HOME

The world's
bestselling
medical
dictionary!

Dorland's Illustrated Medical Dictionary all of its illustrations, tables, and appendices.
is universally acknowledged as the world's At the same time, its state-of-the-art CD-ROM
finest medical dictionary. For 100 years, technology puts tremendous reference power
health care professionals have relied on its and convenience at users' disposal—plus audio
unmatched comprehensiveness, accuracy, pronunciations for over 10,000 primary entries!
clarity, and ease of use. 2000. Single-user CD-ROM for Windows™ or
The 29th Edition presents the very latest Macintosh*. Order #W9493-4.
information from every frontier in health A W.B. Saumlers alle.
care! It offers over 8,100 new terms—
7? Phone:
121,160 in all • over 7,600 new entries— Call toll-free 1-800-545-2522
for a total of 117,469 • over 860 illustra- (8:30-8:00 Eastern Time) to order.
tions—566 brand new • and much more! Be sure to mention DM#66899.
2000. 2112 pp. 864 ills. Order 3W6254-4.
Fax to i-800-568'5136 to order.
Also available on CD-ROM! Be sure to mention DM#66899.

Dorland's Electronic Medical Dictionary, 1^1 Mail:

29th Edition contains all of the definitions, Else vie r Health Sciences
Order Fulfillment Dept.
pronunciations, plural forms, and etymologies 1 1830 Wcsrline Industrial Drive
found in the hardbound dictionary, as well as Saint Louis, MO 63H6-3318

W.B. SAUNDERS
Elsevier Science
EHS 01 DM166»« * Elm*» SCTCHC, ZWä
 Seminars in Orthodontics
EDITOR
P. Lionel Sadowsky, DMD, BDS, DipOrth, MDent
Professor and Chairman
Department of Orthodontics
University of Alabama
1919 Seventh Avenue South
Birmingham, AL 35294
Fax: (205) 975-7590
Seminars in Orthodontics (ISSN 1073-8746) is published
quarterly by W.B. Saunders. Months of issue are March, June,
September, and December. Corporate and Editorial Offices:
The Curtis Center, Independence Square West, Philadelphia,
PA 19106-3399. Accounting and Circulation Offices: 6277 Sea
Harbor Drive, Orlando, FL 32887-4800. POSTMASTER: Send
change of address to: Seminars in Orthodontics, W.B. Saunders,
Customer Service, 6277 Sea Harbor Drive, Orlando, FL 32887-
4800.
Editorial correspondence should be addressed to:
P. Lionel Sadowsky, DMD, BDS, DipOrth, MDent, Editor,
Seminars in Orthodontics, Professor and Chairman, Department
of Orthodontics, University of Alabama, 1919 Seventh Avenue
South, Birmingham, AL 35294-0007; fax: (205) 975-7590.
Correspondence regarding subscriptions or change of
address should be directed to Seminars in Orthodontics, W.B.
Saunders, Periodicals Department, P.O. Box 628239, Orlando,
FL 32862-8239 or e-mail hhspcs@harcourt.com.
Change of address notices, including both the old and new
addresses of the subscriber and the mailing label, should be
sent at least 1 month in advance. Customer Service: 1-800-654-
2452
Yearly subscription rates: United States and possessions:
individual, $133.00; institution, $164.00; student and resident,
$67.00; single issue, $50.00. All other countries: individual
$166.00; institution, $198.00; student and resident, $83.00;
single issue, $50.00. For all areas outside the United States and
possessions, there is no additional charge for surface delivery.
For air mail delivery, add $16.00. To receive student/resident
rate, orders must be accompanied by name of affiliated
institution, date of term, and the signature of program/residency
coordinator on institution letterhead. Orders will be billed at
individual rate until proof of status is received.
Prices are subject to change without notice. Current prices
are in effect for back volumes and back issues. Single issues,
both current and back, exist in limited quantities and are
offered for sale subject to availability. Back issues sold in
conjunction with a subscription are on a prorated basis. Checks
should be made payable to W.B. Saunders and sent to Seminars
in Orthodontics, W.B. Saunders, Periodicals Department, 6277
Sea Harbor Drive, Orlando, FL 32887-4800.
Copyright 2002, Elsevier Science (USA). All rights
reserved. No part of this publication may be reproduced or
transmitted in any form or by any means, electronic or
mechanical, including photocopy, recording, or any information
storage and retrieval system, without permission in writing
from the Publisher. Printed in the United States of America.
Correspondence regarding permission to reprint all or
part of any article published in this journal should be
addressed to Journal Permissions Department, W.B. Saunders,
6277 Sea Harbor Drive, Orlando, FL 32887-4800. Telephone
number: 1-407-345-2500.
W.B. SAUNDERS
The appearance of the code at the bottom of the first page
of an article in this journal indicates the copyright owner's
consent that copies of the article may be made for personal or
internal use, or for the personal or internal use of specific clients,
for those registered with the Copyright Clearance Center, Inc.
(222 Rosewood Drive, Danvers, MA 01923; (508) 750-8400;
www.copyright.com). This consent is given on the condition that
the copier pay the stated per-copy fee for that article through the
Copyright Clearance Center, Inc. for copying beyond that
permitted by Sections 107 or 108 of the US Copyright Law.
This consent does not extend to other kinds of copying, such
as copying for general distribution, for advertising or promotional
purposes, for creating new collective works, or for resale.
Absence of the code indicates that the material may not be
processed through the Copyright Clearance Center, Inc.
Reprint inquiries should be addressed to Ginny
Nicholls, Elsevier Science, The Curtis Center, Independence
Square West, Philadelphia, PA 19106-3399. Telephone (215)
238-5534, fax (215) 238-6423; e-mail: gnicholls@elsevier.com.
Advertising representative: MJ. Mrvica Associates, Inc,
2 West Taunton Ave, Berlin, NJ 08009. Telephone (609) 768-
9360. Fax (609) 753-0064.
Publication of an advertisement in Seminars in Orthodontics
does not imply endorsement of its claims by the Editor(s) or
Publisher of the journal.
The contributors have checked generic and trade names and
verified drug doses for accuracy according to the standards
accepted at the time of publication. The ultimate
responsibility, however, lies with the prescribing physician.
Please convey any errors to the Editor.
The ideas and opinions expressed in Seminars in
Orthodontics do not necessarily reflect those of the Editor or the
Publisher. Publication of an advertisement or other product
mention in Seminars in Orthodontics should not be construed as
an endorsement of the product or the manufacturer's claims.
Readers are encouraged to contact the manufacturer with any
questions about the features or limitations of the products
mentioned. Neither the Editor or Publisher assume any
responsibility for any injury and/or damage to persons or
property arising out of or related to any use of the material
contained in this periodical. The reader is advised to check the
appropriate medical literature and the product information
currently provided by the manufacturer of each drug to be
administered to verify the dosage, the method and duration of
administration or contraindications. It is the responsibility of
the treating physician or other health care professional, relying
on independent experience and knowledge of the patient, to
determine drug dosages and the best treatment for the patient.
Seminars in Orthodontics is indexed in the Cumulative
Index to Nursing and Allied Health Literature® print index
and the Cinahl® database.
SPECIAL
OFFER!
For a limited time, all Year Books published
in 1999 and 1998 are on sale for $35, including
shipping. In most cases that's a savings of
over 50%!

To take advantage of this special offer:

Visit our website at www.mosby.com

Click on the "Special Promotions" link

Browse through the Year Books that interest you

Select the titles you wish to purchase

This offer is good only while supplies last,

so don't delay! ORDER TODAY!

For more information, call toll-free

1-800-654-2452

VH Mosby
• l r B ADivision
Divisionof
ofElsevier
E Science

© 2002 Mosby
<<    
     Article
      >>
Seminars in Orthodontics
EDITOR
P. Lionel Sadowsky, DMD, BDS, DipOrth, MDent
EDITORIAL BOARD
Richard G. Alexander, Arlington, TX
Rolf G. Behrents, Memphis, TN
Samir E. Bishara, Iowa City, IA
Robert Boyd, DBS, San Francisco, CA
Larry M. Bramble, Cypress, CA
John S. Casko, Iowa City, IA
Harry L. Dougherty, Van Nuys, CA
T.M. Graber, Evanston, IL
Robert J. Isaacson, Richmond, VA
Alexander Jacobson, Birmingham, AL
Lysle E. Johnston, Jr., Ann Arbor, MI
Gregory J. King, Seattle, WA
Vincent G. Kokich, Tacoma, WA
Steven J. Lindauer, Richmond, VA
INTERNATIONAL
Zeev Abraham, Herzliya, Israel
W.G. Evans, Johannesburg, South Africa
Roberto Justus, Mexico City, Mexico
Home | TOC |          
Index
James A. McNamara, Jr, Ann Arbor, MI
Robert N. Moore, Grand Island, NE
Ravindra Nanda, Farmington, CT
Perry M. Opin, Milford, CT
Sheldon Peck, Newton, MA
William R. Proffit, Chapel Hill, NC
Cyril Sadowsky, Chicago, IL
David M. Sarver, Birmingham, AL
T. Michael Speidel, Minneapolis, MN
William J. Thompson, Bradenton, EL
James L. Vaden, Cookeville, TN
Robert L. Vanarsdall, Jr., Philadelphia, PA
Katherine Vig, Columbus, OH
C.B. Preston, Buffalo, NY
Shinkichi Namura, Tokyo, Japan
George Skinazi, Paris, France
Björn U. Zachrisson, Oslo, Norway
<<    
     Article
      >> Home | TOC |          
Index

SPECIAL
OFFER!
For a limited time, all Clinics published
in 1999 and 1998 are on sale for $25, including
shipping. In most cases that's a savings of
over 50%!

To take advantage of this special offer:

Visit our website at www.wbsaunders.com

Click on the "Special Promotions" link

Browse through the Clinics that interest you

Select the titles you wish to purchase

This offer is good only while supplies last,

so don't delay! ORDER TODAY!

For more information, call toll-free

1-800-654-2452

W.B. SAUNDERS
A Division of Elsevier Science

) 2002 W. B. Saunders
<<    
     Article
      >> Home | TOC |          
Index

Seminars in Orthodontics
VOL 8, NO 1 MARCH 2002

Clinical Update on Technological Advances in Orthodontics

and Dentofacial Orthopedics
Gregory J. King, DMD, DMSc
Guest Editor
CONTENTS

Introduction
Gregory J. King

Predictive Orthodontics: A New Paradigm in Computer-Assisted Treatment

Planning and Therapy 2
James Mah

Current Concepts in the Biology of Orthodontic Tooth Movement 6

Calogero Dolce, J. Scott Malone, and Timothy T. Wheeler

Teeth in a Genetic Age 13

Heleni Vastardis

Three-Dimensional Cephalometry and Three-Dimensional Skull Models in

Orthodontic/Surgical Diagnosis and Treatment Planning 17
Robert A. W. Fuhrmann

Three-Dimensional Evaluation of Periodontal Remodeling During

Orthodontic Treatment 23
Robert A. W. Fuhrmann

Preliminary Tests of a New Device to Monitor Orthodontic Headgear Use 29

Elizabeth K. Lyons and Douglas S. Ramsay

A Comparison of Skeletal and Dental Changes Between Rigid and Wire

Fixation for Bilateral Sagittal Split Osteotomy 35
Calogero Dolce, John P. Hatch, Joseph E. Van Sickels, Robert A. Bays, and
John D. Rugh

The Invisalign System: Case Report of a Patient With Deep Bite, Upper
Incisor Flaring, and Severe Curve of Spee 43
Ross J. Miller and Mitra Derakhshan
<<    
     Article
      >> Home | TOC |          
Index

Seminars in Orthodontics
Future Issues

Vol 8 No 2 (June 2002)

BIOSTATISTICS FOR THE ORTHODONTIC CLINICIAN
Rose D. Sheats, DMD, Guest Editor
Vol 8 No 3 (September 2002)
RISK ASSESSMENT AND CLINICAL ORTHODONTIC MANAGEMENT
Larry Jerrold, DDA, fl), Guest Editor
Vol 8 No 4 (December 2002)
THE VERTICAL DIMENSION
Timothy E. Wheeler, DMD, Guest Editor

Recent Issues

Vol 7 No 4 (December 2001)

THREE-DIMENSIONAL DIAGNOSIS AND TREATMENT IN ORTHODONTICS
Sheldon Baumrind, DDS, MS, and Robert L. Boyd, DDS, MEd, Guest Editors
Vol 7 No 3 (September 2001)
TOPICS IN BIOMECHANICS
Stanley Braun, DDS, MME, Guest Editor
Vol 7 No 2 (June 2001)
The Alexander Discipline
R.G. Alexander, DDS, MSD, Guest Editor
Vol 7 No 1 (March 2001)
CLINICAL BIOMECHANICS
Steven J. Lindauer, DMD, MDSc, Guest Editor
Vol 6 No 4 (December 2000)
PSYCHOLOGIC ISSUES RELATED TO ORTHODONTIC TREATMENT AND PATIENT COMPLIANCE
Pramod K. Sinha, DDS, BDS, MS, Ram S. Nanda, DDS, MS, PhD, and Roger B. Fillingim, PhD, Guest Editors
Vol 6 No 3 (September 2000)
BIOLOGY OF ORTHODONTIC TOOTH MOVEMENT: CLINICAL IMPLICATIONS
Bhavna Shroff, DDS, MDentSc, Guest Editor
Vol 6 No 2 (June 2000)
MOLAR DISTALIZATION
George f . Cisneros, DMD, MMSc, Guest Editor
Vol 6 No 1 (March 2000)
OBJECTIVES-DRIVEN ORTHODONTICS: EFFECTIVENESS OF MECHANOTHERAPY
Cyril Sadowsky, BDS, MS, Guest Editor
<<    
     Article
      >>
Seminars in Orthodontics
VOL 8, NO 1
Introduction
T oday, advances in technology and biology
promise to change all aspects of human
activity. The ability to process and transmit vast
quantities of data quickly and cheaply has be-
come commonplace. All of the genetic informa-
tion required to make a human being is now
available. Vertebrates can be cloned. Exciting
new ways to acquire and use information are
being developed daily. At this time, it has be-
come abundantly clear that orthodontics and
dentofacial orthopedics will not escape this rev-
olution. In fact, many of these advances show
promise of dramatically altering the practice of
orthodontics.
What will orthodontics look like in the fu-
ture? Computer imaging will make the study
model and articulator obsolete. Robots, given
relevant information by the clinician, will shape
wires. Forces will be delivered to teeth and jaws
by means that are quite different from what we
currently use. Combining biomechanic and bio-
logic signals will safely accelerate the rate of
treatment. The genetic basis of conditions such
as oligodontia will be understood, predicted,
and possibly prevented. Tissues will be engi-
neered at the cellular and molecular levels. The
impact of treatment on clinical outcomes will be
understood as never before because data will be
more readily available and shared, and newer
methods to analyze those data that more closely
Copyright 2002, Elsevier Science (USA). All rights reserved,
doi: 10.1053/sodo. 2002.28185
Seminars in Orthodontics, Vol 8, No 1 (March), 2002: p l
Home | TOC |          
Index
MARCH 2002
mimic the way the brain processes information
will be available. Treatment approaches will be
modeled before beginning treatment. Smart ap-
pliances will be able to store biomechanical in-
formation that will be periodically downloaded
to guide treatment. You may think that many of
these predictions are farfetched. You would be
wrong. At some level today, most of these things
are being done or are rapidly being developed.
Last summer, the Harvard Society for the Ad-
vancement of Orthodontics held its Third Tri-
ennial Conference in Athens, Greece with the
objective of examining the future of orthodon-
tics. Orthodontic clinicians who are currently
actively working on the cutting-edge in imaging,
treatment, biology, and genetics were invited to
present their work and to speculate about how
they may change the nature of orthodontics.
Several of these participants have agreed to
present some of their material in this issue of
Seminars in Orthodontics. The intent of this issue is
to present some of that material, discuss the
challenges that currently exist, and to speculate
about the future. Because of space limitations,
prior publication commitments, and pressures
from other activities, some of the conference
participants were unable to prepare manuscripts
for this issue. What is presented here, however,
will give the reader a taste of the entire meeting.
Gregory J. King, DMD, DMSc
Guest Editor
<<    
     Article
      >> Home | TOC |          
Index

Predictive Orthodontics: A New Paradigm in

Computer-Assisted Treatment Planning
and Therapy
James Mah

The current flow of information in orthodontic diagnosis and treatment

planning begins with the clinical examination, charting, and orthodontic
records. Later, the clinician mentally assembles and processes these data
and derives a diagnosis for the patient. Treatment planning involves infor-
mation recalled from the physician's knowledge and clinical experience.
From here, treatment alternatives are produced, and after consultation with
the patient, a treatment plan is chosen and implemented. The current
system, however, contains several identifiable sites of error or assumptions
in the diagnostic, treatment planning, consultation, and therapeutic func-
tions. These factors along with the spectrum of biological variation and
response to therapy make it particularly challenging to predict exact treat-
ment outcomes for a given patient. Prediction of exact outcomes may be
impossible, but through the use of computer modeling and analysis, predic-
tion may be vastly improved. A new paradigm of orthodontic diagnosis and
treatment planning that includes use of the most current diagnostic modal-
ities to gather accurate anatomic and functional data on the patient and to
construct a craniofacial model is described. Using this model, alternative
treatment planning and hypothesis testing may be performed by using finite
element and other analytical approaches. (Semin Orthod 2002;8:2-5.) Copy-
right 2002, Elsevier Science (USA). All rights reserved.

Introduction return to the patient to verify observations such

T he current process for diagnosis and treat-
ment planning involves the clinician col-
lecting and assembling the patient's medical and
dental history, notes from a clinical examina-
tion, radiographs, photographs, models, and
other diagnostic tests that will help to produce a
mental image of the patient. Such records con-
tain information voids, lack a three-dimensional
(3-D) perspective, and offer extremely limited
functional information. Often the clinician must
From the Department of Orthodontics, University of Southern
California School of Dentistry, Los Angeles, CA.
Address reprint requests to James Mah, DDS, DMSc, Department
of Orthodontics, University of Southern California School of Den-
tistry, 925 W. 34th, #312, Los Angeles, CA 90089-0641.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0801-0001$35.00/0
doi:10.1053/sodo.2002.28164
as mandibular position and movement. These
factors place limits on the use of diagnostic
records, requiring the clinician to make assump-
tions to fill the voids and missing perspectives.
These assumptions can introduce errors that can
lead to less predictable treatment outcomes.
Data are customarily analyzed with knowledge
based on scientific evidence, empirical data, and
various clinical experiences. There is a possibil-
ity that errors may be introduced at this point
because the knowledge base relies on the clini-
cian's training, recollection and clinical experi-
ences. From here, treatment alternatives are
produced, and a plan of treatment is selected
that is in agreement with the patient's wishes
(Fig 1). Errors may also be introduced at this
point because of the combination of the inability
of the clinician to predict an accurate treatment
outcome, to communicate the anticipated result

Seminars in Orthodontics, Vol 8, No 1 (March), 2002: pp 2-5

<<    
     Article
      >> Home | TOC |          
Index

Predictive Orthodontics

addition, without accurate information, the def-

Patient Information inition and evaluation of clinical success and
Medical and Family History outcome is difficult, if not impossible.
Clinical Examination
Imaging and Laboratory Tests
• Radiology, Photography, Models New Technologies
• CT, MRI Recently, new imaging and sensing technologies
• Oral Pathology have been introduced to orthodontics that pro-
vide 3-D images of the craniofacial skeleton, soft
tissues, and dentition, including recordings of
mandibular position and movement.1 Although
the information provided by these technologies
is not all-inclusive, they provide much more and
Data Analysis
better information about the patient than cur-
Knowledge Base
rently available records (Fig 2). These advan-
• Scientific evidence
• Empirical data tages include the third dimension, higher reso-
• "Clinical experience" lution, and also functional information. Ideally,
all essential anatomic, functional, and physio-
logic information regarding a patient is required
to construct the most accurate model of the

Treatment Alternatives
Visualization
3D skeletal imaging
3D facial imaging
3D dental imaging
Mandibular motion
Facial animation
Functional information
Other physiologic data

Definitive Treatment Plan

Model Construction
Image and data refinement
Registration
Integration
Figure 1. The current flow of information in orth- Segmentation
odontic diagnosis and treatment planning. Patient-
specific information is gathered and assembled for
analysis with the knowledge base of the orthodontist.
Treatment alternatives are produced, and with pa-
I
Diagnosis and Treatment Boundary Conditions
tient input, a definitive treatment plan is selected. Planning
Each step contains potential sites of error and also
may include assumptions.
I
Analysis
Finite element analysis
to the patient, and to visualize the expectations Hypothesis testing
Materials selection
of the patient. Clearly, there is abundant room Solutions database
to improve diagnosis and treatment planning by
providing more and better patient information.
Patient records usually are not used directly
to fabricate appliances. However, in recent Treatment
years, orthodontics has seen the introduction of Treatment options
Definitive treatment
a number of computer-made appliances offering Computer assisted therapy
Treatment monitoring
improved precision and efficiency. Treatment results

Considering all of the potential sites of error,

predicting an exact outcome for a patient using Figure 2. Flowchart ofinformation in predictive orth-
the present system can be quite challenging. In odontics.
<<    
     Article
      >>
James Mäh
patient. For example, accurate data on bite
forces, tongue posture and swallowing, para-
function, and respiration are extremely useful in
orthodontics. Certainly, there will be continuing
efforts to improve on existing imaging technol-
ogies and new diagnostic modalities will become
available. As more and better patient informa-
tion becomes available, the computer model of
the patient will also be improved.
The computerized images and data are pro-
cessed using refinement schemes. These pieces
of data are related to each other through regis-
tration, integration, and segmentation. This is
done to create independent objects, like individ-
ual bones or teeth. Geometric model construc-
tion is carried out to incorporate the dentition
of the patient, craniofacial skeleton, and the soft
tissues of the face. In addition, 3-D coordinates
on mandibular position and movement as well as
facial animation may be incorporated into such
a model. Other available functional data may
also be used. At this stage, all available patient
information is in a completely integrated model,
which may then be used for diagnosis and treat-
ment planning. Also at this stage, boundary con-
ditions from a database may be applied to help
select possible treatment alternatives. An exam-
ple of a boundary condition could be the move-
ment of the maxillary central incisor. Beyond
the boundary, growth modification or surgery
would be recommended. Other examples of
boundaries that could be helpful to the clinician
include lack of further growth or inadequate
anchorage. These and the many boundary con-
ditions that pertain to orthodontics would be
included in the database.
The model can also be used for analysis, hy-
pothesis testing and treatment simulations of
any number of treatment scenarios involving var-
ious tissues.2 The model and treatment ap-
proaches can be analyzed by finite element
methods and solved. Treatment alternatives may
be simulated and assessed under a variety of
functional states. Esthetics, occlusion, and other
outcomes may be assessed. The results are then
extracted and stored in a database. In addition,
materials such as arch wires, adhesives, appli-
ances, or implants can be tested on such a
model. From here treatment options would be
presented to the patient and clinical treatment
provided. Accurately predicting treatment out-
come has been shown to be an important aspect
Home | TOC |          
Index
of providing a clear and realistic preoperative
informed consent.3 Interim and posttreatment
clinical results are analyzed by comparison to
the predicted result and to the database of past
clinical experiences. As more information be-
comes available with more patients, a feedback
loop will exist to improve and refine boundary
conditions. In this way, the system will become
more accurate and more predictive. Because
past experiences are stored in a database, this
information can be shared with other clinicians
as well as provide guidelines for orthodontists in
training. In addition to analysis, the computer
model is sufficiently accurate to be used for
implementing therapy. By using computer-as-
sisted design and computer-assisted manufactur-
ing methods and robotics, any number of ther-
apeutic appliances, such as surgical splints,
retainers, surgical guides for implants, templates
and arch wires,4 could be produced. Additional
information from the model also can be used to
create a 3-D scaffold for procedures involving
tissue engineering.5 Computer-assisted therapy
would offer improved precision, reliability, and
more predictable results overall.
Summary
Predictive orthodontics is made possible by new
technologies that allow orthodontists to better vi-
sualize the patient. These include methods to im-
age the craniofacial skeleton by using a newly in-
troduced class of computed tomography scanners
optimized for the head and neck or by methods of
photogrammetry by which 3-D information is pro-
duced from planar radiographs. Three-dimen-
sional images of the dentition can be produced
directly with an intraoral 3-D camera or indirectly
by converting an impression to a computer model.
Images of the face can be produced using a num-
ber of commercially available 3-D cameras. Addi-
tionally, it is now possible to track mandibular
position and facial movements by using ultrasonic
or video-based methods.
Continued improvement in the quality of orth-
odontic services will require implementation of
new diagnostic and treatment approaches. New
technologies will provide improved precision, ac-
curacy, and perspective. This will provide the cli-
nician with the means to optimize the way treat-
ment is planned through computer modeling,
testing, and simulations. In this way, treatment
<<    
     Article
      >>
Predictive Orthodontics
results can be predicted with greater accuracy,
thereby reducing unfavorable outcomes and in-
creasing efficiency. These approaches also can
serve as very powerful tools in orthodontic educa-
tion and patient communication.
References
1. Mah J, Bumann A. Technology to create the 3-dimen-
sional patient record. Semin Orthod 2001;7:199-208.
Home | TOC |          
Index
2. Keeve E, Girod S, Kikinis R, et al. Deformable modeling
of facial tissue for craniofacial surgery simulation. Comp
Aid Surg 1998;3:228-238.
3. Rai AK. Reflective choice in health care: Using informa-
tion technology to present allocation options. Am J Law
Med 1999;25:387-402.
4. Mah J, Sachdeva R. Computer-assisted orthodontic treat-
ment - the SureSmile_ process. Am J Orthod Dentofacial
Orthop 2001;117:622-623.
5. Schultz O, Sittinger M, Haeupl T, et al. Emerging strategies
of bone and joint repair. Arthritis Res. 2000;2:433-436.
<<    
     Article
      >> Home | TOC |          
Index

Current Concepts in the Biology of

Orthodontic Tooth Movement
Calogero Dolce, J. Scott Malone, and Timothy T. Wheeler

When force is applied to a tooth during orthodontic tooth movement,

mechanical stress is loaded on the alveolar bone. Alveolar bone and
the periodontal ligament (PDL) are compressed on one side, while on the
opposite side, the PDL is stretched. The mechanical stress of the
stretched PDL induces alveolar bone modeling (surface apposition of
bone), while the mechanical compression produces bone remodeling
(the turnover of bone in small packets). The steps leading from the
application of an orthodontic force to the appearance of a biologic
response (mechanotranduction) have not been fully elucidated. The
possible mechanisms by which osteoblasts and/or osteocytes can
sense a mechanical stimulus include strain-sensitive ion channels,
shear stress receptors, adhesin/integrin activation, and cytoskeleton
reorganization. Once detected, the signal is internalized and then po-
tentiated in the cytosol by the generation of second messengers and
protein kinases. This, in turn, activates transcription of genes into
messenger RNA that is then translated into proteins to be used by the
cell or exported to modulate the activity of additional enzymes, eventu-
ally leading to altered gene expression. In the nucleus, the induction of
the so-called immediate early genes, which occur shortly after a cell is
stimulated, may also play a critical role in the signal transduction path-
way. (Semin Orthod 2002;8:6-12.) Copyright 2002, Elsevier Science (USA).
All rights reserved.

Introduction down osteoid and then mineralizing it. Osteo-

Bone is composed of four cell types: osteo-
blasts, osteocytes, bone lining cells, and os-
teoclasts. Osteoblasts, bone-lining cells, and os-
teocytes are derived from mesenchymal stem
cells and represent various stages of the devel-
opment of a single cell type.1 Osteoblasts are
round cells with an organelle-rich cytoplasm.
Their function is to form bone, first by laying
From the Department of Orthodontics, College of Dentistry, Uni-
versity of Florida, Gainesville, FL.
Supported by the American Association Orthodontists Founda-
tion.
Address reprint requests to Calogero Dolce, DDS, PhD, Depart-
ment of Orthodontics, Box 100444, JHMHC, Gainesville, FL
32610-0444.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0801-0002$35.00/0
doi: 10.1053/sodo. 2002.28165
cytes are former osteoblasts that have become
entrapped in bone matrix. A striking feature of
osteocytes is the extensive network of cellular
processes that connect them with each other
and with cells at the bone surface.2 As implied by
their name, bone-lining cells cover the bone
surface. Their function is not well understood,
but they may be the pivotal cells that regulate
bone resorption and formation. Osteoclasts are
large, multinucleated cells whose principal func-
tion is to resorb bone. Osteoclasts are derived
from hematopoetic cells and represent the ter-
minal stage in the differentiation of these cells.3
Morphologic features of Osteoclasts include
large size, a ruffled border, multiple nuclei, and
a clear zone where they attach to bone. They
secrete a lysosomal enzyme, tartrate-resistant
acid phosphatase (TRAP), which is also used as
a marker for osteoclast identification.4

Seminars in Orthodontics, Vol 8, No 1 (March), 2002: pp 6-12

<<    
     Article
      >>
Current Concepts in OTM
The Life Cycle of Bone Cells
Differentiation of Osteoblasts
Mesenchymal stern cells differentiate into osteo-
blasts when they are exposed to bone morpho-
genic proteins (BMP). 5 BMPs are part of a large
multigene family, the transforming growth fac-
tor ß (TGFß) superfamily. Our knowledge of the
process involved in the development of an osteo-
blast is very limited. A recent advancement has
been the identification of the core binding fac-
tor alpha 1 (Cbfal} gene during osteoblast de-
velopment and function. Cbfal is a transcription
factor that is expressed by cells of the osteoblas-
tic lineage and is necessary for osteoblast differ-
entiation.6 Studies have shown the lack of osteo-
blasts and consequently bone formation in mice
deficient in the Cbfal protein. Cleidocranial dys-
plasia (CCD) is caused by mutations of the Cbfal
gene.7 It appears that Cbfal may also play a role
in osteoblast cell function because mice express-
ing a dominant-negative version of the gene
show decreased expression of genes that charac-
terize the osteoblast phenotype (eg, collagen
genes, osteopontin, and osteocalcin) .8
Differentiation of Osteoclasts
Osteoclasts differentiate from hematopoetic
cells. This pathway of differentiation is also
shared by macrophages. It is not surprising then
that granulocyte-macrophage colony-stimulating
factor (GM-CSF) and macrophage colony-stimu-
lating factor (M-CSF) are important in regulat-
ing these shared stages of development.4 The
recent discovery of osteopotegerin (OPG) and
its binding ligand (OPG-L) as essential regula-
tors of osteoclasts formation and activity, how-
ever, has created a new paradigm in osteoclast
biology.9
The molecule that inhibits osteoclastogenesis
is known by two different names, OPG (osteopo-
tegerin) and OCIF (osteoclastogenesis inhibit-
ing factor), because it was identified simulta-
neously by two different laboratories working on
different projects.10' n This molecule shares se-
quence homology with the tumor necrosis factor
(TNF) receptor superfamily. OPG is secreted by
osteoblasts and functions to block the formation
of osteoclasts as well as bone resorption. Trans-
genie mice overexpressing OPG develop osteo-
petrosis caused by the lack of osteoclast differ-
Home | TOC |          
Index
entiation, whereas transgenic mice that lack the
OPG gene have severe osteoporosis caused by an
increased number of functional osteoclasts.10
OPG was shown to be as effective as bisphospho-
nates in protecting against bone loss in ovariec-
tomized estrogen-deficient rats.10 When injected
subcutaneously in postmenopausal women,
OPG resulted in a decrease in the bone turnover
marker N-telopeptide.12 In a recently published
study, Kobayashi and colleagues localized the
OPG in alveolar bone surfaces during OTM.13
They showed that in tension areas, as the bone
shifts from resorption to formation, there was a
simultaneous increase in TGF-ß and OPG mes-
senger RNA (mRNA) and the disappearance of
osteoclasts through apoptosis.
Because OPG is a soluble molecule, it was
predicted that its ligand would be a membrane-
bound molecule. Using molecular biology tech-
niques, Suda and colleagues isolated a molecule
that they called osteoclast differentiation factor
(ODF).14 ODF was able to induce osteoclasto-
genesis in bone organ cultures. Lacey et al also
identified a molecule termed OPG-L that dem-
onstrated similar activities as ODF.15 When ODF
and OPG-L were cloned, it was revealed that the
molecules were identical to an already cloned
member of the TNF ligand family, called
TRANCE (TNF-related activation induced cyto-
kine) or RANKL (receptor activator of NF-kB
ligand). RANKL or OPG-L binds to the receptor
activator of NF-kB ligand (RANK), which is
found on the cell membranes of osteoclasts. This
binding can occur in two ways: one is cell bound,
which requires cell-to-cell contact; the other is a
soluble form. In this situation, OPG can act as
decoy receptor, blocking the interaction be-
tween RANK and OPG-L.16
The ratio of OPG/OPG-L regulates the oste-
oclast's lifecycle from differentiation to fusion of
preocteoclasts to function to apoptosis. The hy-
pothesis is that, at steady state, there is a bal-
anced level in the OPG/OPG-L ratio. Cytokines
(TNF, interleukin-1 [IL-1], prostoglandin E2
[PGE2] and growth factors (TGF-B, BMP) are
upstream signals which regulate the OPG/
OPG-L ratio. OPG/OPG-L serve as downstream
regulators of osteoclastogenesis. Thus, when the
balance favors OPG, there are fewer active oste-
oclasts; when the balance favors OPG-L, there is
an increased number of active osteoclasts.
<<    
     Article
      >>
Dolce, Malone, and Wheeler
How Bone Cells Detect Mechanical
Strain
After an orthodontic force is applied, the initial
step is the detection of a mechanical strain. The
cells responsible for sensing mechanical strains
in bone have been considered to be osteoblasts,
osteocytes, or both. Theories on how these cells
sense strain in their environment include
streaming potentials, strain-sensitive ion chan-
nels and cytoskeletal reorganization.
Strain Released Potentials
Mechanical coupling denotes the conversion of
mechanical energy into a form that can be de-
tected by cells. Bone responds to an applied
strain. Strain (e) represents a change in length
(1 fjis represents 1 jam of deformation per meter
of length). Normal strains in humans range
from 400 JLLS to 3000 JJLS depending on the level
of activity. The highest peak strains have been
measured in a galloping horse at 3200 juie.17
Bone begins to fail when strains reach 7000 JULS.
Application of small bending forces to long
bones result in compression on one side and
tension on the opposite side. This produces
a flow of interstitial fluid, through the canalic-
ular network, generating streaming potentials
and/or fluid shear stress. Streaming potentials
occur when electrically charged fluid is forced
over a tissue (cell membrane) with a fixed
charge. For example, bone is partially composed
of proteoglycans, which are entrapped in a col-
lagen network. Because of the negative charge of
proteoglycans, there is an excess of positive mo-
bile ions in the fluid. Charges are symmetrically
arranged so that no net macroscopic electric
field is present. Compression of bone produces
streaming potentials by the displacement of mo-
bile ions relative to charged proteoglycans en-
trapped by collagen. Fluid movement over the
cell surface may directly stimulate bone cells
because it generates shear stress.18
Activation of Ion Channels
Mechanosensitive ion channels in the plasma
membrane have been observed with single chan-
nel recordings in more than 30 cell types from
plants to animals. The stretch-activated ion
channels allow the passage of cations (calcium
and potassium). Application of mechanical
Home | TOC |          
Index
strain to osteoblasts has been shown to acti-
vate stretch-activated calcium channels thereby
producing a large increase in intracellular cal-
cium.19 The mechanism by which a mechanical
stimulus modulates stretch-activated ion chan-
nel gating remains unknown. It is yet to be re-
solved if channel gating is caused by direct me-
chanical perturbation of the channel or if the
channel is activated secondarily because of acti-
vation of stretch-sensitive phospolipase C or D. It
has been suggested that ion channels may be
linked to the cytoskeleton and are opened when
their cytoplasmic tail is phosporylated.20 The
problem that arises from these studies is that it is
difficult to determine the amount of applied
strain and consequently any in vivo relevance. In
addition to the calcium channels, stretch-acti-
vated potassium channels have been identified
in osteoblast-like cells.21
Extracellular Matrix and the Cytoskeleton
The process of cellular adhesion has been impli-
cated in many biologic phenomena, including
cell migration, anchorage, proliferation, differ-
entiation, and mechanotransduction.22 There
are two types of cellular adhesion: cell-to-cell
adhesion and cell-to-extracellular matrix (ECM)
adhesion. Several families of adhesion receptors
have been identified, including integrins. Inte-
grins are cell surface receptors that mediate cell-
to-cell attachment or cell attachment to ECM
molecules, such as fibronectin, laminin, and col-
lagen. Integrins are a family of a/ß het-
erodimeric cell surface receptors composed of at
least 14 distinct a subunits and 8 or more ß
subunits that can associate noncovalently in var-
ious combinations. Although over 100 combina-
tions are possible, the actual diversity is more
limited because many a subunits can associate
with only a single ß subunit. The ß subunit
(80-90 kDa) is more conserved, and each ß sub-
unit defines an integrin family. The a-chain, on
the other hand, is more variable (120-180 kDa)
and confers specificity for the recognition of
different ligands. Osteoblasts have been shown
to express the integrin a 2 ßi and OL5ßl9 the pro-
totype receptors for collagen type I and fi-
bronectin, respectively. Integrins may be puta-
tive mechanotransducers because they have
been found to regulate signaling pathways by
changing intracellular calcium, regulating inosi-
<<    
     Article
      >>
Current Concepts in OTM
tol lipid turnover and phosphorylation of intra-
cellular proteins. Mechanical stretching of the
human osteosarcoma cell line TE-85 has been
shown to induce ßl integrin expression and al-
ter its distribution.23
The specific binding of individual integrins
has been elucidated with the use of cell adhesion
assays, monoclonal antibodies (Mabs), and affin-
ity chromatography. The first receptor-binding
site to be defined was the RGD (arginine-glycine-
aspartic acid) sequence present in fibronectin.24
Since then, other adhesive molecules, such as
vitronectin and collagen, have been shown to
contain the RGD sequence. Proteins containing
the RGD sequence, or variations thereof, have
been engineered to inhibit ECM binding and to
study cellular adhesion. One structural feature
common to each of these peptides is the aspar-
tate residue. It is believed that this residue may
play a critical role in integrin binding because
alteration of the aspartic acid in the peptide
sequences abrogates adhesive activity. For exam-
ple, when RGDS (arginine-glycine-aspartic acid-
serine) and RGES (arginine-glycine-glutamic
acid-serine) were investigated for their ability to
bind to osteoblasts and inhibit cell attachment
to fibronectin, it was found that although both
peptides bound to osteoblasts, only RGDS ef-
fected cell attachment.25 RGES had minimal ef-
fect on cell attachment, whereas RGDS partially
(55% to 60%) inhibited cell adhesion to fi-
bronectin.
The influence of ECM on the survival and
function of bone cells cannot be overempha-
sized. Perturbing the interaction of osteoblasts
with its integrin receptors by using antibodies or
RGD-like fragments suppresses the formation of
mineralized nodules in vitro and delays the ex-
pression of tissue-specific genes, including osteo-
calcin.26' 27 In addition, apoptosis of differenti-
ated osteoblastic cells can be triggered by
interfering with their interaction with fibronec-
tin.28 Osteoclasts also express integrin receptors
including the vitronectin receptor (av/33), which
plays an important role in the adhesion of oste-
oclasts to bone surface. Peptides containing the
RGD motif have been shown to inhibit oste-
oclast-mediated bone resorption in vitro and
prevent osteoporosis in vivo.29 These peptides
could potentially aid in the treatment of osteo-
porosis and possibly in modulating OTM. In
fact, in a recent study, a decrease in the number
Home | TOC |          
Index
of osteoclasts, as well as a decrease in the num-
ber of TRAP+ cells, was demonstrated when the
RGDS peptide was injected locally during
OTM.30 OTM and other parameters of bone
turnover, however, were not measured.
Previously, the osteocytes or osteoblasts were
considered to be the cells that were responsive
to mechanical stimulation. In a recently pub-
lished study, however, Kurata and colleagues
found mature osteoclasts to be responsive to
mechanical strain.31 In this in vitro study, ma-
ture osteoclasts responded to mechanical strain
by significantly increasing mRNA expression of
TRAP and cathepsin. By using a pit formation
assay, they also demonstrated an increase in the
number of pits as well as in the total pit area.
This suggests the up-regulation of bone resorp-
tion by osteoclasts directly from mechanical
stimulation. They attributed the mechanosens-
ing mechanism to a stretch-activated cation (SA-
cat) channel because when godolinium, a SA-cat
channel blocker was used, pit formation and
mRNA levels decreased.
How Bone Cells Translate Mechanical
Strain
Regardless of how a mechanical signal is re-
ceived, it must somehow influence the biochem-
ical machinery of the cell. In the signaling cas-
cade process, receptor activation is followed by
second-messenger generation (adenosine 3',5'
cyclic monophosphate [cAMP] and inositol
trisphosphate [IPS]).32 These advance the signal
to the nucleus through a series of kinases. In the
nucleus, different second messengers account
for the differential pattern of immediate early
gene (IEG) expression. lEGs are among the ear-
liest responses that can be measured at the tran-
scription level. The transcription of the lEGs
(c-fos, c-jun, and egr-1) has been shown to in-
crease when cells are exposed to cytokines,
growth factors,33 or mechanical stimulation.34
Protein products from the c-fos and c-jun genes
form a heterodimeric complex named activator
protein-1 (AP-1). AP-1 is a transcription factor; it
binds to the promoter region of a gene and then
modulates its activity35. Depending on the state
of the cell or in the presence of various stimuli,
AP-1 can produce either cellular proliferation or
differentiation.
We are beginning to understand the pattern
<<    
     Article
      >>
10 Dolce, Malone, and Wheeler
of early gene expression occurring during
stretching of bone and/or bone cells. Bending
forces applied to the tibias of rats were found to
increase c-fos mRNA expression in periosteal
cells, within two hours of force application.36
Alveolar bone is somewhat different from skele-
tal bone, however, because its existence is de-
pendent on the presence of teeth. Its turnover
rate can also be influenced through the stresses
that are produced by the application of forces
(orthodontics) or via the teeth (functional). In
an effort to better understand what processes
are involved in mechanotransduction during
OTM, we have examined whether and how soon
c-fos induction occurs during this process.
With the use of a well-characterized model for
tooth movement,37 we examined the expression
of c-fos during OTM by using semi-quanta tive
reverse transcription- polymerase chain reaction
(RT-PCR). Briefly, 70 young male (40 to 50 days
old) Sprague-Dawley rats were divided into
seven groups corresponding to times 0, 3, 6, 12,
24 hours and 7 and 14 days, with five rats each in
nontreated and OTM groups. From previous
studies, these timepoints best represent the bio-
chemical changes being studied. OTM was
achieved by using a NiTi coil to apply 40 grams
of force between the maxillary first molar and
incisor. One hemimaxilla was used for total RNA
isolation. The RNA was extracted from the roots
and surrounding alveolar bone by using stan-
dards protocols.38 RT-PCR was used to follow
induction of c-fos mRNA.
As shown in Figure 1, c-fos was expressed in
both treated and nontreated groups and at all
timepoints. c-fos expression was observed to be
cyclic in nature (Fig 2). Within 3 hours of apply-
ing a 40-gram force, there was a 1.7-fold induc-
tion in c-fos mRNA expression when compared
to its respective control (P < .05). The next
significant induction of 1.9-fold was detected at
24 hours afater force application. A final 1.5-fold
induction was seen at 7 days after appliance
activation, c-fos mRNA induction was not de-
tected at 6 hours, 12 hours and 14 days.
Animal models that have been used to study
mechanotransduction have loaded skeletal bone
by four-point bending of long bones36 or load-
ing caudal vertebrae.39 Results from our study
are analogous to what has been reported when
other models have been used to study mechano-
transduction. In previous studies in which the
Home | TOC |          
Index
Hours Days
Figure 1. Representative gel of RT-PCR product for
c-fos mRNA expression in alveolar bone (day 0) and at
various time points following OTM (A) or untreated
(B). Glyceraldehyde-3 phosphate dehydrogenase
(G3PDH) was the housekeeping gene. An anti-sense
21-base pair (5'-TCCTACTACCATTCCCCAGCC-3')
probe complementary to mRNA encoding a specific
region in the c-fos gene and its sense probe were
synthesized and. used to amplify cDNA of c-fos. Equal
amounts of RNA were reverse transcribed (RT) by
using the protocol in the Perkin Elmer GeneAmp
RT-PCR Kit (Foster City, CA). The 20 juL RT reaction
then underwent PCR using 1.25 U of Taq polymerase
and primers for the genes in a 50 juL volume. G3PDH
expression was served as a control for variations be-
tween PCR reactions and allow the calculation of a
normalized value for c-fos gene expression. Equal vol-
umes of PCR products were electrophoresed in a
2.0% agarose gel, stained with ethidium bromide,
visualized with an ultraviolet transilluminator, and
scanned.
skeletal bones have been loaded, c-fos induction
was followed less than 24 hours after the me-
chanical stimulation.36 In these studies, c-fos in-
duction peaked between 2 to 4 hours after me-
chanical stimulation. Because orthodontic tooth
movement (OTM) is not linear and multiple
cellular processes are activated, we followed the
induction of c-fos for 14 days after a 40-gram
force was applied to the maxillary molars of male
rats. Our results show a peak c-fos induction in a
cyclic fashion: at three and 24 hours and at 7
days after appliance activation. The c-fos gene
product, a component of the transcription fac-
tor AP-1, has been correlated with cell prolifer-
ation and differentiation.40 During OTM, there
is both the proliferation and differentiation of
cells of the periodontium (alveolar bone and
periodontal ligament).41 We speculate that c-fos
may play a role in both. In the early phase of the
OTM cycle, c-fos may play a role in mechanical
<<    
     Article
      >>
Current Concepts in OTM
14d
Figure 2. Relative mRNA expression of c-fos as deter-
mined by densitometry. Densitometric analysis of gel
pictures by using a Hewlett-Packard desktop scanner
with NIH Image (v. 1.57) software provided a deter-
mination of the intensity of the bands. For each sam-
ple at each time point, the ratio of the integrated
optical densities (IOD) of the c-fos band to G3PDH
band was calculated. In the untreated rats at each
time point, the ratio of o/os/GSPDH was set to 1, and
the expression of the c/os/GSPDH in treated rats was
expressed as a ratio of this value. Mean and standard
deviation of the c-/os/G3PDH ratio for each group
and at each time point was calculated. For compari-
son of two groups at the same time interval, the
unpaired Student's t test was used. P <.05 was consid-
ered significant.
signal transduction because many pathways,
such as tyrosine kinase, p21 ras and MAP ki-
nases, induce its expression.42 In the later stages
of the OTM cycle, as part of AP-1, it may regulate
osteoblast differentiation. AP-1 binding sites are
found in promoter regions of gene that are in-
volved in mineralization, including collagen
type I, osteocalcin, and alkaline phosphatases.
We cannot yet determine whether the induc-
tion of c-fos described in this study occurs di-
rectly or indirectly. Mechanical loading of bones
in vivo has been correlated with increases in PGs,
nitric oxide (NO), insulin-like growth factor,
and transforming growth factor-/3. When me-
chanical loading of bones was performed in the
presence of inhibitors of PGs or NO, <>/asmRNA
expression was partially inhibited.43 This sug-
gests that the application of an orthodontic
force to the teeth may activate more than one
signaling pathway. Ultimately, an understanding
of these pathways may provide novel strategies
on when and how forces should be applied to
teeth.
Home | TOC |          
Index
11
References
1. Aubin JE, Liu F. The osteoblastic lineage, in Bilezikian
JP, Raisz LG, Rodan GA (eds): Principles of Bone Biol-
ogy. San Diego, CAAmerican Press, 1996, chap 5.
2. Doty SB. Morphological evidence of gap junctions be-
tween bone cells. Calcif Tissue Int 1981;33:509-512.
3. Suda T, Udagawa N, Takahashi N. Cells of bone: Oste-
oclasts generation, in Bilezikian JP, Raisz LG, GA Rodan
(eds): Principles of Bone Biology. San Diego, CA:Ameri-
can Press, 1997, chap 7.
4. Roodman GD. Cell biology of the osteoclast. Exp Hema-
tol 1999;27:1229-1241.
5. Wozney JM, Rosen V, Celeste AJ, et al. Novel regulators
of bone formation: Molecular clones and activities. Sci-
ence 1988;242:1528-1534.
6. Komori T, Kishimoto T. Cbfal in bone development.
Curr Opin Genet Dev 1998;8:494-499.
7. Otto F, Thornell AP, Crompton T, et al. Cbfal, a candi-
date gene for cleidocranial dysplasia syndrome, is essen-
tial for osteoblast differentiation and bone development.
Cell 1997;89:765-771.
8. Ducy P, Zhang R, Geoffrey V, et al. Osf2/Cbfal: A
transcriptional activator of osteoblast differentiation.
Cell 1997;89:747-754.
9. Aubin JE, Bonnelye E. Osteoprotegrin and its ligand: A
new paradigm for regulation of osteoclastogenesis and
bone resorption. Osteoporos Int 2000;! 1:905-913.
10. Simonet WS, Lacey DL, Dunstan CR, et al. Osteoprote-
ge rin: A novel secreted protein involved in the regula-
tion of bone density.. Cell 1997;89:309-319.
11. Tsuda E, Goto M, Mochizuki S, et al. Isolation of a novel
cytokine from human fibroblasts that specifically inhibits
osteoclastogenesis. Biochem Biophys Res Commun
1997;234:137-142.
12. Bekker PJ, Holloway D, Nakanishi A, et al. The effect of
a single dose of osteoprotegerin in postmenopausal
women. J Bone Miner Res 2001;16:348-360.
13. Kobayashi Y, Hashimoto F, Miyamoto H, et al. Force-
induced osteoclast apoptosis in vivo is accompanied by
elevation in transforming growth factor beta and osteo-
protegerin expression. J Bone Miner Res 2000; 15:1924-
1934.
14. Yasuda H, Shima N, Nakagawa N, et al. Identity of oste-
oclastogenesis inhibitory factor (OCIF) and osteoprote-
gerin (OPG): A mechanism by which OPG/OCIF inhib-
its osteoclastogenesis in vitro. Endocrinology 1998;139:
1329-1337.
15. Lacey DL, Timms E, Tan HL, et al. Osteoprotegerin
ligand is a cytokine that regulates osteoclast differentia-
tion and activation. Cell 1998;93:165-176.
16. Hofbauer LC, Khosla S, Dunstan CR, et al. The roles of
osteoprotegerin and osteoprotegerin ligand in the para-
crine regulation of bone resorption. J Bone Miner Res
2000;15:2-12.
17. Rubin CT. Skeletal strain and the functional significance
of bone architecture. Calcif Tissue Int 1984;36:S11-S18.
18. Klein-Nulend J, Semeins CM, Burger EH. Prostaglandin
mediated modulation of transforming growth factor-
beta metabolism in primary mouse osteoblastic cells in
vitro. J Cell Physiol 1996;168:l-7.
19. Duncan R, Misler S. Voltage-activated and stretch-acti-
<<    
     Article
      >>
12 Dolce, Malone, and Wheeler
vated Ba2+ conducting channels in an osteoblast-like
cell line (UMR 106). FEBS Lett 1989;251:17-21.
20. Opsahl LR, Webb WW. Transduction of membrane ten-
sion by the ion channel alamethicin. Biophys J 1994;66:
71-74.
21. Davidson RM, Tatakis DW, Auerbach AL. Multiple forms
of mechanosensitive ion channels in osteoblast-like cells.
Pflugers Arch 1990;416:646-651.
22. Ingber DE. Integrins as mechanochemical transducers.
Curr Opin Cell Biol 1991 ;3: 841-848.
23. Carvalho RS, Scott JE, Yen EH. The effects of mechanical
stimulation on the distribution of beta l integrin and
expression of beta l-integrin mRNA in TE-85 human
osteosarcoma cells. Arch Oral Biol 1995;40:257-264.
24. Ruoslahti E, Pierschbacher MD. New perspectives in cell
adhesion: RGD and integrins. Science 1987;238:491-497.
25. Puleo DA, Bizios R. RGDS tetrapeptide binds to osteo-
blasts and inhibits fibronectin-mediated adhesion. Bone
1991;12:27l-276.
26. Akiyama SK, Olden K, Yamada KM. Fibronectin and
integrins in invasion and metastasis. Cancer Metastasis
Rev 1995;14:173-189.
27. Clover J, Dodds RA, Gowen M. Integrin subunit expres-
sion by human osteoblasts and osteoclasts in situ and in
culture. J Cell Sei 1992;103:267-271.
28. Frisch SM, Francis H. Disruption of epithelial cell-matrix
interactions induces apoptosis. J Cell Biol 1994;124:619-
626.
29. Engleman VW, Nickols GA, Ross FP, et al. A peptidomi-
metic antagonist of the alpha(v)beta3 integrin inhibits
bone resorption in vitro and prevents osteoporosis in
vivo. J Clin "invest 1997;99:2284-2292.
30. Terai K, Takano-Yamamoto T, Ohba Y, et al. Role of
osteopontin in bone remodeling caused by mechanical
stress. J Bone Miner Res 1999; 14:839-849.
31. Kurata K, Uemura T, Nemoto A, et al. Mechanical strain
effect on bone-resorbing activity and messenger RNA
expressions of marker enzymes in isolated osteoclast
culture. J Bone Miner Res 2001;16:722-730.
32. Sandy JR, Farndale RW, Meikle MC. Recent advances in
understanding mechanically induced bone remodeling
Home | TOC |          
Index
and their relevance to orthodontic theory and practice.
Am J Orthod Dentofacial Orthop 1993;103:212-222.
33. Lau LF, Nathans D. Genes induced by serum growth
factors, in Foulkes PCAJ (ed): The Hormonal Con-
trol Regulation of Gene Transcription. New York,
Elsevier,1991, pp 257-293.
34. Dolce C, Kinniburgh AJ, Dziak R. Immediate early-gene
induction in rat osteoblastic cells after mechanical de-
formation. Arch Oral Biol 1996;41:1101-1108.
35. Lian JB, Stein GS, Bortell R, et al. Phenotype suppres-
sion: A postulated molecular mechanism for mediating
the relationship of proliferation and differentiation by
Fos/Jun interactions at AP-1 sites in steroid responsive
promoter elements of tissue-specific genes. J Cell Bio-
chem 1991;45:9-14.
36. Raab-Cullen DM, Thiede MA, Petersen DN, et al. Me-
chanical loading stimulates rapid changes in periosteal
gene expression. Calcif Tissue Int 1994;55:473-478.
37. King GJ, Keeling SD, McCoy EA, et al. Measuring dental
drift and orthodontic tooth movement in response to
various initial forces in adult rats. Am J Orthod Dento-
facial Orthop 1991;99:456-465.
38. SambrookJ, Fritsch EF, Maniatis T. Molecular Cloning:
A Laboratory Manual, vols 1-3. Cold Spring Harbor, NY:
Cold Spring Harbor Laboratory Press, 1989.
39. Chambers TJ, Fox S, Jagger CJ, et al. The role of pros-
taglandins and nitric oxide in the response of bone to
mechanical forces. Osteoarthritis Cartilage 1999;7:422-
423.
40. Herrlich P, Ponta H. 'Nuclear' oncogenes convert extra-
cellular stimuli into changes in the genetic program.
Trends Genet 1989;5:112-115.
41. Roberts WE, Chase DC. Kinetics of cell proliferation and
migration associated with orthodontically-induced os-
teogenesis. J Dent Res 1981;60:174-181.
42. Sadoshima J, Izumo S. Mechanical stretch rapidly acti-
vates multiple signal transduction pathways in cardiac
myocytes: Potential involvement of an autocrine/para-
crine mechanism. EMBO J 1993;12:1681-1692.
43. Chow JW, Fox SW, Lean JM, et al: Role of nitric oxide
and prostaglandins in mechanically induced bone for-
mation. J Bone Miner Res 1998;13:1039-1044.
<<    
     Article
      >> Home | TOC |          
Index

Teeth in a Genetic Age

Heleni Vastardis

The outline of the information needed to create a human being, the draft
sequences of the human genome, are now published. Refinement of the
data lies ahead, but the implications for studying our biological selves are
already profound. The Human Genome Project has assisted significantly in
the investigation of dental anomalies and specifically in the genetic roots of
tooth agenesis. Dentistry will gain a great deal in diagnosis, prevention, and
therapeutics from the interpretation and application of the genetic research
data. (Semin Orthod 2002;8:13-16.) Copyright 2002, Elsevier Science (USA). All
rights reserved.

We live in an era when genetics has been
brought to center stage. In talking about
technological advances in dentistry, one cannot
but gaze with awe on the tremendous accom-
plishments of the Human Genome Project
(HGP).
What a long and exciting trip it has been,1
from the time the father of the science of genet-
ics, Gregor Mendel (1822-1884), the man who
created a method to investigate heredity, an-
nounced his theory in a meeting in 1865, to
1901 when biologist William Bateson coined the
word genetics, and finally to 2001 when biology's
gigaproject, the so-called Human Genome Se-
quencing, came into focus at the Nature Web
site (http://www.nature.com). A century after the
first description of Mendelian human genetics,
Nature and Science contain the first reports of the
complete sequence of the human genome. "My
time will come," Mendel said expressing his con-
viction of the significance of his theory, and he
was right again.
The human genome sequence contains the
genetic code that resides at the core of every 1 of
From the Department of Orthodontics, Tufts University School of
Dental Medicine, Boston, MA, and the University of Athens Faculty
of Dentistry, Athens, Greece.
Supported in part by Howard Hughes Medical Institute, Na-
tional Hellenic Scholarship Foundation, Alexander S. Onassis
Foundation, and March of Dimes Birth Defects Foundation.
Address correspondence to Heleni Vastardis, DDS, DMSc, 31
Agias Sophias Street, Neo Psychico 15451, Greece.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0801-0003$35.00/0
doi:10.1053/sodo.2002.287l4
the 10 trillion cells composing each human be-
ing. These 6 feet of DNA per cell, coiled and
tightly packed into 46 chromosomes, withstand
the strenuous task of duplication every time the
cell divides. Additionally, about 50,000 different
kinds of proteins must be supplied "in the right
quantities, at the right times, and to the right
places" for the well-being of the human mind
and body.3 Health is truly a miracle considering
the estimated de novo mutation rate (30 muta-
tions per lifetime) and the hundreds of genetic
mutations we all inherit from our parents. For-
tunately, most of these misprints are safe be-
cause they do not affect the decision-making
parts of a gene.3
A range between 25,000 and 40,000 genes
contains the recipe that defines us. This is sur-
prising because it represents only twice the num-
ber needed to make a fruit fly, a worm, or a
plant. We can blame our genome that the 5.5
billion people on this planet are so similar and at
the same time so different. Eye color, psychiatric
disorders, prognathism, susceptibility to infec-
tion, 32 or fewer teeth, and any other differences
between two unrelated individuals could be con-
tributed to a surprisingly small number of dis-
crete genetic instructions.2'3
Where do teeth stand in this pursuit of the
"code"? From descriptive dental histology and
studies on tooth development, we have known
that the interaction of oral epithelial cells and
the underlying mesenchymal cells of neural
crest descent is required to form a tooth. The
fundamental process is similar for all teeth. Each
tooth develops through successive bud, cap, and

Seminars in Orthodontics, Vol 8, No 1 (March), 2002: pp 13-16 13

<<    
     Article
      >>
14 Heleni Vastardis
bell stages. During these early stages, the tooth
germs grow and expand, and the cells differen-
tiate to form the hard tissues of the teeth. Sig-
nificant biological information has emerged
over the preceding 50 years in the field of odon-
togenesis providing clues on the genetic net-
works regulating tooth development. Not sur-
prisingly, every stage of dental development has
been shown to be under strict genetic control.
Genes do play an important role in orchestrat-
ing dental development.4 However, the focus of
these studies has not been directed toward hu-
mans, and the molecular basis of human dental
development has yet to be defined.
Human molecular genetics provides a new
methodology to answer old questions. The same
methodology can equally be applied to dental
and medical research. Having a dental back-
ground, I have put the following questions to
myself: (1) Why are teeth missing? (2) Why are
some of them missing more often than others?
(3) What accounts for the specificity of tooth
shape and position? and (4) Is tooth agenesis
genetic in origin?
Answering the last question has become easy.
When Nobel laureate Paul Berg of Stanford Uni-
versity in a cancer symposium a few years ago
said "every human affliction is genetic in origin"
people thought he was exaggerating, but it has
turned out that he was not.
Agenesis of one or more teeth is the most
common developmental anomaly of the human
dentition. Although trauma, radiation, and sys-
tematic destruction of the developing tooth bud
are likely origins of this condition, clinical stud-
ies clearly show that a considerable portion of all
cases is inherited. Familial tooth agenesis is
transmitted as an autosomal dominant, reces-
sive, or X-linked condition. There are numerous
reports in the literature on the clinical genetics
of tooth agenesis.5
The primary dentition is less frequently af-
fected (0.1%) than the permanent. The decidu-
ous tooth at stake is usually an incisor followed
by the first upper molar.6'7 In the permanent
dentition, the incidence of tooth agenesis varies
with different teeth; third molars are missing
most frequently (30%) followed by mandibular
second premolars (3.4%) and maxillary lateral
incisors (2.2%). 8 Incisal hypodontia, in particu-
lar, presents a marked variability. In approxi-
mately 1.9% of the population, one of the lateral
Home | TOC |          
Index
incisors is missing, and the other has a conical
shape, the so-called peg lateral. Pegs form bilat-
erally at a frequency of 1.6%; a similar percent-
age appears when a peg is combined with a
well-shaped, but small-sized lateral.9 The inci-
dence drops dramatically in whites when tooth
agenesis affects central incisors, especially the
maxillary, canines, and first permanent mo-
lars.10'12 There are few cases reported in the
literature with simultaneous presence of super-
numerary teeth and missing teeth.12
In the general white population, tooth agen-
esis ranges from 1% to 9.6%, between 6.6% and
9.2% in the Japanese population, and 7.7% in
the black population.13'14 The woman to man
ratio ranges from 3:2 for whites to 2:1 for
blacks.15'16
In the cases of primary dentition agenesis,
there is an increased chance for permanent
teeth failing to develop.17 However, in the ma-
jority of permanent tooth agenesis, the primary
dentition is intact, suggesting different genetic
mechanisms for the two sets of teeth.
Dentistry in the HGP
Back in 1992, the time when the HGP was show-
ing tremendous promise, it was decided to ap-
proach the mystery of odontogenesis with a hu-
man molecular genetic approach and learn
from a genetic error. Medical research had al-
ready set the example. The genes responsible for
the adult form of muscular dystrophy and a form
of dwarfism were identified. Much of the
progress in reading DNA was coming from anal-
yses of genetic defects.
The clinical observation was that tooth agen-
esis is phenotypically variable. On the basis of
this phenotypic variability and the hereditary
character of tooth agenesis, it was hypothesized
that defects in each of the several molecules
associated with the steps of odontogenesis result
in the clinical variation of tooth agenesis. The
objective became the identification of the dental
genes that become mutated in families with miss-
ing teeth.
Application of the HGP methodology de-
pends on availability of human material and spe-
cifically on the willingness of the families with
the condition of interest to donate a small quan-
tity of their blood. After the clinical information
<<    
     Article
      >>
Teeth in a Genetic Age
is gathered and secure diagnoses are made, the
experimental part of the process begins.
Systematic genome-wide search through ge-
netic linkage analyses has identified the location
of a defective dental gene in a family with an
autosomal dominant agenesis of second premo-
lars and third molars on chromosome 4p. The
critical locus included the MSX1 gene, and mu-
tational analysis showed a missense mutation in
the homeodomain, the most highly conserved
region of the MSX1 gene.18 Further biochemical
and functional analyses showed that this specific
mutation renders the gene inactive, and, there-
fore, the reduced amount of MSX1 causes the
particular phenotype in this family.19
Since 1996, when the MSX1 gene was first
associated with familial tooth agenesis, the ge-
netics of tooth agenesis has received so much
attention that it has become one of the most
favored research areas. Significant work is being
carried out, and an increasing number of genes
involved in the regulation of tooth morphogen-
esis is being identified as summarized later.
A family with autosomal recessive hypodontia
associated with various dental anomalies, such as
enamel hypoplasia and failure of eruption, is
linked to a region on chromosome 16q.20 There
is no report on the specific genetic defect re-
sponsible for this type of hypodontia, as yet.
A family with agenesis of permanent molars is
associated with a frame-shift mutation of the
paired domain PAX9 gene through genome-
wide linkage analyses and candidate gene se-
quencing.21
Another MSX1 mutation, (MetGlLys) up-
stream of the homeodomain of MSX1, is identi-
fied in a family presenting hypodontia.22
MSX1 is also associated with syndromic forms
of tooth agenesis. A nonsense mutation in the
MSX1 gene in a family with autosomal dominant
tooth agenesis and combinations of cleft palate
only and cleft lip and palate provides evidence
for the importance of this gene in craniofacial
development and offers an appreciation of the
associations between different organs in the de-
veloping body.23
A nonsense mutation in the homeodomain of
MSX1 has also been found to cosegregate with
the phenotype of Witkop syndrome, also called
tooth and nail syndrome, suggesting that MSX1
is critical for both tooth and nail development.24
In the last six years, two independent groups
Home | TOC |          
Index
15
failed to associate MSX1 with familial or isolated
tooth agenesis.25'26 Their results, however, agree
with our genetic heterogeneity hypothesis that
more than one gene is responsible for the phe-
notypic variability of tooth agenesis.
Conclusions
We are used to describing human beings in an
anatomic manner. The time has come when a
number of genes will describe us. Physical char-
acteristics (phenotype) may matter less than the
set of genes one carries (genotype). Classifica-
tion of characteristics will soon be based on
genetic analyses. Orthodontists may have a diffi-
cult time changing their morphologic classifica-
tions (eg, Angle). However, the move from phe-
notypic to genotypic classification is inevitable.
The HGP has made a promise. If we answer
the following three questions, we will most likely
be able to prevent or treat any of the estimated
4,000 human genetic diseases: (1) What is the
genetic defect responsible for the condition of
interest? (2) What is the normal gene product
(protein)? and (3) Can the defective gene/pro-
tein be fixed or replaced?
Human genetics provides a powerful tool to
study dental patterning. Genetic models will
eventually explain clinical observations such as
the anteroposterior site-specificity of tooth agen-
esis and the occurrence of associated dental
anomalies.27 Mutations of genes have already
showed some of the key genetic interactions that
determine tooth shape and position.28 Func-
tional investigation of the defects could reveal
the mechanisms underlying patterning dysmor-
phologies and could point to better understand-
ing of odontogenesis.
Although the raw genetic information has
limitations, its availability offers new experimen-
tal approaches for studying craniofacial malfor-
mation. What biological insight can be gleaned
from the draft sequence? This question surely
cannot be answered by staring at a genome.
However, it is certain that this work will have
profound long-term consequences in diagnos-
tics and targeted therapeutics.
References
1. Davies K. Cracking the genome: Inside the race to un-
lock human DNA/The sequence: Inside the race for the
<<    
     Article
      >>
16 Heleni Vastardis
human Genome. Weidenfeld and Nicolson 2001, pp
288-320.
2. Available at: http://www.nature.com.genomics/human/
papers.
3. Available at: http://www.hhmi.org/genetictrail/reading/
read.htm.
4. Peters H, Balling R. Teeth. Where and how to make
them. Trends Genet 1999;15(2):59-65.
5. Vastardis H. The genetics of human tooth agenesis: New
discoveries for understanding dental anomalies. Am J
Orthod Dentofac Orthop 2000;117(6):650-656.
6. Bennett CG, Ronck SL. Congenitally missing primary
teeth. Case report. J Dent Child 1980;47:346-349.
7. Daugaard-JensenJ, Nodal M, Kjaer I. Pattern of agenesis
in the primary dentition: A radiographic study of 193
cases. Int J Paediatr Dent 1997;7(l):3-7.
8. Symons AL, Stritzel F, Stamatiou J. Anomalies associated
with hypodontia of the permanent lateral incisor and
second premolar. J Glin Pediat Dent 1993;17:109-111.
9. Le Bot P, Salmon D. Congenital defects of the upper
lateral incisors (ULI): Condition and measurements of
the other teeth, measurements of the superior arch,
head and face. Am J Phys Anthrop 1977;46:231-244.
10. Dermaut LR, Goeffers KR, De Smit AA. Prevalence of
tooth agenesis correlated with jaw relationship and den-
tal crowding. Am J Orthod Dentofac Orthop 1986;90:
204-210.
11. Lapeer GL. Congenitally missing maxillary first perma-
nent molars: A case report. J Can Dent Assoc 1990;56(6):
535-536.
12. Sharma A. A rare non-syndrome case of concomitant
multiple supernumerary teeth and partial anodontia.
J Glin Pediatr Dent 2001;25(2):167-169.
13. Niswander JD, Sujaku C. Congenital anomalies of teeth
in Japanese children. J Phys Anthropol 1963;21:569-574.
14. Salinas CF, Jorgenson RJ. Dental anomalies in a black
population. J Dent Res 1974;53:237 (abstr).
15. Jorgenson Rf. Clinician's view of hypodontia. J Am Dent
Assoc 1980;101:283-286.
16. Egermark-Eriksson I, Lind V. Congenital numerical vari-
ation in the permanent dentition: Sex distribution of
Home | TOC |          
Index
hypodontia and hyperodontia. Odont Revy 197l;22:309-
315.
17. Daugaard-JensenJ, Nodal M, Skovgaard LT, et al. Com-
parison of the pattern of agenesis in the primary and
permanent dentitions in a population characterized by
agenesis in the primary dentition. Int J Paediatr Dent
1997;7(3):143-148.
18. Vastardis H, Karimbux N, Guthua SW, et al. A human
MSX1 homeodomain missense mutation causes selective
tooth agenesis. Nature Genet 1996;13:417-421.
19. Hu G, Vastardis H, Bendall AJ, et al. Haploinsufficiency
of MSX1: A mechanism for selective tooth agenesis. Mol
Cell Biol 1998;18(10):6044-6051.
20. Ahmad W, Brancolini V, Faiyaz ul Haque M, et al. A
locus for autosomal recessive hypodontia with associated
dental anomalies maps to chromosome 16ql2.1. Am J
Hum Genet 1998;62:987-991 (letter).
21. Stockton DW, Das P, Goldenberg M, et al. Mutation of
PAX9 is associated with oligodontia. Nature Genet 2000;
24:18-19.
22. Reising BC. The role of MSX1 mutation in the etiology
of hypodontia. Am J Orthod Dentofac Orthop 2001;
119(5):564-565 (abstr).
23. van den Boorgaard M-JH, Dorland M, Beemer FA, et al.
MSX1 mutation is associated with orofacial clefting and
tooth agenesis in humans. Nature Genet 2000;24:342-
343 (letter).
24. Jumlongras D, Bei M, Stimson JM, et al. A nonsense
mutation in MSX1 causes Witkop syndrome. Am J Hum
Genet 2001;69(l):67-74.
25. Nieminen P, Arte S, Pirinen S, et al. Gene defect in
hypodontia: Exclusion of MSX1 and MSX2 as candidate
genes. Hum Genet 1995;96(3):305-308.
26. Scarel RM, Trevilatto PC, Di Hipolito O Jr, et al. Absence
of mutations in the homeodomain of the MSX1 gene in
patients with hypodontia. Am J Med Genet 2000;92(5):
346-349.
27. Peck S, Peck L, Kataja M. Site-specificity of tooth agen-
esis in subjects with maxillary canine malpositions. Angle
Orthod 1996;66(6):473-466.
28. Thesleff L Genetic basis of tooth development and den-
tal defects. Acta Odontol Scand 2000;58(5):191-194.
<<    
     Article
      >> Home | TOC |          
Index

Three-Dimensional Cephalometry and

Three-Dimensional Skull Models in
Orthodontic/Surgical Diagnosis and
Treatment Planning
Robert A. W. Fuhrmann

In 25 adult patients with severe dentomaxillofacial deformities, we inte-

grated three-dimensional (3-D) cephalometry and 3-D-model surgery with
individually milled or stereolithographically built skull models in our com-
bined orthodontic/surgical diagnosis and treatment planning. After axial
computed tomography scanning and 3-D reconstruction, measurement pro-
cedures could be interactively used to assess skeletal asymmetries quanti-
tatively. The 3-D-model surgery represents a new level of treatment predic-
tion in individual morphology. The orthodontic setup and 3-D-model
surgery permits a verification of the feasibility of the most suitable mobili-
zation and placement of bone segments. The integration of 3-D cephalom-
etry and 3-D-model surgery in patients with severe asymmetric den-
tomaxillofacial deformities allowed a higher precision of diagnosis and
treatment planning. (Semin Orthod 2002;8:17-22.) Copyright 2002,
Elsevier Science (USA). All rights reserved.

Different diagnostic methods, such as lateral
and frontal skull radiographs and dental
casts, are integrated in the treatment of se-
vere dentomaxillofacial deformities. In cephalo-
grams, the position of jaws and teeth in relation
to the base of the skull are measured on a two-
dimensional film plane. All conventional radio-
logic techniques are limited in their ability
to accurately measure the soft tissues and the
skull in all three dimensions. Also, conventional
model surgery has a limited ability to assess max-
illofacial morphology because mounted dental
casts mainly visualize the teeth and alveolar pro-
cesses.
Progress in imaging techniques, especially
computed tomography (CT) and magnetic res-
From the Department of Orthodontics Medical Faculty, Aachen,
Germany.
Address correspondence to Professor Robert A. W. Fuhrmann,
Dr. med, Dr. med. dent., MD, DDS, PhD, Klinik für Kieferortho-
pädie, RWTH-Aachen, 52057 Aachen, Germany.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0801-0004$35.00/0
doi:10.1053/sodo.2002.28167
onance (MR) tomography, now allow images to
be obtained without overlying structures. Three-
dimensional (3-D) images can be interactively
reconstructed by data processing from axial CT
scans.1 3-D imaging allows topographic evalua-
tion of dentomaxillofacial deformities.2 Most
3-D reconstruction procedures require expen-
sive high-capacity computer hardware and soft-
ware; therefore, only radiologic centers or mod-
ern university clinics are able to integrate these
new methods.3-5 The improved performance of
usual personal computers enables CT recon-
struction.6
The data set of the CT scanner can also be
used for the computer-aided manufacturing of
individual 3-D skull models.7 Skull models were
milled directly from a polyurethane foam block
by a milling machine.8 Another method to trans-
fer CT data in skull models is stereolithography.
This technique uses a photocurable monomer
hardened by an ultraviolet laser.
Orthodontic treatment could be a part of
3-D-model treatment simulation.9 In the last 10
years, we have assessed the integration of 3-D
cephalometry and individual 3-D skull models

Seminars in Orthodontics, Vol 8, No 1 (March), 2002: pp 17-22 17

<<    
     Article
      >> Home | TOC |          
Index

18 Robert A. W. Fuhrmann

during combined orthodontic and surgical diag- Case Report

nosis and treatment planning. Does the integra-
tion of a 3-D working concept improve the pre-
cision of the treatment prediction in cases of
severe dentomaxillofacial morphology?
Materials and Methods
In 25 adult patients with severe dentomaxil-
lofacial deformities, contiguous axial CT scans
were made of the region of interest by using a
Siemens Somatom plus CT scanner (Siemens
Inc, Erlangen, Germany). CT scanning was per-
formed at 120 kV and 170 mA, with a slice
thickness and slice increment of 1 to 2 mm in all
cases. The gantry had zero inclination. The CT
data have a scanning matrix size of 256 X 256
pixels, resulting in a maximal resolution of 1
mm.
In a 22-year-old man, a skeletal deviation of
the lower face to the right side was noted. A
hyperplastic condyle and elongation of the as-
cending ramus on the left side of the mandible
was combined with mandibular prognathism. In-
traorally, a crossbite with an asymmetric Angle
Class III dentition was combined with a midline
deviation of the lower arch of 10 mm to the right
side. A lateral open bite was evident on the left
side. The patient also showed an extreme cant of
the maxillary and mandibular occlusal planes
(Fig 1).
After contiguous CT scanning with 2-mm
thick axial slices and transfer of the data set to
the personal computer, several 3-D views of the
skin and skull surface were generated. Simulta-
neous 3-D evaluation of the topographic corre-
lation between the soft and hard tissue was

Generation of 3-D Images

For 3-D reconstruction, the CT data sets of all
25 patients were transferred with disks to a per-
sonal computer. The 3-D-imaging software pro-
cesses all CT data from any scanner for 3-D
reconstructions.
After automatic transfer of the file format, the
CT data can be separated according to tissue
density. Further interactive differentation called
thresholding makes it possible to differentiate
between soft and hard tissues in each scan by
defining the range of Hounsfield units (HU).
For 3-D reconstruction of the skeletal surface, all
voxels representing a tissue density greater than
150 HU were selected from the CT data set. 3-D
surface reconstruction is performed by linear
interpolation between the existing voxels. After
these initial procedures, several software tools
allow a cephalometric evaluation of the skull in
all three dimensions.

Generation of 3-D Skull Models

For 3-D skull fabrication, the CT data sets of
14 patients were transferred to commercial com-
panies. To produce individual 3-D skull models,
two different concepts were used. In 12 cases,
individual life-size polyurethane foam models
Figure 1. Facial appearance of a 22-year-old patient
were milled.7'8 The CT data sets of two patients with an extreme skeletal deviation of the lower face to
were transferred into individual 3-D skull models the right side caused by condylar and mandibular
by stereolithography. hyperplasia on the left side.
<<    
     Article
      >>
3D Cephalometry and 3D Skull Models
performed with transparent skin or bone recon-
struction (Figs 2A and 2B). Superimposition of
different segmented and colored hard and soft
tissue may also be split in any direction to enable
visualization of the internal architecture of the
craniomaxillofacial morphology (Fig 3). Freely
defineable segments were removed digitally, and
the remaining 3-D object could be evaluated in
any direction. This allows quantitative assess-
ment of the real dimensions of the lips and
skeletal structures (Fig 3).
The degree of asymmetry of the mandible was
evaluated quantitatively by superimposing a
Figure 2. Transparent 3-D reconstruction of the skin
surface above the osseus surface contours of the up-
per and lower jaw (A and B).
Home | TOC |          
Index
19
Figure 3. Digital removal of part of the skin surface
over the facial skeleton to assess the topographic cor-
relation and dimension of soft and hard tissues.
freely defineable measurement grid. For assess-
ment of the mandibular deviation, the grid was
orientated along the skeletal midline of the max-
illa. Besides the different size and shape of both
condyles, the whole mandible of this patient had
asymmetric vertical, sagittal, and transversal skel-
etal relations.
After definition of corresponding landmarks
(pogonion, gonion, condylion) at the bone sur-
face of the mandible, several distances were mea-
sured directly in the 3-D reconstruction on the
monitor. The measurements allowed the com-
parison of the different height and length of the
left and right mandibular halves (Fig 4). The
discrepancy in the left and right mandibular
height according to the measurements from the
condyle points to the corresponding gonion
points was 22 mm (Figs 4 and 5). For the length
of both mandible sides, the distance between the
central pogonion point and the corresponding
left and right gonion points differed at 10 mm
(Figs 4 and 5). The defined landmarks and mea-
surement procedures were constantly displayed
at the monitor during further rotation of the
mandible or digital disarticulation of the man-
dible (Fig 5).
From the CT data set of this patient, an indi-
vidual 3-D skull model of polyurethane foam was
milled. The skull model was inserted in a semi-
adjustable articulator. In the 3-D model the im-
precisely milled dental arches of the skull were
<<    
     Article
      >>
20 Robert A. W. Fuhrmann
Figure 4. 3-D cephalometric analysis of the visceral
skeleton with transparent maxillary bone structures;
definition of corresponding measurement points
(pogonion, gonion, condylion) at the bone surface
and measurement distances to compare the length
(Dl, D2) and height (D3, D4) of both halves of the
mandible.
replaced by precise maxillary and mandibular
casts (Fig 6). The treatment prediction was per-
formed step-by-step according to the real clinical
treatment plan. Before the orthodontic and or-
thognathic setup was started, both condyles were
fixed to the base of the skull with wires of de-
fined length. The orthodontic treatment goal
Figure 5. Rotation of the mandible after digital re-
moval of the maxilla; constancy of all measurement
points and distances to compare the symmetry of both
mandible halves from another viewpoint.
Home | TOC |          
Index
Figure 6. Individual 3-D skull model after combined
orthodontic and surgical treatment simulation.
was simulated with a setup of the teeth in the
cast models. The orthodontic treatment was sim-
ulated with a lateral movement of the upper
central incisors for median dissection of the
maxilla. The first left premolar in the lower jaw
was moved distally and derotated according to
the planned vertical osteotomy line.
Surgical treatment simulation began with a
Le-Fort I osteotomy. The maxilla was sagittally
dissected at the midline between the central
incisors. Both segments were tilted to straighten
the maxillary occlusal plane. The left maxillary
segment was distally impacted 3 mm, and the
distal part of the right segment was moved down-
ward 6 mm, fixing with miniplates (Fig 6).
The mandible was first dissected with a bilat-
eral retromolar split osteotomy. In addition, a
mandibular osteotomy was made vertically on
the left side between the canine and the first
premolar. After removal of a 6-mm triangular
area of bone, both mandibular segments were
fixed with two miniplates. After rotation of the
mandible by 10 mm to the left, the occlusion was
stabilized with intermaxillary fixation. The bisag-
ittal retromolar osteotomies were fixed with wire
ligatures (Fig 6).
<<    
     Article
      >>
3D Cephalometry and 3D Skull Models
Results
3-D imaging allows a detailed overview of the
topographic relations and shapes of the den-
tomaxillofacial morphology in all planes of
space from any viewpoint without superimposi-
tioning of anatomic structures. The cephalomet-
ric analysis was not limited to a frontal or lateral
view. The standardized computing system dis-
plays any 3-D reconstructions of the skin and
bone surface of the skull of patients as well as
secondary reformatted slices on the monitor.
Different segmented and colored tissues could
be combined in one 3-D image. After digital
removal of tissue parts or a complete 3-D object,
the remaining anatomic structures can be eval-
uated. This allows the skin, maxilla, and mandi-
ble to be assessed separately or together. After
the definition of several landmarks to measure
the skin and skull relations, the 3-D reconstruc-
tions could be freely rotated. The constancy of
the measurement points prevent any reidentifi-
cation error during 3-D cephalometric evalua-
tion. A precise transfer of well-known cephalo-
metric landmarks, obtained from lateral and
frontal cephalograms, is limited in the 3-D im-
age. Therefore, landmarks were denned in the
3-D reconstructions in the individual.
Several measurement procedures allow 3-D
cephalometric analysis directly on the 3-D ob-
jects. The degree of asymmetry was measured
with two different methods: the skull and the
skin surface and all slices were evaluated quan-
titatively with a superimposable grid. This mea-
surement method facilitates comparison of cor-
reponding anatomic landmarks according to a
defined plane of symmetry. The second method
uses denned distances and angles to assess the
asymmetry. During rotation and alteration of
the skull on the monitor, the defined distances
and angles were constantly visualized on the
screen.
Digital removal of segments, segment dis-
placement, and rearrangement of segments rep-
resent the basic elements of interactive planning
and initial treatment simulation on the monitor.
Treatment prediction with individual 3-D skull
models represent an individual dental and skel-
etal ideal position in each case. The clinical
realization follows step-by-step. The integration
of individual skull models represents a new level
of orthodontic and orthognathic treatment sim-
Home | TOC |          
Index
21
ulation, which allows more realistic 3-D visualiza-
tion and tactile feedback of the complex ana-
tomic morphology, especially skeletal deviations
in relation to the position of the teeth. The
milled or stereolithographic models represent
the individual bony structures and facilitate as-
sessment of the feasibility of planned osteotomy
cuts and the optimal segment displacement in
relation to the aligned dental arches. The need
for presurgical orthodontic treatment can be
simulated with a setup according to the required
positions of the teeth in the cast models. In cases
of unforeseen clinical influences during the
orthodontic pretreatment, the initial dental
models could be replaced by actual casts. The
replacement of the orthodontic situation with
casts into the skull models needs further devel-
opment to avoid vertical, sagittal, or transverse
transfer errors.
The new segment placement can be assessed
according to the skeletal and dentoalveolar sym-
metry. After segment displacement, expected
gaps at osteotomy sites can be visualized, and the
need for bone grafting or configuration of os-
teotomy planes can be evaluated. The defined
fixation of the condyles at the skull base also
allows mobility control in the temporomandibu-
lar joint during the treatment simulation.
Discussion
In comparison with the classic two-dimen-
sional cephalometric radiograph, 3-D images do
not suffer from superimposing of all structures
on a film plane and different enlargement fac-
tors according to the distance from the radio-
graphed structures to the radiograph tube. Gra-
ber10 (1995) has critically noted that with
conventional radiographic techniques we see lit-
tle more than 50% of the real anatomic struc-
tures. His call to integrate 3-D imaging in diag-
nostic and therapeutic efforts has to be
differentiated between skeletal and periodontal
requirements. The CT scanning mode used for
3-D cephalometric assessment in the present
study allowed a maximum resolution of 1.0 mm.
Maximal accuracy is also limited by the partial
volume effect. 3-D cephalometry allows all mea-
surement procedures in space from any view-
point. The topographic correlation and dimen-
sion of segmented hard and soft tissues can be
evaluated true to scale with different measure-
<<    
     Article
      >>
22 Robert A. W. Fuhrmann
ment tools in the 3-D object on the monitor. The
access of 3-D imaging with a readily available
computer system for 3-D cephalometry of the
head will enable a larger number of clinicians to
use 3-D methods in diagnosis and treatment
planning. The offline personal computer and
the user-specified software permit a higher de-
gree of flexibility independent of expensive
workstations or radiologic centers.
The plasticity of the visceral skeleton pro-
vided by a life-size 3-D skull model is a precon-
dition for realistic simulation of the complex
treatment sequence. 3-D skull model-based sur-
gery on milled or stereolithographic built mod-
els permit verification of the most suitable mo-
bilization and placement of jaw segments. The
individual prediction could be copied during
the real clinical treatment. 3-D-model surgery
with integrated orthodontic setups represents a
new dimension of treatment prediction. How-
ever, because of the cost and time consumption,
a 3-D working concept with 3-D cephalometry
and 3-D skull models is presently limited to cases
of severe asymmetric dentofacial disharmonies.
References
1. Moss JP, Grindrod SR, Linney AD, et al.A computer
system for the interactive planning and prediction of
Home | TOC |          
Index
maxillofacial surgery. Amer J Orthod Dentofac Orthop
1988;94:469-475.
2. Vannier MW, Marsh JL, Warren JO. Three-dimensional
CT-reconstruction images for craniofacial surgical plan-
ning and evaluation. Radiology 1984;150:179-184.
3. Zonneveld FG, Normaan van der Dussen F. Three-di-
mensional imaging and model fabrication in oral and
maxillofacial surgery. Oral Maxillofac Surg Clin North
Amer 1992;4:19-33.
4. Zonneveld FG, Fukuta K. A decade of clinical three-
dimensional imaging: A review. Part II: Clinical Applica-
tions. Investigat Radiol 1994;29:574-589.
5. Zonneveld FG. A decade of clinical three-dimensional
imaging: A review. Part III: Image analysis and interac-
tion, display options and physical models. Investigat Ra-
diol 1994;29:7l6-725.
6. Fuhrmann R, Schnappauf A, Diedrich P. Three-dimen-
sional imaging of craniomaxillofacial structures with a
standard personal computer. Dentomaxillofac Radiol
1995;24:260-263.
7. Lambrecht JT, Brix F. Individual skull model fabrica-
tion for craniofacial surgery. Cleft Palate J 1990;27:
382-387.
8. Fuhrmann R, Frohberg U, Diedrich P. Treatment pre-
diction with three-dimensional computer tomographic
skull models. Amer J Orthod Dentofac Orthop 1994;106:
156-160.
9. Fuhrmann R, Feifei H, Schnappauf A, et al. Integration
of three-dimensional cephalometry and 3D-skull models
in combined orthodontic/surgical treatment planning. J
Orofac Orthop 1996;57:32-45.
10. Graber TM. Comment of the editor-in-chief. Amer
J Orthod Dentofac Orthop 1995;107:360 (letter).
<<    
     Article
      >> Home | TOC |          
Index

Three-Dimensional Evaluation of Periodontal

Remodeling During Orthodontic Treatment
Robert A. W. Fuhrmann

In 21 adult patients, two or three high-resolution computed tomography

(HR-CT) examinations were performed before, during and after orthodontic
treatment with fixed appliances. The time between the first, second, and
third CT scanning varied according to the individual treatment or retention
period from 12 to 36 months. Comparison of the CT examinations permits
three-dimensional evaluation of osteoclastic and osteoblastic periodontal
remodeling. The incidence of periodontal lesions, such as bone dehiscenses,
fenestrations and root resorptions were assessed in relation to the initial
periodontal situation and the applied orthodontic biomechanics. Anatomic
risks were a small alveolar process, thin buccal or lingual bone plates,
eccentric positioning of teeth, basally extended maxillary sinus, and pro-
gressive alveolar bone loss. Therapeutic risks were uncontrolled sagittal or
vertical movements of the incisors and cortical or intermaxillary anchorage
preparation. Orthodontically induced bone dehiscences were partly repaired
by osteoblastic periodontal remodeling in the retention period. (Semin
Orthod 2002;8:23-28.) Copyright 2002, Elsevier Science (USA). All rights
reserved.

Introduction the type and direction of tooth movement, scale

T he influences of orthodontic treatment on
the gingiva, marginal periodontium, at-
tachment levels and root integrity have been
previously discussed. 1-3 In the orthodontic treat-
ment of adults, the focus is on involutive ageing
processes of the periodontal ligament and a vary-
ing degree of alveolar bone loss. This periodon-
tal involution increases the risk of bone dehis-
cences, fenestrations, and root resorption.
Various animal experiments have shown that the
loss of thin bone plates may be induced by orth-
odontic tooth movement.1'2 After reorientation
to the initial position, attachment loss can be
osteoblastically regenerated. The genesis of
bone dehiscences or fenestrations depends on
From the Department of Orthodontics Medical Faculty, RWFH-
Aachen, 52057 Aachen, Germany.
Address correspondence to Professor Robert A. W. Fuhrmann,
Dr. med, Dr. med. dent., MD, DDS, PhD, Klinik für Kieferortho-
pädie, KWTH-Aachen, 52057 Aachen, Germany.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0801-0005$35.00/0
doi: 10.1053/sodo. 2002.28168
and duration of force application, volume and
integrity of the surrounding hard and soft tissue,
osteoblastic potential of the periosteum, bone
elasticity, coupling between bone resorption and
apposition, and vascularity of the tissues.
The clinical outcome of extensive bone dehis-
cences is a loss of anchorage and a predisposi-
tion to the development of gingival recession.
Gingival recession may become manifest during
orthodontic treatment or even many years later
through an increasing crown length and ex-
posed cervices.2'3 Buccal and lingual bone plates
or bone dehiscences on the roots of the teeth
cannot be imaged with dental radiologic projec-
tion techniques.4 Graber noted critically that
with dental and panoramic radiographs and lat-
eral cephalograms we see little more than 50%
of the real anatomic structures.5 He calls for
integration of alternative imaging techniques.5
Various in vitro and in vivo studies have
shown the applicability of high-resolution com-
puted tomography (HR-CT) in imaging the
tooth/bone relation true to scale without over-

Seminars in Orthodontics, Vol 8, No 1 (March), 2002: pp 23-28 23

<<    
     Article
      >>
24 Robert A. W. Fuhrmann
lying structures, and in identifying periodontal
lesions.4'6'7
This review summarizes our experiences over
the last 10 years integrating HR-CT scanning
during and after orthodontics. The aim was to
investigate the incidence of bone dehiscences or
fenestrations, root resorption, and osteoblastic
bone remodeling during and after orthodontic
treatment in adult patients.
Material and Methods
Twenty-one adult patients with reduced peri-
odontal bone tissue (ie, narrow alveolar process,
basally extended maxillary sinus, or advanced
alveolar bone loss) were selected for this inves-
tigation. In 15 of the patients, the first CT exam-
ination was performed before the start of orth-
odontic treatment. The remaining 6 patients
were scanned after the first phase of orthodontic
treatment.
All patients were treated with fixed straight
wire appliances (0.022-in slot width) and contin-
uous arch wires. Additional segment arch wires
were integrated during the levelling and guid-
ance phases. The second CT examination was
performed at an interval of 12 to 36 months after
the first CT examination according to the treat-
ment time following removal of the fixed appli-
ances. In 6 patients, a third CT examination was
performed after a retention period of 6 to 36
months after debonding.
Comparability of the patients was limited by
the varying labiolingual width of the alveolar
process, attachment level, and orthodontic treat-
ment concepts. For that reason, a descriptive
evaluation of the CT findings was made with
reference to the individual morphologic situa-
tion and the biomechanic treatment concept to
highlight different anatomic or therapeutic risk
factors in the genesis of periodontal lesions.
CT Scanning Parameters
The patients were placed horizontally on the
table of the CT scanner (Somatom plus, Siemens
Ine, Erlangen, Germany). Each patient's head
was fixed with foamed plastic to avoid movement
artifacts. After an initial lateral topogram, the
axial CT scanning direction was orientated ac-
cording to the maxillary or mandibular occlusal
plane. The gantry inclination was individually
Home | TOC |          
Index
adapted when necessary. The contiguous axial
CT scans of the jaw ran from the cemento-
enamel junction to the apices of the roots with a
scan thickness and table feed of 1.0 mm. Accord-
ing to the curve of Spee and the root length, 15
to 30 scans per jaw were necessary. The CT data
were recorded in high-resolution mode with a
reconstruction matrix of512x512 pixels, a tube
voltage of 120kV at 125 to 165mA and a 2-second
exposure scanning time. For imaging, the axial
CT scans were exposed onto film with a window
width of 4000 Hounsfield units (HU) and a cen-
ter of 600 HU with a laser camera. To facilitate
visualisation of the axial scan sequences and to
simplify qualitative assessment of the bone sur-
face (dehiscence diagnosis), additional three-
dimensional reconstructions were computed
from the primary axial scans for the maxilla and
mandible of selected patients with the recon-
struction software of the CT console.
Case Report
In an 18-year-old patient with a primarily narrow
mandibular alveolar process and crowding of
the lower incisors, the first premolars were ex-
tracted after the pre-treatment CT-examination
(Fig 1). The axial CT scans revealed that the oval
section of the canine roots corresponded to the
maximal width of the alveolar process. However,
no bone dehiscences and fenestrations over the
buccal surface of the lower canines and incisors
could be evaluated in the axial CT scans or
three-dimensional reconstructions (Fig 1A and
B). During the orthodontic treatment, the lower
canines were derotated and distalized with Sen-
talloy springs. The further levelling of the fron-
tal mandibular crowding was accomplished with
the insertion of resilient continuous arch wires.
This resulted in the further protrusion of the
lower incisors.
After two years of orthodontic treatment with
straight wire appliances, a second CT examina-
tion showed extensive vestibular bone dehis-
cences above the root surfaces of the lower ca-
nines and the incisors in the three-dimensional
reconstruction (Fig 1C). Because of the narrow
alveolar process, this uncontrolled protrusion
led to bone dehiscences over the vestibular root
surfaces. The extent of primary bone covering
and secondary subperiosteal apposition were not
sufficient to prevent bone dehiscences. The cor-
<<    
     Article
      >> Home | TOC |          
Index

Periodontal Remodeling During Orthodontics 25

Figure 1. Rotated canines and crowding of the incisors are found in the pre therapeutic mandibular situation
(A). Corresponding pretreatment CT examination; SD-reconstruction: Narrow frontal mandibular alveolar
process with thin facial bone plates (B). Corresponding posttreatment CT examination after two years of
orthodontic treatment; SD-reconstruction: Extensive facial dehiscences above the roots of the incisors and
canines (C). Corresponding clinical situation and CT examination after three years of retention; 3D-reconstruc-
tion: Facial dehiscences above the roots of the lateral incisors and canines were completely remodelled. The
facial dehisences above the roots of the central incisors persisted; the periodontal remodeling at the central
incisors were limited to the apical and lateral area of the dehiscences (D and E).

responding clinical situation, however, showed
no periodontal recession.
After three years of retention with removable
appliances, a third CT examination showed par-
tial periodontal remodeling of the vestibular
bone and dehiscences over the root surfaces.
The dehiscences over the root surfaces of the
lower canines and the lateral incisors were corn-
pletely filled with new alveolar bone. However,
the dehiscenses over the root surfaces of the
central incisors persisted partly. Only the apical
and lateral area of these bone defects was re-
<<    
     Article
      >>
26 Robert A. W. Fuhrmann
paired 5 (FiglE). The clinical examination
showed no differences between the lower inci-
sors (Fig ID).
Results
Comparison of the first, second, and third CT
findings revealed substantial differences in the
positions and angulations of individual teeth as a
result of orthodontic treatment, therefore, the
CT sectioning did not always correspond. There
was no clinical evidence of marginal periodonti-
tis or periodontal recession at the orthodonti-
cally moved teeth in any of the patients before,
during, or after orthodontic treatment.
The CT findings revealed substantial individ-
ual variation, depending on the extension of the
alveolar process and the therapeutic concept. In
cases of translatory tooth movements in a can-
cellous alveolar process, generally there was no
notable root resorption or bone dehiscences re-
vealed.
Expansion of the maxillary arch with a quad-
helix and continuous expanded arch wires initi-
ated a partial resorption of the covering bone
above the buccal root surfaces of the posterior
teeth. The CT scan revealed buccal bone dehis-
cences, especially in the coronal root third of the
posterior teeth. It was conspicuous that the mo-
lars and premolars at the center of maximum
expansion had the highest degree of bone re-
sorption and of subperiosteal bone apposition.
Despite the reactive osteoblastic remodeling of
the buccal bone plate and the primary bone
covering, dehiscences were seen in the second
CT examination. No corresponding lateral root
resorption, nor clinically detectable gingival re-
cession was found.
CT examination of adult patients indicates
various anatomic and/or topographic and ther-
apeutic risk factors in the initiation of periodon-
tal lesions. Some of these were anatomic and/or
morphologic risks arising from a disproportion
between tooth width and labiolingual extension
of the alveolar process or from an eccentric
tooth position, a deep maxillary sinus, or ad-
vanced bone loss. Bone dehiscences or fenestra-
tions were repeatedly found at mandibular
incisors. A small symphysis with reduced labio-
lingual bone width, frontal crowding, and thin
facial or lingual bone plates predispose to bone
dehiscences. Fenestrations or dehiscences were
Home | TOC |          
Index
not observed over the palatal surfaces of the
upper incisors. The retrusion of the upper inci-
sors, however, initiated root resorption.
Therapeutic risks can be observed with un-
controlled tipping tooth movements with con-
tinuous resilient arch wires and with a tooth
movement vertical to the alveolar process. Pro-
trusion, retraction, and intrusion of mandibular
or maxillary incisors were especially critical
tooth movements. Similarly, overloading of the
intraoral anchorage by using intermaxillary elas-
tice or cortical root torque may result in substan-
tial attachment loss.
A complete osteoblastic periodontal remodel-
ing of the orthodontically induced bone defects
was seen above the palatal root surfaces of the
upper premolars and molars. The lowest level of
osteoblastic periodontal remodeling was seen in
the lower frontal area above the buccal and lin-
gual root surfaces of the incisors and canines. In
some teeth, the dehiscences or fenestrations
were totally repaired, in others not. A spontane-
ous reorientation or relapse of single teeth in
the direction of the initial tooth position initi-
ated a complete periodontal remodeling.
Discussion
As the labiolingual diameter of the alveolar pro-
cess decreases and the eccentric positioning of
the tooth increases, the facial/lingual bone plate
above the root surfaces is reduced. The thera-
peutic risk inherent in the application of uncon-
trolled orthodontic force systems may lead to a
breakdown of the bone plate and the covering
soft tissue in terms of periodontal recession. The
bone can be only slightly increased by periosteal
apposition during orthodontic tooth movement.
On extending beyond the primary bone volume
of the alveolar process and the secondary, newly
formed bone apposition, the periodontal liga-
ment fuses with the apposing periosteum, and a
bone dehiscence develops. Additional factors,
such as traumatic lesions resulting from forceful
cleaning techniques or plaque-induced mar-
ginal gingivitis and periodontitis, may accelerate
the development of periodontal recessions.
CT scanning in orthodontically treated pa-
tients allows single-tooth interpretation of re-
modeling and hard-tissue lesions at the peri-
odontal tissue in the wake of various therapeutic
techniques.8 The periradicular osteodynamics
<<    
     Article
      >>
Periodontal Remodeling During Orthodontics
resulting from a therapeutically induced tooth
movement can be evaluated three-dimensionally
by comparing two or three different CT exami-
nations during or after treatment.
Various anatomic, morphologic, and thera-
peutic risk factors are intensified reciprocally
because the side effects of orthodontic therapy
increase with reduced attachment level. When
teeth with a critical ratio between the labiolin-
gual tooth diameter and the width of the alveo-
lar process are derotated, marginal bone dehis-
cences are induced if the bone apposition is
inadequate. In histologic posttreatment exami-
nations of a human specimen dehiscences and
fenestrations of various sizes were found.9'10 Ex-
tensive fenestrations resulting from tooth move-
ment in an atrophied alveolar process or the
dentoalveolar maxillary sinus were detected his-
tologically. Overall, the present CT findings con-
firm these histologic findings. Clearly, there is a
predisposition to attachment loss at these sites.
The degree of bone apposition by the superim-
posed periosteum is not enough to avoid bone
dehiscences in cases of a primarily narrow apical
base or a deep maxillary sinus and extensive
tooth movements. With critical initial findings,
tooth movements should be confined to the an-
atomic limits of the alveolar process. Selecting
controlled biomechanics can most readily re-
duce this anatomic and topographic risk factor.
The initial application of continuous resilient
arch wires for levelling resulted partly in tooth
tipping. Treatment with uncontrolled force sys-
tems corresponding to the widespread ideal
arch concept from straight-wire mechanics is
open to criticism with periodontal involution
because the risk of iatrogenic lesions increases
with decreasing attachment level. Mulie and Ten
Hoeve also demonstrated that bone dehiscences
occur with a narrow frontal alveolar process.11'12
With sagittal incisor movements in the maxilla,
low-grade fenestrations were found above the
facial root apices in cases of uncontrolled tip-
ping. Even after extensive palatal movement, a
continuous bone covering was revealed on the
marginal and apical aspects over the palatal root
surface in the CT scans. Wehrbein et al recorded
extensive osseous remodeling processes directly
above the root during histologic examination of
the periradicular bone of a human specimen
with retracted incisors.9 Despite thin palatal
Home | TOC |          
Index
27
bone apposition, there was no histologic evi-
dence of cortical perforation.
Marginal bone dehiscences were detected
above the buccal root surfaces of teeth in the
expansion center of a maxillary arch. Clinically,
no localised periodontal recessions were found
in these patients. The empirical concept of cor-
tical anchorage preparation has the inherent
risk of a massive attachment loss at the incisors
and at the molars because the roots are being
moved in the dense compact tissue.13 The
present CT findings revealed extensive bone de-
hiscences after cortical anchorage preparation.
In particular, orthodontic uprighting of the in-
cisors to cephalometric norms may be question-
able in cases of a narrow symphysis, thin bone
plates, or advanced alveolar bone loss.
Orthodontically induced bone dehiscences
were partly repaired by osteoblastic periodontal
remodeling or spontaneous reorientation of the
teeth in the retention period. Within the scope
of differential therapeutic assessment of tooth
movements, a reasonable risk-benefit calculation
is needed to assess the extent to which the initial
periodontal situation, the adaptability of the cov-
ering hard and soft tissue and the expected os-
teoblastic remodeling permit extensive orth-
odontic tooth movement.
References
1. Engelking G, Zachrisson BU. Effects of incisor reposi-
tioning on monkey periodontium after expansion
through the cortical plate. Amer J Orthodont 1983;82:
23-32.
2. Wainwright WM. Faciolingual tooth movement: Influ-
ence on the root and cortical plate. Amer J Orthodont
1973;64:278-289.
3. Thilander B, Nyman S, Rarring T, et al. Bone regenara-
tion in alveolar bone dehiscences related to orthodontic
tooth movements. Europ J Orthodont 1983;27:105-114.
4. Fuhrmann R, Bücker A, Diedrich P. Assessment of alve-
olar bone loss with high resolution computed tomogra-
phy. J Periodont Res 1995;30:258-263.
5. Graber TM. Comment of the editor-in-chief. Amer
J Orthod Dentofac Orthop 1995; 107: 360.
6. Fuhrmann R. Three-dimensional interpretation of
alveolar bone dehiscences. An anatomical-radiological
study. J Orofac Orthoped 1996;57:62-74.
7. Fuhrmann R. Three-dimensional interpretation of labio-
lingual bone width of the lower incisors. An anatomical-
radiological study. J Orofac Orthoped 1996;57:168-185.
8. Fuhrmann R. Three-dimensional interpretation of peri-
odontal lesions and remodeling during orthodontic
treatment. J Orofac Orthopedic 1996;57:224-237.
<<    
     Article
      >>
28 Robert A. W. Fuhrmann
9. Wehrbein H, Fuhrmann R, Diedrich P. Periodontal con-
ditions after facial root tipping and palatal root torque of
incisors in man. Amer J Orthodont Dentofac Orthoped
1994;106:455-462.
10. Wehrbein H, Fuhrmann R, Diedrich P. Human tissue re-
sponse after long-term orthodontic tooth movement.
Amer J Orthodont Dentofac Orthoped 1995;107:360-37l.
11. Mulie RM, Ten Hoeve A. The limitations of tooth move-
ment within the symphysis studied with laminography
Home | TOC |          
Index
and standardized occlusal films. J Clin Orthodont 1976;
10:882-899.
12. Ten Hoeve A, Mulie RM. The effect of antero-postero
incisor repositioning on the palatal cortex, as studied with
laminography. J Clin Orthodont 1976;10:804-822.
13. Wehrbein H, Bauer W, Diedrich P. Mandibular incisors,
alveolar bone, and symphysis after orthodontic treat-
ment - A retrospective study. Amer J Orthodont Dento-
fac Orthoped 1996;110:239-249.
<<    
     Article
      >> Home | TOC |          
Index

Preliminary Tests of a New Device to

Monitor Orthodontic Headgear Use
Elizabeth K. Lyons and Douglas S. Ramsay

The orthodontic community has had a long-standing interest in measuring

headgear use characteristics. Progress in microelectronics has made it pos-
sible to incorporate the electronics for a headgear-monitoring device en-
tirely within the body of an orthodontic headgear force module. A novel
headgear-monitoring device was developed recently and subjected to lab-
oratory and clinical testing. Results of these initial tests are described and
suggest that the headgear monitor should be able to measure the temporal
characteristics of headgear wear, estimate how much force is delivered,
detect patient attempts to falsify headgear use, and provide readily acces-
sible feedback to patients, parents, and orthodontists about headgear use.
(Semin Orthod 2002;8:29-34.) Copyright 2002, Elsevier Science (USA). All
rights reserved.

C linical research has shown that orthodontic

headgear treatment is effective for the cor-
rection of Angle Class II malocclusions.1'3 The
removable nature of a headgear places the
proper implementation of the orthodontist's
prescription for its use under the control of the
patient. Patients typically control when and for
how long headgear is worn, and depending on
the appliance, patients can manipulate the
amount of force the headgear delivers. A pa-
tient's willingness to wear a headgear is thought
to be important if treatment is to succeed. Tul-
loch and colleagues4 make this point clearly,
"The chance of improvement in an uncoopera-
tive child is the same as the chance of improve-
ment in an untreated one - small, but not zero."
In contrast, there is little known about how
much patient adherence to the prescribed head-
From the Departments of Orthodontics, Pediatric Dentistry, and
Psychology, University of Washington, Seattle, WA.
Supported by NIH grants KO4-DE00379 and R41-DE12430,
the University of Washington Orthodontic Alumni Association, the
Washington Technology Center, and the University of Washington 's
Royalty Research Fund.
Address correspondence to Douglas S. Ramsay, DMD, PhD,
MSD, Department of Pediatric Dentistry, University of Washington,
Box #357136, Seattle, WA 98195-7136.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0801-0006$35.00/0
doi: 10.1053/sodo. 2002.28170
gear regimen is needed to produce the desired
treatment outcome.5'6 Indeed, there is little re-
search in any health care field that describes
how partial patient adherence relates to clinical
outcome.7 Assuming that patient adherence to a
headgear regimen is a mediator of clinical out-
come then poor adherence may lead to in-
creased treatment time, increased cost, and fail-
ure to correct the malocclusion, which in turn,
may necessitate a less preferred alternative treat-
ment. Being able to identify poorly compliant
patients would allow timely treatment plan mod-
ifications or the implementation of strategies to
enhance patient adherence.6'7 In addition, the
ability to measure important characteristics of
headgear use would make it possible to evaluate
the effect of partial adherence on clinical out-
come. Unfortunately, it is usually difficult to
know how well an individual is adhering to a
treatment regimen.5
Orthodontists have long appreciated the po-
tential value of measuring the time a headgear
has been worn and there have been numerous
attempts to make a device that could provide
this information. Northcutt introduced the first
commercially available timing headgear in
1974.8 Unfortunately, this timing headgear was
both inaccurate9 and susceptible to patient at-
tempts to falsify headgear wear. Continued in-
terest in measuring headgear use lead to the

Seminars in Orthodontics, Vol 8, No 1 (March), 2002: pp 29-34 29

<<    
     Article
      >>
30 Lyons and Ramsay
publication of methods for fabricating headgear
timers.10"12 These timers were bulky and often
necessitated placing the electronics in the neck
strap, which could diminish patient comfort.
Some designs removed the safety disconnect fea-
ture that should be a part of all headgear force
modules. Most designs offered little protection
against patient attempts to falsify headgear use
data and provided only a cumulative measure of
wear time. Finally, these devices provided the
orthodontist with intermittent headgear wear in-
formation, and none gave the patients easily
accessible feedback about their degree of adher-
ence to the headgear wear regimen.5'6
Newer devices to measure headgear use have
been developed. The Ortho Kinetics Corpora-
tion (Vista, CA) has made a commercially avail-
able headgear timer. An optical sensor, which
can detect when the spring in the module is
compressed, activates the timer in this headgear
force module. This device can record each pe-
riod of headgear use, and the record of head-
gear wear can be reviewed during visits at the
orthodontic office. Unfortunately, it is possible
for patients to activate the timer without actually
wearing the headgear. To create a long-term
record of falsified wear data, however, a patient
would need to adhere to this pattern of decep-
tive behavior over a long period (eg, remember
to compress the spring before bed and allow the
spring to relax on awakening). Scientists at the
University of Florida have developed a micropro-
cessor-controlled headgear-timing device that
can record the amount of spring compression
that occurs during headgear use.13 This design
has the advantage of measuring the dynamic
spring compression that occurs during headgear
wear and so can provide information about
headgear use and the amount of force produced
by the spring. The device has limited clinical
utility because it is somewhat fragile and its de-
sign requires connecting the bulky electronics,
located in the neckstrap, to the headgear force
module.
A novel headgear monitor has been devel-
oped at the University of Washington (US Patent
#5,980,026); the rationale underlying its design
is described elsewhere.5'6 In brief, the guiding
design principles were to make the device small,
durable, comfortable, and to maintain the safety
breakaway feature so that it could be used clin-
ically. The device was designed to 1) measure
Home | TOC |          
Index
the temporal characteristics of headgear wear;
2) estimate how much force is delivered by the
headgear; 3) detect patient attempts to falsify
headgear use; and 4) provide readily accessible
feedback to the patients, parents, and orthodon-
tists about headgear use. Initial prototypes were
fabricated and evaluated in laboratory and clin-
ical tests.
The electronics and battery of the prototype
monitor are contained entirely within the head-
gear force module. The monitor uses a micro-
processor with nonvolatile memory, magnet sen-
sors, an infrared light-emitting diode (LED) and
phototransistor for bi-directional communica-
tion and a 4-digit liquid crystal display (LCD)
screen to display output. A lithium battery pow-
ers the monitor. A small cylindrical rare-earth
magnet is embedded in the end of the adjust-
ment strap. The basic unit of measurement is
the position of the magnet within the body of
the force module, which changes as the strap is
extended and the internal spring is compressed.
By measuring the position of the adjustment
strap within the force module, it is possible to
determine when the headgear is in use. In addi-
tion, the force resulting from a known amount
of spring compression can be calculated from
calibration data stored in a look-up table in the
monitor's memory.
Laboratory Testing
Three laboratory tests are described. The first
test measured the accuracy of the clock crystal
that determines when headgear wear occurred.
The second test determined whether the moni-
tor could accurately measure the amount of
spring compression. The third test evaluated
whether the amount of force generated by the
spring during its compression could be esti-
mated by knowing the amount of spring com-
pression.
Clock Accuracy
The microprocessor in the headgear monitor is
equipped with a standard 32,768-Hz crystal os-
cillator internal clock. This clock is used to reg-
ister the headgear use data in real time. To test
the accuracy of the timing unit, the clock was
synchronized with the National Bureau of Stan-
dards' clock through internet access. The head-
<<    
     Article
      >>
Device to Monitor Orthodontic Headgear Use
gear monitors were activated and left running
for 72 hours. After 72 hours, the internal clock
on the microprocessor was compared to the Na-
tional Bureau of Standards' clock. There was
approximately a 2-second delay in the internal
clock out of the running time of 259,200 sec-
onds, indicating 99.999% accuracy for the inter-
nal clock timer.
Measuring Adjustment Strap Position and
Spring Compression
Pulling the headgear adjustment strap through
the force module causes the spring to compress.
The amount of spring compression can be de-
termined by knowing the position of the adjust-
ment strap within the force module. This was
calibrated with the output from the magnet sen-
sors in the headgear monitor by using a MTS
machine (Material Testing System model Sin-
tech 2/S, MTS Systems Corporation, Triangle
Park, NC). The MTS machine allowed the body
of the force module to be mounted to a fixed
lower unit while the adjustment strap was con-
nected to a moving unit attached to a load cell.
The MTS was programmed to provide various
known amounts of adjustment strap extension.
The extension values can be programmed to run
in different increments to a specified extension
distance and to have varying holding times at
any position. The MTS was programmed to
move the adjustment strap at 1-mm increments
through the range of movement. At each 1-mm
increment, the adjustment strap was held in po-
sition, and the output from both magnet sensors
was recorded. The location of the adjustment
strap and the output from both sensors were
entered into a look-up table in the memory of
the headgear monitor. An algorithm was used to
calculate the position of the adjustment strap
based on the output from the two magnet sen-
sors. The accuracy of this method for determin-
ing the position of the adjustment strap was
evaluated. The MTS system was programmed to
move the adjustment strap through a series of
known positions, and the estimated position of
the strap was calculated by using the output
from the sensors. Figure 1 illustrates that there
was excellent agreement between the known po-
sition of the strap as determined by the MTS
machine and the estimated position of the strap
Home | TOC |          
Index
31
25 •
20 -
Estimated
Magnet
Position
(mm)
5 •
0 •
100
80 Differential
60 Hall Output
40 (A-D Values)
20
0 5 10 15 20 25 0 5 101520 25
Known Magnet Position (mm)
Figure 1. A magnet embedded in the end of the
headgear adjustment strap moves through the body of
the headgear force module as the MTS machine pulls
the strap. By using Hall sensors to measure magnetic
field strength, it was possible to obtain excellent res-
olution in estimating magnet position relative to the
known magnet position.
as determined by comparing the sensor output
with the calibration data in the look-up table.
Estimating Force from Knowing the
Adjustment Strap Position
Orthodontists are interested in knowing the
amount of force generated from the spring in
the headgear force module. This force is often
assessed clinically by using a force gauge when
the headgear is being adjusted to a patient.
Therefore, it was important to ascertain whether
the amount of force from the spring could be
estimated by knowing the amount of strap ex-
tension. For this method to work there must be
a direct relationship between headgear strap po-
sition and force. The load cell of the MTS ma-
chine was also able to measure the tensile force
at any given strap extension. Therefore, it was
possible to describe the relationship between
adjustment strap position and force. The MTS
extension program was run with each headgear
module extended to a distance of 20 mm and
returned to its initial position at both slow and
rapid speeds. The force needed to extend the
headgear strap was recorded over the extension
and return range. As shown in Figure 2, the
relationship was, for the most part, linear, al-
though there was some hysteresis. Whereas it is
possible to obtain a reasonable approximation
of the force delivered at a given amount of strap
extension, a goal of future headgear monitor
development is to reduce this hysteresis to im-
prove the accuracy of force estimates. Interest-
<<    
     Article
      >>
32 Lyons and Ramsay
1400
1200
1000 MTS pulling t he
adjustment strap
800
Force
600
400 Spring pushing the
adjustment strap
200
0 5 10 15 20 25
Known Magnet Position (mm)
Figure 2. Using the MTS machine, representative
data collected from a single module demonstrate that
there is a good, but not perfect, relationship between
the position of the adjustment strap and force.
ingly, several commercially available headgear
force modules were also evaluated, and a large
amount of hysteresis was observed in some
models.
Clinical Testing
Two clinical tests are described. The first test
measured the effect of head posture on spring
compression. The second test described how
spring compression changes during headgear
wear.
Effect of Head Posture on Spring
Compression
Research from the University of Florida found
that headgear force varies as a function of head
position.13 Because the headgear monitors at
the University Florida and the University of
Washington both measure an index of spring
compression, it should be possible to replicate
these findings using our different headgear de-
sign. To evaluate the effect of head position on
spring compression, a subject wore a monitor on
each side of the head. The subject then turned
her head through a range of motion, looking
left, right, upward, and downward. Each head
position was held for 30 seconds. The results are
shown in Figure 3. This graph represents an
average of two cycles of head movements with a
Home | TOC |          
Index
collection of 14 data points at each head posi-
tion. Spring compression in the left and right
monitors was inversely related during lateral
head movements. As the head turned to the
right, the right monitor's spring relaxed while
the left monitor's spring was compressed. The
opposite was true when turning the head to the
left. In contrast, a positive relationship in spring
compression was observed for the right and left
monitors when the head moves up or down.
Both right and left springs are compressed when
looking upward and both are relaxed when look-
ing downward. This independent replication of
the relationship between head posture and force
suggests that these two different headgear mon-
itor designs are measuring the same phenome-
non.
Headgear Wear Data Collected During
Patient Use
For clinical data collection, two headgear mon-
itors were worn with a cervical headgear appli-
ance, and data were collected from both the
subject's left and right sides simultaneously. The
subject wore the monitoring headgear for four
episodes during a 24-hour period and carefully
recorded the times the headgear was worn and
*\
Maximum
Spring 25 -i
\
Compresson W
Estimated
Magnet 15 -
20 -
\N
u
^
Position "ffitfftj
JW.
(mm)
10 - \r l* ^
t
5 • gte &5tt*
Passive
Head Position
Left Side Monitor Right Side Monitor
Figure 3. Two headgear monitors were used with a
cervical headgear appliance to collect data from both
the subject's left and right sides simultaneously. This
subject was asked to move her head into different
positions and then to hold still in that position for
approximately 30 seconds while magnet sensor data
were collected. A look-up table algorithm was used to
convert the magnet sensor output into an estimate of
magnet position and thus a measure of spring com-
pression.
<<    
     Article
      >>
Device to Monitor Orthodontic Headgear Use
removed. Output from both magnet sensors
were collected at a rate of four samples per
minute, and these data were converted to an
estimate of adjustment strap position by using
the look-up table. The four episodes of headgear
use were detected accurately by the monitor.
Figure 4 illustrates the position of the adjust-
ment strap for both monitors during an eight-
hour period when the subject was asleep. Al-
though the subject was sleeping, there was
considerable variation in the degree of spring
compression over time, which is indicative of
head movement. These data also support the
observation that the spring compression of the
right and left sides are frequently related (eg, as
one side compresses, the other relaxes). As
could be expected, data collected while the sub-
ject was awake exhibited greater variability than
data collected during sleep.
When the headgear is not being worn, there
is very little variability in the position of the
adjustment strap over time. Similarly, when the
monitors were placed on a mannequin head to
simulate (falsify) headgear use, the data showed
an absence of variability in the movement of the
adjustment strap. Thus, actual wear can be dis-
tinguished from static attempts to falsify head-
gear use by measuring the variability in the po-
sition of the adjustment strap over time. As
shown in Figure 4, there is considerable variabil-
ity in the data, although there was a 45-minute
period between hours 6 and 7 when there was
little variability in the position of the adjustment
strap even though it was being worn. Research
20-
Estimated 15 -
Magnet
Position for 10 -
Left Monitor
(mm)
20 -
Estimated 15 -
Magnet
Position for ,
Right Monitor 10 "
(mm)
Time (hours)
Figure 4. Two headgear monitors were worn with a
cervical headgear appliance to collect data from both
the subject's left and right sides simultaneously. Al-
though the subject was asleep, there was considerable
variation in the degree of spring compression over
time which is indicative of head movement.
Home | TOC |          
Index
33
from the University of Florida has suggested that
an absence of movement over an hour would be
a good criterion to indicate no wear.14
One of our design goals5 was for the monitor
"to provide continuous, accurate, and easily in-
terpretable visual feedback to patients, parents,
and clinicians about headgear use." This is an-
other unique aspect of our design because no
headgear timers have provided feedback directly
to patients in an effort to enhance adherence via
improved self-regulation.6 The LCD on the mon-
itor can be programmed to show the average
daily wear time and/or the cumulative number
of hours of wear since the last visit. This feature
is convenient for the orthodontist to assess the
patient's progress, and it provides the patient
with ongoing feedback to regulate his or her
own behavior. Self-monitoring has been sug-
gested as a useful strategy to enhance patient
adherence.5'7
Conclusion
These initial laboratory and clinical tests were
conducted to validate the functionality of this
novel approach to monitoring headgear wear.
Based on the encouraging findings of these tests,
an improved version of the headgear monitor is
being developed, and these monitors will un-
dergo extensive evaluation in private orthodon-
tic offices. The development of a sophisticated
headgear monitor will be valuable for both sci-
entific and clinical orthodontic applications.
Sufficient flexibility has been designed into this
headgear monitor so that it can meet the needs
of scientists as well as clinicians.
Acknowledgment
We would like to acknowledge the contributions of Chris
Prall, Mani Soma, and Susan Herring to this article.
References
1. GhafariJ, King GJ, TullochJF. Early treatment of Class
II, division 1 malocclusion—Comparison of alternative
treatment modalities. Clin Orthod Res 1998;1:107-117.
2. Keeling SD, Wheeler TT, King GJ, et al. Anteroposterior
skeletal and dental changes after early Class II treatment
with bionators and headgear. Am J Orthod Dentofacial
Orthop 1998; 113:40-50.
3. TullochJF, Phillips C, Koch G, et al. The effect of early
intervention on skeletal pattern in Class II malocclusion:
<<    
     Article
      >>
34 Lyons and Ramsay
A randomized clinical trial. Am J Orthod Dentofacial
Orthop 1997;111:391-400.
4. Tulloch JF, Proffit WR, Phillips C. Influences on the
outcome of early treatment for Class II malocclusion.
Am J Orthod Dentofacial Orthop 1997;111:533-542.
5. Ramsay DS, Soma M, Sarason IG. Enhancing patient
adherence: The role of technology and its application to
orthodontics, in McNamara JA Jr, Trotman CA (eds):
Creating the Compliant Patient. Craniofacial Growth
Series, Center for Human Growth and Development,
University of Michigan, Ann Arbor, MI, 1997;33:141-165.
6. Lyons EK, Ramsay DS. A self-regulation model of patient
compliance in orthodontics: Implications for the design
of a headgear monitor. Semin Orthod 2000;6:224-230.
7. Dunbar-Jacob J, Schlenk E. Patient adherence to treat-
ment regimen, in Baum A, Revenson TA, Singer JE
(eds): Handbook of Health Psychology. Mahwah, NJ:
Lawrence Erlbaum Associates, 2001, pp 571-580.
Home | TOC |          
Index
8. Northcutt ME. The timing headgear. J Clin Orthod
1974;8:321-324.
9. Banks PA, Read MJF. An investigation into the reliability
of the timing headgear. BrJ Orthod 1987; 14:263-267.
10. Cureton SL, Regennitter FJ, Orbell MG. An accurate,
inexpensive headgear timer. J Clin Orthod 1991;25:749-
754.
11. Guray E, Orhan M. Selcuk type headgear-timer (STHT).
Am J Orthod Dentofacial Orthop 1997;111:87-92.
12. Kyriacou PA, Jones DP. Compliance monitor for use
with removable orthodontic headgear appliances. Med
Biol Eng Comput 1997;35:57-60.
13. Johnson PD, Bar-Zion Y, Taylor M, et al. Effects of head
posture on headgear force application. J Clin Orthod
1999;33:94-97.
14. McGorray SP, Dean GS, Keeling SD, et al. Monitoring
orthodontic headgear wear with a micro-electronic data
acquisition system. J Dent Res 1997;76:127 (abstr 909).
<<    
     Article
      >> Home | TOC |          
Index

A Comparison of Skeletal and Dental Changes

Between Rigid and Wire Fixation for Bilateral
Sagittal Split Osteotomy
Calogero Dolce, John P. Hatch, Joseph E. Van Sickels, Robert A. Bays, and
John D. Rugh

For two years, this multisite prospective clinical trial examined longitudinal
skeletal and dental changes after bilateral sagittal split osteotomy for man-
dibular advancement in which either rigid or wire fixation was used. Sub-
jects in the rigid fixation group (n = 78) received 2-mm bicortical position
screws, while the subjects in the wire fixation group (n = 49) received
inferior border wires. Skeletal and dental changes were measured from
cephalometric films taken immediately before surgery, one week after sur-
gery, and at eight weeks, six months, one year, and two years after surgery.
In both groups, the overbite and overjet increase with time, but were not
different from each other. The B-point in the wire group progressively
moved posteriorly, and at two years, it had relapsed 28%. In the rigid
fixation group, there was a transient anterior movement of the B-point
during the first six months and by two years after surgery, the B-point was
unchanged from immediate post surgery. Dental changes occurred in both
groups. These changes, however, were not able to accommodate the skel-
etal changes, resulting in similar increases in both overbite and overjet in
both groups of patients. These results have implications for the orthodon-
tists in management of the postmandibular advancement occlusion. (Semin
Orthod 2002;8:35-42.) Copyright 2002, Elsevier Science (USA). All rights
reserved.

Introduction Since its introduction, it has been modified sev-

T he bilateral sagittal split osteotomy (BSSO),
introduced by Trauner and Oberwise in
1957,1 is the most commonly performed orthog-
nathic surgical procedure in the United States.
From the Department of Orthodontics, University of Florida,
Gainesville, FL; Departments of Psychiatry and Orthodontics, The
University of Texas Health Science Center at San Antonio, San
Antonio, TX; Division of Oral and Maxillofacial Surgery, College of
Dentistry, University of Kentucky, Lexington, KY; Division of Oral
and Maxillofacial Surgery, and the Department of Surgery, Emory
University, Atlanta, GA.
Supported by NIH/NICDR grants DE09630 and 07283.
Address correspondence to Calogero Dolce, DDS, PhD, Depart-
ment of Orthodontics, Box 100444, JHMHC, Gainesville, Florida
32610-0444.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0801-0007$35.00/0
doi:10.1053/sodo.2002.28171
eral times and can now be tailored to correct
either a severe Class II or Class III malocclusion.
Despite the success and popularity of the BSSO,
controversies remain.
One of these areas of controversy is the rela-
tive merit of rigid versus wire fixation with re-
spect to short- and long-term dental and skeletal
changes. Once the osteotomy is performed and
the mandible is moved to its new position, there
are two common methods of fixating together
the proximal and distal bony segments. One
method is to use wire fixation in which inferior
border wires, circumferential wires, and supe-
rior border wires2 are used to approximate the
segments. This technique typically requires 6 to
8 weeks of jaw immobilization. A second method
is to use rigid fixation3 in which either plates,
lag, or position screws fixate and stabilize the

Seminars in Orthodontics, Vol 8, No 1 (March), 2002: pp 35-42 35

<<    
     Article
      >>
36 Dolce et al
bony segments. With rigid fixation, jaw immobi-
lization is eliminated, allowing jaw function
within a few days after surgery.
Many studies have compared both the short-
and long-term stability of rigid and wire fixation.
Short-term relapse has been demonstrated with
wire2'4'5 and with rigid fixation.6'9 Long-term sag-
ittal relapse with wire fixation ranges from 23%
to 45 %,10'12, whereas with rigid fixation it ranges
from 8% to 18%.6' 8 ' 1S All the studies cited above
are retrospective. It is not surprising, therefore,
that the practitioner is confused because con-
clusions based on case reports, case series, or
retrospective designs are often in conflict re-
garding the most basic outcomes of routine pro-
cedures.14
In 1990, the National Institutes of Health
funded a large-scale, multicenter, randomized
clinical trial comparing rigid fixation with wire
fixation. The results of this trial have now clari-
fied many of these controversies. In an earlier
publication,15 we compared relapse in patients
receiving rigid or wire fixation, two years after
surgery, by using the Johnston analysis. We
found that skeletal changes were essentially sta-
ble in the rigid fixation group, whereas the wire
fixation group exhibited a 26% average mandib-
ular symphysis relapse. Both groups experienced
dental changes. In another study,16 we examined
skeletal changes between the two groups from a
cranial base reference. At two years after surgery,
the mandibular symphysis position was un-
changed in the rigid group, but relapsed 30% in
the wire group. The purpose of the present in-
vestigation was to compare the skeletal and den-
tal changes between rigid and wire fixation for
mandibular advancement by using conventional
cephalometrics.
Patients and Methods
This study was a prospective, randomized clini-
cal trial that began in 1990 and was conducted at
The University of Texas Health Science Center
at San Antonio, Texas, Emory University in At-
lanta, Georgia, and the University of Florida in
Gainesville, Florida. Details of this study have
been published elsewhere.15' 16 In brief, subjects
who satisfied all inclusion and exclusion criteria
read and signed an informed consent form. Sub-
jects were stratified on degree of neuroticism,
presurgical over]et (<7 mm or >7 mm), and
Home | TOC |          
Index
institution. Thirty days before the scheduled sur-
gery date, notification of protocol assignment
(either screw or wire) was made by the project
statistician.
The subjects were evaluated 7 to 14 days be-
fore surgery (T2), within 1 week after surgery
(T3), and at 8 weeks (T4), 6 months (T5), 1 year
(T6), and 2 years (T7) after surgery. Some sub-
jects entered the study before the start of orth-
odontic treatment (Tl); however, most were en-
rolled after the start of orthodontic treatment,
largely because of the timing of the orthodon-
tist's referral.
The soft tissue dissection for the sagittal split
was completed according to Bell et al.17 Splitting
of the mandible was achieved by a horizontal
corticotomy on the lingual surface of the ramus
above the mandibular foramen and a vertical
corticotomy on the buccal surface near the sec-
ond molar. A connecting corticotomy was ex-
tended between the previous two. Splitting was
accomplished by using small osteotomes and
gently prying. This was performed bilaterally.
Suspension was achieved by using both circum-
ferential mandibular and intraosseous maxillary
24-gauge wires. Before application of fixation,
an acrylic splint was wired to the maxillary teeth,
and the mandible was placed in its predeter-
mined occlusion and stabilized with maxillo-
mandibular fixation (MMF). Patients random-
ized to the rigid fixation group received three
2-mm bicortical position screws bilaterally and
used elastics for 5 to 7 days. Patients randomized
to wire fixation received bilateral inferior border
wires and were placed in skeletal MMF for 6
weeks. All patients received postoperative phys-
iotherapy instruction after an established exer-
cise protocol. During surgery, the surgeon re-
tained the right to change the method of
fixation if the situation demanded a change
from the assigned study protocol. All protocol
violations were followed-up and described.
Lateral cephalometric radiographs were
obtained using a Wehmer calibration ruler
mounted on the cephalostat. Radiographs ob-
tained immediately after surgery (T3) were the
only radiographs taken with an interocclusal
acrylic splint in place. The clearest radiograph
between T2-T4 was selected, and the details of
the cranial base structures were drawn. An X-Y
cranial base coordinate system was constructed
through Sella with the X-axis drawn parallel to
<<    
     Article
      >>
Skeletal and Dental Changes After a BSSO
Frankfort Horizontal. The Y-axis was drawn pass-
ing through Sella, perpendicular to the X-axis.
In addition, templates were constructed for the
maxilla, mandible, mandibular symphysis, and
the maxillary and mandibular incisors. Details
on template construction for the cranial base,
the maxilla, and the mandibular tooth-bearing
segment have been previously described.18 Reg-
istration crosses were drawn on the templates.
The dental and skeletal templates were oriented
independently on each radiograph by best-fit
visual approximation, and the registration
crosses of each structure transferred to a sepa-
rate acetate sheet for each film. The X and Y
coordinates for the landmarks were registered
by digitization on a Hipad TM 1200 series Digi-
tizer (Houston Instruments, Houston, TX), in-
terfaced with a 386 computer system (KD Tech-
nologies, Gainesville, FL) using the software
program Ceph-Master (Trilobyte Software, Ann
Arbor, MI). The skeletal and dental landmarks
are shown in Figure 1. All tracings were digitized
twice, and means of the two measurements were
used.
Linear and angular measurements were used
to compare skeletal and dental changes over the
following time intervals: surgical change (T3 mi-
nus T2), eight-week change (T4-T3), six-month
change (T5-T3), one-year change (T6-T3), and
two-years change (T7-T3). The analysis of errors
Figure 1. Measuring points and reference lines used
in the cephalometric analysis.
Home | TOC |          
Index
37
was modeled after that used by Houston19 and
has been reported.18 Between group differences
at each time point were examined using un-
paired t tests; within group differences con-
trasted the change at surgery (T3-T2) with
change at each of the succeeding intervals using
paired t tests. All statistical tests were performed
at a = 0.05.
Results
There were 127 subjects who received a BSSO.
Of these, 29 subjects were treated at Emory Uni-
versity, 22 at The University of Florida, and 77 at
The University of Texas Health Science Center
at San Antonio. Our earlier reports followed
intent-to-treat protocol. In this report we analyze
relapse based on treatment received. Among the
study subjects, 49 received wire fixation, whereas
78 received rigid fixation. Differences between
fixation groups with respect to surgical changes
are shown in Table 1. Both fixation groups
showed similar surgical change in all the mea-
sured variables.
The surgical change (T3-T2) in overbite and
overjet was similar in both groups (Table 1).
Changes in overbite and overjet represent stabil-
ity of the occlusion. Immediately after surgery
(T3) (Table 2), the overbite was greater in the
wire group than in the rigid group (1.11 mm v
0.59 mm, respectively; P = .049). This may be
caused by different splint thickness. Thereafter,
the overbite increased with time in both groups
(P < .001); however, there was no difference
between the groups at all time points (Table 2).
One week after surgery (T3), the overjet was
2.61 mm in the rigid group and 2.36 mm in the
wire group (Table 3). Although there was not a
significant difference between the two groups,
the overjet increased with time in both groups.
Sagittal and vertical changes in the position of
the maxillary and mandibular incisor as well as
changes in the B-point were explored to eluci-
date the dental and skeletal contribution to the
changes in overbite and overjet. Skeletal
changes were assessed by measuring sagittal and
vertical changes in the position of B-point. The
surgical procedure (T3-T2) produced a mean
anterior sagittal movement of the B-point of 5.06
mm in the rigid fixation group and 5.38 mm in
the wire fixation group. Analysis of variance
(ANOVA) showed postsurgical changes oc-
<<    
     Article
      >> Home | TOC |          
Index

38 Dolce et al

Table 1. Equivalency of Skeletal and Dental Surgical Changes

Rigid (n = 78) Wire (n = 49) P (R v W)
B-point sagittal (mm) 5.06 (2.74) 5.38 (2.95) NS
B-point vertical (mm) 3.52 (3.05) 3.04 (2.32) NS
Mandibular incisor sagittal (mm) 4.72 (2.56) 4.93 (2.68) NS
Mandibular incisor vertical (mm) 3.28 (2.96) 2.86 (2.05) NS
Maxillary incisor sagittal (mm) -0.15 (0.91) -0.35 (0.65) NS
Maxillary incisor vertical (mm) 0.18 (0.72) 0.08 (0.58) NS
Maxillary incisor— SN (degrees) -0.44 (2.37) -0.86 (2.17) NS
Mandibular incisor— NB (degrees) -1.56 (3.55) 2.53 (2.97) NS
Overbite (mm) 3.11 (2.93) 2.78 (2.16) NS
Overjet (mm) 4.88 (2.72) 5.28 (2.74) NS

NOTE. ( — ) represents posterior movement in the horizontal plane, superior movement in the vertical plane, or retroclination
for angular measurement.
Abbreviations: NS, not significant; SN, Sella turcica-Nasion, Nasion; NB, Nasion B-point.
* Values expressed as mean (standard deviation).

curred both between and within groups. As
shown in Table 4, the wire fixation subjects ex-
perienced sagittal relapse as early as eight weeks
after surgery (T4). At this time point, B-point
had moved posteriorly 0.95 mm or 18% (Figure
2). This horizontal relapse continued, albeit not
at the same rate, and at two years after surgery
(T7), the B-point had moved 1.55 mm posteri-
orly. This was equivalent to a 29% sagittal re-
lapse. The rigid fixation subjects experienced a
transient anterior movement in B-point (Table 4
and Figure 2). The peak of this anterior move-
ment (9%) occurred at six months after surgery
(T5). At two years after surgery (T7), B-point was
unchanged from its immediate postsurgical po-
sition. The BSSO (T3-T2) produced an increase
in the vertical dimension in both wire and rigid
fixation groups (3.52 mm and 3.04 mm, respec-
tively). The vertical changes in the B-point are
shown in Table 5. Both groups experienced a
loss of vertical dimension (the superior move-
ment of the B-point) (P< .01,ANOVA) between
T3 and T7. Although the values of the rigid
group were higher than the wire group, the
values were not statistically significant between
the two groups.
Figures 3 and 4 depict the incisal edge posi-
tions, in both sagittal and vertical planes, of the
mandibular and maxillary incisor, respectively.
In general, the position of the mandibular in-
cisal edges (Figure 3) for both groups mimics
that of the B-point, although not at the same
rate. Unlike the B-point, there was no difference
between groups in the horizontal position of the
mandibular incisor (P < .07, ANOVA). There
were within group differences. At the eight-week
period (T4), the sagittal position of the incisal
edge in the wire group moved only half as much
as the B-point (0.34 mm). Two years after sur-
gery (T7), however, the position of the mandib-
ular incisal edge had moved posteriorly (1.37
mm), nearly the same amount as point B-point
(1.55 mm). This sagittal change was significantly
different (P < .01) than after surgery (T3) at all
time points, except at the eight-week period.
Also, in the rigid fixation group, the incisal edge
initially moved anteriorly, and by two years after
surgery (T7), its sagittal position placed it 0.5
mm more posteriorly than B-point. In the rigid
group, the sagittal position was significantly dif-
ferent from after surgery at the eight-week and
two-year time points.
The vertical position of the mandibular inci-
sor also followed the position of its respective
B-point. There were no between groups differ-
ences (P = .66). Within group differences (P <

Table 2. Measured Overbite

Group Postsurgery (T3) 8 wks (T4) 6 mos (T5) 1 yr (T6) 2 yrs (T7)
Rigid 0.59 (1.4) 1.76* (1.4) 2.20* (1.3) 2.71* (1.4) 2.92* (1.4)
Wire 1.11+ (1.4) 1.37 (1.4) 2.26* (1.3) 2.55* (1.4) 2.87* (1.4)
NOTE. All values expressed as mean (standard deviation).
*Signincantly different within group from postsurgery (T3) (P < .01).
t
Significantly different between groups (P = .049).
<<    
     Article
      >> Home | TOC |          
Index

Skeletal and Dental Changes After a BSSO 39

Table 3. Measured Overjet

Group Postsurgery (T3) 8 wks (T4) 6 mos (T5) 1 yr (T6) 2 yrs (T7)
Rigid 2.61 (1.1) 2.45 (1.4) 2.94 (1.4) 3.15* (1.2) 3.39* (0.9)
Wire 2.36 (1.1) 2.43 (1.4) 2.80 (1.4) 3.28* (1.2) 3.45* (0.9)
NOTE. All values expressed as mean (standard deviation) in millimeters.
* Significantly different within group from postsurgery (T3) (P < .01).

.001) from immediately after surgery were present Discussion

in both groups, except at the eight-week time
The data presented here are part of a larger
point in the wire group. The angulation of the
study that, in addition to the skeletal and dental
mandibular incisal was measured in relation to
changes, examined differences in neurosensory
the Nasion-B-point line (Ll-NB). The mandibu-
deficit, signs and symptoms of temporomandib-
lar incisors retroclined with time in both groups.
ular disorders, masticatory performance, quality
This mandibular incisor retroclination was
of life, patient satisfaction, and cost-effectiveness
greater in the rigid group (Table 6).
of the two fixation techniques. This analysis in-
The changes in the sagittal and vertical posi-
vestigated the short- and long-term changes in
tion of the maxillary incisor are depicted in
overbite and overjet. Skeletal (B-point) and den-
Figure 4. As was the case for the mandibular
tal (maxillary and mandibular incisors) alter-
incisor, the maxillary incisors behave differently
ations were then examined to determine their
for the two groups. In both rigid and wire fixa-
contribution to the resultant overbite and over-
tion groups, the maxillary incisor moved occlu-
jet. In both groups, we found similar increases in
sally with no difference between the groups. At
overbite and overjet overtime. The increase in
the two-year time point, however, within group
overjet in the wire group was attributed to skel-
differences existed in both groups (P < .002).
etal relapse, whereas in the rigid group dental
Sagittally, the trend was for the maxillary incisors
changes contributed most to the increase in
of the two groups to move in opposite directions
overjet.
(P = .067). For the first six months, the maxil-
This was a prospective randomized clinical
lary incisor in the rigid fixation group followed
trial, in which patients were randomized to re-
the mandibular incisor and moved anteriorly
ceive rigid or wire fixation. In a previous report,
(0.4 mm). Thereafter, it moved posteriorly and
we showed that the subjects were balanced be-
finally settled 0.2 mm anteriorly. Unlike the
fore surgery.15 For various reasons, there were 14
mandibular incisors of the wire group, the max-
protocol violations. Consequently, there were
illary incisors in this group moved posteriorly 0.3
more subjects receiving rigid fixation than wire
mm at two years after surgery. The angulation of
fixation. The data, presented in this paper, are
the maxillary incisor was not significantly differ-
ent between the two groups (Table 7). However,
within group differences existed in the rigid 10 7
T6
T5
group. T4
T3 T7

post-sctrg 8 weeks 6 months 1 year 2 years

Table 4. Postsurgical Sagittal Displacement
of B-Point f -10
P
Time Interval Rigid Wire (Rv W) -20 -
8 wks (T4-T3) 0.36 (1.53) -0.95* (1.50) 0.008
6 mos (T5-T3) 0.40 (1.87) -1.07* (1.80) 0.007 T7
1 yr (T6-T3) 0.22 (2.11) -1.29* (1.78) 0.006 -30 -I
2 yrs (T7-T3) -0.02 (2.61) -1.55* (2.00) 0.007
Time
NOTE. (-) represents posterior displacement. ( + ) repre-
sents anterior displacement. All values expressed as mean Figure 2. Percentage change from surgical move-
(standard deviation) in millimeters. ment in the horizontal position of B-point. T3 =
* Significantly different within group from postsurgery (T3) immediate post-surgery; T4 = 8 weeks; T5 = 6
(P < .01). months; T6 = 1 year; and T7 = 2 years.
<<    
     Article
      >> Home | TOC |          
Index

40 Dolce et al

Table 5. Postsurgical Vertical Displacement

of B-Point Rigid
Wire
P
Time Interval Rigid Wire (R v W)
8 wks (T4-T3) -1.09* (1.97) -0.63* (1.75) NS
6 mos (T5-T3) -1.53* (2.06) -0.92* (1.42) NS
1 yr (T6-T3) -1.88* (2.93) -0.84* (1.93) NS
2 yrs (T7-T3) -1.71* (2.29) -0.97* (2.19) NS
NOTE. ( — ) represents superior displacement. ( + ) repre-
sents anterior displacement. All values expressed as mean
(standard deviation).
^Significantly different within group from postsurgery (T3)
(P< .01).
Horizontal position (mm)

on the treatment received. When the data was
analyzed as intent-to-treat, similar conclusions
were drawn.
The skeletal and dental surgical movements
were similar in both groups (Table 1). Skeletal
stability of the mandible was assessed by follow-
ing changes in B-point in both the sagittal and
vertical position. The immediate surgical change
resulted in B-point moving inferiorly 3.5 mm for
the rigid group and 3.0 mm for the wire group.
After surgery, in both groups, B-point moved
superiorly to approach the presurgical vertical
dimension. This was expected in both groups.
The rigid group had the splint removed at two
weeks, whereas the wire group had MMF re-
Figure 4. Horizontal and vertical position of the max-
illary incisor. The area delineated around the time
points includes the 95% CI in both the sagittal and
horizontal direction. T3 (immediate postsurgery was
set at zero for both groups) = immediate postsurgery;
T4 = 8 weeks; T5 — 6 months; T6 = 1 year; and T7 =
2 years. To find the appropriate CI value, extrapolate
either horizontally or vertically to the nearest edge of
the CI area. The CIs are not drawn symmetrically
around the plotted group means.
moved at 6 weeks with splints and suspension
wires removed at 7 weeks. That is one reason why
both groups had radiographs at 8 weeks rather
than the traditional 6 weeks after surgery. An
additional reason for the inferior position of
B-point may be caused by settling of the occlu-
sion. Previous studies20' 21 reported loss of verti-
cal height with both fixation methods, but not to
the extent seen in our study. In contrast, Mom-
maerts et al10 found no changes in the vertical
height with either type of fixation.
The sagittal surgical advancement of B-point
also was similar for the rigid and wire groups, 5.0
mm and 5.4 mm, respectively. In the wire group,
horizontal relapse occurred at a much faster rate
(18%) during the first eight weeks after surgery.

- 2 - 1 0 1 Table 6. Changes in Mandibular Incisor Angulation

Horizontal position (mm) (Ll-NB)

Figure 3. Horizontal and vertical position of the man- Time Interval Rigid Wire W)
dibular incisor. The area delineated around the time
points includes the 95% confidence interval (CI) in 8 wks (T4-T3) 0.08 (2.87) -1.58* (2.68) .020
both the sagittal and horizontal direction. T3 (imme- 6 mos (T5-T3) 1.00 (3.89) -0.24 (4.01) .030
diate postsurgery was set at zero for both groups); 1 yr (T6-T3) 1.67* (4.27) 0.58 (4.42) .027
T4 = 8 weeks; T5 = 6 months; T6 = 1 year; and T7 = 2 yrs (T7-T3) 2.07* (4.58) 1.41* (3.70) .037
2 years. To find the appropriate CI value, extrapolate NOTE. ( — ) represents retroclination. ( + ) represents procli-
either horizontally or vertically to the nearest edge of nation. All values expressed as mean (standard deviation).
the CI area. The CIs are not drawn symmetrically * Significantly different within group from postsurgery (T3)
around the plotted group means. (P< .01).
<<    
     Article
      >>
Skeletal and Dental Changes After a BSSO
Table 7. Changes in Maxillary Incisor Angulation
(Ul-SN)
Time Interval Rigid Wire
8 wks (T4-T3) -1.65* (2.69) 0.59 (2.33)
6 mos (T5-T3) -2.17* (3.98) 0.04 (3.62)
1 yr (T6-T3) -1.58* (4.39) -0.14(3.95)
2 yrs (T7-T3) -1.04* (4.81) 0.60 (4.40)
NOTE. ( — ) represents retroclination. ( + ) represents procli-
nation. All values expressed as mean (standard deviation).
*Significantly different within group from postsurgery (T3)
(P < .01).
Thereafter, relapse continued, but at a much
slower rate. Finally, by two years, the sagittal
relapse had progressed to 28%. In the rigid
group, at the two-year time point, B-point was
stable or unchanged from its postsurgical posi-
tion, even though B-point did move anteriorly at
the earlier time points. The results of this study
mimic what we reported in previous reports in
which relapse was examined using different
analyses15 or points of reference.16 Our study, as
well as those of other investigators,10 support the
idea that the skeletal stability is much greater
where rigid fixation is used. On the contrary,
Watzke et al21 reported no difference between
rigid and wire fixation one year after surgery.
Yet, another study reported relapse with either
type of fixation.20 Most, if not all studies, that
have examined skeletal relapse, have been ret-
rospective. Because of the many limitations of
retrospective studies, their results should often
be interpreted cautiously.
Given the skeletal relapse in the wire group, it
is surprising to find no difference in overbite
and over]et between these two groups. We spec-
ulate that in both groups the dental changes
cannot compensate for the skeletal changes.
The dental changes that take place after surgery,
have been attributed to orthodontic detailing of
the occlusion and/or to compensatory tooth
movement caused by skeletal changes. Changes
in the sagittal and vertical position of the inci-
sors in this study can only be compared with the
study by Watzke et al because other studies used
different selection criteria. In this study, we
found changes in both the maxillary and man-
dibular incisors with both fixation groups. The
sagittal and vertical position of the mandibular
incisor of the wire group followed B-point very
closely. Because the maxillary incisor of the wire
Home | TOC |          
Index
41
group partially (0.28 mm) followed the mandib-
ular incisors, there is an increase in overjet with
P time. In the wire fixation group, Watzke et al
(R v W) reported a -0.6-mm change in overjet; a value
.088 that is about 1 mm less than what we report. In
.074 the rigid fixation group, the change of the man-
.161
.124 dibular incisor was 0.1 mm; a value that is similar
to our findings. At the early time points, the
mandibular and maxillary incisors of the rigid
group also followed B-point and move anteri-
orly. At the two-year time point, however, the
mandibular incisors are further posterior than
B-point. Because the maxillary incisors again
partially follow the mandibular incisors, by two
years an increase in overjet occurs. Although the
final overjet is similar in both groups, the man-
ner in which it occurs in the two groups of
patients is different. In the rigid fixation group,
it is produced from dental changes, whereas in
the wire group, it is the result of skeletal relapse.
A recently published study that compared rigid
fixation with wire fixation also found increases
in overjet after surgery with no differences be-
tween the two groups.20
The results from this study have implications
for any orthodontist preparing his or her patient
for surgery. If the orthodontist is to maintain a
stable occlusal result, then he or she may choose
to set up mandibular advancement cases differ-
ently, depending on the type of fixation to be
used. For example, if rigid fixation will be used,
the maxillary incisors may be left slightly pro-
clined or with some space distal to the lateral
incisors. This space could then be used to ac-
commodate any changes in mandibular incisor
position. In the rigid fixation group, during
treatment, the orthodontist should try not to
adjust the occlusion immediately after surgery to
accommodate the skeletal changes. For a patient
who is to receive wire fixation, the mandibular
incisor may be placed in a retroclined position
before surgery. This may require extraction of
mandibular premolars. Then, as skeletal relapse
occurred, the overjet could be maintained by
the proclination of the incisors. The previously
mentioned study could also be used to deter-
mine retention schemes. For example, perma-
nent retention for the mandibular incisors could
be used with rigid fixation to prevent them from
moving posteriorly.
<<    
     Article
      >>
42 Dolce et al
References
1. Obwegeser HL. Indication for surgical correction of
mandibular deformity by the saggital splitting tech-
nique. Br J Oral Surg 1964;1:157-171.
2. Smith GC, Moloney FB, West RA. Mandibular advance-
ment surgery: A study of the lower border wiring tech-
nique for osteosynthesis. Oral Surg Oral Med Oral
Pathol Oral Radiol Endod 1985;60:467-475.
3. Spiessl LB. The sagittal splitting osteotomy for correc-
tion of mandibular prognathism. Clin Plast Surg 1982;
9:491-507.
4. Schendel SA, Epker BN. Results after mandibular ad-
vancement surgery: An analysis of 87 cases. J Oral Surg
1980;38:265-282.
5. Kohn MW. Analysis of relapse after mandibular advance-
ment surgery. J Oral Surg 1978;36:676-684.
6. Van Sickels JE, Larsen AJ, Thrash WJ. Relapse after rigid
fixation of mandibular advancement. J Oral Maxillofac
Surg 1986;44:698-702.
7. Van Sickels JE, Larsen AJ, Thrash WJ. A retrospective
study of relapse in rigidly fixated sagittal split osteoto-
mies: Contributing factors. Am J Orthod Dentofacial
Orthop 1988;93:413-418.
8. Kirkpatrick TB, Woods MG, Swift JQ, et al. Skeletal
stability following mandibular advancement and rigid
fixation. J Oral Maxillofac Surg 1987;45:572-576.
9. Gassmann CJ, Van Sickels JE, Thrash WJ. Causes, loca-
tion, and timing of relapse following rigid fixation after
mandibular advancement. J Oral Maxillofac Surg 1990;
48:450-454.
10. Mommaerts MY, Lag screw versus wire osteosynthesis in
mandibular advancement. Int J Adult Orthodont Or-
thognath Surg 1991;6:153-160.
11. Poulton DR, Ware WH, Baumrind S. Surgical mandibu-
Home | TOC |          
Index
lar advancement studied with computer-aided cephalo-
metrics. Am J Orthod 1979;76:121-135.
12. Lake SL, McNeill RW, Little RM, et al. Surgical mandib-
ular advancement: A cephalometric analysis of treat-
ment response. Am J Orthod 1981;80:376-394.
13. Van Sickels JE, Flanary CM. Stability associated with
mandibular advancement treated by rigid osseous fixa-
tion. J Oral Maxillofac Surg 1985;43:338-341.
14. Tulloch JFC, Medland W, Tuncay OC. Methods used to
evaluate growth modification in Class II malocclusion.
Am J Orthod Dentofacial Orthop 1990;98:340-347.
15. Keeling SD, Dolce C, van Sickles JE, et al. A comparative
study skeletal and dental stability between rigid versus
wire fixation for mandibular advancement. Am J Orthod
Dentofacial Orthop 2000;117:638-649.
16. Dolce C, Van Sickels JE, Bays RA, et al. Skeletal stability
after mandibular advancement with rigid versus wire
fixation. J Oral Maxillofac Surg 2000;58:1219-1227.
17. Bell WH, Proffit WR, Chase DL, et al. Mandibular defi-
ciency, in Bell W, Proffit W, White R (eds):Surgical
Correction of Dentofacial Deformities. Philadelphia, PA,
WB Saunders, 1980, pp 685-843.
18. Keeling SD, Cabassa SR, King GJ. Systematic and ran-
dom errors associated with Johnston's cephalometric
analysis. Brit J Orthod 1993;20:101-107.
19. Houston WJB. The analysis of errors in orthodontic
measurements. Am J Orthod 1983;83:382-390.
20. Berger JL, Pangrazio-Kulbersh V, Bacchus SN, et al.
Stability of bilateral sagittal split ramus osteotomy: Rigid
fixation verus transosseous wiring. Am J Orthod Dento-
facial Orthop 2000;118:397-403.
21. Watzke IM, Turvey TA, Phillips C, et al. Stability of
mandibular advancement after sagittal osteotomy with
screw or wire fixation: A comparative study. J Oral Max-
illofac Surg 1990;48:108-121.
<<    
     Article
      >> Home | TOC |          
Index

The Invisalign System: Case Report of a

Patient With Deep Bite, Upper Incisor Flaring,
and Severe Curve of Spee
Ross J. Miller and Mitra Derakhshan

The Invisalign System (Align Technology, Santa Clara, CA) is a new treat-
ment method using computer virtual treatments to create a series of clear,
removable, esthetic appliances that can treat a wide range of malocclusions.
We present the treatment of a patient who presents with a deep bite, upper
incisor flaring, and deep curve of Spee with Invisalign. (Semin Orthod 2002;
8:43-50.) Copyright 2002, Elsevier Science (USA). All rights reserved.

Inthe last few years, there has been consider-
able interest in orthodontic treatment for
adult patients. A survey conducted by the AAO
showed an increase in the percentage of patients
over 21 years of age from 4% 10 years ago to
almost 7% today. This is expected to grow to
nearly 11% after another decade.1 Innovations
in appliances have caught the attention of many
adults who would not otherwise seek orthodon-
tic treatment.2 The concept behind using a se-
ries of customized removable appliances to treat
a malocclusion, first appeared in orthodontic
literature in 1945 with an article by Kesling.3 The
purposes of this article are to present an over-
view of the Invisalign System and then present a
case treated with Invisalign.
patient. If the patient has spacing, you must
decide if you are going to simply close space or
work with a restorative dentist to move teeth for
composites or veneers. If the patient has crowd-
ing, you must decide if you are going to strip,
expand, use distalization mechanics, or extract.
This information must be put onto 3 treatment-
planning forms that are sent to Align Technol-
ogy. You can also treatment plan on the Web by
using the Web site and Web treatment planning
in your favorite Web browser. Align Technology
also requires you to take polyvinylsiloxane (PVS)
impressions4 and to send a copy of the pan-
oramic radiograph or full-mouth series of in-
traoral radiographs and cephalograms when ap-
propriate to Align Technology.5

Materials and Methods Align Technology Receives the Case

The basic steps in how the orthodontist inter-
acts with the Invisalign manufacturing process
are shown in Figure 1.
Records, Polyvinylsiloxane Impressions, and
Treatment Planning Forms
As with all orthodontics, diagnosis and treat-
ment planning are critical. The clinician must
have clear ideas about what will be done for the
From Align Technology, Santa Clara, CA.
Address correspondence to Ross J. Miller, DDS, MS, 881 Martin
Avenue, Santa Clara, CA 95050.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0801-0008$35.00/0
doi: 10.1053/sodo. 2002.28182
Align Technology receives the case submis-
sion, and the case is checked for all necessary
information including PVS impressions, photo-
graphs, radiographs, bite, and treatment-plan-
ning forms. Once it is determined to be com-
plete, the impressions are then poured up in
dental die stone. The records are checked by a
clinical team to make sure that the case is suit-
able for treatment and that the treatment plan is
clinically viable. If direct communication is re-
quired to clarify treatment goals, this will be
done at this time.
Scanning, Cutting, and Treating
The models are scanned with white light us-
ing the bite as a reference. Once this informa-

Seminars in Orthodontics, Vol 8, No 1 (March), 2002: pp 43-50 43

<<    
     Article
      >>
44 Miller and Derakhshan
CiinCheck -
web based
tx. viewer
Figure 1. Orthodontist interaction with the Invisalign
System.
tion is acquired to the computer, each arch is
then destructively scanned. Destructive scanning
implies that the object being scanned is de-
stroyed in the process. The models are sliced at
.003 in (.075 mm). Each of the slices is then
reassembled on the computer. The model then
becomes a virtual image in the computer and
can be manipulated using the software program
Treat II. Treat II has many software tools analo-
gous to what one would have in a dental labora-
tory (eg, saws, trimmers, etc). The models can be
manipulated in 6 degrees of freedom. At this
time, individual teeth are separated from each
other and trimmed. Each tooth is now a separate
image that can be moved from one place to
another. A setup is performed, which is similar
to what you would do with a positioner. A deter-
mination of when and how much each tooth will
be moved is planned. Once the treatment setup
is finished, there is a clinical check by a clinical
team.
CiinCheck
Once the case is acceptable, it is made into a
smaller image that can be sent over the Internet
Figure 2. Pretreatment facial photographs.
Home | TOC |          
Index
for viewing by the clinician. If modifications
need to be performed, they are done at this
time. The clinician should be critical of planned
treatments making sure that the movements are
clinically feasible. Movements that are not pos-
sible by using conventional appliances, may also
not be possible with Invisalign.
Stereo Lithography Apparatus and Aligner
Fabrication
The CiinCheck file the clinician receives over
the Internet should be thought of as a 3-dimen-
sional movie, each of the stages or aligners is a
frame in that movie. These computer-generated
frames put together are the steps that create the
treatment. Each of these stages during the man-
ufacturing process is made into a 3-dimensional
physical model on a stereo lithography appara-
tus (SLA), which places new resin layers at a
thicknesses of .006 in (0.15 mm). If there are 10
aligner treatments for one arch, the manufactur-
ing process will create 10 separate SLA parts.
From each of those parts, an aligner will be
made in the usual pressure-forming unit (Bio-
star; Great Lakes Orthodontics, Towanda, NY).
Case Presentation
This 46-year-old man presented for orthodon-
tic treatment with the chief complaint of upper
spacing. A white patch on the right buccal mu-
cosa was noticed and monitored through treat-
ment. Otherwise, the patient was medically
healthy and a nonsmoker. There were no con-
traindications to treatment. Facial analysis re-
<<    
     Article
      >> Home | TOC |          
Index

Lower Incisor Intrusion Utilizing Invisalign 45

Figure 3. Pretreatment intraoral photographs.

vealed a brachycephalic pattern and straight
profile (Fig 2). His lips were competent, and
there was no strain on closure. When smiling,
the patient did not show all of his incisors. In-
traoral photographs showed that the right and
left molars and canines were in a Class I relation-
ship (Fig 3). There was 7 mm of overjet and 5
mm of overbite with incisor contact. The maxil-
lary and mandibular midlines were centered to
the facial midline. There was excess upper space

Figure 4. Pretreatment cephalometric radiograph.

<<    
     Article
      >> Home | TOC |          
Index

46 Miller and Derakhshan

with a large midline diastema. Crowding of the

lower arch was 2 mm, and the curve of Spec was
increased because of overerupted lower incisors.
There was no Bolton discrepancy. Panorex re-
vealed that the soft and hard tissues were within
normal limits. The cephalometric radiograph in-
dicated bimaxillary dentoalveolar protrusion,
excessively proclined upper incisors, and a flat
mandibular plane (Fig 4).

Treatment Plan
The treatment objectives were to close all up-
per spaces and reduce the overjet by retroclining
and retracting upper incisors. Intruding lower
incisors would level the curve of Spec and thus
decrease overbite. Lower crowding was to be
resolved by proclination and interproximal re-
duction if needed. The buccal occlusion and
Class I were to be maintained.

Treatment Progress
PVS impressions were taken, and treatment
began with aligners. These were changed at
2-week intervals. After 4 sets of aligners, the
lower incisors were not responding to the
planned intrusion. We concluded that because
intrusive forces were placed on the lower inci-
sors, there was not enough counter force in the
posterior region to retain the appliances. There-
fore, attachments were bonded on the lower
canines and premolars to provide retention for
the aligners in the posterior as anterior intrusive
forces were being delivered. The first 4 sets of
aligners were redelivered, and intrusion was suc-
cessfully accomplished. Figure 5 shows an in-
traoral photograph and corresponding clin-
check image of intrusion of the lower central
incisors as compared to the lateral incisors.
Treatment continued with 2-week intervals until
all aligners had been delivered. At this point,
there were residual spaces remaining in the up-
per arch that needed to be closed, and approx-
imately 1 mm more intrusion of the lower lateral
incisors and left central incisor was required.
New PVS impressions and finishing aligners
were made including attachments bonded on
the lower canines and lower right central inci-
sor. Incisal plasty was performed on all laterals.
The patient was then placed in retention by
using upper and lower clear retainers to be worn
at nighttime. The total treatment time was 14
Figure 5. Progress intraoral photograph (A) and clin-
check image (B).
months. A total of 20 plus 4 finishing aligners
were required in the upper and 20 plus 3 align-
ers in the lower to complete the treatment.
Treatment Results
Posttreatment facial photos show minimal
change in the facial profile (Fig 6). The initial
Class I molar and canine relationships were
maintained with complete upper space closure
(Fig 7). The overjet was reduced from 7 mm to
4 mm. The upper midline was slightly shifted to
the left because of the fact that there was more
space to be closed on the left than on the right.
The upper and lower arches showed acceptable
alignment. The curve of Spee was decreased,
resulting in a reduction of the overbite. The
posttreatment panoramic radiograph shows
good root approximation (Fig 8). There appears
to be a widened periodontal ligament around
the upper right lateral incisor, which was asymp-
tomatic. The posttreatment lateral cephalogram
shows reduction of the overjet with retraction
and retroclination of the upper incisors (Fig 9).
<<    
     Article
      >> Home | TOC |          
Index

Lower Incisor Intrusion Utilizing Invisalign 47

Figure 6. Posttreatment facial photographs.

Figure 7. Posttreatment intraoral photographs.

Figure 8. Posttreatment panoramic radiograph.

<<    
     Article
      >> Home | TOC |          
Index

48 Miller and Derakhshan

Figure 9. Posttreatment cephalometric radiograph.

Cranial base superimposition showed minimal

mandibular rotation (Fig 10). Maxillary super-
imposition showed that the upper incisor moved
lingually, which contributed to the decrease in
the overjet. Upper molar position was un-
changed. Mandibular superimposition showed
that the lower molar position was unchanged
and the lower incisors were intruded 2 mm and
slightly proclined (advanced) (Fig 11).

Discussion
A thorough treatment plan is imperative with
the Invisalign System. This new treatment mo-
dality requires the clinician to plan out reason-
able sequential tooth movements for every tooth
from beginning to end. The movements that are
shown on the computer will be duplicated in the
aligners. The patient's final tooth setup and
stages of tooth movement were generated by
using a 3-dimensional computer system by Align
Technology Inc. and reviewed by the orthodon- Figure 10. Pretreatment (solid line)/posttreatment
tist on a computer. In this case, because the (dotted line) overall superimpositions.
<<    
     Article
      >>
Lower Incisor Intrusion Utilizing Invisalign
Figure 11. Pre treatment (solid line)/posttreatment
(dotted line) maxillary and mandibular superimpo-
sitions.
objectives were to maintain the buccal segments,
it was decided by the orthodontist to limit move-
ments to anterior teeth only. Therefore in Clin-
check, only the upper and lower canines and
incisors were sequentially moved. Figures 12 and
13 show a schematic representation of the tooth
movements of the lower and upper arches, re-
spectively. Each column represents a tooth, and
each row represents a stage with its correspond-
ing aligners. The black lines span the stages in
which a tooth moves. It can be seen that lower
3(2?)| 2 C 2 6 ) | 1(25)j 1 ( 2 4 ) f 2 ( 2 3 ) I 3(221
Figure 12. Planned tooth movements for the lower
arch.
Home | TOC |          
Index
49
J J
10
12
13
14
16
17
li
19
20
21
22
Figure 13. Planned tooth movements for the upper
arch.
arch treatment was initiated with intrusion to
achieve clearance for retraction of the upper
incisors. Therefore, upper arch treatment was
delayed by 5 stages. Because of the amount of
lower-incisor intrusion that was needed, it was
decided to begin initially with intrusion of the
lower central incisors and then proceed to the
lower laterals. Referring to Figures 12 and 13,
intrusion of the lower centrals began at stage 1
and then the lower laterals began at stage 4. The
intrusion of the central incisors was completed
at stage 5 and the laterals at stage 8. The canines
moved throughout the 20 stages. As previously
discussed, delayed upper movement began with
the laterals moving initially distally to allow
movement of the centrals without causing inter-
arch collisions.
The objectives in this case were achieved. The
upper incisors were retroclined and retracted
during treatment. The lower incisors were in-
truded. The orthodontist asked for intrusive
movements to be initiated with the centrals first
and then followed with the laterals to minimize
anchorage requirements. Throughout this pro-
cess, it was decided that attachments were
needed on the posterior teeth for retention of
the appliance during intrusion of anterior teeth.
Another significant finding in this case was the
need for overcorrection. In this case, 3 to 4
finishing aligners were needed to satisfactorily
complete the treatment. The Invisalign System
<<    
     Article
      >> Home | TOC |          
Index

50 Miller and Derakhshan

requires the orthodontist to actively participate 2. FastlitchJ. Adult orthodontics. J Clin Orthod 1982;16:606-
with Align Technology, Inc. 618.
3. Kesling HD. The philosophy of the tooth positioning
appliance. Am J Orthod 1945;31:297-304.
4. McNamara JA, Brudon J. Orthodontics and Dentofacial
Orthopedics. Ann Arbor, Needham Press, Inc, pp 483-486
References 5. Align Technology. Orthodontist workbook: Embark on a
1. Levitt H. Adult orthodontics. J Clin Orthod 197l;5:130- whole new movement in adult orthodontics. Santa Clara,
155. CA: Align Technology, 2000
<<    
     Article
      >> Home | TOC |          
Index

Take Advantage
of the
Authorities.
W hen leading authorities have mentioned your
contributions in a well-respected publication like
this one, it's time to order reprints.
That's because when praise comes from independent
sources it's the perfect opportunity to share it with your
colleagues and customers.
Reprints are available from W.B. Saunders in quantities of
100 or more and can be customized to meet your
requirements.
So if you are fortunate enough to have this opportunity,
take advantage of it. You simply couldn't say it better
ir Journal Reprints Department
^ * Attn: Ginny Nicholls
W.B. Saunders Company (A Harcourt Health Sciences Company)
The Curtis Center/Independence Square West
Philadelphia, PA 19106-3399
WJB. SAUNDERS Call: (215)238-5534
ElsevierScience [™,™ 238-6423
Email:gnicholls@harcourt.com
<<    
     Article
      >> Home | TOC |          
Index

W. B. S A U N D E R S

JOURNALS Essential Information f or Today's Professionals from theleadingHealth Caw Publisher

ANESTHESIOLOGY NEPHROLOGY PSYCHIATRY

Journal of Cardiothoracic and Vascular Anesthesia Advances in Renal Replacement Therapy—A Journal of Comprehensive Psychiatry—Official Journal of the A\
Journal of Vw—OfficiallJournal of the American Pain Society The National Kidney foundation Psycbopatbological Association
Regional Anesthesia and Pain Medicine American Journal of Kidney Diseases—The Official Journal Seminars in Clinical Neuropsychiatry
Offmal Journal of the American, Asian and (Oceanic, of The National Kidney Foundation
and Latin American Societies of RegionalAnesthesia Journal of Renal Nutrition—The Official Journal of The Council RADIOLOGY
Seminars in Anesthesia on Renal Nutrition of the National Kidney Foundation Seminars in Breast Disease
Seminars in Cardiothoracic and Vascular Anesthesia Seminars in Nephrology Seminars in Nuclear Medicine
Seminars in Pain Medicine Seminars in Radiologie Technology
Techniques in Regional Anesthesia and Pain Management NEUROLOGY Seminars in Roentgenology
Journal of Pain—Official Journal of the American Pain Society Seminars in Ultrasound, CT and MRI
CARDIOVASCULAR DISEASES Journal of Stroke and Cerebrovascular Diseases—Official Journal Techniques in Vascular and Interventional Radiology
Progress in Cardiovascular Diseases of the National Stroke Association and theJapan Stroke Society
Techniques in Intervention^ Cardiology Seminars in Cerebrovascular Diseases and Stroke RHEUMATOLOGY
Seminars in Pediatric Neurology Seminars in Arthritis and Rheumatism
CRITICAL CARE MEDICINE
Journal of Critical Care NURSING SURGERY
Advances in Neonatal Care—Official Journal of the National Journal of Pediatric Surgery—Official Journal of the Section on
DENTISTRY Surgery of the American Academy of Pediatrics, British Association
Journal of Oral and Maxillofacial Surgery—Official Journal of ofPaediatric Surgeons, American Pediatric Surgical Association,
Applied Nursing Research Canadian Association ofPaediatric Surgeons and Pacific
the American Association of Ord and MaxiUofacial Surgeons Archives of Psychiatric Nursing—Official Journal of the SERPN
Journal of Prosthodontics—Official Journal of The American Association of Pediatric Surgeons
Division, International Society of Psychiatric -Mental Health Nurses Operative Techniques in General Surgery
Journal of Pediatric Nursing—Official Journal of the Society of Operative Techniques in Neurosurgery
Seminars in Orthodontics Pediatric Nurses Operative Techniques in Plastic and Reconstructive Surgery
DERMATOLOGY Journal of Pediatric Oncology Nursing—Official Journal of the Operative Techniques in Thoracic and Cardiovascular Surgery—An
American Journal of Contact Dermatitis—The Official Journal Association of Pediatric Oncology Nurses Official Publication of The American Association for Thoracic Surgery
of the American Contact Dermatitis Society Journal of PeriAnesthesia Nursing—Official Journal of the American Seminars in Colon and Rectal Surgery
Seminars in Cutaneous Medicine and Surgery Society of PeriAnesthesia Nurses Seminars in Laparoscopic Surgery
Journal of Professional Nvr^n%—0fficialjournal of the American Seminars in Pediatric Surgery
EMERGENCY MEDICINE Association of Colleges of Nursing Seminars in Spine Surgery
American Journal of Emergency Medicine Pain Management Nursing—Official Journal of the American Society Seminars hi Thoracic and Cardiovascular Surgery—tin Official
Clinical Pediatric Emergency Medicine of Pain Management Nurses Publication of The American Association for Thoracic Surgery
PeriAnesthesia and Ambulatory Surgery Nursing update—Official Seminars hi Thoracic and Cardiovascular Surgery: Pediatric Cardiac
ENDOCRINOLOGY Publication of the American Society ofPeriAnestbesia Nurses Surgery Annual—An Official Publication of The American
Metabolism—Clinical and Experimental Seminars for Nurse Managers Associationfor Thoracic Surgery
Seminars in Oncology Nursing Seminars in Urologie Oncology
GASTROENTEROLOGY/HEPATOLOGY Seminars in Vascular Surgery
Clinical Perspectives in Gastroenterology—The Official OBSTETRICS AND GYNECOLOGY
Clinical PracticeJournal of the American Gastroenterological Clinical Journal of Women's Health TRANSPLANTATION
Association Liver Transplantation-^» Official Publication of the American
Gastroenterology— Official Journal of the American ORTHOPEDICS Association for the Study of Liver Diseases and the International
GastroenterologicalAssociation Armroscopy: The Journal of Arthroscopic and Related Surgery— Liver Transplantation Society
Official Publication of the Artbroscofty Association of North America Transplantation Reviews
the Study of Liver Diseases and the International Society ofArtbroscopy, Knee Surgery, and
liver Transplantation—4» Official Publication of the American Orthopaedic Sports Medicine UROLOGY
AssociationJbr the Study of Liver Diseases and The International Journal of the American Society for Surgery of the üanA— Seminars in Urologie Oncology
Liver Transplantation Society An Officidjwrnal of the Americm Society for Surgery
Seminars in Gastrointestinal Disease of the Hand VETERINARY MEDICINE
Techniques in Gastrointestinal Endoscopy Journal of Hand Surgery-^4» Official Journal of the American Advances hi Small Animal Medicine and Surgery
Society for Surgery of the Hand Clinical Techniques in Equine Practice
HEMATOLOGY/ONCOLOGY Operative Techniques in Orthopaedics Clinical Techniques in Small Animal Practice
Seminars in Hematology Operative Techniques in Sports Medicine Seminars in Avian and Exotic Pet Medicine
Seminars in Oncology Seminars in Armroplasty Veterinary Surgery—The Official Journal of The American
Seminars in Radiation Oncology Seminars in Spine Surgery College of Veterinary Surgeons, Inc. and The European College
Transfusion Medicine Reviews of Veterinary Surgeons
OTORHINOLARYNGOLOGY
INFECTIOUS DISEASES American Journal of Otolaryngology For more information about thesejournals,
Seminars in Infection Control Operative Techniques in Otolaryngology—Head and Neck Surgery please contact:
Seminars in Pediatric Infectious Diseases
Seminars in Respiratory Infections PATHOLOGY Periodicals Marketing
Annals of Diagnostic Pathology W.B. SAUNDERS
MEDICAL TRANSCRIPTION Human Pathology A Division of Elsevier Science
The Latest Word Seminars in Diagnostic Pathology
The Curtis Center, Independence Square West
NEONATAL/PERINATAL MEDICINE PHYSICAL MEDICINE Philadelphia, PA 19106-3399
Newborn and Infant Nursing Reviews Archives of Physical Medicine and Rehabilitation—
Seminars in Perinatology Official Journal of the American Congress of Rehabilitation
Phone (215) 238-5614
Medicine and the American Academy of Physical Medicine Or visit our homepage at:
www.wbsaunders.com
<<    
     Article
      >> | HOME
|           TOC In
<<    
     Article
      >> Home | TOC |          
Index

The world's
bestselling
medical
dictionary!

Dorland's Illustrated Medical Dictionary all of its illustrations, tables, and appendices.
is universally acknowledged as the world's At the same time, its state-of-the-art CD-ROM
finest medical dictionary. For 100 years, technology puts tremendous reference power
health care professionals have relied on its and convenience at users' disposal—plus audio
unmatched comprehensiveness, accuracy, pronunciations for over 10,000 primary entries!
clarity, and ease of use. 2000. Single-user CD-ROM for Windows™ or
The 29th Edition presents the very latest Macintosh*. Order #W9493-4.
information from every frontier in health A W.B. Saumlers alle.
care! It offers over 8,100 new terms—
7? Phone:
121,160 in all • over 7,600 new entries— Call toll-free 1-800-545-2522
for a total of 117,469 • over 860 illustra- (8:30-8:00 Eastern Time) to order.
tions—566 brand new • and much more! Be sure to mention DM#66899.
2000. 2112 pp. 864 ills. Order 3W6254-4.
Fax to i-800-568'5136 to order.
Also available on CD-ROM! Be sure to mention DM#66899.

Dorland's Electronic Medical Dictionary, 1^1 Mail:

29th Edition contains all of the definitions, Else vie r Health Sciences
Order Fulfillment Dept.
pronunciations, plural forms, and etymologies 1 1830 Wcsrline Industrial Drive
found in the hardbound dictionary, as well as Saint Louis, MO 63H6-3318

W.B. SAUNDERS
Elsevier Science
EHS 01 DM166»« * Elm*» SCTCHC, ZWä
<<    
     Article
      >>
Seminars in Orthodontics
EDITOR
P. Lionel Sadowsky, DMD, BDS, DipOrth, MDent
Professor and Chairman
Department of Orthodontics
University of Alabama
1919 Seventh Avenue South
Birmingham, AL 35294
Fax: (205) 975-7590
Seminars in Orthodontics (ISSN 1073-8746) is published
quarterly by W.B. Saunders. Months of issue are March, June,
September, and December. Corporate and Editorial Offices:
The Curtis Center, Independence Square West, Philadelphia,
PA 19106-3399. Accounting and Circulation Offices: 6277 Sea
Harbor Drive, Orlando, FL 32887-4800. POSTMASTER: Send
change of address to: Seminars in Orthodontics, W.B. Saunders,
Customer Service, 6277 Sea Harbor Drive, Orlando, FL 32887-
4800.
Editorial correspondence should be addressed to:
P. Lionel Sadowsky, DMD, BDS, DipOrth, MDent, Editor,
Seminars in Orthodontics, Professor and Chairman, Department
of Orthodontics, University of Alabama, 1919 Seventh Avenue
South, Birmingham, AL 35294-0007; fax: (205) 975-7590.
Correspondence regarding subscriptions or change of
address should be directed to Seminars in Orthodontics, W.B.
Saunders, Periodicals Department, P.O. Box 628239, Orlando,
FL 32862-8239 or e-mail hhspcs@harcourt.com.
Change of address notices, including both the old and new
addresses of the subscriber and the mailing label, should be
sent at least 1 month in advance. Customer Service: 1-800-654-
2452
Yearly subscription rates: United States and possessions:
individual, $133.00; institution, $164.00; student and resident,
$67.00; single issue, $50.00. All other countries: individual
$166.00; institution, $198.00; student and resident, $83.00;
single issue, $50.00. For all areas outside the United States and
possessions, there is no additional charge for surface delivery.
For air mail delivery, add $16.00. To receive student/resident
rate, orders must be accompanied by name of affiliated
institution, date of term, and the signature of program/residency
coordinator on institution letterhead. Orders will be billed at
individual rate until proof of status is received.
Prices are subject to change without notice. Current prices
are in effect for back volumes and back issues. Single issues,
both current and back, exist in limited quantities and are
offered for sale subject to availability. Back issues sold in
conjunction with a subscription are on a prorated basis. Checks
should be made payable to W.B. Saunders and sent to Seminars
in Orthodontics, W.B. Saunders, Periodicals Department, 6277
Sea Harbor Drive, Orlando, FL 32887-4800.
Copyright 2002, Elsevier Science (USA). All rights
reserved. No part of this publication may be reproduced or
transmitted in any form or by any means, electronic or
mechanical, including photocopy, recording, or any information
storage and retrieval system, without permission in writing
from the Publisher. Printed in the United States of America.
Correspondence regarding permission to reprint all or
part of any article published in this journal should be
addressed to Journal Permissions Department, W.B. Saunders,
6277 Sea Harbor Drive, Orlando, FL 32887-4800. Telephone
number: 1-407-345-2500.
W.B. SAUNDERS
Home | TOC |          
Index
The appearance of the code at the bottom of the first page
of an article in this journal indicates the copyright owner's
consent that copies of the article may be made for personal or
internal use, or for the personal or internal use of specific clients,
for those registered with the Copyright Clearance Center, Inc.
(222 Rosewood Drive, Danvers, MA 01923; (508) 750-8400;
www.copyright.com). This consent is given on the condition that
the copier pay the stated per-copy fee for that article through the
Copyright Clearance Center, Inc. for copying beyond that
permitted by Sections 107 or 108 of the US Copyright Law.
This consent does not extend to other kinds of copying, such
as copying for general distribution, for advertising or promotional
purposes, for creating new collective works, or for resale.
Absence of the code indicates that the material may not be
processed through the Copyright Clearance Center, Inc.
Reprint inquiries should be addressed to Ginny
Nicholls, Elsevier Science, The Curtis Center, Independence
Square West, Philadelphia, PA 19106-3399. Telephone (215)
238-5534, fax (215) 238-6423; e-mail: gnicholls@elsevier.com.
Advertising representative: MJ. Mrvica Associates, Inc,
2 West Taunton Ave, Berlin, NJ 08009. Telephone (609) 768-
9360. Fax (609) 753-0064.
Publication of an advertisement in Seminars in Orthodontics
does not imply endorsement of its claims by the Editor (s) or
Publisher of the journal.
The contributors have checked generic and trade names and
verified drug doses for accuracy according to the standards
accepted at the time of publication. The ultimate
responsibility, however, lies with the prescribing physician.
Please convey any errors to the Editor.
The ideas and opinions expressed in Seminars in
Orthodontics do not necessarily reflect those of the Editor or the
Publisher. Publication of an advertisement or other product
mention in Seminars in Orthodontics should not be construed as
an endorsement of the product or the manufacturer's claims.
Readers are encouraged to contact the manufacturer with any
questions about the features or limitations of the products
mentioned. Neither the Editor or Publisher assume any
responsibility for any injury and/or damage to persons or
property arising out of or related to any use of the material
contained in this periodical. The reader is advised to check the
appropriate medical literature and the product information
currently provided by the manufacturer of each drug to be
administered to verify the dosage, the method and duration of
administration or contraindications. It is the responsibility of
the treating physician or other health care professional, relying
on independent experience and knowledge of the patient, to
determine drug dosages and the best treatment for the patient.
Seminars in Orthodontics is indexed in the Cumulative
Index to Nursing and Allied Health Literature® print index
and the Cinahl® database.
<<    
     Article
      >>
Seminars in Orthodontics
EDITOR
P. Lionel Sadowsky, DMD, BDS, DipOrth, MDent
EDITORIAL BOARD
Richard G. Alexander, Arlington, TX
Rolf G. Behrents, Memphis, TN
Samir E. Bishara, Iowa City, I A
Robert Boyd, DDS, San Francisco, CA
Larry M. Bramble, Cypress, C A
John S. Casko, Iowa City, IA
Harry L. Dougherty, Van Nuys, C A
T.M. Graber, Evanston, IL
Robert J. Isaacson, Richmond, VA
Alexander Jacobson, Birmingham, AL
Lysle E.Johnston, Jr., Ann Arbor, MI
Gregory J. King, Seattle, WA
Vincent G. Kokich, Tacoma, WA
Steven J. Lindauer, Richmond, VA
INTERNATIONAL
Zeev Abraham, Herzliya, Israel
W.G. Evans, Johannesburg, South Africa
Roberto Justus, Mexico City, Mexico
Home | TOC |          
Index
James A. McNamara, Jr, Ann Arbor, MI
Robert N. Moore, Grand Island, NE
Ravindra Nanda, Farmington, CT
Perry M. Opin, Milford, CT
Sheldon Peck, Newton, MA
William R. Proffit, Chapel Hill, NC
Cyril Sadowsky, Chicago, IL
David M. Sarver, Birmingham, AL
T. Michael Speidel, Minneapolis, MN
William J. Thompson, Bradenton, FL
James L. Vaden, Cookeville, TN
Robert L. Vanarsdall, Jr., Philadelphia, PA
Katherine Vig, Columbus, OH
C.B. Preston, Buffalo, NY
Shinkichi Namura, Tokyo, Japan
George Skinazi, Paris, France
Björn U. Zachrisson, Oslo, Norway
<<    
     Article
      >> Home | TOC |          
Index

Seminars in Orthodontics
VOL 8, NO 2 JUNE 2002

Biostatistics for the Orthodontic Clinician

Rose D. Sheats, DMD
Guest Editor
CONTENTS

Editorial 51
P. Lionel Sadowsky

Introduction 52
Rose D. Sheats

Statistics as a Second Language: A Brief Overview for the Wary Clinician 54

Lysle E.Johnston

Understanding Distributions and Data Types 62

Rose D. Sheats and V. Shane Pankratz

Sample Sizes and Power: What Is Enough? 67

Ceib Phillips

Common Statistical Tests 77

Rose D. Sheats and V. Shane Pankratz

Regression: Is Your Guess as Good as Mine? 87

Lysle E. Johnston

Sensitivity, Specificity, and Related Concepts 92

Susan P. McGorray

A Clinical Orthodontist Looks at Statistics 102

Sheldon Baumrind
<<    
     Article
      >> Home | TOC |          
Index

Seminars in Orthodontics
Future Issues

Vol 8 No 3 (September 2002)

RISK ASSESSMENT AND CLINICAL ORTHODONTIC MANAGEMENT
Larry Jerrold, DDA, JD, Guest Editor
Vol 8 No 4 (December 2002)
THE VERTICAL DIMENSION
Timothy E. Wheeler, DMD, Guest Editor
Vol 9 No 1 (March 2003)
MOUNTING OF CASTS IN CLINICAL ORTHODONTICS
Richard Kulbersh, DMD, Guest Editor

Recent Issues

Vol 8 No 1 (March 2002)

CLINICAL UPDATE ON TECHNOLOGICAL ADVANCES IN ORTHODONTICS AND DENTOFACIAL ORTHOPEDICS
Gregory]. King, DMD, DMSc, Guest Editor
Vol 7 No 4 (December 2001)
THREE-DIMENSIONAL DIAGNOSIS AND TREATMENT IN ORTHODONTICS
Sheldon Baumrind, DDS, MS, and Robert L. Boyd, DDS, MEd, Guest Editors
Vol 7 No 3 (September 2001)
TOPICS IN BIOMECHANICS
Stanley Braun, DDS, MME, Guest Editor
Vol 7 No 2 (June 2001)
THE ALEXANDER DISCIPLINE
R.G. Alexander, DDS, MSD, Guest Editor
Vol 7 No 1 (March 2001)
CLINICAL BIOMECHANICS
Steven J. Lindauer, DMD, MDSc, Guest Editor
Vol 6 No 4 (December 2000)
PSYCHOLOGIC ISSUES RELATED TO ORTHODONTIC TREATMENT AND PATIENT COMPLIANCE
Pramod K. Sinha, DDS, BDS, MS, Ram S. Nanda, DDS, MS, PhD, and Roger B. Fillingim, PhD, Guest Editors
Vol 6 No 3 (September 2000)
BIOLOGY OF ORTHODONTIC TOOTH MOVEMENT: CLINICAL IMPLICATIONS
Bhavna Shroff, DDS, MDentSc, Guest Editor
Vol 6 No 2 (June 2000)
MOLAR DISTALIZATION
George J. Cisneros, DMD, MMSc, Guest Editor
Vol 6 No 1 (March 2000)
OBJECTIVES-DRIVEN ORTHODONTICS: EFFECTIVENESS OF MECHANOTHERAPY
Cyril Sadowsky, BDS, MS, Guest Editor
<<    
     Article
      >>
Editorial
When selecting topics for the journal, it is
always a temptation to satisfy our desire
for those areas that we are most comfortable
with and in clinical orthodontics specifically is-
sues of how to. We want to improve our clinical
skills and knowledge and get pearls of wisdom or
a take-home message. Perhaps what is more im-
portant is whether the pearl or the message has
a scientific basis. When mentioning the subject
of statistics, the responses range from rolling
one's eyes to quotes that include Disraeli's
"There are three kinds of lies—lies, damned lies,
and statistics,"1 or Andrew Lang's "He uses sta-
tistics as a drunken man uses lamp-posts—for
support rather than illumination."2 Indeed, sta-
tistics is an area that is immensely complicated
and a subject that most clinically oriented stu-
dents do not relish.
And yet how do we improve our clinical skills
and knowledge and on what do we base our
decisions? Whose directions are we to follow?
Whose teachings and advice should we heed? It
would seem logical that if the specialty of orth-
odontics is to advance and improve, then we will
need to make great efforts to apply the scientific
method in as many ways as possible. Most assur-
edly we provide an excellent service to our pa-
tients, and we can all list a number of advantages
of orthodontic treatment. However, it is incum-
bent on us as orthodontic clinicians to con-
stantly assess, challenge, and debate diagnostic
and treatment regimens advocated and to seek
validation, as far as is possible, for the informa-
tion provided.
Hence, it becomes essential for orthodontic
Copyright 2002, Elsevier Science (USA). All rights reserved.
doi:10.1053/sodo.2002.125161
Seminars in Orthodontics, Vol 8, No 2 (June), 2002: p 51
Home | TOC |          
Index
clinicians to be able to understand and appreci-
ate scientific method and its application in the
field of orthodontics and at least to attempt to
base our diagnosis and treatment on sound re-
peatable evidence. Statistics is a vital tool in this
endeavor and although we as clinicians cannot
expect to understand all of the intricate details
of statistical method and application, we should
at least be familiar with statistical applications
and their significance. We should be wary of
those who warn us "not to cloud their conclu-
sions with the facts."
It is my hope that as you read this issue of
Seminars in Orthodontics you will reinforce your
ideas about the essential elements of statistics
and statistical method and appreciate the excel-
lent efforts of the contributors who have labored
long and hard to make this complex area more
easily digestible. An understanding will help us
appreciate what is scientifically sound, what is
still opinion, and what seems to be effective even
though we do not yet have the data to corrobo-
rate the reasons for the outcomes. We will be
more selective in our following unproven or
even partially proven advice. We will develop a
healthy skepticism while still appreciating the
contributions of those who as yet have no scien-
tific proof.
P. Lionel Sadowsky
Editor
References
1. Stevenson B (ed): The Home Book of Quotes (ed 10).
New York, Dodd, Mead & Company, 1967
2. Andrews R, Biggs M, Seidel M, et al (eds): The Columbia
Book of Quotations. New York: Columbia University
Press, 1996
51
<<    
     Article
      >>
Seminars in Orthodontics
VOL 8, NO 2
Introduction
I accepted the task of guest editing this issue
of Seminars in Orthodontics with trepidation.
Dr Sadowsky asked me to assemble a cadre of
clinician-scholars who were not only knowledge-
able about statistics and could relate key princi-
ples to clinical orthodontic practice but who
could also present this material in a clinician
friendly manner. It was a daunting challenge. My
greatest fear was that another dry treatise on
statistics would emerge from the endeavor, and I
immediately sought to identify contributors who
could meet the aforementioned expectations.
All contributing authors are either orthodon-
tists with a measure of expertise in statistics, or
statisticians with a measure of expertise in clini-
cal orthodontics. The intent of this issue of Sem-
inars in Orthodontics is not to teach you to be-
come statisticians or to attempt to present a
comprehensive review of statistical concepts.
Rather I encouraged the authors to address se-
lected topics that they and I hope will be most
beneficial to you as a practicing clinician.
This issue leads off with an article by Dr Lysle
Johnston who, in his eloquent and inimitable
manner, points us in the proper direction. His
discussion of basic statistical concepts is lucid,
relevant, and, dare I say, even humorous at
times. His first article stresses the opportunities
to enhance one's clinical practice by gaining a
familiarity with statistical analysis to critically
evaluate clinical research. The issue closes with
an essay by Dr Sheldon Baumrind that addresses
some of the very same concepts with another
spin. His provocative presentation also focuses
on the application of outcomes from patient-
oriented studies to the individual patient and
discusses some limitations of statistical analysis
of clinical research.
In between these two articles you will find
contributions that elaborate on many of the con-
cepts touched on by Drs Johnston and Baum-
rind. Dr Ceib Phillips tackles the question of
how large a sample size is adequate. As the sta-
tistical consultant for the University of North
52 Seminars in Orthodontics, Vol 8, No 2 (June), 2002: pp 52-53
Home | TOC |          
Index
JUNE 2002
Carolina's prospective study on early treatment
of class II malocclusions, she has been intimately
involved with orthodontic studies and clinical
trials. Her knowledge and appreciation for clin-
ical orthodontics make her a natural choice for
contributing to this issue. Dr Phillips' careful
and detailed discussion in this issue of the im-
pact of different considerations that influence
sample size calculations will help clarify some of
the mystery associated with determining sample
size.
Dr Shane Pankratz (another bona fide statis-
tician) and I review some basic statistics in two
articles, each with a different emphasis although
both address topics that are difficult to separate.
If you end up jumping back and forth between
the two articles, direct your complaints to me
and not to my esteemed colleague.
Dr Johnston devotes a second article solely
to regression analysis, and I was thankful the
"muses moved him" as I had already felt fortu-
nate that he had initially agreed to one contri-
bution! I had been unsuccessful in persuading
various talented colleagues to write this article,
and as this issue was progressing, I attended a
state society meeting where Dr Johnston was the
featured speaker. In presenting data on a clini-
cal question of interest, he slipped in a discus-
sion of regression analysis so effortlessly that I
doubt that the practitioners even knew they were
getting a tutorial on this important statistical
procedure. His skill at this accomplishment so
impressed me that I prevailed on him at the
meeting luncheon to just convert all those slides
into a "quick little paper." Of course, neither he
nor I were fooled by my simplistic request, and
you can now appreciate why he answered me
with the "muse" quote above.
Dr Sue McGorray presents an important con-
cept in her discussion of the sensitivity and spec-
ificity of statistical tests. Like her counterpart at
the University of North Carolina, she has long-
standing experience in analyzing data for the
University of Florida's randomized clinical trial
<<    
     Article
      >>
Introduction
on early treatment of class II malocclusions. It
always amazes me how well she can clarify the
clinical questions and reduce the mass of data
collected in large studies to summary statistics
and explanations that are actually comprehensi-
ble to us clinicians. The concept of sensitivity
and specificity is important to appreciate, and I
was pleased she chose to address this issue.
I suspect I speak for all the authors when I say
that it was a challenge to decide how to limit the
material each of us discussed and how best to
show it with meaningful clinical examples. As Dr
Baumrind points out, statistics is an enormously
complex field. In this issue, introductory statisti-
cal concepts are presented and often repeated
among the articles. I have chosen to accept the
overlap in material because I believe each con-
Home | TOC |          
Index
53
tributor brings a perspective that not only un-
derscores the many facets of statistical analysis
but also may potentially resonate with different
readers. I firmly believe in the teaching concept
of "telling it three times" to make sure the point
is made (ie, tell them what you're going to tell
them, tell them, and tell them what you told
them). I think our various essays complement
each other well in this regard. My hope is that
each of you will value the contributions of the
authors and will come away with a greater appre-
ciation for the role of statistics in your clinical
practice as you strive to separate orthodontic
fact from orthodontic fiction.
Rose D. Sheats, DMD, MPH
Guest Editor
<<    
     Article
      >> Home | TOC |          
Index

Statistics as a Second Language: A Brief

Overview for the Wary Clinician
Lysle E. Johnston, Jr

Orthodontic effects are for the most part gross and measurable. Thus, of all
of dentistry's specialists, it should be easiest for us to deliver optimal,
evidence-based treatments. Unfortunately, the profusion of disparate meth-
ods and philosophies that compete for our attention argues that, for some
reason, our penchant for measuring things has come to naught. Part of the
difficulty stems from a generalized unfamiliarity with the rules—the statis-
tical rules—that govern the process of drawing rational inferences from
numerical data. To assist those who would do better, the present report is
offered up as a relatively nontechnical, jargon-free overview of the basic
methods of statistical analysis and the reasoning behind their application to
the process of clinical decision making. (Semin Orthod 2002;8:54-61.) Copy-
right 2002, Elsevier Science (USA). All rights reserved.

We are said, generally by the editors of pro-
fessional journals, to have entered an era
of evidence-based dentistry. Orthodontics, how-
ever, does not seem to be part of the trend. In
the marketplace of orthodontic ideas, much that
is bought and sold is little more than wishful
thinking shaped by the biases of referring den-
tists, many of whom have a profound distrust
both of orthodontics and of the people who
practice it. How can this be? Orthodontics is a
learned specialty peopled by the best in den-
tistry. Clearly, something has gone wrong.
At least part of the problem stems from the
difficulty of keeping up with a literature that
often can neither be understood nor evaluated
without reference to the methods of statistical
inference. Like it or not, statistics has become
the second language of dental science. To be
unfamiliar with its concepts and techniques is
to be functionally illiterate. Professionals thus
From the Department of Orthodontics and Pediatric Dentistry,
School of Dentistry, The University of Michigan, Ann Arbor, MI.
Supported in part by NIDCR grant DE08716.
Address correspondence to Lysle E. Johnston, Jr, DDS, PhD, FDS
RCS(E), Department of Orthodontics and Pediatric Dentistry, School
of Dentistry, The University of Michigan, Ann Arbor, MI 48109-
1078.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0802-0003$35.00/0
doi:10.1053/sodo.2002.32072
handicapped are the rightful prey of the entre-
preneur. It need not be so.
Some phenomena are perfect and invariant.
Given a temperature in Fahrenheit, there is but
one corresponding value on the Celsius scale. If
you know one, you know the other. In the world
of clinical orthodontics, however, things are not
determined by simple functional relationships.
Instead, a given outcome is the result of many
factors, most of which are unknown. The result
is variation. From patient to patient, a single
treatment never produces the same result; when
compared, the effects of two or more treatments
are never exactly the same. Our problem there-
fore reduces to a reasoned analysis of this inev-
itable variation: is it due to chance or does it
signal a real effect—a difference that might be
of clinical significance? Unfortunately, we must
make these decisions based on incomplete data,
hence the need for that dreaded word, statistics.
We can never look at every patient who has
ever been treated or who ever could be treated.
If we could examine an entire population, deci-
sions would be a matter of simple, albeit infi-
nitely cumbersome, arithmetic. Instead, we must
make do with a relatively small portion of the
population, a representative sample from which
we can take measurements. We then use the
resulting data, not only to describe the outcome
of the study, but also to determine the likelihood

54 Seminars in Orthodontics, Vol 8, No 2 (June), 2002: pp 54-61

<<    
     Article
      >>
Statistics as a Second Language
that it reflects some real property of the popu-
lation, rather than just random variation. It is
this random noise that makes statistical infer-
ence a necessity.
Unless vetoed by the physical laws of the uni-
verse, practically any result, no matter how im-
pressive, can occur purely by chance. There is a
small but finite chance that one monkey of many
sitting at infinite computers will type the Gettys-
burg Address. To an American who is familiar with
Lincoln's words, this monkey's writings might
seem an impressive achievement. To the statisti-
cian, however, it merely means that if you have
enough random events to choose from, an out-
come like this is a certainty. Most of the time,
even the smartest monkey's literary output usu-
ally will be something in the way of "hhs/?jslhpi
v %guwgo nuxouo." Similarly, there is a small,
but finite, chance that a fair coin will come up
heads 100 times in a row. Have we witnessed the
equivalent of a miracle? Perhaps, but a simpler
and much more likely explanation would be that
the coin is not fair and that instead it has two
heads. To interpret the results of any given study
therefore one must come to grips with the cen-
tral issue of statistical inference: what is the prob-
ability that a given outcome could have hap-
pened purely by chance?
If the probability is high, we conclude that
nothing much is going on; if the probability is
small, we say that a statistically significant event
has occurred—one that, although possible,
clearly is not probable. How then do we estimate
the probability of an event? This is when statis-
tics comes to the rescue. For our purposes, sta-
tistics is merely a set of common-sense rules that
permit us to decide whether something out of
the ordinary has occurred. As such, it is a key
part of the process by which biomedical knowl-
edge is generated. All should have at least an
intuitive grasp of the subject.
Admittedly, achieving a basic understanding
of the process of statistical inference is not a
trivial undertaking. On the other hand, statistics
is a good deal more useful and not nearly as
difficult, or as dry, as many of the subjects we all
have had to master during the course of our
professional careers. Stated simply, an apprecia-
tion of the role of statistics in scientific commu-
nication and a familiarity with its seemingly ar-
cane terminology are within the grasp of every
Home | TOC |          
Index
55
orthodontic specialist. If treatment is to be ratio-
nal and evidence based, there is no other way.
The purpose of this article is to provide a
survey of the statistical tools and techniques on
which meaningful professional survival depends
including (1) descriptive statistics, mean, stan-
dard deviation, and so on; (2) inferential statis-
tics, t tests and analysis of variance (ANOVA) to
test hypotheses about between- and among-treat-
ment mean differences; (3) analysis of associa-
tion, correlation and regression to examine the
strength and form of the linear relationship be-
tween two characteristics; and (4) nonparamet-
ric statistics to examine treatment effects when a
normal (ie, symmetrical), bell-shaped, distribu-
tion cannot be assumed.
Perhaps equally important, it is hoped that
this survey will permit someone to read the lit-
erature without being intimidated by the termi-
nology. In the end, it is probably more impor-
tant to know what a statistic does than how it
does it.
Descriptive Statistics
The goal of most investigations is to learn about
some population (Class II malocclusions, high-
angle cases, growth modification treatments,
and so on) in the hope that this knowledge will
aid in making individual treatment decisions.
Perhaps one of the first things one might want to
do with data from a sample is to describe it.
Short of just listing all the numbers ("Here's the
way my study came out: 14.3, 17.2, 16.8, . . . ."),
statistics provides many ways of presenting a
shorthand description of the numerical out-
come of an investigation. First, we can use some
sort of pictorial representation, such as a pie
chart or a bar graph. Alternatively, we can use a
mathematical description that can also serve as a
first step in the process of statistical inference.
Any measurable characteristic of a population
is called a parameter. In the real world, however,
we can never study the presumably infinite mem-
bers of an entire population. Instead, we must
make do with a sample in the hope that it will
serve as a representative surrogate. Things that
we calculate from this sample are called statis-
tics, hence the name and scope of the subject
and the topic of this article. For purposes of
description, we are interested in the general
<<    
     Article
      >>
56 Lysle E. Johnston, Jr
location (central tendency) and scatter (disper-
sion) of the data along the number line.
Central tendency can be described by calcu-
lating such well-known statistics/parameters as
the mean (the arithmetic average), the median
(the middle datum), or the mode (the most
frequent score). Scatter, in turn, can be de-
scribed by calculating the range or, more often,
by an average squared deviation from the mean
(the variance). The square root of the variance is
called the standard deviation and is perhaps the
most common descriptor of dispersion. The for-
mulae for these statistics are beyond the scope of
this article. Suffice it to say, they provide a mean-
ingful description of the data from a sample (or
from an experiment); however, they also serve as
estimates of the mean and standard deviation of
the population from which the sample was
drawn.
A normally distributed population is defined
completely by its mean and standard deviation;
there is nothing else to know. Many effects that
are of clinical significance (eg, growth, cooper-
ation, relapse, treatment time, and so on) are
thought to have a multifactorial origin, which, in
turn, commonly leads to an approximately nor-
mal distribution. If our goal is to learn about
such a population, the mean and standard devi-
ation are not only useful descriptors but also the
starting point in the process of statistical infer-
ence.
Inferential Statistics
Tests on Means
One or two populations (Student t tests). When
we examine a sample of treatment outcomes, we
really are not interested in this particular group
of patients; their treatment is over. Rather, we
study them to draw inferences about patients in
general. Usually, our goal is to test a hypothesis,
which is some sort of guess, conjecture, or rea-
soned assumption about the mean of one or
more populations. These conjectures take the
form of null hypotheses, which are simple de-
clarative statements (eg, extraction and nonex-
traction treatment have the same effect on the
profile, heavy forces and light forces cause the
same amount of root resorption, sliding me-
chanics and sectional mechanics retract canines
equally rapidly, and so on). Although we may
Home | TOC |          
Index
suspect that each of these assertions is untrue,
the neutral form of the null hypothesis (same,
equally rapidly, and so on) emphasizes our will-
ingness to suspend judgment and to be guided
by data rather than by preconceived notions.
To test a null hypothesis, the first step is to
gather a representative sample that can be used
to estimate the parameters of interest. The sam-
ple mean (X) is a good estimate of the popula-
tion mean; the sample standard deviation (S) is
a good estimate of the population standard de-
viation. Good, however, does not mean perfect.
Each population features a certain level of in-
trinsic variability. Depending on sample size, any
statistic we might calculate will tend to reflect
this scatter and thus will differ somewhat from
the population parameter we are trying to esti-
mate. In the present context, good means
(among other things) unbiased: if we were to
repeat the study over and over and over, our
statistics would vary from sample to sample, but
their average would tend to approach exactly the
parameter they estimate. (In contrast, a biased
estimate would tend in the long run to over-
shoot or undershoot the parameter it purports
to estimate.) Therein lies the rub: sampling vari-
ation makes any outcome at least remotely pos-
sible; reality, however, makes some results so
improbable that we are obliged to seek alterna-
tive explanations.
The process of hypothesis testing therefore
involves estimating the probability of obtaining a
given outcome if the null hypothesis is abso-
lutely true. If, under these circumstances, an
outcome is unlikely, we reject (ie, disbelieve)
our null hypothesis. A simple example of what
commonly is called the Student t test can serve as
an example. To test the venerable assertion that
the mandibular incisor to mandibular plane an-
gle should be 90°, pretend that you gather a
presumably representative sample of patients.
Given that we know the sample size (N) and can
calculate an estimate of the intrinsic variability
within the population (commonly, the sample
standard deviation, S), it is possible to calculate
a statistic, the standard error (S/N/7V), which
describes the variability of means calculated
from such a sample. If the sample mean turns
out to be 100°, we need to calculate the proba-
bility of such an outcome if the true population
mean really is 90°. To do so, we standardize the
deviation from hypothesis by expressing it in
<<    
     Article
      >>
Statistics as a Second Language
standard errors (10° divided by S/\/N). If our
sample's mean differs from the hypothesized
mean by more than two or three standard errors
(the exact number is determined with reference
to a table of percentiles for one of the so-called
t distributions), the probability of such an event
is so low that we are forced to doubt our null
hypothesis. But how low does the probability
have to be for us to draw such a conclusion?
Purely by convention, the cutoff is usually set
at less than 1 chance in 20 (significant) or less
than 1 chance in 100 (highly significant), hence
the common expressions P < .05 or P < .01,
which are used to characterize the strength of
the rejection. On the other hand, if the proba-
bility is higher (commonly, P > .05), we fail to
reject. In effect, we reserve judgement—the hy-
pothesis may be false (if you measure carefully
enough, all null hypotheses are false) but not
false enough to be of concern, hence the com-
mon use of the term fail to reject rather than
accept. In this context, it is important to intro-
duce the concept of statistical power.
The process of statistical inference can result
in two kinds of error, both of which flow from
unavoidable sampling variation: we can reject a
null hypothesis that is true (a type I error) or we
can fail to reject a null hypothesis that is false (a
type II error). Some studies are so crude that no
matter how false the null hypothesis, the test
would be unable to detect it. Thus, whenever
there is a failure to reject (ie, a negative result),
one must always ask how much of a deviation
from the stated hypothesis the test could have
detected. In other words, if some competing
alternative were true, the null hypothesis would
be false. Under this circumstance, what is the
probability that the test would have led to the
correct decision, namely to reject the hypothe-
sis?
The ability to reject a false null hypothesis is
called power and is analogous to the resolving
power of a microscope. Power depends on the
inherent variability of the populations from
which we have drawn our sample, the size of the
sample, the level of significance at which the test
was run (ie, the probability that we will reject
even if the null hypothesis is true), and the
discrepancy between the null hypothesis and
some competing alternative about which we are
concerned.
For example, some say that the interincisal
Home | TOC |          
Index
57
angle should be 130°; others argue that it should
be 135°. If we conduct an investigation and fail
to reject the null hypothesis that it equals 130°,
we would like to be reasonably certain that we
would have rejected 130° if, in fact, the true
population value is 135°. Ideally, a research pa-
per will provide an estimate of the study's
power—commonly, the size of the deviation
from hypothesis (in millimeters or degrees) that
a given test has a 90% chance of detecting. In
this context, detection is achieved when a false
null hypothesis is rejected.
The t test is appropriate to tests of hypotheses
about either one population (that the mean
equals some specified value) or two (that the
means of two populations are equal). Often,
however, one may want to test the hypothesis
that three, four, or even more populations have
equal means. It would be possible to use nests to
compare sample means two at a time. Unfortu-
nately, as the number of populations increase,
the number of pair-wise t tests increases even
faster. Because each individual test has a certain
fixed chance of wrongly rejecting a true null
hypothesis (the cutoff point chosen by the inves-
tigator, 1 in 20, 1 in 100, 1 in 1000), a prolifer-
ation of tests makes it progressively more likely
that one or more tests will lead to rejection of
the null hypothesis purely by accident. The re-
sult is an uninterpretable smattering of signifi-
cant outcomes that are in reality just type I er-
rors. Given this problem, it is more appropriate
to choose a method that can test for among-
means equality in a single test and at a single,
fixed level of significance. In the jargon of con-
temporary experimental design, the appropriate
method is some sort of analysis of variance
(ANOVA).
Three or more populations (ANOVA). Within
any given population, there is an intrinsic level
of apparently random variability. Not all edge-
wise patients turn out the same. Not all two-stage
treatments progress at the same speed. Not all
Class II patients have small mandibles. Such vari-
ability commonly is described by the standard
deviation, which, for the purposes of tests on
multiple means, is assumed to be the same in
each of the various populations under compari-
son. Not only does this intrinsic variation serve
to ensure that sample means drawn from several
populations will differ but also to determine by
roughly how much they should differ.
<<    
     Article
      >>
58 Lysle E. Johnston, Jr
ANOVA is used to decide whether three or
more sample means vary more than is likely if
the populations from which they were drawn
really do have equal means. Rejection of the null
hypothesis of equal means implies that two or
more probably differ by more than can be ex-
plained easily by chance. In this event, it is usual
to use some sort of supplemental, after-the-fact
test to decide which of the means are responsi-
ble for the significant outcome. Common exam-
ples are the methods of multiple comparison:
the Fisher exact test (least significant differ-
ence), Tukey test ("honestly significant differ-
ence"), Scheffe test, Duncan multiple range test,
Dunnett test, and others. For both ANOVA and
the various post hoc comparisons, the reasoning
is much the same as for simple t tests. To a first
approximation, they differ only in terms of the
test statistic we calculate from the data and the
table of probabilities (ie, the distribution) we
use to look up the likelihood that it came about
by chance.
In its various forms (randomized block, facto-
rial, repeated measures, Latin squares, and the
like), ANOVA is the centerpiece of modern sta-
tistical inference and is a backbone of contem-
porary experimental design. Not all questions,
however, involve differences between or among
means; often we wish to examine the joint vari-
ation of two or more characteristics.
Analysis of Association
Some treatments relapse, and some are stable.
Some roots resorb, and some do not. Some pa-
tients grow well, and some do not. Much of what
passes for diagnosis and treatment planning in
orthodontics involves an attempt to account for
this variability by generating some sort of predic-
tion. Commonly, these predictions are based on
a presumed association between various pre-
treatment dentofacial characteristics and the de-
tails of treatment on the one hand and impor-
tant outcome variables on the other. We assume,
for example, that changes in intercanine width
or in the angulation of the lower incisors are
related to subsequent stability. Similarly, force
magnitude and the shape of the roots are
thought to predict subsequent root resorption,
the shape of the mandible, and subsequent man-
dibular growth. The list of common-sense pre-
dictors is seemingly endless; it is also largely
Home | TOC |          
Index
without justification, either empirical or theo-
retic.
In any consideration of prediction, two ques-
tions should come immediately to mind: (1) is
there really a relationship between the things we
are trying to predict and the diagnostic charac-
teristics that we use to predict them, and (2) if
there is, what is the form of the relationship? For
answers to these questions, we look to correla-
tion and regression.
Correlation. To test whether or not two vari-
ables bear a linear relationship to each other (ie,
whether or not they vary together, either posi-
tively or negatively), the technique of Pearson
product-moment linear correlation is commonly
used. Given a pair of characteristics suspected of
being related, investigators commonly gather a
sample and then calculate the correlation coef-
ficient (r), a dimensionless index of the extent
to which the two characteristics vary together.
Although many types of relationship (logarith-
mic, exponential, and so on) can be modeled, it
is usual in orthodontics to test for a significant
linear correlation. The resulting correlation co-
efficient can range from +1, denoting a perfect
positive relationship, to — 1, characteristic of a
perfect negative relationship (if one variable is
big, the other is small, and vice versa). Under
the assumption of linearity, r = 0 would signify
complete independence.
Obviously, it is unlikely that data obtained
from any population or from any sample that
might be drawn from it would show either a
perfect correlation (r = +1 or —1) or complete
independence (r = 0). Instead, the correlation
coefficients that we calculate almost always fall
somewhere in between (0.6 or -0.3 or 0.1) and
thus must be subjected to the standard litmus
test of inferential statistics: could a coefficient of
this size come about easily by chance? As with
many other statistical tests, the decision is based
on the size of the deviation from hypothesis
(usually, how much r differs from zero) ex-
pressed in standard errors. Once again, the bor-
derline between chance and reality varies with
the size of the sample and the chosen level of
significance (1 in 20, 1 in 100, and so on).
Correlation, it should be noted, is easily abused
and often misinterpreted. Two cautionary notes
are in order.
First, it must be emphasized that statistical
significance is in part a function of sample size.
<<    
     Article
      >>
Statistics as a Second Language
Thus, given a large sample, a correlation can be
statistically significant (ie, unlikely to have come
about by chance) and yet be so small that knowl-
edge of the underlying relationship it depicts
would be of little or no use clinically. Second, no
matter how high or how statistically significant
the correlation, one cannot infer that one vari-
able causes the other. Linear correlation merely
tests whether two characteristics vary together;
causation must be inferred from our knowledge
of the biology of the system. The growth of the
mandible is correlated with the growth of the big
toe. One, however, does not cause the other.
Instead, both are byproducts of the general pro-
cess of somatic growth. In such a situation, the
correlation would be strong but of little practical
or theoretic significance.
In contrast, there are situations in which the
mere existence of a relationship is all we care
about. For example, we might wish to examine
the possibility of polygenic inheritance by testing
whether the correlation between facial dimen-
sions of siblings is about 0.5. Usually, however,
our ultimate goal in studying correlation is to
find a relationship (causal or otherwise) that is
strong enough to be used clinically. Can we use
a characteristic that we can see (the indepen-
dent variable) to predict some important char-
acteristic or outcome that we cannot (the depen-
dent variable)?
Some treatments relapse, and some are sta-
ble. Some roots resorb, and some do not. It
would be useful to know in advance which out-
come is more likely. Indeed, the process of di-
agnosis and treatment planning is replete with
references to relationships said to be of predic-
tive significance. "Them that has gets." "I looked
at the mother and I looked at the father and I
knew that she was going to be a bad grower."
"He was a hyperdivergent patient with a poor
growth pattern." Talk, however, is cheap. A pre-
cise characterization of the strength and form of
the relationship (if any) is usually beyond the
intuition of even the most perceptive of clini-
cians. Once again, statistics can be of assistance.
Regression. If a linear relationship is signifi-
cant statistically and is strong enough to be of
practical use, the next step is to model it math-
ematically in the form of a prediction equation
so that it can be used clinically. For example, the
size in millimeters of the unerupted mandibular
buccal segments can be estimated by way of a
Home | TOC |          
Index
59
simple linear equation, 10.5 plus half the width
of the permanent lower incisors. Linear regres-
sion commonly is the process by which this mod-
eling is accomplished.
As with other statistical analyses, the first step
is to gather a sample of bivariate data (eg, height
and weight, mandibular plane angle and change
in face height, and so on). Given these data, the
next step is to infer the best possible prediction
equation, which is usually, but not always, a lin-
ear prediction equation. If the dependent vari-
able (Y, the characteristic you wish to predict) is
plotted on the vertical axis of a scattergram, and
the independent variable (X, the characteristic
on which the prediction will be based) is plotted
on the horizontal axis, the methods of regres-
sion are used to calculate a best-fit linear predic-
tion equation of the form, Y = A + $X, in which
A is the Y-intercept, the point at which the line
crosses the ordinate, and B is its slope (rise/run,
increase in Y for a unit increase in X).
If there is a significant linear relationship be-
tween the dependent and independent vari-
ables, the slope of the line will differ significantly
from zero. The prediction error, the scatter of Y
around the regression line, will be less than the
scatter around the overall mean (our best single
prediction in the absence of an equation). As
has been noted, a relationship can be statistically
significant (ie, unlikely to have occurred by
chance) without being clinically significant. In
other words, we always have to look beyond the
initial threshold of statistical significance to de-
cide whether or not a prediction equation is
actually worth using clinically.
Regression and correlation are closely related:
one deals with the strength of a linear relationship
and the other with its form. Indeed, the correla-
tion coefficient squared (r2) can be derived from
the regression analysis and used as an estimate of
the proportion of the variability in Y that can be
eliminated (accounted for) by using the predic-
tion equation. Be warned, however, that r2 has to
be quite large (eg, 0.5-0.6) before the equation
would be of obvious clinical use. In orthodontics, I
can think of few, if any, prediction schemes for
which this criterion is met.
Nonparametric Statistics
The statistical tests described thus far are com-
monly said to be parametric because their null
<<    
     Article
      >>
60 Lysle E. Johnston, Jr
hypotheses deal with population parameters: the
mean incisor to mandibular plane angle is 90°,
the mean growth rate is the same in Class I and
Class II populations, on average three treat-
ments produce the same result, and the correla-
tion between mandibular plane angle and the
pattern of growth is zero, and so forth. Their
ability to generate a valid probability estimate
depends on the truth of a number of assump-
tions including normality, equal variance, and
ratio-scale measurements. Occasionally, these as-
sumptions are not warranted, in which case it is
necessary to use one of the so-called nonpara-
metric, distribution-free statistics. These x2 tests
constitute a comprehensive parallel universe in-
habited by nonparametric versions of most of
the common parametric analyses. Not only are
they insensitive to departures from the funda-
mental assumptions of parametric statistical in-
ference, but also they are nearly as efficient.
That is, for a given sample size, their power (the
ability to detect a false null hypothesis) is almost
as good as that of their parametric counterparts.
When there is doubt about the propriety of a t
test or ANOVA, the various nonparametric tests
are safe, effective alternatives.
Envoi
It is common to argue that the subject of statis-
tics is artificial and irrelevant because it deals
with large samples, whereas treatment, by defi-
nition, involves one patient at a time. Admit-
tedly, this assertion has the ring of common-
sense truth. But what are we to conclude? That
nothing can ever be known? That a clinical in-
vestigation becomes useless academic obfusca-
tion when its sample size goes beyond N = 1?
The one-patient-at-a-time argument seems a rea-
sonable substitute for careful thought until you
realize that the total sample in any given practice
adds up, one at a time, to thousands. Given
thousands of repetitions, a few percentage
points can make a casino rich and its patrons
poor. Why should it be any different in clinical
orthodontics, an activity that often resembles a
game of chance against nature? In the end, an
estimate of the mean response (or perhaps an
interval—a confidence interval—that has a cer-
tain likelihood of containing the true mean) is
probably the best we can ever do.
For a variety of reasons, we probably never
Home | TOC |          
Index
will be able to predict exactly how a given indi-
vidual will grow or cooperate or respond to treat-
ment. We can, however, hope to know how sim-
ilar patients would respond by examining the
average response of progressively more precisely
defined subpopulations (eg, boys, 8-year-old
boys, 8-year-old boys with Class II malocclusions,
8-year-old boys with Class II malocclusions and
long faces, 8-year-old boys with Class II maloc-
clusions and long faces who have small teeth,
and so on). Please note, however, that once we
run out of meaningful pigeon holes (sex, age,
malocclusion, face height, tooth size), our best
bet is an estimate of the mean of what is left, the
average result for a sample of 8-year-old boys
with Class II malocclusions, long faces, and small
teeth.
At this stage, many workers posit clinical rules
of thumb that purport to do better than just
betting on the mean. Unfortunately, it is easy to
prove mathematically that, if clinical rules do
not contribute additional information (ie, some-
thing that would place the patient in a smaller,
more exclusive pigeon hole) and if they lead to
forecasts that differ from the Class mean, they
will on average serve only to increase our overall
prediction error. Thus, although we may always
have to expect/bet on the mean, it can be the
mean of so select a subpopulation that it can be,
in effect, an efficient individual prediction. It
seems to me that this filtering process is a rea-
sonable goal for contemporary orthodontic clin-
ical research. An additional cautionary note is in
order.
Statistics, by definition, deals with samples.
We hope that the samples we read about are bias
free and thus a good approximation of the pop-
ulation we wish to understand. Clinical studies,
however, are notoriously difficult to conduct.
For many of the most interesting questions, ran-
dom allocation of treatments is ethically ques-
tionable, whereas retrospective designs, even
those that seem to feature random sampling,
may be irrevocably tainted by a host of biases.
Although the theory of sampling and experi-
mental design is well beyond the scope of this
article, not to mention the skills of its author, I
would suggest that wary, skeptical clinicians ask a
few simple questions of the literature they read.
If it is a case report, what are you supposed to
infer, that all patients will respond similarly, that
well-trimmed models really look good, that the
<<    
     Article
      >>
Statistics as a Second Language
author is a better orthodontist than you are?
Often, the take-home message is not all that
obvious. If, instead, you are reading a study that
purports to be more than just a case report (ie,
one in which N > 1), you always have to question
the sampling procedure.
Of the thousands of patients who pass
through an orthodontist's practice, how did the
few who figured in the study qualify for inclu-
sion? Often it is because they turned out excep-
tionally well. And why did the treatment work so
well? Commonly, the author of the study would
have you believe that his/her methods were re-
sponsible; however, good results can come about
because the patients were, for some unknown
reason, uniquely susceptible to the treatment
under investigation. For example, some of the
most popular nonextraction philosophies are
just that: more-or-less conventional nonextrac-
tion treatments that work well when applied to
nonextraction patients. If you want to go beyond
the data and apply them to obvious extraction
patients (as you often are tacitly encouraged to
do), that is your business and the patient's re-
gret. In the end, there may be a better chance of
winning the lottery than of being selected to be
the subject of a case report or to be part of the
sample of a clinical study. Suffice it to say, faulty
sampling is just one of a myriad of filters and
biases that can affect the outcome of a study and
distort its interpretation.
Statisticians commonly distinguish between
consumer's risk and producer's risk. With re-
spect to the orthodontic literature, you, the con-
sumer, must be worried that the techniques and
ideas you read about (or listen to on the Chau-
tauqua circuit of orthodontic continuing educa-
tion) are oversold and undertested. The produc-
ers, in turn, commonly are just as worried about
devoting any more time and effort to research
than is needed to convince you of the superiority
of whatever it is they are selling. They want the
benefit of the doubt; more often than we would
like to admit, they get it.
In the end, a passing acquaintance with sta-
tistical inference is a sword and shield for the
skeptical (ie, prudent) professional. In the world
Home | TOC |          
Index
61
of scientific investigation, however, the various
statistics listed here are just tools. There are
many more where these came from. At the risk
of overdoing the second language analogy, sta-
tistical tests are similar to a vocabulary. Words
are important, but they do not automatically add
up to the great American novel. Similarly, in
science, the basic tools of data analysis are im-
portant; however, the design of the experiment
that generates the data is far more important.
Many studies are so poorly conceived that it is
impossible to analyze the data they generate.
Indeed, it is trivial projects and faulty experi-
mental design that so often combine to give
clinical research and statistics a bad name. Given
the evangelical fervor of many who contribute to
our literature, these defects often seem more a
strategy than an accident. As noted by C.E. Ay-
res, "A little inaccuracy saves a world of explana-
tion."1
Ultimately, the problem is not "liars who can
figure" or even researchers who cannot or do
not, but rather a specialty that does not want to
be bothered. Perhaps as a result, contemporary
orthodontic progress is more or less limited to
the materials sciences and is largely driven by
the proprietary efforts of the manufacturers.
Our most interesting questions, however, go well
beyond the choice of brackets, adhesives, and
archwires. What is to be the place of early treat-
ment, gnathology, extraction, expansion, molar
distalization, surgery? The list is very long and
very old.
In the next century, orthodontics will be de-
fined both by its ability and its determination to
answer such questions. An honorable outcome
will require both the generation of valid data
and an ability to interpret it properly. Without a
minimal knowledge of statistics and a respect for
its importance in clinical research, the former is
unlikely and the latter is impossible.
Acknowledgment
The author thanks Dr Charles Kowalski, a real statistician, for
his many suggestions, both technical and stylistic. All were
helpful and all in some form or another made their way into
the final version of this manuscript.
<<    
     Article
      >> Home | TOC |          
Index

Understanding Distributions and Data Types

Rose D. Sheats and V. Shane Pankratz

To comprehend some of the fundamental concepts of statistical analysis, it

is important to appreciate the importance of the distribution of data points
in the sample that is drawn to represent the population. Furthermore, one
must understand the classification of data types. Data type and the distri-
bution pattern of their values influence the choice of appropriate statistical
tests. Emphasis will be placed on the normal, or Gaussian, distribution. This
is an important distribution to understand because the assumption of this
distribution underlies the use of many common statistical tests. The pur-
pose of this article is to familiarize the clinician with some basic biostatis-
tical concepts without delving into the statistical theory or debate associ-
ated with these concepts. There will be no presentation of mathematical
formulas. The interested reader is referred to several texts, which will
provide greater depth and rigor for those seeking additional knowledge.
These texts were chosen specifically for their general readability and clarity
and are referenced at the end of the article. (Semin Orthod 2002;8:62-66.)
Copyright 2002, Elsevier Science (USA). All rights reserved.

L et us begin by describing the types of data

that are collected and analyzed. Data may
be quantitative or qualitative. Quantitative data
measure something with a number. In orth-
odontics, we make myriad measurements. We
quantify such traits as the amount of crowding,
overjet, incisor inclination, and maxilloman-
dibular skeletal discrepancy. However, we also
make qualitative assessments. We record at-
tributes, not meaningfully summarized by a
number, which may influence our diagnostic
and treatment decisions such as the sex of the
patient, judgments regarding the presence or
absence of additional craniofacial growth, sever-
ity of mandibular plane angle (high, normal,
low), likelihood of compliance with headgear or
elastics (yes/no), and so forth.
In biostatistics, data are classified as either
continuous or categorical (Table 1). Continuous
From the Mayo Clinic Rochester, Departments of Dental Special-
ties and Health Sciences Research, Rochester, MN.
Address correspondence to Rose D. Sheats, DMD, MPH, Depart-
ment of Dental Specialties, Mayo Clinic, 200 First Street SW,
Rochester, MN 55902.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0802-0004$35.00/0
doi: 10.1053/sodo. 2002.320 75
data are actual measurements of some variable
such as millimeters overjet, degrees of mandib-
ular plane angle, or megapascals of shear bond
forces. Categorical data are observations that fall
into specified levels such as Angle's classifica-
tion, Adhesive Remnant Index scores,1 or Apical
Root Resorption scores.2
Quantitative data are often continuous but
may be categorical, whereas qualitative data are
always categorical. Categorical data may be fur-
ther classified into ordinal or nominal (Table 1).
Ordinal data have an order to their levels of
assignment, each more (or less) severe than the
previous. The Apical Root Resorption index2 is
an example of categorical quantitative data. It is
measured by an ordinal scale from 1 to 4 that
assigns a severity level to the amount of apical
root resorption that has occurred. Each increas-
ing score level reflects an increasing severity of
apical root resorption. Another example is the
Adhesive Remnant Index,1 which classifies a
tooth according to the amount of adhesive that
remains on the tooth after a bracket is removed.
Each successive score from 1 to 5 indicates lesser
amounts of adhesive left on the tooth.
Nominal data, on the other hand, have no
inherent order to their levels of assignment.
Race and sex are common data that are nominal

62 Seminars in Orthodontics, Vol 8, No 2 (June), 2002: pp 62-66

<<    
     Article
      >>
Understanding Distributions and Data Types
Table 1. Data Types
Data Type Description
Continuous Variables that are measured and can take
on any value along a continuum
Categorical Variables whose values fall into distinct
categories or defined levels
Ordinal Variables for which an order exists in the
levels assigned
Nominal Variables for which there is no
hierarchical order to the category level
in nature. Angle's classification of malocclusion
is another example of nominal data, which do
not arrange themselves into a hierarchy of order
of severity. Angle's Class III malocclusions may
not necessarily be worse than Angle's Class II, or
indeed, some Angle's Class I malocclusions.
The classification of data type is important be-
cause it is a factor in determining the statistical
tests that are appropriate for analyzing the data.
Data types themselves are ordered: continuous
data provide the most information, followed by
ordinal data, and finally by nominal data.3 Some-
times data of a higher level can be rescaled to a
lower level such as when continuous data are cat-
egorized into ordinal or nominal levels.3 An exam-
ple is to group mandibular plane angle cases into
categories using cutoff values of the continuous
measurement in degrees. Categories such as high,
normal, or low angle cases may be created by
grouping them according to their mandibular
plane angle measured in degrees (Table 2). This
continuous variable is rescaled to a categorical vari-
able in which each succeeding category from low
to high represents increasing mandibular plane
angle. Because of this ordered progression, this is
an ordinal variable.
Higher orders of data can be rescaled to
lower orders of data, but the reverse is not pos-
sible. When data are rescaled to a lower order,
information is lost, which influences the power
Table 2. Rescaling Continuous Data to Categorical
Data
Continuous Categorical
MPA < 30° Low angle
30° < MPA < 38° Normal angle
MPA > 38° High angle
Home | TOC |          
Index
63
Examples
Overjet in mm
Mandibular plane angle in °
Shear bond force in megapascals
Adhesive Remnant Index
Apical Root Resorption Index
Race
Sex
Angle's classification of malocclusion
(Class I, II, or III)
of the study to detect differences. There are,
however, instances when it is beneficial to con-
vert continuous data to ordinal or nominal
types. The study statistician will advise when it
may be advantageous to "collapse" the data.
Raw data that are continuous can be summa-
rized by calculating means and standard devia-
tions (SDs), statistics that characterize the sam-
ple and estimate parameters of the larger
population from which the sample was drawn.
Data that are categorical cannot be summarized
by these parameters. Be wary of converting cat-
egorical data into numeric scores and then sub-
jecting them to arithmetic manipulations to ob-
tain means and SDs. Such calculations may be
meaningless (eg, when trying to determine the
mean sex or mean race of a sample). Percent-
ages of observed response categories are useful
summaries of nominal data.
The importance of correctly identifying data
type will be apparent from a subsequent article
in this issue. In brief, its significance lies in the
potential ability to use parametric statistics if the
sample can be summarized with parameters
(means and standard deviations, percentages,
and so on). Nonparametric statistics are avail-
able to analyze data for which the appropriate
parameters cannot be calculated or for those
data that do not meet other requisite assump-
tions for parametric analysis. These concepts will
be addressed further elsewhere in this issue.
Distributions
If you have a collection of data points, begin
your initial analysis by plotting them on a graph
to see how they are distributed. Often these
<<    
     Article
      >> Home | TOC |          
Index

64 Sheats and Pankratz

points can be seen to follow some recognized

pattern or distribution. Many patterns of distri-
butions occur in nature. Frequently, these pat- o
terns can be described by mathematical func-
tions, which then enable you to determine the 0)

likelihood that a data point will fall under a

specific area of the distribution curve. O

Let's say that you place brackets on all your

patients' teeth including first and second molars,
and you would like to know if there is a difference
in the failure rate of brackets according to tooth
type. You examine your records of the last 50
completed patients and plot the number of
bracket failures by tooth type (assume you only
count a tooth once, even if the bracket comes off
more than once). Your data may look like Figure 1.
This is an example of a uniform distribution. Ap-
proximately, the same number of brackets came
off for each tooth type. Regardless of which tooth
type it was (x-axis), the number of failures (y-axis)
was almost the same (uniform).
To see another type of distribution, let's assume
you would like to know at what age the second
permanent molar typically erupts. You have access
to data from a large study of children and plot the
presence of second molars versus the age of the
child. Figure 2 is a hypothetical representation of
how your data may look. You notice two peaks in
your data and, on further analysis, determine that
these peaks correspond to the different ages at
which most girls and most boys have attained their
second molars. This type of distribution, with two
peaks, is known as a bimodal distribution.
o mean value of all the data points. This seemingly
c
8 9 10 11 12 13 14 15 16 17
Age (years)
Figure 2. Bimodal distribution (two peaks): Hypotheti-
cal age at which 2nd molars erupt. The peak on the left
represents mean age of eruption for girls: the peak on
the right represents mean age of eruption for boys.
The Normal Distribution
In many biologic systems, the distribution of
data points for a particular factor (also known as
a variable) often takes, at least approximately,
the form of the normal distribution or Gaussian
distribution. This bell-shaped curve is shown in
Figure 3. Many of you know the bell-shaped
curve from the distribution of scores on a na-
tional examination. It can be seen that the data
cluster around a central point and spread sym-
metrically around this center point. In the nor-
mal distribution, the central point is the mean of
the sample. This central point in a symmetric
distribution such as the normal distribution is
also the median and mode, which are discussed
in greater detail in another article in this issue.
The width of the bell-shaped curve depends
on how much variability there is in the data. One
way to estimate the amount of variability is to
calculate the SD, the square root of the average
squared deviation of each data point from the

o
2 o
CD

M2 M1 PM2 PM1 C I2 l., o-

£
Tooth type

Figure 1. Uniform distribution. Hypothetical num-

ber of bracket failures by tooth type. Abbreviations:
M2, 2nd molars; M1? 1st molars; PM2, 2nd premolars;
PM1? 1st molars; C, canine; I2, lateral incisors; I 1?
central incisors. Figure 3. Normal (Gaussian) distribution.
<<    
     Article
      >>
Understanding Distributions and Data Types
contorted calculation is necessary to account for
deviations both above and below the mean. If
the negative values were not squared, the mean
of the deviations about the mean would always
equal 0. The formula for making this calculation
can be found in any general statistics book and
will not be presented here. Suffice it to say that
the larger the SD is, the greater the variability in
the data. The greater the variability is, the wider
the shape of the curve.
A characteristic of the normal distribution is
that data points that fall under the curve within
1 SD of the mean encompass 68% of all the data.
Thus, the amount of data that fall into the tails
of the curve beyond 1 SD from the mean is
therefore 32% or 16% in each tail because the
curve is symmetric (Fig 4A). The interval de-
fined by the mean ± 2 SD encompasses approx-
imately 95% of the data, with 5% of the data
occurring in the tails or 2.5% in each tail (Fig
4B). Finally, the interval formed by the mean ±
3 SD includes 99.97% of the data.
O"
£ percentage of subjects who have an IMPA
-3 -2 -1 0 +1 +2 +3
Standard deviations from mean
B value of 100° is 2 SD greater than the mean.
{T
£ above the mean. From the normal distribution
-3 -2 -1 0 +1 +2 +3
Standard deviations from mean
Figure 4. Standardized normal (Gaussian) distribu-
tion: areas under the curve. Mean ± 1 SD (A). Mean
± 2SD (B).
Home | TOC |          
Index
65
O
0)
o-
o
80° 85° 90° 95° 100°
IMPA
Figure 5. Hypothetical normal distribution of lower
incisor to mandibular plane angle.
By using such knowledge, one can calculate
the proportion of data points that would fall
under the normal curve above or below the
mean if you know the value of the mean and the
size of the SD. For example, pretend that you
have a sample of 100 untreated subjects, and you
have measured the angle of lower incisor incli-
nation to the mandibular plane (IMPA) in these
subjects. Let's assume that the IMPA is normally
distributed and that the mean inclination in this
sample is 90° with a standard deviation of 5°.
Figure 5 shows what the normal distribution of
these data would look like.
If you want to estimate the proportion or
greater than 100°, how would you calculate this?
It suffices to determine how many SDs from the
mean this specific value is so that you can use
your knowledge of the normal distribution and
proportion of data encompassed by various in-
tervals of the curve. Because the mean is 90° and
the SD is 5° in this example, you know that your
The calculation that you have just made is the z
transformation, an arithmetic technique for con-
verting your data to have a mean of 0 and a SD of
1. Subjects with IMPA > 100° are at least 2 SD
curve, we know that the probability that data will
occur beyond the mean + 2 SD is 2.5% (Fig 4B).
You conclude that the proportion of subjects with
an IMPA > 100° is approximately 2.5% (Fig 5).
What proportion of your sample will have an
IMPA between 85° and 95°? The calculation to
determine the number of SDs above and below
the mean that your specified values fall is again
simple arithmetic. You know that 85° is 5°, or 1
<<    
     Article
      >>
66 Sheats and Pankratz
SD below the mean of 90°, and that 95° is 5°, or
1 SD above the mean of 90°. This interval be-
tween 85° and 95° represents that proportion of
data encompassed by the mean ± 1 SD under
the normal distribution curve. In the normal
distribution, we know that approximately 68% of
the data fall in the interval encompassed by 1 SD
above and 1 SD below the mean (Fig 4A). Thus,
you conclude that 68% of your sample subjects
have an IMPA between 85° and 95° (Fig 5).
Importance of Distributions
Why is it important to evaluate the distribution
of data values? Many statistical tests are based on
parametric assumptions (ie, the data are assumed
to follow a distribution that can be summarized by
parameters) requiring distribution of the data
which is normal (bell-shaped). Many parametric
statistical tests are insensitive to mild departures of
the data from normality, but severe departures
from the normal distribution mandate the use of
distribution-free tests. Such distribution-free tests
are called nonparametric statistics and will be dis-
cussed in greater detail elsewhere in this issue.
Parametric statistics tend to be more powerful
than nonparametric statistics. This means that they
are more likely than nonparametric statistics to
detect a significant significance between samples
when the difference is real, but use of a parametric
test when assumptions are violated is incorrect.
Confirming Normality
How do you know if your data are normally
distributed? You can inspect it visually from your
plot of your data, but it probably comes as no
surprise to learn that there are statistical tests to
check for departure from normality. Tests such
as the Shapiro-Wilk W test can provide informa-
tion to tell you if your data were not likely to
have been sampled from a normal distribution.
If your data are not normal in distribution, it
may be possible to transform your data into a
normal distribution. A log transformation, for
example, may result in a normal distribution of
the transformed data points. Appropriate statis-
tical tests can then be applied to the trans-
formed data although interpretation of the
transformed data may be more difficult. This
level of statistical analysis will most likely require
assistance from a statistician.
Home | TOC |          
Index
Summary
There are many more distributions that are
beyond the scope of this issue and that figure
importantly in statistical analyses. The normal dis-
tribution is one of the most central to statistical
analysis because of its role in the appropriate use
of parametric statistics. The type of data being
analyzed is also important. Continuous data in
general contain more information than categori-
cal data. Of the latter, ordinal data are more de-
sirable than nominal data. The statistical tests se-
lected for data analysis depend on the type of data
available and the distribution pattern of the data.
A review of the most common statistical tests is
presented in another article in this issue. The goal
of this essay was to familiarize the clinician with the
need to comprehend the concepts of data type
and data distribution patterns. Understanding
these concepts will help you in determining if an
appropriate statistical analysis was applied to a
study you find of interest. In this way, you can
decide how confident you are in the conclusions
of the study and judge whether the evidence is
worth considering in future diagnostic or treat-
ment decisions in the care of your individual pa-
tients. Gaining familiarity in the area of statistics
will help you to optimize treatment outcomes
based on knowledge you gain from thoughtful
evaluation of patient-oriented, clinical research.
This practice of striving to provide patients the
best treatment possible using the best available
evidence is the goal of evidence-based medicine.4
References
1. Artun J, Bergland S: Clinical trials with crystal growth
conditioning as an alternative to acid-etch enamel pre-
treatment. Am J Orthod 85:333-340, 1984
2. Malmgren O, Goldson L, Hill C, et al: Root resorption after
orthodontic treatment. Am J Orthod 82:487-491; 1982
3. Riegelman R, Hirsch R: Studying a Study and Testing a
Test (ed 3). Boston, MA, Little, Brown, 1996, pp 259-270
4. Sackett D, Strauss SE, Richardson WS, et al: Evidence-
based Medicine: How to Practice and Teach EBM (ed 2).
New York, NY, Churchill Livingstone, 2000
Suggested Additional References
Brunette DM: Critical Thinking: Understanding and Evalu-
ating Dental Research. Chicago, IL, Quintessence Publishing
Co, Inc, 1996
Dawson-Saunders B, Trapp R: Basic & Clinical Biostatistics.
Norwalk, CT, Appleton & Lange, 1994
Riegelman R, Hirsch R: Studying a Study and Testing a Test.
Boston, MA, Little, Brown, 1996
Sail J, Lehman A: JMP Start Statistics. Belmont, CA, Duxbury
Press, Wadsworth Publishing Company, 1996
<<    
     Article
      >> Home | TOC |          
Index

Sample Size and Power: What Is Enough?

Ceib Phillips

A basic understanding of statistical methodology is essential, both for de-

signing quality research projects and for evaluating the medical literature.
This article deals with the basic principles involved in sample size calcula-
tions and shows the concepts and factors that determine sample sizes for
comparing means, proportions, and time-to-event measures. Special topics
are also discussed including adjustments to sample size calculation and post
hoc analyses of power. (Semin Orthod 2002;8:67-76.) Copyright 2002,
Elsevier Science (USA). All rights reserved.

r\ very researcher who has been involved in a
l J clinical study has an appreciation of the
effort, cost, and often, inconvenience, to both
investigators and study subjects. Even though
investigators often hope for positive findings, in
clinical research true negative findings make
very valuable contributions to clinical practice.
However, false negative findings can occur ei-
ther by chance or because a study is underpow-
ered, which means too few patients have been
studied to allow a clinically meaningful treat-
ment effect or difference to be detected.1 It is
important to realize that designing studies with
inadequate sample sizes may lead to erroneous
results and false conclusions and subsequently to
inappropriate treatment of patients. Such stud-
ies not only drain existing limited resources but
are also unfair to patients who have participated
in them.
Underpowered studies cannot always be
avoided. However, careful sample size calcula-
tions can guide researchers as to what can and
cannot be accomplished in a study with a finite
amount of resources. In fact, National Institute
of Craniofacial Research guidelines for clinical
trials2 now mandate the inclusion of such infor-
mation on all clinical trial applications.
From the Department of Orthodontics, School of Dentistry, Uni-
versity of North Carolina, Chapel Hill, NC.
Supported in part by NIH DE NIH DE 10028 and DE05215.
Address correspondence to Ceib Phillips, MPH, PhD, Depart-
ment of Orthodontics, CB7450 UNC-CH, Chapel Hill, NC 27599.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0802-0005$35.00/0
doi:10.1053/sodo.2002.32074
Sample size calculations are made using vari-
ous assumptions about the anticipated treat-
ment effect or differences between treatments
together with realistic projections concerning
patient accrual and follow-up. Calculating a sam-
ple size requires four things: (1) deciding on the
design of the study; (2) assessing the availability
of resources; (3) specifying distribution assump-
tions; and (4) perhaps most challengingly, de-
fining a clinically relevant effect. Table 1 pro-
vides a set of questions that can help organize
the information needed to calculate an appro-
priate sample size.
The statistical methodology for calculating
sample size has been extensively developed over
the years. Although the sample size calculations
are performed using mathematical methods, the
preparation for the calculation requires both
statistical reasoning and clinical experience. De-
signing Clinical Research3 and Statistics4 are two
introductory textbooks that provide a basic con-
ceptual overview of sample size and power con-
cepts. In this article, the basic principles under-
lying the sample size calculation are reviewed,
and examples of sample sizes are provided.
Three kinds of data will be presented with a
discussion of how each kind influences sample
size calculations. The three kinds of data are: (1)
nominal data, information collected in studies
that can be classified by categories such as man/
woman or Class I/II/III (Angle's classification);
(2) continuous data, information collected in
studies that can be measured on a continuous
scale, such as degrees (incisor inclination, man-
dibular plane angle), millimeters (overjet,
length of mandible), or grams of force (head-

Seminars in Orthodontics, Vol 8, No 2 (June), 2002: p 67-76 67

<<    
     Article
      >>
68 Ceib Phillips
Table 1. Key Questions Before Sample Size
Calculation
1. What is the primary outcome? Clinicians frequently include
multiple outcomes in clinical studies. This creates a
dilemma for calculating a sample size (n) because the n
required will vary from outcome to outcome. One or
two outcomes should be designated as the primary basis
for sample size determinations. Other outcomes
typically are considered for secondary analyses.
2. What's the scale of measurement for the outcome? Outcomes
can be classified as binary or dichotomous (yes/no,
present/absent), categorical (mutually exclusive
categories), ordinal (ranked or rated), or continuous
(measures on which arithmetic operations can be
performed). In general, binary and categorical
measures provide less information and thus lower
power, requiring larger sample sizes.
3. What's the variability of the outcome? For binary measures,
the variability is directly related to the proportion of
positives (yes or present responses). For continuous
measures, the sample size required increases as the
standard deviation increases.
4. What is the desired level of significance and power? Typically,
level of significance is set at 0.05 or 0.01, whereas power
is set at 80% or 90%. Increasing the power or
decreasing the level of significance increases the sample
size required.
5. Are there special features in the study design? These include
multigroup comparisons, clustered outcome data (such
as multiple observations of the same subject or
multicenter studies), long-term follow-up when dropout
or attrition is expected, and inclusion of covariates
(other measures that are believed to somehow affect
the relationship between the treatment and the
outcome). For many studies, software packages such as
nQuery Advisor or Epilnfo can be used to make
preliminary sample size determinations. However,
special features often involve nonstandard calculations
and a statistician should be consulted.
6. What defines a clinically relevant effect (effect size)? This is
perhaps the most challenging question. The clinician
must define the clinically relevant effect in terms of the
primary outcome. What difference expressed in the
values of the outcome measure would alter/change/
modify the way patients are treated) ?
gear force, shear bond strength); and (3) time-
to-event, information collected in studies that
measures how long it takes for a specified event
to occur, such as number of months to correct to
Class I molars, and what proportion of subjects
reach the event in a certain amount of time.
Hypothesis Testing
Clinical research is designed to determine if a
treatment has an effect or if different treatments
produce different outcomes. Suppose we believe
that removable appliances increase mandibular
length. We might assume that in untreated Class
II children, age 7 to 10 years, mandibular length
growth during a 1-year period is a normally dis-
Home | TOC |          
Index
tributed variable (bell-shaped curve) with a
mean ju^ and a standard deviation (SD) crl. The
mean refers to the average amount of mandib-
ular growth in 1 year and is represented by jUq
because we do not know the true amount of
growth. The SD tells us how much variation we
can expect in the amount of annual growth
among all children age 7 to 10 years. The SD is
represented by crl. Another special term is the
standard error of the mean (SEM). SEM is a
measure of variability similar to the SD. SD indi-
cates the spread of the values for a measurement
such as mandibular length changes. SEM tells us
about the spread of values of the means that
would occur if we drew all of the possible sam-
ples of a given size from the population and
calculated the mean for each sample. It is im-
portant to understand the concept of means and
SDs because their estimates are used to calculate
sample sizes.
If we draw a random sample of size (n x ) from
this population of 7- to 10-year-old children and
measure the mandibular length changes (xj),
we would expect that the mean (average) ofthat
sample would be scattered around the true (but
unknown) amount of mandibular growth (jLtj),
with a standard error given by o-j/Vn^ (Fig 1A).
Now assume that we simultaneously draw an-
other random sample of size (n 2 ) from this same
population, but this time the children in the
second sample have been treated with remov-
able appliances. If treatment with a removable
appliance does affect mandibular growth, then
we would expect that the mandibular length
change (x 2 ) measured in the second sample
would be distributed around a different mean
(ju 2 ), with a standard error cr2/Vn^ that de-
pends on the variability (cr2) in the mandibular
length changes in children treated with remov-
able appliances.
We assume that both sample means are valid
estimates of what we would have found if we
could have measured both entire populations of
interest. One population would have consisted
of all untreated Class II children between 7 and
10 years. The other population would have con-
sisted of all Class II children 7 to 10 years who
were treated with a removable appliance. The
greater the difference between x x and x2, the
sample means for the untreated and treated chil-
dren, the more likely it is that the values of the
removable appliance population are not distrib-
uted about fa but around some other value
<<    
     Article
      >>
Sample Size and Power
Figure 1. The curve on the left shows the distribution
of means of random samples of size (n T ) drawn from
a normally distributed population with mean JLII and
SD C7j. The curve on the right shows another normally
distributed population with mean jH2 and SD cr2 from
which random samples of size n2 are drawn. The
horizontal hatched areas show a, the risk of rejecting
the null hypothesis (H(): ^JL} = jn2) when it is in fact
true (false positive). The vertical hatch represents ß,
the risk of accepting the null hypothesis when in fact
it is false (false negative). The alternative hypothesis is
specified as H^ |JLLT — ju,2| = A rather than H^ \JJLI —
jiL2| = 0 because there are infinite possibilities for
l/^i ~ ^2!' and each is associated with a different
likelihood of a false negative when a and the sample
sizes are specified. The figure is drawn so that crl = cr2,
n^ = n2, and a = .05.
(jLL 2 ). In other words, the larger the separation
between the two sample means, the stronger the
evidence that the removable appliance has an
effect on mandibular growth. If removable ap-
pliance therapy had no effect on mandibular
growth, then we would expect very little (less
than some clinically meaningful amount) or no
difference between the two sample mean values
(xj and x 2 ), and we would expect the mandibu-
lar length changes of the treated children to be
distributed, just like those of the untreated chil-
dren, around IJLI.
Unfortunately, the true effect of removable
appliance therapy on mandibular growth is un-
known. The investigator seeks to answer the re-
search question by drawing conclusions (known
as making statistical inferences) based on the
available data and probabilities. Because proba-
bilities are not certainties, these conclusions can
yield four outcomes with regard to the actual
but, unfortunately, unknown truth about the
difference between groups. Either the data has
led the investigator to arrive at a correct decision
relative to what actually happens in the popula-
tion (a true negative or a true positive conclu-
sion) or to an incorrect decision (a false positive
Home | TOC |          
Index
69
or a false negative conclusion). These four out-
comes are depicted in Table 2.
The strategy for arriving at these conclusions
starts with formulating a null hypothesis (H 0 ). If
we were interested in what happens in just one
of the groups, the null hypothesis in our exam-
ple would state that there is no change in man-
dibular length during 1 year (IJLI = 0). If we were
interested in comparing the changes that occur
between the two groups, the null hypothesis
would state that there is no difference in treat-
ment effect (juL2 = jUq or jn2 — ^ = 0). The
expression JLII represents the average growth in
population 1 (untreated children). The expres-
sion ju,2 represents the average growth in popu-
lation 2 (treated children). Thus, ju,2 = juij is
shorthand for saying that the average growth in
population 1 (untreated children) is equal to
the average growth in population 2 (treated chil-
dren). The alternative way to write it, jii2 — fjil —
0, is shorthand for saying the difference in the
average growth between the two populations is
0. In our example, the null hypothesis would be
that the average mandibular growth observed in
the untreated children is the same as that ob-
served in the children treated with removable
appliances.
Next, we assume that the null hypothesis is
true, and we calculate the probability (P) of
observing just by chance an absolute difference
in mandibular growth equal to or greater than
that actually measured for the two samples. The
decision to accept or not accept the null hypoth-
esis is based on the relationship between the
calculated P value and the risk the investigator is
willing to take of incorrectly rejecting the null
hypothesis. This risk level (a) is the level of
significance and is chosen by the investigator
before the start of the study. In other words, a
Table 2. Four Possible Outcomes When Comparing
the Interpretation of a Hypothesis Test and the
True Population Finding Between Treatment and
No Treatment Groups
Truth in the Population
Hypothesis test
interpretation Different
Not different Correct conclusion Type II error
(True negative) (False negative)
(ß)
Different Type I error Correct conclusion
(False positive) (True positive)
(Power = 1 - ß)
<<    
     Article
      >>
70 Ceib Phillips
represents the risk the investigator is willing to
take of incorrectly rejecting the null hypothesis
(and thus incorrectly concluding there is a dif-
ference between the two groups). This is known
as a type I error (or false positive) (Table 2).
Investigators usually set a at 1% or at 5% (a =
.01 or a = .05) to describe the risk they are
willing to take of a false-positive error rate.
When the P value (ie, the probability associ-
ated with finding a difference in growth of the
observed magnitude occurring, just by chance, if
there is no real difference in the population) is
greater than a, a result is said to be not statisti-
cally significant. The shorthand is P > .05 (if a is
chosen by the investigator to be .05) or P > .01
(if a is chosen by the investigator to be .01).
A nonsignificant P value does not imply that
the null hypothesis is true, that there is no dif-
ference between the two treatments, or that the
two treatments are equivalent. A nonsignificant
P value tells you merely that there is inadequate
weight of evidence against the null hypothesis.
Nonsignificant results are inconclusive because
the default position represented by the null hy-
pothesis has been neither confirmed nor re-
jected.
The distinction between failure to reject the
null, which is the correct designation when P is
greater than a, and accepting the null is impor-
tant. The first implies a lack of evidence on
which to reject the null, whereas the latter im-
plies the null is true. Three possibilities exist
when Pis greater than a (eg, P> .05) (Table 2):
(1) there is no difference or no effect (true
negative); (2) you were unlucky and there really
is a difference in the population, but your sam-
ples do not reflect this population difference
(false negative by chance); or (3) the observed
difference is real and clinically important, but
the sample size was too small to reach statistical
significance (false negative because of poor
study design).
If the P value is less than or equal to a (P <
a), the result is said to be statistically significant.
This does not prove with 100% certainty that the
null hypothesis is false nor does it mean that the
difference is clinically important. The null hy-
pothesis is simply rejected because there is too
much evidence against it— the likelihood of
observing such a difference by chance, if there is
no effect or no difference, is less than the ac-
ceptable risk level (a) chosen by the investiga-
tor. Two possibilities exist when P is less than a
Home | TOC |          
Index
(eg, P< .05) (Table 2): (1) there is a difference
or effect (true positive), or (2) you were unlucky
and there really is not a difference in the popu-
lation, but your samples suggest there is (false
positive by chance).
If the decision is made to reject the null hy-
pothesis, then what? An alternative hypothesis
(H x ) becomes the fall-through decision. The al-
ternative hypothesis is generally established as
two-tailed, which means that the direction of the
difference is not specified, (ie, the change in
one group may be either larger than or smaller
than the other group). In other words, if the
means are not the same then they must be dif-
ferent.
When the P value fails to reach statistical sig-
nificance (ie, P > a) even though the underly-
ing groups are truly different, then a type II
error (false negative) has occurred (Table 2).
The probability of committing a type II error is
called ß. ß is the level of risk of accepting a false
negative conclusion: the risk that an investigator
is willing to take of declaring there is not a
difference when there is one (false negative).
This is not the same as a, the risk an investigator
is willing to take of declaring there is a differ-
ence when there is not one (false positive).
The likelihood (or probability) of avoiding a
false-negative error is termed the statistical
power of the study. Power thus expresses the
probability of detecting a true effect, a true pos-
itive. Adequate power traditionally has been de-
fined as 80% or 90% (ie, 0.8 or 0.9) and is equal
to (1 - ß). If power = 0.8 or 0.9 then ß - 0.2 or
0.1. Thus, the risk of committing a type II error
(ß) is 0.2 or 0.1. (Remember that the risk of
committing a type I error is a.)
For specific sample sizes and values for a,
power expresses how likely it is that a study will
detect a difference of a certain magnitude when
that difference really exists in the population. If
power is 90% then we expect 9 out of 10 trials
(identical studies) to indicate a statistically sig-
nificant difference exists when in truth a differ-
ence does exist in the population. If the power is
10%, then we expect that in only 1 out of 10
trials would a statistically significant difference
occur even though a true difference exists.
Another way to explain power is to say that
when the null hypothesis is really false (there is
truly a treatment effect), a study with a power of
10% has only a 10% chance of rejecting the null
hypothesis and a 90% chance of being inconclu-
<<    
     Article
      >>
Sample Size and Power
sive. Thus, if there is a true treatment effect, you
have only a 10% chance of detecting this differ-
ence. Your study is underpowered.
The specification of the alternative hypothesis
as Hp fjil =£ JA?, (two-tailed alternative hypothe-
sis) works in hypothesis testing only if the deci-
sion of interest is the risk we are willing to take
of accepting a false-positive decision (type I er-
ror). In other words, do we only care about
whether or not we have evidence to reject the
null hypothesis? If, however, the risk of a false-
negative decision (type II error) is important
to us or if a sample size or power calculation
is desired, then the difference of interest must
be explicitly stated. The general inequality of
Hp fJL1 3= jLL2 must be replaced by the form
H x : IjUq — /jL2| = 1.25 mm or whatever value the
investigator considers clinically meaningful. In
other words, the type II error rate or power
cannot be calculated for a two-tailed alternative,
which merely says that the two groups are differ-
ent.
Calculating the probability (ß) of a type II
error, risk of accepting a false-negative decision,
requires that the size of the difference between
the two groups be explicitly stated. For example,
in our study we may decide that a true absolute
difference in the average amount of mandibular
growth in the treated and untreated children
would be clinically meaningful if the difference
were 1.25 mm (H^ [/ij — jut2| = 1.25). The size of
this difference should represent the minimum
effect that would be considered clinically rele-
vant.
As the value specified by the alternative hy-
pothesis changes (ie, the clinically relevant ef-
fect) , the sample size required to detect a spec-
ified difference and the power of a study to
detect that difference also changes (if you do
not change the a level and the variability in your
study stays the same). This is discussed further
later.
Clinically Relevant Effect
The specification of the clinically relevant effect
or difference is based on clinical judgment not
statistical inference. The researcher should spec-
ify beforehand what magnitude of treatment ef-
fect would be regarded as the minimum effect
size of interest (ie, the minimum change or
difference that would be clinically meaningful).
Because not all clinicians will share the same
Home | TOC |          
Index
71
opinion regarding the effect that would be re-
quired to change their clinical practice, it may
be prudent to obtain input from a broader base
than a single investigator or even a single aca-
demic care center's faculty.5 Detecting small dif-
ferences will generally require more patients
than showing really strong treatment effects that
result in large differences.
Because it is often difficult to choose the ex-
act magnitude of treatment effect that is of in-
terest, it is important to plan the study so that
there is a high likelihood of rejecting the null
hypothesis if the minimum treatment effect or
difference that would be of therapeutic impor-
tance exists. There are times when a 10% differ-
ence is irrelevant; there also are times when a
10% difference is crucially important. If the rate
of apical resorption after orthodontic treatment
were only 10% higher than in untreated subjects
with a P value of .06, it is likely that the investi-
gator would not be disappointed that statistical
significance was not attained. However, a statis-
tically nonsignificant finding that early orth-
odontic treatment reduced incisor trauma by
only 10% would be disappointing and inconclu-
sive. The goal of any clinical study should be to
have sufficient numbers of subjects so that clin-
ically meaningful effects are also statistically sig-
nificant.
The Role of Variability
The results of a clinical trial are based on a
limited sample of patients, and therefore the
observed treatment difference is simply an esti-
mate of the true treatment difference. Even a
well-designed study can only give an idea of the
answer sought because of random variation in
the sample studied and the variation that would
occur between the samples if several samples
were drawn from the population. Thus, results
from a single sample are subject to statistical
uncertainty, which is inversely related to the size
of the sample. In other words, statistical uncer-
tainty decreases as sample size increases. For
example, you are more likely to have confidence
in an estimate of the rate of incisor trauma if this
estimate is based on a study of 200 rather than
on a study of 20. An estimate based on the larger
sample is likely to be more precise than the one
based on the smaller sample.
In a study measuring continuous variables,
the extent of the variability of the measure
<<    
     Article
      >> Home | TOC |          
Index

72 Ceib Phillips

among the patients will affect the sample size or Sample Size for Comparing Proportions
power calculations. As the variation from patient
Consider a trial designed to compare the rate of
to patient increases, detection of the same level maxillary incisor trauma in orthodontically
of difference between treatments will require treated and untreated children ages of 7 to 10
increasing numbers of subjects. For example, in years old with an over]et of at least 7 mm. Previ-
our mandibular growth study, if we assume that ous studies have estimated that 20% of children
a = .05 and ß = .1, only 24 subjects per group who do not have orthodontic treatment experi-
would be needed to detect a 1.25-mm difference ence incisor trauma. We decided that an abso-
in mandibular length growth if the variability lute difference of 5% (.05) would be considered
was low (SD = 1.3 mm). On the other hand, 73 clinically meaningful. We also set a = .05 (risk of
subjects would be required if the variability was false positive [type I] error) and ß = .20 (risk of
greater (SD = 2.3 mm). Although the extent of false negative [type II] error). Because we are
the variability cannot be known in advance, it comparing proportions of children in each
can be estimated from several sources, includ- group who will experience incisor trauma, we
ing previous studies, pilot data, and educated use the x2 formula in n query6 to compare two
guesses. proportions for the two-tailed alternative hypothe-
sis (Hp (proportion 1 — proportion 2| = .05).
These calculations tell us that 906 patients per
Estimation of Sample Size group or a total of 1,812 children (2 groups X
Requirements 906 subjects/group) are necessary to detect the
difference we consider clinically meaningful.
The methods used to calculate the sample size Setting ß = 0.2 gives us 80% power to detect a
will depend on the analysis plan (ie, what statis- statistically significant difference if it is true that
tical test will be used to analyze the outcome). In there is a 5% difference in the proportion of
addition to the analysis plan, three parameters trauma (effect size) in the population between
must be specified before the calculation of a treated and untreated children 7 to 10 years old
sample size: (1) the effect size of interest, (2) the with an overjet of at least 7 mm.
acceptable risk of a false positive finding (a), Table 3 shows various combinations of values
and (3) the acceptable risk of a false negative for a, ß, and effect size (magnitude of a clinically
finding (ß). meaningful difference) and their influence on
Other information, such as an estimate of the the sample size when the effect size is expressed
variability for a continuous variable, may be re- in proportions. The effect of ß is more easily
quired. The impact of varying the statistical pa- understood when expressed as power (1 — ß). It
rameters and/or the magnitude of the effect size is apparent that to detect a small effect size
on sample size requirements are discussed for
comparisons of the following types of data: two Table 3. Two Proportions Compared
proportions for a dichotomous measure (only
Proportion of Trauma
two responses possible, eg, yes/no, present/ab-
sent) , two means of a continuous measure (mm, Untreated Treated
Children (%) Children (%) Power Sample Size
degrees, grams), and two time-to-event measures (PI)* (P2)f a. Per Group
(%)
(length of time till a specified event occurs).
20 15 .05 80 906
Sample size calculations depicted in the ta- 20 10 .05 80 199
bles to follow were accomplished by using the 20 5 .05 80 76
software package, n Query Advisor Version 4.0 20 15 .05 80 906
20 15 .05 90 1212
(Statistical Solutions, Boston, MA), although in- 20 15 .01 80 1348
vestigator-written computer programs and soft- 20 15 .01 90 1717
ware packages such as Epilnfo are available. Epi- 20 15 .05 80 906
50 45 .05 80 1565
Info is a free software package that includes
sample size and power calculations. Epilnfo is *P1 is the proportion of untreated children who are ex-
pected to experience incisor trauma.
available from the Centers for Disease Control fP2 is the proportion of treated children who are expected
and Prevention (www.cdc.gov). to experience incisor trauma.
<<    
     Article
      >> Home | TOC |          
Index

Sample Size and Power 73

(difference in % incisor trauma = 5%) requires Table 4. Two Means Are Compared
more patients than required to detect a large Mean Annual
effect size (difference in % incisor trauma = Mandibular Growth
15%) if the difference actually exists. It is also (mm)

clear that if more restrictive error rates (that is, Untreated Treated Standard Power Sample Size
decreasing the level of a and/or ß) are chosen Group Group Deviation 0! (%) Per Group

to reduce the risk of observing false-positive and 2 3 1.5 .05 80 37

false-negative results, larger sample sizes are nec- 2 4 1.5 .05 80 10
2 3 1.5 .05 80 37
essary. Finally, it is noteworthy that the sample 2 3 2.0 .05 80 64
size is greatest when the proportion of an effect 2 3 1.5 .05 80 37
2 3 1.5 .05 90 49
(ie, the rate of incisor trauma) in one of the
2 3 1.5 .01 80 55
groups is 0.5 or 50% (as long as a and ß are 2 3 1.5 .01 90 69
constant).

Sample Size for Comparing a Time-to-Event

Sample Size for Comparing Continuous
Measures
Consider the same trial but now assume the
intent is to compare the mandibular growth of
treated and untreated children. Mandibular
growth is measured on a continuous scale and is
therefore summarized for each group by means.
The estimate of mandibular growth (derived
perhaps from other studies or a pilot sample) in
untreated children is 2 mm per year. Assume
that an absolute increase of 50%, or 3 mm per
year, in the treated group would be considered
to be clinically worthwhile. Thus, the effect size
we deem meaningful is the difference between 3
mm and 2 mm growth, or an effect size of 1 mm.
Again the sample size will be calculated given a
two-sided test with an a of .05 and a power of
80% (ß - .2).
Because mandibular growth is a continuous
measure, an estimate of the variability is also
needed. We will assume an SD of 1.5 mm based
on findings from previous studies. By using the
formula for an unpaired t test to compare two
means,6 37 patients per group would be re-
quired to detect an effect size of 1 mm given the
specified a, power, and variability (a — .05, ß =
0.2, S.D. = 1.5 mm). Thus, a total of 74 patients
would be required for the study (2 groups X 37
patients/group). Table 4 shows that for compar-
ison of continuous data, sample size increases as
the effect size decreases (size of the difference in
growth decreases), the variability increases (SD
increases), or the error rates are more restrictive
(a and ß decrease). Remember that power in-
creases (1 — ß) as ß decreases or becomes more
restrictive.
Measure
Time-to-event analysis is useful when the investi-
gator is interested not only in whether some
event occurs but also in how long it takes for the
event to occur. An example of such an event is
the clinical endpoint of correcting the Class II
molar occlusion to a Class I molar occlusion.
Time-to-event analyses are designed for studies
in which patients are entered into a trial and
followed until a specified event occurs (attain-
ment of Class I molars), the patient is lost to
follow-up, or the study ends. In clinical studies in
which the main outcome is the time to an event,
the power of the study depends on the number
of events observed during the trial rather than
the number of patients.
As an example, let us assume that in the Class
II early treatment trial, the measure of interest
was defined as the time required for a child
treated with a removable appliance to reach the
clinical endpoint of Class II molar correction
compared with the time required if a headgear
was used. The event would be thus defined as
reaching the clinical endpoint of attaining Class
I molars. Not all patients will attain Class I mo-
lars in the specified time period of the trial.
Enough patients therefore must be entered and
followed for a sufficient length of time to ob-
serve a critical number of events (ie, a certain
number of molar corrections). Thus, the sample
size involves two stages: (1) deciding on the
required number of events that will be consid-
ered meaningful and (2) calculating the re-
quired sample size based on this number of
events.
Before the calculations can be performed, the
<<    
     Article
      >> Home | TOC |          
Index

74 Ceib Phillips

investigator must specify the event, a reasonable Table 5. Occurrence of Event Are Compared for
length of time within which the event might Two Groups
occur, an estimate of the proportion of children Estimated Accrual Follow-Up
in one of the groups who would have had the Length of Time- Time Time Sample Size*
to-Event (mo) (mo) (mo) PI P2 Per Group
event by the specified time, the magnitude of
the difference in the proportions in the two 6 24 24 .25 .40 119
6 24 24 .25 .35 258
groups, and finally, the length of time available 6 24 24 .25 .30 984
to recruit and follow patients. 12 24 24 .25 .40 155
Assume that we decide that it is important to be 18 24 24 .25 .40 193
6 18 24 .25 .40 102
able to detect an absolute difference of 15% be- 6 12 24 .25 .40 97
tween the two groups in the proportion of chil- 6 24 30 .25 .40 100
dren that reach the clinical endpoint by 6 months. 6 24 36 .25 .40 95
In our example, let's estimate that 25% of the *Sample size estimated using an exponential survival curve
removable appliance group would reach the clin- approach with a specified accrual period.6
fPl is the proportion of subjects in Group 1 to reach defined
ical endpoint (Class I molars) by 6 months. A 15% event.
difference in proportion of the groups to reach JP2 is the proportion of subjects in Group 2 to reach defined
Class I molars means either a 15% increase or a event.
15% decrease in the proportion of headgear pa-
tients to reach Class I molars compared with the
25% of removable appliance patients we expect to nitude of the difference in proportions of events
reach this critical event. The sample size should be decreases, the estimated length of time to event
sufficiently large that if at least 40% of the head- increases, and as the length of time available for
gear group (15% more than the removable appli- recruitment and follow-up of patients decreases
ance group) or only 10% of the headgear group relative to the length of the time to event.
(15% less than the removable appliance group) The difference in the sample sizes that are
reached the endpoint by 6 months, the difference required for these three different types of com-
would almost always be detected as statistically sig- parisons emphasizes an important point: the in-
nificant. (A difference in the two groups between vestigator must declare not only the effect size of
10% to 40% would not be considered clinically interest but also the primary outcome of inter-
meaningful according to our criteria, and there- est. If all three outcomes (proportion of incisor
fore we are not interested in designing our study to trauma, average mandibular growth, and pro-
detect if differences in this range are statistically portion of subjects who attain Class I molars
significant.) Assume that the trial is designed to within 6 months) shown earlier were considered
recruit patients for 24 months (length of time for primary (important to both the investigator and
patient accrual) and the maximum follow-up time the reader), then 906 children per group would
is 24 months. By using our program6 and setting a be the appropriate number of subjects. This will
at .05 and power at 80%, a total of 119 children per ensure that the primary outcome requiring the
treatment group for a total of 238 (2 groups X 119 largest sample size (ie, difference in incisor
children/group) would need to be recruited trauma proportion) is adequately recruited. Al-
within the 24 months to detect an absolute differ- though a study that is based on only 37 patients/
ence of 15% between groups in the number of group may lead to conclusions about the differ-
children who attain Class I molars in the specified ence in mandibular growth for the two groups
time. that are both clinically important and statistically
In studies in which time to event is short, the significant, no valid conclusions could be made
number of patients will be almost the same as with so few patients with regard to the incidence
the number of events because all patients will be of incisor trauma or to the time to reach the
followed until the event. However, in trials in clinical endpoint of molar correction. Thus, the
which an event takes a long time to occur, a primary outcome, which requires the largest
larger number of patients must be recruited to sample size, should be used to determine the
observe the required number of events. Table 5 number of subjects that must be recruited to the
shows that the sample size increases as the mag- study.
<<    
     Article
      >>
Sample Size and Power
Adjustments to Sample Size Calculation
Sample size calculations represent the sample size
per group needed at the end of the study and
therefore underestimate the number of subjects
who need to be enrolled if the study requires that
patients be evaluated on more than one occasion.
No matter how carefully a study is planned, some
patients are likely to be lost between enrollment
and completion (Fig 2). Patients may be dropped
because of violations of the inclusion or exclusion
criteria or patient compliance, protocol errors of
misdiagnosis or incorrect assignment, or patient
attrition resulting from loss to follow-up or missed
examinations.7
Special attention should be paid during the
development of the protocol to mechanisms that
will minimize loss of patients. The attitude of clinic
staff toward patient complaints and the vigorous
pursuit of patients who fail to keep study appoint-
ments are important factors in reducing dropout.8
Strategies that have been used include reminding
patients of forthcoming appointments, assisting
with transportation, minimizing waiting times,
sending newsletters, providing monetary reim-
bursement, providing continuity of care, involving
family members, and maintaining contact with pa-
Do not meet study criteria
(n3)
Patient does not finish study Patient finishes study
(n7) (n8; sample size calculation)
Figure 2. Subsets of patients to be considered when
planning a study. Calculating the sample size to detect a
clinically meaningful effect is not the only consideration
in the number of patients to be screened and enrolled
in a study. The investigator will also need to estimate the
proportion of patients who are likely to meet the study
inclusion criteria, the proportion who will agree to par-
ticipate, and the proportion who are likely to complete
the study protocol (ie, the retention rate). (Modified
from Phillips C: Design principles in oral and maxillo-
facial surgery clinical trials. Oral Maxillofac Surg Clin 13;
237-243, 2001)
Home | TOC |          
Index
75
tients' primary care health professionals. Orr et al9
found, in a multicenter clinical trial, that the fac-
tors most strongly associated with incomplete fol-
low-up could not have been identified before en-
rollment. These included changes in marital or
employment status, motives for enrolling in the
trial, and too little time spent with the study clini-
cian.
Sample size estimates need to be adjusted for
any expected loss from the study. The number of
subjects to be enrolled in the study can be cal-
culated as the sample size required per group
divided by the expected retention rate. For ex-
ample, if the sample size calculation indicates 37
patients per group are needed and if the ex-
pected retention rate for the trial is 80% then
the number of enrolled subjects should be 47
(37/.S = 46.25). As might be expected, the
number of subjects needed for enrollment in-
creases as the projected retention rate decreases.
Post Hoc Analyses of Power
On occasion, well-designed studies with prior
sample size and power calculations result in sta-
tistically nonsignificant results. The concern of
course is that the failure to reject the null hy-
pothesis (ie, no treatment effect) may be the
result of low statistical power. The estimate of
variability may have been too low, or an impor-
tant effect actually may have existed that the
investigator missed by chance, a false-negative
outcome. This has been called the dilemma of
the nonrejected null hypothesis.10 Some authors
have advocated postexperiment power calcula-
tions11 to explain the observed data (ie, calcu-
lating the power associated with the observed
effect or finding the effect that could have been
detected with higher power given the sample
size, level of significance, and variability). How-
ever, these calculations can lead to contradictory
evidence for and against the null hypothesis.10
An alternative approach to determining if
population values (such as the true but un-
known difference) are supported by the data is
the calculation of confidence intervals. For some
statisticians, this approach has more validity than
calculating the power after the study has been
completed. A 95% confidence interval is the
range of values that you are 95% certain con-
tains the true but unknown value of the differ-
ence you are seeking. The interval is bounded by
<<    
     Article
      >>
76 Ceib Phillips
a lower limit and an upper limit and is deter-
mined by rigorous statistical methodology using
specific statistics derived from the sample in
your study. If the value of zero is contained
within the 95% confidence interval, then it is
possible that there is no treatment effect be-
cause the difference between the treated group
and untreated group may equal zero.
For example, using our earlier example that
compared the average annual difference in man-
dibular growth between 37 treated and 37 un-
treated children, let's assume you found that the
mean difference in growth was 1 mm, but the SD
was 2.5 mm. The SD of your actual data turned
out to be much higher than the value you as-
sumed for the sample size calculation. If an un-
paired t test were used to make the comparison
between the treated and untreated groups using
these sample values (mean difference of 1 mm
and SD of 2.5 mm), the P value would be .09,
which tells you that there is no statistically sig-
nificant difference in the average amount of
mandibular growth. If we calculated the 95%
confidence interval for this comparison, we
would have found the interval to be defined by
the limits of ( — 0.16 mm, +2.16 mm).
The difference in mandibular growth was cal-
culated by subtracting the average amount of
growth in untreated children from the average
amount of growth in treated children. Thus, values
greater than 0 tell you that the treated children
grew more than untreated children. Values less
than zero tell you that the treated children grew
less than untreated children. A value of zero
means there is no difference in the growth be-
tween treated and untreated children. Although
we still do not know what the real difference in
growth is, we can say that we are 95% confident
that, in the population, untreated children, on
average, might grow as much as 0.16 mm more
than treated children (this corresponds to —0.16
mm) and that, on average, treated children will
not grow more than 2.16 mm more than untreated
children. The confidence interval includes zero,
but the asymmetry of the negative and positive
boundaries around zero are in the direction of the
hypothesized effect (ie, that treated children will
grow more than untreated children) and could be
supportive evidence for a similar study with a sam-
ple size calculation based on larger variability esti-
mates. If, however, the confidence interval was
Home | TOC |          
Index
symmetrical around zero (eg, —1.5, +1.5) then
the investigator has no evidence to support a di-
rectional effect in the population.
Conclusions
A study with negative results but adequate power
to detect clinically meaningful differences may
be a valuable contribution to the literature. A
negative study with inadequate power is incon-
clusive at best. The question remains: is the sam-
ple size calculation worthwhile? The answer is
yes although sample size calculations are based
on many assumptions and represent our best
guess. These calculations provide a guide to the
feasibility and practicality of undertaking a study
and focus attention on the consideration of what
clinical effect, if it exists, would impact patient
care.
References
1. Cohen J: Statistical Power Analysis for the Behavioral
Sciences (ed 2). Hillsdale, NJ, Lawrence Erlbaum, 1988
2. NIDCR Policies and Procedures for Investigator Initi-
ated Clinical Trials. Available at: www.nidcr.nih.gov/
research/ctp/clinical%5Ftrials.him. Accessed 19 Octo-
ber 2001
3. Hulley SB, Cummings SR: Designing Clinical Research:
An Epidemiologie Approach. Baltimore, MD, Williams
and Wilkins, 1988
4. Freedman D, Pisani R, Purves R, et al: Statistics (ed 2).
New York, NY, WW Norton, 1991
5. Fayers PM, Cuschieri A, Fielding J, et al: Sample size
calculation for clinical trials: The impact of clinician
beliefs. Br J Cancer 82:213-219, 2000
6. Elashoff JD: nQuery Advisor Version 4.1 User's Guide.
Boston, MA, Statistical Solutions, 2000
7. Phillips C, Tulloch JFC: The randomized clinical trial
(RCT) as a powerful means for understanding treatment
efficacy. Semin Orthod Dentofac Orthoped 1:128-138,
1995
8. Goldman JF, Holcomb R, Perry HM, et al: Can dropout
and other noncompliance be minimized in a clinical
trial? Report from the Veterans Administration National
Heart, Lung and Blood Institute cooperative study on
antihypertensive therapy: Mild hypertension. Control
Clin Trials 3:75-89, 1982
9. Orr PR, Blackhurst DW, Hawkins BS: Patient and clinic
factors predictive of missed visits and inactive status on a
multicenter clinical trial. Control Clin Trials 13:40-49, 1992
10. Hoenig JM, Heisey DM: The abuse of power: The per-
vasive fallacy of power calculations for data analysis. Am
Statistician 55:19-24, 2001
11. Berry E, Coustere-Yakir, Grover NB: The significance of
non-significance. QJM 91:647-653, 1998
<<    
     Article
      >> Home | TOC |          
Index

Common Statistical Tests

Rose D. Sheats and V. Shane Pankratz

A recent survey by BeGole of four major orthodontic journals revealed that

76% of the original and scientific articles published in 1997 incorporated
statistical procedures in their methodologies. To evaluate the information
presented in these scientific articles, the clinician needs to have an under-
standing of the statistical techniques used in them. Although a number of
the statistical methodologies were complex, requiring considerable statis-
tical sophistication to understand and apply, the majority of the statistical
methods that were used were more easily understood. The purpose of this
article is to familiarize the orthodontic clinician with those statistical proce-
dures commonly encountered in the orthodontic literature. Actual formulas
and computations are not reviewed in this article and can be found in
standard statistical texts. The focus of this article is to emphasize the
appropriate use of techniques such as the Student t test, analysis of vari-
ance, the x2 test, and the nonparametric alternatives to these methods, which
comprise more than 50% of the statistical tests used in the journals from 1997
examined by BeGole. (Semin Orthod 2002;8:77-86.) Copyright 2002, Elsevier
Science (USA). All rights reserved.

tions. It would also guide us in verifying whether

T he prevailing model of health care delivery
emphasizes evidence-based criteria and
guidelines for developing treatment options and
alternatives.1'2 As practitioners, we understand
the wisdom of this philosophy but find ourselves
challenged to assess the evidence presented to
us at our professional meetings, in our specialty
journals, and by our supply representatives. Not
only do we need to form judgments of the clin-
ical relevance of new technologies, we must also
be able to decide if the enthusiasm for novel
approaches is supported by reasonable evidence.
Our ability to assess critically the information
leveled at us would be greatly enhanced by a
familiarity with biomedical statistics. It would
help us to determine if investigators applied ap-
propriate, if any, statistical techniques to the
data used to address clinical and research ques-
From the Mayo Clinic Rochester, Departments of Dental Special-
ties and Health Sciences Research, Rochester, MN.
Address correspondence to Rose D. Sheats, DMD, MPH, Depart-
ment of Dental Specialties, Mayo Clinic, 200 First Street SW,
Rochester, MN 55905.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8 746/02/0802-0006$35.00/0
doi:10.1053/sodo.2002.32073
or not the conclusions were supported by the
results of the analysis. This is the first step in
practicing evidence-based dentistry.
To help the reader gain an understanding of
common statistical procedures used in orth-
odontic research, this article will discuss key con-
cepts and review statistical tests commonly used
in orthodontic studies. The topics of regression
analysis and correlation merit an article of their
own and are presented in a separate article in
this issue by Dr Lysle Johnston.
Descriptive Versus Inferential Statistics
Statistics provide a tool to summarize the
findings from a specific group of observations.
Ideally, this group would be a representative
collection of observations drawn from a popula-
tion of interest to the practitioner (eg, growing
children or anterior open-bite cases). The pop-
ulation does not have to be limited to just hu-
man beings. It could be a population of all teeth
with standard edgewise brackets or a population
of lateral cephalograms of all anterior open-bite
cases. If the group under investigation is truly a
representative sample of the population, it is

Seminars in Orthodontics, Vol 8, No 2 (June), 2002: pp 77-86 77

<<    
     Article
      >>
78 Sheats and Pankratz
possible to form conclusions about the entire
population based on specific measurements or
observations taken from this representative sam-
ple.
Statistics used to summarize, or describe, the
characteristics of a sample are called descriptive
statistics. Typical summary statistics include the
mean, standard deviation, and variance of a
measurement of interest in a sample and will be
discussed in greater detail later. In theory, these
same measures, or summary statistics, could be
computed from the entire population. When
this is done, the computed population statistics
are often referred to as parameters. Because it is
often impracticable to perform a study on an
entire population of interest, samples from the
population are drawn, and sample statistics of-
ten form the basis for drawing conclusions con-
cerning entire populations. Such conclusions
may simply describe the population, but they
may also make comparisons among populations
or even among subgroups within a specific pop-
ulation.
Statistical procedures that are used to make
inferences (ie, draw conclusions) about the pop-
ulation that the sample represents are called
inferential statistics. Because the sample does
not include all members of the population, it is
unlikely that the measurements on the sample
would be precisely the same as the measure-
ments on the entire population, if it were possi-
ble to measure everyone in the population.
Thus, the true value of the measurement for the
entire population remains unknown. However,
by using the statistics that describe representa-
tive samples, statisticians can estimate with rea-
sonable certainty the range of values (ie, the
interval) that is likely to contain the true mea-
surement of the population. Statisticians can
also make comparisons between or among two
or more groups to determine if the observed
evidence supports the conclusion that a real dif-
ference is present between groups.
We encounter the need for inferential statis-
tics throughout orthodontics. Without such
techniques, how would we decide if one treat-
ment renders a different result from another, if
one adhesive system is superior to another, if
there is an association between a specific maloc-
clusion trait and relapse, or if certain cephalo-
metric values predispose to good or bad treat-
ment outcomes?
Home | TOC |          
Index
Central Tendency and Dispersion
When a group of observations is examined
(humans, rats, teeth, cephalograms, and so on)
in an effort to answer a question, there is a range
of responses (ie, data points). A typical first step
in understanding the data is to plot these data
points on a graph to get a feel for the distribu-
tion of the data. The plot will probably show a
scattering of the data points (hence, the term
scatter plot), but some data values may occur
more frequently than others. Central tendency
refers to the center of the distribution of data
points (the tendency of data from biologic sys-
tems to cluster). This center point is a single
value that is used to describe the set of observa-
tions without listing the values for every subject
(or tooth or cephalogram). Frequently, the scat-
ter plot, or distribution of the data, may look like
a bell-shaped curve. In statistics, this distribution
pattern is called the normal distribution (also
called the Gaussian distribution).
Data that can have an infinite number of
values between any two points (such as millime-
ters overjet or interincisal angle) are called con-
tinuous data because they are measured on a
continuous scale. For continuous data, the most
commonly used measure of central tendency is
the mean. It is calculated by summing the values
of a specific measurement from all observations
and dividing by the number of observations (ie,
calculating the average of that measurement for
the group).
Sometimes data cannot be quantified on a
continuous scale. Examples of such data are sex
or race. Data that are not continuous are cate-
gorical because their values are assigned to cat-
egories. The different values that the variable
can have are the levels of the category. For ex-
ample, the levels of the variable, sex, can be
named male and female. Such categorical data
are called nominal (named) because the levels
of the category are named but have no mean-
ingful order to the levels. Tooth type is another
example of a nominal variable. The values that
this variable could have would include named
levels such as incisors, canines, and premolars.
There is no order from lowest to highest value in
these tooth types, although there is an anatomic
order in the mouth.
Another form of categorical data exists when
there is a meaningful order to the levels. An
<<    
     Article
      >>
Common Statistical Tests
example is the apical root resorption (ARR)
scoring system of Malmgren et al.3 In this system,
the scoring levels of 0 to 4 reflect an increasing
amount of ARR based on guidelines for visually
judging the amount of root resorption by radio-
graphic evaluation rather than actually measur-
ing the amount of resorption with a ruler or set
of calipers. Categorical data such as these, which
have a meaningful order, are ordinal (ordered).
They are similar to nominal data in that their
levels are also named. The difference between
nominal and ordinal data is that the named
levels of ordinal data have a natural order to
them. Note that the ARR scoring system uses
numbers as the names of the levels (ie, scores of
0, 1,2, and so on), but the numbers do not refer
to a measurement on a continuous scale. Instead
these levels refer to a qualitatively increasing
amount of ARR. Malmgren et al3 could have
named the levels of root resorption none, slight,
and moderate. Gingival inflammation is scored on
an ordinal scale (none, mild, moderate, severe).
Orthodontists often classify crowding on a similar
ordinal scale (none, mild, moderate, or severe).
For ordinal data, the median or mode is used to
represent the center of the data because a mean
cannot be calculated as it can for continuous data.
The median is the middle observation where half
the values are lower than the median and half are
higher. The mode is the value that occurs most
frequently in the group observations.
A hypothetic example using a modified gin-
gival index score4 may help clarify the difference
between the median and the mode. Assume
there are 101 patients with gingival index (GI)
scores as summarized in Table 1. From this ta-
ble, it is noted that the most common GI score is
3 because more of the patients (40) scored at
Table 1. Hypothetic Example Using GI Scores to
Compare Median and Mode (n = 101 Patients)
Gingival Index No. of Cumulative No.
Score Patients of Patients
0 5 5
1 20 25
2 30 55
3 40 95
4 6 101
NOTE. The mode — 3 because this is the level of GI score
that occurs most frequently (40 times). The median = 2
because this is the GI score of the middle patient (the 51st
patient) when the patients are ranked by GI score from low
to high.
Home | TOC |          
Index
79
this level than any other level. Thus, GI equals 3
is the most frequent score, the mode. Another
way to summarize the GI score of this group of
patients is to find the score of the patient who
ranked right in the middle of this group (ie, the
score of the 51st patient). From the table, it can
be seen that his score was 2 since patients 26
through 55 have a GI = 2. Thus, the median
score is 2. From this example, it is evident that
the median and the mode are not always identi-
cal. Some statistical tests use the median in their
computations, which will be discussed in more
detail later.
The median is also used as a measure of
central tendency for continuous data that are
skewed (Fig 1A, B) to minimize the effect of
extremely large or small values on the estimate
of the center of the data. Data are skewed to the
left (Fig 1A) when extreme data values occur at
the low end of the scale. The result is that the
mean is lower than the mode (the value under
the peak of the curve because this is the value
that occurs most frequently) or the median
(value of middle observation). In the normal
distribution, an example of a symmetric distri-
bution rather than a skewed distribution,
mean = median = mode (Fig 1C). In skewed
data, the relative positions of the mean, median,
and mode depend on whether the distribution is
skewed to the left or right (Fig 1 A, B).
Nominal data (data that are recorded by lev-
els such as sex or ethnic group) cannot be sum-
marized by means, medians, or modes. As dis-
cussed later, this type of data is analyzed by
looking at the number of observations in each
level of the variable (eg, number of men and
women and the number of whites, blacks, His-
panics, and Asians). The data are summarized by
reporting the proportion or percentage of the
data that are classified in each level. These pro-
portions can be analyzed by examining if the
proportions differ between groups to be com-
pared. A common test for this type of analysis is
discussed later.
In addition to identifying a value to express
the central tendency of data (mean, median, or
mode), a summary measure of the amount of
dispersion or spread of data around the center is
useful. For continuous data, the measure of
spread or dispersion of the data around the
center of the distribution may be expressed in
several ways. A commonly used technique is to
<<    
     Article
      >>
80 Sheats and Pankratz
0)
Mean Mode
Median
Q)
D"
Mean = median = mode
present the minimum and maximum values that
the data span. This is known as the range. This
approach, however, is sensitive to extreme data
values and may not be an accurate reflection of
the spread of data points. An outlier, or extreme
value on either end of the scale, could lead to an
inaccurate perception of how tightly data other-
wise cluster together. Another way to describe
dispersion is to present interquartile ranges,
such as the values for the 25th and 75th percen-
tile level, which are not as likely to be influenced
by the values at the extreme upper and lower
end of the spread of data points.
When you use the mean (average of all the
data points) to represent the sample, you realize,
of course, that most data points will be different
from the mean. The difference between the av-
erage of all the points and each individual point
is called the deviation of that point from the
mean (average). If you wanted to know the av-
erage size of the difference of each point from
the mean (ie, the average deviation from the
mean), one way to determine this difference is
to total all the differences of each observation
from the mean and divide by the number of
Home | TOC |          
Index
51
O
G"
£
Mode Mean
Median
Figure \. Skewed distributions compared with sym-
metric distribution. Skewed to the left, extreme low
values shift mean to the left of (smaller than) the
median or mode. (A) Skewed to the right, extreme
high values shift mean to the right of the median or
node. (B) Normal distribution, mean = median =
mode. (C)
points. If you did it this way, however, you would
quickly realize that the total of the differences
would be zero because some points would be
higher than the mean, and some would be lower
than the mean. To eliminate the effect of the
negative differences, each difference is squared,
and an average of the squared differences is
taken. This is called the variance of the data. The
variance is therefore expressed in squared units.
If you wanted to get back to the average differ-
ence, or deviation, from the mean expressed in
the same units as the original measurement, you
would need to take the square root of the vari-
ance. This value is called the standard deviation.
Thus, the standard deviation is the average
amount by which each observation point varies
from the average (mean) of the points. Tables 2
and 3 summarize these common measures of
central tendency and dispersion.
Parametric Versus Nonparametric
Statistics
Inferential statistics are those statistical proce-
dures that compare groups to see if the groups
<<    
     Article
      >>
Common Statistical Tests
Table 2. Measures of Central Tendency
Measure Description
Mean Average of values of data points
Median Value of data point that falls in the
middle when data are arranged in
order; for even number of
continuous data points, average of
the middle two values gives median
Mode The value that occurs most frequently
when data points are arranged in
order
NOTE. For data that are normally distributed, mean = median = mode.
are significantly different from each other.
There are two kinds of inferential statistics: para-
metric statistics and nonparametric statistics.
The word parametric refers to the use of param-
eters, those calculations of sample means, stan-
dard deviations, and variances that summarize
the sample and provide estimates of the popu-
lation parameters that are unknown. Thus, para-
metric statistics refers to a group of statistical
tests that uses means and a measure of variation
(standard deviation, variance) to help deter-
mine if groups are different from each other.
The mean and standard deviation completely
describe the normal distribution (bell-shaped
curve) in mathematical terms. Further clarifica-
tion about this important mathematical concept
is beyond the scope of this article, but it explains
why certain conditions regarding the data must
be met before the simplest parametric tests,
based on means and standard deviations, may be
validly used. The following three conditions are
necessary: (1) the data must be continuous
(measured on a continuous scale, eg, millime-
ters, pounds, degrees), (2) a scatterplot of the
data must look like a normal distribution (bell-
shaped curve), and (3) the dispersion or spread
Table 3. Measures of Dispersion
Measure Description
Range Minimum and maximum value
of data
Interquartile range Value of data at the 25th and
75th percentile
Variance Measure of the average squared
deviation of each data point
from the mean value of all
the data points
Standard deviation Square root of the variance
Home | TOC |          
Index
81
Use
To summarize continuous data
To summarize ordinal data
To summarize continuous data that are
skewed
To minimize the effect of extreme values
Same as uses for median
To determine the most frequently
occurring value
of data for each variable must be the same in
each group being compared (the size of the
variance or standard deviation of the variable is
the same in each of the groups being com-
pared). Common parametric tests include the
Student t test and analysis of variance (ANOVA)
and will be discussed later.
Data that are ordinal (eg, mild/moderate/
severe, ARR scores of 0,1,2,3,4) cannot be sum-
marized by parameters (means and standard de-
viations). Hence, such data cannot be analyzed
by parametric tests, which rely on means, stan-
dard deviations, and variances. Ordinal data are
analyzed by nonparametric procedures. Non-
parametric statistics use the ranks of the data
rather than means and standard deviations to
make group comparisons.
Rank refers to the position in a hierarchic
ordering of the data that each data value holds
(eg, from highest to lowest or vice versa). A
familiar ranking of data among dental students
is the class rank, often based on grade point
average (GPA). Each GPA is associated with a
rank after the GPAs are ordered from highest to
lowest value. The highest GPA is ranked number
1, the next highest is ranked number 2, and so
Use
Wide ranges imply greater variability;
tight ranges imply lesser variability
Less sensitive to extreme high and low
values
Deviations are squared to eliminate the
effect deviations that are below the
mean (ie, negative in value)
Variability expressed in same units as
original measurements (not squared)
<<    
     Article
      >>
82 Sheats and Pankratz
forth. The lowest GPA holds the rank equal to
the number of students in the class.
From the ranks, a median value can be ob-
tained. Remember that the median is the middle
observation after data have been ordered from
low to high or vice versa. Nonparametric tests
compare medians, rather than means, for anal-
ysis. Common nonparametric tests based on
ranks include the Mann-Whitney U test, the Wil-
coxon signed rank test, and the Kruskal-Wallis
test. These will be discussed later.
Nonparametric statistical tests are also used
for continuous data that are not normally dis-
tributed (bell-shaped curve). Although a mean
can be calculated for such data, as discussed
earlier, the median may be a more appropriate
measure of the center of data that are not nor-
mally distributed in order to avoid the influence
of extreme data on the estimate of the center of
the data. If the median is used to compare
groups, nonparametric tests are used.
Data that are nominal (eg, sex, tooth type)
cannot be summarized by means or ordered into
ranks. By summarizing the numbers of observa-
tions in each level (eg, incisor, canine, premolar,
and so on) of a category (eg, tooth type), one
can determine the ratio or proportion of the
total number of observations that occur in each
level. Group comparisons of such data are made
by analyzing if the same proportion of tooth
types (or men/women, Class I/II/III malocclu-
sions) occur in all the groups. The most com-
mon test to analyze nominal data is the ^2 test,
which will be further detailed later.
Test Statistics
Statistical procedures comparing samples
usually provide a test statistic or critical ratio that
is associated with a probability level (P value).
The probability level, called the P value for
short, is the likelihood or chance that two
groups, representative of the same population,
would be chosen, and that there would be a
difference in the groups at least as big as the one
detected. A P value < .05 means there is an
equal or lower than a 5% chance (1 in 20) that
the two groups could be samples from the same
population. By convention, when P ^ .05, inves-
tigators conclude that their groups are not from
the same population and therefore are statisti-
cally, significantly different. A P value > .05 is
Home | TOC |          
Index
not significant, and investigators conclude that
their groups are not different from each other
(and thus treatment results are not different).
Convention states that P < .05 is statistically
significant; however, this level (called the a
level) may be changed if the investigator believes
a lower or higher probability level is more ap-
propriate for his/her study. It is important, how-
ever, for the investigator to state his/her a level
at the beginning of the study before data analysis
takes place.
Commonly used statistical software packages
provide the value of the test statistic or critical
ratio as well as the P value associated with it. The
P value thus tells whether the evidence is strong
enough to conclude that there is a statistically
significant difference between the groups.
Common Statistical Tests
Parametric Tests
Remember that to use a parametric test prop-
erly, it must be verified that the data are contin-
uous, that the distribution of the data in each
sample is normal (bell-shaped curve), and that
the variance (or standard deviation) of the vari-
able to be examined is the same in each group to
be compared. Two common parametric tests
that are frequently found in the orthodontic
literature are the Student t test and ANOVA.
The Student t test is used when only two
groups are being compared. The Student t test
uses sample means and standard deviations to
calculate the probability or likelihood that the
groups are different. It helps you determine if
the means differ because the two groups repre-
sent two different populations or if the means
differ because the groups have different subjects
but each group represents the same population.
Figure 2A shows what might be expected if the
groups represent two populations (I and II) that
are distinguished by different population means.
Figure 2B shows how two groups could be rep-
resentative of the same population, (III) al-
though the groups have different values for their
group means.
The Student t test actually exists in two forms,
depending on whether the two groups under
comparison are paired (matched) or indepen-
dent of each other. The paired t test (also known
as the matched-group t test), as the name im-
<<    
     Article
      >>
Common Statistical Tests
o Examples of matching criteria include sex, age,
c ethnicity, and cephalometric values, which
D
x1 x2
Value of mean
B of lower incisor position before and after orth-
0)
er
£ and after treatment, the difference would be 0
LL
Value of mean
Figure 2. Location of group means. Two group
means, different populations. X1 and X2 represent
means for group 1 and group 2 when the groups are
representative of different populations (I and II).
Note that the true values of the means for populations
I and II are unknown but would be located on the
x-axis under the peak of each curve. The diagram is
drawn such that X T is coincidental with the mean
value for population I and that X2 is coincidental with
the population mean value for population II. How-
ever, the value of the mean for group 1 could have
occurred anywhere under the area of the curve rep-
resenting population A. Similarly, the value for the
mean of group 2 could also have occurred anywhere
under the area of the curve representing population
II. (A) Two group means, same population. X T and X2
represent means for group 1 and group 2 when both
groups are representative of the same population
(III). Unlike Fig 2A, the value of the means of both
groups does not coincide with the population mean that
would be located under the peak of the curve. (B)
plies, examines groups that are paired or
matched. A common paired design occurs when
a single group of subjects is measured before
and after a procedure to examine the effect of
some intervention (eg, treatment). A matched-
group study design is one in which the outcome
of each subject in the treatment group is com-
Home | TOC |          
Index
83
pared directly to the outcome in another subject
who is as similar as possible to its mate, with the
exception of the treatment under investigation.
should be as close to identical as possible for
each pair of matched subjects. In the absence of
producing clones or having an unlimited supply
of monozygotic twins, matched pairs are often
the best alternative we have for comparisons and
are only as good as our success is in matching
subjects.
An example of a paired study is a comparison
odontic treatment such as with a Herbst appli-
ance. Assume the change in lower incisor posi-
tion for each patient before and after Herbst
treatment is to be examined. If in truth, there is
no change in the lower incisor position before
between pre- and posttreatment incisor position.
Thus, the test is to determine if the average
change in incisor position that is measured in
the sample is significantly different from 0
(when 0 means no change in position). Because
paired samples are used (the patient before
treatment compared with himself/herself after
treatment), the paired t test is appropriate.
The alternative t test to the paired t test is the
two-sample, independent t test. This test is used
to compare independent groups or unmatched
groups. An example is to examine the amount of
skeletal relapse between rigid fixation and inter-
maxillary wire fixation groups in patients who
have undergone bilateral sagittal split osteoto-
mies. The groups are independent because dif-
ferent subjects are used for each group unless
subjects were matched, as discussed previously.
This independent, two-sample study design oc-
curs frequently in retrospective studies in which
subjects are frequently chosen based on records
availability or completeness. Note that in paired
study designs, the number of subjects in both
groups is the same, whereas in the two-sample,
independent design, the size of the two samples
may be different.
If more than two groups are being compared,
the ANOVA is used. Unlike the t test, which uses
the mean and standard deviation of groups for
its computations, ANOVA uses the mean and
variance of groups for computations. The test
statistic rendered by an ANOVA is the F statistic
<<    
     Article
      >>
84 Sheats and Pankratz
so sometimes the test is called the F test. ANOVA
makes a series of pair-wise comparisons for all
the groups in the comparison. For example, if
groups I, II, and III are compared, ANOVA will
compare I to II, I to III, and II to III. A signifi-
cant P value indicates that a difference exists
somewhere between any two comparisons, but
ANOVA does not identify which groups are dif-
ferent. To determine which pairs differ (I-II,
I-III, or II-III), any of several subsequent tests
(called post hoc or a posteriori tests) can be
used to examine the groups in detail and reveal
which groups significantly differ from each
other. Note that these post-hoc tests are not used
if the P value from the ANOVA is not significant.
Common post hoc tests are the Tukey-Kramer
honestly significant difference, Scheffé, Dun-
nett, Duncan, and Newman-Keuls tests. The
choice of test depends on the specific research
design, and consultation with a statistician is
recommended to identify the most appropriate
post hoc test for the study.
Nonparametric Tests
The previous statistical techniques were ex-
amples of common parametric tests. To use
parametric procedures, remember that the con-
tinuous data must follow a normal distribution
(bell-shaped curve) and that parametric tests
rely on means, standard deviations, and vari-
ances. Sometimes the continuous data are not
normally distributed, or there are ordinal data
(data that fall into levels that have a meaningful
order). Ordinal data cannot be summarized by
means and standard deviations. In cases such as
these, parametric statistical tests are not suitable,
and nonparametric procedures are indicated.
They are called nonparametric because they do
not use population parameters (ie, estimates of
means, standard deviations, or variances) in
their computations.
A common nonparametric test for compari-
son of two unpaired samples is the Mann-Whit-
ney Latest also known as the Wilcoxon rank sum
test (not to be confused with the Wilcoxon
signed rank test, which is described later). It is
similar to the unpaired t test described earlier,
which is a parametric procedure that compares
the means of two unpaired groups to see if they
are different. The Mann-Whitney £7 test is a non-
parametric procedure that compares the medi-
Home | TOC |          
Index
ans of two groups, rather than the means, to see
if they are different. The median of each group
is found by ranking the data in each group from
lowest to highest and identifying the middlemost
value (median). The test statistic for the Mann-
Whitney U test is called the U statistic and, like
all test statistics, is associated with a P value that
measures the likelihood that the difference be-
tween the groups could have occurred by chance
alone. The example of grade point averages and
class rank may help clarify the use of the Mann-
Whitney £7 test.
What if the question was to determine if there
was a difference in the academic performance of
female students and male students? Because
GPA is a continuous variable, one way to answer
this question would be to use an unpaired t test
to compare the mean GPA of female students
with male students. But what if one student had
an extremely low GPA? This would unfairly in-
fluence the mean GPA of that group. Eliminate
this unfair influence by comparing the median
of class ranks for men and women. To obtain the
medians, a list of ranks for men and women
based on their GPA rank in the class is created.
From these ranks, the median rank for the men
and for the women is determined. Because the
medians of two independent groups (men and
women) are being compared, the Mann-Whit-
ney U test is the appropriate nonparametric test
to use. If the result of the analysis by the Mann
Whitney U test returns a P value that is not
significant (P> .05), it is concluded that there is
no difference in the median academic ranks
between men and women. But if this test is
statistically significant (P < .05), it is concluded
that the two groups are actually different and
that there is a true difference in the academic
ranks of men and women. Thus, if the median of
the female ranks is significantly lower than the
median of the male ranks (the lower the rank
[no. 1], the better the academic performance),
it would be inferred that the female students
perform better academically than the men. Of
course, it could be the other way around (medi-
an of male ranks lower than median of female
ranks), and the opposite would be concluded.
The comparable nonparametric test to the
paired t test is the Wilcoxon signed rank test. If
there are two matched or paired groups and
ordinal or continuous data are being examined,
this is the appropriate nonparametric procedure
<<    
     Article
      >> Home | TOC |          
Index

Common Statistical Tests 85

to use. An example of the appropriate use of this
test can be found in a recent article comparing
the oral hygiene in upper and lower jaws5 using
several indices which are ordinal such as a mod-
ified gingival inflammation index (GI). 4 The
modified GI is ordinal because it scores gingival
inflammation in five levels ranging from absence
of inflammation to severe inflammation. In the
study cited, the two comparison groups were
upper jaw oral hygiene scores and lower jaw oral
hygiene scores. The groups were paired rather
than independent because the upper jaw was
being compared with the lower jaw in the same
patient. Nonparametric tests were required be-
cause the data were ordinal. If the Wilcoxon
signed rank test were statistically significant, it
would have been concluded that there was a
difference in oral hygiene between the upper
and lower jaws. If the test were not significant, it
would have been concluded that patients
cleaned their upper jaws as effectively as their
lower jaws.
What if there are more than two groups to
compare and it is not appropriate to use para-
metric procedures such as ANOVA (which com-
pares three or more groups)? The nonparamet-
ric test comparable to the ANOVA is the Kruskal-
Wallis procedure. The ANOVA uses means and
variances for its computations. The Kruskal-Wal-
lis test, like the previous nonparametric proce-
dures described, examines intergroup differ-
ences based on ranks.
An example of the use of the Kruskal-Wallis
test can be found in another recent study that
examined three laser energy levels used to bond
brackets on extracted teeth.6 An assessment of
the amount of adhesive remaining on the teeth
after the brackets were removed was made using
the Adhesive Remnant Index (ARI).7 This index
assigns a score from 1 to 5 to represent decreas-
ing amounts of adhesive remaining on the tooth
surface. Thus, the ARI is an ordinal scale and
must be analyzed by a nonparametric test. Be-
cause an ordinal variable was being compared
among three groups (the three laser energy lev-
els), a nonparametric alternative to ANOVA is
necessary (ie, the Kruskal-Wallis test).
Finally, how is nominal data analyzed? Nom-
inal data cannot be summarized by means be-
cause such data are not continuous. These data
cannot be analyzed by medians because there is
no natural order to the levels (ordinal). The
most common way to analyze nominal data is to
use the x2 test. This is a very versatile test that has
many applications, but for the purposes of this
article, discussion is limited only to its use in
comparing nominal data. It is used to compare
the proportion of the data that fall into each
level of the nominal variable. For example, we
might like to know if the number of Class I, II,
and III malocclusions is the same in different
ethnic groups. The levels of Angle class are nom-
inal (Class I, Class II, Class III). The levels of
ethnic group are also nominal (white, black,
Asian, and so on). There is no inherent order to
the levels of either of these variables so they are
not ordinal and therefore not amenable to anal-
ysis by tests that rely on medians or means. The
X^ test compares the proportion of Class I, II,
and III malocclusions for each ethnic group to
determine if these proportions differ by ethnic
group. A statistically significant ^2 test indicates

Table 4. Comparable Parametric and Nonparametric Tests

Parametric Test Nonparametric Test Use

Paired t test Wilcoxon signed rank test To compare two paired

(matched) samples for
equality of means/medians
Two-sample t test for independent samples Mann-Whitney U test (Wilcoxon To compare two independent
rank-sum test) samples for equality of
means/medians
Analysis of variance Kruskal-Wallis To compare more than two
samples for equality of
means/medians
X2 analysis To compare nominal data: to
compare two or more
samples for equivalence in
proportion of data
classified at different levels
of a specific variable
<<    
     Article
      >>
86 Sheats and Pankratz
that there is a difference in the number of Class
I, II, and III malocclusions among the ethnic
groups. A statistically nonsignificant \2 test indi-
cates that the number of Class I, II, and III
malocclusions is the same among ethnic groups.
Table 4 tabulates the procedures discussed
earlier and lists comparable parametric and non-
parametric procedures and their application. It
is by no means comprehensive but may be a
convenient tool to help clarify the indications
for their use. Specific details of how each of
these tests uses sample estimates, distributions,
and proportions will not be addressed in this
article but can be found in any standard statisti-
cal text.
Conclusions/Summary
As our specialty strives to embrace the prac-
tice of evidence-based dentistry to guide our
clinical decisions, expectations will be raised re-
garding the use of appropriate statistical analysis
when applicable. Dr BeGole's enlightening
study, which was cited earlier, urges orthodontic
clinicians and graduate students to become fa-
miliar with common statistical methods. This
presentation of statistical concepts and common
Home | TOC |          
Index
analytic procedures, although not comprehen-
sive, may provide a useful review of the statistical
tests that one is likely to encounter while facing
the challenge of integrating patient-centered
clinical studies with an individual patient's
needs.
References
1. BeGole E: Interpreting the literature on the effectiveness
of clinical procedures. Semin Orthod 6:67-75, 2000
2. Sackett D, Straus S, Richardson W, et al: Evidence-based
Medicine: How to Practice and Teach EBM (ed 2). New
York, NY: Churchill Livingstone, 2000
3. Malmgren O, Goldson L, Hill C, et al: Root resorption
after orthodontic treatment. Am J Orthod 82:487-491,
1982
4. Lobene R, Weatherford T, Ross N, et al: A modified
gingival index for use in clinical trials. Clin Prev Dent
8:3-6, 1986
5. Thienpont V, Dermaut L, Van Maele G: Comparative
study of 2 electric and 2 manual toothbrushes in patients
with fixed orthodontic appliances. Am J Orthod Dentofac
Orthop 120:353-360, 2001
6. Talbot T, Blankenau R, Zobits M, et al: Effect of argon
laser irradiation on shear bond strength of orthodontic
brackets: An in vitro study. Am J Orthod Dentofac Or-
thoped 118:274-279, 2000
7. Artun J, Bergland S: Clinical trials with crystal growth
conditioning as an alternative to acid-etch enamel pre-
treatment. Am J Orthod 85:333-340, 1984
<<    
     Article
      >> Home | TOC |          
Index

Regression: Is Your Guess as Good as Mine?

Lysle E. Johnston, Jr

Much of what is called diagnosis and treatment planning involves predic-

tion. In orthodontics, however, much of our prediction is unabashed guess-
work. Firstly, the strength of the relationship between the thing we want to
predict and the thing we want to predict it with is often weak or poorly
characterized. Secondly, even when a forecast is based on a clinically sig-
nificant biological relationship, the form of the relationship is often largely
unknown. In other words, what is the formula by which knowledge of one
characteristic is to be converted into a specific prediction of another? Finally,
given an appropriate equation, we need to know something of its probable
use. How much error will there be? Will it work well in my hands? Is its
accuracy high enough to warrant routine clinical application? Questions
such as these are central to the topic of prediction; to a first approximation,
their answers are easily inferred from a consideration of regression, a sta-
tistical technique that deals with the analysis of relationships. (Semin
Orthod 2002;8:87-91.) Copyright 2002, Elsevier Science (USA). All rights
reserved.

n the formulation and execution of an orth-
/ odontic treatment plan, the structure of the
decision process often is poorly characterized and
incompletely validated. At each step, therefore, the
clinician is faced with two key questions: does the
decision rule have a real biological basis, and, if so,
is its transfer function (the exact prediction equa-
tion) adequately characterized? Do high-angle pa-
tients always respond poorly to cervical traction?
Or do they do so only occasionally? If there is some
sort of relationship, what is the formula by which
we can estimate an increase in facial height from
initial mandibular plane angle? Does the form of
the mandible tell us how it is going to grow in the
future? If it does, how are we to combine the
various characteristics (symphysis width, condylar
angle, and so on) to calculate the expected growth
increment? Do premolar extractions always or oc-
casionally lead to distal mandibular displacement?
From the Department of Orthodontics and Pediatric Dentistry,
School of Dentistry, The University of Michigan, Ann Arbor, ML
Address correspondence to Lysle E. Johnston, Jr., DDS, PhD, FDS
RCS(E), Department of Orthodontics and Pediatric Dentistry, School
of Dentistry, The University of Michigan, Ann Arbor, MI 48109-
1078.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0802-0007$35.00/0
doi:10.1053/sodo.2002.32189
If so, how much will the mandible be displaced? I
doubt that anyone really knows the answers to
these questions. Given this uncertainty, the best we
can do in many instances is to assume that every
patient will undergo the mean response as we have
come to know it. For example, it is probably fair to
expect that the mandible will outgrow the maxilla,
that molars will come forward if premolars are
extracted, and that there will be clockwise mandib-
ular rotation during treatment. Although we know
that all patients do not respond in the same way,
betting on the mean is probably better than noth-
ing at all and certainly better than a wild guess.
Some, however, argue that we must do more. Im-
plicit in their argument is the assumption that we
can do more.
Many clinicians think that the mean is, by
definition, inappropriate to the individual and
therefore not worth knowing. "I treat my cases
one at a time" is the usual refrain. Much the
same way that athletes, when pressed for some-
thing profound to say, deliver themselves of a
sincere but ultimately trivial determination to
play their remaining games one at a time. But
how are we to render these individualized deci-
sions and on what basis? As it turns out, theory,
intuition, memory, common sense, and the
opinion of helpful experts more often than not

Seminars in Orthodontics, Vol 8, No 2 (June), 2002: pp 87-91 87

<<    
     Article
      >>
88 Lysle E. Johnston, Jr
are inadequate to the task. If we are to do better,
we must return to the twin topics of correlation
and regression introduced earlier in this issue.
A key event in the development of the science
of biometrics was Sir Francis Galton's research into
the inheritance of human physical characteristics.
When he plotted children's heights as a function
of their parents' heights he found that, although
the resulting scattergram depicted a strong rela-
tionship, it also featured what he termed a regres-
sion to the mean: children of tall parents also
tended to be relatively tall but not as much above
average as were their parents. The same pattern of
inheritance—a regression to the mean—was seen
in the offspring of subjects who were shorter than
average. As a result of Galton's pioneering re-
search, the process of modeling and analyzing the
relationship between two characteristics has come
to be called regression. Although not often recog-
nized as such, much of what passes for orthodontic
diagnosis and treatment planning is, at bottom, a
regression problem.
The common practice of predicting the size of
unerupted canines and premolars from the widths
of the mandibular incisors, the likelihood of root
resorption from the form of the roots, or the pat-
tern of future growth from the form of the man-
dible are obvious examples. The venerable prob-
lem of growth prediction—"them that has, gets"—
can serve as a framework for a discussion of the
various aspects of the topic of regression.
Clearly, it would be useful to know how an
individual patient is going to grow. Indeed, con-
trary to the protestations of the one-patient-at-a-
time purists, it would even be useful to know
how the average patient grows. If no significant
predictors can be found, the known changes in
facial form that accompany the usual (average,
mean, normal, expected) pattern of growth con-
stitute a useful alternative to no prediction at all,
or worse, an arbitrary, idiosyncratic stab in the
dark. My mother, for example, may not have
known how much my feet were going to grow
(quite a bit as it turns out); however, she knew
they were going to get bigger. As a result, she
always bought me shoes that were a few sizes too
big. In so doing, she made a practical use of her
knowledge of growth. Regression, however, tries
for more: it attempts to render an individualized
prediction whose error is significantly smaller
than that of any alternative (betting on the
mean, assuming that all patients will be the
Home | TOC |          
Index
same, relying on expert opinion, and so forth).
In the present "them that has, gets" example,
those who believe in common sense forecasting
assume that there is an obvious and clinically
exploitable relationship between present facial
form and the way the face will grow in the future.
Good faces grow well; bad faces grow poorly.
When presented with such an assertion, the wary
clinician should seek answers and evidence be-
fore betting the farm, not to mention the pa-
tient's well-being, on orthodontic old wives' tales
masquerading as diagnostic pearls.
Is the Relationship Significant?
Many of the assumptions that fuel orthodontic
decision making, although popular and occa-
sionally even logical, are of dubious or, at the
very least, unknown significance. As a result, the
first step in evaluating a clinical rule of thumb is
to ask whether it is based on a real (ie, statisti-
cally significant) relationship between the thing
we are trying to predict and the characteristic we
want to predict it with. For example, is there a
relationship between what the patient has (man-
dibular plane angle, antegonial notching, face
height, and the like) and whatever it is we would
like to know about the face in the future (what
the patient gets)? Correlation provides a means
of answering this sort of question.
The possible existence of a clinically useful
relationship commonly is explored by drawing a
representative bivariate sample—a collection of
elements (in this case, patients) from which
both the thing to be predicted (Y, the depen-
dent variable) and the thing to be used to pre-
dict it (X, the independent variable) can be
measured. In passing, it should be noted that a
predictor variable can take many forms: it can be
a single craniofacial characteristic, a change in
some characteristic measured over a period of
observation (as with the so-called Tweed growth
trends inferred from the pattern of change in
A-Na-B), or even a weighted combination of sev-
eral variables. Given a potential predictor, it is
necessary to test whether its variation matches
that of the dependent variable we wish to fore-
cast. In other words, is there a correlation that is
strong enough to be exploited clinically?
A correlation between two variables can take on
many forms, including linear, exponential, power,
and so on. Suffice it to say, a linear relationship is
<<    
     Article
      >>
Regression
Y change will occur. But what about the face? A
20 -i
75- Y = A + ß(X)
10-
rise/run = slope, B
5-
0 10
Y-intercept, A
Figure 1. Y = A + B(X), the formula for a line in
Cartesian coordinates. A, Y-intercept, the value of Y
when X = 0; B, slope constant; X, the value of the
independent (predictor) variable; and Y, the value of
the dependent variable.
perhaps the most common and certainly the easi-
est to interpret. Correlation examines the extent
to which two characteristics vary, either positively
or negatively, in unison. The so-called coefficient
of linear correlation, r, a dimensionless index rang-
ing from —1 to +1, is the usual statistical tool used
to assess the strength of a relationship. A perfect
positive relationship would have r = 1, and a per-
fect inverse relationship would have r = — 1. When
r = 0, the two variables are completely indepen-
dent. In practice, r usually takes on some interme-
diate value, say 0.5 or —0.6, the significance—the
chance that it occurred purely by chance—of
which must be evaluated. If two variables are
highly correlated, then there is a chance that the
relationship can be exploited clinically. In this con-
text, it should be noted that a correlation has to be
surprisingly strong (|r| = 0.6 or 0.7 or even higher)
to be clinically useful. But how do you use one
variable to predict another? Knowing that they are
related is not enough; we need to know how they
are related. We need to model the relationship.
We need to develop a prediction equation.
What Is the Form of the Relationship?
We know that Fahrenheit and Celsius tempera-
tures are related; however, we need to know the
exact formula if we are to calculate one from the
other. Unfortunately, biological forecasts are dif-
ferent; they are imperfect. We do not expect to
see an exact mathematical relationship. Rather,
we hope merely to generate a prediction that has
less error than some alternative approach, com-
monly the simple expectation that the average
Home | TOC |          
Index
89
considerable portion of the face in the future is
just the face today. Change, rather than future
size, would seem to be the logical goal of growth
prediction. If some facial measures were to be
shown to be correlated with future change, by
what formula is this subsequent growth incre-
ment (or any other future event including re-
sorption, relapse, whatever) to be predicted? Al-
though two characteristics can be related in
many ways (linear, power, exponential, and so
on) the usual and often adequate assumption is
that the relationship can be modeled as a
straight line and that a prediction can be derived
from a knowledge of the slope and Y-intercept of
the line (Fig 1). How are we, or rather the
educator/entrepreneur who is lecturing to us at
a meeting, to estimate these two components of
a prediction equation? The same way statistical
parameters are usually estimated; namely, from
a representative bivariate sample (in this case,
probably the one used to establish the existence
of the significant correlation).
As has already been noted, a bivariate sample
consists of elements (patients, for example)
from which both the characteristic to be pre-
dicted and the variable to be used to predict it
can be measured and depicted in the form of a
scattergram (Figs 2 and 3). The trick then is to
use these data to derive an equation, most com-
monly a linear equation (Fig 4), from which
predictions can be generated. If there is a signif-
icant correlation, the error of this approach will
be less than that of any common alternative,
such as betting on the mean (Fig 5). For any
given set of data, however, there are an infinity
of lines that can be used to represent it (Fig 6).
Dependent variable (Y)
20
1
15-
10-
5-
0 —i
2 3 4 5 6 7 10
Independent variable (X)
Figure 2. Scattergram depicting an apparently signif-
icant positive linear relationship between two charac-
teristics, X and Y.
<<    
     Article
      >>
90 Lysle E. Johnston, Jr
Dependent variable (Y)
20
l
75-
10-
5-
O
3 4 5 6 7 8
Independent variable (X)
Figure 3. Scattergram depicting an apparently non-
significant relationship between two characteristics, X
andY.
Intuitively, it is reasonable to assume that we
would pick the line that fits the data best. But
how is best to be defined? In linear regression,
the best is defined as the line that produces the
smallest standard error—the minimum sum of
vertical squared deviations between the points
and the regression line (Fig 5). Given such a
best-fit line, we can use its formula, Y = A +
$(X), to predict outcomes in other cases. The
fun, unfortunately, has just begun.
In regression, it is assumed that each value of
the predictor variable defines a unique subpopu-
lation of dependent variables. The regression
line is assumed to run through the means of
these subpopulations. It is for this reason that
the line is sometimes referred to as the line of
means. In effect, for any given value of our
With regression:
20 -i
15-
Figure 4. Linear regression. Error is measured as the
sum of the squared vertical scatter around the line. If
the regression is significant, this scatter will be less
than the scatter around the mean (see Fig 5).
Home | TOC |          
Index
Without regression:
10
Figure 5. Prediction based on the mean. Error is
measured as the squared vertical scatter around the
mean of Y. If there is a significant relationship be-
tween the dependent and independent variables, this
error will be greater than that of the regression line
(see Fig 4).
predictor variable, we bet that the outcome will
be the mean of the corresponding subpopula-
tion (Fig 7). We are still expecting the mean;
however, now it is the mean of a population with
a smaller standard deviation (the so-called stan-
dard error of regression). If there is a significant
correlation, the standard error of regression will
be smaller than the overall standard deviation of
the dependent variable. In other words, the scat-
ter around the line will be less than the scatter
around the grand mean. Indeed, it can be shown
easily that any estimate that differs from the
subpopulation mean will have, in the long run, a
greater average error. If the standard deviation
Thing to be predicted
20
1
15-
Which line
is best?
10-
5-
0
3 4 5 6 7 10
Predictor variable
Figure 6. Fitting a line to empirical data. For the
regression of Y on X, there are many apparently log-
ical alternatives, only one of which is best.
<<    
     Article
      >>
Thing to be predicted
20-i datum
actual
value
J the error of a forecast generated by a regression
error
predicted^
value
3 4 5 6 7 8
Predictor variable
Figure 7. Regression assumptions. The regression
line of Y on X is assumed to pass through the means
of the various subpopulations of Y, all of which not
only should have the same variance but also should be
normally distributed. For a given X, it may be seen
that the predicted value of Y is the mean of the
corresponding subpopulation. Any scatter within this
restricted distribution therefore constitutes predic-
tion error.
is small enough, the regression will provide us
with an efficient forecast, a prediction that is
better than the simple expectation that each
patient's growth will be equal to the mean
change for the entire population. For example,
if the population standard deviation is 10, but
the standard deviation of the appropriate sub-
population is 5, then the regression equation
has produced a 50% improvement. A few more
comments and caveats are perhaps in order.
There are a number of situations in which it
may be appropriate to use more than one pre-
dictor variable. In the case of multiple regres-
sion, we form a weighted combination of predic-
tors (as with the American-weighted peer
assessment rating index) and use this resultant
as the basis of our prediction. Whether single or
multiple, a regression equation is calculated
from a finite sample. The resulting equation,
therefore, represents the best-fit solution for this
particular set of patients; it probably will not
work as well on a new sample, such as your next
100 patients. It is for this reason that, if it is to be
used clinically, a regression equation has to be
tested on a new sample (validated) to see how
well it is likely to work in your hands and on your
patients. Also, for a variety of reasons, it is ap-
propriate that regression equations be updated
from time to time to ensure that an imperfect/
biased, but still statistically significant, equation
Home | TOC |          
Index
Regression 91
is not used blindly to produce flawed forecasts
year after year. Finally, it should be noted that
equation increases as the individual patient's
predictor variable deviates from the mean. In
other words, regression (all prediction) works
best for the average patient. Whatever the indi-
vidual patient's place in the continuum, how-
ever, a regression equation or some approxima-
9 10 tion thereof can be expected to do better than
any common sense alternative.
Envoi
The purpose of this article is to provide an out-
line of the thinking behind prediction, the pro-
cess of estimating the magnitude of some un-
known characteristic, commonly the way growth
and treatment will affect a given patient. It con-
sists in large measure of classifying each patient
as precisely and meaningfully as possible (male,
age 8, Class II, high angle, and so forth). Regres-
sion is just another means of effecting a detailed
classification by defining restricted subpopula-
tions of patients on the basis of one or more
presumably meaningful predictor variables.
Once a subpopulation has been defined, and
once a patient has been assigned to it, we merely
bet on the mean for that restricted category.
In the end, my message is simple and quite
independent of the details of statistical manipu-
lation: clinical decision making (diagnosis and
treatment planning) must be based on signifi-
cant relationships, the form of which should be
well characterized if they are to be both used
and useful. Some will say that regression is an
unnecessarily complex approach to the problem
of clinical decision making. For decisions that
do not really matter, the popular argument that
orthodontics is an art, not a science, is a harm-
less, comforting fiction; however, for important
questions, there is no valid way of avoiding the
thinking subsumed under the rubric of correla-
tion and regression.
Acknowledgment
The author wishes to thank Dr Charles Kowalski, a card-
carrying statistician, for his ongoing help and encourage-
ment.
<<    
     Article
      >> Home | TOC |          
Index

Sensitivity, Specificity, and Related Concepts

Susan P. McGorray

Few screening or diagnostic tests are perfect in distinguishing individuals

who have a specific disease or condition from those who do not. Descriptive
characteristics of these tests, such as sensitivity, specificity, and predictive
values, show how well the tests perform in the absence of perfection.
Interpretation of test characteristics is dependent on characteristics of the
population on whom the tests are conducted. A review of important test
characteristics and a discussion of their relationship to one another are
presented in this article along with examples to clarify key points. (Semin
Orthod 2002;8:92-101.) Copyright 2002, Elsevier Science (USA). All rights
reserved.

Screening and diagnostic tests play an impor-
tant and ever-expanding role in medicine,
dentistry, and orthodontics in helping us deter-
mine if a disease or condition is present. How-
ever, the interpretation of test results can be
problematic. What exactly is a test result telling
us? How do tests compare? A careful review of
the characteristics associated with test perfor-
mance will improve their use, understanding,
and interpretation.
Basic Definitions
In the simplest case, consider a binary test,
which is a type of test that has only two possible
outcomes, such as positive (+) or negative ( — ) .
A positive test indicates a high likelihood (but
not certainty) that the disease or characteristic is
present, whereas a negative test indicates a low
likelihood (but not certainty) that the disease or
characteristic is present. Assume that we also
know with certainty the true disease status or
characteristic of the subjects. This true informa-
tion, sometimes known as the gold standard,
From the Department of Statistics, University of Florida, Gaines-
ville, FL.
Supported by the National Institutes of Health / National In-
stitute of Dental and Craniofacial Research (grant no. DE08715).
Address correspondence to Susan P. McGorray, PhD, Depart-
ment of Statistics, Box 100212, University of Florida, Gainesville,
FL 32610-0212.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0802-0008$35.00/0
doi: 10.1053/sodo. 2002.00000
may be obtained by more elaborate tests or fur-
ther subject follow-up.
To examine the usefulness of a screening test,
we are interested in the relationship between the
test results and the true disease or condition
status. If we know both test result and true dis-
ease status for each subject, the data can be
displayed as in Table 1. In this table, the total
number of subjects for whom we have both test
results and true status is represented by n. Let's
assume in our illustration that n = 100. Each of
the 100 subjects is classified into one of the four
cross-classification cells of the table based on test
result and disease status. The four cells repre-
sent the numbers of subjects with the following
combinations of test result and disease status:
a = the number with test positive, disease
present ( + ); b = the number with test positive,
disease absent (-); c = the number with test
negative, disease present ( + ) , and d = the num-
ber with test negative, disease absent ( — ) .
From this simple table, a surprising number
of test characteristics can be estimated, and each
characteristic shows a particular feature of the
relationship between test result and disease sta-
tus. To interpret properly a test result, it is im-
portant to understand these characteristics.
To begin, the sensitivity of a test is the prob-
ability that the test is positive for those subjects
who actually have the disease. From Table 1, we
estimate sensitivity by the proportion (ratio)
a/(a + c) because a + c represents the total
number of patients with the disease (28) and a
represents the number in this group who tested

92 Seminars in Orthodontics, Vol 8, No 2 (June), 2002: pp 92-101

<<    
     Article
      >>
Sensitivity, Specificity, and Related Concepts
Table 1. General Case and an Example of
Classification of Subjects by Disease Status
and Test Result
Test
True Disease Status or Characteristic
Result Present (+ ) Absent (-) Total
Positive
(+) a (8) b (10) a + b (18)
Negative
(-) c (20) d (62) c + d (82)
Total a + c (28) b + d (72) n (100)
NOTE. Sample data are listed in parentheses.
positive by the screening test (8). Sometimes this
proportion is represented by the corresponding
percentage (proportion multiplied by 100).
From the example data in Table 1, the sensitivity
of this test is 8/28 or 29%. This indicates that of
those who truly have the disease, only 29% will
have a positive test result. A perfect test will have
a sensitivity of 100%. The sensitivity is also called
the true positive rate.
The specificity of a test is the probability that
the test is negative for those in whom the disease
is absent. From Table 1, this is estimated by
d/(b + d), or the corresponding percentage.
The total number of patients without the disease
is b + d (72), and b is the number of patients in
this group who test negative (62). This is also
called the true negative rate. A perfect test will
have a specificity of 100%. From the example in
Table 1, the specificity estimate is 62/72 or 86%.
By knowing the specificity of this test, we would
expect that 86% of those who do not have the
disease would have a negative test result.
For this hypothetic example, the sensitivity of
our test is 29%, and the specificity is 86%. The
specificity of the test is much better than the
sensitivity. Thus, the test performs better in sub-
jects who do not have the disease or condition
than it does in those subjects who do have the
disease or condition by correctly testing negative
for disease-free individuals in a higher percent-
age of cases than it correctly tests positive for
disease-present individuals.
The false negative rate is defined as the pro-
portion of subjects who have the disease but test
negative. This rate is given in Table 1 by c/(a +
c). The false positive rate is the proportion of
subjects who do not have the disease and test
positive. This rate is given in Table 1 by b/ (b +
d). By using the Table 1 example data, the false
Home | TOC |          
Index
93
negative rate is 20/28 (71%), and the false pos-
itive rate is 10/72 (14%).
A test with high sensitivity is important when
the cost of missing a person who actually has the
disease or condition is high. On the other hand,
high specificity is desirable when the goal is to
rule out a disease on the basis of a test result.
The most useful test has both high sensitivity
and specificity, although unfortunately in the
real world this is often not possible.
To estimate sensitivity and specificity of a new
test, one needs to identify subjects whose disease
status is known by some gold standard. Without
advance knowledge of disease status, the practi-
tioner then performs the new test on these sub-
jects and compares test results to the known
disease status of the subjects. From these results,
the sensitivity and specificity estimates for the
new test can be calculated.
For example, one might wish to examine
methods used to determine temporomandibular
joint (TMJ) disc displacement.1 We might be
interested in knowing the sensitivity and speci-
ficity of a simple screening test, which relies on
clinical evaluation of joint sounds and symptoms
to detect the presence of TMJ disc displacement.
We will assume that the true disease status (pres-
ence or absence of TMJ disc displacement) can
be ascertained by magnetic resonance imaging
(MRI) of the joint. Thus, MRI is considered the
gold standard. Subjects for this study would be
selected based on their MRI results so that we
would know the true TMJ disc status. These sub-
jects would then be evaluated by using the
screening test (clinical examination), and the
sensitivity and specificity of this screening test
would be calculated by using the MRI findings as
the gold standard. If our clinical examination
yielded high sensitivity and specificity, we might
conclude that the expense of MRI evaluation is
not justified to identify patients with the pres-
ence or absence of TMJ disc displacement. On
the other hand, if the sensitivity and specificity
were low, we might conclude that our clinical
examination methods (screening test) are not a
reasonable substitute for the MRI examination
of the TMJ to detect disc displacement.
The true positive and true negative rates (sen-
sitivity and specificity) and the false positive and
false negative rates are calculated using as the
denominator the numbers of patients who are
known to have the disease or to be disease free.
<<    
     Article
      >>
94
From a practical standpoint, however, we often
do not know the true disease status of a patient.
Instead the only information available to us may
be a positive or negative test result. How confi-
dent can we be that the disease is present or
absent based on the test results? Knowing the
sensitivity and specificity of a test may not be
useful for us because these estimates use known
disease status for their denominators. Usually,
we are trying to determine the true disease status
of our patients from our test. Two other test
characteristics, the positive and negative predic-
tive values (defined later), use as their denomi-
nator the numbers of patients who test positive
or test negative and thus are more clinically
relevant indices of test performance. The posi-
tive predictive value (PPV) is the probability of
the presence of the disease for those subjects for
whom the test is positive. In Table 1, this can be
estimated by a/(a + b). Similarly, the negative
predictive value (NPV) is the probability of the
absence of the disease in those subjects for
whom the test is negative, estimated by d/(c +
d). In the Table 1 example, the positive predic-
tive value is 8/18 (44%), whereas the negative
predictive value is 62/82 (76%).
Note that the denominators a + b and c + d
represent the number of positive tests and neg-
ative tests, respectively. Compare this to the de-
nominators a + c and b + d, which represent
the number of true cases with disease and true
cases without disease. The accuracy of a test is
Table 2. Test Characteristics From Two Studies Evaluating the Relationship Between Clinical Assessment of
TMJ and MRI Assessment of TMJ
Clinical Evaluation of TMJ (Test Result)
Study 1
Present ( + )
Absent ( —)
Total
Sensitivity: 40/50 = 0.80 or 80%
Specificity: 30/50 = 0.60 or 60%
Positive predictive value: 40/60 = 0.67 or 67%
Negative predictive value: 30/40 = 0.75 or 75%
Accuracy: 70/100 = 0.70 or 70%
Study 2
Present ( + )
Absent ( —)
Total
Sensitivity: 20/25 = 0.80 or 80%
Specificity: 45/75 = 0.60 or 60%
Positive predictive value: 20/50 = 0.40 or 40%
Negative predictive value: 45/50 = 0.90 or 90%
Accuracy: (20 + 45)/100 = 0.65 or 65%
Home | TOC |          
Index
Susan P. McGorray
the proportion of subjects that are correctly clas-
sified, (a + d)/n, in which a is the number of
true positives and d is the number of true neg-
atives. In our example in Table 1, the accuracy is
70% (true positives = 8, true negatives = 62, n =
100). In other words, the accuracy of the test is
an overall summary of how well the test classifies
those patients with and without the disease.
Relationship Between Test
Characteristics
We have defined several test characteristics by
using the counts or frequencies based on the
four combinations of test result and disease sta-
tus as shown in Table 1. As might be suspected,
relationships exist between these test character-
istics. In this section, we will look at how exactly
they are related.
Table 2 further shows applications of the con-
cepts described in the preceding section with
another twist to add to the mix. We will compare
two studies of TMJ dysfunction (artificial data)
that examine the relationship between a screen-
ing test (clinical evaluation) and the gold stan-
dard (MRI) assessment of disease status. As be-
fore, we will use MRI assessment to ascertain the
true condition with respect to TMJ disc status.
Our task is to evaluate the method of clinical
examination of the TMJ to determine how use-
ful it can be as a screening test to identify disc
displacement. Data presented from study 1
MRI Assessment of TMJ (True Disease Status)
Present (+ ) Absent (- ) Total
40 20 60
10 30 40
50 50 100
20 30 50
5 45 50
25 75 100
<<    
     Article
      >>
Sensitivity, Specificity, and Related Concepts
might result from a study designed to estimate
sensitivity and specificity of the clinical evalua-
tion because equal numbers of subjects with and
without the disease were chosen (50 in each
group). Having equal numbers of subjects with
and without TMJ disc displacement provides the
largest sample size of each and thus will provide
the best estimates of sensitivity and specificity.
Data presented from study 2 might result from
examination of a group of subjects who are seek-
ing general dental care. In the second case, we
must be content with what we get and do not
specify how many subjects we will evaluate with
and without TMJ disc displacement. We will as-
sume that this second group of subjects yielded
25 with disc displacement and 75 without. Table
2 lists the test characteristics for these two stud-
ies. Careful inspection of this table, however, will
show that the test characteristics (sensitivity,
specificity, PPV, NPV, accuracy) are not all iden-
tical for the two studies. How is this possible?
There is one more key component that con-
tributes to the relationship between test charac-
teristics: the prevalence of disease. The preva-
lence of a disease or condition is the frequency
with which it is found in the population under
study. In study 1 in Table 2, TMJ disc displace-
ment was present in 50% of the study subjects,
whereas in study 2, it was present in only 25%.
Positive and negative predictive values depend
on the sensitivity, specificity, and the prevalence
of disease in the setting in which the test is being
used. If we know the sensitivity and specificity of
a test as well as the prevalence of the disease or
condition, we can calculate the PPV and NPV.
The mathematical formulas that define the rela-
tionship between the prevalence, predictive val-
ues, and sensitivity and specificity are as follows:
(prevalence X sensitivity)
PPV = —
[(prevalence X sensitivity)
+ (1 — prevalence) X (1 — specificity)]
and
[(1 — prevalence) X specificity]
NPV-
[(1 — prevalence) X specificity
+ prevalence X (1 — sensitivity)]
Why is it important to know the disease prev-
alence in your population of interest? The PPV
and NPV are the most clinically useful charac-
teristics of a test's performance. But they can
Home | TOC |          
Index
95
vary greatly depending on the prevalence of dis-
ease in the population that is under study. Prev-
alence of the same disease may vary among dif-
ferent populations.
The importance of prevalence rate can be
seen from the two hypothetic studies previously
presented in Table 2. Study 1 may have been
designed to assess the performance of the clini-
cal evaluation of the TMJ as a screening test for
TMJ disc displacement. The prevalence of the
disease in that study was 50% by design (equal
numbers of subjects with and without TMJ disc
displacement were examined). The PPV and
NPV were found acceptable at 67% and 75%,
respectively. Based on these encouraging results,
we will assume the clinical evaluation was then
used as a screening test for TMJ disc displace-
ment on a larger number of subjects, with lower
prevalence of TMJ (25%) than in study 1. The
drop of the PPV (percent of those with positive
clinical evaluation for TMJ dysfunction who ac-
tually had TMJ dysfunction) from 67% to 40%
may have been both surprising and disappoint-
ing. In study 1, the prevalence of the condition,
disc displacement, is 50%, and the correspond-
ing PPV is 67%. In study 2, the prevalence is
25%, and the PPV is 40%. If the prevalence in
another population of interest, perhaps a differ-
ent ethnic or age group, is only 10% and screen-
ing is accomplished by using the clinical exami-
nation, the corresponding PPV would be only
19%. In this last example, 19% of those who test
positive for the condition would actually have
the condition. Furthermore, 81% (approximately
8 of 10 subjects) who test positive based on the
clinical examination would not have the condi-
tion confirmed by MRI. This is a disappointing
finding indeed and shows why new tests that
seem promising at first evaluation may not im-
prove diagnostic capability when put into com-
mon use.
The previously mentioned examples show the
importance of examining a test's performance
characteristics and the disease prevalence when
planning a research study. To aid in understand-
ing the relationship between PPV and NPV and
prevalence, a graphical approach may be help-
ful. Figure 1 shows such a display using data
from our hypothetic example mentioned previ-
ously. Let's assume that our screening test, the
detection of TMJ disc displacement by clinical
examination, has a sensitivity of 80% and speci-
<<    
     Article
      >>
96 Susan P. McGorray
SENSITIVITY = 80 % SPECIFICITY = 60%
25%
8-
LLJ
D
LJJ
>
O
Q
CC
Q.
s- PPV
20 40 60 100
PREVALENCE (%)
NPV = 90 % and PPV = 40 % AT 25 % PREVALENCE
NPV = 75 % and PPV = 67 % AT 50 % PREVALENCE
Figure 1. PPV (dashed) and NPV (solid) as a func-
tion of prevalence for a test with 80% sensitivity and
60% specificity. The squares correspond to the pre-
dictive values in Table 2, study 1, and the circles
correspond to the predictive values in Table 2,
study 2.
ficity of 60% as noted at the top of Figure 1. The
graph allows us to visualize the NPV and PPV
curves for this level of sensitivity and specificity
at different levels of disease prevalence. The
y-axis represents the percentage predictive value
of both the NPV and PPV curves. The x-axis
represents different levels of the disease preva-
lence (TMJ disc displacement). If the prevalence
of disc displacement in our population of inter-
est were 25%, we see that a vertical line at 25%
prevalence (x-axis) would cross the PPV and
NPV curves in this figure at PPV = 40% and
NPV = 90% (closed circles). If, however, the
prevalence of TMJ disc displacement were 50%
in our population, the vertical line at 50% prev-
alence would cross the PPV and NPV curves at
67% and 75%, respectively (closed boxes). Thus,
we see visually that the PPV and NPV change as
disease prevalence changes.
The PPV and NPV curves are unique for spe-
cific sensitivity and specificity combinations (ie,
sensitivity = 80%, specificity = 60%). In Figure
1, one can see that for a given sensitivity and
Home | TOC |          
Index
specificity, the PPV decreases and the NPV in-
creases as the prevalence of a disease or condi-
tion decreases. The reverse occurs as prevalence
increases.
Figure 2 is an example of the PPV and NPV
curves for higher values of sensitivity and speci-
ficity (sensitivity = 95%, specificity = 95%).
Note how the shape of the PPV and NPV curves
differs from Figure 1. For the test shown in
Figure 1 (which has lower sensitivity and speci-
ficity), the prevalence of the disease or condi-
tion has greater impact on the usefulness of the
test than it does for the test depicted in Figure 2
(higher sensitivity and specificity). We can see in
Figure 1 that the PPV drops off quickly as the
prevalence decreases from 50% and the negative
predictive value drops off quickly as the preva-
lence increases from 50%. This pattern is simi-
lar, but not as dramatic, in Figure 2. Thus, over
a wide range of disease prevalence the test
shown in Figure 2 (higher sensitivity and speci-
ficity) would be more useful than the test in
Figure 1 because NPV and PPV values change
less over a range of prevalences. By using graphs
of this sort, one can visually assess how the pre-
SENSITIVITY = 95 % SPECIFICITY = 95%
25%
8-
LU
=>
8-
LU
1
Q
LU
OC
Q_
O 20 40 60 80 100
PREVALENCE (%)
NPV = 98.3% and PPV = 86.4 % AT 25% PREVALENCE
Figure 2. PPV (dashed) and NPV (solid) as a func-
tion of prevalence for a test with 95% sensitivity and
95% specificity. The predictive values that would be
obtained from a study with a 25% prevalence of dis-
ease are indicated by diamonds.
<<    
     Article
      >> Home | TOC |          
Index

Sensitivity, Specificity, and Related Concepts 97

dictive values of a test will change if the test is
used as a screening test in a research study in
which the prevalence of the disease or condition
is lower or higher than that under which the test
was initially evaluated. If you know the test's
sensitivity and specificity, you can consult the
graph with that sensitivity/specificity to estimate
the PPV and NPV of the test for the level of
disease prevalence in your population. This will
help you decide how useful a screening test
would be in identifying disease in your popula-
tion of interest.
The oversimplification of the disease process
may also affect estimates of sensitivity and spec-
ificity. Often diseases or conditions are more
complex than simply being present or absent,
and some criteria or cut point must be specified
to classify subjects as having or not having the
disease. Examples include hypertension, cancer,
and need for orthodontic treatment. Various
sliding scales of need for orthodontic treatment
are commonly used by public health agencies
for prioritizing funds for orthodontic treatment.
For example, if need for orthodontic treatment
is classified as none, optional, recommended, or
necessary, based on some defined set of criteria,
an agency may allocate funds for treatment to
only those patients who meet the criteria for the
necessary level of treatment need. Alternatively,
funds could be allocated for those patients who
meet the criteria for levels of recommended and
above. Thus, the cut point for defining orth-
odontic treatment need in the first instance
would be necessary, and only those patients who
meet those criteria would qualify for care. In the
second instance, the cut point would be recom-
mended, and all those patients who meet criteria
for both the recommended and the necessary
levels would qualify for orthodontic care. When
differing definitions of a disease or condition
are used, sensitivity, specificity, and prevalence
estimates can vary, and studies that use different
definitions cannot be compared in a meaningful
way.
Accuracy and Agreement
The use of one characteristic, such as accuracy,
to summarize the diagnostic capabilities of a test
is appealing but must be approached with cau-
tion. Consider the two studies presented in Ta-
ble 3. We will assume that 1,000 subjects were
tested in each case. Although the sensitivity and
specificity are both higher for study 2, the accu-
racy is higher for study 1. In study 1, the preva-
lence of the disease is low at 50/1,000 (number
of patients with the disease present in popula-
tion of 1,000). When the prevalence of a disease

Table 3. Observed and Expected Cell Frequencies and Test Characteristics for Two Diagnostic Tests
True Disease Status

Test Result Present (+) Absent (—) Total

Study 1 (diagnostic test 1)

Positive ( + ) observed frequency 25 95 120
(expected frequency) (6) (114)
Negative ( — ) observed frequency 25 855 880
(expected frequency) (44) (836)
Total frequency 50 950 1000
Sensitivity: 25/50 = 50%
Specificity: 855/950 = 90%
Accuracy: (25 + 855) /1 000 = 88%
K statistic = 0.24
Study 2 (diagnostic test 2)
Positive ( + ) observed frequency 200 35 235
(expected frequency) (70.5) (164.5)
Negative ( — ) observed frequency 100 665 765
(expected frequency) (229.5) (535.5)
Total frequency 300 700 1000
Sensitivity = 200/300 - 67%
Specificity = 665/700 = 95%
Accuracy = (200 + 665)/1000 = 87%
K statistic = 0.66
NOTE. The expected frequencies are the counts that would be expected if there is no relationship between the diagnostic test
and the true disease status.
<<    
     Article
      >>
98 Susan P. McGorray
is low, many subjects will truly not have the
disease and also test negative (true negatives).
Because of this, the number of true negative
subjects is high and will account for a large
proportion of the accuracy of the test. This can
obscure other test information. True negative
subjects account for 97% of the accuracy (855/
[855 +25]) in study 1, but only 77% of the
accuracy (665/[665 + 200]) in study 2. Disease
prevalence must be similar in both populations
when comparing the accuracy of tests from dif-
ferent studies. Both components of accuracy,
the true negatives and the true positives, should
be examined. When the disease prevalence dif-
fers, it is not appropriate to compare the accu-
racy of two tests.
An alternative approach for assessing the
overall accuracy or agreement of a test involves
taking account of agreement that would occur
purely by chance (ie, if the test results provided
no information about the true disease status).
The statistic that captures this kind of agreement
is called the kappa statistic2 and its calculation
can be found in many general statistics books.
For the purposes of this article, it suffices to
know that, in brief, the kappa statistic depends
on the observed agreement (between the screen-
ing test and gold standard) and the expected
agreement that would occur by chance alone. If
the agreement between the screening test and
the true disease status (determined by the gold
standard) is perfect, the kappa statistic is 1.00. If
the agreement we observe is equal to what would
be expected by chance alone (ie, not very good),
the kappa statistic is 0. Negative kappa values
indicate a very bad situation when agreement is
less than would be expected by chance alone.
Guidelines for interpretation of the kappa statis-
tic have been proposed,3 with kappa values less
than 0.40 indicating poor agreement beyond
chance, 0.40 to 0.75 indicating fair to good
agreement, and greater than 0.75 indicating ex-
cellent agreement.
In our example depicted in Table 3, the
kappa statistic for study 1 is 0.24 and for study 2
is 0.66. Like the PPV and NPV, the kappa statistic
is also affected by prevalence.4 Kappa values can
vary quite dramatically for a given level of accu-
racy, depending on the underlying expected
agreement. For example, consider two cases in
which the observed agreement is relatively high
at 85%. If the expected agreement is 50% in the
Home | TOC |          
Index
first case and 78% in the second case, the kappa
statistics are quite different, 0.70 and 0.32, re-
spectively. It is also possible for the accuracy of
one test to be higher than that of a second test,
but the kappa statistic may be larger for the
second test. This is shown in Table 3, with accu-
racy of 88% and a kappa statistic of 0.24 for study
1 and accuracy of 86.5% and a kappa statistic of
0.66 for study 2. Because of the prevalence of the
disease and the proportion of positive test re-
sults, we would expect more agreement caused
by chance in study 1, leading to higher accuracy.
However, much of the accuracy is caused by
chance alone rather than the relationship be-
tween test result and disease status. These para-
doxes point out that both the accuracy and the
chance-corrected agreement (ie, kappa statistic)
should be considered when evaluating a test.
One measure is not enough to provide a com-
plete picture.
Diagnostic Likelihood Ratios
The use of diagnostic likelihood ratios5 to de-
scribe test performance is becoming more com-
mon. This is a fairly complex concept to under-
stand and is mentioned in this article only for
the sake of completeness. The use of diagnostic
likelihood ratios is appealing because they pro-
vide information useful for ruling in or ruling
out disease, much like predictive values, yet un-
like the previous test characteristics we have dis-
cussed, are not influenced by prevalence of the
disease. In addition to their usefulness as de-
scriptive characteristics of diagnostic or screen-
ing tests, more complex evaluation of screening
tests can be done by statistical modeling. This
level of statistical sophistication, however, is
likely beyond the grasp of most clinicians, and
enlisting the expertise of a statistical consultant
is highly encouraged.
Receiver Operating Characteristic
Curves
The final section of this article discusses a
method that is becoming more widespread for
the interpretation of the sensitivity and specific-
ity characteristics of new diagnostic tests. The
method is based on receiver operating charac-
teristic (ROC) curves.
In the discussion thus far, we have considered
<<    
     Article
      >>
Sensitivity, Specificity, and Related Concepts
screening or diagnostic tests that result in only
two outcomes (eg, positive or negative results).
This is not always the case. Some tests may have
more than two outcomes. For example, the as-
sessment of orthodontic treatment need may
yield ordered categories such as in our example
previously mentioned in which treatment need
may range from none, optional, recommended,
or necessary (four possible outcomes). From this
information, a binary decision (only two choices
such as yes/no) may still be required. In our
treatment need example, the two choices would
be either that the patient should receive treat-
ment or the patient should not receive treat-
ment.
We will show this concept with a hypothetic
study of dentists referring pediatric patients for
orthodontic evaluation. In this example, each
child is classified into one of the following six
levels of a referral scale: definitely refer, proba-
bly refer, possibly refer, possibly not refer, prob-
ably not refer, and definitely not refer. We also
assume that a gold standard is available to iden-
tify which children actually do and do not need
orthodontic treatment. Frequently, this determi-
nation is based on the consensus opinion of
experienced orthodontists who are considered
the gold standard. The method of identifying
treatment need based on pediatric referral pat-
terns is thus considered the screening test. We
will use the sensitivity and specificity of this
screening test to generate ROC curves as de-
scribed later.
Data from this hypothetic study are presented
in Table 4. From the dentists' referral scale of
treatment need, different cut points are exam-
ined to determine which children would receive
an orthodontic evaluation based on the referral
Table 4. Classification of Subjects by Orthodontic
Referral Status (Rows) and True Orthodontic
Treatment Need (Columns)
Treatment Need (True Status)
Do Not Need Need
Orthodontic Referral Status Treatment Treatment
Definitely refer 3 17
Probably refer 4 26
Possibly refer 10 40
Possibly not refer 35 15
Probably not refer 25 5 30
Definitely not refer 19 1 20
Total subjects 96 104
Home | TOC |          
Index
99
level (the screening test). A cut point is the
referral level where all patients on one side of
and including the cut point level are referred
and all patients on the other side are not re-
ferred to an orthodontist. Each possible cut
point level results in a decision rule that clearly
identifies which children should be referred. For
example, if we select probably refer as the cut
point, the decision rule is to refer all children in
the probably refer and definitely refer catego-
ries. Children in the other four categories are
not referred. A different cut point would result
in a different decision rule. By using the gold
standard to determine true need, one can deter-
mine the sensitivity and specificity of each deci-
sion rule (ie, how good is that decision rule in
distinguishing the patients who do and do not
need treatment as compared to the gold stan-
dard [the orthodontists]).
In Table 4, we know that 104 children have
true need based on the opinion of experienced
orthodontists (gold standard). We also see that
96 children have no need according to our gold
standard. If all children in the definitely refer
and probably refer categories are referred (43)
and others are not (89), the sensitivity of this
strategy is 41% (43/104), whereas the specificity
is 93% (89/96). Another cut point at the possi-
bly not refer level results in the following deci-
sion rule: referring all those in the possibly not
refer or higher likelihood of referral categories.
That would mean that 98 patients would be re-
ferred and 44 would not be referred. This yields
a sensitivity of 94% (98/104) and a specificity of
46% (44/96). By calculating the sensitivity and
specificity for each cut point level on the referral
scale, one obtains a series of points that can be
graphically displayed as an ROC curve (Fig 3).
Points in the upper left corner represent the
desirable situation of simultaneous high sensitiv-
ity and specificity. The ROC curve is formed by
joining the points resulting from each decision
rule, as shown in Figure 3. Each point represents
a sensitivity/specificity combination for each of
the referral levels on the referral scale. The y-
Total
axis is the scale of sensitivity values, and the
20 x-axis is the scale of specificity values. If both
30 high sensitivity and high specificity are impor-
50
50 tant (this is usually the case), the best decision
rule corresponds to the point closest to the up-
per left-hand corner of the plot. In our example,
200
the best cut point that jointly maximizes sensi-
<<    
     Article
      >>
100 Susan P. McGorray
8 -
20 40 60 80 100
[1 - specificity] (%)
Figure 3. The ROC curve for the data presented in
Table 4. The starred point represents the most desir-
able cut point on the referral scale indicating the
likelihood to refer for orthodontic evaluation. At this
cut point, patients in the referral groups designated as
definitely refer, probably refer, and possibly refer are
classified as needing orthodontic treatment. Patients
in the referral groups designated as definitely not
refer, probably not refer, and possibly not refer are
classified as not needing orthodontic treatment. The
sensitivity and specificity of this decision rule are 80%
and 82%, respectively. At this cut point, the sensitivity
and specificity of the decision rule are jointly maxi-
mized.
tivity (80%) and specificity (82%) classifies those
with referral status possibly or higher as needing
referral. This point can be easily identified from
the ROC curve displayed in Figure 3.
Other summaries of test performance can be
calculated from ROC curves. One common mea-
sure is the area under the ROC curve. A good
test or classification scale will have a cut point
associated with a point close to the upper left-
hand corner, which indicates 100% sensitivity
and 100% specificity. The closer the point is to
this corner, the larger the area under the ROC
curve. Larger areas correspond to better classi-
fication scales or tests. The area can be inter-
preted as the proportion of correct decisions in
a two-alternative forced choice experiment (eg,
treat/do not treat) .6 Further details about ROC
curves can be found in articles exploiting this
methodology.7'9 Two recent articles in the orth-
odontic literature that rely on ROC curve meth-
Home | TOC |          
Index
odology are by McGorray et al10 and Beglin
et al.11
Conclusion
Evaluation and comparison of test character-
istics is becoming more frequent in the dental
literature. Although characteristics of tests
have been used primarily for descriptive or
comparative purposes, recent articles show fu-
ture directions for their use. The practicing
clinician need not understand the complexi-
ties of statistical modeling that enable the use
of test characteristics in the actual design of
diagnostic tests, but familiarity with funda-
mental concepts will facilitate insight and crit-
ical evaluation of research that relies on such
methodology.
The examination of characteristics of
screening or diagnostic tests can provide
much useful information but must be done
with care. The setting in which a test is used
needs to be considered and comparisons of
characteristics made only when warranted.
Studies to estimate test performance charac-
teristics should be well designed. Aspects to
consider 12 include evaluation of potential bi-
ases such as incomplete diagnostic work-up
and subjects not representative of the popula-
tion on which the test will be used. Adequate
sample sizes are needed of those with and
without the disease or characteristic. Although
the question of does this patient have this
disease seems straightforward, the answer is
often complex, and better understanding and
use of diagnostic and screening tests will result
in an increased ability to answer this key ques-
tion.
References
1. Nebbe B, Major PW: Prevalence of TMJ disc displace-
ment in a pre-orthodontic adolescent sample. Angle
Orthod 70:454-463, 2000
2. Fleiss JL. Statistical Methods for Rates and Proportions
(ed 2). New York, NY, Wiley, 1981
3. Landis JR, Koch GG: The measurement of observer
agreement for categorical data. Biometrics 33:159-174,
1979
4. Feinstein AR, Cicchetti DV: Higher agreement but low
kappa: I. The problems of two paradoxes. J Glin Epide-
miol 43:543-549, 1990
5. Boyko EJ: Ruling out or ruling in disease with the most
<<    
     Article
      >>
Sensitivity, Specificity, and Related Concepts
sensitive or specific diagnostic test: short cut or wrong
turn? Med Decis Making 14:175-179, 1994
6. Green DM, Swets J: Signal Detection Theory and Psycho-
physics. New York, NY, Wiley, 1966
7. Hanley JA: Receiver operating characteristic curves
(ROC), in Armitage P, Colton T (eds): Encyclopedia
of Biostatistics. New York, NY, Wiley, 1998, pp 3738-
3745
8. Pepe MS: Three approaches to regression analysis of
receiver operating characteristic curves for continuous
test results. Biometrics 54:124-135, 1998
9. Beam D, Wright J, Kay E, et al: Perceptions of orth-
odontic treatment need: Receiver operating charac-
Home | TOC |          
Index
101
teristic analysis. Commun Dent Oral Epidemiol 24:
303-306, 1996
10. McGorray SP, Wheeler TT, Keeling SK, et al: Evaluation
of orthodontists' perception of treatment need and the
peer assessment rating (PAR) index. Angle Orthod 69:
325-333, 1999
11. Begun FM, Firestone AR, Vig KW, et al: A comparison of
the reliability and validity of 3 occlusal indexes of orth-
odontic treatment need. Am J Orthod Dentofacial Or-
thop 120:240-246, 2001
12. Gruggenmoos-Holzmann I, van Houwelingen HC: The
(in)validity of sensitivity and specificity. Stat Med 19:
1783-1792, 2000
<<    
     Article
      >> Home | TOC |          
Index

A Clinical Orthodontist Looks at Statistics

Sheldon Baumrind

Statistics is a tool that can help the practitioner judge the clinical impor-
tance of patient-centered studies. Statistical tools have both strengths and
limitations. This article endeavors to clarify for the clinician the role of
statistical analyses in facilitating the application of the outcomes of clinical
studies to the management of individual patients. (Semin Orthod 2002;8:
102-109.) Copyright 2002, Elsevier Science (USA). All rights reserved.

cal reasoning, properly applied, aid us in treat-

A ll clinicians, including myself, have at best
an ambivalent attitude toward statistics.
The subject area is large, complex, highly math-
ematical, and generally outside our area of pro-
fessional competence. Although all orthodontic
specialty programs now incorporate courses on
statistics into their graduate curricula, these
courses rarely take up more than one lecture
hour a week for one or two semesters. Mean-
while, at the University of California Berkeley,
Berkeley, CA, a representative first rank univer-
sity campus, there are at least three different
PhD level programs in statistics, no two of which
overlap in more than 10% of their required
courses. And in addition to the complexity of
the subject, every one of us has heard the old saw
about "liars and statistics." Why should such a
complex and esoteric discipline be of concern to
us?
Statistics is important to clinical orthodontics
only to the extent that it can improve the quality
and efficiency with which clinicians meet the
needs of their patients. That is the criterion by
which we should judge. The fact that statistics
can be used to misrepresent is true but not to
the point. Any tool in the clinical armamentar-
ium can be used badly, and it should not sur-
prise us that statistics is no exception to this rule.
The proper question is rather, "how can statisti-
From the Department of Orthodontics, University of the Pacific,
San Francisco, CA.
Address correspondence to Sheldon Baumrind, DDS, School of
Dentistry, University of the Pacific, 2155 Webster Street, San Fran-
cisco, CA 94115-2399.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0802-0009$35.00/0
doi:10.1053/sodo.2002.32192
ing our patients?"
Given the enormous scope of the question, it
is only possible to touch on a few highlights in
this article. But it is possible to alert the reader
to a few aspects of statistics that are of direct
relevance to those of us who treat orthodontic
patients. From that perspective, this article pro-
poses to comment briefly on the implications of
variability; measurements of central tendency
and dispersion; enumeration versus inference;
the basic principle of inferential statistics; the
relationship between sampling and generaliz-
ability; statistics about groups versus statistics
about individuals; statistical significance, confi-
dence intervals, effect size, and clinical signifi-
cance; and causal inference.
The Implications of Variability
If all patients were the same with respect to some
property of interest (eg, Angle Class), it would
only be necessary to measure one patient to
determine the correct value for all patients for
that property. If all patients were the same with
respect to all properties of interest, we could
learn all we need to know by measuring any one
patient, rendering further investigation unnec-
essary. But in the real world, not all patients are
the same, and it is the reality of variability that
makes it necessary to measure many subjects,
even of the same type.
Measurements of Central Tendency and
Dispersion
When we measure each subject in a group with
respect to any property of interest (eg, total face

102 Seminars in Orthodontics, Vol 8, No 2 (June), 2002: pp 102-109

<<    
     Article
      >>
A Clinical Orthodontist Looks at Statistics
height or mandibular plane angle) we get a
collection of measurements, one for each sub-
ject in the group. Such a collection is called a
distribution. Because distributions for any sub-
stantial group of subjects are necessarily diffuse,
the need arises to develop a shorthand descrip-
tion of the group's properties with respect to
each variable of interest. At minimum, we need
for this purpose an estimate of averageness
(called a measure of central tendency) and an
estimate of variability (called a measure of dis-
persion). The most commonly used estimate of
averageness is the arithmetic mean and the most
common estimate of variability (ie, of the aver-
age deviation from the average), is the standard
deviation. Suppose, for example, we knew (from
well-conducted prior studies) that the mean
value for increase in mandibular body length for
girls between 8.5 and 15.5 years is 6.3 mm with a
standard deviation of 5.2 mm. If we wished to
apply this knowledge to the prediction of the
growth of each of a series of individual female
patients who successively occupy our chair, our
best estimate for the growth of each patient
would by 6.3 mm (the mean). But it is also the
case that if we used this mean value as a predic-
tion of the expected size increase for each indi-
vidual, the average error would be approxi-
mately 5.2 mm (ie, one standard deviation) per
patient.
Thus, the size of the standard deviation gives
the wary clinician an important clue about the
confidence with which he can use the mean to
estimate expected events in growth or treat-
ment. The larger the standard deviation, the
greater the uncertainty is that accompanies the
use of the mean as a predictor in individual
cases. And as the size of the standard deviation
approaches the size of the mean, the mean be-
comes progressively less useful for predicting
expected values in individual patients, although
it may continue to be very useful as a ball-park
estimate.
Enumeration Versus Inference
One major use of statistics is to tally conditions
or relationships within one or more groups as an
end in itself. Examples include bracket inven-
tory, census taking, and vote counting. Such ex-
ercises fall under the heading of enumerative
statistics. Whenever tallies are conducted, issues
Home | TOC |          
Index
103
such as bias in sampling and control for mea-
surement error always arise. And sometimes (as
in our last Presidential election), they assume
major proportions. But beyond these consider-
ations, enumerative statistics is conceptually
fairly straightforward.
There is, however, another general area of
statistics (of greater interest to us here) in which
information gathered from some groups is used
to infer present or future conditions in other
groups. When, for example, we study the growth
trajectory of a sample of subjects for whom metal
implants were placed 40 years ago, we are not
primarily interested in those particular subjects.
Rather, we are interested in what the study of
those earlier subjects can tell us about the likely
growth trajectory of the patients we are going to
treat from now on. The discipline that deals with
valid extrapolation of findings between groups is
called inferential statistics. It is a subject of major
concern to clinical practice and, as might be
expected, its reasoning is considerably more
complex than the reasoning of simple enumer-
ation. The remainder of this article will outline
some aspects of the rationale of statistical infer-
ence.
The Basic Principle of Inferential
Statistics
In the simplest situations in which statistical in-
ferences need to be made, we are usually trying
to identify modal differences between two kinds
of individuals. (For instance, we may be trying to
measure differences between treated individuals
and untreated control individuals or to measure
differences between two groups of individuals
that have received different kinds of treatment.)
For these purposes, it is necessary to sample
without bias the collection (or population) of all
subjects who received each kind of treatment,
and then to compare the samples. If all the
individuals within one sample measure the same
and all the individuals within the second sample
measure the same, then it is only necessary to
measure one member of each sample and then
to compare the two measurements. If the two
measurements do not differ (within our toler-
ances), we may reasonably conclude that there is
no substantial difference between the two
groups. On the other hand, if the two measure-
ments differ substantially, then we can fairly rea-
<<    
     Article
      >>
104 Sheldon Baumrind
sonably conclude that the two samples were
drawn from different populations. (Note that
this is a fair description of what happens when
you take one central incisor bracket from each
of two shipments and compare them to see if the
two shipments are consistent.)
However, identifying differences between two
samples of patients who have received different
orthodontic treatments is not that simple, in
large part because there is variability between
the individual subjects within each treatment
group. Given that variability, it might seem the
most obvious thing to do would be to compare
the averages of the two samples. One might, for
example, measure all the members of each sam-
ple, calculate the average (or arithmetic mean)
for each sample, and then compare the two
means. Unless our measuring instrument had
been quite crude, there would almost certainly
be a difference between the two means, and
some inference could be drawn.
But let us consider further the risks of such an
inference. Clearly, simply as a consequence of
individual variation, even different samples
drawn from the same population should be ex-
pected to have somewhat different means.
Hence, the mere fact of small differences in
mean does not establish that the populations
from which the two samples were drawn were
different. How then can we be at least reasonably
sure that the observed difference between the
means of different samples in any particular ex-
periment is greater than might have occurred by
chance if there really was no difference in the
populations from which they were drawn? This is
equivalent to asking, given the variability in the
populations from which the different samples
were drawn, what is the likelihood that two sam-
ples with means as different as the ones we ob-
serve could have occurred by chance alone?
It seems intuitively obvious that the smaller
the sample sizes and the greater the variability
within the populations, the larger will be the
chance that the means of two or more samples
will differ markedly by chance alone. In any
carefully conducted study, we know the sizes of
the samples, and we can pretty well estimate the
variability of the populations from which they
are drawn by measuring the variability within the
samples themselves. We can therefore establish a
relationship in which the measured difference
between the means (in the numerator) is di-
Home | TOC |          
Index
vided by an estimate of the pooled sample vari-
ability in the denominator. The larger the dif-
ference is between the means as compared with
the measure of variability, the greater the likeli-
hood is that two samples are truly from different
populations.
This critical examination of the relationship
between size of the average differences between
groups and the amount of variability among the
individuals who comprise the groups is the hall-
mark of all the tests that comprise inferential
statistics. The variability observed may be from
actual differences between individuals or from
noise associated with the imprecision of the
measurement process. In either event, all infer-
ential tests, whether simple or complex, are de-
signed to protect the user against the risk of
confusing true between-group differences with
chance occurrences during sampling and mea-
surement.
The Relationship Between Sampling and
Generalizability
Although it may not seem immediately evident,
the manner in which samples for clinical studies
are gathered is of great importance to the clini-
cian seeking to apply findings from the studies
to the treatment of the patient who actually
occupies his or her chair. In evaluating findings,
the clinician inevitably must ask, "are the pa-
tients in this study similar to mine?" If the answer
to this question is either uncertain or in the
negative, then the research findings, although
they may be of academic interest, are of greatly
reduced practical interest to the clinician as a
clinician. For this reason, articles reporting in-
formation on clinical studies have the absolute
requirement of describing sample composition
and sampling methods completely and precisely.
Statisticians describe the property of a study's
findings being directly applicable to the clinical
situation as generalizability. In most medical tri-
als, generalizability is insured by protocols that
specify the criteria for inclusion in the research
sample. Such a protocol typically describes inclu-
sion and exclusion criteria concerning age, gen-
der, precise diagnosis, precise previous course of
illness, and precise previous treatment. The cli-
nician can then directly compare the profile of
his current patient with the protocols of relevant
clinical trials and determine which previous
<<    
     Article
      >>
A Clinical Orthodontist Looks at Statistics
studies have findings that bear directly on the
treatment of each particular patient in his or her
practice. Thus far, very few studies in our spe-
cialty specify the nature of their samples well
enough to allow the clinician to estimate the
applicability of their findings to the treatment of
specific patients with confidence.
As an example, I would like to cite the limi-
tations of one of our own studies that was rela-
tively well received by the specialty a few years
ago.1'2 At that time, we sought to conduct a
prospective randomized clinical trial to study the
differences in outcome of extraction versus non-
extraction strategies in the treatment of border-
line Angle Class I and Class II patients. Figure 1
represents the design of our study. An ethical
and scientific precondition for such randomiz-
ing of patients to either extraction or nonextrac-
tion therapy is the requirement that the investi-
gators establish for each individual patient that
there exists true uncertainty as to which treat-
ment is preferable for that particular patient. On
reflection, the reader will recognize that when-
ever a single clinician makes treatment deci-
sions, there are actually no borderline patients.
Rather, after due deliberation, the clinician al-
ways decides either to extract or not to extract.
In the absence of any consensually accepted
protocol for making the extraction/nonextrac-
tion decision, our method for establishing true
uncertainty was to have each of a number of
experienced clinicians evaluate the pretreat-
ment records for each potential subject inde-
Follow-up
Assign to clinician preferring Tx 1 Assign to clinician preferring Tx 2
(with balancing) (with balancing)
*• Follow-up *
Figure 1. Design of a prospective randomized clinical
trial in which orthodontic patients were randomized
to extraction or nonextraction treatment based on
clinicians' disagreement as to which course of treat-
ment was indicated.3
Home | TOC |          
Index
105
pendently (step 3, Fig 1). The group of patients
for whom some judges advocated extraction
treatment (whereas others advocated nonextrac-
tion treatment) was then considered to consti-
tute a borderline sample.3
Before presenting any cases to the clinician
panel, we attempted to develop a set of objective
criteria that would allow us to segregate out all
or most of the straightforward cases for which
the judges' decision to extract or not to extract
would have been unanimous, (steps 1 and 2, Fig
1.) The construction of such a set of objective
criteria would have yielded two advantages. First,
it would have permitted us to reduce the work-
load of our clinician/judges (each of whom had
an active private practice). Second and more
important, it would have allowed us to codify the
characteristics of the group of patients to which
the results of our study could later be general-
ized appropriately. Unfortunately, although our
study produced a number of interesting and
useful findings, it was not successful in identify-
ing any unique set of objective measures that
defines precisely the individual patients in any
clinician's practice to which the study's results
appropriately generalize.
It should be noted that Johnston and col-
leagues confronting the same problem success-
fully used the method of discriminant analysis to
define the characteristics of a group of subjects
who, on average, could be considered border-
line in the sense that different clinicians would
tend to treat them differently.4'5 This is certainly
a useful contribution and a move in the right
direction, but it does not yet tell the clinician
whether or not the results from the experimen-
tal sample are properly generalizable to the par-
ticular patient in the chair.
Statistics About Groups Versus
Statistics About Individuals
Almost all clinical investigations in medicine and
dentistry involve the investigation of some indi-
viduals treated in the past or present for the
purpose of accumulating information that can
be applied to the care of other individuals who
will be treated in the future. The individuals who
are subjects of our investigations are usually
studied as members of groups called samples.
Users of statistical findings differ in their profes-
sional needs and orientations. Depending on
<<    
     Article
      >>
106 Sheldon Baumrind
the nature of the problems that concern them,
some users focus their interest primarily on the
characteristics of individual subjects, whereas
other users are more interested in the charac-
teristics of the groups to which the individuals
belong. Surprising as it may seem, the kinds of
research design and statistical treatment that are
appropriate to the collection and analysis of data
differ substantially, depending on whether the
investigation is focused primarily on the perfor-
mance of individuals or on the performance of
groups.
Allow me to expand on this surprising asser-
tion. By far the largest investments in clinical
research in dentistry, both in the United States
and elsewhere, are and probably should be in
the domain of public health. Studies in this do-
main have as their main goal not the treatment
of individual patients but rather the establishing
of public health policy and the improvement of
the health of the public in general. For the
purposes of such studies, the investigators are
interested primarily in group trends rather than
in individual outcomes. When, for example, the
US Public Health Service achieved its enor-
mously successful test of the efficacy of fluoride
administration in the reduction of tooth de-
cay,6'8 major concern was not with the caries rate
for any particular individual but rather with the
overall effect of fluoride administration on the
population as a whole. The goal of the investi-
gators was to achieve a measurable reduction in
tooth decay across the entire population, with-
out much concern about which particular indi-
viduals benefited and which did not. The exper-
imental strategy was to set up one of two similar
communities with fluoridated drinking water
and another without and then to measure some
simple measure of difference in decay rate in the
two communities through time. Thus, to estab-
lish the usefulness of the treatment, it was only
necessary to determine that on average the per
capita rate of decayed, missing, and filled teeth
was lower in the fluoridated community than in
the nonfluoridated community.
The general strategy of epidemiologic studies
such as the one just cited is to make some sim-
ple, easily made, and relatively crude measure-
ments across a relatively large number of indi-
viduals in the hope of showing that the average
differences between two groups being compared
are larger than can be accounted for merely by
Home | TOC |          
Index
chance differences in sampling. An example of
an analogous epidemiologic measurement in
orthodontics is the use of the peer assessment
rating index in Britain to monitor the efficacy of
the distribution of the limited funds allocated to
the treatment of malocclusion by the British
national health service.9'12
As we have noted, research designs such as
the ones previously mentioned can be very im-
portant and useful in public health dentistry, but
they are of less use in the understanding of
individual variation or the differential planning
of individual treatment, which is to say for an-
swering the kind of questions that are the imme-
diate focus of interest of clinicians who treat
patients one at a time. For such purposes, one
needs to study in depth the way in which clinical
treatment is actually delivered. Such studies in-
volve sampling of multiple records of different
kinds for the same patient, just as the conscien-
tious clinician does when planning treatment.
This is to say, we must sample a larger portion of
the total variance within each subject patient by
measuring head films and study casts and in-
traoral x-rays and facial photos. Additionally, we
must also measure many more parameters per
individual, and we must accomplish these mea-
surements with much greater precision and ac-
curacy. These tasks involve considerable labor
and tend to be less attractive to public health
funding agencies than are more dramatic single-
issue epidemiologic studies. For this reason and
others, consequential support for clinical studies
on the details of orthodontic treatment over the
next few years will have to come from the orth-
odontic specialty itself, acting through agencies
like the American Association of Orthodontists
Foundation.
Statistical Significance, Clinical
Significance, and Effect Size
One of the most confusing areas in statistics
involves the use of the term significance. This
common English word is defined in my dictio-
nary as "[q]uality of being important; . . . mo-
ment; . . . consequence."13 But that is not at all
what is meant when the term statistically signifi-
cant is used. To examine what the word signifi-
cant means when used in statistics, we must re-
visit the issues of variability and inference that
were discussed earlier. As has been noted (or at
<<    
     Article
      >>
A Clinical Orthodontist Looks at Statistics
least implied), the essential task of inferential
statistics is to determine what can reasonably be
concluded about a population (eg, the popula-
tion of all Class II, high-angle cases) on the basis
of having examined only a sample consisting of
a limited number of individuals drawn from that
population. The sample is thus a relatively small
window through which the investigator hopes to
see the outlines of some larger, more inclusive
reality. In some cases, such a glimpse may truly
represent the larger reality, whereas in other
cases it may misrepresent it, leading the investi-
gator to erroneous conclusions. This is the case
because whenever there is random variability
inherent in the phenomena under investigation,
there is the possibility that what is observed re-
sults from nothing more than chance coinci-
dence—that if a different sample had been
drawn, examination might have led to markedly
different conclusions. Until the possibility that
the findings of a study reflect nothing more than
coincidence has been addressed and rejected,
no conclusions one way or another can reason-
ably be drawn.
By virtue of extensive study of patterns of
variability within groups of subjects, statisticians
have devised methods of conservatively estimat-
ing the variability of populations from the vari-
ability within samples drawn from them. That
being the case, it is possible to estimate mathe-
matically the probability that two samples of
given sizes and internal variability could by
chance alone have differed in mean value as
much as they do, even if both samples had been
drawn from the same homogeneous population.
If the numbers tell us that it is highly unlikely
that two such samples could reasonably have
been drawn from the same population, then we
reject the possibility that there is no difference
between them and conclude that they were
drawn from truly different populations. Of
course, there is always the possibility that two
considerably different samples could in fact be
drawn from the same population, in which case
our assertion of a difference between them is in
error. But by convention, if the numbers show
that the probability of such an error is less than
5% (or 1 chance in 20), we assert that the dif-
ference between two samples (or between one
sample with pre- and post-treatment difference
of zero) is statistically significant at the P < .05
level. Thus, we assert that the two samples were
Home | TOC |          
Index
107
drawn from different populations (ie, that there
is a real difference between the two treatments
that the samples represent). In doing so, we
accept the risk that there is 1 chance in 20 that
our assertion is in error. But there is an impor-
tant caveat to be observed here. If we do many
tests using this 5% yardstick, we are likely by
definition to have one or more spurious findings
of statistical significance merely by chance. For
example, one might (as our laboratory did some
years ago) test the differences associated with
three different strategies for maxillary retraction
compared with an untreated control group.14'15
If one then tests the differences between groups
two at a time, there are actually six tests for any
single variable. In our case, for example, one
could test distalization of the upper first molar
cusp in the following combinations: (1) control
versus cervical, (2) control versus high pull, (3)
control versus activator, (4) cervical versus high
pull, (5) cervical versus activator, and (6) high
pull versus activator. For this reason, one should
try to limit the number of statistical tests to be
performed and to preplan them by generating
testable hypotheses before data acquisition.
Note that the Pvalue, or risk of error we are
willing to accept, is merely a convention; we
may choose alternatively to set the level of
significance at P < .01 or even P < .001 where
we consider appropriate although we are
honor bound to set this standard before we
look at the data. Note also that the discussion
of statistical significance in this section has
really been a restatement, from an alternative
perspective, of the discussion about the basic
principle of inferential statistics earlier in this
essay.
Finally and most importantly, note that statis-
tical significance has nothing to do with impor-
tance. It deals only with the problem of control-
ling for chance errors in sampling. Of course,
one way to minimize chance errors in sample
composition is to use larger samples. But this
strategy can result in a paradox for the clinician
focused on individual case outcomes because as
sample sizes get larger (which orthodontists
have always thought of as an unmitigated boon),
smaller and smaller differences between group
means come to be statistically significant.
Clearly, in addition to assurances that the find-
ings we publish and read are not accidents of
chance, we need information about how impor-
<<    
     Article
      >>
108 Sheldon Baumrind
tant observed differences between experimental
groups really are.
Various criteria for identifying the impor-
tance of differences between groups being com-
pared have been developed during recent years
but it appears fair to say that the area is still in
flux. It does seem clear the finding of statistically
significant differences is of itself insufficient to
establish importance. However, for the examina-
tion of clinical findings with respect to any ma-
ture treatment modality, where it is reasonable
to expect sample sizes of at least 15 or 20 to be
available, the identification of statistically signif-
icant differences at least as strong at the P < .05
level would appear to be the minimum precon-
dition for serious further consideration. Given
the demonstration of statistical significance (ie,
of reasonable assurance that the observed effect
is greater than can be accounted for by chance
alone), one would next seek a reasonable esti-
mate of how large the difference between the
means of two samples (or between the mean of
one sample and pre- and post-treatment differ-
ence of zero) is compared with the variability of
the sample.16 Such estimates are called estimates
of effect size, and a number of different tech-
niques are used under different clinical condi-
tions. A commonly used measure of effect size
for correlations is R2, which identifies the per-
centage of the variance of y accounted for by
knowing x.
In some fields including our own, it has be-
come common in recent years to use the term
clinical significance as a measure of the impor-
tance of a measurement. The trouble with this
term is that it is poorly defined in most fields17
and pretty much undefined in our own. Indeed,
it seems fair to say that in orthodontics, clinical
significance, like beauty, resides in the eye of the
beholder. And, as in the case of beauty, each
beholder has a slightly different concept of ex-
actly what clinical significance is. That being the
case, it seems much more appropriate for orth-
odontic investigators to abandon the term clini-
cal significance entirely at least until the concept
is better defined. It would be much better for
authors to give the reader only the facts, which is
to say the raw means and standard deviations,
and let the clinician-reader decide what he/she
considers important.
Home | TOC |          
Index
Causal Inference
The last point I wish to touch on in this brief
discussion is the problem of causal inference. In
our field as in others, it is often very important to
identify differences between subjects who have
been treated in different ways (eg, differences in
outcome between Class II extraction patients
and Class II nonextraction patients or between
patients treated with a bionator and patients
treated with headgear). In the past, many orth-
odontic investigators have made such compari-
sons and then naively concluded that the ob-
served endpoint differences between groups
were caused by the act of tooth removal or by the
difference in appliance. Johnston has done the
specialty a great service in sharpening our aware-
ness that such causal attribution is warranted
only when one can be sure that the patients
who were treated by each of the methods
tested were equivalent before treatment started,
a state that can only be achieved through care
in sampling.4'5'18'19 (Studies in which the sub-
jects receiving alternative treatments were dif-
ferent before treatment commenced are said
to be characterized by susceptibility or selec-
tion bias.18'19) This does not, however, mean
that we should restrict our investigations only
to the study of subjects who were equivalent
before treatment. No thoughtful investigator
would think of randomizing bimaxillary protru-
sive patients with 12 mm of lower anterior
crowding to nonextraction treatment or Class I
subjects with 6 mm of anterior spacing to extrac-
tion treatment. Rather, we need information
about the central tendencies and variability of
the usual outcomes for many kinds of patients
when they are treated by the methods commonly
in use. But we must learn to differentiate be-
tween those characteristics that are truly caused
by treatment and those associated with the pa-
tient's pre treatment state and potential.
Another extremely common error of causal
attribution occurs when one of two factors that
are correlated with each other is considered to
be the cause of the other solely on the basis of
the correlation. In general, the caveat here is
that mere association between two events is
never of itself sufficient grounds for asserting
that one event caused the other. An easily un-
derstandable example of a strong association
that is not causal would be the correlation be-
<<    
     Article
      >>
A Clinical Orthodontist Looks at Statistics
tween condyle-pogonion distance and reading
comprehension in children between 6 and 16. It
would be improper for an educator to infer from
such a finding that improved reading compre-
hension causes jaw growth, just as it would be
improper for an orthodontist to conclude that
mandibular growth improves reading compre-
hension. Rather, both measures increase sharply
through time as a consequence of physical and
mental growth. (To make the same point, a
British biometrician pointed out in the 1920s
that during the previous 100 years there had
been a highly significant association between the
rise in clergymen's wages and the consumption
of beer. Clearly, it would have been inappropri-
ate without additional information to have con-
cluded that the vicars of the church had spent
their salary increases on alcoholic beverages.) A
similar conceptual problem somewhat more dif-
ficult to grasp occurs when lecturers in orthodon-
tics speak of the dental compensations, which are
presumed to occur as a consequence of certain
patterns of mandibular growth. Here it remains to
be established whether there does or does not
exist a causal relationship. Indeed, there may be
such a relationship but whether or not there is
cannot be established by correlation alone.
This article has attempted to focus on some of
the ideas involved in the use of statistics as a tool
for increasing our understanding of the complex-
ities of growth and treatment with regard to orth-
odontics. Like orthodontics, statistic analysis has
very real limitations. But properly used it does
allow us to identify, estimate, and to some extent
control for deficiencies in our capacity to predict
future events in an uncertain universe. This article
barely scratches the surface of the subject, but it
does seek to point a direction. The clinician-ob-
server needs to look very hard at the implicit and
explicit assumptions that underlie any statistical
analysis. A basic understanding of the logic of sta-
tistical inference is much more important to the
reader of the orthodontic literature than is the
ability to manipulate or remember the numbers.
References
1. Baumrind S, Korn EL, Boyd RL, et al: The decision to
extract: Part 1—inter-clinican agreement. Am J Orthod
Dentofac Orthoped 109:297-309, 1996
2. Baumrind S, Korn EL, Boyd RL, et al: The decision to
extract: Part 2—analysis of clinicians' stated reasons.
Am J Orthod Dentofac Orthoped 109:394-403, 1996
Home | TOC |          
Index
109
3. Korn EL, Baumrind S: Randomized clinical trials with
clinician-preferred treatment. Lancet 337:149-152, 1991
4. Paquette DE, Beattie JR, Johnston LE, Jr. A long-term
comparison of non-extraction and pre-molar extraction
edgewise therapy in "Borderline" Class II patients. Am J
Orthod Dentofac Orthoped 102:1-14, 1992
5. Cassidy DW, Herbosa EG, Rotskoff KS, et al: A compar-
ison of surgery and orthodontics in "borderline" adults
with Class II, Division 1 malocclusions. Am J Orthod
Dentofac Orthoped 104:455-70, 1993
6. Dean HT: The investigation of physiologic effects by the
epidemiological method, in Moulton FR (ed): Fluorine
and Dental Health. Washington, DC, American Associa-
tion for the Advancement of Science, 1942, pp 23-31
7. Dean HT: Epidemiological studies in the United States,
in Moulton FR (ed): Dental Caries and Fluorine. Wash-
ington, DC, American Association for the Advancement
of Science, 1946, pp 5-31
8. McClure FJ: Water Fluoridation: The Search and the
Victory. Bethesda, MD: US National Institute for Dental
Research 1970, pp 109-138
9. Brook PH, Shaw WC: The development of an index of
orthodontic treatment priority. Eur J Orthod 11:309-
320, 1989
10. Shaw WC, Richmond S, O'Brien KD, et al: Indices of
treatment need and treatment standards. Br Dent J 170:
107-112, 1991
11. Richmond S, Shaw WC, Roberts CT, et al: The PAR
Index (Peer Assessment Rating): Methods to determine
outcome of orthodontic treatment in terms of improve-
ment and standards. Eur J Orthod 14:180-187, 1992
12. Shaw WC, Richmond S, O'Brien K: The use of occlusal
indices: A European perspective. Am J Orthod Dentofac
Orthoped 107:1-10, 1995
13. Neilson, William Allan: Webster's New International Dic-
tionary of the English Language. 2nd ed. Springfield,
MA, Merriam-Webster, 1953
14. Baumrind S, EL Korn, Isaacson Rf, et al: Quantitative
analysis of the orthodontic and orthopedic effects of
maxillary traction. Am J Orthod Dentofac Orthoped
84:384-396, 1983
15. Baumrind S, EL Korn, Isaacson RJ, et al: Superimposi-
tional assessment of treatment-associated changes in the
temporo-mandibular joint and the mandibular symphy-
sis. Am J Orthod Dentofac Orthoped 84:443-465, 1983
16. Cohen J: Things I have learned (so far). Am Psychol
45:1304-1312, 1990
17. Kazdin AE: The meanings and measurement of clinical
significance. J Consulting and Clinical Psychology 67:
332-339, 1999
18. Johnston LE Jr, Paquette DE, Beattie JR, et al: The
reduction of susceptibility bias in retrospective compar-
isons of alternative treatment strategies, in Vig KD, Vig
PS (eds): Clinical Research as the Basis of Clinical Prac-
tice. Ann Arbor, MI, Center for Human Growth and
Development, 1991, pp 155-77
19. Johnston LE Jr: Clinical studies in orthodontics: Taking
the low road to Scotland, in Trotman C, McNamara JA Jr
(eds): Orthodontic Treatment: Outcome and Effective-
ness. Monograph 30, Craniofacial Growth Series, Center
for Human Growth and Development. Ann Arbor, MI,
The University of Michigan, 1995, pp 21-41
<<    
     Article
      >> Home | TOC |          
Index

presents two
new approaches to
, W.B. SAUNDERS Preadjusted
- Essential Information for Today's Dentists
Appliances...
from the Leading Health Care Publisher

American Journal of Orthodontics and Coming Soon!

Dentofacial Orthopedics
The Official Publication of the American Orthodontic
Association of Orthodontists, its constituent
societies, and the American Board
• IKE DEMOTION WITH
ImßPffiaBJÖSlID
Management of the
of Orthodontics
1 APPLIANCE
Dentition with
British Journal of Oral <3c Maxillofacial Surgery
the Preadiusted
Official journal of the British Association of Appliance
Oral & Maxillofacial Surgeons and a John C Bennett, PDS, DOrth;
Recognized Journal of the American College of and Richard P. Mclaughlin, DDS
Oral and Maxillofacial Surgeons
This innovative text promotes the individual
Dental Abstracts management of each tooth using the
Preadjusted Orthodontic Appliance. Practical,
Dental Clinics
well-illustrated instruction presents treatment
International Journal of Oral 8c Maxillofacial mechanics for common problems including
Surgery
crowding, spacing, tooth size discrepancies,
Official Publication of the International
Association of Oral & Maxillofacial Surgeons ectopic eruption, and impaction.
April 2002 • Approx. 380 pp., 1,390 illus.
The Journal of Evidence-Based 0-7234-3265-1 • $175.00
Dental Practice

Journal of Oral and Maxillofacial Surgery New!

Journal of Prosthodontics
iSTSTEMJEED
I'OilHODONnC
Systemized
Official Journal of the American Association of
I TREATMENT
IMECHANICS Orthodontic
Oral and Maxillofacial Surgeons Treatment Mechanics
The Journal of Prosthetic Dentistry Richard P. Mclaughlin, DDS;
John C. Bennett, FDS, DOrth;
Journal of Prosthodontics and Hugo Trevisi
Official Journal of the American College This new, up-to-date resource focuses on
of Prosthodontists
correcting malocclusion or misalignment of
Oral and Maxillofacial Surgery Clinics the teeth using the Preadjusted Appliance.
Extensively illustrated with line diagrams and
Atlas of the Oral and Maxillofacial
Surgery Clinics color photographs, this practical manual
provides clear coverage of each treatment stage.
Oral Surgery, Oral Medicine, Oral Pathology,
2002 • 335 pp., 765 illus. • 0-7234-3171-X • $149.00
Oral Radiology and Endodontics

Seminars in Orthodontics IMosby

I A Division of Elsevier Science

Year Book of Dentistry®

3 Easy Ways to Order
For more information, contact Periodicals Marketing at PHONE toll-free: 800-545-2522
(215)238-5614. FAX toll-free: 800-568-5136
Order via the internet at any time.-wwwjnosby.com
<<    
     Article
      >> Home | TOC |          
Index

Available from
W.B. Saunders!

Churchill
Livingstone
Journal of Arthroplasty
Official Journal of the American Association
of Hip and Knee Surgeons

Journal of Cardiac Failure

Official Journal of the Heart Failure
Society of America and the Japanese
Heart Failure Society

Journal of Electrocardiology
Official Journal of the International Society
for Computerized Electrocardiology and the
International Society of Electrocardiology

The Journal of Pain

Official Journal of the American
Pain Society

Seminars in Complementary
Medicine
FOR MORE INFORMATION ABOUT THESE TITLES,
PLEASE CONTACT:
Periodicals Marketing
W.B. SAUNDERS
A Division of Elsevier Science
The Curtis Center, Independence Square West
Philadelphia, PA 19106-3399
Phone: (215) 238-5614
©2002 Elsevier Science.
Or visit our home page at: www.wbsaunders.com
<<    
     Article
      >> Home | TOC |          
Index

W. B. S A U N D E R S

JOURNALS Essential Information f or Today's Professionals from theleadingHealth Caw Publisher

ANESTHESIOLOGY NEPHROLOGY PSYCHIATRY

Journal of Cardiothoracic and Vascular Anesthesia Advances in Renal Replacement Therapy—A Journal of Comprehensive Psychiatry—Official Journal of the A\
Journal of Vw—OfficiallJournal of the American Pain Society The National Kidney foundation Psycbopatbological Association
Regional Anesthesia and Pain Medicine American Journal of Kidney Diseases—The Official Journal Seminars in Clinical Neuropsychiatry
Offmal Journal of the American, Asian and (Oceanic, of The National Kidney Foundation
and Latin American Societies of RegionalAnesthesia Journal of Renal Nutrition—The Official Journal of The Council RADIOLOGY
Seminars in Anesthesia on Renal Nutrition of the National Kidney Foundation Seminars in Breast Disease
Seminars in Cardiothoracic and Vascular Anesthesia Seminars in Nephrology Seminars in Nuclear Medicine
Seminars in Pain Medicine Seminars in Radiologie Technology
Techniques in Regional Anesthesia and Pain Management NEUROLOGY Seminars in Roentgenology
Journal of Pain—Official Journal of the American Pain Society Seminars in Ultrasound, CT and MRI
CARDIOVASCULAR DISEASES Journal of Stroke and Cerebrovascular Diseases—Official Journal Techniques in Vascular and Interventional Radiology
Progress in Cardiovascular Diseases of the National Stroke Association and theJapan Stroke Society
Techniques in Intervention^ Cardiology Seminars in Cerebrovascular Diseases and Stroke RHEUMATOLOGY
Seminars in Pediatric Neurology Seminars in Arthritis and Rheumatism
CRITICAL CARE MEDICINE
Journal of Critical Care NURSING SURGERY
Advances in Neonatal Care—Official Journal of the National Journal of Pediatric Surgery—Official Journal of the Section on
DENTISTRY Surgery of the American Academy of Pediatrics, British Association
Journal of Oral and Maxillofacial Surgery—Official Journal of ofPaediatric Surgeons, American Pediatric Surgical Association,
Applied Nursing Research Canadian Association ofPaediatric Surgeons and Pacific
the American Association of Ord and MaxiUofacial Surgeons Archives of Psychiatric Nursing—Official Journal of the SERPN
Journal of Prosthodontics—Official Journal of The American Association of Pediatric Surgeons
Division, International Society of Psychiatric -Mental Health Nurses Operative Techniques in General Surgery
Journal of Pediatric Nursing—Official Journal of the Society of Operative Techniques in Neurosurgery
Seminars in Orthodontics Pediatric Nurses Operative Techniques in Plastic and Reconstructive Surgery
DERMATOLOGY Journal of Pediatric Oncology Nursing—Official Journal of the Operative Techniques in Thoracic and Cardiovascular Surgery—An
American Journal of Contact Dermatitis—The Official Journal Association of Pediatric Oncology Nurses Official Publication of The American Association for Thoracic Surgery
of the American Contact Dermatitis Society Journal of PeriAnesthesia Nursing—Official Journal of the American Seminars in Colon and Rectal Surgery
Seminars in Cutaneous Medicine and Surgery Society of PeriAnesthesia Nurses Seminars in Laparoscopic Surgery
Journal of Professional Nvr^n%—0fficialjournal of the American Seminars in Pediatric Surgery
EMERGENCY MEDICINE Association of Colleges of Nursing Seminars in Spine Surgery
American Journal of Emergency Medicine Pain Management Nursing—Official Journal of the American Society Seminars hi Thoracic and Cardiovascular Surgery—tin Official
Clinical Pediatric Emergency Medicine of Pain Management Nurses Publication of The American Association for Thoracic Surgery
PeriAnesthesia and Ambulatory Surgery Nursing update—Official Seminars hi Thoracic and Cardiovascular Surgery: Pediatric Cardiac
ENDOCRINOLOGY Publication of the American Society ofPeriAnestbesia Nurses Surgery Annual—An Official Publication of The American
Metabolism—Clinical and Experimental Seminars for Nurse Managers Associationfor Thoracic Surgery
Seminars in Oncology Nursing Seminars in Urologie Oncology
GASTROENTEROLOGY/HEPATOLOGY Seminars in Vascular Surgery
Clinical Perspectives in Gastroenterology—The Official OBSTETRICS AND GYNECOLOGY
Clinical PracticeJournal of the American Gastroenterological Clinical Journal of Women's Health TRANSPLANTATION
Association Liver Transplantation-^» Official Publication of the American
Gastroenterology— Official Journal of the American ORTHOPEDICS Association for the Study of Liver Diseases and the International
GastroenterologicalAssociation Armroscopy: The Journal of Arthroscopic and Related Surgery— Liver Transplantation Society
Official Publication of the Artbroscofty Association of North America Transplantation Reviews
the Study of Liver Diseases and the International Society ofArtbroscopy, Knee Surgery, and
liver Transplantation—4» Official Publication of the American Orthopaedic Sports Medicine UROLOGY
AssociationJbr the Study of Liver Diseases and The International Journal of the American Society for Surgery of the üanA— Seminars in Urologie Oncology
Liver Transplantation Society An Officidjwrnal of the Americm Society for Surgery
Seminars in Gastrointestinal Disease of the Hand VETERINARY MEDICINE
Techniques in Gastrointestinal Endoscopy Journal of Hand Surgery-^4» Official Journal of the American Advances hi Small Animal Medicine and Surgery
Society for Surgery of the Hand Clinical Techniques in Equine Practice
HEMATOLOGY/ONCOLOGY Operative Techniques in Orthopaedics Clinical Techniques in Small Animal Practice
Seminars in Hematology Operative Techniques in Sports Medicine Seminars in Avian and Exotic Pet Medicine
Seminars in Oncology Seminars in Armroplasty Veterinary Surgery—The Official Journal of The American
Seminars in Radiation Oncology Seminars in Spine Surgery College of Veterinary Surgeons, Inc. and The European College
Transfusion Medicine Reviews of Veterinary Surgeons
OTORHINOLARYNGOLOGY
INFECTIOUS DISEASES American Journal of Otolaryngology For more information about thesejournals,
Seminars in Infection Control Operative Techniques in Otolaryngology—Head and Neck Surgery please contact:
Seminars in Pediatric Infectious Diseases
Seminars in Respiratory Infections PATHOLOGY Periodicals Marketing
Annals of Diagnostic Pathology W.B. SAUNDERS
MEDICAL TRANSCRIPTION Human Pathology A Division of Elsevier Science
The Latest Word Seminars in Diagnostic Pathology
The Curtis Center, Independence Square West
NEONATAL/PERINATAL MEDICINE PHYSICAL MEDICINE Philadelphia, PA 19106-3399
Newborn and Infant Nursing Reviews Archives of Physical Medicine and Rehabilitation—
Seminars in Perinatology Official Journal of the American Congress of Rehabilitation
Phone (215) 238-5614
Medicine and the American Academy of Physical Medicine Or visit our homepage at:
www.wbsaunders.com
<<    
     Article
      >> | HOME
|           TOC In
<<    
     Article
      >> Home | TOC |          
Index

The world's
bestselling
medical
dictionary!

Dorland's Illustrated Medical Dictionary all of its illustrations, tables, and appendices.
is universally acknowledged as the world's At the same time, its state-of-the-art CD-ROM
finest medical dictionary. For 100 years, technology puts tremendous reference power
health care professionals have relied on its and convenience at users' disposal—plus audio
unmatched comprehensiveness, accuracy, pronunciations for over 10,000 primary entries!
clarity, and ease of use. 2000. Single-user CD-ROM for Windows™ or
The 29th Edition presents the very latest Macintosh*. Order #W9493-4.
information from every frontier in health A W.B. Saumlers alle.
care! It offers over 8,100 new terms—
7? Phone:
121,160 in all • over 7,600 new entries— Call toll-free 1-800-545-2522
for a total of 117,469 • over 860 illustra- (8:30-8:00 Eastern Time) to order.
tions—566 brand new • and much more! Be sure to mention DM#66899.
2000. 2112 pp. 864 ills. Order 3W6254-4.
Fax to i-800-568'5136 to order.
Also available on CD-ROM! Be sure to mention DM#66899.

Dorland's Electronic Medical Dictionary, 1^1 Mail:

29th Edition contains all of the definitions, Else vie r Health Sciences
Order Fulfillment Dept.
pronunciations, plural forms, and etymologies 1 1830 Wcsrline Industrial Drive
found in the hardbound dictionary, as well as Saint Louis, MO 63H6-3318

W.B. SAUNDERS
Elsevier Science
EHS 01 DM166»« * Elm*» SCTCHC, ZWä
<<    
     Article
      >>
Seminars in Orthodontics
EDITOR
P. Lionel Sadowsky, DMD, BDS, DipOrth, MDent
Professor and Chairman
Department of Orthodontics
University of Alabama
1919 Seventh Avenue South
Birmingham, AL 35294
Fax: (205) 975-7590
Seminars in Orthodontics (ISSN 1073-8746) is published
quarterly by W.B. Saunders. Months of issue are March, June,
September, and December. Corporate and Editorial Offices:
The Curtis Center, Independence Square West, Philadelphia,
PA 19106-3399. Accounting and Circulation Offices: 6277 Sea
Harbor Drive, Orlando, FL 32887-4800. POSTMASTER: Send
change of address to: Seminars in Orthodontics, W.B. Saunders,
Customer Service, 6277 Sea Harbor Drive, Orlando, FL 32887-
4800.
Editorial correspondence should be addressed to:
P. Lionel Sadowsky, DMD, BDS, DipOrth, MDent, Editor,
Seminars in Orthodontics, Professor and Chairman, Department
of Orthodontics, University of Alabama, 1919 Seventh Avenue
South, Birmingham, AL 35294-0007; fax: (205) 975-7590.
Correspondence regarding subscriptions or change of
address should be directed to Seminars in Orthodontics, W.B.
Saunders, Periodicals Department, P.O. Box 628239, Orlando,
FL 32862-8239 or e-mail hhspcs@harcourt.com.
Change of address notices, including both the old and new
addresses of the subscriber and the mailing label, should be
sent at least 1 month in advance. Customer Service: 1-800-654-
2452
Yearly subscription rates: United States and possessions:
individual, $133.00; institution, $164.00; student and resident,
$67.00; single issue, $50.00. All other countries: individual
$166.00; institution, $198.00; student and resident, $83.00;
single issue, $50.00. For all areas outside the United States and
possessions, there is no additional charge for surface delivery.
For air mail delivery, add $16.00. To receive student/resident
rate, orders must be accompanied by name of affiliated
institution, date of term, and the signature of program/residency
coordinator on institution letterhead. Orders will be billed at
individual rate until proof of status is received.
Prices are subject to change without notice. Current prices
are in effect for back volumes and back issues. Single issues,
both current and back, exist in limited quantities and are
offered for sale subject to availability. Back issues sold in
conjunction with a subscription are on a prorated basis. Checks
should be made payable to W.B. Saunders and sent to Seminars
in Orthodontics, W.B. Saunders, Periodicals Department, 6277
Sea Harbor Drive, Orlando, FL 32887-4800.
Copyright 2002, Elsevier Science (USA). All rights
reserved. No part of this publication may be reproduced or
transmitted in any form or by any means, electronic or
mechanical, including photocopy, recording, or any information
storage and retrieval system, without permission in writing
from the Publisher. Printed in the United States of America.
Correspondence regarding permission to reprint all or
part of any article published in this journal should be
addressed to Journal Permissions Department, W.B. Saunders,
6277 Sea Harbor Drive, Orlando, FL 32887-4800. Telephone
number: 1-407-345-2500.
W.B. SAUNDERS
Home | TOC |          
Index
The appearance of the code at the bottom of the first page
of an article in this journal indicates the copyright owner's
consent that copies of the article may be made for personal or
internal use, or for the personal or internal use of specific clients,
for those registered with the Copyright Clearance Center, Inc.
(222 Rosewood Drive, Danvers, MA 01923; (508) 750-8400;
www.copyright.com). This consent is given on the condition that
the copier pay the stated per-copy fee for that article through the
Copyright Clearance Center, Inc. for copying beyond that
permitted by Sections 107 or 108 of the US Copyright Law.
This consent does not extend to other kinds of copying, such
as copying for general distribution, for advertising or promotional
purposes, for creating new collective works, or for resale.
Absence of the code indicates that the material may not be
processed through the Copyright Clearance Center, Inc.
Reprint inquiries should be addressed to Anne
Rosenthal, Elsevier Science, The Curtis Center, Independence
Square West, Philadelphia, PA 19106-3399. Telephone (215)
238-5534, fax (215) 238-6423; e-mail: a.rosenthal@elsevier.com.
Advertising representative: M.J. Mrvica Associates, Inc,
2 West Taunton Ave, Berlin, NJ 08009. Telephone (609) 768-
9360. Fax (609) 753-0064.
Publication of an advertisement in Seminars in Orthodontics
does not imply endorsement of its claims by the Editor(s) or
Publisher of the journal.
The contributors have checked generic and trade names and
verified drug doses for accuracy according to the standards
accepted at the time of publication. The ultimate
responsibility, however, lies with the prescribing physician.
Please convey any errors to the Editor.
The ideas and opinions expressed in Seminars in
Orthodontics do not necessarily reflect those of the Editor or the
Publisher. Publication of an advertisement or other product
mention in Seminars in Orthodontics should not be construed as
an endorsement of the product or the manufacturer's claims.
Readers are encouraged to contact the manufacturer with any
questions about the features or limitations of the products
mentioned. Neither the Editor or Publisher assume any
responsibility for any injury and/or damage to persons or
property arising out of or related to any use of the material
contained in this periodical. The reader is advised to check the
appropriate medical literature and the product information
currently provided by the manufacturer of each drug to be
administered to verify the dosage, the method and duration of
administration or contraindications. It is the responsibility of
the treating physician or other health care professional, relying
on independent experience and knowledge of the patient, to
determine drug dosages and the best treatment for the patient.
Seminars in Orthodontics is indexed in the Cumulative
Index to Nursing and Allied Health Literature® print index
and the Cinahl® database.
<<    
     Article
      >> Home | TOC |          
Index
ÄNESTHESIOLOGY CLINICS
CARDIOLOGY CLINICS
CARDIOLOGY CLINICS: ANNUAL OF DRUG THERAPY
CLINICS IN CHEST MEDICINE FOR MORE INFORMATION ABOUT THESE
CHEST SURGERY CLINICS TITLES, PLEASE CONTACT:
CHILD AND ADOLESCENT PSYCHIATRIC CLINICS
CRITICAL CARE CLIMCS W.B. SAUNDERS
CRITICAL CARE NURSING CLINICS A Division of Elsevier Science
DENTAL CLINICS
DERMATOLOGIC CLINICS Periodicals Department
EMERGENCY MEDICINE CLINICS The Curtis Center
ENDOCRINOLOGY & METABOLISM CLINICS Independence Square West
FACIAL PLASTIC SURGERY CLINICS Philadelphia, PA 19106-3399
CLINICS IN FAMILY PRACTICE Phone: (215) 238-5614
FOOT AND ANKLE CLINICS
GASTROENTEROLOGY CLINICS
GASTROINTESTINAL ENDOSCOPY CLINICS OR VISIT OUR HOME PAGE AT.
CLINICS IN GERIATRIC MEDICINE
http://www.wbsaunders.com
HAND CLINICS
ATMS OF IHE HAND CLINICS
HEMATOLOGY/ONCOLOGY CLINICS
IMMUNOLOGY AND ALLERGY CLINICS
INFECTIOUS DISEASE CLINICS
INFERTILITY AND REPRODUCTIVE MEDICINE Cimics
CLINICS IN LABORATORY MEDICINE
CLINICS IN LIVER DISEASE
MAGNETIC RESONANCE IMAGING CLINICS
MEDICAL CLINICS
NEUROIMAGING CLINICS
NEUROLOGIC CLINICS
NEUROSURGERY CLINICS
NURSING CLINICS
OBSTETRICS AND GYNECOLOGY CLINICS
CLINICS IN OCCUPATIONAL AND ENVIRONMENTAL MEDICINE
CLINICS ATLAS OF OFFICE PROCEDURES
OPHTHALMOLOGY CLINICS
ORAL AND MAXILLOFACIAL SURGERY CLINICS
ATLAS OF THE ORAL AND MAXILLOFACIAL SURGERY CLINICS
ORTHOPAEDIC PHYSICAL THERAPY CLINICS
ORTHOPEDIC CLINICS W.B. SAUNDERS
OTOLARYNGOLOGY CLINICS
A Division of Elsevier Science
PEDIATRIC CLINICS
CLINICS IN PERINATOLOGY
PHYSICAL MEDICINE AND REHABILITATION CLINICS
CLINICS IN PLASTIC SURGERY
CLINICS IN PODIATRIC MEDICINE AND SURGERY
PRIMARY CARE: CLINICS IN OFFICE PRACTICE
PSYCHIATRIC CLINICS
RADIOLOGIC CLINICS
RESPIRATORY CARE CLINICS
RHEUMATIC DISEASE CLINICS
CLINICS IN SPORTS MEDICINE
Clinics
60 Hardcover Periodicals
SURGICAL CLINICS
SURGICAL ONCOLOGY CLINICS
UROLOGIC CLINICS
Designed to Enhance Your
ATIAS OF THE UROLOGIC CLINICS
VETERINARY CLINICS: EQUINE PRACTICE
Clinical Skills
VETERINARY CLINICS: EXOTIC ANIMAL PRACTICE
VETERINARY CLINICS: FOOD ANIMAL PRACTICE
VETERINARY CLINICS: SMALL ANIMAL PRACTICE )02 ELSEVIER SCIENCE
<<    
     Article
      >>
Seminars in Orthodontics
EDITOR
P. Lionel Sadowsky, DMD, BBS, DipOrth, MDent
EDITORIAL BOARD
Richard G. Alexander, Arlington, TX
Rolf G. Behrents, Memphis, TN
Samir E. Bishara, Iowa City, I A
Robert Boyd, DBS, San Francisco, CA
Larry M. Bramble, Cypress, CA
John S. Casko, Iowa City, IA
Harry L. Dougherty, Van Nuys, CA
T.M. Graber, Evanston, IL
Robert J. Isaacson, Richmond, VA
Alexander Jacobson, Birmingham, AL
Lysle E.Johnston, Jr., Ann Arbor, MI
Gregory J. King, Seattle, WA
Vincent G. Kokich, Tacoma, WA
Steven J. Lindauer, Richmond, VA
INTERNATIONAL
Zeev Abraham, Herzliya, Israel
W.G. Evans, Johannesburg, South Africa
Roberto Justus, Mexico City, Mexico
Home | TOC |          
Index
James A. McNamara, Jr, Ann Arbor, MI
Robert N. Moore, Grand Island, NE
Ravindra Nanda, Farmington, CT
Perry M. Opin, Milford, CT
Sheldon Peck, Newton, MA
William R. Proffit, Chapel Hill, NC
Cyril Sadowsky, Chicago, IL
David M. Sarver, Birmingham, AL
T. Michael Speidel, Minneapolis, MN
William J. Thompson, Bradenton, FL
James L. Vaden, Cookeville, TN
Robert L. Vanarsdall, Jr., Philadelphia, PA
Katherine Vig, Columbus, OH
C.B. Preston, Buffalo, NY
Shinkichi Namura, Tokyo, Japan
George Skinazi, Paris, France
Björn U. Zachrisson, Oslo, Norway
<<    
     Article
      >> Home | TOC |          
Index

SPECIAL
OFFER!
For a limited time, all Advances published
in 1999 and 1998 are on sale for $35, including
shipping. In most cases that's a savings of
over 50%!

To take advantage of this special offer

Visit our website at www.mosby.com

Click on the "Special Promotions" link

Browse through the Advances that interest you

Select the titles you wish to purchase

This offer is good only while supplies last,

so don't delay! ORDER TODAY!

For more information, call toll-free

1-800-654-2452

Mosby
A Division of Elsevier Science

© 2002 Mosby
<<    
     Article
      >> Home | TOC |          
Index

Seminars in Orthodontics
VOL 8, NO 3 SEPTEMBER 2002

Management of the Vertical Dimension in Clinical Orthodontics

Timothy T. Wheeler, DMD, PhD
Guest Editor
CONTENTS

Editorial 111
Timothy T. Wheeler

Development of the Vertical Dimension: Nature and Nurture 113

James K. Hartsfield, Jr.

Diagnosis of the Vertical Dimension 120

James L. Vaden and Lloyd E. Pearson

Early Treatment of Hyperdivergent Open-Bite Malocclusions 130

Peter H. Buschang, Wayne Sankey, and Jeryl D. English

Vertical Skeletal and Dental Changes in Early Treatment of Class II

Malocclusion 141
Calogero Dolce, Lisa K. Babb, Susan P. McGorray, Marie G. Taylor,
Gregory J. King, and Timothy T. Wheeler

Biomechanical Considerations in the Management of the Vertical

Dimension 149
Stanley Braun

The Effects of Altering Vertical Dimension on the Masticatory Muscles and

Temporomandibular Joint 155
Charles Widmer

Long-Term Stability of Anterior Open-Bite Therapy: A Review 162

GregJ. Huang

Surgical Modification of Long-Face Problems 173

L'TanyaJ. Bailey, William R. Proffit, Raymond P. White, Jr.,
and David M. Sarver
<<    
     Article
      >> Home | TOC |          
Index

Seminars in Orthodontics
Future Issues

Vol 8 No 4 (December 2002)

RISK ASSESSMENT AND MANAGEMENT IN CLINICAL ORTHODONTICS
Laurance Jerrold, DDS, JD, Guest Editor
Vol 9 No 1 (March 2003)
THE HERBST® APPLIANCE
Hans Pancherz, DMD, Guest Editor
Vol 9 No 2 (June 2003)
MOUNTING OF CASTS IN CLINICAL ORTHODONTICS
Richard Kulbersh, DMD, Guest Editor

Recent Issues

Vol 8 No 2 (June 2002)

BIOSTATISTICS FOR THE ORTHODONTIC CLINICIAN
Rose D. Sheats, DMD, Guest Editor
Vol 8 No 1 (March 2002)
CLINICAL UPDATE ON TECHNOLOGICAL ADVANCES IN ORTHODONTICS AND DENTOFACIAL ORTHOPEDICS
Gregory J. King, DMD, DMSc, Guest Editor
Vol 7 No 4 (December 2001)
THREE-DIMENSIONAL DIAGNOSIS AND TREATMENT IN ORTHODONTICS
Sheldon Baumrind, DDS, MS, and Robert L. Boyd, DDS, MEd, Guest Editors
Vol 7 No 3 (September 2001)
TOPICS IN BIOMECHANICS
Stanley Braun, DDS, MME, Guest Editor
Vol 7 No 2 (June 2001)
THE ALEXANDER DISCIPLINE
R.G. Alexander, DDS, MSD, Guest Editor
Vol 7 No 1 (March 2001)
CLINICAL BIOMECHANICS
Steven J. Lindauer, DMD, MDSc, Guest Editor
Vol 6 No 4 (December 2000)
PSYCHOLOGIC ISSUES RELATED TO ORTHODONTIC TREATMENT AND PATIENT COMPLIANCE
Pramod K. Sinha, DDS, BDS, MS, Ram S. Nanda, DDS, MS, PhD, and Roger B. Fillingim, PhD, Guest Editors
Vol 6 No 3 (September 2000)
BIOLOGY OF ORTHODONTIC TOOTH MOVEMENT: CLINICAL IMPLICATIONS
Bhavna Shroff, DDS, MDentSc, Guest Editor
<<    
     Article
      >>
Seminars in Orthodontics
VOL 8, NO 3
Introduction
T he orthodontic literature and the orth-
odontic lecture circuit are replete with pub-
lications and much discussion pertaining to
treatment of the anteroposterior dimension and
to a lesser extent the transverse dimension of the
dentition. However, there has been very little
discussion and much less research on develop-
ment or treatment of the vertical dimension.
This issue of Seminars in Orthodontics attempts to
present various aspects of this topic ranging
from growth and development to diagnosis and
treatment.
The first article discusses the genetic and en-
vironmental interaction in the growth and de-
velopment of the vertical dimension. The ques-
tion of how environmental and genetic factors
interact is relevant to clinical practice because it
may explain why a particular treatment that al-
ters the environment may be successful in one
compliant patient but not in another. This arti-
cle provides an excellent review of the genetic
considerations that clinicians should understand
when treating vertical problems.
In the second article, Vaden and Pearson re-
view some of the pertinent literature and offer
some diagnostic and treatment-planning sugges-
tions to the orthodontist who deals with the
vertical dimension on a daily basis. This article
brings to the surface those problems that the
practicing orthodontist must be aware of and
plan for when dealing with the vertical dimen-
sion.
Treatment timing of hyperdivergent vertical
problems is addressed in the third article. It
reviews all the factors that must be considered
when considering treatment on a particular pa-
tient. It also points out how the outcomes of this
treatment must be assessed and then interpreted
for application to other patients. As in most
Copyright 2002, Elsevier Science (USA). All rights reserved.
doi:10.1053/sodo.2002.125429
Seminars in Orthodontics, Vol 8, No 3 (September), 2002: pp 111-112
Home | TOC |          
Index
SEPTEMBER 2002
areas of clinical orthodontics, there is much we
need to learn on this topic.
Results from the Class II randomized clinical
trial examining the timing of Class II treatment
are presented in the article by Dolce et al. This
report confirms some ideas held by clinicians
and evaluates when these changes are occurring
and what changes are relapsing. Perhaps with
this information and with better retention meth-
ods, we could improve our outcomes.
The biomechanics involved in the treatment
of patients who exhibit an anterior open bite or
a deep overbite related to excessive or deficient
vertical facial dimensions are discussed in the
article by Braun. Defined objectives must be es-
tablished for a patient, and the clinician must
understand how the biomechanics will impact
on these objectives in the vertical dimension.
In another article, Widmer discusses how the
alteration of the vertical dimension of occlusion
can affect specific components of the jaw closing
muscles and the temporomandibular joint. How
the muscles or the temperomandibular joint is
impacted is rarely considered during orthodon-
tic treatment. Not only is little known on how
these tissues are impacted, but what role the
changes could play in possible relapse need to
be considered.
Anterior open bites are one of the main ver-
tical challenges that orthodontists face. Huang
has written an excellent review of the orthodon-
tic literature on the long-term stability of this
problem. He critically evaluates the literature on
both the orthodontic closure of open bites and
surgical closure of open bites.
The last report deals with the treatment plan-
ning and treatment issues of the orthodontic
and surgical management of long-face vertical
problems.
In summary, although the orthodontist when
treating patients always considers the vertical
dimension, it has been not widely discussed or
published. The topics presented in this issue
were chosen to broadly cover as many aspects as
111
<<    
     Article
      >> Home | TOC |          
Index

112 Timothy Wheeler

possible on the vertical dimension and act as a deal with when concerned with the vertical di-
stimulus for further research and discussion. Al- mension.
though there are many other issues that have
not been covered, I hope this issue helps bring Timothy T. Wheeler, DMD, PhD
to light some of the concerns that we need to Guest Editor
<<    
     Article
      >> Home | TOC |          
Index

Development of the Vertical Dimension:

Nature and Nuture
James K. Hartsfield, Jr

The relevance of analyzing the development of the vertical dimension to

clinical practice is first to determine if there is a vertical dimension compo-
nent to the malocclusion, then ascertain what factors are having the great-
est influence on the vertical dimension problem. Unfortunately, studies on
the genetic and environmental factors that influence the development of
vertical dimension are representative of the samples, not necessarily of any
particular individual. In addition, the extent that a particular trait is influ-
enced by genetic factors may have little if any effect on success of environ-
mental (treatment) intervention. Genetic factors that influenced a trait may
also influence the response to intervention to alter that trait, or other
genetic factors may be involved in the response. Therefore, the possibility
for altering the environment to gain a more favorable dimension is theoret-
ically possible, even in individuals with a relatively high genetic influence on
the vertical dimension. However, the question of how environmental and
genetic factors interact (a question that essentially cannot be answered in
estimates of heritability), is most relevant to clinical practice because it may
explain why a particular alteration of the environment (treatment) in one
compliant patient may be successful and not in another. (Semin Orthod
2002;8:113-119.) Copyright 2002, Elsevier Science (USA). All rights reserved.

onsideration of factors that influence, de- and the environment in which the individual
C termine, or even drive development usually
involves a discussion of nature versus nurture, as
develops.

if they were mutually exclusive. However, devel-

opment is not the result of genetic and environ-
mental (nongenetic) factors working in isolation
or independent of one another. Before proceed-
ing, a couple of basic definitions are required.
Genotype generally refers to the set of genes that
an individual carries and, in particular, usually
refers to the particular pair of alleles (alternative
forms of a particular gene) at a given region of
the genome. In contrast, phenotype is the ob-
servable properties and physical characteristics
of an individual,1 as determined by genotype
From the Indiana University School of Dentistry, Indianapolis,
IN.
Address correspondence to James K. Hartsfield, Jr, DMD, MS,
MMedSci, PhD, Indiana University School of Dentistry, 1121 W
Michigan, Indianapolis, IN 46202-5186.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0803-0002$35.00/0
doi:10.1053/sodo.2002.125430
Even Gene Mutations For Dominant
Traits Are Not Predetermining
The craniosynostosis syndromes (along with
their effect on craniofacial growth and develop-
ment) are autosomal dominant traits associated
with single-gene mutations. They provide good
examples of how, even with the strong influence
of a single gene, the phenotype can vary mark-
edly. Contrary to an earlier presumption that a
particular mutation in a given gene would always
result in a specific syndrome, several identical
mutations in the fibroblast growth factor recep-
tor 2 gene have been found in patients diag-
nosed with the three clinical entities of Crouzon,
Pfeiffer, and Jackson-Weiss syndrome.2'3
Another example of the individual variability
of these autosomal dominant phenotypes associ-
ated with a single-gene mutation occurred in
individuals with the classic phenotypes of
Pfeiffer and Apert syndrome, as well as in seven

Seminars in Orthodontics, Vol 8, No 3 (September), 2002: pp 113-119 113

<<    
     Article
      >>
114 James K. Hartsfield, Jr
other individuals with a facial resemblance to
Crouzon syndrome that occurred in the same
family.4 The phenotype may be so variable that
an individual may appear to be clinically normal
yet have the same gene mutation associated with
Crouzon syndrome in three of his children and
two of his grandchildren. Only through cepha-
lometry was a minimal expression of features
suggestive of Crouzon syndrome evident.5
The phenotypic variation present in these ex-
amples may be caused by modifying factors such
as environment and other (modifying) genes in
the genome that interact with the effect of a
specific mutation associated with a dominant
trait. In fact, the concepts of variable expressivity
and reduced penetrance are applied to domi-
nant traits or conditions, acknowledging the po-
tentially variable phenotype that may not be ev-
ident at all in an individual with the gene
mutation.6 These examples give a clear message
that even for a generally extreme autosomal
dominant phenotype, simply discovering the
gene mutation will very likely indicate that there
will be an effect on craniofacial growth and de-
velopment, but it does not give a precise picture
of what that effect will be only what it may tend
to be.
Estimating the Influence of Genetic and
Environmental Factors on Phenotype
A discussion of the methods and assumptions
made to estimate heritability, defined as the pro-
portion of the total phenotypic variance in a
sample that is contributed by genetic variance,7
is beyond the scope of this article. For more
information on these methods and assumptions,
the reader may start with Genetics and Analysis of
Quantitative Traits by Lynch and Walsh.8 A trait
with a heritability of 1 is said to be expressed
without any environmental influence, whereas a
trait with a heritability of 0.5 would have half its
variability (from individual to individual) influ-
enced by environmental factors and half by ge-
notypic factors. Values over 1 may occur because
the twin methodology provides an estimate of
heritability, under several simplifying assump-
tions, that may be incorrect. Still, the estimation
of heritability can provide an indication of the
relative importance of genetic factors. Confirm-
ing that there is a certain degree of genetic
influence on a trait is a preliminary step to fur-
Home | TOC |          
Index
ther specific genetic linkage studies (using DNA
markers) to determine areas of the genome that
appear to be associated with the characteristics
of a given trait.9
A few points should be kept in mind when
reviewing heritability estimates. First of all, they
refer to a specific sample and do not necessarily
pertain to a given individual even from within
the sample. Thus, they do not allow one to tell to
what degree a particular trait was determined by
genetic or environmental factors in a single in-
dividual. In addition, heritability estimates are
descriptive of variances within a sample at a
given time, and they are not predictive.9 Herita-
bility estimates can change with age; for exam-
ple, a longitudinal analysis of 30 sets of siblings
that had not undergone orthodontic treatment
showed a significant increase in heritability esti-
mates between the ages of 4 and 14 years for 29
craniofacial skeletal variables, including in-
creases for total anterior face height, upper an-
terior face height, total posterior face height,
and upper posterior face height. Despite the
general trend for all the craniofacial skeletal
variables to increase, there was a decrease for
lower posterior face height. When a comparison
was made of the craniofacial skeletal heritability
estimates at age 14 years and 20 years, there was
an insignificant upward trend for some of the
traits. However, there was a decrease in the her-
itability estimate from the age of 14 for upper
anterior face height and an increase for lower
posterior face height to that estimated at age 4
years.10
The heritability of a trait cannot necessarily
be extrapolated from one sample and set of
environmental conditions to another.7 An ad-
verse environment can alter the phenotypic ex-
pression that the genes would have promoted
under more favorable conditions. An extreme
example of this principle is the delayed growth
seen from the effects of famine associated with
war.11 Therefore, a high heritability does not
prevent a trait from being substantially influ-
enced by subsequent changes in environmental
conditions in that sample.12
Estimation Of Vertical Dimension
Heritability
Clinical consideration of the vertical dimension
may include the evaluation of the ratio of the
<<    
     Article
      >>
Development of the Vertical Dimension
upper anterior face height to lower anterior face
height, as well as anterior to posterior face
heights. Investigation of the anterior upper face
height to anterior lower face height in 30
monozygotic twin pairs (ranging in age from
12.0 to 18.8 years, with a mean of 15.9 years),
and 30 dizygotic like-sex twin pairs (ranging in
age from 12.4 to 21.0 years, with a mean of 15.5
years) resulted in a heritability estimate of
0.52.13
Lateral cephalographs of 33 monozygotic and
46 dizygotic twins, who ranged in age from 9 to
16 years (mean, 12.1 years) and had not under-
gone orthodontic treatment, were used in ge-
netic model fitting to determine the heritability
of anteroposterior and vertical facial propor-
tions.14 The analysis indicated that additive
genes and the specific environment influenced
all the facial proportions. The heritability was
0.71 for upper to lower anterior face height and
0.66 for anterior to posterior face height.
The better-fitting model, with additive as op-
posed to dominant gene influence, indicates
that genetic influence was the sum of some num-
ber of approximately equal gene effects. The
specific environment aspect of the fitted model
implies that the environmental influences were
of a more individual, as opposed to a more
common environmental, nature. This is consis-
tent with the prevailing concept that malocclu-
sion has a multifactorial origin (combination of
a number of genetic and environmental factors)
and implies that specific environmental (treat-
ment) factors might have some effect on the
traits.
Analysis of the soft tissue associated with an-
terior vertical height, as measured on the facial
profile of lateral photographs taken of 42 pairs
of monozygotic twins and 37 pairs of dizygotic
twins, produced a heritability estimate of 0.66
for anterior face height.15 Lateral cephalographs
were used in a path analysis study to compare
the heritability of horizontal and vertical dis-
tances (as opposed to proportions or ratios) on
55 pairs of twins of the same gender, ranging
from 13 to 20 years of age who had not under-
gone orthodontic treatment.16 The mean age of
the monozygotic twins was 15.2 years, whereas
the mean age for the dizygotic twins was 14.7
years. Although the heritability estimate for both
the lower anterior face height and total anterior
face height was 0.86, it was markedly lower for
Home | TOC |          
Index
115
upper anterior face height (0.16) and posterior
face height (0.26). These two relatively low her-
itability estimates were probably artifacts caused
by random variation in a limited sample. The
likelihood of some random variation, expressed
in the path analysis used, was reinforced when
markedly higher estimates of heritability for up-
per anterior face height (0.81) and posterior
face height (0.88) were determined in the same
sample by using weighted means of monozygotic
and dizygotic twin estimates instead of path anal-
ysis.17
Although attributed to random error in the
path analysis method, the dichotomy of the her-
itability between the upper anterior face height
and the lower anterior face height echoed the
findings of an earlier study on 35 pairs of
monozygotic twins and 21 pairs of like-sex dizy-
gotic twins (ranging in age from 18-55 years,
with a median age of 24) in which there was a
significant difference in the intrapair differences
(variance) between the monozygotic and dizy-
gotic twins for total anterior face height and
lower anterior face height but not for upper
anterior face height. The upper anterior face
height was essentially the same between each
pair of twins, regardless of their zygosity. It was
concluded that the dichotomy of the heritability
between the upper anterior face height and the
lower anterior face height, along with the rela-
tively high heritability of the total anterior face
height, infers that it is the lower anterior face
height that is primarily responsible for the her-
itability of the total anterior face height.18
The dichotomy was also suggested when lat-
eral cephalometric measurements were made of
67 monozygotic twin pairs and 29 dizygotic twin
pairs and investigated through factor analysis
with subsequent estimation of heritability. The
factor with the largest heritability estimate
(0.76) included the lower anterior face height
and total anterior face height. The heritability
estimate for the factor that included upper an-
terior face height was 0.48, whereas that for total
posterior face height and lower posterior face
height was 0.31.19
If there are relatively high common inheri-
tance estimates for the upper anterior face
height and posterior face height as compared to
the lower anterior face height and total anterior
face height, this suggests that some common
(cultural) environmental factor(s) have greater
<<    
     Article
      >>
116 James K. Hartsfield, Jr
influence on these traits.16 One conjecture
might be diet consistency20 on the posterior face
height and the development of nasal airway pa-
tency on the upper anterior face height; how-
ever, the presence of random variation may also
be an explanation for these relatively high com-
mon inheritance estimates.16-17 In support of the
environmental effect on nasal development is
the rejection of the null hypothesis that there is
no common sibling effect on nasal height. These
results are based on path analysis of family re-
semblance using craniofacial anthropometric
measurements of 1,763 individuals in 399 fami-
lies from a rural community in Andhra Pradesh,
India.21
Considering the effect of breathing on the
vertical dimension, the most striking difference
was found in a comparison of cephalometric
facial dimensions in 25 white children with pe-
rennial allergic rhinitis. These apparent mouth
breathers were compared with their 25 siblings
who apparently were not mouth breathers and
did not have perennial allergic rhinitis and 14
nasal breathing control subjects. The analysis
revealed that the allergic children had a more
divergent facial pattern.22 This is consistent with
the findings in a study of 100 11-year-old Finnish
children in which there was an increase in the
vertical dimension in moderate and severely al-
lergic subjects.23 An additional report with sim-
ilar findings was based on a study of 37 children
with perennial allergic rhinitis (ages 5-10 years)
and matched controls.24
If an increase in total anterior face height and
lower anterior face height, in particular, are as-
sociated with perennial allergic rhinitis and
mouth breathing, why do some (although not
all) studies indicate a dichotomy in the estimates
of heritability of the upper anterior face height
and the lower posterior face height? One hy-
pothesis is that the lower anterior face height
may have a relatively greater heritability than the
upper anterior face height in some individuals
unless increased nasal obstruction, resulting in
mouth breathing, becomes a predominating fac-
tor. Again, heritability is a descriptive statistic for
a particular sample under defined environmen-
tal conditions.
Studies that estimate heritability of craniofa-
cial structures may have a bias because they have
generally been performed with subjects who had
not undergone orthodontic treatment; thus,
Home | TOC |          
Index
subjects with an extreme malocclusion tend to
be excluded.25 In a thought-provoking study of
the heritability of cephalometric and occlusal
variables in siblings with overt malocclusions, it
was found that, in contrast to a series of similar
subjects with naturally occurring good occlu-
sion, the heritability estimates for craniofacial
skeletal variables in the subjects with overt mal-
occlusions were significantly lower, and the her-
itability estimates for occlusal variations were sig-
nificantly higher.25 To quote King et al,25 "We
propose that the substantive measures of intersib
similarity for occlusal traits reflect similar re-
sponses to environmental factors common to
both siblings. That is, given genetically influ-
enced facial types and growth patterns, siblings
are likely to respond to environmental factors
(eg, reduced masticatory stress, chronic mouth-
breathing) in similar fashions." Malocclusions
appear to be acquired, but the fundamental ge-
netic control of craniofacial form predisposes
siblings into comparable physiologic responses,
that often lead to development of similar maloc-
clusions.25
Does Knowing The Heritability Matter
In Treatment?
It has been stated that, "Variables with a lower
genetic determination are more open to influ-
ence by, for example, orthopedic correction
than are variables with a high genetic determi-
nation, which are not so easily changed by the
environment." 14 This implies that the genetic
influence is a predetermining, unalterable fac-
tor. However, as has already been discussed, it
may be altered under different environmental
(treatment) conditions. Certainly, treatment de-
pends on the origin of a disorder if that cause is
known and specific. However, it has also been
pointed out that contrary to popular opinion the
extent that a particular trait is influenced (or if
you wish even determined) by genetic factors
may have little if any effect on the success of
environmental (treatment) intervention.26 What
is important is the response of the individual to
the environmental (treatment) intervention,
which may be similar for comparable genotypes
(or at least the genes that are going to influence
the response to the particular intervention). Ge-
netic factors that influence a trait may also in-
fluence the response to intervention designed to
<<    
     Article
      >>
Development of the Vertical Dimension
alter that trait, but that is not inherently known
in the estimation of the heritability of a trait.
The question is, depending on the ability of the
individual to respond to a given environment
(treatment), will the interaction of the new or
altered environment, with the genetic factors
present, result in a change in the phenotype?
Nature, Nurture, or Both?
It has been stated that analyses of craniofacial
structures have led to the conclusion that they
have moderate to high estimated heritabilities
and that they are primarily a consequence of
nature rather than nurture.25 In a sense, this is
true for what estimates of the heritability repre-
sent, but it is often interpreted that genetic fac-
tors are influencing development independently
of the environment and that genetic factors have
controlled or determined the development. The
basic interpretation of the estimation of herita-
bility is that the genetic and environmental fac-
tors are separate, thus they can be portioned and
do not interact. This has been typified by the
phrase nature versus nurture, which by its very
construction defines a separation and even op-
position.
To say, for example, based on heritability es-
timates of a particular sample that the anterior
face height is 70% genetic and 30% environ-
mental gives a misleading dichotomy between
genetic and environmental factors and obscures
the fact that most if not all human disease (and
development) results from the interaction be-
tween genetic susceptibility and environmental-
moderating factors.27 This is true even for con-
ditions in which an environmental influence is
known to be strong, such as in smoking and
oropharyngeal cancer. Everyone who smokes
does not develop cancer, which indicates an in-
teraction of smoking with other factors, includ-
ing genetic susceptibility.28 Essentially all aspects
of normal and abnormal development are in
some way the result of the interaction of genetic
and environmental factors; thus, there is no
compelling reason to label a trait or condition as
being either genetic or environmental.27
Searching For Genetic Factors
Heritability estimates can indicate the relative
contribution or influence that genetic factors
Home | TOC |          
Index
117
have had on a trait, which is a consideration with
regard to the feasibility of a search for identify-
ing those factors. The search for DNA markers
linked with certain phenotypes may indicate ar-
eas of the genome that have a gene or genes that
influence the phenotype; however, this process
does not necessarily precisely define what gene
in the area is contributing or what allele of that
gene may be more influential than others. How-
ever, the search for markers linked with certain
phenotypes can indicate areas of the genome
that contain influential genes that were previ-
ously not known or even suspected to have an
influence on the phenotype. Once a particular
gene or genes in an area of the genome are
identified, they become candidate genes for spe-
cific analysis of their structure to pinpoint the
relevant allele (s).
Study of the influence of particular genetic
factors on development may be performed by
using a candidate gene chosen for the function
of its associated protein. An example is a study of
the association of the Pro561Thr (P56IT) vari-
ant in the growth hormone receptor gene
(GHR), which is considered to be an important
factor in craniofacial and skeletal growth. Out of
a normal Japanese sample of 50 men and 50
women, those who did not have the GHR P56IT
allele had a significantly greater mandibular ra-
mus length (condylion-gonion) than did those
with the G/ffiP56IT allele. The average mandib-
ular ramus height, in those with the GHR P56IT
allele, was 4.65 mm shorter than the average for
those without the GHR P56IT allele. This signif-
icant correlation between the GHR P56IT allele
and shorter mandibular ramus height was con-
firmed in an additional 80 women.29
Interestingly, the association was with the
mandibular ramus height but not mandibular
body length, maxillary length, or anterior cra-
nial base length. These data suggest an effect
that is site, area, or region specific. Although it
was concluded that the GHR P56IT allele may be
associated with decreased growth of mandibular
height and can be a genetic marker for it, it is
not clear if the effect is directly on the mandible
and/or on another nearby tissue or matrix. It
would also be interesting to see what effect dif-
ferent diet consistencies have on individuals with
and without the GHR P56IT allele, as a way of
looking at genetic and environmental interac-
tion.
<<    
     Article
      >>
118 James K. Hartsfield, Jr
Summary and Conclusion
The relevance of analyzing the development of
the vertical dimension to clinical practice is first
to determine if there is a vertical dimension
component to the malocclusion and then ascer-
tain what factors are having the greatest influ-
ence on the vertical dimension problem in that
individual. Unfortunately, at this time, studies
on the genetic and environmental factors that
influence the development of vertical dimension
are representative of the samples studied and
not necessarily of any particular individual. In
addition, the extent that a particular trait is in-
fluenced by genetic factors may have little if any
effect on success of environmental (treatment)
intervention. It may be that genetic factors that
influenced a trait will also influence the re-
sponse to intervention to alter that trait, or other
genetic factors may be involved in the response.
Therefore, the possibility for altering the envi-
ronment to gain a more favorable dimension is
theoretically possible, even in individuals in
which the vertical dimension does have a rela-
tively high genetic influence. However, the ques-
tion of how environmental and genetic factors
interact (a question that essentially cannot be
answered in estimates of heritability) is most
relevant to clinical practice because it may ex-
plain why a particular alteration of the environ-
ment (treatment) in one compliant patient may
be successful and not in another. Study of these
environmental and genetic factors has been dif-
ficult at the clinical level because of the relatively
small sample sizes and lack of markers to analyze
genetic diversity from one patient to the next.
Animal studies using inbred strains compare
the different responses of an environmental fac-
tor against consistent genotypes and the effect of
different background genotypes on the pheno-
typic expression of a specific gene mutation.30'31
Although mice, the most commonly used mam-
malian species for genetic inbred strain studies,
have been used and will continue to be used for
the study of genes that cause disease and aber-
rations in mammalian development, their differ-
ent craniofacial morphology (ie, the presence of
snouts and single dentition with incisors and
molars only) may not be readily applicable to
some of the clinical questions orthodontists have
regarding craniofacial growth in humans. Stud-
ies using mating crosses of various inbred strains
Home | TOC |          
Index
of mice help estimate the number of genes that
influence a phenotype. The development of the
mouse genome project, not far behind the hu-
man genome project, will increase the number
of known DNA markers that may be used in the
study of putative relevant genetic factors and
genetic-environmental interactions, which may
then be tested for in the human population.
The human genome project resulted in not
only a single human genome sequence com-
posed of overlapping parts from many humans
but also cataloged some 1.4 million sites of vari-
ation in the human genome sequence. This
increased number of variations (or polymor-
phisms) may be used as markers to perform
genetic (including genetic-environment interac-
tion) analysis in an outbred population such as
human beings. Our genome varies from one
individual to the next, most often in terms of
single-base changes of the DNA called single-
nucleotide polymorphisms. The main use of this
human single-nucleotide polymorphism map
will be to determine the contributions of genes
to diseases (or nondisease phenotypes) that
have a complex, multifactorial basis. Although
the scale of such studies could be daunting and
there are still problems to solve, the potential for
studying how natural variation leads to each one
of our qualities is significant. This approach may
be the best opportunity yet to better understand
the roles of nature and nurture rather than na-
ture versus nurture in development.32
Acknowledgment
The author thanks Dr W. Eugene Roberts for reviewing the
manuscript and Ms. Robyn Tibbs, Ms. Madeline Hawkins,
and Ms. Claudette Maurer for retrieving and copying refer-
ences.
References
1. Baltimore D: Our genome unveiled. Nature 409:814-816,
2001
2. Mulvihill JJ: Craniofacial syndromes: no such thing as a
single gene disease. Nature Genet 9:101-103, 1995
3. Park Wf, Bellus GA, Jabs EW: Mutations in fibroblast
growth factor receptors: phenotypic consequences dur-
ing eukaryotic development. Am J Hum Genet 57:748-
754, 1995
4. Escobar V, Bixler D: On the classification of the acro-
cephalosyndactyly syndromes. Clin Genet 12:169-178,
1977
5. Everett ET, Britto DA, Ward RE, et al: A novel FGFR2
gene mutation in Crouzon syndrome associated with
<<    
     Article
      >>
Development of the Vertical Dimension
apparent nonpenetrance. Cleft Palate-Craniofacial J 36:
533-541, 1999
6. Jorde LB, Carey JC, Bamshad MJ, et al: Medical Genetics
(ed 2). St. Louis, Mosby, 2000
7. Goodenough U: Genetics (ed 3). Philadelphia, Saim-
ders, 1984
8. Lynch M, Walsh B: Genetics and Analysis of Quantitative
Traits. Sunderland, Sinauer, 1998
9. LaBuda MC, Gottesman, II, Pauls DL: Usefulness of twin
studies for exploring the etiology of childhood and ad-
olescent psychiatric disorders. Am J Med Genet 48:47-59,
1993
10. Harris EF, Johnson MG: Heritability of craniometric and
occlusal variables: a longitudinal sib analysis. Am J
Orthod Dentofacial Orthop 99:258-268, 1991
11. Tanner JM: Hormonal, genetic, and environmental fac-
tors controlling growth, in Harrison GA, Weiner JS,
Tanner JM, et al (eds): Human Biology (ed 2). Oxford,
Oxford, University Press, 1977, pp 343
12. Vogel F, Motulsky AG: Human Genetics: Problems and
Approaches (ed 2). New York, NY, Springer-Verlag, 1986
13. Lobb WK: Craniofacial morphology and occlusal varia-
tion in monozygous and dizygous twins. Angle Orthod
57:219-233, 1987
14. Savoye I, Loos R, Carels C, et al: A genetic study of
anteroposterior and vertical facial proportions using
model-fitting. Angle Orthod 68:467-470, 1998
15. Vanco C, Kasai K, Sergi R, et al: Genetic and environ-
mental influences on facial profile. Aust Dent J 40:104-
109, 1995
16. Lundström A, McWilliam JS: A comparison of vertical
and horizontal cephalometric variables with regard to
heritability. Eur J Orthod 9:104-108, 1987
17. Lundström A, McWilliam J: Comparison of some ceph-
alometric distances and corresponding facial propor-
tions with regard to heritability. Eur J Orthod 10:27-29,
1988
18. Horowitz SL, Osborne RH, DeGeorge FV: A cephalomet-
ric study of craniofacial variation in adult twins. Angle
Orthod 30:1-5, 1960
19. Nakata M, Yu PL, Davis B, et al: Genetic determinants of
cranio-facial morphology: A twin study. Ann Hum Genet
37:431-443, 1974
20. Corruccini RS: An epidemiologic transition in dental
Home | TOC |          
Index
119
occlusion in world populations. Am J Orthod 86:419-
426, 1984
21. Byard PJ, Poosha DV, Satyanarayana M, et al: Family
resemblance for components of craniofacial size and
shape. J Craniofac Genet Dev Biol 5:229-238, 1985
22. Trask GM, Shapiro GG, Shapiro PA: The effects of pe-
rennial allergic rhinitis on dental and skeletal develop-
ment: a comparison of sibling pairs. Am J Orthod Dento-
facial Orthop 92:286-293, 1987
23. Hannuksela A: The effect of moderate and severe atopy
on the facial skeleton. Eur J Orthod 3:187-193, 1981
24. Sassouni V, Friday GA, Shnorhokian H, et al: The influ-
ence of perennial allergic rhinitis on facial type and a
pilot study of the effect of allergy management on facial
growth patterns. Ann Allergy 54:493-497, 1985
25. King L, Harris EF, Tolley EA: Heritability of cephalomet-
ric and occlusal variables as assessed from siblings with
overt malocclusions. Am J Orthod Dentofacial Orthop
104:121-131, 1993
26. Smith RJ, Baut HL: Problems and methods in research
on the genetics of dental occlusion. Angle Orthod 47:
65-77, 1977
27. Khoury MJ, Thrasher JF, Burke W, et al: Challenges in
communicating genetics: A public health approach.
Genet Med 2:198-202, 2000
28. Amador ÄG, Righi PD, Radpour S, et al: Polymorphisms
of xenobiotic metabolizing genes in oropharyngeal car-
cinoma. Oral Surg Oral Med Oral Path 93:440-445, 2002
29. Yamaguchi T, Maki K, Shibasaki Y: Growth hormone
receptor gene variant and mandibular height in the
normal Japanese population. Am J Orthod Dentofacial
Orthop 119:650-653, 2001
30. Bourgeois P, Bolcato-Bellemin AL, Danse JM, et al: The
variable expressivity and incomplete penetrance of
the twist-null heterozygous mouse phenotype resemble
those of human Saethre-Chotzen syndrome. Hum Mol
Genet 7:945-957, 1998
31. Everett ET, Hartsfield JKJr: Mouse models for craniofa-
cial anomalies, in Davidovitch Z, MahJ (eds): Biological
Mechanisms of Tooth Movement and Craniofacial Ad-
aptation. Boston, Harvard Society for the Advancement
of Orthodontics, 2000, pp 287-298
32. Chakravarti A: To a future of genetic medicine. Nature
409:822-823, 2001
<<    
     Article
      >> Home | TOC |          
Index

Diagnosis of the Vertical Dimension

James L. Vaden and Lloyd E. Pearson

The vertical dimension problem is complex and multifactorial. Not only

must the clinician recognize a vertical discrepancy abnormality, he/she must
be able to recognize its numerous components and understand their inter-
relationships. Many scientific investigators and orthodontic clinicians have
contributed to the body of knowledge to which we have access. This article
reviews some of the pertinent literature and offers some diagnostic and
treatment planning suggestions to the clinical specialist who struggles with
the vertical dimension enigma on a daily basis. (Semin Orthod 2002;8:
120-129.) Copyright 2002, Elsevier Science (USA). All rights reserved.

T he human face has been the subject of

study since man could first express himself.
As civilizations have risen and subsequently
faded away, one thing that has remained is art,
in most cases, drawings, paintings, and so on of
faces. During the Renaissance, da Vinci, Michae-
langelo, and Duhrer led other artists to study
faces. Facial proportion was discovered; there
were standards set for balance and harmony of
the lower face. In our specialty of orthodontics,
Angle was vitally concerned about the face. In
his sixth edition,1 he states, "One of the evil
effects of malocclusion is the marring or distort-
ing of the normal facial lines. It follows that, in
the application of the principles of orthodontia,
our efforts should be so directed as to mold and
modify these lines of inharmony to those of
harmony and facial beauty so far as lies within
the range of the possibilities of art, and of the
type and temperament of the individual. Our
opportunities for benefiting humanity are very
great in this field, far exceeding those offered by
any other branch of dental science, for patients
with facial lines so distorted as sometimes to be
a marked deformity and a source of constant
humiliation to themselves and their friends may
From the Department of Orthodontics, University of Tennessee,
Memphis, TN; and a Private Practice, Edina, MN.
Address correspondence to James L. Vaden, DDS, MS, Depart-
ment of Orthodontics, University of Tennessee, Health Science Cen-
ter, 875 Union Avenue, Memphis, TN 38163.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0803-0003$35.00/0
doi:10.1053/sodo.2002.125431
now be so treated as to bring about a complete
transformation of the facial expression, even to
the establishment of lines of beauty."
Tweed revolutionized orthodontic diagnosis
because of his concern for the balance and har-
mony of the lower face. Many in our specialty
have studied the face,2-6 developed diagnostic
guidelines for quantifying facial balance,7'11 and
proposed treatment regimens that give the orth-
odontic clinician a greater certainty that facial
balance and harmony is an attainable goal for
their patients.
The underlying theme that surfaces from all
artists and orthodontic investigators is the con-
cept that there cannot be good balance and
harmony in the lower face unless the vertical
dimension is within normal limits. The most
important prerequisite for facial balance is a
normal vertical dimension of the lower face.
Poulton12 conducted a study on cervical traction
and found that large lower anterior facial
heights were most often associated with a dis-
pleasing face. In their article on soft-tissue pro-
file preference, DeSmit and Dermaut13 created
three different series of nine profile photo-
graphs so that a total of more than 200 profiles
could be ranked by graduate dental students.
They found that differences in gender and orth-
odontic knowledge of the students seemed to
have no significant influence on their esthetic
preference. The results of their study confirmed
the importance of anteroposterior deviations
but suggested that unaesthetic facial profiles
that were a result of anteroposterior deviations
were completely overshadowed by long-face fea-

120 Seminars in Orthodontics, Vol 8, No 3 (September), 2002: pp 120-129

<<    
     Article
      >>
Diagnosis of the Vertical Dimension
tures—the long-face feature being more anaes-
thetic. Because of the challenge of the vertical
dimension, the subject of this article is of ex-
treme importance to the orthodontic specialist.
Not only must the specialist recognize the prob-
lem, the specialist must understand the diagno-
sis of the problem so that all facets and compo-
nents of the vertical dimension enigma are
understood. The clinician must be able to rec-
ognize the various components of a vertical di-
mension abnormality and understand the inter-
relationship of all the elements of the problem.
Before discussing the abnormal, it is prudent
to understand the normal. Two of the most ac-
cepted descriptions or publications of vertical
facial proportions have been published by Fra-
kas14 and Frakas and Munro.15 In these, they
describe the ideal face as vertically divided into
equal thirds by horizontal lines that approxi-
mate the hairline, the bridge of the nose, the ala
of the nose, and menton (Fig I). 16 Figure 1 also
shows that in the ideal vertically proportioned
face there is a further division of the lower one
1/3
1/3
1/3
Figure 1. The ideal facial proportions as described
by Frakas14 and Frakas and Munro.15 The frontal
view of the face is divided into equal thirds by
horizontal lines that approximate the hairline, the
bridge of the nose, and the ala of the nose and
menton. The lower third is further divided into an
upper third and a lower two third.
Home | TOC |          
Index
121
third of the face into an upper one third and a
lower two third. These divisions of the face can
be used by the clinician to help diagnose vertical
dimension problems. For example, does a pa-
tient have a disproportionately long lower facial
height because of vertical maxillary excess or to
excessive chin height? Conversely, is a short fa-
cial height caused by vertical maxillary defi-
ciency or a short chin height?17 By using these
accepted proportions as a guide, the patient
shown in Figure 2 has an excessive lower ante-
rior facial height, whereas the patient shown in
Figure 3 has diminished lower anterior facial
height. Although it is evident that both have
vertical dimension abnormalities by looking at
the face, measuring the facial proportions con-
firms this intuitive conclusion. A careful deter-
mination of the vertical proportions of the face
is therefore the first step in the diagnosis of a
vertical dimension problem.
Role of Skeletal and Dental
Relationships
After examining the face and quantifying its pro-
portions, the skeletal pattern and the teeth and
their relationships to each other must be scruti-
nized. However, a diagnosis of the vertical di-
mension is more complicated because vertical
discrepancy malocclusions are multidimen-
sional. For example, dentoalveolar abnormali-
ties can impact the skeletal pattern, and poor
skeletal patterns can cause dentoalveolar com-
pensations that are difficult for the clinician to
correct. The following variations can be present,
either alone or in combination: (1) maxilla:
maxillary posterior alveolar excess and inferiorly
positioned maxilla and (2) mandible: mandibu-
lar posterior alveolar excess and short mandib-
ular rami. Other abnormalities may include
superiorly positioned condylar fossa, obtuse cra-
nial base angle, and condylar resorption.
Any of these conditions, with or without ab-
errant mandibular growth rotation, can be a
causative factor in the vertical discrepancy mal-
occlusion.
Condylar Growth
A common scenario affecting the skeletal prob-
lem is mandibular growth and growth rotation,
which unfavorably impacts dentoalveolar devel-
<<    
     Article
      >>
122 Vaden and Pearson
oprnent in both the maxilla and mandible.
Bjork18'21 and Bjork and Skieller23'24 have per-
formed numerous studies that have shown that
the most common direction of condylar growth
is vertical, with some anterior component. Pa-
tients with a pronounced short lower anterior
facial height (Fig 4A and B) generally exhibit
upward and forward condylar growth (Fig 5).
These individuals generally have a deep vertical
overbite with a deep mentolabial sulcus and a
strong overdosed appearance.25 In contrast, pa-
Figure 3. An example of a
patient with a diminished
lower facial height.
Home | TOC |          
Index
Figure 2. An example of a
patient with an excessive
lower facial height.
tients with long-face syndrome (Fig 6A and B)
have a more posteriorly directed growth pattern
of the mandibular condyle (Fig 7). 26 > 27 These
backward growth rotators have increased ante-
rior facial height, a more posterior position of
the chin, and in extreme cases, an anterior open
bite may develop. Serial images of the patient
taken to monitor the direction of condylar
growth would be very useful for the diagnosis of
vertical growth. At the present time, serial imag-
ing poses certain concerns, most significantly
<<    
     Article
      >>
Diagnosis of the Vertical Dimension
Figure 4. A patient with a pro-
nounced short lower anterior fa-
cial height. (A) The cephalomet-
ric radiograph is also shown. (B)
radiation exposure. Advances in imaging tech-
nology may, in the future, permit the clinician to
use these methods for diagnostic purposes with
greater safety.
An understanding of the maxillomandibular
growth rotation of the patient would be most
helpful in the diagnosis of vertical variations.
Bjork28 has contributed information that offers
some guidelines for the clinician to assist in the
determination of the growth rotation of the
mandible so that the concomitant vertical
changes are more easily understood. Bjork's
Home | TOC |          
Index
123
method of prediction of condylar growth rota-
tion from a cephalogram offers the clinician
some guidelines. Bjork identified seven specific
structural features that might develop as a result
of remodeling during a particular type of growth
rotation. Bjork's suggestions for predicting con-
dylar rotation have, however, not been widely
used by the specialty because (1) some of the
indicators cannot be easily seen on the average
cephalogram, (2) the use of the indicators is very
time-consuming for the clinician, and (3) there
has been no scientific validation of the suggested
<<    
     Article
      >>
124
Figure 5. An example showing upward and forward
condylar growth.
indicators because of difficulties encountered in
study design. Some in the specialty also question
whether several of the suggestions are valid in-
dicators of a particular type of growth rotation.
However, when used for their intended purpose,
as guidelines only, the indicators have some use-
ful clinical applications in the diagnosis of the
patient with vertical dysplasia (Table 1).
Using Bjork's guidelines, it is interesting to
study Figures 4B and 6B. Figure 4B, the forward
rotator, exhibits several of Bjork's indicators in-
cluding observations that (1) the condylar head
curves forward, (2) the mandular canal is
curved, (3) the symphysis has a backward cant,
(4) the interincisal angle is obtuse and, (5)
lower anterior facial height is short. Figure 6B,
Table 1. Bjork's Seven Structural Guidelines28
Inclination of the condylar head
Curvature of the mandibular canal
Shape of the mandibular lower border
Inclination of the symphysis (Anterior aspect just below "B" point)
Interincisal angle
Interpremolar or intermolar angles
Anterior lower face height
Home | TOC |          
Index
Vaden and Pearson
the backward rotator exhibits (1) a straight in-
clination of the condyle, (2) a relatively straight
mandibular canal, (3) the symphysis slopes for-
ward and, (4) lower anterior facial height is
long.
Isaacson,29 Isaacson et al,30 and Schudy,31 fol-
lowing on Bjork's reports, studied jaw rotation
caused by vertical condylar growth. A succinct
summary of the findings of these investigators is
that a forward mandibular rotation occurs when
vertical condylar growth exceeds the sum of the
vertical growth of the maxillary sutures and the
maxillary and mandibular alveolar processes. If
growth of the maxillary sutures and the maxil-
lary/mandibular alveolar processes exceeds ver-
tical condylar growth, a backward rotation
occurs, and the face becomes longer. An under-
standing of the effect of condylar growth on
mandibular position is fundamental if the clini-
cian is to adequately and appropriately diagnose
a vertical dimension abnormality.
Anterior and Posterior Facial Height
Vertical dimension skeletal abnormalities are
not solely caused by condylar growth direction.
They are also caused by differences in anterior
facial height and posterior facial height develop-
ment. These differences in height development
can lead to rotational growth or to changes in
mandibular position that greatly influence the
position of the chin.30 Etiologies influencing un-
favorable differences in development of anterior
and posterior facial height are multifactorial.
These factors can, for simplicity, be subdivided
into those caused by (1) dentoalveolar develop-
ment and (2) environmental factors.
Dentoalveolar Development
Issacson et al33 studied dentoalveolar develop-
ment in three groups of subjects—those with
Forward Rotator Backward Rotator
Curves forward and back Straight or slopes up
Curved Straight
Curved downward Notched
Slopes backward Slopes forward
Vertical or obtuse Acute
Vertical or obtuse Acute
Short Tall
<<    
     Article
      >>
Diagnosis of the Vertical Dimension
Figure 6. An example of a pa-
tient with long-face syn-
drome.
short anterior facial height, those with average
anterior facial height, and those with excessive
anterior facial height. The amount of maxillary
posterior alveolar development was found to de-
crease as the MP-SN angle decreased. In patients
with long anterior facial height (high MP-SN
angles), the mean distance from the occlusal
plane to the inferior edge of the palate was 22.50
mm. This distance decreased to 19.6 mm for the
average group and 17.1 mm for the group with
Home | TOC |          
Index
125
short anterior facial height (low MP-SN angles).
This difference of 5.1 mm of dentoalveolar de-
velopment between the high angle and low an-
gle groups is of significance.
Mandibular posterior alveolar development
similarly decreased with decreases in the
MP-SN angle but much less dramatically than
those found in the maxilla. Mandibular height
showed a mean of 31.2 mm for the long ante-
rior face height group, 28.2 for the average
<<    
     Article
      >>
126 Vaden and Pearson
Figure 7. An example showing a posterior-directed
growth pattern of the mandibular condyle.
group, and 28.3 for the short anterior face
height group.
The findings of the Issacson et al33 study were
confirmed in a study performed by Janson et
al.34 These investigators found that all dentoal-
veolar heights were significantly greater in long
anterior facial height patients than in patients
with normal facial height. Also, in the short
lower anterior facial height, all dentoalveolar
heights were significantly shorter than in the
normal lower anterior facial height group.
The differences in dentoalveolar develop-
ment, most particularly in the maxilla, have a
significant impact on the anterior facial height
of the orthodontic patient. Moller and Inger-
vall35 and Thilander36 have postulated that ex-
cessive maxillary posterior dentoalveolar devel-
opment is associated with weaker masticatory
musculature in high-angle patients compared
with the strong musculature commonly associ-
ated with short anterior facial height patients.
Enviornmental Role -Swallowing and
Tongue Posture
The role of tongue posture, swallowing, and
breathing are still subjects of debate, argument,
and study in orthodontics. Their respective im-
pact on the vertical dimension are in need of
continued study and research.
Home | TOC |          
Index
Mouth breathing. The relationship between
mouth breathing, altered posture, and the de-
velopment of malocclusion is not as clear cut as
the theoretical outcome of shifting to oral respi-
ration might appear at first glance.37 Recent ex-
perimental studies have only partially clarified
the situation. Current experimental data for the
relationship between malocclusion and mouth
breathing are derived from studies of the nasal/
oral ratio in normal versus long-face children.38
The data from the study show that both normal
and long-face children are likely to be predom-
inantly nasal breathers under laboratory condi-
tions. A minority of the long-face children had
less than 40% nasal breathing, whereas none of
the normal children had such low nasal percent-
ages. When adult long-face patients are exam-
ined, the findings are similar: the number with
evidence of nasal obstruction is increased in
comparison to a normal population, but the
majority are not mouth breathers in the sense of
predominantly oral respiration.
Airway problems, such as large adenoids, ton-
sils, or blocked airways caused by septum devia-
tions, large conchae, or allergies are frequently
observed in high-angle patients and may affect
mandibular posture, allowing more freedom for
posterior eruption. This hypothesis is supported
by Linder-Aronson39'40 who showed closing of
the mandibular plane angle and reduction in
the anterior face height after removal of ade-
noids and tonsillectomy.
It appears that research on respiration, up to
the present time, has resulted in two opposing
views: (1) total nasal obstruction is highly likely
to alter the pattern of growth and lead to mal-
occlusion in experimental animals and humans,
and individuals with a high percentage of oral
respiration are overrepresented in the long-face
population, but (2) the majority of individuals
with the long-face pattern of deformity have no
evidence of nasal obstruction and must there-
fore have some other etiologic factor as the prin-
cipal cause.
In conclusion, it appears that mouth breath-
ing may contribute to the development of orth-
odontic problems but is difficult to indict as a
frequent etiologic agent. Clinically, most orth-
odontists refer mouth breathers to an otolaryn-
gologist for an evaluation. This problem should
be carefully evaluated during the diagnosis of a
patient with excess vertical dimension.
<<    
     Article
      >>
Diagnosis of the Vertical Dimension
Swallowing and tongue posture. One viewpoint
holds that tongue thrust swallowing is seen in
(1) younger children with reasonably normal
occlusion in whom it represents only a transi-
tional stage in normal physiologic maturation
and (2) in individuals who have displaced in-
cisors. In the latter, it is an adaptation to
the space between the teeth. Others argue that
tongue thrust swallowing simply has too
short a duration to have an impact on tooth
position. Pressure by the tongue against the
teeth during a typical swallow lasts for approx-
imately 1 second. A typical individual swal-
lows about 800 times per day while awake but
has only a few swallows per hour while asleep.
The total per day, therefore, is usually under
1,000. One thousand seconds of pressure, of
course, totals only a few minutes, not nearly
enough time, it is argued, to affect the equi-
librium.41
Most clinicians believe that if a patient has a
forward resting posture of the tongue, the dura-
tion of this pressure, even if very light, could
affect tooth position, vertically or horizontally.
Tongue-tip protrusion during swallowing is
sometimes associated with a forward tongue pos-
ture.
During the diagnosis of the patient with a
vertical dimension problem, the clinician must
understand that condylar growth, sutural lower-
ing of the maxillary complex, dentoalveolar de-
velopment, dental eruption, and the patient's
oral environment/habits are interrelated. There
is not generally a single causative factor that
predisposes the patient to too much or too little
vertical development of lower facial height. To
simplify, one might conclude as a general rule,
that when vertical condylar growth exceeds
tooth eruption (alveolar development), forward
mandibular rotation occurs. The result is in-
creased posterior facial height and an increase
in the ratio of posterior facial height to ante-
rior facial height. Conversely, if dentoalveolar
growth and tooth eruption are greater than ver-
tical condylar growth, the resultant mandibular
change is backward rotation. The anterior facial
height/posterior facial height ratio decreases.42
Environmental factors can play a role, but the
role is, at times, difficult to assess and varies from
patient to patient.
Home | TOC |          
Index
127
Diagnostic Considerations
Steep Excess Vertical Pattern:
The Backward Rotator
During differential diagnosis of the high-angle
patient, two questions must be asked. First,
where should the teeth be positioned? For the
patient with long anterior facial height, the man-
dibular anterior teeth are most often positioned
in a more retracted posture over basal bone. Lip
procumbancy can be best resolved if the man-
dibular anterior teeth are upright. The amount
of uprighting that must be achieved is a matter
of (1) clinical preference and must be deter-
mined during the treatment planning phase of
the treatment protocol or (2) the dictates of the
malocclusion. If indeed the facial profile of the
patient with excess vertical dimension is long, a
vertical reduction genioplasty can be effective
for facial esthetics. It is fundamental for the
clinician to be able to visualize the posttreat-
ment positions of the mandibular anterior teeth
during treatment plan preparation. Secondly,
will extractions be necessary? For many patients
with excessive lower anterior facial height, ex-
tractions may be necessary. The question of
which teeth should be extracted can be an-
swered only after a thorough and accurate dif-
ferential diagnosis.
The Overdosed Forward Rotator
Patients with short anterior vertical facial height
have a unique set of problems that require dif-
ferent diagnostic considerations. The following
diagnostic guidelines should be considered
when a patient with this skeletal pattern is
treated without surgical intervention.
Mandibular incisors, if well aligned before
treatment, can be allowed to remain in their
pre treatment position. Uprighting of mandibu-
lar incisors has an adverse impact on facial es-
thetics of the low-angle patient. However, the
mandibular incisors, if malaligned, should not
be proclined beyond their bony support for the
purpose of alignment.
Some overdosed forward rotator malocclu-
sions are characterized by a deep vertical over-
bite, maxillary incisor protrusion, and/or crowd-
ing. Correction of the overbite for these patients
is best accomplished by intrusion and retraction
of the maxillary incisors.
<<    
     Article
      >>
128 Vaden and Pearson
Treatment Concerns
During the diagnosis of the vertical dimension
problem, the clinician must be attentive to the
force systems that are planned for treatment and
understand that undesirable reactions to incor-
rectly applied force systems are disastrous. Pos-
terior facial height must be carefully controlled
for the high-angle patient because an increase in
posterior facial height will result in an increase
in anterior facial height.43"45 An increase in an-
terior facial height of high-angle patients is ca-
lamitous.
An important mechanical tooth manipulation
that must be accomplished during the treatment
of the patient with excess vertical dimension is
prevention of extrusion of the mandibular pos-
terior teeth, assuming that the maxillary poste-
rior vertical dimension is controlled by intrusive
forces (ie, headgear or other methods). Extru-
sion in the molar areas will prevent successful
correction of the malocclusion with excess ver-
tical dimension and long lower anterior face
height. It is important for the clinician to under-
stand these concepts during diagnosis and treat-
ment planning so that extraoral traction can be
planned to help control the vertical dimension
during treatment. There should be intrusive
forces to the posterior segments of both arches.
Additionally, Class II elastic wear can be one of
the most detrimental force applications that is
applied to a patient with long lower anterior
facial height. If Class II elastics are used indis-
criminately on the high-angle patient, the man-
dible drops down and back and increases the
sagittal discrepancy. Therefore, Class II elastic
use, or the absence of it, must be planned for
during diagnosis and treatment planning.
Pearson46 has published his results using ver-
tical pull chin cups and has provided evidence
that their use can create some effective skeletal
changes for the long-face patient. A thoughtful
diagnostician must consider the use of whatever
means is necessary to impact treatment and pre-
vent the lengthening of lower anterior facial
height during the course of treatment.
Summary
The orthodontic clinician must make a careful
differential diagnosis for each patient who seeks
his or her care. The diagnosis must analyze all
Home | TOC |          
Index
three components of a malocclusion—facial,
dental, and skeletal. Each component must be
carefully studied and understood so that (1) the
proper questions are asked and (2) the correct
diagnostic decisions are made to lead to an ef-
fective treatment plan.
Diagnosis of the vertical dimension is a com-
plex problem. Yet, it can be as simple as studying
a face and applying common sense diagnostic
tools to ascertain the reason that the lower face
is too long or too short. The vertical dimension
has been a subject of study and debate since
orthodontics became a specialty. Researchers in
the field of vertical dimension diagnosis, includ-
ing Bjork,18 Schudy,27 Nielsen,25 Isaacson,29
Pearson,26 and others, have provided the spe-
cialty many useful guidelines and concepts that
can be used by every orthodontic clinician as
they diagnose a malocclusion that is complicated
by a vertical dimension discrepancy. Orthodon-
tists should continue to use the work of these
researchers and clinicians for a foundation as
more studies are undertaken that will yield more
knowledge so that diagnosis of the vertical di-
mension becomes less art and more science.
References
1. Angle EH: Malocclusion of the Teeth and Fractures of
the Maxillae (ed 6). Philadelphia, PA, SS White Co,
1900, p 15
2. Tweed CJ: Indications for the extraction of teeth in
orthodontic procedure. Am J Orthod 30:405-428, 1934
3. Peck H, Peck S: A concept of facial esthetics. Angle
Orthod 40:284-317, 1970
4. Burstone CJ: Lip posture and its significance in treat-
ment planning. Am J Orthod 53:262-284, 1967
5. Hulsey CM: An esthetic evaluation of lip-teeth relation-
ships present in the smile. AmJ Orthod 57:132-144, 1970
6. Peck S, Peck L: Selected aspects of the art and science of
facial esthetics. Semin Orthod 1:105-126, 1995
7. Merrifield LL: The profile line as an aid in critically
evaluating facial esthetics. AmJ Orthod 52:804-821, 1966
8. Czarnecki ST, Nanda R, Currier F: Perceptions of a
balanced facial profile. Am J Orthod, Dentofacial Or-
thop 104:180-187, 1993
9. Ricketts RM: Divine proportions in facial esthetics. Clin
Plastic Surg 9:401-422, 1982
10. Steiner CC: Cephalometrics for you and me. Am J
Orthod 39:729-755, 1953
11. Holdaway RA: A soft tissue analysis and its use in orth-
odontic treatment planning: Part I. AmJ Orthod 84:1-
28,1983
12. Poulton DR: The influence of extraoral traction. AmJ
Orthod 53:8-18, 1967
13. DeSmit A, Dermaut L: Soft-tissue profile preference.
AmJ Orthod 86:67-73, 1984
<<    
     Article
      >>
Diagnosis of the Vertical Dimension
14. Frakas LG: Anthropometry of the Head and Face in
Medicine. New York, NY, Elsevier Science, 1981
15. Frakas LG, Munro JR: Anthropometric Facial Propor-
tions in Medicine. Springfield, IL, Charles C. Thomas,
1987
16. Proffit WR: Diagnosis and Treatment Planning in Con-
temporary Orthodontics. St. Louis, MO: Mosby, 2000
17. Sarver D, Proffit W, Ackerman J: Diagnosis and treat-
ment planning in orthodontics from orthodontics, in
Graber TM, Vanarsdall RL (eds): Current Principles and
Techniques (ed 3). St. Louis, MO, Mosby, 2000
18. Bjork A: Facial growth in man, studied with the aid of
metallic implants. Acta Odontol Scand 13:9-34, 1955
19. Bjork A: Variations in the growth pattern of the human
mandible: longitudinal cephalometric study by the im-
plant method. J Dent Res 400-411, 1963
20. Bjork A: Sutural growth of the upper face studied by the
implant method. Acta Odontol Scand 24:109-129, 1966
21. Bjork A: The use of metallic implants in the study of
facial growth in children, method and application. Am J
Phys Anthropol 29:243-254, 1968
22. Skieller V: Cephalometric analysis in the treatment of
overbite. Rep Congr Eur Orthod Soc 147-157, 1967
23. Bjork A, Skieller V: Facial development and tooth erup-
tion, an implant study at the age of puberty. Am J
Orthod 621:339-383, 1972
24. Bjork A, Skieller V: Normal and abnormal growth of the
mandible: A synthesis of longitudinal cephalometric im-
plant studies over a period of twenty five years Eur
J Orthod 5:1-46, 1983
25. Neilsen I L: Vertical malocclusions: Etiology, develop-
ment, diagnosis and some aspects of treatment. Angle
Orthod 61:247-260, 1991
26. Pearson LE: Vertical control in treatment of patients
having backward rotational growth tendencies. Angle
Orthod 48:130-140, 1978
27. Schudy FF: Vertical growth vs anteroposterior growth as
related to function and treatment. Angle Orthod 34:75-
93, 1964
28. Bjork A: Prediction of mandibular growth rotation. Am J
Orthod 55:585-599, 1969
29. Isaacson RJ: The geometry of facial growth and its effects
on the dental occlusion and facial form. J Charles H.
Tweed Int Found 9:21-38, 1981
30. Isaacson RJ, et al: Effects of rotational jaw growth on the
occlusion and profile Am J Orthod 3:276-286, 1977
31. Schudy FF: The rotation of the mandible resulting from
growth: Its implication in orthodontic treatment. Angle
Orthod 35:36-50, 1965
32. Proffit WM: The development of orthodontic problems,
in Proffitt WM, Fields HW (eds): Contemporary Orth-
odontics. St. Louis, MO, Mosby, 2000, pp 102-106
Home | TOC |          
Index
129
33. Isaacson JR, Isaacson RJ, Speidel TM, et al: Extreme varia-
tion in vertical facial growth and associated variation in
skeletal and dental relationships. Angle Orthod 41:219-
228, 1971
34. Janson G, Metaxas A, Woodside D: Variation in maxillary
and mandibular molar and incisor vertical dimension in
12 year old subjects with excess, normal, and short lower
anterior face height. Am J Orthod Dentofacial Orthop
106:409-418, 1994
35. Moller E: The chewing apparatus. Acta Physiol 69:571-
574, 1966
36. Ingervall B, Thilander B: Relationship between facial
morphology and activity of the mastecatory muscles.
J Oral Rehap 1:131-147, 1974
37. Vig KWL: Nasal obstruction and facial growth: the
strength of evidence for clinical assumptions. Am J
Orthod Dentofacial Orthop 113:603-611, 1998
38. Fields HW, Warren DW, Black K, et al: Relationship
between vertical dentofacial morphology and respira-
tion in adolescents. Am J Ortho Dentofacial Orthop
99:147-154, 1991
39. Linder-Aronson S: Effects of adenoidectomy on the den-
tition and facial skeleton over a period of five years, in
Cook JT (ed): Transactions of the Third International
Orthodontic Congress. St Louis, MO, Mosby, 1975
40. Woodside DG, Linder-Aronson S, Lundström A, et al:
Mandibular and maxillary growth after changed mode
of breathing. Am J Orthod Dentofacial Orthop 100:1-18,
1991
4L Proffitt WR: The etiology of orthodontic problems. Con-
temporary Orthodontics (ed 3). St. Louis, MO: Mosby, p
136
42. Schudy FF: The rotation of the mandible resulting from
growth—Its implications in orthodontic treatment. The
Vertical Dimension of the Human Face. Houston, TX: D
Armstrong & Co, 1992, pp 151-179
43. Pearson LE, Pearson BL: Rapid maxillary expansion with
incisor intrusion: a study of vertical control. Am J
Orthod Dentofacial Orthop 115:576-582, 1999
44. Klontz HA: Facial balance and harmony: an attainable
objective for the patient with a high mandibular plane
angle. Am J Orthod Dentofacial Orthop 114:176-188,
1998
45. Vaden JL: Alternative nonsurgical strategies to treat
complex orthodontic problems. Semin Orthod 2:90-113,
1996
46. Pearson LE: The management of vertical dimension
problems in growing patients, from the enigma of the
vertical dimension. In McNamara JA Jr (ed): Craniofa-
cial Growth Series 36, Center for Human Growth and
Development. Ann Arbor, MI, The University of Michi-
gan, 2000
<<    
     Article
      >> Home | TOC |          
Index

Early Treatment of Hyperdivergent Open-Bite

Malocclusions
Peter H. Buschang, MA, PhD, Wayne Sankey, DDS, MS, and
Jeryl D. English, DDS, MS

This article establishes the morphologic attributes that characterize hyper-

divergent open-bite (Hyp-OB) subjects, reviews associated etiologic factors,
compares the various treatment modalities, and provides a rationale and a
possible approach for early treatment. Treatment of Hyp-OB patients must
address three-dimensional dentoalveolar and skeletal problems in both
jaws. For early treatment to be successful, it must effectively deal with the
etiology of the problem and the resulting mandibular skeletal dysmorphol-
ogy. Of the various treatment approaches, including high-pull headgear,
extractions, and bite blocks, the vertical chin cup holds the greatest poten-
tial for mandibular skeletal modification. Although early treatment can be
theoretically justified based on psychosocial benefits and growth potential,
more clinical and experimental research is required to optimize the treat-
ment approach and define the long-term consequences. (Semin Orthod
2002;8:130-140.) Copyright 2002, Elsevier Science (USA). All rights reserved.

ten correct spontaneously.1'2 Because spontane-

T reatment is ultimately dependent on an
appropriate diagnosis, which in turn re-
quires an unambiguous description of the
problem. Before the advent of cephalometrics,
orthodontists necessarily focused on dental rela-
tionships and defined open-bite malocclusion
based on the vertical relationships of the maxil-
lary and mandibular teeth. By the early 1960s, it
became evident that distinctions needed to be
made between open-bite malocclusions that in-
cluded a skeletal component and those that did
not.
Simple open-bite malocclusions that do not
include skeletal components present less of a
challenge for orthodontists; such open-bites of-
ous correction occurs in up to 80% of mixed-
dentition open-bite cases, it has been suggested
that interceptive treatments are of little or no
value.3 Successful treatment approaches are also
available for persistent cases that do not self-
correct.4'5
In contrast, open-bite malocclusions that in-
clude skeletal components have proven to be
extremely challenging for orthodontists. Most
investigators have simply referred to them as
skeletal open bites6'7; Schudy8 characterized
them as hyperdivergent, which reflects the skel-
etal phenotype.

From the Department of Orthodontics, Baylor College of Den-
tistry, The Texas A&M University System Health Science Center,
Dallas, TX; Private Practice, Flower Mound, TX; and Department
of Orthodontics, University of Texas Health Science Center Houston,
Houston, TX.
Address correspondence to Peter H. Buschang, Department of
Orthodontics, Baylor College of Dentistry, The Texas A&M Univer-
sity System Health Science Center, 3302 Gaston Avenue, Dallas, TX
75246.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0803-0004$35.00/0
doi:10.1053/sodo.2002.125432
Morphologic Characteristics
The typical hyperdivergent open-bite (Hyp-OB)
patient presents with three-dimensional skeletal
and dentoalveolar problems pertaining to both
the maxilla and mandible. Variation in expres-
sion of traits among subjects should be ex-
pected. Within-subject differences clearly show
that most of the dysmorphology occurs in the
mandible.
Table 1 summarizes the problem list for the
maxilla. The most consistent findings across

130 Seminars in Orthodontics, Vol 8, No 3 (September), 2002: pp 130-140

<<    
     Article
      >> Home | TOC |          
Index

Management of the Vertical Dimension 131

Table 1. Maxillary Morphologic Characteristics of Untreated Open Bites

Dentoalveolar
Heights Heights
Author Ant Post Length SNA PPA Incisor Molar Width
Subtelny and Sakuda6 0 0 0 p 0 t t P

Bell105 p p p > p t 1
Isaacson et al106 0 p P i;> p t T 1
Nahoum et al107 i ? ip ip 0 0 P

Schendel et al108 0 p ip Tp p P

Lundström and Woodside109 1 0 p p P P

Fields et al110 0 0 p 0 p T t
P

Nanda95-96 1 0 ? p ip P p P

Haralabakis et al111
Trouten et al112
0
0
0
T
ip p
p i
tP tp 0
p
Lopez-Gavito et al113
Janson et al114
ip 0 ?
p
ip ip T T p
p
0 t t
Abbreviations: t , increase; j , decrease; O, no difference; ?, not reported.

studies pertain to anterior and posterior den-
toalveolar heights, which tend to be excessive.
When evaluated, the palatal plane angles have
been reported to be flatter because of decreased
anterior upper facial heights; posterior upper
facial heights do not appear to be affected.
There is also a tendency for maxillas to be
shorter and slightly recessive. When transverse
dimensions have been evaluated, the maxilla is
often reported to be narrow with an increased
incidence of posterior cross-bites.
The most consistent mandibular characteris-
tics are increased lower anterior facial heights,
steeper mandibular planes, and larger gonial
angles (Table 2). These traits combine to pro-
duce increased lower to upper and lower to total
anterior facial height ratios. Most studies also
report decreased posterior facial heights caused
by smaller ramus heights. It has also been shown
that the glenoid fossa is positioned more supe-
rior relative to sella turcica. Mandibular dentoal-
veolar heights are also generally reported to be
excessive in Hyp-OB patients.
Etiology of Hyp-OB Malocclusion
It is well established that abnormal muscles and
habits are associated with Hyp-OB malocclu-
sion.6 Finger, thumb, and tongue habits are per-
haps the best known physical factors that pro-
duce open-bite malocclusions by lowering the
mandible and preventing normal eruption.9 In-
dividuals with prolonged nonnutritive-sucking
habits have been repeatedly shown to have a

Table 2. Mandibular Morphologic Characteristics of Untreated Open Bites

Dentalveolar
Heights Corpus Gonial Heights
Author Ant Post Lt Angle MPA Retrogn Incisor Motar

Schudy 8
T 1 p p T t p t
Subtelny and Sakuda6 T 1 0 T T t 0 0
Bell105 t i p tP t T TP ?
Isaacson et al106 t 4 p T t? t
Nahoum et al107 t 1 i tp T 0
p i
Schendel et al108 t 1p p t T Tp
Lundström and Woodside109
Cangialosi115
T
t i
p
p
tt T
t
tp P
? p
Fields et al110
Nanda95'96
T
tP
1
0
0
p
tT tp 0
p
Tp Tp
Haralabakis et al111
Trouten et al112 P
1p 1*
p
T
t?
p

t
?
p
tp 0
p
Lopez-Gavito et al113
j
t tp p
? Tp T 0 0
Janson et al114 p p
T t
Abbreviations: f , increase; J,, decrease; O, no difference; ?, not reported.
*Females only.
<<    
     Article
      >>
132 Buschang, Sankey, and English
decrease overbite,10'14 increased overjet,10'11'13
decreased palatal width,11'15 and increased lip
incompetence.12
There is considerable evidence showing
smaller, less active, muscles and weaker bite
forces among hyperdivergent subjects. Masseter
and medial pterygoid muscle volumes and cross-
sectional areas have been positively correlated
with posterior face and ramal heights and neg-
atively correlated with the mandibular plane and
gonial angles.16'21 Kiliaridis and Kalebo22 showed
that women with thin masseter muscles had pro-
portionately longer faces. Bakke et al23 found
that muscle thickness at the most voluminous,
superficial portion of the masseter was signifi-
cantly correlated with bite force, anterior face
height, and mandibular inclination. Moreover,
the amplitudes of masseter and temporalis elec-
tromyographic (EMG) activity at rest, during
swallowing, chewing, and maximal biting have
also been negatively related with mandibular hy-
perdivergence in children 9 to 11 years of age.24
Bakke and Michler25 showed that EMG activity
during maximal voluntary contraction was neg-
atively correlated with anterior face height, man-
dibular inclination, vertical jaw relation, and go-
nial angle.
The smaller, less active muscles that charac-
terize Hyp-OB subjects produce lower than nor-
mal bite forces.26 In adults, stronger molar and
incisor bite forces are highly correlated with
smaller gonial angles, lower mandibular plane
angles (MPA), and larger posterior face
heights.27'28
The relationship between weaker bite forces
and hyperdivergence is not as strong for chil-
dren as adults. Proffit and Fields29 originally
showed no significant differences in occlusal
forces between long-face and normal children 6
to 11 years of age. By using more refined analyt-
ical techniques, Garcia-Morales et al30 recently
found that children (age 9.3 ± 3.6 years) with
greater skeletal hyperdivergence had poorer me-
chanical advantage and lower maximum bite
forces.
Although the association is well established, it
remains controversial whether lower occlusal
forces in long-face individuals produce hyperdi-
vergence26 or whether weaker occlusal forces are
a biomechanical result of the long vertical facial
proportions.31 This controversy is at least par-
tially resolved by studies showing typical Hyp-OB
Home | TOC |          
Index
features among individuals with genetically de-
termined neuromuscular diseases that weaken
masticatory muscles. Patients with myotonic dys-
trophy have 2 to 3 times less EMG activity of the
temporalis and masseter muscles during maxi-
mum clenching, lower maximum bite forces,
anterior open bite malocclusions, and hyperdi-
vergent growth patterns.32'33 Similarly, individu-
als with spinal muscular atrophy have open-bite
malocclusions, increased vertical skeletal dimen-
sions, steep mandibular plane angles, and hyper-
divergent facial patterns.34 They also have
weaker, less efficient, masticatory muscles that
fatigue more quickly than those of matched con-
trols.35
Numerous experimental animal studies and
human clinical studies have also shown relation-
ships between the Hyp-OB phenotype and
chronic upper, middle, and lower airway ob-
structions. Rhesus monkeys forced to become
mouth breathers showed greater than expected
increases in anterior face height associated with
mandibular lowering, decreases in arch width,
and changes in tongue position and shape.36 In
fact, both light and heavy nasal obstruction pro-
duces downward and backward mandibular ro-
tation, increased gonial angulation, and anterior
open bite in primates37 and rodents.38
The relationship between airway obstruction
and skeletal hyperdivergence is well established
in humans.39-40 Compared with their nose-
breathing counterparts, chronically allergic
mouth-breathing children 6 to 12 years of age
have narrower maxillas; greater incidence of
posterior cross-bites; longer anterior facial
heights; steeper palatal, occlusal, and mandibu-
lar planes; larger gonial angles; and more ret-
rognathic mandibles.41 Mouth breathers with hy-
pertrophied adenoids have narrow maxillas,
lower tongue positions, proclined incisors, and
increased lower anterior facial heights.42 Nor-
malization of incisor inclination, maxillary arch
width, depth of the bony nasopharynx, and man-
dibular plane inclination has been reported for
children after adenoidectomies to correct severe
nasopharyngeal obstruction.43'44 Children with
enlarged tonsils also have more retrognathic
mandibles, larger lower anterior facial heights,
and larger MPAs.45 The controversy concerning
airway relates more to our ability to accurately
diagnose airway problems for individuals rather
<<    
     Article
      >> Home | TOC |          
Index

Management of the Vertical Dimension 133

Table 3. Outcomes of Various Treatments Modalities for Hyperdivergent Open-Bite Malocclusion

Effect Site HPHG HPGH + Splint
Condylar growth/amt - -
Condylar growth/dir 0 0
Mandibular AP Position 0 0
Posterior Face Height - -
Anterior Face Height 0 0
Skeletal AP Relations + +
U6 Position ++ ++
L6 Position - 0
Overbite 0 0
Overjet + +
Abbreviations: 0, no difference; —, not significant; +, significant; + + , very significant; + + +, extremely significant.
Extract HPHG + Extract Passive PBB Active PBB vcc
0 0 0 0 +
0 0 0 +
0 0 + + +
0 0 + 0 +
0 0 + + 0
0 0 + + +
- ++ + + +
- - + + +
+ ++ +++ + +
+ ++ +
+ +

than in the relationship between obstruction
and skeletal hyperdivergence.
The common denominator among Hyp-OB
subjects with habits, weak muscles, and airway
obstructions appears to be a lowered mandibu-
lar posture. An individual's response to airway
obstruction is immediate and predictable; the
activity of the neck and masticatory muscles,
head posture, and mandibular posture change
quickly.46'47 Habits and weak elevator muscles
might also produce changes in mandibular pos-
ture that, during growth, necessitate develop-
mental adaptations leading to the Hyp-OB dys-
morphology.
Treatment Modalities
Based on the forgoing, treatment of Hyp-OB
patients must address three-dimensional prob-
lems pertaining to the dentoalveolar and skele-
tal structures of both jaws. Although hyperdiver-
gent Class III open-bite cases are perhaps the
most difficult to treat nonsurgically, they occur
less frequently. In contrast, the Class II is the
most common type of hyperdivergent patient. A
severe hyperdivergent Class II patient requires
reductions in dentoalveolar height throughout
the maxilla and mandible (Table 3); reductions
in gonial angulation; increased palatal plane an-
gulation; maxillary expansion; and true mandib-
ular autorotation to increase posterior mandib-
ular height, redirect condylar growth, decrease
anterior lower facial height, and reposition the
chin forward (Fig 1).
High-Pull Headgear
High-pull headgear (HPHG) has traditionally
been the appliance of choice for treating
Hyp-OB patients because it has been shown to
effectively hold maxillary sutural growth and ver-
tical dentoalveolar development.48'53 Although
animal studies have reported absolute distal and
superior displacement of metallic implants in
the maxilla,54'55 human studies do not support
increased autorotation.49'56 In fact, Baumrind
and coworkers51'57 found that high-pull patients
displayed relative increases in the MPA and re-
duced condylar growth.
Acrylic splints with HPHG create one large
anchor unit that prevents unfavorable tipping of
the upper molars.54'55'58'61 Caldwell and collabo-
rators62 showed that this approach produced a
superior and distal displacement of the maxilla,
reductions of the Sella-Nasion-A (SNA) angle,
clockwise rotation of the palatal plane, and rel-
ative intrusion of the upper molars. Importantly,
they also reported increased lower molar erup-
tion, decreased mandibular growth, and in-
creased Sella-Nasion-B (SNB) angulation.
Extractions
Extraction therapy for hyperdivergent patients is
predicated on the belief that molars moved me-
sially out of the occlusal wedge increase man-
dibular autorotation, decrease anterior facial
height, and reduce open-bite malocclusions.
However, Yamaguchi and Nanda63 reported no
differences in molar position or the A point-
Nasion-B (ANB) angulation between extraction
and nonextraction patients treated with high-
pull face bows. Staggers64 showed no significant
differences between Class I nonextraction and
extraction cases in changes of anterior facial
height, upper to lower face height ratios, MPAs,
posterior to anterior facial height ratios, or the
distances of the molars to the palatal and man-
<<    
     Article
      >>
134 Buschang, Sankey, and English
OPEN-BITE
MALOCLUSIONS
Hyperdivergent
C lass II
Most common
Figure 1. Problem lists for open-bite malocclusions. (AFH = anterior facial height; ant = anterior; post
posterior; md — mandible; mx = maxilla)
dibular planes. After extractions, Class II pa-
tients show decreases in the ANB angle, suprae-
ruption of the lower molars, and increases in the
MPA.65 Pearson66 evaluating cases with moder-
ately steep MPAs treated with extractions and
occipital headgears also reported significant su-
praeruption of the lower molars. Clearly, for-
ward mandibular rotation and loss or even main-
tenance of vertical facial and dentoalveolar
heights does not occur in the extraction pa-
tients.
Combined HPHG and extraction treatment
produces results similar to extractions only. The
primary differences pertain to the molars; the
vertical movements of the upper molar are bet-
ter controlled, but the lower molar shows even
greater compensatory supraeruption.65'66 The
positive effects of combined HPHG and extrac-
tion treatment are confined to the maxillary
dentoalveolar regions. If the goal is to improve
the orientation, position, and shape of the man-
dible, then other treatment approaches may be
required.
Posterior Bite Blocks
Posterior bite blocks (PBB) have been shown to
effectively modify vertical skeletal patterns in an-
Home | TOC |          
Index
Dental
Skeletal
C lass I Class III
Least common
imal models67'73 and humans.74"81 However,
PBBs hinge the mandible open beyond its rest-
ing position by varying amounts, which tends to
increase the gonial angle.69'72'79 McNamara79
concluded that the maxillary complex was most
affected by PBB, although changes have been
reported to occur throughout the craniofacial
complex.67
Animal studies evaluating repelling magnets
embedded in bite-block appliances show superi-
or-anterior maxillary displacement and molar
intrusion. However, they also show greater po-
tential for root resorption, deviated mandibular
jaw posture that could produce skeletal asymme-
tries, and lateral open bites.70'73'75'76 Kalra and
Burstone,76 evaluating fixed magnetic bite
blocks, reported increased mandibular length,
decreased facial convexity, intrusion of the lower
and upper molars, improved overjet and molar
relationships, and small decreases in the man-
dibular plane angle and y axis. After treatment,
posterior tooth eruption closed the posterior
open bite, and the transverse jaw deviations self-
corrected. More pronounced treatment effects
have been reported for magnetic than spring-
loaded bite blocks.80
Iscan and associates74 compared the effects of
<<    
     Article
      >>
Management of the Vertical Dimension
a spring-loaded bite block (SLBB) worn for 6
months to a passive bite block and a vertical chin
cup worn for 8 months. Both groups showed
similar amounts of forward maxillary displace-
ment, increases in mandibular length, posterior
molar intrusion, mandibular autorotation, in-
creased overbite, and reductions of anterior fa-
cial height. The SLBB group showed greater
reductions in the ANB angle and lower molar
intrusion, whereas the passive bite block and
vertical chin cup group showed greater improve-
ment in lower facial height and overbite. The
SLBB also showed gonial angle increases, an
undesirable effect.
Bellinger's77 active vertical corrector, with re-
pelling magnets embedded in bite blocks and
acrylic shields to prevent lateral jaw deviations, is
used in conjunction with a vertical chin cup.
Intrusion of the posterior teeth, mandibular au-
torotation, and reductions in anterior height
have been shown after 4 to 7 months of treat-
ment. Bellinger78 has also reported good long-
term stability for five treated cases. Barbre and
Sinclair,81 evaluating the effects of the active
vertical corrector without the vertical chin cup,
showed increased overbite, decreased lower an-
terior facial height, and reductions in mandibu-
lar plane angulation associated with upper and
lower molar intrusion.
Vertical Chin Cup
Pearson82-84 has used the vertical chin cup in the
mixed and permanent dentition to reduce the
mandibular plane angle and limit increases in
anterior facial height. A case treated with a ver-
tical-pull chin cup in conjunction with a Kloehn
cervical headgear showed significant dental and
skeletal alterations; upper molar eruption and
descent of the maxilla were inhibited while man-
dibular growth was redirected toward a more
horizontal direction.85 Treatment success was at-
tributed to the increase in posterior facial
height. Chin cups have also been used during
active rapid palatal expansion (RPE) therapy to
minimize the vertical displacement of the max-
illa and control the opening of the mandibular
plane angle.86 Importantly, the vertical chin cup
is the only appliance shown to effectively alter
mandibular shape by increasing posterior
heights, redirecting condylar growth, and de-
creasing gonial angulation.
Home | TOC |          
Index
135
Timing of Treatment
Early treatment can be best justified based on
psychosocial benefits and growth potential for
correction. Because facial appearance is the
most important determinant of physical appear-
ance87 and the oral region contributes most to
overall facial appearance,88 abnormalities in the
oral region negatively affect interpersonal rela-
tions,89 how individuals are perceived by oth-
ers,90-91 and self-perception.92 Early treatment
might also obviate the costs and risks associated
with relatively complex twojaw surgical proce-
dures typically performed to correct Hyp-OB
malocclusions.
In terms of growth potential, early treatment
of Hyp-OB is predicated on the knowledge that
(1) it can be diagnosed early, (2) the phenotype
does not self-correct, (3) potential for mandib-
ular rotation and associated remodeling is great-
est during childhood, (4) certain characteristics
may require long periods of growth to fully cor-
rect, and (5) treatments are available that can
correct complex three-dimensional configura-
tion of problems.
The Problem Is Apparent Early and Does
Not Self-Improve
Longitudinal studies agree that the Hyp-OB phe-
notype develops early and, on average, does not
worsen with age. Longitudinal comparisons
show a strong tendency to maintain long, aver-
age, and short facial type; most individuals
(77%) present with the same facial type at 5 and
25.5 years of age.93 Although there are growth
differences between open- and deep-bite sub-
jects in anterior face height,94 the overall pattern
of development is established early, even before
the eruption of the first permanent molars, and
maintained during growth. The mandibular
plane and gonial angles of both open- and deep-
bite groups decrease with age, which tend to
decrease the absolute magnitude of skeletal im-
balance for open-bite subjects.95 Untreated ver-
tical growers show less increase of Sella-Nasion-
Pogonion (S-N-Pg), less decrease in the MPA,
and less decrease in the gonial angle than hori-
zontal growers.96 If treatment mechanics of
long-face individuals require orthopedic modifi-
cation, it has been suggested that treatment
should be initiated before the adolescent spurt.
<<    
     Article
      >> Home | TOC |          
Index

136 Buschang) Sankey, and English

Based on various measures of vertical devel-

opment, Garcia-Morales and Buschang97 re-
cently showed that untreated subjects classified
as hyperdivergent at 6 years of age maintain
their dysmorphology relative to average or hypo-
divergent subjects through 15 years of age (Fig
2). Correlations between the individuals' 6- and
15 year-old phenotypes ranged between 0.5 to
0.75. Approximately 64% of the 6-year-old sub-
jects classified as having a high MPA were also
classified with high MPAs at 15 years; 28% had
average MPAs and 8% had low MPAs (Fig 3A). At 6 Yrs Changes 6-15
Changes that occurred were related to the 15-
year-old phenotype (correlations range 0.5-0.7)
but not to the 6-year-old phenotype. In other
words, knowing what a young child looks like
helps predict his/her adult status but not his/
her growth changes. Although most 6-year-old
subjects with high MPAs improved, 24% became
more hyperdivergent (Fig 3B). Together, these
data suggest that the apparent stability displayed
between group averages should not be expected
Figure 3. (A) Frequencies of subjects with high, me-
for individual patients. These data emphasize dium, and low MPAs (S-N/Go-Me) at 6 and 15 years of
that early interceptive treatment of Hyp-OB pa- age. (B) Frequency of MPA (S-N/Go-Me) increasing
tients cannot be justified because we cannot cur- and decreasing between 6 to 15 years for subjects
rently predict whose malocclusions will worsen. classified with high,medium and low MPA (S-N/Go-
Me) at 6 years.
Childhood as a Period of Greater Potential
Overall growth potential and the capacity of cer-
forward rotation than low angle boys, with the
tain characteristics to change are greater during
difference being most pronounced during child-
childhood than adolescence. It is also possible
hood.98 Changes in the MPA angle also show
that younger children would be more coopera-
significant group differences during childhood
tive and more willing to undergoing complex
but not during adolescence. Spady et al"
long-lasting treatment regimens than adoles-
showed that true mandibular rotation was signif-
cents.
icantly greater during childhood than adoles-
Longitudinal analyses show that boys with
cence; rotation was particularly marked during
high angles display significantly less true or total
the transition from primary to early mixed den-
titions.
• Low Medium - - High]
As shown, Hyp-OB patients often present with
excessive maxillary and mandibular dentoalveo-
lar heights. Relative intrusion of the teeth, espe-
cially mandibular teeth that show less eruption
i)37 potential than their maxillary counterparts, may
33
require considerable time for extreme cases. For
example, the average mandibular incisor erupts
5.0 mm for boys and 3.7 mm for girls between 8
10 11 to 15 years of age; the first molar erupts 0.5 to
Age
0.7 mm more than the incisor.100 Relative intru-
Figure 2. Future growth changes for 6-year-old chil- sion over extended periods of growth could
dren with low (<1 SD), medium ( ± 1 SD) and high serve as an important and potent treatment mo-
(>1 SD) mandibular plane angles (S-N/Go-Me). dality.
<<    
     Article
      >>
Management of the Vertical Dimension
Early Treatment Results
It is becoming increasingly clear that treatments
for Hyp-OB patients should be directed toward
the mandible. Naumann et al101 have shown that
mandibular skeletal changes were twice as im-
portant as mandibular dental changes and 2.5
times as important as maxillary changes in affect-
ing overbite changes. Vertical mandibular
growth was shown to be more important than
mandibular rotation in determining overbite
changes. Their models suggest using appliances
that limit anterior vertical mandibular growth
and augment its forward rotation, rather than
appliances that restrict maxillary growth.
Complex problems require complex treat-
ment approaches. Sankey and coworkers102 re-
cently reported outcomes for patients treated
with lip seal exercises, a lower Crozat/lip
bumper, a bonded palatal expander constructed
to serve as a bite block, and a high-pull chin cup.
A sample of children 8.2 ± 1 . 2 years with aver-
age MPAs of 40.1° ± 1.2° performed lip seal
exercises for 60 consecutive minutes each day.
The Crozat/lip bumper appliance was cemented
in place with 2 to 3 mm activation (after 8 weeks,
it was reactivated 1 mm every 8 weeks). At the
same time, the upper arch was expanded slowly
(1/4 turn per week, 1 mm per month) for ap-
proximately 6 months. The expander infringed
on the freeway space approximately 2 to 3 mm
and was ramped to produce progressively
thicker occlusal coverage on the palatal half of
the appliance. A high-pull chin cup (force di-
rected approximately 45° to occlusal plane) de-
livering 16 to 20 oz was worn at least 14 hours
per day. The patients were compared with con-
trols matched for age, gender, and MPA.
After 1.3 ± 0.3 years of treatment, condylar
growth had been increased and changed toward
a more anterosuperior direction. Maxillary ex-
pansion did not increase vertical dimensions.
The mandible showed almost three times more
forward rotation than expected, posterior facial
height increased significantly more, the molars
showed relative intrusion, the articular angle in-
creased, the gonial angle decreased, and the
chin moved forward almost twice as much as in
the controls. Overjet decreased and overbite in-
creased, especially for the patients with severe
open-bite malocclusions. The aggregate of
changes implies corrections in all three planes
Home | TOC |          
Index
137
of space that combine to produce a remarkable
treatment result. Interestingly, a subsample of
patients that performed clenching exercises suf-
ficient to produce marks on the expander's
acrylic did not wear high-pull chin cups but
showed similar treatment outcomes.
The potential importance of clenching exer-
cise in the treatment of Hyp-OP patients should
not be underestimated. Tran and coworkers103
recently completed a prospective clinical trial
evaluating the effects of light clenching exer-
cises as adjunctive treatment for HPHG and
rapid palatal expansion in young children. Al-
though HPHG without exercise produced den-
toalveolar effects in the maxilla only, HPHG in
combination with exercises also increased true
mandibular autorotation and produced signifi-
cant reductions in the ANB and gonial angles.
Because there was no evidence of increased mus-
cle strength, it was suggested that exercise might
have altered the postural position of the mandi-
ble, which in turn influenced its rotational and
remodeling patterns.
Conclusions
Although early treatment of hyp e r diverge n t
open-bite cases is theoretically appealing and
practically possible, it remains poorly under-
stood and must be approached with caution.
More clinical and experimental research is re-
quired to establish the psychosocial benefits and
to precisely define the long-term consequences
of early treatment. The stability of early treat-
ment will undoubtedly be related to the orth-
odontist's ability to correct the original cause of
the problem. To that end, open-mouth posture
associated with habits, weak muscles, or respira-
tory obstructions must be eliminated to avoid
relapse. This implies additional studies necessary
to develop sensitive and specific diagnostic indi-
cators. Growth is clearly a critical period that
holds great potential for orthopedic and orth-
odontic corrections as well as for relapse toward
the original condition.
References
1. Hellman M: Open bite. Int J Orthod 17: 421-436, 1931
2. Korkhaus G: The frequency of orthodontic anomalies
at various ages. Int J Orthod 14:120-129, 1928
3. Worms FW, Meskin LH, Isaacson RJ: Open-bite. Am J
Orthod 59:589-595, 1971
<<    
     Article
      >>
138 Buschang, Sankey, and English
4. Parker JH: The interception of the open bite in the
early growth period. Angle Orthod 41:24-44, 1971
5. Johnson ED, Larson BE: Thumb-sucking: Classification
and treatment. J Dent Child 60:392-398, 1993
6. SubtelnyJD, Sakuda M: Open-bite: Diagnosis and treat-
ment. Am J Orthod 50:337-358, 1964
7. Sassouni V: A classification of skeletal facial types. Am J
Orthod 55:109-123, 1969
8. Schudy FF: Vertical growth versus anteroposterior
growth as related to function and treatment. Angle
Orthod 34:75-92, 1964
9. Graber TM: Thumb and finger-sucking. Am J Orthod
45:257-264, 1959
10. Larrson E: Dummy- and finger-sucking habits with spe-
cial attention to their significance for facial growth and
occlusion. 4. Effects on facial growth and occlusion.
Sven Tandlak Tidskr 65:605-634, 1972
11. Hawkins AC: A constructive approach to thumbsucking
habit. J Glin Orthod 12:846-848, 1978
12. Bowden DB: A longitudinal study of the effects of digit-
and dummy-sucking. Am J Orthod 52:887-901, 1966
13. Meisen B, Stensgaard K, PedersenJ: Sucking habits and
their influence on swallowing pattern and prevalence
of malocclusion. EurJ Orthod 1:271-280, 1979
14. Nanda RS, Khan I, Anand R: Effect of oral habits on the
occlusion in preschool children. J Dent Child 39:449-
452, 1972
15. Willmot DR: Thumb sucking habit and associated den-
tal differences in monozygotic twins. Br J Orthod 11:
195-199, 1984
16. Weijs WA, Hillen B: Relationships between muscle
cross-section and skull shape. J Dent Res 63:1154-1165,
1984
17. Gionhaku N, Lowe A: Relationship between jaw muscle
volume and craniofacial form. J Dent Res 68:805-809,
1989
18. Hannam AG, Wood WW: Relationships between the
size and spatial morphology of human masseter and
medial pterygoid muscles, the craniofacial skeleton,
and jaw biomechanics. Amer J Phys Anthrop 80:429-
445, 1989
19. Van Spronsen PH, Weijs WA, ValkJ, et al: Relationships
between jaw muscle cross-sections and craniofacial
morphology in normal adults, studied with magnetic
resonance imaging. Eur Orthod Soc 13:351-361, 1991
20. Van Spronsen PH, Weijs WA, ValkJ, et al: A Compari-
sion of jaw muscle cross-sections of long-face and nor-
mal adults. J Dent Res 71:1279-1285, 1992
21. Benington PC, Gardener JE, Hunt NP: Masseter muscle
volume measured using ultrasonography and its rela-
tionship with facial morphology. EurJ Orthod 21:659-
670, 1999
22. Kiliaridis S, Kalebo P: Masseter muscle thickness mea-
sured by ultrasonography and its relation to facial mor-
phology. J Dent Res 70:1262-1265,1991
23. Bakke M, Tuxen A, Vilmann P, et al: Ultrasound image
of human masseter muscle related to bite force, elec-
tromyography, facial morphology, and occlusal factors.
Scand J Dent Res 100:164-171, 1992
24. Ingervall B, Thilander B: Relation between facial mor-
phology and activity of the masticatory muscles. J Oral
Rehab 1:131-147, 1974
Home | TOC |          
Index
25. Bakke M, Michler L: Temporaiis and masseter muscle
activity in patients with anterior open bite and cranio-
mandibular disorders. Scand J Dent Res 99:219-228,
1991
26. Proffit W, Fields H, Nixon W: Occlusal forces in nor-
mal- and long-face adults. J Dent Res 62:566-571, 1983
27. Ringqvist M: Isometric bite force and its relation to
dimensions of the facial skeleton. Acta Odont Scand
31:35-42, 1973
28. Kiliaridis S, Johansson A, Haraldson T, et al: Craniofa-
cial morphology, occlusal traits, and bite force in per-
sons with advanced occlusal tooth wear. Am J Orthod
Dentofacial Orthop 107:286-292, 1995
29. Proffit W, Fields H: Occlusal forces in normal and
long-face children. J Dent Res 62:571-574, 1983
30. Garcia-Morales P, Buschang PH, Throckmorton GS, et
al: Maximum bite force, muscle efficiency and mechan-
ical advantage in children with vertical growth pattern.
EurJ Orthod 2002
31. Throckmorton G, Finn R, Bell WH: Biomechanics of
differences. I. Lower face height. Am J Orthod 77:410-
420, 1980
32. Harvold EP, Vargervik K, Chierici G: Primate experi-
ments on oral sensation and dental malocclusions.
Am J Orthod 63:494-508, 1973
32. Odman C, Kiliaridis S: Masticatory muscle activity in
myotonic dystrophy patients. J Oral Rehab 23:5-10,
1996
33. Kiliaridis S, Mejerso C, Thilander B: Muscle function
and craniofacial morphology: a clinical study in pa-
tients with myotonic dystrophy. EurJ Orthod 11:131-
138, 1989
34. Houston KD, Buschang PH, lannaccone ST, et al:
Craniofacial morphological characteristics in patients
with spinal muscular atrophy. Pediatric Res 36:265-269,
1994
35. Granger M, Buschang P, Throckmorton G, et al: Mas-
ticatory muscle function in patients with spinal muscu-
lar atrophy. Am J Orthod 115:697-702, 1999
37. Yamada, T, Tanne K, Miyamoto K, et al: Influences of
nasal respiratory obstruction on craniofacial growth in
young Macaca fuscata monkeys. Am J Orthod Dentofa-
cial Orthop 111:38-43, 1997
38. Scarano E, Ottaviani SE, DeGirolamo S, et al: Relation-
ship between chronic nasal obstruction and craniofa-
cial growth: an experimental model. Int J Pediatr Oto-
rhinolaryngol 2:125-131, 1998
39. Subtelney JD: The significance of adenoid tissue in
orthodontia. Ang Orthod 25:59-69, 1954
40. McNamara JA: Influence of respiratory pattern on
craniofacial growth. Angle Orthod 51:269-300, 1981
41. Bresolin D, Shapiro PA, Shapiro GG, et al: Mouth
breathing in allergic children: its relationship to dento-
facial development. Am J Orthod 83:334-340, 1983
42. Linder-Aronson S: Adenoids: Their effect on mode of
breathing and nasal airflow and their relationship to
characteristics of the facial skeleton and the denition.
Acta Otolaryngologica Suppl 265:1-132, 1970
43. Linder-Aronson S: Respiratory function in relation to
facial morphology and the dentition. Brit J Orthod
6:59-71, 1979
44. Woodside DG, Linder-Arsonson S, Lundström A, et al:
<<    
     Article
      >>
Management of the Vertical Dimension
Mandibular and maxillary growth after changed mode
of breathing. Am J Orthod Dentofacial Orthop 100:1-
18, 1991
45. Behlfelt K, Linder-Aronson S, McWilliam J, et al:
Cranio-facial morphology in children with and without
enlarged tonsils. Eur J Orthod 12:233-243, 1990
46. Hellsing E, Forsberg CM, Linder-Aronson S, et al:
Changes in postural EMG activity in the neck and
masticatory muscles following obstruction of the nasal
airway. Eur J Orthod 8:247-253, 1986
47. Tourne LP, Schweiger J: Immediate postural responses
to total nasal obstruction. Am J Orthod Dentofacial
Orthop 110:606-11, 1996
48. Armstrong MM: Controlling the magnitude, direction,
and duration of extraoral force. Am J Orthod 59:217-
243, 1971
49. Watson WG: A computerized appraisal of the high-pull
facebow. Am J Orthod 62:561-573, 1972
50. Badell M: An evaluation of extraoral combined high-
pull traction and cervical traction to the maxilla. Am J
Orthod 69:431-445, 1976
51. Baumrind S, Korn, EL, Molthen, BS, et al: Changes in
facial dimensions associated with the use of forces to
retract the maxilla. Am J Orthod 80:17-30, 1981
52. Baumrind S, Korn EL, Isaacson RJ, et al: Quantitative
analysis of the orthodontic and orthopedic effects of
maxillary traction. Am J Orthod 84:384-398, 1983
53. Firouz M, Zernik J, Nanda R: Dental and orthopedic
effects of high-pull headgear in treatment of Class II
Division I malocclusion. Am J Orthod 102:197-205,
1992
54. Elder JR, Tuenge RH: Cephalometric and histologic
changes produced by high-pull traction to the maxilla
in Macaca mulatta. Am J Orthod 66:599-617, 1974
55. Meldrum RJ: Alterations in the upper facial growth of
Macaca mulatta resulting from high-pull headgear.
Am J Orthod 67:393-411, 1975
56. Worms FW, Isaacson RJ, Speidel TM: A concept and
classification of centers of rotation and extraoral force
systems. Angle Orthod 43:384-401, 1973
57. Baumrind S, Molthen R, West EE, et al: Mandibular
plane changes during maxillary retraction. Am J
Orthod 74:32-40, 1978
58. Thurow RC: Craniomaxillary orthopedic correction
with en masse dental control. Am J Orthod 68:601-624,
1975
59. Joffe L, Jacobsen A: The maxillary orthopedic splint.
Am J Orthod 75:54-69, 1979
60. Fotis V, Meisen B, Williams S, et al: Vertical control as
an important ingredient in the treatment of severe
sagittal discrepancies. Am J Orthod 86:224-232, 1984
61. Orton HS, Slättery DA, Orton S: The treatment of
severe 'gummy' class II division I malocclusion using a
maxillary intrusion splint. Eur J Orthod 14;216-223,
1992
62. Caldwell SF, Hymas TA, Timm TA: Maxillary traction
splint: A cephalometric evaluation. Am J Orthod 85:
376-384, 1984
63. Cangialosi TJ: Skeletal features of anterior open bite.
Am J Orthod 85;28-36, 1984
63. Yamaguchi K, Nanda RS: The effects of extraction and
Home | TOC |          
Index
139
nonextraction treatment on the mandibular position.
Am J Orthod 100:443-452, 1991
64. Staggers JA: Vertical changes following first premolar
extractions. Am J Orthod Dentofacial Orthop 105:19-
24, 1994
65. Dougherty HL: The effect of mechanical forces upon
the mandibular buccal segments during orthodontic
treatment. Am J Orthod 54:29-49, 1968
66. Pearson LE: Vertical control through use of mandibu-
lar posterior intrusive forces. Angle Orthod 43:194-200,
1973
67. Sergl HG, Farmand M: Experiments with unilateral bite
planes in rabbits. Angle Orthod 45:108-114, 1975
68. Altuna G: The effect of excess occlusal force on the
eruption of the buccal segments and maxillary and
mandibular growth direction in the Macaca monkey.
Master's thesis, University of Toronto, Toronto, 1979
69. Altuna G, Woodside DG: Response of the midface to
treatment with increased vertical occlusal forces. Treat-
ment and post-treatment effects in monkeys. Angle
Orthod 55:251-263, 1985
70. Woods MG, Nanda RS: Intrusion of posterior teeth with
magnets—an experiment in growing baboons. Angle
Orthod 58:136-150, 1988
71. Woodside DG, Linder-Aronson S: Progressive increase
in lower anterior facial height and the use of posterior
occlusal bite-block in its management, in Graber LW
(ed): Orthodontics: State of the Art, Essence of the
Science, St Louis, MO, Mosby, 1986, pp 209-218
72. Rowe, TK, Carlson, DS: The effect of bite-opening ap-
pliances on the mandibular rotational growth and re-
modeling in the rhesus monkey (Macaca mulatta).
Am J Orthod Dentofacial Orthop 98:544-549, 1990
73. Meisen B, McNamara JA, Hoenie DC: The effect of
bite-blocks with and without repelling magnets studied
histomorphologically in the rhesus monkey (Macaca
mulatta). Am J Orthod Dentofacial Orthop 108:500-
509, 1995
74. Iscan HN, Akkaya S, Koralp E: The effects of the spring-
loaded posterior bite block on the maxillo-facial mor-
phology. Eur J Orthod 14:54-60, 1992
75. Kiliardis S, Egermark I, Thilander B: Anterior open bite
treatment with magnets. Eur J Orthod 13:447-457, 1990
76. Kalra V, Burstone CJ: Effects of a fixed magnetic appli-
ance on the dentofacial complex. Am J Orthod Dento-
facial Orthop 95:467-478, 1989
77. Dellinger EL: A clinical assessment of the active vertical
corrector—A nonsurgical alternative for skeletal open-
bite. Am J Orthod Dentofacial Orthop 89:428-436, 1986
78. Dellinger EL: Active vertical corrector treatment-Long-
term follow-up of anterior open bite treated by the
intrusion of posterior teeth. Am J Orthod Dentofacial
Orthop 110:145-154, 1996
79. Lundström A, Woodside DG: Longitudinal changes in
facial type in cases with vertical and horizontal mandib-
ular growth directions. Eur J Orthod 5:259-268, 1983
79. McNamara JA: An experimental study of increased ver-
tical dimension in the growing face. Am J Orthod 71:
382-395, 1977
80. Kuster R, Ingervall B: The effect of treatment of skeletal
open bite with two types of bite-blocks. Eur J Orthod
14:489-499, 1992
<<    
     Article
      >>
140 Buschang, Sankey, and English
81. Barbre RE, Sinclair PM: A cephalometric evaluation of
anterior open bite correction with the magnetic active
vertical corrector. Angle Orthod 61:93-109, 1991
82. Pearson LE: Vertical control in treatment of patients
having backward-rotational growth tendencies. Angle
Orthod 48:132-140, 1978
83. Pearson LE: Vertical control in fully-banded orthodon-
tic treatment. Angle Orthod 56:205-224, 1986
84. Pearson LE: Treatment of a severe openbite excessive
vertical pattern with an eclectic non-surgical approach.
Angle Orthod 61:71-76, 1991
85. Haas AJ: A biological approach to diagnosis, mechanics
and treatment of vertical dysplasia. Angle Orthod 50:
279-300, 1980
86. Majourau A, Nanda R: Biomechanical basis of vertical
dimension control during rapid palatal expansion ther-
apy. Am J Orthod Dentofacial Orthop 106:322-328,
1996
87. Alley TR: Social and applied aspects of face perception:
An introduction, in Alley TR (ed): Social and Applied
Aspects of Perceiving Faces. Hillsdale, NJ, Lawrence
Erlbaum Assoc, 1988, pp 1-8
88. Berschied E, Walster E, Bohrnstedt G: The happy
American body: A survey report. Pyschol Today 7:119-
131, 1973
89. Macgregor EC: Social and psychological implications of
dentofacial disfigurement. Angle Orthod 40:231-233,
1970
90. Dion KK, Berscheid E, Walster E: What is beautiful is
good. J Pers Soc Psychol 24:285-293, 1972
91. Shaw WC: The influence of children's dentofacial ap-
pearance on their social attractiveness as judged by
peers and lay adults. Am J Orthod 79:399-415, 1981
92. Kiyak HA, McNeill RW, West RA, et al: Personality
characteristics as predictors and sequelae of surgical
and conventional orthodontics Am J Orthod Dentof
Orthop 89:383-392, 1986
93. Bishara SE, Jakobsen JR: Longitudinal changes in three
normal facial types. Am J Orthod 88:466-502, 1985
94. Nanda SK: Patterns of vertical growth in the face. Am J
Orthod Dentofacial Orthop 93:103-116, 1988
95. Nanda SK: Growth patterns in subjects with long and
short faces. Am J Orthod Dentofacial Orthop 98:247-
258, 1990
96. Lundström A, Woodside DG. Longitudinal changes in
facial type cases with vertical and horizontal mandibu-
lar growth directions. Eur J Orthod 5:259-268, 1983
97. Garcia-Morales P, Buschang PH: Longitudinal stability
of divergent growth patterns. J Dent Res 81:388, 2002
98. Karlsen AT: Craniofacial growth differences between
low and high MP-SN angle males: a longitudinal study.
Angle Orthod 65:341-350, 1995
99. Spady M, Buschang PH, Demirjian A: Mandibular ro-
Home | TOC |          
Index
tation and angular remodeling during childhood and
adolescence. Am J Human Biol 4:683-689, 1992
100. Watanabe E, Demirjian A, Buschang PH: Longitudinal
post-eruptive mandibular tooth movements of males
and females. Eur J Orthod 21:459-468, 1999
101. Naumann SA, Behrents RG, Buschang PH: Vertical
components of overbite change: A mathematical
model. Am J Orthod Dentofacial Orthop 117:486-495,
2000
102. Sankey WL, Buschang PH, English J, et al: Early treat-
ment of vertical skeletal dysplasia: The hyperdivergent
phenotype. Am J Orthod Dentofacial Orthop 118:317-
327, 2000
103. Tran MT, English JD, Throckmorton GS, et al: The
adjunctive treatment effects of light masticatory muscle
training on hyperdivergent open-bite patients. A pilot
study. 2002 (in press)
104. Bell WH, Creekmore TD, Alexander RG: Surgical cor-
rection of the long face syndrome. Am J Orthod 71:40-
67, 1977
105. Isaacson JR, Isaacson RJ, Speidel TM, et al: Extreme
variation in vertical facial growth and associated varia-
tion in skeletal and dental relations. Angle Orthod
41:219-229, 1971
106. Nahoum HI, Horowitz SL, Benedicto EA: Varieties of
anterior open-bite. Am J Orthod 61:486-492, 1972
107. Schendel SA, Eisenfeld J, Bell WH, et al: The long face
syndrome: Vertical maxillary excess. Am J Orthod 70:
398-408, 1976
108. Lundström A, Woodside DG. A comparison of various
facial and occlusal characteristics in mature individuals
with vertical and horizontal growth direction expressed
at the chin. Eur J Orthod 3:227-235, 1981
109. Fields H, Proffit W, Nixon W, et al: Facial pattern
differences in long-faced children and adults. Am J
Orthod 85:217-223, 1984
110. Haralabakis NB, Yiagtzis SC, Toutountzakis NM: Ceph-
alometric characteristics of open bite in adults: A three
dimensional cephalometric evaluation. Int J Adult
Orthod Orthognath Surg 9:223-231, 1994
111. Trouten JC, Enlow DH, Rabine M, et al: Morphologic
factors in open bite and deep bite. Angle Orthod 53:
192-211, 1983
112. Lopez-Gavito G, Wallen TR, Little R, et al: Anterior
open-bite malocclusion: A longitudinal 10-year postre-
tention evaluation of orthodontically treated patients.
Am J Orthod 87:175-186, 1985
113. Janson GR, Metaxas A, Woodside DG: Variation in
maxillary and mandibular molar and incisor vertical
dimension in 12-year-old subjects with excess, normal,
and short lower anterior facial height. Am J Orthod
106:409-418, 1994
114. Congialosi TJ: Skeletal morphologic features of ante-
rior open-bite. Am J Orthod 85:28-36, 1984
<<    
     Article
      >> Home | TOC |          
Index

Vertical Skeletal and Dental Changes in Early

Treatment of Class II Malocclusion
Calogero Dolce, Lisa K. Babb, Susan P. McGorray, Marie G. Taylor,
Gregory J. King, and Timothy T. Wheeler

Increases in the vertical dimension can be detrimental in certain facial types

or unstable in others. By using cephalograms, this study examined the
skeletal and dental changes as well as their permanence in a group of
children participating in a randomized clinical trial on the timing of Class II
orthodontic treatment. Children (mean age, 9.6 years) were assigned to
either a headgear/biteplane (n = 94), bionator (n = 86), or observation (n =
82) group. Cephalometric data were obtained from tracings at baseline, after
Class I molar was obtained or 2 years after treatment had elapsed, and after
1 year of retention or observation. In general, the skeletal changes, such as
total anterior and posterior face height and mandibular plane angle, ob-
served during the active phase of early treatment persisted until the end of
phase I. Phase I treatment also produced dental changes, such as molar
eruption and tip. With the exception of the eruption of the mandibular
molars in the bionator group, these changes could not be maintained until
the end of phase I. Whether the subjects were in retention or not had little
impact on any of the studied parameters. Mandibular plane angle, pretreat-
ment, and retention had an affect on the mandibular molar tip over the
course of the study. (Semin Orthod 2002;8:141-148.) Copyright 2002, Elsevier
Science (USA). All rights reserved.

ontrolling the vertical dimension is critical
C to functional appliance treatment because
clockwise rotation can have deleterious effects in
patients who have a high mandibular angle; like-
wise counter clock rotation in those patients,
who exhibit a low mandibular plane angle may
be unstable. Numerous studies have reported on
the skeletal and dental vertical changes pro-
duced by functional appliance treatment.
Jakobsson1 was among the first to point out
that activator treatment altered the vertical di-
mension, which was represented by an apparent
clockwise rotation of the palatal plane, an in-
From the Department of Orthodontics, College of Dentistry, Uni-
versity of Florida, Gainesville, FL.
Supported by NIH/NICDR DE08715
Address correspondence to Calogero Dolce, DDS, PhD, Depart-
ment of Orthodontics, Box 100444, JHMHC, Gainesville, FL
32610-0444.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0803-0005$35.00/0
dot: 10.1053/sodo.2002.125433
creased lower anterior face height, a downward
positioning of A-point, and an increase in the
total mandibular length for both the headgear
and activator groups relative to the observation
group. There was also a significant increase in
the mandibular plane angle associated with ac-
tivator treatment. Interestingly, this study did
"not support the hypothesis that activator treat-
ment can bring the mandible forward and affect
condylar growth."1
Overall, the available literature is often times
the result of retrospective efforts that are com-
monly plagued by selection biases and insuffi-
ciently matched controls. It is also common to
find relatively small experimental groups and a
narrow, if not shortsighted, selection of experi-
mental parameters. In light of these and other
shortcomings, some findings seem to be fairly
constant among the reviewed investigations. Of
the studies that explored vertical ramus height
and/or vertical position of the lower molar (rel-
ative to the mandible), there was a predominant

Seminars in Orthodontics, Vol 8, No 3 (September), 2002: pp 141-148 141

<<    
     Article
      >>
142 Dolce et al
agreement that both of these parameters in-
crease with early treatment involving either
headgear or activator.2"8 With regard to anterior
and posterior facial height, about half of the
time the total anterior face height was studied
with no concurrent study of posterior facial
height.1'5'6 When the proportion posterior lower
facial height to anterior lower facial height
was investigated, the ratio was greater than one
if a posttreatment observation period was in-
cluded.4'9'10 During treatment itself, a ratio less
than one was seen. This means that posttreat-
ment changes in growth are occurring relative to
the observation group in the vertical dimension.
This study evaluated the vertical treatment
effects of the headgear/biteplane (HG/BP) and
bionator during the early treatment of the Class
II malocclusion. More specifically, the aim of
this work evaluated changes in (1) the posterior
and anterior facial height, (2) the alveolar
height of the maxillary and mandibular molars,
and (3) the orientation of the palatal and man-
dibular plane as well as the molars that may
occur within the periods studied for all the
groups.
Materials and Methods
The details of this randomized clinical trial have
been previously described by Keeling et al.11
Briefly, this was a prospective randomized clini-
cal trial designed to examine the timing of Class
II orthodontic treatment. The inclusion and ex-
clusion criteria have already been described.11 A
stratified block randomization was used to assign
a subject into one of the following groups:
bionator, HG/BP, or observation. Subjects were
stratified according to the following criteria: (1)
severity of molar Class II malocclusion (mild,
bilateral Vz cusp; moderate, one side 34 cusp;
and severe, one side full cusp), (2) mandibular
plane angle (low, <30°; normal, 30° - 40°; and
high, >40°), (3) race, (4) gender, and (5) re-
tention. Each subject had complete orthodontic
records taken at every data collection point as
they progressed through the study. The end of
phase I treatment occurred when two project
orthodontists independently agreed that a bilat-
eral Class I molar relationship was achieved or 2
years had elapsed from the start of treatment.
Treated subjects were then randomized to reten-
tion or no retention. The retention scheme was
Home | TOC |          
Index
to continue to wear the bionator or HG/BP
every other night for 6 months, followed by ob-
servation for 6 months. The nonretention
scheme lasted 1 year and began immediately
after treatment. All subjects had the following
data collection points (DC): (1) DC1, initial
records; (2) DC3, end of early Class II treatment
or observation; and (3) DC5, end of retention
period and observation period.
Cephalometric Analysis, Calibration,
and Digitization
All lateral cephalograms were taken on the same
machine. The superimpositions of lateral ceph-
alometric tracings from DC 1 to DC5 were used
to detect vertical changes in the sample popula-
tion. All skeletal measurements were performed
relative to an xy-axis superimposed on the struc-
ture of Sella (S) with the x component of the
grid drawn parallel to the Frankfort horizontal
line. Additional templates were then con-
structed for the maxilla, mandible, upper and
lower molars, condyle, and gonial angle. The
details on the template construction have previ-
ously been described.12 All points were regis-
tered with crosses on both the templates and the
final tracing for each film. Figure 1 depicts the
angular measurements, and Figure 2 depicts the
linear measurements that were taken on the
tracings.
Before tracing, four crosshatches were estab-
lished on each film, acetate paper was taped to
the left-hand margin, and the crosshatches were
transferred to the acetate. The dental and skel-
etal templates were oriented independently on
each radiograph by best-fit visual approxima-
tion, and the registration crosses of each struc-
ture were transferred to the final tracing for
each film. In addition, points representing the
anterior and posterior extent of the functional
occlusal plane (FOP) were marked on the ace-
tate. The FOP was identified as the line of max-
imum intercuspation of the first permanent mo-
lars, permanent first bicuspids, and/or primary
first molars.
The following measurements were executed:
(1) mandibular plane (MP) angle, the angle
formed by the mandibular plane and the x-axis;
(2) palatal plane (PP) angle, the angle formed
by the PP and the x-axis; (3) occlusal plane (OP)
angle, the angle formed by the FOP and the
<<    
     Article
      >>
Class II Malocclusion
Figure 1. Diagram of the angular measurements ex-
ecuted in the study relative to the x-axis. UGtip, max-
illary molar tip relative to the palatal plane; L6tip,
mandibular molar tip relative to the mandibular
plane.
x-axis; (4) tip of the upper molar, the angle
formed by the tangent to the anterior border of
the upper molar and the PP; (5) tip of the lower
molar, the angle formed by the tangent to the
anterior border of the lower molar and the man-
dibular plane; (6) total posterior face height
(TPFH), the distance between the x-axis and
gonion as projected onto the y-axis; (7) upper
posterior face height (UPFH), the distance be-
tween the x-axis and condylion as projected onto
the y-axis; (8) lower posterior face height
(LPFH), the distance between condylion and
gonion as projected onto the y-axis; (9) total
anterior face height (TAFH), the distance be-
tween the x-axis and men ton as projected onto
the y-axis; (10) upper anterior face height
(UAFH), the distance between the x-axis and the
anterior nasal spine (ANS) as projected onto the
y-axis; (11) lower anterior face height (LAFH),
the distance between ANS and menton as pro-
jected onto the y-axis; (12) upper molar erup-
tion (U6PP), the perpendicular distance be-
tween the mesial contact point of the upper
molar and the PP; and (13) lower molar erup-
tion (L6MP), the perpendicular distance be-
tween the mesial contact point of the lower mo-
lar and the mandibular plane.
Home | TOC |          
Index
143
Reliability and Statistical Analysis
Reliability was evaluated by estimation of the
intraclass correlation coefficient of reliability for
each vertical cephalometric variable. This coef-
ficient can be interpreted as the proportion of
the variance that is caused by subject-to-subject
variability. The estimate can range from 0.00 to
1.00, with higher values indicating more reliable
measurements.
Intraexaminer reliability was assessed for 20
cephalometric radiographs, which were traced
twice. The results of the intraexaminer reliability
for the 14 measurements in this study range
from 0.933 for upper molar tip to 0.998 for total
posterior facial height.
Descriptive statistics were used to examine
differences in the cephalometric variables be-
tween the treatment groups. These differences
were identified analysis of variance. Chi-square
tests were used to detect differences between
treatment groups between categoric variables
such as gender and race. Linear regression mod-
els were developed to examine the joint impact
of race, gender, mandibular plane angle, age,
initial molar class severity, pretreatment, reten-
tion, and treatment group on change of the
UPFH
\
LAFH
Figure 2. Diagram of the linear measurements in-
volved in the study. Presented here are the UAFH,
LAFH, UPFH, and the lower posterior facial height
LPFH. Also depicted are the perpendicular heights of
the maxillary molar as measured from the PP (U6PP)
and mandibular molars as measured (L6MP) molars
as measured from the MP.
<<    
     Article
      >>
144 Dolce et al
cephalometric variables between DC l to DC5. A
P value of less than .05 was considered statisti-
cally significant.
To standardize the data, the results presented
in Tables 1 through 4 are annualized, that is the
difference between time point 1 and time point
2 is divided by the time between measurements
to obtain a rate of change. This method was
chosen so as to correct for the time needed to
achieve a given correction. In this manner, a
given amount of change is weighed more heavily
if it occurred over a shorter period of time.
Results
Baseline characteristics for the 262 subjects are
presented in Table 5. With the exception of the
alveolar height of the mandibular molar, all
other measured variables were not significantly
different at the beginning of treatment. The
mean active treatment (DC1-DC3) time, in
years, for the bionator, HG/BP, and observation
group was 2.2 ± 0.7, 2.1 ± 0.8, and 2.2 ± 0.4,
respectively. Although the total phase I (DC1-
DC5) treatment time, in years, for the bionator,
HG/BP, and observation group was 3.4 ± 0.6,
3.2 ± 0.7, and 3.2 ± 0.4, respectively
The annualized skeletal and dental effects of
treatment (DC1-DC3) are shown in Tables 1 and
2. Overall, the HG/BP group experienced sev-
eral statistically significant changes in the mea-
sured parameters that were not observed by in
the bionator group when compared with the
observation group. The dental effects included
the distal tipping of the maxillary molars (U6
tip) in the HG/BP group. Interestingly, the
mandibular molar (L6 tip) tipped dis tally in all
the three groups; however, the greatest amount
of change was seen in the HG/BP group (Table
1). Eruption of the maxillary (U6PP) molars
occurred in all three groups, the changes in the
HG/BP group were statistically significant from
the other groups. The final observed dental
Table 1. Annualized Angular (°) Changes From the Treatment Phase (DC1-DC3)
Bionator HG/BP
Mean + SD Mean ± SD
Palatal plane -0.02 -+- 0.6 0.03 ±
MP -0.23 -+- 0.8 0.50 ±
Occlusal plane -0.31 •+- 1.3 0.01 ±
U6 tip 0.35 -f- 1.9 -1.22 ±
L6 tip -0.42 + 2.3 -1.66 ±
Home | TOC |          
Index
change was the eruption of the mandibular
(L6MP) molars. Similar eruption of the mandib-
ular molars occurred in both the HG/BP and
bionator group when compared with the obser-
vation group (Table 2). The observed skeletal
changes were mainly in the HG/BP group and
included the increase in mandibular plane angle
and lengthening of the UAFH. Skeletal changes,
which were observed in both the HG/BP and
bionator group, were increases in LAFH and
TAFH. Again the HG/BG group showed the
greatest amount of change. TPFH increased
equally in the HG/BP and bionator groups
when compared with the observation group (Ta-
ble 2).
Tables 3 and 4 depicts the changes for the
total phase I time period (DC1-DC5). When the
treatment phase and the 6 months of retention/
nonretention and 6 months of observation are
combined, many of the dental effects of treat-
ment for the headgear group have relapsed. For
example, the change in the maxillary and man-
dibular molar tip have disappeared. However, all
of the skeletal effects seen during treatment per-
sisted to the end of phase I for the HG/BP
group with the exception of the anterior upper
face height. All treatment effects for the biona-
tor group persisted to the end of the phase I
time period.
Because other factors besides treatment
group may play a role in cephalometric changes,
linear regression models were used to examine
the joint impact of treatment and gender, race,
mandibular plane angle, age, molar severity,
pre treatment, and retention during the DC1 to
DC5 time period. Accounting for the additional
variables did affect the treatment group differ-
ences by analysis of variance in three cases.
Change in mandibular plane angle did not dif-
fer significantly between the HG/BP group and
the other two groups after including the addi-
tional variables. Tip back of the lower molar was
affected by the following variables: retention
Control
Mean ± SD P Value Difference
0.9 0.12 ± 0.7 .39
1.0 -0.39 ± 0.5 .0001 H/BC
1.6 -0.51 ± 1.2 .06
2.9 0.84 ± 1.9 .0001 H/BC
2.3 -0.24 ± 1.4 .0001 H/BC
<<    
     Article
      >> Home | TOC |          
Index

Class II Malocclusion 145

Table 2. Annualized Linear (mm) Changes From the Treatment Phase (DC1-DC3)
Bionator HG/BP Control
Mean ± SD Mean ± SD Mean ± SD P Value Difference
UPFH 0.66 ± 0.8 0.42 ± 0.9 0.39 ± 0.7 .06
LPFH 1.56 ± 0.9 1.73 ± 1.1 1.37 ± 0.8 .07
TPFH 2.22 ± 0.9 2.15 ± 0.9 1.76 ± 0.7 .0024 BH/C
UAFH 1.25 ± 0.7 1.56 ± 0.7 1.13 ± 0.7 .0001 H/BC
LAFH 1.35 ± 0.9 1.66 ± 0.9 0.92 ± 0.6 .0001 H/B/C
TAFH 2.61 ± 1.1 3.23 ± 1.2 2.05 ± 0.8 .0001 H/B/C
U6PP 0.65 ± 0.4 0.92 ± 0.5 0.69 ± 0.4 .0001 H/BC
L6MP 0.72 ± 0.6 0.76 ± 0.5 0.49 ± 0.3 .0011 HB/C

(P = .04), mandibular plane angle (P < .0001),
pre treatment (P = .0413), and treatment group
with HG/BP differing from controls (P =
.0177). UPFH change was affected by age at start
of DC1 (P = .0245), and when additional vari-
ables were included, treatment group achieved
only borderline significance (P = .0545). For all
other variables, the inclusion of additional vari-
ables (covariates) did not change the treatment
group differences we observed with univariate
analysis.
Regression models were developed to exam-
ine the impact of treatment group and other
variables (length of treatment, gender, race,
mandibular plane angle, age, molar severity,
pretreatment, and retention) on the amount of
change. The initial mandibular plane angle had
an impact on the outcome for the change in the
OP with those who initially had an average or
high angle experiencing more change in the OP
angle. The amount of time that the subject was
in treatment had a positive effect on the erup-
tion of the molars. Those who had a longer time
span between DC1 and DC5 experienced more
upper and lower molar eruption. Those subjects
whose baseline lower molar height was larger
experienced more eruption, and those who had
retention experienced less mandibular molar
eruption. Both the baseline value and initial
mandibular plane angle had a positive effect on
the amount of molar tip seen. For both the
upper and lower molar, the larger the baseline
molar tip (the less it was tipped back), the more
it tipped back during phase I. The initial man-
dibular plane angle had a similar effect with the
subjects with a moderate (between 30° and 40°)
mandibular plane angle and those with a high
angle (greater than 40°) experiencing more tip-
ping back of both the upper and lower molar
during the DC1 to DC5 time period. The larger
the baseline value for the UPFH, the smaller the
change was seen. In addition, male patients and
those who had a longer span of time between
DC1 to DC5 experienced more of an increase in
the TPFH. Gender had an effect on the upper
anterior face height, with males experiencing
more change. Also, the greater the amount of
time that elapsed between DC1 and DC5, the
greater the amount of change seen for the an-
terior upper, lower, and total face heights for all
subjects
Discussion
The findings of this study show that HG/BP or
bionator treatment produce vertical changes in
skeletal and dental structures, which are signifi-
cantly different from the observation group. Al-
though the resultant skeletal changes are stable
1 year after treatment irrespective of retention,
the dental changes have a propensity to relapse.

Table 3. Annualized Angular (°) Changes From Phase I Treatment (DC1-DC5)

Bionator HG/BP Control
Mean ± SD Mean ± SD Mean ± SD P Value Difference

Palatal plane 0.02 ± 0.5 0.01 ± 0.5 0.10 ± 0.5 .25

MP -0.31 ± 0.7 -0.05 ± 0.6 -0.33 ± 0.6 .005 H/BC
Occlusal plane -0.33 ± 0.9 -0.62 ± 1.0 -0.53 ± 0.8 .15
U6 tip 0.93 ± 1.1 0.66 ± 1.4 0.82 ± 1.4 .28
L6 tip -0.37 ± 1.4 -0.62 ± 1.3 -0.18 ± 1.1 .14
<<    
     Article
      >> Home | TOC |          
Index

146 Dolce et al

Table 4. Annualized Linear (mm) Changes From Phase I Treatment (DC1-DC5)

Bionator HG/BP Control
Mean ± SD Mean ± SD Mean ± SD P Value Difference
UPFH 0.48 ± 0.6 0.46 ± 0.5 0.29 ± 0.5 .0349 BH/C
LPFH 1.67 ± 0.7 1.60 ± 0.7 1.48 ± 0.6 .22
TPFH 2.15 ± 0.8 2.06 ± 0.7 1.78 ± 0.6 .0024 HB/C
UAFH 1.30 ± 0.5 1.41 ± 0.5 1.23 ± 0.6 .09
LAFH 1.19 ± 0.7 1.18 ±0.5 0.89 ± 0.5 .0008 BH/C
TAFH 2.49 ± 0.9 2.59 ± 0.7 2.12 ± 0.7 .0003 BH/C
U6PP 0.76 ± 0.4 0.85 ± 0.3 0.72 ± 0.3 .0520
L6MP 0.66 ± 0.4 0.57 ± 0.3 0.50 ± 0.3 .0177 B/C

Posterior Face Height
In this study, TPFH was depicted as the vertical
distance from the x-axis to condylion (UPDF) plus
the distance from condylion to gonion (LPFH).
Although statistical significance was not met for
UPFH and LPFH, the TPFH significantly in-
creased during active treatment with either
HG/BP or bionator. This skeletal change was last-
ing because it was present at the end of the reten-
tion phase. We speculate that the condyle de-
scended relative to the x-axis to a greater extent
than the ramus increase in length for the following
reasons. First, the bionator group nearly reached
significance for UPFH for both treatment (P —
.06) and overall differences that were detected
through the univariate analysis (P = .54) when
accounting for other variables. Second, the LPFH
changes did not compare with the UPFH. These
results contradict other studies that have attrib-
uted increases in TPFH to an increase in ramal
length.2-4'15'16 However, they support Jakobsson's1
results showing the vertical descent of the condyle
during early treatment with the headgear or biona-
tor. Similar, animal studies have shown that con-
tinuous, heavy distal forces to the upper molars
can alter resorption at craniofacial sutures in grow-
ing monkeys.13'14 Specifically, the zygomaticomax-
illary, zygomaticotemporal, and the junction be-
tween the maxillary tuberosity, pterygoid plates of
the sphenoid, and the pyramidal process experi-
ence more resorption with continual heavy head-
gear force compared with an untreated observa-
tion group. Given these observations and the fact
that Jakobsson1 found a significant increase in the
UPFH for his headgear group, it is possible that
the glenoid fossa itself is affected by headgear
treatment in the human. Because the fossa itself is
part of the temporal bone, it may be physically
moved or rotated by forces transmitted through
the zygomatico temporal suture, thereby displacing
the condyle form the glenoid fossa enough to
cause an adaptive response. Alternatively, the tran-
sitory bite opening that resulted from the molar
tipping could trigger an adaptive response in the
TMJ. Whatever the case, an increase in the TPFH
appears to be a lasting effect 1 year after headgear
treatment was discontinued.
Whether this increase in the TPFH and pos-

Table 5. Group Comparison of the Studied Variables at Baseline

Bionator HG/BP Control
Mean ± SD Mean ± SD Mean ± SD P Value
Angular measures (°)
Palatal plane 3.10 ± 2.4 2.83 ± 2.0 3.18 ± 2.3 .54
Mandibular plane 25.20 ± 5.9 25.98 ± 5.1 26.06 ± 5.0 .51
Occlusal plane 12.99 ± 4.3 13.17 ± 4.2 13.45 ± 4.4 .79
U6 tip 78.10 ± 4.8 77.97 ± 5.1 79.10 ± 4.9 .26
L6 tip 88.46 ± 5.6 89.33 ± 5.1 88.62 ± 4.3 .46
Linear measures (mm)
UPFH 17.75 ± 2.8 17.97 ± 2.6 17.64 ± 2.8 .72
LPFH 51.43 ± 4.0 50.54 ± 3.1 51.46 ± 3.0 .12
TPFH 69.18 ± 5.3 68.51 ± 4.0 69.10 ± 4.0 .54
UAFH 40.77 ± 4.9 41.39 ± 4.8 41.46 ± 4.4 .67
LAFH 59.93 ± 4.6 59.72 ± 4.1 60.09 ± 4.4 .85
TAFH 100.70 ± 6.6 101.11 ± 5.8 101.25 ± 5.8 .83
U6PP 16.99 ± 2.0 17.26 ± 1.9 17.55 ± 1.9 .17
L6MP 27.38 ± 2.1 26.28 ± 2.3 27.60 ± 2.1 .01
<<    
     Article
      >> Home | TOC |          
Index

Class II Malocclusion 147

sible descent of the condyle in the bionator greater increase of the anterior face height over
group reflects primarily a hard- or soft-tissue the observation group. Jakobsson1 in his study
adaptation within the TMJ is debatable. Ani- failed to show a significant increase in the man-
mal studies involving an anterior positioning of dibular plane angle.
the mandible with a fixed functional appli-
ance have found both types of effects. The dep- Molar Eruption and Facial Height
osition of new bone at the posterior and supe-
As the mandibular molar erupts, it typically mi-
rior aspects of the glenoid fossa and similar
grates mesially.29 This and the fact that that Class
adaptive changes in the condyle involving the
II correction is facilitated if there is counter-
intermediate zone of condylar cartilage at the
clockwise rotation of the OP are often viewed as
posterior aspect have been noted.17"21 Further-
desirable events during Class II orthodontic
more, when the mandible is protruded forward
treatment. In this study, lower molar eruption
and forced to function in this position full-time,
was encouraged by the selective removal of
the fibrous retrodiscal tissue has been found to
acrylic occlusal to the lower molar in the biona-
proliferate posteriorly.20
tor group and by the use of an anterior biteplane
in the headgear group. Our data show that al-
Vertical Control With High Pull Headgear though both treated groups experienced lower
molar eruption compared with the observation
Controlling the vertical dimension has always been
group during the treatment phase (DC1-3), only
of utmost importance during orthodontic treat-
the bionator group retained this change until
ment. Although much has been written on con-
the end of phase I (DC1-5), with no correspond-
trolling the vertical dimension during headgear
ing change in the angle of the OP. Conse-
treatment,2'22"27 most of these articles are either
quently, although it is possible that eruption of
anecdotal25"27 or fail to follow-up with what hap-
the lower molar is contributing to the molar
pens after the headgear is removed.2'22"24 It is spec-
correction in some way, it does not necessarily
ulated that headgear treatment may result in an
have an effect on the inclination of the OP. The
increase in the anterior face height and mandibu-
maxillary molars erupted during DC1-DC3 in
lar plane angle with a correspondingly negative
the HG/BP group with the changes from DC1-
esthetic impact on the higher mandibular plane
DC5 being nearly significant (P = .052).
angle Class II patients. In the present study, during
Both the bionator and HG/BP group experi-
the DC1-DC3 treatment phase the HG/BP group
enced significant mean increases in both the total
experienced an increase of 0.50° in mandibular
and lower anterior face height during the treat-
plane angle, whereas the bionator and observation
ment phase, which persisted through the end of
group saw a decrease of 0.23° and 0.39°, respec-
phase I. It is possible that mandibular molar erup-
tively. This decrease in mandibular plane angle
tion in the bionator group contributed to the in-
was maintained during the 1-year retention/obser-
creases in the lower anterior face height, whereas
vation period. Interestingly, during this retention/
in the HG/BP group, the previously mentioned
observation period the changes in the mandibular
changes have been attributed to a combined effect
plane angle reversed in the HG/BP group. Con-
from both upper and lower molars.
sequently, phase I treatment (DC1-DC5) essen-
tially resulted in no annualized change in mandib-
Headgear's Effect on Lower Molar Tip
ular plane angle.
All of the high mandibular plane angle Many of the dental treatment effects for the
(>40°) subjects were treated with a high-pull HG/BP group were transient and relapsed by
headgear with no corresponding increase in the the end of phase I of the study. Among these was
annualized change of the mandibular plane an- the distal tipping of the molars. As expected, the
gle at the end of phase I. Because the initial use of a headgear produced distal tipping of the
mandibular plane angle did not have an impact maxillary molars. What was surprising was the
on the increases in the total and lower anterior distal tipping of the mandibular molars. Because
face height for phase I, we can conclude that no appliances were used in the lower arch dur-
when high mandibular plane angle cases are ing treatment with the HG/BP, the tipping ef-
treated with a high-pull headgear, there is no fect on the lower molar may be because of an
<<    
     Article
      >>
148 Dolce et al
altered occlusal vector from the upper molar.
Alternatively, because the subjects who were in
the headgear group were instructed to wear the
biteplane full time, it may be that the biteplane
itself has an effect on the lower molar tip. There
is no evidence in the literature that indicates
that the biteplane wear can result in a tip back of
the lower molar. The headgear's effect of tip-
ping back the molars may be responsible for the
increase of the LAFH, increase in the TAFH,
and increase of the mandibular plane angle dur-
ing the active treatment phase (DC1-DC3).
Conclusions
Considerable variability is present in the vertical
response to Class II orthodontic treatment with
the bionator and HG/BP. The HG/BP groups
experienced more changes in vertical parame-
ters than the bionator or observation group.
Also, the observed skeletal changes remained
stable until the end on phase I, whereas the
dental changes seen at the end of active treat-
ment were lost by the end of phase I.
References
1. Jakobsson SO: Cephalometric evaluation of treatment
effect on Class II, division I malocclusions. Am J Orthod
53:446-457, 1967
2. Baumrind S, Korn EL, Molthen R, et al: Changes in
facial dimensions associated with the use of forces to
retract the maxilla. Am J Orthod 80:17-30, 1981
3. Bolmgren GA, Moshiri F: Bionator treatment in Class II,
division 1. Angle Orthod 56:255-262, 1986
4. Derringer K: A cephalometric study to compare the
effects of cervical traction and Andresen therapy in the
treatment of Class II division 1 malocclusion. Part
1—Skeletal changes. Br J Orthod 17:33-46, 1990
5. Mills JR: The effect of functional appliances on the
skeletal pattern. Br J Orthod 18:267-275, 1991
6. Chang HF, Wu KM, Chen KG, et al: Effects of activator
treatment on Class II, division 1 malocclusion. J Glin
Orthod 23:560-563, 1989
7. Hultgren BW, Isaacson RJ, Erdman AG, et al: Mechanics,
growth, and class II corrections. Am J Orthod 74:388-
395, 1978
8. Derringer K: A cephalometric study to compare the
effects of cervical traction and Andresen therapy in the
treatment of class II division 1 malocclusion. Part
2—Dentoalveolar changes. Br J Orthod 17:89-99, 1990
9. Vaden JL, Harris EF, Sinclair PM: Clinical ramifications
of posterior and anterior facial height changes between
treated and untreated Class II samples. Am J Orthod
Dentofacial Orthop 105:438-443, 1994
10. Ben Bassat Y, Baumrind S, Korn EL: Mandibular molar
Home | TOC |          
Index
displacement secondary to the use of forces to retract
the maxilla. Am J Orthod 89:1-12, 1986
11. Keeling SD, Wheeler TT, King GJ, et al: Anteroposterior
skeletal and dental changes after early Class II treatment
with bionators and headgear. Am J Orthod Dentofacial
Orthop 113:40-50, 1998
12. Keeling SD, Cabassa SR, King GJ: Systematic and ran-
dom errors associated with Johnston's cephalometric
analysis. Br J Orthod 20:101-107, 1993
13. Elder JR, Tuenge RH: Cephalometric and histological
changes produced extra-oral high pull traction to the max-
illqa in Macaca Mulatta. Am J Orthod 66:599-617, 1974
14. Sproule W: Dentofacial changes produced by extraoral
cervical traction to the maxilla of the Macaca Mulatta: A
histological and serial cephalometric study. Am J Orthod
56:532-545, 1969
15. McNamara JA, Bookstein FL, Shaughnessy TG: Skeletal
and dental changes following functional regulator ther-
apy on class II patients. Am J Orthod 88:91-110, 1985
16. Haynes S: A cephalometric study of mandibular changes
in modified function regulator (Frankel) treatment.
Am J Orthod Dentofacial Orthop 90:308-320, 1986
17. Stockli PW, Willert HG: Tissue reactions in the temporo-
mandibular joint resulting from anterior displacement
of the mandible in the monkey. Am J Orthod 60:142-
155, 1971
18. Hin ton RJ, McNamara JA: Temporal bone adaptations
in response to protrusive function in juvenile and young
adult rhesus monkeys (Macaca mulatta). Eur J Orthod
6:155-174, 1984
19. Woodside DG, Altuna G, Harvold E, et al: Primate ex-
periments in malocclusion and bone induction. Am J
Orthod 83:460-468, 1983
20. Woodside DG, Metaxas A, Altuna G: The influence of
functional appliance therapy on glenoid fossa remodel-
ing. Am J Orthod Dentofacial Orthop 92:181-198, 1987
21. McNamara JA, Carlson DS: Quantitative analysis of tem-
poromandibular joint adaptations to protrusive func-
tion. Am J Orthod 76:593-611, 1979
22. Cook AH, Sellke TA, BeGole EA: Control of the vertical
dimension in Class II correction using a cervical head-
gear and lower utility arch in growing patients. Part I.
Am J Orthod Dentofacial Orthop 106:376-388, 1994
23. Baumrind S, Molthen R, West EE, et al: Mandibular
plane changes during maxillary retraction. Am J Orthod
74:32-40, 1978
24. Creekmore TD: Inhibition or stimulation of the vertical
growth of the facial complex, its significance to treat-
ment. Angle Orthod 37:285-297, 1967
25. Poulton DR: The influence of extraoral traction. Am J
Orthod 53:8-18, 1967
26. Kühn RJ: Control of anterior vertical dimension and
proper selection of extraoral anchorage. Angle Orthod
38:340-349, 1968
27. Pearson LE: Vertical control in fully-banded orthodontic
treatment. Angle Orthod 56:205-224, 1986
28. Schudy FF: Cant of the occlusal plane and axial inclina-
tion of teeth. Angle Orthod 14:69-82, 1963
29. Richardson ME: Mesial migration of lower molars in
relation to facial growth and eruption. Aust Orthod J
14:87-91, 1996
<<    
     Article
      >> Home | TOC |          
Index

Biomechanical Considerations in the

Management of the Vertical Dimension
Stanley Braun

The vertical dimension may be altered purposefully or unintentionally dur-

ing therapy by dental extrusion, intrusion, growth modification, or orthog-
nathic surgery. Vertical changes result in the mandible rotating either open
or closed with corresponding alterations in interarch dental relationships,
facial esthetics, and lip and tongue function. The biomechanics involved in
the treatment of patients who exhibit an anterior open bite or a deep
overbite related to excessive or deficient vertical facial dimensions are
discussed. (Semin Orthod 2002;8:149-154.) Copyright 2002, Elsevier
Science (USA). All rights reserved.

T he vertical dimension has been defined by

some1'3 as the relationship between the
Frankfort and mandibular planes and by others4-7
as the relative relationship between the anterior
and posterior facial heights. Alterations in the
vertical dimension can occur purposefully or un-
intentionally during therapy by dental extrusion,
intrusion, growth modification, or surgical inter-
vention. Vertical changes result in the mandible
rotating either open or closed with corresponding
alterations in interarch dental relationships, facial
esthetics, and lip and tongue function. This article
examines the biomechanics involved in the treat-
ment of patients who exhibit an anterior open bite
or a deep overbite related to excessive or deficient
vertical facial dimensions.
The Anterior Open Bite Related to an
Excessive Vertical Dimension
This may be treated by surgically impacting the
dentomaxillary complex. As a consequence, the
mandible will rotate, contributing to a reduction
of the vertical dimension.8'9 Ancillary mandibu-
lar surgery may be required depending on the
resultant degree of mandibular rotation because
From Vanderbilt University Medical Center, Nashville, TN.
Address correspondence to Stanley Braun, DDS, MME, 7940
Dean Road, Indianapolis, IN 46240.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0803-0006$35.00/0
doi:10.1053/sodo.2002.125434
it tends to drive the dental occlusion toward a
Class III relation. If, however, the treatment ob-
jective requires both the maxillary and mandib-
ular dentitions to be positioned more forward,
the dentomaxillary complex may also be posi-
tioned anteriorly during the surgical impaction
process. Thus, it is possible that either one or
both jaws may require surgical procedures de-
pending on the pre treatment facial objectives
established. It should be noted that postsurgical
vertical growth has been reported; therefore,
timing of surgery is vital.10'11
In a patient undergoing circumpubertal skel-
etal growth, it is possible to alter the vertical
component of maxillary growth through the use
of extraoral orthopedic forces. In Figure 1, a
high-pull headgear is shown; it's force is directed
through the center of resistance of the den-
tomaxillary complex.12'13 It should be noted that
the direction (angle) of the headgear force is
designed to produce a larger maxillary intrusive
component than a distally directed one. By re-
stricting the vertical component of maxillary
growth in this manner, the normal vertical com-
ponent of mandibular growth may also be
masked through mandibular rotation. This com-
bined effect reduces the vertical dimension.
The line of action of the headgear force may
alternatively be positioned distal to or anterior
to the center of resistance of the dentomaxillary
complex (Fig 2). This would tend to rotate the
dentomaxillary complex during growth about a
point anterior or posterior to its center of resis-

Seminars in Orthodontics, Vol 8, No 3 (September), 2002: pp 149-154 149

<<    
     Article
      >>
150 Stanley Braun
Q Dentomaxillary
Center of Resistance
Figure 1. High-pull headgear force through the den-
tomaxillary complex center of resistance.
tance, depending on the point of application of
the headgear force relative to its center of resis-
tance.13 Rotation of the dentomaxillary complex
in this manner may be desirable if it is initially
abnormally angulated so as to produce an in-
creased vertical dimension. The extraoral forces
of the headgear may also produce some dentoal-
veolar response while altering the growth of the
dentomaxillary complex because the forces are
applied through the dentition.
Functional appliances with bite blocks have
been used to diminish the eruption of the poste-
rior teeth and the vertical growth component of
the dentomaxillary complex.14'15 If the bite regis-
tration is designed to bring the mandible forward,
then a headgear effect is obtained wherein the
anteroposterior growth component of the den-
tomaxillary complex may also be diminished.16
Dentomaxillary
Center of Resistance
Figure 2. High-pull headgear forces anterior or pos-
terior to the dentomaxillary center of resistance.
Home | TOC |          
Index
More recently, simultaneous high-pull headgear in
combination with a bite block have been used in
the reduction of vertical excesses.17
If extractions are required in a patient with an
excessive vertical dimension, the biomechanics of
extraction-site closure must be carefully controlled
to avoid eruption of the posterior teeth. One com-
monly used method of extraction-site closure is to
use horizontal chains or coil springs while the
teeth are engaged on a relatively light archwire.
This does not apply defined moment-to-force ra-
tios (M/F) to the teeth anterior and posterior to
the extraction site(s). For teeth to translate, a con-
sistent M/F approximating 10:1 is required. This
cannot be provided by chains and/or coil springs
while the teeth are engaged on an archwire.18'21 In
this approach to the closure of extraction site (s),
the anterior and posterior teeth characteristically
tip into the extraction site(s) during closure. The
M/F ratios are less than 10:1. Consequently, sec-
ondary reverse curve of Spec archwires are then
introduced with the concept that two equal and
opposite moments will be produced to upright the
tipped teeth. These uprighting moments must be
equal and should be in the range of 3,000 gr.mm
each for efficient root movement with minimal
root resorption.12 This is seldom the case, and
M p> M a
MP-Ma=Fd
Figure 3. Moments and vertical forces resulting from
placement of reverse curve of Spee archwires after
extraction-site closure.
<<    
     Article
      >>
Management of the Vertical Dimension
when there is an inequality of uprighting mo-
ments, vertical forces occur, resulting in the poten-
tial eruption of the posterior teeth. The spring rate
primarily varies inversely as the cube of the inter-
bracket wire length across the extraction site(s) (R
a l/ L3). Thus, the rate is high, and the anterior
and posterior moments easily become unequal be-
cause the posterior and anterior tooth movement
velocities are not necessarily equal. When the pos-
terior moment exceeds the anterior moment, an
undesirable eruptive force will occur (Fig 3). This
is undesirable in the patient already exhibiting an
increased vertical dimension. The same problem
also exists with the placement of archwire V bends
at the extraction site(s) related to closure. Again,
the applied moments must be equal throughout
the range of activation, and because this approach
invariably involves a high spring rate, it makes
inequalities in the moments applied to the ante-
rior and posterior teeth a likely occurrence as the
anterior and posterior teeth approach each other.
To reduce this problem, frequent in-office appli-
ance adjustments are required. One should also
consider that the increased angulation of the arch-
wire at the brackets' interfaces might result in
significant undesirable frictional effects.
A controlled, efficient means of extraction-
site closure has been outlined previously18"20
wherein moment alterations along with related
vertical forces between the anterior and poste-
rior teeth are under control of the orthodontist
throughout extraction-site closure. It should be
noted that in the most stringent requirement
wherein the entirety of the extraction site(s)
must be occupied by the anterior teeth a maxi-
mum vertical eruptive force of 22.7 g is pro-
duced at the occlusal surfaces of the six poste-
rior teeth.18 Functional (biting) forces easily
negate this vertical force preventing their erup-
tion. This is true for the patient with a steep
mandibular plane angle as well.21
It has been suggested that extractions be con-
sidered for the purpose of protracting the pos-
terior teeth into the extraction sites, thereby
reducing the vertical dimension.22'23 Recent
studies do not substantiate this.24-25
A maxillary protraction device commonly used
in the treatment of Class III malocclusions consists
of bilateral forces emanating from a face mask and
applied to the first molar or canine regions. The
occlusogingival angle of the protraction forces in
the sagittal view is determined by the commissure
of the lips and the intraoral points of attachment
Home | TOC |          
Index
151
(Fig 4). Any constrained body (a tooth, a group of
teeth, or an osseous structure joined to other os-
seous structures through viable sutures) will react
to the forces applied to it relative to its center of
resistance.26'27 The locations of the centers of re-
sistance of the dentomaxillary complex have re-
cently been determined.12 They are located on a
line drawn perpendicular to the functional occlu-
sal plane (FOP) through the distal contacts of the
maxillary first molars as seen in a sagittal cephalo-
gram. They are further identified on the afore-
mentioned perpendicular line at one half the dis-
tance from the FOP to a line drawn parallel to the
FOP through the inferior border of the orbit.
There are two centers of resistance of the den-
tomaxillary complex when reviewed in the frontal
aspect. This is because the dentomaxillary com-
plex is essentially made up of two bones, a right
and a left maxilla each containing one half the
dental arch. Each maxillary bone articulates with
the other at the median palatine suture and rela-
tively symmetrically on each side with the fron-
tomaxillary suture, the nasomaxillary suture, the
zygomaticomaxillary suture, and the transverse
palatine suture. Because each of the protraction
forces (one on each side) results in a moment that
tends to stress the midpalatine suture greater at its
distal area than at its anterior area, each half of the
dentomaxillary complex acts somewhat indepen-
dently. Thus, two centers of resistance are identi-
fied. However, if protractive forces are applied in
the presence of a stiff 0.036-in stainless steel trans-
palatal arch or a sutural expander, the left and
<p = Dentomaxillary center of resistance
F = Protraction force applied to teeth
F' = F = Protraction force equivalent at center of resistance
M = F(Y) = Tipping moment at center of resistance
FOP = Functional occlusal plane
F(sin 6) = Eruptive component of protraction force
Figure 4. The force system produced by a commonly
used protraction device.
<<    
     Article
      >>
152 Stanley Braun
Figure 5. Protraction force adjusted to pass through
the centers of resistance of the dentomaxillary com-
plex bilaterally.
right maxillary bones act as one for the separating
moments are negated by the presence of either of
these devices or their equivalent. In the sagittal
view, only one center of resistance is identified
since the two are superimposed. Consequently, the
aforementioned protraction forces (attached to
either the molar or canine regions near the plane
of occlusion and forced to exit through the com-
misure of the lips) will cause a counterclockwise
moment and eruptive forces relative to the centers
of resistance of the dentomaxillary complex (in
the sagittal view). As a result, the dentomaxillary
complex will rotate about a point close to its cen-
ters of resistance.19 It is therefore important for the
clinician to recognize that this type of protraction
device does not cause protraction. It causes den-
tomaxillary rotation, which results in an increased
lower facial height and a decrease in the upper
facial height.26 In a recent thesis,27 it has been
shown that if the points of protraction force attach-
ments are relatively high in the canine vestibular
regions, their lines of action pass through, or ex-
tremely close to, the centers of rotation of the
dentomaxillary complex (Fig 5). Consequently, ro-
Home | TOC |          
Index
tation of the dentomaxillary complex is eliminated
or significantly reduced. However, an undesirable
eruptive component of the protraction force re-
mains, which results in an increased vertical di-
mension with the consequences of mandibular ro-
tation and a decreased overbite. For true
protraction to occur, the line of action of any
protractive forces must pass through the centers of
resistance of the dentomaxillary complex and be
parallel to the occlusal plane.
Recently, a protraction device design has
been reported that permits true protraction
forces to be applied through the centers of re-
sistance of the dentomaxillary complex.13 This
consists of a standard facebow contoured to in-
sert in the maxillary molar tubes from the distal
(Fig 6). The outer bow may be adjusted so that
the lines of action of the protraction forces pass
through the centers of resistance bilaterally as
shown in Figure 7, resulting in true protraction
(translation) of the dentomaxillary complex. It
is important to note that in the case of vertical
maxillary excess the outer bow can be adjusted
to have an intrusive (impaction) component. An
extrusive component can be similarly obtained
in the case of a maxillary vertical deficiency. The
protractive forces may also be located superior
to the center of resistance (in the sagittal view),
resulting in a clockwise moment accompanying
the protractive force. This resulting moment has
the potential of rotating the anterior portion of
the maxilla downward without posterior maxil-
lary extrusion. Thus, it may be possible to obtain
an improved incisor/lip relationship if desired.
As a result, any alterations in the vertical dimen-
sion, purposeful or none at all, are under the
clinician's control.
The Deep Overbite Related to a
Deficient Vertical Dimension
In a growing individual, the mandible may be
rotated open by supereruption of the mandibu-
Figure 6. Inner bow of a protraction headgear in-
serted from the distal of the molar headgear tubes.
<<    
     Article
      >>
Management of the Vertical Dimension
Figure 7. Line of action of protraction force passing
through the center of resistance of the dentomaxillary
complex, resulting in translation.
lar posterior teeth when using reverse curve of
Spee mechanics. This will result in a reduction
of a deep overbite while increasing the mandib-
ular plane angle, the anterior facial height, and
facial convexity. This approach often requires
ancillary growth modification of the dentomax-
illary complex with high distal pull headgear.
Failing this, the Angle Class II tendency created
by rotating the mandible open may not be cor-
rected through growth alone. Additionally, fail-
ure to alter the continued downward and for-
ward growth of the dentomaxillary complex may
result in an undesirable anterior facial height.20
To avoid mandibular incisor flaring, which ac-
companies reverse curve of Spee mechanics, the
clinician should consider using segmented arch
mechanics, wherein the clinician determines the
location of the occlusal plane rather than the
appliance.18'20'21'28-29
Passive eruption of the mandibular posterior
teeth may also be encouraged through the use of
a bite plate. This tends to prevent the normal
eruption of the mandibular anterior teeth while
Home | TOC |          
Index
153
permitting the posterior teeth to passively erupt.
The bite plate should be in place continuously as
well as after the eruption of the mandibular
posterior teeth to allow mandibular growth to
catch up. High-pull headgear to the dentomax-
illary complex to reduce its eruption and to
compensate for the resultant mandibular rota-
tion, which tends to drive the dentition in a Class
II direction, should be considered.
If posterior dental eruption is used to reduce
a deep overbite, to increase the mandibular
plane angle, and to increase anterior facial
height in the nongrowing patient or in a patient
with limited dentofacial growth, compensatory
mandibular orthognathic surgery or distraction
osteogenesis to alter the ramal and/or body
length of the mandible is necessary.30'31
The lower facial height may also be increased
in the growing individual through encourage-
ment of vertical growth and dental eruption of
the maxillary complex. This may be accom-
plished with cervical headgear. Its orthopedic
forces may be through the center of resistance of
the dentomaxillary complex (Fig 8) or anterior
or posterior to it, depending on the type of
alteration desired in the C axis.32 The accompa-
nying downward mandibular "rotation" that oc-
curs tends to alter the occlusion in a Class II
direction. The distal component of the cervical
headgear, accompanied by mandibular growth,
is needed to reduce this transient effect.
Over time, interarch elastics will, in the case
of a Class III traction, cause eruption of the
maxillary posterior teeth and in the case of a
Dentomaxillary
Center of Resistance
Figure 8. Potential lines of action of cervical head-
gear forces relative to the center of resistance of the
dentomaxillary complex.
<<    
     Article
      >>
154 Stanley Braun
class II traction, cause eruption of the mandib-
ular posterior teeth, thereby increasing the ver-
tical dimension. This is also accompanied by
rotation of the occlusal plane.33 Accordingly,
skeletal growth is critical to accommodate these
treatment changes. Use of Class II or Class III
interarch elastics in patients with a vertical ex-
cess is contraindicated.
Conclusion
Before beginning active orthodontic therapy, it
is vital that the clinician clearly establish post-
treatment objectives for each patient. This in-
cludes anterior facial height, mandibular plane
angle, location of the occlusal plane, and the an-
teroposterior position of the dentition, among
others. Once this has been performed, a de-
tailed individualized mechanical plan to support
these objectives should be outlined. Various pro-
cedures and their biomechanical considerations
have been outlined for the clinician to consider
to achieve these objectives.
References
1. Broadbent BH: The face of the normal child. Angle
Orthod 7:183-207, 1937
2. Brodie AG: On the growth patterns of the human head
from the third month to the eighth year of life. Am J
Anat 68:209-262, 1941
3. Wylie W, Johnson E: Rapid evaluation of facial dysplasia
in the vertical plane. Angle Orthod 22:165-182, 1952
4. Tweed CH: A philosophy of orthodontic treatment. Am J
Oral Surg 31:74-103, 1945
5. Merrifield LL: Edgewise sequential directional force tech-
nology. J Charles H. Tweed Int Found 14:22-37, 1986
6. Shudy FF: Vertical growth vs. anteroposterior growth as
related to function and treatment. Angle Orthod 34:75-
93, 1964
7. Horn AJ: Facial height index. Am J Orthod 101:180-186,
1992
8. Ellis E: The nature of vertical maxillary deformities.
J Oral Maxillofac Surg 43:756-762, 1985
9. Fish LC, Wolford LM, Epker BM: Surgical-orthodontic
correction of vertical maxillary excess. Am J Orthod
73:241-257, 1978
10. Pancherz H, Fackel U: The skeletofacial growth pattern
pre-and post-dentofacial orthopedics: A long term study
of Class II malocclusions treated with the Herbst appli-
ance. Eur J Orthod 12:209-218, 1990
11. Nanda SK: Patterns of vertical growth of the face. Am J
Orthod Dentofacial Orthop 93:103-116, 1998
12. Lee K, Ryu W, Park Y, et al: A study of holographic
interferometry on the initial reaction of the maxillofa-
cial complex during protraction. Am J Orthod Dentofa-
cial Orthop 111:623-632, 1997
13. Braun S, Lee K, Legan HL: A re-examination of various
Home | TOC |          
Index
extraoral appliances in light of recent research findings.
Angle Orthod 69:81-84, 1999
14. Frankel R: A functional approach to orofacial orthope-
dics. Br J Orthod 7:41-51, 1980
15. Harvold EP, Vargervik K: Morphogenic response to ac-
tivator treatment. Am J Orthod 60:478-490, 1971
16. Proffit WR, Fields HW Jr (eds): Contemporary Orth-
odontics (ed 2). Philadelphia, PA, Mosby, 1992
17. Lagerström LO, Nielsen IL, Lee R, et al: Dental and skel-
etal contributions to occlusal correction in patients treated
with high-pull headgear-activator combination. Am J
Orthod Dentofacial Orthop 97:495-504, 1990
18. Braun S: Diagnosis driven vs. appliance driven treatment
outcomes, in Sachdeva RC, Bantleon HC, White LW, et
al (eds): Orthodontics for the Next Millenium. Glen-
dora, CA, Ormco Corp, 1997
19. Marcotte MR: Biomechanics in Orthodontics. Philadel-
phia, PA, BC Decker, 1990
20. Burstone CJ: Modern Edgewise Mechanics and the Seg-
mented Arch Technique. Glendora, CA, Ormco Corp,
1995
21. Braun S, Sjursen RCJr, Legan HL: On the management
of extraction sites. Am J Orthod Dentofacial Orthop
112:645-655, 1997
22. Tulley WJ: The role of extractions in orthodontic treat-
ment. Br Dent J 107:199-205, 1959
23. Wyatt NE: Preventing adverse effects on the TMJ
through orthodontic treatment. Am J Orthod 91:493-
499, 1987
24. Kacaderali I: The effect of first premolar extraction on
the vertical dimension. Am J Orthod Dentofacial Orthop
116:47-50, 1999
25. Casinano C, McLaughlin RP, Zernik JH: Effect of first
bicuspid extraction on facial height in high angle cases.
J Clin Orthod 27:594-598, 1993
26. Nanda R, Bruce G: Biomechanical approaches to the
study of alterations of facial morphology. Am J Orthod
78:213-226, 1980
27. Leone P: The effect of one type efface mask therapy on
the 'C' axis: The growth axis for the dentomaxillary
complex. Master's thesis, Saint Louis University, St
Louis, MO, 2001
28. Braun S, Hnat WP, Johnson BH: The curve of Spec
revisited. Am J Orthod Dentofacial Orthop 110:206-210,
1996
29. Braun S: An examination of the effects of leveling with
nickel titanium archwires combined with torqued inci-
sor brackets. Semin Orthod 7:215-220, 2001
30. Bell WH, Proffit WR, Chase DL, et al: Mandibular defi-
ciency, in Bell WH, Proffit WR, White RP (eds): Surgical
Correction of Dentofacial Deformities. Philadelphia, PA,
Saunders, 1980, pp 684-843
31. Guerroro CA, Bell WH, Meza LS: Intraoral distraction
osteogenesis. Atlas Oral Maxillofac Surg Clin North Am
7:111-151, 1999
32. Braun S, Rudman RT, Murdoch HJ, et al: The 'C' Axis:
A growth vector for the maxilla. Angle Orthod 116:264-
270, 1999
33. Braun S, Legan HL: Changes in occlusion related to the
cant of the occlusal plane. Am J Orthod Dentofacial
Orthop 111:184-188, 1997
<<    
     Article
      >> Home | TOC |          
Index

The Effects of Altering Vertical Dimension on

the Masticatory Muscles and
Temporomandibular Joint
Charles G. Widmer

Increasing the vertical dimension of occlusion by surgical repositioning of

the maxillary or mandibular segments or by prosthetic approaches such as
interocclusal appliances can affect specific components of the jaw closing
muscles and the temporomandibular joint. It has been assumed that the
relapse observed after orthognathic surgery is due to an increased active
and passive vertical force that is generated by the muscle and the connec-
tive tissue respectively. This review identifies multiple potential sources
including the innate characteristics of the muscle fibers, muscle spindle
activation, the architecture of the jaw closing muscles and biomechanical
changes that can contribute to the increase in vertical force after lengthen-
ing the jaw closing muscles or increasing mandibular height. It is con-
cluded that future advancement into the interactions between the hard
and soft tissue will depend on a more complete understanding of the jaw
muscle anatomy and function at the molecular, architectural and biome-
chanical levels. (Semin Orthod 2002;8:155-161.) Copyright 2002, Elsevier
Science (USA). All rights reserved.

Vertical dimension of occlusion has been of
interest for a long period of time in the
areas of prosthodontic reconstruction, orth-
odontics and orthognathic surgery. It is interest-
ing to note that even though a number of spe-
cialties are involved in the manipulation of
vertical dimension, there have been very few
studies on the impact of these changes on the
functional masticatory status of the patient. Most
published studies provide empirical descriptions
of the clinical outcome of changing the vertical
dimension, but these studies lack the necessary
experimental design to adequately test whether
the effect was caused by changes in vertical
From the Department of Orthodontics, College of Dentistry, Uni-
versity of Florida, Gainesville, FL.
Supported in part by grant No. DE 12207 from the National
Institute of Dental and Craniofacial Research.
Address correspondence to Charles G. Widmer, DDS, MS, Dept of
Orthodontics, Box 10044, JHMHSC, 1600 SW Archer Road, Uni-
versity of Florida, Gainesville, FL 32610-0444.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0803-0007$35.00/0
doi:10.1053/sodo.2002.125435
height, tooth contacts, or other factors.1'2 When
the vertical dimension is increased, the stability
of this change is unclear and seems to be depen-
dent on whether the soft tissue adapts to this
new position or the hard tissue adapts to the
influences of the soft tissue (ie, relapse). The
purpose of this overview is to examine recent
literature to provide new insight into the
changes that may occur in the jaw muscles and
temporomandibular joint (TMJ) with changes in
the vertical dimension.
What Is Vertical Dimension?
Vertical dimension by the simplest definition is
the vertical relationship between the maxilla and
mandible. Terms such as vertical dimension of
occlusion (VDO) and vertical dimension at rest
(VDR) are prosthodontic terms that refer to the
vertical dimension measured with the maxillary
and mandibular teeth in occlusion and at the
postural rest position of the mandible respec-
tively. Vertical dimension can also be described
as lower facial height using the distance between

Seminars in Orthodontics, Vol 8, No 3 (September), 2002: pp 155-161 155

<<    
     Article
      >>
156 Charles G. Widmer
the anterior nasal spine (ANS) and gnathion.
These definitions indirectly describe the func-
tional length of the jaw closing muscles either
when the teeth are in contact or in the rest
position. It is important to emphasize the rela-
tionship between the jaw muscles and maxillo-
mandibular relationships because the jaw mus-
culature acts as a primary determinant of vertical
dimension or lower facial height.3 The defini-
tion of these measures must be modified, how-
ever, after mandibular surgical procedures such
as bilateral sagittal split osteotomy (BSSO) when
the mandible is split into a proximal and distal
segment. The traditional measures of maxillo-
mandibular relationships do not reflect the mus-
cle length in this case because there is dissocia-
tion between the segment where the muscles are
attached (proximal segment) and the occlusal
or bony landmarks that are associated with the
distal segment. The relationship of the proximal
segment to the cranial base would be a more
accurate representation of the jaw closing mus-
cle functional length.4'5 The need for emphasiz-
ing the relative position of the proximal segment
has been recognized for a number of years, and
various methods have been proposed to position
this segment according to the treatment plan.5'6
VDR
The postural position of the mandible (VDR)
while the patient is sitting or standing is main-
tained in an antigravity position by the jaw clos-
ing muscles including the masseter, temporalis,
and medial pterygoid muscles. This postural po-
sition or vertical dimension of rest is controlled
by the activation of muscle spindles (stretch re-
ceptors) that are located in parallel to the mus-
cle fibers. When the mandible is stretched open
even by just a few tenths of a millimeter, these
exquisitely sensitive receptors will activate the
jaw-closing muscles to slightly close the jaw and
restore the mandibular postural position. These
receptors also provide feedback to the central
nervous system (CNS) regarding muscle length
so that as the mandible is opened, the activation
of these receptors is more profound, and the
CNS can receive information regarding the po-
sition of the mandible. The sensitivity of the
muscle spindle can be influenced by many fac-
tors such as peripheral pressure or pain in the
oral region, by higher centers such as the limbic
Home | TOC |          
Index
system associated with anxiety and stress, and by
the central pattern generator for mastication
located in the reticular formation. It is common
to find that the mandibular postural position
can vary from day to day and even within the
same day so that there is no specific measure
that will accurately define this relationship.
Many techniques have been used to estimate this
position particularly in edentulous patients and
include functional assays such as speech patterns
and swallowing techniques.7'8 Techniques that
use electrical stimulation over the right and left
sides of the face have been shown to only stim-
ulate the superficial masseter muscle and do not
simulate the postural position of the mandible.9
VDO
Muscular forces from the jaw muscles and, to a
lesser extent, tongue muscles, facial muscles,
and passive elastic properties of the soft tissues
establish a vertical dimension of occlusion in a
patient with a normal dentition. One example of
this equilibrium can be seen when a slow attri-
tion of the dentition occurs secondary to
chronic bruxism and the vertical dimension of
occlusion is maintained by the slow eruption of
the abraded teeth and the alveolus so that there
is no VDO loss. However, the same does not
occur with attrition of teeth in complete den-
tures because there cannot be a structural com-
pensation. In these cases, there is a true loss of
vertical dimension that occurs over a number of
years and may result in masticatory musculoskel-
etal pain in some patients.
The fact that vertical dimension of occlusion
does not change significantly in the adult sug-
gests that equilibrium is established between the
hard (teeth, alveolus, and compact bone) and
soft tissues (jaw muscles, tongue muscles, facial
muscles, and connective tissue). Any disruption
of these components will affect the balance of
the forces, and change will occur in the form of
bony remodeling and/or soft-tissue adaptation.
To understand the role and adaptive capacity of
these components, it is necessary to first under-
stand normal form and function. Recent re-
search has provided new insight into the normal
form and function of the masticatory muscles,
particularly the masseter muscle, and these new
findings may have a significant impact on the
understanding of vertical dimension.
<<    
     Article
      >> Home | TOC |          
Index

Effects of VDO on Masticatory Muscles/TMJ 157

Masseter Muscle Architecture dibular form in the rabbit during a power stroke
showed this differential effect and also provided
Most of the extensive work identifying the archi-
insight into the TMJ torque changes associated
tecture of the masseter has been performed in
with increasing the vertical dimension of the
the mouse, rabbit, and pig.10'15 The masseter
mandible. A 20% vertical increase in the man-
muscle is composed of three layers arranged in a
dibular height (measured from the center of the
medial-lateral dimension, and each superficial,
condyle to the mental foramen in which a three-
intermediate, and deep layer is composed of
dimensional force transducer was attached)
multiple anatomical partitions. Each partition,
caused an 8.7% increase in the jaw-closing
by definition, has a unique tendon of origin or
torque as well as a 6.8% increase in negative
insertion and may contain multiple neuromus-
mandibular roll (rotating the condyle on an an-
cular compartments. Because neuromuscular
terior-posterior axis so that the mandibular
compartments are composed of exclusive collec-
teeth move lingually). Therefore, it appears (at
tions of motor units, they constitute the smallest
least in the rabbit) that the structural change of
elements within muscle that can be indepen-
increasing the vertical dimension alone without
dently activated.16 These output elements have
changing the muscle or soft-tissue contributions
distinctly different force trajectories and pro-
had an effect of increasing the jaw closing and
duce torques about the temporomandibular
mandibular roll torques produced about the
joint with different magnitudes and/or direc-
TMJ. Increasing these torques may require ad-
tions.17 Neuromuscular compartments can be
aptation in the TMJ in the form of bony remod-
activated in different combinations to form a
eling or increased tension on ligaments such as
diverse repertoire of torques about the TMJ
the collateral ligaments stabilizing the articular
from a single muscle.
disc. However, these data are based on the rabbit
The human masseter has many similar prop-
model and may not accurately represent the
erties to the masseter described in other species.
changes that would occur in the human TMJ.
For example, the human masseter has the same
three distinct layers (superficial, intermediate,
and deep) ,18 and each layer is composed of mul- Effect of Lengthening the Masseter
tiple partitions.19 The masseteric nerve-branch- Muscle after VDO Increase
ing patterns innervating the human masseter is
A muscle fiber generates differential force de-
similar to the general pattern observed in the
pending on the sarcomere length. As a muscle
rabbit and pig.20'21 There are obvious differ-
fiber is stretched from a relaxed position, cross-
ences between the human mandible and other
bridging between the actin and myosin filaments
species in the height of the mandible, form of
increases until they are maximal and then de-
the temporomandibular joint, arrangement of
creases. The maximum cross-bridging is consid-
the tooth row, and nonfused symphysis. How-
ered the optimal sarcomere length. Optimal sar-
ever, these animal models do provide a first level
comere lengths will generate the maximal force
of approximation into the architecture and and, for the jaw-closing muscles, this optimal
function of the human masticatory muscles and
length is found when the jaw is opened some-
allow more extensive studies to be made of the
what and not at the intercuspal position.23 For
CNS organization of the neurons innervating
example, tetanic tension was measured from the
the jaw muscles.
intermediate layer of the masseter muscle in the
rat while progressively increasing the vertical di-
mension, and it was found that the maximum
Increasing Mandibular Height Increases
force was produced at an average vertical dimen-
the Jaw-Closing Torque About the TMJ sion of 8 mm (50% of maximum opening).
The torques produced about the TMJ by indi- Thus, increasing the vertical dimension will have
vidual rabbit masseter compartments are differ- the effect of stretching the jaw-closing muscle
entially affected by variation in mandibular fibers closer to the optimal sarcomere length
form.22 This is not unexpected because of the and will produce higher force even though there
differences among the compartment-generated was no change in the muscle activation pattern.
torque trajectories. Modeling variation of man- It is important to realize that increasing the
<<    
     Article
      >>
158 Charles G. Widmer
vertical dimension has a differential effect on
stretching the jaw-closing muscle fibers such as
the masseter. A recent study of different regions
of the human masseter muscle has shown that
the largest increases in muscle fiber length oc-
curred in the medial part of the deep masseter
followed by the anterior, superficial masseter,
whereas the least change was found in the pos-
terior, superficial masseter.24 Similar differential
changes were also reported in the rabbit masse-
ter and, when sarcomere lengths were compared
with the timing of activation of the muscle re-
gion, the optimal sarcomere length (2.4-2.8 jam)
coincided with the peak electromyographic ac-
tivity.25 In summary, after an acute increase in
vertical dimension, the stretched muscle fiber
can generate higher forces because of optimiz-
ing sarcomere length even without changing the
activation pattern of the muscle and could con-
tribute to early adaptive changes of hard tissues
because of the increased load.
Myosin Composition of the Masseter
Muscle
Myosin heavy chain (MyHC) is one of the con-
tractile proteins found in muscle fibers and is
associated with contraction speed as well as the
fatigue properties of the muscle fiber. Nine dif-
ferent MyHC isoforms have been identified and
consist of developmental isoforms (embryonic
and neonatal); slow, fatigue-resistant isoforms
(type I, a-cardiac); and fast isoforms (type Ila,
IIx, lib, Urn, Hextraocular). Type Urn and Hextraocular
are specialized isoforms that have been found in
masticatory muscles of cats and canines (Urn)
and extraocular and laryngeal muscles.26'28 The
presence of different MyHC isoforms in a par-
ticular muscle will vary depending on a variety of
factors including the particular species, the func-
tional demand of the muscle, and hormonal
status of the animal. Generally, for jaw-closing
muscles, small species such as mice and rats have
fast MyHC type Ha, IIx, and lib isoforms that are
arranged in discrete locations within the mus-
cle.14 Larger species such as rabbits, pigs, and
humans have slower, more fatigue-resistant fi-
bers such as type I, a-cardiac, Ha, and IIx fibers
that are generally concentrated with the slower
fibers located more anteriorly and faster con-
tracting fibers located more posteriorly.29'31 The
functional implication for this general arrange-
Home | TOC |          
Index
ment of fibers is still unclear but may be related
to the different parts of the muscle that are more
highly stretched (eg, the anterior masseter in
which the fibers are generally slower in contrac-
tion speed and would maintain posture com-
pared with the posterior masseter in which the
muscle is stretched the least and faster contract-
ing fibers are located). Interestingly, the region
in which the stretch is maximal is also the loca-
tion of the muscle spindles in the masseter so
they are positioned ideally for detecting small
changes of mandibular posture.
The MyHC phenotype profile within the mas-
seter muscle has been found to be different
between genders. The male masseter has a
higher proportion of faster MyHC isoforms than
the female masseter.30 Although a logical func-
tional explanation may be considered in the
larger mammals in which the male has larger
canines and uses his teeth for fighting, it is not as
clear why gender differences exist in small mam-
mals such as the mouse and rat. This disparity in
fiber types between genders has been found to
be dependent on the hormone testosterone.32
For example, if a young male rabbit is castrated
without testosterone supplementation, the
MyHC phenotype profile of his masseter muscle
when he reaches adult is no different from that
observed in female rabbits. These differences in
male and female jaw-muscle contractile charac-
teristics may affect the timing and amount of
skeletal relapse observed after orthognathic sur-
gical experiences. Other hormones such as thy-
roid hormone also can change the phenotype of
muscles, causing slower MyHC phenotypes to
switch to faster isoforms when given in large
doses.33
Another method that promotes muscle fiber
phenotype switching is changing the activity pat-
tern of the muscle. For example, slow muscles
have a prolonged firing pattern that extends
over many hours usually for postural mainte-
nance, whereas fast twitch fibers fire in brief
bursts. When a fast muscle is activated for a
prolonged period similar to slow muscle activity,
the phenotype of the fast muscle eventually
switches to a MyHC phenotype that is slower in
contraction time and more fatigue resistant.34
Phenotype switching is one means by which mus-
cles such as the masseter can adapt to the func-
tional demands of the masticatory system. For
example, placement of a bite-raising appliance
<<    
     Article
      >>
Effects of VDO on Masticatory Muscles/TMJ
in the pig increases the proportion of slow
MyHC expression and is probably because of the
increased grinding to attempt to remove the
appliance (see later).
Early Effects of a VDO Increase on the
Jaw-Closing Muscles
Any appliance or prostheses that causes an in-
crease in VDO has the potential to cause
changes in the jaw-closing muscles. There are
two lines of experimental evidence that are
markedly different and are dependent on
whether the vertical dimension was increased
but the muscle was splinted as in intermaxillary
fixation or whether the VDO was increased and
the jaw was freely moving, such as the use of an
interocclusal appliance. Immobilization of the
muscle with an increase in muscle length results
in the formation of new sarcomeres at the ends
of the muscle fiber and atrophy of the muscle
fibers. This atrophy is independent of the mus-
cle fiber phenotype because both type I and type
II fibers were affected.35 When functional length
of the muscle was restored after the removal of
the splint, the muscle fiber reduced the number
of sarcomeres to adapt the muscle fiber to the
original length, and the atrophic fibers were
restored to their original size. Alternatively, al-
lowing the mandible to be freely movable with
function on the appliance seems to have the
effect of adding sarcomeres to lengthen the
muscle fiber as well as resulting in the hypertro-
phy of the muscle fiber.
Based on recent studies, changes at the mo-
lecular level also occur fairly rapidly in muscle
fibers after placement of a bite-raising appli-
ance. For example, studies using these appli-
ances in rats and pigs have shown that an up-
regulation in the RNA message for slower MyHC
isoforms occurs within a few days after appliance
use.36'37 This is now known to be caused by the
release of mechano-growth factor (MGF), an iso-
form of insulin growth factor-1, which is pro-
duced locally in the muscle fibers and is signifi-
cantly upregulated in exercised and damaged
muscle.38'39 Within 7 to 14 days, new MyHC pro-
tein is formed in muscle fibers, changing the
profile of the muscle toward a slower muscle.
However, interpretation of these results as they
apply to the human jaw-closing muscle response
to a bite-raising appliance should be made with
Home | TOC |          
Index
159
caution. All three animal models bruxed in an
attempt to remove the appliance. This increased
muscle activity in the jaw-closing muscles may
have been a primary factor in the changes ob-
served in the MyHC isoform switch to a slow
phenotype. Prolonged activity levels over ex-
tended periods of time such as with bruxism will
influence phenotype switching as described ear-
lier. So it appears that an increase in vertical
dimension along with an increase of jaw-closer
muscle function such as bruxism will promote
MyHC phenotype switching to a slower isoform.
Does Increasing the Vertical Dimension
Cause Bruxism?
Placement of an interocclusal appliance will af-
fect many components including the vertical di-
mension, tooth contacts, tongue posture, and
the patient's awareness.40 The response to any of
these factors may be a modification of the nor-
mal muscle activity pattern either reflexively
(such as to maintain mandibular posture and a
patent airway) or voluntarily. In the case of the
animal studies, the intent to remove the foreign
material by bruxing was a primary behavioral
drive of the animal. Bruxing behavior has not
been a significant response by patients that use
an interocclusal appliance, and, in fact, this type
of appliance tends to initially reduce jaw-closing
muscle activity with a gradual return to pretreat-
ment levels.41
Vertical Dimension Increases and
the TMJ
There is a paucity of studies that have examined
the effects of increasing the vertical dimension
on the TMJ. Studies that have cemented interoc-
clusal appliances over the teeth of animals cause
the animal to brux, and this increase in force
generation will be reflected by changes of load
on the condyle.42 Indeed, changes were ob-
served in the chondroblastic and prechondro-
blastic layers of the condylar cartilage support-
ing a bony remodeling secondary to increased
load. However, it is difficult to determine if the
condylar changes observed in these animals are
because specifically of the increase in vertical
dimension or to the change in the loading pat-
tern in the joint because of bruxing. In addition,
it is possible that biomechanical changes such as
<<    
     Article
      >>
160 Charles G. Widmer
an increased jaw-closing torque because of an
increase in mandibular height can produce an
altered function in the TMJ (see earlier). Fur-
ther studies are necessary to distinguish the ef-
fect of vertical dimension on the temporoman-
dibular joint.
Can Skeletal Stability be Maintained
After Increasing the VDO?
Vertical relapse is observed after orthognathic
surgical procedures that increase the vertical di-
mension and seems to occur at different rates
depending on the type of fixation that was used.
Rigid fixation has a lower amount of relapse, at
least initially43 and may be because of the de-
mand for muscle fibers to function at a new,
longer length, thus requiring additional sarco-
meres to be added as an adaptive response. The
lengthening of the muscle fibers will also nor-
malize the output of the muscle spindles. Wire
fixation, however, allows the bony segments to
be flexed initially and allows muscle fibers to
function nearer their original length so that lit-
tle muscle adaptation would be required. In this
case, the bony segments would adapt rather than
the muscle. After 5 years, however, it appears
that the vertical dimension stabilizes at the same
vertical height regardless of the type of fixa-
tion.44 Therefore, the mechanism for long-term
stabilization of the facial skeleton is still unclear
but is probably because of the equilibrium of
forces from a variety of muscles including the
jaw closers, the tongue, and the facial muscles.
Summary
It has been hypothesized that increasing the
vertical dimension will cause an increase in jaw-
closing force because of active and passive forces
of the muscles and soft tissues. However, it has
never been clear how this increased force is
generated. Based on our current knowledge, an
increased vertical force is caused by the combi-
nation of tonic low-level muscle activity caused
by muscle spindle activation combined with the
increased force produced by the stretched mus-
cle fiber. However, muscle fibers are not
stretched homogeneously within a jaw muscle as
seen in the masseter muscle so the effect of
muscle stretch is regional. Finally, increasing the
mandibular height has the biomechanical effect
Home | TOC |          
Index
of increasing torques produced about the TMJ.
These factors contribute to the active mainte-
nance of the vertical dimension. Therefore, the
key to successful stabilization of facial skeletal
morphology after increasing the vertical dimen-
sion may lie in understanding the complex bio-
mechanical effects of the jaw muscles at multiple
levels and controlling the adaptive change of
muscle and bone.
References
1. Sato S, Hotta TH, Pedrazzi V: Removable occlusal over-
lay splint in the management of tooth wear: A clinical
report. J Prosthet Dent 83:392-395, 2000
2. Ormianer Z, Gross M: A 2-year follow-up of mandibular
posture following an increase in occlusal vertical dimen-
sion beyond the clinical rest position with fixed restora-
tions. J Oral Rehabil 25:877-883, 1998
3. Proffit WR, Turvey TA, Phillips C: Orthognathic surgery:
A hierarchy of stability. Int J Adult Orthodon Orthog-
nath Surg 11:191-204, 1996
4. Leonard M: Preventing rotation of the proximal frag-
ment in the sagittal ramus split operation. J Oral Surg
34:942, 1976
5. Epker BN, Wylie GA: Control of the condylar-proximal
mandibular segments after sagittal split osteotomies to
advance the mandible. Oral Surg Oral Med Oral Pathol
62:613-617, 1986
6. Burye MT, Stella JP: An innovative method for accurate
positioning of the proximal segment in sagittal split
osteotomies. Int J Adult Orthodon Orthognath Surg
15:59-63, 2000
7. Miralles R, Dodds C, Palazzi C, et al: Vertical dimension.
Part 1: Comparison of clinical freeway space. Cranio
19:230-236, 2001
8. Woda A, Piochon P, Palla S: Regulation of mandibular
postures: Mechanisms and clinical implications. Grit Rev
Oral Biol Med 12:166-178, 2001
9. Dao TTT, Feine JS, LundJP: Can electrical stimulation
be used to establish a physiologic occlusal position. J
Prosthet Dent 60:509-514, 1988
10. Schumacher GH: Funktionelle morphologic der kau-
muskulatur. Jena: VEB Gustav Fischer, 1961
11. Weijs WA, Dantuma R: Functional anatomy of the mas-
ticatory apparatus in the rabbit. Neth J Zool 31:99-147,
1981
12. Herring SW, Wineski LE, Anapol FC: Neural organiza-
tion of the masseter muscle in the pig. J Comp Neurol
280:563-576, 1989
13. Widmer CG, Klugman D, English AW: Anatomical par-
titioning and nerve branching patterns in the adult rab-
bit masseter. Acta Anat (Basel) 159:222-232, 1997
14. Eason JM, Schwartz GA, Pavlath GK, et al: Sexually di-
morphic expression of myosin heavy chains in the adult
mouse masseter. J Appl Physiol 89:251-258, 2000
15. Widmer CG, Morris-Wiman JA, Nekula C: Spatial distri-
bution of myosin heavy chain isoforms in mouse masse-
ter. J Dent Res 81:33-38, 2002
16. English AW, Letbetter WD: Anatomy and innervation
<<    
     Article
      >>
Effects of VDO on Masticatory Musctes/TMJ
patterns of cat lateral gastrocnemius and plantaris mus-
cles. Am J Anat 164:67-77, 1982
17. English AW, Carrasco DI, Widmer CG: Torques pro-
duced by different compartments of the rabbit masseter
muscle. J Appl Biomech 15:348-360, 1999
18. Mosolov NN: On the anatomy of human masticatory
musculature, in Schumacher GH (ed): Morphology of
the Maxillo-Mandibular Apparatus. Leipzig, VEB Georg
Thieme, 1972, pp 65-69
19. Bui AT, Widmer CG. Anatomical architecture of the
human masseter muscle. J Dent Res 75:242, 1996 (abstr)
20. Bui AT, Widmer CG: Innervation patterns of anatomical
partitions within the human masseter muscle. J Dent Res
76:29, 1997 (abstr)
21. Wineski LE, Herring SW: Innervation and function in
the masseter complex of the pig. Fortschr Zool 30:285-
287, 1985
22. Widmer CG, English AW, Carrasco DI, et al: Modeling
rabbit temporomandibular joint torques during a power
stroke. Angle Orthod (in press)
23. Nordstrom SH, Yemm R: The relationship between jaw
position and isometric active tension produced by direct
stimulation of the rat masseter muscle. Arch Oral Biol
19:353-359, 1974
24. Goto TK, Langenbach GE, Hannam AG: Length
changes in the human masseter muscle after jaw move-
ment. Anat Rec 262:293-300, 2001
25. Weijs WA, van der Wielen-Drent TK: Sarcomere length
and EMG activity in some jaw muscles of the rabbit. Acta
Anat (Basel) 113:178-188, 1982
26. Rowlerson A, Pope B, Murray J, et al: A novel myosin
present in cat jaw-closing muscles. J Muse Res Cell Motil
2:415-438, 1981
27. Briggs MM, Schachat F: Early specialization of the super-
fast myosin in extraocular and laryngeal muscles. J Exp
Biol 203:2485-2494, 2000
28. Shiotani A, Flint PW: Expression of extraocular-super-
fast-myosin heavy chain in rat laryngeal muscles. Neuro-
report 9:3639-3642, 1998
29. Thornell LE, Billeter R, Eriksson PO, et al: Heteroge-
neous distribution of myosin in human masticatory mus-
cle fibers as shown by immunocytochemistry. Arch Oral
Biol 29:1-5, 1984
30. English AW, Eason J, Schwartz G, et al: Sexual dimor-
phism in the rabbit masseter muscle: Myosin heavy chain
composition of neuromuscular compartments. Cells Tis-
sues Organs 164:179-191, 1999
31. Anapol F, Herring SW: Ontogeny of histochemical fiber
types and muscle function in the masseter muscle of
miniature swine. AmJ Phys Anthropol 112:595-613, 2000
Home | TOC |          
Index
161
32. Eason JM, Schwartz G, Shirley KA, et al: Investigation of
sexual dimorphism in the rabbit masseter muscle show-
ing different effects of androgen deprivation in adult
and young adult animals. Arch Oral Biol 45:683-690,
2000
33. Baldwin KM, Haddad F: Effects of different activity and
inactivity paradigms on myosin heavy chain gene expres-
sion in striated muscle. J Appl Physiol 90:345-357, 2001
34. Schuler M, Pette D: Fiber transformation and replace-
ment in low-frequency stimulated rabbit fast-twitch mus-
cles. Cell Tissue Res 285:297-303, 1996
35. Mayo KH, Ellis E, III, Carlson DS: Histochemical char-
acteristics of masseter and temporalis muscles after 5
weeks of maxillomandibular fixation—An investigation
in Macaca mulatta. Oral Surg Oral Med Oral Pathol
66:421-426, 1988
36. Ohnuki Y, Saeki Y, Yamane A, et al: Quantitative changes
in the mRNA for contractile proteins and metabolic
enzymes in masseter muscle of bite-opened rats. Arch
Oral Biol 45:1025-1032, 2000
37. Gedränge T, Luck O, Hesske G, et al: Differential ex-
pression of myosin heavy-chain mRNA in muscles of
mastication during functional advancement of the man-
dible in pigs. Arch Oral Biol 46:215-220, 2001
38. McKoy G, Ashley W, Mander J, et al: Expression of
insulin growth factor-1 splice variants and structural
genes in rabbit skeletal muscle induced by stretch and
stimulation. J Physiol 516:583-592, 1999
39. Owino V, Yang SY, Goldspink G: Age-related loss of
skeletal muscle function and the inability to express the
autocrine form of insulin-like growth factor-1 (MGF) in
response to mechanical overload. FEBS Lett 505:259-
263, 2001
40. Clark GT: A critical evaluation of orthopedic interocclu-
sal appliance therapy: Effectiveness for specific symp-
toms. J Am Dent Assoc 108:364-368, 1984
41. Pierce CJ, Gale EN: A comparison of different treat-
ments for nocturnal bruxism. J Dent Res 67:597-601,
1988
42. Sim Y, Carlson DS, McNamara JA Jr: Condylar adapta-
tion after alteration of vertical dimension in adult rhesus
monkeys, Macaca mulatta. Cranio 13:182-187, 1995
43. Dolce C, Van Sickels JE, Bays RA, et al: Skeletal stability
after mandibular advancement with rigid versus wire
fixation. J Oral Maxillofac Surg 58:1219-1227, 2000
44. Dolce C, Hatch JP, Van Sickels JE, et al: Rigid versus wire
fixation for mandibular advancement: Five-year skeletal
and dental changes. Am J Orthod Dentofac Orthop (in
press)
<<    
     Article
      >> Home | TOC |          
Index

Long-Term Stability of Anterior Open-Bite

Therapy: A Review
Greg J. Huang

This article reviews the orthodontic literature with respect to long-term

stability after orthodontic or combined orthodontic-surgical treatment of
anterior open bite. Efforts were made to review the studies in a manner that
allowed comparison of treatment success as well as the subsequent stabil-
ity. The existing literature suggests that approximately 80% of anterior
open-bite subjects will have positive overlap at the latest follow-up,
whether they undergo only orthodontic therapy or a combination of orth-
odontic-surgical therapy. However, orthodontic therapies appear to have
slightly lower treatment success but better stability than surgical therapy. In
other words, fewer subjects achieve positive incisor overlap with orthodon-
tic therapy alone, but almost all that do maintain it. The level of evidence
provided by the current literature is not conclusive because many surgical
and nonsurgical open-bite studies are characterized by small samples and
the potential for selection bias. Thus, well-designed studies are necessary
to improve our knowledge of the etiology, therapies, and stability of ante-
rior open bite. (Semin Orthod 2002;8:162-172.) Copyright 2002, Elsevier
Science (USA). All rights reserved.

A ll orthodontists have been faced with the

difficult task of treating patients with ante-
rior open-bite and the subsequent challenge of
retention. The orthodontic literature has nu-
merous case reports and studies reporting good
results at the end of treatment. These are help-
ful, but the success of open-bite therapy is ulti-
mately measured by long-term stability. Thus,
the purpose of this article is three-fold: (1) to
identify and review studies that address the long-
term stability of anterior open-bite therapy, (2)
to compare the long-term stability reported by
the nonsurgical and surgical studies, and (3) to
critically assess the level of evidence presented
by these studies with suggestions for future re-
search.
To identify studies, PubMed was searched by
using combinations of the key words stability,
From the Department of Orthodontics, University of Washington,
Seattle, WA.
Address reprint requests to GregJ. Huang, DMD, MSD, MPH,
Box 357446, University of Washington, Seattle, WA 98195-7446.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0803-0008$35.00/0
doi:10.1053/sodo.2002.125436
orthodontics, and either openbite, open-bite, or
open bite. Additionally, reference lists from the
open-bite literature were searched to identify
citations not retrieved by PubMed. Of the hun-
dreds of open-bite articles published in the En-
glish language, 21 were identified that specifi-
cally investigated stability. Among these, six
investigated orthodontic therapy, whereas 15 in-
vestigated combined orthodontic and surgical
therapy. Studies that did not specifically address
anterior open bite were excluded, as were those
that had less than five subjects.
Studies Assessing Orthodontic
Therapies
One of the earliest orthodontic studies was a
retrospective case series, comparing a sample of
13 relapsed open-bite subjects with a sample of
13 relapsed deep-bite subjects.1 All subjects had
fixed appliances, and overbite was measured as
the distance between the upper and lower inci-
sor tips perpendicular to the occlusal plane. In
this study, open-bite relapse was associated with

162 Seminars in Orthodontics, Vol 8, No 3 (September), 2002: pp 162-172

<<    
     Article
      >>
Anterior Open-Bite Therapy
an increase in posterior maxillary facial height,
resulting in downward rotation of the mandible.
Fränkel and Fränkel2 investigated a func-
tional approach to the treatment of skeletal
open bite.2 (This study did not specifically ad-
dress stability, but because it followed subjects
for approximately 8 years, it was included.) In a
cohort study, 30 subjects treated with functional
regulators and lip-seal exercises were compared
with 11 untreated subjects. Subjects were se-
lected based on a hyperdivergent skeletal pat-
tern, as well as clinical observation of lip incom-
petence and postural weakness of the orofacial
muscles. The treated subjects were followed
from a mean age of 7 until a mean age of 15,
with the last follow-up at least 4 years postre-
tention. The corresponding means for the un-
treated group were 8 and 16 years. The treated
subjects had significant closure of the mandibu-
lar plane and gonial angles during the study
period. However, the mean pretreatment char-
acteristics of the two groups were not reported
so it is difficult to assess whether the treated and
untreated subjects were comparable before the
start of the study. Additionally, although the
normalization of the skeletal characteristics in
the treated subjects was impressive, no overbite
values were reported, making it difficult to assess
whether the open bites were successfully closed
and stable.
Lopez-Gavito et al3 evaluated stability of orth-
odontic treatment (fixed appliances and head-
gear) in subjects at least 9 years and 6 months
after the discontinuation of retention devices.
Subjects were selected from the pös tre ten tion
files of the University of Washington based on
the availability of records. In this retrospective
case series, Lopez-Gavito et al reported that 35%
of their sample relapsed, as defined by a cepha-
lometric distance >3 mm from the tip of the
lower incisor to the nearest hard tissue (tooth or
palate) measured along the long axis of the
lower incisor. Although these results suggest
a high relapse rate, anteroposterior skeletal
change may have influenced this distance in the
absence of vertical changes. In fact, a subsequent
master's thesis, using this same sample but a
different method of measuring overbite (dis-
tance between upper and lower incisor tips mea-
sured relative to the nasion-menton line), found
that all of these subjects had positive overlap at
the postretention period, although 60% of these
Home | TOC |          
Index
163
subjects did not have incisal contact.4 Thus, the
relapse rate varied considerably depending on
the method chosen to measure overbite.
Another study of orthodontic interventions
assessed the impact of tongue cribs.5 In this
study, subjects were selected based on a negative
pretreatment overbite; the use of a tongue crib;
and the availability of records at the pretreat-
ment, posttreatment, and at least 1-year post-
treatment times. This retrospective case series
identified 26 adolescent (growing) patients and
7 adult (nongrowing) patients. The mean pre-
treatment overbite, which was measured relative
to the nasion-menton line, was —2.8. The overall
success rate for achieving a positive overbite was
88%, and of these successful subjects, none re-
lapsed to negative overbite during the posttreat-
ment period. Although the stability of the suc-
cessfully treated cases was impressive, a criticism
of this study was the potential for selection bias.
In a retrospective case series, Katsaros and
Berg6 assessed the stability of 20 subjects treated
with fixed (n = 19) or functional (n = 1) appli-
ances at least one year after discontinuation of
retention. Selection of the sample was based on
the existence of a pretreatment open bite, as-
sessed on plaster casts, and the availability of
records at the pretreatment, posttreatment, and
postretention periods. Cephalometric measure-
ments were also taken, with overbite measured
perpendicular to the nasion-menton line. This
sample exhibited typical open-bite characteris-
tics, with a mean pretreatment openbite of —1.9
mm. Success, based on incisal contact measured
on plaster casts in habitual occlusion or in a
forward position, was achieved in 15 of the 20
subjects (75%). None of the successfully treated
subjects had negative overbite at the postreten-
tion period. Again, the stability of successful
cases is impressive, but limitations include the
potential for selection bias, as well as a definition
of success that depends on the articulation of
models rather than a cephalometric measure-
ment.
Kim et al7 reported that the multiloop edge-
wise arch-wire technique is a stable method for
orthodontic correction of open bites. In a recent
study, 55 open-bite subjects treated with the mul-
tiloop edgewise arch-wire technique technique
were identified, with 27 of these having final
records at the 2-year follow-up period. The in-
vestigators divided their sample into growing
<<    
     Article
      >>
164 Greg J. Huang
and nongrowing groups. Among the growing
subjects with complete records (n = 17), all
achieved a positive overbite at the end of treat-
ment, and one (6%) relapsed to negative over-
bite at the 2-year follow-up. Among the non-
growing subjects with complete records (n =
10), all achieved a positive overbite at the end of
treatment, and one (10%) relapsed to negative
overbite at the 2-year follow-up. Thus, the overall
relapse rate was 7% (2/27). These results are
encouraging but could be influenced by selec-
tion bias and a considerable loss to follow-up.
We can learn several things from these studies
(Table 1). First, there are relatively few studies
that evaluate the long-term stability of orthodon-
tic therapies. Second, the sample sizes for these
studies are not large. Third, definitions for open
bite differ, as well as methods to assess success
and stability. In an effort to compare these stud-
ies, the following conventions were used. Sub-
jects were required to have an open bite (specif-
ically, no incisal overlap) before treatment.
Treatment success was defined as the achieve-
ment of positive overlap at the end of therapy,
and stability was defined as the maintenance of
positive overlap at the last follow-up period. The
overall success rate was the product of the per-
centage of treatment success and the percentage
of stable cases at the last follow-up. For example,
if a study reported 80% treatment success and
90% stability, the overall success would be 72%.
By using these conventions, the overall success
of orthodontic therapies is between 75% and
93%, which seems quite good (Table 2). How-
ever, several other factors should also be consid-
ered.
First, were the open bites of these subjects
likely to improve without therapy? Some studies
have reported from 40% to 80% spontaneous
closure of untreated open-bite subjects.8'9 How-
ever, their conclusions were based on cross-sec-
tional data using subjects as young as 6 to 8 years
old. At that age, open-bite malocclusions are
difficult to define because of the normal transi-
tion of incisors. Most orthodontists would prob-
ably agree that open bites persisting into the late
mixed dentition rarely close without interven-
tion. This view is supported by the National
Health and Nutrition Estimates Survey III data,
which reports the prevalence of openbite to be
3.6% in 8-11 year-olds, 3.5% in 12-17 year-olds,
and 3.3% in 18-50 year-olds.10 Additionally,
Home | TOC |          
Index
some believe skeletal open bites are less likely
than dental open bites to spontaneously im-
prove. Thus, given the lack of controls, pretreat-
ment characteristics are useful in assessing the
severity and nature of the open-bite subjects,
which, in turn, may provide information on the
likelihood of spontaneous correction. The open-
bite subjects in the previous studies all had steep
mandibular planes and moderate open bites,
with the exception of Lopez-Gavito et al's sub-
jects. (Although the mean open bite they re-
ported was —5 mm, the mean overbite was 0.3
mm if measured relative to the nasion-menton
line.) Additionally, all of the subjects were prob-
ably in permanent dentition, except for the ad-
olescent crib therapy group. Thus, the majority
of the subjects in the studies mentioned pre-
viously had permanent dentitions, significant
open bites, and skeletal components, and would
not be expected to improve without therapy.
The second issue that should be addressed is
sample selection. Although selecting subjects
based on the availability of records might seem
unbiased, we must keep in mind that having
models and radiographs at various time points
requires some degree of cooperation from pa-
tients and that willingness to participate in these
recall visits may differ among successfully and
unsuccessfully treated patients. Additionally, the
likelihood of recalling successful and unsuccess-
ful cases may differ as well as the effort placed
into recalling these subjects. If successfully
treated subjects are more likely to be recalled by
orthodontists and they are also more likely to
participate in recalls, an overly optimistic esti-
mate of stability will be reported. Thus, every
effort should be made to identify either a ran-
dom sampling of subjects or all consecutively
treated subjects. In either case, it is also useful to
know how many subjects were treated with a
particular therapy and what percentage was re-
called for the study.
In summary, the orthodontic therapies do not
always result in positive overlap of the incisors at
the end of treatment. However, for subjects who
achieve a positive overlap, the stability of the
open-bite correction seems to be quite good.
Although other nonsurgical therapies have been
prescribed, such as myofunctional therapy, pos-
terior bite blocks, magnets, and chin cups, no
studies assessing the long-term stability of those
therapies could be located. As with many orth-
<<    
     Article
      >> Home | TOC |          
Index

Anterior Open-Bite Therapy 165

Table 1. Studies Assessing Orthodontic and Combined Orthodonic/Surgical Therapies

Pre-TXAge Overbite
Year Design Sample (y-mo) Measure Intervention Pre-TX OB Pre-TX MP
Orthodontic studies
Nemeth and 1974 Case Series 13 Relapsed 13-0 (at post- Occlusal FA (+HG?) NR NR
Isaacson1 cases TX) Plane
Fränkel and 1983 Cohort 30 Treated Followed age NR Functional regulator/ NR NR
Fränkel2 7 to 15 Lip exercises
11 Untreated Followed age NR No treatment NR NR
8 to 16
3
Lopez-Gavito et al 1985 Case Series 41 Adolescents 12-6 (median) Long-axis FA + HG -5.1 (.3)* 39.8
MN
incisor
Huang et al5 1990 Case Series 26 Adolescents 9-7 Nasion- Crib (+FA + HG) -2.9 37.4
menton
7 Adults 20-10 Nasion- Crib FA (+HG) -2.7 41.5
menton
Ratsaros and Berg6 1993 Case Series 20 Adolescents 11-8 From casts 19 w/FA, 1 -1.9 39
w/Functional
Kim et al7 2000 Case Series 17 Adolescents 13-5 NR MEAW -2.3 30.7 (to FH)
10 Adults 26-1 NR MEAW -2.2 32.7 (to FH)

Combined orthodontic/surgical studies Pre-op OB Pre-op MP

Turveyetal 11 1976 Case series Range: 14-27 NR MX impaction ( + MN NR NR
surgery + FA)
Denison et al12 1989 Cohort 28 openbite 24-0 (median) Nasion- MX impaction + FA NR NR
menton
24 overlap 27-0 (median) Nasion- MX impaction + FA NR NR
menton
14 contact 32-2 (median) Nasion- MX impaction + FA NR NR
menton
Reitzik et al13 1990 Consecutive 16 of 20 NR NR Reverse-L osteotomy NR NR
+ FA
Haymond et al14 1991 Case series 38 26-4 Nasion- MX impaction + FA NR NR
menton ( + MN surgery)
McCance et al15 1992 Case series 21 NR NR Bimaxillary surgery + -5.5 37.6 (to PP)
FA
Oliviera and 1997 Case series 10 NR NR BSSO + FA NR 39.7
Bloomquist16
Ermel et al17 1997 Consecutive 20 of 44 26-0 NR MX impaction + FA NR 37.5
( + MN surgery)
Hoppenreijs 1997 Cohort 267 26-6 Occlusal MX impaction (+MN -1.24 37.6 (to PP)
et al18 plane surgery + FA)
Hoppenreijs 1998 Cohort 130 23.1 Occlusal MX impaction (+MN NR NR
et al18 plane surgery + FA)
Lo and Shapiro20 1998 Cohort 19 extrusion 19-2 Nasion- MX impaction + FA -1.7 NR
menton ( + MN surgery)
21 no 26-7 Nasion- MX impaction + FA -2 NR
extrusion menton ( + MN surgery)
Arpornmaeklong 2000 Case series 17 21-4 Occlusal MX impaction + FA -1.7 35.5 (to FH)
and Heegie21 plane
10 22-0 Occlusal Bimaxillary surgery + -1.4 33.8 (to FH)
plane FA
22
Proffit et al 2000 Case series 28 21.8 True MX impaction + FA >2 mm open NR
vertical
26 24.5 True Bimaxillary surgery + >2 mm open NR
vertical FA
23
Moldez et al 2000 Cohort 13 NR Occlusal MX impaction + MN -1.7 46.7
plane surgery + FA
10 NR Occlusal MX rotation + MN -3.2 44.4
plane surgery + FA
11 NR Occlusal Bimaxillary surgery + -0.1 38.8
plane FA
24
Fischer et al 2000 Consecutive 58 of 58 23-0 at surgery NR Bimaxillary surgery + -0.9 46.2
FA
25
Swinnen et al 2001 Consecutive 38 of 38 -20 Relative to LeFort I intrusion -0.6 NR
S'-N ( + MN surgery +
FA)
11 of 11 -20 Relative to LeFort I extrusion -1.9 NR
S'-N ( + MN surgery +
FA)

NOTE. Interventions in parentheses were performed for some subjects.

Abbreviations: Pre-TX, pretreatment; OB, overbite; MP, mandibular plane relative to sella-nasion; FA, fixed appliances; HG,
headgear; FH, Frankfort horizontal; MX, maxillary; MN, mandibular; PP, palatal plane; S'-N, sella-nasion construct; NR, not
reported; pre-op, preoperative.
*If measured relative to nasion-menton.
<<    
     Article
      >> Home | TOC |          
Index

166 Greg J. Huang

Table 2. Assessment of Treatment Success, Stability, and Overall Success

TX Overall
Mean Follow-up Follow-up Success Stable Success
Study (y-m) Range Fixation (%) (%) (%) Open-Bite Confirmation

Orthodontic studies
Lopez-Gavito 11-6 post- 9-6 to 18-0 NA 93 100 93 15 subjects, no pre-TX overlap
etal 3 retention* on cephs
5
Huang et al 5-3 post-ortho 1-0 to 14-0 NA 88 100 88 33 subjects, no pre-TX overlap
on cephs
Katsaros and >1 year post- 1-0 to 19-6 NA 75 100 75 20 subjects, no pre-TX overlap
Berg6 retention on casts
Kim et al7 2-8 post-ortho 2-0 to ? NA 100 93 93 27 subjects, no pre-TX overlap
on cephs
Combined orthodontic/surgical studies (TX success assumed to be 100% for these studies)
Denison 3-0 post-ortho* 1-0 to 11-9 NR 100 79 79 28 subjects, no pre-TX overlap
etal 12 on cephs
13
Reitzik et al >1 year post-op 1-0 to 7-2 Rigid 100 100 100 16 subjects, pre-TX openbite
not reported
Raymond >1 year post-op 1-0 to 5-0 Rigid 100 84 84 38 subjects, no pre-op overlap
14
et al on cephs
Ermel et al17 7-3 post- 1-0 to 15-0 Rigid 100 88 88 20 subjects, pre-TX openbite
retention not reported
Hoppenreijs 5-9 post-op 1-8 to 17-6 Mixed 100 81 81 267 subjects (168 confirmed
etal18 with no pre-TX overlap on
cephs)
Lo and 5-10 post-ortho 2-0 to ? NR 100 75 75 40 subjects, no pre-TX overlap
Shapiro20 on cephs
Proffit et alf22 5-4 post-op 2-6 to 13-7 Mixed 100 88 88 54 subjects with >2 mm pre-TX
openbite
Moldez et al23 >5 year post- 5-0 to ? Mixed 100 91 91 23 subjects, clinical diagnoses
ortho of pre-TX openbite
Fischer >2 year post-op 2-0 to ? Mixed 100 71 71 58 subjects, clinical diagnoses
24
et alt of pre-TX openbite
Swinnen 1-0 post-op 1-0 to 1-2 Rigid 100 89 89 49 subjects, no pre-TX overlap
et al25 on cephs
Abbreviations: NA, not applicable; NR, not reported.
^Median values.
fStable defined as <2 mm decrease in overbite.
^Reported range of pre-TX openbite from —7.9 to +5.8.

odontic studies, the limitations of the studies
reviewed earlier are small sample sizes, selection
bias, loss to follow-up, and a lack of untreated
controls.
Studies Assessing Combined
Orthodontic-Surgical Therapies
Surprisingly, there are considerably more stud-
ies assessing stability of open bite with combined
orthodontic-surgical therapy. One of the earliest
studies of stability after surgical correction of
open bites focused on tongue function and
speech changes.11 This retrospective case series
classified nine subjects into simple (canine to
canine), compound (premolar to premolar),
and infantile (extending posterior to premolars)
open bites, and a variety of maxillary and man-
dibular surgeries were performed. They re-
ported that six of nine subjects had excellent
stability (no appreciable dental or skeletal move-
ment was detected), two had fair stability (de-
tectable movement but no effect on the cor-
rected overbite), and one had poor stability (a
reopening of the bite). These results are difficult
to evaluate because the definitions of open-bite
stability are not quantitatively defined.
Denison et al12 used a cohort design to com-
pare vertical stability in 66 subjects who had
maxillary LeFort I osteotomies. These subjects
were selected from the practices of various fac-
ulty members of the University of Washington
Department of Orthodontics, Seattle, WA. Over-
bite was measured relative to the nasion-menton
line, and records were available at pre treatment,
posttreatment, and at least 1 year posttreatment.
The subjects were divided into three groups: (1)
an open-bite group (no incisal overlap, n = 28),
<<    
     Article
      >>
Anterior Open-Bite Therapy
(2) an incisal overlap group (positive overbite
but no contact of incisors, as assessed on the
cephalometric headfilm (ceph), n = 24), and
(3) an incisal contact group (as assessed on the
ceph, n = 14). All subjects had maxillary supe-
rior repositioning surgery and were followed for
a minimum of 1 year after the completion of
orthodontic treatment. Although the incisal
contact and incisal overlap subjects did not show
any significant changes in overbite during the
posttreatment period, 43% of the open-bite sub-
jects had significant increases in facial height,
eruption of maxillary molars, and decreases in
overbite during the posttreatment phase. In
21% of the open-bite subjects, the overbite re-
lapsed to a negative value.
Reitzik et al13 reported on a mandibular pro-
cedure for open-bite correction—the reverse-L
osteotomy with rigid fixation. Of 20 consecu-
tively treated cases, the records of 16 were avail-
able at the preoperative, postoperative, and >1-
year postoperative time periods. The surgical
procedure resulted in a mean closure of 9° for
the gonial angle and 4.3° for the mandibular
plane angle. At 1 year postoperative, the gonial
angle had relapsed 0.3° and the mandibular
plane angle 0.6°. The investigators also reported
that incisal contact had been maintained in all
subjects, but overbite measurements were not
reported. Based on these results, the investiga-
tors concluded that the reverse-L procedure
with rigid fixation is a stable method for open-
bite correction. Their results are impressive, but
the short follow-up time may have allowed some
subjects to be just completing orthodontics. Ad-
ditionally, cephalometric characteristics of the
sample were not described at the pretreatment
time.
Haymond et al14 investigated the stability of
open-bite correction using rigid fixation for
maxillary procedures alone (n = 15), maxillary
procedures in combination with mandibular ad-
vancement (n = 16) or set-back (n — 4), and
isolated mandibular set-back (n = 3). In this
case series, records were available at the preop-
erative, postoperative, and 1 to 5 year postoper-
ative periods; 6 of the 38 subjects (16%) had
negative incisal overlap at the latest follow-up.
The investigators reported that dentoalveolar
changes were primarily responsible for relapse,
especially in 3 cases that were expanded orth-
odontically. They did not note relapse in sub-
Home | TOC |          
Index
167
jects with surgically assisted or surgical expan-
sion. From this data, they concluded that
surgical correction of openbite using rigid fixa-
tion is relatively stable, and that orthodontic
expansion may contribute to relapse in the ver-
tical dimension. As in many of the surgical stud-
ies, pretreatment cephalometric characteristics
were not described, and the earliest follow-up
period may have occurred before removal of
orthodontic appliances.
McCance et al15 reported on 21 high-angle
subjects who underwent fixed appliances and
bimaxillary surgery for open-bite correction.
Rigid fixation was used in the maxilla, with wire
fixation in the mandible, and intermaxillary fix-
ation. Records were available at the preopera-
tive, postoperative, and 1-year postoperative
times. He reported variable postoperative move-
ments of the maxilla and mandible. In the Class
III subjects (n = 11), overbite decreased form
— 6 mm to 3.1 mm postoperatively but relapsed
to 2.4 at the follow-up time. The overbite values
for the Class II subjects (n — 10) were —4.6,
— 1.6, and —1.6, respectively. The results of this
study are difficult to interpret because the
method for measuring overbite was not de-
scribed nor was the actual number of subjects
who achieved or maintained positive overlap re-
ported. Additionally, the follow-up time was
short.
Oliveira and Bloomquist16 investigated the
stability of bilateral sagittal split osteotomy
(BSSO) with rigid fixation in 10 subjects who
had records from the preoperative, postopera-
tive, and > 1-year postoperative time points (all
subjects completed orthodontic treatment be-
fore the last follow-up). The mandibular plane
angle was evaluated as an indicator of surgical
stability, with a mean operative change of 3.9° of
closure followed by a mean increase of 1.3° dur-
ing the postoperative period. Two subjects had
mandibular plane angles at the latest follow-up
that were greater than the preoperative value.
However, all 10 subjects had positive incisal over-
lap at the latest follow-up. Although these results
are encouraging for correcting open bites with
isolated mandibular surgery, pretreatment char-
acteristics were not reported, and the minimum
follow-up period was quite short.
Ermel et al17 reported on the Schuchardt
technique for open-bite correction, which is a
segmented maxillary osteotomy involving impac-
<<    
     Article
      >>
168 Greg J. Huang
tion of the buccal segments. Of 44 subjects who
underwent this procedure at the University of
Tubingen (Germany) from 1980 to 1994, 26
subjects agreed to participate, but only 20 had
complete records at the preoperative, postoper-
ative, and long-term follow-up period (1-15 years
postretention). Fifteen of the 26 subjects also
had BSSO, and in all subjects, rigid fixation was
used. At the last follow-up visit, 3 of the 26
subjects (12 %) exhibited a negative overbite,
and the investigators concluded that the Schu-
chardt technique is a relatively stable method to
address anterior open bite when the maxillary
anterior teeth are in good relationship with the
upper lip. Pre treatment overbite was not re-
ported, although the mean mandibular plane
angle was 39°.
In the largest study of surgically treated ante-
rior open bite, Hoppenreijs et al18 reported on a
total of 267 subjects in a multicenter study.18
Rigid fixation was used in 114 subjects, and wire
fixation was used in 153 subjects. In 144 subjects,
only a LeFort I was performed, with the remain-
ing 123 subjects receiving a combined LeFort I
with mandibular BSSO. Although records were
collected at the pretreatment (Tl), preoperative
(T2), postoperative (T3), 6 to 19 months post-
operative (T4), and 20 to 210 months postoper-
ative periods (T5), not all subjects had complete
records from all time points. All subjects that
had Tl cephalometric characteristics had no in-
cisal overlap at that time, as measured relative to
the occlusal plane, and at T5, all subjects were at
least 1 year out of orthodontic appliances. In
comparing the differing surgical procedures and
methods of fixation, the investigators concluded
that the maxilla exhibited good vertical stability
overall, and that the mandible exhibited better
stability when rigid fixation was used. Their over-
all relapse rate was 19% when incisal overlap was
used as the criterion for success at T5. Again,
pretreatment cephalometric characteristics were
not reported.
Hoppenreijs et al19 also reported on trans-
verse stability after LeFort I surgery with and
without BSSO by using methodology similar to
the previous study mentioned. They compared
77 subjects with orthodontic expansion and one
piece LeFort I osteotomy with 53 subjects who
underwent multisegment LeFort I osteotomy
and found no difference in the transverse stabil-
ity of the two groups at the latest follow-up time,
Home | TOC |          
Index
with only 20% of the subjects in either group
exhibiting good transverse stability. For multi-
segment LeFort I procedures, the stability was
better with rigid fixation than with wires. Al-
though the transverse relapse was associated
with clockwise rotation of the mandible, it was
not well correlated with changes in overbite,
suggesting that compensatory eruption oc-
curred in some subjects.
In a study investigating maxillary incisor ex-
trusion and vertical relapse in surgically treated
open bites, Lo and Shapiro20 collected records
for 40 subjects who had maxillary LeFort I sur-
gery, with some also undergoing mandibular
surgical procedures. All subjects had records
from the pretreatment, preoperative, postoper-
ative, and >2-year posttreatment time periods.
These subjects were divided into a group that
underwent presurgical extrusion of the maxil-
lary incisors (n = 19) and a group that did not
(n = 21). The mean pretreatment open bite for
these groups was —1.7 and —2.0, respectively.
The investigators did not find a significant dif-
ference in the amount of relapse because both
groups had 5 subjects with no incisal overlap at
the final follow-up time. Thus, presurgical extru-
sion of the incisors did not appear to affect
stability of open-bite correction in these subjects,
and the overall rate of relapse was 25%.
Arpornmaeklong and Heggie21 compared 17
subjects undergoing isolated maxillary superior
repositioning with 20 subjects who underwent
bimaxillary surgery with rigid fixation. Records
were available at the preoperative, postopera-
tive, and > 1-year postoperative times. Good
maxillary stability was reported in both groups.
In the bimaxillary surgery group, 3 subjects
(15%) exhibited considerable anteroposterior
relapse of mandibular advancement. The mean
overbite for the LeFort I group was 0.9 at the
final follow-up and 0.7 for the bimaxillary
surgery group. Unfortunately, the number of
subjects with vertical relapse was not reported.
Additionally, pretreatment cephalometric char-
acteristics were not reported, and the final fol-
low-up period was relatively short.
Proffit22 reported on the long-term stability of
all open-bite subjects in the University of North
Carolina database who had maxillary osteoto-
mies (n = 28) or bimaxillary osteotomies (n =
26), along with complete records. All subjects
had at least 2-mm open bite at the pretreatment
<<    
     Article
      >>
Anterior Open-Bite Therapy
time, and records were available at the preoper-
ative, postoperative, 1 year postoperative, and at
least 3 years postoperative time periods. He
found that 12% of the subjects had decreases of
2 to 4 mm in overbite during the follow-up
period. Although facial height increased in
more than 35% of the subjects, this was usually
accompanied with incisor eruption, which main-
tained the overbite relationship. The actual
overbite measurements at various time points
are not reported, and therefore it is difficult to
compare this study with results from other stud-
ies. Also, although relapse <2 mm was consid-
ered stable, it is difficult to know if those subjects
maintained positive incisal overlap.
Moldez et al23 compared a random selection
of 23 Class III open-bite subjects who underwent
maxillary impaction and BSSO to 11 Class III
subjects without open bite but similar surgical
correction. The open bites were further divided
into two groups: one with maxillary impaction
but no rotation (n = 13) and one with clockwise
rotation of the maxilla (n = 10). In this well-
documented study, cephalometric characteris-
tics were reported at the pretreatment (Tl),
preoperative (T2), postoperative (T3), postorth-
odontic (T4), 1-year postorthodontic (T5), and
5-year postorthodontic (T6) time periods. The
pretreatment overbites for the impaction, rota-
tion, and non-open-bite groups were —2.2,
— 3.8, and 2.0, respectively. Moldez et al re-
ported that 2 of 13 subjects (15%) in the max-
illary impaction group exhibited relapse of the
open bite, whereas no subjects from the rotation
group or non-open-bite group displayed verti-
cal relapse. Thus, they concluded that the Le-
Fort I osteotomy with rotation, in which the
vertical dimension was maintained, was more
stable than LeFort I osteotomy with impaction,
in which the mandible rotated in a counter-
clockwise manner. The proposed rationale is the
temporary alteration of freeway space in the im-
paction subjects.
Perhaps the most interesting finding from the
latter study is the change in overbite during the
postoperative periods. From T3 to T4 (postop-
erative to debanding), both the rotation and
impaction group displayed increases in the over-
bite. This could be due to dental movements in
the finishing stages of orthodontics. Many surgi-
cal studies report stability at the 1-year postop-
erative time, which may be very close to the time
Home | TOC |          
Index
169
of orthodontic appliance removal. Therefore,
the overbite measurements at that time may re-
flect orthodontic finishing rather than true sta-
bility. From T4 to T5 (debanding to 1 year pos-
torthodontics), both the rotation group and
impaction group showed vertical relapse. From
the T5 to T6 period (1 to 5 years after orthodon-
tics), the overbite continued to decrease slightly.
Thus, it would seem that the minimal period
required to monitor overbite stability is at least 1
year after the removal of orthodontic appli-
ances.
Fischer et al24 reported on 58 consecutive
subjects who were diagnosed with retrognathia
and open bite and underwent bimaxillary sur-
gery. Records were available at the pretreatment,
preoperative, postoperative, and >2-year post-
operative periods. They found that 17 of the 58
subjects (29%) had no incisal overlap at the final
follow-up. Of these, 8 had multisegment LeFort
I osteotomies. They concluded that maxillary
stability was good but that the mandible tended
to rotate downward. It is interesting that the
relapse rate was so high, especially because the
range of overbite at the pretreatment period was
-7.9 to 5.8, with a mean of -0.8.
In the most recent surgical study, Swinnen et
al25 evaluated 49 consecutively treated open-bite
subjects.25 Of these, 38 had maxillary intrusion,
whereas 11 had maxillary extrusion. Records
were available at the pretreatment, preopera-
tive, postoperative, 20-weeks postoperative, and
1-year postoperative time points, and all subjects
had rigid fixation. Both the intrusion and extru-
sion groups exhibited good maxillary skeletal
stability at the 1-year postoperative period, but
11% of the subjects had no incisal overlap at that
time.
The surgical studies display considerable
variation in methodology. The comparison of
pretreatment characteristics is difficult because
preoperative, rather than the pretreatment,
characteristics are more commonly reported
(Table 1). Although the mean mandibular plane
angles from these studies tended to be steeper
than the nonsurgical studies, the mean open-
bite measurements were not more severe (Table
1). Thus, the surgical subjects were typically
adults with very steep mandibular planes and 1
to 2 mm of open bite. Again, these are subjects
in which spontaneous improvement is unlikely.
Another issue that arises when assessing the
<<    
     Article
      >>
170 Greg J. Huang
surgical studies is the follow-up period. In the
orthodontic studies, all subjects were out of
treatment at least 1 year. In the surgical studies,
more than half use the 1-year postoperative time
as the minimum follow-up. As previously men-
tioned, this time point is likely to coincide with
the removal of orthodontic appliances, and
therefore the overbite would be expected to be
positive. The study by Moldez et al23 shows in-
creased overbite from the postoperative period
to the end of orthodontic treatment and then a
decrease during the first year after debanding.
Therefore, although it may be possible to evalu-
ate surgical stability 1 year postoperatively, over-
bite stability, which reflects both skeletal and
dental movements, cannot truly be assessed until
a sufficient period of time has passed after the
removal of appliances. From the work by Moldez
et al, 12 months would seem to be an appropri-
ate period.
An attempt was made to measure overall suc-
cess for the surgical studies by using the conven-
tions from the nonsurgical studies. Surgical ther-
apy should allow positioning of the maxilla or
mandible to achieve a positive overlap of the
incisors so the treatment success was assumed to
be 100%. With this assumption, the overall suc-
cess, which varied from 71% to 100%, only de-
pended on the subsequent stability (Table 2).
The four surgical studies that documented a
negative overlap at the pretreatment or preop-
erative time and had a minimum follow-up of 1
year postorthodontics (Table 2 in bold) re-
ported a range of overall success from 75% to
88%. This is very similar to the range of 75% to
93% reported for the nonsurgical studies. How-
ever, the nonsurgical studies tended to report
less treatment success and better stability. It is
difficult to predict what the overall success of the
surgical studies would have been if incisor over-
lap were not achieved in all surgical subjects
because the treatment success would have de-
creased, but the stability of subjects who were
successfully treated may have improved. Unfor-
tunately, this assumption was necessary because
most surgical studies did not report the percent-
age of subjects with incisal overlap postopera-
tively or posttreatment.
It is important to critically assess the level of
evidence provided by these studies. The gener-
ally recognized hierarchy of clinical evidence,
from strongest to weakest, is (1) randomized
Home | TOC |          
Index
controlled trials, (2) cohort studies, (3) case/
control studies, (4) case series, and (5) expert
opinion. The small number of nonsurgical stud-
ies is all case series, with relatively small samples
and the potential for selection bias. Thus, we
cannot be sure that the nonsurgical therapies
are truly as successful or as stable as the current
literature suggests. The surgical studies consist
of case series and cohort studies that compare
different therapies or dental characteristics. Al-
though most of the surgical studies also have
small samples and the potential for selection
bias, the large, multicenter study of Hoppenreijs
et al18 suggests that the stability of surgical ther-
apy is approximately 80%. The surgical studies
provide a higher level of evidence than the orth-
odontic studies based on their greater numbers,
the case series that report on consecutively
treated subjects, and the work of Hoppenreijs
et al.18
The lack of untreated controls in open-bite
studies can be justified by the fact that these
subjects usually do not improve without therapy,
especially after the eruption of the permanent
incisors. Therefore, unless a study is assessing an
early intervention, it is probably more important
to compare different types or combinations of
orthodontic and surgical therapies in cohort or
randomized designs. The use of incisal overlap
as a criterion for measuring open bite should
also be discussed. This measurement has the
advantage of being cephalometrically defined. It
also is a good indicator of whether the lack of
incisor contact is due to a true vertical problem
or an anteroposterior discrepancy. Additionally,
two studies that have compared subjects with
incisal contact, incisal overlap, and no incisal
overlap found that the incisal overlap subjects
reacted very similarly to the incisal contact
group, in which minimal vertical relapse oc-
curred.4'12 Finally, from a clinical standpoint, a
lack of incisal overlap is usually associated with
an inability to incise and therefore indicates an
abnormal functional relationship of the incisors.
Skeletal stability of the maxilla tended to be
good with both wire and rigid fixation because it
typically was positioned superiorly. In bimaxil-
lary surgeries, mandibular stability seemed to be
more variable than maxillary stability, and Hop-
penreijs et al18 reported that rigid fixation
seemed to decrease relapse. Two studies re-
ported that poor transverse stability of the max-
<<    
     Article
      >>
Anterior Open-Bite Therapy
illa contributed to vertical relapse.14-19 All of
these findings are generally consistent with the
statements in an article by Proffit26 reviewing the
stability of surgical movements. Relapse of open
bites may also occur because of tongue size or
posture, unfavorable growth patterns, orofacial
musculature, respiratory problems, and dental
movements. Additionally, condylar resorption
after orthognathic surgery has been reported to
be a factor in relapse.27
As stated at the beginning of this article,
open-bite therapy is still a challenge to the orth-
odontic profession. The current literature sug-
gests that 80% of anterior open-bite subjects
will maintain positive overlap after treatment,
whether it is with orthodontics or combined
orthodontics and surgery. However, these find-
ings are based on studies that usually have small
samples and the potential for selection bias. Re-
viewing the existing studies allows us to improve
future investigations of anterior open bite. For
instance, Hoppenreijs et al18 show the feasibility
of large, multicenter studies. If well designed,
these types of studies can substantially increase
sample size, which results in greater statistical
power when comparing different therapies. Sev-
eral of the surgical studies report on consecu-
tively treated or randomly selected subjects,
which minimizes selection bias. Also, the careful
documentation by Moldez et al23 helps to show
the importance of an adequate follow-up period
after all treatment has been completed. To com-
pare the overall success of orthodontic and sur-
gical therapies, standardized information is nec-
essary. This includes complete records from the
pretreatment period to describe the sample (in-
cluding age and gender); consistent and repro-
ducible methods to measure overbite; and care-
ful documentation of therapies with interim, final,
and long-term records. Well-designed studies are
necessary to better understand the effectiveness
and stability of orthodontic and surgical thera-
pies as well as the mechanisms for relapse. This
will provide orthodontists with the evidence nec-
essary to recommend the most appropriate op-
tions to their anterior open-bite patients.
References
1. Nemeth KB, Isaacson RJ: Vertical anterior relapse. Am J
Orthod 651:565-585, 1974
2. Frankel R, Frankel C: A functional approach to treat-
ment of skeletal open bite. Am J Orthod 841:54-68, 1983
Home | TOC |          
Index
171
3. Lopez-Gavito G, Wallen TR, Little RM, et al: Anterior
open-bite malocclusion: A longitudinal 10-year postre-
tention evaluation of orthodontically treated patients.
Am J Orthod 871:175-186, 1985
4. Zuroff JP: Orthodontic treatment of anterior openbite
malocclusion: Stability ten-years post-retention. Master's
thesis, University of Washington, Seattle, WA, 1990
5. Huang GJ, Justus R, Kennedy DB, et al: Stability of
anterior openbite treated with crib therapy. Angle
Orthod 601:17-24, 1990; discussion 25-26
6. Katsaros C, Berg R: Anterior open bite malocclusion: A
follow-up study of orthodontic treatment effects. Eur
J Orthod 151:273-280, 1993
7. Kim YH, Han UK, Lim DD, et al: Stability of anterior
openbite correction with multiloop edgewise archwire
therapy: A cephalometric follow-up study. Am J Orthod
Dentofacial Orthop 1181:43-54, 2000
8. Kantorowicz A, Korhaus G: The self-correction of orth-
odontic anomalies. Tr First Intl Orthod 40, 1926
9. Worms FW, Meskin LH, Isaacson RJ: Open-bite. Am J
Orthod 591:589-595, 1971
10. Proffit WR, Fields HW Jr, Moray LJ: Prevalence of mal-
occlusion and orthodontic treatment need in the United
States: Estimates from the NHANES III survey. Int J
Adult Orthodon Orthog Surg 131:97-106, 1998
11. Turvey TA, Journot V, Epker BN: Correction of anterior
open bite deformity: A study of tongue function, speech
changes, and stability. J Maxillofac Surg 41:93-101, 1976
12. Denison TF, Kokich VG, Shapiro PA: Stability of maxil-
lary surgery in openbite versus nonopenbite malocclu-
sions. Angle Orthod 591:5-10, 1989
13. Reitzik M, Barer PG, Wainwright WM, et al: The surgical
treatment of skeletal anterior open-bite deformities with
rigid internal fixation in the mandible. Am J Orthod
Dentofacial Orthop 971:52-57, 1990
14. Haymond CS, Stoelinga PJ, Blijdorp PA, et al: Surgical
orthodontic treatment of anterior skeletal open bite
using small plate internal fixation. One to five year
follow-up. Int J Oral Maxillofac Surg 201:223-227, 1991
15. McCance AM, Moss JP, James DR: Stability of surgical
correction of patients with skeletal III and skeletal II
anterior open bite, with increased maxillary mandibular
planes angle. Eur J Orthod 141:198-206, 1992
16. Oliveira JA, Bloomquist DS: The stability of the use of
bilateral sagittal split osteotomy in the closure of ante-
rior open bite. Int J Adult Orthodon Orthognath Surg
121:101-108, 1997
17. Ermel T, Hoffmann J, Alfter G, et al: Long-term stability
of treatment results after upper jaw segmented osteot-
omy according to Schuchardt for correction of anterior
open bite. J Orofac Orthop 601:236-245, 1999
18. Hoppenreijs TJ, Freihofer HP, Stoelinga PJ, et al: Skel-
etal and den to-alveolar stability of Le Fort I intrusion
osteotomies and bimaxillary osteotomies in anterior
open bite deformities. A retrospective three-centre
study. Int J Oral Maxillofac Surg 261:161-175, 1997
19. Hoppenreijs TJ, van der Linden FP, Freihofer HP, et al:
Stability of transverse maxillary dental arch dimensions
following orthodontic-surgical correction of anterior
open bites. Int J Adult Orthodon Orthog Surg 131:7-22,
1998
<<    
     Article
      >>
172 Greg J. Huang
20. Lo FM, Shapiro PA: Effect of presurgical incisor extru-
sion on stability of anterior open bite malocclusion
treated with orthognathic surgery. Int J Adult Orthodon
Orthog Surg 131:23-34, 1998
21. Arpornmaeklong P, Heggie AA: Anterior open-bite mal-
occlusion: Stability of maxillary repositioning using rigid
internal fixation. Aust Orthod J 161:69-81, 2000
22. Proffit WR, Bailey LJ, Phillips C, et al: Long-term stability
of surgical open-bite correction by Le Fort I osteotomy.
Angle Orthod 701:112-117, 2000
23. Moldez MA, Sugawara J, Umemori M, et al: Long-term
dentofacial stability after bimaxillary surgery in skeletal
Class III open bite patients. Int J Adult Orthodon Or-
thog Surg 151:309-319, 2000
24. Fischer K, von Konow L, Brattstrom V: Open bite:
Home | TOC |          
Index
Stability after bimaxillary surgery—2-year treatment
outcomes in 58 patients. Eur J Orthod 221:711-718,
2000
25. Swinnen K, Politis C, Willems G, et al: Skeletal and
dento-alveolar stability after surgical-orthodontic treat-
ment of anterior open bite: A retrospective study. Eur
J Orthod 231:547-557, 2001
26. Proffit WR, Turvey TA, Phillips C: Orthognathic surgery:
A hierarchy of stability. Int J Adult Orthodon Orthog-
nath Surg 111:191-204, 1996
27. Hoppenreijs TJ, Freihofer HP, Stoelinga PJ, et al: Con-
dylar remodelling and resorption after Le Fort I and
bimaxillary osteotomies in patients with anterior open
bite. A clinical and radiological study. Int J Oral Maxil-
lofac Surg 271:81-91, 1998
<<    
     Article
      >> Home | TOC |          
Index

Surgical Modification of Long-Face Problems

L'Tanya J. Bailey, William R. Proffit, George H. Blakey, and
David M. Sarver

Long-face individuals make up approximately one fourth of the dentofacial

deformity group who seek surgical treatment, almost 20% more than would
have been expected from the presumed population incidence. Just fewer
than 25% of patients who undergo surgical treatment are in the long-face
group, and this group is slightly more likely to accept and receive surgical
treatment than their proportion of the surgical population. When the sever-
ity of vertical deformity is so great that reasonable correction cannot be
obtained by growth modification or camouflage, the combination of orth-
odontics and orthognathic surgery may provide the only viable treatment
option. Because many long-face patients will also present with an antero-
posterior problem involving prominent incisors and a receding chin, it is
often easy to dwell on the Class II aspect of the problem and ignore the
vertical contribution. Should a camouflage treatment plan be implemented
by extracting maxillary first premolars and retracting the maxillary incisors,
quite unacceptable esthetics may result because the nasolabial angle will
increase. The option of vertical maxillary impaction now allows the orth-
odontist to avoid the negative facial esthetics that were once considered an
inevitable consequence in improving the dental occlusion. Careful planning
involving both the orthodontist and the oral maxillofacial surgeon now
provides patients with an option that results in both desirable esthetics and
good occlusion. (Semin Orthod 2002;8:173-183.) Copyright 2002, Elsevier
Science (USA). All rights reserved.

When the severity of vertical deformity is so
great that reasonable correction cannot
be obtained by growth modification or camou-
flage, the combination of orthodontics and or-
thognathic surgery may provide the only viable
treatment option. Epidemiologie data suggest
that there are approximately 1.8 million people
in the United States with dentofacial dispropor-
tions too severe to be corrected by orthodontics
alone.1'2 It seems likely that demand for treat-
From the Department of Orthodontics and the Department of
Oral and Maxillofacial Surgery, School of Dentistry, University of
North Carolina, Chapel Hill, NC; and a private practice, Birming-
ham, AL.
Address correspondence to L 'Tanya J. Bailey, DDS, MS, Depart-
ment of Orthodontics, School of Dentistry, University of North Caro-
lina, Chapel Hill, NC 27599-7450.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0803-0009$35.00/0
dot: 10.1053/sodo.2002.125437
ment would be greater for patients with partic-
ular characteristics. A survey of some 1,000 pa-
tients who sought evaluation in the Dentofacial
Clinic at the University of North Carolina,
Chapel Hill, NC, in the 1980s concluded that
individuals with a long face or Class III problem
were more likely to seek and receive surgical-
orthodontic treatment than those who had a
Class II problem.3
Long-face individuals make up approximately
one fourth of the dentofacial deformity group
who seek surgical treatment, almost 20% more
than would have been expected from the pre-
sumed population incidence (Table 1). 4 Of the
2,074 patients evaluated through the University
of North Carolina Dentofacial Clinic from 1979
to 1998, 464 (22.4%) were judged to have exces-
sive face height. Anterior open bite >4mm was
found in about one in six of the long-face group.
Just fewer than 25% of the patients who under-

Seminars in Orthodontics, Vol 8, No 3 (September), 2002: pp 173-183 173

<<    
     Article
      >> Home | TOC |          
Index

174 Bailey et al

Table 1. Skeletal/Dental Characteristics of Dentofacial Population (% of Sample)

1979-1989 1990-1998 Total
(n = 1,202) (%) (n = 872) (%) (n = 2,074) (%)
Skeletal Class
Class I (Number, %) 154 (11.9) 164 (14.2) 318 (12.9)
Class II 718 (60.4) 426 (51.6) 1144 (56.7)
Over] et ^6 mm 38.4 31.7 35.6
Class III 330 (27.7) 282 (34.2) 612 (30.4)
Reverse overjet 11.4 14.9 12.9
Long face
Number 270 (22.5) 194 (22.2) 464 (22.4)
Midface 6.3 7.7 6.9
Lower face 81.5 76.8 79.5
Both 12.2 15.5 13.6
Anterior open bite 16.6 17.0 16.8
Short face
Number (%) 371 (30.9) 262 (30.0) 633 (30.6)
Midface 4.9 11.8 7.8
Lower face 91.1 83.2 87.8
Both 4.0 5.0 4.4
Anterior deepbite 18.5 15.6 17.1

went surgical treatment were in the long-face for approximately 15% to 20% of the population
group (Table 2) (ie, long-face patients were (Fig 1). Both short- and long-face problems have
slightly more likely to accept and receive surgical been more often identified in the lower face (Fig
treatment than their proportion of the clinic 2). The frequency of long lower-face problems
population). has ranged from 16% to nearly 40%. The occur-
The percentage of patients with greater than rence of mid-face and combined mid- and lower-
5 mm of overbite and those with an open bite face problems has consistently been at or below
(reverse overbite) have been constant and equal 5% of the sample. The long-face condition often
throughout the past 2 decades, each accounting is combined with other problems, especially lack
of chin prominence, thereby making it difficult
Table 2. Ske letal /Den tal Characteristics of Surgery to be certain exactly what aspects are most im-
Subsample (% of Sample) portant in leading a patient to seek treatment.
Had The chief complaint often includes the gummy
Total Surgery No Surgery smile and/or anterior open bite that are the
(n = 1305) (n = 551) (n = 754) hallmarks of the long-face condition. The com-
bination of aesthetic and functional problems in
Skeletal Class the long-face condition seems to be harder to
Class I (Number,
%) 174 (11.9) 73 (9.8) 111 (13.3)
Class II* 714 (54.7) 320 (58.1) 394 (52.3)
Overjet >6
Overbite
mm* 35.6 42.1 30.9
Class III 417 (32.0) 168 (30.5) 249 (31.8)
Reverse overjet 13.9 13.4 14.2
Long face
Number (%) 342 (26.2) 155 (28.1) 187 (24.8)
Midface 8.2 8.4 8.0
Lower face 77.2 78.1 76.5
Both 14.6 13.5 15.5
Anterior open bite 16.9 17.8 16.2
Short face
Number (%) 419 (32.1) 187 (33.9) 232 (30.8)
Midface 8.8 12.3 6.0
Lower face 85.9 83.4 87.9 79-80 81-82 83-84 85-86 89-90 91-92 93-94 95-% 97-98
Both 5.3 4.3 6.1 Year (2-year increments)
Overbite >5 mm 17.5 18.0 17.2
indicates statistically significant difference when comparing Figure 1. Graph showing distribution of dentofacial
had surgery group to no surgery group (P value < .05). population for overbite in 2-year increments.
<<    
     Article
      >>
Long face Problems
79-80 81-82 Si-84 85-86 87-88 89-90
Year (2-year increments)
Figure 2. Graph showing distribution of dentofacial
population for long-face problems in 2-year incre-
ments.
live with than, for instance, mandibular defi-
ciency without excess face height.
The climate for surgical-orthodontic treat-
ment changed in the 1990s. This was for the
better in several ways because significant ad-
vances in treatment occurred. For example,
computer software to generate surgical predic-
tions has greatly improved communication with
patients and their understanding of treatment
options, and the near-total replacement of inter-
maxillary fixation with rigid internal fixation
now makes patients more comfortable during
the postsurgical recovery period. It was for the
worse for many patients, however, because
health insurance coverage for the surgery was
restricted in many plans and largely eliminated
by health maintenance organizations. Many
procedures are now being performed on an am-
bulatory care basis. This article discusses the
clinical management of patients with severe
long-face problems requiring surgical-orthodon-
tic treatment.
Treatment Planning
Adolescents With Questionable Growth
Potential
When evaluating orthodontic patients with pri-
marily vertical problems, it is easy to dwell on
anteroposterior problems because most long-
face patients will have a receding chin and Class
II malocclusion. Often the chief complaint is
that the upper incisors are too prominent. An
astute patient may be perceptive enough to de-
scribe the gummy smile as a problem that should
Home | TOC |          
Index
Long-Face Problems 175
be corrected but most know only that they do
not like the prominence of their upper incisors,
without differentiating vertical and horizon-
tal components. A camouflage treatment plan
based on retraction of the upper incisors may be
suggested if the orthodontist views the problem
primarily as Class II malocclusion, without rec-
ognizing the contribution of the skeletal discrep-
ancy.
Correcting the overjet by retraction of inci-
91-92 93-94 95-96 97-98
sors for long-face adolescents is very unattrac-
tive. Extraction of premolars to retract the upper
incisors will cause them to elongate even further
and increase the nasolabial angle. Because this
correction is often accomplished through the
use of Class II elastics, the mandible is likely to
further rotate down and back, accentuating an
already long-face pattern. Before the option of
vertical maxillary impaction became available,
the negative facial aesthetics that resulted were
considered an inevitable consequence of im-
proving the dental occlusion. The continuation
of growth well into the late teens tends to worsen
the deformity; however, the fact that there is
remaining growth provides an opportunity for
implementation of growth modification tech-
niques. Growth modification after the adoles-
cent growth spurt is theoretically possible but
actually improbable because very few adoles-
cents will wear a functional appliance with bite
blocks or a headgear.
Anterior open bite in adolescents (or adults)
often can be corrected with orthodontic treat-
ment by intruding the posterior teeth, but that is
almost impossible without surgery. A multiloop
edgewise appliance, in conjunction with ante-
rior vertical elastics, claims to produce posterior
intrusion and improvement of the skeletal prob-
lem.5 Recent reports show that the open bite
correction occurs almost totally by elongating
the incisor teeth.6 Successful camouflage of a
long-face problem is a function of both the pa-
tient's perception of the success of treatment
and the soft-tissue adequacy (the fuller the lips,
the better they can cover the teeth). If the chief
complaint is excessive display of the teeth and a
gummy smile, elongating the incisors will not
correct it. However, if the patient's major con-
cern is the open bite, an increased display of the
anterior teeth may be tolerable. The orthodon-
tist must be sure the patient understands the
aesthetic implication of the latter decision.
<<    
     Article
      >>
176 Bailey et al
Patients who are treated with other extrusive,
nonextraction orthodontic approaches may end
up with their occlusion reasonably well cor-
rected, but both facial aesthetics and long-term
stability are questionable. In this particular situ-
ation, the lower incisors will remain too protru-
sive relative to the chin for good stability, the
chin will still be deficient, and the lip incompe-
tence will still be present. If the nonextraction
treatment is considered unsuccessful and the
patient is retreated with extractions, esthetics
will be compromised even more. For borderline
cases such as these, a lower border osteotomy of
the mandible to bring the chin upward and
forward can greatly improve both dental and
facial aesthetics because the lower lip relaxes
and moves up as the chin is elevated. The lower
border osteotomy is not a complicated proce-
dure and may be accomplished in an ambulatory
care setting.
Adults With Little or No Growth Potential
Long-face patients with no prospect for success-
ful growth modification have no real alternative
to surgery for a successful and stable outcome.
Orthodontic camouflage is not a real option in
long-face problems. It may be better not to treat
a patient with a true vertical problem who re-
fuses to consider surgical correction because aes-
thetics are likely to be severely compromised.
Planning Surgical-Orthodontic
Treatment
As with mandibular deficiency and other skeletal
problems, the surgical approach must be deter-
mined before the orthodontic approach can be
established.
Surgical Approach
Excessive lower-face height is the primary distin-
guishing clinical characteristic of long-face pa-
tients. It follows logically that a decrease in face
height is needed. This can be accomplished in
three ways:
1. The first is superior repositioning of the
maxilla, or at least the posterior part of the
maxilla, by total or segmental maxillary os-
teotomy. When the maxilla moves up, the
mandible rotates around the horizontal con-
Home | TOC |          
Index
dylar axis to move up with it so that the chin
moves upward and forward. Indirectly, the
maxillary surgery repositions the mandible.
Often it is important not to elevate the an-
terior maxilla, and it may be necessary to
even rotate it downward so that an aesthetic
smile arc can be maintained.7
2. The second is mandibular surgery to bring
the lower jaw forward and upward, which
could be accomplished in an open-bite pa-
tient by tilting the body of the mandible up
after a ramus osteotomy. This would correct
an anterior open bite, but the position of
the maxilla would not be altered at all so this
treatment approach implies that the prob-
lem is largely in the mandible, which rarely
is the case.
3. Superior repositioning of the chin by a man-
dibular lower-border osteotomy. This proce-
dure is a useful adjunct to either of the other
two surgical possibilities but is unlikely to be
adequate by itself in an adult.
The guideline for choosing between maxil-
lary and mandibular surgery is quite clear: in
patients whose face height should be reduced,
maxillary surgery is the primary procedure. A
mandibular ramus osteotomy is recommended
only as a secondary procedure after the maxilla
has been repositioned vertically. This is just the
opposite of the guideline for short-face patients:
for them, mandibular surgery is preferred when
face height must be increased, and a maxillary
osteotomy is a secondary procedure if required.
The maxilla is the focus of surgical treatment
in long-face patients for two major reasons. First,
the maxilla nearly always has excessive vertical
development, whereas the mandible may not be
involved beyond the indirect rotation that max-
illary surgery corrects. Neither normal jaw and
lip function nor good aesthetics can be achieved
without correcting the maxillary deformity for
most patients. Second, moving the maxilla up
produces a stable surgical correction. Rotating
the mandible at the ramus osteotomy site in a
counterclockwise direction stretches soft tissues
posteriorly and is notoriously unstable. In pa-
tients with a normal mandible that has been
rotated downward and backward (ie, who could
be characterized initially as Class I rotated to
Class II), superior repositioning of the maxilla to
correct the vertical discrepancy also corrects the
<<    
     Article
      >>
Long-Face Problems
anteroposterior problem because the mandible
rotates at the horizontal condylar axis. If the
mandible is both small and rotated, a ramus
osteotomy for further advancement is needed in
addition to superior repositioning of the max-
illa. In patients with a large but rotated mandible
(who might have been characterized initially as
Class III rotated to Class I), correcting the verti-
cal position of the maxilla causes the mandible
to rotate into a prognathic position, and a ramus
osteotomy to shorten it is required.
Some surgeons have advocated counterclock-
wise rotation of the maxilla and mandible, which
theoretically could lead to neuromuscular adap-
tation in a way that isolated mandibular surgery
would not. This would allow the posterior max-
illa to be moved downward to improve smile
aesthetics and would improve the gonial angle.
Currently, there are only anecdotal case reports
to document this response. Long-term remodel-
ing at the gonial angle is a potential problem,
and long-term outcomes of this treatment ap-
proach have not been clearly established.
Many long-face patients have excessive erup-
tion of the lower incisors (ie, the distance from
the incisal edge to the chin is too great). In
addition, the incisors tend to be flared forward,
which produces poor chin-lip balance. Both of
these problems can be addressed with a mandib-
ular lower-border osteotomy. The bony cuts are
angled up anteriorly, allowing the chin to be
moved up and forward, and a wedge of bone is
removed above the chin if further vertical short-
ening is needed.
During maxillary surgery, with the maxilla in
the down-fractured position, dentoalveolar seg-
ments can be created readily. The usual indica-
tion for two segments, created by a parasagittal
osteotomy, is to allow the maxilla to be widened
as it is moved superiorly. Three segments, two
posterior and one anterior, usually are used to
correct a vertical step in the arch, typically by
moving the posterior segments up more than
the anterior. The canines can be in either the
anterior or posterior segments depending on
their initial position.
A major reason for segmenting the maxilla,
although not the only one, is to expand it trans-
versely. Problems with the stability of transverse
expansion have led some surgeons to recom-
mend two-stage surgical treatment when a long-
face patient needs both vertical repositioning
Home | TOC |          
Index
177
and expansion of the maxilla. The first stage is
surgically assisted palatal expansion to widen the
maxilla during the presurgical orthodontics,
and then a LeFort I osteotomy is performed to
reposition it in one piece.8 In our view, better
clinical success with the two-stage surgery has
not been shown, and the increased morbidity
and cost of two surgical procedures rather than
one cannot be justified.9 We recommend one-
stage surgery with segmentation when it is
needed.
In summary, the surgical approach to long-
face patients almost always includes a LeFort I
osteotomy to superiorly reposition the maxilla.
Maxillary segments, mandibular ramus osteot-
omy to advance or set back the mandible, and
lower-border osteotomy to reposition the chin
are added as the requirements of the individual
case dictate.
Orthodontic Approach
As with any surgical-orthodontic case, the orth-
odontic approach is oriented toward positioning
the teeth presurgically in all three planes of
space so their position will facilitate the surgical
plan and the teeth will fit appropriately when
the surgery is completed. To accomplish this
goal, the orthodontist must know the general
surgical plan and two things quite specifically:
(1) whether the maxilla will be kept in one piece
or segmented and (2) whether chin-lip balance
is to be achieved by orthodontically reposition-
ing the incisors or lower-border osteotomy.
Long-face patients rarely have a severely ex-
aggerated curve of Spee in the mandibular arch,
even if a deep overbite is present. In general, it
is preferable to level the lower arch before sur-
gery. Like the guideline for maxillary versus
mandibular surgery, this guideline is in sharp
contrast to the one for short-face patients, whose
leveling often is done postsurgically. The reason
is the same, however. Postsurgical leveling makes
it easier to increase face height, whereas presur-
gical leveling is better when decreased face
height is the goal.
On the other hand, a long-face patient with
severe anterior open bite often has an extreme
curve of Spee in the upper arch to the point that
vertical steps exist in the arch. Usually, the steps
are distal to the canines, but they may occur
between the lateral incisors and canines. The
<<    
     Article
      >>
178 Bailey et al
more severe the steps, the more advantageous it
is to segment the maxilla during the surgery and
level the arch by repositioning the dentoalveolar
segments rather than by moving the teeth orth-
odontically. The orthodontist's role when surgi-
cal segments are planned is to level presurgically
within the segments but not across the osteot-
omy sites and to be sure that there is enough
space between the roots of the involved teeth to
allow interdental osteotomies.
Similar thinking guides the decision as to
whether to expand a narrow maxillary arch orth-
odontically, either with dental expansion (arch
wires only) or orthopedic separation of the mid-
palatal suture (jackscrew appliance or equiva-
lent), or to defer this for segmental osteotomy.
The more severe the narrow maxilla and the
older the patient, the better the decision to ex-
pand surgically. If the patient is young enough
to open the suture orthopedically, presurgical
expansion with a jackscrew appliance is accept-
able. As noted previously, we do not recommend
surgically assisted palatal expansion to widen a
narrow maxilla in a patient who will be sched-
uled for LeFort I osteotomy later nor is there any
reason to expand orthodontically if the dentoal-
veolar segments are needed to correct vertical
steps because the segments also can be ex-
panded at the time of surgery.
Presurgical Orthodontics
There are three points of special interest for
long-face patients.
1. A LeFort I osteotomy requires a long inci-
sion in the maxillary vestibule. If a lower-
border osteotomy is part of the surgical
plan, an incision in the mandibular anterior
vestibule will be necessary. These incisions
tend to stress the gingival attachment of ad-
jacent teeth, presumably because scar con-
traction during healing pulls the attachment
apically. It is much easier to prevent strip-
ping of gingival tissues and unsightly expo-
sure of roots than it is to correct it afterward.
When the gingival attachment is question-
able, the attachment should be augmented
by placing gingival grafts in doubtful areas at
least 2 to 3 months before the orthognathic
surgery (preferably, but not necessarily, be-
fore the orthodontic treatment begins). Gin-
Home | TOC |          
Index
gival stripping is most common around
lower incisors after a vestibular incision, but
loss of gingival attachment around the up-
per incisors and canines can and does occur
in patients having maxillary osteotomy, and
can be an aesthetic problem.
2. In a patient with an anterior open bite
whose plan calls for a segmental maxillary
osteotomy with anterior and posterior den-
toalveolar segments, it is important not to
level the upper arch during the presurgi-
cal orthodontics. Instead, the orthodontist
should level only within the segments. Al-
though continuous arch wires with steps at
the planned osteotomy sites can be used
until just before surgery if desired, usually it
is easier and better to use separate arch wire
segments throughout the presurgical treat-
ment. The size of the arch wire segments
and their sequence would be the same as the
continuous wires for any other patient, cul-
minating with full-dimension rectangular
stabilizing segments. An auxiliary arch wire
fitted to the model surgery casts is placed in
the headgear tubes at surgery. It is a mistake
to level the upper arch presurgically in pa-
tients with severe open bite because this pro-
duces a relapse tendency. The leveling
would occur primarily by elongating the up-
per incisors. When the orthodontic appli-
ance is removed postsurgically, the incisors
would tend to relapse apically to some ex-
tent and, of course, that would lead to re-
opening of the bite anteriorly. As the time of
surgery approaches, if it is observed that the
maxillary incisors have been elongated dur-
ing the presurgical orthodontics, the contin-
uous arch wire should be removed and re-
placed by anterior and posterior segments.
This should be performed several weeks be-
fore the surgery to allow any relapse ten-
dency to express itself before the final sur-
gical planning is performed.
3. Similar considerations guide the approach
to maxillary width. If the arch will be ex-
panded orthodontically, this should be per-
formed at the very beginning of the presur-
gical orthodontics so the expansion can be
maintained as long as possible before the
expansion appliance is eventually removed.
If a LeFort I osteotomy with separate poste-
rior dentoalveolar segments is planned and
<<    
     Article
      >>
Long-Face Problems
the expansion will be accomplished surgi-
cally, the orthodontist should be careful not
to produce any orthodontic expansion. In-
deed, in these patients, any presurgical
transverse tooth movement should be con-
traction rather than expansion of the arch
so that orthodontic relapse will not contrib-
ute to any tendency to posttreatment nar-
rowing of the arch. The surgical segments
inevitably will relapse somewhat toward the
midline, and teeth tend to move back to-
ward the midline after dental expansion.
Final Presurgical Planning
Complete records—panoramic and lateral ceph-
alometric radiographs, other radiographs if in-
dicated (eg, periapical radiographs in areas in
which interdental osteotomy for segmental sur-
gery is planned), facial and intraoral photo-
graphs, and dental casts—are required immedi-
ately before surgery. Because these long-face
patients will have maxillary surgery, a facebow
transfer to a semiadjustable articulator is neces-
sary.
The planning at this stage is largely a repeti-
tion of what was done before treatment started.
The goals are to verify the original plan, pre-
cisely quantify the movements needed at sur-
gery, and complete the model surgery so that
occlusal splints can be made. The first step is a
cephalometric prediction. From this, the mea-
surements that are necessary for model surgery
are taken, and the casts are repositioned on the
articulator.
There are two critical elements in the plan-
ning at this stage: (1) how far the maxilla is
moved up and (2) if there would be residual
overjet with a straight vertical movement of the
maxilla, whether the maxilla is moved forward
or back to correct overjet or the mandible
lengthened or shortened by ramus osteotomy.
Moving the maxilla up too far is as harmful to
facial aesthetics as leaving the long face uncor-
rected. A long-face patient should not be over-
treated because a few millimeters of lip incom-
petence is normal. It is a mistake to elevate the
maxilla enough to bring the lips into contact at
rest. Younger patients can tolerate more upward
movement of the maxilla (aesthetically and psy-
chologically) than can older patients.
It is better to leave approximately 4 mm of lip
Home | TOC |          
Index
179
separation in the younger patients, perhaps even
more in those over age 30. Incisor display is a
youthful characteristic and decreasing it makes
an individual look older.7 The associated soft-
tissue changes accentuate this effect. When the
maxilla is moved up, the soft tissues of the
cheeks are relaxed, which is good for stability
but bad for aesthetics because the wrinkles that
inevitably accompany aging become more prom-
inent. Patients beyond their early twenties do
not appreciate suddenly looking older.
Exposure of 30% to 40% of the clinical crown
of the maxillary incisor beneath the lip is aes-
thetically pleasing, whereas completely covering
it is not. Because the upper lip increasingly cov-
ers the upper incisors with advancing age, de-
creasing the amount of exposure of the incisors
may be undesirable as the patient gets older
even if it looks good initially. Rotating the ante-
rior maxilla down as the posterior goes up to
establish the best exposure of the teeth may be
needed for optimum aesthetics.
Long-face patients who have a moderate de-
gree of mandibular deficiency can be a treat-
ment-planning dilemma because as the mandi-
ble rotates upward, it also comes forward, which
means that if the maxilla were moved up
enough, the mandible could come forward to a
normal anteroposterior position. Or, if the max-
illa were moved back as it moved up, the overjet
could be corrected without so much vertical
movement. For many patients, moving the max-
illa just a little more could prevent having to
operate on both jaws.
All other things being equal, one-jaw surgical
procedures have less morbidity, better stability,
and lower cost compared with two^jaw surgery.
On the other hand, moving the maxilla up too
much is bad aesthetically, and moving it back is
worse. Often the maxilla should be advanced
somewhat to obtain the best lip support and
esthetics. Almost never should it be retracted. It
is much better to accept the need to do two-jaw
surgery to obtain a good result than to signifi-
cantly compromise aesthetics to keep the surgery
within one jaw. The planning is even more of a
problem in a patient who habitually postures the
mandible forward so that there is more than a 1-
to 2-mm difference between intercuspal position
(centric occlusion) and retruded contact posi-
tion (centric relation). Articulator mountings
and the rotation of the mandible around the
<<    
     Article
      >> Home | TOC |          
Index

180 Bailey et al

Figure 3. (A and B) JW,

age 45.4, pretreatment fa-
cial appearance. Patient
has a severe anterior open
bite, a narrow maxillary
arch, and a step in her max-
illary arch between the an-
terior and posterior seg-
ments. She required a
three-piece maxillary os-
teotomy for correction of
her vertical and transverse
discrepancies.

horizontal condylar axis are predicated on a

clinically repeatable position of the mandible,
and planning cannot be accurate if this repeat-
able position cannot be found. After patients
have been under orthodontic treatment for
some months, the dental cues that lead to con-
sistent forward shifts of the mandible usually are
disrupted. It is easier to obtain an accurate
mounting just before surgery than at the begin-
ning of treatment. If there is any doubt about
the need for both LeFort I and mandibular sur-
gery as the plan for treatment is being devel-
oped, the orthodontist and surgeon should dis-
cuss the possible need for two-jaw surgery with
the patient. The revised plan should be pre-
sented as a desirable alternative, showing the
patient why and requesting approval.
Often, mandibular lower-border osteotomy is
added to maxillary surgery with or without man-
dibular ramus surgery to correct long-face prob-
lems. In the prediction, if it is apparent that
better facial balance can be achieved by length-
ening the mandible at the lower border and
removing bone to decrease vertical height in the
lower third of the face, then this procedure Figure 4. JW, pretreatment of cephalometric tracing.
<<    
     Article
      >> Home | TOC |          
Index

Long-Face Problems 181

Figure 5. JW, age 46.10,

(A) posttreatment facial
and (B) dental appearance.
Posttreatment facial photo-
graphs showing improved
facial balance and ideal
overjet and overbite.

should be included. Inclusion of this short sur- bilizing wires are removed, they should be re-
gical procedure produces little risk of additional placed at the same appointment with working
morbidity and provides stable and predictable arch wires and light vertical elastics.
results.
The final step in the presurgical planning is
preparation of the splint or splints. If two^jaw
surgery is planned, it is helpful to have an inter-
mediate splint that fits the result of the first
(maxillary) stage of surgery and also a final
splint.

Postsurgical Orthodontics
As with all orthognathic surgery patients, it is
recommended for long-face patients that the
orthodontist, not the surgeon, remove the splint
when the patient is ready for postsurgical orth-
odontics. The splint should not be removed un-
til the patient is ready to have the stabilizing arch
wires removed so that finishing orthodontics can
proceed. It is the surgeon's judgment as to when
that step is appropriate. With maxillary surgery
only and rigid internal fixation, orthodontic
treatment sometimes can resume as early as 2 to
3 weeks postsurgically. With two-jaw surgery, a
longer healing time seems prudent, even with
the use of rigid internal fixation. When the sta- Figure 6. JW, posttreatment cephalometric tracing.
<<    
     Article
      >>
182 Bailey et al
Figure 7. JW, superimposition showing pretreatment
to posttreatment changes.
The most difficult part of postsurgical orth-
odontics for long-face patients is maintaining
transverse maxillary expansion, particularly sur-
gical expansion. If the maxillary buccal seg-
ments relapse medially after the surgery, not
only is there relapse toward crossbite but also
the bite tends to open anteriorly because of
cuspal interferences posteriorly. It takes at least
6 months after surgery for the maxillary dentoal-
veolar segments to stabilize transversely so they
must be held in their expanded position during
the finishing orthodontics. The easiest way to do
this is to use a heavy labial auxiliary wire in the
headgear tubes along with the light working
arch wires. If segmental surgery was performed,
this can be the same auxiliary wire that was used
at surgery. If dental expansion was performed
presurgically, it may be wise to make a new labial
auxiliary wire for use at the postsurgical stage.
An alternative for maintaining width is a
transpalatal lingual arch, which has the signifi-
cant advantage of excellent torque control. (Al-
Home | TOC |          
Index
though a removable palate-covering appliance
also might be possible, this tends to interfere
with finishing, and it is not recommended.) A
lingual arch cannot be in place at surgery before
the segments are moved, and it is difficult to
place one in the operating room at the conclu-
sion of the surgical procedure. A labial auxiliary
is best until the time of splint removal. At that
point, if a lingual arch is desired, it can be
added. The new lingual arch system that uses a
redesigned 32 X 32 tube can make this easier.10
Because the arches were leveled either pre-
surgically or during the surgery, postsurgical
orthodontics for long-face patients is often ac-
complished quickly. The teeth usually fit quite
well when the patient returns from surgery, and
it is only necessary to settle them into position
before proceeding to retainers. If orthodontic
appliances are removed a few months postsurgi-
cally, it is important that the patients who had
transverse expansion must wear their maxillary
retainer diligently (full-time for several months
at least).
Case Report
JW, age 45.4, was seen on referral from her local
dentist. Facial examination showed a good rela-
tionship of the maxillary teeth to the lips, with
chin deficiency relative to the face and lower lip
(Fig 3A and B). On intraoral examination, the
teeth were well aligned in both arches, with a
narrow and constricted maxillary arch. Ceph-
alometric analysis confirmed mandibular defi-
ciency primarily because of a mandible that was
rotated down and back and a skeletal open-bite
pattern because of vertical maxillary excess
(Fig 4).
Because of the mandibular deficiency and
mandibular incisor protrusion, the treatment
plan was extraction of maxillary first premolars;
LeFort I osteotomy to elevate the posterior but
not the anterior maxilla would allow the mandi-
ble to rotate up and forward for closure of the
anterior open bite without lengthening the ra-
mus and stretching the pterygomaxillary sling.
Two presurgical predictions were presented
to the patient: a LeFort I and a LeFort I and a
lower-border osteotomy. The patient elected the
LeFort I only because there was very little differ-
ence in the two images. At the completion of
treatment, she had very good occlusion with
<<    
     Article
      >>
Long-Face Problems
ideal overjet and overbite (Fig 5A and B). Post-
treatment cephalometric analysis showed im-
proved facial balance (Fig 6). The superimposi-
tion showing pretreatment to posttreatment
change is shown in Figure 7.
References
1. Proffit WR, White RP: Who needs surgical-orthodontic
treatment? Int J Adult Orthod Orthog Surg 5:81-89,
1990
2. Bailey LJ, Proffit WR, White RP: Assessment of patients
for orthognathic surgery. Semin Orthod 5:209-222,
1999
3. Proffit WR, Phillips C, Dann C IV: Who seeks surgical-
orthodontic treatment? Int J Adult Orthod Orthog Surg
5:153-160, 1990
4. Bailey LJ, Haltiwanger LH, Blakey GH, et al: Who seeks
Home | TOC |          
Index
183
surgical orthodontic treatment-a current review. Int J
Adult Orthod Orthognath Surg 16:1-13, 2001
5. Kim YH: Anterior open bite and its treatment with mul-
tiloop edgewise archwire. Angle Orthod 57:290-321,
1987
6. Kim YH, Han UK, Lim DD, et al: Stability of anterior
openbite correction with multiloop edgewise therapy: A
cephalometric follow-up study. Am J Orthod Dentofac
Orthop 118:43-54, 2000
7. Sarver DM: The importance of incisor positioning in the
esthetic smile: the smile arc. Am J Orthod Dentofac
Orthoped 120:98-111, 2001
8. Silverstein K, Quinn PQ: Surgically-assisted rapid palatal
expansion for management of transverse maxillary defi-
ciency. J Oral Maxillofac Surg 55:725-727, 1997
9. Bailey LJ, White RP, Proffit WR, et al: Segmental LeFort
I osteotomy to effect palatal expansion. J Oral Maxillofac
Surg 55:728-731, 1997
10. Burstone CJ: The precision lingual arch: hinge cap at-
tachment. J Clin Orthod 28:151-158, 1994
<<    
     Article
      >> Home | TOC |          
Index

Mosby presents two

new approaches to
, W.B. SAUNDERS Mosby Preadjusted
- Essential Information for Today's Dentists
Appliances...
from the Leading Health Care Publisher

American Journal of Orthodontics and Coming Soon!

Dentofacial Orthopedics
The Official Publication of the American
Association of Orthodontists, its constituent
societies, and the American Board
of Orthodontics
British Journal of Oral &c Maxillofacial Surgery
Official Journal of the British Association of
Oral & Maxillofacial Surgeons and a
Recognized Journal of the American College of
Oral and Maxillofacial Surgeons
Dental Abstracts
Dental Clinics
International Journal of Oral 5c Maxillofacial
Surgery
Official Publication of the International
Association of Oral & Maxillofacial Surgeons
The Journal of Evidence-Based
Dental Practice
Orthodontic
I THE DKWlTfUN WITH
ITHEPREaBlISIBD
Management of the
I APPLIANCE
Dentition with
the Preadjusted
Appliance
John C. Bennett, FDS, DOrth;
and Richard P. Mclaughlin, DDS
This innovative text promotes the individual
management of each tooth using the
Preadjusted Orthodontic Appliance. Practical,
well-illustrated instruction presents treatment
mechanics for common problems including
crowding, spacing, tooth size discrepancies,
ectopic eruption, and impaction.
April 2002 • Approx. 380 pp., 1,390 illus.
0-7234-3265-1 • $175.00

Journal of Oral and Maxillofacial Surgery New!

Journal of Prosthodontics föITHÖMMlC

Systemiied
Official Journal of the American Association of
IltEATMßHr-
1 MECHANICS Orthodontic
Oral and Maxillofacial Surgeons Treatment Mechanics
The Journal of Prosthetic Dentistry Richard P. Mclaughlin, DDS;
John C. Bennett, FDS, DOrth;
Journal of Prosthodontics and Hugo Trevisi
Official Journal of the American College This new, up-to-date resource focuses on
of Prosthodontists
correcting malocclusion or misalignment of
Oral and Maxillofacial Surgery Clinics the teeth using the Preadjusted Appliance.
Extensively illustrated with line diagrams and
Atlas of the Oral and Maxillofacial
Surgery Clinics color photographs, this practical manual
provides clear coverage of each treatment stage.
Oral Surgery, Oral Medicine, Oral Pathology,
2002 • 335 pp., 765 illus. • 0-7234-3171-X • $149.00
Oral Radiology and Endodontics

Seminars in Orthodontics I Mosby

I A Division of Elsevier Science

Year Book of Dentistry®

3 Easy Ways to Order
For more information, contact Periodicals Marketing at PHONE toll-tree. 800-545-2522
FAX toll-free: 800-568-5136
(215)238-5614. Order \ri& the Internet at any dme: www.mosliy.coin
<<    
     Article
      >> Home | TOC |          
Index

Available from
W.B. Saunders!

Churchill
Livingstone
Journal of Arthroplasty
Official Journal of the American Association
of Hip and Knee Surgeons

Journal of Cardiac Failure

Official Journal of the Heart Failure
Society of America and the Japanese
Heart Failure Society

Journal of Electrocardiology
Official Journal of the International Society
for Computerized Electrocardiology and the
International Society of Electrocardiology

The Journal of Pain

Official Journal of the American
Pain Society

Seminars in Complementary
Medicine
FOR MORE INFORMATION ABOUT THESE TITLES,
PLEASE CONTACT:
Periodicals Marketing
W.B. SAUNDERS
A Division of Elsevier Science
The Curtis Center, Independence Square West
Philadelphia, PA 19106-3399
Phone: (215) 238-5614
©2002 Elsevier Science.
Or visit our home page at: www.wbsaunders.com
<<    
     Article
      >> Home | TOC |          
Index

W. B. S A U N D E R S

JOURNALS Essential Information f or Today's Professionals from theleadingHealth Caw Publisher

ANESTHESIOLOGY NEPHROLOGY PSYCHIATRY

Journal of Cardiothoracic and Vascular Anesthesia Advances in Renal Replacement Therapy—A Journal of Comprehensive Psychiatry—Official Journal of the A\
Journal of Vw—OfficiallJournal of the American Pain Society The National Kidney foundation Psycbopatbological Association
Regional Anesthesia and Pain Medicine American Journal of Kidney Diseases—The Official Journal Seminars in Clinical Neuropsychiatry
Offmal Journal of the American, Asian and (Oceanic, of The National Kidney Foundation
and Latin American Societies of RegionalAnesthesia Journal of Renal Nutrition—The Official Journal of The Council RADIOLOGY
Seminars in Anesthesia on Renal Nutrition of the National Kidney Foundation Seminars in Breast Disease
Seminars in Cardiothoracic and Vascular Anesthesia Seminars in Nephrology Seminars in Nuclear Medicine
Seminars in Pain Medicine Seminars in Radiologie Technology
Techniques in Regional Anesthesia and Pain Management NEUROLOGY Seminars in Roentgenology
Journal of Pain—Official Journal of the American Pain Society Seminars in Ultrasound, CT and MRI
CARDIOVASCULAR DISEASES Journal of Stroke and Cerebrovascular Diseases—Official Journal Techniques in Vascular and Interventional Radiology
Progress in Cardiovascular Diseases of the National Stroke Association and theJapan Stroke Society
Techniques in Intervention^ Cardiology Seminars in Cerebrovascular Diseases and Stroke RHEUMATOLOGY
Seminars in Pediatric Neurology Seminars in Arthritis and Rheumatism
CRITICAL CARE MEDICINE
Journal of Critical Care NURSING SURGERY
Advances in Neonatal Care—Official Journal of the National Journal of Pediatric Surgery—Official Journal of the Section on
DENTISTRY Surgery of the American Academy of Pediatrics, British Association
Journal of Oral and Maxillofacial Surgery—Official Journal of ofPaediatric Surgeons, American Pediatric Surgical Association,
Applied Nursing Research Canadian Association ofPaediatric Surgeons and Pacific
the American Association of Ord and MaxiUofacial Surgeons Archives of Psychiatric Nursing—Official Journal of the SERPN
Journal of Prosthodontics—Official Journal of The American Association of Pediatric Surgeons
Division, International Society of Psychiatric -Mental Health Nurses Operative Techniques in General Surgery
Journal of Pediatric Nursing—Official Journal of the Society of Operative Techniques in Neurosurgery
Seminars in Orthodontics Pediatric Nurses Operative Techniques in Plastic and Reconstructive Surgery
DERMATOLOGY Journal of Pediatric Oncology Nursing—Official Journal of the Operative Techniques in Thoracic and Cardiovascular Surgery—An
American Journal of Contact Dermatitis—The Official Journal Association of Pediatric Oncology Nurses Official Publication of The American Association for Thoracic Surgery
of the American Contact Dermatitis Society Journal of PeriAnesthesia Nursing—Official Journal of the American Seminars in Colon and Rectal Surgery
Seminars in Cutaneous Medicine and Surgery Society of PeriAnesthesia Nurses Seminars in Laparoscopic Surgery
Journal of Professional Nvr^n%—0fficialjournal of the American Seminars in Pediatric Surgery
EMERGENCY MEDICINE Association of Colleges of Nursing Seminars in Spine Surgery
American Journal of Emergency Medicine Pain Management Nursing—Official Journal of the American Society Seminars hi Thoracic and Cardiovascular Surgery—tin Official
Clinical Pediatric Emergency Medicine of Pain Management Nurses Publication of The American Association for Thoracic Surgery
PeriAnesthesia and Ambulatory Surgery Nursing update—Official Seminars hi Thoracic and Cardiovascular Surgery: Pediatric Cardiac
ENDOCRINOLOGY Publication of the American Society ofPeriAnestbesia Nurses Surgery Annual—An Official Publication of The American
Metabolism—Clinical and Experimental Seminars for Nurse Managers Associationfor Thoracic Surgery
Seminars in Oncology Nursing Seminars in Urologie Oncology
GASTROENTEROLOGY/HEPATOLOGY Seminars in Vascular Surgery
Clinical Perspectives in Gastroenterology—The Official OBSTETRICS AND GYNECOLOGY
Clinical PracticeJournal of the American Gastroenterological Clinical Journal of Women's Health TRANSPLANTATION
Association Liver Transplantation-^» Official Publication of the American
Gastroenterology— Official Journal of the American ORTHOPEDICS Association for the Study of Liver Diseases and the International
GastroenterologicalAssociation Armroscopy: The Journal of Arthroscopic and Related Surgery— Liver Transplantation Society
Official Publication of the Artbroscofty Association of North America Transplantation Reviews
the Study of Liver Diseases and the International Society ofArtbroscopy, Knee Surgery, and
liver Transplantation—4» Official Publication of the American Orthopaedic Sports Medicine UROLOGY
AssociationJbr the Study of Liver Diseases and The International Journal of the American Society for Surgery of the üanA— Seminars in Urologie Oncology
Liver Transplantation Society An Officidjwrnal of the Americm Society for Surgery
Seminars in Gastrointestinal Disease of the Hand VETERINARY MEDICINE
Techniques in Gastrointestinal Endoscopy Journal of Hand Surgery-^4» Official Journal of the American Advances hi Small Animal Medicine and Surgery
Society for Surgery of the Hand Clinical Techniques in Equine Practice
HEMATOLOGY/ONCOLOGY Operative Techniques in Orthopaedics Clinical Techniques in Small Animal Practice
Seminars in Hematology Operative Techniques in Sports Medicine Seminars in Avian and Exotic Pet Medicine
Seminars in Oncology Seminars in Armroplasty Veterinary Surgery—The Official Journal of The American
Seminars in Radiation Oncology Seminars in Spine Surgery College of Veterinary Surgeons, Inc. and The European College
Transfusion Medicine Reviews of Veterinary Surgeons
OTORHINOLARYNGOLOGY
INFECTIOUS DISEASES American Journal of Otolaryngology For more information about thesejournals,
Seminars in Infection Control Operative Techniques in Otolaryngology—Head and Neck Surgery please contact:
Seminars in Pediatric Infectious Diseases
Seminars in Respiratory Infections PATHOLOGY Periodicals Marketing
Annals of Diagnostic Pathology W.B. SAUNDERS
MEDICAL TRANSCRIPTION Human Pathology A Division of Elsevier Science
The Latest Word Seminars in Diagnostic Pathology
The Curtis Center, Independence Square West
NEONATAL/PERINATAL MEDICINE PHYSICAL MEDICINE Philadelphia, PA 19106-3399
Newborn and Infant Nursing Reviews Archives of Physical Medicine and Rehabilitation—
Seminars in Perinatology Official Journal of the American Congress of Rehabilitation
Phone (215) 238-5614
Medicine and the American Academy of Physical Medicine Or visit our homepage at:
www.wbsaunders.com
<<    
     Article
      >> | HOME
|          
<<    
     Article
      >>
Seminars in Orthodontics
EDITOR
P. Lionel Sadowsky, DMD, BBS, DipOrth, MDent
Professor and Chairman
Department of Orthodontics
University of Alabama
1919 Seventh Avenue South
Birmingham, AL 35294
Fax: (205) 975-7590
Seminars in Orthodontics (ISSN 1073-8746) is published
quarterly by W.B. Saunders. Months of issue are March, June,
September, and December. Corporate and Editorial Offices:
The Curtis Center, Independence Square West, Philadelphia,
PA 19106-3399. Accounting and Circulation Offices: 6277 Sea
Harbor Drive, Orlando, FL 32887-4800. POSTMASTER: Send
change of address to: Seminars in Orthodontics, W.B. Saunders,
Customer Service, 6277 Sea Harbor Drive, Orlando, FL 32887-
4800.
Editorial correspondence should be addressed to:
P. Lionel Sadowsky, DMD, BDS, DipOrth, MDent, Editor,
Seminars in Orthodontics, Professor and Chairman, Department
of Orthodontics, University of Alabama, 1919 Seventh Avenue
South, Birmingham, AL 35294-0007; fax: (205) 975-7590.
Correspondence regarding subscriptions or change of
address should be directed to Seminars in Orthodontics, W.B.
Saunders, Periodicals Department, P.O. Box 628239, Orlando,
FL 32862-8239 or e-mail hhspcs@harcourt.com.
Change of address notices, including both the old and new
addresses of the subscriber and the mailing label, should be
sent at least 1 month in advance. Customer Service: 1-800-654-
2452
Yearly subscription rates: United States and possessions:
individual, $133.00; institution, $164.00; student and resident,
$67.00; single issue, $50.00. All other countries: individual
$166.00; institution, $198.00; student and resident, $83.00;
single issue, $50.00. For all areas outside the United States and
possessions, there is no additional charge for surface delivery.
For air mail delivery, add $16.00. To receive student/resident
rate, orders must be accompanied by name of affiliated
institution, date of term, and the signature of program/residency
coordinator on institution letterhead. Orders will be billed at
individual rate until proof of status is received.
Prices are subject to change without notice. Current prices
are in effect for back volumes and back issues. Single issues,
both current and back, exist in limited quantities and are
offered for sale subject to availability. Back issues sold in
conjunction with a subscription are on a prorated basis. Checks
should be made payable to W.B. Saunders and sent to Seminars
in Orthodontics, W.B. Saunders, Periodicals Department, 6277
Sea Harbor Drive, Orlando, FL 32887-4800.
Copyright 2002, Elsevier Science (USA). All rights
reserved. No part of this publication may be reproduced or
transmitted in any form or by any means, electronic or
mechanical, including photocopy, recording, or any information
storage and retrieval system, without permission in writing
from the Publisher. Printed in the United States of America.
Correspondence regarding permission to reprint all or
part of any article published in this journal should be
addressed to Journal Permissions Department, W.B. Saunders,
6277 Sea Harbor Drive, Orlando, FL 32887-4800. Telephone
number: 1-407-345-2500.
W.B. SAUNDERS
Home | TOC |          
Index
The appearance of the code at the bottom of the first page
of an article in this journal indicates the copyright owner's
consent that copies of the article may be made for personal or
internal use, or for the personal or internal use of specific clients,
for those registered with the Copyright Clearance Center, Inc.
(222 Rosewood Drive, Danvers, MA 01923; (508) 750-8400;
www.copyright.com). This consent is given on the condition that
the copier pay the stated per-copy fee for that article through the
Copyright Clearance Center, Inc. for copying beyond that
permitted by Sections 107 or 108 of the US Copyright Law.
This consent does not extend to other kinds of copying, such
as copying for general distribution, for advertising or promotional
purposes, for creating new collective works, or for resale.
Absence of the code indicates that the material may not be
processed through the Copyright Clearance Center, Inc.
Reprint inquiries should be addressed to Anne
Rosenthal, Elsevier Science, The Curtis Center, Independence
Square West, Philadelphia, PA 19106-3399. Telephone (215)
238-5534, fax (215) 238-6423; e-mail: a.rosenthal@elsevier.com.
Advertising representative: MJ. Mrvica Associates, Inc,
2 West Taunton Ave, Berlin, NJ 08009. Telephone (609) 768-
9360. Fax (609) 753-0064.
Publication of an advertisement in Seminars in Orthodontics
does not imply endorsement of its claims by the Editor (s) or
Publisher of the journal.
The contributors have checked generic and trade names and
verified drug doses for accuracy according to the standards
accepted at the time of publication. The ultimate
responsibility, however, lies with the prescribing physician.
Please convey any errors to the Editor.
The ideas and opinions expressed in Seminars in
Orthodontics do not necessarily reflect those of the Editor or the
Publisher. Publication of an advertisement or other product
mention in Seminars in Orthodontics should not be construed as
an endorsement of the product or the manufacturer's claims.
Readers are encouraged to contact the manufacturer with any
questions about the features or limitations of the products
mentioned. Neither the Editor or Publisher assume any
responsibility for any injury and/or damage to persons or
property arising out of or related to any use of the material
contained in this periodical. The reader is advised to check the
appropriate medical literature and the product information
currently provided by the manufacturer of each drug to be
administered to verify the dosage, the method and duration of
administration or contraindications. It is the responsibility of
the treating physician or other health care professional, relying
on independent experience and knowledge of the patient, to
determine drug dosages and the best treatment for the patient.
: Seminars in Orthodontics is indexed in the Cumulative
Index to Nursing and Allied Health Literature® print index
and the Cinahl® database.
<<    
     Article
      >>
Seminars in Orthodontics
EDITOR
P. Lionel Sadowsky, DMD, BDS, DipOrth, MDent
EDITORIAL BOARD
Richard G. Alexander, Arlington, TX
Rolf G. Behrents, Memphis, TN
Samir E. Bishara, Iowa City, I A
Robert Boyd, DDS, San Francisco, CA
Larry M. Bramble, Cypress, CA
John S. Casko, Iowa City, I A
Harry L. Dougherty, Van Nuys, CA
T.M. Graber, Evanston, IL
Robert J. Isaacson, Richmond, VA
Alexander Jacobson, Birmingham, AL
Lysle E.Johnston, Jr., Ann Arbor, MI
Gregory J. King, Seattle, WA
Vincent G. Kokich, Tacoma, WA
Steven J. Lindauer, Richmond, VA
INTERNATIONAL
Zeev Abraham, Herzliya, Israel
Jack Dale, Toronto, Ontario, Canada
W.G. Evans, Johannesburg, South Africa
Roberto Justus, Mexico City, Mexico
Home | TOC |          
Index
James A. McNamara, Jr, Ann Arbor, MI
Robert N. Moore, Grand Island, NE
Ravindra Nanda, Farmington, CT
Perry M. Opin, Milford, CT
Sheldon Peck, Newton, MA
William R. Proffit, Chapel Hill, NC
Cyril Sadowsky, Chicago, IL
David M. Sarver, Birmingham, AL
T. Michael Speidel, Minneapolis, MN
William J. Thompson, Bradenton, FL
James L. Vaden, Cookeville, TN
Robert L. Vanarsdall, Jr., Philadelphia, PA
Katherine Vig, Columbus, OH
C.B. Preston, Buffalo, NY
Shinkichi Namura, Tokyo, Japan
George Skinazi, Paris, France
William A. Wiltshire, Winnipeg, Manitoba, Canada
Björn U. Zachrisson, Oslo, Norway
<<    
     Article
      >> Home | TOC |          
Index

Seminars in Orthodontics
VOL 8, NO 4 DECEMBER 2002

Risk Assessment and Management in Clinical Orthodontics

Laurance Jerrold, DDS, JD
Guest Editor

CONTENTS

Introduction: What's It All About Alfie 185

Laurance Jerrold

Understanding the Basics of the Dentist-Patient Relationship, the Standard

of Care, and Informed Consent in the Orthodontic Clinical Setting 187
Laurance Jerrold

Dollars and Sense: Collecting Fees and Dismissing Patients 198

Eric Ploumis

The Scope of Vicarious Liability and Referral Liability in the Orthodontic

Setting: Am I My Brother's Keeper? 205
Burton R. Pollack

Why Orthodontists Get Sued 210

Elizabeth Franklin

The Anatomy of an Orthodontic Lawsuit: From the Summons Through

Trial 216
Toni Reale

Suing the Orthodontist: A View From the Plaintiffs Side of the Bar 220
Joel Kotick

Defending the Orthodontist: A View From the Defendant's Side of the Bar 228
Arthur V. Pearson
<<    
     Article
      >> Home | TOC |          
Index

Have Gun Will Travel: The Role of the Expert Witness Before and During
Litigation 234
Malcolm Meister and Richard Masella

David Versus Goliath: The State Against the Doctor—Administrative

Liability 238
T. Michael Speidel

Transitioning the Orthodontic Practice: Seller's Concerns and Perspectives 243

Randall K. Berning

Transitioning the Orthodontic Practice: Buyer's Concerns and Perspectives 249

Randall K. Berning
<<    
     Article
      >> Home | TOC |          
Index

Seminars in Orthodontics
Future Issues

Vol 9 No 1 (March 2003)

THE HERBST® APPLIANCE
Hans Pancherz, DMD, Guest Editor
Vol 9 No 2 (June 2003)
MOUNTING OF CASTS IN CLINICAL ORTHODONTICS
Richard Kulbersh, DMD, Guest Editor

Recent Issues

Vol 8 No 3 (September 2002)

MANAGEMENT OF THE VERTICAL DIMENSION IN CLINICAL ORTHODONTICS
Timothy T. Wheeler, DMD, PhD, Guest Editor
Vol 8 No 2 (June 2002)
BIOSTATISTICS FOR THE ORTHODONTIC CLINICIAN
Rose D. Sheats, DMD, Guest Editor
Vol 8 No 1 (March 2002)
CLINICAL UPDATE ON TECHNOLOGICAL ADVANCES IN ORTHODONTICS AND DENTOFACIAL ORTHOPEDICS
Gregory}. King, DMD, DMSc, Guest Editor
Vol 7 No 4 (December 2001)
THREE-DIMENSIONAL DIAGNOSIS AND TREATMENT IN ORTHODONTICS
Sheldon Eaumrind, DDS, MS, and Robert L. Boyd, DDS, MEd, Guest Editors
Vol 7 No 3 (September 2001)
TOPICS IN BIOMECHANICS
Stanley Braun, DDS, MME, Guest Editor
Vol 7 No 2 (June 2001)
THE ALEXANDER DISCIPLINE
R.G. Alexander, DDS, MSD, Guest Editor
Vol 7 No 1 (March 2001)
CLINICAL BIOMECHANICS
Steven ]. Lindauer, DMD, MDSc, Guest Editor
Vol 6 No 4 (December 2000)
PSYCHOLOGIC ISSUES RELATED TO ORTHODONTIC TREATMENT AND PATIENT COMPLIANCE
Pramod K. Sinha, DDS, BDS, MS, Ram S. Nanda, DDS, MS, PhD, and Roger B. Eillingim, PhD, Guest Editors
Vol 6 No 3 (September 2000)
BIOLOGY OF ORTHODONTIC TOOTH MOVEMENT: CLINICAL IMPLICATIONS
Bhavna Shroff, DDS, MDentSc, Guest Editor
<<    
     Article
      >>
Seminars in Orthodontics
VOL 8, NO 4
Introduction: What's It All About Alfie
L egal and risk management concerns in orth-
odontic practice have been fairly hot topics
in continuing dental education over the last de-
cade. All of the contributors for this issue of
Seminars in Orthodontics are experts in their re-
spective fields. The topics selected were done so
with the intention of bringing a different per-
spective to the field of orthodontic risk manage-
ment. It is vitally important to remember that
even though much of what is written was done so
by lawyers, no author contemplated the exis-
tence of an attorney-client relationship through
these writings, and all actions taken by the
reader as a result of having read this issue should
first be discussed with local counsel as the laws
on each topic vary from state to state.
In the first article, I provide the reader with
the legal basis for establishing the standard of
care to which we are held. I also discuss the
nature of the doctor-patient relationship and
one of the most important obligations, the duty
to obtain a patient's informed consent. The
reader may use the letters and forms used in this
article, either in part or in their entirety.
Dr Ploumis, then discusses the financial side
of practice. Because financial disputes often pre-
cipitate risk and practice management issues in
clinical practice, his essay on how we should go
about collecting our fees and how to comport
ourselves when doing so is vital information for
both the novice and seasoned practitioner.
Dr Pollack's article is a classic primer on un-
derstanding the clinical nuances concerning the
topic of vicarious liability in the orthodontic
setting. Practicing good risk management in-
volves having a working understanding of this
legal doctrine. Along with noting the distinc-
tions between employee status and that of an
independent contractor, coupled with the liabil-
Copyright 2002, Elsevier Science (USA). All rights reserved.
doi:10.1053/sodo.2002.127863
Seminars in Orthodontics, Vol 8, No 4 (December), 2002: pp 185-186
Home | TOC |          
Index
DECEMBER 2002
ity exposure related to each relationship, he also
discusses referral liability in everyday practice.
Elizabeth Franklin, the claims manager of the
American Association of Orthodontists Insur-
ance Company, our own insurance company,
then discusses several actual cases culled from
her files in an effort to explain why it is that
orthodontists get sued. The age-old adage of "he
who cannot remember the past is doomed to
repeat it" brings home the lesson that if it hap-
pened to someone else, it can happen to me.
Recognizing the fact patterns, the issues in-
volved, and the appropriate responses makes
this article a most valuable addition.
Toni Reale, the claims manager for another
self-insured, dentist-owned insurance company,
discusses what actually happens internally in
your insurance carrier's office from the first tele-
phone call onward. Administratively, this is
when risk management begins, and it is vital to
have an understanding of this process.
Dr Kotick provides a poignant story of how an
orthodontist became a plaintiffs attorney. His
recounting of cases that he has personally han-
dled against orthodontists will leave you in awe
and provide a basis for understanding just how
vulnerable we are to the throws of litigation. It is
interesting to note how recurrent the fact pat-
terns are that inevitably lead to orthodontic liti-
gation.
From the plaintiffs side to the defendant's
side, Arthur Pearson, a nationally renowned de-
fense attorney, provides the reader with insight
into what really goes on in the mind of a defense
attorney during the process of evaluating a mal-
practice case and ultimately, defending it from
the inception of the suit, through discovery, and
ultimately dealing with the issue of bringing it to
trial.
Drs Meister and Massella provide the reader
with a detailed analysis of the role of an expert
witness in the legal process. From what experts
do to how they are selected, the information
they provide is must know, and their approach
185
<<    
     Article
      >>
186 Laurance Jerrold
to this topic is quite fascinating for the uniniti-
ated. From this point on, this issue takes a turn
and looks at orthodontic risk management from
a different perspective.
Dr Spiedel's essay on administrative liability
will knock your socks off. If you only think of risk
management and practice management as deal-
ing with the prevention and defense of a mal-
practice suit, be forewarned; don't read this be-
fore going to bed. It will give you nightmares.
Astute readers will note that administrative lia-
bility is far more dangerous to us than are causes
of action based in the civil arena.
Randall Berning has been a major contribu-
tor to the dental literature, particularly on topics
dealing with practice valuations and transitions.
Looking at our careers along a time line through
risk management glasses, his contribution gra-
Home | TOC |          
Index
ciously provides the reader with the long view
regarding our ultimate evolution from clinical
practitioner to professional citizen and the cor-
responding insight into the concerns we all
share in transitioning one's practice from the
perspective of both the buyer and the seller.
I hope you enjoy this issue and keep it handy
for reference purposes. Adhering to sound in-
terpersonal skills, as well as practice and risk
management principles, is what enables us to
practice day in and day out without driving our-
selves crazy. Every practitioner can attest to the
fact that it is not the orthodontics that does us
in. Instead, it is the daily business grind that
ultimately wears on our professional souls.
Laurance Jerrold DDS, JD
Guest Editor
<<    
     Article
      >> Home | TOC |          
Index

Understanding the Basics of the Dentist-

Patient Relationship, the Standard of Care,
and Informed Consent in the Orthodontic
Clinical Setting
Laurance Jerrold

The foundation of owing one's patient a duty to conform to an accepted

standard of care is predicated on the existence of a mutually consensual
doctor-patient relationship. The nuances of establishing and terminating
this relationship are discussed in detail. What the standard of care is, how it
is established, and by whom, as well as the meaning of it, is also covered in
depth. Finally, an in-depth review of the doctrine of informed consent along
with various mechanisms for implementing this legal mandate into one's
practice is discussed. Because a working understanding of the law relating
to professional liability is imperative if one is to practice prudent risk man-
agement, this article will provide the reader with the requisite familiarity
regarding this issue. (Semin Orthod 2002;8:187-197.) Copyright 2002,
Elsevier Science (USA). All rights reserved.

rthodontists often misunderstand the na-
O ture of the doctor-patient relationship. Es-
sentially, it is nothing more than a simple con-
tract wherein on one hand, patients seek
treatment from doctors with the expectation
that their professional needs will be addressed,
hopefully resulting in a cure of some type,
whereas doctors, on the other hand, consensu-
ally agree to treat these patients with the dual
expectations of affecting such a cure and with
receiving compensation for the professional ser-
vices they render. As with most contracts, there
exist responsibilities and obligations on the part
of each party. The lion's share of these duties
falls on the dentist. In essence, we owe our pa-
tients the following:
1. The dentist will use competent personnel,
training and supervising them appropriately,
From the Program Director, Orthodontic Residency Program, St.
Barnabas Hospital, Bronx, NY.
Address correspondence to LauranceJerrold, DDS,ß), 82 Laurel
Drive, Massapequa Park, NY 11762.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0804-0002$35.00/0
doi:10.1053/sodo.2002.127864
and all persons will be properly credentialed
2.
and licensed.
The dentist will not use experimental proce-
dures on patients and will not undertake pro-
cedures that he/she is unqualified to per-
form.
3. The dentist and staff will keep current with
scientific and technologic advances within
their fields.
4. The dentist shall obtain the patient's in-
formed consent before treatment, keep pa-
tients apprised of their clinical progress, and
complete care in a timely manner.
5. The dentist will be reasonably available for
emergencies and will not abandon the pa-
tient.
6. The dentist will charge a reasonable fee for
services rendered.
7. Appropriate consultations and referrals will
be made, and neither the dentist nor his staff
will practice beyond the scope of duties al-
lowed by their license as defined by law.
8. The dentist will maintain appropriate and
accurate records of the treatment rendered
and will maintain the confidentiality of the
patient.
9. The dentist and staff will comply with all reg-

Seminars in Orthodontics, Vol 8, No 4 (December), 2002: pp 187-197 187

<<    
     Article
      >>
188 Laurance Jerrold
ulatory rules and regulations pertaining to
the scope of his practice and will abide by the
Code of Ethics.
Contrast the duties we owe our patients, with the
following duties that our patients owe us:
1. All instructions will be followed (ie, wearing
elastics, diet control, seeking recommended
consultations, maintaining oral hygiene, and
so on).
2. All scheduled appointments will be kept.
3. Fees for services rendered will be paid.
4. Patients will conform to generally accepted
modes of behavior.
5. Patients will be truthful regarding their
health history.
It is in the application of these contractual obli-
gations that the dentist-patient relationship
takes on meaning.
It is quite easy to establish a dentist-patient
relationship. The bottom line is, if you consen-
sually offer someone professional advice (the
physical setting is irrelevant) intending for them
to rely on the advice, and they do so, you have
just established a doctor-patient relationship.
On finding the existence of such a relationship,
any professional opinion offered or treatment
rendered related to the patient's concern will
now be judged according to the appropriate
standard of care. It does not matter whether or
not you formally examined the patient, where
you did it, or whether or not you charged a fee
for your services.
Must you treat everyone who presents to your
office for treatment as a patient? No! Our courts
have routinely stated that dental practitioners
are not akin to common carriers or innkeepers
and therefore need not open their doors to all
who seek their services. You cannot, however,
discriminate against or refuse services to any
person based solely on the fact they fall into one
of many legally protected classifications such as
race, religion, gender, sexual orientation, na-
tional origin, handicapping condition, and so
on. You are permitted to legally discriminate on
the basis of such criteria as limiting your practice
to a particular specialty eg, the inability of the
patient to assume the financial obligations asso-
ciated with treatment, a patient's inability to
abide by reasonably constructed office protocol,
and so on.
Home | TOC |          
Index
A big problem that practitioners face is not
knowing under what circumstances it is legal to
terminate the dentist-patient relationship and
how to appropriately do so. If handled properly,
saying goodbye to a patient is a very effective
practice and risk management tool.
Legally, the ways to end a professional rela-
tionship are (1) both parties agree to end it, (2)
the patient is cured or a course of treatment is
completed, (3) the dentist or the patient dies,
(4) the patient decides to unilaterally terminate,
or (5) the dentist decides to unilaterally termi-
nate.
An example of the first scenario is the patient
whose spouse or parent gets transferred across
country in the middle of treatment. The second
and third are fairly self-descriptive. In the fourth,
the patient is usually dissatisfied with some as-
pect, clinical or administrative, of the care they
received, and there is often some acrimony
when they leave. It is the last one that vexes most
practitioners. To unilaterally terminate a patient
from your practice, the following procedure
should be followed.
First, the patient must be given sufficient no-
tice of the intent to withdraw as their practitio-
ner of record. A letter should be sent both by
certified mail with return receipt requested and
by regular mail by using a certificate of mailing.
By using both methods, the dentist assures that
the patient was either notified or that an attempt
to do so was made.
Second, in this letter, the patient should be
informed of the reason (s) you are terminating
your professional relationship with them (Ap-
pendix A). The legally recognized reasons for
allowing doctors to take this action occur when
the patient breaches one or more of the contrac-
tual obligations they owed you as previously
noted.
Next, if the patient is at a point in treatment
such that continued care is still required, he/she
should be strongly urged to seek it. State in this
letter that you will provide adequate time for the
patient to seek substitute or alternative care.
Give the patient or parent notice of a specific
time frame during which they should seek out a
new dentist (ie, 30 or 45 days). Inform the pa-
tient that during this period of time you will be
available for emergency care only or to offer a
referral if necessary. If you have placed active
appliances in the mouth such as retractive me-
<<    
     Article
      >>
Understanding the Basics of the Dentist
chanics, reverse curve arches, or distalizing
springs, you should inform the patient to come
in immediately so these appliances can be re-
moved, citing the possibility of negative sequel-
lae if this is not done.
Finally, inform the patient that on request a
copy of his/her records will be forwarded to
them or to a subsequent treating practitioner.
Although you may legally be entitled to charge
the patient a reasonable fee for the duplication
of their records, you cannot withhold the pa-
tient's records because they owe you money. Do
not place a financial stumbling block in the
patient's way while attempting to end the doctor-
patient relationship.
It should now be evident why it is important
to continually document all instances of (1) the
patient not following instructions or cooperat-
ing in his/her care; (2) the patient breaking,
missing, and/or canceling his/her appoint-
ments; (3) times when the patient was abusive or
disruptive to the staff or others; and (4) the
patient's failure to meet his/her financial obli-
gations to compensate you for the professional
services you rendered.
Obviously, a copy of this correspondence,
along with the proof of mailing, should be
placed into the patient's record. Does dismissing
patients from one's practice constitute the tort
of abandonment? The answer is a qualified no.
Abandonment may be defined as the following:
1. Not giving the patient any further appoint-
ments or refusing to treat him/her before
his/her course of treatment is completed
without having a legally recognized reason to
do so. Thus, unless a patient of record is in a
life-threatening situation, is bleeding pro-
fusely, or is in excruciating pain, all rarities in
orthodontic practice, you have a legally rec-
ognized right to terminate the doctor-patient
relationship on proper notice.
2. Not being available to a patient who requires
follow-up therapy to a procedure performed
is also a form of abandonment. Thus, not
being available and not providing for substi-
tute or emergency coverage should a prob-
lem materialize has been viewed as abandon-
ing a patient.
3. Constructively abandoning the patient, which
often takes the form of extending or prolong-
Home | TOC |          
Index
189
ing treatment for nonclinical (financial) rea-
sons.
As long as all of these guidelines and the dis-
missal protocol are followed, it will be difficult
for patients to sustain any claim that they were
abandoned.
By virtue of the existence of the doctor-pa-
tient relationship, once we agree to treat a pa-
tient, we now have an obligation, a legally rec-
ognized duty, to conform our ministrations to a
minimally accepted standard of care. What is
this standard of care, and how is it determined?
In its barest sense, the standard of care is mea-
sured by what a reasonable practitioner, practic-
ing in the same or in a similar community, would
have done or refrained from doing under the
same or similar circumstances. Reasonable prac-
titioners are permitted latitude accounting for
the fact that although diagnostics have the po-
tential for some minimal differential variability,
there often exists even greater variability relative
to a practitioner's mechanotherapeutic ap-
proach to a given problem. How far can one go
and still be considered reasonable? Generally,
you are safe if what you are doing is also being
done by a respectable minority of your peers. In
other words, if you are the only one, or one of
only a few, who is espousing a particular philos-
ophy or form of treatment, you are skating on
thin ice from a standard-of-care perspective.
What is this community thing? In essence, it
was never considered fair to compare the pro-
fessional behavior of a practitioner who prac-
ticed in a metropolis and had the benefits of
ultrahigh-tech tools, a stable of super specialists
to rely on, the benefits of major university and
teaching hospitals' support and continuing edu-
cation programs, and patients who could afford
to pay for state of the art cutting edge treatment
with that of the doctor who practiced in a small
town in which the community may be several
years behind in terms of the equipment available
for use, no specialists for miles, a local commu-
nity hospital at best, less access to educational
advancements, and patients who paid for ser-
vices by bartering odd jobs.
Historically, this disparity favored the devel-
opment of the locality rule to be used in deter-
mining the standard of care to be applied. Doc-
tors in small local communities were not held to
the standards of their big city brethren. As times
<<    
     Article
      >>
190 Laurance Jerrold
changed, so did the law. With more schools turn-
ing out more doctors, the numbers of specialists
increased. More sophisticated equipment was
being made available geographically, coupled
with communications and travel evolving to the
point when continuing education was more ac-
cessible; the community per se began to shrink.
Although the locality rule still exists in some
states for general practitioners, for specialists in
the United States today, the community to which
we are compared is now national is scope.
Who determines this community/national
standard? We do. Every doctor, even nonorth-
odontists, by virtue of his/her skill, knowledge,
education, expertise, and experience, can go
into court and offer to be an expert witness. It is
up to the court to qualify that person as an
expert, and many states have different legislative
and precedential criteria for making this deter-
mination. Once the qualification has been
made, the function of the expert witness is mul-
tifaceted.
The first task is to act as an interpreter, mak-
ing sure all the facts of the case are understand-
able to lay jurors. The remaining tasks are to
educate and inform the jury because jurors are
the ones who will determine whether the doctor
was liable or not. This education and informa-
tion dissemination will focus on whether or not
the standard of care was or was not breached
and if it was, in what way. The next task is to give
an opinion as to the extent and severity of any
injuries claimed by the plaintiff. Finally, the ex-
pert must give an opinion as to whether or not
the defendant's breach of duty actually, directly,
or proximately caused the injuries that the plain-
tiff sustained.
These duties are important because they par-
allel the burden of proof that the plaintiff must
sustain to prevail in a malpractice suit. The
plaintiff must prove all four of the elements that
constitute the requisites of a malpractice claim:
(1) the doctor failed to conform his/her profes-
sional actions to a certain standard of care; (2)
the doctor breached this duty; (3) there was an
injury sustained; (4) the breach of duty was the
actual, direct, or proximate cause of the injury
suffered.
Although it is relatively straightforward to ap-
ply this information to the clinical setting, one of
the biggest problems facing orthodontists in the
legal arena today is that of defending a claim for
Home | TOC |          
Index
lack of informed consent. Breaking down this
legal doctrine into its individual elements is crit-
ical to understanding the risk management ap-
plication of this basic fundamental legal, ethical,
and practice management principle.
First, one has to start by realizing that there is
a difference between consent and informed con-
sent. To put it into simple terms, consent is a
threshold issue. In other words, before you can
obtain one's informed consent to render any
specific treatment, you first must obtain his/her
general consent for you to be his/her doctor.
For a patient's consent for treatment to be valid,
it must first be given voluntarily. Second, the
person granting the consent must be of legal
age. Finally, this person must be of sufficient
mental capacity. Let's look at these a little closer.
The issue of one's mental capacity is usually
restricted to the elderly and to those with easily
recognizable mental deficiencies; fortunately,
we rarely encounter this issue in clinical orth-
odontics. However, consider, for example, the
patient or parent who shows up at the consulta-
tion visit acting in a manner that would lead you
to believe he/she was impaired secondary to
drug or alcohol abuse. If he/she respond irra-
tionally to the treatment plan you suggest, the
issue of his/her capacity to grant consent, for
himself/herself or for others, may come into
play.
The more common problems relating to con-
sent fall into the other two categories. The legal
age issue arises in two contexts. The first in-
stance occurs when the patient is a minor, some-
one under the legal age of majority (usually 18).
Because infancy generally precludes the patient
from being able to grant consent, you must ob-
tain the legal guardian's consent to treat that
child. If grandma, an aunt, the housekeeper, or
even the noncustodial parent accompanies the
minor to your office, you do not technically have
legal consent to allow you to render treatment.
You may obtain consent by telephone, letter, or
health care proxy; just make sure the appropri-
ate party grants it to you.
A second example arises when you are treat-
ing a minor and right in the middle of orth-
odontic therapy, the patient turns 18, the legal
age of majority. The patient looks at you and says
"take my braces off." Mom retorts "over my dead
body." What do you do? The patient, not mom,
has the final say. Leaving the braces on after the
<<    
     Article
      >>
Understanding the Basics of the Dentist
patient requests their removal is technically a
battery, defined as unauthorized, harmful, or
offensive contact with another.
Finally, whether the consent granted was vol-
untary or not may turn on a patient's claim that
they were mentally coerced into accepting treat-
ment. This can arise when patients are told
something like "if you don't undergo orthodon-
tic therapy you will lose all of your teeth from
periodontal disease, suffer TMD problems, etc."
These types of remarks should not be made if
there is no clinical evidence to substantiate them
or if they cannot be supported by scientific evi-
dence. Nor should you imply that a patient's self
image or facial esthetics is such that they will be
socially ostracized or scorned because of their
"crooked teeth." You cannot make untrue or
unfounded statements that "bully or guilt" pa-
tients into accepting orthodontic treatment.
Once we have received a patient's consent, we
must then obtain his/her informed consent to
our ministrations. Patients must be told in a
language they can comprehend such things as
what the problem is, what you are going to do,
how you are going to do it, how long treatment
will take, the costs, and so on. Sounds simple
right? Maybe, maybe not. Let's take each item
separately. But before we do that, there's just
one more thing we have to know; which legal
standard do we use for disclosing this informa-
tion? There are basically two standards for dis-
closure, and which one you use depends on
which state you practice in. The more traditional
standard is professionally based. Under this ap-
proach, the only information you need to dis-
close to your patients is the information usually
disclosed by other doctors under the same or
similar circumstances. In other words, the doc-
tor decides what and how much to tell the pa-
tient. The more modern approach is the patient
need to know standard. The idea here is to give
the patient all of the information that a reason-
able person in the patient's position would
deem material to make a decision to accept or
reject the proposed treatment. From an ethical
perspective, the patient-based standard is the
preferred alternative.
Looking further at the individual elements,
one can see that patients who have a problem
with English should be spoken to in their native
tongue for them to understand what is being
said. Next, speak plainly. To tell your patient
Home | TOC |          
Index
191
that they have a retrognathic mandible, aplasia
of tooth no. 7, and that the recommended treat-
ment is a sagittal split ramus osteotomy followed
by an implant prosthesis in the maxilla may be
technically correct. However, it is infinitely more
comprehensible to say that the patient's bottom
jaw is too small, they are missing an upper front
tooth, and that the best way to solve their prob-
lem is by surgically bringing the bottom jaw
forward and replacing the missing tooth with an
implanted fake one. Replacing English with
Dentalese is a very common mistake. Many prac-
tice management experts have recommended
that your treatment coordinator, a nondentist,
take over the consultation duties because they
are less apt to fall into the language trap. The
information can be given by anyone in the office
and in almost any medium. Each of us should
develop our own style of information transmis-
sion. Some will go high tech by computer imag-
ing and prepackaged interactive CD programs
or proprietarily produced videotapes. Others
will find that educational brochures or forms
work best. For some, a good old-fashioned con-
versation with the patient is the way to go. How
it is done and by whom is irrelevant; the key is to
ensure that regardless of how it's done or who
does it, the patient or parent understands the
information being transmitted.
Whatever treatment you are proposing may
be clear cut in your own mind, but you must
learn to put yourself in the patient's position. If
any doctor were to recommend a certain proce-
dure to be performed on you, I'm certain that
being the intelligent person you are, you would
want to know what your other treatment options
or alternatives were. You would want to know
what the risks are for each one, what the conse-
quences are, and what limitations exist. Why
should our patients be different from us? They
are not, and just because we are talking about
teeth, does not make it any different. Patients
should be apprised of all viable treatment alter-
natives, even if you do not perform them. Re-
member, it's their bodies we are invading; only
they can give us the right to trespass and in what
way. Patients also have the right to know what
will happen if they choose not to undergo treat-
ment, for in many cases this is, at least from their
perspective, another viable alternative.
How long treatment will take and how much
it will cost are also factors involved in obtaining
<<    
     Article
      >>
192 Laurance Jerrold
one's informed consent. It is a common pre-
sumptive mistake for the doctor to make unsub-
stantiated judgements as to what patients can or
should be able to afford. The amount of time
that treatment will take may be of major signifi-
cance to some whereas far less crucial for others.
Your estimate of how long treatment will take
should not "guild the lily" in the hope that the
patient will not be turned off. We should not be
engaging in paternalism by believing that we
know that the treatment we are recommending
is the best thing for the patient. The reason for
this is that often our best intentions are not the
best thing for that particular patient because of
financial or temporal concerns, whether real or
perceived.
Finally, the patient must have the opportunity
to ask and have answered all of their questions.
Remember, you do not give informed consent,
the patient grants you his/her informed consent
for the treatment being proposed and/or ren-
dered. In other words, it's a two-way street; doc-
tors must provide their patients with sufficient
information that in turn allows them to accept
or reject the course of therapy being recom-
mended.
A frequently asked question is, what is the
best way to document that you have obtained the
patient's informed consent? Although a few
states require a signed writing, think of docu-
mentation as a risk-management ladder. The
higher rung on the ladder that you stand on, the
greater the risk-management protection. Video-
taping the consultation would be the best de-
scriptor, the highest rung on the risk manage-
ment ladder. The next rung down on the risk
management ladder is audiotaping what was
said. Both of these offer wonderful protection to
prove what was said, but they also prove what was
not said. The third rung down is to have the
patient write in his/her own hand what was rec-
ommended and what the risks and alternatives
are. By doing this, the patient can never claim
he/she did not understand what you told him/
her; after all, he/she wrote it. Next, the most
commonly used tool is to distribute a printed
pamphlet or form that is acknowledged via the
patient or parent's signature. Following that
would be a notation in the patient's chart that
this form or booklet of information was given
but without his/her signature or initials. The
next rung down is documenting in the chart
Home | TOC |          
Index
what information you told the patient or parent,
again evidenced by their signature or initial.
This would again be followed by the same entry
without such acknowledgement. There is a le-
gally accepted presumption that if something
was written in the chart it was said or done; the
corollary also applies, if it was not written, it was
not said or done. Toward the bottom of the risk
management ladder, your chart might say that
the risks and alternatives were discussed without
actually elaborating on the discourse. The next
rung might note that informed consent was dis-
cussed. Finally, the last rung would depict a
chart devoid of any evidence related to a consul-
tation discussion—your word against that of the
patient's.
Many doctors err by believing that if informed
consent was obtained at the onset of treatment,
their patients no longer have to be informed if
negative sequellae start to occur. If, in the mid-
dle of treatment, you discover incipient root
resorption, periodontal disease, or decalcifica-
tions, you know that at some point in time a
decision may have to be made to discontinue
active treatment prematurely, even if the treat-
ment goals have not been realized. Who makes
that decision? If you answered the doctor, you
are being paternalistic. If you say the patient,
how are they to know how far to go before
calling it quits, and what are the ramifications of
doing so? It should be obvious that the process
of informed consent is an ongoing give and take
between the doctor and the patient based on
what is occurring at any given point along the
treatment time line. Ideally, informed consent
should be continually obtained allowing you to
either continue on with treatment, or one's con-
sent can be withdrawn should the patient decide
to terminate treatment. Either way, this decision
is solely dependent on the sufficiency of the
information provided.
The following are two good ways to obtain
informed consent. The first involves a sending a
preliminary letter followed up by a conventional
informed consent form that is signed and kept
with the chart (Appendix B). This is a good
method for poor verbal communicators. The
second method (Appendix C) uses a checklist to
identify the factors to be discussed at the consul-
tation visit. It is for better communicators; how-
ever, it requires that the doctor memorialize the
<<    
     Article
      >>
Understanding the Basics of the Dentist
informed consent discussion with the patient in
the patient's record.
If you are like most orthodontists, when a new
patient calls for an appointment, you send them
a welcome to our office packet. It probably in-
cludes a letter thanking the patient for selecting
your office as well as confirming the time of their
appointment. You might also include a brochure
about the doctor, staff, mission, etc., a map
showing your location, and possibly you include
a health questionnaire to be filled out in ad-
vance. Let me suggest that you start obtaining
informed consent before you even meet the pa-
tient. Along with the other data, use the follow-
ing two letters, "Your Right to Know" and
"Information Concerning Your Treatment" (Ap-
pendix B). These two communiques set the
stage for an interactive relationship between the
doctor and the patient. Send the following letter
(Appendix B) with your "Welcome to Our Of-
fice" packet of information.
Your Right to Know
It's been said "an informed consumer is our best
customer." This has never been truer than in the
provision of health care services. What should
you expect when you first visit an orthodontist?
You should be told what the problem appears to
be, what tests are necessary to properly diagnose
the problem and formulate a treatment plan;
and, in the case of a child, whether now is the
appropriate time to begin therapy or whether it
is best to wait until a future date to initiate
treatment. Let's look at these individually.
When a patient either desires to have his/her
teeth straightened or is informed that orthodon-
tic therapy is recommended, he/she should be
told why it is in their best interest to undergo
such treatment. The most common reasons for
seeking treatment are (1) to improve cosmetics
(crooked teeth, overbite, and so on), (2) to fa-
cilitate other necessary dental work (to properly
position teeth for capping, bridges or implants),
(3) to correct a jaw discrepancy or skeletal dis-
harmony (one's bite is off), or (4) to help pa-
tients maintain their periodontal status (the
health of the supporting gums and bone).
The next step is the gathering of necessary
information by obtaining diagnostic records.
The following diagnostic materials are the one's
usually obtained: (1) radiographs of the teeth to
Home | TOC |          
Index
193
determine how sound they are and whether the
bony support for them is adequate, (2) a film of
the skull to see the relationship of the teeth to
the jaws and the jaws to one another, (3) pho-
tographs of the face to assess the profile and the
effect that any proposed tooth or jaw movement
may have on it, (4) models of your teeth for both
diagnostic purposes and to use as a baseline for
assessing one's progress and the results
achieved, (5) a clinical examination to check for
cavities and gum disease, and (6) a review of
your medical history as there are many underly-
ing physical problems that can impact on the
success or failure of orthodontic therapy. A con-
sultation between the doctor and the patient will
then follow.
At the consultation visit, your doctor should
explain to you what the actual problem is in a
language you can understand. You should be
told why correction is advised and how it is to be
achieved. You should also be informed of all
reasonable alternative methods of resolving your
particular problem. Because no form of medical
treatment is without the potential for some risks,
limitations, or compromises, you should be
made aware of those that pertain to your specific
situation. Next, you should be told what result
you can expect to achieve, and, finally, what will
occur if no treatment is undertaken.
At this time, the fee for the services to be
rendered should be discussed in full and suit-
able financial arrangements made between you
and your doctor. Make sure that you have had
the chance to ask and have answered all ques-
tions regarding your treatment as well as how
long it will take and the financial responsibilities
you are assuming. It is now time to begin treat-
ment.
The appliances or braces will be specifically
tailored to your particular problem. You may
require permanent and/or removable appli-
ances. Once they are attached to your teeth,
make sure you have been given careful instruc-
tions on how to care for them. Check to be sure
that your doctor will either be available or will
provide for coverage if an emergency (break-
age) arises. You also need to be informed of your
responsibilities if the best results possible results
are to be achieved. Some of these are maintain-
ing good oral hygiene, wearing rubber bands or
a headgear as instructed, and keeping regularly
scheduled appointments. You must continue to
<<    
     Article
      >>
194 Laurance Jerrold
see your general dentist at least twice a year
unless your orthodontist recommends other-
wise.
At the completion of active treatment you will
undergo a period of retention care. This phase
of treatment is necessary to monitor and help
maintain the results achieved. An appropriate
retainer will be fabricated to maximize the sta-
bility of the finished result. Remember, nothing
lasts forever, and some movement of your teeth
over the years is normal and should be expected.
Orthodontic therapy carries many benefits.
Evaluating these can only be done if you have
been given sufficient information on which to
base your decision to undergo care. You have a
right to know; to be educated is to be able to
choose wisely.
After the patient's diagnostic appointment,
when they are rescheduled for the consultation
appointment, give them two copies of the next
letter, "Information Concerning Your Treat-
ment" (Appendix B). Ask them to read it at their
leisure and bring one signed copy with them to
the consultation appointment.
Finally, for those who do not like the formal-
ity and associated stigma of the signed form
route, Appendix C is a checklist of risk factors
that should be identified and individualized for
each patient. In other words, not all of these
risks will happen to everyone. The doctor should
review this list when the diagnosis and treatment
plan is established, thus any potential risks will
not be transmitted in so over broad a fashion
Home | TOC |          
Index
that they become meaningless to any one indi-
vidual. In other words, the risk factors should be
case specific. Because it's not always possible to
know in advance what changes will occur, it
reenforces the notion of ongoing informed con-
sent. This checklist can then be placed in the
patient's chart indicating that the identified fac-
tors were discussed.
Summary
Understanding the doctor patient relationship is
the key to practicing good medicine. It is from
this relationship that the duty to conform to a
standard of care arises. The standard of care is
merely a euphemism for exercising sound clini-
cal judgment in a reasonable manner. Obtaining
a patient's informed consent, when done appro-
priately, can be a very rewarding practice, pa-
tient, and risk-management tool. It fosters a
close doctor patient relationship as all parties
are now on the same track regarding the diag-
nosis, treatment plan, and prognosis. Your pa-
tients will understand and appreciate more of
the complexities that are involved with their
orthodontic treatment, thereby having a greater
respect for the fees charged. It gives the doctor
an aura of openness, honesty, and caring. All in
all, if used as a sword and not a shield, appro-
priately obtaining informed consent benefits the
patient, the doctor, and the profession through
open communication.
<<    
     Article
      >> Home | TOC |          
Index

Understanding the Basics of the Dentist 195

Appendix A: Dismissal Letter for Active Patients

Dear :
Due to the fact that (choose from category 1-5 below), we must inform you that we are withdrawing from further
professional attendance to (Patient's Name) orthodontic needs.
Since her/his/your dental condition still requires further treatment, we urge you to continue your treatment with
another orthodontist without delay.
If you wish, we will be available to attend to (Patient's Name) orthodontic needs for the next (30, 45, 60)
days for emergencies or referrals only. This should give you ample time to select another orthodontist.
Should you authorize the release of (Patient's Name) records, we will be happy to forward them to you or an
orthodontist of your choice along with any other clinical information concerning the diagnosis and treatment rendered
by this office. (If you are going to charge a fee for the duplication of the records, state that amount here).
We regret having to take this action but the situation as noted above has left us no other option.
Respectfully,
(1) there has been a lack of cooperation on (Patient's Name) part which has been very detrimental to her/his/your
dental health and/or the achievement of an adequate orthodontic result
(2) we are unable to coordinate appointments and treatment with you and have been unable to do so for some time now
which has resulted in a compromising of our treatment goals
(3) you have not kept up with the financial obligations you agreed to regarding the professional services rendered
(4) you have not been honest and forthright in dealing with our office thus compromising our ability to render
appropriate professional services
(5) there are significant interpersonal differences and/or problems between you and members of our office staff which
have created disharmony or disruption to our office routine and activities resulting in interference with our patient care
activities

Appendix B: Information Concerning Your Treatment

GENERAL INFORMATION
Every medical or dental procedure carries some degree of risk and orthodontics is no exception.
Fortunately, most of the risks involved with undergoing orthodontic therapy don't even come close
to canceling out the tremendous benefits that patients achieve by undergoing orthodontic
treatment. Please indicate to us that you are aware of the following possibilities that are associated
with correcting your bite, straightening your teeth, and providing you with a beautiful smile, by
initialing the paragraphs listed below.
CAVITIES AND DECALCIFICATIONS Patient's/Parent's Initials
Braces don't cause cavities - poor oral hygiene does. Proper brushing will help reduce the likelihood
of developing cavities or decalcifications (white scarring of the enamel around the outside edges of
the braces). You must continue to see your regular dentist for check-ups at least twice a year.
Broken or looses appliances can also lead to these problems, so report any breakages to your
doctor.
PERIODONTAL (GUM) DISEASE Patient's/Parent's Initials
Some patients will experience various degrees of breakdown or loss of the bone and/or gums
supporting their teeth. While some patients are more predisposed to this condition than others,
most of the time, it is due to poor oral hygiene. If you experience this problem you may have to
see your dentist or a periodontist for treatment three or four times per year. On rare occasions,
your treatment may have to be interrupted or even discontinued if this condition cannot be
effectively treated or controlled.
ROOT RESORPTION Patient's/Parent's Initials
It is very common for the roots of some of the teeth to shrink in size during orthodontic treatment.
This root shrinkage even occurs in people who never wear braces. As long as your teeth and
supporting structures are healthy, a small degree of root shrinkage does not matter to your overall
dental health. However, as you grow older, if you develop periodontal disease as described above,
the combination of the two might affect the longevity of those teeth affected. If root resorption
becomes severe (a rare occurrence) treatment may have to be discontinued before it is completed.
<<    
     Article
      >> Home | TOC |          
Index

196 Laurance Jerrold

Appendix B: Information Concerning Your Treatment (Cont'd)

TMJ DISEASE Patient's/Parent'sinitials_

Your temporomandibular joint (TMJ) may become affected during treatment. Clicking, locking,
limitation of movement and/or pain in the jaw joints, or in the muscles of the head and neck have
all been reported to occur. This problem also occurs in people who have never undergone
orthodontics; and often, orthodontic therapy is often recommended to help correct a TMJ
problem. Usually this condition is temporary, disappearing after treatment is completed. If it
persists, you will need to see other doctors to address this problem.

ROOT CANAL Patient's/Parent'slnitials

On very rare occasions the nerve of a tooth may die during orthodontic therapy. This is almost always
caused by trauma (even from years ago) to the affected tooth or from a deep filling. If this occurs,
a root canal procedure to remove the dead nerve and save the tooth may be required.

RELAPSE Patient's/Parentslnitials
Throughout life, all tissues in the body change with the aging process. Your teeth and supporting
structures are no different. The retention phase of therapy can minimize the movement of your
teeth after treatment is completed. Teeth can shift position for a number of reasons such as
excessive and/or unanticipated growth, or uncorrected oral habits. Wearing your retainers as
instructed can help maintain the results achieved.

INJURIES Patient's/Parent'slnitials
All orthodontic appliances have the potential to cause injuries. Properly used and worn, this is usually
not a problem. Removable appliances may cause reactions to those allergic to acrylics. The braces
themselves may affect those with nickel allergies; and ceramic braces have been associated with
injuries to a tooth's enamel surface. The latex gloves we wear for your protection may affect those
sensitive to latex. If you experience oral trauma, the braces themselves have been known to make
some injuries worse while at other times, they have actually prevented more serious injuries from
occurring. Finally, headgears (night braces) and elastics (rubber bands) have been associated with
eye injuries; and there have been reports of patients swallowing parts of their braces. Your braces
are not toys; proper care and maintenance will minimize the possibility of any untoward
occurrences.

POOR RESULTS Patient's/Parent'slnitials

Orthodontics, like every branch of medicine, carries no guarantees. Sometimes, despite our best
efforts, the results achieved are less than anticipated; although they almost always result in
significant improvement. The most common reasons for achieving a less than ideal result are: the
decision to treat a limited problem as opposed to the entire problem; underlying skeletal or
periodontal limitations; limitations with regard to treatment rendered by other doctors, delays in
beginning treatment when recommended; and most often, poor cooperation by the patient as far
as not following treatment instructions and recommendations, not keeping appointments as
scheduled, and chronically breaking, loosening or losing the appliances.

NECESSARY SECONDARY TREATMENT Patient's/Parent'slnitials

Other treatment is often required in conjunction your orthodontic therapy such as routine dental
check-ups, caps, extractions, or gum treatments. You are responsible to ensure that these
procedures are performed in a timely fashion. Any fee associated with these treatments is separate
from the orthodontic fee charged.

STOPPING THERAPY BEFORE COMPLETION Patient's/Parent'slnitials

The doctor patient relationship needs to be respected by both parties. We will use our best efforts
and judgment in exercising our skill, knowledge, and experience to provide you with a quality
orthodontic result and experience. We will respect your confidentiality, your time, and your values.
In return we ask for your cooperation with regard to keeping scheduled appointments, following
all instructions, and promptly paying for services rendered. We reserve the right to discontinue
your treatment even though treatment may not be completed if problems of this nature persist
without adequate resolution on your part.

SUMMARY
The information noted above is solely so you, an informed consumer, can better appreciate that all
medical treatment, including orthodontics, carries some small downside risks. Fortunately these
risks are minimal, rarely happen, and can be easily dealt with should they occur. The tremendous
benefits associated with undergoing orthodontic therapy far outweigh any potential negative
occurrences associated with treatment. We encourage you to ask questions of us before, during and
after treatment so that you become one of our most important assets . . . a happy and informed
orthodontic consumer.

Signature of Patient or Parent if Patient is a

minor Date
<<    
     Article
      >> Home | TOC |          
Index

Understanding the Basics of the Dentist 197

Appendix C: Checklist for Informed Consent Regarding Risks, Compromises, and Limitations
[] Hygiene Related Problems
[] Caries and Decalcification
[] Root Resorption - Generally
[] Specific Morphology Prone to Resorption
[] Scout films necessary
[] Periodontal Complications
[] Rebound or Relapse Vs. Normal Tooth Movement
[] TMJ/MPD
[] Endodontic Involvement
[] History Of Trauma
[] Deep Decay/Restorations
[] Cuspid or First Premolar Roots Perforating Buccal Plate
[] Allergies
[] Acrylic Appliances
[] Latex
[] Nickel
[] Ceramic Brackets
[] Attrition
[] Opposing Cusp Fracture or Wear
[] Debonding Fracture
[] Removable Appliances
[] Ingestion
[] Aspiration
[] Headgear
[] Oral Surgery
[] Inability To Close Extraction Spaces
[] Inability To Close Osteotomy Sites
[] Potential Consequences Of Going After Impacted Teeth
[] Growth
[] Excessive
[] Unanticipated
[] Insufficient
[] Patient Cooperation
[] Secondary Restorative Treatment Needed
[] Skeletal Component with Associated Dental Compensations
[] Retention
[] Long Term, Lifetime
[] Fixed
[] Removable
[] Notice As To When Treatment Is Over
[] Prognosis Long Term
[] Limited Treatment Regarding Achieving Specific Objectives Only
[] Tooth Size/Arch Length Discrepancies
[] Need to Leave Spaces Post Treatment
[] Need for Restorative Post Treatment
[] Continued Deleterious Habits and Effect on Stability
[] Temporal Factors Relating to Treatment
[] Fees Associated With Treatment
[] What Is Included and What Isn't
[] Additional Fees
[] Amount and What For
<<    
     Article
      >> Home | TOC |          
Index

Dollars and Sense: Collecting Fees and

Dismissing Patients
Eric Ploumis

Money and ethics often conflict. But being effective in collecting what you
are owed is not in conflict with being ethical. There is nothing unethical
about charging a fair fee, rendering competent services, and collecting your
fee. This article explores issues that impact on fee collection efforts from
three different perspectives: legal, practical, and ethical. The legal nuances
associated with fee collections and the process for terminating nonpaying
patients from your practice is discussed. (Semin Orthod 2002;8:198-204.)
Copyright 2002, Elsevier Science (USA). All rights reserved.

A patient who does not know what to pay will

be less likely to pay that amount in a timely
fashion. Informing your patients of the cost of
orthodontic treatment is a prerequisite to suc-
cessful collections. It is also a federal law. Fed-
eral Truth-in-Lending Regulations (15 USCA
sec. 1601 et seq.) are federal consumer protec-
tion statutes first passed in 1968. These statutes
apply to most consumer transactions, including
the purchase of dental services. For the purposes
of the truth-in-lending statute, the orthodontist
is a creditor, and the patient a consumer. Com-
pliance requires that patients be clearly in-
formed in writing of their financial obligations.
This consumer protection statute is designed
to safeguard a consumer from creditors who
hide excessive charges in the fine print or fail to
inform the consumer of charges associated with
a transaction. Compliance with the federal truth-
in-lending statute does not require elaborate
forms. You should have a written contract with
your patients discussing your mutual obligations,
and a section of this agreement should review
your fee and the terms for payment to ensure
statutory compliance.
The regulations apply whenever a dentist in-
tends to charge a patient interest, service fees, or
From the New York University College of Dentistry, New York,
NY.
Address correspondence to Eric Ploumis, DMD, ]D, 110 Lock-
wood Avenue, New Rochelle, NY 10801.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0804-0003$35.00/0
doi:10.1053/sodo.2002.127865
late fees on any unpaid balance. The regulations
also apply when a dentist's contract with a pa-
tient permits payment over four or more install-
ments, not including the downpayment. Even if
there are no specific finance charges associated
with the installments, the law assumes these
charges are built in, and the dentist is still a
creditor for the purposes of the regulation. If
you discount your fee for payment in full up
front, the patient who pays in installments
should be made aware of this option in writing.
The Fair Debt Collection Practices Act
What steps are you legally allowed to take in an
effort to collect your fee? Collections efforts
must comply with federal, state, and local guide-
lines. Most state and local statutes are modeled
on a 1977 federal statute known as the Con-
sumer Credit Protection Act, also known as the
Fair Debt Collection Practices Act (FDCPA) (15
USCA sec. 1692 et seq.). Many state and local
debt-collection statutes are more restrictive than
the federal one; however, if they provide less
protection to the debtor, the safeguards built
into the federal statute prevail.
Knowledge of the FDCPA provides guidance
as to what constitutes abusive debt collection
practices and provides a reference point to more
local statutes. Understanding and avoiding abu-
sive debt-collection practices helps implement
collection protocols that are legal, ethical, and
effective. The FDCPA begins by stating, "There
is abundant evidence of the use of abusive, de-

198 Seminars in Orthodontics, Vol 8, No 4 (December), 2002: pp 198-204

<<    
     Article
      >>
Dollars and Sense
ceptive, and unfair debt collection practices. Ex-
isting laws and procedures for redressing these
injuries are inadequate to protect consumers."
Following the strictures of the statute, even when
they may not specifically apply to us, will provide
some guidance on how to proceed with collect-
ing fees. Under the FDCPA and many state col-
lection statutes, a dentist attempting to collect a
fee is prohibited from the following actions:
1. A dentist may not communicate, or threaten
to communicate, with the patient's employer
that a patient is in arrears. You are permitted
to call a patient at work, but you may not do
so if the patient or the patient's employer asks
you not to.
2. You may not threaten to take action against
the patient that is illegal, that you have no
intention of doing, or that you do not ordi-
narily take. For example, you may not
threaten to sue a patient for a fee if you have
no intention of doing so and, as a matter of
course, never sue to collect a fee. You may,
however, threaten to turn the patient over for
collection if you regularly do so. You may
never threaten a patient with physical harm.
3. You may not communicate with the patient or
any member of his/her family or household
with such frequency or at such unusual hours
or in such a manner as can reasonably be
expected to abuse or harass the patient. You
are only allowed to call a patient between the
hours of 8:00 AM and 9:00 PM unless you know
these hours to be disruptive; however, you
may call a patient at other hours if attempts to
reach him during the permitted hours are
unsuccessful. Although there is no generally
accepted calling frequency, legal precedent
suggests that contacting the patient more
than twice a week about an overdue account
is excessive.
4. You may not communicate with anyone other
than the patient, the patient's spouse, parent
(if the patient is a minor), guardian, execu-
tor, or administrator. You may not discuss
past due accounts with a patient's fiance, sib-
ling, child, coworker, or roommate.
5. You may not knowingly attempt to collect or
assert a right to any collection fee, attorney's
fee, court cost, or expense, unless such
charges are justly due and legally chargeable
to the patient. Unless your original contract
Home | TOC |          
Index
199
or truth-in-lending form specifies that you
will seek reimbursement for any fees incurred
as a consequence of your collection efforts,
you may not threaten to or add them on at a
later date.
Although the purpose of this statute is to
protect the consumer from unethical and overly
aggressive collections tactics, we shall see that it
is possible to comply assiduously with the letter
and spirit of the law and still pursue a delin-
quent account aggressively.
Statute of Limitations
All too often, suing a patient to collect a fee
triggers a retaliatory malpractice claim. By wait-
ing until they think the patient is no longer
within the statutory time frame to sue for mal-
practice, the savvy practitioner may think it is
now safe to initiate a suit to recover an unpaid
fee. It is only after misinformed dentists are
served with papers alleging malpractice that they
realize that their attempts to collect a small sum
have backfired. The following are a couple of
tolling provisions that extend the period of time
that patients have to initiate a lawsuit.
Toll for Infancy
If the patient is an infant, in most jurisdictions
this means under the age of 18, the statute may
be tolled (suspended) until the child reaches
majority. In other words, a child does not lose
his/her right to sue you for malpractice until
how ever long the statutory period is after his/her
18th birthday. This toll for infancy is very impor-
tant to us as orthodontists because we treat a large
number of minors. Often, we wrongly assume that
because the statute of limitations for malpractice
appears to have expired, aggressive fee collection
efforts may be pursued against the responsible
party, usually the parents without any malpractice
repercussions. Any collection effort should be tem-
pered with the knowledge that the parents can sue
on the child's behalf until the child reaches the
legal age or majority. After that, the child has the
right to sue on his/her own behalf, often for sev-
eral more years.
Continuous Treatment
In many jurisdictions, another exception to the
basic statute of limitations governing malprac-
<<    
     Article
      >>
200 Eric Ploumis
tice is called the continuous treatment doctrine.
Simply put, any continual treatment pertaining
to the initial therapy prevents the clock from
running for statute of limitations purposes. The
statute of limitations on a patient's claim does
not begin to tick down from the day you remove
the braces; it begins to run from when you either
formally dismiss a patient from your practice or
they come in for that last retainer check. We all
have patients who, despite being chronically be-
hind on their account, shamelessly manage to
come in month after month. Be advised that
each time they return to your office, they may be
resetting the clock on the statute of limitations,
hence making it more difficult to pursue their
delinquent account safely.
Dismissal Versus Abandonment
One of the thorniest problems we face is how to
dismiss a patient who fails to pay for treatment.
Misconceptions abound regarding how best to
dismiss an active patient or even whether we can
do so legally. The fact of the matter is that
dismissing nonpaying patients is an extremely
effective collections tool. Following established
legal guidelines will reduce the risk of being
accused of having abandoned the nonpaying
dismissed patient. When a dentist agrees to treat
a patient, a legal contract is formed. Each party
to this contract agrees to conform to both the
express and implied terms of that contract. For
example, the dentist agrees to competently per-
form certain procedures in exchange for remu-
neration from the patient.
The doctor-patient contract may be termi-
nated at any time during treatment. Although
the patient may end treatment at any time for
any reason, a dentist must be able to point to a
specific reason why treatment is being termi-
nated. Acceptable reasons to terminate treat-
ment usually involve the failure of the patient to
fulfill any of the express or implied contractual
duties he/she owes the dentist. If treatment is
being provided for an agreed on fee, nonpay-
ment of that fee is a breach of a patient's duty
and is a legally acceptable reason for a dentist to
terminate the doctor-patient relationship. If a
dentist decides to dismiss a patient for failure to
pay for dental treatment, he/she should take
steps to avoid a claim of abandonment by the
patient. Inform the patient of the arrears and
Home | TOC |          
Index
give them an opportunity to cure the default. If
a resolution is not forthcoming, send a certified
letter to the patient informing him/her of your
decision to discontinue treatment. The letter
should include the following: (1) the reason you
are discontinuing treatment, (2) an offer to treat
the patient on an emergency basis for a reason-
able period of time until alternative care can be
secured, (3) the importance of seeking contin-
ued care with an alternative provider and the
risks involved in the failure to do so, and (4)
notification that the patient's records are avail-
able to either the patient or to a subsequent
treating practitioner.
The key element in a dismissal letter is notice.
The patient must have notice that he/she is
being dismissed, when the dismissal becomes
effective, and notice of the consequences for
failing to seek alternative care. By following
these legal requisites when dismissing a patient
for nonpayment of a fee, you will minimize the
risk of a claim of abandonment by the patient.
Practical Considerations: Fundamentals
of a Fee-Collection Protocol
Having examined some of the laws relating to
collecting fees, it is now time to use this knowl-
edge to create a simple and effective protocol to
collect outstanding account balances. Before
any attempt is made to collect a delinquent ac-
count, indeed before an account even becomes
delinquent, three background essentials need to
be in place to insure successful collections. They
are (1) a clear understanding of who the patient
is and who the financially responsible party is,
(2) a clear understanding of when an account is
delinquent, and (3) a clear understanding of
what the fees are.
A Clear Understanding of Who the Patient
and Who the Responsible Party is
Before we can collect a fee, we need to know
who is responsible for paying it. If we need to
undertake a collections effort, we need to know
as much about the responsible party as we pos-
sibly can by the end of the initial visit. You will
never have a better information-gathering op-
portunity than during the initial encounter.
Look over your patient information form. At a
minimum, for collections purposes you should
<<    
     Article
      >>
Dollars and Sense
be asking the following: name of patient; age of
patient; address of patient; name of financially
responsible party (even for adults); address of
financially responsible party; telephone num-
bers, home, work (if a minor, mother and father
work), emergency; name of person who referred
patient; name of dentist; name of physician; and
social security number.
Each of these has a specific collections pur-
pose. Obviously, you need to know the patient's
name. His/her age is helpful for several reasons.
If the patient is a minor, he/she lacks the capac-
ity to form a legal contract with you. If you begin
treatment, a minor has no legal obligation to pay
you, and you may be committing a battery by
doing so. A patient's age also alerts you to the
statute of limitations issues.
The address tells you where to send the bill
and can be an instant credit check. Although
prejudging a patient's financial status is never a
sure thing, a patient's address can be an indica-
tor of his/her ability to pay and can suggest that
a customized payment plan may be advisable. If
possible, inquire how long a patient has been at
their current address.
Even if the responsible party is the patient,
asking who the responsible party is may rein-
force the fact that somebody is responsible for
paying you. Is the responsible party's address
different that the patient's? Is it a post office
box? If so, why? Is this a divorce case? If the latter
is the case, you need more information. Many
collections headaches arise from divorce issues,
and you do not want to be caught in the cross-
fire.
Get as many different telephone numbers as
possible. In the event that you need to pursue a
fee, calling at work usually gets more of a re-
sponse than calling at home. When you call
someone at work, you usually do not get an
answering machine, and, more often than not,
you will be put through directly.
Having the name of the person who referred
the patient, as well as that of the patient's dentist
and physician, will prove helpful later if you
cannot find a nonpaying patient. Chances are
that even if a deadbeat patient is not taking your
calls or answering your letters, he/she has not
severed all contact with their friends and other
health care professionals. In the event that you
need to skiptrace a delinquent account, these
numbers will assist you.
Home | TOC |          
Index
201
A person is not required to reveal his/her
social security number, but when someone
chooses not to, you have another credit alert.
Usually people who refuse to provide a social
security number on an information form do so
because they are aware of the value of the num-
ber in skip tracing and enforcing judgments.
The same goes for people who refuse to give a
street address or a telephone number and pro-
vide you only a postoffice box. They may be
privacy freaks, but they may also be poor payers.
With all of this in mind, review and revise
your patient history form. Make sure you are
asking enough information to enable you to put
a collections effort into play if necessary.
A Clear Understanding of When an Account
is Delinquent
Have a clear office policy of what constitutes a
delinquent account and what steps you need to
take to address one. The farther behind a pa-
tient's account gets, the less likely you are to
collect it. As orthodontists, we usually bill
monthly. An account that is two billing cycles
behind should be considered delinquent and
deserves immediate attention.
You should personally look over your state-
ments every month. It is your money that is not
being collected. Your staff gets paid each week
regardless of who has not paid you. If you stay on
top of your accounts and begin your collections
efforts promptly, the number of accounts you
have to deal with will shrink to next to nothing.
Then, and only then, can you delegate the job of
reviewing your statements to a trusted staff mem-
ber.
A Clear Understanding of What the Fees
Are
When we discussed legal considerations, we re-
viewed the need for a truth-in-lending form.
Either you or a trained staff member should
review the fee plan with the patient as carefully
as you do the treatment plan. Very often, pa-
tients listen more attentively to your fee presen-
tation than your case presentation. My prefer-
ence is to have the doctor discuss the fees with
the patient. If you delegate this duty, train your
staff to pick up the clues that may indicate that a
patient may be a slow payer. No matter who
<<    
     Article
      >>
202 Eric Ploumis
discusses the fee structure, make sure you leave
nothing to the patient's imagination.
The Collections Effort
By using our knowledge of the legal issues in-
volved and building on the fundamentals we just
discussed, the actual collections efforts should
go smoothly. You need to be systematic and
timely to be effective. The following is a sug-
gested protocol.
First Effort
Your first contact should be when a patient's
account is 30 days past due. Begin your collec-
tions effort by jotting a note on their statement
that reads something like; "Please call if you have
any questions about your account" or "Please
bring your account up to date." The doctor, not
a staff member, should write the note and either
initial or sign it. The response from those pa-
tients who "just forgot" is dramatic. This will be
your single most effective collections tool.
Often, the doctor is seen as distant from the
financial concerns of the practice. Often, the
doctor either feels discussing money is unseemly
(something I will address in the final section), or
he/she feels embarrassed discussing it. Patients
pick up on this, and those looking for a reason
not to pay you will have it. Unless you let your
patients know that you expect them to pay you
promptly, you should not expect to be paid
promptly. That is why it is important that you,
not a staff member, initial this first collections
effort. Try not to do this with a telephone call. A
telephone call takes too much of your time, is
usually delegated to a staff member who really
does not relish the job and will not do it well,
and more often than not requires repeated ef-
forts to successfully contact the debtor. All you
want is for the patient to bring the account up to
date. A letter, as opposed to a telephone call,
can be designed to invite only a positive re-
sponse.
Second Effort
Although the response to your first effort will
bring in many of your easily collected overdue
accounts, you will still have a handful of ac-
counts that require your attention. At this point,
you are in the 60- to 90-day past due range, and
Home | TOC |          
Index
you need to be a bit more decisive (actuarially,
after 90 days an uncollected dollar is worth only
72 cents). At this point you need to send a real
collections letter. It should inform the patient
how much they owe you and welcome them to
call your office to discuss his/her account. Be
sure to keep copies of all letters you send the
patient. Keep in mind that no matter what your
fee, patients think orthodontic work is expen-
sive. Your statement will often be permanently
positioned at the bottom of the patient's pile of
bills, and unless you take some action to move
yourself to the top you may never get paid. Un-
like the utility, cable, luxury car, or private
school bill, our bill does not have a must-pay
imperative attached to it. We have not condi-
tioned our patients to take it seriously, and there
is very little penalty attached to paying us late.
Remember, the truth-in-lending statute does
not allow for retroactively adding a penalty for
paying late. You should inform the patient, in
writing, at the initial fee presentation that there
is a penalty for paying late, either through an
interest charge or a late payment fee. You can
always waive the penalty, but you cannot add it.
Virtually every other creditor charges late fees,
and people are conditioned to paying them.
There is no reason we should not keep the op-
tion available to us and use it.
Third Effort
If your patient becomes more than 90 days plus
in arrears, you need to send a very decisive letter;
one that spells out the potential penalty in store
for failure to pay. Like the last letter, this should
not say too much, but it should inform the pa-
tient that you will not be able to continue treat-
ment unless the account is brought up to date.
Be careful not to give patients the idea that you
are abandoning them, but do let them know that
you reserve the right to dismiss them. What usu-
ally happens at this point is that the responsible
party calls to discuss developing a new payment
schedule. You are not obligated to accept this
overture and often fairly but firmly insisting on
being paid will get you paid. If you do choose to
work something out with the patient, do not
accept smaller payments over a greater number
of months.
Legally, what is occurring is that you and the
patient are forming a new contract. You should
<<    
     Article
      >>
Dollars and Sense
be prepared to dictate the new terms. Whatever
you agree to should be backed up with some
guaranteed payment method such as a series of
post-dated checks, a nonrecourse credit card au-
thorization, or permission to draw the money
directly from the patient's checking account
each month. A promise to pay that is not backed
up by one of these initiatives will put you right
back where you started when you send your next
statement out.
I do not like collections agencies. Why should
you give up 20% to 50% of your money to let
them collect for you? Collections agencies do
not know or care about your relationship with
the patient. Their heavy handedness often pro-
vokes patients, invoking a retaliatory lawsuit or
resulting in a complaint to a disciplinary or reg-
ulatory agency. Following the simple protocol
outlined earlier should be just as effective as
using a collections agency; the difference being,
you get to keep all of the money and still main-
tain a good rapport with your patient.
What if your patient promises to pay but in-
stead of paying the $200 a month you agreed on,
begins to send you check for $20 a month? If you
accept this check on a regular basis without pro-
test, you have reached what is legally known as a
new "accord and satisfaction." This means that
the patient has unilaterally renegotiated his/her
contract with you and you have accepted these
terms. What if your patient owes you $1,000 and
sends you a check for $500 marked "payment in
full." If you accept the check, you have indeed
been paid in full. There is nothing you can write
on the check that will negate the words paid in
full if clearly written. Your choices are to either
cash the check accepting it as full payment or
return it to the patient.
These efforts at collecting overdue accounts
will resolve most of your outstanding account
balances, but remember successful resolutions
decrease exponentially with the amount of time
the account is in arrears. If you still have not
been successful in working something out by this
point, you either have to accept the fact that you
are working for nothing or cut your losses and
dismiss the patient from your practice.
Dismissing the Patient
You have no obligation to continue to work for
nothing. If you follow the guidelines discussed
Home | TOC |          
Index
203
previously, you will be perfectly within your
rights to inform the patient you will not con-
tinue his/her treatment. In the legal section on
dismissal, we learned that notice is the key ele-
ment in patient dismissal. To be sure that the
patient gets fair notice, this letter must be sent
by certified mail, return receipt requested. The
legal presumption is that the addressee received
the letter even if it is returned to you. The
addressee is responsible for knowing what is in
that letter and failure to pick it up does not
absolve a responsible party of that burden. You
should send another copy of the dismissal letter
at the same time you send the certified one, but
place this copy in a plain white envelope and
handwrite their address. More often than not, it
is this letter to which the patient responds.
Virtually without exception, your office will
get a call right after the letter is received. Most
often, the response is one of incredulity, "I never
knew I owed you money," or "I never received a
bill." Take it in stride; insist that the account be
brought up to date and continue with treatment.
Dismissing the patient is a tactic of last resort,
but it is one that is certain to get results as long
as you do it by the book.
Suing a Patient
Suing a patient to collect your fee is not an
option in my collections protocol. If you imple-
ment this collections protocol as previously de-
scribed, you should never have an account that is
delinquent enough to warrant filing a lawsuit,
even in small claims court. Begin your collection
efforts early and cut loose the nonpayers early.
Structure your fee schedule so that even if you
have to dismiss a patient, you are at least at a
break-even point. A nonpayer understands why
he/she is being dismissed and will call you to
make arrangements to pay. If you sue that same
person to collect your fee, you are needlessly
antagonizing him/her, inviting him/her to re-
spond with a complaint to the state dental board
or to file a retaliatory lawsuit. Even if you win
your suit, you are often left with a judgment that
is difficult, if not impossible, to collect.
There is no reason that orthodontists, who
have the luxury of presenting clear fee arrange-
ments for an ongoing treatment plan, ever have
to resort to a lawsuit to collect our fee.
<<    
     Article
      >>
204 Eric Ploumis
Ethical Considerations
Can we, as health care professionals, reconcile
our duty to provide ethical and compassionate
care with aggressive fee collections? I submit that
it is the ethical practitioner who takes money out
of the doctor-patient equation. Having an office
full of patients who are up to date financially
creates an environment that allows us to focus
on what we do best—creating beautiful smiles.
Parents are aware when their child's account
is in arrears. They often do not come in with the
child for his appointment, break appointments,
or fail to respond to your efforts to discuss treat-
ment. In the end, the child suffers as treatment
drags on. Doctors have a tendency to get frus-
trated with the broken appointments and lack of
cooperation, ultimately just wanting to "get it
over with." Some doctors tell me they insert a
"financial wire" in the hope that by dragging out
treatment, they will eventually get paid. Not only
is this unethical, but it places the doctor in jeop-
ardy with the regulatory bodies in his state. Once
financial issues begin to impact on treatment
issues, you have begun sliding down the slippery
Home | TOC |          
Index
slope of ethical ambiguity. Taking money out of
the treatment equation is liberating for both the
doctor and the patient. We need to stop think-
ing that the business side of orthodontics has
nothing to do with the clinical side. They are
intimately related, and the sooner we acknowl-
edge this intimacy, the better we can render
top-quality care to our patients.
I am an attorney, but like you, I am also an
orthodontist. Because I wear two hats, I am privy
to the "secrets" of many of our peers. The pre-
ceding advice is the distillation of hundreds of
consultations with other professionals. It is field
tested. I urge each of you to consult with a local
attorney who is well versed in collections issues.
Have him/her assist you in shaping an effective
collections policy for your office.
This article looked at collections from three
distinct perspectives. Reread it at your leisure.
Pick it apart, customize it, and adapt it to fit your
personality and office structure. If you would
like samples of any of the letters or forms I have
mentioned, send me a stamped, self-addressed
envelope.
<<    
     Article
      >> Home | TOC |          
Index

The Scope of Vicarious Liability and Referral

Liability in the Orthodontic Setting: Am I My
Brother's Keeper?
Burton R. Pollack

The legal theory of holding an orthodontist liable to a patient who is injured

through the negligent act of another is described. The transfer of liability
from an employee to the employer under the legal doctrine of respondeat
superior is presented. The transfer of liability for referrals, having its basis in
the legal doctrine of vicarious liability is likewise discussed. Liability in
partnership practice, a major risk in the transfer of liability, is also described
in the section. Finally, various dental organizational entities, such as the
Professional Corporation or Association and the Limited Liability Company,
are discussed with regard to their effectiveness in avoiding the transfer
of liability. (Semin Orthod 2002;8:205-209.) Copyright 2002, Elsevier
Science (USA). All rights reserved.

he answer to the question asked in the title
T of this article, "Am I My Brother's Keeper,"
is yes but only when the brother has a specific
professional relationship with the brother's
keeper. Therefore, the purpose of this article
is to describe the relationships between the
brothers and when an innocent party (one
brother) will be held liable to an injured patient
as a result of the negligence of another (his
brother). This transfer or substitution of liability
is described in case law and is based on the
common law. It is not as a result of legislative
acts or the rules and regulations of administra-
tive agencies.
The discussion begins, as it must, with the
definition of terms used to describe the complex
legal doctrine of vicarious liability and a special
form of vicarious liability know as respondeat
superior that is limited to a master-servant or
employer-employee relationship between the
parties. A master or employer is one who en-
gages another to perform a service and who
controls or has the right to control that person
regardless of whether or not that control is ac-
tually exercised. A servant or employee is the
person directed to perform the service. An In-
dependent contractor is one performing a ser-
vice for another but is not under the control of
the employer as to the time, manner, or meth-
ods used to complete the service.
Two cases decided in Minnesota provide us
with legal definitions of vicarious liability and
respondeat superior. In the first case, decided in
1961, the court stated,
Vicarious liability is based on a relationship between the
parties, irrespective of participation, either by act or omission,
of the one vicariously liable, under which it has been deter-
mined as a matter of policy that one person should be liable
for the act of the other. Its true basis is largely one of public
or social policy under which it has been determined that,
irrespective of fault, a party should be held to respond for the
acts of another.1

From the School of Dental Medicine, State University of New
York at Stony Brook, Stony Brook, NY.
Address correspondence to Burton R. Pollack, DDS, JD, MPH,
School of Dental Medicine, State University of New York at Stony
Brook, Stony Brook, NY 11794.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0804-0004$35.00/0
doi:10.1053/sodo.2002.127866
Thus, vicarious liability is the transfer or sub-
stitution of liability, making an innocent party
liable for the negligent act of another. For the
substitution to take place, there must be a rela-
tionship between the parties. However, if the
relationship is one of master (employer) and
servant (employee), it becomes a subset of the

Seminars in Orthodontics, Vol 8, No 4 (December), 2002: pp 205-209 205

<<    
     Article
      >>
206 Burton R. Pollack
generic doctrine of vicarious liability and is
known as respondeat superior.
In an earlier case, the court defined Respon-
deat Superior by stating,
Under the doctrine of respondeat superior, according to
the generally accepted view, vicarious liability to third per-
sons is imposed upon the master [employer] for his servant's
[employee's] torts, not because the master is at fault, or
because he authorized the particular act, or because the
servant represents him, but because the servant is conducting
the master's business, and because the social interest in the
general security is best maintained by holding those who
conduct enterprises in which others are employed to an
absolute liability for what their servants do in the course of
the enterprise. Where the doctrine of respondeat superior is
relied on as a basis for recovery by a third person, the tortious
act of the servant committed in the scope of his employment,
and not the master's fault or the absence of it in hiring or
retaining the servant, is the basis of liability.2
In summary, vicarious liability is the transfer
of liability to an innocent party for the negligent
act of a guilty party; it covers all transfers of
liability regardless of the relationship of the par-
ties; and it is a generic term. Respondeat supe-
rior is the same transfer of liability but applies
solely to the employer-employee relationship.
Let us first examine respondeat superior as it
applies to the orthodontic practice setting. In
the typical orthodontic practice, there is likely to
be, in addition to the dentist/owner of the prac-
tice, one or more dental assistants, and in some
situations a dentist employee. These are the ser-
vants (employees) of the master (employer)
dentist/owner.
The following are several hypothetical situa-
tions that show the transfer of liability under the
doctrine of respondeat superior. In the first, the
orthodontist/employer on hiring a dental assis-
tant instructs him/her to exercise care in the
performance of his/her assigned duties. While
assisting the dentist in the placement of fixed
appliances, and as a result of the dental assis-
tant's negligence, the patient's cheek is burned
from the acid etch material requiring repair by a
plastic surgeon. The patient experienced consid-
erable expense, and the injury is permanent,
leaving a slight scar. The patient then com-
mences a malpractice suit naming the dentist
and not the assistant. In his/her defense, the
dentist states that the injury to the patient was
the fault of the assistant; the assistant was told to
be careful, and he/she was not. In this situation,
Home | TOC |          
Index
the court will find the dentist liable for the injury
to the patient despite his/her claim of inno-
cence for the assistant's negligent act. The the-
ory that courts apply to transfer liability is (1) the
dentist/employer hired the employee (assistant)
and thus placed him/her in a position to com-
mit a negligent act, (2) the employer benefits
(profits) from the services performed by the
employee, (3) the employer has the right of
control over all acts performed by the employee,
and (4) the patient played no part in the selec-
tion of the employee and was not able to control
the acts of the employee.
In other words, the major factor is that of
control, whether exercised or not.
In our second hypothetical situation, the
orthodontist/employer, when hiring the dental
assistant informs him/her of all the laws regulat-
ing the practice of dental assisting in the state in
which his/her practice is conducted. However,
the dentist informs the assistant that he/she
does not want him/her to perform a specific
service even though the statute regulating the
practice of dental assisting permits the task. Dur-
ing the dentist's absence, the assistant does what
the dentist had prohibited, and the patient suf-
fers an injury. A lawsuit follows. The dentist is
sued, not the assistant. The dentist's defense is
that he/she specifically prohibited the act the
assistant performed. Again, as in the prior situ-
ation, the court will find the dentist liable, ap-
plying the same legal logic.
Now for the third hypothetical set of facts.
The dentist, in hiring the assistant, gives the
assistant a copy of the law defining his/her per-
mitted duties and tells the assistant not to per-
form any duty not permitted by the law. During
the dentist's absence the assistant does, and the
patient is injured. The dentist is sued, not the
assistant. The dentist's defense is that the assis-
tant was aware of the laws regulating his/her
profession and violated the law despite the den-
tist's admonitions; thus, the assistant should
have been named as the sole defendant. Follow-
ing the same logic of the first set of facts de-
scribed earlier, the court will hold the dentist
liable. In summary, under the doctrine of re-
spondeat superior, the dentist hired, he/she
profited, he/she controlled, which is why he/
she should be responsible.
However, under common law, that is law
based on moral and ethical principles that
<<    
     Article
      >>
Am I My Brother's Keeper?
guides courts in decision making, a party injured
by the negligent act of another should be com-
pensated for the injury, and further, the party
guilty of causing the injury should be made to
provide the compensation. In the hypothetical
situations presented previously, how is the com-
mon law satisfied? Clearly, the injured party has
been compensated, but the guilty party, the as-
sistant, has not been held accountable. The ap-
parent defect can be corrected in that the em-
ployer is able to sue the negligent employee for
indemnification of his/her losses. This relief is
rarely, if ever, taken advantage of in the scenar-
ios described, but it does occur in a hospital
setting when the hospital is held liable for the
negligent act of a nurse, and the hospital's in-
surance company then sues for indemnification
of its losses from the nurse's insurance company.
The fact that the relief is available, even if rarely
exercised, satisfies the common law principle.
What courts rarely admit is that they often
look for the one with the "deepest pocket" in the
transfer of liability. An employee is not likely to
have the financial resources available to com-
pensate an injured party; the employer has the
"deepest pocket." An easily understood situa-
tion, unrelated to dentistry, is that if your auto-
mobile was struck by a Macy's truck that ran a
red light, who would you expect the court to
hold liable, Macy's or the driver of the truck?
Clearly, Macy's, despite that it can be shown that
all Macy's truck drivers were instructed to obey
the motor vehicle laws when making deliveries.
The issue of fixing liability on the party having
the "deepest pocket" will be addressed in some
of the following examples.
The employer-employee relationship as it re-
lates to nondentist employees in the transfer of
liability has been covered. However, there are
situations in dental practice, including orth-
odontic practice, in which the employee is an-
other dentist. In another hypothetical set of
facts, the employee/dentist performs a service
that results in an injury to one of the patients of
the practice. The injured patient alleging negli-
gence on the part of the employee/dentist has
the following options: (1) sue the employee
without suing the employer, (2) sue the em-
ployer without suing the employee, or (3) sue
them jointly.
In legal terms, the parties share joint and
several liability for each other's negligent act.
Home | TOC |          
Index
207
Under the doctrine of respondeat superior, the
employer/dentist may be held solely liable for
the negligent acts of the employee. Likewise, the
employee/dentist can be held solely liable to the
injured patient for his/her own negligence. The
joint suit option has some advantage for the
patient. In questionable situations, when it is not
clear to the plaintiffs attorney who will be held
liable by the court, it may be prudent to sue both
and let the court decide who is to compensate
his/her client for the injury. Another more com-
pelling reason to enter suit against both is when
the plaintiffs attorney discovers that one of the
possible defendants does not posses enough in-
surance coverage to compensate the patient for
the injury. It may be that to find the "deep
pocket" both will have to be sued.
There are other situations in which the doc-
trine of vicarious liability, exclusive of respon-
deat superior, may apply. These are not related
to the employer-employee relationship. They oc-
cur regarding referrals to and from other prac-
titioners.
First, let us examine liability issues in referrals
made by an orthodontist to an orthognathic sur-
geon for treatment when the surgeon is negli-
gent in his/her care of the orthodontic patient,
and the patient, as a result of the negligence,
suffers an injury. The attorney for the patient
discovers that the surgeon has insufficient insur-
ance to fully compensate his/her client for the
injury. So, the attorney joins the orthodontist in
the suit as a primary defendant. The jury may
find the orthodontist not liable unless (1) the
orthodontist knew, or should have known, that
the surgeon to whom the referral was made was
impaired or incompetent; (2) the orthodontist
took an active part in the treatment provided by
the referred to practitioner; or (3) the orth-
odontist benefited in some material way from
the referral.
Little need be said about the liability of a
referring practitioner when the referral is made
to a practitioner who is known by the referring
practitioner to be incompetent or impaired or if
the orthodontist took an active part in the treat-
ment provided by the referred to practitioner, as
in both cases, liability is easily attached. How-
ever, the third basis for the transfer of liability
requires some further examination. Clearly, if
the referring practitioner receives a financial
benefit from the referral, such as receiving a
<<    
     Article
      >>
208 Burton R. Pollack
percentage of the fee collected by the practitio-
ner to whom the referral was made, commonly
known as a split fee, the courts will transfer
liability to the referring practitioner. Another
situation to transfer liability is if the referring
practitioner receives a set amount of money for
each referral made, commonly called a kick-
back. A more complex situation may take place
when a generalist owns the office building in
which a specialist rents space; the rental is not
comparable to that of professional space in the
community, and it is shown that the generalist
refers all his patients needing specialty care to
the specialist who rents space in the building.
Thus, the owner practitioner benefits materially
(the increased rent that is tied to a percentage of
the specialist's earnings). Under these circum-
stances, the referring dentist may be held vicar-
iously liable for the negligent act of the special-
ist.
Gifts given at the end of the year from a
specialist to his/her referral sources are not con-
sidered such a sufficient material benefit to sup-
port a transfer of liability unless the gifts are
excessive in value.
Associate practice relationships also may re-
sult in the application of vicarious liability. Part-
nerships are particularly risky in the transfer of
liability. Suppose, while partner A is absent from
the office and without his knowledge partner B,
because of his negligence, injures a patient. Can
A be held liable to the patient? The answer the
courts have provided is yes. The legal logic is
that the partners are united in interest because
each partner profits from the services provided
by the other partner. Therefore, under the doc-
trine of vicarious liability, they may be held
jointly or severally liable as described earlier; the
plaintiffs lawyer has the choice of whom to sue
or both.
In another practice relationship two dentists
who are not partners share an office, and in it
they share a receptionist, the waiting room, the
dark room, and at times use each other's treat-
ment rooms. In addition, their names are on the
common door to the office. As a result of one
dentist's negligence a patient is injured. Under
these set of facts, will the courts treat them as a
partnership? Will the innocent dentist be held
either jointly or severally of liable? The innocent
dentist will claim they are not partners; in fact,
they are not. However, courts have ruled that if
Home | TOC |          
Index
it looks like a partnership to the patient, it will
be treated as a partnership by the court, and
each dentist may be held jointly or severally
liable.
This transfer of liability does not take place if
the practitioners are organized as a professional
corporation or professional association (title de-
pending on state law). Only the guilty party and
the corporation or association is held liable. A
new legal entity, the professional limited, liabil-
ity company, now often takes the place of the
professional corporation or association. It also
protects individual members from the transfer
of liability and is available to members of the
dental, medical, legal, and other professions us-
ing the abbreviations PLLC or similar designa-
tion as permitted under local law.
Some courts have held that when the relation-
ship is one between a principal and an indepen-
dent contractor, the principal may be held liable
for the negligent acts of the independent con-
tractor. As defined previously, an independent
contractor is one performing a service for an-
other but who is not under the control of the
employer as to the time, manner, or method to
complete the service. However, courts look care-
fully into the relationship, often rejecting the
independent contractor relationship in favor of
one of employer-employee, applying the doc-
trine of respondeat superior in transferring lia-
bility.
Summary
Under the legal doctrine of respondeat supe-
rior, an orthodontist will be held liable for the
negligent acts of his/her employees, including
acts of an orthodontist employee (but not nec-
essarily an orthodontist who is an independent
contractor); he is his brother's keeper.
Under the legal doctrine of vicarious liability,
an orthodontist may be held liable for the neg-
ligent acts of a practitioner whom the orthodon-
tist has referred his patient to (if he/she exer-
cises a degree of participation or control over
the referred to doctor's treatment or, if the re-
ferral was negligently made); he/she may be his
brother's keeper.
Under the same doctrine of vicarious liability,
an orthodontist can be held liable for the negli-
gent acts of his/her partner orthodontist (or
those doctor's who the public might ostensibly
<<    
     Article
      >> Home | TOC |          
Index

Am I My Brother's Keeper? 209

believe are united in interest with him/her); he
may be his brother's keeper, at the option of the
injured patient's attorney.
Orthodontists who practice as a professional
corporation, professional association, or profes-
sional limited liability company are immune
from vicarious liability for the acts of other mem-
bers of the professional association or profes-
sional limited liability company; he is not his
brother's keeper,
References
1. Nadeau V. Melin, 110 N.W. 2d 99, (Minn. 1961).
2. Porter V. Grennan, 16 N.W. 2d 906, (Minn. 1945).
<<    
     Article
      >> Home | TOC |          
Index

Why Orthodontists Get Sued

Elizabeth Franklin

The question, Why do orthodontists get sued?, is a vexing one for most
orthodontists. The circumstances that evolve to make one person challenge,
verbally attack, or sue another is a very individual issue that involves
multiple aspects of the interpersonal relationship. Doctor-patient relation-
ships are not infallible. They involve the personalities of each of the partic-
ipants, the activity on which the relationship is founded, the relative success
or failure of the treatment, communication or the lack thereof, and possibly
psychological issues. The bottom line is that orthodontic lawsuits can and
do happen. This article describes a number of orthodontic lawsuits, the
basis for each suit, and the risk management lessons that one should take
home today and put into use tomorrow. (Semin Orthod 2002;8:210-215.)
Copyright 2002, Elsevier Science (USA). All rights reserved.

As claims manager for the American Associa-
tion of Orthodontists Insurance Company
(a Risk Retention Group), which writes profes-
sional liability insurance, I have had a detailed
view into the lawsuits that are filed against orth-
odontists. During the handling of those lawsuits,
I have been able to look into the practices of the
involved doctors and have made some general
observations about the causes and effects that
most frequently appear to have contributed to
the filing of the actions. Often, these patterns
repeat themselves, notwithstanding major varia-
tions in a patient's diagnoses and treatment, the
locations of the practices, and the doctors in-
volved.
What are some of the major issues that I see?
Communication breakdown is an important
one. I am referring to listening and speaking
skills, providing critical information to the pa-
tient, and the coordination, or lack thereof, with
other health care practitioners. Secondly, good
initial records are obviously needed to formulate
appropriate diagnoses and treatment planning;
they are also the front line in the defense of a
From the American Association of Orthodontists Insurance Com-
pany, St Louis, MO.
Address correspondence to Elizabeth Franklin, AAOIC, 401
North Lindbergh Blvd, St Louis, MO 63141.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0804-0005$35.00/0
doi:10.1053/sodo.2002.127867
lawsuit. Next is the securing of detailed in-
formed consent consultations with the patient
and/or family, followed by placing a signed ac-
knowledgment of this act in the patient's record.
Finally, being able to empathize with the patient
and focus on their treatment needs are also
critical in this analysis.
To illustrate these general observations, I will
summarize the details of several actual lawsuits
handled through our office, minus, of course,
any identifying information to protect confiden-
tiality. The summaries will examine the nature
of the patient's complaint. What did the doctor
say and/or do preceding the filing of a com-
plaint? Were office staff members or office pro-
cedures involved and in what way? What treat-
ment was performed, and what was the
outcome? Was the outcome a surprise to the
patient or family; was it a surprise to the doctor?
What was the demeanor of the patient during
the course of treatment? Were there signs of
dissention, such as arguments with the doctor or
staff, requests for records, or concerns about
fees? One doctor told me that orthodontists
love to work on teeth; the problem is, he said,
the teeth have humans attached. It is the
breakdown of the humanity inherent within
the relationship that results in lawsuits. Let's
look at some routine human problems to see if
they can be tempered enough to minimize
potential exposure.

210 Seminars in Orthodontics, Vol 8, No 4 (December), 2002: pp 210-215

<<    
     Article
      >>
Why Orthodontists Get Sued
Case One
Facts
The first claim involved inadequate treatment
progress and exacerbation of periodontal dis-
ease in a 40-year-old professional woman. Her
treatment lasted 3 years. The initial problems
included a 4-mm skeletal open bite, with protru-
sion of the upper central incisors. The orth-
odontist prescribed orthognathic surgery. The
orthodontist's chart contained a referral to an
oral surgeon, and there was also a letter to the
surgeon in the records. There was no informed
consent documentation present in the records;
the doctor and the patient dispute whether it
was actually given. The patient declined the sur-
gical option, and brackets were placed. Rubber
bands and elastomeric chains were used to cor-
rect the malocclusion.
As treatment progressed, the open bite closed
but then relapsed. The patient's gums became
inflamed and bled. The patient's general dentist
sent her to a periodontist who diagnosed peri-
odontal disease, and he recommended termina-
tion of the orthodontic treatment. The patient
then sought a second opinion from another
orthodontist. The second doctor determined
that, despite 27 months of treatment, the patient
was only biting on her molars. He found the
upper central incisors to be "rabbited in," the
upper right canine and premolar depressed, and
an open bite of 4 mm extending from the upper
right central incisor to the upper right first mo-
lar. He observed that the patient had experi-
enced serious root resorption, significant bone
loss, and periodontal disease.
Two years after these observations, the patient
filed a lawsuit against the first orthodontist, ask-
ing for damages of $500,000. She alleged she
would lose eight teeth because of root resorp-
tion and bone loss. She also alleged that the
extraction of the left first bicuspid during the
course of treatment was inappropriate, and the
loss of that tooth was, therefore, also part of the
damages. In addition, because a number of
teeth were no longer in contact, she needed
multiple units of prosthetics to correct her oc-
clusion.
By the time the lawsuit was filed, the insured
orthodontist had passed away. His office had
closed; all employees had taken other jobs, and
none were available to testify on behalf of the
Home | TOC |          
Index
211
doctor. His estate was, therefore, unable to de-
fend him during the proceeding beyond what
was written in his records. The case went to trial.
After 2 days of testimony, the jury returned a
verdict in excess of $300,000 in favor of the
plaintiff.
Obviously, this was an adverse result. The
plaintiff presented evidence that root resorption
was visible in the initial pretreatment radio-
graphs. This should have been an indication to
the doctor to monitor the teeth carefully during
the treatment and carefully assess the forces
used during treatment. During her deposition,
the patient testified that given her periodontal
problems, she believed she was not a good can-
didate for orthodontic treatment and that the
doctor should have advised her of this fact. In
addition, the patient's testimony indicated that
the orthodontist's listening and empathy skills
were lacking. The plaintiff indicated she contin-
ually complained of pain and discomfort, but
the orthodontist did not seem to listen. She
testified that the doctor was unsympathetic, un-
concerned, and unduly critical of her because of
her complaining. In addition to missing the root
resorption on the initial radiographs, he was also
accused of overlooking signs of bone loss and
exacerbating periodontal disease as the treat-
ment progressed, and he missed the opportunity
to make an appropriate referral to a periodontal
specialist.
The subsequent treating orthodontist testi-
fied that the power chains used by the insured
were inappropriate and caused the unnecessary
pain experienced by the patient. He also be-
lieved that the placement of appliances contrib-
uted to the gingivitis and periodontal disease.
The records were unsupportive in that there was
no signed informed consent form, and there was
no documentation regarding important conver-
sations that had occurred and treatment deci-
sions made between the doctor and the patient.
Risk Management Recommendations
From the standpoint of managing an orthodon-
tic practice, it is imperative to obtain good
records, including radiographs, before treat-
ment begins. Once treatment begins, one must
also continue to maintain thorough, legible, and
clear records. They not only facilitate good treat-
ment but also help discourage litigation in the
first place. Finally, they are a primary source of
<<    
     Article
      >>
212 Elizabeth Franklin
defense in the event that a malpractice claim is
made.
Good communication with the patient or
family is another effective way a doctor can pre-
clude a lawsuit. Listen to the patient. Is there
pain, bleeding, a compromise in function? Ad-
dress these issues satisfactorily and explain them
fully. A patient who understands what is taking
place, the reason why specific treatment or re-
ferrals are recommended, and believes that the
doctor is concerned and sympathetic, is less
likely to sue.
Always have an informed consent discussion
with the patient. Emphasize those risks that are
most important to his/her particular treatment
and repeat these concerns during treatment,
redocumenting this discussion. Have the patient
sign a form memorializing the conversation and
keep it with the records. This process manages a
patient's expectations and minimizes unpleasant
surprises. A patient who realizes they have been
clearly apprised of potential problems will have
far more difficulty suing should that problem
materialize.
Case Two
Facts
The patient in this case was a 10-year-old child at
the start of treatment in 1992. The diagnosis was
an anterior protrusion. His treatment lasted 5
years, and over that period of time, he presented
with more than 60 instances of loose bands or
broken appliances. He missed a significant num-
ber of appointments. The doctor documented
poor oral hygiene several times in the records.
In 1997, a general dentist examined the patient
and found extensive decay and demineralization
in 50% to 80% of the patient's teeth. Most of this
damage was visible to the naked eye. After the
orthodontic appliances were removed, the areas
of decay and lines of decalcification were consis-
tent with the placement of the bands and brack-
ets. The patient's current general dentist opined
that this condition should have been obvious to
any dentist, especially an orthodontist. He im-
mediately cleaned the teeth and applied a resin
to provide temporary relief in the sensitive areas.
He anticipated that multiple units of restora-
tions would be necessary in the future and that
some root canals would also likely be required.
The family filed a lawsuit against the orthodon-
Home | TOC |          
Index
tist and the general dentist who had seen the
patient only a few times during that 5-year pe-
riod. The dental records did contain a signed
informed consent form.
The expert witness who examined the records
on behalf of the doctor opined there was evidence
of neglect and recommended the case be settled.
He based this decision on several factors. First, the
depth and degree of the decay substantiated its
presence for a number of years. Secondly, because
the caries was located under the bands, he be-
lieved that washout was evident and no amount of
adequate tooth brushing would have prevented
the damage; and the significant number of re-
placed or recemented bands also indicated inade-
quacies with cementation. In addition, he also
noted that the records were written in code with
very little detail. Notations of "poor oral hygiene"
were not explained. The plaintiff also contended
that much of the treatment was performed by staff
rather than by the doctor.
The expert also believed there was another
significant issue that encouraged early settle-
ment of this case. The patient had testified that
late in 1996 he began taking Acutane (Roche
Labs, Hoffman-LaRoche Inc., Nutley, NJ) for
treatment of severe acne. This drug is known to
cause xerostomia that can contribute to gingival
inflammation and tooth decay. The boy said he
apprised the orthodontist of this fact; however,
the orthodontist disputed this. The expert be-
lieved that the presence of acne in a teenager
should alert a doctor to the possible use of this
drug. Although he opined that the orthodontist
should have made an inquiry, because the dam-
age was under the bands, he did not believe the
Acutane was the primary cause of the decay.
Until the time this case was settled, the pa-
tient's family had spent approximately $20,000
on restorations. The initial demand had been
$350,000; the case was settled for $150,000. What
went wrong here? The first issue appears to be
one of office organization. The doctor reported
that he was frequently absent from his office
handling personal matters. He believed that it
was his female partner who may have done some
of the work; however, the plaintiff could not
differentiate whether it was she or any of the
dental assistants who was performing a signifi-
cant amount of the treatment. The patient tes-
tified that the doctor was not paying sufficient
attention to his care. The records did not reveal
who performed which procedures. In short, they
<<    
     Article
      >>
Why Orthodontists Get Sued
contained little detail and were incapable of pro-
viding much support during the defense of the
case. This was especially true regarding the doc-
umentation of poor oral hygiene. Those entries
were not clarified, and the parents testified that
they were never made aware of any problems in
that regard. Had there been details provided in
the notes, perhaps even initials from the young
man acknowledging these discussions, prefera-
bly with copies of the letters sent to the family
apprising them of the problems, there might not
have been a lawsuit. Better records would have at
least made the case easier to defend.
The matter of the excessive number of bro-
ken appliances was a shouting match between
the patient and the doctor. The doctor believed
the boy was to blame for eating inappropriate
foods. The boy, of course, denied that he did so.
The expert believed that the number of repairs
and replacements was far too extensive to be
caused by anything other than poor bonding
and banding technique. One very important
point to be considered from a review of this case
is the possible need for early termination of
treatment secondary to noncooperation. If the
doctor believed there was poor oral hygiene and
that too many bands needed replacing, perhaps
he should have stopped the treatment when the
problems continued. This would have prevented
the exacerbation of the damages.
Orthodontists are charged with the responsi-
bility of recognizing and addressing (usually by
referral to other specialists or general dentists)
routine dental issues, such as caries or periodon-
tal disease. Merely having a sign on the office
wall advising patients to routinely see their gen-
eral dentist for cleaning and examination does
not insulate the orthodontist from responsibility
if the patient fails to comply. It is the orthodon-
tist's responsibility to determine compliance
when significant problems persist. Communi-
cate directly with referral doctors to ensure it is
safe and appropriate to continue orthodontic
treatment. Refusal to continue treatment until
there has been compliance must be seriously
considered. That may interfere with your pro-
jected treatment schedule, but it will possibly
save you from having to defend a lawsuit later
after the damage occurs and the patient forgets
that you told them to have regular checkups. In
the end, you will be held accountable.
The issue involving Acutane provides another
interesting observation from a risk management
Home | TOC |          
Index
213
perspective. It appears that information about
the effects of this drug on dental health is com-
monly available. Because acne is a frequent teen-
age concern, a careful orthodontist needs to be
proactive and make an appropriate inquiry, es-
pecially if decay seems to be a problem. Any
change in the patient's complexion should be
readily visible. In this case, the doctor seemed to
be unaware of many details regarding his pa-
tient, perhaps because of overdelegation. Better
focus and attention to detail might have pre-
cluded the lawsuit.
Risk Management Recommendations
Beware of overdelegation of duties to assistants
as small but important details can go unnoticed.
Regardless of the problem, if a patient is not
keeping appointments, has poor oral hygiene, is
eating inappropriate foods, or has excessive
damage to the appliances, the orthodonist
should advise the patient and his parents of the
consequences. Communicate with the patient,
and in the case of minors, at least occasionally
with the parents. In addition, notify the general
dentist of any problems with the patient's treat-
ment. Document all of these communications in
the patient's chart. Ensure that the patient is
regularly visiting his general dentist for cleaning
and checkups; document this fact, as well.
Case Three
Facts
In this case, the patient was a young adult fe-
male. Her diagnosis was a Class III malocclusion
secondary to a combined mandibular progna-
thism and slight maxillary retrusion. The exam-
ination revealed a facial asymmetry, abnormally
shaped lower anterior crowns, a midline devia-
tion, and partially impacted third molars. The
treatment time was projected to last 2 to 2V£
years. The plan was to slenderize the abnormally
shaped teeth. The doctor did not want to extract
lower premolars because he believed that would
flatten her profile and subject her to other den-
tal problems such as root resorption or peri-
odontal disease. He anticipated extracting the
third molars at the end of treatment, which he
believed would allow for settling of the upper
and lower arches.
After 3 years of treatment, the braces were
removed, and the patient's maxillary anterior
<<    
     Article
      >>
214 Elizabeth Franklin
teeth were significantly flared and tipped labi-
ally. An anterior open bite was present, and lip
closure was difficult. The patient alleged her
dentition was deformed, and she was facially
disfigured. When she expressed consternation
to the doctor, she said she was abruptly dis-
missed from the orthodontist's practice. At that
time, he gave her a small fee refund of less than
$500. She immediately saw another orthodontist
who extracted eight teeth, retreated for 2 more
years, and achieved an excellent final outcome.
She then sued the first orthodontist.
The plaintiffs expert alleged that the doctor
failed to conduct a tooth mass to arch length
analysis and that there was insufficient space
without resorting to extractions. He also opined
that 3 years of treatment should have been suf-
ficient time to achieve success of this patient's
problem. The expert witness who reviewed the
records on behalf of the doctor opined he was
satisfied with the original treatment plan be-
cause he could see that the insured intended to
extract the third molars at the end of treatment
to add to the space. He agreed with the propo-
sition that root resorption and periodontal dis-
ease could possibly be a problem for the young
woman if the premolars were extracted. How-
ever, when he saw the plaster casts showing the
severe flaring, he admitted that extractions did
appear to be indicated. Nevertheless, he had no
complaint with the length of treatment and be-
lieved that had the patient remained in the in-
sured's practice, she would have achieved a good
final result.
Why then, did this patient sue? Was the initial
diagnosis unsatisfactory? The allegation of the
tooth mass/arch length discrepancy and the
lack of such an evaluation was challenging to
overcome. The records were devoid of such a
computation. The doctor eventually produced
one well after the records had been submitted to
the plaintiff. It was written on a small post-it
note, and was not impressive evidence. It had
the appearance of having been created after the
fact. That negative impression no doubt pro-
vided an incentive to the plaintiff to sue.
In response to that allegation, the insured
testified that he had always been aware of the
tooth mass/arch length discrepancy. He charac-
terized his treatment plan as conservative and
appropriate, given the diagnosis. Even though
the insured made an adequate witness, the fact
that he might have created evidence after the
Home | TOC |          
Index
fact would have diminished his credibility. The
young plaintiff on the other hand was a sympa-
thetic witness. She testified that when she ex-
pressed dissatisfaction with the outcome of her
treatment, she was immediately dismissed from
the practice. She felt abandoned and very em-
barrassed by the condition of her teeth.
The damages in this case were not significant;
the plaintiff alleged expenses of $7,000, includ-
ing subsequent orthodontic care and the costs
for the extractions. She was upset at having to be
in braces for a total of 6 years. She initially
demanded $125,000 to compensate her for the
additional treatment, her embarrassment, pain,
and inconvenience. After investigation, defense
counsel seriously recommended settlement of
this case before trial. Although both parties were
good witnesses, the post-it note addendum to
the records had a negative effect. In addition to
that, once the expert witness testified that ex-
tractions were indicated, the fact that the teeth
were badly flared at the end of 3 years became
more difficult to defend. The dismissal of the
patient from the practice was also a problem.
The case was settled for $37,000.
Risk Management Recommendations
From a risk management perspective, records
and their quality are once again seen as critical
to the prevention of and subsequent defense of
a case. As we have said before, incomplete or
inadequate records provide little support against
a plaintiffs allegations. Records that have an
appearance of being altered are deadly. They
create such doubt in the minds of jurors that
they can erase any other good supportive factors
in a case. Never engage in records alteration.
As we saw in the other two cases, communi-
cation deficiencies appeared to influence this
case. Was the treatment plan carefully explained
so that the patient understood why the doctor
chose to treat without extraction? Was a good
informed consent discussion held, and did the
patient sign a form to that effect? When the
treatment time extended past the original esti-
mate, was an explanation provided? What aspect
of communication broke down so unequivocally
that the doctor felt compelled to dismiss the
patient? It seems easy to understand in this case
why the patient felt dissatisfied and abandoned
by the original orthodontist.
<<    
     Article
      >>
Why Orthodontists Get Sued
Tying It All Together
We have now analyzed three very diverse cases.
One occurred on the West coast, another on the
Eastern seaboard, and the third happened in the
South. One treatment period was extremely pro-
tracted; the other two were normal or close to
normal. The patients were different—one was a
middle-aged adult female, another a male child,
the third a young adult female. All of these
patients sued their orthodontist.
There are many factors that go into the actual
filing of litigation about which we cannot be
aware. It is commonly believed that a conversa-
tion with a lawyer at a cocktail party or one living
across the backyard fence may encourage some
litigious behavior. Television and other media
forms also frequently emphasize success in mal-
practice litigation and that motivates real-life lit-
igation. In other words, some lawsuits will be
filed, no matter what. Those of us involved in the
handling of professional liability claims and law-
suits contend that knowing the bases for some
problem patterns of behavior and interrupting
them can prevent some lawsuits from occurring.
This ultimately provides some protection to
orthodontists.
Let's review the most obvious forms of pro-
tection. One is to improve communication skills.
Listen to the patient or their parents. Speak to
them, carefully explaining what is expected to
happen, and what is actually happening. Explain
negative behavior or other problems that are
impeding the success of the treatment. Contin-
ually communicate with referral doctors to be
sure that the overall dental health of the patient
is the primary focus, not just the orthodontic
treatment.
Focus on the treatment. If it is not progress-
ing satisfactorily, take note. Is it taking too long?
Is the occlusion not being corrected? Make
changes if necessary. Focus on the patient. What
issues about them are occurring that might af-
fect the success of the treatment? Empathize
with the patient. Realize that each person has a
different perspective to pain, motivation for
treatment, and esthetics. Try to place yourself in
their position and understand what they are ex-
periencing. Let them know that you understand
and are concerned.
Take good records. Take enough radiographs
Home | TOC |          
Index
215
before, during, and after treatment so you know
what is happening physiologically. Record
clearly and concisely all critical aspects of the
treatment, pertinent conversations, complaints,
and occurrences. Document all instructions and
advice that you give and keep copies in the
patient record. Make it a habit to have an in-
formed consent consultation with every patient,
tailored to their specific needs. Then, place a
signed form in the records as proof of the con-
sultation. Renew the informed consent if the
treatment changes.
If there is one thing that I have learned over
the years of handling dental malpractice claims,
it is that no dentist enjoys being sued. Some are
more willing than others to go through the trial
process, but none of them enjoy it. It is time
consuming at the least and antagonistic and of-
fensive at its worst. It requires hours of answer-
ing written and oral questions about every detail
of the office procedures, the diagnosis and treat-
ment, the records, and the communication or
lack thereof. Doctors who are sued spend a sig-
nificant amount of time working with an attor-
ney to formulate and present a defense. They
spend hours in preparation, pretrial settlement
forums, and listening to constant accusations of
wrongdoing. The trial itself is more of the same,
only more intense, with one witness after an-
other alleging practice below the standard of
care.
If the emotional toll of litigation is not bad
enough, the financial cost is worse. Not only is
defense expensive in terms of legal charges and
expert witness support, but in addition, all of the
time a practitioner spends defending a lawsuit is
time away from a money-making practice. Most
doctors I know object to that.
As I speak with dentists, I find that the moti-
vation to understand and make necessary
changes to preclude or minimize exposure to
lawsuits is generally high these days. Our society
has become increasingly litigious in some areas
of the country, with some cities being worse than
others. Orthodontists must therefore be vigilant
and recognize their own role in the process.
They must make appropriate changes in their
practice and in their relationships with patients
to protect themselves from a malpractice lawsuit.
<<    
     Article
      >> Home | TOC |          
Index

The Anatomy of an Orthodontic Lawsuit:

From the Summons Through Trial
Toni Reale

Being sued for malpractice is one of the most harrowing experiences any
practitioner can go through. Understanding all of the steps involved in the
legal process is the first step toward being able to deal with the angst that
accompanies such an action. The article will cover a multitude of facets of
medical malpractice litigation such as the service of process, jurisdiction,
notification, maintenance of records, depositions, pretrial motions, the trial
itself, and a cadre of other facts that practitioners should be aware of.
(Semin Orthod 2002;8:216-219.) Copyright 2002, Elsevier Science (USA). All
rights reserved.

It starts out as any other day in the office, and
then your receptionist calls you out to the
waiting room. The person standing there calls
out your name and thrusts papers into your
hand stating, "Doctor, you have been served." At
this point, if you have not had any experience
with the legal system, numbness and then panic
sets in. First, take a deep breath then, call your
professional liability insurance carrier. Do not
call the patient or his/her attorney in an attempt
to explain that there must be a mistake. Such a
call will do no good and may actually be detri-
mental to your interests. Rarely, if ever, is a
defendant released from a lawsuit after talking
to the plaintiffs attorney.
When you call your malpractice carrier, have
in front of you your policy number and the
patient's records. You will be asked for the dates
of treatment and a description of the treatment
you provided to the plaintiff. The dates are im-
portant to determine if the lawsuit has been
brought in a timely manner; each state has a
time limit pertaining to when the plaintiff can
institute a lawsuit. This is called the statute of
limitations. These time periods been enacted to
From Medical Liability Mutual Insurance Company, East
Meadow, NY.
Address correspondence to Toni Reale, Esq., 23 Fern Street,
Floral Park, New York, NY 11001.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0804-0006$35.00/0
doi: 10.1053/sodo. 2002.12 7868
encourage lawsuits to be filed in a timely man-
ner.
It is important to maintain treatment records
concurrent with when the treatment was ren-
dered. Keep all entries intact and unchanged. If
you need to correct an entry, the date of the
correction should be noted and the entry made
at that time. It is not necessary to write that you
spoke to your insurer in the record. Remember,
a dental record is a recognized business record
of the treatment provided a patient and all facts
pertinent to that treatment. If there were any
communications with the patient, regarding dis-
satisfaction with treatment or any other com-
plaints, these should be noted at the time they
were made, as well as any problems encountered
during the course of the treatment. Do not make
any changes or explanations on the record in
hopes of clarifying your notations at any time
after the original entry was written. Any changes
in a record may be a criminal act, a felony in
some states, and are surely actionable against
one who does so. If not actually a crime, a state's
Dental Practice Act usually prohibits records al-
terations and doing so may result in censure or
even suspension of your license.
The course of a lawsuit is generally a long
one. There are several stages to the litigation
process. Once you file a report with your carrier,
you set in motion the mechanism for protecting
your interests. Initially, you will be required to
send your carrier, along with the legal papers
you received, copies of your complete records.

216 Seminars in Orthodontics, Vol 8, No 4 (December), 2002: pp 216-219

<<    
     Article
      >>
The Anatomy of an Orthodontic Lawsuit
This includes everything contained in the pa-
tient's chart, including the dental and medical
history forms, treatment plan(s), all consent
forms, prescriptions, any telephone messages,
referral forms, reports to or from other dentists
or physicians, all radiographs, models, corre-
spondences to or from the patient or anyone on
their behalf, insurance forms, insurance approv-
als for treatment, explanations of benefits, and
even the outside of the record jacket if it con-
tains notations.
Once the insurer receives these documents,
an attorney will be assigned to you. This attor-
ney, although retained by the carrier, represents
your interests. You are the client, and the attor-
ney must not act in any way that is detrimental to
you. The attorney will meet with you to discuss in
detail the treatment you provided to the plain-
tiff. You can and should be candid and open at
this time. Any information that you give to your
attorney is protected by the attorney-client priv-
ilege and cannot be divulged. Therefore, com-
plete honesty is essential in preparing your de-
fense.
There have been an increasing number of
suits against orthodontists, particularly with the
increase in orthodontic treatment of adults.
These claims often involve failure to obtain the
desired results, causation or exacerbation of
temporomandibular joint problems, misapplica-
tion of orthodontic appliances causing bite
problems, failure to diagnose and/or treat decay
at the site of appliance placement, causing root
resorption, and causing periodontal disease.
The one claim that is in almost every lawsuit is
lack of informed consent. This makes obtaining
informed consent a very important defense tac-
tic. It is not sufficient to orally give the patient a
treatment plan. It should be done in writing and
include options as well as any potential compli-
cations for each viable treatment option.
A lawsuit is usually initiated by service of a
summons and a complaint. Some jurisdictions
may require that these documents be filed be-
fore service. The summons sets forth the names
of the parties and the venue, whereas the com-
plaint elaborates, usually in general terms, the
nature of the allegations against you, the dam-
ages allegedly suffered by the plaintiff, and may
also contain a demand for monetary damages.
To obtain jurisdiction over an individual,
there are certain prescribed rules that must be
Home | TOC |          
Index
217
followed. They may vary from state to state, but
one thing is constant: failure of a plaintiff to
properly serve a defendant can result in the
dismissal ofthat lawsuit. It is important to keep a
written record of how and when you were served
to help your attorney determine if the service of
the legal papers was proper.
In addition, there are specific times under the
law for defendants, through their attorneys, to
answer the allegations specified in the com-
plaint. An answer will address each allegation
and will also set forth any affirmative defenses.
The burden of proof for these defenses rests
with the defendant. One defense usually raised
is the comparative or contributory negligence of
the plaintiff. This means that the injuries alleged
were added to or caused by the plaintiff. Failure
to timely serve an answer may allow a plaintiff to
obtain a default judgment against you. In es-
sence, this means that you will not have an op-
portunity to defend yourself against the allega-
tions of malpractice made by the plaintiff.
The course of a lawsuit is often long and
protracted. Each stage may take several months
or even years. Generally, the courts oversee this
process. The defense attorney will request autho-
rizations to obtain the records of prior, concur-
rent, and subsequent dentists and physicians
who may have treated the plaintiff for the al-
leged injuries. Once a patient places his/her
condition in issue, the defendants are entitled to
obtain all records relating this. Your attorney
may discuss with you some of the salient facts
ascertained from these treatment records.
Once the records are obtained, an examina-
tion before trial, also known as a deposition, of
the plaintiff will be held. The defense attorney
will question the plaintiff under oath, with a
court reporter present, as to both their medical
and dental history, the treatment that was pro-
vided, what the plaintiff was told about the treat-
ment in issue, the benefits sustained, compli-
ance with the treatment plan, and any damages
or injuries claimed as a result of the treatment.
Because most plaintiffs swear that they did every-
thing you advised them to do, it will be to your
benefit to have notations in your records indict-
ing each time that this did not occur.
Often, your only defense in a lawsuit that
occurs long after the treatment was provided is
your records. Accurate, concise, complete
records can sometimes be your only defense.
<<    
     Article
      >>
218 Toni Reale
Plaintiffs memories become sharper and clearer
the closer they get to trial. If your records, which
were kept concurrent with the treatment, differ,
then the defense attorney can cast serious doubt
on the plaintiffs version of the facts.
After the defense attorney questions the
plaintiff, the plaintiffs counsel then questions
the defendant(s). Before this, the defense attor-
ney will meet with you to go over your records
and explain the deposition process. Again, your
records are invaluable in this instance. They will
provide backup for your recall. The plaintiffs
attorney will most likely question you directly
from your records. You will be asked about each
notation. Plaintiffs attorneys pay particular at-
tention to what a patient was told regarding
treatment. If you do not recall the answer to any
question, just say so; do not make up an answer.
The opposing attorney may also ask you ques-
tions about your education and training. By do-
ing this, he/she may be laying the foundation
for claiming a lack of education or experience in
performing a particular procedure. The most
important advice you need to remember while
preparing for and attending your deposition is
that your lawyer is there with you to protect your
interests. Follow his/her instructions, and do
not volunteer any information. Keep your an-
swers short, and just answer the question asked.
Once your testimony is taken, you will receive
a copy of the transcript to review for errors or
inaccuracies. You may make corrections, with
the advice of your attorney on the forms pro-
vided and are encouraged to do so. The plain-
tiffs attorney will use this transcript against you
at trial if your trial testimony differs from the
deposition testimony. Any deviations can under-
mine your case and your credibility. Your depo-
sition transcript is one of the best opportunities
to assist in your own defense. The attorney rep-
resenting you knows the pitfalls of the examina-
tion before trial process and will be on the alert
to protect you from them.
At some point during the course of litigation,
a defendant is entitled to have the plaintiff ex-
amined by a doctor of his/her (his/her attor-
ney's) choosing to determine the condition of
the plaintiffs mouth. This usually occurs after
the plaintiffs deposition. However, if the plain-
tiff has had all of the dental work done by the
defendant replaced, the decision may be made
to waive this examination. Once a report is pro-
Home | TOC |          
Index
vided, it will be exchanged with the plaintiffs
attorney. You are also entitled to have the re-
ports from the plaintiffs examining doctors as
well. In today's court system, there is an open
exchange of information. Long gone are the
days of the "Perry Mason" surprise witness. If the
opposition does not furnish you with advance
copies of reports or records, they may not be
admitted into evidence at the time of trial.
After all parties have been deposed, and an
independent physician or dentist has examined
the plaintiff, the discovery phase of the lawsuit is
completed. At some point, after all treatment
records have been obtained, your attorney will
have the file reviewed by an expert, or several
experts, if the claims or damages warrant it. One
expert will be an orthodontist. The attorney's
purpose will be to determine if there were any
departures from good and accepted practice re-
garding your treatment. Additional experts may
be retained to discuss the issue of damages.
To prevail in a malpractice action, the plain-
tiff must prove that the defendant had a duty of
care to the plaintiff, that there was a breach of
this duty, and that this breach was the proximate
cause of the plaintiffs damages. The experts
from both sides will consider all of the treatment
and damages documented to determine if the
plaintiff presented valid claims or not. To prove
the allegations of malpractice, the plaintiff must
present expert testimony setting forth the stan-
dard of care, the defendant's treatment did not
meet that standard, and that the breach of duty
caused the damages alleged.
Standards of care may also be promulgated to
some degree by professional societies or devel-
oped through regulation. Some of these stan-
dards are mandatory; others, are merely guide-
lines. Individual practitioners should keep
abreast of any changes in their own specialties.
After the discovery phase, the case is placed
on the trial calendar. The plaintiffs attorney
may request a trial by jury or a bench trial in
front of a judge without a jury. If he/she does
not request a jury trial, your attorney may do so.
Depending on the court calendar in a given
jurisdiction, this pretrial period can be several
weeks to many years. If the defense has been
able to find an expert to defend your treatment,
the case will now proceed to trial. If not, you may
be approached by your carrier to discuss settle-
<<    
     Article
      >>
The Anatomy of an Orthodontic Lawsuit
ment and to sign consent to settle, if one is
required by your policy.
Your attorney will be preparing for trial by
meeting with you, your experts, and getting any
exhibits made if necessary. In addition to your
records, you will also have your deposition tran-
script for review. Don't be dismayed if there are
long periods of inactivity; this is a common oc-
currence.
The plaintiff has the burden to prove all his/
her claims. This must be done by a preponder-
ance of the evidence. This means that he/she
must persuade the judge or jury that the facts
are more likely than not to support his/her
claim (s). The defendant has only to prove any
affirmative defenses refuting the plaintiffs
claims. The jury is the sole judge of the credibil-
ity of the witnesses, the plaintiff, the defendant,
and all of the experts. If contributory or com-
parative negligence of the plaintiff is proven, the
plaintiff may receive nothing or have any award
reduced by the degree of his/her negligence.
Immediately before trial, you may be served
with a subpoena to testify at trial. This is the
plaintiffs method of making sure that you testify
during presentation of the case. A trial proceeds
in a proscribed manner. Both attorneys have the
opportunity to question and select jurors from a
panel to hear this case. After this, there are
conferences with the assigned judge, after which
the attorneys present their opening argument to
the jury. The plaintiff puts on his/her case first.
He/she will call the plaintiff to testify and any
other witnesses, possibly family members. Then
he/she will call the defendant to testify, if
he/she has served the defendant with a sub-
poena (if not, the defendant will testify when the
defense presents its case). Finally, the plaintiff
will have the testimony of his/her experts and
examining or treating dentists or physicians.
The defense attorney will have the opportunity
to cross-examine all witnesses in an effort to
disprove or discredit their testimony. Once all of
Home | TOC |          
Index
219
plaintiffs' witnesses have testified, the plaintiff
rests his case.
The defense now puts on its case. You may
not be called to take the stand for a second time
if you have already testified during the plaintiffs
case. Your experts will now be called to refute
the plaintiffs assertions. When all witnesses have
testified, the defense also rests. Both lawyers are
then given the chance to persuade the jury for
the final time when presenting their summation.
Here, the defense goes first; the final words the
jury will hear will be from the plaintiffs attorney.
After the summations, the judge will instruct
the jury as to their role as jurors, he/she will
explain the law, and their role to determine the
true facts as the jury believes they were pre-
sented. He/she will then read the jury charge,
which sets forth the alleged departures for ac-
cepted practice and the claimed damages. Based
on the evidence, they must determine if any
departures caused the injuries.
As a defendant, you may be expected to be in
court each day or as specified by your attorney.
While in the courtroom, you will be under the
constant scrutiny of the jury. Your demeanor
and cooperation as well as the testimony of the
experts will be an integral factor in any success
in the courtroom.
In summary, the following are examples of
what to do and what not to do. You should (1)
report service to your insurer, (2) provide com-
plete records to the carrier, (3) be accessible to
meet with your attorney and/or investigator for
the carrier, (4) cooperate with your attorney
during all phases of the action, and (5) feel free
to contact your defense attorney with any ques-
tions or concerns.
On the other hand, you should not (1) call
the patient or their attorney, (2) alter your
records, (3) discuss the case with any other de-
fendants, or (4) do not offer gratuitous state-
ments or explanations when testifying.
<<    
     Article
      >> Home | TOC |          
Index

Suing the Orthodontist: A View From the

Plaintiff's Side of the Bar
Joel Kotick

From the perspective of a plaintiff's attorney, there is good and bad news for
the orthodontic profession. The bad news is there is no shortage of orth-
odontic malpractice. Borrowing a phrase from Ralph Nader, "The root of the
malpractice problem is malpractice." The good news is that few attorneys
understand and/or accept orthodontic malpractice cases. This article chron-
icles my experiences as an orthodontist and how a series of events led me
to become a plaintiff's attorney. It also provides insight into what I see as
the more common reasons for orthodontic litigation. (Semin Orthod 2002;8:
220-227.) Copyright 2002, Elsevier Science (USA). All rights reserved.

rthodontic malpractice represents a small
O percentage of my dental malpractice case-
load because the injuries complained of are not
usually severe, and orthodontic cases do not jus-
tify the cost of bringing a case to trial. From
purely a business perspective, better medical
malpractice attorneys do not accept cases that
have the potential to generate less then a certain
value. Most orthodontic actions fall far below
this threshold. It is also difficult to find orth-
odontic expert witnesses because the orthodon-
tic community still operates within an "old boy's
club" mindset. Orthodontists are less willing to
testify against their colleagues than are doctors
in other dental and medical specialties. This
observation is truer outside of major metropoli-
tan cities but does not apply if the defendant is
a general dentist or a pediatric dentist rendering
orthodontic services.
Another additional obstacle to attorneys ac-
cepting orthodontic cases is, depending on the
jurisdiction, the instructions that the judge gives
to the jury to apply to the facts. In New York, for
instance, juries are instructed that an orthodon-
tist is not a guarantor of a good result nor is
he/she liable for mere errors in judgment as
long as he/she exercised his/her best judgment.
Address correspondence to Joel Kotick, DMD, ]D, 501 East 79th
Street, New York, NY 10021.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0804-0007$35.00/0
doi:10.1053/sodo.2002.127869
Best judgment is defined as that used by a rea-
sonably prudent orthodontist practicing within
the community. This instruction is archaic, con-
voluted, confusing, and places a tremendous
burden on the plaintiff.
There is further good news for orthodontists;
national statistics reveal that the defendant pre-
vails in the overwhelming majority of cases that
go to trial. In orthodontics, one reason for this
statistic is that the average personal injury attor-
ney representing a plaintiff is not technically
conversant with orthodontic art and science. A
second reason is that most orthodontic treat-
ment occurs over several years, and the mechan-
ics are difficult for a lay jury to follow. The third
reason is that the defendant and the expert
witnesses can and often do invent the medicine.
Finally, as previously stated, the legal application
of the law presents an additional hurdle.
Because I have been asked to tell this story
from the perspective of a plaintiffs attorney, I
will begin with my introduction to orthodontic
malpractice and end with a current recital of the
fact patterns from orthodontic cases in which I
have had personal involvement during the years
2000 and 2001 because these cases represent
what is currently happening.
Sometime in the late 1960s, approximately 12
years before my entering law school, I received a
telephone call from the head of the hospital in
which I was on the orthodontic staff. He asked
permission to give my name to an attorney who
needed someone to review an orthodontic mal-

220 Seminars in Orthodontics, Vol 8, No 4 (December), 2002: pp 220-227

<<    
     Article
      >>
Suing the Orthodontist
practice case pending in a neighboring state
because no one else would do it. My credentials
at the time were not overly impressive, and I did
not know then what I know today: juries do not
differentiate between fair credentials and signif-
icant credentials. The plaintiffs attorney noted
that I was the codirector of the temporomandib-
ular joint clinic at a local dental school, I was an
attending on a local hospital staff where I was
the chief of the section of orthodontics and
temporomandibular joint, I had published two
scientific papers, I was past president of my den-
tal school alumni association, and at that time I
was the president of the local dental society. To
a lay jury, it was enough.
On examination of the patient and the
records, I learned that this 30-year-old woman
had four premolars extracted and was being
treated with the Begg appliance. While on vaca-
tion with her husband one June, she thought she
felt her teeth with her tongue protruding into
her palate. She asked her husband to look, and
indeed her roots were palpable. Three months
before, the orthodontist had taken a midtreat-
ment panoramic radiograph that revealed se-
vere root resorption. Instead of stopping treat-
ment, the doctor initiated the third stage of
Begg treatment and torqued the roots through
the palate. In my opinion, there was no malprac-
tice up until March. However, once the resorp-
tion was diagnosed, there was no excuse to begin
the third stage of Begg. The panoramic radio-
graph was the smoking gun. I agreed to testify
for the plaintiff.
I was young and very nervous; my credentials
were weak, and I had no clinical experience in
Begg mechnaotherapy. I had taken a 2-week
Begg course and had attended some Begg lec-
tures. My major anxiety was appearing stupid.
Contrary to my fears, I discovered the orthodon-
tist's attorney had little understanding of the
medicine. He could not impeach me, and there
was no good defense for this case. The case
settled for $45,000 within minutes after I left the
witness stand. Even back then, $45,000 seemed
too little for the loss of four anterior teeth, along
with the extraction of four first premolars to a
30-year-old woman. I drove home proud that I
had the courage to stand up for justice.
I had no involvement with malpractice until
some 12 years later when I was in law school. I
authored a paper entitled Does Dental Malpractice
Home | TOC |          
Index
221
Litigation Influence the Quality of Dental Care. I
reached the conclusion that dental malpractice
litigation does not influence the quality of care.
I do not agree with that conclusion today be-
cause I truly believe that the threat of malprac-
tice has made all of us better practitioners. The
best prevention for malpractice suits is conscien-
tious and caring practice.
While in law school, I was referred a 19-year-
old patient who was occluding only on his pos-
terior teeth. Four first premolars had been ex-
tracted, and he presented with a prognathic
mandible and an anterior open bite. It was a
functional and esthetic nightmare. I sent him for
a surgical consult, and, naturally, his parents
were upset. They then asked me to examine his
16-year-old brother who was also prognathic, was
also developing an open bite, and was in the
middle of four bicuspid extraction therapy. I
explained that the skeletal problem could only
be corrected with a combination of surgery and
orthodontics. The parents then asked me to ex-
amine their 13-year-old daughter. She was ready
to begin treatment with the same orthodontist
for essentially the same problem. He again had
prescribed the extraction of four first premolars,
and I explained that extraction would result in
the same poor ending. They accepted my treat-
ment plan, and treatment was successfully com-
pleted in 18 months without any extractions.
Seeing the simple progress of their daughter,
her parents asked me to refer them to a mal-
practice attorney. I referred them to one of my
professors, who initially agreed to represent
them and then changed his mind. The parents
asked me if I would represent them. By that
time, I had recently graduated law school, taken
the bar, and was awaiting the results. I explained
I had no experience and the statute of limita-
tions would toll if I did not pass the bar. They
said they would gamble on me passing the bar
and my inexperience. I passed the bar and had
my first two clients.
The orthodontist was board certified, had
complete records, and was able to impress the
jury with his knowledge. The expert for the de-
fense was low keyed and simply told the jury that
growth was unpredictable, and the removal of
four premolars was the standard for the commu-
nity. I attempted to explain that the role of a
good diagnostician is to be able to predict the
potential for adverse growth. The cephalometric
<<    
     Article
      >>
222 Joel Kotick
arguments were lost on the jury. I was not per-
mitted to tell the jury this orthodontist was about
to make the same mistake for the third time and
that the daughter had been treated successfully
without extractions. The judge gave the usual
New York State instructions that an orthodontist
is not a guarantor of a good result, and the jury
came back quickly with a finding for the defen-
dant. I was personally disappointed. I believed
his misdiagnosis, not once but three times, was
malpractice. The orthodontist left court with an
air of self-satisfaction, and the two young men
each left court with a poor functional occlusion
and an unsightly profile.
The lessons that I learned was that you do not
have to be a good orthodontist to win; it is
important to have good credentials and make a
good personal appearance, and one must have
good records. I realized I was naive in expecting
the jury to comprehend the nuances of growth
prediction. I learned some good lessons.
Compare this case with another case I lost. It
was also a four-premolar extraction case. The
defendant orthodontist had weak credentials,
his records were poor, and his mechanics bad. I
argued the poor occlusion resulted in a tem-
poromandibular joint problem. The plaintiff
had not yet undergone subsequent orthodontic
therapy or paliative treatment for her joint prob-
lem. She had been to an oral surgeon who tes-
tified that joint surgery was necessary. The de-
fense attorney emphasized that because she did
not have the surgery nor any treatment to date,
how bad could the problem be? The defendant
produced a very effective witness who used a
skull to explain the temporomandibular joint,
"lying" about the medicine. He convinced the
jury that my client did not have a temporoman-
dibular problem. The jury found that the defen-
dant was negligent in his orthodontic treatment
but did not find that his negligence was the
proximate cause of the temporomandibular
joint injury.
From this case, I learned about venues. The
case was tried in a very conservative venue with a
bias toward the defendant. Some venues are
much more favorable to the plaintiff than oth-
ers. In other words, the same fact pattern in one
jurisdiction will prevail in one court and lose in
another. The second lesson was that there are
experts who will say anything and consider lying
to be part of their role. The third lesson was the
Home | TOC |          
Index
importance of convincing a jury that the very
negligence complained of is the proximate
cause of the injury. At that time, the instructions
regarding proximate cause were archaic and
confusing. The judge instructed the jury that an
act is the proximate cause of an injury if a rea-
sonable person by a fair preponderance of the
evidence would believe that the act was a sub-
stantial factor in bringing about the injury. No-
body speaks like that, and I am convinced that
juries have no idea what the judge is talking
about. Today, although still confusing, the prox-
imate cause instruction is less onerous, and the
jury is instructed it must be substantial factor in
bringing about the injury. The final lesson for
me was that a really bad orthodontist can get
away with providing a horrible result.
Juxtapose these cases with the infamous Mich-
igan case of the mid-1980s. As I recall, the orth-
odontist treated a Class II Division I malocclu-
sion by extracting the upper first premolars. The
orthodontist had excellent credentials, the
records were complete, and the treatment was
reported to be excellent standard edgewise ex-
traction therapy. The patient developed tem-
poromandibular joint problems requiring sur-
gery. The expert for the defendant was
nationally recognized; the expert for the plain-
tiff was a general dentist who lectures around
the country on functional orthodontics and
against extraction. The negligence alleged was
that the extraction of the upper premolars re-
sulted in the patient's temporomandibular prob-
lem. The verdict was over $1,000,000 for the
plaintiff.
The profession was up in arms. It was sug-
gested in the dental literature that the case
would be reversed on appeal. It was not. The
case prompted the first annual conference on
orthodontics and the temporomandibular joint.
Many of our leaders drafted a position paper on
orthodontics and the temporomandibular joint
that was subsequently published. The position
paper was self-serving and concluded orthodon-
tic treatment does not cause temporomandibu-
lar joint problems.
Like many of you, I attended the conference.
While in Michigan, I visited with the plaintiffs
attorney. He has a sign over his private office
door "TMJ Specialist." He was generous with his
time and enjoyed discussing this case and other
temporomandiular joint cases he had won. He
<<    
     Article
      >>
Suing the Orthodontist
speaks to a jury like a used car salesman in a
manner I do not believe would prevail in other
jurisdictions. He attributed his win to the pom-
posity of the defendant. The orthodontist admit-
ted he observed the patient's temperomandibu-
lar joint symptoms during treatment but did not
remove the active wires and stop treatment. I do
not know the facts well enough to determine if it
was a just result. The lesson I took away is that
juries are willing to ignore nationally recognized
experts and sometimes as John Adams is quoted
as saying "those stubborn facts."
A case deserving of significant discussion con-
cerns a 35-year-old plaintiff who sought orth-
odontic treatment for correction of rotated in-
cisors. The extraction of two upper premolars
exacerbated the patient's undiagnosed and un-
treated periodontal condition. To defend
against the claim of not recognizing or treating
the patient's periodontal disease before initiat-
ing treatment, the defendant orthodontist pro-
duced a three page letter he said he wrote be-
fore beginning treatment in September of 1985.
The letter was allegedly sent to the patient's
general dentist and the patient advising her of
her periodontal disease and the need for treat-
ment. The plaintiff was adamant about never
having received this letter. The first two pages
were dated 1985 but the third was dated 1986.
The defendant had great credentials, board cer-
tified, had written a textbook, and so on. At trial,
I asked him why he dated his letters 1986 during
1985. He fumbled, explaining it was a secretarial
error. The jury understood that not many peo-
ple date their letters 1986 in September of 1985.
I followed with why his letterhead bore a street
address that he was not practicing at in 1985. I
believed I had him on the ropes. He responded
that he had moved in 1984. The jury was con-
fused. I challenged him to bring in telephone
bills, a lease, or tax support documentation per-
taining to the move.
Coincidentally, only days before this trial, the
mother of a child treated by this same orthodon-
tist was referred to me regarding possible orth-
odontic malpractice. I determined that although
there was malpractice, the damages were ex-
ceedingly small, and I told the mother I would
have to think about taking the case. Coming
back to the case at hand, after the orthodontist
lied about when he moved, I called this child's
mother and asked if she would be a witness as to
Home | TOC |          
Index
223
the orthodontist's office address when her child
received treatment from him in 1985; she
agreed.
In those days, I was still practicing orthodon-
tics full-time at the same time I was developing
my legal practice. I was constantly running back
and forth between the courthouse and my office.
My client, incensed at the outright lies, took over
part of the lawyering while I treated patients.
The plaintiff went to the doorman of the defen-
dant's building to secure his testimony as to what
year the defendant occupied the premises. She
also found old telephone books listing his old
office location for 1985. The judge would not
allow the telephone listing to be admitted with-
out securing a telephone company employee's
testimony as to their authenticity. I figured the
doorman's testimony was sufficient so I did not
subpoena the employee. The trial was no longer
about the medicine. I was sure that the jury
believed that the defendant was lying; we
couldn't lose.
After resting our case, the defendant pre-
sented his case. The defense attorney intro-
duced the general dentist who testified that she
received the letter in 1985 and that she recalled
the orthodontist having moved in 1984. The
defendant then called one of his experts who
coincidentally rented space to the defendant be-
fore his move. My client was now very con-
cerned. The defendant then testified about the
move and stated it occurred in 1984. When
asked how he could be so sure about the date,
the defendant replied that his wife had died in
1984, and his son had to be institutionalized that
same year. All of it was true, except the part
about the move. The testimony was both power-
ful and devastating.
When questioned about whether or not he
was able to provide documentation as to the
move, the defendant produced a subrider to his
lease indicating the move occurred in 1984. My
client and I were the only people in the court-
room who knew he was lying. I made a long
record outside the presence of the jury telling
the judge that the subrider was a fraud. I said if
it goes to the jury I will lose the case. I asked the
judge for a representation from the defendant's
attorney that he did not know it was a fraud. The
attorney said nothing. The judge said he would
not ask an attorney for such a representation
and counseled that if it was a fraud, the attorney
<<    
     Article
      >>
224 Joel Kotick
was well aware of the consequences. I had made
my record for an appeal.
My client was of foreign extraction and did
not make a sympathetic witness. The defense
attorney told the jury she was bright, shrewd,
cunning, and she made up this whole story
about not receiving the letter and the different
address to get tax-free money. The verdict was
for the defendant. One juror, an editor from a
national magazine was the one dissenting juror.
In New York, you only need five out of the six
jurors to win. He told me he was unsure about
the letter and address but instinctively believed
the defendant was lying. He said the other jurors
believed the lease and the testimony of defen-
dant's expert.
The following day, I personally served a sub-
poena for the original lease on the defendant's
expert and served a subpoena for the original
leases from the dentist relating to his various
practice locations; something I should have
done at trial instead of returning to my office to
treat my orthodontic patients. Neither dentist
complied. I ran to court and had the judge sign
an ordered subpoena and again, I personally
served them. I was under a time gun because I
had to make my posttrial motion within 2 weeks.
I learned later that the defendant's attorney,
aware of the time restraints, instructed both den-
tists not to comply. I then went to court and
procured an order to show cause, again, served
personally. Both dentists had to appear in court
with the leases or go to jail. I was convinced that
the judge believed I was correct. Neither dentist
appeared. The judge was livid. Instead, they en-
gaged an attorney who came to court with the
original leases. The judge read the leases and
gave me a copy that I attached as an exhibit with
my motion papers. The defendant had copied
the subrider to the lease, changed the dates, and
had the expert dentist sign the changed sub-
rider.
The judge reversed the verdict. The insur-
ance company settled. The defendant's expert
frightened he might lose his license called me
apologizing for his stupidity in lying. I called the
nice mother that had testified and said I would
take her daughter's case even though the dam-
ages were small. Truth being stranger than fic-
tion, I immediately initiated the lawsuit. At the
deposition with this same defendant, I asked
him what he meant when he wrote to his attor-
Home | TOC |          
Index
ney, "Dear Mr. Attorney, enclosed are my
records, please destroy the original records I
sent." I thought his attorney would fall off her
chair. By accident, his attorneys sent me a copy
of that letter with his treatment chart. The in-
surance company now settled a minor case for
well over six figures. The defendant orthodontist
almost went to jail. His attorney was investigated.
The offshoot of this debacle was a well-attended,
full-day seminar cosponsored by a component
orthodontic society and the insurance company.
The format was a mock trial about an adult
orthodontic case with underlying periodontal
disease. The most common orthodontic cases in
my legal practice are adults who are treated in
the presence of periodontal disease.
The lessons learned are obvious. First, I could
no longer practice law part-time; it was all or
nothing. Second, a good attorney cannot be lazy;
you dot every "i" and cross every "t". As for the
orthodontic lessons, we have all been taught
never to change records. Here the orthodontist
wrote a letter after the fact, he bungled the
dates, used false letterhead, and then enrolled
other colleagues to lie for him. All could have
lost their licenses. I see altered records all the
time. When I convince a jury about the alter-
ation, it is devastating to the defendant. You
have to be dumb and then dumber to alter
records.
The occurrences stated previously are what
brought me to become a plaintiffs attorney.
Make no mistake about it, I am still an orthodon-
tist at heart, but I believe that occasionally we do
commit malpractice and on those occasions
when a patient is injured, they deserve represen-
tation. I made a lot of mistakes at the beginning
of my legal career, the same way we all made
many mistakes at the beginning of our orth-
odontic careers. I am now much better at what I
do, and, in the remainder of this article, I will
discuss the fact patterns in orthodontic cases I
have personally been involved with as plaintiffs
attorney during the last 2 years. During this
time, I have tried six orthodontic cases, five go-
ing to verdict, and one ending in a nonsuit,
which means that the case has to be started all
over again. I prevailed in all of the five cases that
went to verdict. Additionally, I have settled other
orthodontic cases during this time frame before
they went to trial. There is no shortage of orth-
odontic malpractice.
<<    
     Article
      >>
Suing the Orthodontist
Case One
The patient was a brachyfacial woman in her late
20s who presented with a Class II Division I
deep-bite malocclusion. Treatment involved the
extraction of upper first premolars. The negli-
gence alleged was that the case required orthog-
nathic surgery and that there was a failure to
monitor for root resoprtion. The injuries
claimed were that the extraction spaces could
not be closed, there was no reduction in either
the overjet or the overbite, and significant root
resorption had occurred. The defense claimed
that root resorption happens all the time, the
patient would not lose any teeth, and that surgi-
cal treatment was too extreme. The case was
tried in Virginia, and we received a verdict of
$300,000. The verdict was appealed and af-
firmed. To my surprise, it was reported to be the
largest dental verdict to date in that state, which
has a cap of $1,000,000 for all medical malprac-
tice actions.
Lessons
You must consult with surgeons in skeletal cases
and midtreatment radiographs are critical to
monitor for root resorption. In this case, the
attorney for the defendant showed the orth-
odontic expert a panoramic radiograph and
asked him if he saw any root resorption. The
expert told the jury no. On the panoramic ra-
diograph, root resorption could not be seen. It
was a poor-quality film with the spinal column
superimposed over the anterior teeth. This "de-
ception" was exposed.
Case Two
The patient was a 35-year-old woman who pre-
sented with an anterior open bite and spacing.
The treatment was the extraction of one upper
premolar because the patient was missing one
on the other side. The negligence alleged was
that the open bite required a surgical correc-
tion, extractions were unnecessary given the an-
terior spacing, the mechanics were poor, and
there was a failure to monitor the dentition with
midtreatment radiographs. The defense was that
premolars are routinely extracted. The injuries
claimed were significant root resorption and
that the open bite was made worse. The original
Home | TOC |          
Index
225
verdict was for the plaintiff; the defendant ap-
pealed and then settled for $377,000 before we
received a decision.
Lesson
Skeletal problems in adults, particularly open
bites, require obtaining a surgical consultation
as well as the necessity to take midtreatment
radiographs. As an aside, this case was tried in 2
days. We began trial at 9:00 AM on a Monday
morning. The judge told us that if your next
witness is not ready when your last witness com-
pletes testimony your case is dismissed. I asked
what the court hours were as this was another
out-of-state case and unlike New York, each
judge is free to set his own schedule. The judge
said, "In my courtroom, we work from 8:00 AM
and go through midnight if the jury is willing."
We worked until 9:30 PM Monday and had a
verdict by 10:00 PM on Tuesday. I tried the same
type of case with the same number of witnesses
recently in New York. The case lasted for 4
weeks.
Case 3
The patient was a 35-year-old woman with a Class
1 bimaxillary protrusion who underwent four
premolar extraction therapy. The venue (the
county in which the case was heard) was a "plain-
tiff friendly" (the juries often side with the plain-
tiff) locale in New York. The negligence claimed
was sloppy mechanics treated in the presence of
periodontal disease, coupled with a prolonged
treatment time. The defense claimed that the
poor patient cooperation for months at a time
was not contributory, and the defendant as-
sumed that the general dentist was treating the
patient's periodontal condition. The injuries
claimed significant bone loss and future loss of
teeth. The plaintiff did not undergo any subse-
quent treatment. We received a verdict of
$225,000; the case was not appealed. Frankly, I
was disappointed with the award particularly
with a favorable venue.
Lesson
There is no excuse for treating cases in the
presence of periodontal disease and certainly no
defense for extracting teeth without first resolv-
ing the underlying periodontal disease.
<<    
     Article
      >>
226 Joel Kotick
Case 4
Plaintiff was a 35-year old with a Class II maloc-
clusion. The treatment involved the extraction
of upper first premolars. The negligence alleged
prolonged treatment and failure to monitor for
root resorption with periodic radiographs. The
injury claimed was root resorption of the upper
anterior teeth. The defendant was board certi-
fied and arrogant, arguing that his treatment
was exemplary. The case returned a verdict of
$472,000. It was appealed and was settled for
$400,000.
Lesson
In my opinion, both orthodontic and legal, any
adult case that takes longer then 2 years must be
carefully reevaluated. There is no excuse for not
taking periodic radiographs to monitor for the
presence of root resoprtion.
Case 5
The plaintiff was a 35-year-old brachyfacial
woman with a Class II Division I deep-bite mal-
occlusion. Treatment involved the extraction of
upper first premolars to retract the upper ante-
rior teeth and extraction of the lower left and
right third molars to upright the second molars
that had drifted into the first molar position.
The negligence claimed was that the second and
third molars were functioning well and had no
significant bone loss, thus both the lower extrac-
tions and the prosthetics were unnecessary. The
defendants countered that the extraction of the
third molars in the position of the second mo-
lars was indicated to improve the axial inclina-
tion of the second molars in preparation for
fixed bridges. The plaintiff argued that she was
told upper arch treatment was necessary or she
would lose her teeth. The defendant claimed
that the plaintiff desired treatment for cosmetic
reasons. The plaintiff further argued even if
treatment was desired, it did not require the
extraction of two teeth, and, therefore, treat-
ment was totally improper.
The defendant was a graduate orthodontist
and taught orthodontics at a local dental school.
He began treatment in the upper arch with an
.012 steel wire. He replaced it with an .014 steel
wire. For the next 26 months, the only treatment
Home | TOC |          
Index
rendered was power chain from the upper first
molar to the upper first molar. The defendant
produced two experts. The first, a periodontist,
told the jury that extracting third molars was
appropriate because they are very often ex-
tracted. He also attacked the integrity of the
subsequent treating periodontist and prosth-
odontist, saying that the subsequent treatment
was not necessary. The defendant's other expert
was an orthodontist who told the jury that cepha-
lopmetric radiographs are not the standard of
care in orthodontics and that placing a power
chain on an .014 wire is appropriate treatment.
The injuries claimed were temporomandibular
joint pain; tipped upper anterior teeth with a
total loss of anchorage and lack of stability; and
the need for subsequent periodontic treatment,
orthodontic treatment to upright the teeth, and
full-mouth rehabilitation. The jury returned a
verdict of $450,000. This case is on appeal.
Lesson: Abandoning Every Principal Learned
in School Will Result in a Losing Case
Tactics attacking the integrity of the opposing
experts and subsequent treating dentists and ly-
ing about the medicine is sometimes effective. I
have lost cases to these tactics. It is a sad com-
mentary, but some dentists will say anything un-
der oath to help support the defense. No orth-
odontist with any self-respect who teaches
students should come to court and tell a jury
that a cephalometic radiograph is not the stan-
dard of care and that an .014 steel wire and
power chain is the standard of care in an extrac-
tion case.
Lastly, I would like to discuss three other cases
that I was involved with during this time period
to drive home the message concerning root re-
sorption. All three were very serious cases, al-
though I was the attorney of record in only one
of them. The case in which I was the attorney of
record was venued in California. It is very expen-
sive to bring cases to trial there, and there is a
monetary cap on pain and suffering. Discovery is
extensive, and depositions are taken of the ex-
perts and the treating dentist. The case con-
cerned a 50-year-old woman who underwent
orthodontics for cosmetic reasons. The records
were excellent, as was the treatment, and the
doctor had excellent credentials. Pretreatment
radiographs were not carefully reviewed because
<<    
     Article
      >>
Suing the Orthodontist
they showed evidence of periodontal destruc-
tion. After 2 years, severe root resorption had
occurred on approximately 18 teeth. No
midtreatment radiographs were taken. The case
settled for a large sum just before trial.
In another case, a legal colleague asked if I
would look at an orthodontic case he was han-
dling. After our meeting, he asked if he could
pay me to take the deposition of the defendant.
I agreed. I had successfully sued this orthodon-
tist once many years earlier. He was not aware
that I was involved until he walked into the room
the day of the deposition. Again, the injury was
severe root resorption, the worst I have seen in
my career. A young 18-year-old girl is going to
lose almost every tooth in her mouth. Because I
still maintain a limited orthodontic practice, I
was upset when I saw this destruction. I can only
assume that this girl must have had some meta-
bolic predisposition to resorption.
Unfortunately, the orthodontic treatment was
very sloppy. A 2-year case took 4 years. I assume
that it was round tripping mechanics that caused
the extensive resorption. However, the smoking
gun was the defendant's lack of radiographs. For
some reason, he only took periodic bite-wing
radiographs. On these films, a careful diagnosti-
cian could see lateral root resorption of the
premolars. At the deposition, he was very honest
and admitted it was root resorption that he did
not notice or record at the time. There is no
question that during treatment he should have
taken anterior radiographs. The attorney for the
defendant has offered a very large settlement. I
have advised my legal colleague that the ten-
dered amount is not enough, but I suspect that
this case will settle.
The last case was unusual in that it is fairly
uncommon for me to consult on defense cases,
although, as a very small part of my practice, I do
represent defendants who have no insurance, if
I believe in their case. I also represent many
colleagues and friends as their personal attor-
neys by monitoring and helping their insurance
company—appointed attorney with their defense.
This case presented a special, sticky situation
Home | TOC |          
Index
227
prompting an insurance company to engage me.
An orthodontist, the head of a major teaching
institution, had treated a young boy with a cleft
palate for several years. It was exemplary cleft
palate orthodontic treatment. The problem was
that the patient, now a teenager, was found to
have severe root resoprtion and would lose his
anterior teeth. The insurance company had the
case reviewed by an orthodontic expert in Cali-
fornia who advised settling. The orthodontist
said he did nothing wrong and would not settle.
In reviewing the case, I read his deposition. He
claimed although he has read about root resorp-
tion he has never seen it. This comment trou-
bled me as well as the insurance company. It is
hard to believe that a competent orthodontist, a
teacher and a leader in the field, has never seen
root resorption, and, therefore, there was no
reason to take any midtreatment radiographs to
monitor for this condition. I did not agree and
said so to the insurance company representative.
The lesson is that if you personally have not
experienced root resorption you are fortunate;
you are also in the distinct minority. There is
sufficient documentation in the literature for
every orthodontist to acquire in-depth knowl-
edge concerning root resorption. As an aside, it
is not the fact that root resorption has occurred
that constitutes negligence. It is knowing its po-
tential for existence and not monitoring for its
occurrence that constitutes negligent treatment.
Many orthodontists have the impression that
there are many personal injury attorneys ready,
willing, and able to sue orthodontists. Just the
opposite is true. Very few attorneys accept these
types of cases. Good medical malpractice attor-
neys are willing to learn the medicine in medical
malpractice cases because the damages are po-
tentially large. They are not willing in most den-
tal cases, particularly orthodontic ones, because
the injuries are usually not severe and the den-
tistry too technical. However, as you can see,
orthodontic malpractice exists, and I have at-
tempted to identify the more typical cases from
my own practice.
<<    
     Article
      >> Home | TOC |          
Index

Defending the Orthodontist: A View From the

Defendant's Side of the Bar
Arthur V. Pearson

There is a tremendous difference in how opposing attorneys evaluate their

cases, view their prospects, decide on settlement verses trial, and prepare
cases for trial. This article describes how these considerations are evaluated
from the defense perspective. Why we recommend what we do and the
factors that go into making these decisions are vitally important for the
clinician to understand. Once these evaluations, actions, and reasons are
appreciated, the risk management messages they spawn can be put into
proper context for future use. (Semin Orthod 2002;8:228-233.) Copyright 2002,
Elsevier Science (USA). All rights reserved.

L awyers who specialize in representing plain-

tiffs in dental malpractice cases have the
luxury of picking and choosing which cases they
wish to accept, risking their time, effort, and
expense for the prospect of seeking a return for
themselves and their clients. That screening pro-
cess allows the plaintiffs attorney to take on only
those cases in which liability is close enough and
damages are large enough to warrant the invest-
ment of significant amounts of time and money.
Attorneys who specialize in representing defen-
dants in dental malpractice cases generally do
not enjoy the same luxury of picking and choos-
ing their clients because they receive the assign-
ment to defend the doctor from the doctor's
insurance carrier with whom they probably re-
ceive a volume of assignments. Defense attor-
neys must play the hand they have been dealt;
declining the case is rarely an option.
The relationship between the defense attor-
ney, the defendant, and the defendant's insur-
ance carrier is unique. The attorney has two
clients in such a relationship, the doctor and the
doctor's insurance carrier; however, virtually all
jurisdictions in the United States recognize that
the attorney's primary duty is to the defendant
doctor and that the attorney cannot act in any
way that would favor the insurance carrier over
the doctor or prejudice his/her rights in that
relationship. Despite this tension in the relation-
ship, it is one that works extremely well. This can
be attributed largely to the fact that the claims
professionals working for these insurance carri-
ers generally have a lot of experience in dental
malpractice cases. They serve as an important
part of the defense team in assessing liability,
damages, selection of experts, and the overall
direction of the defense. They know their role,
the duties of the defense attorney, and what to
expect.
When an attorney is assigned to defend an
orthodontist, experienced counsel will take a
series of actions, each of which is built on those
that precede it and that ultimately will take the
case to its resolution, either through settlement,
disposition by motion, or trial before a judge or
jury. What are these steps, and how do they
impact the disposition of a case? They can be
grouped into four larger collections of activities:
the initial assessment, official information gath-
ering (otherwise known as discovery), pretrial
motion practice, and trial.

From Murphy, Pearson, Bradley & Feeny, San Francisco, C A.

Address correspondence to Arthur V. Pearson, JD, Murphy,
Pearson, Bradley & Feeny, 88 Kearny Street, 10th Floor, San
Francisco, CA.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0804-0008$35.00/0
doi:10.1053/sodo.2002.127870
Initial Assessment
One of the most important skills a defense attor-
ney can bring to his/her clients is judgment
based on years of experience in handling the
same or similar kinds of litigated claims. That

228 Seminars in Orthodontics, Vol 8, No 4 (December), 2002: pp 228-233

<<    
     Article
      >>
Defending the Orthodontist
skill is never more beneficial than when assess-
ing the case at its outset, knowing what informa-
tion to gather, how to weigh it, and how that
information fits into an overall feel for the case.
What factors go into the initial assessment?
Defendant's Records
A review of the defendant's records usually re-
veals a great deal about the doctor and his/her
practice. All of the records are pertinent in this
review and include patient health-history forms;
informed consent forms and related informed
consent booklets or other information; incom-
ing or outgoing referral slips and correspon-
dence; fee for service and insurance billing data;
pretreatment records such as models, photo-
graphs, full-mouth or panoramic radiographs,
cephalometric plates, tracings, and analysis; and,
finally, the patient's diagnosis and treatment
plan(s). These records indicate to the attorney,
possibly aided by an expert consultant, how thor-
ough was the doctor's information gathering
process; whether problems, actual or incipient,
were evident; and whether there are weaknesses
present, or conversely, strong defenses evident
within the records. There may be problems on
issues such as informed consent, diagnosis, treat-
ment planning, poor responses to treatment, the
development of negative sequellae, or evidence
of contributory negligence. On a macro basis,
the overall organization of the records speaks
volumes about how well a practice is managed
and in turn how well patients are managed.
The chronological treatment record is also
examined carefully for consistency of entries,
sanctity, missed, cancelled, or broken appoint-
ments by the patient; level of oral hygiene com-
pliance; indications of problems during treat-
ment; length of treatment; progress of
treatment; and the doctor's response (s) to any
problems. Consistency among records, and the
ability to rely on them as accurate representa-
tions of who did, or said what, and when by
recording all visits and events, have a high im-
portance.
Interview With the Defendant Orthodontist
The evaluation of the defendant is multifaceted.
Certainly, identifying whether the defendant
makes a good witness is a significant part of this
evaluation. This evaluation measures: the ability
Home | TOC |          
Index
229
to comprehend questions and provide answers
that are responsive and to the point and that will
be understood by lay people; the ability to testify
consistent with a given theme or approach; the
ability to build word pictures from technical in-
formation; and, whether or not one has that
elusive animal personality and likeability as a
witness. A further part of this evaluation is
whether any deficiencies as a witness are capable
of being cured.
Assessing the defendant's knowledge of the
science and professional standards applicable to
the treatment rendered is an early indicator of
whether an error may have been made. Asking
what the doctor believes may have gone wrong
and how it could have been prevented may pro-
vide the defense attorney with significant insight
into what the problem's cause may have been.
The doctor's management of a problem at the
very outset of the case can be an indicator of
trouble. This information can assist in knowing
which experts to seek review from and may serve
as an early indicator of a defendant who is well
behind current thinking in diagnosis, treatment
planning, treatment armamentarium, and some-
one who is weak in interpersonal and risk man-
agement skills. All of this information is taken
into account in the micro and macro evaluation
of the defendant doctor.
The Injury Involved
The nature of the injury and its impact on the
plaintiff carries significant weight in the overall
evaluation of the case. Clearly, a case involving
an extensive or aggressive oral cancer or a case
that involves the loss of a substantial number of
teeth is evaluated differently than a case that has
few life-altering effects on the plaintiff. This in-
formation can be learned from several sources.
The defendant doctor may have insight that the
plaintiff shared with him/her before their rela-
tionship ended. The plaintiffs attorney may
have written a demand letter describing the in-
jury and its impact on the plaintiff. A telephone
conversation with the plaintiffs attorney may
also serve to elicit this information. Although
this information must be verified and con-
firmed, it is nonetheless valuable and must be
taken into account in the initial assessment.
<<    
     Article
      >>
230 Arthur V. Pearson
Skill of Adverse Counsel
The level of skill and experience possessed by
the plaintiffs lawyer is another factor in assess-
ing a case. Attorneys inexperienced in orthodon-
tics specifically, or health care malpractice gen-
erally, will tend to overlook important facts, will
waste time and resources by following the wrong
leads or incorrect perceptions of technical infor-
mation, will frequently select the wrong experts.
These attorneys can frequently cause litigation
cost to be more than if an experienced attorney
handled the case. It is a cost factor that must be
taken into account. That does not mean that a
good personal injury attorney with little experi-
ence in dental malpractice cannot successfully
achieve a large verdict. However, the good ones
usually work extensively with and can afford to
hire consulting experts from the very start, min-
imizing the risks of overlooking important infor-
mation or wandering off in the wrong direction.
Attorneys who have not taken many cases all
the way through jury trial, or who have not en-
joyed much success at doing so, may be unable
to use the threat of going to trial effectively.
However, even the best defense attorney knows
the best defense case he/she has ever seen is
never absolutely guaranteed to succeed before a
jury. The most inexperienced attorney can still
win. Overall, a plaintiffs attorney who possesses
a greater degree of skill and experience will
make the plaintiffs case more credible, al-
though this equation does not necessarily hold
true in the reverse.
Conclusion of the Initial Assessment
The object of the initial assessment is to place
the defendant's case into one of several catego-
ries. These categories are: cases that are subject
to immediate disposition, cases presenting with
a significant injury and an assessment of likely
liability, cases that present with significant inju-
ries but doubtful liability, cases with a run-of-the-
mill injury coupled with an assessment of likely
liability, and cases with run-of-the-mill injuries
but carrying doubtful liability.
Cases with facts that present opportunities for
an immediate disposition should be and are
acted on first, whether or not the case carries a
high or low exposure. The early elimination of
the case may be on a procedural basis, such as
statute of limitations or other statutory defenses,
Home | TOC |          
Index
or on a substantive basis, such as lack of causa-
tion. The early elimination of the case avoids
significant costs to the client, as well as exposure,
by preventing it from getting in front of a jury.
The four remaining-categories of cases can be
reduced into two groups: those cases in which
the assessment of liability is likely and those
cases in which the assessment of liability is
doubtful. This is a rational division, because an
old saying in litigation is that bad cases don't get
better with age; bad cases need to be identified
early on, and resolved early on in the process.
Cases in which assessment of liability is high.
Good defense attorneys and well-advised insur-
ance carriers devote a significant effort to iden-
tifying cases in which there is little doubt that
liability will be established by the plaintiff against
the defendant. These cases are often put on a
fast track toward settlement after the appropri-
ate verification of the patient's injuries, the mon-
etary damages sustained, the need for remedial
or subsequent therapy, and so on. There are
some members of the plaintiffs bar who believe
that carriers purposely stall the resolution of a
case to either wear the plaintiffs down or to
continue to earn money from their investments
instead of paying the claim. While that could be
true if there was no cost associated with delaying
tactics of that sort, the reality is that such is not
the case.
Once litigation commences, the defense at-
torney has to perform the tasks necessary to
establish a defense and that costs money. In
almost every case, these costs should ultimately
exceed any potential return on investment the
carrier could enjoy. Therefore, there is a real
incentive on the part of the defense to quickly
resolve the cases in which liability is likely to be
established. This process may get bogged down
when the demand is simply too high, higher
than any reasonable estimate of what a jury
would likely return, if the case were to be tried.
In those cases the strategy is to defend, despite a
likely finding of liability, and the decision to do
so is based on the theory that, if the plaintiff
wins, he/she will win less than he/she was de-
manding. Indeed, the defendant may even try
the case based on an admission of fault and
force by the plaintiff to convince the jury that
his/her injuries are worth what is being claimed.
Cases in which assessment of liability is doubtful.
These are cases that are highly likely to be de-
<<    
     Article
      >>
Defending the Orthodontist
fended vigorously. They usually end up being
tried or settled for highly discounted amounts
based on the weak liability question. They are
fought with greater zeal by the defense because
they have a realistic possibility of a total victory at
trial. As a result, they are likely to be expensive
cases to pursue, for both the plaintiff and defen-
dant, and are not likely to resolve before trial
unless the plaintiff takes a very realistic view of
the case and deeply discounts its value.
There is another category of cases, those in
which there just are not sufficient facts available
at the outset, during the assessment period, to
know what category the case belongs in. These
cases are put into sort of a holding pattern with
the idea that the category they are likely to wind
up in is dependent on adducing and verifying
certain information.
Discovery
Once the case is assessed, some discovery will be
conducted, either to verify information with the
hope of early settlement or dismissal or to sup-
port those cases that are likely to go to trial.
There are generally three types of discovery de-
vices that are useful for gathering information:
written discovery, subpoenas of third-party
records, and depositions of parties and wit-
nesses.
Written Discovery
Almost every discovery phase begins by serving
on the opposing party a series of written ques-
tions called interrogatories coupled with a for-
mal request to produce documents in the pos-
session of that adverse party. Interrogatories are
answered with the assistance of counsel, and
counsel can be and usually is creative in phras-
ing responses so that they seem to say a great
deal but in fact provide little meaningful or con-
crete information. Consequently, interrogato-
ries are generally not very useful to discover
much beyond basic information such as job and
earning history, expenses incurred for treat-
ment, a description of the injury, names and
addresses of doctors who provided treatment,
and familial relationships. The responses gener-
ally produce foundational information from
which other, more pertinent information can be
obtained.
Home | TOC |          
Index
231
Record Subpoenas
Subpoenas for a patient's medical records are a
critical part of the discovery phase because it is
the first opportunity for the defense to see what
the plaintiff has actually been doing after their
departure from the defendant/doctor's office
and care. Generally, these records provide a de-
tailed view of the subsequent treating doctors'
opinions of the injury, its causal connection to
the defendant's treatment, recommended steps
for remedial treatment, the costs involved with
that remedial treatment, and how the doctor is
likely to testify (on whose side of the case the
doctor will line up) if called as a witness at trial
or at deposition.
In addition to the written records, there are
radiographs, photographs, models, and other
objective diagnostic information that can be ob-
tained, analyzed, and provided to the defense
expert consultants for their evaluation of the
necessity for the remedial or subsequent treat-
ment, its efficaciousness, and its relationship to
treatment of the defendant doctor.
Depositions
The deposition is the most effective tool of any
attorney because it provides the well-prepared
attorney the opportunity to inquire in depth and
detail the observations, recollections, and opin-
ions of parties, lay, and expert witnesses. Al-
though an attorney assisting a client can be quite
artful in crafting a written response to a written
interrogatory so that it discloses little of anything
of importance, the same cannot be done at dep-
osition. There, the witness is on his/her own and
must answer questions that peel back the layers
of obfuscation, beliefs built on assumptions, in-
accurate perceptions, missed conclusions, out-
right lies, and statements with no corroborating
facts. It is a matter of style whether the deposi-
tions are taken of the various players in litigation
at the early, middle, or late stage of the litiga-
tion. Each attorney has his/her own reasons and
rationale for taking depositions in a particular
order. However, the earlier in a case that a key
witness is deposed, the sooner any changes in
assessment can be made.
Regardless of the form in which the discovery
is conducted, its goal is always the same. Are facts
being adduced that favor or disfavor a finding of
liability? Do any of the facts support or detract
<<    
     Article
      >>
232 Arthur V. Pearson
from affirmative defenses, such as statute of lim-
itations, contributory negligence, or lack of cau-
sation? Did the discovery support or suggest that
the plaintiffs damages are overstated, over-
blown, or related to other conditions that were
not associated with the defendant's treatment?
As discovery progresses, good defense attor-
neys and well-advised insurance carriers are
looking to see whether the facts adduced during
discovery change the initial assessment of the
case and, in turn, how the case should be han-
dled from that point. Changes may affect the
timing of settlement overtures, the amounts au-
thorized to obtain a settlement, and how the
defense will be conducted. In other words, well-
handled cases are constantly reevaluated as new
information comes to light. It is part of the skill
set of the defense attorney to know how to
search for any and all information that bears on
every issue in the case. No one wants surprises at
trial. The failure to cast a wide net creates a risk
that harmful information will be learned for the
first time in front of a jury. The plaintiff has the
advantage because it is generally information
within the plaintiffs knowledge that the defense
seeks.
Pretrial Motion Practice
The pre trial period is typically the last 90 to 120
days before a set trial date. It is a time in which
the plaintiff and the defendant generally know
all the facts and circumstances surrounding a
claim, what the witnesses are likely to say at trial,
whether the party witnesses are likely to be good
or bad witnesses, what the likelihood of success
or failure at trial may be, and the estimated
verdict ranges. The parties may not agree on
these, but they have the same information and
have formed their own opinions. This period is
clearly the last best chance for both sides to
settle the case. A case that is incapable of settle-
ment during the pretrial stage usually means
that one of the parties has not evaluated their
case correctly, one of the clients is being unre-
alistic, or there are critical facts that cannot be
agreed on by the parties and must be decided by
the trier of fact, the jury. Because most of the
facts that will be presented at trial have been
developed and documented before the pretrial
stage, mediation can sometimes be a useful tool.
It should be kept in mind, by both parties,
Home | TOC |          
Index
that the longer unrealistic positions are taken,
the longer the case will last. This also leads to
increasing each party's investment in the case,
sometimes dramatically. Unfortunately, these
factors make the party's positions more polar-
ized and more difficult to bring to a compro-
mise. Contemporary thinking among most pro-
fessional liability carriers is that they are not
afraid to try cases in which they believe the
plaintiffs demand is simply unrealistic or has a
low percentage for success. The plaintiff should
keep in mind that this is not always a dollars and
cents game, and they cannot worry the carrier
into settlement by aggressively working a case.
When the plaintiffs position is beyond the reach
of compromise, the carrier may simply take the
very reasonable position of making the plaintiff
work for his/her money by having to convince
the jury, not only as to the merits, but also the
value of their case.
Trial
The end game of most orthodontic malpractice
cases is settlement. Probably something close to
95% of all health care malpractice cases, and
orthodontic cases do not vary from this statistic,
are ended by settlement. Thus, trial is usually the
rare default option. Because trial can be enor-
mously expensive for both the plaintiff and de-
fendant and risky because not even the best
attorney can guess the outcome each and every
time, it should be reserved for those cases in
which no other method of disposition is possi-
ble. Indeed, some practitioners view a matter
going to trial as evidence of failure of the parties
to adequately assess their case.
All trial attorneys know that trial is not a
search for truth. It is a pre-scripted, well-re-
hearsed presentation to create an impression or
emotional response that is favorable to their side
of the case. The critical players in orthodontic
malpractice cases are the parties, followed by the
attorneys, percipient witnesses, and finally the
expert witnesses. In my view, that is the order of
importance in the jury's eyes regarding who con-
vinces them and whom they accept information
from. The defendant doctor is at the top of the
list of witness importance because it is usually
based on what he/she says and what the records
show, from which the jury must make a number
of conclusions such as the level of attention that
<<    
     Article
      >>
Defending the Orthodontist
was paid to the patient, the degree of expertise
the doctor had in recognizing, diagnosing and
treating the patient, and what the standard of
care was. Equally important, the jury must de-
cide whether they like and respect the defen-
dant doctor on a personal as well as on an ob-
jective level. It is the defense attorney's job to
prepare the doctor for trial in order for his/her
testimony to have a favorable impact on the jury,
whether it is under direct or cross examination.
Attorneys add or subtract to the case by bring-
ing clarity to facts and terms that may not be
familiar to the jury, by being speedy and efficient
in their presentation of the case, by treating the
jury with respect for their intellectual capacity,
by not being unduly argumentative, and by pre-
senting lucid facts and clear arguments. They
can severely prejudice their clients' cases by pre-
senting a sloppy case, treating parties or wit-
nesses with disrespect, being overly aggressive
advocacy, failing to answer obvious questions
that the jury must have based on the evidence
presented, constantly repeating the same evi-
dence during direct or cross-examination of wit-
nesses or parties, and not giving the jury a rea-
son to side with their interpretation of the facts.
Third-party witnesses can be swing witnesses
in that they are frequently not invested in the
case as parties or hired expert witnesses. Thus,
they have a patina of objectivity or at least a lack
of obvious bias. Depending on the anticipated
testimony, defense counsel should be careful,
particularly with subsequent treating doctors, in
avoiding introducing expert opinions regarding
Home | TOC |          
Index
233
standards of care and other detrimental testi-
mony from these witnesses. Their testimony
should stick to what they observed, what they did
and why, rather than what they think the defen-
dant may have done.
Finally, expert witnesses are frequently seen
as canceling each other out. However, when the
plaintiff has the burden of proof, particularly in
areas in which the science is not well developed
and there are significant differences of opinion
by well-qualified experts, having one expert can-
cel the other out can be devastating to the plain-
tiffs case. In fact, it is one of the instances when
his/her case is most vulnerable. Selecting ex-
perts who know their field and who can convey
technical information to lay people in a lan-
guage that they can comprehend are very desir-
able skills.
Conclusion
It is likely that both parties view the workup of
their cases quite differently because they ap-
proach them differently, both from an economic
and a management perspective. Nevertheless,
there are significant procedural similarities re-
garding the life cycle of a case and how the
attorneys and their clients have to assess their
position at each step along the way. After all,
they are dealing with the same law, generally the
same set of facts, and the same jury or trier of
fact. Good attorneys on either side of the case
usually reach similar conclusions, which proba-
bly explains why nearly all malpractice cases are
settled.
<<    
     Article
      >> Home | TOC |          
Index

Have Gun Will Travel: The Role of the Expert

Witness Before and During Litigation
Malcolm Meister and Richard Masella

The role of the expert witness has been maligned and misunderstood by the
dental community at large. Expert witnesses play an integral role in the legal
process and often make or break a malpractice case. This article explores the
topic of the expert witness from why they are needed, who they are, how
they are qualified, and what role they play. (Semin Orthod 2002;8:234-237.)
Copyright 2002, Elsevier Science (USA). All rights reserved.

T he American legal system uses medical ex-

pert witnesses to establish a standard of care
in health professional practice and to address the
meaning of medical and scientific evidence.1
Expert opinion may be needed to educate judges
and juries on the type and degree of risk inherent
in various treatments, the frequency with which a
risk occurs, the probabilities of success or failure
regarding a given treatment modality, and the ex-
istence and appropriateness of viable alternative
therapies.2 In courtroom proceedings, medical ex-
perts serve as translators who allow the judge and
jury to see and understand the medical issue in
question. Ideally, the expert witness then acts as a
bridge to justice.3 In this light, forensic medical
testimony can be viewed as a social responsibility of
health care professionals.
Who Is a Medical Expert Witness?
Legal references use the umbrella term medical
expert to describe physician, dentist, and other
health care professional witnesses. Implicit in
the following citations will be the equation of
dental or orthodontic expert to medical expert
witness and physician with that of dentist or
orthodontist. Federal and state statutes, profes-
sional association principles of ethics and codes
From the Department of Orthodontics, Nova Southeastern Uni-
versity College of Dental Medicine, Ft Lauderdale, FL.
Address correspondence to Malcolm Meister, DDS, MSM, JD,
Dept of Orthodontics, Nova Southeastern University College of Den-
tal Medicine, 3200 S University Drive, Ft Lauderdale, FL 33328.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0804-0009$35.00/0
doi:10.1053/sodo.2002.127871
of conduct, and individual interpretations help
define the legal system's expert witness. As stated
in Rule 702 of the Federal Rules of Evidence, "if
scientific, technical, or other specialized knowl-
edge will assist the trier of fact to understand the
evidence or to determine a fact in issue, a wit-
ness qualified as an expert by knowledge, skill,
experience, training, or education, may testify
thereto in the form of an opinion or otherwise."4
State law commonly defines the medical ex-
pert witness as "a person duly and regularly en-
gaged in the practice of his profession who holds
a health care professional degree from a univer-
sity or college and has had special professional
training and experience or one possessed of
special health care knowledge or skill about the
subject upon which he is called to testify or
provide an opinion."5 The broad definition of
medical expert witness is emphasized by the dec-
laration that "any health care provider may tes-
tify as an expert in any action if he is a similar
health care provider" to the plaintiff or defen-
dant, or "[if] not a similar health care pro-
vider . . . , to the satisfaction of the court pos-
sesses sufficient training, experience, and
knowledge as a result of practice or teaching
in a related field of medicine . . . ."5
Although the American Association of Orth-
odontics' Principles of Ethics and Code of Pro-
fessional Conduct do not reference expert wit-
ness services, the American Dental Association
counterpart states, "dentists may provide expert
testimony when that testimony is essential to a
just and fair disposition of a judicial or adminis-
trative action."6 Perhaps the most concise defi-
nition of an expert witness states that they are

234 Seminars in Orthodontics, Vol 8, No 4 (December), 2002: pp 234-237

<<    
     Article
      >>
Have Gun Will Travel
" . . . people believed to possess information be-
yond the ordinary knowledge of judges and ju-
ries."7 From these general characteristics, it is
clear that courts have great discretion in quali-
fying medical expert witnesses. Anyone possess-
ing a medical or dental degree may be granted
expert witness status by a court. This holds even
if the dentist or physician is not a specialist in the
field in question.8 It is possible, then, for an
orthodontist accused of negligence to have ex-
pert opinions on the standard of care and
breach thereof rendered by nonorthodontists.
Types of Medical Witnesses
A fact witness may only testify about facts directly
observed, whereas the expert witness may offer
an opinion about facts not directly observed. A
treating physician, whether he/she treated the
patient in the past, presently, concurrently, or
subsequently to the defendant's treatment, and
who is called as a witness, is legally obligated to
testify about the treatment rendered to a patient
of record. In this capacity, he/she may only serve
as a fact witness. However, the treating physician
may also qualify as an expert witness, and it is in
this capacity that he/she may then offer opin-
ions about the facts at hand.8
A nontreating medical expert witness is not
compelled to provide legal testimony. An exam-
ple of such a practitioner is the physician hired
by an insurance company to perform an inde-
pendent medical evaluation regarding an acci-
dent, a disability claim, an insurance applica-
tion, and so on. In this case, a patient of record
or treatment relationship is not formed, and
thereby a patient-doctor relationship is not es-
tablished except in unique circumstances.
Medical Expert Standards
Honesty and objectivity are essential principles
of ethical testimony.9 The American Dental As-
sociation Principles of Ethics and Code of Pro-
fessional Conduct advises that "it is unethical for
a dentist to agree to a fee contingent upon the
favorable outcome of the litigation in exchange
for testifying as a dental expert."6 Rappeport10
notes the prime responsibility and most impor-
tant legal service that medical experts serve is to
promote the search for the truth.
Because possession of a health care degree in
Home | TOC |          
Index
235
and of itself usually confers expert witness status
in most courtrooms, the judicial process of qual-
ifying medical experts is rightly seen as "a course
sieve for sorting knowledgeable witnesses from
poor ones."11 Unfortunately, in many instances,
the lack of medical knowledge by judges and
juries often causes them to focus on the expert's
style of presentation or the elaboration of im-
pressive sounding credentials instead of the pos-
session of true expertise.1
Amendment to Federal Rule of Evidence
702: Establishing Expert Witness
Reliability
The course sieve of expert witness courtroom
qualifying may become finer in coming years
because of new instructions given to trial judges
contained in recent amendments to Federal
Rules of Evidence. Amended Rule 702 should
help the courts fulfill their role as gatekeepers of
the expert witness on a more consistent basis.
Now, before the offering of any expert testi-
mony, trial judges are instructed to make a de-
termination of expert opinion's reliability. Be-
fore the formalization of the gatekeeper role,
some trial courts had been remiss in carrying out
this responsibility. Challenges by opposing par-
ties as to the reliability of expert opinions are
also more likely to be heard by the trial judge.
An in-depth investigation into the rationale of
an expert's opinion should be undertaken as
soon as possible in the litigation process. Coun-
sel must explore all information on which the
expert relies, whether it is admissible or not. If
the trial judge appears reluctant to conduct a
reliability assessment, counsel can suggest the
appointment of an independent expert to assist
in the evaluation. At pretrial conferences, the
attorney should stress the need for the reliability
assessment of any anticipated expert testimony,
as well as encouraging disclosure of all witnesses
who might render expert testimony. The trial
court should be given adequate time to make
the determination of expert witness reliability
but not so far before trial that an excluded ex-
pert witness can be replaced by another of the
same ilk. All too often the credentials, character,
and background of the expert are perhaps more
important than his/her actual testimony. An ex-
pert who possesses a distinguished record in
<<    
     Article
      >>
236 Meister and Masella
these areas provides the best chance of securing
honest and objective testimony.1
Medical Expert Liability
Although actions against privately hired medical
experts and court-appointed medical evaluators
are uncommon, even in cases of bizarre or bla-
tantly dishonest testimony, a trend is emerging
toward holding expert witnesses liable for their
professional errors.12 Eight states have ad-
dressed the issues of expert witness immunity.
Among them, the Supreme Court of New Jersey
held that a forensic expert is not immune from
liability for deviating from accepted standards of
practice. A California appeals court stated that
expert witness immunity "does not encourage
witnesses to testify truthfully; indeed, by shield-
ing a negligent expert witness from liability, it
has the opposite effect."12 Increased judicial rec-
ognition of the relatively new tort of expert wit-
ness malpractice may discourage unethical testi-
mony and elevate the quality of courtroom
testimony.
Beware the Hired Gun
Fletcher13 described the "hired gun" as an ex-
pert witness in a legal dispute whose opinions
are for sale to either party.13 Usually working
exclusively for the plaintiff or the defendant,
hired guns earn most or all of their professional
incomes from testifying. This testimony can be
"incomplete, deliberately shaped, . . . incompe-
tent, or transparently dishonest."11 For hired
guns, the desire for financial gain supercedes
professional ethics. Tsushima and Nakano8
wrote, "these physicians [or dentists] do great
harm to themselves and the medical [dental]
profession, and fail to safeguard rights and free-
doms of plaintiffs or defendants." The opposing
attorney has the responsibility of alerting the
jury to contradictions presented by the hired
gun. Cross-examination is the best method of
revealing hired guns on the witness stand and
casting doubt on their credibility.
Battle of the Experts
The adversarial nature of the American legal
system confers equal rights to plaintiffs and de-
fendants to present medical experts to enhance
Home | TOC |          
Index
their arguments. Attorneys perform direct exam-
ination on experts they have hired, with oppos-
ing attorneys subjecting the expert to cross-ex-
amination. This system is considered the best
route to truth and justice.
Annas,14 Bayer,15 Jerrold,16 and Laskin17 re-
mind us that instead of building bridges to truth
and justice, contradictory medical testimony can
create mountains of confusion between justice
and the judge and jury's understanding of the
medical issues and facts. Although some legal
analysts believe that lay juries come to under-
stand complex medical and scientific issues,18
others argue that juries are easily swayed by non-
factual or scientifically weak testimony, known as
junk science.19 In the ideal world, courts would
rule junk science as inadmissible, and juries
would decide guilt or innocence based on truth-
ful expert testimony and not be swayed by emo-
tional appeals. Reality forces the acknowledge-
ment that decisions by judges and juries might
not rest on the truth but on the best perfor-
mance by expert witness or lawyer, along with
exploitations of courtroom emotions.
Skillful cross-examination by an opposing at-
torney provides an important check on the use
of junk science. The cross-examiner has the op-
portunity not only to impugn an expert's credi-
bility but also to cast doubt on the reliability and
significance of the medical expert's evidence.1
Neutral Expert
Hope for improved quality of medical expert
testimony may be found in the independent or
neutral scientific expert or panel. A neutral
medical expert can be appointed by a judge, as
opposed to the plaintiff or defendant selecting
his own. A neutral expert's work is separate from
the adversary proceedings in the courtroom,
and his/her findings may override other court-
room expert testimony (Rule 706, Federal Rules
of Evidence).20'21 Neutral panels or experts offer
the prospect of reduced jury confusion about
medical facts and issues and elimination of the
battle of the experts. The excitement over this
approach, however, is tempered by the inherent
difficulty of changing an established and power-
ful legal system.
<<    
     Article
      >>
Have Gun Will Travel
Summary
A social and ethical dilemma exists in the desire
to protect the rights of injured people to sue for
professional negligence, while discouraging
false claims of injury. Part of resolving this di-
lemma lies in obtaining honest and objective
medical expert witnesses. When held to this stan-
dard, such witnesses provide a valuable contri-
bution to society. When medical experts exhibit
bias and intentionally serve the interests of only
one party in a legal dispute, professional ethics
and public interest suffer equally.
Tsushima and Nakano8 offer sound advice for
health care professionals serving as expert wit-
nesses: "physicians called on to testify should be
concerned with truth and justice, and the re-
sponsibility of determining what the truth is and
what justice is rests with the individual physician.
Doctors should accept responsibility for testify-
ing in court, support testimony with firm scien-
tific evidence, acknowledge limits of the evi-
dence as well as their expertise, and remain
objective and impartial."
References
1. Masella RS, Meister M: The ethics of health care profes-
sionals' opinions for hire. J Am Dent Assoc 132:361-367,
2001
2. Kallay T: Expert evidence should be taken into account
in 'prudent patient' jurisdictions. Med Malpract Law and
Strategy 18(12):2-3, 2001
3. Nishiyama A: Ethical issues in criminal forensic evalua-
tion. Jpn J Psychiatry Neurol 48:63-70, 1994 (suppl)
4. Fed R Evid 702 (Amended Dec. 1, 2000)
5. Fla Stat Ch 766.202(5), 102(2)(c)l,2 (1993)
Home | TOC |          
Index
237
6. American Dental Association Council on Ethics, Bylaws
and Judicial Affairs: Principles of ethics and code of
professional conduct. Chicago; IL, American Dental As-
sociation, 2000, p 6
7. Miller RD: Professional versus personal ethics: methods
for system reform? Bull Am Acad Psychiatry Law 20(2):
164, 1992
8. Tsushima WT, Nakano KK: Effective Medical Testifying:
A Handbook for Physicians. Boston, MA, Butterworth-
Heinemann, 1998, pp 2-5
9. Simon RI, Wettstein RM: Toward the development of
guidelines for the conduct of forensic psychiatric exam-
inations. J Am Acad Psychiatry Law 25:17-30, 1997
10. Rappeport JR: Ethics, the expert witness, and the search
for a higher truth. Hosp Community Psychiatry 44:390-
391, 1993
11. Appelbaum PS: Forensic psychiatry: The need for self-
regulation. Bull Am Acad Psychiatry Law 20:153-162,
1992
12. Hansen M: Experts are liable, too. Amer Bar Association
J 86:17-18, 2000
13. Fletcher JC: Bioethics in a legal forum: confessions of an
"expert" witness. J Med Philos 22:297-324, 2000
14. Annas GJ: Burden of proof: Judging science and protect-
ing public health in (and out of) the courtroom. Am J
Public Health 89(4):490-493, 1999
15. Bayer R: Editor's note: Science, justice, and breast im-
plants. Am J Public Health 89:483, 1999
16. Jerrold L: Hard science versus soft science. Am J Orthod
Dentofacial Orthop 114:467-468, 1998
17. Laskin DM: The ethics of expert testimony. J Oral Max-
illfac Surg 47(11):1131, 1989
18. Horsley JE: Testifying in Court: A Guide for Physicians
(ed 2). Medical Economics Press, Oradell, New Jersey,
1983, pp 127-132
19. Brent RL: Bringing science to the courtroom. Pediatrics
95:954-956, 1995
20. Fed R Evid 706 (Amended Dec. 1, 2000)
21. Macklin R: Ethics, epidemiology, and law: The case of
silicone breast implants. Am J Public Health 89:487-489,
1999
<<    
     Article
      >> Home | TOC |          
Index

David Versus Goliath: The State Against the

Doctor—Administrative Liability
T. Michael Speidel

Dental practice is subject to regulation by many agencies. The most formi-

dable regulatory body for many practitioners is their State Licensing Board
whose charge is protecting the well-being of the public. Because of the great
power differential between the practitioner and the regulatory body, cou-
pled with the limited due process requirements afforded the practitioner, it
is important to know your board's policies and procedures and to respond to
board inquiries appropriately. Practitioners would be wise to obtain appro-
priate legal counsel when responding to inquiries by most regulatory bodies
and especially to inquiries made by their State Licensing Board. The objec-
tive of this article is to provide a general overview of the regulatory process
and raise issues that you, as a health professional, must consider when
confronted by an inquiry from a regulatory agency affecting the practice of
dentistry. Although the major focus will be on a State Licensing Board, the
general principles are applicable to most regulatory bodies. (Semin Orthod
2002;8:238-242.) Copyright 2002, Else vier Science (USA). All rights reserved.

enerally, regulatory agencies are either
G federal or state. In addition to your State
Licensing Board, there is the State Department
of Health, the Department of Human Services,
the Pollution Control Agency or Environmental
Protection Agency, the Department of Com-
merce, Occupational Safety and Health Admin-
istration, Medical Privacy Rules, the State Attor-
ney General's Office, and countless other
alphabet soup regulations and organizations.
There are also regulatory bodies that are neither
federal nor state but have considerable influ-
ence on a dental practice such as your dental
society's peer review committee, a hospital's
peer review committee, or a managed care orga-
nization's peer review committee. Even though
their actions may not have the force of law, they
can be as important in affecting your practice as
From Hagglund, Weimer and Speidel, Minneapolis, MN.
Presented in part at the 1999 Annual Session Risk Management
Program in San Diego, CA.
Address correspondence to T. Michael Speidel, DDS, MSD, JD,
Hagglund, Weimer and Speidel, 4000 Water Park Place, 5101
Olson Memorial Highway, Minneapolis, MN 5.5422.
Copyright 2002, Ehevier Science (USA). All rights reserved.
1073-8746/02/0804-0010$35.00/0
doi:10.1053/sodo. 2002.127872
agencies that do have the force of law behind
them. Your response to a nongovernmental
body should be similar to your response to a
state or federal regulatory agency.
At the outset, one should understand that
regulatory agencies' interests are often in con-
flict with the practitioner's interests. Regulatory
agencies' primary interests are to protect the
well-being of the public, to set standards of care,
to affect quality improvement, to contain costs
and encourage the use of scarce resources
wisely, to maximize cost/benefit ratios to the
patient, to discourage behavior on the part of
the practitioner that society deems inappropri-
ate, and to rehabilitate practitioners whose in-
terests in autonomy and profitability are in con-
flict with the regulatory body's interests.
Most dentists think of their State Licensing
Board as primarily responsible for the testing
and granting of licenses to practice. Although
licensing is an important role, the State Licens-
ing Board is also charged with disciplining vio-
lations of the state statutes related to dental
practice. The grounds for discipline may be in-
competence, negligence, impairment, fraud and
dishonesty, mental incapacity, sexual impropri-
eties, or miscellaneous infractions such as, in

238 Seminars in Orthodontics, Vol 8, No 4 (December), 2002: pp 238-242

<<    
     Article
      >>
David Versus Goliath
some states, failure to complete mandatory con-
tinuing education and failure to pay income
taxes or child support. In many states, the ma-
jority of complaints are in the category of incom-
petence, welfare fraud, or drug and alcohol
abuse.
The reason that the relationship between the
practitioner and the regulatory agency is similar
to David versus Goliath is based on the relative
differences between the two in experience, re-
sources, and the limited requirements for fair-
ness or due process. Most people think of due
process as something guaranteed under the
United States Constitution. That high standard
is only applicable when governmental action will
impair an individual's life, liberty, or property.
When governmental or quasi-governmental ac-
tions do not constitute a denial of an individual's
life, liberty or property, the government or other
agency may not be required to give the person
any hearing or process. Obviously, State Licens-
ing Board's actions do not deprive one of his/
her life or liberty, and it is unclear whether or
not one has a property right in a dental license.
Furthermore, having a license and having a re-
stricted license are quite different things.
Regulatory agencies use a balancing test in
determining how many due process require-
ments must be provided to individuals who ap-
pear before them. The regulatory agency bal-
ances the worth of the procedure to the
individual against the cost to society as a whole.
In most cases, the respondent (you, the lic-
ensee) will get limited due process. It is the
belief that the government's interest in protect-
ing the public would be hindered if every inves-
tigation had to be conducted with full due pro-
cess protections. Therefore, the due process the
respondent is accorded is usually limited to no-
tice, an opportunity to present your case, an
opportunity to present evidence, an opportunity
to have an attorney present, and a transcript of
the proceeding.
In contrast to a formal judicial proceeding,
the respondent may or may not get a neutral
decision maker. Usually, the respondent will not
be able to strike jurors or the judge. The respon-
dent may not be able to present witnesses or
expert witnesses, exclude irrelevant or immate-
rial evidence, and confront and cross-examine
witnesses. Often you do not know, and cannot
find out, who made the complaint against you.
Home | TOC |          
Index
239
Furthermore, you will not be able to compel the
appearance of witnesses, conduct pre trial discov-
ery, have a public hearing, have a jury trial, or
invoke privilege against self-incrimination. You
also may not be entitled to a decision based on
the record with a statement of supporting rea-
sons.
In addition to the absence of some of the
customary requirements of constitutional due
process, there are some other limitations to the
administrative hearing. The rules of civil proce-
dure do not apply. The rules of evidence do not
apply; evidence that might be inadmissible in a
courtroom because it is not reliable or relevant,
may be admissible. The hearing body may bring
up issues that were not in the original complaint,
issues of which you had no notice. You may have
to testify in a narrative manner, thus risking
self-incrimination. The decisions may be made
by individuals who lack expertise in all areas of
dentistry, especially in a dental specialty area.
The members of the hearing body may be pri-
marily political appointees. Finally, when judg-
ing you, the hearing body may subject you to
unrealistically high personal or aspirational stan-
dards of care, rather than the much lower legal
standard of care.
In most states, the legislature has given the
licensing boards broad subpoena power. Stat-
utes mandate cooperation with board investiga-
tions, and the prohibition against self-incrimina-
tion attaches only in a criminal case. It is small
consolation that evidence and testimony consid-
ered in the disciplinary hearing usually cannot
be used in a criminal case, but, in a subsequent
proceeding, the evidence and prior testimony is
often discovered. These factors place the dentist
at a considerable disadvantage in the hearing
process.
It is important to consider the source of any
complaints to the dental board. Complaints can
arise from patients, employees (more likely,
former employees), other dentists, or investiga-
tors employed by the Office of the Attorney
General or the Board of Dentistry itself. Usually,
complaints must be submitted in writing, and
the board is required to investigate and respond
to all complaints it receives. If the subject of the
complaint is nonjurisdictional, such as an alle-
gation of criminal conduct, the complaint will
be forwarded to the appropriate jurisdiction for
disposition.
<<    
     Article
      >>
240 T. Michael Speidel
If the subject of the complaint is within the
jurisdiction of the licensing board, the board
may gather additional information from the
complainant or the respondent (licensee). Usu-
ally the board will request from the respondent
a written response to the allegations in the com-
plaint. The board may send an investigator for
record reviews, interviews, or site inspections.
The investigators are no t your friends; be cordial
but no more helpful than is necessary.
Some complaints are fact specific and present
little or no proof problem (eg, example the
absence of a properly functioning sterilizer).
Others are more difficult to adjudicate such as
allegations of incompetence that are especially
problematic because bad results may stem from
one having actually rendered negligent treat-
ment or substandard care, may occur through
no one's fault, may be satisfactory but fall below
the patient's or another complainant's expecta-
tions, or may simply be a result of poor commu-
nications between patient and practitioner.
After the board has finished its investigation,
the board may dismiss the complaint, issue an
advisory letter, or request a conference with the
respondent. The conference may be informa-
tional or investigational and-either relatively in-
formal or quite formal. All conferences should
be considered formal by the respondent and
treated as such. After the hearing, the board will
determine whether the complaint should be dis-
missed or whether corrective action is war-
ranted.
In addition to dismissal of the complaint, the
Board of Dentistry may take a number of correc-
tive actions. The most common are as follows:
(1) admonish or warn; (2) reprimand or cen-
sure; (3) impose conditions to continue prac-
tice, such as remedial course work, practice un-
der supervision, or limitations that restrict the
scope of one's practice; (4) required provision
of nonremunerative service; (5) submission to
periodic office inspection; (6) temporary sus-
pension of one's license; or (7) permanent re-
vocation of one's license. In the case of an ap-
plicant for licensure, licenses may be either
denied or conditioned.
In determining a disciplinary response, the
board will consider several factors including the
existence of or clarity of a practice standard, the
seriousness of the misconduct, the frequency or
duration of the misconduct, the perceived intent
Home | TOC |          
Index
of the practitioner, any corrective action (s)
taken by the practitioner, the potential for de-
terring future misconduct, the potential for re-
habilitating the practitioner, the costs of investi-
gations and hearings incurred by the board, and
the practitioner's ability to pay.
The board has the burden of proof when
taking adverse action against a licensee; how-
ever, the standard of proof is usually the rela-
tively low preponderance of the evidence test. In
the case of an application for licensure, the bur-
den is on the applicant, and the standard of
proof is the same.
Licensees should be aware that the board's
findings, other than dismissal, are usually public
and may require voluntary or involuntary report-
ing to other agencies. Adverse findings may re-
sult in bad publicity for you or your practice, a
limitation on current or future employment,
prohibition from future employment or inde-
pendent practice, and a limitation to your pro-
fessional liability insurance coverage or a change
in insurance premium cost.
When one receives a notice of complaint
from the Board of Dentistry, a peer review com-
mittee, or even from a plaintiffs attorney, it is
very important to respond in a proactive man-
ner. The following suggestions may be helpful in
responding to the notice of complaint.
Do not delay; read the entire notice of com-
plaint and respond in a timely manner. To do
otherwise may result in a default action against
you.
Inform your professional liability insurance
carrier. You may or may not have insurance
coverage; however, most policies require you to
do so promptly. To do otherwise, may jeopardize
any coverage you may have.
Call an attorney who has experience in deal-
ing with your dental board and also possesses a
knowledge of dentistry, if possible. This is impor-
tant because every licensing board is slightly dif-
ferent. Make sure your attorney knows the
board's rules and the Dental Practice Act. Even
if you are asked to do no more than write an
explanatory letter to the board, make sure you
know and follow any required format. Let an
experienced attorney review your response as he
may be able to minimize the charges and keep
you from making any damaging admissions.
If you have coverage under your professional
liability policy, the insurer will provide an attor-
<<    
     Article
      >>
David Versus Goliath
ney for you. Usually, you will have several indi-
viduals from which to choose. Interview them,
and make sure you are comfortable with the
attorney. If you are not, find your own, and ask
the insurer to approve their representation of
you. If you wish, you can have your personal
attorney work with the attorney provided by the
insurance company. Often it is a good idea to
have someone who represents only your inter-
ests. The attorney provided by the insurance
company may have somewhat divided loyalties.
In the event that you are required to appear
for a hearing, it is extremely important to be
represented by appropriate counsel. Many law-
yers expect a hearing to be like a formal judicial
proceeding and may irritate the hearing panel
members by trying to impose their idea of ap-
propriate procedure on a board composed of
dentists and others who are totally unfamiliar
with formal judicial procedures.
The aforementioned recommendations may
appear self-serving, but because of the relative
power differential between you and the regula-
tory agency, to appear pro se (representing one-
self), may jeopardize your defense and, in turn,
your livelihood.
Refrain from conversations with the board
members and the complainant (if you know who
the complainant is). If you do not know who the
complainant is, do not discuss the complaint
with anyone until you consult with your attorney.
If required to write a response to a board's in-
quiry, make sure your attorney approves it, and,
if necessary, your insurance company as well
before sending it to the board.
You must be absolutely candid with your at-
torney. To do otherwise may jeopardize your
defense and subject your attorney to the possi-
bility of being blindsided by damaging evidence
unknown to him/her. Do not be concerned
about making damaging admissions to your at-
torney; your admission will be protected by the
attorney/client privilege or the work-product
doctrine. The foregoing presumes that you also
maintain confidentiality. If you disclose pro-
tected information to third parties, you may be
considered to have waived the attorney/client
privilege.
Assemble all necessary records, including
records from prior treating dentists if appropri-
ate. Hopefully, you did not accept a transfer
patient without obtaining the prior treating den-
Home | TOC |          
Index
241
tist's records, and you have kept the type of
records necessary for a good defense rather than
just the minimum records. If you cannot easily
obtain necessary records from a prior treating
dentist, you will definitely need an attorney's
help to obtain them.
Do not alter any records! To do so, may result
in evidence being construed against you and
may subject you to civil penalties. If your records
are difficult to read, include shorthand abbrevi-
ations or typewritten transcript, with explanatory
text in parenthesis, along with a copy of the
original.
Appear before the board if necessary. The
hearing will be transcribed by a court reporter,
tape recorded, or both. The board will ask you
questions, not your attorney. Your attorney may
not answer for you. In answering questions, lis-
ten carefully to the question. Make sure you
understand the question. If you do not under-
stand, ask the questioner to repeat or clarify the
question. If you answer, it will be presumed that
you understood the question. If you do not know
the answer to a question, say so. Do not guess. If
you must guess, make sure the record reflects
that it was a guess.
Tell the truth. Tell your side of the story in as
succinct a manner as possible. Before speaking,
take time to formulate the answers in your mind,
then answer and stop speaking. Do not volun-
teer any information not asked for as you may
make a damaging admission or inadvertently
give the questioners additional potentially dam-
aging avenues to explore. One of the question-
er's favorite tricks is to sit back and let you
ramble, giving you enough rope to hang your-
self. Do not volunteer information unless you
are sure it will help you. Do not try to be helpful
or a nice guy. Use specially prepared exhibits,
such as a time line chart, if it will help tell your
story.
Honest admissions of misconduct, along with
good explanations and credible evidence of any
remedial action taken, may minimize or prevent
adverse action by the board. In answering ques-
tions, being vague or evasive damages your cred-
ibility. If you do not think a question is fair, often
you can rephrase it and then answer the ques-
tion. Finally, do not try to answer compound
questions because the correct answer for one
part of the question may be incorrect for an-
other part of the question.
<<    
     Article
      >>
242 T. Michael Speidel
Most lawyers have videotapes available on
how to conduct yourself at a deposition; take
time to learn the principles. Watch the tapes a
few times, or spend a few more dollars and ask
your attorney to prepare you for the hearing. It
will be one of the best investments you ever
make.
Even formal hearings are somewhat informal,
like a deposition. Although the rules of evidence
may be relaxed, the proceedings will be re-
corded. Do not let the seeming informality of
the proceeding lull you into adopting a casual
approach. Every word you utter is important.
Be serious, humble, and cordial at all times.
Do not become indignant or defensive in your
manner. Remember, the board members are
responsible to the public, and they are just doing
their job.
After the hearing is completed, you will re-
ceive a notice of the board's proposed action. If
you disagree with the proposed action, your at-
torney may be able to suggest to the board some
reinterpretation of the facts or a modification to
the proposed action. The board may or may not
accept the counter-proposal. If the board does
not accept your proposal, there are limited op-
portunities for formal appeal to an administra-
tive law judge or a lower court. Appeals are very
tricky and should not be attempted without com-
petent legal help. Before appealing, try to ascer-
tain the likelihood of a successful appeal in your
jurisdiction because appeals are often unsuccess-
ful and simultaneously expose you to additional
expenses for the appeal process.
Home | TOC |          
Index
In sum, actions by regulatory agencies are
truly David versus Goliath proceedings. This is
not a blanket condemnation, just a fact of life.
You can give yourself an advantage by knowing
what they do and how they do it. You must be
aware of the fact that you are involved in a
process that is somewhat informal and does not
have the built-in protections for the respondent
that formal judicial proceedings have. Even
though the proceeding is informal, you must
take it seriously and prepare adequately. The
results are just as important as they would be if it
were a formal judicial proceeding. Finally, your
attorney, depending on the jurisdiction in which
you practice, is not particularly welcome at ad-
ministrative proceedings and cannot protect you
as well as in a formal proceeding. Therefore, you
have to protect yourself by thorough prepara-
tion and obtaining and following good legal ad-
vice.
This general overview of the basic functions
of any regulatory agency and their relationship
with you, as a dental practitioner, cannot be
considered legal advice from the author, editor,
or publisher. The opinions expressed are those
of the author. The principals expressed may not
be applicable to a specific situation. All agencies
function slightly differently from one another.
The facts differ from situation to situation, and
the law is constantly changing. If you as the
reader have questions about your particular sit-
uation, you are encouraged to seek competent
and timely counsel.
<<    
     Article
      >> Home | TOC |          
Index

Transitioning the Orthodontic Practice:

Seller's Concerns and Perspectives
Randall K. Berning

Sellers have many choices to make in planning and implementing the

transition of their orthodontic practice. The following comments will
present a number of the more important concerns. Who is in charge of the
transaction, how to determine a value for the assets being sold, what type
of transition is best, why the terms can be more important than the price,
developing a covenant not to compete, and cases in progress are among the
items discussed. Although the listing is not intended to be comprehensive,
it is intended to provide an overview of these topics in a sequence that
should prove helpful for any seller. (Semin Orthod 2002;8:243-248.) Copy-
right 2002, Elsevier Science (USA). All rights reserved.

he sale of all or part of one's practice can seller should lean toward a more immediate
T be one of the more vexing issues for orth-
odontic practice owners who love their chosen
transfer.

profession and want to continue practicing but
for fewer days per month. This can also be a
significant concern for the practitioner who has
built a very large practice in either a single or
multilocation format. The relevant consider-
ation is whether the practice will benefit from a
long-term succession plan or a short-term tran-
sition. If it is apparent that a short-term transi-
tion will impair the practice by not enabling the
next generation of practitioner(s) to gain ad-
ministration, finance, personnel, and clinical ca-
pability sufficient to effect a successful transfer,
the doctor/owner should lean toward a succes-
sion plan with the sale of incremental equity
interests. By way of contrast, if it is apparent that
either a smaller practice, or each of the offices of
a multilocation practice that will be split, cannot
support two practitioners even with a reduced
schedule for the doctor/owner, or the available
space is cramped or the patient flow limited, the
Address correspondence to Randall K. Berning, JD, LLM, Pres-
ident, Berning & Affiliates, Inc, National Administration Office,
3400 Tamiami Trail North, Suite 201, Naples, FL 34103-3717.
www. berningaffiliates. com
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0804-0011$35.00/0
doi:10.1053/sodo.2002.127873
Concern for the Buyer or Not?
This is related to the previously mentioned issue
but is different. In a professional practice trans-
fer and sale, unlike a more routine commercial
business sale, one of the major differences is, or
should be, the continuity of care for the patients
who have trusted a chosen practitioner. Viewed
from this perspective, the question arises, how
does the exiting practice founder, owner, or
partner assure continuity of a similar commit-
ment to treatment, both philosophically and
mechanotherapeutically? In other words, for
some doctors, it is very important to find a buyer
or new partner who shares their philosophy of
care and commitment to the mission of the prac-
tice. Others are far less concerned about philos-
ophy and more concerned about technique and
how cases will finish. If a seller is very concerned
about either of these issues then it is important
to get the issue out in the open and spend
adequate time qualifying a prospective buyer to
reduce this risk of remorse during or after the
transition.
Who's in Charge?
In many routine commercial business sales, out-
side of professional practice transfers, the buyers

Seminars in Orthodontics, Vol 8, No 4 (December), 2002: pp 243-248 243

<<    
     Article
      >>
244 Randall K. Berning
and their advisors are in charge. They make the
offer, they question or assert value, they do their
due diligence, and they propose terms and often
take control of document drafting. I have found
that generally because of the inexperience of the
new orthodontist in building a team of advisors
and the wide range of items to undertake cou-
pled with limited financial resources, orthodon-
tic practice buyers generally do not take control
of the process. Therefore, sellers and their advi-
sors very often put the transaction together.
However, as noted later, these transactions can
fail if the deal's terms and documents are or
appear to be unduly one-sided in favor of the
seller and the buyer does not feel that an equi-
table transaction is occurring. A word to the wise
is to be solicitous in inviting comments on the
valuation methodology and the development of
the terms for the transaction. Direct your legal
counsel to drop unnecessary legalese and draft
documents in an evenhanded manner. This will
go far in making the deal come together quicker
and in a pleasant manner.
Seller's Fundamental Considerations
How to Get to Practice Value?
Of all of the many considerations leading up to
a consummated practice transfer or equity inter-
est purchase, valuation was once, and still can
be, the toughest. Thankfully, in contrast to 20
some years ago when I started in the area of
practice transfer, there are now good resources
to guide both the buyer and seller, first to un-
derstand the theory of professional practice val-
uation and second how to apply it to formulate
the value of the practice. In the case of orth-
odontists, the American Association of Orth-
odontists has developed the Practice Transition
Alternatives binders, one for owners and one for
younger professionals. In each binder, there is a
monograph entitled Guide to Orthodontic Practice
Valuation.1 The guide discusses valuation options
but emphasizes the earnings based methods,
particularly the capitalized earnings method.
The reason for this emphasis is that by using a
practice's determined net income to arrive at
value, a buyer can identify how much of that net
they can use to pay for the debt service and
therefore be able to project how much income
they will have to live on.
Home | TOC |          
Index
Consequently, today many buyers and their
advisors are very interested in pinning down the
net and using an earnings method to arrive at
value. For sellers to apply any other method for
arriving at value simply begs the question of how
can the buyer pay for the practice debt and meet
living expenses, which will have to be answered
sooner or later and will ultimately result in using
some form of net analysis to arrive at value. In
short, it is my advice to sellers that it is much
easier and more direct to simply use a profes-
sional earning-based valuation at the start and
be done with it.
At the same time that sellers grapple with how
to arrive at value, they need to bear in mind that
value is not price. Value can be arrived at using
a valuation methodology or a group of compa-
rables. But price is what is negotiated when the
value is known and usually takes into account
other considerations such as the tax-planning
strategies of the parties, the goals of each party,
and the terms of the deal.
What Form of Transaction to Use?
Form is a basic decision but one that must be
made and in the case of orthodontic practice
transfers often seems to be made by the sellers.
What is being sold and how it is to be purchased
leads to the time-honored debate that sellers
and their advisors enter into year in and year
out. Namely, is the form of the transaction best
set up as a sale of assets, in the case of an
incorporated practice a sale of the stock of the
professional corporation, or if the practice is a
partnership, the sale of either the partnership's
assets or the partnership interest of the part-
ner (s)? The various factors favoring an asset,
stock, or partnership interest sale depend on
who is making the decision, and the answer can
be dramatically different for the seller and the
buyer engaged in the same transaction. The rea-
son I find sellers generally making the call is that
in orthodontic transfers, particularly of larger
practices that have a long-term buyout, immedi-
ate asset sales often make little sense because the
longer transfer time frame usually pushes the
sale into either an equity (stock or partnership
interest) transfer. Further, sellers generally will
prefer to sell stock or a partnership interest be-
cause they receive capital gains treatment on any
receipts from the transfer. However, the pur-
<<    
     Article
      >>
Transitioning the Orthodontic Practice
chaser of an equity interest takes the practice
subject to all liabilities including tax, contingent,
and unknown. Purchaser's counsel will often
seek to exclude certain liability categories be-
tween the buyer and seller via indemnity (it will
not apply to third-party creditors), and this is
generally a reasonable request.
Often overlooked, but of equal importance, is
what is not going to be sold. This is sometimes
thought of in the context of orthodontic cases
that will be completed by the seller in a few
months after closing or personal items that will
be taken by the seller. Attorneys refer to what is
not to be sold as the liabilities and responsibili-
ties a purchaser retains after the sale. My expe-
rience is that both areas must be addressed early
in the process of coming to terms for the parties'
proposed transaction. Cases to be completed ties
directly to risk-management concerns by the
purchaser. For example, it is far better for the
purchaser to never touch a case in the last 4 to 6
months of treatment than to have a purchaser
share portions of the patient's treatment. In the
worst-case scenario, the parties get involved with
finger pointing regarding who did what, should
the patient be disgruntled or a professional mal-
practice action occur. Active and retention cases
to be transferred should be listed in an exhibit
to the transition agreement along with all as-
sumed liabilities.
Being mindful of tax considerations. Preliminar-
ily, sellers need to bear in mind that although
many of the once common tax-planning tech-
niques are history because of legislative tax re-
form changes, some planning options still re-
main. However, if either side pushes too hard
for their best tax result, it can result in breaking
the deal. Very different tax consequences can
result from a purchase of assets or stock, but they
are best left for discussion within the context of
your personal situation with your accountant or
tax advisor.2
Rather than rehash general tax consider-
ations, it is better to focus on a couple of areas
that can make a significant impact. One has
been eluded to, specifically, admonish your tax
advisor or transaction advisor to come up with a
balanced tax approach that does not severely
disadvantage either seller or purchaser. Next,
pay attention to all allocations of the purchase
price but not late in the discussion, as many
seem to do. Instead, do it early on. A good time
Home | TOC |          
Index
245
is actually after you have secured your practice
valuation and started to weigh what the actual
purchase price will be.
If you are using a long-term transition, partic-
ularly if it involves the sale of stock, significant
balancing of interests can be done to allocate a
portion of the purchase price to the interest
equity (allowing for capital gains) and another
amount to management services (generally ordi-
nary income but currently deductible to the
payor). This type of allocation is not done as
fluff but in fact recognizes the time-consuming
mentoring that is necessary for a new practitio-
ner to educate himself/herself to sufficiently
take over the management aspects associated
with running a significant practice such as the
necessary orientation to practice administration
and clinical skill refinement. Comparable ap-
proaches can be fashioned for partnership trans-
fers. One should have a third-party arms length
negotiation of any allocation should any after
closing taxing authority question the asset allo-
cation. Obviously, the ultimate allocation should
be stated clearly in the transition agreement.
Will You Get Paid?
In most orthodontic practice transfers, other
than those that are immediate purchase and
sales, regardless of how they are to be financed,
debt-encumbered purchasers are often at a loss
to come up with much of a downpayment. Tran-
sitions that are immediate purchase and sales, if
fairly valued and if structured properly, find the
seller to be generally well protected using rela-
tively standard agreement provisions. For those
that are not immediate transfers, the lack of a
down payment throws out the old axiom that a
purchaser should put 10% to 20% down. Fur-
ther, although the amount of money a pur-
chaser can put down should have a bearing on
the sale price, transition experiences over the
last 5 years show it does not. This is because in
part of the fact that a virtual purchaser's market
exists (due to fewer purchasers) where purchas-
ers and their advisors are driving harder bar-
gains, despite the buyer's lack of resources.
An illustration of the sequence of steps to
follow in practice transitions and how they might
progress follows. First, the seller should secure a
third-party professional valuation. This valuation
must ultimately be agreed to by both parties.
<<    
     Article
      >>
246 Randall K. Berning
Second, the parties must negotiate terms for a
multiyear buyin/out in which the cash flow is
tied to the valuation and the terms. Next, each
parties tax advisors should agree on the asset
allocation vis a vis the purchase price. This is
followed by the document drafting. It is impor-
tant that the placement of a downpayment be
deemphasized; for, in reality, it is nothing more
than one more of the transition's terms. This
does not mean that a seller is not well advised to
pay attention to a buyer's creditworthiness, char-
acter, personality (yes personality), and capabil-
ity to take over the size of the practice under
consideration. These are areas that if the seller
fails to investigate, he/she does so at his/her
peril. I have seen transfers flounder with an
introvert purchaser taking over an extrovert sell-
er's practice because in part of the fact that the
referral sources and the patient base have sub-
stantially different expectations of the purchaser
than was in fact the case. Often, transitions in-
volve a structure that is detailed, well docu-
mented, and carry a purchase period for an
initial equity interest followed by a sale of the
remaining equity interest. Likewise, any com-
pensation and purchase payments must be prop-
erly structured and documented to take advan-
tage of existing tax laws.
What Is the Best Process?
For those sellers who are concerned about how
to make a deal or those wanting to avoid elemen-
tary mistakes relating to process, other commen-
tary exists.3'4 Specific helpful pointers on good
process to follow in structuring a proposed prac-
tice transfer, whether it be a buyin/out or a
more immediate variation of transferring or
transitioning from an associateship relationship
into an equity interest are provided below.
Terms not documents. In my view, doctors are
well advised to first get down what is important
to them in the form of a terms listing. This could
start with the seller identifying key items that
make it worthwhile to go through all the cost
and time of developing the practice transition. It
can also start with both parties using a facilitator
to guide them in raising key issues and items for
their mutual consideration. Only after the terms
are fairly well established and agreed to in prin-
ciple should the seller proceed to have a set of
documents prepared to reflect the parties'
Home | TOC |          
Index
terms. Drawing up the documents before agree-
ing to the terms is time-consuming and expen-
sive, and starting with them followed by subse-
quent refinement after review by both parties
and their advisors is a far more torturous path.
Starting with a term sheet ensures that the buyer
and seller have agreed on the key points before
incurring costs for attorneys and other advisors
to analyze and document the deal.
Advisors. The previous statements raise the
critical issue of getting the right advisors in the
first place. It can be tempting for a seller to use
a long-time advisor, even if that person has no
experience in this area. This can be a mistake
simply by virtue of using someone who has a
lesser degree of expertise concerning what is
required to effect the transition. This will have a
direct impact on the time and cost of getting the
transition accomplished. It is often better to use
the services of a transition consultant and/or a
corporate, business law, tax law, or health care
law attorney. After consultation with experi-
enced advisors, if desired, have the long-time
advisor review matters with you. Buyers some-
times are tempted to use family members or go
it alone until the last minute when they become
overwhelmed and finally seek counsel. The par-
ties usually fare better if they have competent
counsel, ready to review in a timely manner, the
terms that are worked out by the parties them-
selves rather than to wait too long. Both parties
should avoid strident my-way-only advisors. Deal
makers are usually better representatives than
deal breakers. Finally, have a clear understand-
ing of the roles that your advisors will play right
from the beginning. Namely, it should be the
seller's transition consultants, accountant, law-
yer, and other advisors who prepare all financial,
tax, and valuation materials for the review by the
buyer or prospective partner. A buyer's review
should not include remaking the deal that par-
ties have articulated in principle. It can take
strong nerves and competent advisors on the
seller's side to stay the course and not get side
tracked.
There is a difference. Should you seek an asso-
ciateship first for a longer period of time or a
more immediate buyin/out? Without a long ex-
planation, it is worthwhile to explore with a sell-
er's advisors, given the particulars of a practice's
situation and the desires of the seller, how best
to approach the planned transition. There is no
<<    
     Article
      >>
Transitioning the Orthodontic Practice
question that certain practice transition situa-
tions benefit from a more immediate transfer
approach. If this is the case, more needs to be
done in a shorter period of time. This includes
preparing a well-developed transition plan that
is tied to a timeline indicating to buyer and all of
his advisors when valuation, cash flow, terms,
financing, document drafts, and closing dates
are expected or anticipated.
Seller's Additional Issues
When any seller of an equity interest in a prac-
tice or sale of the entire practice moves past
valuation and negotiation of preliminary terms,
there are still additional issues that need to be
addressed prior to drafting the written agree-
ment. Some of the more important areas for a
seller are as follows.
Covenant not to compete. Nothing other than
reasonableness dicates a geographic radius or
other area and an amount of time for the cove-
nant. What considerations go into the terms of
such a covenant? In part, an appropriate geo-
graphic and temporal restriction should be suf-
ficient to protect the asset value of the practice.
Consideration needs to be made if there is any
area that a seller would want to practice in after
the sale or a buyout. Should a covenant against a
seller still apply if the purchaser defaults on the
purchase? The answer is generally no. These
issues illustrate the range of items that should be
specifically addressed in the transition docu-
ments. Be aware that most buyers want the ben-
efit of their bargain, and they will want a geo-
graphic area and time period sufficient to
protect the purchase of the goodwill they have
made. Be prepared to be reasonable.
Disputes and their resolution. In any long-term
transition with a phased buyin/out, having a
stated approach to resolving disputes is advanta-
geous. Rather than resorting to litigation, a bet-
ter strategy might be meditation followed by
binding arbitration. My reasoning, in contrast to
some members of the litigation bar, is that the
community of orthodontists is too small of a
professional population and patients too atten-
tive to their doctors' circumstances to warrant
allowing a dispute of this sort to become public
via a trial. There are a variety of arbitration
forms available these days and your legal advisor
Home | TOC |          
Index
247
can recommend a specific form based on his/
her experience.
Representations and warranties of the parties.
Some sellers, and some advisors too, unfortu-
nately do not give the weight necessary to tailor-
ing the representations and warranties made to
specific transactions. In fact, it is these provisions
that are the foundation on which each side relies
on as evidence of the basis on which their bar-
gain was made. In other words, it is the repre-
sentations or promises that each side makes that
form the basis for the transition under consid-
eration. Particular care should be made, aside
from more routine items, in representing the
financial status of the practice (what is true and
accurate as to any financial, patient, or treat-
ment status for cases completed, in progress, or
in recall, in addition to whether or not the staff
is staying with the practice, the referral base is
stable, and so on). Sellers too should seek rep-
resentations and warranties from the purchaser.
One such significant item is that a buyer under-
stands that an ownership change in the practice
is a major event, and, particularly in the situation
of the sale of an entire practice, there is no
assurance that any projected result will occur.
This is particularly the case in the specialty of
orthodontics because most practices are heavily
dependent on referral relationships that have
been either cultivated over time, age relative, or
those that are personality dependent. This as-
pect must be recognized since loyalties often
change when ownership changes occur.
Orthodontic cases in treatment and a listing of
patients. Sellers are well advised to protect their
records. There is seldom reason to leave all
records at the practice premises when leaving
the practice. Although it can be nice not to have
to store records, it has the potential to leave
sellers particularly vulnerable. All active patients,
defined as those with an appliance in the mouth;
recall patients, defined as patients who have
been examined but who have not started treat-
ment yet and are subject to regularly scheduled
recall examinations; and retention patients, de-
fined as those patients who are within 12 months
after the end of a full-treatment case, should be
listed on an exhibit to the appropriate transac-
tion document. All others patient records
should be removed to storage. Furthermore, if
the purchaser ever deems it appropriate to dis-
pose of records that were active when the pur-
<<    
     Article
      >>
248 Randall K. Berning
chase was made, notice to the seller should be
given coupled with a window of time during
which the records can be secured before their
destruction.
Conclusion
It takes a good deal of advanced planning, con-
sidering all of the choices sellers have to make,
for the transition of one's orthodontic practice
to be well thought out and executed wisely.
Without taking adequate time to plan, transi-
tioning a practice can be a harrowing experi-
ence and a rough road to travel. Far too often
over my 22 years of facilitating transitions, I have
encountered doctors who think the practice will
transfer with little problem and that they will be
hired back as employees or consultants for as
long as they desire. One common problem is the
seller's perception that any buyer should be able
to see and appreciate the value of his/her prac-
tice and that any young professional will agree to
almost any terms just to be a part of the best
practice around. It often comes as a rude awak-
ening that buyers today are better prepared, are
more capable of assessing the tenor of a prac-
Home | TOC |          
Index
tice, and can be down right picky about accept-
ing a stated value and negotiating terms than at
any time in past. Believing otherwise tends to
protract the negotiation and document prepara-
tion period. In short, advance planning, having
a thoughtful approach to meeting both parties'
needs, and the securing of competent advisors
can result in a wonderful and rewarding experi-
ence for both sellers and buyers.
References
1. Berning RK, Domer LR: "Guide to Orthodontic Practice
Valuation for Residents and Practitioners", Practice Tran-
sition Alternatives for Owners, American Association of
Orthodontists, St. Louis MO, 1999
2. For a discussion and illustration of various tax aspects for
the C corporation see: Schiff AM, Jakubowitz LM, Janes
PE: Tax Ramifications of Selling a Dental Practice: Sole Propri-
etorship, Partnership or Corporation, The Expert Series for
Dentists™, Berning & Affiliates, Inc., Naples FL, pp 12-14,
2002
3. Berning RK: "Psychology of the Deal", Practice Transition
Alternatives for Owners, American Association of Orthodon-
tists, St. Louis MO, 1999
4. Berning RK: "Common Mistakes to Avoid", Practice Tran-
sition Alternatives for Owners, American Association of
Orthodontists, St. Louis MO, 1999
<<    
     Article
      >> Home | TOC |          
Index

Transitioning the Orthodontic Practice:

Buyer's Concerns and Perspectives
Randall K. Berning

Buyers are in an enviable position in today's orthodontic practice transition

market. Depending on the area of the country one wants to practice in and
the desire for a particular type of practice situation, there are generally a
significant number of practice opportunities available. The following com-
ments and checklists are applicable to a prospective buyer assessing a
practice transition opportunity (associateship leading to a buyin/buyout or
immediate sale). They are not intended to address the issues encountered
when an orthodontist chooses to start a practice from scratch. Furthermore,
the comments are intended to be process oriented because finding, evalu-
ating, and negotiating a practice equity purchase and transition tends to fall
into a more or less routine set of steps. Finally, all comments should be
discussed with the prospective buyer's legal and tax advisors. (Semin
Orthod 2002;8:249-252.) Copyright 2002, Elsevier Science (USA). All rights
reserved.

T here are practice opportunities every-

where. Why do some younger practitioners
say I only want to practice in this town versus any
town? The reasons are many and varied. It may
be because of where a person was raised, went to
school, or the proximity to a particular environ-
ment such as the mountains or a desire to be on
either the East or West coast. If a particular
location is what drives the decision of where a
doctor wants to practice, it is important for the
doctor to recognize that this decision may be
based as much on emotion as on rational choice.
Once recognized, the younger professional
should act appropriately and shy away from op-
portunities that sound to good to be true partic-
ularly when these opportunities are located in
places you do not want to be.
Why limit your responses? Because it can be a
very time-consuming experience to meet and
discuss practice opportunities for both the pro-
spective buyer and seller. Also, there are cost
Address correspondence to Randall K. Berning, JD, LLM, Pres-
ident, Berning Affiliates, Inc, National Administration Office, 3400
Tamiami Trail North, Suite 201, Naples, FL 34103-3717. www.
berningaffiliates. com.
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0804-0012$35.00/0
doi:10.1053/sodo.2002.127874
considerations. Add to this mix a high probabil-
ity of disappointment when one of the parties is
already disinclined to practice in a particular
area and you have your answer to the benefits of
limiting one's responses to potential opportuni-
ties.
On the other hand, if a younger professional
is not committed to a particular locale, then
he/she needs to evaluate and compare opportu-
nities, weighing the various pros and cons of a
particular practice, only one item of which is
geographic location. In the latter case, as noted
previously, there are practice opportunities ev-
erywhere, and the task then becomes how to
evaluate these opportunities.
Is the Decision Based on a Particular
Seller, Practice, or Profile?
Some younger professionals are clear about
what they want in a practice. They may want to
associate with an established practitioner, seek
out a partnership opportunity, or purchase a
practice from a particular doctor. It can be a
doctor that he/she met in their residency pro-
gram or at an orthodontic meeting, or someone
from their home town. If this is the case, and
there is a reciprocated interest, both parties

Seminars in Orthodontics, Vol 8, No 4 (December), 2002: pp 249-252 249

<<    
     Article
      >>
250 Randall K. Berning
must carefully build on that personal interest to
develop, as noted later, the practice's financial
and valuation information along with the nego-
tiated terms that will allow for a fully developed
business transaction relationship to be finalized.
Others may have an interest in a particular
style of practice, a large versus a small practice,
one based on a particular philosophy of care, or
some other desired practice profile. To the de-
gree that a younger professional can elucidate
the practice style or profile he/she seeks and
then proactively search for just that format, the
more the likelihood that it will be found. Most
younger professionals fall somewhere in be-
tween and at times seem unclear about exactly
what they would like in a practice. In short, the
more specificity regarding the attributes of a
practice and or practitioner that a younger pro-
fessional can identify ahead of time when seek-
ing a practice opportunity, the better.
Following A Consistent Process
Qualifying An Opportunity
If a younger professional is actively searching for
practice opportunities, there will be a number of
interested practice owners open to discussing
their practice and the expectations that they
have in mind. However, unless the buyer uses a
consistent series of steps or a checklist to follow
when qualifying various opportunities, he/she
can quickly miss key items in evaluating and
comparing opportunities. The following brief
listing is prepared to assist prospective buyers.
Buyers should feel free to tailor it to their own
needs. It is important to bear in mind that in
many cases a buyer is on the receiving end of a
thought process that an owner or group practice
has been considering for some time. Buyers of-
ten have less time to consider each offer and
therefore the task becomes one of evaluating the
opportunity in a timely manner and, as may be
needed, adding provisions that balance the
terms between the parties resulting in meeting
both the buyer's and the seller's particular
needs.
Key questions. Start by qualifying the doctor/
owner/partner. It can be helpful to have several
meetings with the practice owner or the doctors
comprising a group practice. Visit their offices
and spend more than 1 or 2 hours. In effect, you
Home | TOC |          
Index
will be interviewing them to determine if this is
the type of person that you want to forge a
personal and business relationship with or
whether this person is a "my way or the highway"
individual or group. Discuss treatment philoso-
phy and the protocol for finishing cases. Discuss
management style and philosophy. Do you share
these common philosophies? Evaluate the per-
sonality and character of the doctor or the
group. Is this a mentoring kind of person or
group, or are you expected to be on your own
immediately after the transaction? Evaluate your
personality and character. Do you perceive there
to be a good fit of personality and character
between the parties?
Starting a relationship. If you are asked to be-
gin with the practice as an associate, request a
draft of an employment agreement. Do not start
working in the practice until the agreement is
presented and you have time to review it with
your advisors, approve it, and sign it. Do you
understand the provisions?1 A good test of your
understanding is seeing if you could explain the
contractual provisions to your friend or signifi-
cant other? Is the compensation and benefits
package clearly defined? Has the agreement
been reviewed to ensure there are no ambiguous
provisions? Has your risk exposure been evalu-
ated? Your advisors should be able to conclude
that the agreement is both fair and balanced,
thus helping to ensure that the relationship gets
started on the right track.
Buyin or Buyout versus an outright sale. If the
associateship is likely to lead to a buyin/buyout
or an immediate sale with a potential hire back,
ask and get a commitment as to when you will
meet with the owner or partners to discuss
terms. Often, the terms for the transition will be
agreed to in the initial documents requiring
only a triggering event such as a period of time,
achieving a certain gross or production, and so
on to conclude the transaction. A contract's
terms can be thought of as the core items relat-
ing to building an equity relationship. For exam-
ple, discuss what will be the format for valuation,
how will the price be paid, and what happens in
the event of disability or death. In my view, the
process of developing the contract's terms shows
compatibility, decision-making ability, and a
shared interest between the parties. It can be
helpful to secure the services of a facilitator to
raise key terms for both parties.
<<    
     Article
      >>
Transitioning the Orthodontic Practice
For any proposed equity purchase-related
transaction, get the practice valuation prepared
by a third-party professional valuation firm. One
unbiased approach is to have the firm apply the
capitalized earnings method as described in the
Guide to Orthodontic Practice Valuation for Residents
and Practitioners in the American Association of
Orthodontists Practice Alternatives Program
binder.2
In certain situations, the offer may be for a
very short period of associateship or an immedi-
ate purchase and sale, either containing provi-
sions for a hire back. In such a case, you will
need to move quickly past vague statements of
how the deal will be structured to more solid
terms and financial information. Ask for a
binder or packet of materials that contain at
least the following information: first, a third-
party practice valuation as per the Guide to Orth-
odontic Practice Valuation for Residents and Practi-
tioners as noted earlier or an indication of when
it will be prepared. Second, a cash flow projec-
tion showing, without guaranteeing, that you
can earn a comfortable living and pay the debt
load based on this valuation. Third, a list of the
terms or the name of the professional who will
be helping you and the practice owner or group
develop the terms. Next, be sure to check the
background of your advisor(s). Finally, you must
have reference material on the practice such as
financial and tax information. You may be asked
to sign a confidentiality agreement before being
given this information; make sure that this
agreement is also reviewed.
Buyer's Fundamental Considerations
If Entering Into an Associate Period, for
How Long, How Are You Paid, and Are
There Any Restrictions?
Because many practice transitions occur with a
preliminary associateship period, it can be help-
ful to have a brief list of items to discuss between
the parties. Generally, the parties should discuss
or develop a clear statement of the duties of the
prospective employee. These items should in-
clude but are not limited to (1) what treatment
and patient management standards the practice
maintains and what is expected from the em-
ployee in this regard; (2) a clear understanding
concerning the days and hours to be worked; (3)
Home | TOC |          
Index
251
how billings and collections are to be managed
and by whom; (4) the office policies and proce-
dures of the practice to be followed including
the requirements concerning record keeping
must be made clear; (5) the confidential nature
of information that an employee is exposed to
while working at the practice; (6) any prohibi-
tions on soliciting patients or referral sources
and any covenant not to compete; (7) the exact
compensation package, its format relating to tax
status, an understanding of any benefits relating
to payments for health or professional liability
insurance, along with a clearly defined vacation
policy; and (8) the terms for termination.
All of these should be part of a well-prepared
associate-employee agreement.
If a Buyin/Buyout or a Transitional or an
Immediate Sale What Information Is
Needed?
There are two groups of important information
for a prospective partner or buyer to receive and
evaluate. The first group is nuts and bolts ori-
ented and often represents information deemed
important by the doctors. The second are struc-
tural such as legal and tax aspects and are im-
portant to the advisor(s). The first grouping
varies widely by doctor and practice transition. A
minichecklist might include such items as the
following.
1. What is the number of patients in treatment,
in retention, and on observation; has this
number increased or decreased dramatically,
and what is the mechanism to transfer them
most effectively?
2. Who are the practice's referrers, ranked from
most favorable to least, and how is each re-
ferral group to be oriented to the new prac-
titioner?
3. Who on the staff is effective and likely to
continue with the practice, and what is the
best way to ensure a smooth transition of
staff?
4. At the point of an eventual buyout or at the
sale of the equity interest of the current
owner, what is an appropriate set of restric-
tions (nonsolicitation, covenant not to com-
pete, and so on) to ensure that the good will
of the practice transfers to the new owner
without threat?
5. How should the practice location identity be
<<    
     Article
      >>
252 Randall K. Berning
maintained, with regard to extending the cur-
rent lease; if there is ownership of the real
estate ownership by the seller should there be
an option to buy, right of first refusal, and
method to arrive at value of the real estate
negotiated in tandem with the practice equity
purchase?
6. If there is to be a period of partnership be-
fore a buyout, what details should be dis-
cussed regarding a buy back of either party's
interest in the event of either a permanent or
temporary disability or the death of either
party?
7. If there is to be a period of partnership, what
provisions will provide for effective dispute
resolution in the event the parties reach an
impasse on a critical practice, patient, staff,
financial, or tax issue?
8. If there is to be a period of partnership be-
fore a buyout, what provisions will be negoti-
ated now regarding decision making author-
ity relating to the ongoing affairs of the
practice during the partnership period?
The second grouping is more standardized in
that various experienced advisors should be ca-
pable of guiding each party through the consid-
erations relating to the issues noted later. This
checklist should be discussed and amended with
the advisors that a prospective associate, partner,
or buyer selects and should then be tailored to
the particular transaction under consideration.
In all proposed equity interest or business trans-
fers, it is important to thoroughly understand
the underlying practice. An in-depth investiga-
tion of the seller's practice and business prac-
tices must be completed. This is often referred
to as a buyer doing his/her due diligence, and
no young professional should feel reticent to ask
for and review any piece of financial, practice,
and tax information relating to the practice. It
can be helpful to secure the following items
among others.
1. The practice's financial statement along with
federal and state income tax returns for the 3
previous years.
2. The balance sheet and income statements for
the 3 previous years.
3. The schedule of production and collections
for the practice with contracts receivable and
accounts receivable broken out and coordi-
Home | TOC |          
Index
nated with all accounts written off for the 3
previous years.
4. The practice valuation coupled with support-
ing patient, practice, and financial informa-
tion including clear title to all assets refer-
enced.
5. Tax considerations relating to asset, stock, or
partnership form of transaction.
6. Liabilities being assumed, if any, and, pend-
ing litigation, if any.
7. A proposed covenant not to compete and any
nonsolicitation provisions.
8. The representations and warranties of the
seller. This is one of the most important areas
on which buyers should focus. These are as-
surances that a seller has, in effect, told you
that you can rely on regarding all of the
statements or information provided.
Conclusion
It takes a nimble professional to move quickly
from an academic environment or an unsatisfac-
tory current associateship to a new practice re-
lationship and to have met the challenge of
doing so in a business like fashion, informed,
well advised, and conversant with all available
options. To do otherwise, however, is to risk a
transaction that is out of financial, legal, or prac-
tice balance. Once in a binding relationship, it is
difficult to undo it. Just as for cautious sellers,
advanced planning with an emphasis on looking
for a well thought-out approach with supporting
financial, valuation, and material terms will of-
ten make all the difference to the buyer. There
is too much at stake with a career and lifestyle in
the balance to do otherwise. Too often over my
22 years of facilitating transitions, I have encoun-
tered younger professionals who rely on emo-
tion and the "feel-good" impressions of a seller.
Today that is not enough. This is the time to do
your homework, and, if you do, in the vast num-
ber of instances, the transition will be fine.
References
1. Berning RK: Evaluating employment agreements, in A Guide for
New Dental Practitioners. California Dental Association, Commit-
tee on the New Dental Professional, Sacramento, CA, 1997
2. Berning RK, Domer LR: Guide to orthodontic practice
valuation for residents and practitioners, in Practice Alter-
natives Program. American Association of Orthodontists,
St. Louis, MO, 1998
<<    
     Article
      >> Home | TOC |          
Index

Author Index
Babb, L.K., 141 Johnston, L.E., 54, 87 Ramsay, D.S., 29
Bailey, L'.T.J., 173 Reale, T., 216
Baumrind, S., 102 King, G.J., 1 Rugh,J.D., 35
Bays, R.A., 35 Kotick,J., 220
Berning, R.K, 243, 249
Sadowsky, P.L., 51
Braun, S., 149 Lyons, E.K., 29
Sankey, W., 130
Buschang, P.H., 130
Sarver, D.M., 173
McGorray, S.P., 92, 141
Sheats, R.D., 52, 62, 77
Derakhshan, M., 43 Mah,J., 2
Speidel, T.M., 238
Dolce, C., 6, 35, 141 Malone, J.S., 6
Masella, R., 234
English, J.D., 130 Meister, M., 234 Taylor, M.G., 141
Miller, R.J., 43
Franklin, E., 210
Vaden, J.L., 113
Fuhrmann, R.A.W., 17, 23 Pankratz, V.S., 62, 77
Van SickelsJ.E., 35
Pearson, A.V., 228
Vastardis, H., 13
Hartsfield, J.K., Jr., 113 Pearson, L.E., 113
Hatch, J.P., 35 Phillips, C., 67
Huang, G.J., 162 Ploumis, E., 198 Wheeler, T.T., 6, 111
Pollack, B.R., 205 White, Raymond P., Jr., 173
Jerrold, L., 185, 187 Proffit, W.R., 173 Widmer, C., 155

Subject Index
Abandonment, of patient, vs. dismissal, 200, 203 Bill payment. See Fee collection
Accuracy, of tests, 97-98 Bimodal distribution, in statistics, 64
Activator protein-1, in strain translation, 9 Binary test, definition of, 92
Adherence, to treatment, with headgear, moni- Biomechanics, in vertical dimension manage-
toring of, 29-34 ment, 149-154
Administrative liability, 238-242 Bionator, for Class II malocclusion, vertical treat-
Agenesis, tooth, agenesis of, 14-15 ment effects of, 141-148
Agreement, of tests, 97-98 Biostatistics. See Statistics
Airway obstruction, in hyperdivergent open bite, Bite blocks
132-133 for anterior open bite in excessive vertical
Align Technology, Invisalign System of, 43-50 dimension, 150
Allergy, breathing in, vertical dimension devel- posterior, for hyperdivergent open bite, 134-
opment and, 116, 126 135
Alternative hypothesis, 71 Bite forces, in hyperdivergent open bite, 132-
Analysis of association, in statistics, 58-59 133
Analysis of variance (ANOVA), 57-58, 83-84 Bite plane, headgear with, for Class II malocclu-
Apert syndrome, phenotypic differences in, 113
sion, vertical treatment effects of, 141-
Appliances, functional, for anterior open bite,
148
stability of, 163
Bjork guidelines, for condylar rotation predic-
Arches and arch wires, for anterior open bite,
tion, 123-124
stability of, 163-164
Associateship, in practice transfer, 246-247, 250-251 Bone, strain on
detection of, 8-9
Bilateral sagittal split osteotomy, rigid vs. wire translation of, 9-11
fixation of, skeletal and dental changes Bone morphogenic proteins, in osteoblast differ-
in, 29-34 entiation, 7

Seminars in Orthodontics, Vol 8, No 4 (December), 2002: pp 253-260 253

<<    
     Article
      >> Home | TOC |          
Index

254 Subject Index

Breathing, mouth, vertical dimension develop- Data

ment and, 126 distribution of, 56, 62-66
Bruxism, vertical dimension increase and, 159 normality confirmation of, 66
Buying, of orthodontic practice, 243-244, 249- types of, 62-63
252 Defendant, in lawsuit, 228-233
Dentist-patient relationship, 191
Calcium ion channels, activation of, in strain, 8 communication in, 191-192
Categorical data, 62, 78-79 documentation of, 189
Causal interference, in statistics, 108-109 establishing, 188
Cell adhesion, in strain detection, 8-9 general consent for, 190
Central point, in distribution, 64 with minor patient, 190-191
Central tendency, in statistics, 56, 78-80, 102-103 mutual obligations in, 187-188
Cephalometry refusal to establish, 188
for Class II malocclusion, 141-148 right to know in, 193-194
three-dimensional, for treatment planning, termination of, 188-189
17-22 Dentoalveolar development, vs. facial height,
Chi (x)2 test, in statistics, 85-86 124-126
Chin cup, vertical, for hyperdivergent open bite, Deposition, in lawsuit, 217, 231-232
135 Descriptive statistics, 55-56, 77-78, 78
ClinCheck, in Invisalign System, 44 Diagnostic likelihood ratios, 98
Clinical significance, 106-108 Discipline, by state licensing boards, 238-240
Collection, of fees. See Fee collection Discovery, in lawsuits, 231-232
Communication Dismissal, of patient, 188-189, 200, 203
in dentist-patient relationship, 191-192 Dispersion, in statistics, 78-80, 102-103
on fee collection, 199 Distributions, in statistics, 56, 62-66
Computed tomography confirmation of, 66
for periodontal remodeling evaluation, 23-28 importance of, 66
for treatment planning normal, 56, 64-66, 81
in model preparation, 2-5 Documentation
in three-dimensional model construction, for board hearing, 241
17-22 of dentist-patient relationship termination,
Computers, in treatment planning, 2-5 189
Consent, informed. See Informed consent ofinformed consent, 192, 195-197
Consumer Credit Protection Act (Fair Debt Col- for lawsuits, 216-218, 229
lection Practices Act), 198-199 for practice transfer, 247-248
Continuous data, 63, 78, 82
Continuous measures, comparison of, sample Enumerative statistics, 102-103
size for, 73 Environmental factors, in vertical dimension de-
Core binding factor alpha 1, in osteoblast differ- velopment, 114, 126-127
entiation, 7 Epilnfo software, for sample size calculation, 72
Correlation, in statistics, 58-59 Error
Covenant not to complete, in practice transfer, 247 in causal inference, 108-109
Craniosynostosis syndrome, genetic factors in, in regression, 89-90
113-117 Ethical considerations
Cribs, tongue, for anterior open bite, stability of, in expert witnessing, 235
163 in fee collection, 204
Crouzon syndrome, phenotypic differences in, Expert witnesses
113-114 definition of, 234-235
Curve of Spee, severe, with deep bite and upper dishonest, 236
incisor flaring, Invisalign System for, liability of, 236
43-50 neutral, 236
Cytokines, in osteoclast differentiation, 7 on opposing sides, 236
<<    
     Article
      >> Home | TOC |          
Index
Subject Index, 255

Expert witnesses (Continued) Guide to Orthodontic Practice Valuation, 244, 251

reliability of, 235-236
role of, 190, 233 Headgear
standards for, 235 for anterior open bite, stability of, 163
types of, 235 bite plane with, for Class II malocclusion, ver-
Extracellular matrix, in strain detection, 8-9 tical treatment effects of, 141-148
Extractions evaluation of, 29-34
for anterior open bite in excessive vertical high-pull
dimension, 150-151 for anterior open bite in excessive vertical
for hyperdivergent open bite, 133-134 dimension, 149-152
for Class II malocclusion, vertical treatment
F test (ANOVA), 57-58, 83-84 effects of, 141-148
Face for hyperdivergent open bite, 133
ideal proportions of, 120-121 Hearings, before state licensing boards, 239,
vertical dimension of. See Vertical dimension 241-242
Fair Debt Collection Practices Act, 198-199 High-pull headgear. See under Headgear
Federal Rules of Evidence, on expert witnesses Human Genome Project, relevance of, to orth-
(Rule 702), 234-236 odontics, 13-16
Fee collection, 198-204 Hypodontia, genetic factors in, 14-15
collection agencies for, 203 Hypothesis
in dismissal vs. abandonment, 200 alternative, 71
ethical considerations in, 204 null, 56-57, 69-71
failure of, patient dismissal in, 203 in statistics, 56
Fair Debt Collection Practices Act, 198-199 testing of, sample size for, 68-71
protocol for, 200-203
responsible party for, 200-201 Independent t test, 83
statute of limitations for, 199 Individuals, statistics about, vs. groups, 105-106
suing for, 203 Infancy, toll for, in fee collection, 199
telephone numbers for, 201 Inferential statistics, 56-59, 78, 80-81, 103-104
three efforts for, 202-203 Informed consent
Financial issues, in practice transfer, 245-246, coercion in, 191
252 continuing nature of, 192
Fixation, of bilateral sagittal split osteotomy, disclosure for, 191-192
rigid vs. wire, skeletal and dental documentation of, 192, 195-197
changes in, 29-34 on fees, 198
Forces lack of, in lawsuit, 217
in headgear use, measurement of, 29-34 mental capacity for, 190
in hyperdivergent open bite, 132-133 for minors, 190-191
in tooth movement, bone cell reactions to, obtaining, 192-193, 195-196
8-11 vs. general consent, 190
Fos proteins, in strain translation, 9-11 Inheritance, of vertical dimension development,
113-117
Gaussian (normal) distribution, in statistics, 56, Integrins, in strain detection, 8-9
64-66, 81 Invisalign System, for deep bite, upper incisor
Generalizability, vs. sampling, 104-105 flaring, and severe curve of Spee, 43-50
Genetic factors Ion channels, activation of, in strain, 8
in orthodontics, 13-16
in vertical dimension development, 113-117 Jackson-Weiss syndrome, phenotypic differences
Gold standard test, 92 in, 113
Groups, statistics about, vs. individuals, 105-106
Growth hormone receptor gene, variants of, ver- Kappa statistic, 98
tical dimension and, 117 Kruskal-Wallis procedure, in statistics, 85
<<    
     Article
      >> Home | TOC |          
Index
256 Subject Index

Lawsuits Long-face problems, 173-183 (Continued)

anatomy of, 216-219 case report of, 182-183
answering allegations in, 217 prevalence of, 173-174
attorney assignment for, 217 treatment planning in
from defendant's side, 228-233 in adolescents with questionable growth po-
discovery in, 231-232 tential, 175-176
initial assessment in, 228-231 in adults with little or no growth potential,
interview in, 229 176
pretrial motion practice in, 232 orthodontic therapy for, 177-178
in trial, 232-233 postsurgical, 181-182
deposition for, 217, 231-232 presurgical, 178-179
discovery in, 231-232 presurgical, 178-181
documentation for, 216-218, 229 surgical approach in, 176-177
in excessive decay and demineralization, 212-
213 Malocclusion
expert witnesses in. See Expert witnesses backward rotator, in excess vertical dimen-
for fee collection, 203 sion, 127
initial assessment in, for defendant, 228-231 Class II, treatment of, vertical changes in, 141-
length of, 217-218 148
in periodontal disease development, 211-212 forward rotator, in deficient vertical dimen-
from plaintiff’s side, 220-227 sion, 127
pretrial period in, 232 open-bite, hyperdivergent, 130-140
reasons for, case studies of, 210-215 Malpractice
standards of care and, 218 statute of limitations for, 199-200
subpoenas in, 231 suits for. See Lawsuits
summons for, 216 Mandible
in treatment failure, 213-214 asymmetrical hyperplasia of, three-dimen-
trial procedure in, 218-219, 232-233 sional model of, 18-22
in vicarious liability, 205-209 bilateral sagittal split osteotomy of
Le Fort I osteotomy for anterior open bite, orthodontic therapy
for anterior open bite, orthodontic therapy with, 167-169
with, stability of, 166-169 fixation methods for, 29-34
for long-face problems, 178-179 condyle of. See also Temporomandibular joint
Legal concerns. See Risk management and legal growth of, vertical dimension and, 121-124
concerns morphology of, in hyperdivergent open bite,
Liability 130-131
administrative, 238-242 osteotomy of, in long-face problems, 176-180
in associate practices, 208 reverse-L osteotomy of, for anterior open bite,
in corporation, 208 orthodontic therapy with, stability of, 167
doubtful potential for, 230-231 Mann-Whitney U test, 84
for employees' negligence, 205-207 Masseter muscles
of expert witnesses, 236 activity of, in hyperdivergent open bite, 132
high potential for, 230 architecture of, 157
for independent contractors, 208 myosin composition of, 158
in limited liability association, 208 vertical dimension alteration effects on, 157—
in practice transfer, 245 160
in referrals, 207-208 Maxilla
vicarious, 205-209 morphology of, in hyperdivergent open bite,
Licensing, liability issues affecting, 238-242 130-131
Likelihood ratios, diagnostic, 98 osteotomy of
Long-face problems, 173-183 for anterior open bite, orthodontic therapy
appearance of, 174-175 with, 167-169
<<    
     Article
      >>
Subject Index
Maxilla (Continued)
for long-face problems, 176-180
protraction device for, for anterior open bite
in excessive vertical dimension, 151-
152
Mean, in statistics, 80
Median, in statistics, 56, 79, 82
Mental capacity, in informed consent, 190
Minors
fee collection and, 199, 201
informed consent from, 190-191
Mode, in statistics, 56
Models
skull, three-dimensional, for treatment plan-
ning, 17-22
for treatment planning, 2-5
Money, collection of. See Fee collection
Monitoring, of headgear use, 29-34
Mouth breathing, vertical dimension develop-
ment and, 126
MSX1 gene, mutations of, tooth agenesis in, 15
Negative predictive value, of test, 94-97
Negative test, definition of, 92
Negligence, liability for. See Liability
Nominal data, 62-63, 78-79, 82
Nonparametric statistics, 59-60, 80-82, 84-86
Normal distribution, in statistics, 56, 64-66, 81
n Query Advisor software, for sample size calcu-
lation, 72
Null hypothesis, in statistics, 56-57, 69-71
Occlusion, vertical dimension of, definition of,
155-156
Open bite
anterior
in excessive vertical dimension, 149-152
spontaneous improvement of, 164
treatment of, 162-172
combined orthodontic-surgical, 165-171
orthodontic, 162-164, 166
hyperdivergent, 130-140
etiology of, 131-133
extractions for, 133-134
high-pull headgear for, 133
morphology of, 130-131
posterior bite blocks for, 134-135
treatment of, 133-137
methods for, 133-135
results of, 137
timing of, 135-136
vertical chin cup for, 135
Home | TOC |          
Index
257
Ordinal data, 62, 79
Ortho Kinetics Corporation, headgear timer of,
29-34
Osteoblasts
description of, 6
differentiation of, 7
mechanical strain detection and translation
by, 8-11
Osteoclastogenesis inhibiting factor, in osteo-
clast differentiation, 7
Osteoclasts
description of, 6
differentiation of, 7
mechanical strain detection and translation
by, 8-11
Osteocytes, description of, 6
Osteopotegerin, in osteoclast differentiation, 7
Osteotomy
Le Fort I
for anterior open bite, orthodontic therapy
with, stability of, 166-169
for long-face problems, 178-179
mandibular
bilateral sagittal split
for anterior open bite, orthodontic ther-
apy with, 167-169
rigid vs. wire fixation of, skeletal and den-
tal changes in, 29-34
in long-face problems, 176-180
reverse-L, for anterior open bite, orthodon-
tic therapy with, stability of, 167
maxillary
for anterior open bite, orthodontic therapy
with, 167-169
for long-face problems, 176-180
Overbite, deep, in deficient vertical dimension,
152-154
P value, in statistics, 69-70, 82
Paired t test, 83
Parameters, in statistics, 55-56, 80-84
Pearson product-moment linear correlation, 58
Peer review committees, 238
Periodontal disease, lawsuit over, 211-212
Periodontic remodeling, three-dimensional eval-
uation of, 23-28
Pfeiffer syndrome, phenotypic differences in,
113
Plaintiff, in lawsuit, 220-227
Positive predictive value, of test, 94-97
Positive test, definition of, 92
<<    
     Article
      >>
258 Subject Index
Posterior bite blocks, for hyperdivergent open
bite, 134-135
Potassium ion channels, activation of, in strain, 8
Power
in hypothesis testing, 70
post hoc analysis of, 75-76
to reject false null hypothesis, 57
Practice management, risk management in. See
Risk management and legal concerns
Predictive orthodontics, 2-5
Price, of orthodontic practice, for transitioning,
244
Probability level, in statistics, 69-70, 82
Proportions, comparison of, sample size for,
72-73
Protraction device, maxillary, for anterior open
bite in excessive vertical dimension,
151-152
Qualitative data, distribution of, 62
Quantitative data, distribution of, 62
Range, in statistics, 80
Rank, in nonparametric statistics, 81-82
RANKL (receptor activator of nuclear factor-KB
ligand), in osteoclast differentiation, 7
Raw data, distribution of, 63
Receiver operating characteristic curves, 98-100
Receptor activator of nuclear factor-KB ligand
(RANKL), in osteoclast differentiation, 7
Records. See Documentation
Referrals, liability in, 207-208
Regression, in statistics, 59, 87-91
Regulations, governmental, 238-242
Relationships
dentist-patient. See Dentist-patient relation-
ship
in statistics
form of, 89-91
significance of, 88-89
Remodeling, periodontic, three-dimensional
evaluation of, 23-28
Representations, in practice transfer, 247
Respondeat superior doctrine, 205-209
Rest, vertical dimension at, definition of, 155-
156
Reverse-L osteotomy, of mandible, for anterior
open bite, orthodontic therapy with,
stability of, 167
Rigid fixation, of bilateral sagittal split osteot-
omy, skeletal and dental changes in,
29-34
Home | TOC |          
Index
Risk management and legal concerns
dentist-patient relationship and, 187-197
dismissal of patient, 188-189, 200, 203
fee collection, 198-204
governmental regulations, 238-242
informed consent. See Informed consent
lawsuits
defendant's view of, 228-223
expert witness role in. See Expert witnesses
for fee collection, 203
plaintiff’s view of, 220-233
procedure for, 216-219
reasons for, 210-215
liability, 205-209
administrative, 238-242
standard of care. See Standard of care
transitioning of practice
buyer's perspective of, 249-252
seller's perspective of, 243-248
Sagittal split osteotomy, bilateral
for anterior open bite, orthodontic therapy
with, 167-169
rigid vs. wire fixation of, skeletal and dental
changes in, 29-34
Samples and sampling, in statistics, 55-56
bivariate, 89
size of, 67-76
adjustments to, 75
calculation of, 67-68, 72-75
clinical relevance of, 71
for comparing continuous measures, 73
for comparing proportions, 72-73
for comparing time-to-event measure, 73-74
hypothesis testing and, 68-71
post hoc analysis of, 75-76
variability and, 71-72
vs. generalizability, 104-105
Scatter, in statistics, 56
Schuchardt technique, for anterior open bite,
orthodontic therapy with, 167-168
Screening tests, usefulness of, 92
Selling, of orthodontic practice, 243-248
Sensitivity and specificity, in statistics
accuracy and, 97-98
agreement and, 97-98
diagnostic likelihood ratios and, 98
receiver operating characteristic curves and,
98-100
terminology of, 92-94
test characteristics and, 94-97
<<    
     Article
      >> Home | TOC |          
Index

Subject Index 259

Skull, models of, three-dimensional, for treat- Teeth (Continued)

ment planning, 17-22 movement of. See Tooth movement
Specificity, in statistics. See Sensitivity and specificity Temporomandibular joint, vertical dimension
Standard deviation, in statistics, 56, 64-65 alteration effects on, 157-160
Standard of care, 187-190 Termination, of dentist-patient relationship,
expert witness and, 234 188-189, 200, 203
guidelines for, 187-188 Tests, sensitivity and specificity of, 92-101
in lawsuits, 218 Three-dimensional imaging
local community aspects of, 189-190 for periodontal remodeling evaluation, 23-28
State licensing boards, 238-242 for treatment planning, 17-22
Statistical significance, 58-59, 106-108 Thumb sucking, hyperdivergent open bite in,
Statistics 132
about groups vs. individuals, 105-106 Time-to-event measure, sample size for, 73-74
about individuals vs. groups, 105-106 Tongue cribs, for anterior open bite, stability of,
clinical applications of, 102-109 163
descriptive, 55-56, 77-78 Tongue posture, vertical dimension develop-
distributions in, 56, 62-66 ment and, 127
enumerative, 102-103 Tooth movement, 6-12
inferential, 56-59, 78, 80-81, 103-104 after bilateral sagittal split osteotomy, fixation
nonparametric, 59-60, 80-82, 84-86 methods affecting, 29-34
overview of, 54-61 bone cell life cycle and, 7
power in, 57, 70, 75-76 in Invisalign System, 43-50
regression in. See Regression mechanical strain detection and translation
sample size in, 67-76 in, 8-11
sensitivity and specificity in, 92-101 TRANCE (TNF-related activation induced cyto-
tests for, 77-86 kine), in osteoclast differentiation, 7
uses of, 77-78 Transitioning of orthodontic practice
Statute of limitations, for fee collection, 199-200 buyers' concerns and perspectives in
Stereolithography, for three-dimensional model associateship, 250-251
construction, for treatment planning, control of process, 243-244
17-22 decision basis, 249-250
Stock, sale of, in practice transition, 244-245 finances, 252
Strains information needs, 251-252
detection of, 8-9 process for, 250-251
translation of, 9-11 questions to ask, 250
Student t tests, 56-57, 82-83 seller's concerns and perspectives in, 243-248
Subpoenas advisors, 246
for medical records, 231 articles not included in sale, 245
from state licensing boards, 239 associateship, 246-247
Surgery best process, 246-247
for anterior open bite, with orthodontic ther- continuity of care, 243, 247-248
apy, 165-171 covenant not to complete, 247
for long-face problems. See Long-face problems dispute resolution, 247
simulated, on three-dimensional models, 17-22 documentation transfer, 247-248
Swallowing, vertical dimension development finances, 245-246
and, 127 personnel involved in, 243-244
representations, 247
t tests, 56-57, 82-83 taxes, 245
Tax considerations, in practice transfer, 245 terms, 246
Teeth transaction form, 244-245
agenesis of, genetic factors in, 14-15 valuation, 244
development of, genetic factors in, 13-16 warranties, 247
<<    
     Article
      >> Home | TOC |          
Index

260 Subject Index

Treatment planning Vertical dimension (Continued)

computer-aided, 2-5 genetic factors in, 113-117
with Invisalign System, 43-50 diagnosis of, 120-129
statistics for. See Statistics backward rotator, 127
three-dimensional cephalometry and skull condylar growth in, 121-124
models for, 17-22 dental relationships in, 121
Trial, procedure for, 218-219, 232-233 facial height in, 124-127
Tumor necrosis factor-related activation induced forward rotator, 127
cytokine (TRANCE), in osteoclast differ- skeletal relationships in, 121
entiation, 7 excessive. See also Long-face problems
anterior open bite in, 149-152
University of Florida headgear monitor, 29-34 condylar growth in, 122
University of Washington headgear monitor, treatment of, 127
29-34 of occlusion, definition of, 155-156
posterior facial height growth and, 124-127
Valuation, of orthodontic practice, for transi- at rest, definition of, 155-156
tioning, 244, 251 treatment of, 128
Variability, in statistics, 71-72, ;102 biomechanical considerations in, 149-154
Vertical chin cup, for hyperdivergent open bite, Class II malocclusions, 141-148
135 hyperdivergent open-bite malocclusions,
Vertical dimension 130-140
allergy and, 116, 126 long-term stability of, 162-172
alteration of masticatory muscle effects of, 157-159
masseter muscle changes in, 157-160 surgical, 173-183
TMJ effects of, 157-160 TMJ effects of, 157-160
anterior facial height growth and, 124-127 Vicarious liability, 205-209
deficient
condylar growth in, 122 Warranties, in practice transfer, 247
deep overbite in, 152-154 Wilcoxon signed rank test, 84-85
treatment of, 127 Wire fixation, of bilateral sagittal split osteot-
definition of, 149, 155-156 omy, skeletal and dental changes in,
development of, 113-119 29-34
environmental factors in, 114, 126-127 Witness, expert. See Expert witnesses
 <<    
     Article
      >> Home | TOC |          
Index
United States Postal Servic
Statement of Ownership, Management, and Circulation
1. Publication Title 2. Publication Number 3. Filing Date

Seminars in Orthodontics 1 0 7 3 1 - 8 7 4 6 9/15/02

4. Issue Frequency 5 . Number of Issues Published Annually 6. Annual Subscription Price

Mar, Jun, Sep, Dec 4 $133.00

e,andZIP+4) Contact Person
Elsevier Science Amy Snyder
6277 Sea Harbor Drive, Orlando, FL 32887-4800, Orange County Telephone

215-238-8319
8. Complete Mailing Address of Headquarters or General Business Office of Publisher (Not printer)

Elsevier Science, 1 1830 Westline Industrial Drive, St. Louis, MO 63146-3318_

9. Full Names and Complete Mailing Addresses of Publisher. Editor, and Managing Editor (Do not leave blank) _ , W.B.SAUNDERS RflMosby
Publisher (Name and complete mailing address)

Jane Ryley, Elsevier Science, 1 1830 Westline Industrial Drive, St. Louis, MO 63146-3318 _
Editor (Name and complete mailing address)

P. Lionel Sadowsky, DMD, BDS, DipOrth, MDent, Professor and Chairman, Department of Orthodontics, University of Alabama, 1919 • Essential Information for Today's Dentists
Seventh Avenue South, Birmingham, AL 35294
Managing Editor (Name and complete mailing address)
_ from the Leading Health Care Publisher
Michael Miller, Elsevier Science, The Curtis Center, Independence Square West, Philadelphia, PA 19106-3399
10. Owner (Do not leave blank. If the publication is owned by a corporation, give the name and address of the corporation immediately followed by the
names and addresses of all stockholders owning or holding 1 percent or more of the total amount of stock. If not owned by a corporation, give the
names and addresses of the individual owners. If owned by a partnership or other unincorporated firm, give its name and address as well as those of
each individual owner. If the publication is published by a nonprofit organization, give its name and address.)
Complete Mailing Address
American Journal of Orthodontics and
Dentofacial Orthopedics
Wholly owned subsidiary of 4520 East-West Highway
The Official Publication of the American
Reed/Elsevier, US holdings Bethesda, MD 20814
Association of Orthodontists, its constituent
societies, and the American Board
of Orthodontics
11. Known Bondholders, Mortgagees, and Other Security Holders Owning or
Holding 1 Percent or More of Total Amount of Bonds, Mortgages, or
Other Securities. If none, checkbox ~ British Journal of Oral £c Maxillofacial Surgery
Complete Mailing Address
Official Journal of the British Association of
Oral & Maxillofacial Surgeons and a
Recognized Journal of the American College of
Oral and Maxillofacial Surgeons
12. Tax Status (For completion by nonprofit organizations authorized to ma& at nonprofit rates) (Check one)
The purpose, function, and nonprofit status of this organization and the exempt status for federal income tax purposes:
Has Not Changed During Preceding 12 Months Dental Abstracts
Has Changed During Preceding 12 Months (Publisher must submit explanation of change with this statement)
(See Instructions on Reverse)
PS Form 3526, October 1999 Dental Clinics

International Journal of Oral & Maxillofacial

Surgery
Official Publication of the International
Association of Oral & Maxillofacial Surgeons

The Journal of Evidence-Based

3. Publication Title
Seminars in Orthodontics
14. Issue Date for Circulation Data Below
June 2002
Dental Practice
15. Average No. Copies Each Issue No. Copies of Single Issue
Extent and Nature of Circulation During Preceding 12 Months Published Nearest to Filing Date

2600 2600
Journal of Oral and Maxillofacial Surgery
a. Total Number of Copies (Net press run)
Paid/Requested Outside-County Mail Subscriptions Stated on 1382 1232
(1) Form 3541. (Include advertiser 's proof and exchange copies)
Journal of Prosthodontics
Requested
Circulation
(2) (Include advertiser 's proof and exchange copies)
Official Journal of the American Association of
516 488
(3) Counter Sales, and Other Non-USPS Paid Distribution Oral and Maxillofacial Surgeons
(4) Other Classes Mailed Through the USPS
c. Total Paid and/or Requested Circulation
[Sum oflSb. (I), (2), (3), ana (4)]
^
^
1898 1720 The Journal of Prosthetic Dentistry
d. Free 130 131
Distribution (1) Outside-County as Stated on Form 3541
by Mail
(Samples, Journal of Prosthodontics
compliment- (2) In-County as Stated on Form 3541
ary, and
otherfne)
Official Journal of the American College
(3) Other Classes Mailed Through the USPS
e. Free Distribution Outside the Mail
of Prosthodontists
(Carriers or other means)

130 131
f. TWI^K-fl-taea.*/*.-/*) ^ Oral and Maxillofacial Surgery Clinics
2028 1851
g. Total Distribution (Sum oflSc. and 15f.) *•
h. Copies not Distributed 572 749 Atlas of the Oral and Maxillofacial
i. Total (Sum of 15g. and h.) 2600 2600 Surgery Clinics
j. Percent Paid and/or Requested Circulation 94% 93%
(J5c. divided by I5g. times 100)
Publication of Statement of
Publication required. Will be printed in the December 2002 issue of thi Publication not required
Oral Surgery, Oral Medicine, Oral Pathology,
of Editor, Publisher, Business Manager, or Owner
Oral Radiology and Endodontics
I COSTS' that all information furnished on this form is true and complete. I understand that anyone who furnishes false or misleading information on this form
or Who omits material or information requested on the form may be subject to criminal sanctions (including fines and imprisonment) and/or civil sanctions Seminars in Orthodontics
(including civil penalties).

Year Book of Dentistry®

For more information, contact Periodicals Marketing at

(215)238-5614.
<<    
     Article
      >> Home | TOC |          
Index

W.B. SAUNDERS

«NÅLS
ANESTHESIOLOGY NEPHROLOGY PSYCHIATRY
Journal of Cardiothoradc and Vascular Anesthesia Advances in Renal Replacement Therapy—AJournal of Comprehensive Psychiatry—Official Journal of tbe American
The National Kidney Foundation PsycbofMtbological Association
Regional Anesthesia and Pain Medicine American Journal of Kidney Diseases—Tbe Official Journal Seminars in Clinical Neuropsychiatry
Official Journal of the American, Asian and Oceanic, of The National Kidney Foundation
and Latin American Societies of Regional Anesthesia Journal of Renal Nutrition—The Officialjournal of The Council RADIOLOGY
Seminars in Anesthesia on Renal Nutrition of the National tSdney Foundation
Seminars in CardJothoracic and Vascular Anesthesia Seminars in Nephrology Seminars in Nuclear Medicine
Seminars in Pain Medicine Seminars in Radiologie Technology
Techniques in Regional Anesthesia and Pain Management NEUROLOGY Seminars in Roen tgenology
Journal of Pain—Official Journal of the American Pain Society Seminars in Ultrasound, CT and MRI
CARDIOVASCULAR DISEASES Journal of Stroke and Cerebrovascular Diseases—Officialjournal Techniques in Vascular and Interven tional Radiology
Progress in Cardiovascular Diseases oftbe National Stroke Association and theJapan Stroke Society
Techniques in Intervention^ Cardiology Seminars in Cerebrovascular Diseases and Stroke RHEUMATOLOGY
Seminars in Pediatric Neurology Seminars in Arthritis and Rheumatism
CRITICAL CARE MEDICINE
Journal of Critical Care NURSING SURGERY
Advances in Neonatal Care—Officialjournal of the National Journal of Pediatric Surgery—Officialjournal of tbe Section on
DENTISTRY Surgery of tbe American Academy of Pediatrics, British Association
Journal of Oral and Maxülofeäal Surgery—Qflfcwf Joumalof ofPaediatric Surgeons, American Pediatric Surgical Association,
Applied Nursing Research Canadian Association ofPaediatric Surgeons and Pacific
the American Association of Ord and Maxitbfacial Surgeons Archives of Psychiatric Hur$w%—official Journal of tbe SKRPH
Journal of Prosthodontics—Officialjournal of The American Association of Pediatric Surgeons
Division, International Society of Psychiatric -Mental Health Nurses Operative Techniques in General Surgery
College of Prosthodontists Journal of Pediatric Nursing—Official Journal of tbe Society of
Seminars in Orthodontics Operative Techniques in Neurosurgery
Pediatric Nurses Operative Techniques in Plastic and Reconstructive Surgery
DERMATOLOGY Journal of Pediatric Oncology Nursing—Officialjournal of the Operative Techniques in Thoracic and Cardiovascular Surgery-^4»
American Journal of Contact Dermatitis—The Officialjournal Official Publication ofTbeAmericanAssociationfor Thoracic Surgery
of the American Contact Dermatitis Society Journal of PeriAnesthesia Nursing—Officialjournal of tbe American Seminars in Colon and Rectal Surgery
Seminars in Cutaneous Medicine and Surgery Society ofPeriAnestbesia Nurses Seminars in Laparoscopic Surgery
Journal of Professional Nursing—Official Journal of tbe American Seminars in Pediatric Surgery
EMERGENCY MEDICINE Association of Colleges of Nursing Seminars in Spine Surgery
American Journal of Emergency Medicine Pain Management Nursing—Officialjournal of tbe American Society Seminars in Thoracic and Cardiovascular Surgery-^to Official
Clinical Pediatric Emergency Medicine of Pain Management Nurses Publication ofTbeAmericanAssociationfor Thoracic Surgery
PeriAnesthesia and Ambulatory Surgery Nursing update—Official Seminars in Thoracic and Cardiovascular Surgery: Pediatric Cardiac
ENDOCRINOLOGY Publication of tbe American Society ofPeriAnestbesia Nurses Surgery Annual—«In Official Publication of Tbe American
Metabolism—Clinical and Experimental Seminars for Nurse Managers Associationfor Thoracic Surgery
Seminars hi Oncology Nursing Seminars in Urologie Oncology
GASTOOEM-EROLOGY/HEPATOLOGY Seminars in Vascular Surgery
Clinical Perspectives in Gastroenterology—The Official OBSTETRICS AND GYNECOLOGY
Clinical PracticeJournal of the American Gastroenterological Clinical Journal of Women's Health TRANSPLANTATION
Liver Transplantation-^!» Official Publication of Ore American
Gastroenterology— Official Journal of the American ORTHOPEDICS
Gastmenterological Association Arthroscopy: The Journal of Arthroscopic and Related Surgery— Liver Transplantation Society
Hepatology—OfficiäJournaloftbeAmericanAssociationfor Official Publication of tbe Artbroseopy Association of North America Transplantation Reviews
Ute Study of Liver Diseases and tbe International Society of Artbroseopy, Knee Surgery, and
Liver Transplantation-^» Official Publication of the American Orthopaedic Sports Medicine UROLOGY
Associationfor the Study of Liver Diseases and The International Journal of the American Society for Surgery of the Hand— Seminars in Urologie Oncology
Liver Transplantation Society An Official Journal of the American Society for Surgery
Seminars in Gastrointestinal Disease of tbe Hand VETERINARY MEDICINE
Techniques in Gastrointestinal Endoscopy Journal of Hand Surgery-^» Official Journal of tbe American Advances in Small Animal Medicine and Surgery
Society for Surgery of tbe Hand Clinical Techniques in Equine Practice
HEMATOLOGY/ONCOLOGY Operative Techniques in Orthopaedics Clinical Techniques in Small Animal Practice
Seminars in Hematology Operative Techniques in Sports Medicine Seminars in Avian and Exotic Pet Medicine
Seminars in Oncology Seminars in Arthroplasty Veterinary Surgery—Tbe Officialjournal of Tbe American
Seminars in Radiation Oncology Seminars in Spine Surgery College of Veterinary Surgeons, Inc. and Tbe European College
Transfusion Medicine Reviews of Veterinary Surgeons
OTORHINOIARYNGOLOGY
INFECTIOUS DISEASES American Journal of Otolaryngology
Seminars in Infection Control Operative Techniques in Otolaryngology—Head and Neck Surgery please contact:
Seminars in Pediatric Infectious Diseases
Seminars in Respiratory Infections PATHOLOGY Periodicals Marketing
MEDICAL TRANSCRIPTION
Annals of Diagnostic Pathology WJB. SAUNDERS
Human Pathology A Division of Elsevier Science
The Latest Word Seminars in Diagnostic Pathology
The Curtis Center, Independence Square West
NEONATAL/PERINATAL MEDICINE PHYSICAL MEDICINE Philadelphia, PA 19106-3399
Newborn and Infant Nursing Reviews Archives of Physical Medicine and Rehabilitation—
Seminars in Perinatology Officialjournal of the American Congress of Rehabilitation
Phone (215) 238-5614
Medicine and tbe American Academy of Physical Medicine Or visit our homepage at:
©2002 EISEVIER Soera. www.wbsaunders.com