polymers, Appl. Phys. Lett. 46, 828-830 (1985). Srinivasan, R. et al., Photochemical cleavage of a polymeric solid: details of the ultraviolet laser ablation of poly (methyl methacrylate) at 193 and 248 nm, Macromolecules 19, 916-921 (1986). Puliato, C.A., K. Wong, and R.F. Steinert, Quantitative and ultrastructural studies of excimer laser ablation of the cornea at 193 and 248 nanometers, Lasers Surg. Med. 7, 155-159 (1987). Kitai, M.S. et al., The physics of UV laser cornea ablation, IEEE J. Quantum Electron. QE-27, 302307 (1991). Kaufman, H.E. et al., The Cornea, Churchill Livingstone, New York (1988). Krueger, R.R. and S. L. Trokel, Quantitation of corneal ablation by ultraviolet laser light, Arch. Ophthalmol. 103, 1741-1743 (1985). Aron-Rosa, D.S. et al., Excimer laser surgery of the cornea: quantitative and qualitative aspects of photoablation according to the energy density, J. Cataract Refract. Surg. 12, 27-33 (1986). Seiler, T., T. Bende, and J. Wollensack, Astigmatismus-korrektur mit dem Excimer Laser, Klin. Monatsbl. Augenheilkd. 191, 179-183 (1987). Fantes F. E. and G. O. Waring, Effect of excimer laser radiant exposure on uniformity of ablated corneal surface, Lasers Surg. Med. 9, 533-542 (1989). Fuxbruner, A. et al., Controlled lens formation with unapertured excimer lasers: use with organic polymers and corneal tissue, Appl. Opt. 29, 5380-5386 (1990). Van Saarlos, P.P. and I.J. Constable, Bovine corneal stroma ablation rate with 193-nm excimer laser radiation: quantitative measurement, Refract. Corneal Surg. 6, 424-429 (1990). Unkroth, A., et al., Ablation of the cornea by using a low-energy excimer laser, Graefes Arch. Clin. Exp. Ophthalmol. 231, 303-307 (1993). Puliato, C.A. et al., High-speed photography of excimer laser ablation of the cornea, Arch. Ophthalmol. 105, 1255-1259 (1987). Pettit, G.H. et al., Transmission of polyimide during pulsed ultraviolet laser irradiation, Appl. Phys. A 58, 573-579 (1994). Kper, S. and M. Stuke, Femtosecond UV excimer laser ablation, Appl. Phys. B 44, 199-204 (1987). Davis, G.M. and M.C. Gower, Time resolved transmission studies of poly (methyl methacrylate) lms during ultraviolet laser ablative photodecomposition, J. Appl. Phys. 61, 2090-2092 (1987). Pettit, G.H. and R. Sauerbrey, Pulsed ultraviolet laser ablation, Appl. Phys. A 56, 51-63 (1993). Pettit, G.H. and M.N. Ediger, Corneal tissue reectance during argon uoride excimer laser irradiation, Lasers Light Ophthalmol. 6, 143-147 (1994). Ediger, M.N., G.H. Pettit, and D.W. Hahn, Enhanced ArF laser absorption in a collagen target under ablative conditions, Lasers Surg. Med. 15, 107-111 (1994). Dlott, D.D., Ultrafast vibrational energy transfer in the real world: laser ablation, energetic solids, and hemeproteins, J. Opt. Soc. Am. B. 7, 1638-1652 (1990). Venugopalan, V., N.S. Nishioka, and B.B. Mikic, The thermodynamic response of soft biological tissue to pulsed ultraviolet laser irradiation, Biophys. J. 69, 1259-1271 (1995). Dyer, P.E. and R.K. Al-Dhahir, Transient photoacoustic studies of laser tissue ablation, Proc. SPIE 1202, 46-60 (1990).

The Physics of Ultraviolet Laser Ablation

131

52. Dyer, P.E. and J. Sidhu, Excimer laser ablation and thermal coupling efciency to polymer lms, J. Appl. Phys. 57, 1420-1422 (1985). 53. Bende,T., T. Seiler, and J. Wollensak, Side effects in excimer corneal surgery; corneal thermal gradients, Graefes Arch. Clin. Ophthalmol. 226, 277-280 (1988). 54. Wolgin, M. et al., Excimer ablation of human intervertebral disc at 308 nm, Lasers Surg. Med. 9, 124-131 (1989). 55. Schwarzmaier, H.J. et al., Optical density of vascular tissue before and after 308-nm excimer laser irradiation, Opt. Eng. 31, 1436-1440 (1992). 56. Albagli, D. et al., Inertially conned ablation of biological tissue, Lasers Life Sci. 6, 55-68 (1994). 57. Edwards, G. et al., Tissue ablation by a free-electron laser tuned to the amide II band, Nature 371, 416-419 (1994). 58. Cross, F.W. et al., Time-resolved photoacoustic studies of vascular tissue ablation at three laser wavelengths, Appl. Phys. Lett. 50, 1019-1021 (1987). 59. Srinivasan, R., P.E. Dyer, and B. Braren, Far-ultraviolet laser ablation of the cornea: photoacoustic studies, Lasers Surg. Med. 6, 514-519 (1987). 60. Dyer, P.E. and R. Srinivasan, Nanosecond photoacoustic studies on ultraviolet laser ablation of organic polymers, Appl. Phys. Lett. 48, 445-447 (1986). 61. Simon, P., Time-resolved ablation-site photography of XeCl-laser irradiated polyimid, Appl. Phys. B. 48, 253-256 (1989). 62. Paraskevopoulos, G. et al., Time resolved reectivity as a probe of the dynamics of laser ablation of organic polymers, J. Appl. Phys. 70, 1938-1945 (1991). 63. Pettit, G.H., M.N. Ediger, and R.P. Weiblinger, Dynamic optical properties of collagen-based tissue during ArF excimer laser ablation, Appl. Opt. 32, 488-493 (1993). 64. Pettit, G.H. et al., Atherosclerotic tissue analysis by time-resolved XeCl excimer laser reectometry, Lasers Surg. Med. 13, 279-283 (1993). 65. Singleton, D.L., G. Paraskevopoulos, and R.S. Taylor, Dynamics of excimer laser ablation of polyimide determined by time-resolved reectivity, Appl. Phys. B 50, 227-230 (1990). 66. Ediger, M.N. and G.H. Pettit, Time-resolved reectivity of ArF laser-irradiated polyimide, J. Appl. Phys. 71, 3510-3514 (1992). 67. Hahn, D.W., G.H. Pettit, and M.N. Ediger, Optical properties of polyimide during ArF excimer laser ablation, J. Appl. Phys. 76, 1830-1832 (1994). 68. Ball, Z. et al., Transient optical properties of excimer-laser-irradiated polyimide, Appl. Phys. A 61, 547-551 (1995). 69. Pettit, G.H., M.N. Ediger, and R.P. Weiblinger, Excimer laser corneal ablation: absence of a significant incubation effect, Lasers Surg. Med. 11, 411-418 (1991). 70. Ediger, M.N., G.H. Pettit, and R.P. Weiblinger, Noninvasive monitoring of excimer laser ablation by time-resolved reectometry, Refractive and Corneal Surgery 9, 268-275 (1993). 71. Dingus, R.S., Grneisen-stress induced ablation of biological tissue, Proc. Laser-Tissue Interaction II, SPIE vol. 1427, 45-55 (1991). 72. Albagli, D. et al., Interferometric surface monitoring of biological tissue to study inertially conned ablation, Lasers Surg. Med. 14, 374-385 (1994). 73. Zweig, A.D. and T.F. Deutsch, Shock waves generated by conned XeCl excimer laser ablation of polyimide, Appl. Phys. B 54, 76-82 (1992). 74. Dingus, R.S., Momentum induced by laser-tissue interaction, Proc. SPIE (1993). 75. Bor, Z. et al., Time resolved study of surface shock wave formation during excimer laser ablation of the cornea, Conference on Lasers and Electro-Optics (CLEO), May 10-15, 1992, Anaheim, CA, Technical Digest Series 10, 514 (1992). 76. Srinivasan, R. et al., Ultrafast imaging of ultraviolet laser ablation and etching of polymethylmethacrylate, Appl. Phys. Lett. 55, 2790-2791 (1989).

132

Lasers in Medicine

77. Zweig, A.D. and T.F. Deutsch, Shock waves generated by XeCl excimer laser ablation of polyimide in air and water, Conference on Lasers and Electro-Optics (CLEO), May 10-15, 1992, Anaheim, CA, Technical Digest Series 10, 512 (1992). 78. Hahn, D.W., M.N Ediger, and G.H. Pettit, Dynamics of ablation plume particles generated during excimer laser corneal ablation, Lasers Surg. Med. 16, 384-389 (1995). 79. Kermani, O. et al., Mass spectroscopic analysis of excimer laser ablated material from human corneal tissue, J. Cataract Refract. Surg. 14, 638-641. 80. Singleton, D.L., G. Paraskevopoulos, and R.S. Irwin., XeCl laser ablation of polyimide: inuence of ambient atmosphere on particulate and gaseous products, J. Appl. Phys. 66, 3324 (1989). 81. Hahn, D.W., M.N. Ediger, and G.H. Pettit, Ablation plume particle dynamics during excimer laser ablation of polyimide, J. Appl. Phys. 77, 2759-2766 (1995). 82. Pettit, G.H.et al., Electron paramagnetic resonance spectroscopy of free radicals in corneal tissue following excimer laser irradiation, Lasers Surg. Med. 18, 367-372 (1996). 83. Nakagawa, K. Direct observation of laser generated free radicals from a myocardium target site, Free Radical Biol. Med. 12, 241-242 (1992). 84. Jain, S. et al., Modulation of corneal wound healing following 193-nm excimer laser keratectomy using free radical scavengers. Invest. Ophthalmol. Vis. Sci. 35, 2015 (1994). 85. Danielzik, B. et al., Velocity distribution of molecular fragments from polymethylmethacrylate irradiated with UV laser pulses, Appl. Phys. Lett. 48, 212-214 (1986). 86. Larciprete, R. and M. Stuke, Direct observation of excimer-laser photoablation products from polymers by picosecond-UV-laser mass spectroscopy, Appl. Phys. B 42, 181-184 (1987). 87. van Leeuwen, T.G. et al., Origin of arterial wall dissections induced by pulsed excimer and midinfrared laser ablation in the pig, J. Am. Coll. Cardiol. 19, 1610-1618 (1992). 88. van Leeuwen, T.G. et al., Intraluminal vapor bubble induced by excimer laser pulse causes microsecond arterial dilation and invagination leading to extensive wall damage in the rabbit, Circulation 87, 1258-1263 (1993) 89. Srinivasan, R., K.G. Casey, and J.D. Haller, Subnanosecond probing of the ablation of soft plaque from arterial wall by 308 nm laser pulses delivered through a ber, IEEE J Quant. Electron. 26, 2279-2283 (1990). 90. Srinivasan, R. et al., The signicance of a uence threshold for ultraviolet laser ablation and etching of polymers, J. Appl. Phys. 67, 1604-1606 (1990). 91. Lustmann, J. et al., Photoacoustic injury and bone healing following 193 nm excimer laser ablation, Lasers Surg. Med. 12, 390-396 (1992). 92. Yashima, Y. et al., Laser-induced photoacoustic injury of skin: effect of inertial connement, Lasers Surg. Med. 11, 62-68 (1991). 93. Martinez, M. et al., A comparison of excimer laser (308 nm) ablation of the human lens nucleus in air and saline with a ber optic delivery system, Refract. Corn. Surg. 8, 368-374 (1992). 94. Dingus, R.S., Laser-induced contained vaporization in tissue, Proc. LaserTissue Interaction III, SPIE vol. 1646, 266-274 (1992). 95. Zeldovich, Y.B. and Y.P. Raizer, Physics of Shock Waves and High-Temperature Hydrodynamic Phenomena, 45-49, Academic Press, New York (1966). 96. Doukas, A.G. et al., Non-invasive determination of shock wave pressure generated by optical breakdown, Appl. Phys. B 53, 237-245 (1991). 97. Harris, P. and H.N. Presles, Reectivity of a 5.8 kbar shock front in water, J. Chem. Phys. 74, 68646866 (1981). 98. Sutcliffe, E. and R. Srinivasan, Dynamics of UV laser ablation of organic polymer surfaces, J. Appl. Phys. 60, 3315-3322 (1986). 99. Kper, S. and M. Stuke, UV excimer laser ablation of polymethylmethacrylate at 248 nm: characterization of incubation sites with Fourier transform IR and UV spectroscopy, Appl. Phys. A 49, 211-215 (1989).

The Physics of Ultraviolet Laser Ablation

133

100. Srinivasan, R., B. Braren, and K.G. Casey, Nature of incubation pulses in the ultraviolet laser ablation of polyimide, J. Appl. Phys. 68, 1842-1847 (1990). 101. Ball, Z. et al., Transient optical properties of excimer-laser-irradiated polyimide II: carbon cluster scattering, Appl. Phys. A 61, 575-578 (1995). 102. Dyer, P.E., S.D. Jenkins, and J. Sidhu, Development and origin of conical structures on XeCl laser ablated polyimide, Appl. Phys. Lett. 49, 453-455 (1986). 103. Taylor, R.S. et al., The effect of debris formation on the morphology of excimer laser ablated polymers, J. Appl. Phys. 64, 2815-2818 (1988). 104. Krajnovich, D.J. and J.E. Vzquez, Formation of intrinsic surface defects during 248 nm photoablation of polyimide, J. Appl. Phys. 73, 3001-3008 (1993). 105. Dyer, P.E. and R.J. Farley, Periodic surface structures in the excimer laser ablative etching of polymers, Appl Phys. Lett. 57, 765-767 (1990).

6
Tissue Diagnostics Using Lasers
6.1 6.2 Introduction ....................................................................... 135 Light Interaction with Tissue ............................................ 136
Molecular Absorption of Laser Light Laser-Induced Tissue Heating Laser-Induced Photochemistry Laser-Induced Fluorescence Elastic Scattering Raman Scattering

6.3

Spectroscopic Diagnostics of Malignant Tumors ............ 141

Basic Considerations Instrumentation for Point Monitoring of Laser-Induced Fluorescence Point Monitoring of Cancer using Laser-Induced Fluorescence Equipment for Imaging Multi-Color Fluorescence Imaging of Tumors

6.4

Spectroscopic Diagnostics of Atherosclerotic Plaque...... 151

Basic Considerations Plaque Discrimination Using LaserInduced Fluorescence Time-Resolved Monitoring Raman Spectroscopy for Tissue Diagnostics Tissue Reectance Monitoring

6.5

Light Scattering and Tissue Transillumination................ 157

Doppler-Flow Monitoring Transillumination Imaging Interpretation of Time-Resolved Tissue Recordings

Sune Svanberg
Lund Institute of Technology

6.6 Outlook ............................................................................... 163 Acknowledgments......................................................................... 164 References ...................................................................................... 165

6.1 Introduction
While therapeutic aspects of lasers in medicine have been very dominant, tissue diagnostic techniques using laser spectroscopic techniques are now emerging strongly. Applications include endoscopic cancer detection using uorescence, plaque monitoring in the circulatory system including coronary arteries, and tissue transillumination for optical mammography. Several factors contribute to the increasing interest in laser tissue diagnostics: Optical monitoring is non-intrusive in nature. Non-ionizing radiation is employed. Real-time data representation is possible. Spectroscopy allows molecular specicity in analysis. Point monitoring or imaging capability can be employed. Integration of laser diagnostics and therapy is possible.

0-8493-1146-2/02/$0.00+$1.50 2002 by CRC Press LLC

135

136

Lasers in Medicine

Sometimes the term optical biopsy, which also gives the avor of the attractive features of the new emerging methods, is used. In comparison with well established diagnostic modalities such as x-ray radiography, NMR imaging, ultrasound imaging and positron emission tomography,1 the optical spectroscopic techniques are in an early stage of development. However, through extensive work at several research centers, the optical techniques are developing swiftly, and clinically useful equipment is now becoming available. The organization of the present chapter is as follows: Light interaction with tissue, forming the basis for optical tissue diagnosis, is reviewed in Section 6.2. In a following section, malignant-tumor diagnostics based on laser-induced uorescence is treated, where point monitoring and multi-spectral imaging, as well as time-resolved aspects, are included. Monitoring of the interior of vessel walls and of the heart muscle is treated in Section 6.4, with a discussion of the possibilities of integration of diagnostics with a treatment modality. Finally, light scattering in tissue transillumination is treated in Section 6.5, which includes Doppler perfusion monitoring, optical dosimetry for photodynamic therapy and the detection of deep-lying malignant breast tumors using time-gated viewing. An outlook for the future concludes the chapter.

6.2 Light Interaction with Tissue

Optical diagnostics for medical purposes is based on the interaction of light with tissue. The interactions can be resonant or non-resonant in nature: Resonant Molecular absorption of laser light Laser-induced tissue heating Laser-induced photochemistry Laser-induced uorescence Non-Resonant Elastic scattering Raman scattering

Lightmatter interaction mechanisms, spectroscopic instrumentation and measurement techniques are treated in text books on atomic and molecular spectroscopy.2 Here, we will now discuss the different modes of interaction between light and tissue.

6.2.1

Molecular Absorption of Laser Light

When tissue is irradiated by laser light, a small fraction of the light is reected, but most of it penetrates into the tissue, where it is either absorbed of scattered by the molecules. In certain wavelength regions, absorption strongly dominates over scattering. The stronger the absorption, the smaller the penetration depth into the tissue will be. A schematic diagram of the absorption of some important biological constituents and chromophores, such as water, proteins, nucleic acids, hemoglobin and melanin is shown in Figure 6.1.3 Here also, the emission wavelengths of some important medical lasers are included. As can be seen, water has two regions of strong absorption, one in the deep ultraviolet (UV) region and one in the infrared (IR) region. The absorption maximum in the case of the aromatic rings of proteins and the nucleic acids is in the UV region between 260 and 280 nm. Thus, in the UV region, the laser light is highly absorbed by the water and proteins in the tissue, resulting in a poor light penetration into the tissue. The same is true in the IR region, starting at the second water absorption region for water at about 1.3 m. Hemoglobin in the blood absorbs light in a broad wavelength region up to red light (about 600 nm), with a pronounced absorption maximum at about 400 nm and additional weaker peaks in the green/yellow region. Above 600 nm, the absorption of blood is weak. Melanin, the most important chromophore in the epidermis,

Tissue Diagnostics Using Lasers

137

ENERGY 102
MOLAR EXTINCTION COEFFICIENT e(1031/mol-cm)

10 8

E(av) 1 0.8 0.6

0.4

0.2

0.1

104 10-17 H2O

CROSS SECTION (cm2)
DOPAMelanin HbO2 Adenine

106

10-18

108

10-19 H2O 10-20

10

Krf Nd-YAG (4 w) XeCl

Argon Copper Argon Nd-YAG (4 w) Copper Gold He-Ne

HPD
Ga-As Nd-YAG

Dye

HbO2
CO2

10-21

Arf

10-1

10-22

10-2 100

200

300

600 700 1000 2000 WAVELENGTH l(nm)

4000

10000

FIGURE 6.1 Absorption properties of biological molecules: water, hemoglobin, melanin, hematoporphyrin derivative (HPD) and adenine (From Ref. 3).

absorbs light in a region from UV to near IR. Between 600 nm and 1.3 m, the attenuation coefcients of the tissue molecules are comparatively small, resulting in a wavelength region with favorable light penetration properties, with increasing penetration toward the IR region. Considering the total tissue absorption, it can be noted that the human body is as transparent as can be at about 1 m. The absorptive properties of a surface can also be monitored indirectly in reectance. This is a wellknown phenomenon that forms the basis for the human perception of colors. For example, the red color of blood can be understood from the absorption curve given in Figure 6.1, where all colors below 600 nm in the impinging white light are strongly absorbed by the hemoglobin molecules. Red light, however, is not absorbed, but can be scattered, reaching the eye of the observer. In practice, reectance spectroscopy performed by the doctors eye and coupled to the processing of an experienced brain is a most important means of assessing human tissue status. The technique can be rened by using a spectrometer, which also gives access to the IR region, where characteristic absorption bands due to molecular vibrational transitions modify the impinging radiation. This technique can basically yield the same type of information as Raman spectroscopy, discussed below. Analysis of reected light forms the basis for global earth-resource monitoring using passive satellite sensors.4

6.2.2 Laser-Induced Tissue Heating

Laser-induced tissue heating occurs through the conversion of electromagnetic energy into thermal energy. This photothermal process takes place when biomolecules have absorbed light quanta to arrive in an excited state and return nonradiatively to lower excited levels, transferring their energy to the tissue. In the thermal mode of tissue interaction the choice of wavelength, as well as the tissue type, determine the light penetration depth. Heating leads to protein coagulation (60 C), body uid boiling (100 C) and tissue carbonization (200 C) that can be used for hemostasis and tissue surgery. This is a major application of lasers in medicine, having much impact in opthtalmology, dermatology and general surgery. The carbon dioxide laser (10.6 m) has a very low tissue penetration due to water absorption, as has the argon-ion laser (488, 515 nm), due to hemoglobin. If a dye laser is tuned to 580 nm, it can be made to be absorbed quite selectively by the hemoglobin in small vessels (see Figure 6.1). Thus,

138

Lasers in Medicine

hemangioma, such as port-wine stains, can be efciently treated while sparing surrounding, normal tissue. Maximum penetration occurs for the Nd:YAG laser (1.06 m), which induces wide coagulation zones around surgical cuts. The various clinical applications of lasers operating in the thermal mode are discussed in many chapters of this book. For the purpose of tissue diagnostics using laser techniques, thermal effects should be avoided by keeping the laser uence sufciently low. We include this brief discussion of laser thermal interaction only to clarify the limitations of optical tissue diagnostics. For UV wavelengths, the penetration depth of laser light is only of the order of 1100 m. Such radiation can be used to induce uorescence in the biomolecules, which provides an important means of tissue diagnostics. However, high incident uences of such radiation in short pulses leads to explosive tissue ablation, and must thus be avoided in all diagnostic work. Laser tissue ablation, on the other hand, has important medical applications for myopia correction by corneal reshaping and as a modality for clearing occluded vessels using transluminal optical bers, as discussed in Chapters 85,6and 9.

6.2.3 Laser-Induced Photochemistry

Optically excited molecules can, instead of releasing their excess energy as heat, transfer their energy to other molecules, thus inducing chemical reactions of therapeutic interest. Optical treatment of psoriasis patients or newborn infants with an excess of bilirubin are well-known examples of medical photochemistry. Photodynamic tumor therapy (PDT) using tumor-seeking agents is a further example of special interest in our context. The most commonly used tumor sensitizer is hematoporphyrin derivative (HPD), which is commercially available under the trade name of Photofrin. The absorption curve of HPD is included in Figure 6.1. When injected intravenously, such molecules are selectively retained to a higher degree in malignant tumor tissue than in other surrounding tissue. It can be noted in the gure that the absorption maximum of HPD is around 400 nm, similar to hemoglobin in the blood, but also, that HPD exhibits four more absorbing peaks, with the last one at 630 nm, where blood absorption is strongly reduced. Therefore, to penetrate deeper into the tumor tissue, laser light of 630 nm instead of 400 nm is used. A schematic diagram showing the energy levels of the HPD molecule with absorption and uorescence emission wavelengths is shown in Figure 6.2.7 Following light irradiation, HPD molecules become excited. The molecules can then be de-excited following two main pathways. In the rst, the molecules are transferred nonradiatively to the bottom of the rst excited band, from whence a radiative decay follows with the emission of a dual-peaked uorescent light distribution in the red spectral region. This emission can be used for the diagnostics of malignant tumors, because the characteristic red uorescence follows only from irradiated tissue that has retained the HPD molecules. This diagnostic aspect will be discussed extensively in the next subsection. In a second de-excitation pathway, the molecules are transferred into the triplet state. From here, the excitation energy can be transferred to ground-state tissue oxygen molecules that are excited into their singlet state. Singlet state oxygen is known to be an extremely toxic agent for living cells. As the HPD in the cells mediates this energy transfer, which results in singlet-oxygen formation, the cells are transformed from viable malignant cells to nonviable necrotic cells. This is the essence of photodynamic therapy, which is now being pursued on an experimental basis at many centers throughout the world. The therapeutic aspect, as well as the diagnostic, are schematically illustrated in Figure 6.3.7 Reviews of clinical PDT have been given in Refs. 8 and 9. The eld is further treated in Chapter 10 of this volume.10

6.2.4 Laser-Induced Fluorescence

Laser-excited molecules can return to the ground state by emitting uorescent light. Fluorescence spectroscopy is treated in detail in Refs. 11 and 12. The uorescence always occurs for longer wavelengths (Stokes shift). For large biomolecules, the uorescence normally exhibits a rather structureless intensity distribution as a function of wavelength, reecting the distribution of substates in the ground electronic level. However, by choosing the excitation wavelength properly and accurately analyzing the uorescent light distribution, it is frequently possible to obtain diagnostic information for the tissue molecules. Using short-pulse exci-

Tissue Diagnostics Using Lasers

139

HPD
S2

O2

S1
405 nm 500 nm 530 nm 570 nm 630 nm 690 nm

T1

ENERGY TRANSFER

a1g
1270 nm

S0

x 3g
SINGLET OXYGEN INDUCES TISSUE NECROSIS

600

700 nm

HPD FLUORESCENCE

FIGURE 6.2 Energy level diagram for the HPD molecule and illustration of HPD uorescence emission and energy transfer to oxygen molecules.7

tation, it is also possible to measure the decay time for the uorescent light, yielding additional information. Many different chromophores may contribute to the tissue signal: endogenous molecules such as NADH, NAD+, collagen and elastin, giving rise to so-called auto-uorescence, and additional tumor-seeking agents such as hematoporphyrin derivative, phtalocyanines and chlorins, all featuring sharp, characteristic signals in the red and near-IR region. Examples of uorescence spectra for different pure tissue substances and of tumor-seeking agents injected into the same type of tumor-bearing rats are shown in Figure 6.4.13,14

6.2.5 Elastic Scattering

Light scattering in tissue is important for red and near-IR light, which is not strongly absorbed by the tissue constituents. The scattering is normally elastic, i.e., no change in wavelength occurs. If absorption is low, many scattering processes may occur in a more or less random way before the photon is ultimately absorbed or escapes from the tissue. The fate of an individual photon is determined in a complex interplay between scattering and absorption. For certain geometries, the process can be described by solving the diffusion equation, specifying values for the absorption and scattering coefcients a and s, respectively. Of special interest is the study of transillumination through a thick tissue sample (few cm). Many of the photons have been scattered many times, resulting in effective pathlengths of tens of centimeters. If large computer capacity is available, the photon propagation can be simulated using Monte Carlo techniques, where photon trajectories are followed for a large number of events to build up an intensity distribution. In such calculations, the mean cosine value between the incoming and scattered photon directions of propagation is also included.1519 The presence of scattering makes tissue absorption measurements much more involved, as discussed above. The pathlength in tissue that has been traversed cannot be assessed in steady-state measurements, making a direct use of the Beer-Lambert absorption relation impossible. Time-resolved monitoring of picosecond pulses, on the other hand, allow well-dened measurements, which are of particular interest for measuring tissue oxygenation, using absorptive spectral features in the red/near-IR region as shown for oxygenated hemoglobin in Figure 6.1. De-oxygenated hemoglobin has a different spectral structure.

140

Lasers in Medicine

Inject HPD 3 mg/kg

Wait 2-3 days

HPD is selectively retained in tumors

Autofluorescence UV Laser 400 630 nm 600 HPD signal nm

HPD

O2 Singlet Oxidation Triplet

FIGURE 6.3 Schematic diagram of tumor marking by sensitizing molecules and the subsequent use for uorescence diagnosis and photodynamic therapy.7

If light is scattered against a moving particle, a very small frequency shift proportional to the in-line velocity occurs due to the Doppler effect. The only important moving particles in the body are the red blood cells. In the supercial capillary vascular bed accessible for non-intrusive light scattering measurements, the ow velocities are low, typically 1 mm/s. For the helium-neon laser wavelength (633 nm) this corresponds to a frequency shift of only about 3 kHz out of about 5 1014 Hz. Measurements of such small shifts can be accomplished by a heterodyne technique, where the beat frequency between the frequency-shifted light and the unshifted light scattered from the xed cell structures is observed. Since the scattering geometry is undened, an average of different ow velocities and ow directions is obtained, yielding an effective signal that is related to the supercial blood perfusion of the investigated tissue. Laser Doppler techniques for tissue perfusion measurements are treated in Refs. 20 and 21.

6.2.6. Raman Scattering

Raman scattering is inelastic, leading to scattered photons of strongly changed wavelength. It may be seen as an excitation to a virtual (non-resonant) state and an immediate transfer from this state to a lower state of different energy from that of the initial state. In cases of medical interest, the lower state splittings are due to vibrational energy. The resulting shifts are characteristic for vibrations in specic molecular groups. Although Raman scattering provides a much higher molecular specicity than laserinduced uorescence, there is a sensitivity problem, because Raman scattering is many orders of magnitude weaker than laser-induced uorescence. Thus, unless proper precautions are taken, the signals tend to be swamped by the uorescence signals.

Tissue Diagnostics Using Lasers

141

COLLAGEN NADH ELASTIN FLUORESCENCE INTENSITY

CAROTHENE

a)

Trp x1/2

350

400

450

500 550 WAVELENGTH (nm)

600

650

700

b)
FLUORESCENCE INT. (arb. units)

HP

DHE PHE

BPD-MA

TSPc

450

550 650 WAVELENGTH (nm)

750

FIGURE 6.4 Fluorescence spectra for (a) pure tissue constituent molecules and (b) a number of tumor sensitizing agents, accumulated in the same type of experimental rat tumor. (From Refs 13,14). Trp: tryptophan, HP: hematoporphyrin, DHE: dihematoportphyrin ether (or Photofrin), PHE: polyhematoporphyrin ester, BPD-MA: bensoporphyrin derivative mono acid, TSPc: tetra sulphonated phtalocyanine.

6.3 Spectroscopic Diagnostics of Malignant Tumors

6.3.1 Basic Considerations
This section will focus on tumor diagnostics using laser-induced uorescence, and will comment only briey on Raman spectroscopy, which is presently less developed. At rst glance, it might seem a formidable task to derive any information from tissue uorescence, considering the many types of complex organic molecules that make up living tissue. We will rst present typical tissue spectra in Figure 6.522,23 and discuss the type of information it contains. The sample chosen is a rat tumor in a muscle environment,24 and we show spectra for tumor as well as for surrounding normal tissue and for two excitation wavelengths, 337 and 405 nm. The rat was injected with a dose of hematoporphyrin derivative 2 days earlier. For both excitation wavelengths, a broad intensity distribution extending from the blue to the near IR region is obtained. In the measurements, the strong, elastically scattered light from the primary laser excitation light is suppressed by a suitable lter, completely blocking shorter wavelengths but letting all longer wavelengths through basically unattenuated.

142

Lasers in Medicine

Fluorescence Intensity

Fluorescence Intensity

TUMOR exc = 337 nm

MUSCLE

exc

= 337 nm

500

600

700

nm

500

600

700

nm

Wavelength

Wavelength

Fluorescence Intensity

TUMOR = 405 nm exc

Fluorescence Intensity

MUSCLE = 405 nm exc

500

600

700

nm

500

600

700

nm

Wavelength

Wavelength

FIGURE 6.5 Laser-induced uorescence spectra for an experimental tumor (a human colon adenocarcinoma inoculated in rat muscle), and normal surrounding tissue. The rat had been injected with a dose of hematoporphyrin derivative about 2 days before the investigation. Two excitation wavelengths, 337 and 405 nm, are chosen to illustrate the inuence of the wavelength choice (From Ref. 23).

In the tumor spectra, we can see the characteristic dual-peaked uorescence signal from HPD in the red wavelength region (see Figures 6.2 and 6.4). This signal is less prominent in the normal tissue due to the selective retention of the drug in tumor tissue. The signal in the blue-green region is called the autouorescence, as it is due to the natural chromophores in the tissue. This signal peaks at about 470 nm. The blue-green uorescence is much stronger and the red HPD signal much weaker when 337 nm is chosen as the excitation wavelength. This can be understood from the absorption curves in Figure 6.1. 405 nm corresponds to the absorption maximum (the Soret band) of the HPD molecules, and excitation here is more efcient than at 337 nm. Many chromophores contribute to the signal in the blue-green wavelength region (examples are given in Figure 6.4a). The absorption for most of these molecules increases for shorter wavelengths. An interesting observation relating to Figure 6.5 is that the intensity of the blue-green signal is reduced in tumor tissue as compared with normal tissue. This was noted already in 198425 and seems to be due primarily to a transformation of strongly uorescing NADH in normal tissue to NAD+ with a much weaker uorescence.2628 These observations form the basis for efcient tumor detection and demarcation using laser-induced uorescence after injection of a tumorseeking drug: the red uorescence increases and the blue-green uorescence decreases in a tumor. Both effects can be utilized in a convenient way by dividing the red signal intensity by the blue intensity.25,29 This is illustrated in Figure 6.630 for the case of a scan through a tumor inoculated in a rat brain. The rat had been injected with Photofrin (1 mg/kg bodyweight) 24 hours before the animal was sacriced and the uorescence investigation was performed. Typical spectra for normal and tumor regions are included in the gure, with the red intensity at 630 nm denoted by A and the blue intensity at 470 denoted by B. The background-free HPD signal at 630 nm is denoted by A. In the scan, the red increase and the blue decrease in the tumor are very evident. Contrast enhancement is obtained by forming the ratio AB. The use of such a dimensionless ratio also has other important advantages, particularly in practical clinical work, as dimensionless quantities are immune to: Distance changes between tissue and measurement equipment

Tissue Diagnostics Using Lasers

143

FUNCTION VALUE arb. units

INTENSITY rel. units

10 8 6 4 2 400 500 600 700 WAVELENGTH / nm

INTENSITY rel. units TUMOR TISSUE A' A'

80 60 40 20

NORMAL BRAIN TISSUE

A A A/B -2 0 2 4 6

400

500

600 700 WAVELENGTH / nm

brain tumor brain

DISTANCE / mm

FIGURE 6.6 Fluorescence data obtained in a scan across an experimental glioma (TCVC) in a rat injected with 1 mg/kg b.w. Photofrin 24 hours earlier. Laser-induced uorescence spectra of tumor tissue and surrounding normal brain parenchyma are also shown. The background-free HPD-related signal is denoted A, while the total signal at 630 nm is denoted i.e., A and the autouorescence B. The dimensionless ratio A/B shows a strong contrast enhancement (From Ref. 30).

Variations in angle of incidence of radiation on tissue Fluctuations in illumination source and detection system efciency The ratio signal is basically sensitive only to the intrinsic properties of the tissue. In particular, if the drug-related signal is weak and sitting on a substantial slope of autouorescence intensity, to retain contrast, it is important to use the background-free signal intensity.

6.3.2 Instrumentation for Point Monitoring of Laser-Induced Fluorescence

This section will present equipment suitable for uorescence studies of tissue. A sealed-off nitrogen laser, emitting at 337 nm, is a very convenient excitation source. This laser typically generates 3 ns-long pulses at 10 Hz, and can also be employed to pump a compact dye laser, converting the pump radiation to a longer, selectable wavelength. This is the excitation source used in the recording of the data in Figures 6.5 and 6.6. Fluorescence is best analyzed with an optical multichannel analyzer system consisting of a spectrometer equipped with an intensied diode-array detector in the focal plane. For each exciting laser pulse, the whole uorescence spectrum is captured and can be directly displayed on a monitor and stored in a computer. The image intensier, providing amplication of the weak light signals, is gated synchronously with the laser pulses. In this way, background radiation due to the light from operating lamps can be suppressed in the recordings. In laboratory investigations, the excitation light can be directed to the sample by means of mirrors and the resulting uorescent light can be collected by an optical set-up of mirrors and lenses. The uorescence thus captured can be directed into the entrance slit of the spectrometer. However, for clinical measurements, a more exible uorosensor is needed. Such instrumentation is shown in Figure 6.7.31 The excitation laser light from the laser is focused into an optical ber that can be sterilized when used in surgical procedures or used through the biopsy channel of an endoscope. The uorescence is captured through the same ber separated from the excitation light by a dichroic mirror. The uorescent light is

144

Lasers in Medicine

ELECTRICAL SIGNALS TO AND FROM OMA MAINFRAME

DETECTOR NITROGEN LASER

POLYCHROMATOR
LENS MIRROR FILTER FLIP-IN MIRROR

CUT-OFF FILTER LENS OPTICAL FIBER DICHROIC MIRROR LIGHT TO AND FROM TARGET MIRROR

DYE LASER
LENS LENS FILTER

FLIP-IN MIRROR

MIRROR

FIGURE 6.7

Lay-out of a clinical uorescence diagnostic system (From Ref. 31).

focused onto the slit of a multichannel analyzer system. The uorosensor is constructed as a mobile system and can easily be transported between different clinics. Two systems of this construction are used for uorescence studies at various clinics at the Lund University Medical Laser Center. Most of the examples of uorescence spectra shown in this chapter have been recorded with this type of equipment. Interference lters can also be used to select the interesting wavelengths for uorescence signal detection. The uorescent light can then be divided by dichroic beamsplitters and directed toward individual photomultipliers. One detection channel can be centered at the 630 nm peak, while another one is set at 470 nm. An additional channel can be centered at 600 nm to provide a signal for background subtraction. Gated integrators are used to capture the signal during a brief time interval, followed by signal processing to provide the desired ratio signal. It is also possible to use a mercury lamp as an excitation source instead of a laser. Such a system is described in Ref. 32, and was further developed and tested as described in Ref. 33. A chopper wheel equipped with interference lters analyzes the uorescent light beam and lters out two excitation wavelengths, i.e., 365 and 405 nm. The ltered light is reected by a dichroic beam splitter and is focused into the investigation ber. The uorescent light comes back through the ber. Here, interference lters sequentially select the useful wavelengths, i.e., 470, 600 and 630 nm for both excitation wavelengths in cases of tumor diagnostics utilizing HPD sensitization. The light is recorded by a photomultiplier tube, and the signal is electronically integrated during the passage of every lter. The result is digitized and fed to a computer. In this way, three uorescence wavelengths can be investigated at two excitation wavelengths for every tissue spot under study.

6.3.3 Point Monitoring of Cancer using Laser-Induced Fluorescence

This section will give some examples of point monitoring of uorescence related to malignant tumor detection. Data for supercial tumors in humans are shown in Figure 6.8.34,35 In the left part of the gure, uorescence spectra and evaluated data from a scan are shown for a basal cell cancer in a patient who was injected by 2 mg/kg body weight of Photofrin 3 days before the uorescence measurements. In the right part of the gure, data for a breast carcinoma metastasis are shown in a similar way. This patient was studied 1 day after the same dose of Photofrin. A good demarcation of the tumor is obtained for both tumors, particularly in the A/B signal.

Tissue Diagnostics Using Lasers

145

FLUORESCENCE INTENSITY TUMOR A B 500 600 700 SKIN

FUNCTION VALUE FLUORESCENCE arb. units INTENSITY TUMOR B A B 500 600 700 SKIN

FUNCTION VALUE B

A/B A/B A A 500 600 700 20 10 0 10 20 distance/cm 500 600 700 -6 -4 -2 0 2 4 WAVELENGTH/nm skin tumor skin WAVELENGTH/nm DISTANCE/mm

FIGURE 6.8 Fluorescence spectra and evaluated data from two different human malignant tumors and normal surrounding skin. To the left two uorescence spectra and evaluated data from a scan across a basalioma 3 days after Photofrin incjection (2 mg/kg body weight). To the right two corresponding spectra and data from a scan across a breast carcinoma metastasis 1 day after Photofrin injection at the same dose.34,35

Extensive measurements on low-dose-injected bladder cancer patients were made at the Department of Urology, University of Leuven.33 The patients were injected with 0.35 or 0.50 mg/kg bodyweight, which is a low enough dose to avoid sun sensitization, which, at therapeutic dose injection (2 mg/kg bodyweight), requires that patients be kept at reduced light level for about 46 weeks. Spectra for a papillary bladder tumor are shown in Figure 6.9 for 337 and 405 nm excitation. It can be seen that there is a very strong reduction in the blue-green uorescence for tumor tissue, and also a prominent change in the spectral shape when 337 nm is employed. This change allows malignant tumor detection without using the HPD signal, which appears clearly only for 405 nm excitation, optimizing the conditions for HPD detection while simultaneously reducing the autouorescence intensity. Clinically, the most important task is to be able to identify dysplasia and carcinoma in situ, and it was shown that such tissue has sufciently different spectral characteristics to be distinguished from normal tissue. Spectral curves for mild and severe dysplasia are included in Figure 6.9. Data for tissues from 21 patients are shown in Figure 6.10. Fluorescence monitoring on tumors of the prostatic gland has also been performed after low-dose Photofrin injection and surgical resection. As can be seen in Figure 6.11, tumors are characterized by an increase in the red uorescence and a decrease in the blue-green uorescence.36,37 The lung is, like the bladder, an important location where early tumor detection can have a major inuence on the course of disease. This is particularly true for patients with a positive sputum test but negative biopsy sampling. We have studied the spectral characteristics of lung tissue during bronchoscopy with regard to autouorescence31 and after Photofrin injection.36,38 Spectra for a low-dose-injected patient are shown in Figure 6.12.38 Tissue diagnostics based on uorescence can sometimes be performed without any injection of a drug by utilizing the properties of endogenous chromophores only. Early such work was performed by Chinese researchers,39,40 particularly regarding the gastrointestinal tract, in which endogenous porphyrins frequently localize in tumors. However, endogenous uorescence can also be utilized to detect lung and colon tumors, and also malignancies at many other locations.4145 The unwanted side effect of skin sensitization can be avoided by using topical application in the case of supercial tumors instead of systemic injection of a sensitizing drug. Topically applied -amino levulinic acid (ALA) can be used for PDT of supercal skin tumors.46 ALA penetrates the damaged keratin in the dermis but not the undamaged parts. ALA is a precursor in the biosynthetic pathway of the blood pigment hemoglobin. ALA is a non-photoactive substance, but within the hem cycle it is transformed into protoporphyrin, which is a photodynamically very potent agent. This process is illustrated in Figure 6.13.47 The steps from ALA to protoporphyrin appear comparatively fast, while the last step to hemoglobin is relatively slow. Thus, an accumulation of protoporphyrin can be induced in the tumor cells. Subsequent exposure to photoactivating light leads to a selective destruction of the lesions. In a rst publication, a 90% response rate for basal cell carcinoma following a single treatment was found.46 We have reported the same response

146

Lasers in Medicine

FLUORESCENCE INT.

FLUORESCENCE INT.

NORMAL

405 nm

NORMAL TUMOR x 10

337 nm

TUMOR 500 550 600 650 700 WAVELENGTH (nm) 750

TUMOR

450

350 400 450 500 550 600 650 700 750 WAVELENGTH (nm)

FLUORESCENCE INT.

NORMAL

MILD DYSPLASIA

SEVERE DYSPLASIA

450

500

550 600 650 WAVELENGTH (nm)

700

750

FIGURE 6.9 Fluorescence spectra for a papillary bladder tumor and normal surrounding mucosa for a patient, injected with a dose of 0.35 mg/kg bodyweight of Photofrin 48 hours prior to the investigation (top). Spectra are shown for 337 and 405 nm excitation. Spectra for mild and severe dysplasia are also included.33

FLUORESCENCE INT.

FLUORESCENCE INT.

NORMAL

405 nm

NORMAL TUMOR x 10

337 nm

TUMOR 500 550 600 650 700 WAVELENGTH (nm) 750

TUMOR

450

350 400 450 500 550 600 650 700 750 WAVELENGTH (nm)

FLUORESCENCE INT.

NORMAL

MILD DYSPLASIA

SEVERE DYSPLASIA

450

500

550 600 650 WAVELENGTH (nm)

700

750

FIGURE 6.9 Fluorescence spectra for a papillary bladder tumor and normal surrounding mucosa for a patient, injected with a dose of 0.35 mg/kg bodyweight of Photofrin 48 hours prior to the investigation (top). Spectra are shown for 337 and 405 nm excitation. Spectra for mild and severe dysplasia are also included.33

rate for different kinds of supercial skin tumors, including basal cell and squamous cell carcinoma.47 The conversion of ALA into protoporphyrin in tumors can be readily followed using laser-induced uorescence.47,48 The superior tumor demarcation properties of ALA are illustrated in Figure 6.144749 for the case

Tissue Diagnostics Using Lasers

147

NORMAL FLUORESCENCE INTENSITY

HUMAN LUNG SPECTRA DHE 1 mg/kg b.w.

405 nm exc.

fibre

TUMOR

B3

400

450

500

550 600 WAVELENGTH (nm)

650

700 A = 630 nm B = 470 nm

750

FIGURE 6.12 Fluorescence spectra for a bronchial carcinoma and adjacent normal mucosa in a patient that had received a dose of Photofrin at a concentration of 1 mg/kg bodyweight 48 hours before the uorecence investigation.36,38

Succinyl CoA

Glycine

Haem Fe

hv

ALA

Feedback
Protoporphyrin

Singlet

O2
Triplet

Porphobilinogen Protoporphyrinogen Linear Tetrapyrrole Coproporphyrinogen Uroporphyrinogen

Tumor

Laser 630 nm
FIGURE 6.13 Schematic diagram of PDT using ALA. By exchanging the laser wavelength to about 400 nm and recording the uorescence from the tumor, its extent can be clearly determined.47

of a human basalioma. In the lower part of the gure, the corresponding uorescence signals after PDT (60 J/cm2) are shown. A strong photo-bleaching of the drug occurs, which can be used for dosimetry to ascertain that the correct light dose has been delivered to all parts of the tumor.48 The bleaching of the drug

148

Lasers in Medicine

Pre PDT Border 80

0

Pre PDT Tumor 60

0

70 0

Pre PDT Border

Pre PDT
30 0 20 0

500 600 700

500 600 700 Tumor 17 mm

500 600 700 60

Normal Normal skin 500 600 700 Normal

Pre PDT Normal

Normal skin

0 30 0

500 600 700 Normal

500 600 700 Post PDT 10 0

Border

10 0

Tumor

10 0

500 600 700 Border Post PDT

500 600 700 Post PDT

500 600 700 Post PDT

500 600 700 Post PDT

FIGURE 6.14 Fluorescence data from scans through a basal cell carcinoma before (top) and after (bottom) PDT (60 J/cm2). An ALA cream had been topically applied over the tumor 6 hours prior to the uorescence investigation.47,49

in the upper strongly exposed tissue layer also allows a more efcient treatment of deeper lying layers by continued light irradiation, because the upper layer cannot be overtreated.50 As already mentioned, there is an ongoing search for even better sensitizers, frequently concentrating on the behavior in PDT procedures. Most sensitizers also have uorescence properties that are useful for tumor demarcation. A list of new sensitizers with wavelengths for PDT excitation and uorescence detection is given in Table 6.1.
TABLE 6.1 Evaluation Photosensitizing Drugs Presently Under Different Stages of Clinical
Abbrev DHE BPD ALA MACE THPC ZnPC SnET2 nm 630 690 635 670 650 675 660

Name Photofrin Bensoporphyrin -amino levulinic acid Mono aspartyl Chlorin e6 Tetra hydr. phenylchlorin Zinc phtalocyanine Tin ethyl etiopurpurin

Note: Common abbreviations and approximate wavelengths for PDT irradiation and uorescence detection are given.

6.3.4 Equipment for Imaging

It is of great diagnostic interest to extend uorescence monitoring with incorporation of the contrastenhancement techniques illustrated in Figures 6.6 and 6.11 into imaging measurements. This has been accomplished in our group, which has developed a multi-color uorescence imaging system starting with a one-dimensional proof-of-principle demonstration51 later extended into full two-dimensional imaging.5254 To simultaneously obtain spatial and spectral resolution, the uorescent light is divided using a multi-mirror arrangement as illustrated in Figure 6.15. By using a spherical mirror, which is

Tissue Diagnostics Using Lasers

149

False Color Image Video Signal

Trig Signal Fiber Coupler

CCD Camera

Image Intensifier HV Supply Split Mirror Telescope

PC 2 1 4 3

Passband Filters

Dichroic Mirror Pulsed UV Source Quartz Lenses Object Plane

FIGURE 6.15 Diagram showing the basic construction of a multi-color uorescence imaging system (From Ref. 53).

divided into four individually adjustable sectors, four identical images can be sent to an intensied matrix detector, a CCD camera. By placing optical lters in front of the different mirror sectors, uorescence images in selected colors are obtained simultaneously. In the four different images, computerized calculations of dimensionless function values can be performed for each spatial location of the tissue investigated and a generalized image in an optimized contrast function can be produced. This function would normally be the ratio between the background-free 630 nm peak intensity A and the blue-green intensity B at about 470 nm. A properly weighted recording at 600 nm (kD) provides the background intensity. The weighting factor k also takes differences in optical efciency into account and is adjusted to result in a zero intensity value for A = (A-kD) when no tumor is present. A photograph of a fully clinically adapted uorescence imaging system based on the principles given here is shown in Figure 6.16. This system can be used together with endoscopes of different types allowing, e.g., lung and bladder investigations.

6.3.5 Multi-Color Fluorescence Imaging of Tumors

Fluorescence images at 630, 600 and 470 nm are shown in Figure 6.1754 for a human nodular basal cell carcinoma. Also included is the resulting image using the contrast function (A-kD)/B, showing the tumor with a very clear demarcation toward surrounding normal tissue. A nitrogen-laser-pumped dye laser was employed for uorescence excitation. The data were obtained with a laboratory instrumentation requiring substantial processing time to produce the nal image. In an industrial prototype, a considerably higher processing rate was achieved using a vector processor for image processing. Up to eight images per second could be obtained allowing basic live recording of tumors. An example of a processed image from that system is shown in Figure 6.18.56 This is a direct photograph of the monitor screen, showing a tumor, in a rat that had been injected with Photofrin at a concentration of 15 mg/kg bodyweight. Registrations of a malignant tumor in resected breast tissue, using the system shown in Figure 6.16, is given in Figure 6.19.56 Here, several images are presented where the processed tumor image is superimposed on a normal color video image using a video mixer. The patient had been given Photfrin at a dose of 0.35 mg/kg body weight 1 day before the operation. Further uorescence imaging systems are described in Refs. 5760.

150

Lasers in Medicine

FIGURE 6.16 Photograph of a clincal multi-color uorescence system adapted for bronchoscope applications (Courtesy: Spectraphos AB, Lund).

FIGURE 6.17 Imaging uorescence recordings of a nodular basal cell carcinoma on the head of a patient that had been subject to topical ALA application 6 hours before the investigation. Individual images at 630, 600 and 470 nm are shown together with a processed image using the data in the three individual images.55

Tissue Diagnostics Using Lasers

151

FIGURE 6.18 Photograph of the computer monitor screen showing a processed image of an experimental rat tumor of the same kind as was illustrated in Figure 6.5. The excitation source was a nitrogen laser, operating at 337 nm.55

6.4 Spectroscopic Diagnostics of Atherosclerotic Plaque

6.4.1 Basic Considerations
Cardiovascular disease is, besides cancer, the major cause of death in Western countries. Several risk factors in the development of atherosclerosis, such as hypertension and smoking, have been identied. When the process of atherosclerosis starts in the artery, the innermost part of the vessel wall, the intima, with the endothelium cell lining is thickened and new constituents are deposited in the tissue. During the initial stage, a brotic plaque develops, with increasing amount of collagen and elastin. Later in the process of the disease, cholesterol crystals and lipid droplets are deposited. In the nal stage, calcium is incorporated and the former elastic vessel tube is transformed into a rigid, inexible, partly obstructed and brittle vessel. Sometimes, the surface of the plaque turns into a wound, which can easily be an area for growth of dangerous blood clots. An atherosclerotically affected vessel is schematically shown in Figure 6.20.61 The peripheral as well as coronary arteries can be affected by atherosclerotic disease. The

152

Lasers in Medicine

FIGURE 6.19 Imaging recordings of a malignant tumor in breast cancer. The instrument shown in Figure 6.16 was utilized. The processed tumor image is superimposed on the normal color CCD output using a video mixer. Different balances between the two types of images have been used.56

Vessel lumen Blood Stream Intima Media

Adventitia Endothelium cell Smooth muscle cell Collagen Macrophage Lipid droplets Cholesterol crystal Capillaries Lymphocyte Calcium Thrombosis

FIGURE 6.20 Schematic diagram of an atherosclerotically transformed vessel.61

change in chemical composition in the atherosclerotic vessel wall forms the basis for uorescence diagnostics of this disease. The surgical treatment of severe atherosclerosis includes aggressive procedures such as open-heart surgery with coronary bypass. Several new modalities are now being investigated as possible replacements for such extensive procedures. A new modality is laser-based transluminal angioplasty, in which ber-based catheters through a peripheral artery, e.g., the femoral artery, are used as an entrance for intravasal treatment. A schematic diagram of such a procedure is shown in Figure 6.21.62 The eld is treated in detail in Chapter 116 of this volume. Different kinds of lasers have been used. Nonthermal excimer laser ablation is of special interest for the treatment of coronary arteries, as thermal

Tissue Diagnostics Using Lasers

153

Time-Integrated Diagnostics

EXCIMER LASER
Fluorescence Int.

Dichroic Mirrors TISSUE SPECTRAL DIAGNOSIS

Plaque Normal 400 500 600 WAVELENGTH/nm

Time-Resolved Diagnostics "Late"/"Early" Fluorescence

DIAGNOSTIC LASER

Plaque THRESHOLD Normal INTENSITY 0 5

10 15 (ns) TIME

Plaque

Aortic arch

Plasma Emission Diagnostics Normal 400 Emission Int. 500 600

Heart

Left Coronary

Ca and Ca+ Lines

Sodium D Line Plaque

400

500

600

FIGURE 6.21 Schematic diagram showing transluminal laser ablation of atherosclerotic plaque with three varieties of spectroscopic guidance: time-integrated uorescence spectroscopy, time-resolved uorescence monitoring and laser plasma emission spectroscopy (From Ref. 62).

lasers induce artery spasm that might be avoided with UV ablation. Laser angioplasty is discussed in Refs. 6366. Although promising, laser-based percutaneous angioplasty has some problems that must be overcome. In addition to arterial spasm, vessel wall perforation and dissection are complications that must be avoided. Another problem that follows all procedures performed in the vessel wall is restenosis, which is caused by an overreaction from the treated tissue. This might be minimized by radical treatment. Because the procedure is performed inside the vessel walls, a guiding system for the laser ring would improve the possibility of monitoring the procedure and inhibiting the laser power when the ablation has to be interrupted to avoid complications. A crude diagnostic signal is provided from the plasma emission signal, as shown in Figure 6.21. A calcied plaque features strong lines of Ca and Ca+, which disappear when the calcication has been ablated away.67,68

6.4.2 Plaque Discrimination Using Laser-Induced Fluorescence

It has been shown by several groups,6973 including ours,28,68,74 that the uorescence from the tissue itself, the autouorescence, can be used in demarcating atherosclerotic lesions from non-diseased vessel wall, depending on the various chromophore content in the tissue types. The spectral shape of the tissue uorescence can be used in demarcating the lesions, as the shape reects the different molecular contents. Fluorescence provides a completely nonintrusive way of probing prior to ring a high-power pulse. When excited with the light from a UV laser, the normal vessel wall emits strong uorescence in the region 390500 nm, while atherosclerotic plaque has a prominent peak at 390 nm, with a fall-off toward longer

154

Lasers in Medicine

wavelengths. This is illustrated by the spectra included in Figure 6.21. The signals are built up mainly from contributions of elastin and collagen (see Figure 6.4a), with a relative increase in collagen in atherosclerotic plaque being the origin of the spectral difference. In both spectra, a minimum in the uorescence signal can be seen at 420 nm, representing the strong absorption of hemoglobin in blood. Weaker absorption peaks occur at 540 and 580 nm. The identication of these features becomes very clear in a comparison with blood absorption spectra as shown in Figure 6.22.28 Clearly, the inuence of the blood reabsorption must be handled properly by a reliable uorescence diagnostic system.
60 50 TRANSMISSION % 40 30 20 10 0 50 40 30 20 10 0 400 500 600 700nm
VENOUS BLOOD ARTERIAL BLOOD

400

500

600

700nm

FIGURE 6.22 Absorption spectra for arterial and venous blood. Wavelength pairs with equal blood absorption are indicated.28
F1 = I(390) / I(480) F2 = I(415) / I(480)
1.5 6 3 1.0 4 2 0.5 2 1 10 1 1 20 2 2 30 3 3

F3 = I(580) / I(600)

F4 = I(390) / I(600)

F5 = I(380) / I(437)

F6 = I(390) / I(431)

IVIII II I 0

IVIII II I 0

IVIII II I 0

IVIII II I 0

IVIII II I 0

IVIII II I 0

FIGURE 6.23 Demarcation of ve different types of vessel wall using uorescence spectroscopy. Data for 6 different demarcation criteria are shown. Criteria F5 and F6 are not affected by the presence of blood.28

In our investigations we have observed that the signature in vitro from a plaque region differs depending on the stage of the atherosclerotic process. The early brotic lesions can be separated from the plaque regions with higher fat content and also from calcied regions. We have been able to spectroscopically separate four different classes of atherosclerotic lesions, depending on the severity of the disease. This is illustrated in Figure 6.23.28 The intensity ratios at selected wavelengths have been formed for a large number of samples. Group O is the normal non-diseased vessel wall tissue. Group I includes the least damaged tissue, with only small amounts of brotic constituents. In group II, more brotic constituents are present. Group III includes lesions with calcium content in the plaque, while,

Tissue Diagnostics Using Lasers

155

in group IV, calcium occurs at the surface. In the latter group, the lesions with endothelial damage with or without thrombotic material on the damaged surface are included. In the gure, all functions except F3 show a substantial demarcation among the tissue types. Functions F5 and F6 are particularly interesting, as they are immune to the presence of blood. This is because the wavelength pairs chosen correspond to points of equal blood absorption, as indicated in Figure 6.22. Thus, in the ratio, the inuence of the absorption is largely eliminated.28,68,75 Similar techniques can be used for studying myocardial tissue for assessing the status of the heart muscle. From a large number of biopsy samples, it was found that normal myocardial tissue can be distinguished from scar or fatty tissue, as shown in Figure 6.24, where two uorescence intensity ratios have been used for the demarcation.76
5 R1 390/465 4,5 4 3,5 3 2,5 2 1,5 1 ,5 0 ,5 1 1,5 2 2,5 3 R3 465/525

NORMAL FAT SCAR 3,5 4 4,5

FIGURE 6.24 Diagram showing differentiation of different kinds of myocardial tissue using uorescence (From Ref. 76).

FIGURE 6.25 Photograph of a uorosensor for plaque monitoring employing photomultipliers observing uorescent light, that has been selected for wavelength by interference lters (Courtesy: Spectraphos AB).

Some in vivo investigations were performed in connection with coronary bypass procedures using a clinically adapted uorosensor system, as shown in Figure 6.25. The system displays the ratio of the signal intensities for the blood-independent wavelength pair 380 and 440 nm. The ber probe of the system was placed where the coronary arteries were opened to be attached to the new vessels. Investigations were also performed inside the coronary arteries in the peripheral direction, with the probe inserted into the

156

Lasers in Medicine

I(380)

RATIO =

I(380) I(440)

I(440)

RATIO

NORMAL PLAQUE

NORMAL

PLAQUE NORMAL

FIGURE 6.26 Recording from the instrument shown in Figure 6.25 for normal tissue and plaque in human aorta.77
(a) Intensity (relative units) Normal vessel (b) Intensity (relative units) Normal tissue

10 15 Time (ns)

10

20 30 Time (ns)

Intensity (relative units)

Plaque

Intensity (relative units)

Malignant tumor

10 15 Time (ns)

10

20 30 Time (ns)

FIGURE 6.27 Time-resolved uorescence decay curves for a) atherosclerotic plaque and normal vessel wall, and b) an experimental tumor in an Photofrin-injected animal, and normal surrounding tissue.75,78

lumen toward known occlusions. Fluorescence was also measured from normal aortic wall close to the heart. A recording using this system is shown in Figure 6.26.77

6.4.3 Time-Resolved Monitoring

In all the uorescence measurements discussed so far, the total uorescence intensity following pulsed laser excitation has been used. However, the temporal decay characteristics of the uorescent light can also be utilized for tissue characterization, e.g., for distinguishing atherosclerotically diseased vessel wall from normal vessel wall.12,28,75,78 The uorescence from the diseased region has a longer decay time than the normal vessel wall tissue, as shown in Figure 6.27. This is basically because collagen, more abundant in plaque, has a longer lifetime than elastin.13 These recordings were taken with an advanced set-up incorporating a mode-locked picosecond laser source in conjunction with time correlating photoncounting electronics. By a detailed analysis, it is possible to unfold the decay into three components with different decay times.78 However, since the lifetimes involved are of the order of ns, for a clinically practical device it is also possible to use a normal nitrogen laser with a pulse duration of about 3 ns and to use gated integrators to capture late uorescence (515 ns) and divide it by early uorescence (05 ns). A recording obtained in this way when alternating the ber tip between plaque and normal wall is

Tissue Diagnostics Using Lasers

157

included in Figure 6.21. Because a single detection wavelength is used and because blood is non-uorescing, the data obtained are immune to the presence of blood, if the blood layer is not so thick that the whole signal is blocked out. A system based on the spectroscopic characterization of diseased tissue for guiding purposes in percutaneous laser angioplasty could, of course, include time-integrated as well as time-resolved aspects to improve the classication ability. In fact, an instrumentation of the type shown in Figure 6.25 allows both these aspects to be utilized. Decay curves detected at 630 nm for an experimental tumor and surrounding normal issue are also included in Figure 6.27. The animal had earlier been injected with Photofrin. It can be seen that the tumor curve has a fast and a slow decay component, while normal tissue has only a fast decay. The fast component corresponds to the tissue autouorescence, whereas the more long-lived component is due to the HPD drug. As was noted in Ref. 28, gating on late uorescence yields only suppression of the autouorescence and contrast enhancement using a single detection wavelength.

6.4.4 Raman Spectroscopy for Tissue Diagnostics

Raman spectroscopy has long been a powerful tool used by chemists in the analysis of complex organic molecules. The techniques of Raman spectroscopy are now being adopted for tissue spectroscopy.7981 As mentioned above, the main difculty with this technique is the weakness of the signals and the competing process of uorescence, which normally tends to completely swamp the Raman signals. An efcient way to reduce the inuence of uorescence is to use a laser with a near-IR wavelength, which does not excite uorescence very efciently. A CW Nd:YAG laser ( = 1064 nm) in combination with a Fourier transform spectrometer, provides an efcient way of recording Raman spectra.82 A clinically more practical way is to use a near-IR diode laser in combination with a cooled near-IR CCD array.80 A different way to suppress uorescence is to use a picosecond laser source and limit the detection to the duration of these short pulses. Then uorescence, which typically has a lifetime of a few ns, is reduced while all the Raman photons are recorded, because the latter process is basically instantaneous. This procedure can, however, hardly replace the use of a near-IR laser. Raman spectra of plaque and its molecular constituents are particularly impressive. Such recordings are shown in Figure 6.28.81 Here, near-IR Fourier transform spectra of brous plaque, atheromatous plaque and cholesterol powder (bottom) are shown, featuring strong Raman peaks corresponding to the vibrational shifts. The 1667 cm-1 vibrational mode from C = C (stretch), and the 1440 cm-1 mode of CH (bend) are particularly prominent. Raman spectra are also being investigated for the diagnosis of malignant tumors; the results seem promising. Through the increased specicity of Raman spectra over uorescence spectra, an interesting potential for powerful tissue diagnosis exists. Because of the weak signals and the need for a respectable spectral resolution, the possibilities for real-time Raman tissue imaging are limited.

6.4.5 Tissue Reectance Monitoring

We have already discussed visible light tissue reectance in Section 6.2.1 and noted that this is the normal means by which color on the object is determined. However, by going toward the IR region, much more specic information can be obtained, as molecular overtone vibrational transitions occur here, making an imprint in the reectance spectrum. Thus, in a certain sense, information similar to Raman spectroscopy can be obtained in IR reectance spectroscopy. This has been used for tissue spectroscopy.83

6.5 Light Scattering and Tissue Transillumination

The basic principles of elastic light scattering in tissue were discussed in Section 6.2.5. Light scattering in tissue complicates normal light propagation, as straight beam lines cannot be obtained. Thus, measurements of tissue oxygnation using differential light absorption at the near-IR absorption peaks of hemoglobin cannot assume a well-dened path length through the tissue probed, but a considerably

158

Lasers in Medicine

1662

1448

1340

1265

(a)
1005

Intensity (arb. units)

1440

1667

(b)
1300 1130 959 701

1671

1439

882

1328

(c)
1130 957 878 700

1800

1600

1400

1200

1000
-1

800

600

400

Frequency (cm )

FIGURE 6.28 Raman spectrum of (a) brous plaque, (b) atheromatous plaque and (c) cholesterol monohydrate powder, excited with a 1064 nm CW Nd:YAG laser.81

longer effective path length is obtained. Also, in transillumination, no sharp images can normally be obtained because of the strong scattering. Using time-resolved spectroscopy, it is possible to follow the photon paths through the tissue. Two basic geometries are illustrated in Figure 6.29.84 In transillumination, there is a shortest path corresponding to the very unlikely event of non-scattered radiation, giving rise to a distinct signal (ballistic photons). Most photons arriving at the detector come much later, and only after several scattering events. If the transmitter and the detector are at the same side of the sample, all photons received must have been scattered more or less. If the position of transmitter and receiver coalesce, a histogram of arrival times for photons that have sampled a certain volume below the measurement point is obtained. This mode of operation is common in Doppler measurements of supercial blood ow, which we will discuss rst.

6.5.1 Doppler-Flow Monitoring

A measure of the blood perfusion in tissue can be obtained by analyzing the signal due to slightly Doppler-shifted photons that encountered scattering against moving red blood cells. In such measurements, normally a CW helium-neon laser operating at 633 nm, or a dark red nearIR diode laser is used. Obviously, an objective full quantication is difcult to obtain, because a complex averaging is performed over photons scattered at different depths. Still, using experience, valuable information can be obtained, in particular for assessing reperfusion in connection with transplant surgery. The techniques and clinical application of laser Doppler blood owmetry can be found in Refs. 20 and 21.

Tissue Diagnostics Using Lasers

159

a)
A A B

Time

Time

b)
Absorbing Tumor

Input Pulse

Time-Integrated light

Time
Intensity
Time-Gated light

FIGURE 6.29 (a) Illustration of time-resolved signals received in elastic photon scattering measurements on tissue using two different geometries. (b) Enhanced viewing through a scattering medium using detection of the rst arriving photons in transillumination (From Ref. 84).

By performing sequential measurements with a system where the laser beam is scanned in a matrix pattern over the tissue, it is possible to generate images of the supercial blood ow.86 A schematic diagram of such a system is shown in Figure 6.30.86 An example of the application of such a system is shown in Figure 6.31,87 where the blood ow in a tumor region is shown in connection with PDT. A basal cell carcinoma was treated after topical application of ALA. The perfusion image is shown on four occasions. First, an image before the treatment was taken. Directly after PDT (60 J/cm2), given at such a low laser light uence rate that tissue heating is avoided, an increased blood ow is seen in the tumor and surrounding area, indicating an immediate inammatory tissue response. One week after the treatment, the central tumor area is covered by a crust surrounded with a high perfusion area. Finally, 8.5 months after the treatment, a renewed imaging of the area shows a reduction of the blood ow in the tumor to the same level as the surrounding normal tissue.

160

Lasers in Medicine

FIGURE 6.30 Arrangement for scanning laser Doppler owmetry for tissue.86

FIGURE 6.31 Imaging measurements of supercial blood perfusion in connection with ALA PDT. Recordings are shown for the following situations: a) before PDT, with the tumor still covered by a crust, b) after removal of the crust, c) immediately after PDT, and d) 2 weeks after treatment.87

6.5.2 Time-Resolved Scattering Spectroscopy

As mentioned above, time-resolved transillumination allows enhanced viewing through tissue. X-ray diagnostic techniques are the conventional means for such imaging that has been brought to a very high level of sophistication. However, ionizing radiation is accompanied with a certain risk of mutagenicity.88 It would be very desirable to use optical radiation for transillumination of tissue to avoid this risk and also to allow spectroscopic recordings of molecular-specic absorption. Red light penetrates tissue quite well, as can be evidenced when shining light from a ashlight through the hand. The reason is the quickly decreasing absorption of hemoglobin beyond 600 nm, as is illustrated by the corresponding curve in Figure 6.1. The low intensity of the transmitted light through thick tissue is not a problem, as very sensitive light detectors exist, but rather, the strong multiple scattering from the cell structures, washing

Tissue Diagnostics Using Lasers

161

FIGURE 6.32 Transillumination scan through a human hand with data for the total time-integrated light and the early light.94

out all the spatial information. Scattering prevents the observation of the bones in the hand as shadows. The scattering has limited the application of optical transillumination (diaphanography) for mammographic investigations.8991 Only supercial tumors close to the skin in the breast parenchyma can be made visible. Recently, much research has been initiated to overcome these problems,85,92 e.g., by using time-resolved photon-counting techniques employing picosecond red laser pulses.9396 Because of the multiple scattering, most of the photons penetrating the tissue have spent a long time in the tissue, while only very few photons arrive to the detector at the nominal propagation time without disruptions. However, by selecting a very small detection time window and recording the very rst photons, it is possible to look at just the fastest photons coming out of the tissue. If an obstacle, such as a bone or a tumor blocks the passage, a shadow will be created for the selected subgroup of very early photons. For thicker samples, there are no unscattered photons. Still, the rst photons have been minimally scattered, producing a clearer image, as schematically shown in Figure 6.29b. An example of a scan through a human hand is shown in Figure 6.32.94 A cavity-dumped dye laser, synchronously pumped by an argon-ion laser, was used in these measurements. Time-gated single-photon counting was used to record early photons in a time window 80 ps wide. Clear shadows of the bones can be seen in the timegated scan, while the contrast disappears in the time-integrated recording.

6.5.3 Transillumination Imaging

By scanning the tissue under investigation, a transillumination image can be created. Recently, we have shown that such measurement can also be performed using a pulsed diode-laser source, making the

162

Lasers in Medicine

FIGURE 6.33 Two-dimensional transillumination scans through a resected breast sample containing a ductal carcinoma. The tumor clearly emerges in the early light, while it remains hidden in time-integrated monitoring.96

X - Y Translator Glass Slide Laser

Diode Laser Driver

Sample CFD Reference Fibre TAC CFD Amp Detector

Optical Fibre

PC MCA

FIGURE 6.34 Set up for two-dimensional time-gated transillumination imaging through tissue, using a pulsed diode laser and time-correlated photon counting electronics.96

technique very realistic.96 Images of a ductal breast carcinoma in a newly resected breast are given in Figure 6.33. A diode laser, operating at 815 nm and producing 30 ps-long pulses at a repetition rate of 10 MHz was used in these recordings, performed with the set-up shown inFigure 6.34. The transmitting and receiving bers are scanned together under computer control over the tissue, which is compressed between glass plates. In the time-integrated light, the tumor cannot be seen. It was also difcult to detect using conventional x-ray mammography. In the time-gated image, the tumor can be clearly seen. To eliminate possible artifacts, it is advantageous to divide the early light by the total light, as also shown in the gure. Many groups are now working in this new eld of research, developing a variety of techniques for suppressing excessive scattering using streak-camera,97 Raman-amplier98 and frequency-doubling gating.99 Fast modulation of the source followed by detection of phase shifts and loss of modulation depth can give similar information.100102 The coherence properties of non-scattered light are used in

Tissue Diagnostics Using Lasers

163

heterodyne103 and light-in-ight holography.104,105 Imaging in scattering media can also be performed with spatial domain reectometry.106,107

6.5.4 Interpretation of Time-Resolved Tissue Recordings

This section will focus on the intepretation of time-resolved recordings of tissue to understand, e.g., how tumors can be detected in the early light. This is best performed by examining the results of a model experiment95 performed with picosecond pulses traversing a 30 mm-thick cuvette containing a mixture of Intralipid (Kabi Vitrum) and ink in water. Intralipid is a strongly scattering, non-absorbing liquid, while ink has a strong absorption but negligible scattering. By mixing these two substances, the values of the scattering and absorption coefcients, s and a, can be varied freely. In the upper part of Figure 6.35, time-resolved recordings can be seen, where no ink but increasing amounts of Intralipid have been added. It can be seen that, at early times, the time dispersion curves are strongly inuenced, while the decay constant at late times is little affected. If, on the other hand, increasing amounts of ink are added for a constant amount of Intralipid, it can be seen in the lower part of Figure 6.35 that the early light is basically unaffected, while the decay constant for the late light is quickly shortened for increasing absorption. In the gure, an early xed detection window is indicated. Using these curves, the increasing signal for a tumor in Figure 6.33 (brighter areas) can be interpreted as a reduction in the scattering in a tumor rather than a change in the absorption. This means that the early light is not useful for absorption spectroscopy changing the wavelength. On the other hand, wavelength-dependent scattering changes might prove useful for tissue identication. The decay constant for the late light is related in a simple way to the absorption coefcient.108 However, for this late light, little image information is retained. By using the whole temporal dispersion curve, maximal tissue information can be extracted using wavelength-dependent properties of different chromophores and tissue structures. Focusing on the properties of the late light, spectroscopic analysis has been performed15 for assessing the concentration of photosensitizers in tumors for PDT light dosimetry. Then, light with a wavelength on the sensitizer absorption peak, as well as light off that peak, is analyzed. The rst time-resolved measurements on human tissue concerned brain oxygenation assessment, e.g., in premature infants, by using differences in oxy- and deoxyhemoglobin absorption108110 (see Figures 6.1 and 6.23). At the end of this section, we should note that tissue information can be obtained also when timeintegrating instrumentation is used, e.g., when a CW light source is employed. As we have noted, the pathlength is not well dened, due to multiple scattering in the tissue. Still, equipment used in a standardized way can yield reliable data. An example of this is the blood oxygenation measurements performed with a pulse oximeter, where light is sent through a patients ngertip, probing the differential absorption of oxy- and deoxyhemoglobin.111 Normally, light at three different wavelengths is used, generated by light-emitting diodes or diode lasers. 805 nm provides a reference point, where oxy- and deoxyhemoglobin absorb equal amounts (the isobestic point). At 660 and 940 nm, the two types of hemoglobin have very different absorptions. The oxygenation value (oxygen tension) obtained with a pulse oximeter has been shown to be inuenced by the hematocrit value (density of red blood cells) and by complex formation in the blood of smokers. Even if the absolute value is subject to various inuences, the instrument has an important application in monitoring of patient status from the time before anesthesia through an operation, or during bronchoscopy. A further example is a technique for cancer diagnostics, where the elastically scattered light from a broadband source is analyzed at some distance from the point of light injection into the tissue.112 Because the scattering properties vary for different types of tissue, the signal recorded can be compared with catalogue spectra obtained for well characterized tissues. Thus, a tissue diagnosis can be obtained in this empiric way.

6.6 Outlook
As illustrated in this chapter, many new possibilities for tissue diagnostics using lasers are emerging, addressing important elds such as cancer detection, cardiovascular monitoring and tissue transillumi-

164

Lasers in Medicine

a 0 mm-1 s(1-g)=0.30 mm-1 s(1-g)=0.45 mm-1

Intensity (Arb. Units)

s(1-g)=0.60 mm-1 s(1-g)=0.75 mm-1 s(1-g)=0.90 mm-1

0.5 1 120 psec time-gate

1.5 Time (nsec)

2.5

s(1-g)=0.38 mm-1 a 0 mm-1

Intensity (Arb. Units)

a=0.0045 mm-1 a=0.009 mm-1 a=0.014 mm-1 a=0.018 mm-1

0.5 1 120 psec time-gate

1.5 Time (nsec)

2.5

FIGURE 6.35 Experimental time-dispersion curves for light propagation through a 30 mm thick cuvette lled with a scattering and absorbing liquid. From the curves it is evident that the amount of early light is mainly inuenced by the scattering and not by the absorption.95

nation imaging without using ionizing radiation. The rapid development of more and more practical and cost-effective laser sources, detectors and computers facilitates the clinical implementation of the new techniques. The noninvasive nature of the techniques and the real-time data acquisition are attractive features of the optical methods. Optical spectroscopy, in principle, allows molecular analysis, providing information not normally accessible with other modalities. For the future, an increased use of combined diagnostic and therapeutic equipment can be anticipated. Ideally, each single cell should be optically diagnosed, followed by an immediate decision on whether that cell should be eliminated through the immediate use of laser ablation, heating or photochemistry.

Acknowledgments
The author gratefully acknowledges a most stimulating collaboration with a large number of colleagues and graduate students within the Lund University Medical Laser Center, as well as with many foreign groups. This work was supported by the Swedish Research Council for Engineering Sciences, The Swedish Board for Technical and Industrial Development, the Swedish Cancer Foundation and the Swedish Medical Research Council.

Tissue Diagnostics Using Lasers

165

References
1. Mackay, R.S., Medical Images and Displays: Comparison of Nuclear Magnetic Resonance, Ultrasound, X-rays and other Modalities, Wiley, New York (1984). 2. Svanberg, S., Atomic and Molecular Spectroscopy, 2nd ed., Springer, Heidelberg (1992). 3. Boulnois, J.-L., Photophysical processes in lasertissue interactions, in: Laser Applications in Cardiovascular Diseases, (R. Ginsberg, Ed.), Futura, New York (1987). 4. Chen, H.C., Space Remote Sensing Systems, Academic, Orlando (1985). 5. Thompson, K., Lasers in ophthalmology, in: Lasers in Medicine, (G. Pettit and R.W. Waynant, Eds.), Wiley, New York (1993). 6. Haller, J., Lasers in cardiology, in Lasers in Medicine, (G. Pettit and R.W. Waynant, Eds.), Wiley, New York (1993). 7. Svanberg, S., Medical applications of laser spectroscopy, Physica Scripta T26, 90-98 (1989). 8. Marcus, S.L., in Photodynamic Therapy of Human Cancer: Clinical Status, Potential and Needs, (C. Gomer, Ed.), pp. 1, SPIE Press, Bellingham (1990). 9. Pass, H.I., Photodynamic therapy in oncology: mechanisms and clinical use, J. National Cancer Inst. 85, 443-456 (1993). 10. Marcus, S.L., Lasers in photodynamic therapy, in: Lasers in Medicine, (G. Pettit and R.W. Waynant, Eds), Wiley, New York (1993). 11. Lakowicz, J.R., Principles of Fluorescence Spectroscopy, Plenum, New York, (1983). 12. Wolfbeis, O.S. (Ed.), Fluorescence Spectroscopy; New Methods and Applications, Springer, Heidelberg (1992). 13. Andersson-Engels, S. et al., An investigation of possible uorophores in human atherosclerotic plaque, Lasers in Life Science 5, 1-11 (1992). 14. Andersson-Engels, S. et al., Fluorescence imaging and point measurements of tissue: applications to the demarcation of malignant tumors and atherosclerotic lesions from normal tissue, J. Photochem. Photobiol. 53, 897-814 (1991). 15. Wilson, B.C. et al., Tissue optical properties in relation to light propagation models and in vivo dosimetry, in Photon Migration in Tissue, (B. Chance, Ed.), pp. 24-42, Plenum, New York (1989). 16. Patterson, M.S., B.C. Wilson, and D.R. Wyman, The propagation of optical radiation in tissue. I. Models of radiation transport and their application, Lasers Med. Sci. 6, 155-168 (1991). 17. Wilson, B.C. and G. Adam, A Monte Carlo model for the absorption and ux distributions of light in tissue, Med. Phys. 10, 824-830 (1983). 18. Jacques, S.L., Time resolved propagation of ultrashort laser pulses within turbid tissue, Appl. Opt. 28. 2223-2229 (1989). 19. Patterson, S., B. Chance and B.C. Wilson, Time resolved reectance and transmittance for the noninvasive measurement of optical properties, Appl. Opt. 28, 2331-2336 (1989). 20. berg, P.., Laser-Doppler owmetry, Critical Reviews in Biomedical Engineering 18, 125-163 (1990). 21. Shepherd, A.P. and P.. berg (Eds.), Laser Doppler Blood Flowmetry, Kluwer Academic, Boston (1990). 22. Andersson-Engels, S. et al., Aspects of tumor demarcation in rats by means of laser-induced uorescence and haematoporphyrin derivatives, Lasers in Med. Sci., 3, 239-248 (1988). 23. Montn, S., On the use of laser-induced uorescence in medical and industrial applications, Ph.D. dissertation, Lund Reports on Atomic Physics LRAP-75, Lund University (1987). 24. Hedlund, G. and H.O. Sjgren, Induction of transplantation immunity to rat colon carcinoma isografts by implantation of intact fetal colon tissue, Int. J. Cancer 26, 1-73 (1980). 25. Ankerst, J. et al., Laser-induced uorescence studies of hematoporphyrin derivative (HpD) in normal and tumor tissue of rat, Appl. Spectroscopy 38, 890-896 (1984). 26. Lohmann, W. et al., Native uorescence of the cervix uteri as a marker for dysplasia and invasive carcinoma, Eur. J. Obstet. Gynecol. Reprod. Biol. 31, 249-253 (1989).

166

Lasers in Medicine

27. Lohmann, W. and E. Paul, Native uorescence of unstained cryo-sections of the skin with melanomas and nevi, Naturwissenshaften 76, 424-426 (1989). 28. Andersson-Engels, S. et al., Malignant tumor and atherosclerotic plaque diagnosis using laserinduced uorescence, IEEE J. Quant. Electron. 26, 2207-2217 (1990). 29. Svanberg, K. et al., Fluorescence studies of hematoporphyrin derivative in normal and malignant rat tissue, Cancer Res. 46, 3803-3808 (1986). 30. Andersson-Engels, S. et al., Identication of brain tumors in rats using laser-induced uorescence and hematoporphyrin derivatives, Lasers in Med. Sci. 4, 241-249 (1989). 31. Andersson-Engels, S. et al., Clinical recordings of laser-induced uorescence spectra for evaluation of tumor demarcation feasibility in selected clinical specialities, Lasers in Med. Sci. 6, 415-424 (1991). 32. Andersson, P.S. et al., Fluorescence endoscope instrumentation for improved tissue characterization, Med. Phys. 14, 633-636 (1987). 33. Baert, L. et al., Detection of supercial bladder cancer utilizing laser-induced uorescence and lowdose haematoporphyrin derivative, Urology 41, 322-330 (1993). 34. Andersson-Engels, S. et al., Photodynamic therapy and simultaneous near-infrared light-induced hyperthermia in human malignant tumors a methodological case study, Proc. ICALEO 87, 60, 67-74 (1987). 35. Andersson-Engels, S. et al., Photodynamic therapy alone or in conjunction with near-infrared light-induced hyperthermia in human malignant tumors: a methodological case study, Proc. SPIE 908, 116-125 (1988). 36. Andersson-Engels, S. et al., In vivo uorescence spectra of malignant lesions in various clinical specialties, In: Photodynamic Therapy and Medical Laser Applications, (P. Spinelli, Ed.), Elsevier, Amsterdam (1992). 37. Andersson-Engels, S. et al., Photofrin distribution in malignant tumours of the prostate, Progress Report 1992, Lund University Medical Laser Centre, p. 32, Lund University (1993). 38. Andersson-Engels, S. et al., Progress Report 1992, Lund University Medical Laser Centre, p. 30, Lund University (1993). 39. Yanming, Ye et al., Single pulse cancer diagnosis by laser excited autouorescence, Chinese Physics Lasers 14, 284 (1987). 40. Yuanlong, Yang et al., Characteristic autouorescence for cancer diagnosis and its origin, Lasers Surg. Med. 7, 528-532 (1987). 41. Alfano, R.R. et al., Laser induced uorescence spectroscopy from native cancerous and normal tissue, IEEE J. Quant. Electr. QE-20, 1507-1511 (1984). 42. Glassman, W. S. et al., Ultraviolet excited uorescence spectra from non-malignant and malignant tissues of the gynecological tract, Lasers Life Sci. 5, 49-58 (1992). 43. Richards-Kortum, R. et al., Spectroscopic diagnosis of colonic dysplasia, J. Photochem. Photobiol. 53, 777-786 (1991). 44. Hung, J. et al., Autouorescence of normal and malignant bronchial tissue, Lasers Surg. Med. 11, 99-105 (1991). 45. Alfano, R.R. et al., Optical spectroscopy may offer novel diagnostic approaches for the medical profession, in Laser Non-Surgical Medicine, l. Goldman (Ed.), pp. 55-123, Technomic, Lancaster (1991). 46. Kennedy, J.C., R.H. Potter and D.C. Pross, Photodynamic therapy with endogenous protoporphyrin IX: Basic principles and present clinical experience, J. Photochem. Photobiol. 6, 143-148 (1990). 47. Svanberg, K. et al., Photodynamic therapy of non-melanoma cancers of the skin utilizing topical -amino levulinic acid application and laser irradiation, submitted to Brit. J. Dermatol. 130, 743 (1994). 48. Berg, R. et al., Photodynamic therapy in interplay with uorescence diagnostics in the treatment of human supercial malignancies, SPIE Vol.1645, 187 (1992).

Tissue Diagnostics Using Lasers

167

49. Svanberg, K. et al., Photodynamic therapy of human skin malignancies and laser-induced uorescence diagnostics utilizing Photofrin and -amino levulinic acid, in: Photodynamic Therapy and Medical Laser Applications, (P. Spinelli, Ed.), Elsevier, Amsterdam (1992). 50. Potter, W.R., T.S. Mang and T.J. Dougherty, The theory of photodynamic therapy dosimetry: consequences of photodestruction of sensitizer, Photochem. Photobiol. 46, 97-101 (1987). 51. Montn, S., K. Svanberg and S. Svanberg, Multicolor imaging and contrast enhancement in cancertumor localization using laser induced uorescence in hematoporphyrin-derivative-bearing tissue, Opt. Lett.,10 , 56-58 (1985). 52. Andersson, P.S., S. Montn and S. Svanberg, Multispectral system for medical uorescence imaging, IEEE J Quant. Electr. 23, 1798-1805 (1987). 53. Andersson-Engels, S., R. Berg, and K. Svanberg, Multicolor uorescence imaging in connection with photodynamic therapy of -amino levulinic acid (ALA)-sensitized skin malignancies, Bioimaging 3, 134-143 (1995). 54. Andersson-Engels, S., J. Johansson and S. Svanberg, Medical diagnostic system based on simultaneous multi-spectral uorescence imaging, to appear. 55. Andersson-Engels, S. et al., Fluorescence and transillumination imaging for tissue diagnostics, Proc. CLEO 91, Baltimore, April 1991, Opt. Soc. Am., Washington DC (1991). 56. Svanberg, K. et al., Clinical multi-color uorescence imaging of malignant tumors initial experience. Acta Radiol. 38, 2 (1998). 57. Proo, A.E., Review of uorescence diagnosis using porphyrins, SPIE 907, 150-156, SPIE Optical Engineering, Bellingham, Wash. (1988). 58. Wagnires, G., et al., Photodetection of early cancer by laser-induced uorescence of a tumorselective due: apparatus design and realization, Proc. SPIE 1203, 43-52 (1990). 59. Palcic, B. et al., Detection and localization of early lung cancer by imaging techniques, Chest 99, 742-43 (1991). 60. Lam, S. et al., Detection of dysplasia and carcinoma in situ with a lung imaging uorescence endoscope device, J. Thorac. Cardiovasc. Surg. 105, 1035-40 (1993). 61. Andersson-Engels, S. et al., Laser spectroscopy in medical diagnostics, in: Photodynamic Therapy: Basic Principles and Clinical Aspects, (Th. J. Dougherty and B.W. Henderson, Eds), pp. 387-424, Marcel Dekker Inc., New York (1992). 62. Svanberg, K. and S. Svanberg, Lasers in medicine, La Recherche 255, 686, (1993). 63. Isner, J.M., P.G. Steg and R.H. Clarke, Current status of cardiovascular laser therapy, IEEE J Quant. Electr., QE-23, 1756-1771 (1987). 64. Karsch, K.R. et al., Percutaneous coronary excimer laser angioplasty: initial clinical results, Lancet Sept. 16 , 647-650 (1989). 65. Litvack, F. et al., Percutaneous excimer laser coronary angioplasty, Lancet July 8, 102-103 (1989). 66. Litvack, F., J. Margolis, and T. Linnemeier, Percutaneous excimer laser coronary angioplasty: results of the rst 110 procedures, J. Am. Coll. Cardiol. 15, 25A (1990). 67. Hohla, K.G. et al., Simultaneous tissue identication and ablation with excimer laser, Proc. SPIE 908, 129 (1988). 68. Andersson-Engels, S. et al., Laser-induced uorescence used in localizing atherosclerotic lesions, Lasers Med. Sci. 4, 171-181 (1989). 69. Kittrell, R. et al., Diagnosis of brous arterial atherosclerosis using uorescence, Appl. Opt. 24, 2280-2281 (1985). 70. Baraga, J.J. et al., Ultraviolet laser-induced uorescence of human aorta, Spectrochim. Acta 45A, 95-99 (1989). 71. Hoyt, C.C. et al., Remote biomedical spectroscopic imaging of human artery wall, Lasers Med. Surg. 8, 1-9 (1988). 72. Deckelbaum, L.I. et al., Discrimination of normal and atherosclerotic aorta by laser induced uorescence, Lasers Surg. Med. 7, 330-335 (1988).

168

Lasers in Medicine

73. Orayevsky, A.A. et al., Laser spectral analysis of human atherosclerotic vessels, in: Laser Spectroscopy VIII, (W. Persson and S. Svanberg, Eds.), pp. 370-371, Springer, Heidelberg, (1987). 74. Andersson, P.S. et al., Diagnosis of atherial atherosclerosis using laser induced uorescence, Lasers Med. Sci. 2, 261-266 (1987). 75. Andersson-Engels, S. et al., Time-resolved laser-induced uorescence spectroscopy for enhanced demarcation of human atherosclerotic plaques, J. Photochem. Photobiol. B4, 363-369 (1990). 76. Aganauskiene, J. et al., Characterization of myocardial tissue utilizing laser-induced uorescence spectroscopy, unpublished report. 77. Johansson, J. et al., Fluorescence characterization of atherosclerotic plaques and normal vessel walls in vitro and in vivo, private communication. 78. Andersson-Engels, S., J. Johansson and S. Svanberg, The use of time-resolved uorescence for diagnosis of atherosclerotic plaque and malignant tumours, Spectrochima Acta 46A, 1203-1210 (1990). 79. Alfano, R.R. et al., Lasers Life Sci. 4, 23 (1991). 80. Baraga, J.J., M.S. Feld and R.P. Rava, Rapid near-infrared Raman spectroscopy of human tissue with a spectrograph and CCD detector, Appl. Spectr. 46, 187-190 (1992). 81. Baraga, J.J., M.S. Feld and R.P. Rava, In situ histochemistry of human artery using near infrared Fourier transform Raman spectroscopy, Proc. Natl. Acad. Sci. USA (1992). 82. Hirschfeld, T. and B. Chase, Appl. Spectroscopy 40, 133-139 (1986). 83. Nilsson, A.M.K. et al., Near-infrared diffuse reection and laser-induced uorescence spectroscopy for myocardial tissue characterization, Spectrochim. Acta A 53, 1901/1997. 84. Berg, R., S. Andersson-Engels and S. Svanberg, Time-resolved transillumination imaging, to appear in Optical Tomography, (G. Mller et al., Eds.), SPIE Institute, Vol. 11, SPIE, Bellingham (1993) . 85. Chance, B. and G. Mller (Eds.), Optical Tomography, SPIE, Bellingham (1993). 86. Wrdell, K., A. Jacobsson and G.E. Nilsson, Imaging of tissue perfusion by dynamic light scattering, Trans. Biomed. Eng. IEEE, (1991). 87. Wang, I. et al., Supercial blood ow following photodynamic therapy of malignant non-melanoma skin tumors measured by laser Doppler perfusion imaging, Brit. J. Dermatology 136, 184-189 (1997). 88. Swift, M. et al., Incidence of cancer in 161 families affected by ataxia-telangiectasia, New Engl. J. Med. 325, 1831-1836 (1991). 89. Bartrum, R.J. and H.C. Crow, Transillumination lightscanning to diagnose breast cancer: A feasibility study, Am. J. Rad. 142, 409-414 (1984). 90. Proo, A.E., G.A. Navarro and O.W. Sartorius, Scientic basis of breast diaphanography, Med. Phys. 16, 60-65 (1989). 91. Monsees, B., J.M. Destouet and W.G. Totty, Light scanning versus mammography in breast cancer detection, Rad. 163, 463-465 (1987). 92. Chance, B. (Ed.), Photon Migration in Tissue, Plenum, New York (1989). 93. Andersson-Engels, S. et al., Medical application of laser spectroscopy, in Laser Spectroscopy IX, (M. Feld, Ed.), pp. 500-504, Academic Press, New York (1989). 94. Andersson-Engels, S. et al., Time-resolved transillumination for medical diagnostics, Opt. Letters 15, 1179-1181 (1990). 95. Andersson-Engels, S., R. Berg and S. Svanberg, Effects of optical constants on time-gated transillumination of tissue and tissue-like media, J. Photochem. Photobiol. 16, 155-167 (1992). 96. Berg, R., O. Jarlman and S. Svanberg, Medical transillumination imaging using short pulse diode lasers, Appl. Opt. 32, 574-579 (1993). 97. Alfano, R.R., P.P. Ho and K.M. Yoo, Photons for prompt tumor detection, Phys. World 5, 37-40 (1992) 98. Duncan, M.D. et al., Time-gated imaging through scattering media using stimulated Raman amplication, Opt. Lett. 16, 1868-1870 (1991).

Tissue Diagnostics Using Lasers

169

99. Yoo, K.M., Q. Xing and R.R. Alfano, Imaging objects hidden in highly scattering media using femtosecond second-harmonic-generation cross-correlation time gating, Opt. Lett. 16, 1019-1021 (1991). 100. Fishkin, J. et al., Diffusion of intensity modulated near-infrared light in turbid media, in: TimeResolved Spectroscopy and Imaging of Tissue, (B. Chance, Ed.), pp. 122-135, SPIE 1431 (1991). 101. Jacques, S.L., Principles of phase-resolved optical instruments, in: Future Trends in Biomedical Applications of Lasers, (L.O. Svaasand, Ed.), pp. 143-153, SPIE 1525 (1991). 102. Patterson, M.S. et al., Frequency-domain reectance for the determination of the scattering and absorption properties of tissue, Appl. Opt. 30, 4474-4476 (1991). 103. Toida, M., T. Ichimura and H. Inaba, The rst demonstration of laser computed tomography achieved by coherent detection imaging method for biomedical applications, IEICE Trans. E 74, 1692-1694 (1991). 104. Spears, K.G. et al., Chrono-coherent imaging for medicine, IEEE Trans. Biomed. Eng. 36, 12101221 (1989). 105. Chen, H. et al., Two-dimensional imaging through diffusing media using 150-fs gated electronic holography techniques, Opt. Lett. 16, 487-489 (1991). 106. Swanson, E.A. et al., High-speed optical coherence domain reectometry, Opt. Lett. 17, 151-153 (1992). 107. Hee, M.R. et al., Femtosecond transillumination optical coherence tomography, Opt. Lett 18, 950952 (1993). 108. Chance, B.J. et al., Comparison of time-resolved and -unresolved measurements of deoxyhemoglobin in brain, Proc. Natl. Acad. Sci. USA 85, 4971-4975 (1988). 109. Sevick, E.M. et al., Anal. Biochem., 195, 330 (1991). 110. Delpy, D.T. et al., Estimation of optical pathlength through tissue from direct time of ight measurement, Phys. Med. Biol. 33, 1433-1442 (1988). 111. Payne, J.P. and J.W. Severinghaus (Eds), Pulse Oximetry, Springer, Heidelberg (1986). 112. Bigio, I.J. et al., Optical diagnostics based on elastic scattering: recent clinical demonstrations with the Los Alamos Biopsy System, in Proc. Biomedical Optics Eur. 93, EUROPTO 2081, SPIE, Bellingham (1994).

7
Low-Power Laser Effects
7.1 7.2 Introduction ....................................................................... 171 Primary and Secondary Mechanisms of the Action of Monochromatic Visible and Near Infrared Radiation on Cells .............................................................. 172
Photoacceptors for Low-power laser Effects Primary Mechanisms of Light Action Secondary Reactions of Light Action

7.3

Explanation of Controversies and Limitations of LowPower Laser Effects on Cellular Level ................. 187
Magnitude of Low-Power Laser Effects Diversity of LowPower Laser Effects Biological Limitations of Low-Power Laser Effects Dependence on the Physiological Status of the Host Organism

7.4

Clinical Applications of Low-Power Laser Effects ........... 196

Focus of Application Is Low-Power Laser Therapy a Mature Modality? Possible Optimal Light Parameters for Low-Power Laser Therapy

Tiina Karu
Laser Technology Research Center of Russian Academy of Science

7.5 Summary............................................................................. 200 References ...................................................................................... 201

7.1 Introduction
The most frequently used mechanism of photon energy conversion in laser medicine is heating. Average heating of irradiated samples occurs with all methods of tissue destruction (cutting, vaporization, coagulation, ablation). Many of these surgical laser techniques are reviewed elsewhere in this book. At low light intensities, the photochemical conversion of the energy absorbed by a photoacceptor prevails. This type of reaction is well known for specialized photoacceptors such as rhodopsin or chlorophyll. In medicine, light absorption by non-specialized photoacceptor molecules (i.e., molecules that can absorb light at certain wavelengths, but are not integral to specialized light reception organs) is used rather extensively (Figure 7.1). The absorbing molecule can transfer the energy to another molecule, and this activated molecule can then cause chemical reactions in the surrounding tissue. This type of reaction is successfully used in photodynamic therapy (PDT) of tumors (also discussed elsewhere in this book). Alternatively, the absorbing molecule in a light-activated form can take part in chemical reactions, as occurs in treatment of skin diseases with psoralens and UVA radiation (PUVA). Importantly, in both PDT and PUVA therapy, the photoabsorbing molecules are articially introduced into a tissue before irradiation. Irradiation of cells at certain wavelengths can also activate some of the native components. In this way, specic biochemical reactions, as well as whole cellular metabolism, can be altered. This type of reaction is believed to form the basis for low-power laser effects (Karu, 1989b; Smith, 1991). One should note that light therapy methods, based on photochemical conversion of photoabsorbing molecules

0-8493-1146-2/02/$0.00+$1.50 2002 by CRC Press LLC

171

172

Lasers in Medicine

LOW-POWER (LASER) LIGHT IN MEDICINE

[ exogenous (artificially introduced) photoacceptors]

[ endogenous (natural) photoacceptors]

PHOTODYNAMIC THERAPY (PDT)

PHOTOTHERAPY WITH UV

PHOTOCHEMOTHERAPY (PUVA)

LOW-POWER LASER THERAPY (LASER BIOSTIMULATION)

FIGURE 7.1

Methods of light therapy based on endogenous and exogenous photoacceptors in cells.

(Figure 7.1), are not laser-specic methods. Conventional sources generating the appropriate wavelength can also be used (as is done in PUVA and UV therapy). Laser sources are just handy tools providing many practical advantages (e.g., efcient ber optic coupling to irradiate interior body parts, high monochromaticity and easy wavelength tunability, simplicity in use and electrical safety in the case of semiconductor lasers). However, for therapeutic effect in deeper tissue levels, coherent light could provide some additional benets (Karu, 2002). Low-power laser effects are the topic of the present chapter. During the past few years, numerous reviews on this topic have been written describing various aspects of the problem: history (Karu, 1987,1989a), controversies (Karu, 1989a; Smith, 1991; Belkin et al., 1988; Harris, 1988; Berns and Nelson, 1988; Basford, 1989), quantitative laws of visible light action on cells (Karu, 1987, 1989b, 1991a), photobiological fundamentals (Karu, 1989a, 1998), molecular mechanism (Karu 1988). Specic data about the effects of irradiation on various cells has also been reviewed, including: cell cultures in vitro (Karu,1990,1991b), Escherichia coli (Tiphlova and Karu, 1991a), microorganisms (Karu, 1996a), and human lymphocytes (Karu, 1992b, 1996b). In addition, clinical applications of low-power laser therapy have been surveyed (Ohshiro and Calderhead, 1988, 1991; Calderhead et al., 1991; Baxter, 1994; Tunr and Hode, 1999). As a rule, the material from those previous reviews is not repeated here. The present review is designed in the following way. First, primary and secondary mechanisms of action of visible and near IR radiation on cells are described (Section 7.2). Second, controversies and limitations are considered (Section 7.3). Third, a short glance into clinical applications of low-power laser therapy is presented (Section 7.4).

7.2 Primary and Secondary Mechanisms of the Action of Monochromatic Visible and Near Infrared Radiation on Cells
A photobiological reaction involves the absorption of a specic wavelength of light by the functioning photoreceptor (photoacceptor) molecule. To distinguish specialized photoreceptor molecules such as rhodopsin, phytochrome, bacteriorhodopsin and chlorophylls from nonspecialized chromophores, the photoacceptors take part in a metabolic reaction in a cell that is not connected with a light response. After absorbing the light of the wavelength used for irradiation, this molecule assumes an electronically excited state from which primary molecular processes can lead to a measurable biological effect in certain circumstances. To work as a photoacceptor taking part in photobioregulation, this molecule must be part of a key structure that can regulate a metabolic pathway. Redox chains are example of this type of key structures, which suit to these requirements.

7.2.1 Photoacceptors for Low-Power Laser Effects

Several pieces of evidence show that mitochondria are sensitive to irradiation with monochromatic visible and near infrared (IR) light. The illumination of isolated rat liver mitochondria increased adenosine triphosphate (ATP) synthesis and the consumption of O2 (Kato et al., 1981; Passarella et al., 1984; Gordon and Surrey, 1960). Irradiation with light at wavelengths of 415 nm (Kato et al.,

Low-Power Laser Effects

173

1981), 602 nm (Vekshin and Mironov, 1982), 632.8 nm (Passarella et al., 1984), 650 nm and 725 (Gordon and Surrey, 1960) enhanced ATP synthesis. Light at wavelengths of 477 and 554 nm (Kato et al., 1981) did not inuence the rate of this process. Oxygen consumption was activated by illuminating with light at 365 and 436 nm, but not at 313, 546 and 577 nm (Vekshin and Mironov, 1982). Irradiation with light at 633 nm increased the mitochondrial membrane potential () and proton gradient (pH), caused changes in mitochondrial optical properties, modied some NADH-linked dehydrogenase reactions (NADH is a reduced form of nicotinamide adenine dinucleotide) (Passarella et al., 1983) and increased the rate of ADP/ATP exchange (ADP is adenosine diphosphate) (Passarella et al., 1988a) as well as RNA and protein synthesis in the mitochondria (Hilf et al., 1986). In the case of state 4 respiration, the 351 nm and 458 nm laser irradiations accelerated the oxygen consumption of rat liver mitochondria; such acceleration was not observed with 514.5 nm irradiation. On the contrary, in the case of state 3 respiration, the 514.5 nm laser irradiation activated the oxygen consumption of mitochondria. Activation did not occur with 458 nm irradiation, and 351 nm irradiation reduced the oxygen consumption in state 3 (Morimoto et al., 1994). The 660 nm irradiation increased state 3 oxygen consumption at both coupling II and III sites, as well as increasing the respiratory control ratio (Yu et al., 1996). It is also believed that mitochondria are the primary targets when the whole cells are irradiated with light at 630 nm (Hilf et al., 1986), 632.8 nm (Karu et al., 1995a; Bakeeva et al., 1993; Manteifel et al., 1997) or 820 nm (Herbert et al., 1989). Irradiation with light at 812 nm (Loevschall and ArenholdtBindslev, 1994a) or 632.8 nm (Anders et al., 1995) altered the rhodamine 123 uptake by broblasts. These results were interpreted by the authors as inducting the perturbation of mitochondrial energy production (Loevschall and Arenholdt-Bindslev, 1994a) and membrane potential (Anders et al., 1995). The question is, which molecule in a mitochondrion is responsible for the effects mentioned above? When considering the cellular effects, this question can be answered with the aid of action spectra. We know that, within certain limits, an action spectrum follows the absorption spectrum of the photoacceptor molecule (Hartman, 1983). On the other hand, the action spectrum is insufcient to distinguish between potential photoacceptor pigments with very similar absorption. Moreover, the absorption spectrum for the photoacceptor pigment may depend strongly on its environment, but this environment remains unknown as long as the photoacceptor pigment itself is unknown. Because of these inadequacies of the photoacceptor pigment, some other criteria for identication are also used. Action spectra for the DNA and RNA synthesis rate in HeLa cells of the exponential and plateau phase of growth, as well as those for the adhesive properties of HeLa cellular membranes, were published by Karu et al. (1984a,b, 1996a). Figure 7.2 shows a generalized action spectrum for HeLa cells (a sum of the four spectra from the paper of Karu et al. (1984a,b). Recall that in the wavelength range 310500 nm, a maximum stimulating effect was obtained with a radiation dose one order of magnitude less than in the longer-wave spectral range (Karu et al., 1984a, b). Figure 7.2 shows that the action spectrum in the range 580860 nm consists of two series of doublet bands in the range 620680 nm and 760895 nm with well-pronounced maxima at 620, 680, 760 and 825 nm. In the violet-blue region there is one maximum at 400 nm with the edge of the envelope near 450 nm. It is known that the action spectrum is roughly the same shape as the absorption spectrum of the photoacceptor (Hartman et al., 1983). Therefore, the bands in the action spectra were identied by analogy with the metal-ligand systems absorption spectra characteristic of this spectral range (Wilkinson et al., 1987; Siegel, 19711981; Hughes, 1987). The regions 400450 nm and 620680 nm are characterized by the bands pertaining to complexes with charge transfer in a metal-ligand system, and within 760830 nm, these are d-d transitions in metals. The region 400420 nm is typical of -* transitions in a porphyrin ring (Gouterman, 1978). Comparative analysis of spectral data for transition metals and their complexes on one hand, and biomolecules participating in the regulation of cellular metabolism on the other, allows us to suggest that multinuclear enzymes containing Cu(II) may be participating (Hughes, 1987; Lichtenstein, 1979). Analysis of the electron excitation transitions of participating molecules containing Cu(II) (Hathaway,

174

Lasers in Medicine

FIGURE 7.2 The generalized action spectrum of proliferation increase of HeLa cells for = 330860 nm (Karu et al., 1984a, b). Curve 1-dose 10 J/m2, curve 2-dose 100 J/m2. A possible belonging of peaks to absorbing chromophores is marked as suggested by Karu and Afanasyeva, 1995.

1987; Lontie, 1984; Spiro, 1981; Karlin and Zubieta, 1983) shows that metal-ligand transitions in the range at 400450 nm correspond to the Nimidasole Cu transition, at 620 nm, to Scysteine Cu transition, and at 680 nm to Smethionine Cu transition. Comparing the lines of possible d-d transitions and charge-transfer complexes of Cu (Hughes, 1987; Lontie, 1984; Spiro, 1981; Karlin and Zubieta, 1983; Brunori and Chance, 1987) with our action spectrum (Figure 7.2) allows us to assume that the photoacceptor molecule has different types of centers containing Cu(II) in the ranges 420450 nm and 760830 nm. In the range 420450 nm, this may be a combination of centers of types I and II (for the characteristics of centers of types I, II and III see Lichtenstein, 1979) though a center of type I may be present. At 330 nm, a center of type III may be present, and in the range 760820 nm centers of types I and III coexist. Within 620680 nm, there is a center of type I and a combination of centers of different types is unlikely. The above analysis allows us to conclude that all bands in the action spectrum in Figure 7.2 may be related to the cytochrome c oxidase. The fact that the photoacceptor is a component of the respiratory chain was considered earlier (Karu, 1989a). Cytochrome c oxidase (or cyt a/a3), is the terminal enzyme of the respiratory chain in eukaryotic cells (Figure 7.3a), which mediates the transfer of electrons from cyt c to molecular oxygen. Ferrocytochrome c is oxidized, dioxygen is reduced, and protons are pumped vectorially from the mitochondrial matrix to the cytosol. Free energy resulting from this redox chemistry is converted into an electrochemical potential across the inner membrane of the mitochondrion, which ultimately drives the production of ATP. Accordingly, cytochrome c oxidase plays a central role in the bioenergetics of the cell. Cytochrome c oxidase of mammalian cells is a large multicomponent membrane protein of considerable structural complexity. Two heme moieties (heme a and heme a3), two redox active copper sites (CuA and CuB), one zinc and one magnesium, are located in 13 subunits (molecular size of 200 kilodaltons). Recently, the high resolution three-dimensional x-ray structure of cytochrome c oxidase of bovine heart (Tsukahara et al., 1995, 1996) and Paracoccus denitricans (Iwata et al., 1995) were reported. These studies indicated that CuA is a dinuclear copper center with an unexpected structure similar to a [2Fe-2S] type iron-sulfur center in which the Fe ions and inorganic sulfur atoms are replaced with Cu ions and cysteine sulfur atoms respectively. The O2 binding site contains heme a3

Low-Power Laser Effects

175

FIGURE 7.3 (a) Mitochondrial respiratory chain of eukaryotic cells. (b) A scheme of the catalytic cycle of cytochrome c oxidase (after Gennis and Ferguson-Miller, 1995).

iron and CuB; there is no detectable bridging ligand between iron and copper atoms. Heme a is coordinated with two imidazoles of histidine residues. The fth ligand of heme a3 is an imidazole, whereas CuB is coordinated by three imidazoles of histidine. CuA (Cu-Cu) center is coordinated by residues of two cysteins, two histidines, one methionine and one peptide carbonyl of a glutamate (Tsukahara et al., 1995). In the catalytic cycle of cytochrome c, oxidase electrons are transferred sequentially from water-soluble cytochrome c to CuA, then to heme a and to the binuclear center a3-CuB, where oxygen is reduced to water (Figure 7.3b). Oxygen binds to heme a3 and is reduced to water through a series of short-lived elusive intermediates. Singular value decomposition analysis indicated the presence of at least seven intermediates (Sucheta et al., 1997). The best characterized species until now are ferrous-oxycomplex and peroxy species (Babcock and Wikstrm, 1992; Verkhovsky et al., 1996; Sucheta et al., 1997). Generally speaking, the cytochrome c oxidase can be fully oxidized (four redox active metal centers: CuA, CuB, irons in hemes a and a3, are in their common higher oxidation state; 3+ for iron and 2+ for copper), or fully reduced (four metal centers are in their common lower oxidation state; 2+ for iron and 1+ for copper). Partially reduced enzymes, usually called mixed-valence one, have some metal centers in their higher oxidation state and the remainder in their lower oxidation state. There are also a number of forms of oxidized enzyme: fast enzyme (reacts relatively rapidly with cyanide), slow enzyme (reacts at about 1% of the rate of the fast enzyme, also called resting enzyme), pulsed enzyme (obtained by reducing slow enzyme and oxidizing it with oxygen under conditions in which the production of H2O2 is avoided), oxygenated enzyme (subjected to a cycle of reduction and reoxidation under conditions in which H2O2 is produced) (Capaldi, 1990; Palmer, 1993). These details are given to illustrate how complicated and controversial the overall picture of cytochrome c oxidase functioning still is. Coming back to the comparative analysis of the action spectrum in Figure 7.2 and available spectroscopic data on cytochrome c oxidase cited above, it was suggested (Karu and Afanasyeva, 1995) that the 820 nm band belongs to the oxidized CuA, the 760 nm band to the reduced CuB, the 680 nm band to the oxidized CuB, and the 620 band to the reduced CuA (Figure 7.2). The 400450 nm band is more likely to be the envelope of a few absorption bands in the range 350500 nm (i.e., a superposition of several bands). The band with a maximum near 404420 nm can be assigned to the oxidized heme, whereas the longer-wave edge of the envelope at 450 nm (due to its asymmetry), should evidently be assigned to the

176

Lasers in Medicine

reduced CuB. The participation of the heme in the action spectra is conrmed by the optimal dose ratio (10 and 100 J/m2 respectively for 404 and other visible region maxima, Karu et al., 1984a, b). It should be noted that the Soret band of heme compounds (i.e., the band in the range 400420 nm) is more intense by an order of magnitude than the absorption bands of these compounds in the visible region (Gouterman, 1978). The weak band at 330 nm may belong to the oxidized CuB. Thus, the bands at 330, 404420, 680 and 825 nm can be attributed to the oxidized form of cytochrome c oxidase; the edge of the blue-violet band at 450 nm and the distinct bands at 620 and 760 nm belong to the reduced form of the enzyme. Analysis of the band shapes in the action spectra (Figure 7.2) and the line intensity ratios enables us to conclude that cytochrome c oxidase cannot be considered as a primary photoreceptor when it is fully oxidized or fully reduced, but only when it is in one of the intermediate forms (partially reduced, or mixed-valence enzyme). The conclusion that the action spectra (Figure 7.2) reect the absorption spectrum of one of the intermediate forms of the enzyme complex is supported by the results of experiments using dichromatic irradiation (Figure 7.4). As well as irradiating the cells with light of different wavelengths as normal (shown on the abscissa), they were simultaneously irradiated with light at 632.8 nm. This wavelength was close to the position of one maximum in the action spectrum in Figure 7.2 (620 nm). The technique of this experiment is described in a paper by Karu et al. (1985). The light doses in both irradiation processes were either optimal (100 J/m2 in the region 600850 nm, (Figure 7.4, curve 3), and 10 J/m2 (Figure 7.4, curve 1)), or increased (25 J/m2 in the blue-green region, Figure 7.4, curve 2). So, in the case of simultaneous dichromatic irradiation, a new action spectrum is formed. The comparison of the action spectra in Figures 7.2 and 7.4. reveals essential differences between them: the shift of the blue-violet maximum from 404 nm to 450 nm or its absence in the spectrum; new bands in the green region (550560 nm); the absence of bands at 680, 760, 825 nm (Figure 7.4). It is known (Palmer et al., 1978) that the shift of the band from 400 to 450 nm can be observed in the course of cytocrome c oxidase reduction. It should be noted that, in the range 550560 nm, one can observe an absorption band of one of the intermediate forms (Brunori and Chance, 1988). These results enable us to conclude that simultaneous dichromatic irradiation changes the ratio of the reduced and oxidized forms of the enzyme as compared with ordinary irradiation (Karu et al., 1984a,b). The suggestion that an intermediate form of cytochrome c oxidase might be the primary photoacceptor is also supported by the results of Pastore et al. (2000). The fully oxidized form of the enzyme appeared to be insensitive to He-Ne laser radiation as revealed by absorption spectra. The irradiation increased the absorption of the partially reduced enzyme as well as its proton pump activity. It is worth noting that when the cellular monolayer was irradiated simultaneously with = 633 nm and various wavelengths of visible light, the red-far red peaks at 680 and 760 nm disappeared (Figure 7.4). When the cells were irradiated consecutively with wavelengths 633 and 760 nm, and the time interval between the two irradiation events was varied, the DNA synthesis rate depended on the order in which the wavelengths were used (Figure 7.5a). Irradiation rst with the light at 760 nm and then with the red light ( = 633 nm) stimulated the DNA synthesis, whereas irradiation in the reverse order (633 nm followed by 760 nm), inhibited it. These effects reached their maxima when the time interval between the successive irradiation events was between 1 and 3 min and became progressively less pronounced with a further increase in the interval. It should be noted that the effects were not equal in magnitude: stimulation amounted to 60%, while inhibition was 20% (Figure 7.5a). This result supports the conclusion that the 620 nm and 760 nm bands do not belong to the one photoacceptor, but to its different absorbing centers (probably CuA and CuB). Electronic excitation of these centers in a different sequence may inuence the electron transfer in cytchrome c oxidase in a different way, one that has an effect on the nal photobiological response, namely DNA synthesis. When the consecutive irradiation was performed with red (633 nm) and blue (404 nm) light, the sequence 633 nm followed by 404 nm had no effect on DNA synthesis, but the sequence 404 nm followed by 633 nm stimulated it (Figure 7.5b). Again, as in the previous case (Figure 7.4a), the effects did not occur until the interval between the two irradiation events reached 10 seconds, then increased as the

Low-Power Laser Effects

177

FIGURE 7.4 Action spectra of simultaneous dichromatic irradiation with = 632.8 nm and (wavelengths shown on the abscissa) on the DNA synthesis rate in exponentially growing HeLa cells (adapted from Karu et al., 1985).

FIGURE 7.5 Stimulation of DNA synthesis rate in HeLa cells measured after consecutive irradiation with (a) far red and red, or (b) blue and red light as dependence of irradiation wavelength sequence and the time interval between the two irradiation events (shown on abscissa) (adapted from Karu et al., 1985). The doses were 100 J/m2 for radiation at 760 and 633 nm, and 10J/m2 for radiation at 404 nm.

178

Lasers in Medicine

interval was increased. The difference between the results of two experiments presented in Figure 7.4a,b is that in the case of far red-red light-irradiations (Figure 7.5a) the effect disappeared (was reduced to control levels) when the time interval was increased. In the second case, when irradiation was performed in the sequence 404 nm + 633 nm (Figure 7.5b), the effect was still maximal at the same time interval of 104 s. It is quite possible that chromophores absorbing at 404 and 633 nm do not belong to one molecule but to different molecules of the same redox chain. Let us recall here three results indicating the antagonistic action of red and far red light. First, when irradiating isolated rat liver mitochondria, Gordon and Surrey (1960) found that red light at 650 nm increased oxidative phophorylation but far red light at 725 nm inhibited it. Irradiating hamster broblasts with red (632.8 nm) and far red (760 nm) light caused a change in the intracellular concentration of cAMP (Karu et al., 1987a), the cAMP concentration behaviors in the former case being opposite to that in the latter case (increase and decrease, respectively). Irradiation at 670 and 830 nm stimulated the proliferation of the Schwann cells but irradiation at 780 nm inhibited it (van Breugel et al., 1993). These results could be explained by taking into account the fact that the wavelengths mentioned above are absorbed by different chromophores in a different redox state: max at 620 nm by CuA (reduced), max at 680 nm by CuB (oxidized), max at 760 nm by CuB (reduced), and max 820 nm, by CuA (oxidized). Possibly, different absorbing chromophores may play a different role in driving the metabolism. One important step in identifying the photoacceptor molecule is to compare the absorption and action spectra (Hartmann, 1983). Recording an absorption spectrum of a cellular monolayer or individual cell is not an easy task. The absorption spectra of individual cells were recorded years ago with the aim of identifying respiratory chain carriers (Nicholls and Elliot, 1974). Absorption spectrum of eight parallel monolayers of human broblasts was recorded using a commercial double beam spectrophotometer (van Breugel and Br, 1992). For recording the absorption of one layer with an aim to study the irradiationinduced changes in absorption of cell chromophores, a sensible multichannel registration method was developed (Karu et al., 1998, 2000). The rst results of these experiments are presented in Figure 7.6 a,b.

FIGURE 7.6 Absorption spectrum of monolayer of dry HeLa cells. The experimental details are described in the paper by Karu et al. (1998).

First, we recorded the absorption spectra of a dry monolayer (the monolayer was held in air at room temperature for 30 min. after being rinsed with Hanks solution). An example of such a spectrum is presented in Figure 7.6. The spectrum exhibits distinct absorption bands with maxima at 620 nm, 680 nm (with a shoulder manifest at 665 nm), 810 nm, and 870 nm, weak bands being observed to occur at 715 nm, 730 nm, and 765 nm. Irradiating this type of monolayer with laser radiation had no effect at all. This circumstance is explained by the fact that virtually not one cell survived the drying, which was veried through staining with trypan blue. Further experiments were aimed at recording the absorption spectra of a wet monolayer immediately after rinsing with Hanks solution. To improve sensitivity, the spectra in this series were recorded in the wavelength range 640710 nm (585900 nm in the preceding series). In the given wavelength range (640710 nm), there are clearly manifest absorption bands at 670, 718, and 744 nm, and also a less distinct band or a band shoulder in the vicinity of 750 nm (curves 1 in Figs. 7.7a and b).

Low-Power Laser Effects

179

FIGURE 7.7 Absorption spectrum of the HeLa monolayer immediately after the removal of the nutrient medium (curve 1) and following exposure to radiation with = 670 nm for the rst time (curve 2), second time (curve 3), and third time (curve 4) (each exposure 10 s in a dose of 6.3 103 J/m2) (Karu et al., 1998); (b): absorption spectrum of the HeLa monolayer immediately after the removal of the nutrient medium (curve 1) and following exposure to radiation with = 820 nm for the rst time (curve 2), second time (curve 3), and third time (curve 4) (each exposure 10 s in a dose of 6.3 103 J/m2, as described by Karu et al., 1998).

Exposing the sample for 10 s to laser radiation with a wavelength of 670 nm and a dose of 6.3103 J/m2 once (curve 2 of Figure 7.7a), twice (curve 3 of Figure 7.7b), or thrice (curve 4 of Figure 7.7a) caused changes in its absorption bands around 670, 750, and 774 nm, the absorption band at 718 nm remaining unchanged (see Figure 7.7a). In the action spectra, the band in the neighborhood of 670680 nm belongs supposedly to the chromophore CuB in the oxidized state, while that in the vicinity of 760770 nm belongs to the chromophore CuB in the reduced state (Karu and Afanasyeva, 1995). If there is a correspondence between the action spectra bands (Figure 7.2) and the absorption spectra bands presented in Figure 7.6, the results presented in Figure 7.7a are quite natural; as laser irradiation increases absorption in the band at 670 nm, hence the concentration of the chromophore in the oxidized state, absorption near 750770 nm (and the concentration of the reduced chromophore) decreases. It is interesting to compare the responses of the monolayer to the rst, second, and third exposure to laser radiation. The

180

Lasers in Medicine

rst, with a wavelength of 670 nm, causes no substantial changes in the 670-nm absorption band (curve 2) as compared with that of the unexposed monolayer (curve 1). Only the subsequent two exposures result in the intensication of the 670-nm (band curves 3 and 4, respectively). An entirely different effect is observed to occur in the complex absorption band in the region of 745780 nm. The rst exposure (curve 2) brings about a sharp change in both the intensity and shape of the complex band prole, practically no changes taking place as a result of the second and third exposures (curves 3 and 4, respectively), which is very similar to the saturation effect. Two bands, one with = 745 and a shoulder at = 755 nm, and the other with = 775 nm, are clearly dened in the absorption spectra of the exposed monolayer (curves 24 in Figure 7.7a). The shape of the band points to the presence of two conformers in the hypothetical reduced chromophore. The exposure of the cellular monolayer to laser light with = 820 nm (Figure 7.7b) was also observed to cause changes in the absorption bands in the vicinity of 670 and 775 nm. Recall that the action spectra featured a band at around 825 nm, which is supposedly associated with the oxidized chromophore CuA (Figure 7.2). The exposure of the monolayer to light with = 820 nm was carried out in the same way as in the case of laser light with = 670 nm in the preceding series of experiments. The sample was exposed for 10 s to the radiation in a dose of 6.3 103 J/m2 once (curve 2), twice (curve 3), or thrice (curve 4). Following the rst exposure (curve 2), a sharp increase of absorption is observed to occur in the band near 670 nm (and the correspondingly sharp reduction of absorption in the band near 770 nm) in comparison with the intact monolayer (curve 1). The second (curve 3) and the third (curve 4) exposure cause no sharp changes in absorption, which is likely due to an equilibrium being established between the oxidized and reduced forms of the chromophore CuB. Clearly manifest in Figure 7.7b (curves 24) are two bands, one with 1 = 750 nm and the other with 2 = 770 nm, as was also in the case with irradiation at = 670 nm (Figure 7.7a). One can therefore state with certain reservations that radiation with a wavelength of 670 nm or 820 nm has most likely no effect on the chromophore reduction mechanism. In both cases ( = 670 nm and = 820 nm), irradiation reduces the absorption of radiation and the concentration of the reduced form of the chromophore, and gives rise to its two conformably ordered forms. On the contrary, the behavior of the 670-nm band in the case of exposure to laser light with = 820 nm (Figure 7.7b) differs drastically from that in Figure 7.7a (i.e., in the case of irradiation at = 670 nm). The very rst exposure to light with = 820 nm increases the intensity of the 670-nm band (curve 2 of Figure 7.7b). Following the second exposure (curve 3), this band changes but little, and practically no changes are observed to occur following the third exposure (curve 4) (there takes place something like saturation). Thus, the laser wavelength and the number of exposures have different (opposite) effects on the (supposedly) oxidized form of the chromophore, with the absorption maximum at = 670 nm. Radiation with = 820 nm oxidizes the chromophore CuB stronger in the rst exposure (i.e., oxidation proceeds in a stepwise manner), whereas radiation with = 670 nm has practically no effect on the oxidized form of CuB in the rst exposure (Figure 7.7a). It is only after the second and the third exposure that the concentration of the oxidized form of the chromophore grows higher, no saturation being reached in our experiments. The 718-nm band suffers virtually no changes following exposure to laser light differing in wavelength. It should be noted that this band is not manifest in the action spectra (Figure 7.2). The absence of changes in absorption following exposure to radiation with two wavelengths (Figure 7.7a, b) also seems quite logical and gives reason to believe that the chromophore absorbing in this region takes no part in the photoregulation process responsible for the action spectrum. The rst results of the experiments on irradiating the cellular monolayer bear witness to the fact that the method developed (Karu et al., 1998) allows one to measure the weak absorption of a live cellular monolayer and holds much promise for detailed research into the primary changes in the photoacceptor molecule (cytochrome c oxidase) consecutive upon the absorption of light by its various chromophores (CuA and CuB, for example). New results of this type of measurement are presented in the paper of Karu et al., 2001.

Low-Power Laser Effects

181

7.2.2 Primary Mechanisms of Light Action

The primary mechanisms of light action on the photoacceptor molecules have not yet been established. Below, four main possibilities discussed in the literature so far are considered. It has been suggested that one possible process involves an acceleration of electron transfer in the respiratory chain due to a change in the redox properties of the carriers following photoexcitation of their electronic states (Karu, 1988; 1989a). A similar principle governs the function of photosynthetic reaction centers (Govindjee, 1982) and is involved in many so-called blue light responses (Senger, 1980, 1984). It is well known that electron excitation changes the redox properties of molecules (Marcus and Sutin, 1985). The results of the recent measurements (Karu et al., 1998, 2000; Figure 7.7), as well as the discussion from Section 7.2.1, allow one to propose that the photoexcitation of certain reaction centers in cytochrome c oxidase molecule (like CuA and CuB or hemes a and a3) or in cyt bd and bo complexes of Escherichia coli (Afanasyeva et al., 1995) inuences the redox state of these centers, and consequently, the rate of the electron ow in the molecule. It was suggested, for example, that irradiation shifts the cytochromes to more reduced forms that naturally promote electron transport (Adar and Yonetani, 1978). The laser irradiation and activation of electron ow in the molecule of cytochrome c oxidase can reverse the partial inhibition of its catalytic center by NO (Karu, Tyatibrat, Kalendo, 2001). Recently, the redox absorbance changes of the respiratory chain components of Escherichia coli were measured following He-Ne laser irradiation. A dose of 4.3 104 J/m2 led to partial oxidation of cyt b and cyt d, while avoproteins were found to be slightly reduced (Dube et al., 1997). During light excitation of electronic states, a noticeable fraction of the excitation energy is inevitably converted to heat, which causes a local increase in the temperature of the absorbing chromophores (Letokhov, 1991). Any appreciable time- or space-averaged heating of the sample can be prevented by controlling the irradiation intensity and dose appropriately. However, there is still the possibility of localized transient heating of absorbing chromophores. The local transient rise in temperature of absorbing biomolecules may cause structural (e.g., conformational) changes and trigger biochemical activity (secondary dark reactions) such as activation or inhibition of enzymes. To evaluate the contribution of local transient heating of light-absorbing microregions to biochemical activity, Escherichia coli (Karu et al., 1991a) and HeLa cells (Karu, 1992a) were irradiated using femtosecond laser pulses ( = 620 nm, p = 3 10-13 s, f = 0.5 Hz, Ep = 1.1 10-3 J/cm2, Iav =5.5 10-4 W/cm2, Ip = 109 W/cm2), as well as CW laser radiation ( = 632.8 nm, I = 1.3 W/cm2). The irradiation dose required to produce a similar biological effect (an increase in the clonogenic activity of irradiated cells compared with the non-irradiated control) was found to be a factor of about 102103 lower for pulsed radiation than for CW radiation. The minimum size of the microregions transiently heated by irradiation with femtosecond laser pulses was estimated at about 10 , which corresponds to the size of the chromophores of the hypothetical primary photoacceptors the respiratory chain components (Karu et al., 1991a). The effects of slight local heating and a substantial local temperature gradient (Letokhov, 1991) occur for both CW and femtosecond pulses. The difference is that the density of such local heating sites in the case of CW radiation is extremely low because of the low radiation intensity. This density is lower than the density of the absorbing chromophores by a factor of approximately 1071010, depending on the relaxation time and excitation cross section of the chromophores. In the case of femtosecond pulses, the chromophores become excited, which causes a stronger biological response. Therefore, it can be concluded that higher than average local transient heating of microregions with a size characteristic of absorbing chromophores is possible. The evaluation of the transient heat intensity of these microregions and the diffusion of heat from them requires a more detailed analysis. One should note here that the above considerations involved individual cells. The problem of local heating of absorbing chromophores takes entirely new dimensions and signicance when whole tissues are irradiated. Some considerations of this problem are found in papers by Letokhov (1991), Karu et al. (1997), and Karu (2002). The principal use of oxygen in a respiratory chain involves its 4-electron reduction to water. In normal metabolic processes, as well as in a number of nonenzymatic biological reactions, both univalent and

182

Lasers in Medicine

Ca2+ Uptake or ATP Production

NADH Transhydrogenase Q2

cyt b

cyt c1 cyt c SOD R

cyt a/a3
Catalase

H2 O

H2 O2

ROOH GSH GSPX

GSH Reductase
Ca2+ Release

NADPH

or Uncoupling

FIGURE 7.8 The connection of antioxidant enzymes to electron transport, oxidative phosphorylation and Ca2+ ux. Abbreviations: GSH glutathione; GSPx glutathione peroxidase; SOD superoxide dismutase; Cat catalase; R reduced lipid; ROOM lipid peroxide (after Forman and Boveris, 1982).

bivalent reductions of molecular oxygen also occur (Figure 7.8). For example, free radicals like HO.(hydroxyl) and O2 (superoxide) appear as a result of 1-electron reduction. It has been shown that, in mitochondrial electron transport, the superoxide radical is produced, and the production of O2, as well as the product of its dismutation, H2O2, primarily depend on the metabolic state of the mitochondria (Forman and Boveris, 1982). By activating electron ow in the respiratory chain, one can also expect increasing O2 production. Some data in the literature also suggest that, besides the above-mentioned method of generating O2, there is another NADH-related way to generate O2 in the mitochondria (Nohl, 1987). It has been demonstrated experimentally that mitochondria possess a mechanism for the reabsorption of O2, and O2 may be a source of electrons for the oxydative phosphorylation of ADP under physiological conditions (Mailer, 1990). Experimental data show that liver mitochondrial ATP synthesis can be inhibited and promoted by the UV generation of O2 (Dmitriev et al., 1990). Recall also that even a small increase in O2 concentration in a cell results in multiple responses like increase in [Ca2+]i and pHi (alkalization), release of arachidonate, activation of Na+/H+ antiport and Ca2+ ATPase, alteration of Na+ Ca2+ exchange (Murphy et al., 1988; Shibanuma et al., 1988; Kaneko et al., 1990). Experiments measuring the luminol-amplied chemiluminescence of murine splenocytes after near IR irradiation (Karu et al., 1993b) do not exclude the possibility of increased O2 concentration. A comparison of the action spectrum of chemiluminescence stimulated after irradiation with various wavelengths of light and the absorption spectrum of the cyt c oxidase indicated some similarity between these two. As the absorption band of cyt a/a3 at 830 nm is thought to be due to its copper component, and the chemiluminescence emitted by mitochondria is believed to be copper-dependent, one should consider the possibility that some low-power laser effects can be related to increased O2 production. This possibility, discussed in a paper of Karu et al. (1993b), becomes more serious if we take into account the fact that H2O2, the product of O2 dismutation, is involved in a complex set of reactions linking the H2O2 production in mitochondria to the regulation of cellular metabolism (Forman and Boveris, 1982). As shown in this reference and also in Figure 7.8, many antioxidant enzymes are involved in this link. One of the enzymes taking part in the pathway is catalase. Catalase has been shown to be involved in the increase in protein synthesis induced by He-Ne laser light in yeast cells (Karu et al., 1993c). In that study, the activity of catalase was measured immediately after the irradiation of Torulopsis sphaerica and the amount of synthesized protein was tested 18 h later. The activity of catalase as well as the amount of synthesized protein were increased in the irradiated cells. The inhibition of catalase activity by 3-amino1,2,4-triazole at the moment of irradiation suppressed protein synthesis. At the same time, the protein synthesis was not affected by 3-amino-1,2,4-triazole in nonirradiated cells. Possible specic links between the increase in catalase activity and protein synthesis in irradiated cells were discussed in the paper of Karu et al. (1993c).

Low-Power Laser Effects

183

Certain photoabsorbing molecules like porphyrins and avoproteins (some respiratory chain components belong to these classes of compounds) can be reversibly converted to photosensitizers (Spikes, 1989; Giese, 1980). Based on monochromatic visible light action spectra for DNA synthesis in HeLa cells, and spectroscopic data for porphyrins and avins, a hypothesis was advanced that the absorption of light quanta by these molecules is responsible for the generation of singlet oxygen, 1O2 (Karu et al., Kalendo1984a). Chemically active 1g and +gstates of oxygen with energies of 1.0 and 1.5 eV might play an active mediator role in achieving the biological effects of irradiation. This possibility has been considered for some time as a predominant reaction during irradiation of cells with high doses and intensities of light (Karu, 1988; 1989b). The possible role of singlet oxygen in low-power laser effects was discussed in several other papers (Danilov et al., 1990; Friedman et al., 1991). However, experiments demonstrating the generation of singlet oxygen in cells or tissues after irradiation as well as examination of the possible secondary (dark) biochemical reactions are to date nonexistent. Figure 7.9 presents two principal pathways for 1O2 generation, as discussed in the papers cited above. The pathway on the left includes strong permitted transitions and, for this reason, is highly probable (classical photodynamic mechanism). This pathway was discussed in papers by Karu, 1988; Karu et al., 1984a; Friedman et al., 1991. In contrast, the pathway on the right, showing direct excitation of the oxygen molecule, which was discussed in the paper by Danilov et al. (1990), has extremely low probability due to forbidden transitions. The light effects of respiration are oxygen-dependent and prevented by anaerobiosis; it is generally believed that photodynamic reactions promoted by certain respiratory chain components (avins, hemes and Fe-S centers) are associated and occur in aerobic conditions only (Epel, 1973; Giese, 1980; Spikes, 1989; Kim and Jung, 1995). Irradiating yeasts in aerobic and anaerobic conditions with an He-Ne laser increased protein synthesis (which was measured as the nal photobiological macroeffect) in both cases, the only difference being the dose range (Karu et al., 1993d). This nding indicates that, at least in this particular case, the 1O2-connected mechanism is not involved.

FIGURE 7.9 Two principal ways for generation of singlet oxygen in a cell: (a) photodynamic action; (b) direct excitation of triplet oxygen. Details can be found in the text.

Taken together, there is certainly more than one reaction involved in the primary mechanisms of lowpower laser effects. The various possible reactions discussed above are summarized in Figure 7.10. There are no grounds to believe that only one of these processes occurs when a cell is irradiated. An important question for the future is which of these reactions is responsible for a certain low-power laser effect. However, recent experimental results of measurements of redox absorbance changes of living cells after the irradiation (Dube et al., 1997; Karu et al., 1998, 2000) clearly indicate that mechanism based on changes in redox properties of terminal enzymes of respiratory chains might be crucial.

184

Lasers in Medicine

changes in redox properties and acceleration of electron transfer

S1

O2

O2

O2 one-electron auto-oxidation
1

O2

photodynamic action hv Thermal relaxation changes in biochemical activity induced by local transient heating of chromophores

S0

FIGURE 7.10 Possible primary reactions with a photoacceptor molecule after promotion to excited electronic states. This theoretical scheme does not mean that all relevant reactions occur from the rst singlet state.

7.2.3 Secondary Reactions of Light Action

Sections 7.2.1 and 7.2.2 considered photoacceptor molecules and possible primary reactions occurring under irradiation. On the other hand, it is well known that many biochemical reactions in cells occur hours and even days after the irradiation procedure. Usually, the irradiation lasts in a timescale of seconds and minutes. The biological responses of cells occurring later, when the radiation is switched off, are called secondary reactions. The specicity of light action is believed to be due to the quanta absorption by a photoacceptor, cytochrome c oxidase molecule (Section 7.2.1). Also, a avoprotein like NADHdehydrogenase has been discussed as a possible photoacceptor for blue and red light (Karu, 1988, 1989a). If the photoacceptors are located in the mitochondria (Section 7.2.2), then how are the primary reactions occurring in the respiratory chain (Section 7.2.2) connected with DNA synthesis in the nucleus? A scheme presented in Figure 7.11b can explain this. This scheme is based on the fact that a redox chain such as the respiratory chain is capable of controlling cellular homeostasis. The photoexcitation induces changes in cyt a/a3 (Section 7.2.2) or in avinic components of the chain (e.g., NADH-dehydrogenase, Karu, 1988, 1989a); this event can, in turn, cause other redox changes and modulations of biochemical reactions through a photosignal transduction and amplication chain, which leads to a photobiological macroeffect, such as increased proliferation (on Figure 7.11b,c marked by DNA synthesis). The absorption of light quanta by the respiratory chain components causes a short-term activation of the respiratory chain and oxidation of the NADH pool. It is known that oxidation of the NADH pool leads to changes in the redox state of both mitochondria and cytoplasm (Krebs and Veech, 1970; Kohen et al., 1983). The activation of the electron transport chain also results in an increase with proton motive force (pmf, H+), electrical potential of mitochondrial membrane (), ATP pool, and acidication of the cytoplasm. It has been conrmed experimentally that this is also true in the case of respiratory chain activation by irradiation: changes in pmf, , and pH as well as extrasynthesis of ATP have been achieved by irradiating mitochondria with a He-Ne laser (Passarella et al., 1984). By irradiating cells with a wideband visible light of > 400 nm, the enhancement of the activity of ATP-synthase was observed (Nedelina et al., 1985). The ATP level was also found to be raised after irradiation of whole cells: human lymphocytes at 820 nm (Herbert et al., 1989), R3230AC adenocarcinoma cells at 630 nm (Hilf et al., 1986), HeLa cells at 632.8 nm (Karu et al., 1995a). Irradiating human lymphocytes with a He-Ne laser also caused changes in the ultrastructure of mitochondria, such as the formation of giant mitochondria, which was interpreted by the authors as an intensication of the energy metabolism (Bakeeva et al., 1993). A perturbation of the mitochondrial energy production was measured by rhodamine 123 uptake, when human oral broblasts were irradiated at 812 nm (Loevschall and Arenholt-Bindslev, 1994a).

Low-Power Laser Effects

185

FIGURE 7.11 A possible mechanism of light-enhanced proliferation of mammalian cells. Monochromatic visible and near IR radiation initially absorbed by mitochondria (a) eventually causes stimulation of DNA synthesis (c) after numerous intervening dark reactions (b). Eh -shift of cellular redox potential to moe oxidied direction; the arrows and indicate increase or decrease of respective values, the brackets[] indicate the intracellular concentration of respective chemicals.

The acidication of the cytoplasm (rise of intracellular H+ concentration), caused by the activation of the respiratory chain, controls allosterically the activity of the Na+/H+ antiporter situated in the cytoplasmic membrane (Pouyssegur et al., 1985). This enzyme plays a key part in the alkalization of the intracellular medium. A short-term increase in the intracellular pH (pHi) is one of the necessary components involved in the transmission of mitogenic signals in the cell (Hesketh et al., 1985, Moolenaar, 1986; Pouyssegur et al., 1985). A change in pHi (alkalization of the cytoplasm after irradiation) has been measured experimentally. The pH of rat brain was measured during 1 week following photodynamic therapy (Chopp et al., 1990). In a control (light only) experiment, the tissue pH was elevated after irradiation at 70 and 140 J/m2, as compared with nonirradiated control (pH = 0.20.3 units). In an eukaryotic cell, a change in the redox state of the mitochondria causes changes in the redox state of the cytoplasm or, in other words, changes the overall redox state of a cell. In Figure 7.11b, this event is marked by the NAD/NADH couple. More exactly, the overall redox state of the cell represents the net balance between stable and unstable reducing and oxidizing equivalents in dynamic equilibrium and is determined by three couples: NAD/NADH, NADP/NADPH, and GSH/GSSG (GSH = glutathione). These pairs are mutually dependent, and altering the ratio in one couple causes changes in the others. Intracellular pH (pHi) is also closely connected to these ratios. Experiments described in papers by Keyse and Tyrrell (1987, 1989) support the suggestion of the crucial role of redox changes in alterations of cellular metabolism. It has been demonstrated that agents that reduce the level of available glutathione in human skin broblasts, i.e.. that shift the cellular redox

186

Lasers in Medicine

state to a more oxidized direction, also induce heme oxygenase. In other words, it appears that the level of the heme oxygenase is fully regulated by the redox state of the cell (Tyrrell, 1992). Among the agents causing this effect were UVA radiation, visible light at 405 nm, hydrogen peroxide, cadmium chloride, iodoacetamine, and menadione (Keyse and Tyrrell, 1987, 1989). The importance of this nding is as follows: heme oxygenase appears to be a 32-kDa stress protein. Stress proteins are synthesized in cells as a response to the action of a wide variety of chemical and physical factors. The most studied stress proteins are the heat shock proteins (Schlesinger et al., 1982). Factors such as glucose deprivation and hypoxic stress induce the synthesis of stress proteins unrelated to the heat shock proteins, and the 32-kDa stress protein is also believed to be distinct (Keyse and Tyrrell 1989). Keyse and Tyrrell (1989) have proposed that the induction of heme oxygenase may be a general response to an oxidant stress. Modulation of cellular redox state affects gene expression via mechanisms of cellular signalling e.g., NF -B and AP-1 (Figure 7.11b). There is also an indirect indicator that He-Ne laser irradiation alters the redox state of cells. It is known that any shift in the redox state and metabolic activity of a cell will necessarily alter its radiosensitivity, adaptive response to radiation or other forms of endogeneous on exogeneous free radical toxicity (Greenstock, 1986). In experiments when the -irradiated monolayer of HeLa cells was irradiated with a HeNe laser at various time intervals before exposure to ionizing radiation, the viability of the cells previously irradiated with a laser increased signicantly as compared with -irradiated cells (Karu et al., 1994a). The Na+/H+ antiporter is not the only cell membrane enzyme to participate actively in photosignal transduction and amplication chain. Other ion carriers such as Na+, K+-ATPase and enzymes controlling cAMP level in a cell are also activated. The activation of the Na+, K+-ATPase following He-Ne laser irradiation of human erythrocytes (Moroz, 1983) and diode laser ( = 830 nm) irradiation on a rat saphenous nerve (Kudoh et al., 1989) has been established. Experimental data on changes in cellular cAMP level after irradiation with light of various wavelengths provide reason to believe that the action of light on proliferation may be connected with the regulation of cell metabolism via cAMP (Karu et al., 1987a). Far more than one cellular response to irradiation are connected with the plasma membrane. The depolarization of spontaneously active neurons in subesophageal ganglia of Helix pomatia occurred after irradiation with a He-Ne laser (Balaban et al., 1992). The intracellular Ca+ concentration was found to rise during the rst few minutes after He-Ne laser irradiation of human lymphocytes (Karu et al., 1991b) and neutrophils (Loevschall et al., 1994c) as well as bovine sperm cells (Lubart et al., 1997) and rat Schwann cells (van Breugel et al., 1993). Depending on the time elapsed after irradiation and the wavelength used. Irradiation increased both the cellcell and cellglass adhesion (Karu et al., 1996a). Direct measurement of ionic currents through the plasma membrane of both excitable (cardiomyocytes, neurons) and nonexcitable (glial) cells using the patch-clamp technique under He-Ne laser irradiation showed the activation of background channels, probably ATP-dependent K+-channels or Ca+-dependent K+-channels (Karu et al., 1996b). First, the ATP-dependence of the light-sensitive background single channel currents supports the scheme of photosignal transduction chain under discussion (Figure 7.11b). Furthermore, in the same series of experiments, it was established that the light-sensitive ion currents were recorded only in a cell-attached conguration of the patch pipette, i.e., in conditions of cellular integrity, but not in a whole-cell conguration. Indeed, the cascade of biochemical reactions (photosignal transduction and amplication chain) depends on the cellular homeostasis and can occur only in conditions of cellular integrity. The photosignal transduction and amplication chain in its part from plasma membrane to nucleus is not specic for light signal but includes also a standard way of controlling cell proliferation (cAMP level, changes in intracellular contentation of H+, K+, Na+, Ca2+). This is a complicated area in cell biology and the regulation mechanisms are not fully clear. The interested reader is referred to papers of Boyton and Whiteld (1983), Cone (1971), Kaplan (1978), Rozengurt (1986), Rozengurt and Mendoza (1980), Hesketh et al. (1985), Hlser and Frank (1971), Moolenaar (1986), Pouyssegur et al. (1985). The alternation of the cellular homeostasis parameters leads to a parallel shift of different reactions, and it is not easy to establish the causal relationships.

Low-Power Laser Effects

187

In Figure 7.11b, two principal regulation paths by light were suggested. The rst is the existence of a connection between the light-activated redox functions of mitochondria, changes in redox state of cytoplasm, activation of certain redox sensitive transcription factors, the depolarization of cellular membrane, and the rise of intracellular pH (alkalization of cytoplasm). The second is the photoacceptor control over the level of intracellular ATP. As we know, even small changes in ATP level can alter the cellular metabolism signicantly (Brown, 1992).

7.3 Explanation of Controversies and Limitations of Low-Power Laser Effects on Cellular Level
Despite almost 30 years of clinical use, until now, a considerable degree of scepticism concerning the use and clinical efcacy of low-power lasers existed. This is, in part, a consequence of the generally poor quality of many original publications, as well as the lack of critical scientic scrutiny of the many claims made for these devices. In recent years, this situation has begun to change. Many high-quality publications within the medical and scientic literature attest to the clinical usefulness of light sources (lasers and LEDs) and quantify their photobiological effects in the laboratory. Years ago, the usual perception was that laser biostimulation occurred in Eastern laboratories but not in Western. At that time, the actual situation was indeed rather close to this. Very few Western scientists took low-power laser effects seriously and even fewer performed experiments. There have been changes in this attitude recently, and a number of well-designed experiments have been performed on various cells. The reader will nd references to these throughout this book. In some cases, the effects of visible and near IR light on various cellular functions have been described in papers that were not dealing with the issue of low-intensity laser effects at all (e.g., Kato et al., 1981; Marchesini et al., 1989; Chopp et al., 1990; Albrecht-Behler, 1991). This gives a greater reliability to those results for the eld of low-power laser effects. Based on the literature data gathered so far on the cellular level, the main controversies and limitations will be analyzed in this chapter. Some that are no longer topical were mentioned in the Introduction.

7.3.1 Magnitude of Low-Power Laser Effects

One confusing point about low-power laser effects has been their possible degree (a strong effect, a weak effect, or no effect at all). Indeed, even when the same culture of cells is irradiated, the degree of biological response can differ. Based on the discussion and data from Section 7.2, one can clarify this issue with the aid of Figure 7.12. It was concluded in Section 7.2.3 that one key event among the secondary reactions of cellular responses was the change in overall redox state of the irradiated cell. Recall that the overall redox state represents the net balance between stable and unstable reducing and oxidizing equivalents in dynamic equilibrium. The main idea of the diagram in Figure 7.12 is that the cellular response is weak or absent when the overall redox potential of a cell is optimal or near optimal for the particular growth conditions, and stronger when the redox potential of the target cell is initially shifted to a more reduced state (and intracellular pHi is lowered commensurably). This explains why the degrees of cellular response can differ markedly in different experiments, and can sometimes be nonexistent. This suggestion is supported by the following experimental results. In two sets of experiments, the cellular redox state was changed articially before irradiation. In the rst case, the redox state of the HeLa cells was lowered (shifted into a more reduced direction) by oxygennitrogen transition. By such a transition, the NADH uorescence increases, which means a shift to the reduced direction (Chance et al., 1962; upper corner of Figure 7.13). When these cells were irradiated with light of various wavelengths marked on the abscissa of Figure 7.13, the new action spectrum appeared to be quite different from the action spectrum in conditions of normal O2 partial pressure. As seen in Figure 7.13, the stimulative effect of light at 620630 nm on DNA synthesis is stronger than in normal cultivation conditions. A new peak appears at = 570 nm.

188

Lasers in Medicine

FIGURE 7.12 Schematic illustration of the action of monochromatic visible and near IR radiation on a cell. The magnitude of the irradiation effect is determined by redox potential of the cell at the moment of irradiation. TABLE 7.1 Changes in DNA Synthesis Rate in Exponentially Growing HeLa Cells 1.5 h after Irradiation with He-Ne Laser in Various Doses with or without the Presence of 1 mM Methylene Blue (MB) (Modied from Karu, 1989a)
Dose, J/m2 0 6 Irradiated Cells MB absent MB present 102 5% 200 9% 150 8% 227 11% 183 12% 213 8% 153 9% 189 6% 30 102 103

Nonirradiated Cells MB absent MB present 100% 202 6%

In the second set of experiments, the redox state of HeLa cells was shifted before irradiation to oxidized direction using methylene blue solution. Methylene blue participates readily in a number of oxidationreduction reactions as an electron acceptor modulating the intracellular redox state. This supravital dye has been used as an anti-inammatory and antibacterial agent, as an antidote for cyanide and carbon monooxide poisoning, and also as a radio-protectant. It has been shown that methylene blue affects the concentration of intracellular reducing agents (Hrushesky et al., 1985) and markedly increases oxygen consumption of fresh tissue homogenates in the presence of adequate NAD(P)H (Hrushesky et al., 1988). As seen in Table 7.1, irradiating cells pretreated with methylene blue with a He-Ne laser stimulated the DNA synthesis only a little (a maximum of 25% at a dose of 30 J/m2 ) because the control level was already elevated twice before the irradiation (202% as compared with nontreated control, 100%) (Karu, 1989a). During in vitro measurements of superoxide dismutase and catalase activity after He-Ne laser irradiation (both enzymes have absorption bands near 633 nm), no changes were detected when the enzyme solutions had a pH corresponding to maximal catalytic activity. In contrast, very pronounced photore-

Low-Power Laser Effects

189

FIGURE 7.13 Action of visible monochromatic light on DNA synthesis in HeLa cells in normal cultivation conditions and after 60 s O2/N2 transition. DNA synthesis rate was measured 2.5 h after irradiation of a dose of 100 J/m2 as described by Karu (1989). In the upper corner: a microuorimetric recording of the kinetics of uorescence observed in oxygen-nitrogen transition of kidney cells (adapted from the paper of Chance et al., 1962).

activation effects were detected when the enzymes were irradiated under non optimal lowered pH conditions (Zubkova, 1978; Gorbatenkova et al., 1988). Recall also that a jump in intracellular pH (pHi) after irradiation has been measured experimentally. pHi is a parameter of cellular homeostasis, which is closely connected with cellular redox state. These experiments proved that pHi was increased by 0.20 units in mammalian cells due to irradiation with red light (Chopp et al., 1990) and 0.32 units in E. coli (Quickenden et al., 1995). These two works concluded that, by means of irradiation, it is indeed possible to shift the pHi and the overall redox state of cells into a more oxidized direction, as was suggested in Figure 7.12. Let us now see if this working hypothesis for cellular phenomena can be applied to clinical low-power laser effects as well. Under normal conditions, mammalian cells maintain a very precise pHi, usually between 7.0 and 7.2 (Roos and Boron, 1981). The normal pH of arterial blood is 7.4 and healthy tissues usually have pH values between 7.0 and 7.4 (Vaupel, 1977). Growth factors, neurotransmitters, and direct cellcell interactions can modify the regulation of pHi in receptive cells (Roos and Boron, 1981). The two most important areas of low-power laser therapy are wound healing and treatment of chronic inammations. Both these conditions are characterized by decreased oxygen tension (decreased pO2, hypoxia) and acidosis (decreased pH) (Kittlick, 1986). Most normal tissues have a pO2 about 40 mm Hg, while, in hypoxic states, the pO2 is in the range of 05 mm Hg (Vaupel, 1977; Freitas, 1991). In the case of normal regeneration, wound hypoxia is a transient condition eventually replaced by a normal pO2 during the nal stages of regeneration. The situation is different with chronic injuries. Chronic inam-

190

Lasers in Medicine

mation is characterized by continued aerobic glycolysis and by a redox shift (measured as the NAD+/NADH ratio) toward a reduced state. The pO2 of rheumatoid synovial uid, for example, was found to fall as low as 0 mm Hg (Kittlick, 1986). As also noted by Kittlick (1986), local hypoxia is considered to be one of the primary causes of rheumatism. When irradiating fresh wounds, the effect of irradiation can be minimal or nonexistent. This happens in cases where the proliferation is active and the regeneration of tissue integrity occurs at a more or less maximal (normal) rate. This may be the reason there is often no phototherapeutic effect observed when irradiating fresh experimental wounds. This problem was discussed earlier (Karu, 1987, 1989). Two examples illustrate this conclusion: He-Ne laser irradiation failed to accelerate healing of fresh wounds in two rat models (Allendorf et al., 1997). Laser radiation at 630 nm signicantly increased the healing of chronic wounds of genetically diabetic mice (Yu et al., 1997). It is known that intracellular pH affects the contractile function of the heart, its metabolic reactions, ion exchange and calcium homeostasis. A fall in extracellular pH, whatever the mechanism, causes a decrease in the hearts ability to contract, as occurs, for example, in ischemia (Poole-Wilson, 1989). It should be noted here that low-power lasers are specically used in treating ischemia (Kashuba, 1981; Korochkin, 1988). The working hypothesis that an alteration of the intracellular redox state plays the crucial role in lowpower laser effects may have a broader signicance than simply that of explaining the laser action on wound healing and inammation. When a pharmacological agent interacts with a receptor, usually situated on the outer cell membrane, it sets in motion a chain of events known as a stimulusresponserecovery cycle. These processes are complicated and consist of many biochemical reactions, not to mention a multitude of agonists and receptors. There is, however, a common step involved in the stimulusresponserecovery cycle: a change in the redox state and pHi. Analyzing the data from the extensive review by Roth et al. (1983), one can see that this phenomenon is involved in almost every branch of pharmacology. The working hypothesis introduced above can be extended as follows: the irradiation-altered redox state of cells modulates their response to biochemical agonists. To illustrate how sensitive and diverse cellular responses to changes in the cellular redox state may be, refer to Puppi et al. (1968). The authors of that study articially altered the redox state in their tissue models (frog ileum and frog heart) and then studied the action of acetylcholine and adrenaline. Oxidation with methylene blue caused acetylcholine to act in a stimulatory manner, but, after reduction with ascorbate, acted in an inhibitory fashion. In contrast, adrenaline was stimulative in oxidizing states, and inhibitive under reducing conditions. An attempt was made to quantify the magnitude of the irradiation effect as a dependence on the metabolic status of the cells at the moment of exposure (Tiphlova and Karu, 1991b). The E. coli culture was cultivated using various nutrient conditions; minimal growth medium M9 supplemented with glucose, glycerol, or arabinose (Figure 7.14b,c,d) or Hottinguer broth, (Figure 7.14a). The kinetics of growth curves of intact cells appeared to be quite different in different conditions. The population grown with glucose (Figure 7.14b, curve 1) started dividing almost without a latent period. The specic rate of the exponential growth was equal to 0.55h-1. Irradiating this culture caused an increased specic rate of exponential growth k = 0.78 h1 (Figure 7.14b, curve 2). Here k is determined from Xt = X0ekt; X0 denotes the number of cells in the sample at the beginning of the exponential phase of growth and Xt, denotes the number after t hours of growth (Andersen and von Meyenburg, 1980). The difference in the number of cells in the exposed and unexposed cultures after an incubation period of 60 min was 126% (Figure 7.14b). In the nutrient medium with glycerol, the control culture had a latent period of about 15 min, and the specic rate of exponential growth was 0.64 h1 (Figure 7.14c, curve 1). In an irradiated culture, the lag-period was almost fully reduced and the specic rate of growth was increased (k = 0.80 h1). The difference in the number of cells in the exposed and unexposed cultures at 1-h incubation was 130% (Figure 7.14c).

Low-Power Laser Effects

191

FIGURE 7.14 Growth curves of cultures of E. coli WP2 (1) control and (2) irradiated with a He-Ne laser at a dose of 4 103 J/m2, cultivated in (a) Hottinguer broth or M9 medium supplemented with (b) glucose, (c) glucerol, and (d) L-arabinose (adapted from Tiphlova and Karu, 1991b). Growth stimulation is measured as the ratio of number of irradiated to control cells after 1 h incubation (in %).

When arabinose was used as the carbon source, the latent period of unexposed culture was longer than 1 h, and the specic growth rate was 0.74 h-1 (Figure 7.14d, curve 1). The effect of irradiation in this case manifested itself rst of all in the almost complete disappearance of the lag period and an abrupt increase in the number of cells throughout the rst hour of incubation. For this curve, k = 0.80 h-1, and the difference in the number of cells in the exposed and unexposed cultures came to 145% (Figure 7.14d). A comparison of curves 1 and 2 in Figure 7.14 makes it possible to reveal the characteristic features of inuence of irradiation with red light on E. coli cultures under different conditions. The difference between the number of cells in exposed and unexposed cultures is small for cultures with a short latent period (Figure 7.14b, c) and large if the control culture begins to divide after a long lag period (Figure 7.14d, as well as Figure 7.14a). The specic rate of exponential growth of exposed culture is almost the same in all cases (k = 0.780.80 h-1). These results show that irradiation under these specic experimental conditions provides the same maximal k for a bacterial culture no matter what the parameters of the growth curve were. This is a biological limit on growth stimulation.

192

Lasers in Medicine

The effect of irradiation, however, shows up for just a short time, being maximal during the rst 45 to 60 min of incubation. Later (2 to 4 h after the start of incubation), there is no difference in the number of cells in exposed and control cultures. This regulatory mechanism is characteristic of all models studied (Figure 7.14). The experimental results obtained indicate that the magnitude of photostimulation of E. coli division depends on the metabolic state of the initial culture or, in other words, on the presence and duration of the latent period. On the other hand, there is a maximal specic rate of exponential growth (in our experimental conditions, 0.80 h1 ), which limits the degree of photostimulation effect.

7.3.2 Diversity of Low-Power Laser Effects

One major controversy surrounding low-power laser effects has always been their apparent diversity and universality. When viewed from the outside, low-power laser therapy seems to be a panacea. If one considers that low-power laser effects are attributable to light action on the respiratory chain, their diversity as well as their universality can be explained quite easily. The respiratory chains for complex eukaryotic (Figure 7.3a) and prokaryotic cells differ markedly. The carriers can have different structures, and, in the case of prokaryotic cells, the respiratory chains usually branch with alternative electron transfer pathways to multiple terminal acceptors. There are also some minor differences in respiratory chains of complex (e.g., mammalian) cells and primitive eukaryotic cells (yeasts). Despite these differences, the functional principles of the chains as well as the operation of carriers are universal. Precisely this fact makes low-power laser effects so diverse, at least on a cellular level. An irradiation-induced alteration in the overall cellular redox state is considered a key step in secondary reactions (Section 7.2.3), but it does not explain all irradiation effects at a rst glance. It is well documented that more than one function of cells can be inuenced by irradiation. For example, some authors have described an increase in the proliferation of broblasts (Boulton and Marshall, 1986; van Breugel and Dop Br, 1992; Nara et al., 1992; Loevschall and Arenholt-Bindslev, 1994b) and keratinocytes (Steinlecher and Dyson, 1993; Loevschall and Arenholt-Bindslev, 1996). Others have found that irradiation increased the motility of broblasts (Noble et al., 1992) and keratinocytes (Haas et al., 1990), collagen synthesis in broblasts (Lam et al., 1986; Balboni et al., 1986), taxis of broblasts (AlbrechtBheler, 1991) and autocrine production of bFGF (a growth factor) by broblasts (Yu et al, 1994). Analysis of this data clearly showed that when one cellular function was altered by irradiation, the others remained unchanged. For example, when the motility of keratinocytes changed, no effects on their proliferation were noticed (Haas et al., 1990). When the proliferation of keratinocytes was increased, no signicant effects on their migration were found (Loevschall and Arenholt-Bindslev, 1996). The broblasts that responded to light by increasing collagen synthesis did not increase proliferation (Lam et al., 1986; Balboni et al, 1986). In fact, the production of collagen type I by broblasts was affected by irradiation in an inverse manner to cell proliferation; when cell proliferation increased, collagen type I production decreased (van Breugel and Dop Br, 1992). It should be noted that collagen synthesis by irradiated broblasts was found to be ascorbate-dependent, i.e., connected to redox events (Labbe et al., 1990). Another fact in connection with this: In experiments with broblasts, procollagen production was, on average, enhanced approximately fourfold by He-Ne laser radiation. The greatest enhancement (36-fold) was noted in cultures that initially synthesized procollagen at a relatively low level, while an insignicant effect was achieved in cultures that already actively synthesized the procollagen (Lam et al., 1986). Explaining such a diversity of possible responses of one cell type to monochromatic light is not trivial. A possible explanation in the case of a prokaryotic E. coli cell could be as follows. Stimulation of E. coli growth by monochromatic visible light has been suggested to be a pH-dependent genetic process (Tiphlova and Karu, 1991a). However, growth stimulation is not the only process utilizing the pH gradient. A priority in the utilization of pH for ATP synthesis, nutrient transport, taxis, and initiation of genetic processes has been proposed. This priority enables the metabolic processes utilizing pH to be regulated and coordinated (Tiphlova and Karu, 1991a).

Low-Power Laser Effects

193

How the priority between motility, secretory and replicative functions in broblasts or keratinocytes is regulated to enable a cell to respond in a certain way is not yet understood. Solving this problem will have a great impact on the understanding of low-power laser effects.

7.3.3 Biological Limitations of Low-Power Laser Effects

It is well known that cultured cells are heterogeneous with regard to proliferative activity of subpopulations (Nias, 1968; Azzarone and MacLeira-Coelho, 1982; Hassel and Stanek, 1983). There is no reason to believe that all subpopulations and even all individual cells respond to irradiation in an absolutely similar way. The following examples illustrate this suggestion. One conclusion from experiments with cultured cells was that only the proliferation of slowly growing subpopulations can be stimulated by irradiation (Karu et al., 1987b). Also, the experiments with HeLa cells demonstrated that one of the effects of He-Ne laser irradiation of these cells is a decrease of the duration of the G1 period (Karu et al., 1987b). It means that the cells in the G1 phase of the cellular cycle certainly responded to irradiation in these experimental conditions. The results of experiments with E. coli grown in various growth media showed that all E. coli batches contained a subpopulation that, in response to irradiation, rapidly began a new cycle of replication and division (Figure 7.14). The number of cells in this subpopulation depended on the cultivation conditions, being smaller in faster-growing populations (e.g., the glucose-grown culture), and larger in slowergrowing populations (e.g., the arabinose-grown culture). First, one can suppose that, in cells of this light sensitive subpopulation, the particular metabolic state necessary for the division could be established for the same reason, and second, that the irradiation helped them to achieve this active state. Also, this set of experiments (Tiphlova and Karu, 1991b) clearly proved that there is a limit in the specic growth rate of all populations (0.80 h-1) that is not dependent on growth conditions, and it is not possible to stimulate the populations that are already growing at this rate. This was also found to be the reason E. coli growth was not stimulated in summer but the effect was maximal in winter. In autumn and winter, the intact culture featured a relatively slow growth. In spring and summer, when the growth of that culture accelerates and the growth rate of the control culture is almost comparable to that of the culture exposed to the optimum dose of red light in the autumn-winter period, the irradiation had but little effect (Karu, 1989a). In principle, a similar effect of a biological limit was found when the luminol-amplied chemiluminescence was measured in murine spleen cells after He-Ne laser irradiation and treatment with an object of phagocytosis Candida albicans (Karu et al., 1989). The irradiation effect was detectable in these cases only when the cells were treated rst with a low concentration of Candida albicans (5107 particles/ml in our experimental conditions). At the concentration of 1 108 particles/ml the chemiluminescence was maximally activated, and no additional activation by light was possible. Electron microscopy studies of human lymphocytes demonstrated that, after He-Ne laser irradiation, changes in the structures of nuclei existed in 70% of the cells (Manteifel and Karu, 1992). Only 2547% of 3T3 broblasts responded to near IR radiation by extending their pseudopodia toward the monochromatic light source. Importantly, in these experiments, the percentage of reactive cells depended on the irradiation wavelength (Albrecht-Breheler, 1991). Some cell types were found to not respond to the irradiation. For example, the silent neurons of Helix pomatia did not respond to He-Ne laser irradiation, while the spontaneously active neurons had a strong response under the same experimental conditions (Balaban et al., 1992). In some cases, it is possible to cause only a partial activation of cells. An example here is the mitogenic activation of human peripheral lymphocytes. A summary of these experiments is presented in Figure 7.15. The irradiation of lymphocytes with a He-Ne laser (56 J/m2, 5.6 W/m2, 10 s) induced short-term changes in these cells, both qualitatively and quantitatively similar to those caused by the mitogen phytohemagglutinin (PHA). There was a two- to threefold increase in intracellular Ca2+ concentration 2 min after either irradiation or PHA treatment. Enhanced acridine orange binding to chromatin (reecting the changes in the structure of chromatin) and increased RNA synthesis were both recorded, with maximal

194

Lasers in Medicine

FIGURE 7.15 Comparison of lymphocyte activation by phytohemagglutinin and He-Ne laser radiation (Fedoseyeva et al., 1988b, Karu et al., 1991b; Manteifel and Karu, 1992; Shliakhova et al., 1996).

values occurring 1.52 h after the treatment. These ndings indicate an enhanced template activity of chromatin. Ultrastructural changes in the nucleoli of both irradiated and PHA-treated cells evidenced an intensication of rRNA metabolism, including its synthesis, processing and transport. It was concluded that there is a similarity in the transcription activation of r-genes under PHA action and He-He laser irradiation. Both the irradiation and PHA treatment caused an accumulation of preliminary terminated proto-oncogene c-myc RNA in the lymphocytes, but caused no variation in the amount of full-length c-myc RNA. Despite similarities in the early cell response to PHA and irradiation, the irradiated lymphocytes did not enter the S-phase, i.e., no full mitogenic activation occurred in the irradiated lymphocytes. In contrast to the PHA that is continuously present in the cellular suspension throughout mitogenic activation, the laser light is on for only 10 s. This period is apparently not long enough for the light stimulus to cause the entire cascade of reactions needed for blast transformation. Detailed data about lymphocyte activation can found in the papers of Karu (1992b, 1996b). Direct measurement of ionic currents through a cell membrane under He-Ne laser radiation using the patch-clamp technique showed that no light effects on the voltage-activated whole-cell ionic currents were found in any type of the cells studied (rat spinal cord neurons, rat hippocampus pyramidal neurons, guinea pig cardiomyocytes, rat glial cells, bovine pulmonary artery endothelial cells). The only ionic

Low-Power Laser Effects

195

current type inuenced by the radiation in all the cells studied was the background single-channel current recorded in the cell-attached conrmation of the patch pipette (Karu et al., 1996b). A greater quantity of nitric oxide (NO) was produced by mouse peritoneal macrophages after 660 nm irradiation only in the cases where the cells were cultivated in the presence of a special NO inducer medium. Irradiation without the NO inducer medium had no effect on enhancing NO production (Naim et al., 1996). Only partially inactivated Na+, K+-ATPase was reactivated by irradiation with a 904 nm diode laser, while, at the same time, laser irradiation did not stimulate the activity of the native enzyme (Bolognani and Volpi, 1991). The photoreactivation of superoxide dismutase and catalase by radiation = 632.8 nm was signicant only in conditions where the pH of these solutions was reduced before irradiation (Zubkova, 1978; Gorbatenkova et al., 1988). These examples clearly prove that it is not possible to activate a process that is activated already or occurring at speed near maximal. This conclusion underlines, once more, that laser biostimulation is not a general phenomenon, but occurs only in certain circumstances. Bertoloni et al. (1993) established that not all strains of E. coli displayed an appreciable photoresponse to He-Ne laser irradiation. Only those cells whose normal rate of growth was particularly slow due to the presence of factors inhibiting cell reproduction were stimulated. Table 7.2 presents the E. coli strains studied by Bertoloni et al. (1993). The strains whose growth was stimulated by He-Ne laser radiation are also marked in Table 7.2. It is quite possible that some genetic factors are involved in the photosensitivity of E. coli strains. This suggestion is supported by data from Voskanyan et al. (1985, 1986). They found that the radio-protective action of He-Ne laser radiation was genotype-dependent in the case of E. coli K-12 mutants AB1157 and Gam 444. Also, a fast-growing wild strain of E. coli was not stimulated by He-Ne laser radiation (G.Bertoloni and G.Jori, personal communication).
TABLE 7.2 E. coli Strains Derived from K12 in which Photosensitivity was Investigated (Modied from Bertoloni et al., 1993)
Growth Stimulation No. 1. 2. 3. 4. 5. 6. 7. Strains db1344 921 J53 AB1157 db1229 db1245 CGSC6405 Genotypes araD, argF-lac, b, pts, relA, rpsL, lamB(am), deoC thr, leu, met, las, nal, tonA, hsdR pro, met thr, leu, pro, arg, his, thi, lac, gal, rpsl, supE his, trp, las, Sm, Tn10 Sm, ura, arg, thi, his, ade, lac araD, argF-lac, b, non-9, gyrA, relA, resL, metE, btu: Tn10, thi, deoC Log-Phase Culture + + + + + + Plateau Phase Culture + +

In the case of yeasts, not all strains were stimulated equally. It was found that the degree of protein synthesis stimulation was in agreement withthe liability of metabolism, i.e., with the possibility of reconstructing the metabolism (Fedoseyeva et al., 1988a). Also, not all mammalian cell lines cultivated in vitro responded to irradiation in equal measure; some did not respond at all (Marchesini et al., 1989). All this data indicates, again, that laser biostimulation is not a general phenomenon, and serious biological limits exist. Recall also that the effects depend on light parameters (Karu 1989a, 1998), and last, but not least, biochemical and morphological changes in the irradiated cells depend on the time elapsed after the irradiation. This means that the measurements should be made at the right time. For example, a signicant increase in DNA synthesis of E. coli WP2 trp was detectable only during the rst 10 min after inoculation (Tiphlova and Karu, 1991a) and growth stimulation was maximal 1 h after the irradiation and vanished later on (Tiphlova and Karu, 1988). When irradiating the yeasts, the irradiation effect was revealed in the exponential phase of culture growth (Fedoseyeva et al., 1984). At the same time, the growth stimulation effect of E. coli WP2 was detectable only in the lag phase of growth (Tiphlova

196

Lasers in Medicine

and Karu, 1988,1991a,b). All these limits, limitations and difculties should be taken into account when planning an experiment.

7.3.4 Dependence on the Physiological Status of the Host Organism

Such a complex substance as human blood reacts to irradiation as a dependence on the physiological (immunological) status of the host organism. The measurements performed on whole blood of 28 clinically healthy people, showed that laser irradiation (820 nm, 292 Hz, 1104 J/m2) had no statistically signicant effect (stimulation or inhibition) on the luminol-amplied chemiluminescence (CL) of healthy human blood (Karu et al., 1995b). The situation was quite different when the CL of the blood from a healthy female donor was recorded during periods of acute viral respiratory illness (Karu et al., 1993b). Note that CL reects the intensity of free radical reactions (Baggiolini and Wyman, 1990). On the organism level, where complex regulation mechanisms are involved and more than one cell type is irradiated simultaneously, the situation is more complicated. In some cases, the effects of irradiation can be not only stronger or weaker, as discussed above, but can even change its nature (i.e., being alternately suppressive or stimulative for radiation with constant parameters). In the example given below, the irradiation effect (both its nature and magnitude) is correlated with percentage variation of cellular components in a complex mixture (Karu et al., 1993a). A murine spleen suspension consists mostly of lymphocytes (6085%) and neutrophils (1015%). Other cell types (monocytes, eosinophils, basophils, plasmacytes, etc.) are generally present at 0.12%. These cellular percentages vary during illnesses and with the age of the organism (Cheville, 1983). Spleen suspension from A/Sn mice of two different age groups (1.52 months and 89 months old) was irradiated with a laser diode at 820 nm (292 Hz, 1.1 103 J/m2) and luminol-amplied CL was recorded. It was possible to nd correlations between the cellular composition in the spleen suspension and the irradiation effect. The linear regression analysis revealed direct correlations between the irradiation effect and the percentage of plasmacytes (p < 0.001), neutrophils (p < 0.001), myelocytes (p < 0.01), and an inverse correlation between the irradiation effect and the percentage of lymphocytes (p < 0.001) (Figure 7.16). Thus, an increase in the percentage of neutrophils, neutrophil precursors, and plasmacytes, as well as a decrease in the percentage of lymphocytes, was found to increase the irradiation effect (increased CL). It is known that plasmacytes develop from activation and transformation of B lymphocytes on contact with antigens. The amount of neutrophils and plasmacytes in spleen is known to increase with aging and as a result of chronic infections, as well as other diseases (Cheville, 1983). As one can see from the correlations shown in Figure 7.16, there are regions (i.e., respective cellular compositions of the spleen) of CL stimulation and inhibition as well as a rather broad range of cellular compositions between the two. One should also note that, using the correlation equations shown in Figure 7.16, it is possible to predict the effect of irradiation (820 nm, 292 Hz) when the cellular composition is known. On the other hand, it is also possible to make some predictions about cellular composition when the irradiation effect is known.

7.4 Clinical Applications of Low-Power Laser Effects

7.4.1 Focus of Application
Low-power laser therapy is now used by physiotherapists (treatment of a wide variety of acute and chronic musculoskeletal aches and pains), dentists (treatment of inamed oral tissue, healing of diverse ulcerations), dermatologists (treatment of edema, indolent ulcers, burns, dermatitises), rheumatologists (attenuation of pain, treatment of chronic inammations and autoimmune diseases), and by other specalists as well as general practitioners. This modality is also used rather widely in veterinary medicine

Low-Power Laser Effects

197

EFFECT OF IRRADIATION %

220

160

100

r=0.65 p<0.001 y=7x+61 0 6 12 18 24

40 220

NEUTROPHILS, %

160

100

r=0.74 p<0.001 y=34x+72 0 1 2 3 4 5

PLASMACYTES, %

40 220

160

100

r=0.51 p<0.01 y=5x+84 0 10 20

MYELOCYTES, %

40 220

30

160

100

r=-0.59 p<0.001 y=2.6x+326 40 60 80 100

LYMPHOCYTES, %

40

FIGURE 7.16 Linear regression analysis between percentage of (A) neutrophils, (B) plasmacytes, (C) myelocytes, or (D) lymphocytes in cellular suspension and the effect of the irradiation (820 nm, 292 Hz, 1103 J/m2) r denotes the coefcient of the correlation (after Karu et al., 1993a).

(especially in racehorse training centers) and in sports medicine and rehabilitation clinics (reduction of swelling and hematoma, relief of pain and improvement of mobility, treatment of acute soft tissue injuries). For more details, the following books are recommended: Baxter (1994); Oshiro and Calderhead (1988,1991); Calderhead et al. (1991); Triner et al. (1999). Some areas where the most experimental and clinical work has been done are in wound healing, injured nerve regeneration, pain attenuation and treatment of diverse rheumatological conditions. The clinical effects of light can be classied as either direct or indirect, depending on whether the light causes an effect to occur within the irradiated tissue or whether a nervous or neuroendocrine signal is generated in the irradiated area and causes a systemic effect in another part of the body. As a whole, the eld of low-power laser effects is diverse. The versatility of this kind of treatment has been a controversial and poorly explained feature, making laser biostimulation an uncertain medical modality. Some possible explanations for the controversies of low-power laser effects were presented in Section 7.3.

198

Lasers in Medicine

7.4.2 Is Low-Power Laser Therapy a Mature Modality?

The clinical eld of low-power laser therapy is characterized by a variety of diverse methodologies, and often by a lack of double-blind tests or even control groups. Moreover, the laser exposure parameters used (dose, wavelength, intensity, CW or pulsed mode, pulse duration and repetition rate) vary widely. In many cases, it has not been proven that laser therapy is better than alternative therapeutic treatments. Moreover, the operative mechanism on an organism level has not been identied. The conclusion from the preceding is that low-power laser therapy is not yet a mature modality. It is still developing.

7.4.3 Possible Optimal Light Parameters for Low-Power Laser Therapy

In the early days of low-power laser therapy, He-Ne lasers in the 130 mW power range were used exclusively. These are still employed today, although currently, a large range of additional equipment varying from single diode probes to multiple diode heads and multihead systems is manufactured and marketed as well. Such intense activity by the commercial manufacturers is surprising for two reasons. First, lowpower laser therapy, not yet a mature medical modality, is out of the mainstream of medical applications. Second, the optimal parameters of light needed (wavelength, intensity, CW or pulsed mode, pulse width, repetition rate, and duty cycle) are not clear, and quite probably vary for different medical applications. On the laser therapy equipment market, one can now nd semiconductor lasers operating at a wide variety of wavelengths (6601300 nm), pulse frequencies (230,000 Hz), and output powers (1100 mW). The combined action of various wavelengths is used in numerous diode heads and other integrated equipment (e.g., an He-Ne laserIR diode laser combination). Multi-wavelength effects have been very poorly investigated to date, and the possible benet of such equipment is not clear. In fact, one should be careful in using simultaneous irradiation at certain wavelengths. For example, it has been found that the simultaneous dichromatic irradiation of a cellular monolayer with light in the red and far red regions produced a substantially reduced or completely absent response (Figure 7.4) despite the fact that each wavelength-stimulated proliferation when used alone (Karu et al., 1984a,b). Much experimental work should be done before optimal parameters for clinical applications can be recommended. Discussed below are some considerations stemming from experiments at the cellular level. An ideal laser for low-power therapy must emit a beam at a wavelength capable of reaching not only supercial tissue but also deeper layers. In addition, this wavelength must provide a photoexcitation of photoacceptor molecules to give a maximal photobiological response. These two objectives can be met by analyzing the action spectra of various photobiological effects, absorption spectra of both photoacceptor molecules responsible for these photobiological effects and the tissue chromophores, and wavelength penetration depth data. As no action spectra for clinical effects have yet been produced, action spectra for cellular effects must be used. Figure 7.17a presents three of these, and Figure 7.17b illustrates the absorption spectrum of a possible photoacceptor for low-power laser effects. Figure 7.17c presents absorption spectra of the main tissue chromophores in redfar red regions, and Figure 7.17d depicts skin penetration depth data for light at these wavelengths. Analysis of the data presented in Figure 7.17 indicates that the optimal wavelengths are near 760 nm and between 810840 nm. In both regions, the tissue chromophores have a weak absorption (Figure 7.17c) and light penetration into the skin is near maximal (Figure 7.17d). To illustrate the deep penetration of light into tissues, recall that ~2% of incident light at 580600 nm reaches the uterine lumen in pregnant rats, and this is believed to be enough to inuence physiological processes related to circadian rhythms (Jacques et al., 1987). The cytochrome c oxidase (cyt a/a3) has an absorption maximum in the oxidized form at 830 nm (extinction coefcient = 1.4 mM cm-1) due to the copper components of this molecule (Wharton and Tzagaloff, 1964). Three points should be emphasized here:

Low-Power Laser Effects

199

1. The absorption of cyt a/a3 near 830 nm is weak, but sufcient to cause photochemical reactions. In some tissues (e.g., myocardium, brain), the density of mitochondria is high, which also results in a higher cyt a/a3 concentration. This circumstance enables redox states of the mitochondria to be measured in situ using a beroptic based spectrophotometric technique (Rea et al., 1985). 2. Figure 7.17b presents the absorption spectrum of cyt a/a3 (cytochrome c oxidase) in the oxidized form insofar as the absorption spectra of its redox intermediates are not yet available. Recall that some absorption spectra of a monolayer of HeLa cells were presented in Figures 7.6 and 7.7, as well as in Karu et al. (2001). It was supposed that these spectra belong to a redox intermediate of cytochrome c oxidase. Recall also that the fully oxidized form of cytochrome c oxidase appeared to be insensitive to He-Ne laser radiation (Pastore et al. 2000). 3. Recall that, to be light sensitive, the electron ow and ATP synthesis must be coupled (Kato et al., 1981). Irradiation experiments should be performed with cells or mitochondria with functioning respiratory chain. It could be expected that the irradiation of cristallic cytochrome c oxidase (or its solution), which became available recently (Gennis and Ferguson-Miller, 1995), is not a useful experiment to explain the mechanism of low-power laser therapy.

FIGURE 7.17 An evaluation of potentially optimal wavelengths for low-power laser therapy: (a) action spectra for three photobiological effects (DNA and RNA synthesis rate in plateau-phase HeLa cells (Karu et al., 1984b) and chemiluminescence (CL) emittance by murine splenocytes (Karu et al., 1993b); (b) absorption spectra of suggested photoacceptor in eukaryotic cells, cytochrome c oxidase (Wharton and Tzagaloff, 1964), (c) absorption of principal tissue chromophores; (d) penetration depth of various light wavelengths into tissue (Sliney and Wolbarsht, 1980).

200

Lasers in Medicine

Taking all this into consideration, laser devices emitting at 820830 nm seem to have wavelengths sufcient for low-power therapy. Lasers operating at another promising wavelength, 760 nm, only recently became commercially available. Also, 680 nm wavelength is good, with the additional benet for clinicians that this wavelength is visible by eye, which facilitates the irradiation procedure. Quite possibly, for different applications, different wavelengths are optimal. In experiments with mammalian cellular monolayers, optimal doses in the red and far red regions were found to be near 102 J/m2 (Karu et al., 1984a,b). During irradiation of real tissues, the doses are necessarily higher due to light losses through reection and scattering. For example, it has been suggested that, in the case of peptic ulcer treatment, the clinical dose providing an equivalent response to the laboratory 102 J/m2 effect should be approximately 400 times higher (Karu, 1989a). Experiments with cellular cultures, E. coli and identied neurons have shown that, in many cases, the intensity of light is a more important parameter than the total dose (reviews of Karu, 1989a, 1998). In other words, these experiments demonstrated that one of the rules of photochemistry, the reciprocity rule (Bunsen-Roscoe law), is often not valid. The reciprocity rule says that an effect does not depend on the intensity or the irradiation time when the dose is constant. This issue of the importance of the light intensity was specically investigated and it was found that at least two reaction channels (mechanisms) are involved in producing the same photobiological macroeffect (Karu et al., 1994b). One channel was activated by a certain dose, and the reciprocity law was valid in that case. A second channel was activated by a certain irradiation time (independent of dose), and therefore, the reciprocity rule was not valid in that case. In experiments where the reciprocity rule was not valid, a strict threshold (e.g., near 6 W/m2 at 633 nm (Karu et al., 1984a) and a very narrow range of optimal intensities was recorded in the light intensity vs. cellular response relationship. The issue of optimal intensities at different wavelengths for various medical applications has not been addressed. On an empirical basis, it has been proposed that, for CW 830 nm diode lasers, the ideal power output is ~60 mW (Moore and Calderhead, 1991). Other unanswered questions include whether medical devices should employ continuous or pulsed laser radiation, and, if pulsed radiation has some benet, what are the optimal pulse frequencies, durations, and duty factors? Examples presented by Karu (1998) for various cellular systems clearly demonstrate the importance of laser parameters associated with pulsed operation. However, it is not yet clear that a pulsed source has a specic clinical benet over a CW device producing the same wavelength and average power density. Quite probably, the optimal laser light parameters associated with pulsed operation are different for different clinical applications.

7.5 Summary
Biological responses of cells to visible and near-lR (laser) radiation occur due to physical or chemical changes in photoacceptor molecules, components of respiratory chains like NADH-dehydrogenases and cytochrome c oxidase. As a result of the photoexcitation of electronic states, the following physical or chemical changes can occur: Alteration of redox properties and acceleration of electron transfer No release from the catalytic center of cytochrome c oxidase Changes in biochemical activity due to local transient heating of chromophores One-electron auto-oxidation and O2 production Photodynamic action and 1O2 production

The primary physical or chemical changes induced by light in photoacceptor molecules are followed by a cascade of biochemical reactions in the cell that need no further light activation and occur in the dark (photosignal transduction and amplication chains). These reactions are connected with changes in cellular homeostasis parameters. The crucial step here is thought to be an alteration of the cellular

Low-Power Laser Effects

201

redox state a shift toward oxidation is associated with stimulation of cellular vitality, and a shift toward reduction is linked to inhibition. Cells with a lower than normal pH, where the redox state is shifted in the reduced direction, are considered to be more sensitive to the stimulative action of light than those with the respective parameters optimal or near optimal. This circumstance explains the possible variations in observed magnitudes of low-power laser effects. Light action on the redox state of a cell via the respiratory chain also explains the diversity of low-power laser effects. Beside explaining many controversies in the eld of low-power laser effects (i.e., the diversity of effects, the variable magnitude or absence of effects in certain studies), the proposed redox-regulation mechanism may be a fundamental explanation for some clinical effects of irradiation, for example, the positive results achieved in treating indolent wounds, chronic inammation, and ischemia, all characterized by acidosis and hypoxia. One unsolved mystery is integral to low-power laser effects. It is rather well documented that more than one of the cellular functions (e.g., the replicative, motile, and secretory functions of broblasts or keratinocytes) can be inuenced by radiation at the same wavelength. Even more interesting is that, when one of these cellular functions is altered by irradiation, the others remain unchanged (Karu, 1998). This nding clearly indicates the existence of some unknown regulatory mechanism establishing priorities in a cell. Identifying this priority regulation system (i.e., the preferential utilization of light-generated proton motive force, ATP, etc.) will have profound importance for future studies of low-power laser effects. Finally, a remark about the reliability of low-power laser effects should be made. Years ago, a rather usual perception existed that laser biostimulation occurred in Eastern but not Western laboratories. At that time, the actual situation was indeed rather close to this. Very few Western scientists took laser biostimulation seriously, and even fewer performed experiments. However, lately there have been changes in this attitude, and a number of well-designed experiments have been performed on various cells. These data are summarized by Karu (1998). In some cases, the effects of visible and near IR light on various cellular functions have been described in papers not dealing with the issue of laser biostimulation at all (e.g., Kato et al., 1981; Chopp et al., 1990; Albrecht-Behler, 1991). This circumstance gives a higher reliability to those results for the eld of low-power laser effects. Alternatively, these ndings clearly indicate that the responsivness of mammalian cells to various wavelengths of monochromatic visible and near IR radiation can be used to understand some physiological processes (e.g., adaptation). Laser biostimulation is only one eld where the cellular sensitivity (which seems to be increased in injured or otherwise stressed systems) to monochromatic visible and near IR radiation is used. Albrecht-Behler (1991) described the phenomenon of broblast phototaxis (pseudopodia moving toward a monochromatic light source). In the paper of Kato et al., (1981) it was suggested that mitochondria in a special part of bird brain could work as photoreceptors for a photobiological process relating to gonadal growth. To complete this chapter, two remarks from a classical paper by K. Smith (1980) should be remembered. First, just because humans cant see through the human body does not mean that the human body is opaque to all wavelengths of light. Second, because light (especially red light) can penetrate deep into human tissues, and because absorbed light can cause photochemistry, it is appropriate to be concerned with the biological consequence of such absorbed light in the tissues of man.

References
Adar, F. and Yonetani, T. (1978) Resonance Raman spectra of cytochrome oxidase. Evidence for photoreduction by laser photons in resonance with the Soret band. Biochem. Biophys. Acta, 502, 8086. Afanasyeva, N.I., Karu, T.I. and Tiphlova, O.A. (1995) Oxidases bd and bo as primary photoacceptors when Escherichia coli cells are irradiated with monochromatic visible (laser) radiation. Doklady Akad. Nauk Doklady Biophysics (Moscow), 345, 404406.

202

Lasers in Medicine

Albrecht-Behler, G. (1991) Surface extensions of 3T3 cells toward distant infrared light sources. J.Cell Biol., 114, 494502. Allendorf, J.D.F., et al. (1997) Helium-Neon laser irradiation of uences of 1.2 and 4 J/cm2 failed to accelerate wound healing as assessed by both wound contracture rate and tensile strength. Lasers Surg. Med., 20, 340345. Anders, J.J., Mysliwiec, V. and Preston, S. (1995) Low-power laser irradiation alters the uorescence intensity of rhodamine 123 labeled mitochondria in human broblasts. Lasers Surg. Med., Suppl., 7, p.78 (abstr. 30). Andersen, K. and von Meyenburg, K. (1980) Are growth rates of Escherichia coli in batch cultures limited by respiration? J. Bacteriol., 144, 114118. Azzarone, B. and Macieira-Coelho, A. (1982) Heterogeneity of the kinetics of proliferation within human skin broblastic cell populations. J. Cell Sci., 57, 177187. Babcock, G.T. and Wikstrm, M. (1992) Oxygen activation and the conservation of energy in cell respiration. Nature, 356, 302309. Bakeeva, L.E. et al. (1993) Formation of giant mitochondria in human blood lymphocytes after He-Ne laser irradiation. Molecular Biology (Moscow) 27, 608617. Baggiolini, H. and Wyman, M.P. (1990) Tuning on the respiratory burst. Trends Biochem. Sci., 15, 6972. Balaban, P. et al. (1992) He-Ne laser irradiation of single identied neurons. Lasers Surg. Med., 12, 329337. Balboni, G.C. et al. (1986) Effects of He-Ne /IR laser irradiation on two lines of normal human broblasts in vitro. Arch. Ital. Anat. Embriol., 91, 179188. Basford, J.R. (1989) Low-energy laser therapy: controversies and new research ndings. Lasers Surg. Med., 9,1-5. Baxter, G.D. (1994) Therapeutic Lasers. Theory and Practice. Edinburgh, London: Churchill Livingstone. Belkin, M., Zaturunsky, B. and Schwartz, M. (1988) A critical review of low energy laser bioeffects. Lasers Light Ophthalmol. 2, 63-71. Berns, M.W. and Nelson, J.S. (1988) Laser applications in biomedicine.Part I: biophysics, cell biology, and biostimulation. J. Laser Applications, 1,34-39. Bertoloni, G., et al. (1993) Biochemical and morphological changes in Escherichia coli irradiated by coherent and noncoherent 632.8 nm light. J. Photochem. Photobiol., B: Biology, 18, 191196. Bolognani, L. and Volpi, N. (1991) Reaction of partially inactivated enzymes by low-power laser beams (IPL). La Radiol. Medica, Suppl., 1 (= 4), 8186. Boulton, M. and Marshall, J. (1986) He-Ne laser stimulation of human broblast proliferation and attachment in vitro. Lasers Life Sci., 1, 125134. Boynton, H.L. and Whiteld, J.F. (1983) The role of cyclic AMP in cell proliferation: a critical assessment of the evidence. In: Advances in Cyclic Nucleotide Research. P. Greengard and G.A. Robinson, Eds., New York: Raven Press, v. 15, pp.195294. Brown, G.C. (1992) Control of respiratory and ATP synthesis in mammalian mitochondria and cells. Biochem. J, 284, 171182. Brunori, M. and Chance, B., Eds. (1988) Cytochrome oxidase: structure, function and physiopathology. Ann. N.Y. Acad. Sci., v. 550. Calderhead, R.G., Moore, K.C. and Nagasawa, A. (1991). Low Reactive-Level Laser Therapy: Practical Application. Chichester: J. Wiley Sons. Capaldi, R.A. (1990) Structure and function of cytochrome c oxidase. Ann. Rev. Biochem., 59, 569596. Chance, B. et al. (1962) Intracellular oxidation-reduction states in vivo. Science, 137, 499508. Cheville N.F. (1983) Cell Pathology. 2nd ed. Ames (Iowa, USA): The Iowa State Univ. Press. Chopp, H. et al. (1990) Chronic metabolic measurement of normal brain tissue response to photodynamic therapy. Photochem. Photobiol., 52, 10331036. Cone, C.D. (1971) Unied theory on the basic mechanism of normal mitotic control and oncogenesis. J. Theor. Biol., 30, 151181.

Low-Power Laser Effects

203

Danilov, V.P., Zakharov, S.D. and Ivanov, A.V. (1990) Photodynamic injury of cells in red and IR absorption peaks of endogenous oxygen. Doklady Akad. Nauk SSSR (Moscow), 311, 12551258. Dmitriev, L.F., Ivanova, M.V. and Ivanov, I.I. (1990) Liver mitochondrial ATP synthesis can be promoted by UV generation of superoxide. Biol. Membr., (Moscow) 7, 961965. Dube, A., Gupta, P.K. and Bharti, S. (1997) Redox absorbance changes of the respiratory chain components of E.coli following He-Ne laser irradiation. Lasers Life Sci., 7, 173180. Epel, B.L. (1973) Inhibition of growth and respiration by visible and near ultraviolet light. In: Photophysiology. A.C. Giese, Ed., New York: Acad. Press, v. 8, pp. 209229. Fedoseyeva, G.E., et al. (1984) Effect of the He-Ne laser radiation on the reproduction rate and protein synthesis of the yeasts. Laser Chemistry, 5, 2733. Fedoseyeva, G.E., et al. (1988a) The activation of yeast metabolism with He-Ne laser radiation. I. Protein synthesis in various cultures. Lasers Life Sci., 2, 137146. Fedoseyeva, G.E., et al. (1988b) Human lymphocyte chromatin changes following irradiation with HeNe laser. Lasers Life Sci., 2, 197205. Forman, H.J. and Boveris, A. (1982) Superoxide radical and hydrogen peroxide in mitochondria. In: Free Radicals in Biology, A. Pryor, Ed., New York and London: Acad. Press, vol. 5, pp. 6590. Friedman, H., et al. (1991) A possible explanation of laser-induced stimulation and damage of cells cultures. J. Photochem. Photobiol. B: Biology, 11, 8795. Freitas, I. (1991) Tumor hypoxia, reoxygenation and oxygenation strategies. Possible role in photodynamic therapy. J. Photochem. Photobiol., B: Biology, 11, 330. Gennis, R. and Ferguson-Miller, S. (1995) Structure of cytochrome c oxidase, energy generator of aerobic life. Science, 269, 10631064. Giese, A.C. (1980) Photosensitization of organisms with special reference to natural photosensitizers. In: Lasers in Biology and Medicine, F. Hillenkampf, R. Pratesi and C. Sacchi, Eds., pp. 299314, New York: Plenum Press. Gorbatenkova, E.A. et al. (1988) Mechanism of photoreaction of superoxide dismutase with He-Ne laser radiation. Doklady Acad. Nauk SSSR Doklady Biophysics (Moscow), 299, 9951000. Gordon, S.A. and Surrey, K. (1960) Red and far red light action on oxidative phosphorylation. Radiat. Res., 12, 325339. Gouterman, M. (1978) Optical spectra and electronic structure of porphyrins and related rings. In: The Porphyrins, D. Dolphin, Ed. New York: Acad. Press, v. 3, pp. 1165. Govindjee, Ed., (1982) Photosynthesis, vol. 1, New York: Acad. Press. Greenstock, C.L. (1986) Radiation-induced aging and induction and promotion of biological damage. In: Radiation, Free Radical Damage, and Aging, J.E. Johnson et al., Eds., New York: Alan Liss, pp. 197230. Haas, A.F., et al. (1990) Low-energy He-Ne laser irradiation increases the motility of cultured human keratinocytes. J. Invest. Dermatol., 94, 822826. Harris, D.M. (1988) Laser biostimulation: review and hypothesis. Laser Topics, 10 (3), 9-14. Hartmann, K.M. (1983) Action Spectroscopy. In: The Biophysics, W. Hoppe, W. Lohmann, H. Marke and H. Ziegler, Eds., Chapt. 3.2.7, pp. 115144. Hassel, T.M. and Stanek, E.J. (1983) Evidence that healthy human gingiva contains functionally heterogeneous broblast subpopulations. Arch. Oral. Biol., 28, 617625. Hathaway, B.J. (1987) Copper. In: Comprehensive Coordination Chemistry. G. Wilkinson, R.D. Gillard and J.A. McCleverty, Eds., Oxford: Pergamon Press, vol. 5, pp.534774. Herbert, K.E. et al.(1989) Effect of laser light at 820 nm on adenosine nucleotide levels in human lymphocytes. Lasers Life Sci., 3, 3746. Hesketh, T.R. et al. (1985) A common sequence of calcium and pH signals in the mitotic stimulation of eukaryotic cells. Nature, 313, 481484. Hilf, R. et al. (1986) Relationship of mitochondrial function and cellular adenosine triphosphate levels to hematoporphyrin derivative-induced photosensitization in R3230AC mammary tumors. Cancer Res., 46, 211217.

204

Lasers in Medicine

Hrushesky, W.J.M. et al. (1985) Modifying intracellular redox balance: an approach to improving therapeutic index. Lancet, 8428, 565567. Hrushesky, W.J.M. et al. (1988) Methylene blue protects intestinal mucosa from free radical-mediated sublethal damage. Free Radical Biol. Med., 5, 207213. Hughes, M. (1987) Coordination Compounds in Biology. In: Comprehensive Coordination Chemistry. G. Wilkinson, R.D.Gilland and J.A. McCleverty, Eds., Oxford: Pergamon Press, vol. 6, pp. 541753. Hlser, D. and Frank, W. (1971) Stimulating von Kulturen embryonaler Rattenzellen durch eine ProteinFraction aus ftalem Klberserum. I. Electrophysiologische Messungen an den Oberhenmembranen. Z. Naturforsch., 266, 10451048. Iwata, S. et al. (1995). Structure of 2.8 resolution of cytochrome c oxidase from Paracoccus denitricans. Nature, 376, 660669. Jacques, S.L., Weaver, D.R. and Reppert, S.H. (1987) Penetration of light into the uterus of pregnant rats. Photochem. Photobiol., 45, 637641. Kaneko, M., Singal, P.K. and Dhalla, N.S. (1990) Alteration in heart sarcolemmal Ca2+-ATPase and Ca2+binding activities due to oxygen free radicals. Basic Res. Cardiol., 85, 4554. Kaplan, J.G. (1978) Membrane cation transport and the control of proliferation of mammalian cells. Annu. Rev. Physiol., 40, 1941. Karlin, D. and Zubieta, J., Eds. (1983) Copper Coordination Chemistry. New York: Acad. Press. Karu, T.I. (1987) Photobiological fundamentals of low-power laser therapy. IEEE J. Quantum Electr., QE23, 17031717. Karu, T.I. (1988) Molecular mechanism of the therapeutic effect of low-intensity laser radiation. Lasers Life Sci., 2, 5374. Karu, T.I. (1989a) Photobiology of Low-Power Laser Therapy. Chur, London: Harwood Acad. Publ. Karu, T.I. (1989b) Photobiology of low-power laser effects. Health Physics, 56, 691-704. Karu, T. (1990) Effects of visible radiation on cultured cells. Photochem. Photobiol, 52, 1089-1099. Karu, T.I. (1991a). Stimulation of metabolic processes by low-intensity visible light: a scientic basis for biostimulation, In: Laser Applications in Biology and Medicine (Wolbsrsht, M. Ed.), vol.5, pp.l-47. New York: Plenum Press. Karu, T. (1991b) Effects of visible (laser) radiation on cultured cells. In: Laser Systems for Photobiology and Photomedidne (Martellucci, S., Chester, A.N., Eds.), pp.89-113. New York: Plenum Press. Karu, T.I. (1992a) Local pulsed heating of absorbing chomophores as a possible primary mechanism of low-power laser effects. In: Laser Applications in Medicine and Surgery, G. Galletti, L. Bolognani and G. Ussia, Eds., Bologna: Monduzzi Editore, pp. 253258. Karu, T. (1992b) Derepression of the genome after irradiation of human lymphocytes with He-Ne laser. Laser Therapy, 4, 5-24. Karu, T. (1996a) Activation of metabolism of nonphotosynthezing microorganism with monochromatic visible (laser) light: A critical review. Lasers in the Sciences, 7 11-34. Karu, T.I. (1996b) Derepression of genome and alteration of DNA template activity in nonproliferating cells by means of UV and visible laser radiation. In: Biomedical Optical Instrumentation and LaserAssisted Biotechnology. Verga-Scheggi, A.M. et al., Eds. Dordrecht, Boston, London: Kluwer Acad. Publ., pp. 332. Karu, T. (1998) The Science of Low-Power Laser Therapy. London: Harwood Acad. Publ. Karu, T.I. (2002) Low Power Laser Effects, in Biomedical Photonics Handbook. Vo-Dinh, T., Ed., Boca Raton: CRC Press, Ch. 48b (in press). Karu, T.I. and Afanasyeva, N.I. (1995) Cytochrome oxidase as primary photoacceptor for cultured cells in visible and near IR regions. Doklady Akad. Nauk Doklady Biophysics (Moscow), 342, 693695. Karu, T.I., Kalendo, G.S. and Letokhov, V.S. (1981) Control of RNA synthesis rate in tumour cells HeLa by action of a low-intensity visible light of copper laser. Lettere al Nuovo Cimento, 32, 5559. Karu, T.I. et al. (1984a) Biostimulation of HeLa cells by low intensity visible light. II. Stimulation of DNA and RNA synthesis in a wide spectral range. Il Nuovo Cimento D, 3, 309318.

Low-Power Laser Effects

205

Karu, T.I. et al. (1984b) Biostimulation of HeLa cells by low intensity visible light. III. Stimulation of nucleic acid synthesis in plateau phase cells. Il Nuovo Cimento D, 3, 319325. Karu, T.I., Letokhov, V.S. and Lobko, V.V. (1985) Biostimulation of HeLa cells by low-intensity visible light. IV. Dichromatic irradiation. Il Nuovo Cimento D, 5, 483496. Karu, T.I. et al. (1987a) Effect of irradiation with monochromatic visible light in cAMP content in Chinese hamster broblasts. Il Nuovo Cimento D, 9, 12451251. Karu, T.I., Pyatibrat, L.V. and Kalendo, G.S. (1987b) Biostimulation of HeLa cells by low-intensity visible light. V.Stimulation of cell proliferation in vitro by He-Ne laser radiation. Il Nuovo Cimento D, 9, 14851494. Karu, T.I. et al. (1989) Induced by He-Ne laser radiation respiratory burst of phagocytic cells. Lasers Surg Med., 9, 585588. Karu, T.I. et al. (1991a) Comparison of the effects of visible femtosecond pulses and continous wave laser radiation of low average intensity on the clonogenicity of Escherichia coli. J. Photochem. Photobiol. B: Biology, 10, 339344. Karu, T., Smolyaninova, N. and Zelenin, A. (1991b) Long-term and short-term responses of human lymphocytes to He-Ne laser radiation. Lasers Life Sci., 4, 167178. Karu, T., Andreichuk, T. and Ryabykh, T. (1993a) Changes in oxidative metabolism of murine spleen following diode laser (660950nm) irradiation: effect of cellular composition and radiation parameters. Lasers Surg. Med., 13, 453462. Karu, T., Andreichuk, T. and Ryabykh, T. (1993b) Suppression of human blood chemiluminescence by diode laser radiation at wavelength 660, 820, 880 or 950 nm. Laser Therapy, 5, 103109. Karu, T.I., Lyapunova, T.S. and Pomoshnikova, N.A. (1993c) The activation of yeast metabolism with He-Ne laser radiation. IV. Relationship between the activity of catalase and protein synthesis. Lasers Life Sci., 5, 251258. Karu, T.I., Kutomkina, E.V. and Pomoshnikova, N.A. (1993d) The activation of yeast metabolism with He-Ne laser radiation. III. Protein synthesis in Saccharomycodes ludwigii grown in aerobic and anaerobic conditions. Lasers Life Sci., 5, 259266. Karu, T., Pyatibrat, L. and Kalendo, G. (1994a) Irradiation with He-Ne laser can inuence the cytotoxic response of HeLa cells to ionizing radiation. Int. J. Radiat. Biol., 65, 691697. Karu, T. et al. (1994b) Two different mechanisms of low-intensity laser photobiological effects, J. Photochem. Photobiol. B: Biology, 24, 155162. Karu, T., Pyatibrat, L. and Kalendo, G. (1995a) Irradiation with He-Ne laser increases ATP level in cells cultivated in vitro. J. Photochem. Photobiol. B: Biol., 27, 219223. Karu, T.I., Andreichuk, T.N. and Ryabykh, T.P. (1995b) On the action of semiconductor laser radiation ( = 820 nm) on the chemiluminescence of blood of clinically healthy humans. Lasers Life Sci., 6, 277282. Karu, T.I. et al. (1996a) Effects of monochromatic low-intensity light and laser irradiation on adhesion of HeLa cells in vitro. Lasers Surg. Med., 18, 171177. Karu, T. et al. (1996b) He-Ne laser radiation inuences single-channel ionic currents through cell membranes: a patch-clamp study. Lasers Life Sci., 7, 3548. Karu, T.I., Ryabykh, T. and Letokhov, V.S. (1997) Different sensitivity of cells from tumor-bearing organisms to continuous-wave and pulsed laser radiation (=632.8 nm) evaluated by chemiluminescence test: III Effect of dark period between pulses. Lasers Life Sci. 7, 141-156. Karu, T.I. et al. (1998) Changes in absorption spectra of monolayer of living cells after irradiation with low intensity laser light. Doklady Akad. Nauk Doklady Biophysics (Moscow), 360, 267-270. Karu, T.I. et al. (2001) Changes in absorbance of monlayer of living cells by laser radiation at 633, 670 and 820 nm. J. Selected Topics Q. Electron., in perss. Karu, T.I., Tyatibrat, L.V. and Kelando, G.S. (2001) Donors of NO and pulsed radiation at = 820 nm exert effects on cell attachment to extracellular matrices. Toxicol. Lett., 121, 5761. Kashuba, V.A. (1985) The use of lasers in cardiology. Kardiologiya (Moscow), 4, 123129.

206

Lasers in Medicine

Kato, M., Shinizawa, K. and Yoshikawa, S. (1981) Cytochrome oxidase is a possible photoreceptor in mitochondria. Photobiochem. Photobiophys., 2, 263269. Keyse, S.M. and Tyrrell, R.M. (1987) Both near ultraviolet radiation and the oxidizing agent hydrogen peroxide induce a 32-kDA stress protein in normal human skin broblasts. J. Biol. Chem., 262, 1482114825. Keyse, S.M. and Tyrrell, R.M. (1989) Heme oxygenase is the major 32-kDA stress protein induced in human skin broblasts by UVA radiation, hydrogen peroxide, and sodium arsenite. Proc. Natl. Acad. Sci., 86, 99103. Kim, C.S. and Jung, J. (1995) Inactivation of the respiratory chain in plant mitochondria by visible light: the primary target for photodamage and endogeneous photosensitizing chromophores. J. Photochem. Photobiol., B: Biol., 29, 135139. Kittlick, P.D. (1986) Inammation, glycoliytic metabolism, and glucosaminoglucans. Exp. Pathol., 30, 19. Kohen, E. et al. (1983) Metabolic control and compartmentation in single living cells. Cell Biochem. Funct., 1, 316. Korochkin, N.H. (1988) Clinical and pathological aspects of efcacy of low-power laser therapy by treating of heart ischemia. Sov. Med. (Moscow), No 11, 2327. Krebs, H.K. and Veech, R.L. (1970) Regulation of the redox state of the pyridine nucleotides in rat liver. In: Pyridine Nucleotide-Dependent Dehydrogenases, H. Sund, Ed. Berlin, Heidelberg and New York: Springer, pp. 413434. Kudoh, Ch. et al. (1989) Effects of 830 nm GaAlAs diode laser radiation on rat saphenous nerve K+, Na+, ATPase activity: a possible pain attenuation mechanism examined. Laser Therapy, 1, 63-68. Labbe, R.F. et al. (1990) Laser photobiological mechanisms: in vitro studies using ascorbic acid uptake and hydroxyproline formation as biochemical markers of irradiation response. Lasers Surg. Med., 10, 201207. Lam, T. et al. (1986) Laser stimulation of collagen synthesis in human skin broblast cultures. Lasers Life Sci., 1, 6167. Letokhov, V.S. (1991) Effects of transient local heating of spatially and spectrally heterogeneous biotissue by short laser pulses. Il Nuovo Cimento D, 13, 939948. Lichtenstein, G.I. (1979) Multinuclear Redox Metalloenzymes. Moscow: Nauka (in Russian). Loevschall, H. and Arenholt-Bindslev, D. (1994a) Low level laser therapy effect on mitochondrial rhodamine 123 uptake in human oral broblasts. Laser Therapy, 6, 3031. Loevschall, H. and Arenholt-Bindslev, D. (1994b) Effects of low level diode laser (GaAlAs) irradiation of broblasts of the human oral mucosa in vitro. Laser Surg. Med., 14, 347354. Loevschall, H. et al. (1994c) Effects of low level laser irradiation on cytosolic Ca2+ in human neutrophils in vitro. Laser Therapy, 6, 31. Loevschall, H. and Arenholt-Bindslev, D. (1996) Near-infrared low level laser effect on human oral keratinocyte migration and proliferation in vitro. Lasers Life Sci., 7, 129140. Lontie, R., Ed. (1984) Copper Proteins and Copper Enzymes. Boca Raton: CRC Press, v. 13. Lubart,R. et al. (1997) Changes in calcium transport in mammalian sperm mitochondria and plasma membranes caused by 780 nm radiation. Lasers in Surg. Med., 21, 493-499. Mailer, K. (1990) Superoxide radical as electron donor for oxidative phosphorylation of ADP. Biochem. Biophys. Res. Comm., 170, 5964. Manteifel, V.M. and Karu, T.I. (1992) Ultrastructural changes in human lymphocytes under He-Ne laser radiation. Lasers Life Sci., 4, 235248. Manteifel, V., Bakeeva, L. and Karu, T. (1997) Ultrastructural changes in chondriome of human lymphocytes after irradiation with He-Ne laser: appearance of giant mitochondria. J. Photochem. Photobiol. B: Biology, 38, 2530. Marchesini, R. et al. (1989) Effect of low-energy laser irradiation on colony formation capability in different human tumor cells in vitro. Lasers Surg. Med., 9, 5962. Marcus, R.A. and Sutin, N. (1985) Electron transfers in chemistry and biology. Biochem. Biophys. Acta, 811, 265322.

Low-Power Laser Effects

207

Moolenaar, W.H. (1986) Regulation of cytoplasmic pH by Na+/H+ exchange. Trends Biochem. Sci., 11, 141143. Moore, K.C. and Calderhead, R.G. (1991) The clinical application of low-incident power density 830 nm GaAlAs diode laser radiation in the therapy of chronic intractable pain: a historical and optoelectronic rationale and clinical review. Int. J. Optoelectr., 6, 503520. Morimoto, Y. et al. (1994) Effect of low-intensity argon laser irradiation on mitochondrial respiration. Lasers Surg. Med., 15, 191199. Moroz, A.M. (1983) Na, K,-ATPase activity after the laser irradiation. Ukr. Biochem. Zh. 55, 674-676. Murphy, J.G., Smith, T.W. and Marsh, J.D. (1988) Mechanisms of reoxygenation-induced calcium overload in cultured chick embryo heart cells. Am. J. Physiol., 254, H1133H1141. Naim, J.O. et al. (1996) The effect of low level laser irradiation on nitric oxide production by mouse macrophages. Lasers Surg. Med., Suppl. 8, p. 7, (abstr. 28). Nara, J. et al. (1992) Stimulative effect of He-Ne laser irradiation on cultured broblasts derived from human dental pulp. Lasers Life Sci., 4, 249256. Nedelina, O.S., Brzevskaya, O.N. and Kayushin, L.P. (1985) Redox regulation in ATP synthesis. Biophysics (Moscow), 30, 179191. Nias, A.H.W. (1968). Clone size analysis: a parameter in the study of cell population kinetics. Cell Tissue Kinet., 1, 153165. Nicholls, P. and Elliot, W.B. (1974) The cytochromes. In: Iron Biochemistry and Medicine. A. Jacobs and M. Worwood, Eds. New York and London: Acad. Press., pp. 221-272. Noble, P. et al. (1992) Locomotory characteristies of broblasts within a three-dimensional collagen lattice: modulation by a He-Ne soft laser. Lasers Surg. Med., 12, 669674. Nohl, H. (1987) A novel superoxide radical generator in heart mitochondria. FEBS Lett., 214, 269273. Oshiro, T. and Calderhead, R.G. (1988) Low Level Laser Therapy Chichester; John Wiley and Sons. Oshiro, T. and Calderhead, R. G., Eds. (1991) Progress in Laser Therapy. Chichester, London: John Wiley. Palmer, G. (1993). Current issues in the chemistry of cytochrome c oxidase. J. Bioenergetics Biomembr., 25, 145151. Palmer, G. et al. (1978) Electronic states of heme in cytochrome oxidase. In: Mechanisms of Oxidizing Enzymes. T.P. Singer, Ed., Amsterdam: Elsevier, v. 1, pp. 221238. Passarella, S. et al. (1983) Evidence of changes induced by He-Ne laser irradiation in the biochemical properties of rat liver mitochondria. Bioelectrochem. Bioenerg., 10,185198. Passarella, S. et al. (1984) Increase of proton electrochemical potential and ATP synthesis in rat liver mitochondria irradiated in vitro by He-Ne laser. FEBS Lett., 175, 9599. Passarella, S. et al. (1988a) Increase in the ADP/ATP exchange in the rat liver mitochondria irradiated in vitro by He-Ne laser. Biochem. Biophys. Res. Commun., 156, 978986. Pastore, D. et al. (2000) The effect of He-Ne laser light on the mitochondrial cytochrome c oxidase. Int. J. Radiat. Biol. 76, 863869. Poole-Wilson, P.A. (1989) Regulation of intracellular pH in the myocardium, relevance to pathology. Mol. Cell. Biochem., 89, 151155. Pouyssegur, J. et al. (1985) Cytoplasmic pH, a key determinant of growth factor-induced DNA synthesis in quiescent broblasts. FEBS Lett., 90, 115119. Puppi, A. et al. (1968) Effect of environmental oxydoreduction on the action of acetylene. Acta Biol. Hung., 19, 517. Quickenden, T.R., Daniels, L.L. and Byrne, L.T. (1995) Does low-interesting He-Ne laser radiation affect the intracellular pH of infact E. coli cells? In: Laser-Tissue Interaction VI. S.L. Jacques and Katzir, Eds., SPIE Proc. v. 2391, pp. 535545. Rea, P.A. et al. (1985) Noninvasive optical methods for the study of cerebral metabolism in the human newborn: a technique for the future? J. Med. Eng. Technol., 9, 160166. Roos, A. and Boron, F. (1981) Intracellular pH. Physiol. Rev., 61, 296334. Roth, Z., Chayen, N. and Dikstein, S. (1983) The involvement of the intracellular redox state and pH in the metabolic control of stimulus-response coupling. Int. Rev. Cytology, 85, 3961.

208

Lasers in Medicine

Rozengurt, E. (1986) Early signals in the mitogenic response. Science, 234, 161166. Rozengurt, E. and Mendoza, S. (1980) Monovalent ion uxes and the control of cell proliferation in cultured broblasts. Ann. N.Y. Acad. Sci., 339, 175190. Schlesinger, M.J., Ashburner, M. and Tissiers, A., Eds. (1982) Heat Shock: From Bacteria to Man. Cold Spring Harbor, New York: Cold Spring Harbor Lab. Senger, H. Ed. (1980, 1984) Blue Light Effects in Biological Systems. Berlin: Springer-Verlag. Shibanuma, M., Kuroki, T. and Nose, K. (1988) Superoxide as a signal for increase in intracellular pH. J. Cell. Physiol. ,136, 379383. Shliakhova, L.N. et al. (1996) Expression of c-myc gene in irradiated at 670 nm human lympocytes. A preliminary report. Laser Life Sci., 7, 107114. Siegel, H., Ed. (1971-1981) Metal Ions in Biological Systems. New York, Basel: M. Dekker, v. 113. Sliney, D. and Wolbarsht, H. (1980) Safety with Lasers and Other Optical Sources. New York: Plenum Press. Smith, K.C. (1980) Common misconceptions about light. In: Lasers in Photomedicine and Photobiology. Eds. R. Pratesi and C. Sacchi, Berlin: Springer-Verlag, pp. 2325. Smith K.S. (1991) The photobiological basis of low-level laser radiation therapy, Laser Therapy 3,19-25. Spikes, J.D. (1989) Photosensitization. In: The Science of Photobiology, 2nd edition. K.C. Smith, Ed., New York, Plenum Press: pp. 79111. Spiro, G. Ed. (1981) Copper Proteins. New York: Wiley Interscience. Steinlecher, C.W.B. and Dyson, M. (1993) The effects of low level laser therapy on the proliferation of keratinocytes. Laser Therapy, 5, 6573. Sucheta, A., Georgiadis, K.E., and Einarsdottir, O. (1997) Mechanism of cytochrome c oxidase-catalyzed reduction of dioxygen to water: evidence for peroxy and ferryl intermediates at room temperature. Biochemistry, 36, 554565. Tiphlova, O.A. and Karu, T.I. (1988) Escherichia coli growth stimulation by low-intensity monochromatic visible light. J. Photochem. Photobiol., 46, 467473. Tiphlova, O. and Karu T. (1991a) Action of low-intensity laser radiation on Escherichia coli. CRC Crit. Rev. Biomed. Eng., 18, 387412. Tiphlova, O. and Karu, T. (1991b) Dependence of Escherichia coli growth rate on irradiation with HeNe laser and growth substrates. Lasers Life Sci., 4, 161166. Tsukahara, T. et al. (1995) Structures of metal sites of oxidized bovine heart cytochrome c oxidase at 2.8 . Science, 269, 10691074. Tsukahara, T. et al. (1996). The whole structure of the 13-subunit oxidized cytochrome c oxidase at 2.8 . Science, 272, 11361144. Tunr, J. and Hode, L. (1999) Low Level Laser Therapy: Clinical Practice and Scientic Background. Stockholm: Prima Books. Tyrrell, R.M. (1992) UVB radiation induction of heme oxygenase gene expression clues as to the mechanism and signicance of the response. Photochem. Photobiol., 55S, 75S. van Breugel, H.H.F.I. and Dop Br, P.R. (1992) Power density and exposure time of He-Ne laser irradiation are more important than total energy dose in photobiomodulation of human broblasts in vitro. Laser Surg. Med., 12, 528537. van Breugel, H.H.F.I., Engels, C. and Br, P.R. (1993) Mechanism of action in the laser induced photobiomodulation depend on the wavelength of the laser. Lasers Surg. Med., Suppl. 5, p. 9 (abstr. 36). Vaupel, P. (1977). Hypoxia of neoplastic tissue. Microvasc. Res., 13, 339408. Vekshin, N.L. and Mironov, G.P. (1982) Flavin-dependent oxygen uptake in mitochondria under illumination. Biophysics (Moscow), 27, 537539. Verkhovsky, M.I., Morgan, J.E. and Wikstrm, M. (1996) Redox transitions between oxygen intermediates in cytochrome c oxidase. Proc. Acad. Sci. USA, 93, 1223512339. Voskanyan, K.S.H.et al. (1985) Effect of He-Ne laser radiation on x-ray sensitivity of Escherichia coli K12 cells. Radiobiologiya (Moscow), 25, 557559. Voskanyan, K.S.H. et al. (1986) Modication of the damaging effect of -particles on E. coli K-12 by low-intensity laser radiation. Radiobiologiya (Moscow), 26, 375377.

Low-Power Laser Effects

209

Wilkinson, G., Gillard, R.D. and McCleverty, J.A., Eds. (1987) Comprehensive Coordination Chemistry. Oxford: Pergamon Press, vol. 17. Wharton, D.C. and Tzagaloff, A. (1964) Studies on the electron transfer system. LVII. The near infrared absorption band of cytochrome oxidase. J. Biol. Chem., 239, 20362041. Zubkova, S.M. (1978) About mechanism of biological action of He-Ne laser radiation. Nauchn. Dokl. Vyshei Shkoly (Moscow), N=17, 3037. Yu, W., Naim, J.O. and Lanzafame, R.J. (1994) The effect of laser irradiation on the release of bFGF from 3T3 broblasts. Photochem. Photobiol., 59, 167171. Yu, W. et al. (1996) Mechanism of low level laser biostimulatory effects. Lasers Surg. Med., Suppl., 8, p. 8 (abstr. 31). Yu, W., Naim, J.O. and Lanzafame, R.J. (1997) Effects of photostimulation on wound healing in diabetic mice. Lasers Surg. Medicine, 20, 5663.

8
Therapeutic and Diagnostic Application of Lasers in Ophthalmology
8.1 8.2 8.3 8.4 Introduction ....................................................................... 211 Basic Ocular Anatomy and Physiology............................. 212 Transmission and Absorptive Properties of Ocular Tissues................................................................ 214 Photothermal Laser Applications...................................... 217
Mechanisms of Photothermal LaserTissue interaction Photothermal Laser Applications Beam Delivery Systems for Photocoagulation

8.5 8.6

Photodisruptive Laser Applications .................................. 231

Photodisruptive LaserTissue Interaction Photodisruptor Applications Photodisruptor Delivery Systems

Photochemical Laser Applications: Photoablation and Photodynamic Therapy .............................................. 233

Photoablation Mechanism Delivery Systems for Photoablative Lasers Photodynamic Laser Applications

8.7

Diagnostic Laser Applications ........................................... 239

Visual Acuity Measurement Laser Doppler Velocimeter Scanning Laser Ophthalmoscope Spectroscopic Diagnosis of Ocular Diseases

Keith P. Thompson, M.D. Qiushi S. Ren, Ph.D. Jean-Marie Parel, Ing., ETS-G

8.8 Summary............................................................................. 240 Acknowledgments......................................................................... 241 References ...................................................................................... 241

8.1 Introduction
The most widespread medical application for laser technology in medicine has been in ophthalmology. Since the introduction of the ruby laser more than three decades ago, ophthalmic laser applications have experienced rapid growth with use of the argon, krypton, argon pumped dye, Nd:YAG, and, most recently, diode lasers. Lasers have achieved remarkable clinical successes in the treatment of retinal diseases, glaucoma, and lens capsule opacication following cataract surgery. The 1990s saw another period of rapid growth of ophthalmic laser technology with the then newly devised excimer and midinfrared lasers. The reasons for the prevalence of laser applications in ophthalmology are easy to grasp. The eye is one of the most accessible human organs, and its media (cornea, aqueous humor, lens, and vitreous) are

0-8493-1146-2/02/$0.00+$1.50 2002 by CRC Press LLC

211

212

Lasers in Medicine

transparent to visible light, allowing direct inspection of its internal structures for diagnosis and treatment. Many intraocular tissues contain pigments (such as melanin) that allow absorption of laser energy for photothermal lasertissue interactions. Transparent intraocular structures (such as the posterior lens capsule and vitreous), have also been amenable to laser treatment with Nd:YAG photodisruptors. Another factor that has facilitated the growth of ophthalmic laser applications is the ability of ophthalmologists to evaluate objectively the efcacy of new treatment modalities such as laser therapy. Whereas efcacy in other medical specialties is more difcult to assess (evaluating new treatments for patients with cancer or cardiovascular disease requires many patients and years of follow-up), ophthalmologists can observe disease processes directly and functional parameters such as visual acuity, visual eld, and intraocular pressure can be measured accurately and objectively. These factors have facilitated the introduction of new laser treatments for eye diseases and the rapid determination of their efcacy. This chapter will review the current status of clinical and experimental applications of laser technology in ophthalmology. Because a basic understanding of ocular anatomy, physiology, and some disease processes is necessary to understand the rationale for laser applications, a brief overview of these subjects for the non-physician readership is provided. Also discussed are therapeutic applications of lasers according to photothermal, photodisruptive, and photochemical mechanisms of lasertissue interaction. Current diagnostic applications of lasers in ophthalmology are also reviewed.

8.2 Basic Ocular Anatomy and Physiology

The eye is a slightly ovoid organ measuring about 24 mm in length and 23mm in diameter (Figure 8.1).1 It is composed of three distinct tunics or layers. The sclera is the tough outer coat of the eye that makes up the posterior four fths of the globe. The sclera is white, because its dense collagen brils have variable diameters that cause diffuse scattering of visible light. Anteriorly, the sclera is contiguous with the cornea, the clear watchglass-like covering where the collagen brils have more uniform diameters and are arranged in orderly layers, permitting transparency (T 95% for 400-900 nm; refractive index, n 1.3765

Figure 8.1

Anatomical drawing of the eye.

Therapeutic and Diagnostic Application of Lasers in Ophthalmology

213

0.0005, Tensile strength, T 106 Nm-2). The middle layer of the eye, composed of a rich network of blood vessels that supplies nutrition to the eye is termed the uveal tract. In the posterior segment of the eye, the uvea is termed the choroid. In age-related macular degeneration, the boundary between the retina and choroid may break down and allow abnormal blood vessels to grow beneath the retina, causing hemorrhage and loss of vision.2 Anteriorly, the uveal tract is continuous with the iris, the colored aperture of the eye, which allows light to pass through its central opening (pupil). The diameter of the pupil varies between 1.5 mm to 8 mm, depending upon the intensity of light reaching the retina. The eye is divided into three cavities by intraocular structures. The anterior chamber lies between the cornea and iris. It is lled with a clear water-like uid, the aqueous humor (viscosity 1.0 cs; n 1.3335). Behind the iris and anterior to the lens lies the posterior chamber. The lens of the eye, composed of specialized crystalline proteins, is clear at birth (n 1.40-1.42, T 104 Nm-2) and often becomes progressively cloudy or cataractous with age. The lens is approximately 0.15 to 0.25 cm3 in volume, 3.5 mm in thickness, 8 mm in diameter and enveloped in a tough, thin (1014 m), transparent, collagenous capsule (n = 1.396, T 107 Nm-2). The posterior portion of this capsule is left intact following removal of the lens substance during modern cataract surgery to help support a plastic intraocular lens. The large cavity posterior to the lens is lled with a transparent jelly-like substance, the vitreous humor (n 1.335, viscosity 10-100 cs). The vitreous body (5.5 cm3) is often surgically removed for the treatment of diabetic eye diseases and the cavity then lls with aqueous humor, with no apparent adverse effects upon the eye. The innermost layer of the eye is the neurosensory retina, an outgrowth of brain tissue that converts visible photon energy into electrochemical impulses that are processed (in both a digital and analog fashion) in the inner retinal layers and relayed to the brain via the optic nerve. The outer portion of the retina contains the photoreceptors and receives its nutrition from the uveal tract, while the inner retina contains its own retinal vascular supply. Normally, the retinal arterioles are sealed tightly, and allow no leakage of blood uids or proteins into the surrounding retina. In diabetes, and several other vascular diseases, the blood vessels become abnormal and allow uid to leak into the macula, the posterior region of the retina that is responsible for central vision, causing retinal dysfunction and loss of vision. Fragile new blood vessels may also grow onto the surface of the retina, causing both leakage of uid and hemorrhage into the eye.3 The eye is internally pressurized with a normal intraocular pressure of about 13 3 mm Hg above ambient pressure. Internal pressure is created by a clear uid termed aqueous humor, produced by the ciliary body, a specialized secretory tissue that is contiguous with the retina posteriorly. The aqueous humor ows anteriorly though the pupil to drain through the trabecular meshwork, a specialized ltration structure located in the angle between the cornea and iris. Abnormalities of the pressure regulating system of the eye may result in abnormally high ( 22 mm Hg) intraocular pressure, a condition termed glaucoma, which damages the delicate nerve bers in the optic nerve, causing a loss of vision. The eye has a focal length of about 16.7 mm in air.4 Ophthalmologists prefer to use optical units of diopters (l diopter = l/F meters) when measuring the refractive error of the eye or the power of the eyes optical elements. These are usually determined according to Gulstrands reduced schematic eye (Figure 8.2a).4,5 For instrument design and computational purposes, opticists use the model described in Figure 8.2b.6 Two thirds of the eyes optical power, or about 44 diopters, results from the curvature of the cornea; the remainder is provided by the crystalline lens. The lens can change shape and increase its focusing power for near vision (accommodation) until about the fth decade of life, when, for most people, reading glasses become a necessity for clear viewing of close objects. Because the cornea is the predominant optical element of the eye, a small change in its curvature causes a large effect on the total optical power of the eye, a principle utilized by corneal refractive surgeons.7,8 An emmetropic eye, or one without refractive error, focuses a distant point source of light on the fovea, the region of the macula with the highest density of cones (1 m diameter with 1.3 m spacing), the photoreceptors that provide the highest resolution (1530 min. of arc). Nearsighted or myopic eyes have too much optical power for their length, resulting in displacement of the focal point anterior to the

214

Lasers in Medicine

n=1.33 GULLSTRAND REDUCED SCHEMATIC EYE F N 16.67mm 5.65mm 16.67mm 22.6mm IMAGE PLAN

Figure 8.2

(a) Optical model of the human eye. (b) Littman-Gullstrand schematic eye with accommodation.

retina. Farsighted or hyperopic eyes have insufcient power for their length, with a focal plane that is located behind the retina.

8.3 Transmission and Absorptive Properties of Ocular Tissues

Light entering the eye can be reected, scattered, transmitted, or absorbed. Reected or scattered light contains information that can be used for noninvasive diagnostic purposes, discussed in Section 8.7. In ophthalmic laser surgery, the surgeon must choose a proper laser wavelength that is transmitted through intervening ocular layers to become incident upon the target tissue. In the visible and near infrared spectrum (400 nm to 1,400 nm), the absorption characteristics of ocular tissues are determined by a group of chromophores within the tissue. Ocular chromophores that absorb light in the visible spectrum include melanin, located in the retinal pigment epithelium, iris pigment epithelium, uvea, and trabecular meshwork; hemoglobin, located in red blood cells within blood vessels; and xanthophyll, located in the inner and outer plexiform layers of the retina

Therapeutic and Diagnostic Application of Lasers in Ophthalmology

215

10,000 514.5 nm 1 2 Extinction coefficient (cm-1) 1000 3 647.1 nm 577 nm

n n go go Dye Ar Ar
488.0 nm

yp Kr

ton

ND:YAG 1064 nm

100 4

10

1 400 500 600 700 800 Wavelength (nm) 900 1000 1100

Figure 8.3 Transmission spectra of the major ocular chromophores: melanin (1), reduced hemoglobin (2), oxygenated hemoglobin (3), and macular xanthophyll (4). Also shown are the wavelengths of four commonly utilized ophthalmic lasers: argon, krypton, dye and Nd:YAG. (Permission pending.)

in the macular region. The light absorption spectra of these pigments is illustrated in Figure 8.3. In the mid-infrared spectrum (1.8 m10.6 m), the major molecule absorbing incident photons is water (Figure 8.4). The retina contains at least six pigments: melanin, hemoglobin, macular xanthophyll, rhodopsin, cone photopigments, and lipofuscin. The rst three of these are of major importance in laser photocoagulation of the retina. The cornea consists of ve principal layers, the epithelium, Bowmans layer, stroma, Descemets membrane and the endothelium (Figure 8.5). The cornea is protected continuously by a tear lm that has a high lipid content and a high index of refraction (n = 1.40). Due to transmission characteristics, the cornea provides an effective window for vision, photocoagulation, photodisruption, imaging and examination of intraocular structures (Figure 8.6). The transmission and absorption characteristics of the crystalline lens change with age (Figure 8.7). The total transmittance at the various anterior surfaces is given in Figure 8.8.
104

absorption coefficient (cm-1)

H2O 103

102

10

1000

2000

3000 4000 5000

7000 10000

(nm)

wavelength
Figure 8.4 Absorption spectra of water.

216

Lasers in Medicine

Figure 8.5 nea.

Detailed structure of the cor-

100

80 1 PERCENT TRANSMITTANCE 60 2

40 CORNEA TRANSMITTANCE TOTAL 1 DIRECT, 4 1/2 YRS. 2 DIRECT, 53 YRS.

20

300

Figure 8.6

Corneal transmittance spectrum.

400 500 600 800 1000 1200 WAVELENGTH MILLIMICRONS

1600

2000

100

80

1 2

TRANSMITTANCE

60

LENS TRANSMITTANCE TOTAL , 4 1/2 YRS. DIRECT, 4 1/2 YRS. 1 DIRECT, 53 YRS. 2 DIRECT, 75 YRS. 3 3

40

PERCENT

20

Figure 8.7 Lens transmittance spectrum (aging effect).

300

400

500 600 800 1000 1200 WAVELENGTH MILLIMICRONS

1600

2000

Therapeutic and Diagnostic Application of Lasers in Ophthalmology

217

100 1 80 2 3 PERFECT TRANSMITTANCE 60 4

40 TOTAL TRANSMITTANCE AT THE VARIOUS ANTERIOR SURFACES 1 AQUEOUS 3 VITREOUS 2 LENS 4 RETINA

20

300

400

500 600 800 1000 1200 WAVELENGTH MILLIMICRONS

1600 2000

Figure 8.8

Total transmittance of the entire eye.

8.4 Photothermal Laser Applications

8.4.1 Mechanisms of Photothermal LaserTissue interaction
Photothermal lasertissue interactions occur when laser energy is absorbed by the target tissue and converted into heat. The effect on the tissue depends on both the magnitude and rate of temperature elevation. Low (less than 10C) temperature elevations that occur over many seconds to minutes (photoheating) usually cause cell damage or death without causing structural alterations to the tissue. Greater temperature elevations (2030C), occurring over shorter time intervals (approximately 1 sec.), cause thermal coagulation of tissue (photocoagulation), with cellular death and irreversible structural damage to the tissue due to denaturation (loss of the three-dimensional molecular structure) of tissue proteins.9 Laser energy depositions that rapidly (much less than 1 sec.) heat the tissue above its boiling point cause thermally mediated tissue removal by explosive vaporization (photovaporization). The wavelength of laser radiation, the absorption coefcient of the tissue, the power (or energy) density, and the duration of laser radiation determine which of these photothermal effects predominates. To minimize unwanted thermal damage to surrounding ocular tissue during laser treatment, a wavelength should be selected that is preferentially absorbed by the target tissue, and the laser exposure duration should be shorter than the thermal relaxation time of the tissue, given by:

T h = d2 / 4k
where d represents the absorption depth of the radiation, and k is the thermal diffusivity of the tissue. For example, mid-infrared (2.94 :m) Er:YAG and Er:YSGG lasers are under investigation for cutting the cornea. Water (k = 1.5 x 10-3 cm2/sec) is the primary absorbing molecule at this wavelength (d = 1 m) and the thermal relaxation time is calculated to be 1.7 s. Therefore, to achieve efcient photovaporization with minimum damage to the surrounding corneal tissue, the pulse duration of the laser must be signicantly less than 1.7 s. Decreased thermal tissue damage with shorter laser pulse duration has been demonstrated by comparing the tissue damage adjacent to corneal excisions created by a Q-switched (pulse duration = 100 ns) to a free running (pulse duration 250 s) Er:YSGG laser (Figures 8.9a and b).

218

Lasers in Medicine

Figure 8.9 (a) Er:YSCG normal mode (250 s pulses). (b) Er:YSCG Q-switched mode (100 ns pulses). Light microscopy of corneal excisions produced by a mid infrared Er:YSCG laser demonstrates the effect of pulse duration on adjacent thermal damage to tissue. Excisions produced in the normal mode (250 s) causes signicantly more thermal damage than when operated in the Q-switched mode (100 ns).

Therapeutic and Diagnostic Application of Lasers in Ophthalmology

219

8.4.2 Photothermal Laser Applications

Tables 8.1a and 8.1b summarize the present clinical and research applications for photothermal lasers in ophthalmology, respectively.
Table 8.1a Photothermal Laser Applications in Clinical Use
Disease Proliferative retinopathies Diabetic macular edema, vein occlusions Age-related macular degeneration Laser/Wavelength (nm) Argon-488/514 Krypton 647 Dye-577 Argon-488/514 Krypton 647 Dye-577 Argon Green-514 Dye-577 Rationale Destruction of peripheral retina decreases stimulus for new blood vessel growth Focal treatment decreases leakage of uid in retina Destruction of new vessel membranes prevents complication from subretinal hemorrage and leaking vessels Focal photocoagulation of trabecular meshwork improves aqueous outow, decreasing pressure Hole in iris allows alternative path for aqueous ow, preventing angle closure Destruction of ciliary body decreases intraocular pressure Comment Widespread application Widespread application Widespread application

Procedure Pan-retinal photocoagulation Focal macular photocoagulation Photocoagulation of subretinal neovascular membranes Traeculoplasty

Open angle glaucoma

Argon-488/514

Peripheral iridotomy Cyclophotocoagulation

Narrow angle glaucoma Advanced glaucoma

Argon-488/514

Compared favorably to medications in recent study Has replaced need for intraocular surgery

Nd:YAG-1064 free running thermal mode

Table 8.1b

Photothermal Laser Applications in Experimental Development

Disease Glaucoma Laser/Wavelength (nm) TmHoCr:YAG2.1m HF 2.9 m Er:yAG-2.9 m Ho:YAG Line-tuned HF 2.55 m Rationale Small incision, minimal trauma may increase survival of ltration bleb Nonmechanical cutting may decrease postoperative astigmatism Shrinkage of collagen steepens central cornea Fusion of collagen and connective tissue Comment

Procedure Filtration surgery

Corneal trephination Photothermal keratoplasty Laser tissue welding

Corneal scarring and disease Hyperopia Wound closure

Cataract removal

Cataract

HF and other midIR 2.53.0 m

Cataract removal while maintaining capsular bag

Full thickness trephination possible in 7 s Early clinical studies Unsuccessful in clinical applications to date Will facilitate development of accommodating lens

Laser-induced hyperthermia

Ocular tumors

Mid-infrared

Selective heating causes tumor necrosis

8.4.2.1 Photocoagulation Therapies for Retinal Diseases The use of lasers as a selective heat source to coagulate retinal tissue is one of the most important and widely adopted ophthalmic laser applications. The 694 nm ruby laser was the rst laser introduced to ophthalmology in the early 1960s.10 Energy from the ruby laser was absorbed by the retinal pigment epithelium and caused photocoagulation of the outer retina. Unfortunately, this early laser performed

220

Lasers in Medicine

poorly in the clinical setting due to its short pulse duration ( 100300 sec), its deep penetration depth ( 400m), variable power output, and crude beam delivery system. In 1968, the continuous wave argon laser was introduced by LEsperance.12 This laser emitted at both 488 and 514.5 nm and offered improved beam stability. It quickly surpassed the xenon arc lamp and ruby laser as the preferred methods for retinal photocoagulation in the 1970s. The irradiance of the Ar+ laser treatment is approximately 100 W/cm2 with an application time of 0.11 second. Photocoagulation has proven most useful in the treatment of proliferative diabetic retinopathy. In this disorder, the retina becomes ischemic (starved for oxygen and nutrients) and releases chemical messengers that are thought to cause the growth of fragile new blood vessels in an attempt to nourish the oxygen-starved tissue. The abnormal new blood vessels, and their associated proliferating brous tissue, are a major cause of sight-threatening complications in diabetic eye disease. By destroying a portion of the peripheral retina with the laser, retinal metabolic demands are decreased and the stimulus for new vessel formation is reduced.3 This treatment, termed panretinal photocoagulation (Figure 8.10) can signicantly decrease the risk for vision-threatening complications from new blood vessel growth.1214 The side effects of panretinal photocoagulation loss of some night vision and constriction of visual eld are far outweighed by the preservation of central vision offered by treatment. The continuous wave Kr+ laser, introduced in 1972, was also used to photocoagulate the retina effectively, with the additional advantage that the 568 nm wavelength can penetrate ocular media clouded by cataract or vitreous hemorrhage more efciently than the 488 nm argon laser.15 In addition, because 568 nm is not absorbed by the xanthophyll pigment located in the macula, the krypton laser is best suited to treat microaneurysms located close to the fovea. The widespread use of the argon and krypton lasers by ophthalmologists in the treatment of proliferative diabetic retinopathy has been directly responsible for preserving the vision of many thousands of diabetic patients.

Figure 8.10 Pan retinal photocoagulation. Multiple 500 m diameter laser burns have been applied to the peripheral retina of this diabetic patient using a 488/514 nm argon laser. Several areas of focal treatment in the macula are also present.

Laser treatment of the macula was also found to be effective for decreasing the leakage of uid into the retina (diabetic macular edema), which also can occur in diabetic eye disease.16 Laser treatment within the macular region requires special consideration for the absorptive properties of the tissue. Within the macula, the inner retinal layers contain xanthophyll pigment, which absorbs strongly between 450 and 500 nm. Use of the argon blue wavelength (488 nm) causes heating and destruction of the nerve ber layer in the inner retina, resulting in loss of vision (Figure 8.11). By using the argon green wavelength (514 nm), or a longer wavelength from a krypton or dye laser, absorption by xanthophyll and damage to the inner retina are minimized.

Therapeutic and Diagnostic Application of Lasers in Ophthalmology

221

Figure 8.11 Histology of macula following photocoagulation with the argon blue (488 nm) wavelength. Note the damage to the inner retinal layers due to absorption of laser energy by xanthophyll pigment. Inner retinal damage can be avoided by treatment with the argon green (514 nm) or krypton (647 nm) wavelengths.

Treatment of subretinal neovascular membranes that occur in some forms of age-related macular degeneration is another major application for laser photocoagulators. Macular degeneration is the leading cause of acquired blindness in the elderly. Here, the surgeon uses a photocoagulator to destroy the invading vascular membrane, leaving a chorioretinal scar (Figure 8.12). The benet of photocoagulation to preserve vision in this disorder has also been demonstrated by a large clinical study.1721 Tunable dye lasers, introduced to ophthalmology in 1981, gave added exibility to ophthalmologists and allowed them to select a wavelength for a given photothermal laser application.22 Ideally, photon penetration depth for a laser wavelength used in vessel closure should be approximately the same length as the vessels diameter, such that effective bulk heating of the blood column occurs without supercial damage and perforation of the vessel wall at the radiation site. Treatment of neovascular membranes is best done with a wavelength maximally absorbed by hemoglobin. 8.4.2.2 Photo-Thermal Laser Treatment for Glaucoma In addition to treatment of retinal diseases, photothermal lasers have also found widespread clinical application in the treatment of glaucoma. Continuous wave argon, krypton, or dye lasers are used to burn a small hole through the iris when narrow angle or angle closure glaucoma is present. Creating a small opening in the peripheral iris allows an alternate pathway for aqueous humor when its normal pathway through the pupil is impeded by the lens (angle closure glaucoma). Melanin is the major absorptive pigment in the iris, and the argon 488 nm wavelength is absorbed quite effectively (Figure 8.3). In individuals with lightly colored irises, little pigment is present, decreasing the effectiveness of the laser. In these cases, the use of a Q-switched Nd:YAG photodisruptor, discussed below, offers the clinician an alternative laser that is independent of pigment absorption for its effect.23 Creating a hole in the iris with the laser is easily performed with the patient seated at the slitlamp (Figure 8.13) in the outpatient clinic. This treatment has saved many patients and surgeons a trip to the operating room and an intraocular surgical procedure with its associated risks, complications, and costs. Laser trabeculoplasty is performed in eyes with open-angle-type glaucoma by applying focal photocoagulation to the trabecular meshwork located in the angle of the eye.24 For this treatment, the 13 W

222

Lasers in Medicine

Figure 8.12 (a) Subretinal neovascular membranes (arrows) in the macula in a patient with age related macular degeneration. (b) Same eye after photocoagulation treatment with the argon green laser.

argon (488 nm and 514 nm) laser is aimed through a mirrored contact lens and focused into a 50100 m spot. Application times of 0.3 to 1 sec. are used. Although the mechanism of action is uncertain, it is thought that laser photocoagulation may stretch open the structures of the trabecular meshwork, facilitating outow of aqueous humor and decreasing intraocular pressure.25 Laser trabeculoplasty may play a major role in the management of patients with open-angle glaucoma, estimated to be 1.6 million in the United States alone.26 When medical management fails, glaucoma surgery is indicated to control the intraocular pressure. The objective of glaucoma surgery is to provide an alternate pathway for uid to egress from the eye. The trend in techniques has been toward smaller incisions and less invasive procedures. Recently, a pulsed TmHoCr:YAG laser (Figure 8.14) has been applied subconjunctivally through a quartz ber optic probe for a transcleral ltering procedure.27 In this procedure, a small snip incision is made in the conjunctiva, approximately 5 mm posterior to the junction of the cornea and sclera (limbus) and the ber optic laser probe is passed through the opening and directed over the pupil until the tip approaches the opposite angle. The beam is shaped and focused by a cylindrical prism built at the end of the probe and directed toward the limbus to create a full-thickness stula. The exposure is made with energy ranging from 60150 mJ per pulse, with pulse duration of 300 ms and a repetition rate of 5 Hz. Total energy levels required to produce a full-thickness channel ranged from 1.35 to 6.6 J. Compared with conventional surgical procedures, this technique may avoid the need for large conjunctival aps and intracorneal

Therapeutic and Diagnostic Application of Lasers in Ophthalmology

223

cw : Ar+, Kr+, Nd:YAG, Ar+ pumped Dye G630 AIGaAs DIODE and other LASERS

CCD 35mmSLR

SURGEON PATIENT BIOMICROSCOPE FILTERS Tu-HALOGEN SLIT-ILLUMINATOR

SLIT APERTURE

Figure 8.13 (a) Typical appearance of a slitlamp laser delivery system. A Keeler diode laser is integrated into the slitlamp optical system. (b) Optical schematic of slitlamp laser delivery system in widespread clinical use. Laser energy is conducted to the slitlamp by a beroptic cable and directed toward the eye by a coaxial beamsplitter. (continued)

224

Lasers in Medicine

CCD MACRO ZOOM OPTICS

35mm SLR

SURGEON

Tu-HALOGEN 50m FIBER ILLUMINATOR 0.1-0.2 NA STERILIZABLE PLUG-IN PREFOCUSSED CONNECTOR 400m FIBER 0.2 NA LASER ENDOPROBE M2

SAFETY SHUTTER STEREO ZOOM OPERATION MICROSCOPE DM JOY-STICK

cw: Ar+ HeNe Kr+ AIMING Nd:YAG BEAM Ar+DyeR630 Al Ga As & others POWER LASERS

EYE

FOOT-PEDAL

Figure 8.13 (continued) (c) Typical hand-held operating-room-delivery system.

manipulations, reduce tissue trauma and hemorrhage, and limit the outow of wound-healing agents that lead to scarring and failure of the stula.28 Clinical investigation of this new technique is pending. Eyes with advanced forms of glaucoma unresponsive to conventional surgical or medical management may undergo photocoagulation of the ciliary body (cyclophotocoagulation) with a Nd:YAG laser operating at 1064 nm in the free running (520 ms) mode. Both pulsed (10 msec) and continuous wave Nd:YAG lasers are used to heat and destroy portions of the aqueous humor-secreting ciliary body in an attempt to decrease intraocular pressure. Diode lasers have also shown promise for cyclophotocoagulation.29 8.4.2.3 Mid-Infrared Laser Corneal Surgery Although the healthy cornea is transparent to visible light and does not interact with argon, krypton or dye photothermal lasers, wavelengths in the mid-infrared and infrared spectrum (1.9 m10.6 m) are strongly absorbed, largely due to the absorption of water, which composes 75% of corneal tissue (Figure 8.15). Within this spectral region, laser parameters such as wavelength, pulse duration, energy density, etc. can be chosen to cut, shrink, and weld tissue by photothermal mechanisms. Although midinfrared corneal laser surgery is still in its infancy, with most techniques still in laboratory or early human studies, there are many promising potential clinical applications. 8.4.2.3.1 Mid-IR Laser Corneal Cutters (Photocutting) Corneal and refractive surgery demands a high degree of spatial accuracy and reproducibility. Existing surgical techniques using metal and diamond scalpels rely heavily on the surgeons manual abilities and experience. Even with the help of high power operating microscopes, manual precision is limited to 100200 m, contributing to the unpredictability of postoperative refractive outcome. Postoperative astigmatism following corneal transplant surgery continues to be a major problem.30 A mechanical trephine (the circular blade used by the surgeon to cut corneal tissue for transplantation) produces a torsional moment on the cornea during cutting, leading to an uncertainty in the precise dimensions of the trephined tissue. A tissue size disparity between the donor and host of 100 m produces about 4 diopters of astigmatism. The application of photothermal cutting lasers to corneal transplant surgery provides the potential for a noncontact corneal trephination system. Beckman et al.31 demonstrated that the cornea could be excised with a rapidly pulsed (60300 Hz) CO2 laser operating at 10.6 m, emitting a peak power output of 25,000 watts and an average output power of 150 watts, but the amount of adjacent thermal damage to the corneal tissue was too excessive for clinical use.31 More recently, Parel, Jeffers et al. selected the

Therapeutic and Diagnostic Application of Lasers in Ophthalmology

225

Figure 8.14

(a) Sunrise Technologies laser system for glaucoma and (b) its special beroptical tip design.

mid-infrared region to investigate laser corneal surgery, and, with a pulsed HF laser (2.73.0 m, 50200 nsec, 2.4 J/cm2), created 70 m-wide uniform corneal excisions with much less thermal damage than that caused by the CO2 laser.37 Therefore, we postulate that the laser might signicantly decrease the amount of undesirable postoperative astigmatism,34 and we have begun comparative studies in animal models. For clinical application, solid state lasers generating mid-infrared wavelengths are desirable because of their simplicity, reliability and safety in the operating room. A Q-switched Er:YAG solid state laser system (2.94 m, 100 ns, 1.5 J/cm2) coupled to an axiconlens combination was used successfully in preliminary corneal trephination studies,35 and research to develop this technology is continuing.

226

Lasers in Medicine

LIGHT ABSORPTION COEFFICIENT vs. WAVELENGTH

105 Absorption Coefficient (1/cm) 104 Co:MgF2 tuning range (a) (b) Er:YAG Er:YSGG 102 pulsed HF range Fully Hydrated Cornea (+ -20%) Dehydrated Cornea (+ -20%) Liquid Water (+ -10%) Human Cornea (+ -30%) HYDRATED, DEHYDRATED HUMAN CORNEA, and WATER

COAG

LPTK

CUTTING

103

101 Nd:YAG 100 1.0 1.5 2.0

THC:YAG

2.5

3.0

3.5

4.0

Wavelength (um)

Figure 8.15 Light absorption coefcient vs. wavelength for the hydrated, dehydrated human cornea and water. Absorption rises sharply for wavelengths longer than 2.5 um with a peak absorption coefcient occurring near 2.9 um. Because the normal cornea is nearly 75% water by weight, the corneas IR absorption spectra closely parallels that of H2O.

Figure 8.16 Circular corneal button excised with pulsed HF laser (2.9 um, 50-200 nsec., 2.4 J/cm2). Edges of the excised button are smooth and uniform. 360 full thickness trephination is not possible because aqueous uid egressing from the rst penetration site lls the excision and attenuates energy.

8.4.2.3.2 Laser Photo-Thermal Keratoplasty (LPTK or Photoshrinking) Because collagen, the major structural component of the cornea, shrinks when heated to 5558 C9 the shape of the cornea and consequently, the refractive power of the eye, may be changed by creating focal burns in the stroma. Excessive mechanical and thermal damage to the cornea rendered previous attempts to thermally alter the shape of the cornea by heated metal probes (100600 C) unsuccessful (Figure 8.17a). Seiler et al.3638 recently demonstrated that the refractive power of the cornea can be adjusted by shrinkage of corneal collagen bers with lasers emitting around 2 m.39 A small series of hyperopic patients treated with the holmium laser coupled to a ber optic probe delivery system have shown reasonable predictability and fair stability of the refractive outcome. By heating the corneal stroma locally with a non-contact mid-infrared TmHoCr:YAG laser system operating at 2.1 m in the free-running mode ( 300 sec, 1 Hz), fresh cadaver eyes and the eyes of rabbits, cats, and non-human primates have been studied topographically and histologically (Figure 8.17b). The energy density applied to the target varied from 5 to 20 J/cm2 and the total number of pulses from 5 to 50. Changes of 10 diopters were obtained with ve 10 J/cm2 pulses placed in a semi-

Therapeutic and Diagnostic Application of Lasers in Ophthalmology

227

Figure 8.17 (a) Corneal burns induced by a heated copper lament. Note ring of coagulation damage extending 0.5 mm from center of in injury. Topographic changes are evident by stress lines in the cornea seen by retroillumination in the pupil. (b) Appearance of cornea immediately following focal stromal heating with a non-contact 300 usec, 2.1 um TmHoCr:YAG laser. Large dioptric changes are possible with gross coagulation injury.

annular pattern designed to correct astigmatism (Figure 8.18). With non-contact treatment, the treatment zone was found to be limited to an inverted cone having an epithelium base of 300 m in diameter and a stromal depth of 250 m; the endothelium was spared and the damage to the adjacent stromal and epithelium tissue was limited to 50 m of the treatment zone. 8.4.2.3.3 Laser Tissue Welding (Photowelding) Laser tissue welding involves the use of laser energy to fuse tissue directly or activate biological agents to bond tissue. It may allow sutureless wound closure and reduction of surgical time, and nd application in corneal, cataract, glaucoma, and vitreoretinal surgery. Laser welding has been applied successfully in other surgical elds, including vascular, dermatology, urology, otolaryngology, and neurology.4042 The pathophysiologic mechanisms of laser tissue welding are not yet fully understood; however, it is postulated that tissue fusion is due to the denaturation and homogenization of collagen with interdigitation of individual brils that occur when tissue is heated by absorption of laser radiation.43 Welding synthetic collagen lenticules to the cornea with and without the use of solders has been attempted using a milliwatt CO2 laser.41 Direct welding, either lenticule to lenticule, lenticule to cornea, or cornea to cornea was not successful without the use of bio-solders. Because the mechanism for laser tissue welding is probably heat dependent, the short penetration (10 m) of CO2 laser radiation tends to overheat and char tissue, producing weak welds. More recently, unique matches

228

Lasers in Medicine

Figure 8.18 (a) Histology of laser photothermal keratoplasty burn in rabbit cornea. The injury is conical in shape and spares the endothelium. (b) Color-coded map of topographic changes (in diopters) induced in human cadaver eye treated by LPTK with a TmHoCr:YAG laser operating g at 2.1 m (300 s, 1Hz).

between laser and tissue characteristics have been identied that may lead to improvements in tissue welding. 8.4.2.4 Cataract Surgery with Endo-Laser System The future goal of cataract surgery is to restore accommodation.44,45 This can be accomplished by removal of the cataractous lens through a small opening while preserving the lens capsule and its zonular attachments. The empty lens capsule is then lled with a bio-compatible and optically suitable synthetic

Therapeutic and Diagnostic Application of Lasers in Ophthalmology

229

gel, creating a physiologically compatible articial intraocular lens in situ (Figure 8.19). This procedure will demand a more precise and less damaging method for cataract removal, which is currently performed by ultrasonic probes (phacoemulsiers). Endo laser systems for removing the cataractous lens are the most promising candidates. By directing laser energy into a exible ber, cataract removal could be performed endoscopically while minimizing iatrogenic trauma to healthy tissue. Bath demonstrated the successful use of the 308 nm XeCl excimer laser for cataract removal.46,47 Despite several advantages, such as a rapid tissue removal rate (25 m/pulse at 6.5 J/cm2), limited thermal damage to adjacent tissue (50 m) and the availability of inexpensive silica bers with good transmission properties, the uorescence induced by 308 nm laser pulses may cause signicant retinal damage.48 Potential carcinogenesis and cataractogenesis of UV light (308 nm) to the operator and the patient also raises serious concern about the use of this wavelength. Mid-infrared (2.5 m3.5 m) lasers offer an alternative possibility for laser cataract removal. In this spectral region, the optical penetration depth can be controlled to less than 1 m by choosing an IR wavelength that corresponds with the absorption peak of lens tissue (2.93.0 m). The laser energy will then be conned to a very small volume, preventing photons from reaching other intraocular structures such as the retina or the corneal endothelium. The short penetration depth of mid-IR laser radiation also minimizes thermal damage to adjacent tissue, providing the pulse duration is shorter than the thermal relaxation time of the lens tissue. In addition, laser excitation in the mid-IR region precludes electronic absorption, eliminating undesirable uorescence. Transmitting mid-infrared (2.8 m3.1 m) short laser pulses ( 500 ns) through an optical ber is a challenging technical problem.49 It may require the use of exotic materials such as zirconium uoride (ZnF4) and specially designed probes (Figures 8.20 and 8.21). Efforts to rene this approach and to integrate a micro irrigationaspiration system that will be practical for clinical use are under way by a number of investigators.

Figure 8.19 Phaco-Ersatz. The cataractous lens is removed through a small incision with an integrated endolaserirrigation-aspiration system. The capsular bag is retained, allowing injection of a suitable polymer that has optical and mechanical properties similar to those of the natural lens. Such a procedure may allow restoration of accommodation, the ability of the lens to increase its optical power and focus on near objects.

8.4.2.5 Laser Induced Hyperthermia for Ocular Tumors Ocular tumors include uveal melanoma in adults and retinoblastoma in children. Conventional treatments for these tumors include enucleation, ionizing radiation, external beam or brachytherapy, photocoagulation and cryotherapy. Tumor-selective hyperthermia (41 C to 45 C), and nonselective hyperthermia, (> 45 C), are techniques that show potential synergistic effects with other forms of tumor treatment such as ionizing radiation, chemotherapy and photodynamic therapy.50 Present methods of producing hyperthermia in ocular tumors include the use of microwaves, radio frequency waves, and ultrasonic energy sources. The localization of the heating by electromagnetic radiation,

230

Lasers in Medicine

Figure 8.20 Photovaporization of a dense human lens nucleus in vitro with a pulsed HF (350 nm, 2.9 um, 10 Hz) laser conducted through a shielded ZrF ber. Practical clinical use will require coupling the laser ber to an irrigationaspiration system to remove lens byproducts and prevent excessive heat build up.
Silica Sheath ZrF Fiber 480 m 560 m

Figure 8.21

Silica clad ZrF ber used for conducting 2.9 um Hf laser energy for experimental cataract removal.

either transcorneal or transcleral, is very difcult to control and can lead to signicant side effects such as cataract formation. Near infrared light (700 nm to 1200 nm) can also induce signicant tissue heating. Additionally, such wavelengths can penetrate ocular tissue effectively. Infrared laser wavelengths are scattered in most lightto medium-pigmented tissue, and absorbed by heavily pigmented tissues. Additionally, the cornea, lens and vitreous are relatively transparent to infrared wavelengths. These two factors make infrared laser wavelengths a potential source for producing localized heating in ocular tumors.51

8.4.3 Beam Delivery Systems for Photocoagulation

In ophthalmology, four types of optical delivery systems are currently used. Two are used in the outpatient setting: (1) A joy-stick-controlled delivery system mounted on the slit-lamp-biomicroscope for treatment of the patient in the sitting position (Figure 8.13a, b) and (2) an indirect ophthalmoscope delivery system worn on the surgeons head and used with the patient in a 4570 reclined position. Two other modalities are used in the operating room with the patient in the supine position: (3) a joy-stickcontrolled system mounted on the binocular operating microscope to treat readily visible and accessible tissues and, (4) a hand-held ber optic probe system used for endocular and endo-orbital surgical treatment (Figure 8.13c). All types of delivery systems are usually connected to medium power (5 to 8 W) constant wave argon krypton, dye, pumped dye, constant wave Nd:YAG, and (1 to 3W) AlGaAs diode lasers via a 500 m diameter monolament silica ber having a numerical aperture of 0.1 to 0.2. Articulated arm delivery systems are no longer used in ophthalmic photocoagulator designs. The optics for the rst three types of delivery systems are designed to produce a small focal spot on the target (range 501000 m). The endophotocoagulator probe uses a 300400 m core diameter ber that produces a 500 m2 mm spot, depending on the distance of the probe tip to the tissue. Recently, Rol et al. introduced a series of novel endoprobes equipped with micro-optics to further reduce the spot diameter and focalize the laser beam

Therapeutic and Diagnostic Application of Lasers in Ophthalmology

231

on target.52 This systems output power is in the range of 100 mW2W and the energy delivered to target is 10010 J. With most delivery systems that are available commercially, the laser power output (0100%) and pulse duration (0.1 to 1 sec, in 0.1 sec increments) are preset on the control panel by the surgeon who activates the laser using a foot-pedal switch. For safety, several interlocks are provided by the manufacturer in accordance with laser safety regulations. To prevent contamination and permit mobility of surgical equipment, single-phase (110 or 220 VAC) and air cooled laser operation is preferred in the clinic and almost mandatory in the aseptic operating room. As with many medical instruments, miniaturization of the laser and delivery systems are prerequisites for ophthalmological use. The use of subminiaturized endoscopes designed for intraocular photocoagulation of the ciliary body and anterior retina under visual control have been introduced recently.45 Thus far, this technique has been restricted to in vitro and animal studies. For other ophthalmic surgical procedures, subminiature beroptic endoscopes provide the surgeon with a good view and illumination and lead to more precise surgical procedures than the more conventional open-sky surgical techniques. At the present time, endolaser surgical procedures require different types of bers for tissue ablation, coagulation, shrinkage, welding, and sensitization of tissue, and illuminating and imaging of the surgical eld. In the future, these different functions may be integrated into a single beroptic delivery system, facilitating surgeon control. Diode lasers, introduced in 1987, are the newest photocoagulators.53 These lasers have the advantage of greatly decreased size, cost, and reduced maintenance. Presently, AlGaAs diode lasers emitting at 805 nm with output power 13 W have been used for the treatment of retinal vascular disease. Clinical advantages of the diode laser include minimal absorption and scattering in cataract or mild vitreous hemorrhage compared with the argon green wavelength and low absorption by intraretinal hemorrhage. Excellent laser penetration through macular edema and serous retinal thickening have also been demonstrated. Technical disadvantages of the diode laser include a limited power output and a more divergent beam cone angle than the argon and krypton lasers used presently. Because semiconductor laser technology has advanced rapidly in recent years, the future outlook for semiconductor lasers is very promising. Diode lasers and diode-pumped solid state lasers with multi-watt power output at wavelengths that are desirable for various clinical applications will be available to replace the more expensive and bulky argon, krypton, and dye lasers for ophthalmic applications. Q-switched and mode-locked diode lasers with sufcient energy density in the nanosecond to picosecond range may also replace the Nd:YAG laser as a photodisruptor for iridectomy, trabeculoplasty, and cyclophotocoagulation.

8.5 Photodisruptive Laser Applications

8.5.1 Photodisruptive LaserTissue Interaction
A disadvantage to laser applications utilizing photothermal mechanisms is that the laser wavelength and absorption properties of the tissue must be carefully matched. Tissues transparent to the incident wavelength will be unaffected. In the late 1970s, short pulsed 1064 nm Nd:YAG lasers ( 30 ps for mode-locked and 10 ns for Qswitched) were introduced to create optical breakdown and photodisruption of ocular tissues. The Nd:YAG photodisruptors are now in common clinical use. The laser beam is brought through a high numerical aperture beam expanderfocusing lens system (2040 cone angle) to a sharp focus ( 1050 m diameter) to achieve extremely high irradiances (1081011 W/cm2) such that ionization of molecules occurs at the target. The irradiated material disintegrates into plasma, a collection of ions and electrons, that reaches peak temperatures of 104 C. Shock waves emanating from the site of plasma formation cause mechanical disruption of structures adjacent to the site of optical breakdown. Because optical breakdown is a threshold phenomenon that depends on the strength of the electrical eld, photodisruption is not dependent on absorption of the wavelength by the target tissue, and photodisruptors can be used in tissues that are transparent to the incident laser wavelength. For optical breakdown to occur, it

232

Lasers in Medicine

is only necessary for the photodisrupting laser delivery system to achieve the required high irradiance within a small volume of tissue.

8.5.2 Photodisruptor Applications

Photodisruptor laser applications are useful for cutting, incising, or vaporizing intraocular tissues. Table 8.2 summarizes the clinical and research applications for photodisruptors in ophthalmology.
Table 8.2 Photodisruptor Laser Applications
Disease Capsular opacication after cataract surgery Laser/Wavelength Nd:YAG-106-1 Rationale Photodisruption allows small opening to be created without IOL damage Adhesions can be severed without entering eye Removing tissue from stroma changes corneal refractive power non-contact cutting Comment Widespread use replaced need for intraocular procedure

Procedure Posterior capsulotomy

Lysis of vitreous strands Peripheral iridotomy

After trauma, surgery Refractive errors

Nd:YAG-106-1

Nd:YAG-106-1 2nd harmonic-562

Experimental

Endolaser vitreous surgery

Diabetes, trauma

Nd:YAG-106-1 Er:YAG-2.9 m

Experimental

8.5.2.1 Posterior Capsulotomy By far the most common and successful photodisruptor application is the creation of a posterior capsulotomy with a Nd:YAG laser, performed several months or years following extracapsular cataract surgery when the posterior capsule, left intact by the surgeon to support a plastic intraocular lens (IOL), opacies due to proliferation of retained lens epithelial cells.54 The patient, who experienced improvement of vision following surgery, suffers slowly declining vision as the capsule opacies. The surgeon uses the Nd:YAG photodisruptor to create a central opening in the capsule, restoring vision. The surgeon must be careful to focus on or slightly behind the posterior capsule, or undesirable pitting or cracking of the polymethyl methacrylate intraocular lens can occur. The use of the Nd:YAG laser in ophthalmology was simultaneously introduced by glaucoma specialist Franz Fankhauser in Switzerland (Q-switched) and cataract specialist Danielle Aaron-Rosa in France (mode-locked) in 1980. Two years later, the rst international congress dedicated to the theory and clinical use of these lasers was held in Munich and the technique gained rapid acceptance thereafter. 8.5.2.2 Vitreo-Retinal Surgery The Nd:YAG photodisruptor has also been used for cutting vitreous strands in the vitreous cavity and anterior chamber, creating peripheral iridectomies, severing trapped IOL haptic loops, and cutting sutures. These lasers allow outpatient surgical treatment and have reduced postoperative complications such as inammation and infections associated with intraocular surgical procedures. In 1980, Parel and Fankhauser designed a 1 mm-diameter Q-switched Nd:YAG laser miniature endoprobe delivery system for severing vitreous and proliferative retinal membranes (Figure 8.22). In these early experiments, Fankhausers group and ours failed to transmit short 10 nsec pulses through the silica bers that were available at the time. After designing a novel laser-to-ber focusing mechanism, and making repetitive attempts, Rol and Fankhauser were successful in transmitting 10 ns (Q-switched), and 250 s to 20 ms (free-running) Nd:YAG pulses of sufcient energy for photodisruption and photocoagulation of intraocular tissues using high purity silica bers. More recently, Margolis et al. successfully photoablated intraocular proliferative membranes using a free-running Er:YAG coupled to a bare ZrF4 ber optic.55 However, hygroscopic, mechanical and optical problems, including laser beam spatial

Therapeutic and Diagnostic Application of Lasers in Ophthalmology

233

stability, remain formidable and further research is required in ber and laser technology before a clinically usable endolaser photodisruptor instrument is available. 8.5.2.3 Intrastromal Corneal Photodisruption Recently, picosecond and nanosecond photodisruptors have been designed for use in corneal surgery. Focused in the middle portion of the central cornea, the high repetition rate (l03104 Hz) pulsed laser beam is scanned parallel to the corneal plane. Photodisruption transforms the thin lamellae of the corneal stroma into gas that, after absorption, may change the outer corneal curvature, resulting in a change in the eyes refractive power. The theoretical advantage to this approach is that the epithelium and Bowmans layer are left intact and, therefore, the unpredictable refractive effect of corneal wound healing may be minimized. In addition, this technique is thought to minimize thermal damage to the collagen lamellae adjacent to the photodisrupted zone. Two systems are presently under development; a frequency doubled (562 nm) mode locked (30 ps) Nd:YAG (562 nm, 30 ps) laser and a Q-switched 1064 nm (5 ns) Nd:YAG. Both use very high numerical aperture aberration-free optical systems to minimize the volume of photodisrupted tissue. Although this approach is intriguing in theory, thus far no data has been published indicating intrastromal photodisruption can alter the refractive status of the eye effectively or predictably. Basic laboratory and clinical research with this technique is continuing.

8.5.3 Photodisruptor Delivery Systems

Because it is technically difcult to transmit Q-switched laser pulses and impossible to transmit modelocked laser pulses with sufcient energy for tissue photodisruption through optical bers, the optical delivery of the photodisruptor laser beam is accomplished via sets of high quality mirrors and high precision articulated arms. The laser optics are normally integrated physically in a conventional slit-lamp biomicroscope or operating microscope in a fashion similar to that shown for the photocoagulating instruments (see Figure 8.12). Because focusing of the photodisrupting laser beam on target is very critical, a set of two or more confocal He-Ne laser beams are focused on the target tissue while observing under magnication (2040 ). To facilitate focusing, the He-Ne laser beams are made to rotate or synchronously alternate so as to produce a single stable spot when focus on target is reached. To minimize the optical aberrations produced by the cornea and off-axis treatment, special contact lenses are normally used. These lenses further increase the numerical aperture of the optical delivery system and therefore minimize the spot diameter and improve overall efciency. With such lenses, capsulotomy can be performed with as little as 0.5 mJ using a 10 ns Q-switched Nd:YAG laser. The systems designed for corneal intrastromal ablation also incorporate an eye-tracking device to achieve the high spatial resolution necessary for this procedure. These systems can also be used for conventional photodisruptor applications such as capsulotomies and iridotomies.

8.6 Photochemical Laser Applications: Photoablation and Photodynamic Therapy

8.6.1 Photoablation Mechanism
A new form of lasertissue interaction was discovered in 1983 by Trokel and Srinivasan, who were working at IBM.56 Srinivasan was studying the far-UV (193 nm) argon uoride excimer laser for photoetching applications when Trokel, an ophthalmologist, noted that corneal tissue could be removed discretely and precisely without any histological evidence of thermal damage to the adjacent corneal tissue. Trokel recognized the potential of the excimer laser to offer a new approach to corneal surgery: corneal sculpting. Photoablation occurs because the cornea has an extremely high absorption coefcient at 193 nm, and the photons at 193 nm are highly energetic, possessing more energy than the carboncarbon bonds interlinking the protein molecules of the cornea. Consequently, 193 nm laser photons rupture, intermolecular bonds and a discrete volume of corneal tissue is removed with each pulse of the laser

234

Lasers in Medicine

Figure 8.22 Endo-laser probe. An aspheric microlens focuses the laser beam above the vitreous uid aspiration port. Endoscopic illumination is provided by a conventional coaxial 12 m ber bundle. The hook is designed to lift and separate the membranes to be cut from the delicate retinal tissue and provides a shield that prevents laser radiation and plasma-induced secondary radiation from reaching healthy tissues. An electromechanical beam switcher (Nd:YAG to argon) located in the remote console allowed intraoperative photocoagulation of intraocular bleeding vessels.

(Figure8.23).57,58 The depth of the ablation is dependent on the radiant exposure and typically varies from 0.1 to 0.5 m per pulse at a radiant exposure of 50 to 250 mJ/cm2 (Figure 8.24).59 The lasercorneal tissue interactions of other far-UV excimer wavelengths including 157, 248, 308, and 351 nm have been investigated, and the 193 nm wavelength generated by the argon uoride excimer has been found to achieve the best quality corneal excisions for clinical use (Figure 8.25). The 157 nm produced by the uorine dimer produced similar-quality excisions,60 but its use is not practical clinically because of high atmospheric absorption at this wavelength. Initially, it was suggested to use the excimer laser as a laser scalpel for corneal surgery in operations such as radial keratotomy.61 However, the excimer laser was found to be a poor replacement for a cutting scalpel, because the laser removes rather than cutting a dened volume of tissue. Even with high quality aberration-free optics, excisions smaller than 60 m were difcult to produce due to multimode operation and poor beam homogeneity. Clinical attempts to use the excimer laser with a set of linear slit-masks for radial keratotomy resulted in even wider (250300 m) excisions (Figure 8.26). Because of the wide excisions and poor corneal stromal wound healing that followed after deep laser excisions, the use of the excimer laser for this type of surgery has generally been abandoned. The most exciting application of the excimer laser was found to be its ability to sculpt or reshape the outer de-epithelialized surface of the cornea to correct refractive errors. This relatively new surgical procedure was named photorefractive keratectomy (PRK) by its developers (Figure 8.27).62 PRK underwent considerable scrutiny because treatment of the central cornea the most important image-forming element of the eye was involved. Early animal studies were encouraging,63,64 and as the laser and delivery system technology matured, condence was gained that sighted human eyes could undergo excimer laser corneal sculpting without vision threatening complications. The rst sighted-eye procedure was performed in 1988,65 and, since that time, several thousand have been performed in Europe and several hundred in the USA as part of a carefully controlled FDA study conducted by three manufacturers: Summit Technology (Watertown, Massachusetts), Taunton Technologies, and VisX (Sunnyvale, California), (the latter two companies have recently merged). Present data indicate that PRK appears to be a safe and effective method for reducing myopia up to 6 diopters with few side effects or complications.6669 Treatment of myopia greater than six diopters has been less successful due to regression of the refractive effect caused by corneal wound healing.70 To date, attempts to steepen the cornea to treat hyperopia with the excimer laser have

Therapeutic and Diagnostic Application of Lasers in Ophthalmology

235

Figure 8.23 193 nm argon uoride laser pulses result in a unique photochemical interaction with the cornea termed photoablation. The laser pulse (a) is highly absorbed in the rst several microns of corneal tissue, primarily by protein chromophores (b). Because the 193 nm photons have energy in excess of the carbon-carbon intermolecular bonds of corneal tissue, the bonds are photoelectrically decoupled, resulting in the ejection of many small fragments from the surface (c). The photoablation process is completed after 15 ns, leaving clean, precise edges with only 0.3 um of adjacent tissue alteration.

been unsuccessful because regeneration of the corneal stroma and thickening of the epithelium cause the curvature to return to its preoperative shape, negating the refractive effects of the procedure. Now approved for general use, the laser may offer an alternative to glasses and contact lenses for millions of myopic individuals, and may represent the most widespread medical application for a laser.71 Longer term studies to assess the overall predictability, stability, and safety of the procedure are under way. Although the excimer laser is entering clinical use, this gas laser has several disadvantages. The apparatus is large and bulky, complicating its use in the clinical environment. Fluorine gas is toxic (ED50 = 0.3 ppm) and requires special safety considerations. Therefore, a solid state laser source in the far UV is desirable. Ren et al. have demonstrated generation of 213 nm from a frequency quintupled Nd:YAG laser using nonlinear frequency multiplying crystals (Figure 8.28).72 They demonstrated high quality photoablation

236

Lasers in Medicine

ABLATION DEPTH/PULSE (mM/pulse)

1.0 0.8 0.6 0.4 0.2 0

10

20

30

40

50

60

70

80

90

RADIANT EXPOSURE (mJ/cm2)

Figure 8.24 Ablation depth of the human cornea per pulse vs. radiant exposure with the 193 nm excimer laser. Tissue ablation rate will vary with the hydration of the cornea.

Figure 8.25 High quality graded corneal excisions produced with the 193 nm excimer laser using the Hanna-IBM moving slit delivery system. Single epithelial cells (Ep) have been cleaved with the laser. Note the recontouring of Bowmans layer (BM) possible with the laser (S = corneal stroma).

of corneal tissue, similar to 193 nm excimer excisions with the 213 nm wavelength. Whether the frequency multiplied solid state lasers can become practical clinically remains to be seen.

8.6.2 Delivery Systems for Photoablative Lasers

Several beam delivery systems have been developed to homogenize, shape, transmit, and focus the ArF excimer laser for corneal ablation. Methods of beam homogenization include moving lens arrays and cylindrical optics. Beam shaping strategies include the use of one or several apertures whose dimension, shape and position relative to the corneal apex are normally controlled by a computer. Rotating wheels that contain apertures of several shapes and sizes,73 expanding or constricting iris-diaphragms,7476 mathematically dened slits, and moving perforated masks have been used to regulate the energy delivered to

Therapeutic and Diagnostic Application of Lasers in Ophthalmology

237

Linear Excision
Figure 8.26 Deep linear corneal excisions produced with a 193 nm excimer laser and mask system. The excisions lled with an epithelial plug and were found to heal poorly, resulting in abandonment of this approach.

Photoablative Keratectomy for Myopia

Figure 8.27 Photorefractive keratectomy (PRK) for myopia with the 193 nm excimer laser. The stromal surface of the cornea is recontoured with the laser following removal of the epithelium by the surgeon. Following reepithelialization, the anterior corneal curvature is atter, reducing myopia. Preliminary evidence indicates that this technique is safe and effective for reducing myopia up to 6 diopters.

the cornea in optical delivery systems. A different approach utilizes a preshaped photoablatable synthetic mask positioned immediately above the corneal surface. A uniform irradiation beam pattern is required with this technique. All of these concepts are based on surface ablation using commercially available excimer laser systems with quasi-rectangular beam patterns (usually 25 mm 7 mm) and low repetition rates of 10 to 20 Hz (Figure 8.28). The low repetition rates require that these systems treat relatively large areas of the cornea with each pulse to complete the treatment within a reasonable clinical time frame (3060 seconds). Therefore, with the exception of the synthetic photoablatable mask, the present delivery systems are limited to creating simple spherical or sphero-cylindrical optical corrections. A laser diode pumped, rapidly pulsed (KHz range) UV solid state laser (frequency-quintupled Nd:YAG laser emitting at 213 nm) might offer an alternative laser source for corneal sculpting. State-of-the-art scanning technology could be used to paint the cornea in any desired pattern, which may facilitate aspheric and

238

Lasers in Medicine

Ablatable Rotating Disc

Diaphragm

Scanning

Figure 8.28 Different methods of shaping the excimer laser beam include an ablatable polymer mask, computer controlled diaphragms, rotating discs, and movable elliptical masks.

spatially irregular corneal sculpting.77 However, multiple technical challenges that face this technology, including low crystal damage thresholds, small spot sizes, and unstable power outputs, must rst be overcome.

8.6.3 Photodynamic Laser Applications

Photodynamic therapy (PDT) utilizes a selective laser wavelength to activate a photosensitizing agent that, in turn, causes damage or destruction to malignant or abnormal tissue. A photosensitizer is a molecule or compound that, when excited by an appropriate wavelength, can transfer energy to oxygen in tissue to produce singlet oxygen, or can dissociate to produce free radicals. Phototoxicity results from the biochemical interaction of tissues with the singlet oxygen and radicals and is a function of the photosensitizers quantum efciency and absorption spectra, its concentration in tissue and the wavelength and uence of the irradiation. Only recently has photochemical reaction with cellular and vascular structures been introduced as a means of therapy.78 Dougherty recognized the potential of di-hematoporphyrins ether (DHE) in the treatment of malignancies. Following intravenous injection, these photosensitizing agents remain longer in rapidly proliferating cancerous tissues, and allow a localized phototoxic effect to be induced by use of adequate photo irradiation. Dougherty developed this procedure, which he named DHE photodynamic therapy (DHE-PDT), more than 20 years ago. The treatment method consisted of injecting 25 g/ml of DHE (Photofrin II) per Kg body weight followed by a 100250 J/cm2 irradiation at 630 nm with a dye laser 48 hours after injection. Irradiation was performed near a DHE absorption peak at 630 nm because this wavelength has a deeper tissue penetration than DHEs other absorption peaks. For photoactivation of DHE, a constant wave argon pumped 630 dye laser with 12 W output is normally used. Postoperative complications of photodynamic therapy include long lasting (up to 3 weeks) hyper-photosensitization in certain patients. Several centers are investigating the use of other photosensitizers, including phtallocyanides, that can be excited at longer wavelengths (also see Chapters 6 and 10). Apart from the treatment of basal cell carcinomas of the face and eyelids, ophthalmic applications of photodynamic therapy have been limited to attempts to treat ocular melanomas and conjunctival carcinoma.79 Because HDP-PDT has failed to demonstrate efcacy to date, several authors are investigating the use of other photosensitizers, including rose-bengal,81 and phtallocyanide.53 In addition, other authors are investigating photodynamic therapy for the treatment of proliferating vascular disorders of the cornea81 and retina,82 the treatment of fungal infections,84 and the treatment of proliferative lens epithelium following cataract surgery.8488 As the later type of treatment is supercial in nature, the constant wave argon 514.5 nm wavelength has also been used successfully. However, because the peak absorption of rose bengal is highest at 562 nm, a solid state laser operating near this wavelength would facilitate ophthalmic applications.

Therapeutic and Diagnostic Application of Lasers in Ophthalmology

239

The laser delivery systems used for PDT are similar optically to those shown in Figure 8.12a, b; a timing mechanism connected to the laser foot-switch control is used to calculate the total radiant dose given to the patient. Clinicians have been plagued by technical problems associated with the argon pumped dye laser. Present systems have not incorporated means to adjust and control the output wavelength in the laser console and, in many instances, irradiation is performed with an off-peak wavelength. Furthermore, the laser output power tends to uctuate during treatment. These problems lead to sub-threshold dose and inadequate treatment. In addition, dye laser breakdown is too common in the clinical setting. Because DHE is injected 48 h prior to treatment, instrumentation breakdown might cause serious medical complications because the clinician must wait 5 days to several weeks for drug clearance in hypersensitive patients before repeating treatment. A solid-state laser providing an output of 1 W at the wavelengths used for photodynamic therapy would be of tremendous clinical value and facilitate the investigation of this new therapeutic modality.

8.7 Diagnostic Laser Applications

The high radiance, monochromaticity, and spatial and temporal coherence make the laser a unique light probe for noninvasive diagnostic applications in ophthalmology. The information contained in laser light reected or scattered by intraocular structures can be detected and analyzed for diagnostic purposes. In general, light scattering is a phenomenon inversely proportional to the fourth power of the wavelength, thus, blue light is scattered more than red light.

8.7.1 Visual Acuity Measurement

In this application, the spatial coherence of lasers is used to form interference fringes on the retina. These are dark lines with variable spacing and orientation whose formation is largely insensitive to the optical clarity of the intervening media. The retinas functional health may be evaluated even when it is obscured by a cataract or other cloudy media existing in the eye. The spacing of the fringes is varied by adjusting the angle of two interference beams and the patient indicates whether they are visible, facilitating prediction of the patients acuity after cataract surgery. Fringe contrast can also be varied by adjusting the relative intensity of two laser beams, yielding the retinal modulation transfer function (MTF) termed the contrast sensitivity function when plotted reciprocally. The light source used for interferometry is generally a He-Ne laser, and the retinal irradiance is in the range of 1 W/cm2. This technology has widespread clinical use today and is very helpful to the ophthalmologist in deciding whether cataract surgery will be of benet to the patient.

8.7.2 Laser Doppler Velocimeter

In laser Doppler velocimetry (LDV), laser light scattered by moving blood cells is shifted in frequency and is used to measure the rate of blood ow in the veins and arteries of the retina. In this application, the temporal coherence of lasers is used. The light scattered at a particular angle by the owing particles, as well as the light scattered from the vessel wall, is collected by a photodectector. The light scattered from the vessel wall is unshifted in frequency and thus acts as the reference beam. Interference between the reference beam and the scattered light can be used to generate a difference in frequency that is directly related to the maximum red blood cell velocity at the measurement site. In this application, it is important to keep a well dened spot on the site to be measured and to know reliably the angle between the incident and scattered light. LDV is a research tool that is providing important information to scientists studying retinal blood ow.

8.7.3 Scanning Laser Ophthalmoscope

The scanning laser ophthalmoscope (SLO) was developed by Webb for viewing the retina and its supporting structures including blood vessels, nerve bundles, and underlying layers.8991 In scanning laser

240

Lasers in Medicine

ophthalmoscopy, a laser beam about 1 mm in diameter at the eyes pupil is focused to a 10-m spot on the retina. The beam is scanned over the retina without changing its location at the pupil (it pivots within the entrance pupil). At any instant, only a 10-m spot on the retina is illuminated, and that instant may last only 100 ns. Light in the illuminating spot may be absorbed or scattered. If absorbed, the detector at that instant records no response, and the display shows black. If the light is scattered, however, some of it will reach the detector, the detector records a voltage and the display shows a spot with the corresponding gray level intensity. The SLO uses a highly collimated laser beam for illumination, thus requiring a smaller entrance aperture and less sensitive collection optics than conventional optical imaging systems. The image detected by the SLO is temporally rather than spatially coded, as with an indirect ophthalmoscope. The SLO has opened a new dimension in psychophysics, because it allows direct observation of the retinal location on which the stimulus is presented. Future applications include use as a therapeutic device (photocoagulator) and a retinal eye tracker.

8.7.4 Spectroscopic Diagnosis of Ocular Diseases

Spectroscopic techniques such as uorescence spectroscopy and Raman spectroscopy are under investigation as noninvasive means to detect various abnormal states of ocular tissues and ocular diseases. Much of the research effort has been concentrated on detection of early cataract formation and certain vitreous disorders. The transparency of the lens changes normally with age, and sometimes degenerates entirely with the formation of a cataract. Laser Raman spectroscopy has been employed as a noninvasive probe to study the metabolic and structural features of the lens.9294 With newly developed near infrared (NIR) Fourier transform (FT) Raman spectroscopy, uorescence interferences from heavily pigmented ocular lenses can be eliminated and spectroscopic information can be remotely acquired via beroptic probe and analyzed in real time.95 An automated laser-scanning-microbeam system has been developed for the uorescence and Raman imaging of the human lens. Age-related uorophor conversion (from blue to green) and intensity change in spatial distribution have been identied.96,97 This spectroscopic information has added to our understanding of the biochemical changes associated with cataract formation.

8.8 Summary
During the past two decades, laser technology has revolutionized many aspects of ophthalmic surgery. Millions of patients have beneted from the restoration or preservation of vision from laser treatment. The use of lasers to successfully treat retinal diseases, glaucoma, and capsular opacication following cataract surgery is now the standard of care. The end result of the interactions among basic scientists, clinicians, and industry that have produced this revolution has markedly improved the ability of ophthalmologists to help their patients. Nevertheless, potential for abuse and exploitation of laser technology exists and should be resisted. The use of expensive lasers should be limited to procedures that they can do uniquely or do more effectively than simpler, inexpensive technology. Use of lasers for dubious indications for a medical marketing advantage not only escalates the cost of health care unnecessarily, but further elevates the publics unrealistic expectation of what laser technology can actually deliver. Still, the future of laser technology in ophthalmology looks very bright. The excimer laser may offer improvements in the safety and efcacy of refractive surgery, and could give millions of myopic individuals a viable alternative to dependence on glasses and contact lenses for clear vision. Like parallel advances in computer technology, existing lasers are expected to become smaller, more capable, and less expensive as solid state laser technology matures.

Therapeutic and Diagnostic Application of Lasers in Ophthalmology

241

Acknowledgments
We are grateful to Drs. Edward W.D. Norton, Victor Curtin, Gary Margules, Takashi Yokura, William Q. Jeffers, Michael Berry, Shiro Takizawa, Gerard Kervick, Tom Patrick, Pascal Rol, Michael T. Gill, Ms. Mary Ann Taylor, and Ms. Barbara French for continuous scientic and moral support. Supported in part by: Research to Prevent Blindness, Inc. by P30 EY06360 (a NIH Departmental Core Grant), Florida Lions Eye Bank, Topcon Research Institute, Helena Rubinstein Foundation, NEI #EY02180 and EY07803-01, Florida High Technology and Industry Council, Miami Veterans Administration Medical Center, T32 EY07092 NIH Training Grant, and P30 EY06360 Departmental Core Grant from NIH.

References
1. FW Newell, Ophthalmology: Principles and Concepts. St. Louis, MO: CV Mosby Co., 1979, pp. 3-79. 2. FM Bessette, LC Nguyen, Laser light: Its nature and its action on the eye, Can. Med. Assoc. J., 141:1141-1148, 1989. 3. RN Frank, On the pathogenesis of diabetic retinopathy. A 1990 update. Ophthalmology, 98:586593, 1991. 4. ML Rubin, Optics for Clinicians, Gainesville, FL: Triad Scientic Publishers, 1974. 5. KN Ogle, Optics: An Introduction for Ophthalmologists, Springeld, IL: C.C. Thomas Publishers, 1968. 6. J-M Parel, GW Crock, LJ Pericic, The optics of the ophthalmoscope and related instrument, J. of Physics E., 13:1242-1253, 1980. 7. GO Waring, Making sense of keratospeak: A classication of refractive corneal surgery, Arch. Ophthalmol., 103:1472-1477, 1985. 8. GO Waring, Development and evaluation of refractive surgical procedures. 1: Five stages in the continuum of development, J. Refract. Surg., 3:142-157, 1987. 9. H Stringer, J Parr, Shrinkage temperature of eye collagen, Nature, 204:1307, 1964. 10. D Ross, G Zeidler, Pumping new life into ruby lasers, Electronics, 39:115-118, 1966. 11. FA LEsperance, Jr., Treatment of ophthalmic vascular disease by argon laser photocoagulation, Trans. Am. Acad. Ophthalmol. & Otolaryngol., 73:1077-1096, 1969. 12. The Diabetic Retinopathy Study Research Group, Preliminary report on effects of photocoagulation therapy, Am. J. of Ophthalmol. 81:383-396, 1976. 13. The Diabetic Retinopathy Study Research Group, Photocoagulation treatment of proliferative diabetic retinopathy: The second report of the Diabetic Retinopathy Study ndings, Ophthalmology, 85:82-106, 1978. 14. The Diabetic Retinopathy Study Research Group, Photocoagulation treatment of proliferative diabetic retinopathy: Clinical application of Diabetic Retinopathy Study (DRS) ndings, DRS Report Number 8. Ophthalmology, 88:583-660, 1981. 15. FA LEsperance, Jr., Ophthalmic Lasers, Photocoagulation, Photoradiation, and Surgery, St. Louis, MO: CV Mosby, pp. 179, 1983. 16. RJ Oik, Modied grid argon (blue-green) laser photocoagulation for diffuse diabetic macular edema, Ophthalmology 93:938-950, 1986. 17. Macular Photocoagulation Study Group, Argon laser photocoagulation for neovascular maculopathy. Three-year results from randomized clinical trials, Arch. Ophthalmol., 104:694-701, 1986. 18. Macular Photocoagulation Study Group, Recurrent choroidal neovascularization after argon laser photocoagulation for neovascular maculopathy, Arch. Ophthalmol., 104:503-512, 1986. 19. Macular Photocoagulation Study Group, Krypton laser photocoagulation for idiopathic neovascular lesions. Results of a randomized clinical trial, Arch. Ophthalmol., 108:832-837, 1990.

242

Lasers in Medicine

20. Macular Photocoagulation Study Group, Krypton laser photocoagulation for neovascular lesions of age-related macular degeneration. Results of a randomized clinical trial, Arch. Ophthalmol. 108:816-824, 1990. 21. Macular Photocoagulation Study Group, Persistent and recurrent neovascularization after krypton laser photocoagulation for neovascular lesions of age-related macular degeneration, Arch. Ophthalmol., 108:825-831, 1990. 22. FA LEsperance, Jr., Clinical applications of the organic dye laser, Ophthalmology, 92:1592-1600, 1985. 23. MR Moster et al., Laser iridectomy. A controlled study comparing argon and neodymium:YAG, Ophthalmology, 93:20-24, 1986. 24. GR Reiss, JT Wilensky, EJ Higginbotham, Laser trabeculoplasty, Surv. Ophthalmol., 35:407-428, 1991. 25. EMV Burskirk, Pathophysiology of laser trabeculoplasty, Surv. Ophthalmol., 33:264-272, 1989. 26. JM Tieisch et al., Racial variations in the prevalence of primary open-angle glaucoma, The Baltimore Eye Survey, JAMA, 266:369-374, 1991. 27. HD Hoskins et al., Subconjunctival THC:YAG laser limbal sclerostomy ab externo in the rabbit, Ophthalmic Surg. 21:589-592, 1990. 28. GL Skuta, RK Parrish, II, Wound healing in glaucoma ltering surgery, Surv. Ophthalmol., 32:149170, 1987. 29. T Jennings et al., Transcleral contact retinal photocoagulation with an 810 nm semiconductor diode laser, Ophthalmic Surg., 21:492-496, 1990. 30. KP Thompson et al., Potential use of lasers for penetrating keratoplasty, J. Cataract and Refract. Surg., 15:397-403, 1989. 31. H Beckman et al., Limbectomies, keratectomies, and keratostomies performed with a rapid-pulsed carbon dioxide laser, Am. J. Ophthalmol., 71:1277-1283, 1971. 32. H Loertscher et al., Preliminary report on corneal incisions created by a hydrogen uoride laser, Am. J. Ophthalmol, 102:217-221, 1986. 33. H Loertscher et al., Noncontact trephination of the cornea using a pulsed hydrogen uoride laser, Am. J. Ophthalmol., 104:471-475, 1987. 34. AG Cartlidge, J-M Parel, T Yokokura, A laser surgical unit for photoablative and photothermal keratoplasty, SPIE Proc. Ophthal. Technol., 1423:167-174, 1991. 35. QS Ren, R Birngruber, Axicon: A new laser beam delivery system for corneal surgery, IEEE J. Quantum. Electronics, 26:2305-2308, 1990. 36. G Horn et al., A new refractive method for laser thermal keratoplasty with the CO:MgF2, laser, J. Cataract. Refract. Surg., 16:611-616, 1990. 37. J-M Parel et al., Laser photothermal keratoplasty (LPTK), Invest. Ophthalmol. Vis. Sci., 32(suppl):995, 1991. 38. DD Koch et al., HF chemical laser photothermal keratoplasty, Invest. Ophthalmol. Vis. Sci., 32:994, 1991. 39. T Seiler, M Matallana, T Bende, Laser thermokeratoplasty by means of a pulsed holmium: YAG laser for hyperopic correction, Refract. Corneal Surg., 6:335-339, 1990. 40. RP Gailitis et al., Laser welding of synthetic epikeratoplasty lenticules to the cornea, Refract. Corneal Surg., 6:430-436, 1990. 41. RA Burger et al., Laser welded vascular anastomosiscomparison of CO2 and Neodymium YAG laser techniques, Urol. Res., 16:127-131, 1988. 42. AF Flemming et al., Laser assisted microvascular anastomosis of arteries and veins: Laser tissue welding, Br. J. Plast. Surg., 41:378-388, 1988. 43. K Weadock, RM Olson, FH Silver, Evaluation of collagen crosslinking techniques, Biomaterials, Medical Devices and Articial Organs, 11:293-318, 1983. 44. J-M Parel et al., Phacoersatz: Cataract surgery designed to preserved accommodation, Graefes Arch. Clin. Exp. Ophthalmol., 224:165-173, 1986.

Therapeutic and Diagnostic Application of Lasers in Ophthalmology

243

45. E Haeiger et al., Accommodation of an endocapsular silicone lens (Phaco-Ersatz) in the nonhuman primate, Ophthalmology, 94:471-477, 1987. 46. PE Bath, G Mueller, DJ Apple, Excimer laser application for cataract surgery, SPIE Proc.: Laser Interaction with Tissue, 908:72-74, 1989. 47. PE Bath, Laserphaco: An introduction and review, Ophthalmic Laser Therapy, 32:75-82, 1988. 48. RH Keates et al., Absorption of 308-nm excimer laser radiation by balanced salt solution, sodium hyaluronate, and human cadaver eyes, Arch. Ophthalmol., 108:1611-1613, 1990. 49. H Loertscher et al., Ocular tissue ablation by a pulsed hydrogen-uoride laser transmitted through an optical ber, Lasers in Surg. and Med., 7:120, 1987. 50. AC Steger et al., Interstitial laser hyperthermia: A new approach to local destruction of tumours, Br. Med. J., 299:362-365, 1989. 51. LO Savaasand, CJ Gomer, AE Proo, Laser induced hyperthermia of ocular tumors, Appl. Opt., 38:2280-2287, 1989. 52. PO Rol, D Beck, P Niederer, Endocular ophthalmoscope: Miniaturization and optical imaging quality, SPIE Proc. Ophthal. Technol., 1423:84-88, 1991. 53. MW Balles et al., Semiconductor diode laser photocoagulation in retinal vascular disease, Ophthalmology, 97:1553-1561, 1990. 54. O Geyer, M Lazar, Laser therapy of eye diseases, Lasers in Surg. Med., 6:423-426, 1986. 55. TI Margolis et al., Erbium-YAG laser surgery on experimental vitreous membranes, Arch. Ophthalmol., 107:424-428, 1989. 56. SL Trokel, R Srinivasan, B Braren, Excimer laser surgery of the cornea, Am. J. Ophthalmol., 96:710715, 1983. 57. R Srinivasan, E Sutcliffe, Dynamics of the ultraviolet laser ablation of corneal tissue, Am. J. Ophthalmol., 103:470-471, 1987. 58. R Srinivasan et al., Mechanism of the ultraviolet laser ablation of polymethyl methacrylate at 193 and 248 nm: Laser-induced uorescence analysis, chemical analysis, and doping studies, J. Opt. Soc. Am. B. Opt. Phys., 3:785-791, 1986. 59. RR Krueger, SL Trokel, HD Schubert, Interaction of ultraviolet light with the cornea, Invest. Ophthalmol. Vis. Sci., 26:1455-1464, 1985. 60. KP Thompson, J Trentacoste, RK Parrish, II, Corneal surgery with pulsed UV lasers, Arch. Ophthalmol., 105:3, 1987. 61. A Tenner, T Neuhann, E Schroder, Excimer laser radial keratotomy in the living human eye. A preliminary report, J. Refract. Surg., 4:5-8, 1988. 62. J Marshall, SL Trokel, S Rothery, Photoablative reproling of the cornea using an excimer laser: photorefractive keratotomy, Lasers Ophthalmol. 1:21-48, 1986. 63. FE Fantes et al., Wound healing after excimer laser keratomileusis (photorefractive keratectomy) in monkeys, Arch. Ophthalmol., 108:665-675, 1990. 64. N SundarRaj et al. Healing of excimer laser ablated monkey corneas: An immunohistochemical evaluation, Arch. Ophthalmol., 108:1604-1610, 1990. 65. MB McDonald et al., Excimer laser ablation in a human eye, Case report, Arch. Ophthalmol. 107:641-642, 1989. 66. MB McDonald et al., Clinical results of 193-nm excimer laser central photorefractive keratectomy (PRK) for myopia: The partially sighted and sighted eye studies, Invest. Ophthalmol. Vis. Sci., 31(suppl):245, 1990. 67. MB McDonald et al., Central photorefractive keratectomy for myopia: The Blind Eye Study, Arch. Ophthalmol., 108:799-808, 1990. 68. T. Seiler, G Kahle, M Kriegerowski, Excimer laser (193 nm) myopic keratomileusis in sighted and blind human eyes, Refract. Corneal Surg., 6:165-173, 1990. 69. RW Zabel et al., Myopic excimer laser keratectomy: A preliminary report, Refract. Corneal Surg., 6:329-334, 1990.

244

Lasers in Medicine

70. JC Liu et al., Myopic excimer laser photorefractive keratectomy: An analysis of clinical correlations, Refract. Corneal Surg., 6:321-328,1990. 71. KP Thompson, Will the excimer laser resolve the unsolved problems with refractive surgery? Refract. Corneal Surg., 6:315-317, 1990. 72. QS Ren et al., Ablation of the cornea and synthetic polymers using a UV (213 nm) solid state laser, IEEE J. Quant. Electron., 26:2284-2288, 1990. 73. PR Yoder et al., Application of the excimer laser to area recontouring of the cornea, SPIE Excimer Lasers and Appl., 1023:260-267,1988. 74. CR Munnerlyn, SJ Koons, J Marshall, Photorefractive keratectomy: A technique for laser refractive surgery, J. Cataract Refract. Surg., 14:46-52, 1988. 75. KD Hanna et al., Scanning slit delivery system, J. Cataract Refract. Surg., 15:390-396, 1989. 76. RG Caro, DF Muller, A medical excimer laser system for corneal surgery and laser angioplasty, SPIE. Lasers in Med., 712:95-98, 1986. 77. QS Ren, RP Gailitis, KP Thompson, Corneal refractive surgery using an ultra-violet (213 nm) solid state laser, SPIE Proc. Ophthal. Technol., 1423:129-139, 1991. 78. JD Spikes, Chlorins as photosensitizers in biology and medicine, J. Photochem. Photobiol. B., 6:259274,1990. 79. RW Lingua, Photodynamic therapy for ocular tumors, J. Photochem. Photobiol. B., 9:119-122, 1991. 80. TS Olsen, NA Lassen, A dynamic concept of middle cerebral artery occlusion and cerebral infarction in the acute state based on interpreting severe hyperemia as a sign of embolic migration, Stroke, 15:458-468, 1984. 81. AJ Huang et al., Photothrombosis of corneal neovascularization by intravenous rose bengal and argon laser irradiation, Arch. Ophthalmol., 106:680-685, 1988. 82. ML Lewis et al., Photochemical thrombosis of retinal and choroidal vessels using rose bengal: In Laser Surgery: Advanced Characterization Therapeutics and Systems, S.N. Joffe, N.R. Goldblatt, K. Atsumi, Eds., SPIE Proceedings Series Progress in Biomedical Optics, 1066:1989. 83. YC Chan et al., An in-vitro evaluation of the effectiveness of photodynamic therapy in the treatment of acanthamoeba polyphaga cysts, Invest. Ophthalmol. & Vis. Sci., 32(suppl):421, 1990. 84. RW Lingua et al., Photodynamic therapy to retard lens epithelial proliferation after lensectomy, Laser and Light in Ophthalmol., 2:103-113, 1988. 85. J-M Parel et al., Endocapsular lavage with Photofrin II as a photodynamic therapy for lens epithelial proliferation, Lasers Med. Sci., 5:25-30, 1990. 86. RW Lingua, et al., Photodynamic treatment of lens epithelial cells for cataract surgery, Biomedical Optics 91 Proc. SPIE, 1423:58-61, 1991. 87. KF Li, RW Lingua, J-M Parel, Optimal parameters for photodynamic destruction of rabbit lens cells in vitro, Invest. Ophthalmol. & Vis. Sci., 31(suppl):201, 1990. 88. KF Li, R Lingua, J-M Parel, 14.5 nm DHE-PDT and lens epithelial proliferation in vitro: A pilot study, Photochem. Photobio., 51:75, 1990. 89. RH Webb, GW Hughes, O Pomerantzeff, Flying spot TV ophthalmoscope, Appl. Opt., 29:29912997, 1990. 90. RH Webb, GW Hughes, Scanning laser ophthalmoscope, IEEE Trans. Biomed. Eng., 28:488-492, 1981. 91. MA Mainster et al., Scanning laser ophthalmoscopy:clinical applications, Ophthalmology, 89:852857, 1982. 92. S. Lerman, R Borkman, Spectroscopic evaluation and classication of the normal, aging, and cataractous lens, Ophthalmic Res., 8:335-353, 1976. 93. NT Yu, BC Barron, Vision research: Raman/Fluorescence studies on aging and cataract formation of the lens, Supramolecular Structure and Function, Ed. Greta Pifat-Mrzljak, Springer-Verlag Berlin Heidelberg, pp.104-128, 1986. 94. NT Yu, M Bando, JFR Kuck, Fluorescence/raman intensity ratio for monitoring the pathologic state of human lenses, Invest Ophthalmol. Vis. Sci., 26:97-101, 1985.

Therapeutic and Diagnostic Application of Lasers in Ophthalmology

245

95. SM Nie, KL Bergbauer, JJ Ho, Applications of near-infrared Fourier transform Raman spectroscopy in biology and medicine, Spectroscopy, 5:24-32, 1990. 96. NT Yu et al., Automated laser-scanning-microbeam uorescence/Raman image analysis of human lens with multichannel detection: Evidence for metabolic production of a green uorophor, Proc. Natl. Acad. Sci. USA, 85:103-106, 1988. 97. NT Yu, BC Barron, JFR Kuck, Distribution of two metabolically related uorophors in human lens measured by laser microprobe, Exp. Eye. Res., 49:189-194, 1989.

9
Cardiovascular Applications of Lasers
9.1 9.2 9.3 9.4 9.5 9.6 9.7 Introduction ....................................................................... 247 History ............................................................................... 248 Anatomy and Pathology .................................................... 249 Physics ................................................................................. 251 Angioplasty (with Mark H. Wholey, M.D.) ..................... 262 Transmyocardial Laser Revascularization (TMLR) ......... 268 Other Surgical Applications of Lasers in Cardiology (with Osvaldo J. Yano, M.D, and Mehmet C. Oz, M.D.) ........................................................ 271
Valvulotomy and Debridement Myotomy, Myomectomy, and Myoplasty Septostomy Endarterectomy Vascular Anastomoses: Laser Welding in the Cardiovascular System Aids to Welding: Tissue Solders Clinical Examples of Tissue Welding

Jordan D. Haller, M.D

Columbia University School of Public Health

References ...................................................................................... 278

9.1 Introduction
This chapter will review the cardiovascular applications of lasers from their initial experimental use shortly after the invention of the laser, through a period of unrealistic expectation and enthusiasm, and to the present period of skepticism. The medical and scientic bases for the application of lasers are summarized so that the causes for failure as well as the potential for benet of cardiovascular uses can be understood in developmental perspective. Medical terms and conditions are explained for the physical scientist. Lasertissue interaction is stressed. Engineering and ancillary physical science topics such as ber optics are omitted for lack of space. In the course of their medical practice, ophthalmologists and dermatologists regularly deal with the effects of light on patients health, so it is not surprising that they were the rst medical specialists to turn to lasers as potential diagnostic and therapeutic aids very shortly after the invention of the laser. The cardiovascular system, however, is not exposed to light, and cardiovascular medical specialists are not usually familiar with the special biologic effects of light and the physical and chemical properties and limitations of various lasers. Access to the cardiovascular system poses an additional challenge. Not only does the cardiovascular physician usually not have a background in the physical-chemical properties of light, but, to be able to treat the disease process, additional technologies such as ber optics and catheter delivery must be utilized and understood to bring the laser light into the cardiovascular system. Lack of fundamental scientic knowledge led to unnecessary confusion and allowed some cardiovascular investigators to pursue impossible goals, while extravagant marketing claims were made for laser angioplasty. The initial unrealistic expectations for laser use in the cardiovascular system have largely been

0-8493-1146-2/02/$0.00+$1.50 2002 by CRC Press LLC

247

248

Lasers in Medicine

replaced by more objectivity. There remains, however, great need for the cardiovascular specialist to understand the limits as well as the opportunities posed by the physical science. The contributions of physical scientists to this educational process are essential and cannot be over estimated.

9.2 History
The rst recorded series of laser experiments on the cardiovascular system were by McGuff et al. in 19631 This group exposed fresh postmortem atherosclerotic plaques to short pulses (0.5 to 3.0 ms) of a ruby laser at 10 to 13 J and noted that the athero-matous plaques were affected more than the calcic plaques. McGuff concentrated on the effect of laser energy on tumors and summarized his work with the cardiovascular system in 1966,2 observing, Calcic plaques appeared to be more resistant to the laser energy and were grossly little damaged by a 10 joule burst. He noted, as had physical scientists working with inert materials, that shock waves of supersonic velocity spread through adjacent biologic tissues, but concluded (erroneously, as results would later show) that they were not of sufcient strength to cause damage to non-ocular tissues. McGuff also noted that bone, which contains inorganic calcium salts similar to that which occurs in atherosclerotic plaque, was more resistant to laser energy than any other biologic tissue. Unfortunately, these basic observations went largely unnoticed and subsequently unheeded by later cardiovascular investigators. Inappropriate applications of laser energy, mainly in various forms of laser angioplasty caused needless pain and suffering, while millions of dollars were spent and ultimately lost in failed business investments. The results of laser angioplasty are reviewed in Section 8.5. Yahr, Strully and Hurwitt reported in 1964 on the enhancement of laser action by selective stains which affect the relative transparency of tissue.3 Subsequent efforts to enhance or tag tissue to obtain selective absorption of laser energy by certain cardiovascular chromophores are discussed in the appropriate sections. This group also created the rst laser-welded vascular anastomoses and reported additional vascular studies with the Nd:YAG and CO2 lasers in 1966.4 The next reports on the use of a laser to aid in vascular anastomoses were by Morris and Carter at the Orthopedic Research Society in 19805 and by Jain and Gorisch in 1979.6 Both groups described the use of Nd:YAG to weld small-diameter vessels in experimental animals.6 Although laser-welded anastomoses of blood vessels have subsequently been shown to be possible and may even have some theoretical advantages over the traditional suture-created anastomoses, the additional time required has mitigated against their use. This topic is reviewed in Section 9.7.5. In 1974, Arapov reported that, under direct vision, he had used a laser to incise a pulmonary valve whose leaets were fused in the closed position. 7 However, his paper was published in Russian and went largely unnoticed. More recently, however, lasers have been used in combination with balloons to open narrowed (stenotic) valves and, in fact, valves congenitally fused closed in newborns (neonates).8 The era of laser angioplasty was ushered in by the report of Macruz et al. in January 1980,9 in which they described the experimental use of an argon ion laser to remove atherosclerotic plaque. This paper was written in Portuguese and published in Brazil, but it contained an English abstract that was widely circulated.10 In May 1988, at a meeting in China, Choy10 described a catheter to deliver laser energy to open blood vessels that were occluded by either atherosclerosis or thrombosis. He had led a patent for this laser catheter delivery system in 1978, but this was not disclosed until June, 1980. Macruz and associates reported on additional experiments with an argon laser at the Fourth Congress of the International Society for Laser Surgery in Tokyo (November, 1981). They also described the use of a laser to weld closed incisions that had been made in arteries (arteriotomies) and to create arteriovenous and arterio-arterial anastomoses (functioning connections). Using beroptic catheter delivery, they reported the performance of laser catheter pulmonary valvulotomy and angioplasty in living dogs at the October 1981 American Heart Association Meeting.11

Cardiovascular Applications of Lasers

249

FIGURE 9.1 Vascular anatomy and access sites for endovascular diagnosis (angiography) and treatment by percutaneous transluminal angioplasty (PTA), including LASER angioplasty

9.3 Anatomy and Pathology

An overview of the cardiovascular system is shown schematically in Figure 9.1 The artery at the elbow, the brachial, and the artery at the groin, the femoral, are large and supercial. Their pulses can be felt easily so they can be entered relatively easily by a needle puncture without a surgical incision. The Seldinger technique,12 by making access possible to any artery in the body, as well as to the heart itself, including its individual coronary arteries, permitted the development of the diagnostic era of angiography, which, in turn, has led to the current endovascular therapeutic era. After placing a hollow needle into either the brachial or femoral artery, a exible guide wire is inserted through the lumen of the needle. Finger pressure is held over the puncture site until a slightly larger-diameter exible plastic catheter can be threaded over the guide wire. The larger catheter minimizes blood leakage from the smaller needle hole. The catheter is then guided under uoroscopy into the vessel that is to be studied or treated. The guide wire is removed, radiographic contrast material is injected to visualize the area and a series of xray lms are taken, on video or on lm, with a rapid cassette changer. The diagnostic catheter can then be removed and replaced by the laser delivery catheter or a balloon catheter or, in fact, any endoluminal mechanical device to perform the percutaneous transluminal angioplasty (PTA), PTCA for coronary arteries, or PTLA when the laser is used. The anatomy of the heart is shown in Figures 9.2 and9.3. Blood enters the right atrium (RA) of the heart from the superior and inferior vena cavae (SVC, IVC). It passes through the tricuspid valve into the right ventricle (RV) and from there is pumped out into the pulmonary arteries (PA) to the lungs to be oxygenated. The entire pulmonary artery, from its origin in the right ventricle to its bifurcation into right and left branches, has been removed in Figure 9.2 to show the more posterior ventricular septum (IVS) and aortic valve. Oxygenated blood returns from the lungs via the pulmonary veins (PV) into the left atrium (LA) and passes through the mitral valve into the left ventricle (LV). This is the thickest, most muscular, chamber of the heart. The left ventricle pumps blood through the aortic valve into the arterial

250

Lasers in Medicine

FIGURE 9.2 Simplied drawing of the heart. (From The Articial Heart: Prototypes, Policies, and Patients, J.R. Hogness and M. VanAntwerp, Eds., National Academy Press, Washington, D.C., 1991).

system to the entire body. After passing through the individual organ and tissue capillaries, the deoxygenated blood returns through the veins to the vena cavae and the cycle is repeated. Note the origin of the coronary arteries at the base of the aorta, just above the aortic valve. Although the heart is lled with the blood it pumps to the entire body, none of this blood can be used to nourish the heart, which receives its own blood supply through coronary arteries. See Section 9.6 for the Transmyocardial Laser Revascularization studies being done to try to take advantage of connections between the ventricular chambers and small vessels within the ventricular walls. Several types of congenital deformities prevent the ow of blood through the heart. Death of the newborn will occur quickly unless an opening can be made in the thin interatrial septum (septostomy) or unless a malformed intracardiac valve can be opened (a valvulotomy note: the term valvotomy is sometimes used interchangeably; grammarians have been debating which term is correct for several centuries and have not reached a conclusion; priority of rst use seems to favor valvulotomy). Abnormal thickening or hypertrophy of any part of the heart muscle, most commonly the left ventricle, may be treated by incision of the muscle bers (myomotomy or myotomy) or by excision (myomectomy). In all of these conditions, the tissue is either brous or muscular and no inorganic calcium particles are interspersed. The tissue is pure and homogeneously organic. Laser treatment of many of these conditions has been attempted and is discussed in Section 9.7. The most common acquired diseases of the cardiovascular system are due to aging and deterioration, leading to atherosclerosis in blood vessels and stenosis and insufciency of the aortic and mitral valves. These are complex biochemical and metabolic conditions and, once they are severe enough to cause symptoms, they always contain inorganic crystals mixed with the original basic organic tissues. They pose the greatest therapeutic challenge to physician and scientist alike. Laser angioplasty is discussed in Section 9.5. Figure 9.3 shows the three-dimensional relationships of the cardiac anatomy diagrammatically in two dimensions. The relationships of the interatrial (IAS) and interventricular septa (IVS) are shown behind the aorta and pulmonary artery and between the attachments of the tricuspid and mitral valves. Understanding of these important structures and relationships is essential to the understanding of the application and use of lasers within the heart as described in the text.

Cardiovascular Applications of Lasers

251

FIGURE 9.3

Two-dimensional diagram of three-dimensional relationships of cardiac anatomy.

Figure 9.4 shows the passage of a catheter through the arterial tree from the brachial or femoral artery as described in Figure 9.1, in the opposite direction of the arterial blood ow. The catheter can be passed retrograde through the aortic valve into the left ventricular chamber to interrupt abnormal conduction pathways high on the ventricular septum, as in position (b), or to drill channels through the ventricular wall as in position (a). Thickened cardiac muscle can also be incised (cut) or excised (cut out, or cut away) with a laser catheter (see Section 9.6.2). A transvenous approach is shown in Figure9.5. The laser catheter is passed along the direction of the ow of blood in the veins after needle puncture of a peripheral vein. The catheter enters the right side of the heart from the superior or inferior vena cava. This approach is ideal for performing an atrial septostomy ( position (b) ) or tricuspid or pulmonic valvulotomy (position (a) ), as well as to destroy abnormal conduction bers high on the right side of the ventricular septum, or, more commonly, along their pathway from the superior vena cavalatrial junction to the interventricular septum (position (a)). In addition to creating a sufciently large atrial septostomy for blood to ow, the septostomy can be kept small to allow passage of the catheter into the left side of the heart (position (c)). Pediatric cardiologists have used such approaches for many years for diagnostic and some therapeutic procedures. (See text for description and Sections 9.7.

9.4 Physics
The lasertissue interactions that occur as laser energy is transferred into living tissue are complex; in addition, the biologic reactions that result in each type of tissue are many and varied, and our understanding of them is, at best, rudimentary. Fortunately, the medical investigator can judge his results from simple observations of patient complications, recurrences, and failures. However, to avoid needless and possibly dangerous applications, the physician should have as thorough an understanding of as many aspects as possible. The greater the understanding of basic science the physician has, and the greater the understanding of the biologic response the physicist has, the more likely the success of an innovation.

252

Lasers in Medicine

FIGURE 9.4 Passage of a catheter through the arterial tree from the brachial or femoral artery in the opposite direction of the arterial blood ow.

Such knowledge is also useful to the physician and basic scientist in understanding the marketing claims of different equipment manufacturers. We describe these physicalchemical factors as they apply to the cardiovascular system, particularly to laser angioplasty, and relate them to current knowledge of the initiation of vessel wall injury and the hyperplastic biologic response that results in restenosis. In general, the transfer of the light energy into a tissue depends on three factors: 1. The wavelength of the light 2. Its power density 3. The chemical and physical properties of the target Figure9.613 shows the energy contained in photons of different wavelengths. The differences in the absorption properties of the tissues of the blood vessel wall and the organic portions of atherosclerotic plaque are very small, especially at the energy density levels needed to remove tissue, although investigators have attempted to utilize such differences between normal vessel wall and the atheroma to aim or direct the laser beam.1420 However, for patients to have symptoms of atherosclerotic disease, the lesions must be ow obstructing, and such lesions always contain inorganic calcium phosphate particles dispersed throughout. Symptomatic, or complex atheromas are, therefore, mixed, and contain both organic and inorganic portions. The inorganic portions contain less water than the surrounding organic tissue. This is important because the energytissue interaction of such a sample is far different from that of purely organic components. The absorption of the laser radiation by the tissue is given by the equation:

I' ( intensity of light transmitted ) = I ( incident ) ( e

-L

Cardiovascular Applications of Lasers

253

FIGURE 9.5

A transvenous approach.

FIGURE 9.6 Photon energy in electron-volts of some lasers currently in use or under investigation for medical applications.

where I is the intensity of the light beam, L is the thickness of the tissue, and is the absorptivity of the target tissue. This relationship has an important consequence in the nature of the lasertissue interaction. If the absorption is strong, the photons penetrate to a depth of only 1/10th to 1/100th of a mm from the tissue surface. Strong absorption tends to minimize the thermal damage to the underlying areas because the diffusion of heat from the laser photons is minimized. Shortening the lasers pulse also minimizes the lateral spread of heat or thermal damage. The wavelengths of infrared photons correspond in energy to vibrational and rotational sublevels of excitation of the molecules of all biological tissues.

254

Lasers in Medicine

This mode of energy transfer is known as heat, and is rapidly redistributed among the molecules and raises the temperature of the target tissue. At power densities of a kilowatt/cm2 or more, especially when the exposure time is in milliseconds (ms) or longer, the temperature of the exposed region can rise to well above the boiling point of water. Here, the primary absorber of the photon becomes an important consideration. Short-pulsed solid state lasers in the near-infrared such as the Er:YAG and the Ho:YAG are well suited by the wavelength to be absorbed selectively by the water component in tissue. The output of the CO2 laser is also in the infrared at a wavelength strongly absorbed by water, but the lack of a suitable means to deliver this energy inside the blood vessel has limited its applications. Because chemical changes that are brought about by ultraviolet photons are pervasive in our lives, a eld of study called photochemistry has evolved over the past hundred years. These chemical processes include effects such as photosynthesis and sun tanning, as well as less desirable effects such as air pollution (photochemical smog) and destruction of the ozone layer in the stratosphere. The absorption process of ultraviolet lasers (mainly excimer lasers) is fundamentally different (Figure 9.7) from the absorption process of the visible and IR wavelengths. With UV, absorption leads to excitation of the electrons that hold the atoms of the structural molecules together. This physical phenomenon is called electronic excitation. It conveys the energy of the photon directly to the bonding electrons and facilitates the breakup of these bonds. The process of excitation itself can cause the bond break, or it can happen subsequently in a few billionths of a second. The typical bond of an organic molecule has energy of 3 to 4 electron volts. The energy of a photon at 193 nm (such as ArF) is 6.2 eV. Although electronic excitation is due to a specic excitation process at a given wavelength, the reactions that follow need not be equally specic. Heating effects as a result of UV excitation may occur, due either to an inefcient bond-breaking process in which photons that fail to break a bond merely heat the molecule, or the bond energy is considerably less than the energy of the photon so that the excess is available for heating.

FIGURE 9.7 A schematic representation and comparison of the interaction of ultraviolet photons with tissue (top) and with the interaction of infrared photons with tissue (bottom). The mesh represents laments of structural protein molecules interspersed with the water that constitutes the major part of the tissue. When the photons are of ultraviolet wavelengths. Absorption is exclusively by the protein, which then undergoes bond-breaking and ablation. The water is expelled mostly in a liquid state, along with the pieces of the protein and its gaseous products. When the photons are of infrared wavelengths, it is the water that absorbs strongly to produce steam. The resulting explosion tears the protein laments apart.

Cardiovascular Applications of Lasers

255

The lasertissue interaction is a collective process in which a stream of photons interacts with a slab of tissue. The density of the photon stream, or power density, is one of two factors that control the density of excited molecules in the tissue the instant the light beam strikes. At the power levels at which lasers operate, the lasertissue interaction is by no means limited to a given molecules being excited by just one photon. Two, and even multi-photon excitations, become possible. Figure 9.8 is a plot of the matrix of power densities and pulse widths currently employed in the gamut of lasertissue interactions. This is explained by: Fluence = Energy deposited/area,

where E is expressed in Joules (4.2 J = 1 Calorie) and Power density = Energy deposited /time so that Power density = Fluence /time.
The shorter the time, i.e., pulse width in which a given amount of energy, uence, is delivered to a target, the greater the likelihood that it will excite molecules by more than one photon. Ablative photodecomposition and photodisruption strongly depend on such multi-photon processes. The same amount of energy, if delivered in a longer pulse (toward the right in Figure 9.8) results in simple single-photon reactions and heating effects. The wavelength of the photon should always be kept in mind because the wavelength determines the energy of the photon and therefore affects, and, in fact, may determine, the nature of the excitation process. The mode of action for UV photons is specic and different from photons of longer wavelengths. Only two types of interaction between the photon beam and the arterial wall (including the atheroma) need be considered in the clinical application of coronary laser angioplasty. These are the interactions brought about by a pulsed infrared laser whose photons are absorbed predominantly by the water in the tissue, and by a pulsed ultraviolet laser whose photons are absorbed exclusively by the organic bodies. Other modes of lasertissue interaction that were tried and have largely been abandoned are: CW argon and CW Nd:YAG, either alone through a ber, or through a metal tipped ber in which the beam really did not interact with the tissue directly, but merely served to heat the metal tip in short periods of time. This so called hot-tip is not a laserpulse interaction. In a modication of the capping of these bers, the laser beam both heated the tip and partially escaped to interact with the tissue. Here again, the output of the laser was in the visible spectrum, so that the interaction of the photons with the tissue

FIGURE 9.8 Power density and pulse width effects (Ready, 1978).21 A matrix showing how different lasertissue interactions relate to the power densities and pulse widths of the laser. Both scales are logarithmic. The diagonal lines connect points of constant energy density (uence) in joules/cm2.

256

Lasers in Medicine

was entirely thermal in nature. (An electrical source of heat energy that was reminiscent of a heating, or soldering, iron was actually tried experimentally but was quickly abandoned due to excessive thermal injury. Nonetheless, the thermal use of the laser continued for several years.) The overall process of using the photon energy of a pulsed laser to cause a transformation in a tissue can be sorted into three individual steps (Figure 9.9). These are: 1) absorption of the photon, 2) bond breaking, and 3) ablation of the products. The last step does not always follow, as the rst two steps can leave a transformed material in place. The process of absorption comes rst and is conned to the surface layers of the tissue. The term ablation refers to a specic miniature explosion that results from the extremely rapid buildup of pressure from any cause, including photo-decomposition, which is non-thermal as well as thermal. The term ablative photodecomposition refers specically to the action of UV photons and is not a thermal reaction. In contrast, the term vaporization refers specically to a thermal process such as occurs with non-UV laser reactions. It is exemplied by boiling water so that it turns into a vapor stream. These terms are often used interchangeably by those inexperienced in the eld, which has led to confusion among cardiovascular physicians. Ophthalmologists and dermatologists are generally more aware of the exact meaning of these terms and, in fact, the former group have found a near optimal application of an excimer laser to ablate essentially pure organic tissue. The 193 nm argon uoride excimer laser was recently approved by the FDA to correct myopia and astigmatism (PRK and PTK) by ablating not vaporizing corneal tissue. The pressure rise that occurs with ablation can momentarily be greater than 1000 atmospheres. When ablation does occur, there is a shock wave that travels in a direction exactly opposite to the direction of the expelled products: the shock wave travels into the tissue. Its presence has been detected and its passage timed by a piezo-electric pressure sensor that can be attached to an isolated piece of tissue.22 The shock wave shows the time scale in which ablation occurs. If the photons, especially in the ultraviolet, are

FIGURE 9.9 Transfer of photon energy into target issue. Hypothetical steps in the interaction of a laser beam with tissue. Top: the laser radiation, which is dened by a mask, is absorbed; Middle: chemical bonds are broken in the tissue by the photon energy; Bottom: the products ablate at supersonic velocities, leaving an etched sample.

Cardiovascular Applications of Lasers

257

delivered to the tissue in a sub-microsecond pulse, the shock wave also builds up in a similar scale of time, demonstrating that the process of tissue breakup is quite rapid. An equally important point of concern is the potential for damage to the artery wall that can be shock-induced. The shorter the laser pulse, the higher the power density, (for a given energy content, see Figure 9.8) and the more violent the force of ablation. Because ablation and its shock wave are paired in magnitude by a physical law, the ablation that is brought about by pulses of high power density cause shock damage that is readily detected. In biomedical applications, the power densities should be kept as low as possible to achieve the desired result, but the cumulative action of numerous ablative pulses must also be taken into account. Laser ablation of atheromas is achieved with a succession of laser light pulses delivered to the site by means of an optical ber. A single pulse removes (ablates) only a minute volume of the atheroma, the cross-section of this volume being dened by the cross-section of the beam, and the depth being related to the uence of the laser pulse (Figure 9.10). The normal arterial wall is also susceptible to these pulses, and its etch depth is also shown as a function of uence. A threshold uence must always be exceeded before there is any ablation, and this threshold varies with the absorption characteristics of the tissue. Most, if not all, of these complex physical phenomena that occur with laser energy transfer had been observed and were well known to scientists long before physicians began to apply lasers to cardiovascular tissue. In addition to the obvious thermal effects and vaporization, scientists were aware of many nonthermal effects such as recoil, pressure wave formation, ionization, free radical formation, photochemically induced oxidation, and even multiphoton absorption processes and mechanically induced shock waves.2325 The development of the excimer laser in the mid 1970s, however, gave scientists a new tool capable of heretofore unattainable ultraviolet wavelengths. Srinivasan was one of the rst to study the effect of these new excimer lasers. In 1982, he described chemical bond breaking and introduced the term ablative photodecomposition to describe the interactive process between far ultraviolet radiation and protein molecules, resulting in a photochemical breakdown of molecular bonds with virtually no thermal effect.26,27 Excimer lasers with wavelengths typically between 151 and 351 nanometers (0.151 to 0.351 m) contain photons with 10 to 100 times more energy than any lasers previously available; these wavelengths correspond to electronic and ionizing excitation. As part of a study of the basic energy transfer process, Srinivasans group specically studied the process of ablation of vascular tissue by high-speed photography. The set-up is shown schematically in Figure 9.11 and the photo results in Figures 9.129.15.

FIGURE 9.10 Etch-curve for normal (continuous line) and atheromatous (dashed line) aortic wall at 249 nm and 351 nm. The energy density is the total value for a train of 250 pulses.

258

Lasers in Medicine

FIGURE 9.11 Schematic diagram of experiment.

A sample of artery was obtained at postmortem, and laser energy at 308 nm was channeled through a quartz ber of 600 m cross-section to ablate the inner surface of the artery in air. The laser pulse had a half-width of 30 ns. To freeze the motion of the ablated products at any time t following the ablation pulse, a dye laser that produced visible light pulses of less than 1 ns duration was used to provide illumination for the photo. A timer (trigger) connected to both lasers allowed the dye laser (which provided the ash illumination for ash photography) to be red at a controlled delay after the ultraviolet laser pulse had been red. The purpose of these photos was to investigate the ablation process itself when the optical ber was placed well above the tissue surface (Figure 9.12) and when the ber was in contact with the target tissue (Figure 9.14). Ablation of the normal wall and atheroma was also compared. Many studies by other investigators have since conrmed Srinivasans work and advanced our understanding of this ablation process and the consequences of the blast wave that is propagated.

FIGURE 9.12 Ablation of intimal wall by a single 308 nm pulse seen in prole. The vertical cylinder in the middle of the frame is the distal end of a 600 m thick optical ber stripped of its outer protective plastic coating. The frame covers approximately 3 mm by 3 mm. The delay between the excimer pulse and the dye laser pulse: 750 ns.

Cardiovascular Applications of Lasers

259

FIGURE 9.13 Ablation of soft plaque by a single 308 m pulse seen in prole. The delay time is 1s.

FIGURE 9.14 Ablation of intimal wall by a single 308 m pulse seen in prole. The distal end of the ber is just in contact with the surface of the tissue. The delay time is 1000 ns.

FIGURE 9.15 Ablation of soft plaque, predominantly organic tissue with little or no inorganic crystals, by a single 308 m pulse seen in prole. The distal end of the ber is just in contact with the surface of the tissue. The delay time is 30 s.

260

Lasers in Medicine

Because ablation is an explosion brought about by the generation of an enormous pressure inside the tissue, the products emerge from the surface at high speed, which can exceed the velocity of sound in air. When this stream meets the air at atmospheric pressure, a blast wave is created that becomes visible as a gas bubble because of the great difference in the refractive indices. This blast wave is usually followed within microseconds by a stream of debris referred to as a plume, which consists of solid particles of tissue or atheroma and water droplets that are driven by the gas stream. Both the blast wave and the plume slow down as they expand and cool down. The blast wave is seen to emerge from the surface of normal artery in less than 750 ns in Figure 9.13. and expand progressively. Its average velocity is 7.5 105 cm/second, which is in the supersonic range. The prole of the blast wave is complex and actually shows an inverted pattern. It is possible that the ber end, which is only 0.5 mm from the tissue surface, causes a reection of the blast wave. The velocity for the front edge of the wave is 4 104 cm/second, slightly greater than the velocity of sound in air. In Figure 9.14, the ablation of soft plaque is seen to be much different from the ablation of the artery wall. The blast wave is only faintly seen. The plume emerges with a velocity similar to that noted in the artery wall, but its density (opacity) is far greater, which suggests that the amount of solid and liquid material ablated is also greater. Material is also ejected from the surface for a much longer time up to 15 s. The effect of contact between the ber end and the intimal wall on the ablation is pictured in Figure 9.14. As the blast wave is propelled by rapidly expanding gases at high pressure, it manages to escape from the connement of the ber end. In polymer ablation, it has been estimated that the gas pressure built up during ablation might be of the order of 100 to 1000 atmospheres. It is not surprising that gases at such high pressure would escape in the interstices between the ber and the tissue surface. However, the plume is almost totally suppressed, with just a trace of debris being visible. Figure 9.15 shows that, when the ber tip is gently pressed against the surface of the soft plaque, the result is different from that seen when the tip is barely in contact with the intimal wall. There is a violent upheaval of the surface around the ber tip as the conned gases burst out of the limited volume in which they are formed. During laser angioplasty, the ber tip is steadily advanced into the plaque to open a channel. A series of high-speed photos were taken at a constant time delay of 60 sec between the ultraviolet pulse and the ash pulse to determine how the progress of the tip into the plaque affects the ablation caused by successive pulses. The rst pulse was red with the ber tip pressed gently to the surface of the plaque. Violent bursting of the surface is observed. A second pulse was delivered with the ber tip not advanced. A similar outburst was not seen. However, when a pulse was red after the tip was advanced until contact with the tissue surface was again felt, and with the ber end is buried inside the plaque, the ablation is once again as violent as with the rst pulse. This series of photographs reveals an additional important problem in the performance of laser angioplasty. The interventionalist positions the distal end of the laser catheter against the obstruction and uses the tactile response from rm contact between the ber end and the tissue surface as guidance. As the catheter moves farther into the plaque, a tight channel is formed. Kar and Biamino28 also noted this phenomenon. The enclosure of the distal end of the catheter into the plaque traps the expanding gases until only a violent ablation can permit their escape. This can cause an unacceptable stress to the walls of the vascular system. Although these photographs were obtained using a UV laser, similar explosive reactions occur with all lasers. The pressure buildup preceding ablation may be due to different lasertissue reactions, but the effects of the ablation depend only on the explosive power behind the phenomenon. The effect of an excimer laser on the inorganic molecules of calcium phosphate that are mixed throughout the organic stroma of plaque appears to be mechanical (See Figure 9.16). The inorganic crystals appear to be pulverized (see Figure 9.17). Several mechanisms and processes involved in pulverizing the calcium include:29

Cardiovascular Applications of Lasers

261

FIGURE 9.16 Effect of excimer laser on pure organic portion of human atherosclerotic plaque. Because of explosive mechanical processes, even pure organic portions are ripped and torn.

1. Although there is far less water in calcium phosphate than in the organic part of the plaque, the water that is present is disrupted by secondary and possibly thermal means to produce steam. This steam explosion can shatter the adjacent crystalline material. 2. Absorption of laser energy by the remaining dehydrated inorganic mineral crystals can result in differential byproducts expansion, which, in turn, can cause shattering and disruption. 3. Photochemical disruption of the chemical bonds that bind the inorganic calcium phosphate molecules together is now believed to occur. 4. Mechanical vibrational disruption of inorganic molecules has long been known to occur in nonbiologic systems. 5. Energy that has been stored within the inorganic plaque as it was formed is released. This includes additional mechanical and possibly some thermal energy. The potential medical application of ablative photodecomposition, rst described by Srinivasan in 1982 and summarized by him in 1986,30 1989,31 and 1994,32 was very quickly recognized as a possible means of avoiding thermal damage and efforts were directed to develop a beroptic delivery system. To transmit the very high peak powers of these very short pulses, the pulse duration was stretched. While threshold for ablative photodecomposition may no longer be achieved, the energy can be delivered through the ber in very short pulses that are still measured in nanoseconds. Absorption depth is very shallow so that thermal damage has not been noted. Mechanical shock wave damage, however, is now believed to be responsible for the high incidence of restenosis.

262

Lasers in Medicine

FIGURE 9.17 Effect of 308 excimer laser on section of hard or calcied plaque. Note the fragmentation into smaller particles but the absence of evidence of melting of these fragments, consistent with mechanical energy transfer.

9.5 Angioplasty (with Mark H. Wholey, M.D.)

The overwhelming cause of incapacitating occlusive vascular disease is the metabolic disorder termed atherosclerosis. (The older term, arteriosclerosis, means literally just hardening of the arteries). The normal arterial structure is depicted graphically in Figure 9.18. The nonsurgical endovascular treatment of occlusive arterial disease including the use of lasers is generally acknowledged to have begun with Andreas Gruentzigs report in 197933 that he had successfully opened both coronary and peripheral arteries by inating a balloon attached to a thin exible catheter. Charles Dotter34 had reported as early as 1964 on a method of progressive dilation of peripheral arteries with exible but uncompressible rods or catheters. Although Dotters report provided the stimulus for Gruentzig, it generated more controversy than enthusiasm with other medical investigators. Nonetheless, it is now recognized that soft plaque is often pushed aside by stiff catheters. This has been termed Dottering and may occur with any rigid device including hot tips and laser catheters. The balloon dilation method of Gruentzig however, combined with the Seldinger technique, provided a simplied means of opening vessels of any size by using only a needle puncture followed by the insertion of a single narrow catheter. Many other mechanical devices have been developed since, but balloon dilation remains the standard nonsurgical technique. As a result of the success of percutaneous transluminal peripheral and coronary angioplasty (PTA/PTCA), experiments with the commercially available lasers at that time, the CW Nd:YAG, argon and CO2 were done on excised human arteries or on cholesterol plaques that had been induced by diet in experimental animals.3538 Dead tissue does not react and animal tissue bears no true resemblance to human atherosclerosis. All of these lasers removed atherosclerotic material but the human endovascular reaction, however, could not be assessed or even appreciated at that time. Despite the lack of scientic background, human experiments were begun in 1983 with CW argon lasers coupled to bare silica bers.39,40 Occluded peripheral and coronary arteries were opened, proving the feasibility of the method, but the perforation rate was unacceptably high,41 and occlusion usually recurred within several months.

Cardiovascular Applications of Lasers

263

b c d e

FIGURE 9.18 Schematic of a normal artery cross-section. (a) intima, inner-lining layer of 1-3 cells thickness. (b) internal elastic lamella; this is a tissue plane within the media that demarcates the subjacent intima from the media; it provides a natural surgical cleavage plane that enables the surgeon to peel away the plaque, a procedure called endarterectomy. (c) media. (d) external elastic lamella, the tissue plane that separates the media from the exterior adventitia. (e) outermost rim of media subjacent to the most outer layer, (f) adventitia, the most outer layer of the vessel wall.

The perforation rate was decreased by capping the tip of the ber with metal ,but now there was no true lasertissue interaction because the laser merely served to heat this hot tip.42 Nonetheless, with the encouragement of marketing and sales claims, some physicians actually performed the equivalent of running hot pokers through the vascular system to cook fatty droplets. These initial clinical experiments were reported enthusiastically in the medical literature and lay press and were vigorously promoted by the marketing and advertising efforts of some newly formed entrepreneurial companies. This period has been described as overly aggressive and premature clinical use of poorly characterized laser angioplasty systems fueled by exaggerated expectations and a limited understanding of laser microsurgery.43 Rutherford summarized deviations from what had been traditional and acceptable standards of ethical medical practice,44 noting that they included, Problematic reporting practices .elimination of initial failures not distinguishing between primary and secondary patency counting re-treated patients as successes using subjective rather than objective criteria comparison with historical controls selected with self-serving bias (and) failure to follow uniform standards in evaluating and reporting results . He concluded that, Clearly some practical constraints must be imposed to curb this tendency toward unreasonable commercialism. Unfortunately, this was the atmosphere into which laser angioplasty was born. Zarins45 has also reviewed the lack of constraints during this period, especially in regard to conicts within the medical specialties, and characterized the issues raised between the various medical specialties as The Vascular Wars of 1988. Enthusiasm fueled by the unrealistic optimism of marketing hype reached the point that some patients actually demanded laser angioplasty and searched for physicians and hospitals with any type of a laser. Rutherford46 reviewed the literature at this time and estimated that, despite a success rate of less than 50% in lesions longer than 7 cm (which is a reasonable assessment of the majority of clinical problems and extent of the disease) in less than 2 years, more than 20,000 hot-tip angioplasties were performed in more than 300 laser centers in the United States alone. In addition to the many adverse effects of the hot tip, more problems are not so obvious. Iodine has been shown to be liberated from the contrast media during the procedure, both in the form of soluble free iodine and in solid elemental iodine debris that can break off and embolize. Free iodine embolizes immediately. In addition, both blood and contrast agents have been shown to boil in proximity to the probe tip. Blood so heated exhibits markedly accelerated clotting properties.47

264

Lasers in Medicine

The initial accomplishment of pushing metal-capped laser-heated ber beyond an obstruction now seems more probably the result of technical advances in guidewire and catheter technology plus the effect of Dottering, rather than the result of any laser phenomenon. Surprisingly, however, FDA approval was quickly granted for peripheral use, mainly on the basis of the distorted claims of acute, or immediate, technical success. Gradually, as long-term results were reported, it became apparent that there was no signicant improvement in long-term patency, and that restenosis quickly occurred. Results with some subsets of patients were not only worse than with either surgery or PTA, but worse than the natural history of the disease untreated.48 Today, the hot-tip has essentially been abandoned as a peripheral vascular treatment in favor of PTA or one of the other FDA-approved mechanical devices, or for surgical bypass. Attempts to use these thermal systems within the coronary arteries were quickly abandoned, as the heart very rapidly demonstrated that it cannot tolerate the acute thermal insult.4952 Srinivasan had almost immediately recognized the potential medical application of the non-thermal ablative photodecomposition of excimer lasers that he had described.53,54 In a review of some of the commercial development of one of the excimer lasers for angioplasty, Forrester et. al.55 stated, As late entrants into this eld we found that there was a major limitation for vascular application: the thermal energy frequently caused vascular spasm and thrombosis disappointed by our in-vitro results, we decided to review the materials literature to our amazement we found that the excimer laser produced cuts through inorganic material without thermal damage. Although Forrester et. al. credit Srinivasan and Leigh as the source of this information, they cite the date of their publication incorrectly as 1984 (it was 1982) and misidentify the pure organic target material that these IBM researchers used as the basis for their initial ablation experiments (see Reference # 27). The correct information is obviously of great importance, not only to establish proper priority, but to understand the evolution of the research in this eld, the rationale behind it, and, ultimately, to understand the science. Atherosclerotic tissue contains a mixture of inorganic salts in an organic matrix. The transfer of the excimers high-intensity energy into this mixed tissue is complex and has been shown to result in ripping and tearing of adjacent tissue. This mechanical effect is discussed in Section 9.4. Forrester et. al. go on, we packed some tissue in a suitcase and ew to the Argonne National Laboratories in Illinois to use their excimer laser on our tissue. The result was spectacular . The basic scientists at the Argonne National Labs were not only aware of the IBM scientists observations and this new theory of energy transfer that they had termed ablative photodecomposition, but the Argonne scientists understood the signicance of these observations. The Argonne scientists carried out the initial experiments with the 308 nm excimer (xenon chloride) for laser angioplasty and received the patent56 for this achievement. Forresters group at Cedars-Sinai Hospital in Los Angeles then collaborated with scientists from the nearby Jet Propulsion Labs (JPL) in Pasadena to form the Advanced Interventional Systems Company (AIS) for the purpose of developing and commercializing a laser angioplasty system. Laudenslager,57 a former JPL scientist who became the chief scientist at AIS, has acknowledged that, it was Srinivasans work with excimer lasers that was the experimental model for all of our work on lasertissue interactions. AIS has since merged with the Spectranetics Company. Fiber optic delivery systems were developed to transmit the very high peak powers of the micro pulses of excimer lasers.58,59 Technical improvements with guide wires, catheters, and control systems including ber optic visualization (angioscopy) and intraluminal ultrasound, resulted in a very high rate of immediate or technical success. Unfortunately, patency was not maintained and restenosis rates were again disappointingly high. In addition, the laser catheters are unable to create sufciently large lumens, meaning that PTA/PTCA injury is still required for most patients. This dilemma of superimposing a second PTA/PTCA injury on the laser injury to achieve an adequate lumen has not been solved. A report60 that restenosis seemed to be lower for laser treatment alone than for laser plus PTCA has not been conrmed and, in fact, the reverse has also been claimed..61 Nonetheless, in October 1990, the Medical Advisory Panel to the FDA recommended approval of the AIS Companys excimer laser system with a 1.6 mm diameter catheter for ostial lesions and diffuse coronary artery obstructions greater than 20 mm in length. Clinicians have continued to try to identify

Cardiovascular Applications of Lasers

265

subsets of disease patterns for which this laser system might offer some improvement (efcacy) over the poor results achieved with PTCA alone,6265 including heavily calcied long66or ostial lesions,67 failed PTA/PTCA, vein grafts,68 total occlusions,69 and thrombosis. Experience with the AIS system through August,1990 at 15 institutions on 1,519 lesions in 1284 patients was reviewed by Margolis and Mehta.70 Experience with this laser system in 3000 patients from 33 laser centers was reported by Litvack, et al. in 1994. While total numbers were greater, results were essentially the same as in the earlier reports. In November 1991, the FDA Medical Advisory Panel recommended approval of the Spectranetics Companys excimer laser system with 1.4- and l.7-mm-diameter catheters for the following subsets of coronary obstructions: 1. 2. 3. 4. 5. 6. Diffuse and greater than 10 mm in length Aorto-ostial Calcied Total PTCA failures Saphenous vein bypass grafts

The clinical studies on which this decision was based have been published.71,72 Haase et al. summarized results with the Technolas Excimer Laser system.73 This German system has not been presented to the FDA for U.S. approval. The company has modied its original system in an attempt to decrease the accompanying mechanical shock, or blast, wave.74 This group acknowledged that the results of excimer laser angioplasty ablation did not result in an improved patency rate due to the high incidence of restenosis. They attributed this to pressure-wave generation. The catheter was modied by Technolas physicists and engineers to contain eight sections of 20 bers each, for a total of 160 bers of 50 microns each. With the use of a scanner, the individual sections were red one at a time, thus reducing the amount of energy delivered with each pulse. Mean uence was 49+/-2 mJ/mm2. This effectively reduced the shock wave in bench tests. Delivering laser pulses into a glass of water, the pulse wave of the ordinary delivery system feels like a hammer blow, while the pulse wave from the SELCA catheter is barely detectable.75 Animal experiments also showed that mechanical acoustical injury and the proliferative response were both reduced, while ablation was maintained.76 This system was then used in 28 patients with a 1.5mm catheter and four patients with a 1.8mm catheter. The mean reduction in the coronary artery stenosis all 32 patients was from pretreatment of 85+/-10% to postlaser treatment of 57+/-20%. However, half of the patients required additional immediate balloon PTCA, ve of these being necessary because of abrupt closure after the laser treatment. Dissection was observed in 10 patients. Restenosis, however, was noted in 14 of the 24 patients who agreed to angiography within 6 months posttreatment. There was no difference in subgroup analysis between those patients who had laser only and those with laser plus balloon dilation. These authors point out the importance of delivering the energy axially by catheters of the proper size. They conclude that larger-diameter catheters need to be developed to properly evaluate their new system. Despite the failure of any type of laser to achieve the hoped-for improvement in patency over PTA/PTCA, clinical success has been claimed in selected subsets of patients77 who have been able to delay surgical bypass (see Figures 9.19 and 9.20). The arrows point to areas of atherosclerotic occlusive disease. Note the benecial effect of the laser treatment. Efforts continue to salvage the excimer laser and to nd some subset of patients for whom it might offer an advantage. At the same time, it has become apparent that not only is the POBA (plain old balloon angioplasty) adequate for the vast majority of lesions, but that there are some subsets of lesions for which PTCA is superior to the laser. In addition, Appleman et al.78 showed that the excimer yields an adverse clinical outcome of tandem lesions. Spectranetics is also sponsoring clinical studies using their device to help remove pacemaker electrode leads that have become brosed within the right ventricle. This use of a laser catheter was rst suggested by Rao and Flaherty,79 who also demonstrated its feasibility in animal experiments. This may be an appropriate use for the excimer because the electrode is trapped within essentially pure brous tissue;

266

Lasers in Medicine

FIGURE 9.19 Pre-angioplasty. Arrows point to areas of atherosclerotic occlusive disease.

FIGURE 9.20 Post-angioplasty. Note the benecial effects of laser treatment.

there is no inorganic calcium salt to interfere with the lasers mode of ablation of tissue. The company is also preparing a separate clinical study using its excimer laser to reopen thrombosed stents (Figure 9.23). This, too, could be an ideal application of the technology, because the occlusion of stents within vessels is an urgent medical problem and is usually recognized quickly; these obstructions are almost always either thrombus or early hyperplasia, and neither contain inorganic calcium.

Cardiovascular Applications of Lasers

267

FIGURE 9.21 Arterial and venous vessels within the muscular layers of the ventricular wall, and the interrelationship of the arterioles, venules and capillary vascular bed to the ventricular wall.

FIGURE 9.22 Vineberg Procedure. The internal mammary artery is dissected free and divided and then pulled into a tunnel made into the ischemic wall of the ventricle. The side branches are not ligated, but are allowed to bleed freely to communicate with the intramural vessels.

The totally occluded vessel remains a therapeutic problem that usually requires surgical bypass. Although a guidewire can often be passed through short occlusions, it is not often possible to introduce a dilating balloon or device. In an effort to take advantage of the interventionalists ability to pass a guidewire through such occlusions, the Specranetics Company has designed a laser tip guidewire. Reports of the initial clinical studies with this device indicate that it could be passed in 58% to 63% of lesions. The perforation rate was 21% to 27%, and all patients required additional endoluminal treatment, either by laser or balloon angioplasty, artherectomy or stent replacement. Treatment success in a subset of 56 patients was claimed to be 61%.80 The term laser-assisted balloon angioplasty requires special mention. It has been used by some authors to refer to the fact that balloon angioplasty is usually needed after laser angioplasty, but it had a very specic meaning in a series of clinical experiments when a laser was used to deliver thermal energy, or heat, to smooth and stiffen the ruptured walls of vessels following the balloon angioplasty. This latter application is more related to tissue welding than angioplasty. The concept was suggested by Spears,81 who noted that CW Nd:YAG energy could fuse or weld the ssures and cracks that result as part of the necessary plaque disruption with PTCA. From this observation, Spears developed the concept that delivery of controlled laser energy, essentially as heat, to the tissue surrounding the PTCA area, could modify or chemically alter the collagen to become a biologic stent. Clinical trials conrmed that the

268

Lasers in Medicine

tissue surrounding the PTCA could be modied by the heat, but unfortunately, the almost innite variability of the composition and thickness of the tissue receiving the heat resulted in an unacceptable incidence of occlusions81 and the procedure has been abandoned.82

9.6 Transmyocardial Laser Revascularization (TMLR)

In 1977, Mirhoseini83 reported using a CO2 laser to create channels through the ventricular wall into the left ventricular cavity of dogs to nourish ischemic myocardium. He showed patent, endothelialized connections between the ventricular cavity and intramural arterioles 2 1/2 months postop. Although this procedure has an anatomic and physiologic basis, there is considerable confusion, controversy and even skepticism about its potential. Direct coronary artery connections through intramyocardial sinusoids and arterioles to the intracardiac blood were rst described by Vieussens (1705) and Thebesius (1708). These connections have intrigued surgeons and cardiologists at least since 1898, when Pratt kept a cats heart beating by perfusing blood into the ventricle. In 1933, Wearn et al.84 identied two types of connections, arterio-sinusoidal and arterio-luminal, and pointed out that, because they communicate between the cardiac chambers and the coronary vessels, they might serve as a source of blood for ischemic myocardium. These vessels, long before open-heart surgery was possible, quickly became the basis for many attempts at revascularization, most prominently Vinebergs implantation of an open, bleeding internal mammary artery directly,8587 and Becks arterialization of the coronary sinus88,89 and coronary sinus retroperfusion operations.90,91 These procedures were done on beating hearts long before the advent of heart-lung bypass or selective angiography. Vineberg summarized his personal experience over more than 30 years, including post op angiography and proposed explanations for why others had not achieved comparable results.92 Shrager recently reviewed the remarkably hard fought debate that raged over the value of IMA implantation, but recent reports of long-term graft patency and clear patient benet reinforce the belief that Vineberg should be given more credit and that IMA implantation should not be relegated to the status of historical curiosity. Between 1946 and 1975, more than 10,000 of these procedures had been performed, but follow-up data was contradictory and this raging debate was rendered moot, however, by the advent of coronary bypass grafting. the debate over the Vineberg procedure was never resolved, only abandoned. Once bypass was possible, the medical community lost interest in implantation.93 Likewise, the Beck operations were done on thousands of patients in this era. Bailey94 reported his own series of 53 patients with vein segments anastomosed between the aorta and the coronary sinus plus 18 more direct anastomoses, all done on a beating heart before the era of heart-lung bypass. In his encyclopedic review of cardiac surgery, Shumacker95 recounts additional efforts to utilize these intramyocardial anastomoses by arterializing the coronary sinus in patients. These surgeries were undertaken without the benet of coronary angiography, which had not yet been developed. It seems clear now that the major problem was the inability to identify the extent of disease and therefore to make the proper patient selection. Also, it was ultimately shown that it takes several weeks for the new vessels to be able to carry signicant amounts of blood. Many patients died during this period, although those long-term survivors who eventually had selective angiograms demonstrated functioning anastomoses and the value of the method for the appropriate patient. In 1956, Goldman et al.96 suggested a method of using the intracavitary blood to revascularize the heart directly. They cut multiple holes in a segment of artery or plastic tubing that was passed through the ventricular wall into the ventricular cavity and then back out through the wall. Although animal results were encouraging, there is no record that this ingenious procedure was ever performed in patients. In 1965, Sen pointed out that the reptilian heart derives its blood entirely from the ventricular cavities through many small channels into the sinusoidal-capillary bed.97 Following extensive animal experimental studies, he performed multiple transmyocardial needle punctures in four patients with acute infarction, but none survived. He noted that other surgeons had tried the procedure as a last resort for acute infarction also without success.98 A year later, Hershey and White did report a survivor who was successfully resuscitated by transmyocardial needle punctures after having been refractory to all other forms of treatment.99,100

Cardiovascular Applications of Lasers

269

However, in 1969, Pifarre et al.101 claimed that, even if the anastomoses remained patent, it was impossible for an adequate amount of blood to perfuse the ventricular mass because of the hearts systolic contraction. This, coupled with the development of the bypass operation, brought transmyocardial revascularization efforts to a halt until Mirhoseinis work. Some surgeons, however, continued to use retrograde coronary sinus perfusion.102105 Studies of retrograde perfusion have shown that, under certain conditions, anastomotic collateral ow does supply at least a part of the heart muscle.106109 Many patients have such diffuse disease that they are not candidates for either coronary artery bypass surgery or balloon dilation and do not respond well to drugs. They suffer from recurrent angina and congestive heart failure and face early demise. Cardiac transplantation may help some of these patients; however, the 2200 donor hearts that become available each year110 are inadequate to meet the needs of potential recipients. The lack of adequate treatment for these patients stimulated research into alternative means for revascularizing ischemic myocardium and renewed interest in some of the previously abandoned methods. Mirhoseini111115 created 20 to 30 CO2 laser channels in ischemic myocardium of dogs and demonstrated signicantly increased survival in those treated (83%) compared with controls (0%). Patency after 6 months was also observed, suggesting that laser channels eventually reendothelialize. Patency of laser channels 5 years post-op has been reported by ventriculography and histology.116119 In 1986, Mirhoseini120 reported the rst clinical use of this technique in a patient who underwent quadruple coronary by-pass surgery and supplemental transmyocardial laser revascularization to an ischemic area that could not be bypassed. Neither arrhythmias nor signicant cardiac enzyme increase were reported. Although this result was encouraging, Mirhoseini did not study myocardial contractility or myocardial perfusion. Solid evidence of feasibility and safety was therefore lacking, and the medical community remained unconvinced of the efcacy of laser myocardial revascularization. A 2.4 micron Thulium-Holmium-Chromium:YAG (THC:YAG) laser decreased the infarct size area fourfold when compared with untreated animals.121 Because this laser can be conducted by optic bers, an endocardial approach was chosen. A guiding catheter was introduced into the left ventricle across the mitral valve and the laser was applied from the endocardial surface to create nontransmural channels. The versatility of this laser and the potential utilization of optic bers in a closed-chest procedure encouraged pursuit of further studies. This group then studied the effect of the THC:YAG laser on regional myocardial performance using regional preload recruitable stroke work (RPRSW), which is a load-insensitive index of contractility.122,123 A group of dogs was divided into two sets, one of controls and one of laser-treated animals. Left ventricular pressure was measured with micro-manometers. Myocardial segment length was measured with micrometry transducers and cardiac output was measured by thermo-dilution. The area at risk was identied prior to performing the laser channels by brief occlusion of the left anterior descending coronary artery. Intramural laser channels were then made with a ber 600 micron in diameter from the endocardial surface with parameters as follows: 600 mW,10 Hertz, and 20 repetitions, totaling 12 joules/channel and averaging 3 channels/ cm with the total of channels per animal being approximately 40. The left main anterior descending artery was ligated for 90 minutes and then reperfused for 6 hours. Pre-load was varied by temporary inow occlusion. Although the area at risk in control animals became dyskinetic, in laser-treated animals, RPRSW remained positive and contractility was preserved. Myocardial dyskenisis was prevented and contractility was preserved. Microscopic exam demonstrated patent channels but with collateral thermal damage. Subsequently, experiments124 using an acute infarct model showed that contractility was improved in the ischemic region when compared with controls. These studies demonstrate that small conduits from the ventricular cavity can preserve regional myocardial contractility, indirectly implying that myocardial perfusion was present. However, there are also negative reports with this procedure.125 Whittaker et al.126 used Ho:YAG to make l mm diameter channels from the epicardial surface of dog hearts into cyanotic myocardium after ligating the coronary artery, similar to Sens attempts to treat patients with acute infarcts. With an 80

270

Lasers in Medicine

watt CO2 laser, Landreneau et al. created l mm diameter channels every 3 to 5 mm in the area supplied by the left anterior descending coronary artery. They found no benet following acute ischemia.127 The numerous differences among experimental models not only makes comparison difcult, there is even controversy over what the microscopic sections reveal.128 Clinical trials with a specially made CO2 laser are currently under way at several centers. Encouraging results have been reported using computer-controlled 35- to 60-millisecond pulses of an 850 watt CO2 laser to bore 15 to 30 1 millimeter diameter transmural channels. Anginal class has been reduced signicantly with some preliminary evidence by nuclear radiography of improved perfusion of previously ischemic areas.129,130 The technique can actually be used without the heart-lung machine (placing the patient on cardiopulmonary bypass), another advantage for these very sick patients.131 Specic subsets of those patients who might benet and those who would probably not are being identied.132 In Germany, the Laser-Medicine-Zentrum Research Institute and the Technolas Company have collaborated with surgeons in Berlin with an operative approach using an excimer laser. Results are also encouraging.133 The mechanism of the improved perfusion is thought to be due to the created channels, however, alternative hypotheses such as stimulated angiogenesis from the laser treatment have also been suggested.134 The average hospital stay has been only 9 days with an operative mortality of approximately 10%. None of the deaths appear to be secondary to the laser intervention. A carbon dioxide laser was also used to create transmyocardial channels in ve canine hearts, using 800W, 40J/pulse, and an average of 12 channels per heart, or approximately 1 channel per 2 cm2. Hearts were then arrested and excised, the aortic valve was closed to prevent any LV ejection, the main coronaries were cannulated, and perfusion was established from a second dog heart. A cannula was inserted into the LV to control LV lling and afterload pressure. The LAD and epicardial vessels were ligated to make the channel region ischemic. Colored microspheres were injected into the perfusate and the number of spheres per gram of myocardium was measured. Blood from within the LV did not perfuse the myocardium through the TML channels.135,136 Whittaker and Kloner used a 600 micron diameter ber and pulse energy of 9mJ at 20Hz to make three TML channels in rat hearts. These were the parameters that they had found made the widest channels with the least thermal damage in vitro. After allowing for healing, these hearts were challenged with 90 min of coronary artery occlusion. The authors measured the amount of necrosis and counted the open channels. They found that open channels had less brosis surrounding them than did the closed channels. These investigators proposed that the less thermal injury as manifested by brosis, the better the chances for the patency of the channels.137,138 De Guzman et al.139,140 attached the CO2 laser output beam to a thorascope and were able to create TMLR channels in ve living pigs. Like the transvenous catheter approach, this technique avoids the need for an incision into the chest (thoracotomy). The thoracoscope is introduced into the pericardial cavity through a small intercostal incision. Additional follow-up is necessary to support the initial encouraging results. Future research projects should be designed to search for direct evidence of myocardial perfusion by the laser channels. Important issues, such as the relationship between channel morphology, the radius of perfusion from each channel, and the timing of perfusion through channels in relation to the cardiac cycle should be better understood. The optimal wavelength is not yet known; there are theoretical advantages to an IR laser in terms of short penetration depth and highly selective absorption, yet shock wave damage could be critical to patients with already badly damaged hearts. The UV offers the possibility of less thermal and less shock wave injury. If the technique can be shown to be of value, a transvenous catheter system would be much preferred to an operative open-chest approach. All of these factors remain to be worked out. Nevertheless, this technology remains one of the most promising, albeit challenging, applications of lasers in the cardiovascular system.

Cardiovascular Applications of Lasers

271

9.7 Other Surgical Applications of Lasers in Cardiology (with Osvaldo J. Yano, M.D. and Mehmet C. Oz, M.D.)
The attention given to laser angioplasty has almost obscured other potential cardiovascular applications where relatively homogeneous tissue may actually afford a better opportunity for success. Lasers are being used either at surgery, or by an endocavitary catheter to create transmyocardial channels to increase blood ow to ischemic myocardium (TMLC), to ablate arrhythmogenic foci, for valvulotomy, valvuloplasty, septostomy, myotomy, myomectomy, welding and endarterectomy. The cardiovascular structures are shown elsewhere and described in the appropriate text.

9.7.1 Valvulotomy and Debridement

Congenital narrowing of the cardiac valves can be treated by simple incision along the fused cusps. The use of a laser catheter to open these valves was suggested by Macruz et al. in 1981,141 Riemenschneider et al.142 and Gessman et al.143,144 in 1983 and 1984. Rao145,146 has recently reviewed transcatheter methods of relieving pulmonary stenosis, including the use of the laser. This treatment is still experimental, but appears to be a promising alternative to surgery in neonates with these specic conditions; success continues to be reported sporadically. Since1991, pediatric cardiologists at Leeds, UK have used a laser to open a congenitally closed pulmonic valve in nine infants aged 1 to 70 days weighing 2.1 to 4.7 kgm. The distal pulmonary artery trunk beyond the imperforate valve was shown to be present. The laser catheter was manipulated into the right ventricle, below the imperforate pulmonic valve with a 0.018 in guide wire. One to three bursts of CW laser energy of 3 to 5 Watts lasting 3 to 5 seconds allowed passage of the guide wire beyond the valve. This was followed by passage of a coronary artery dilation balloon to dilate the PV.147 Long-standing mitral and aortic stenosis and insufciency are far more distorted and repair is much more difcult. Until satisfactory valve prostheses became available, valvulotomy and various plastic procedures, including debridement (the manual peeling away and excision of brous and calcied tissue from the underlying valve tissue), were the only means of correction.148152 Results were generally poor due to scarring, even when the procedure was done perfectly. With the development of reliable valve prostheses in the early 1960s, these unreliable and time consuming procedures are performed far less often. However, lasers have been proposed as a means of achieving more precise debridement with less damage to the underlying supporting tissue. But studies of excised valves, however, showed both the CO2 and argon lasers to be ineffective.153 Using a variety of near IR lasers in the range of 2 to 3 m wavelengths, Nuss, et.al.154 showed these lasers to be effective in ablating bone with minimal thermal or shock wave damage. The absorption peaks of calcium phosphate and hydroxyapatite (the chemical structure of bone as well as the calcium deposits in atherosclerotic plaque)155159 had long been known to be at 3.1 m and 2.94 m respectively.160 This knowledge is the basis for testing the near IR lasers on calcied atherosclerotic plaque and heart valves. The 2.79 m Erbium:yttrium-scandium-gallium garnet laser (Er:YSGG) was the most effective in removing calcium without signicantly shattering the adjacent tissues161 at settings of 0.2 J per pulse (6 Hz) and uence of 238 Joules per cm2. Freshly removed aortic valve tissue can be meticulously and expeditiously debrided using a Thulium-Holmium-Chromium:YAG laser.161 Use of lasers with lower absorption coefcients for inorganic salts and collagen require higher uences and results in undesired collateral thermal and shock wave injury, leading to unacceptable scarring similar to that seen with ultrasonic and mechanical-manual debridement.162 Laser debridement of acquired calcied stenotic heart valves remains highly experimental, with an uncertain future.

9.7.2 Myotomy, Myomectomy, and Myoplasty

Catheter-guided laser resection or incision has been suggested as a possible alternative to open heart surgery. In 1984, Cleveland163 used as an argon laser system to remove an abnormally enlarged section

272

Lasers in Medicine

Applications and Regulatory Approvals (Oct. 2001) U.S. FDA European Coronary angioplasty Yes Yes Pacemaker electrode removal
Yes Yes

Lower extremity angioplasty In clinical trials (PELA; LACI) Yes Coronary stent restenosis In clinical trial (LARS) Yes
FIGURE 9.23 Spectranetics Model CVX-300: This xenon chloride (excimer) laser system has a wavelength of 308 nanometers; it delivers 40 pulses per second at a pulse width of 125 to 200 ns. The output uence is 30 to 60 millijoules per square millimeter.

of heart muscle that was obstructing blood ow through the aortic valve, a condition known as hypertrophic subaortic stenosis. Conventional surgical treatment for patients with this hypertrophic cardiomyopathy who are refractory to medical therapy is to place the patient on an open heart bypass, open the aorta, visualize the left ventricular outow tract below the aortic valve, and then to make an incision into the thickened heart muscle (septal myotomy or myomectomy).164166 Clevelands cardiologist colleague Isner chose the argon laser, rst because its wavelength (454-514 nm) is in the visible range so it could be used to illuminate the intraventricular operative eld, and second, the wavelength is well matched to the absorption spectrum of myoglobin, the principal constituent of cardiac muscle. As a result, absorbed light energy is converted efciently into heat that can initiate an intense, localized thermal reaction and vaporize the target myocardial tissue into the equivalent of a myomectomy. A 200 micron ber coupled to an argon laser was used to deliver 1.5 watts in less than 4 min to create a 4 cm long, 1 cm deep, 0.5 cm wide incision. The patient had an uncomplicated post-op course and was well 5 years later. This group found no experimental evidence of toxic byproducts or particulate debris.167 Lasers have also been used to make surgical incisions into the heart muscle (cardiomyotomies) for other purposes and for the dissection of the scar tissue that surrounds the heart and great vessels (aorta and pulmonary artery) during reoperations, in an attempt to limit bleeding and minimize trauma to surrounding tissues. The clinical experience for these purposes is very limited and consists mostly of anecdotal reports. Figure 9.23 refers to the laser currently approved for angioplasty.

9.7.3 Septostomy
In certain congenital anomalies, such as transposition of the great vessels, a fatal condition in which the infant is born with aorta and pulmonary artery recessed, total anomalous pulmonary venous return, and pulmonary and tricuspid atresia (a form of very severe stenosis or absorption) with intact septums, surgical correction is often not possible or carries a great mortality in neonates. A palliative mixing procedure is often done, either a surgical septostomy or a Rashkind balloon septostomy.168 Such procedures may allow the infant to survive for several years and to grow to a size that is compatible with survival after surgical correction. Riemenschneider,169 Bommer et al.170,171 and Ben-Sacher et al.172 performed atrial septostomy in human infant autopsy hearts and in living dogs using an argon laser delivered transvenously by a ber. They thought that the method was feasible and could provide a means of performing a septostomy in patients in whom balloon septostomy is either difcult or not possible, such as older infants or those lacking a foramen ovale. To date, there are no reports of this theoretically promising technique in patients.173

Cardiovascular Applications of Lasers

273

9.7.4 Endarterectomy
Attempts to restore blood ow through obstructed arteries and veins by removing the obstructing material had invariably failed prior to dos Santos report in 1947174 of a radical procedure in which he removed not only the occluding arterial atherosclerotic plaque, but all tissue to the internal elastic lamella. While this procedure, termed thrombo-endarterectomy, is generally considered to mark the advent of modern vascular surgery, it has largely been replaced by bypass operations with either a vein or a prosthesis. Nonetheless, thrombo-endarterectomy remains the procedure of choice for certain specic obstructions, most notably in the carotid artery in the neck. The healing of endarterectomy wounds created with different lasers was compared in experimental animals.175177 Thermal injury with resultant thrombogenesis was readily apparent. It was minimized by delivering the laser energy in short pulses. In October 1984, Fasano et al. performed the rst endarterectomy in a patient. They used an argon laser coupled to a 1 mm ber to deliver 21 pulses of 3.5 watts at 0.7 second duration to aid in the removal of a carotid artery plaque in a 51-year-old. There was good clinical recovery.178 Eugene and associates have reported a careful comparison study of CO2, Nd:YAG and argon and concluded that argon provides the most satisfactory tissue response.179,180 This group had earlier attempted to do endarterectomy and weld the endpoints using a xenon chloride excimer laser (308 nm). They found that, with a 50 mJ/pulse at 120 ns pulse duration and either 15 or 20 Hz frequency, they could dissect the atheroma quickly, precisely and easily, but they could not weld the end points. These ndings are consistent with the non-thermal mode of action of this excimer laser. The non-thermal ablation of minute amounts of tissue makes the excimer laser an ideal scalpel, precisely because of its non-thermal mode of action.181 Eugenes group has reported on two clinical trials with argon laser light delivered through a 300 m quartz ber at an average power of 1.0 watt. The surgical technique is shown in Figure 9.24. After incising the artery across the plaque, the ber is positioned at a distance to make a spot size of approximately 0.5 mm in diameter. For perforation, the beam is positioned perpendicular to the surface; for dissection of the plaque, the beam is delivered tangentially, and for welding, it is kept moving over the surfaces while a saline drip is used to cool the tissues. Eighteen endarterectomies ranging in length from 6cm to 60cm were done in 16 patients arteries: aortic, iliac, supercial femoral, profunda femoral and popliteal-posterior tibial. The atheromatous plaque was dissected free within the classic plane at the internal elastic lamella. The endpoints were then

FIGURE 9.24

Laser endarterectomy.

274

Lasers in Medicine

welded to the remaining vessel wall by the laser and the incision was closed by welding. Three patients had early thrombosis requiring thrombectomy. All patients had good clinical results. The anklearm ratio (the ratio of the blood pressure in the ankle to the arm) went from a pre-op mean of 0.53 to 0.97 post op (normal is considered over 0.8). The 1-year patency was 88%.182 Ten carotid endarterectomies were done in nine patients; all were well clinically and the vessels were patent on duplex ultrasonic imaging at 1 year.183 Although these results are satisfactory and the procedure appears to be safe, no real advantage has been shown over the classical surgical endarterectomy. There is great potential value, however, in tacking the endpoints with the laser instead of mattress sutures, and a tapering of large protruding ledges could reduce turbulence and promote patency.

9.7.5 Vascular Anastomoses: Laser Welding in the Cardiovascular System

Lasertissue welding (Figures 9.25 and 9.26) has been investigated for possible advantages of speed, ease and reduction in inammatory response. An additional advantage of laser welds is the ability to allow growth of the anastomosis with the growth of the patient. This is of great importance for vascular repairs in infants and children. Continuous suture repair does not allow for growth, and even interrupted sutures may limit growth. Obstacles to acceptance of this new technology are the excellent results of longestablished suture techniques and hemostatic compounds in most cardiovascular applications, and the added time required to set up to do the laser weld. The exact mechanism of lasertissue welding remains uncertain. Theories have ranged from a denial of even protein denaturation to steadfast support for the formation of covalent bonds as the only means by which sufcient weld strength could be attained. Most investigators assume that tissue welding is accomplished by a photothermal mechanism. However, some have speculated that laser light may induce other effects in tissue independent of heating that are responsible for welding, such as photochemical bond cleavage or bond formation. Such mechanisms occur in the lens of the human eye secondary to UV light absorption from sunlight.184

FIGURE 9.25 Technique for performing laser welding of the distal anastomosis of a vein-to-artery bypass. Sutures are placed at the apices of the incisions and at the middle of the posterior wall (A). Tension on the suture (solid arrow) at the middle of the posterior wall apposes the edges of the repair for welding (B). A suture is placed in the middle of the anterior surface of the anastomosis (C) and apposes the edges for welding (D). Broken arrow represents site of laser energy delivery. (Reprinted from White RA et al., Mechanism of tissue fusion in argon laser welded veinartery anastomoses, Lasers Surg Med. 8:83-89, 1988, without permission.

Cardiovascular Applications of Lasers

275

FIGURE 9.26 Laser welding. Schematic of the concept of argon lasertissue fusion. (a) Apposition of collagen ber in a tissue incision. (b) Mild heating of the collagen bers by laser energy causes unwinding of the helical conguration. (c) Formation of covalent bonds in the annealing tissues with cooling of the site by saline irrigation. (Reprinted from White, RA and Kopchok, GE, Laser vascular tissue fusion: Development, current status and future perspective, J Clin Laser Med. Surg 8:47-54, 1990, without permission.

Several investigators have studied the effects of laser welding on structural proteins such as collagen. Schober et al.185 demonstrated loss of periodicity, increased caliber and interdigitation of collagen brils in rat carotid arteries anastomosed with the Nd:YAG laser. Using a diode laser, Bass et al.186 welded rat tail tendon, type I collagen, the principal structural protein in skin, blood vessels and connective tissue and observed no novel covalent bonds or cleavage of bonds. However, protein was denatured. Disulde bonds are not possible in type I collagen so the role of this type of bonding was not evaluated. Noncovalent bonding or interaction among denatured collagen molecules is suggested as the likely mechanism. Lemole et al. correlated tensile strength of bovine tendon welds with tissue temperature187 by using water bath heating of the tendons and avoiding the use of a laser. The optimum temperature for welding in this model was 62C, suggesting that collagen is denatured, as no covalent bonds are broken. Temperatures high enough to cause protein denaturation in the target tissue are required for tissue welding. The formation of covalent bonding or disulde bonding may play a role in weld strength, but adequate tissue welding can be obtained without the presence of either of these effects. Further denition of tissue level parameters for tissue bonding is required. Kung et al.188 studied rat and rabbit vessels with wall thickness that approximately matched absorption depth of 123 um for the l.9 um laser they chose to use. They delivered 150 mW through a silica ber held approximately l mm from the target, which resulted in a spot of approximately 0.7 mm.The linear delivery rate was about 0.3mm/sec. They found that the acute burst pressures of the welds was the reciprocal of the vessel radius, suggesting that the product of the weld strength and the optical absorption depth is constant over the range of the vessel sizes studied. The weld strength for a weld thickness equal to the optical absorption depth was determined to be 4 x 106 dynes/cm2, which is comparable to the strength of a sutured anastomosis. This group concluded that an optimal weld can be achieved when the tissue absorption of the laser is matched to the vessel wall thickness. The l.9 um M~ Raman shifted NdYAG laser therefore seems well suited for small-vessel welding. A laser with greater penetration will cause

276

Lasers in Medicine

spread of heat and a laser with more shallow penetration will not achieve a weld. This study raises the question of the optimal wavelength for anastomoses of vessels of different thicknesses, such as for an end-to-side anastomosis. Broader clinical acceptance and application of tissue welding techniques depend on simplication of the process and more convincing evidence for its value over conventional suture techniques. If the procedure can be made easier and less dependent on operator skill and judgment, it might nd special application in laparoscopic surgery. Several trends and advances may contribute to the integration of laser welding into surgical practice. These include special devices to facilitate and automate welding,189,190 the use of computer controlled laser dosimetry,191 and compact, relatively inexpensive diode lasers. The development of tissue solders is another advance that makes the welding process more forgiving. The chosen solder can compensate for defects in tissue apposition and imperfections in dosimetry by absorbing most of the laser energy and protecting the underlying tissues to be bonded. A soldered anastomosis also reduces the chance of excessive heating and irreversibly damaging host tissues. If initial attempts with solder fail, reapplication of solder allows repeated trials without thermally damaging the host tissue. In addition lasertissue fusions created with brinogen and other protein solders are stronger than bonds formed without the solder.192,193 Use of a dye that absorbs laser energy enhances selective localization of heating with less collateral injury and allows use of less energy for the weld.194,195 Laser-dye pairs can be chosen so that the lasers output wavelength matches the absorption peak of the dye closely, thereby providing efcient and targetspecic laser energy delivery. This use of an exogenous dye is an alternative to trying to nd a best universal wavelength for all native tissues, because the dye dominates the lasertissue interaction regardless of water content or endogenous pigmentation. A tissue solder consisting of human brinogen mixed with indocyanine green dye and an 808 nm solid state diode laser has been evaluated very recently.196198 No exogenous clotting activators such as thrombin or calcium are used with the brinogen solder. Indocyanine green has a large absorption peak close to the 808 nm. Another feature of the 808 nm wavelength is that it is within an optical window199 for vascular and other tissue. This wavelength fails to show any tissue effects even at the highest power outputs available (9.6 W/cm2) unless the indocyanine dye is present. Thus, despite elevation of the temperature of the dye-enhanced brinogen during laser exposure, the underlying vessel wall is minimally damaged. These concepts are illustrated in a rabbit aortotomy model.200 The immediate bursting pressures of welds created without brinogen (262 +/ 29 mmHg) were signicantly less than repairs with brinogen (330+/-75mm Hg, p<0.05). When exposed to urokinase (25,000 IU), the bursting pressures of repairs performed with brinogen were not signicantly different from baseline(290+/-74 mm Hg). Sutureclosed aortotomies did not burst, but leaked at pressures signicantly below those of laser-closed vessels (l65+/-9 mm Hg, p<0.0l). Fibrinogen-soldered arteriotomy repairs in rabbits were examined from 1 to 120 days postop. No anastomotic ruptures, thromboses or aneurysms were identied. Soldered sites rapidly regenerated a new intimal surface and healed by myobroblast proliferation. No signicant foreign body response was identied; the brinogen was eventually completely reabsorbed. The inability of urokinase to alter the strength of brinogen-soldered aortotomies suggests that the activation of brinogen by the normal clotting mechanism is not the dominant mechanism for producing bond strength.201 Several problems complicate the procurement of brinogen. Although possible, it is logistically complex to take the patients own blood during the preoperative period to produce cryoprecipitate. The use of homologous plasma products is complicated by the risks of blood infection. In addition, naturally occurring glues have varying physical properties such as viscosity and stickiness, which complicates the application. Ideally, a totally synthetic source of the appropriate protein for soldering should be developed. These difculties might all be solved by the development of multicomponent glues made of a protein component, a ground substance component and an energy absorbing dye. By altering the proportions and characteristics of the components, the physical and optical properties of the glue can be tailored to specic applications.

Cardiovascular Applications of Lasers

277

9.7.7 Clinical Examples of Tissue Welding

Jain202,203 and Yamagata et al.204 both used Nd:YAG in 1979 to create anastomoses in 0.3 mm- to 1.0 mmdiameter vessels of animals. Jain claimed that heat caused fusion of collagen within the media and noted advantages of speed and consistency. After achieving 92.5% angiographic patency for up to 6 months postop he used this laser to weld a supercial temporal artery to a middle cerebral artery in a patient205 with cerebral artery insufciency. Clinical experimentation with CO2, laser welding for the creation of arteriovenous stula was initiated in 1985 by Okada et al.206 His rst patient was a 44-year-old woman in chronic renal failure, for whom an arteriovenous (radial artery-cephalic vein) stula was created using low power (2040mW) CO2, laser and a few (typically four) stay sutures. In 1989, Okada et al.207 stated that 35 patients had undergone CO2, laser arteriovenous or arterio-arterial anastomoses without complications due to the laser, including four who underwent femoro-popliteal or proximal popliteal-distal popliteal bypass using a saphenous vein graft. In 1988, John V. White208 used a low-power CO2 laser to create an arteriovenous stula. Further clinical experience was obtained by Rodney A. White et al.,209 who reported in 1990 on a series of 10 patients who underwent forearm Brescia-Cimino arteriovenous stula creation using the argon laser. These stulas were welded using a technique in which four-quadrant sutures divide the anastomoses into four equal segments of approximately 5 mm in length.210 Each segment required about 75 to 100 seconds to fuse using approximately 0.5 watts of argon laser power. Whites group used continuous saline irrigation to prevent tissue overheating. Patients have been followed by physical exam and duplex scanning. After 4.5 years of follow-up, seven of the patients continue to have functioning stulae without evidence of hematomas, stenosis or false aneurysms. Three patients required delayed stula revision, which was thought to be due to inadequate development of the stula or proximal vein thrombosis not related to the laser welding process. Histology of the stulae at time of revision demonstrated good healing. A second clinical series is under way combining argon laser fusion and absorbable sutures. Whites group has also begun a clinical study of laser fusions of the distal anastomoses of femoral-popliteal bypasses. Tissue glues or solders have recently been applied clinically in a small clinical series of arteriovenous stulae.211 Blood loss, operative time, and patency were similar to sutured anastomoses. However, no signicant advantages of the lasered anastomoses over sutured ones were identied. The feasibility of using laser activated tissue solders to reinforce human arteriovenous stulae created with PTFE graft material has also recently been demonstrated.212 Such solder reinforcement could be advantageous in situations in which an anticoagulated patient is bleeding from a site for which placement of an additional suture is difcult or could be complicated by entrapment of the back wall of the anastomosis. These long-term clinical studies have reinforced the experimental observations that laser assisted vascular anastomoses are capable of excellent long-term healing. The initial strength of the weld, however, remains a concern, because all of these examples had reinforcement of the anastomosis with sutures. It is not known if these anastomoses could have been accomplished without the additional immediate strength provided by the sutures. Bass et al.213 have pointed out that completely sutureless anastomosis of microvessels is entirely possible with immediate bursting pressures above 300 mm of Hg. This very high time-zero bursting pressure is probably due, in part, to the favorable consequences of LaPlaces law, when applied to vessels of small diameter. From the clinical standpoint, it is in these microvessels that laser vascular welding makes the most sense, as it is for these vessels that the technical difculty of placing sutures is the greatest, and also for which the hemodynamic ow disruption and healing disturbances caused by sutures have the greatest relative impact. Although the results of these studies have been generally good, signicant obstacles to clinical acceptance remain, including the need for precise tissue apposition, concerns over the initial strength of fusion, difculty of clinical recognition of the optimum endpoint for energy application, logistical problems with laser utilization, and limited availability of biocompatible materials to constitute appropriate tissue

278

Lasers in Medicine

glues or solders. Progress in automated welding techniques, renement and development of better glues or solders and availability of logistically simple diode lasers should eventually overcome these obstacles. The FDA recently approved a beroptic catheter device manufactured by Cardiofocus for treatment of chronic atrial brillation. The catheter device creates a linear atrial ablation lesion that is an improvement over radiofrequency ablation catheters in blocking conduction that would otherwise allow tachyarrythmias to recur.214

References
1. McGuff PE et al., Studies of the surgical applications of laser (light amplication by stimulated emission of radiation), Surgical Forum 14:l43-145, l963. 2. McGuff PE, Surgical Applications of Laser. Thomas, Springeld, IL, 1966. 3. Yahr WZ, Strully KJ, and Hurwitt ES, Non-occlusive small arterial anastomosos with a neodymium laser, Surgical Forum 15:224-5, 1964. 4. Yahr WZ and Strully KJ, Blood vessel anastomosis by laser and other biomedical applications, JAMA, 1:1-,1966. 5. Morris JR and Carter M, Laser-assisted microvascular anastomosis (LAMA), (abstract) Orthopedic Research Society, Las Vegas, 1980. Cited by Frazier OH et al., Laser-assisted microvascular anastomoses; angiographic and anatomopothologic studies on growing microvascular anastomoses: a preliminary report, Surgery 97:585-589, 1985. 6. Jain KK and Gorisch W, Repair of small blood vessels with the neodymium-YAG laser: a preliminary report, Surgery 85:684-8,1979. 7. Arapov AD et al., Initial experience in using the laser ray in heart surgery, Eksp Khir Anesteziol, 1974, July-August (#4), p.10-12, cited by Macruz R et al., Laser Application on Cardiovascular System.Experimental data, p. 288-295, in New Frontiers in Laser Medicine and Surgery, in K.Atsumi, Ed., Excerpta Medica, Elsevier, Amsterdam and New York, 1983. 8. Ozun O et al., Laser valvotomy with balloon valvuloplasty for pulmonary atresia with intact ventricular septum, JACC: vol 27 Suppl., p.120A, 1996. 9. Macruz R et al., Possibilidades terapeuticas do raio laser em ateromas, Arq. Bras. Cardiol vol. 34, p.9, 1980 (Portuguese). 10. Choy DSJ, History of Lasers in Angioplasty, in First German Symposium on Laser Angioplasty. Advances in Laser Medicine I, G.Biamino and G Muller, Eds., Ecomed, Berlin 1988, p.56-64. 11. Macruz R et al., The use of laser beam as a surgical tool for correction of experimental pulmonary stenosis during catheterization procedure, Circ. 64:Suppl. IV:67,1981. 12. Seldinger SG, Catheter replacement of the needle in percutaneous arteriography ACTA Radiol.39:368, 1953. 13. Haller JD and Srinivasan R, LaserTissue Interactions in Laser Angioplasty, in Coronary Laser Angioplasty, Karsch KR and Haase KK, Steinkopf, 1991., p. 19-42. 14. Deckelbaum LI, Lam JK, and Cabin HS, Discrimination of normal and atherosclerotic aorta by laser-induced uorescence, Lasers Surg Med. vol. 7, pp.330-335, 1987. 15. Prince GM et al., Preferential ablation of calcied arterial plaque with laser-induced plasmas, IEEE J. Quantum Electron, vol. 23, pp. 1783-1786, 1987. 16. Murphy-Chutorian D et al., Selective absorption of ultraviolet laser energy by human atherosclerotic plaque treated with tetracycline, Am J Cardiol, vol. 55, pp. 1293-1297. 17. Kittrell C, Diagnosis of brous arterial atherosclerosis using uorescence, Appl Optics, vol. 24, pp. 2280-2281, 1985. 18. Spears JR et al., Fluorescence of experimental atheromatous plaques with hematoporphyrin derivative, J Clin Invest, vol 71, pp. 395-399, 1983. 19. Leon MB, Fluorescence-guided laser assisted balloon angioplasty in patients with femoropolitical occlusion, Circulation, vol. 81, pp. 143-155, 1990.

Cardiovascular Applications of Lasers

279

20. Singleton DI et al., Excimer lasers in cardiovascular surgery: Ablation products and photoacoustic spectrum of arterial wall, App Phys Lett, vol 48, pp. 878-880, 1986. 21. Ready JF, Interaction of high-power laser radiation with materials, (Chapter 13), p. 336-357, in Industrial Applications of Lasers, Academic Press, New York, 1978. 22. Srinivasan R, Dyer PE and Braren B, Far-UV Laser ablation of cornea: photo acoustic studies, Laser Surg Med 6: 514-519, 1987. 23. Polanyi TG, Physics of the surgical laser, Internat. Advances in Surg Laser Oncol, 1:205-215, 1978. 24. Bruma MS, Leffet laser, Gazette Medicale de France 1 (4); 1809-1826, 1967. Cited by Goldman and Rockwell, New York, 1971, p. 201. 25. Goldman L and Rockwell RJ, Jr., Lasers in Medicine, Gordon and Breah, New York, 1971, p. 329-352. 26. Srinivasan R, Mayne-Banton V, Self developing photoetching of poly (ethylene terephthalate) lms by far ultraviolet excimer laser radiation, Appl Phys Lett 41 576 (1982). 27. Srinivasan R, Leigh WJ, Ablative photodecomposition: action of far-ultraviolet (193nm) laser radiation on poly (ethylene terephthalate) lms, J Am Chem Soc 104, 6784-5 (1982). 28. Kar H, and Biamino G, Delivery systems for laser angioplasty particularly considering the safety aspects in Lasers in Medicine II, G Muller and H-P Berlien, Eds., Ecomed Landsberg, 1989, p. 155166. 29. Haller JD et al., Physical and chemical effects of ultraviolet excimer laser radiation on human atherosclerotic plaque: therapeutic implications, Lasers Med Surg, vol 3, pp. 98-102, 1987. 30. Srinivasan R, Ablation of polymers and biological tissue by ultraviolet lasers, Science, 31 October 1986, Volume 234, 559-565. 31. Srinivasan R, Lasertissue interactions, Ber Bunsenges Phys Chem 93, 1989, p. 265-269. 32. Srinivasan R, Interaction of Laser Radiation with Organic Polymers, (Chapter 5) Laser Ablation, in Springer Series in Materials Science, Volume 28, Springer-Verlag, Berlin, Heidelberg, 1994, p. 107-133. 33. Gruentzig AR, Senning A and Siegenthaler WE, Non-operative dilation of coronary artery stenosis: Percutaneous transluminal coronary angioplasty, N Eng J Med vol. 301, pp.61-66, 1979. 34. Dotter CT and Judkins MP, Transluminal treatment of arteriosclerotic obstruction. Description of a new technique and a preliminary report of its application, Circ vol.30, p.654, 1964. 35. Macruz R et al., Possibilidades terapeuticas do raio laser em ateromas, Arq Bras Cardiol, 34:9,1980 (Portuguese). 36. Lee G et al., Laser-dissolution of coronary atherosclerotic obstruction, Am Heart J vol .102 ,p. 1074,1981. 37. Abela GS et al., Effects of carbon dioxide, Nd-YAG and argon laser radiation on coronary atheromatous plaques, Am J Cardiol 50:1199-1205,1982 38. Choy DSJ et al., Laser coronary angioplasty: experience with nine cadaver hearts, Am J Cardiol vol. 50, pp.1209-1211, 1982. 38. Ginsburg R et al., Percutaneous transluminal laser angioplasty for treatment of peripheral vascular disease, Radiol vol.156, pp.619-624, 1985. 40. Choy DSJ et al., Human coronary laser recanalization, Clin Cardiol, vol.7,pp.377-381, 1984. 41. Sanborn TA, Observations on thermal angioplasty, (Chapter 12) in Interventional Cardiology, Vogel JHK and King SB, III, Eds., St. Louis, MO: C.V. Mosby, 1989, pp.145-156. 42. Abela GS and Barbeau GR, Laser angioplasty: potential effects and current limitations, (Chapter 37) in Textbook of Interventional Cardiology, Topol EJ, Ed. Philadelphia, W.B. Saunders, l990, pp. 724-737. 43. Bonner RF et al., New sources for laser angioplasty, (Chapter 8) in Interventional Cardiology, Vogel JHK and King SB, III, Eds., St Louis, C.V. Mosby, 1989, pp.101-118. 44. Rutherford RB, Political issues in endovascular surgery, Surg Clin N. Am., 72:757-765,1992 45. Zarins CKZ, The vascular wars of 1988: the enemy is met, JAMA 261: 416-417,1989.

280

Lasers in Medicine

46. Rutherford RBR, Durham JD, and Kumpe DA, Endovascular interventions for lower extremity ischemia, (Chapter 58) in Vascular Surgery, 4th ed., Rutherford RBR, Ed., W.B. Saunders and Company, Philadelphia, 1992, p.858-874. 47. Dawson P, Booth A, and Machan L, Laser angioplasty: enough to make your blood boil! Dr J Radiol (1994 Apr) 67 (796): 346-8. 48. Hertzer NR, The natural history of peripheral vascular disease: implications for its management, Circ 83, Suppl. I: p1-12, 19, 1991. 49. Nordstrom LA, Experimental background preceding direct laser-assisted angioplasty in the human coronary anatomy, Texas Heart Inst J, vol.16, pp.158-162, 1989 50. Blebea J et al., Laser angioplasty in peripheral vascular disease: symptomatic versus hemodynamic results, J Vasc Surg 13:222-230, 1991. 51. Pilger E et al., Nd:YAG laser with sapphire tip combined with balloon angioplasty in peripheral arterial occlusions.Long term results Circ 83:141-147,1991. 52. Dietrich EB, Laser angioplasty: a critical review based on 1849 clinical procedures, Angiol 41:75767, 1990. 53. Linsker RR et al., Far-ultraviolet laser ablation of atherosclerotic lesions. Lasers Surg Med, 4:201206,1984 54. Trokel, SL, Srinivasan, R and Braren, B, Excimer laser surgery of the cornea. Am J Ophthal 96:710715, 1983. 55. Forrester JS et al., Coronary excimer laser, J Invasive Cardiol 4:75-82, 1992. 56. Gruen, DM, Young, CE and Pellin, MJ, Excimer laser phototherapy for dissolution of abnormal growth. Patent #4,686,979, led Jan. 9, 1984, issued August 18, 1987. 57. Laudenslager J, Personal communication to JD Haller 11/11/1992. 58. Grundfest WS et al., Laser ablation of human atherosclerotic plaque without adjacent injury, J Am Coll Cardiol vol. 5, pp.929-933, 1985. 59. Bohley TK et al., An excimer laser system for coronary angioplasty, SPIE Proc. vol. 1200, pp. 480485, 1990. 60. Karsch KR et al., Percutaneous coronary excimer laser angioplasty in patients with stable and unstable angina pectoris, Circ 81:1849-1859,1990. 61. Kent KM et al., Stand alone excimer laser angioplasty, Circ vol.84, Suppl. II, p.363, 1991. 62. Sanborn TA et al., Percutaneous coronary excimer laser-assisted angioplasty; initial multicenter experience in 141 patients, J Am Coll Cardiol 17:169B1-73B,1991. 63. Bresnahan JF et al., Excimer laser coronary angioplasty: initial results of a multicenter investigation in 958 patients, J Am Coll Cardiol vol.17, Suppl.I, 1991, p. 30A. 64. Mccarthy WJ et al., Excimer laser-assisted femoral angioplasty; early results, J Vasc Surg vol.13, pp.607-614, 1991. 65. Cook SL et al., Percutaneous excimer laser coronary angioplasty of lesions not ideal for balloon angioplasty, Circ vol.84, pp.632-643, 1991. 66. Holmes DR et al., Excimer coronary laser anagioplasty (ELCA) registry; lesion length and outcome, Circ vol. 84, Suppl II, p 362,1991. 67. Eigler NL et al., Excimer laser coronary angioplasty of aorto-ostial stenosis: results of the ELCA registry, Circ vol. 84, suppl. II, p.251, 1991. 68. Untereker WJ et al., Excimer laser coronary angioplasty of saphenous vein grafts, Circ vol .84, Suppl. II, p. 249, 1991. 69. Rothbaum A et al., Excimer laser angioplasty in total coronary occlusions: a registry report, Circ 84:Suppl.II, p.744, 1991. 70. Margolis JR, and Mehta S, Excimer Laser Coronary Angioplasty: American Experience with AIS System, pp.75-102 in Coronary Laser Angioplasty, K.R. Karsch and K.K. Haase, Eds, Springer-Verlag, New York 1991. 71. Sanborn TA et al., Coronary excimer laser angioplasty: the Spectranetics registry, pp.103-ll0 in Coronary Laser Angioplasty, Karsch KR and Haase KK, Eds. Springer-Verlag, New York, 1991.

Cardiovascular Applications of Lasers

281

72. Sanborn TA et al., Lack of effect of lesion severity on clinical success and complication rates with percutaneous excimer laser coronary angioplasty (PELCA), Circ vol. 84, Suppl II, p.362, 1991. 73. Haase KK, Baumbach A and Karsch KR, European experience: coronary excimer laser angioplasty: Technolas study group, pp.111-124, in Coronary Laser Angioplasty, Karsch KR and Haase KK, Eds., Springer-Verlag, New York, 1991. 74. Haase KK et al., Initial clinical experience with a modied excimer laser for coronary angioplasty, Lasers in Med Sci 9:7-15,1994. 75. Personal communication, C.Hohla demonstrated this to JDH. 76. Xie DY et al., In vitro evaluation of ablation parameters of normal and brous aorta using smooth excimer laser coronary angioplasty (SELCA) Lasers Surg Med 13:618-24,1993. 77. Serruys PW, Recanalization of chronic total occlusions using a laser guide wire: a pilot study. J Am Coll Cardiol 1997, Sept 30 (3): 649656. 77a. Hamburger JN et al. Am J Cardiol, 1997 80: 141923. 78. Appleman Y et al., Excimer laser angioplasty versus balloon angioplasty in longer coronary lesions: a multivarient analysis, Circ 92:8 October 15, 1995, Suppl, p. I-94. 79. Rao,G and Flaherty,P.Use of laser to extract unwanted pacemaker leads Texas Heart Inst J16:1638, 1989. 80. Pompa JJ et al. Angiographic outcomes after arterial recanalization of refractory occlusions with the PRIMA Excimer laser wire, JACC 27 (#2, Suppl. A, Feb.), 1996, p.152A. 81. Glazier, J et al. Laser balloon angioplasty combined with local intracoronary heparin therapy. Am Heart J 1997, Aug: 134(2 pt 1): 266273. 82. Deckelbaum LI, Cardiovascular applications of laser technology, Lasers Surg Med., 15:315-341, 1994. 83. Mirhoseini M, Discussion of Techniques and Modalities for Myocardial Revascularization, in Second Henry Ford Hospital International Symposium on Cardiac Surgery, JC. Davila, Ed., AppletonCentury-Crofts, New York, 1977, p.595-577. 84. Wearn JT et al., Am Heart J 9:143-6, 1933. 85. Vineberg AM, Development of an anastomosis between the coronary vessels and a transplanted internal mammary artery, Can Med Ass J 55: 117-9, 1946. 86. Vineberg AM, Clinical and experimental studies in the treatment of coronary artery insufciency by internal mammary artery implant, J Int Coll Surg 1954; 22: 503-8. 87. Vineberg, AM, Myocardial revascularization: a 30-year journey from laboratory to man and back again, in Second Henry Ford Hospital International Symposium on Cardiac Surgery, by JC Davila, Ed., Appleton-Century-Crofts, New York, 1977, p.58-74. 88. Beck CS and Mako AE, Venous stasis in coronary circulation Am Heart J 21:767,1941 89. Beck CS et al., Revascularization of heart by graft of systemic artery into coronary sinus, JAMA 137:436- , 1948. 90. Lillehei CW et al., The direct vision correction of calcic aortic stenosis by means of a pumpoxygenator and retrograde coronary sinus perfusion, Dis Chest 30:123, 1956. 91. Gott VL et al., Retrograde perfusion of the coronary sinus for direct-vision aortic surgery, Surg Gyn Obs l04:319-23, 1957. 92. Vineberg AM, Myocardial revascularization: a 30 year journey from laboratory to man and back again, in Second Henry Ford Hospital International Symposium on Cardiac Surgery, ed. by JC. Davila, Appleton-Century-Crofts, New York, 1977, p.58-74. 93. Shrager JB, The Vineberg procedure: the immediate forerunner of coronary artery bypass grafting, Ann Thorac Surg 57:1354-64, l994. 94. Bailey CP, Surgery of the Heart, Philadelphia: Lea and Febiger, 1955, p. 956-983. 95. Shumacker HB, Jr, The Evolution of Cardiac Surgery, Indiana Univ., 1992, p. 133-138. 96. Goldman A et al., Experimental methods for producing a collateral circulation to the heart directly from the left ventricle, J Thorac Surg 31:364-374 1956. 97. Sen PK et al., Transmyocardial acupuncture, J Thorac Cardiovas Surg 1965; 50: 181-9.

282

Lasers in Medicine

98. Sen PK et al., Further studies in multiple transmyocardial acupuncture as a method of myocardial revascularization, Surgery 64:861-, 1968. 99. Hershey JE and White M, Transmyocardial puncture revascularization, a possible emergency adjunct to arterial implant surgery, Geriatrics 24: 101-108 ,1969. 100. Mirhoseini M, Laser revascularization of the heart, in New Frontiers in Laser Medicine and Surgery, K. Atsumi, Ed., Excerpta Medica, Elsevier, Amsterdam and New York, 1983, p. 296-303. 101. Pifarre R et al., Myocardial revascularization by transmyocardial acupuncture. A physiologic impossibility, J Thorac & CV Surg 58: 424-431, 1969. 102. Menasche P et al., Retrograde coronary sinus perfusion: a safe alternative for ensuring cardioplegic delivery in aortic valve surgery, Ann Thor Surg 34:647- 1982. 103. Menasche P, Piwnica A, Cardioplegia by the coronary sinus for valvular and coronary surgery, J Am Coll Cardiol 1991; 18:628-36. 104. Hochberg MS and Austen WO, Selective retrograde coronary venous perfusion Ann Thor Surg 39:578,1980. 105. Yau TM et al., Which techniques of cardioplegia prevent ischemia? Ann Thorac Surg 56:1-1028,1993. 106. Gates RFL et al., Gross and microvascular distribution of retrograde cardioplegia in explanted human hearts, Ann Thorac Surg 56:410-417, 1993. 107. Tschabitscher M, Anatomy of Coronary Veins, in The Coronary Sinus, Proc 1st Int Symp Myocardial Protection via Coronary Sinus, pp. 8-25. W. Mohl et al., Eds., Steinkopf/Springer, New York, 1984. 108. Heintzberger CFM, The vascularization pattern in the ventricular wall in different species during development, in The Coronary Sinus, Proc 1st Int Symp Myocardial Protection via Coronary Sinus, pp. 47-52. W Mohl et al., Eds., Steinkopf/Springer, New York, 1984. 109. Ratajczyk-Palalska E, and Kolff WJ, Anatomic Basis for Coronary Venous Outow, in The Coronary Sinus, Proc 1st Int Symp Myocardial Protection via Coronary Sinus, p.40-46. W. Mohl et al., Eds., Steinkopf/Springer NewYork, 1984. 110. United Network for Organ Sharing, UNOS Update 7 (May):2,1991. 111. Mirhoseini M, Cayton MM, Revascularization of the heart by laser, J Microsurg 1981; 2: 253-60. 112. Mirhoseini M, Muckerheide M, Cayton MM, Transventricular revascularization by laser, Lasers Surg Med 1982: 2: 187-98. 113. Mirhoseini M, Shelgikars S, Cayton MM, New concepts in revascularization of the myocardium, Ann Thorac Surg 1988: 45: 415-420. 114. Mirhoseini M et al., Clinical report: laser myocardial revascularization, Lasers Surg Med 1986: 6:459-61. 115. Mirhoseini M, Shelgikar S, Cayton M, Clinical and histological evaluation of laser myocardial revascularization, J Clin Lasers Med Surg 6: 73-890 116. Okada M et al., Alternative method of myocardial revascularization by laser: experimental and clinical study, Kobe J Med Sci 1986: 32: 151-61. 117. Okada M et al., A new method of myocardial revascularization by Co2 Laser, Kyobu Geka 1984, 37:100-105. 118. Hardy RI et al., A histologic study of laser-induced transmyocardial channels, Lasers Surg Med 6: 563-73 1987. 119. Graham O et al., Rransmyocardial laser revascularization in ischemic cardiomyopathy, J Heart Lung Transplant, 2001, Jun: 20 (6) 687691. 120. Mirhoseini M et al., Clinical report: laser myocardial revascularization, Lasers Surg Med 1986: 6:459-61. 121. Jeevanandam V et al., Myocardial revascularization by laser induced channels, Surg Forum l99l: 41:225-7. 122. Yano OJ et al., Prevention of acute regional ischemia with endocardial laser channels Ann.Thoracic Surgery, 1993:56; 46-53. 123. Yano OJ et al., Endocardial Laser Myocardial Revascularization. Progress in Biomedical OpticsDiagnostic and Therapeutic Cardiovascular Interventions III, Proc. Int Soc Optic Engrg (SPIE), 1993: 1878: 183-205.

Cardiovascular Applications of Lasers

283

124. Horvath KA et al.,Improved short- and long-term recovery after acute myocardial infarct treated by transmyocardial laser revascularization Surg Forum 43:220-3, 1993. 125. Krabatsch T et al., Histologic ndings after transmyocardial laser revascularization, J Card Surg 1996, SeptOct 11 (5) 326333. 126. Whittaker P, Kloner PA, Przyklenk K, Laser-mediated transmural myocardial channels do not salvage acutely ischemic myocardium, J Am Coll Cardiol 22: 302-309, 1993. 127. Landreneau L et al., Direct CO2 Laser revascularization of the myocardium, Lasers Surg Med 11:3542, 1991. 128. Beranek JT, Laser-mediated treatment channels induce a hyalin degeneration of neighboring myocardium, (letter to the editor), J Am Coll Card 23: 1518, 1994. 129. Crew JR et al., Transmyocardial laser revascularization, J Am Coll Cardiol, 1994:lA-484A, p.l5lA (abstract). 130. Mirhoseini M, Cayton MM, Shelgikar S, Transmyocardial laser revascularization, J Am Coll Cardiol, 1994:lA-484A,p.416A (abstract). 131. Mueller G, Laser-Medizin-Zentrum (LMZ), Berlin, personal communication to JDH, August, 1994. 132. Graham O et al., Transmyocardial laser revascularization in ischemic cardiomyopathy, J Heart Lung Transplant, 2001, Jun: 20 (6) 687691. 133. Hohla C, Technolas, Inc. Munich, personal communication to JDH, August, 1994. 134. Whittaker P, Zheng S and Kloner RA, Chronic response to direct myocardial revascularization; a preliminary study. Diagnostic and therapeutic cardiovascular interventions III Proc Biomed Optics, Circ 88:I-435, 1993 (abstract). 135. Whittaker P, Zheng S and Kloner RA, Chronic response to direct myocardial revascularization; a preliminary study. Diagnostic and therapeutic cardiovascular interventions III Proc Biomed Optics, SPIE l878:l60-5, l993. 136. Kohmoto K, et al., Does blood ow through transmyocardial CO2 laser channels? (abstract) JACC 27: (#2, Suppl A., Feb.) 1996, p. 13A. 137. Kohmoto, K et al., Physiology of chronic transmyocardial laser channels created with CO2 laser, (abstract) Circ 92: #8, Oct. 15, 1995 Suppl. I, p. 1-176. 138. Whittaker P and Kloner P, Excimer laser channels protect against myocardial ischemia (abstract). JACC 27: (#2, Suppl A., Feb.) 1996, p. 13A. 139. de Guzman, Circ 92 #8 Oct. 15, 1995 Suppl. P. I-176. 140. de Guzman et al., Thorascopic transmyocardial laser revascularization, Ann Thorac Surg, 1997 Jul: 64, (1) 171174. 141. Macruz R et al., The use of laser beam as a surgical tool for correction of experimental pulmonary stenosis during catheterization procedure, Circ 64: Suppl. IV: 67,1981. 142. Riemenschneider TA et al., Laser irradiation of congenital heart disease: potential for palliation and correction of intracardiac and intravascular defects, Am Heart J 106:1389-1393, 1983. 143. Gessman L, Reno C, Chang K et al., Feasibility of laser catheter valvotomy for rheumatic aortic and mitral stenosis, (abstract) Circ: 68:21, 1983. 144. Gessman L et al., Feasibility of laser catheter valvulotomy for aortic and mitral stenosis, Am J Cardiol 54:1375-1377, 1984. 145. Rao PS, Transcatheter treatment of pulmonary outow tract obstruction: a review, Prog Cardiovasc Dis 35:119-158, 1992. 146. Rao PS, Transcatheter management of cyanotic congenital heart disease: a review, Clin Cardiol 15:483-486, 1992. 147. Uzun O, et al., Laser valvotomy with balloon valvuloplasty for pulmonary atresia with intact ventricular septum, JACC 27 (#2, Suppl. A, Feb.) 1996, p.120A. 148. Harkin DE et al., The surgical correction of calcic aortic stenosis in adults: results in the rst 100 consecutive transaortic valvuloplasties, J Thor Surg 36:759-73, 1958. 149. Kirklin JW and Mankin HT, Open operation in treatment of calcic aortic stenosis Circ 21:578,1960.

284

Lasers in Medicine

150. Hurley PJ, Lowe JB, Barratt-Boyes BG, Debridement valvotomy for aortic stenosis in adults: a follow-up of 76 patients, Thorax 22:314-319, 1967. 151. Ellis Jr., FH, and Kirklin JW, Aortic stenosis, Surg Clin. N. Am. 35:1029, 1955. 152. Bailey CP, Redondo-Ramirez HP, and Lazelere HB, Surgical treatment of aortic stenosis, JAMA 150:1647, 1952. 153. Isner JM et al., Laser-assisted debridement of aortic valve calcium, Am Heart J, 109:448-452, 1985. 154. Nuss RC et al., Infrared laser bone ablation, Lasers Surg Med 8:381-391, 1988. 155. Haller JD et al., Effects of 193 nm and 248 nm excimer laser on inorganic calcic mineral portion of human atherosclerotic plaque, Proc Conf Lasers and Electro-Optics (CLEO) Postdeadline papers, P2, May, 1985. 156. Haller, JD et al., Physical and chemical effects of ultraviolet excimer laser on human atherosclerotic plaque, Radiology 157: (abstract) P65, 1985. 157. Haller JD et al., Ablation of human atherosclerotic plaque by 193 nm and 248 nm wavelength nanoseconds delivered laser energy, Proc. Med & Biol Symp (ICALEO), Laser Institute of America, 49:11, 1986. 158. Haller JD et al., Physical and chemical effects of ultraviolet excimer laser radiation on human atherosclerotic plaque: therapeutic implications, Lasers Med Surg (Munich, Germany) 3:98- , 1987. 159. Spector M et al., Atherosclerotic plaque: X-ray diffraction investigation, Science 65: 711, 1969. 160. Miller FA and Wilkins CH, Infrared spectra and characteristic frequencies of inorganic ions, Analyt Chem 24:1253-1259 and 1292-1294, 1952. 161. Williamson WA et al., Lasers Surg Med 1993, 13 (4): 421-428. 162. Lilge L, Radtke W and Nishioka NS, Lasers Surg Med 1989, 9(5) 458460. 163. Isner, JM et al., Laser myoplasty for hypertrophic cardiomyopathy. Initial in vitro experience in human postmortem hearts and in vivo experience in canine model (transarterial) and human patient (intraoperative), Am J Cardiol 53:1620, 1984. 164. Morrow AG, Hypertrophic subaortic stenosis. Operative methods utilized to relieve left ventricular outow obstruction, J Thoracic Cardio Surg 76:423, 1978. 165. Bigelow WG et al., The ventriculomyotomy operation for muscular subaortic stenosis. a reappraisal, J Thoracic Cardio Surg 52:524, 1966. 166. Jeffrey DL et al., Left ventricular myotomy. Physiologic approach to surgical therapy for IHSS, Chest 80:550- , 1981. 167. Isner JM et al., Identication of photoproducts liberated by in vitro laser argon irradiation of atherosclerotic plaque, calcied cardiac valves and myocardium, Am J Cardiol, 55:1192, 1985. 168. Rashkind WJ and Miller WW, Creation of an atrial defect without thoracotomy JAMA 196:173, 1966. 169. Riemenschneider TA et al., Am Heart J, 1983. 170. Bommer WJ et al., Laser atrial septostomy, Am Heart J 106:1152-1156, 1983. 171. Bommer WJ et al., Atrial septostomy using argon laser beroptic catheter and two-dimensional echography, Circ 68: Suppl. III-87, (abstract #348), 1983. 172. Ben-Shacher G et al., Laser atrial septostomy in live canines, Circ 74:II-360, 1986. 173. Ben-Shacher G et al., Am Heart J, 1985. 174. Dos Santos JC, Sur la desobstruction des thromboses arterielles anciennes Mem Acad Chir 73:409,1947. 175. Treat MR et al., Laser endarterectomy in vivo, Circ 66: Suppl. II, (abstract #1465), 1982. 176. Treat MR et al., Effect of CO2 laser on the luminal surface of blood vessels in vivo, Lasers Surg Med, 3:247-254, 1983. 177. Fasano VA et al., Effects of laser sources (Argon, Nd:YAG,CO2) on the elastic resistance of the vessel wall:histological and physical study, Lasers Surg Med 2:45-54, 1983. 178. Fasano VA et al., First observations on carotid artery recanalization by argon laser, Laser Surg Med. (abstract #152), 1985.

Cardiovascular Applications of Lasers

285

179. Eugene J, Baribeau Y and Berns MW, Laser Endarterectomy, (Chapter 49), in Endovascular Surgery, 2nd ed, Ahn SS and Moore WS, Eds. , WB Saunders, Philadelphia, 1992, p.439-450. 180. Eugene J, Laser Endarterectomy, (Chapter 25) in Primer on Laser Angioplasty, 2nd ed., Ginsburg R and Heschwind HJ Eds., Futura, Mt. Kisco, NY, 1992; p.437-465. 181. Baribeau Y et al., Excimer laser radiation for endarterectomy of experimental atheromas, J Invest Surg 4: 247-258, 1991. 182. Eugene J et al., Initial trial of argon ion laser endarterectomy for peripheral vascular disease, Arch. Surg, 125: 1007-1011, 1990. 183. Eugene J, Ott RA and Nudelman KL, Initial clinical evaluation of carotid artery laser endarterectomy, J Vasc Surg 3:284-287, 1990. 184. Goosey UD, Zigler US Jr, Kinoshita JH, Cross-linking of lens crystallins in a photodynamic system: a process mediated by singlet oxygen, Science 208: 1278-1280, 1980. 185. Schober P et al., Laser-induced alteration of collagen substructure allows micro-surgical tissue welding, Science; 232: 1421-2, 1986. 186. Bass LS et al., Electrophoretic mobility patterns of collagen following laser welding, Proc SPIE 1422:123-7, 1991. 187. Lemole GM, Anderson RR, DeCoste S, Preliminary evaluation of collagen as a component in the thermally induced weld, Proc SPIE 1422: 116-122, 1991. 188. Kung RTV et al., Absorption characteristics at 1.9 um: effect on vascular welding, Lasers Surg Med 13:12-17, 1993. 189. Sauer JS, Hinshaw JR, A ber optic exoscope for lasertissue welding, Lasers Surg Med 6: 219, 1986. 190. Sauer JS, Hinshaw JR, McGuire KP, The rst sutureless, laser-welded, end-to-end bowel anastomosis Lasers Surg Med 9:70-73, 1989. 191. Dew DK, Review and status report on lasertissue sealing, Proc SPIE 1200:38, 1990. 192. Grubbs PE et al., Enhancement of CO2 laser microvascular anastomoses by brin glue, J Surg Res 45:112 9 1988. 193. Poppas OP et al., Laser welding in urethral surgery: improved results with a protein solder, J Urol 1988;139:415-417. 194. Chuck RS et al., Treat MR, dye-enhanced lasertissue welding, Lasers Surg Med l989: 9:47l-477. 195. Oz MC et al., Indocyanine green dye enhanced vascular welding with the near infrared diode laser, Vasc Surg 24: 564-570, 1990. 196. Oz MC et al., Treat MR, Indocyanine green dye enhanced vascular welding with the near infrared diode laser, Vasc Surg 24: 564-570, 1990. 197. Oz MC et al., Tissue soldering using indocyanine green dye enhanced brinogen with the near infrared diode laser, J Vasc Surg 11:718-725 1990. 198. Oz MC et al., Autologous brin glue from intraoperatively collected platelet-rich plasma Ann Thorac Surg 53:530-l, 1992. 199. Anderson RR and Parrish JA, Selective photothermolysis, precise microsurgery by selective absorption of pulsed radiation, Science 220:524-527, 1983. 200. White RA et al., Argon laser welded arteriovenous anastomoses, J Vasc Surg 6:447-53, 1987. 201. Oz MC et al., Urokinase modulation of weld strength in laser vascular anastomosis, Lasers Surg Med 10: 393-395, 1990. 202. Jain KK and Gorisch W, Repair of small blood vessels with the neodymium-YAG laser: a preliminary report, Surgery 85:684-688,1979. 203. Jain KK, Sutureless microvascular anastomosis using neodymium-YAG Laser J Microsurg1:436-9, 1980. 204. Yamagata S et al., Experimental nonsuture microvascular anastomosis using a soluble PVA tube and plastic adhesive, J Microsurg 1:208-215, 1979. 205. Jain KK, Sutureless extra-cranial anastomosis by laser (letter), Lancet 2:(8406):8l6-7, 1980. 206. Okada M et al., An alternative method of vascular anastomosis by laser: experimental and clinical study, Lasers Surg Med 7: 240-248, 1987.

286

Lasers in Medicine

207. Okada M et al., Experimental and clinical studies on the laser application in the cardiovascular surgery: analysis of clinical experience of 112 patients, Nippon Geka Gakkai Zasshi 90: 1589-93, 1989. 208. White UV, Laser instrumentation: a microtenaculum for lasertissue fusion Lasers Surg Med 8:433434, 1988. 209. White RA, Kopchok GE, Laser vascular tissue fusion: development, current status and future perspectives, J Clin Laser Med Surg: 47-54, 1990. 210. White RA et al., Argon laser welded arteriovenous anastomoses, J Vasc Surg 6:447-53, 1987. 211. Oz MC et al., Initial clinical experience with laser assisted solder bonding of human vascular tissue, Proc SPIE 1422: 147-51,1991. 212. Oz MC et al., Clinical experience with laser enhanced tissue soldering of vascular anastomoses, Lasers Surg Med Supp 3:74, 1991. 213. Bass LS et al., Sutureless microvascular anastomosis using the THC:YAG laser: a preliminary report, Microsurgery 10: 189-193, 1989. 214. Keane D and Ruskin JN, Circulation, 1999: 100, e59e60.

10
Lasers in Photodynamic Therapy
10.1 Introduction ....................................................................... 287
PDT Registration and Drug Plus Device Co-Dependence

10.2 Tissue Optical Properties, Photobleaching, and Light Delivery Systems................................................ 289 10.3 PHOTOFRIN PDT for Supercial Bladder Cancer......... 291 10.4 PHOTOFRIN PDT in the Treatment of Endobronchial Lung Cancer........................................................................ 296 10.5 PHOTOFRIN PDT in Gynecologic Malignancies ........... 300 10.6. PHOTOFRIN PDT in Head and Neck Cancer................ 302 10.7 PHOTOFRIN PDT in Gastrointestinal Cancer ............... 304
PHOTOFRIN PDT for Esophageal Cancer PHOTOFRIN PDT for Colorectal Cancer PHOTOFRIN PDT for Early Gastric Cancer

10.8 PHOTOFRIN PDT for Intraoperative Abdominal or Thoracic PDT ................................................................ 306 10.9 PHOTOFRIN for Intraoperative PDT in the Treatment of Intracranial Tumors....................................................... 307 10.10 PHOTOFRIN PDT in the Treatment of Ocular Cancer . 310 10.11 PHOTOFRIN PDT in the Treatment of Cutaneous and Subcutaneous Tumors ................................................ 311 10.12 New Photosensitizers for PDT .......................................... 313

Stuart L. Marcus, M.D., Ph.D.

DUSA Pharmaceuticals, Inc.

VISUDYNE (Verteporn) LEVULAN (Aminolevulinic Acid HCl, ALA) PDT

10.13 Conclusions ........................................................................ 317 References ...................................................................................... 317

10.1

Introduction

Photodynamic therapy (PDT), a subset of photochemotherapy, is a therapeutic method requiring a drug (photosensitizer), a device (light source and light delivery device), and the presence of molecular oxygen for the treatment of a variety of human clinical conditions.13 Clinical photodynamic therapy (PDT) is a two-stage process, with the rst stage consisting of the delivery of a photosensitizer, either systemically or topically. The rst photosensitizers used historically for this purpose were hematoporphyrin derivative (HpD, PhotoFRIN I) and pormer sodium (PHOTOFRIN, formerly known as dihematoporphyrin ether/esters, DHE, or PhotoFRIN II). PHOTOFRIN (QLT, Inc), is a partially puried preparation of the photoactive ingredients within HpD, with a proportionate increase in potency. The only reported reproducible side effect of drug injection is cutaneous (skin) photosensitivity, which lasts approximately 46 weeks. Following a period of time during which PHOTOFRIN is mostly cleared from a variety of tissues

0-8493-1146-2/02/$0.00+$1.50 2002 by CRC Press LLC

287

288

Lasers in Medicine

(4872 h) and retained in tumor, skin, and organs of the reticuloendothelial system (including liver and spleen), the tumor is illuminated with 630 nm wavelength light, a process that constitutes the second and nal stage in the therapy.13 PDT-induced target tissue destruction requires the presence of oxygen and is thought to be due to the production of singlet oxygen for direct cytotoxic effect; a microvascular shutdown effect is also observed that results in tumor death from oxygen starvation (ischemic necrosis).1,4 Vascular endothelium may selectively take up HpD or PHOTOFRIN within tumor tissue.5,6 Laser Doppler measurements were performed to determine blood ow velocity in normal and in tumor blood vessels in mice bearing spontaneous mammary tumors and in rats with chondrosarcoma (cartilage tumor) implants.7 PHOTOFRIN was administered to the tumor-bearing animals, followed 24 h later by light exposure. Decreases in blood ow occurred within 10 sec of light treatment and were maximal within 5 min.7 The rapidity of response suggests a thrombotic phenomenon, and in a recent report,8 thromboxane was detected systemically in rats given PDT within minutes of light exposure. Tumor selectivity in treatment is thought to occur through a combination of selective retention of PHOTOFRIN selective delivery of light, and light dose selection such that normal (non-tumor) tissue is clinically unaffected.13 PDT toxicity is local, and dependent upon the organ being irradiated. Fluorescence produced by photosensitizers has been used as a means of detecting tumors or dening tumor margins.9,10 This use of photosensitizers will not be covered in this chapter, which deals solely with therapy. This chapter will primarily discuss those patient treatments using PHOTOFRIN and HpD as prototype photosensitizers for PDT. Devices in development which may render the application of PDT easier or more user-friendly will also be discussed. Following this discussion of PDT using PHOTOFRIN or HPD, newer approved photosensitizers such as VISUDYNE (verteporn for injection), and LEVULAN (aminolevulinic acid HCl) will be briey discussed.

10.1.1 PDT Registration and Drug Plus Device Co-Dependence

Clinical drug development for anticancer drugs usually proceeds in three phases. The rst clinical experiments in man, termed Phase I studies, are primarily concerned with drug safety. In Phase I studies, the maximum tolerated dose of drug- and dose-limiting side effects are determined in a small number (approximately 2050) of human subjects. Phase II studies target specic diseases for potential therapeutic effect by the new drug, and serve to separate drugs of genuine potential from drugs that are inactive in specic cancers or that are very toxic in specic patient populations. Phase II studies may involve between 100 and 200 cancer patients, although they may be larger if many tumor types are to be screened as potential targets. Once efcacy in therapy is shown, the drug may be compared with standard therapy (if a comparator drug has been approved and is available) in large clinical trials that might involve hundreds of patients (Phase III studies). In such studies, the patient is usually randomized to the standard vs. experimental therapy. The Phase III study is the most rigorous and extensive type of scientic clinical investigation of a new drug or therapy. PDT with HpD or PHOTOFRIN has been reported effective in publications citing thousands of patients treated.3 The unique characteristics of PDT have made the registration process complex for various reasons. In countries such as the U.S., a combined drug and device (in this case specic ber optic entities) product license ling application must be prepared, and may require review by a combination of agencies that normally deal with reviewing either a drug or a device. It is important to note that, despite the large number of anecdotal reports on the clinical success of PDT, no matter how large a patient series by an original clinical investigator is published, such publications are often viewed as multiple clinical anecdotes. Such anecdotal reports are usually not acceptable as data for registration purposes unless objective evidence of patient condition, treatment methods, and response as well as information on the safety of the investigational agent is provided. Three photosensitizers have now received marketing approval for the treatment of diseases in the United States. In 1995, PHOTOFRIN was the rst PDT agent approved for the palliation of malignant

Lasers in Photodynamic Therapy

289

dysphagia caused by esophageal cancer. Since then, it has also been approved for the treatment of recurrent supercial endobronchial lung carcinoma. In 1999, an endogenous photosensitizer, LEVULAN (aminolevulinic acid HCl, DUSA Pharmaceuticals Inc, marketed by Schering AG) became the rst PDT agent approved for topical use, together with blue light, for the treatment of actinic keratoses (precancerous skin lesions) in the United States. Most recently, in 2000, VISUDYNE (verteporn, QLT Inc) was approved in the United States as the rst drug therapy for age-related macular degeneration, a common cause of blindness.

10.2 Tissue Optical Properties, Photobleaching, and Light Delivery Systems

Upon the absorption of light by a photosensitizer, energy is passed (in a nonradiative process) to an oxygen molecule, increasing the energy state of the oxygen molecule from a triplet conguration to the excited singlet state. The singlet oxygen may react with many cellular molecules, and interaction with the plasma and mitochondrial membranes is currently thought to be the method of cytotoxic action of the photodynamic effect.2 The optical properties of tissue, for light propagation at the 600 to 1100 nm wavelength range, are dominated by scattering effects (see Ref. 2 for review). Thus, the propagation of light in tissue is best described by diffusion theory using the scattering coefcient, , and the total attenuation coefcient , which includes tissue absorption as well as scattering. It is to be noted that both and are wavelength dependent, and that absorption varies signicantly among various tissues that contain differing chromophores (mainly hemoglobin and melanin).2 One of the ndings that will be notable throughout the clinical section of this chapter is the selective toxicity of PDT for tumor tissue. While this can be explained, in part, by selective retention of PHOTOFRIN within the tumor, a phenomenon termed photobleaching appears to also provide an effective amplication of tumor selectivity. Photobleaching is the permanent photochemical loss of the chromophore used in PDT. If the photosensitizer used in PDT were completely stable and could therefore pass energy to oxygen molecules as long as light was present, the maximum depth of tumor necrosis that could be achieved without damaging normal tissues would be determined entirely by the ratio of drug in tumor to that in surrounding (and also irradiated) normal tissues. If the concentration of drug in normal skin surrounding a tumor is one third the concentration in tumor, and the photobleaching rate is the same for both tissues, then the maximum photodynamic dose in normal skin is one third that of the tumor and no amount of light will damage the normal skin but the tumor will undergo necrosis. Dougherty and Marcus2 have shown examples of the clinical application of this concept in the treatment of skin tumors. The wavelength recommended for activation of PHOTOFRIN by light is 630 3 nm using monochromatic, coherent light (laser light), although light of shorter wavelengths can also activate the drug. The 630 3 nm wavelength was chosen because it is the longest wavelength capable of activating PHOTOFRIN and, therefore, has the deepest tissue penetration (3 mm to 8 mm)3,4 for use of this photosensitizer. Non-laser light sources have been used to activate PHOTOFRIN but lasers are preferred for applications requiring endoscopic delivery of light into the body because the properties of laser light provide efcient coupling into small optical bers. The ber optic tips used for the delivery of laser light in preregistration studies using PDT with PHOTOFRIN in North America and Europe are shown in Figure 10.1. The microlens-tipped ber optic provides a uniform eld of light for front surface illumination. Cylindrical-diffuser-tipped ber optics provide approximately 95% of diffused light in a uniform, cylindrical distribution perpendicular to the axis of the ber tip; these bers can be used intraluminally (placed within the lumen of a cylinder-shaped organ part) for delivery of light to the wall of a hollow viscus such as the bronchus or esophagus when it is not possible to point a microlens-tipped ber directly at a lesion. Five lengths of cylindrical tips are provided, from 0.5 to 2.5 cm, in increasing increments of 0.5 cm. A tumor longer than a cylindrical tip

290

Lasers in Medicine

FIGURE 10.1 Examples of ber optics used to deliver laser light for photodynamic therapy. (top to bottom) Spherical light diffuser for isotropic light delivery to spherical or near spherical body cavities; microlens-tipped ber for uniform front surface illumination; Cylinder diffusers for delivery of light intraluminally or interstitially; 0.5-cm length pointed cylinder diffuser; 1.0 cm length pointed cylinder diffuser; 2.5 cm length pointed cylinder diffuser. Reprinted from Ref. 3 with permission of the publisher.

may be treated in non-overlapping segments using different sized ber optics in sequence. The cylindrical diffuser tipped bers are prepared with pointed ends, so that they can also be inserted directly into the tumor substance for efcient, tumor-specic interstitial light delivery, particularly for thick lesions. The spherical diffuser-tipped ber optic provides an isotropic, spherical distribution of light and is used for PDT of the entire urinary bladder, but could be applied to any indication where a spherical or nearspherical distribution of light is required. Modications of these general ber optic delivery systems will be discussed under the specic category of malignancy for which they have been developed. Lasers necessary for producing light to be delivered via ber optic devices must be considered in plans for product license applications. In the Phase III studies carried out and sponsored by QLT, Inc. within North America for PHOTOFRIN the only laser type used is the argon ion laser-pumped tunable dye laser (APDL). The continuous-wave output APDL may be contrasted with pulsed output lasers such as the gold vapor laser and the copper vapor laser-pumped dye laser.11 These lasers have pulsed outputs in the 10- to 100-kW range, although power is usually presented as average output power and they have been used in a fashion identical to the APDL. The excimer laser-pumped dye laser system has been used extensively in Japan, and this was the laser system used for some Phase III registration studies in Japan.

Lasers in Photodynamic Therapy

291

The excimer-pumped dye laser is an extremely high-power pulsed laser system in which peak power may approach the megawatt range. There have been no published controlled studies comparing clinical PDT efcacy and safety results in the APDL system with pulsed-output laser systems. By comparing the safety and efcacy of APDL lasers with copper vapor laser-pumped dye laser systems in otherwise identical clinical trials for lung and bladder cancer in different European centers, data will be obtained that can be used to determine whether such laser systems are indeed equivalent. Solid state laser systems such as the frequency-doubled Nd:YAG laser-pumped dye systems or visible diode lasers are now being used in PDT systems. Advancements in diode technology have now allowed the production of relatively inexpensive lasers emitting in the wavelength range of 630700 nm at powers currently considered therapeutically useful, in the 1- to 4-watt output range. The availability and applicability of new diode lasers to PDT is discussed in later sections of this chapter. Light delivery systems as they currently exist for PDT, although adequate, may be modied in the future to allow in situ dosimetry of delivered light. Because of the geometry of certain organs or the size of areas to receive light for photosensitizer activation, current methods of uence determination are unquestionably suboptimal. Devices that have been developed to provide such in situ methods of dosimetry have been developed for PDT of the entire urinary bladder (see next section) and for PDT of intrathoracic and abdominal spaces. The potential advantage of such systems would be to prevent underor overtreatment of target organs with activating light. Discussions of such needs and those applications of PDT in which in situ dosimetry might be most appropriately be utilized will be discussed in sections dealing with specic cancer treatment. In each of the following sections, the characteristics of each cancer (incidence, prognosis), together with current standard methods of therapy and results of such therapy, will be briey discussed. The current status of PDT in the treatment of each particular form of cancer will be critically reviewed as well as any technical advancements that might increase the potential for PDT effectiveness.

10.3 PHOTOFRIN PDT for Supercial Bladder Cancer

Supercial bladder cancer will be diagnosed in approximately 50,000 patients in the United States during 2001. Such disease can present as papillary tumors (tumors presenting as small nipplelike or frond-like processes) or as carcinoma in situ (CIS). The standard of therapy for supercial papillary tumor treatment is surgical excision performed via endoscopy through the urethra (a transurethral resection of the bladder tumor or TUR-BT). Because the penetration of 630 nm light at treatment power can approach 5 mm in certain tissues, PDT is theorized to be able to eradicate most, if not all, supercial bladder cancer, which can extend into the mucosa or submucosal layers, but not into the muscular layers of the urinary bladder. Nseyo et al.12 have recently reviewed the use of PDT in this disease entity, and the results of clinical trials published in the literature1326 are summarized in Table 10.1. The reports each include between ve and 34 patients, some of whom were treated for multiple lesions. Whole bladder PDT, in which light is delivered via a spherical diffuser-tipped ber optic centered within a uid-distended bladder, is generally used for treatment of CIS or for the prevention of tumor recurrence following papillary tumor removal. Focal therapy using microlens-tipped ber optics to deliver light to papillary tumors has also been used. Partial or complete removal of papillary tumors occurred in 70100% of sites treated (Table 10.1). Although the incidence of side effects in bladder cancer PDT is not always reported, there appears to be a general post-PDT syndrome characterized by irritative bladder symptoms of urinary frequency, urgency and dysuria (burning feeling on urination) that is transient but that may vary in duration, from a few days to several weeks.12 Irreversible bladder contraction has been reported occasionally. Such results appear to be due to high light dosing, bladder overdistension, or vigorous treatment of patients already having brosis (scarring) as the result of prior radiation therapy.12,25 In those patients who received focal tumor illumination, the severe toxicities reported by some authors may be due to lack of correcting for backscattered light reaching the entire bladder, which could have raised the total light dose to the whole bladder from 50% to 300%, depending upon the number of focal lesions treated. Alternatively, the drug and light doses that might prove optimal in terms of efcacy and

292

TABLE 10.1 Summary of Published Results with PDT in Bladder Cancer

Therapy Drug Dose (mg/kg) PDT 150 J/cm Flat tip with microlens Bulb diffuser NS 150 to 250 mL 5 150 to 300 mW/cm2 100 to 200 J/cm2 200 mL NS at tip 1 0 0 0 0 0 2 2 0 0 -* 1 1 I 15 12 0 0 0 0 8 2 CR PR NR RE HPD 2.5 Patient Procedures Follow-up (months) 6 to 32 (mean 7) Patient Response (lesions

Reference

Patients/ Lesions

Description of Disease

Benson [13,14]

27/31

15-focal CIS 12-diffuse CIS 2-T2 2-Ta HPD 4 to 5 25 to 45 J/cm2 3 or 48 h HPD 5 HPD 2 to 3.2

Ha et al. [15]

Hisazumi et al. [16,17]

9/48

Transitional cell Ta-T1 <1 cm 12 cm 2-3 cm >3 cm PHOTOFRIN 2 1 2 10 to 69 J/cm2 Bulb diffuser; water or NS; lling press 30-60 cm H2O NS150 MI Flat tip; Bulb diff. Microlens or Cylinder 22 4 6 0 0 2 2 0 4 3 5 0 -

Nseyo and Dougherty [18]

10/13

4 TIs

2 to 12

16 HPD 2.5 PHOTOFRIN{ 2 16 to 200 J/cm2 120 to 260 J/cm2

PHOTOFRIN 2

< 20 J/cm2

Shumaker and Hetzel [19] Tsuchiya et al. [20]

Nseyo et al. [21,22]

8 G HI 23

1 Ta 6 T1 2 T2 12TIs 4Ta Ta-T2

7 2 8

0 2 0

6 to 24 6 to 15 6 to 18 2 to >12 2 1

Prout et al. [23,24]

20/50

6 TIs 8 Ta 8 T1 3 T2 1 T3 3 TIs PHOTOFRIN 2

5.5 to 10J/cm2 100200J/cm2

0 0 Bulb diff. Microlens

1 3 2 1 0 3 12

1 2 5 2 1 0 25

0 13 3

Lasers in Medicine

50 Ta or T1

Harty et al. [25]

PHOTOFRIN 2

Microlens Bulb Diffuser

3 1 1

0 0 0

0 1 1

1 0 0

Dugan et al. [26]

24

3 Ta 2 T1 2 TIs 21 Ta,3T1 PHOTOFRIN 2 after TUR spherical bulb diffuser Not reached for PDT arm 12 patients observed Median time to recurrence 93 days observation arm

Lasers in Photodynamic Therapy

100 J/cm2 focal +25 J/cm2 whole bladder 12 patients 15 J/cm2 whole bladder

CIS = carcinoma in situ NR = no response CR = complete response NS = normal saline PR = partial response RE = recurrence TIs, Ta, R1, T2 = clinical stage of bladder cancer from CIS to supercial muscle invasion

* Data not provided.

293

294

Lasers in Medicine

safety may not have been deduced during clinical trials. None of the studies listed were true Phase I studies, in that careful stepwise elevation of light dose (starting at a low dose of photoensitizer) was not attempted. Therefore, the clinically optimal dose for drug and light has really never been determined in this indication. The severity of toxicities observed in PDT may also vary according to the amount of bladder mucosa involved with disease, as well as with the position of the light delivery device. Centering of the intravesical light delivery device within the bladder must be accomplished to avoid undertreatment or overtreatment of differing mucosal areas. In Phase III protocols (see below), centering of spherical diffuser-tipped ber optics was specied by both sounding and ultrasound-directed methods. Two devices have been redeveloped, each for the dual purpose of centering the ber optic and determining in situ light dosimetry.27,28 The device described by Marynissen et al.27 consists of a modied cystoscope and three translucent nylon catheters that are unfolded inside the bladder and into each of which is inserted a dosimetry probe (Figure 10.2). As bladder shape is not normally spherical, equalizing the light output reaching each probe provides a method of centering the light source as far as can be possible. In situ dosimetry also provides a method of determining real-time changes in the light reaching the bladder wall caused by variation in laser output or changes in the transparency of the intravesical uid due to bleeding or mucosal sloughing. However, this device requires a modied cystoscope and considerable set-up time, as a) essential preforming of the nylon catheters is accomplished outside the bladder after determining the neck-to-dome distance and b) light must be checked on three devices as the detectors are moved through the catheters across the entire bladder wall. Nseyo, et al.28 have obviated the need for multiple detectors for ber optic tip centering by using an intravesical laser catheter (IVLC) delivery system to convert a bladder into as near a spherical organ as possible (Figure 10.3). The IVLC consists of a translucent balloon that is inated with uid to produce spherical distension of the bladder; a central lumen within the catheter automatically centers the ber optic tip. A light sensor on the IVLC monitors light reaching the bladder wall and displays real-time irradiance (mW/cm2) as well as accumulated light energy (J/cm2) and can be used to automatically monitor and adjust treatment time based on changes in light uence. Outlet ports in the device allow removal of urine produced during the procedure, thus keeping the bladder distended to a constant volume, and provide a means of irrigation to remove blood or other light-interfering secretions without altering bladder volume. Although devices such as the IVLC appear to increase ease of whole-bladder PDT, the effect of such devices on the safety of whole bladder PDT awaits results of their testing in clinical trials. A phase III controlled clinical trial of whole bladder PDT for the prophylaxis of recurrent supercial bladder cancers (papillary type) has been carried out.26 After transurethral resection (TUR) of bladder tumors (up to stage T1G3), patients were randomized to either a single course of PDT using PHOTOFRIN and a whole bladder light dose of 15 J/cm2 or observation. Follow-up cytologies and cystoscopies were rigorously carried out every 3 months. An unscheduled interim analysis performed after accrual of 24 patients well matched in demography after a median follow-up of 116 days revealed that the median time to recurrence for the observation group was 93 days but had not been reached for patients receiving PDT (p <0.001).26 These results are encouraging, for no agent has been approved for prophylaxis of bladder cancer and no agent appears to provide efcacy similar to PDT with PHOTOFRIN following a single intravesical treatment. These data26 formed part of a product license package that resulted in the approval of PHOTOFRIN for the prophylaxis of supercial bladder cancer by Canadas board of health in 1993. Long-term prophylaxis of tumor recurrence in PDT may be the result of macrophage and lymphocyte stimulation within the bladder. Preliminary studies have shown that cytokines such as interleukin-1 and tumor necrosis factor appear in the bladder following PDT (but not TUR or infection) and may be detected in urine as long as 2 months after PDT even though the patient is asymptomatic.28 CIS that is refractory to at least two courses of intravesical (applied directly to the bladder) chemotherapy or immunotherapy has a strong potential for life-threatening muscle or prostate invasion and subsequent or concurrent metastasis.3,12 Data from six studies reported from 1983 to 1988 have been

Lasers in Photodynamic Therapy

295

FIGURE 10.2 Device developed by Dr. W. Star for centering of ber optic within the urinary bladder and for in situ dosimetry, 27 (a) dosimetry unit. (b) Cystoscope with telescope inserted via dosimetry unit for inspection of unfolded catheters. (c) Telescope replaced by light source for positioning and irradiation (inset = magnied view of light detector in place within catheter). Figure originally published in. Ref. 27 and reprinted with permission of The Journal of Urology.

FIGURE 10.3 Line drawing of intravesical laser catheter described by Nseyo et al.28 8 = catheter outlet port. Reprinted from. Ref. 28 with permission from The Journal of Urology.

296

Lasers in Medicine

reviewed (Table 10.1).3 In the initial studies of 47 patients with refractory CIS reported in the literature, 4647 (97.8%) were reported to have a complete response (negative cytology or biopsy) with follow-ups from a median of 3 months to 12 months. More recently, the results of PDT with PHOTOFRIN for refractory CIS have been reviewed from articles in which 215 patients received from 1.5 to 2 mg/kg PHOTOFRIN via the intravenous route, with an overall response rate of 66%. Estimated time to recurrence in the few large studies with long-term follow-up ranged from 37 to 84 months.29

10.4 PHOTOFRIN PDT in the Treatment of Endobronchial Lung Cancer

Non-small-cell lung cancer, which occurs in approximately 120,000 patients in the United States annually, often produces symptoms arising from that portion of the tumor present within the airway (endobronchial or endotracheal component). Symptoms may present as hemoptysis (coughing up of blood), cough, or dyspnea (shortness of breath), the latter caused by direct obstruction of the airway by the tumor within the airway (endobronchial obstruction) or by compression of the airway by tumor outside of the airway (extrinsic obstruction). Symptoms caused by endobronchial tumor obstruction are currently treated with the Nd:YAG laser. Because the Nd:YAG laser produces a nonspecic thermal laser beam to vaporize or coagulate tumor tissue, removal of tumor to the bronchial wall would risk perforation. Therefore, only part of the tumor can be removed using this therapy, although it can be sufcient to improve ventilation. Patients with recurrent endobronchial lung cancers who, because of concurrent physical conditions such as emphysema or insufcient lung reserve, cannot be treated at such an early stage with the ND:YAG laser. However, because of the selective nature of PDT, such patients theoretically can achieve complete endobronchial tumor removal (complete response or CR) using PDT. Since the early 1980s, more than 500 patients have been reported in the literature with endobronchial lung cancer treated with PDT using either HPD or PHOTOFRIN The results of various investigators3043 are remarkably consistent in that complete and partial response rates ranged from 70% to 100% (Table 10.2). Best results were obtained with early stage lung cancers. 33,36,38,44 HPD was used at a dose of 2.0 to 5.0 mg/kg intravenously and PHOTOFRIN at a dose of 2.0 mg/kg intravenously. Laser light doses ranged from 20 to 500 J/cm2, with power intensity at the ber tips ranging from 20800 mW. One is struck by the wide range of light doses with which at least a partial response can be achieved. However, as no standardized method of power measurement was practiced among the various institutions, the consistency of response in the presence of widely varying light doses must be interpreted with caution. It is important to note that, in certain studies, patients in very poor condition because of respiratory distress due to endobronchial obstruction were treated with PDT.45,46 None of the patients died as a result of therapy and 80% improved clinically, suggesting that the modality might be safe to use in patients too sick to be considered for the usual clinical trial population. In the studies in which disease cure was the goal of therapy, follow-up of response ranged from 3 days to 41 months, with the average duration of follow-up approximately 12 months. The rst patients with early disease who were inoperable have now been reported surviving more than 5 years with no other treatment but PDT.47,48 When symptomatic improvement or improved operability was the therapeutic endpoint, a positive response was noted in 106 of 111 patients following PDT treatment44-52 (Table 10.3). All responses are ideally characterized on the basis of a local (i.e., endobronchial site of treatment) response, with a complete response (CR) dened as the absence of bronchoscopically visible tumor. Partial response (PR) is dened as a decrease of 50% in intrinsic obstruction for the tumor site treated. Stable disease (SD) is dened as decreases or increases in intrinsic obstruction insufcient to qualify as either a PR or progressive disease. Progressive disease (PD) is dened as an increase in intrinsic obstruction 30% attributed to endobronchial tumor growth as compared with the percent intrinsic obstruction at the time of best response.

TABLE 10.2 Summary of Published Clinical Studies Using PDT in Lung Cancer
Follow-up Drug (mg/kg) HPD 2.5 to 5.0 S/1 8-34 6 7 6 9 Method (mos) CR PR NR Died Side Effects Phototoxicity, cough obstruction Response

Reference

Patients/Lesions

Condiction (no. patients)

Cortese and Kinsey [31,32]

19/20 SC 17

Resected or inoperable

Lasers in Photodynamic Therapy

Edell and Cortese [33]

Resected or inoperable

HPD 2.0 to 5.0

6 days to 56 mos.

14

15

11

25

1_ burn Hemoptysis Obstruction

Hayata et al. [31]

HPD 2.5 to 4.0

1 0 0 1 2 1 8 2 2 0 3 6 0 0 0

0 1 0 8 6 0

0 0 0 0 0 0

0 2 4

Phototoxicity 1_ burn

Hayata et al. [31,36] 13-41 7-30 4-41 HPD 2.0 to 5.0 S/I

38/40 SC 37 SCC 1 Cylindroma 1 Fibrosarcoma 1 14/20 SC 6 AC 2 LC 3 SCC 2 metaplasia 21 SC 20 LC 1 HPD S/I

Kato et al. [37]

Kato [38]

Phototoxicity 1_ burn possible PDT complications 36-phototoxicity 2 obstr. pneumonia. 2 bronchial stula 11/23 cleared obstruction 6/9 pts operable after PDT 3 1 0 1 0 2 3 0 3 8 15 10 0 11 0 0 0 0 0 0 1

Keller et al. [39]

43/54 SC 29 AC 5 LC 4 SCC 2 metaplasia 2 other 1 15 HPD 2.0 to 3.0 PHOTOFRIN{ 1.5 to 2.0 S/I 17

Early stage (1) Stage I (1) Stage II (0) Stage III (8) Stage IV (3) metaplasia (1) Early stage unresected (8) resected (5) Stage I (8) 23 obstructed early stage (3) Stage I (4) Stage II (8) Stage III (16) Stage IV (10) met (2) Obstructed

continued

297

298

TABLE 10.2 (CONTINUED)

Follow-up Drug (mg/kg) HPD 5.0 S 3-7 3 17 4 1 Method (mos) CR PR NR Died Side Effects Fever, hemoptysis, progression of in 8 Response

Summary of Published Clinical Studies Using PDT in Lung Cancer

Reference

Patients/Lesions

Condiction (no. patients)

Li et al.[40]

All advanced

Vincent et al. [41]

Vincent and Dougherty [42]

21/24 SC 18 AC 2 SCC 1 17 SC 11 AC 5 HPD 2.5 to 3.0 S/I 5-210 days (mean 40) 0 13 4 8

Recurrent, advanced obstructed

Fever, pneumonia, hemoptysis candidiasis hypersecretion

LC 1 HPD 5.0 SC = squamous cell AC = adenocarcinoma LC = large cell SCC = small cell S = supercial I = interstitial 1-6 7 2 3

Karanov et al. [43]

12

2 Stage II 10 Stage III

CR = complete response PR = partial response NR = no response

* Data not provided

TABLE 10.3 Summary of Published Studies Using PDT for Palliation or Improved Operability of Lung Cancer
Follow-up Drug (mg/kg) HPD 3.0 S/I Method (mos) to 20 cleared or improved Results 80/81 6 other Died 38 cancer Side Effects 5 pulmonary hemorrhage 3 staph pneumonia

Reference

Patients/ Lesions

Condition (no. patient)

Balchum and Doiron [44]

81/135 SC 56 AC 5 LC 4 SCC 4 other 5 metastatic 7

Stable, good prognosis; 72 obstructed

Lasers in Medicine

Hugh-Jones and Gardner [46]

15/15 SC 15

All advanced symptomatic

S/I

to 24 CR-3 PR-9 80% Clinically improved *11/15 improved 3 cancer 4 other

9 cancer 2 obstruction

2 burns

Kato et al. [49]

15 SC 11 AC 2 LC 2

HPD 3.0-4.9 DHE 1.9-2.3 HPD 2.5 to 5.0 S/I

Lasers in Photodynamic Therapy

Lam et al. [50]

5/5 *

Preoperative therapy Stage I (5) Stage II (2) Stage III (7) Stage IV (1) All symptomatic DHE 2.0 followed by radiation therapy HPD 3.0 PHOTOFRIN{ 2.0 PHOTOFRIN{ 2.5 S/I 1-10 CR-2 PR-11 MR-2 6/10 clinically improved 16 10/10 clinically improved 6-cancer S/I 0.534 CR-12 PR-14 61% clinically improved 14 cancer 2 other 1 pneumonia 2 burns 1 obstruction 1 hemoptysis 2 obstruction 1 pneumothorax 10-cancer 1 mild anasarca 2 burns S/I 3 to 10 = 100% All clinically improved CR + PR 1 cancer

McCaughan et al. [45]

Pass et al. [51]

18/26 SC 10 AC 3 LC 1 other 4 10/15 Lung*4

Obstructed symptomatic Stage III (15) CIS (1) metastatic (2) Obstructed 6 symptomatic

LoCicero et al. [52]

Metastatic 11 10 PHOTOFRIN{ 2.0 I = interstitial CIS = carcinoma in situ SC = squamous cell S/I

Obstructed all symptomatic

CR = complete response PR = partial response MR = minor response S = supercial

AC = adenocarcinoma LC = large cell SCC = small cell

* Data not provided

299

300

Lasers in Medicine

Palliative randomized, controlled Phase III studies were carried out in North America and Europe to determine the safety and efcacy of PDT in the treatment of endobronchial lung cancer. Patients (generally at an advanced stage of lung cancer) presenting with symptoms of dyspnea, hemoptysis or cough referable to a specic endobronchial lesion were randomized to receive either Nd:YAG laser thermal ablation therapy, or to PDT with PHOTOFRIN. The majority of patients receiving PDT had light delivered via interstitial light treatment, which consists of the direct insertion of a ber optic tip into the tumor mass and includes point insertion (PI, using a at-cut ber tip), point insertion via hollow bronchoscopic needle for tumors too hard to be penetrated directly by ber (PIN), and insertion of a cylindrical diffusing tip (see Figure 10.1), (cylindrical insertion, CI). Intraluminal light treatment consists of the supercial (or surface-only) photoradiation of tumors, carried out via cylindrical (CS) microlens-tipped ber (Figure 10.1), or forward surface (FS) illumination using a at-cleaved ber tip. A total of 211 patients were randomized (102 to PDT, 109 to Nd:YAG) in these studies and, after the rst month of follow-up, objective tumor response was observed in 60% of the PDT-treated patients vs. 41% of the Nd:YAGtreated patients. Atelectasis improvement in the same follow-up period was observed in 35% of the PDTtreated patients vs. 20% of the Nd:YAG-treated patients. Clinically signicant symptom improvements were observed that were equivalent in both treatment arms for symptoms of dyspnea and cough; at month 1 or later, improvement in hemoptysis was seen in 79% of the PDT-treated patients vs. 35% of the Nd:YAG-treated patients. These results helped support the marketing approval for PHOTOFRIN PDT for palliation of symptoms caused by endobronchial obstruction in the United States in 1995.53 On the basis of three noncomparative studies that accrued a total of 62 inoperable patients, PHOTOFRIN PDT has also been approved for the treatment of microinvasive endobronchial non-small-cell lung cancer in patients for whom surgery and radiotherapy are not indicated. For these patients, the biopsy-proven CR rate 3 months after treatment was 50%, median survival was 2.9 years and median time to tumor recurrence was 2.7 years; disease-specic survival was 4.1 years.53 It appears that PDT has the potential to palliate patients with endobronchial obstruction in a manner similar to the Nd:YAG laser. However, because of the necessity of waiting 2 to 3 days for selective retention of PHOTOFRIN patients with acute onset of severe shortness of breath at rest may best be served by immediate Nd:YAG laser therapy, followed perhaps by PDT to attempt complete removal of the endobronchial obstruction. It is probable that patients with recurrent lung cancer of a supercial nature can have excellent long-term results using PDT alone, if surgery is not indicated or not possible due to their physical condition.

10.5 PHOTOFRIN PDT in Gynecologic Malignancies

Gynecologic malignancies include cancers of the vulva, vagina and cervix, all of which are accessible to endoscopic or surface treatment with light for PDT. Vulvar and vaginal cancer together account for a total of 5% of female malignancies. Preferred treatment for supercial cancer of both areas (disease that does not penetrate deeply into tissue and does not involve local lymph nodes) is surgery. Although invasive cancer of the cervix affects approximately 15,000 women annually, supercial cervical cancer (carcinoma in situ) is annually diagnosed in approximately 50,000 women and can be treated by conization (removing a cone-shaped area of the cervix with the cervical opening at the center of the large surface of the cone). Cryotherapy and CO2 laser surgery have been used in selected patients with carcinoma in situ of the cervix. Complications of such therapy include scarring, unaccepable cosmetic effects, and, in the case of conization, narrowing of the cervical opening affecting fertility or delivery. Clinical reports describing the use of PDT with HPD or PHOTOFRIN for gynecologic cancer are summarized in Table 10.4. Patients had either primary gynecologic cancer or metastatic lesions from other primary tumor sites. All patients were treated with HPD with a dose ranging from 2.3 to 5.0 mg/kg or PHOTOFRIN (2.0 mg/kg) and patients were often retreated if the initial response was incomplete. One researcher administered 20 or more treatments to two patients with local endometrial and ovarian cancer, and used a Kodak projector lamp as the light source rather than an argon-pumped dye laser. 59

Lasers in Photodynamic Therapy

301

TABLE 10.4 Summary of Published Results with PDT in Gynecologic Malignancies

Reference Corti et al. [56] Patients (sites) 11 cervix, + 2 uterus, + 2 rectal 1 metastasis; ovary 1 CIS vulva 1 cervix, recurrence 3 cervix 1 vaginal; papillary 1 endometrium 1 ovary 21 (26) - All recurrent 7 perineum, vulva, abdomen (skin) 11 vagina + cervix 7 recurrent (45 sites) Perianus vagina vulva 2 primary, vagina 1 vulva 2 metastatic Recurrent, vagina 1 endometrium 3 cervix 1 vagina 1 Bertholin's gland 3 Intraepithelial neoplasia HPD Dose (mg/kg) HPD 2.5 Response CR PR NR 8 6 1 PR CR NR PR CR PR PR CR-7 CR-2 PR-2 NR-7 CR 34 PR 8 NR 3 Follow-up (months) 3 to 25 Comments 7/8 patients with supercial disease had a CR

Forbes et al. [57] Dahlman et al. [58] Ha et al. [15] Kennedy, [58] Lele et al, [59]

HPD 5 HPD 3 HPD 5 HPD 2.5 Kodak light PHOTOFRIN{ 2

4 * 1 to 28 Both treated multiple times for recurrence Pain at treatment site in all patients requiring analgesics

Lobraico et al. [61]

PHOTOFRIN {1.5 to 2.0

1 second degree burn

McCaughan et al. [62] Rettenmeier et al. [63]

HPD 2.3 to 3 HPD 3.0

0CR-2 PR-1 CR-1, PR-1 3 CR-2 PR-3 NR-1 rst degree burns

HPD 5%

Soma and Nutahara [63]

5 vagina 5 CIS cervix 3 metastatic; vagina 5 vagina recurrent

HPD 2.5

PR-1 NR-1 No FU-1 CR PR NR 4 1 0 3 2 0 2 0 0 8 to 24

-Photosensitive

Ward et al [64]

HPD 5

CR-2 PR-3

0.2 to 12 (mean 4.5)

CIS = carcinoma in situ CR = complete response PR = partial response * Data not provided.

NR = no response FU = follow-up

Table 10.4 shows results of a total of 100 women who have received PDT with PHOTOFRIN for treatment of gynecologic cancers. Of those patients for whom a response is given, a complete response was reported in 26 patients, partial response in another 24 patients, and no response in four patients; four patients were not available for subsequent analysis. The duration of follow-up varied considerably between studies. Edema, erythema, and necrosis accompanied by various levels of pain occurred in treated sites 12 to 48 hours post-treatment. PDT appears effective for the treatment of gynecological cancer, and is capable of achieving complete eradication of visible tumors in patients with supercial disease. Improvements in and standardization

302

Lasers in Medicine

of light delivery and dosimetry to irregularly shaped anatomic areas such as the vagina, vulva and areas of junction between cervical and vaginal mucosa are required for optimization of this form of therapy.

10.6 PHOTOFRIN PDT in Head and Neck Cancer

More than 40,000 cases of head and neck cancer (excluding skin cancer) are diagnosed annually in the United States. Surgery and radiation therapy are the only curative treatments for cancer arising in the head and neck. Cancers involving the lip, oral cavity (mouth, tongue, cheek), pharynx and larynx are often treated by radical surgery with reconstruction wherever possible. In such patients, recurrence is often followed by additional mutilating surgery, or surgery may not be possible. Radiation therapy is also often given following surgical excision of tumor and, as maximal radiation therapy is often given initially, follow-up radiation cannot be given on recurrence. Data are shown in Table 10.5 for 198 patients treated with PDT with PHOTOFRIN or HPD for head and neck cancer.6678 Patients who had a mixture of primary, recurrent, and metastatic lesions were included, with the histologic type of cancer unspecied in four reports. Cancerous lesions involved most structures of the head and neck region. Patients were treated with HPD 2.0 to 5.0 mg/kg, or PHOTOFRIN 1.5 to 2.0 mg/kg. Many patients were retreated with PDT if the response was inadequate after the initial treatment, and two to four treatments were used in each of three patients in one study.68 In most studies, complete response was dened as absence of tumor; partial response was dened as a reduction in tumor mass; and no response was dened as stable disease, disease progression, or patients not available for analysis. Follow-up of response was continued for up to 34 months; however, most patients were followed for less than one year after PDT. Grossweiner et al.69 reported PDT treatment with PHOTOFRIN of head and neck squamous cell cancers in nine patients (11 sites) with primary and one (one site) with recurrent, advanced disease. Laser light treatment was delivered 24 hours following PHOTOFRIN injection. This paper is unique in that theoretical dosimetry calculations were attempted using dimensions to account for the irregularity of the tumor surface, and compared with the actual light uence delivered to the sites. A CR was obtained in 10 sites; the single partial response was obtained in a patient with an advanced stage tumor that was thicker and relatively inaccessible to PDT in the nasopharynx. The site reported as no response had been treated with lidocaine containing epinephrine (a vasoconstrictor) just prior to treatment with laser light, which might have blocked oxygen delivery to the tumor. Optimal light delivery consisted of a combination of surface illumination (using a microlens-tipped ber) and interstitial treatment with cylindrical diffuser insertion. Surface illumination was given rst, as interstitial cylindrical diffuser placement can cause bleeding that can affect absorbed light energy. A 1 mm change in maximum tumor depth was found to have a signicant effect on calculated light dose. Dosimetry calculations thus varied up to a factor of 3 with respect to actual light uence delivered to tumor tissue. A recent review of PDT for the treatment of oral, laryngeal, and head and neck cancers analyzed results obtained for > 600 patients.80 Twenty-ve patients with CIS and T1 squamous cell carcinomas of the true vocal cord treated with PDT for cure obtained a CR after a single PDT treatment, and the CR rate to a 79-month follow-up period remains 95%. A review of 217 patients with early squamous cell carcinomas of the head and neck treated with PDT in the literature demonstrated an 89.5% CR rate. The varying geometries and difculty with determining true dosimetry in these cases render generalization difcult. However, it appears that the role of PDT in head and neck cancer may lie in eradicating the supercial, early recurrence and condemned mucosal (pre-cancerous appearing tissue) appearance of this disease, and thus this treatment modality has the potential to save patients

Lasers in Photodynamic Therapy

303

TABLE 10.5 Summary of Published Results with PDT in Head and Neck Cancer
Reference Cai et al. [67] Patients (sites) 26 primary or recurrent 2 adenoid 3 early 8 advanced 9 recurrent 13 metastatic Drug Dose (mg/kg) HPD 5 CR 11 PR 8 NR 7 Response Comments Follow-up to 34 months; better responses in untreated patients Follow-up Follow-up to 18 months; all early stage had CR 1 burn; 3 ulcer or necrosis; 1 fatal hemorrhage; follow-up to 16 months 2 laryngectomy; 3 with tumor; follow-up <(29 squamous) HPD combined with radiation provided enhanced response Only 4 recurrences; 4 developed nodal metastases 3 1 CRs duration 6 Mos. to 48 Mos. Duration of CRs 3-7 months. Duration of response 12 to 48 months Duration of response 840 Mos. Duration of response 22 and 36 months PHOTOFRIN{ 2 2 7 1 Duration of response 12 to 48 Mos. 1 Duration of response 5-51 Mos.

Keller et al. [39]

Schuller et al. [70]

HPD 3 HPD 2-3 or PHOTOFRIN{ 1.5-2 HPD 3-5

2 4 2

1 6 2

0 1 17

Taketa and Imakiire [71]

6 squamous T1 to T3 31 primary and recurrent

HPD 3

HPD 5 and 2.5

15

16

Zhao et al. [74]

50 Lip

HPD 5

50

Schweitzer [75] Schweitzer and Visscher [76] Freche and de Corbierre [76] Gluckman [78]

8 5 oral

PHOTOFRIN{ 2 PHOTOFRIN{ 2 Kaposi's sarcoma HPD or PHOTOFRIN{ HPD 3 or PHOTOFRIN{ 2 PHOTOFRIN{ 2 3 2

32 vocal cord

25

8 advanced, recurrent 13 early Ca of oral cavity and oropharynx 2 vocal cord 8 condemned mucosa

11 2

Wenig et al. [79]

26 early disease (various sites)

PHOTOFRIN{ 2 PHOTOFRIN{ 2

20

CR = complete response PR = partial response NR = no response

from additional mutilating surgical procedures or to effectively treat patients with early disease who are not eligible for or who refuse such surgery. Standardization of light treatment, and particularly dosimetric methods, is necessary prior to the development of controlled clinical trials. It is possible that combined light delivery and uence measurement devices may eventually be incorporated into a single device to simplify dosimetry in situ. PDT may also be of benet as an adjuvant intraoperative treatment of Stages III and IV tumors of the head and neck in conjunction with surgery and radiation therapy to improve cure rates.

304

Lasers in Medicine

10.7 PHOTOFRIN PDT in Gastrointestinal Cancer

10.7.1 PHOTOFRIN PDT for Esophageal Cancer
Approximately 10,000 cases of esophageal cancer are diagnosed annually in the United States. While early stage cancer (supercial disease without spread to lymph nodes) may be cured by surgery, 50% of esophageal cancer patients are advanced to the point of inoperability at the time of diagnosis. Difculty swallowing because of obstruction of the esophagus by cancer (malignant dysphagia) is often seen as an initial symptom of esophageal cancer. Although surgery may relieve malignant dysphagia, its use in patients with advanced disease may not be justied because of high postoperative mortality rates (10 to 20%). Palliation of malignant dysphagia may be achieved using radiation therapy, intraluminal placement of a tubular prosthesis, or removal of obstructing tumor endoscopically using Nd:YAG thermal ablation therapy (as described above in the section on endobronchial lung cancer). Patients who have been treated with PDT using HPD or PHOTOFRIN were of two types; those with supercial or early esophageal cancer for whom surgery was not indicated8285 and those with malignant dysphagia due to advanced cancer and seeking palliation.8690 Seventy-one patients with supercial (early) esophageal cancer were treated with PDT using HPD or PHOTOFRIN.8083 Complete responses were observed in 44 of 71 (62%) patients. Patients were treated using cleaved bers, microlens or cylindrical diffuser-tipped ber optics, with side effects including cutaneous photosensitivity. One group85 developed a new light delivery device for the treatment of early esophageal cancer lesions. Monnier et al.85 reported on the PDT treatment of 15 patients with supercial esophageal cancer using HPD at 3 mg/kg or PHOTOFRIN at 2 mg/kg. PDT was carried out 72 hours after injection using a special irradiator attached proximally to a Savary-Gillard dilator; 630 nm laser light dose varied from 60150 J/cm2 (Figure 10.4). In this report, necrosis of esophageal areas not involved with tumor is reported. While the use of the device described in Figure 10.4 allows dilatation of a normally collapsed esophageal lumen and presumably uniform illumination, the use of semicircular mirrors to allow 180degree rather than 360-degree illumination may have increased uence and thus total light dose beyond that desired. It is also possible that even mild stretching of the esophageal wall by an inelastic device may serve to compress blood vessels and compromise blood ow slightly, which could increase the sensitivity of all tissues to the photodynamic effect. Studies that have used bare cylinder diffuser-tipped bers (Figure 10.1) or a cleaved ber placed within a thin-walled balloon (see below) have not reported nonspecic necrosis. Complete responses were observed in 15 of 15 cases (100%), but local recurrence was observed within 13 months in three patients. The remaining patients were locally tumor-free for from 960 months. PDT has been used for the treatment of malignant dysphagia in uncontrolled studies on 55 patients.8690 Relief of dysphagia has been recorded in 52 of 55 (95%) treated patients. Complications included cutaneous photosensitivity (including painless tanning), and esophageal scarring resulting in strictures (narrowing) managed by mechanical dilatation. Two patients with esophageal obstruction so advanced that a guide wire could not be passed, which rendered Nd:YAG laser treatment inadvisable, were treated with PDT and obtained resolution of their malignant dysphagia, which had been so advanced they were unable to swallow their own saliva.90 Thus, PDT appears to have efcacy both in the treatment of early esophageal cancer and as a method of palliation for malignant dysphagia. PHOTOFRIN has been approved for the treatment of patients with completely obstructing esophageal lesions using data from 17 patients provided by a Phase II singlearm protocol for treatment.90 Following a single course of therapy, 94% of patients obtained an objective tumor response, and 76% of the patients experienced at least some palliation of their malignant dysphagia.53 Before treatment, these patients, on average, had difculty swallowing liquids, even saliva. The median duration of clinically meaningful benet to these patients was 69+ days. For the treatment of long esophageal tumors, the development of exible but strong cylinder diffusers of 5 to 10 cm length would allow treatment of long tumors in a short time. Such long cylinder diffusers

Lasers in Photodynamic Therapy

305

Optical fiber

Graduated tube

Removable PTFE diffusing tube (see inset) Savary-Gillard Fiber end dilator Pin photodiode X-ray marker

Chromel-const. thermocouple Ag thermal contact PMMA bulb

Al mirrors Positioning spring

FIGURE 10.4 Line drawing of light-diffusing cylinder developed by Monnier et al.85 for the delivery of light to early cancers of the pharynx and esophagus. Reprinted with permission of the publisher, Lasers in Medical Science.

would have to maintain homogeneity of light dispersion along their lengths. High-output lasers would be needed for such treatments because, with the current recommended power density of 400 mw/cm ber, lasers would have to deliver a stable output of 2 to 4 watts at the ber tip. PHOTOFRIN has also been used to treat the precancerous condition of Barretts esophagus with highgrade dysplasia. Barretts esophagus is an acquired condition in which the normal squamous epithelium of the lower esophagus becomes replaced with a columnar epithelium with intestinal metaplasia.92 It is named after Norman Barrett who, in 1950, described a case of columnar epithelium in a congenitally short esophagus.93 The denition has been rened over the years, but its signicance arises from its potential for malignant transformation. Adenocarcinoma of the esophagus is increasing in incidence more rapidly than any other cancer in the Western World.94 The major risk factor is Barretts metaplasia, which is thought to arise as a healing mechanism in response to chronic injury from gastroesophageal reux (GERD).95,96 In a proportion of cases, this metaplastic tissue can progress to dysplasia and malignancy.97,98 Barretts esophagus is a relatively common condition, with an estimated incidence of 0.4% of the general population99 and 12% in patients with chronic gastroesophageal reux.100 The metaplastic epithelium confers a lifetime risk of 10% of developing adenocarcinoma,101 which equates to one in 52 to one in 441 patient years, or 30125 times the risk of the general population.98,102104 Furthermore, this risk increases signicantly with increasingly severe dysplasia.102,105 Overholt et al. have published the largest series of patients treated with PDT.106 They have treated 100 patients, 87 of whom had dysplasic Barretts esophagus. They were all treated with PHOTOFRIN and red light (630nm) and followed up for a mean of 19 months (range 484 months). The results were encouraging, with 43% having eradication of the entire Barretts segment, and there being a 7580% replacement with squamous epithelium in the rest. Dysplasia was eliminated in 78 patients, although two developed high-grade dysplasia underlying the neo-squamous epithelium. Signicant oesophageal strictures that required formal dilatation developed in 34% of patients. In almost 20% of cases over the period of follow-up, there was recurrence of dysplasia that required further treatment with PDT.

10.7.2 PHOTOFRIN PDT for Colorectal Cancer

Approximately 150,000 cases of colorectal cancer are diagnosed in the United States each year, representing 15% of all cancers. Surgery is the only potentially curative form of treatment, with degree of spread found at the time of surgery (i.e., limited to within the colon, spread to adjacent lymph nodes, or spread beyond the colon or rectum) predictive of the risk of recurrence. Chemotherapy following removal of all visible tumor, (a procedure termed adjuvant chemotherapy) with 5-uorouracil and levamisole is currently recommended as a means of increasing the time to recurrence of disease. Results have been obtained that show accumulation of HPD in colon cancer,89 and which suggest that PDT may be superior to thermal ablation therapy for this class of tumors in light of the resistance of the colon to perforation despite over-treatment with light and HPD combinations.90 Exposure of normal rat colon to PDT at light- and drug-dose combinations sufcient to achieve full thickness necrosis had no

306

Lasers in Medicine

effect on the mean bursting pressure of the colon even 96 hours post-PDT. Thermal necrosis of comparable extent is associated with a 22% perforation rate and weakening of the bowel wall. The authors postulate that lack of mechanical damage to submucosal collagen allows retention of mechanical strength; healing was by regeneration and was complete 2 weeks after treatment. Barr et al.108 reported on a pilot study of PDT of 10 patients with colorectal cancer using HPD (2.5 mg/kg) as photosensitizer. All patients presented with symptoms (bleeding, overow diarrhea, tenesmus, or discharge). All tumors were 5 cm or less in length, and were treated with 630 nm laser light 48 hours after HPD injection using an interstitial method with a cleaved ber optic with the cladding removed from the distal 1mm. Power at the ber tip was 50-l00mW to avoid any thermal damage, and a total light dose of 50 J was given with multiple doses such that light elds just overlapped to treat the entire tumor. The authors conclude that PDT has promise in the treatment of small, or residual, tumors of the gastrointestinal tract and that a combination of Nd:YAG laser (for tumor debulking) and PDT might produce optimal results. While the data on HPD localization91 suggests that adenomas as well as adenocarcinomas can be effectively treated with PDT, it is apparent that the use of a cleaved ber provides suboptimal light distribution. Future work should, if possible, use cylinder diffuser-tipped ber optics, which can provide light to sufcient depths to rapidly treat such tumors. Large, sessile adenomatous polyps or villous adenomas may prove to be effective targets for PDT treatment as well.

10.7.3 PHOTOFRIN PDT for Early Gastric Cancer

The Japanese tradition of screening for gastric cancer has allowed the PDT treatment of patients who refuse surgery or for whom surgery is contraindicated. All of the 53 patients thus treated with HPD and cleaved-ber endoscopic laser light delivery had Type II cancers, dened as ulcerative and circumscribed.109113 Thirty-two patients were treated with PDT alone, and of these, 28 (82%) had no remaining tumors visible by endoscopy or biopsy. Partial responses in this group tended to occur in patients with tumors greater than 4 cm in diameter. Epigastric pain following PDT appears to be the only complication of treatment. PHOTOFRIN has been approved and is marketed in Japan for this indication. While treatment of early gastric cancer has great potential for treatment with PDT, several technical problems may need to be overcome to allow optimal treatment. The shape of the stomach, ongoing peristalsis and, perhaps most important, deep gastric mucosal folds (rugae) which may prevent even light distribution, must all be taken into account. It may be possible to use some device to smooth the mucosal folds while providing light illumination through a diffusing medium, similar to that described by Muller and Wilson for the treatment of brain tumor cavities.114

10.8 PHOTOFRIN PDT for Intraoperative Abdominal or Thoracic PDT

In the evolution of clinical PDT, one of the more promising areas appears to be in the eld of intraoperative PDT, in which PDT is used as an adjunct to surgical resection. Nambisan et al.115 pioneered in this area. Ten patients with recurrent retroperitoneal sarcomas were injected intravenously with either HPD (2.55 mg/kg) or PHOTOFRIN. If no extensive organ involvement was found during surgery, an attempt was made to completely resect the tumor, and areas of uorescence discovered using a UV lamp were also resected. After resection, the eld was illuminated with 630 nm laser light using a microlenstipped ber optic to a total light dose of 30 to 288 J/cm2. In 8 out of 10 cases, complete surgical resection was possible, making the treatment of these patients of adjuvant type. No complications or side effects were noted except for photosensitivity. McCaughan et al.116 reported on the intraoperative therapy of extrahepatic bile duct tumors in a 57-year-old female. Over the course of 4 years, the patient underwent a total of seven treatments, with the last six administered via a U-tube tract. Median life expectancy for this condition ranges from 5 to 11 months if curative resection is not performed. Although anecdotal, this report indicates that interventional PDT may be a treatment option for cholangiocarcinoma.

Lasers in Photodynamic Therapy

307

The National Cancer Institute of the United States National Institutes of Health (NIH) has reported on a Phase I study of PDT with PHOTOFRIN for disseminated intraperitoneal neoplasms.117 Patients (n = 23) were injected with PHOTOFRIN prior to laparotomy and ,48 to 72 hours after injection, underwent debulking to leave behind no neoplastic nodules > 5 mm in diameter. Following debulking, they were treated with 630 nm laser light, using in situ dosimetry in which four sterile photodiodes were sewn into the peritoneal cavity and a microcomputer was used to provide baseline reading of uence and light dose. Dilute lipid suspension was used (0.020.05% Intralipid) within the abdomen to enhance light diffusion. The authors thought that the maximal tolerated dose of PDT to the entire peritoneal surface had not been determined; the highest doses of drug and light reported used in the study were 2.5 mg PHOTOFRIN/kg and 3.0 J/cm2 630 nm wavelength laser light. At the National Cancer Institute, NIH, intraoperative PDT has also been extended to the treatment of pleural malignancies.118 A Phase I study has been reported on the use of PDT for intrathoracic treatment after thoracotomy with debulking of all gross tumor to 5 mm thickness. Eligible patients had disease conned to one hemithorax. All patients received 2 mg/kg PHOTOFRIN and, 48 hours after injection, thoracotomy was performed with resection of gross disease. Laser light dosimetry was performed using seven photodiodes sewn in situ throughout the chest; 0.02% Intralipid was used as a light diffusion aid. All three patients reported treated were discharged 710 days after surgery, and no complications were reported. Lofgren et al.119 have reported on transthoracic endoscopic PDT as treatment for malignant mesothelioma (tumor originating in the pleural lining of the lung) in a patient. A model of the pleural cavity was used to determine light dosimetry; the patient was given PHOTOFRIN 2 mg/kg and 48 hours after injection, thoracoscopy was performed. The entire surface of the pleural cavity (as could be reached via thorascopy and ber optics) was reportedly exposed to a total light dose of 20J/cm2 from a gold vapor laser. The entire procedure took 6 hours under epidural anesthesia; in situ dosimetry was performed using a probe inserted via a separate puncture in the thoracic wall. The patient was followed for 10 months without signs of tumor progression; all laboratory values returned to normal. Intraoperative PDT may, in the future, become a reality for primary tumor treatment. At the present time, limitations on the treatment relate to the large areas (particularly for intraperitoneal therapy) that require treatment with light. Intraperitoneal therapy may require delivery of light to a total area of up to 20,000 cm2. Laser power is therefore, at the present time, a limiting variable in the expansion of this indication of PDT. However, the above mentioned work suggests that adjuvant intraoperative therapy of tumors at risk of recurrence, as well as methods to convert surgical partial responses to complete tumor responses (and therefore cures) should be the goals of intraoperative PDT. Development of longerwavelength photosensitizers as well as alternative methods of activation of photosensitizers should be explored in order to minimize the time necessary for the surgical PDT procedure.

10.9 PHOTOFRIN for Intraoperative PDT in the Treatment of Intracranial Tumors

Primary intracranial tumors (which arise within the central nervous system), the majority of which occur in persons over the age of 45, represent a small percent of total cancer incidence (approximately 8,000 cases annually). Malignant intracranial tumors are generally removed only by surgery. Patient prognosis and survival depend greatly upon the type of tumor and how much can be removed by surgical excision. Tumors of minimal malignant potential are often encapsulated and easily removed in toto. The most malignant and aggressive of intracranial tumors is glioblastoma multiforme, representing up to 9% of all malignant intracranial tumors found in persons over the age of 45. Glioblastoma is not encapsulated, and individual tumor cells indistinguishable among normal brain tissue at the time of surgery may spread several centimeters away from the tumor bed. Despite surgery and postoperative radiation therapy (the standard method of treatment), median survival remains approximately 1 year following diagnosis. Upon recurrence, survival is only 4 to 6 months. PDT has been examined for its ability to treat and prevent recurrence in glioblastoma and non-glioblastoma tumors in man.120127

308

Lasers in Medicine

To laser

Protective tubing

Fiber holder Tightening screw

O - ring Out

In

1 cm

Retaining flange

Optical fiber Rubber 'balloon' Nutralipid

FIGURE 10.5 Line drawing of the balloon device developed by Muller and Wilson for delivery of light to the tumor bed left following craniotomy and tumor resection.107 Reprinted with permission of the publisher, Lasers in Medical Science.

Kaye and Hill120 reviewed progress in this area.121127 The rst 10 patients treated with PDT for intracranial tumors were reported 21 years ago.122,123 Laser light was delivered via ber optic probes inserted directly into the tumor or cyst cavity intraoperatively, but total light doses did not exceed 10 J/cm2. A new device for delivery of light to post-resection tumor beds was developed by Muller and Wilson, who rst reported on its use in the treatment of 32 patients with malignant supratentorial gliomas by intraoperative PDT.114 These authors have reported on a series of 50 patients.125 Eighteen to 24 hours following photosensitizer injection, craniotomy (opening of the skull) with maximal tumor resection was performed followed by photoradiation of the resultant cavity with an inatable balloon applicator lled with 0.1% lipid emulsion to scatter the light produced by a cleaved optical ber inserted into the balloon (Figure 10.5). Patients were treated with laser light doses of 868 J/cm2. Complete responses (disappearance of residual tumor by computed tomographic scans) were observed in 12 of 45 primarytumor patients (19%); four of these patients had recurrent tumors. The authors concluded that the PDTassociated response rate was 35%.

Lasers in Photodynamic Therapy

309

There were no signicant differences in median survival for patients with de novo tumors as compared with patients who had recurrent tumors. While this series is too small and heterogeneous with respect to pathology and primary or recurrent disease to draw conclusions, it is interesting to note that there was a signicant difference in survival between patients receiving high (>1500 J total) and low (<1500 J total) light doses in the nonglioblastoma multiforme group. Those patients receiving the high-dose light had a median survival of 18 months (n = 11) compared with 6.6 months for those patients receiving the low-dose light (n = 11). Using an intrinsic light detection device, Muller and Wilson114,121,125 determined the light penetration depth to be 2.9 mm in tumor (1.24.5 mm) and 1.5 mm in normal brain. Kaye and Hill120 reviewed their experience on the treatment of 45 patients with intraoperative PDT, 44 of whom had high-grade gliomas (15 recurrent glioblastoma). All patients received HPD (5 mg/kg) and 24 hours later underwent craniotomy with radical tumor excision followed by PDT using a at-cut ber for light delivery of up to 240 J/cm2 to the tumor bed as adjuvant therapy, although residual tumor was left at the excision site in 19 patients. Laser light was produced by an argon-pumped dye laser for the treatment of nine patients, and by gold-vapor laser for 36 patients. If a cavity resulted from tumor resection, it was lled with 0.5% (v/v) Intralipid to scatter light evenly throughout the cavity. Total light dose varied from 70240 J/cm2; the temperature of the tumor bed was kept at 37C by frequent irrigation. Patients in this series were followed up to 39 months, and survival curves in various groups were compared. As in the studies of Muller and Wilson,125 PDT was found to have no signicant impact on survival time in patients with recurrent tumors (median survival time 6.8 months). However, median survival of patients with primary tumors treated with resection and PDT followed by postoperative radiotherapy (45 Gy in 20 divided doses) had not yet been reached at the 39-month follow-up point. It is also interesting to note that , in this series,103 patients who received < 120 J/cm2 had a median survival of approximately 12 months, while those who received >120 J/cm2 had not yet reached median survival at the 39-month follow-up period. However, numbers of patients in the low- and high-power-treatment groups were not provided, nor is it known how many had primary or recurrent disease. No postoperative complications or evidence of increased intracranial pressure were observed. In a follow-up study121 of 56 patients with recurrent tumors, all of whom had failed radiation therapy, the mean survival time for patients receiving PDT with PHOTOFRIN for glioblastoma, malignant astrocytoma, and mixed astrocytoma-oligodendroglioma was 30, 44, and > 61 weeks, respectively. For patients undergoing surgery alone, median survival was only 20 weeks. The survival of patients with malignant astrocytoma was again related to light dose, with those receiving a total of > 1800 J. surviving longer (64 weeks median) than patients who received < 1800 J (27 weeks median survival). Powers et al.127 reported on the stereotaxic placement of cylindrical diffuser-tipped bers (Figure 10.1) for the treatment of recurrent inoperable malignant brain tumors with PDT using PHOTOFRIN . Six patients had recurrent supratentorial malignant gliomas (four anaplastic (undifferentiated)) astrocytomas, one glioblastoma and one gliosarcoma and one patient had an intracranial metastatic melanoma. Patients received 2 mg/kg PHOTOFRIN intravenously 24 hours prior to laser light exposure at 630 nm wavelength. Cylindrical diffuser length was chosen to closely match axial length through the epicenter of the tumor. A power of 200 mw/cm of diffuser was delivered to a total light dose of 400 J/cm diffuser tip. The tumor response to PDT after light treatment was estimated by pre- and post-PDT computed tomography (CT) and magnetic resonance imaging (MRI) scans; 31P MR spectroscopy was used to determine metabolic changes in tumor tissue. Following PDT, the solid tumor volume dened by radiographic appearance, changed for all six gliomas from a homogeneously enhancing to a centrally non-enhancing lesion with a greater total volume than that originally dened by contrast enhancement in the preoperative CT images. The patient with metastatic melanoma showed no CT or clinical evidence of tumor response. Two patients, both with anaplastic astrocytomas, were stable without evidence of active disease at 45 and 35 weeks post-PDT; radiologic evidence of recurrence was seen in the other ve patients, generally within 6 to 8 weeks of treatment. Glioma regrowth occurred only outside the contrastenhancing rim noted on post-PDT CT scans. It is clear from this study127 that intratumoral PDT produces tumor necrosis but that additional treatment of the surrounding tissue is required to destroy inltrating cells presumably responsible for tumor recurrence.

310

Lasers in Medicine

The results from the above studies can only be described as preliminary and suggestive, particularly with respect to differences in survival between groups of patients receiving high and low light doses.120,125 Nearly all of the published reports have suffered from lack of stratication of patients on the basis of primary versus recurrent tumors, mixing of patients with different kinds of intracranial tumors, and lack of uniformity in light delivery and dosimetry methods. Despite the fact that Muller and Wilson125 measured the penetration depth of light into brain tumor as 2.9 1.5 mm, they and Powers et al.127 note that the depth of tumor necrosis may be several times the apparent light penetration depth. The reason for this discrepancy is not clear at the moment, but may relate to photobleaching, vascular occlusion and local inammatory or cytokine-mediated effects. It is clear that new light-delivery devices must be developed to adequately treat large areas of the brain to depths of several centimeters. The balloon device described by Muller and Wilson114 has been criticized as inappropriate for uneven cavities or for resections resulting in a at surface following tumor resection.120 Additional clinical trials need to be performed to determine the potential of PDT in the treatment of brain tumors. Perria et al.128 have proposed that all patients with primary brain tumors should be admitted into such a study at rst presentation. Patients would receive surgical tumor excision and PDT followed by radiation therapy, and would be compared with patients undergoing standard resection followed by radiotherapy. A randomized trial comparing the two methods of treatment would be the ideal (and only) method of determining the potential contribution of PDT to the other modalities used in neurosurgical oncology. Such multicenter trials must use uniform methods of light delivery and light dose escalation to a uniform tumor type. The ndings of Powers et al.127 should stimulate the initiation of additional trials utilizing stereotaxic placement of ber optics for treatment of unresectable tumors to fully explore the use of PDT in this indication. The potential of steroid use given after administration of laser light to minimize cerebral edema effects should also be investigated.

10.10 PHOTOFRIN PDT in the Treatment of Ocular Cancer

Malignant melanoma accounts for approximately 70% of all eye (ocular) malignancies, and occurs at an incidence approximately 12% that of skin melanomas (about 20,000 cases per year, but varying with geographic location). Initially, therapy for ocular melanomas consisted of enucleation (surgical removal of the globe of the eye in its entirety). More recently, however, attempts have been made to use radiation therapy, brachytherapy and laser photocoagulation for selective destruction of the tumor and preservation of the eye. The equivalence of newer methods of treatment with surgery, however, remains to be proven. PDT has been used for the treatment of ocular melanoma.129133 Bruce and McCaughan133 treated the largest series, 24 patients, with uveal melanoma over a 2-year period and provided follow-up for as long as 7 years. HPD was administered to all patients at 2.5 mg/kg body weight. Light produced by an argon-pumped dye laser at 630 nm wavelength was delivered via the transpupillary route as well as directly over the base of the tumors. Total light doses ranged from 300 to 3000 J/cm2. Complications included transient initially reduced vision, cataract, exudative retinal detachment and neovascular glaucoma as well as photosensitivity. Tumors were divided into three groups based on size; small (< 500 mm3); medium (5001000 mm3) and large (> 1000 mm3). All groups responded well initially with avascular necrosis. In the small and medium-sized tumor groups, the tumor mass was completely replaced by scar tissue. In the large-tumor group, tumor destruction was incomplete, despite additional PDT treatments, and enucleation was required in each case. During 7 years of follow-up, six of the 24 patients died, with three deaths due to metastatic melanoma. Ten patients underwent enucleation for recurrence of tumor at the edge of the treated area or regrowth after a partial response. Thus, PDT appears to have good promise for the treatment of ocular melanomas of 1000 mm3 volume.133 It has been suggested that an alternate method of ocular melanoma treatment involve treatment of the choroid surrounding the tumor to close off the choroidal blood supply, rather than concentrating on direct light treatment of the tumor itself.129 Although the collective data suggest that high light doses (in excess of 1000 or 2000 J/cm2), might be optimal for treatment of melanoma, clinical trials for dose-ranging studies are required to test this possibility. The fact that melanoma tends to be a heavily

Lasers in Photodynamic Therapy

311

pigmented tissue and thus attenuates light rapidly over a short distance, is supported by the complete response seen in amelanotic melanoma at a light dose that produced only partial responses, at best, in pigmented melanomas.132 Development of devices specic for delivery of light to ocular structures may be able to increase the complete response rates seen for ocular melanoma. Retinoblastoma is the most common primary ocular tumor in children, with an incidence of approximately 200 per year. It arises from cells within the nuclear (outer) layer of the retina, and 80% of cases are diagnosed in children less than 4 years old. Surgery (enucleation) or radiation therapy may be used for treatment, although approximately 50% of patients receiving radiation therapy alone eventually require enucleation. Patients with small, unilateral tumors may be treated with cryosurgery or laser photocoagulation to preserve both the eye and vision. However, even patients with large tumors and distant metastasis can have a 5-year survival approaching 90%, although the chance for complete disease control is only about 30% in such cases. In retinoblastoma, PDT appears to be able to delay time of enucleation but, in the studies reported,132,134 did not produce long-term responses. The vitreous cancer seedings, which receive nourishment not from blood, but from vitreous humor, may not contain sufcient photosensitizer to allow a good response at the light doses used in these trials, and might account for the poor results that were obtained. Future trials may consider injection of small quantities of photosensitizer directly into the vitreous in the vicinity of the tumor. The presence of lipophilic photosensitive dyes may allow selective uptake by the tumors and improve the tumor response.

10.11 PHOTOFRIN PDT in the Treatment of Cutaneous and Subcutaneous Tumors

The most common forms of skin (cutaneous) cancer are basal cell and squamous cell carcinoma, which generally occur on sun-exposed areas such as the hands, neck, and head. More than 500,000 such skin cancers are reported annually in the United States. Surgery is generally performed and is curative when performed in early stages of either cancer. However, curettage, electrodessication and laser vaporization can be used as alternative methods of therapy for supercial lesions. For the treatment of difcult and recurrent skin cancers, Mohs micrographic surgery is used, a specialized technique in which serial horizontal sections of excised tissue are mapped and examined by frozen section examination, with excision continued until margins are microscopically free of tumor. Cryosurgery, topical 5-uorouracil and radiation therapy have also been used for selected patients with supercial skin cancer. In patients with recurrent or aggressive skin cancer, PDT has been examined as a potential treatment modality as a nonsurgical treatment on patients in whom surgery is not indicated because of poor health, and in patients with skin cancers in sites where surgical excision might lead to scarring upon healing. Rare forms of aggressive skin cancer that occur over many sites simultaneously (Bowens disease, basal cell nevus syndrome) have also been treated experimentally with PDT due to the difculty of adequate therapy with existing procedures. Cutaneous or subcutaneous cancer can also occur as a consequence of metastatic disease from other sites. Local recurrence in the chest wall following mastectomy may occur in up to 27% of patients. Up to 80% of these patients may have an initial response to radiation therapy, but up to two-thirds will develop another local recurrence. Systemic chemotherapy and hormonal therapy have also been used to treat these patients. PDT has been investigated as a potential therapy in these patients as well. PDT has been used to effectively treat surface lesions, whether of primary skin cancer origin or metastatic from a distant primary source. A summary of publications135143 on the use of PHOTOFRIN PDT in this indication is given in Table 10.6. Schuh et al.140 reported on the treatment of 14 patients with breast cancer metastatic to chest wall with PDT using PHOTOFRIN (1 or 2 mg/kg). All patients had been previously treated with multiple modalities. It is perhaps important to note that light doses as high as 288 J/cm2 did not cause necrosis of normal tissue. However, in previously heavily irradiated areas skin necrosis was observed to occur at light doses

312

Lasers in Medicine

TABLE 10.6 Summary of Published Studies Using PDT for Treatment of Cutaneous and Subcutaneous Tumors
Method of Dose of Light Delivery S 60-120 J/cm2 Results CR 26 PR 16 NR 7 PD 3 Follow-up (months (Evaluable)

Reference Bandieramonte et al.[117]

Patients (sites) Disease 7 (61) (43 Basal cell) (18 Metastatic breast cancer) 6 (34) Squamous cell, lung cancer 27 Basal cell cancer (7) Squamous cell cancer (3) Melanoma (3) Metastatic breast cancer (12) 6 (53) Basal or squamous cell cancer 2 (>500) Bowen's Disease 14 Metastatic breast cancer

Drug (mg/kg) HPD (3)

Gilson et al. [118] McCaughan et al. [119]

PHOTOFRIN 1-2 mg/kg HPD (3) PHOTOFRIN (2)

S 25-100 J/cm2 S (20-30 J/cm2) 1

16

NA

NA

NA

(3)

48

19

NA

NA

(12)

Pennington et al. [120]

HPD (5)

S (30 J/cm2)

36

NA

NA

NA

(12)

Robinson et al. [121]

PHOTOFRIN (2) PHOTOFRIN (1-2)

S (25 J/cm2)

>5 00

(6)

Schuh et al. [122]

S (36-288 J/cm2) 1 (350-440 J/cm2) S (38-180 J/cm2 S (8-60 J/cm2)

(6)

Tse et al. [111] Waldow et al. [123]

3 (40) Basal cell nevus Syndrome 10 (155) Metastatic breast cancer Basal cell cancer Bowen's disease 37 (151) Basal Cell Cancer 20 Metastatic breast cancer

HPD (3) HPD (3) PHOTOFRIN (2)

33

(12-14)

11 7

30

(8-24)

Wilson et al. [124]

PHOTOFRIN (1)

S (180-233 J/cm2)

13 3

18

(Minimum 12)

Sperduto et al. [125]

PHOTOFRIN { (1.5)

S (20-359 J/cm2)

(1-18)

S-Surface; I-Interstitial; NA-Not Available; CR-Complete Response; PR-Partial Response; NR-No Response; PD-Progressive Disease

Lasers in Photodynamic Therapy

313

in the 104144 J/cm2 range. No attempt was made to correlate degree of skin photosensitivity with extent or dose of prior radiation therapy. It is apparent that large numbers of surface lesions can be treated repeatedly with PDT to get a desired effect. Robinson et al.139 described the treatment of two patients who together had > 500 lesions of Bowens disease (intraepidermal carcinoma). Up to a total of ve treatment sessions were given, resulting in complete clearing of the lesions by the last 6-month observation period post-treatment. The authors noted that halving the dose of PHOTOFRIN for one treatment session did not diminish the photosensitivity reactions but signicantly reduced the efcacy of treatment. Tse et al.129 treated multiple lesions of the basal cell nevus syndrome in three patients. Nevoid basal cell carcinoma syndrome (NBCS) is a rare inherited autosomal-dominant disease. Nevoid basal cell carcinomas are multiple, have an early age of onset, and are indistinguishable microscopically from basal cell carcinoma. Of the 40 lesions treated, 33 had complete responses and seven partial responses (CR + PR = 100%). CRs were followed for 1214 months, and four recurrences (10.7%) were observed. The National Cancer Institute experience using PDT with PHOTOFRIN for the treatment of 20 patients with locally recurrent (chest wall) breast carcinoma has been described.143 Complications included pain, bruising, blistering, ulceration and necrosis in areas of tumor involvement. It is to be noted that healing was observed in all cases. The authors conclude that the depth of tumor involvement in chest wall metastases from breast cancer generally exceeds the penetration of 630 nm wavelength light applied by surface treatment, and that this may account for the short periods of response as well as the low numbers of patients responding. Responses did not show a clear relationship to power density or to total light dose. No attempts were made to deliver therapy via interstitial ber placement in this study. Thus, it appears that PDT has the potential of becoming a preferred method for the treatment of certain cutaneous and subcutaneous cancers. In addition to efcacy at removing lesions, the healing process of lesions occurring as the result of PDT treatment (which may take from 2 to 12 weeks depending on the size and depth of the necrosis), results in excellent cosmetic results often mentioned as free of scar formation.136 Cutaneous photosensitivity and pain during and after treatment at the site of treatment are generally mentioned side effects. For PDT to reach its potential in this indication, standardization of drug and light dose through the use of multicenter trials will be necessary. Standardization of the method of light delivery, particularly if interstitial light delivery is to be used, must also be carried out. The contribution of prior radiation therapy to normal skin necrosis at usual therapeutic doses of light and drug must also be determined if treatment of cutaneous breast cancer metastases is to become a standard indication for PDT treatment.

10.12 New Photosensitizers for PDT

PHOTOFRIN, although the rst approved photosensitizer for use in PDT, has characteristics that could be improved upon for greater acceptance in certain clinical indications. With PHOTOFRIN, a waiting period of 40 to 48 hours has been recommended to allow the photosensitizer to wash out of normal tissue to avoid a lack of selectivity in tissue destruction when laser light is applied.3 Cutaneous photosensitivity may persist for 4 to 6 weeks or even longer, limiting the mobility of the treated individual to periods during which the sun is down or cloudy weather predominates (unless complete covering up in protective garments is carried out.2 PHOTOFRIN is also a complex molecular mixture, containing thousands of species of oligomers, thus rendering it extremely difcult to determine which species are responsible for antitumor efcacy or cutaneous photosensitivity. Photosensitizing agents, which are single molecular entities, provide brief periods of cutaneous photosensitivity, and allow treatment of patients with light shortly after administration (preferably on the same day as photosensitizer dosing), have been sought. The following is a brief discussion of individual compounds that have been approved for treatment of various indications and represent the second generation of PDT photosensitizers.

314

Lasers in Medicine

10.12.1 VISUDYNE (Verteporn)

VISUDYNE (QLT, Inc.) is a chlorin produced, as is PHOTOFRIN, from hemoglobin as starting material. It is a chlorin with the longest wavelength absorbance peak at 690 nm and is more hydrophobic than PHOTOFRIN, necessitating formulation in organic solvent or liposomes.144 Peak tissue levels of VISUDYNE are achieved 3 hours following intravenous injection in rodents, and the compound appears to wash out of normal skin 3 or 4 days following injection. It is hoped that the 690 nm wavelength of activation will allow deeper tissue penetration for greater efcacy in bulky tumors than is seen with PHOTOFRIN, which uses 630 nm wavelength light for activation of the PDT process. It is believed that VISUDYNE PDT mechanisms of tumor necrosis are similar to those of PHOTOFRIN and involve damage to tumor vasculature. A phase I study in patients with primary skin cancer or cutaneous metastases of cancer from other primary sites145 reported that patients were rst given an intravenous infusion of 0.20.5 mg/kg of VISUDYNE in a liposomal formulation. The tumors were then irradiated with 50150 J/cm2 of 690 nm light from an argon-pumped tunable dye laser via front surface illumination 3 to 4 hours after VISUDYNEl injection. Of 64 skin tumors from the rst 15 patients, the overall (complete + partial) tumor response was 74%, with higher response rates observed using high dose ranges of either drug or light. Side effects of tumor and normal skin treatment (normal skin was included in the eld of illumination) included edema and erythema of normal skin. 10.12.1.1 VISUDYNE PDT for the Treatment of Age-Related Macular Degeneration Age-related macular degeneration (AMD), a deterioration of the central portion of the retina, is the major cause of severe, irreversible vision loss and blindness in the United States and other developed countries (see Ref. 146 for review). There are two forms the atrophic and the neovascular, exudative. The neovascular, exudative form includes serous or hemorrhagic detachment of retinal pigment epithelium and choroidal neovascularization, which lead to leakage and brovascular scarring. VISUDYNE PDT received marketing approval in the United States and other countries during 2000 for the treatment of AMD in patients with predominantly classic subfoveal choroidal neovascularization. For PDT of AMD, VISUDYNE is injected intravenously to a total dose of 6 mg/m2 and 689 nm laser light is delivered at an intensity of 600 mW/cm2 over 83 sec to provide a recommended light dose of 50 J/cm2.147 The treatment spot size was 1000 microns larger than the greatest linear dimension of the lesion on the retina to allow a 500-micron border and to ensure full coverage of the lesion. The phase III trials were double-masked, placebo-controlled, randomized studies enrolling a total of 609 patients (VISUDYNE 402, placebo 207). During these studies, retreatment was allowed every 3 months if uorescein angiograms showed any recurrence or persistence of leakage. The placebo control consisted of intravenous administration of Dextrose 5% in water, followed by light application identical to that used for VISUDYNE PDT. A planned analysis of safety and efcacy conducted at 1 year statistically favored VISUDYNE for the visual acuity endpoints. The subgroup of patients with predominantly classic choroidal neovascularization lesions was more likely to exhibit a treatment benet. For the primary efcacy endpoint (percentage of patients who lost < 3 lines of visual acuity), these patients showed a difference of 28% between treatment groups (67% for VISUDYNE PDT patients compared with 39% for placebo patients, P < 0.001). The most frequently reported adverse events to VISUDYNE included headaches, injection site reactions (including extravasation and rashes) and visual disturbances, including blurred vision, decreased visual acuity and visual eld defects. Severe vision decrease of four lines or more within 7 days of treatment was reported in 14% of patients.

10.12.2 LEVULAN (Aminolevulinic Acid HCl, ALA) PDT

Levulan (DUSA Pharmaceuticals, Inc.) is itself not a photosensitizer, but a sort of prodrug, as it is the natural metabolic precursor of the endogenous tissue photosensitizer protoporphyrin IX (PPIX).128 PPIX is an immediate biosynthetic precursor of heme. The apparent rate-limiting step in the synthesis of PPIX is the availability of ALA.148 Therefore, by providing large quantities of exogenous ALA, certain cells will

Lasers in Photodynamic Therapy

315

produce sufcient amounts of PPIX to allow for a PDT effect when illuminated with an activating quantity of light. Different types of cells appear to have different capabilities for synthesizing PPIX from exogenously provided ALA.148,156 Topical application of ALA to supercial basal cell carcinomas and squamous cell carcinomas induced PPIX uorescence and photosensitization restricted to the area of ALA application. Kennedy et al. have reported treating more than 300 supercial basal cell carcinomas in patients using topically applied ALA and broad spectrum red light with a complete response rate at 3 months of approximately 79%.148 No cutaneous photosensitivity was reported in sites other than those at which ALA application was given. Recent clinical trials using topically applied ALA appear to corroborate the initial clinical reports.149,150,156 Wolf et al.149 reported a phase II study in which patients with actinic keratoses, early invasive squamous cell carcinomas, supercial basal cell carcinomas, noduloulcerative basal cell carcinomas and cutaneous metastases of malignant melanoma were treated with ALA PDT. ALA was topically applied as a 20% oilin-water emulsion for 4 to 8 hours. Approximately 4 hours after application of ALA, photoactivating visible light was provided by a slide projector with or without a lter, which eliminated wavelengths below 570 nm. Total applied uence varied from 30 to 100 J/cm2. CRs were observed in 9 of 9 actinic keratoses, 5 of 6 (83%) early invasive squamous cell carcinomas, 36 of 37 (97.3%) supercial basal cell carcinomas, 1 of 10 (10%) noduloulcerative basal cell carcinomas, and 0 of 8 metastatic malignant melanoma lesions. With a median follow-up of 7 months (range 3 to 12 months), only one recurrence among complete responders (in a patient with supercial basal cell carcinoma) was seen. Shanler et al.150 reported that topical application of 2040% ALA using occlusive dressings in humans for 45 hours resulted in PPIX concentrations (as measured by uorescence) that were 36-fold higher in actinic keratoses and basal cell carcinomas than in normal skin. Treatment with laser light at 630 nm to a total uence of 150200 J/cm2 was found to be clinically effective for the complete resolution of supercial lesions. Topical LEVULAN in solution, together with blue light, was approved for marketing in the United States in 1999 for the treatment of non-hyperkeratotic actinic keratoses of the face and scalp. 10.12.2.1 LEVULAN PDT for Treatment of Actinic Keratoses of the Face and Scalp LEVULAN Topical Solution, 20%, plus blue light at 610.9 J/cm2, has been used to treat actinic keratoses in 232 patients in six clinical trials.151 Phase III studies were two identically designed, multicenter twoarm studies using LEVULAN KERASTICK for Topical Solution applicators plus blue light for 1000 seconds (16 min, 40 sec) for a nominal exposure of 10 J/cm2. Excluded from these studies were patients who had a history of cutaneous photosensitization, porphyria, hypersensitivity to porphyrins, photodermatosis, or inherited or acquired coagulation defects. A minimum of four and a maximum of 15 clinically typical, discrete, non-hyperkeratotic, target actinic keratosis lesions were identied. Target lesions on the face or on the scalp, but not in both locations in the same patient, received treatment. The patients were randomized to receive treatment either with LEVULAN 20% topical solution plus blue light or Vehicle plus blue light. Patients were randomized at a 3 to 1 LEVULAN to Vehicle ratio. Altogether, 243 patients were enrolled in the two identical phase III studies and pooled efcacy results at the end of 12 weeks (lesions remaining at the week 8 follow-up period could be retreated) indicated that 88% of LEVULAN PDT-treated patients responded as compared with 20% of Vehicle-treated patients (P < 0.001). Patients did not receive follow-up past 12 weeks after the initial treatment. No non-cutaneous adverse events were found to be consistently associated with topical LEVULAN application followed by blue light exposure. The constellation of transient local symptoms of stinging or burning, itching, erythema and edema as a result of LEVULAN KERASTICK for topical solution plus blue light treatment was observed in all controlled clinical studies of LEVULAN photodynamic therapy for actinic keratoses treatment. Stinging or burning subsided between 1 minute and 24 hours after the BLU-U Photodynamic Therapy Illuminator was turned off, and appeared qualitatively similar to that perceived by patients with erythropoietic protoporphyria upon exposure to sunlight. There was no clear drug-dose- or light-dose-dependent change in the incidence or severity of burning and stinging. The most common changes in lesion appearance after LEVULAN PDT were erythema and edema. In 99% of active treatment patients, some or all lesions were erythematous shortly after treatment, while in 79%

316

Lasers in Medicine

of Vehicle treatment patients, some or all lesions were erythematous. In 35% of active treatment patients, some or all lesions were edematous, while no Vehicle-treated patients had edematous lesions. Both erythema and edema resolved to baseline or improved by 4 weeks after therapy. LEVULAN Topical Solution applications to photodamaged perilesional skin resulted in photosensitization of photodamaged skin and a photodynamic response. 10.12.2.2 ALA PDT for Lesions of the Gastrointestinal Tract Systemic administration of ALA appears to result in the selective synthesis of PPIX within certain mucosal surfaces. The effects of intravenously administered ALA at various doses on the accumulation of PPIX in rat colon152 and stomach were studied.153 Fluorescence levels in normal colonic mucosa peaked at 45 hours after administration of ALA. The ratio of uorescence in the mucosa to that seen in muscle or submucosal regions was 6:1 at all times,152 and a similar selective PPIX synthesis by rat gastric mucosa was also observed.153 Selective killing of the rat colon tumor model was seen in this study152 using ALA PDT, with only normal mucosa showing damage from PDT, while muscle and submucosal tissues remained intact.152 The remarkable selectivity demonstrated by this system suggests the possibility of future ALA PDT treatment for mucosal carcinomas with an improved therapeutic index as compared with PHOTOFRIN. Systemic dosing of patients with ALA has been reported.156 Three patients with histologically proven carcinomas of the oral cavity were given ALA orally as a bolus dose (3080 mg/kg). Peak uorescence of PPIX was measured in tumor biopsy specimens and was found to peak between 4 and 6 hours after ALA administration, and returned to background autouorescence levels in 24 hours. One patient had 60 mg/kg ALA readministered by the oral route followed in 5 h by 630 nm laser light at 100 J/cm2, which resulted in selective tumor necrosis. A number of clinical studies have been reported for ALA PDT in Barretts esophagus (BE), with only the largest series reported here. Gossner et al.154 treated 32 patients with ALA-PDT using 60 mg/kg of ALA given orally and 635nm laser light delivered 4-6 hours after drug dosing through a 2 cm cylinder diffuser-tipped ber optic at a power density of 100 mW/cm2 to a light dose of 150 J/cm2. Ten patients had high-grade dysplasia (HGD), and the remainder mucosal cancer. Patients were maintained on omeprazole for the duration of the study. Dysplasia was eradicated in all patients with HGD, and there was complete remission of the cancers in 17 of 22 patients (77%). Of note was that all the tumors that were not eradicated were greater than 2mm in depth. This remission was maintained during follow-up of 130 months (mean 9.9 months). Re-epithelialization of the Barretts mucosa by histologically normal squamous epithelium was seen in 68%. However, the presence of non-dysplasic specialized columnar epithelium under the neo-squamous epithelium was noted in some patients. A mean of 1.7 treatment sessions were required for the eradication of mucosal tumors. Side effects included transient nausea up to 6 hours after PDT, managed by symptomatic therapy. Mild transient increases in hepatic aminotransferases (ALT and AST) were noted that returned to normal levels within 1 week and remained normal throughout follow-up. No patients reported dysphagia or exhibited strictures or systemic photosensitivity. Ackroyd et al. used green (514nm) light for the treatment of Barretts esophagus with LGD.155,157,158 This group also used a lower dose of ALA (30mg/kg) than most other studies. In the only double-blind, placebo-controlled study reported for BE, 36 patients with low-grade dysplastic BE on chronic omeprazole therapy were randomized to receive oral ALA (30 mg/kg) or placebo, followed 4 hours later by endoscopy and treatment of up to 6 cm of BE with green laser light (514 nm) to a total light dose of 60 J/cm2 through a Perspex cylinder.155 Follow-up endoscopy and biopsy was performed at 1, 6, 12, and 24 months. A response was seen in 16 of 18 patients (89%) in the ALA group, with a median area decrease of BE of 30% (range = 0-60%). In the placebo group, a median area decrease of 0% was seen (range 010%). In the ALA PDT group, the 16 patients maintained regression to normal squamous epithelium over the entire 24-month follow-up period, and no dysplasia was observed in the treated areas of BE that remained in any patient. In the placebo group, persistent low-grade dysplasia was found in 12 of 18 patients (67%) followed for 24 months (P < 0.001). The only consistent side effect of ALA PDT seen in all treated patients was some degree of pain in the chest during light treatment, but analgesia was required in only 3 of the 18 (17%). The discomfort persisted for 35 days following treatment and was aggravated

Lasers in Photodynamic Therapy

317

by swallowing or coughing. One patient developed a mild rash after exposure to sunlight on the day after treatment. None of the patients developed strictures or complained of dysphagia during the course of the study. No laboratory abnormalities were reported. Improved methods of circumferential light delivery may be required for complete ablation of BE. In the normal rat colon, the area of mucosal necrosis caused by ALA PDT could be increased by a factor of 3 by interrupting the light dose for 150 sec, after 20% of the total light energy had been delivered.159 It was hypothesized that fractionation of the light regime for even a short time could result in replenishing of molecular oxygen, allowing for production of additional singlet oxygen, whereas continuous illumination caused rapid depletion of oxygen, limiting the PDT effect. In a recent paper,160 the partial pressure of oxygen (pO2) was measured using microelectrode technology in rat colon mucosa before, during, and after ALA PDT 1 mm and 3 mm away from the site of irradiation with a ber optic. Although a fall in pO2 was observed adjacent to the ber during continuous and fractionated light delivery, pO2 fell at the more distant site only during the fractionation regime. Therefore, light fractionation might increase the efcacy of BE mucosal ablation if the model is consistent with human photophysiology.

10.13 Conclusions
PDT is now an approved therapy in several areas of medicine. However, PDT continues to evolve and to have mechanisms of treatment rened and extended. This is most clearly seen in the sections on intraoperative PDT (abdominal, intrathoracic, and intracranial indications). PDT may have a potential role in treating peritoneal carcinomatosis and pleural malignancy, but it is in these indications, which involve treatment of large surface areas during operative procedures, that the need for high output lasers and improved methods of light delivery are evident. Still, development of new photosensitizers with increased singlet oxygen production efciencies and higher wavelengths of activation than 630 nm continues, which could allow the effective use of existing lasers in even these challenging indications. The role of PDT in the treatment of head and neck cancers appears to be focused realistically on treatment of supercial recurrent tumors of the oral mucosa and oropharynx as well as the treatment of condemned mucosa. Treatment of intracranial tumors remains an exciting area of investigation for PDT, and recent results suggest a tantalizing connection between light dose and increased survival in non-glioblastoma brain tumors. However, this area of research requires much work, particularly in the standardization of light delivery systems and in the development of randomized multicenter clinical trials with stratication based on tumor type and location. The number of potential second-generation photosensitizers in clinical trials is increasing, and certain clinical advantages of some compounds may prove sufciently attractive to replace PHOTOFRIN in the future. However, it is likely that different photosensitizers will nd optimal uses in certain indications. For use in intraoperative PDT indications, which require several hours of exposure to laser light (using present-day technology) it is likely that the ability to be retained in tissues for many hours during which photosensitizing ability remains maximal (as with PHOTOFRIN) will prove useful. In contrast, for ambulatory patients with skin lesions, a rapidly clearing topically applied photosensitizer such as LEVULAN is clinically desirable. The next few years may be those in which PDT achieves more of its potential, with advances in hardware (diode lasers, for example, or the development of useful non-laser light sources) putting PDT within reach of many more clinical and basic researchers.

References
1. C.J. Gomer, Photodynamic therapy in the treatment of malignancies, Seminars in Hematology 26, 27-34, (1989). 2. T.J. Dougherty and S.L. Marcus, Photodynamic Therapy, Eur. J. Cancer 28A, 1734-1742 (1992). 3. S.L. Marcus, Photodynamic therapy of human cancer, Proc. IEEE, 80, 869-889 (1992).

318

Lasers in Medicine

4. B.W. Henderson, Probing the effects of photodynamic therapy through in vivoin vitro methods, in Photodynamic Therapy of Neoplastic Disease, D. Kessel (Ed.), Boca Raton, FL., CRC Press, pp. 169-188 (1990). 5. P.J. Bugelski, C.W. Porter and T.J. Dougherty. Autoradiographic distribution of hematoporphyrin in normal and tumor tissue of the mouse, Cancer Res, 41, 4606-4610, (1981). 6. C. Zhou, New trends in photobiology (invited review): mechanisms of tumor necrosis induced by photodynamic therapy, J. Photochem. Photobiol. 3, 299-318, (1989). 7. T.J. Wieman et al., Effect of photodynamic therapy on blood ow in normal and tumor blood vessels, Surgery 104, 512-517, (1988). 8. V.H. Fingar, T.J. Wieman and K.W. Dunk, Role of thromboxane and prostacyclin release on photodynamic therapy-induced tumor destruction, Cancer Res., 50, 2599-2603, (1990). 9. D.R. Doiron et al., Fluorescence bronchoscopy for detection of lung cancer, Chest 76, 27-32, (1979). 10. A.E. Proo, Fluorescence diagnosis and dosimetry using porphyrins, in Photodynamic Therapy of Neoplastic Disease, D. Kessel (Ed.), Boca Raton, FL, CRC Press, pp. 77-90, (1990). 11. N.D. Ainsworth and J.A. Piper, Laser systems for photodynamic therapy, in Phototherapy of Cancer, G. Morstyn and A.H. Kayo (Eds.), New York, Harwood Academic Publishers, pp. 37-72, (1990). 12. U. Nseyo, R.K. Whalen and S. Lundahl, Photodynamic therapy in the management of bladder cancer, a critical review, Las. Inst. Amer. 64 ICALEO, 57-65, (1988). 13. R.C. Benson, Jr. Hematoporphyrin photodynamic therapy for transitional cell carcinoma of the bladderan update, in Methods in Porphyrin Photosensitization, D. Kessel (Ed.), New York, Plenum Press, pp 3-12, (1985). 14. R.C. Benson, Jr. Treatment of diffuse transitional cell carcinoma in situ by whole bladder hematoporphyrin derivative photodynamic therapy, J. Urol. 134, 675-678, (1985). 15. X.W. Ha et al., Clinical use of hematoporphyrin derivative in malignant tumors. Chin. Med. J. 96, 754-758, (1983). 16. H. Hisazumi, T. Misaki and N. Miyoshi, Photodynamic therapy of bladder tumors, in Lasers and Hematoporphyrin Derivative in Cancer, Y. Hayata and T.J. Dougherty (Eds.), Tokyo, Igaku-Shoin, pp. 85-87, (1983). 17. H. Hisazumi, T. Misaki and N. Miyoshi, Photoradiation therapy of bladder tumors, J. Urol. 130, 685-687, (1983). 18. U.O. Nseyo and T.J. Dougherty, Photodynamic therapy in the management of resistant bladder cancer (abstract) Lasers Surg. Med. 6, 228, (1986). 19. B.P. Shumaker and F.W. Hetzel Clinical Laser photodynamic therapy in the treatment of bladder carcinoma, Photochem. Photobiol. 46, 899-901, (1987). 20. A. Tsuchiya, et al., Hematoporphyrin derivative and laser photo-radiation in the diagnosis and treatment of bladder cancer, J. Urol. 130, 79-82, (1983). 21. U.O. Nseyo, T. J. Dougherty and L. Sullivan, Evaluation of photodynamic therapy in the management of bladder cancer, Lasers Surg. Med. 7, 107, (1987). 22. U.O. Nseyo, T.J. Dougherty and L.L. Sullivan, Photodynamic therapy in the management of resistant lower urinary tract carcinoma, Cancer 60, 3113-3119, (1987). 23. G.R. Prout Jr., P. Grifn and J.J. Daly, The outcome of conservative treatment of carcinoma in situ of the bladder, J. Urol. 138, 766-770, (1987). 24. G.R. Prout Jr. et al., Photodynamic therapy with hematoporphyrin derivative in the treatment of supercial transitional call carcinoma of the bladder, N. Engl. J. Med. 317, 1251-1255, (1987). 25. J.I. Harty et al., Complications of whole bladder dihematoporphyrin ether photodynamic therapy, J. Urol. 141, 1341-1346, (1989). 26. M. Dugan et al., A randomized trial of observation (OBS) vs. photodynamic therapy (PDT) after transurethral resection (TUR) for supercial papillary bladder carcinoma (SPBC), Proc. Ann. Mtg. Am. Soc. Clin. Onc., 10, 554, (1991).

Lasers in Photodynamic Therapy

319

27. J.P.A. Marynissen, H. Jansen and W.M. Star, Treatment system for whole bladder wall photodynamic therapy with in vivo monitoring and control of light dose rate and dose, J. Urol. 142, 13511355, (1989). 28. U.O. Nseyo, S.L. Lundahl and D.C. Merrill, Whole bladder photodynamic therapy: critical review of present-day technology and rationale for development of intravesical laser catheter and monitoring system, Urology 36, pp. 398-402, (1990). 29. M.M. Walther, The role of photodynamic therapy in the treatment of recurrent supercial bladder cancer. Urologic Clin. N. Am. 27, 163-170 (2000). 30. U. Nseyo et al., Urinary cytokines following photodynamic therapy for bladder cancer, Urology 36, 167-171, (1990). 31. D.A. Cortese and J.H. Kinsey, Bronchoscopic phototherapy using hematoporphyrin derivative, Surg. Clin. N. Am. 64, 941-946, (1984). 32. D.A. Cortese and J.H. Kinsey, Hematoporphyrin derivative phototherapy in the treatment of brochogenic carcinoma, Chest 86, 8-13, (1984). 33. E.S. Edell and D.A. Cortese, Bronchogenic phototherapy with hematoporphyrin derivative for treatment of localized bronchogenic carcinoma: A 5-year experience, Mayo Clin. Proc. 62, 8-14, (1987). 34. Y. Hayata et al., Photoradiation therapy with hematoporphyrin derivative in early and stage 1 lung cancer, Chest 86, 169-177, (1984). 35. Y. Hayata et al., Photoradiation therapy in early stage cancer cases of the lung, esophagus and stomach, in Porphyrins in Tumor Phototherapy, A. Andreoni and R. Cubeddu (Eds.), New York, Plenum Press, pp 405-412, (1984). 36. Y. Hayata et al., Hematoporphyrin derivative and laser photoradiation in the treatment of lung cancer. Chest 81, 269-277, (1982). 37. H. Kato, C. Konaka and J. Ono, Effectiveness of HPD and radiation therapy in lung cancer, in Methods in Porphyrin Photosensitization, D. Kessel, and T.J. Dougherty (Eds.), New York, Plenum Press, pp 23-39 (1983). 38. H. Kato, Lung cancer, in Lasers and Hematoporphyrin Derivative in Cancer, Y. Hayata and T.J. Dougherty (Eds.), Tokyo, Igaku-Shoin, pp 39-56, (1983). 39. G.S. Keller, D.R. Doiron and G.U. Fisher, Photodynamic therapy in otolaryngology - head and neck surgery, Arch. Otolaryngol. 111, 758-761, (1985). 40. J.H. Li, Y.P. Chen and S.D. Zhao, Application of hematoporphyrin derivative and laser-induced photodynamical reaction in the treatment of lung cancer: A preliminary report of 21 cases, Lasers Surg. Med. 4, 31-37, (1984). 41. R.G. Vincent et al., Photoradiation therapy in advanced carcinoma of the trachea and bronchus, Chest 85, 29-33, (1984). 42. R. Vincent and T.J. Dougherty, Photoradiation therapy in the treatment of advanced carcinoma of the trachea and bronchus, in Porphyrin Localization and Treatment of Tumors, D.R. Doiron and C.J. Gomer (Eds.), New York, Alan R Liss Inc, pp. 759-766, (1984). 43. S. Karanov et al., Photodynamic therapy in lung and gastrointestinal cancers, J. Photochem. Photobiol., B: Biology, 6, 175-181, (1990). 44. O.J. Balchum, D.R. Doiron, Photoradiation therapy of endobronchial lung cancer: large obstructing tumors, nonobstructing tumors, and early-stage bronchial cancer lesions, Clin. Chest Med. 6, 255275, (1985). 45. J.S. McCaughan Jr, T.E. Williams and B.H. Bethel, Photodynamic therapy of endobronchial tumors, Lasers Surg. Med. 6, 336-345, (1986). 46. P. Hugh-Jones and W.N. Gardner, Laser photodynamic therapy for inoperable bronchogenic squamous carcinoma, Quart. J. Med. 64, 565-582, (1987).

320

Lasers in Medicine

47. H. Kato et al., Five-year disease-free survival of a lung cancer patient treated only by photodynamic therapy, Chest 90, 768-770, (1986). 48. J.S. Nelson, R.D. Fairshter and M.W. Berns, Long-term survival of a lung cancer patient treated with photodynamic therapy. Case Report, Lasers in Surg. & Med. 10, 208-210, (1990). 49. H. Kato et al., Preoperative laser photodynamic therapy in combination with operation in lung cancer, J. Thorac. Cardiovasc. Surg. 90, 420-429, (1985). 50. S. Lam et al., A randomized comparative study of the safety and efcacy of photodynamic therapy using PHOTOFRIN II combined with palliative radiotherapy versus palliative radiotherapy alone in patients with inoperable obstructive non-small cell bronchogenic carcinoma, Photochem. Photobiol., 46, 893-897, (1987). 51. H.I. Pass et al., Bronchoscopic phototherapy at comparable dose rates: early results, Ann. Thoracic Surg. 47, 693-699, (1989). 52. J. LoCicero, M. Metzdorff and C. Almgren, Photodynamic therapy in the palliation of late-stage obstructing non-small-cell lung cancer, Chest 98, 97-100, (1990). 53. PHOTOFRIN (Pormer sodium) package insert (approved labeling), Axcan Scandipharm, Inc. (2000). 54. P.A. Cowled, L. Mackenzie and I.J. Forbes, Potentiation of photodynamic therapy with haematoporphyrin derivatives by glucocorticoids, Cancer Lett. 29, 107-114, (1985). 55. J.M. Budinger, Untreated bronchogenic carcinoma, Cancer 11, 106-116, (1958). 56. L. Corti et al., Photodynamic therapy in gynecological cancer, Lasers Med. Sci. 4, 155-158, (1989). 57. I.J. Forbes et al., Phototherapy of human tumors using haematoporphyrin derivative, Med J. Aust. 2, 489-493, (1980). 58. A. Dahlman et al., Laser photoradiation therapy of cancer, Cancer Res, 43, 430-434, (1983). 59. J. Kennedy, HPD photoradiation therapy for cancer at Kingston and Hamilton, in Porphyrin Photosensitization, D. Kessel and T.J. Dougherty (Eds.), New York, Plenum Press, pp 53-62, (1983). 60. S.B. Lele et al., Photodynamic therapy in gynecologic malignancies, Gynecolog. Oncol. 34, 350-352, (1989). 61. R.V. Lobraico et al., Photodynamic therapy for cancer of the lower female genital tract, Colposcopy Gynecol. Laser Surg. 2, 185-199, (1986). 62. J.S. McCaughan Jr. et al., Photodynamic therapy of gynecologic neoplasms after presensitization with hematoporphyrin derivative, Lasers Surg. Med. 5, 491-498, (1985). 63. M.A. Rettenmaier et al., Gynecologic uses of photoradiation therapy, in Porphyrin Localization and Treatment of Tumors, D.R. Doiron and C.J. Gomer (Eds.), New York, Alan R Liss Inc., pp. 767775, (1984). 64. H. Soma and S. Nutahara, Cancer of the female genitalia, in Lasers and Hematoporphyrin Derivative in Cancer, Y. Hayata and T.J. Dougherty (Eds.), Tokyo, Igaku-Shoin, pp 97-109 (1983). 65. B.G. Ward et al., The treatment of vaginal recurrences of gynecologic malignancy with phototherapy following hematoporphyrin derivative pretreatment, Am. J. Obstet. Gynecol. 112, 356-357, (1982). 66. J.S. McCaughan, Jr., Photoradiation of malignant tumors presensitized with hematoporphyrin derivative, in Porphyrin Localization and Treatment of Tumors, D.R. Doiron and C.J. Gomer (Eds.), New York, Alan R Liss Inc, pp. 805-827, (1984). 67. W.M. Cai et al. Photodynamic therapy in the management of cancer: An analysis of 114 cases, in Methods in Porphyrin Photosensitization, Kessel (Ed.), New York, Plenum Press, pp 13-19, (1985). 68. J.A.S. Carruth and A.L. McKenzie, Preliminary report of a pilot study of photoradiation therapy for the treatment of supercial malignancies of the skin, head and neck, Eur. J. Surg. Oncol. 11, 47-50, (1985). 69. L.I. Grossweiner, J.H. Hill and R.V. Lobraico, Photodynamic therapy of head and neck squamous cell carcinoma: optical dosimetry and clinical trial, Photochem. Photobiol. 46, 911-917, (1987). 70. O.E. Schuller, J.S. McCaughan, Jr., R.P. Rock, Photodynamic therapy in head and neck cancer, Arch. Otolaryngol. Vol. 111, pp. 351-355, (1985).

Lasers in Photodynamic Therapy

321

71. C. Taketa and M. Imakiire, Cancer of the ear, nose, and throat, in Lasers and Hemtoporphyrin Derivative in Cancer, Y. Hayata and T.J. Dougherty (Eds.), Tokyo, Igaku-Shoin, pp. 70-78, (1983). 72. A.G. Wile et al., Photoradiation therapy of head and neck cancer, in Porphyrin Localization and Treatment of Tumors, C.J. Gomer and D.R. Doiron (Eds.), New York, Alan R Liss Inc, pp. 681-691, (1984). 73. F.Y. Zhao et al., Use of haematoporphyin derivative as a sensitizer for radiotherapy of oral and maxillofacial tumors: A preliminary report, Lasers Med. Sci. 1, 253-256, (1986). 74. F.Y. Zhao et al., Treatment of 510 cases of oral squamous cell carcinoma, Am. Acad. Med. 18, Singapore 533-536, (1989). 75. V.G. Schweitzer, Photodynamic therapy for treatment of head and neck cancer, Otolaryngol. Head Neck Surg. 102, 225-232, (1990). 76. V.G. Schweitzer and D. Visscher, Photodynamic therapy for treatment of AIDS-related oral Kaposis sarcoma. Otolaryngol. Head Neck Surg. 102, 639-649, (1990). 77. C. Freche and S. DeCorbiere, Use of photodynamic therapy in the treatment of vocal cord carcinoma, J. Photochem. Photobiol. 6, 291-296, (1990). 78. J.L. Gluckman, Hematoporphyrin photodynamic therapy: is there truly a future in head and neck oncology? Reections on a 5-year experience, Laryngoscope 101, 36-42, (1991). 79. B.L. Wenig et al., Photodynamic therapy in the treatment of squamous cell carcinoma of the head and neck, Arch. Otolaryngol.. Head Neck Surg. 116, 1267-1270, (1990). 80. M. A. Biel, Photodynamic therapy and the treatment of head and neck cancer, The Laryngoscope 108, 1259-1268 (1998). 81. A.G. Wile et al., Laser photoradiation therapy of cancer: an update of the experience at the University of California, Irvine, Lasers Surg. Med. 4, 5-12, (1984). 82. M.E. Tian, S.L. Qui and Q. Ji, Preliminary results of hematoporphyrin derivative-laser treatment for 13 cases of early esophageal carcinoma, Adv. Exp. Med. Biol. 193, 21-25, (1985). 83. M. Fujimaki and K. Nakayama, Endoscopic laser treatment of supercial esophageal cancer, Semin. Surg. Oncol. 2, 248-256, (1986). 84. F. Calzavara et al., Oesophageal cancer treated by photodynamic therapy alone or followed by radiation therapy, J. Photochem. Photobiol, B:Biology, 6, 167-174, (1990). 85. P. Monnier et al., Photodetection and photodynamic therapy of early squamous cell carcinoma of the pharynx, esophagus and tracheobronchial tree. Lasers Med. Sci. 5, 149-169, (1990). 86. R.J. Thomas et al., High-dose photoradiation of esophageal cancer, Ann. Surg. 206, 193-199, (1987). 87. J.S. McCaughan Jr., T.A. Nims and J.T. Guy, Photodynamic therapy for esophageal tumors, Arch. Surg. 124, 74-80, (1989). 88. A. Segalin et al., Surgical and endoscopic palliation of esophageal carcinoma, Ann. Thoracic Surg. 48, 267-271, (1989). 89. T. Patrice et al., Endoscopic photodynamic therapy with hematoporphyrin derivative for primary treatment of gastrointestinal neoplasms in inoperable patients, Digestive Dis. & Sci. 35, 545-552, (1990). 90. H.M. Likier, J.G. Levine and C. Lightdale, Photodynamic therapy for completely obstructing esophageal carcinoma, Gastro. Endosc. 37, 75-78, (1991). 91. M. Dal Fante, G. Bottiroli, and P. Spinelli, Behavior of hematoporphyrin derivative in adenomas and adenocarcinomas of the colon: a microuorometric study, Lasers Med. Sci., 3, 165-171, (1988). 92. R.C. Haggitt, Barretts esophagus, dysplasia, and adenocarcinoma, Hum. Pathol. 25, 982-993 (1994). 93. N. Barrett, Chronic peptic ulcer of the oesophagus and oesophagitis. Br. J. Surg. 38, 175-182 (1950). 94. S.S. Devesa, W.J. Blot, J.F. Fraumeni Jr., Changing patterns in the incidence of esophageal and gastric carcinoma in the United States, Cancer, 83(10), 2049-53, (1998). 95. P. Gillen et al., Experimental columnar metaplasia in the canine oesophagus, Br. J. Surg. 75, 1135, (1988). 96. G.W. Falk, Reux disease and Barretts esophagus, Endoscopy, 31(1), 9-16, (1999).

322

Lasers in Medicine

97. W.A. Williamson et al., Barretts oesophagus. prevalence and incidence of adenocarcinoma, Arch. Int. Med. 151, 2212-2216, (1991). 98. M. Miros, P. Kerlin, N. Walker, Only patients with dysplasia progress to adenocarcinoma in Barretts oesophagus, Gut 32(12), 1441-6, (1991). 99. A.J. Cameron et al., Prevalence of columnar lined Barretts oesophagus: comparison of population based clinical and autopsy ndings. Gastroenterology 99, 918-922, (1990). 100. C. Winters et al., Barretts esophagus: a prevalent occult complication of gastroesophageal reux disease, Gastroenterology 92, 118-124, (1987). 101. R.W. Sjogren, L.F. Johnson, Barretts esophagus: a review, Am. J. Med. 74, 313-21, (1983). 102. W. Hameeteman et al., Barretts esophagus: development of dysplasia and adenocarcinoma, Gastroenterology 96(5), 1249-56, (1989). 103. A.J. Cameron, B.J. Ott, W.S. Payne, The incidence of adenocarcinoma in columnar-lined (Barretts) esophagus, N. Engl. J. Med. 313(14), 857-9, (1985). 104. D.J. Drewitz, R.E. Sampliner, H.S. Garewal, The incidence of adenocarcinoma in Barretts esophagus: a prospective study of 170 patients followed 4.8 years, Am. J. Gastro. 92(2), 212-5, (1997). 105. A.P. Burke et al., Dysplasia of the stomach and Barretts oesophagus: a follow-up study, Modern Pathol. 4, 336-41, (1991). 106. B.F. Overholt, M. Panjehpour, J.M. Haydek, Photodynamic therapy for Barretts esophagus: followup in 100 patients, Gastro. Endosc. 49, 1-7, (1999). 107. H. Barr et al., The signicance of the nature of the photosensitizer for photodynamic therapy: quantitative and biological studies in the colon, Br. J. Cancer 62, 730-735, (1990). 108. H. Barr et al., Photodynamic therapy for colorectal cancer: a quantitative pilot study, Br. J. Surg. 71, 93-96, (1990). 109. M. Aida and M. Kawagulhi, Gastric cancer, in Lasers and Hematoporphyrin Derivative in Cancer Y. Hayata and T. Dougherty (Eds.). NY/Tokyo Igaka-Shoin, pp. 65-69, (1983). 110. Y. Hayata et al., Photodynamic therapy with hematoporphyrin in cancer of the upper gastrointestinal tract, Seminars Surg Onc. 1, 1-11, (1985). 111. S. Okuda et al., Experimental and clinical studies on HPO-photoradiation therapy for upper gastrointestinal cancer, in Porphyrins in Tumor Phototherapy, Adreoni, A. and Cubedda, R. (Eds.), NY/London, Plenum Press, pp. 413-421, (1984). 112. H. Tajiri et al., Photoradiation in early gastrointestinal cancer, Gastro. Endosc. 33, 88-90, (1987). 113. T. Nakamura et al., Photodynamic therapy for early gastric cancer using a pulsed gold vapor laser, J. Clin. Laser Surg. (Oct.), 63-67, (1990). 114. P.J. Muller, and B.C. Wilson, Photodynamic therapy of malignant brain tumors: clinical effects, postoperative ICP, and light penetration of the brain, Photochem. Photobiol. 46, 929-935, (1987). 115. R.N. Nambisan et al., Intraoperative photodynamic therapy for retroperitoneal sarcomas, Cancer 61, 1248-1252, (1988). 116. J.S. McCaughan Jr. et al., Photodynamic therapy to treat tumors of the extrahepatic biliary ducts, Arch. Surg. 126, 111-113, (1991). 117. W.F. Sindelar et al., Technique of photodynamic therapy for disseminated intraperitoneal malignant neoplasms: Phase I Study, Arch. Surg. 126, 318-324, (1991). 118. H. Pass et al., Intraoperative photodynamic therapy after resection of pleural malignancies, Abstract XX/5. 3rd Biennial Meeting, Int. Photodynam. Assoc., (1990). 119. L. Lofgren et al., Transthoracic endoscopic photodynamic treatment of malignant mesothelioma, Lancet 337, 359, (1991). 120. A.H. Kaye, and J.S. Hill, Photoradiation therapy of brain tumors: laboratory and clinical studies, in Phototherapy of Cancer, Morstyn, G. and Kaye, A. (Eds.), New York, Harwood Academic Publishers, pp. 101-118, (1990). 121. P.J. Muller and B.C. Wilson, Photodynamic therapy for recurrent supratentorial gliomas, Semin. Surg. Oncol. 11, 346-354 (1998)

Lasers in Photodynamic Therapy

323

122. C. Perria et al., First attempts at the photodynamic treatment of human gliomas, J. Neurosurg. 24, 119-129, (1980). 123. E.R. Laws et al., Photoradiation therapy in the treatment of malignant brain tumors: a Phase I (feasibility) study, Neurosurg. 9, 672-678, (1981). 124. G.A.J. McCullough et al., Phototherapy in malignant brain tumors, in Porphyrin Localization and Treatment of Tumors, D.R. Doiron and C.J. Gomer (Eds.), New York, Alan R Liss Inc, pp. 709-717, (1984). 125. P.J. Muller and B.C. Wilson, Photodynamic therapy of malignant brain tumors, Lasers Med. Sci. 5, 245-251, (1990). 126. H. Kostron, E. Fritsch and V. Grunert, Photodynamic therapy of malignant tumors: a phase I/II trial, Br. J. Neurosurg. 2, 241-248, (1988). 127. S.K. Powers et al., Stereotaxic intratumoral photodynamic therapy of recurrent malignant brain tumors, Neurosurgery 29, 688-695 (1991). 128. C. Perria, G. Casu and E. Sgaramella, Proposal of a protocol for the photodynamic therapy of malignant brain tumors, Photochem. Photobiol. 6, 443-445, (1990). 129. D.T. Tse, R.C. Kersten and R.L. Anderson, Hematoporphyrin derivative photoradiation therapy in managing nevoid basal-cell carcinoma syndrome. Preliminary report, Arch. Ophthalmol. 102, 990994, (1984). 130. D.T. Tse et al., Hematoporphyrin photoradiation therapy for intraocular and orbital malignant melanoma, Ophthalmology 102, 833-838, (1984). 131. T.W. Sery et al., Photodynamic therapy of human ocular cancer, Ophthalmic Surg. 18, 413-418, (1987). 132. A.L. Murphree, M. Cote and C.J. Gomer, The evolution of photodynamic therapy techniques in the treatment of intraocular tumors, Photochem. Photobiol. 46, 919-923 (1987). 133. R.A. Bruce and J.S. McCaughan, Lasers in uveal melanoma. Ophthalmol. Clin. N. Am. 2, 597-604, (1989). 134. Y. Ohnishi, Y. Yamana and M. Minei, Photoradiation therapy using argon laser and a hematoporphyrin derivative for retinblastoma: a preliminary report, Jpn. J. Ophthalmol. 30, 409-419, (1986). 135. C. Bandieramonte et al., Laser phototherapy following HpD administration in supercial neoplastic lesions, Tumori 70, 327-334, (1984). 136. D. Gilson et al., Therapeutic ratio of photodynamic therapy in the treatment of supercial tumors of skin and subcutaneous tissues in man, Br. J. Cancer 58, 665-667, (1988). 137. J.S. McCaughan Jr. et al., Photodynamic therapy for cutaneous and subcutaneous malignant neoplasms, Arch. Surg. 124, 211-216, (1989). 138. D.J. Pennington, M. Waner and A. Knox, Photodynamic therapy for multiple skin cancers, Plastic& Reconst. Surg. 82, 1067- 1071, (1987). 139. P.J. Robinson, J.A.S. Carruth, and G.M. Fairris, Photodynamic therapy: a better treatment for widespread Bowens disease, Br. J. Derm. 199, 59461, (1988). 140. M. Schuh et al., Photodynamic therapy for palliation of locally recurrent breast carcinoma, J. Clin. Onc. 5, 1766-1770, (1987). 141. S.N. Waldow et al., Photodynamic therapy for treatment of malignant cutaneous lesions, Lasers Surg. Med. 7, 451-456, (1987). 142. B.W. Wilson et al., Use of photodynamic therapy for the treatment of extensive basal cell carcinomas, Facial Plast. Surg. 6, 185-189, (1990). 143. P.W. Sperduto et al., Photodynamic therapy for chest wall recurrence in breast cancer, Int J. Radiation Oncol. Bio. Phys. 21, 441-446, (1991). 144. D. Kessel, In vitro photosensitization with a benzoporphyrin derivative. Photochem. Photobiol. 49, 579-582 (1989). 145. H. Lui et al.. Photodynamic therapy of malignant skin tumors with benzoporphyrin derivative monoacid ring A (BPD-MA). Proc. Am. Soc. Photobiol. (Meeting Abstract), in press (1993). 146. S.L. Fine et al., Age-related macular degeneration, N. Eng. J. Med. 342, 483-492 (2000)

324

Lasers in Medicine

147. VISUDYNE (Verteporn for Injection) Package Insert (Approved Labeling), QLT Inc. and CIBA Vision, Inc. (2000). 148. J.C. Kennedy and R. Pottier, Endogenous protoporphyrin IX, a clinically useful photosensitizer for photodynamic therapy. J. Photochem. Photobiol. B: Biol 14, 275-292 (1992). 149. P. Wolf, E. Rieger and H. Kerl, Topical photodynamic therapy with endogenous porphyrins after application of 5-aminolevulinic acid. J. Am. Acad. Dermatol. 28, 17-21, (1993). 150. S.D. Shanler et al., Localization of endogenous protoporphyrin IX and effects of photodynamic therapy in human and murine carcinomas. Proc. Am. Assoc. Cancer Res. Abstract #2161, 34, 363 (1993). 151. LEVULAN KERASTICK Package Insert (Approved Labeling), DUSA Pharmaceuticals, Inc. and Schering, AG/Berlex Laboratories (2000). 152. J. Bedwell et al., Fluorescence distribution and photodynamic effect of ALA-induced PPIX in the DMH rat colonic tumor model. Br. J. Cancer 65, 818-824 (1992). 153. C.S. Loh et al., Photodynamic therapy of the normal rat stomach: A comparative study between di-sulphonated aluminum phthalocyanine and 5-aminolevulinic acid. Br. J. Cancer 66, 452-462 (1992). 154. L. Gossner et al., Photodynamic ablation of high-grade dysplasia and early cancer in Barretts esophagus by means of 5-aminolevulinic acid. Gastroenterology, 1998; 114:448-55. 155. R. Ackroyd et al., Photodynamic therapy for dysplastic Barretts oesophagus: a prospective, double blind, randomized, placebo controlled trial. Gut. 2000; 47:612-7 156. S.L. Marcus et al., Photodynamic therapy (PDT) and photodiagnosis (PD) using endogenous photosensitization induced by 5-aminolevulinic acid (ALA): Current clinical and development status. J. Clin. Laser Med. Surg. 1996;14:59-66. 157. R. Ackroyd et al., 5-Aminolevulinic acid photosensitization of dysplastic Barretts esophagus: a pharmacokinetic study. Photochem. Photobiol. 1999; 70:656-62. 158. C.J. Kelty et al., Photodynamic therapy for dysplastic Barretts oesophagus: Long term follow up. Br. J. Surg. 2001; 88:(in press). 159. H. Messman et al., Enhancement of photodynamic therapy with 5-aminolaevulinic acid-induced porphyrin photosensitization in normal rat colon by threshold and light fractionation studies. Br J. Cancer 1995;72:589-94. 160. A. Curnow, J.C. Haller and S.G. Bown, Oxygen monitoring during 5-aminolaevulinic acid induced photodynamic therapy in normal rat colon. Comparison of continuous and fractionated light regimes. J. Photochem. Photobiol. Biology 2000;58:149-155.

Index
A
3-amino-1,2,4-triozole, 182 abdominal space, 291 ablation, 109, 112 114, 116, 120, 123 127, 138, 164, 171, 231, 234, 236, 256-260, 266, 278 arrhythmogenic foci, 271 plume, 122, 260 threshold uence, 110-111, 118-119, 122, 125, 128 absorbance, 90, 138 absorption, 172, 231, 252 cellular, 92 coefcient, 111-117, 120,122, 135, 217, 233 cross section, 117 deoxyhemoglobin, 162 depth, 217 Dopa-Melanin, 94 mammalian cells, 94 molecular, 91, 112 multiphoton, 50, 113, 257 oxyhemoglobin, 94, 162 spectra bands, 179, 182, 215 tissue, 95, 137, 275 two photon absorption, 51, 112 urocanic acid, 94, 102 water, 137 absorptivity, 112, 253 acetylcholine, 190 acidication, 184 acidosis, 189, 201 acridine orange, 193 actinic keratoses, 315 action, spectra bands, 179 spectrum, 175, 198 spectroscopy, 96, 102 acousto-optical, 125 acuity, 239 adenosine triphosphate (ATP), 172-174, 182, 184, 186187, 192, 199 adinosine dinucleotide (ADP), 173, 182 adrenaline, 190 aerobic, 183 albedo, 29, 31-36, 62-63 ALA, (d-amino levulinic acid) 145-147, 149, 159, 314317 amino acids, 112-113, 115 amplier, Raman, 162 anaerobic, 183 anastomoses, 269, 274, 277 arterio-arterial, 277 anesthesia, 163 aneurysms, 276-277 angina, 269 class, 270 angiogenesis, stimulated, 270 angioplasty, 155, 262, balloon, 262-265, 267 coronary(also PTCA), 262 laser, 248, 250, 252, 255, 260, 263, 264-265, 267, 271 transluminal, 152, 262-265 percutaneous, 153, 157 percutaneous transluminal (PTA), 262, 264-265 peripheral, 262 anisotropic factor (g), 35 aorta, 113, 250, 268 aortotomy, 276 arabinose, 191 aromatic ring, 112 arrhythmias, 269 arterial spasm, 153 arterioles, retinal, 213 arteriotomy, 276 artery, 258, 260 brachial, 249, 251 coronary, 151, 155, 250, 262 corotid, 275 femoral, 249, 251 peripheral, 262 pulmonary, 250, 272 retinal, 239 astigmatism, 227, 256 astrocytoma, 309 atherosclerosis, 151150, 262 atherosclerotic, plaque, 248, 252 atrium, left, 249 right, 249 attenuation coefcient, 28, 137 autocrine, 192
B

backscatter factor, 62 bacteria, 92, 97 bacteriorhodopsin, 172

325

326

Lasers in Medicine

balloon dilation, 262, 271 basal cell cancer, 142,145, 159 basiloma, 147 basophils, 196 Barretts esophagus, 305, 316 Beers Law, 28, 36, 42, 90, 113, 116, 118, 139 bending bend radius, 18 benzene, 112 bladder, 145, 290 bladder cancer, 141-142, 145, 291 supercial, 294 blackbody, 51 blood, 122, 155, 157, 158, 196 clots, 151 vessels, 220, 239, 248, 254 Boltzman constant, 73 distribution, 1 bonds, peptide, 112 bone, 161 bowel wall, 306 Bowens disease, 311 Bowmans layer, 215, 233 brachytherapy, 229. 310 brain, 213 bronchoscopy, 145, 163 bronchus, 289 breast carcinoma, 144 bubble formation, 124

C
Ca, 153 Ca+ , 153 intracellular, 186 Ca2+, 182, 193 cadmiun chloride, 186 calcium, 151, 260 calcium phosphate, 260-261 calcium salts, 266 cancer, bladder, 291 cervical 300 colorectal, 305-306 endobronchial lung, 296, 300 endometrial, 300 esophageal, 289, 304 gastrointestinal, 304 gynecological, 300- 301 head, 302, 317 larynx, 302 lip, 302 neck, 302, 317 non-small-cell lung, 296, 300 oral cavity, 302 ovarian, 300

pharynx, 302 skin, 302 supercial bladder, 294 cancer detection endoscopic, 133, 300 capsulotomy, posterior, 232 carbohydrates, 113 carcinogenesis, 102-103, 229 carcinoma, basal cell, 103, 146, 159, 238, 311, 313, 315 breast, 144, 313 conjunctival, 238 ductal breast, 162 in situ, 142, 291, 294, mucosal, 316 refractory, 296 squamous, 103, 146, 157, 302, 311, 315 cardiologist, pediatric, 251 cardiomyocytes, guinea pig, 193 cardiomyopathy, 272 cardiovascular, applications, 247 catalytic activity, 188 cataract, 211, 220, 239, 310 formation, 230 surgery, 228 cataractogenesis, 229 catheter, 248, 260, 262 balloon, 249 beroptic, 248 intravesical laser, 294 laser delivery, 249 cell death, 100 endothelium, 151, 193 epithelial, 232 human HeLa, 93 membrane effects, 101 monkey kidney, 93 mutation, 100 CCD, 149 near IR, 157 chamber, anterior, 213 posterior, 213 ventricular, 250-251 chemiluminescence, luminol-amplied, 182, 193, 196 chemotherapy, 294, 305 chopper acousto-optic, 50 beam, 54 charge-coupled device, 57 mechanical, 50 wheel, 144

Index

327

cholangiocarcinoma, 306 cholesterol, 151, 157 chlorin, 139 chlorophyll, 96, 171-172 chromatin, 193 chromophores, 95, 101-102, 112, 115, 117, 118, 120, 152, 162, 172, 178, 214, 289 cardiovascular, 248 endogenous, 145 chromatography, 123 choroid, 213, 310 choroidal neovascularization, 314 C. albicans, 102 cAMP, 178, 186 C2, 123 ciliary body, 213, 224 circadian rhythms, 198 clonogenic activity, 181 CN, 123 CO2, 123 CO, 123 coagulation, 171, 231 coherence, temporal, 239 collagen, 115,120,139, 151, 154,154, 192, 227, 267, 275, 277 brils, 212, 226 lenticules, 227 submucosal, 306 concentration molar, 113 conservation of mass, 127 contractile function, 190 compressive stress, 127 colon, 305 tumor, 145 cones, 213 conization, 300 coronary artery, left anterior descending, 269 cornea, 115,120, 123, 125-126, 211-213, 215, 222, 226, 230, 233, 237-238 corneal sculpting, 234, 256 cough, 300 cryotherapy, 229, 300, 311 cutting, 171 CuA (dinuclear copper), 174, 176, 181 CuB (dinuclear copper), 175-176, 179-181 cyclophotocoagulation, 224, 231 cysteins, 175 cytochrome c oxidase, 174-176, 180-182, 184, 198-200 cytokines, interleukin-1, 294 tumor necrosis factor, 294 cytoplasm, 98, 184, 186 cytosol, 174 cytotoxic action, 289

D
decay, constant, 163 time, 156 density, energy, 217, 224, 252 power, 217, 252, 255, 257 dentists, 196 detection endoscopic cancer, 135 modulation depth, 160 phase shift, 160 detectors indium gallium arsenide, 52 intensied diode array, 143 intensied matrix, 149 light, 51 photodiode, 51 photomultipliers, 52, 53, 144 photothermal, 51 silicon, 52 dermatologists, 196, 256 dermatology, 137 dermis, 95 Descemets membrane, 215 delayed contact hypersensitivity, 102 delivery systems, 230, 236, 264, 287, 289 denaturation, 121 dichroic beam splitter, 144 mirror. 141 diffuse reectance, 67 diffusion approximation, 34-35, 40, 42-43 constant, 40 diaphanography, 161 di-hematoporphyrins ether (DHE), 238-239, 287 diffraction grating spectrograph, 57 diopters, 213, 226 disection, 153 dispersion, 163 dissociation, 118 DNA, 87, 92-93, 96-100, 113, 173, 176, 183-184, 187 Doppler effect, 140, 158 laser, 140 perfusion monitoring, 136 dosimetry light, 294 optical, 47 probe, 294 drug, 287 safety, 288 dye, 276 (also see laser, dye) indocyanine green, 276 dysphagia, 289, 304-305

328

Lasers in Medicine

dysplasia, 145, 192 dyspnea, 296, 300 dysuria, 291

E
edema, 301, 314, 316 elbow, 249 elastin, 139, 151, 154, 156 emmetropic eye, 213 endarterectomy, 271, 273-274 aortic, 273 iliac, 273 poplitea-posterior tibial, 273 profunda femoral, 273 thrombo, 273 supercial fundus, 273 endothelium, 215, 227 cell, 149, 155 endoscope, 143, 291, 316 energy, density, 217 threshold, 114, 125 enucleation, 229 enzyme, activation, 181 antioxidant, 182 cardiac, 269 fast, 175 inhibition, 181 native, 195 slow, 175 eosinophils, 196 epidermis, 95 epithelium, 215, 227, 233, 235 iris pigment, 214 retinal pigment, 214, 219 erythema, 301, 314, 316 erythrocytes, 186 Escherichia coli, 181, 192-193, 195, 200 esophagus, 289 Barretts, 305 eukaryotic cells, 174, 185, 192 extinction coefcient, molar 112-113, 117 extracellular pH, 190

focal volume, 15 ber, bundles, 19 chalcogenide, 23 cleaved, 304, 306, 308 connectors, 19 contact tips, 19 cylindrical-diffuser, 289-290 cylindrical microlens-tipped, 300, 304 damage, 19 uoride glass, 22 infrared tramsmitting, 22 launching energy into, 19 metal-tipped, 255 microlens, 304 optics, 16, 232, 257 polychrystalline, 23 silver halide, 23 single crystal sapphire, 23 spherical diffuser-tipped, 291, 294 tips, 289 brils, collagen, 212 broblasts, 173, 178, 184, 192-193, 201 human skin, 185 brinogen, human, 276 stula, 222 avins, 183 avoproteins, 183-184 uence, 48, 113 rate, 27, 50, 61 threshold, 114 uorescence, 138, 141, 149, 153, 157, 229, 288, 306, 315-316 auto, 139, 142-145, 153, 157 endogenous, 145 imaging, 149 probe-beam laser induced, 124,136, 139-141, 153 uorescent probes, 60 Fluorine, 235 uorosenser, 143 uoroscopy, 249 free radical, formation, 257 reactions, 196

F
FDA, 256, 264-265, 278 femtosecond lasers, 25 Ferrocytochrome c, 174 ber lasers, 25 eld, visual, 212 lters, 53 interference, 144 f-number, 15

G
gastroesophageal reux (GERD), 305 gastrointestinal lesion, 316 Gaussian, beam prole, 49 pulses, 49, 50 gating frequency doubling, 162 glaucoma, 211, 221-222, 224, 240, 310

Index

329

open-angle, 222 glioblastoma, 307, 309 glioma, 308-309 glucose deprivation, 186 glue, 276 glutathione (GSH), 99-100, 185 glutinin, 193 glycerol, 190 glycolysis, 190 graft, saphenous vein, 277 groin, 249 growth rate, 193 Gruneisen stress, 122 GSSG, 185 guide wire, 249, 267 Gulstrands, schematic eye, 213

aqueous, 211, 213, 221-222, 224 vitreous, 211, 213, 311 hydrogen peroxide (H2O2), 186 hyperopia, 234 hyperopic, 214 hyperplasia, 266 hyperthermia, 229 interstitial laser, 69, 73 hypertrophy, 250 hypoxia, 189, 201

I
illuminating, 231 intima, 149 intensier image, 144 interference fringes, 239 intrathoracic space, 291 imaging laser ash, 123 multi-spectral, 136 radiological, 69-70 time-gated, 162 incubation, 128 indium-antimonide, 70 indocyanine green, 39 infrared (IR), 136-137, 157, 172, 193, 200-201, 214 ink, 163 inorganic salts, 264 imaging, 215, 231 uorescence, 149 multi-spectral, 136 imidazole, 175 immune effects, 101-103 immunotherapy, 294 impulse coupling coefcient, 125-126 specic, 125, 127 inammation, chronic, 189 intracellular pH, 189-190 intraocular, lens, 213 pressure, 213, 222 structures, 215 integrating spheres, 53 intima, 151 Intralipid, 163, 307 Iodine, 263 iodoacetamine, 186 ion exchange, 190 ionization, 257 ionizing radiation, 157, 164, 229 iridecrtomy, 231-232 iris, 213, 221 iron-sulfur center, 174, 183 irradiance, 28, 30,32

H
hand, 161 Hanks solution, 178 haptic loops, 232 heart failure, congestive, 269 heat diffusion, 1, 10 heating, 164 HeLa cells, 173, 181, 183-184, 186, 188, 199 Helix pomatia, 186, 193 hemangioma, 138 hematocrit hematoma, 277 heme heme a, 174-175, 183 heme a3, 174-175, 183 oxygenase, 186 hemoglobin, 136-139, 145, 157-160, 163, 214, 221, 289 oxygenated, 139 hemoptysis, 296 HPD (hematophorphyrin derivative), 138-142, 144145, 157, 287-288, 296, 300, 302, 304-305, 309-310 Herpes simplex, 102 heterodyne, 102, 140, 163 histidine, 175 HIV, 102 HO (hydroxyl), 182 H2O, 123 H2O2, 175, 182 HCN, 123 holography light-in-ight, 163 homeostasis, calcium, 190 cellular, 184, 186, 200 hot-tips, 262-264 humor,

330

Lasers in Medicine

continuous, 28 pulsed, 28 irradiation, dichromatic, 176, 198 ischemia, 190. 201, 220, 270 isobestic point (805nm), 162

K
keratinocytes, 192-193, 201 keratoplasty, 226

L
lamella, internal elastic, 273 Laser Alexandrite (Cr:BeAl2O4), 9 argon uoride (ArF), 109, 111-113, 115-116, 123125, 128, 233-234, 254, 256 argon-ion, 137, 161, 211, 220-221, 224, 230-231, 238, 248, 255, 262, 271, 277, 290, 300, 309, 314 carbon dioxide, 6, 8, 224-225, 227, 248, 262, 268, 270271, 273, 277 cavity-dumped, 159 copper vapor, 290-291 continuous wave (cw) damage, 19 diode, 158, 161-163, 186, 195-196, 211, 224, 230231, 276, 317 dye. 137, 211, 220-221, 224, 230, 239, 290-291, 300, 314 endo systems, 229 erbium YAG, 6, 217, 225, 232, 254 erbium YSGG, 217, 271 excimer, 211, 235, 254, 257, 264-265, 270, 290-291 free electron, 14 femtosecond, 181 gold vapor, 290, 307, 309 HF (hydrogen uoride), 225 holmium, 6, 226, 254, 269 helium-neon, 158, 176, 181-184, 186, 188, 190, 192194, 198-199, 233, 239, 289 infrared, 6 krypton, 220-221, 224, 230-231 krypton uoride (KrF), 115-116, 123 mid-infrared, 211, 224-225, 229 mode-locked, 156 near-ir (NIR), 158, 224 neodymium, 1, 157, 211, 224, 230-233, 235, 237, 248, 255, 262, 267, 273, 275, 291, 296, 300, 304, 306 nitrogen, 143, 149, 156 picosecond, 161 power, 15 probe, 222 ruby, 1, 219

semiconductor, 198, 231 TmHoCr:YAG, 222, 226, 269, 271 titanium, 9 uv ablation, 109 vaporization, 311 xenon chloride (XeCl), 109, 111, 121-122, 124-125, 229, 273 lamp, slit, 233 xenon arc, 220 laparatomy, 307 La Places law, 277 Larynx, 302 lens, 211, 230 capsule, 228 cataractous, 228-229 interocular, 213, 232 posterior capsule, 212 lesion atherosclerotic, 153 brotic, 154-155 LEVULAN, 288, 314-317 ligand, 175 light scattering, 134 light source, 287 limbus, 222 lip, 302 lipids, 112-113, 151 lipofusion, 215 liver, 288 lock-in amplier (LIA), 54, 56 lung, cancer, 291 tumor, 145 lymph nodes, 300, 304-305 lymphocytes, 193-194 lymphoma, murine cells, 101 T-cells, 149

M
macular degeneration, 221, 314 age-related, 314 hemorrhagic, 314 serous, 314 macular edema, 220, 231 region, 215, 220 macrophages, mouse peritoneal, 195 magnesium, 174 malignancies, gynecologic, 300 malignant, 238, 288 tumor diagnostics, 136, 145 mamalian cells, 92. 96, 189, 192 mammography

Index

331

optical, 135 x-ray, 162 materials organic, 113 melanin, 136, 212, 214-215, 289 melanoma, 103, 238 malignant, 310 ocular, 310 membrane cellular, 187 mitochondrial, 289 plasma, 186, 289 retinal, 232 subretinal neovascular, 221 menadione, 186 mercury-cadmium telluride, 70 mercury lamp, 144 mesothelioma, 307 metabolism, 185 metabolic pathways, 172 metal-ligand systems, 173 metaplasia, intestinal, 305 metastasis, bladder cancer, 294 breast carcinoma, 144 methionine, 175 Methylene blue, 188, 190 Microwaves, 229 mid-infrared, 215 miniaturization, 231 minimally invasive, 69-70 mirror, dichroic, 143 mitochondria, 172, 182, 184-185, 187, 199, 201 matrix, 174 membrane potential, 173, 184 optical properties, 173 mitogenic signals, 185 modulation transfer function (MTF), 239 Mohs micrographic surgery, 311 monochromatic, 172 monomers, 118 monochromator, 53 monocytes, 196 mucosa, 291 Barretts, 316-317 cervical, 302 colonic, 316 deep gastric, 306 necrosis, 317 vaginal, 302 multichannel analyzer optical, 143 multiphonon absorption, 50 edge, 50 excitation, 255

processes, 110 murine spleen cells, 193 muscle, cardiac, 251 musculoskeletal, aches, 196 pains, 196 mutation, 97, 100-101 mutagenesis, 98, 99, 160 myocardium, contractility, 269 ischemic, 268-269 perfusion, 269 myelocytes, 196 myoglobin, 272 myopia, 256 myopic, 213, 235 myomectomy, 250, 271 myotomy, 250, 271

N
nanosecond, 233 NAD, 139, 142, 185 NAD+/NADH ratio, 190 NADH, 139, 142, 173, 182, 184-185, 200 NADP, 185, 188 NADPH, 185, 188 National Cancer Institute (NCI), 307, 313 National Institutes of Health (NIH), 307 necrosis, 270, 301 ischemic, 288 nerve Bundle, 239 optic, 213 regeneration, 197 neurons, depolarization, 186 rat hippocampus pyramidal, 194 rat spinal cord, 194 neutrophils, 196 nevus syndrome, 311, 313 nitric oxide (NO), 195 nitrogen laser, 143, 149, 156 nucleic acid, 112, 136 nucleus, 186 numerical aperture, 16, 58

O
O2 , 172 occlusion, 262, 265 ocular, anatomy, 212 physiology, 212 opacication, lens capsule, 211, 240 ophthalmology, 137, 211, 256

332

Lasers in Medicine

ophthalmoscope, 230, 240 scanning laser (SLO), 239-240 optic nerve, 213 optical biopsy, 136 breakdown, 231 density, 90 mammography, 133 multichannel analyzer, 57 optically thick, 21 optimal dose, 294 organic, 117-118 organelles, 92 ostial lesions, 264-265 oxidation, 257 oxygen, 175, 181, 288-289 molecular, 287 singlet state, 138, 183, 288-289 tension, 189 oxygenation, 157, 163

P
pacemaker, electrode leads, 265 pain attenuation, 197 palliation, malignant dysphagia, 304 peptide, bonds, 92, 112-113, 115, 120 carbonyl, 175 perforation, vessel wall, 153, 262-263 perfusion, 158 peristalsis, 306 PHA, 194 phacoemulsiers, 229 pharynx, 302 photo ablation, 233, 235-236 absorbing, 171 acceptor, 171, 172,184, 198, 200 acoustic, 121-122, 125 activating, 146, 238 activating, retinal, 71 activating, enzyme, 103 bleaching, 289, 310 chemistry, 164, 171, 199, 201, 212, 254, 274 chemotherapy, 287 coagulation, 69, 215, 217, 219-221, 224, 229-232, 240, 310-311 cutting, 224 decomposition, 122, 255-257, 261, 264 dematosis, 315 disruptive, 212, 215, 221, 231-233, 255 dynamic therapy (PDT), 69,138, 145, 147, 159, 163, 171, 185, 229, 238-239, 287-291, 294, 296, 300-302, 304, 305, 307, 309-311, 313-317

etching, 233 excitation, 181, 198 heating, 217 multipliers, 144 pigments, 215 radiation, 308 refractive kerotectomy (PRK), 234 receptor, 172, 201, 213 sensitivity, 195, 287, 304, 310 sensitizers, 162, 238, 287, 291, 294, 306-307, 311, 313-315 stimulation, 192 synthetic reaction, 181 thermal, 212, 217, 221, 224, 231 toxicity, 238 vaporization, 217 welding, 227 Photofrin, 138, 142, 144, 144-145, 149, 157, 287-290, 294, 296, 300, 302, 304, 307, 309, 311, 313314, 317 photon counting, 158-159, 161 density, 119 phtallocyanide, 238 phtalocyanines, 139 physiotherapists, 196 phytochrome, 172 picosecond, 233 piezoelectric, 125, 256 pigments, 95,100 Planks Radiation law, 70, 151-150, 152,156 plaque. 135, 149-150, 152, 157, 260-262, 267, 271, 273 plasmacytes, 196 polymethylmethacralate (PMMA), 115, 124, 128 porphyria, 315 porphyrin, 173, 183 port-wine stains, 135 polyimide, 116, 120-125,128 power density, 48 pressure, 257, 260 interocular, 212-213, 222 piezoelectric sensor, 256 prokaryotic cells, 192 prosratic gland, 143 proteins, 92-93, 100,112-113, 120, 136, 173, 183, 275-276 stress, 186 synthesis, 195 proton motive force, 184 protoporphyrin, 145, 314 psoralens, 171 pulse duration, 224 pulsed photothermal radiometry (PPTR), 68 pump sources chemical reactions, 5 discharge, dc, 4 discharge, rf, 4 electron beam, 5

Index

333

electric current, 5 ashlamp, 3 laser, 4 nuclear, 5 pupil, 213 PUVA, 171-172 pyrimidine dimer, 96, 98, 100 pyrometer, 70

Q
quantum efciency, 238 quantum yield, 90

R
radiance, 48 radiation, hardening, 128 ionization, 229 therapy, 302 radicals, 123 radiant exposure, 28 radiation therapy, 311 radio frequency, 229 radiologic imaging, methods, 79 Raman amplier, 162 spectroscopy, 137, 141, 157 Rayleigh scatter, 21 reciprosity rule, 200 rectum, 305 redox, chain, 172 chemistry, 174, 181 light-activated, 187 shift, 190 state, 185-186, 189-190, 192, 200-201 reectance, spectroscopy, 137, 157 reection, 29 Fresnel, 17, 29-30 Specular, 29, 32 total internal, 16 reectometry, spatial domain, 163 total internal, 16 refractive index, 113, 212, 215 rehabilitation clinics, 197 resonators, 6 stable, 6-7 unstable, 6-7 resonant cavity, 1 resistive temperature detectors, 74 resolution,

spatial, 148 spectral, 148 restenosis, 264-265 reticuloendothelial system, 288 retina, 213, 215, 219-220, 238-239 retinal detachment, 310 retinaoblastoma, 229, 311 retinopathy, proliferative diabetic, 220 revascularization, transmyocardial laser (TMLR), 268 rheumatologists, 196 rheumatism, 190 rhodamine 123, 173 rhodopsin, 171-172, 215 RNA, 113, 173, 193-194 rose-bengal, 238 rRNA, 194

S
saturation, 180 scattered light, 28 scattering, coefcient, 163 elastic, 139, 157-160 Raman, 140 Schwann cells, 178 sclera, 212, 222 Seebeck, voltage, 74 coefcient, 74 Seldinger technique, 249 sensors, integrated circuit, 74 sensitivity, 140, 178 sensitization, 231 sensitizers, 148 septa, interatrial, 250 interventricular, 250 septostomy, 250-251, 271-272 atrial, 272 septum, 250 interventricular, 251 posterior ventricular, 249 shock wave, 125-127, 231, 248, 256-257, 261, 265, 270-271 shrinkage, 231 single photon counting, 60 sinus, coronary, 268 coronary retroperfusion, 268 simulations Monte Carlo, 29, 31, 34-36, 38, 65 slitlamp, 221, 223, 230 Soret band, 176 spasm, 151

334

Lasers in Medicine

spleen, 288 Snell's law, 16, 29, 31 solder, 276-277 specicity, 140. 157 spectrometer, 143 spectrophotometer, double beam, 178 spectroscopy uorescence, 240 Fourier transform, 156 infrared, 123 mass, 123 31P magnetic resonance, 309 Raman, 135, 141, 157-157, 240 FT Raman, 240 reectance, 137, 157 time of ight, 124 time-resolved, 160 specular reection, 29, 33, 41, 58 sports medicine, 197 Steinhart-Hart equation, 75 stain, port-wine, 138 stenosis, 277 aortic, 271 subaortic hypertrophic, 272 mitral, 271 stents, 266 stimulation, lymphocyte, 294 macrophage, 294 squamous cell carcinoma, 146, 157 strain-to-failure, 18 stress, oxidant, 186 protein, 186 stroma, 215, 226, 260 corneal, 226, 235 Stefan-Boltzmann, law, 70 constant, 70 stratum corneum, 95 streak-camera, 162 superoxide, 182 dismutase, 188 surgery, general, 137 cataract, 227 corneal, 227 glaucoma, 227 ophthalmic, 240 vitreoretinal, 227 sutures, 232, 277 systemic, 146

temperature probes ber optic, 77 thermal conductivity, 41, 126 damage, 229, 233, 253, 271 diffusivity, 121, 217 relaxation time, 110, 122, 217 response, 41 thermography, 69 thermotherapy, interstitial, 69 thermisters, 74-76 thermocouples, 73-74, 76 thoracoscope, 270 thoracotomy, 307 thrombectomy, 274 thrombosis, 248, 265, 274 thromboxane, 288 tissue, 113 myocardial, 155 optical properties, 289 tomography, computed, 309 topical, application, 146 trabeculoplasty, 221, 231 trabecular meshwork, 213-214, 221 transcleral, 222 transverse mode, 49 transillumination, 157-158, 160-161 transmyocardial, 271 transurethral, laser coagulation, 72 trephine, 224 triplet state, 138 trypan blue, 178 tryptophan, 112 tumor, 149, 159, 161, 306 colon, 145 cutaneous, 311 endobronchial, 296 endotracheal, 296 esophageal, 304 intrachranial, 310, 317 malignant, 138, 144, 157 malignant brain, 309 malignant intercranial, 307 margins, 288 papillary, 291 subcutaneous, 311 tyrosine, 112

U
ultrasonic, probes, 229 ultrasound, intraluminal, 264

T
tamped conditions, 126

Index

335

ultraviolet, 85-87, 98, 100, 102-105,109, 113, 115, 120, 122,126,138, 152, 172, 229, 234, 254-257, 260, 270, 274, 306 A (UV-A), 85, 88-89, 95-100, 102-104, 171, 186 absorption, 89-90 B (UV-B), 85, 87,89, 92, 97-100, 102-104 C (UV-C),85,87. 89, 92, 94, 96-100, 102-104, 234-235 direct effects, 96 uorescent sources, 88-89 germicidal sources, 88-89 halogen lamps, 89 ionizing, 91 indirect effects, 96 laser, 88 non-ionizing, 91 solar simulators synchrotrons, 88-89 sources, 88-89 terestrial solar, 87-89, 102 vaccum (VUV), 85, 87, 89, 94, 96-100, 102-103 umbonate, 94 urethera, 291 urinary bladder, 291 uvea, 214 uveal, melanoma, 229 tract, 213

cavity, 268 left, 249 right, 249, 271 wall, 268 vibrational, shifts, 156 transitions,157 viruses, 92 visible, 172, 186, 200-201, 214 vision, 215 visual eld, 212 VISUDYNE, 288-289, 314 vitreous, 232 hemmorage, 231 humor, 212-213 vocal cords, 302 vulva, 300, 302 cancer of, 300

W
waveguides, hollow, 24 wavelength, 252 water, 112, 136, 215 Wein Displacement Law, 70 welding, 231, 271, 274-277 dermatology, 227 neurology, 227 otolaryngology, 227 urology, 227 vascular, 227 wound healing, 189, 197

V
vagina, 300, 302 cancer of, 300 valve, aortic, 249 mitral, 250 prostheses, 271 tricuspid, 250 valvuloplasty, 271 valvulotomy, 271 pulmonic, 251 tricuspid, 251 vaporization, 69, 171 veins, 239, 250 pulmonary, 249 vein grafts, 265 velocimetry, laser Doppler (LDV), 239 vena cavae, 250 inferior, 249, 251 superior, 249, 251 ventricle,

X
xanthophyll, 214 macular, 215 pigment, 220 xeroderma pigmentosum cells, 98, 104

Y
yeast, 182-183, 195

Z
zinc, 174 zonules, 228