Documente Academic
Documente Profesional
Documente Cultură
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.
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 The appearance of the code at the bottom of the first page
quarterly by W.B. Saunders. Months of issue are March, June, of an article in this journal indicates the copyright owner's
September, and December. Corporate and Editorial Offices: consent that copies of the article may be made for personal or
The Curtis Center, Independence Square West, Philadelphia, internal use, or for the personal or internal use of specific clients,
PA 19106-3399. Accounting and Circulation Offices: 6277 Sea for those registered with the Copyright Clearance Center, Inc.
Harbor Drive, Orlando, FL 32887-4800. POSTMASTER: Send (222 Rosewood Drive, Danvers, MA 01923; (508) 750-8400;
change of address to: Seminars in Orthodontics, W.B. Saunders, www.copyright.com). This consent is given on the condition that
Customer Service, 6277 Sea Harbor Drive, Orlando, FL 32887- the copier pay the stated per-copy fee for that article through the
4800. 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
Editorial correspondence should be addressed to: as copying for general distribution, for advertising or promotional
P. Lionel Sadowsky, DMD, BDS, DipOrth, MDent, Editor, purposes, for creating new collective works, or for resale.
Seminars in Orthodontics, Professor and Chairman, Department Absence of the code indicates that the material may not be
of Orthodontics, University of Alabama, 1919 Seventh Avenue processed through the Copyright Clearance Center, Inc.
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. Reprint inquiries should be addressed to Ginny
Saunders, Periodicals Department, P.O. Box 628239, Orlando, Nicholls, Elsevier Science, The Curtis Center, Independence
FL 32862-8239 or e-mail hhspcs@harcourt.com. Square West, Philadelphia, PA 19106-3399. Telephone (215)
Change of address notices, including both the old and new 238-5534, fax (215) 238-6423; e-mail: gnicholls@elsevier.com.
addresses of the subscriber and the mailing label, should be
sent at least 1 month in advance. Customer Service: 1-800-654-
2452 Advertising representative: MJ. Mrvica Associates, Inc,
2 West Taunton Ave, Berlin, NJ 08009. Telephone (609) 768-
9360. Fax (609) 753-0064.
Yearly subscription rates: United States and possessions: Publication of an advertisement in Seminars in Orthodontics
individual, $133.00; institution, $164.00; student and resident, does not imply endorsement of its claims by the Editor(s) or
$67.00; single issue, $50.00. All other countries: individual Publisher of the journal.
$166.00; institution, $198.00; student and resident, $83.00; The contributors have checked generic and trade names and
single issue, $50.00. For all areas outside the United States and verified drug doses for accuracy according to the standards
possessions, there is no additional charge for surface delivery. accepted at the time of publication. The ultimate
For air mail delivery, add $16.00. To receive student/resident
responsibility, however, lies with the prescribing physician.
rate, orders must be accompanied by name of affiliated
institution, date of term, and the signature of program/residency Please convey any errors to the Editor.
coordinator on institution letterhead. Orders will be billed at
individual rate until proof of status is received. The ideas and opinions expressed in Seminars in
Prices are subject to change without notice. Current prices Orthodontics do not necessarily reflect those of the Editor or the
are in effect for back volumes and back issues. Single issues, Publisher. Publication of an advertisement or other product
both current and back, exist in limited quantities and are mention in Seminars in Orthodontics should not be construed as
offered for sale subject to availability. Back issues sold in an endorsement of the product or the manufacturer's claims.
conjunction with a subscription are on a prorated basis. Checks Readers are encouraged to contact the manufacturer with any
should be made payable to W.B. Saunders and sent to Seminars questions about the features or limitations of the products
in Orthodontics, W.B. Saunders, Periodicals Department, 6277 mentioned. Neither the Editor or Publisher assume any
Sea Harbor Drive, Orlando, FL 32887-4800. responsibility for any injury and/or damage to persons or
property arising out of or related to any use of the material
Copyright 2002, Elsevier Science (USA). All rights contained in this periodical. The reader is advised to check the
reserved. No part of this publication may be reproduced or appropriate medical literature and the product information
transmitted in any form or by any means, electronic or currently provided by the manufacturer of each drug to be
mechanical, including photocopy, recording, or any information administered to verify the dosage, the method and duration of
storage and retrieval system, without permission in writing administration or contraindications. It is the responsibility of
from the Publisher. Printed in the United States of America. 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.
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, Seminars in Orthodontics is indexed in the Cumulative
6277 Sea Harbor Drive, Orlando, FL 32887-4800. Telephone Index to Nursing and Allied Health Literature® print index
number: 1-407-345-2500. and the Cinahl® database.
W.B. SAUNDERS
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%!
VH Mosby
• l r B ADivision
Divisionof
ofElsevier
E Science
© 2002 Mosby
<<
Article
>> Home | TOC |
Index
Seminars in Orthodontics
EDITOR
P. Lionel Sadowsky, DMD, BDS, DipOrth, MDent
EDITORIAL BOARD
Richard G. Alexander, Arlington, TX James A. McNamara, Jr, Ann Arbor, MI
Rolf G. Behrents, Memphis, TN Robert N. Moore, Grand Island, NE
Samir E. Bishara, Iowa City, IA Ravindra Nanda, Farmington, CT
Robert Boyd, DBS, San Francisco, CA Perry M. Opin, Milford, CT
Larry M. Bramble, Cypress, CA Sheldon Peck, Newton, MA
John S. Casko, Iowa City, IA William R. Proffit, Chapel Hill, NC
Harry L. Dougherty, Van Nuys, CA Cyril Sadowsky, Chicago, IL
T.M. Graber, Evanston, IL David M. Sarver, Birmingham, AL
Robert J. Isaacson, Richmond, VA T. Michael Speidel, Minneapolis, MN
Alexander Jacobson, Birmingham, AL William J. Thompson, Bradenton, EL
Lysle E. Johnston, Jr., Ann Arbor, MI James L. Vaden, Cookeville, TN
Gregory J. King, Seattle, WA Robert L. Vanarsdall, Jr., Philadelphia, PA
Vincent G. Kokich, Tacoma, WA Katherine Vig, Columbus, OH
Steven J. Lindauer, Richmond, VA C.B. Preston, Buffalo, NY
INTERNATIONAL
Zeev Abraham, Herzliya, Israel Shinkichi Namura, Tokyo, Japan
W.G. Evans, Johannesburg, South Africa George Skinazi, Paris, France
Roberto Justus, Mexico City, Mexico 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%!
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
Introduction
Gregory J. King
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
Recent Issues
Seminars in Orthodontics
VOL 8, NO 1 MARCH 2002
Introduction
Copyright 2002, Elsevier Science (USA). All rights reserved, Gregory J. King, DMD, DMSc
doi: 10.1053/sodo. 2002.28185 Guest Editor
Predictive Orthodontics
Treatment Alternatives
Visualization
3D skeletal imaging
3D facial imaging
3D dental imaging
Mandibular motion
Facial animation
Functional information
Other physiologic data
James Mäh
patient. For example, accurate data on bite of providing a clear and realistic preoperative
forces, tongue posture and swallowing, para- informed consent.3 Interim and posttreatment
function, and respiration are extremely useful in clinical results are analyzed by comparison to
orthodontics. Certainly, there will be continuing the predicted result and to the database of past
efforts to improve on existing imaging technol- clinical experiences. As more information be-
ogies and new diagnostic modalities will become comes available with more patients, a feedback
available. As more and better patient informa- loop will exist to improve and refine boundary
tion becomes available, the computer model of conditions. In this way, the system will become
the patient will also be improved. more accurate and more predictive. Because
The computerized images and data are pro- past experiences are stored in a database, this
cessed using refinement schemes. These pieces information can be shared with other clinicians
of data are related to each other through regis- as well as provide guidelines for orthodontists in
tration, integration, and segmentation. This is training. In addition to analysis, the computer
done to create independent objects, like individ- model is sufficiently accurate to be used for
ual bones or teeth. Geometric model construc- implementing therapy. By using computer-as-
tion is carried out to incorporate the dentition sisted design and computer-assisted manufactur-
of the patient, craniofacial skeleton, and the soft ing methods and robotics, any number of ther-
tissues of the face. In addition, 3-D coordinates apeutic appliances, such as surgical splints,
on mandibular position and movement as well as retainers, surgical guides for implants, templates
facial animation may be incorporated into such and arch wires,4 could be produced. Additional
a model. Other available functional data may information from the model also can be used to
also be used. At this stage, all available patient create a 3-D scaffold for procedures involving
information is in a completely integrated model, tissue engineering.5 Computer-assisted therapy
which may then be used for diagnosis and treat- would offer improved precision, reliability, and
ment planning. Also at this stage, boundary con- more predictable results overall.
ditions from a database may be applied to help
select possible treatment alternatives. An exam-
Summary
ple of a boundary condition could be the move-
ment of the maxillary central incisor. Beyond Predictive orthodontics is made possible by new
the boundary, growth modification or surgery technologies that allow orthodontists to better vi-
would be recommended. Other examples of sualize the patient. These include methods to im-
boundaries that could be helpful to the clinician age the craniofacial skeleton by using a newly in-
include lack of further growth or inadequate troduced class of computed tomography scanners
anchorage. These and the many boundary con- optimized for the head and neck or by methods of
ditions that pertain to orthodontics would be photogrammetry by which 3-D information is pro-
included in the database. duced from planar radiographs. Three-dimen-
The model can also be used for analysis, hy- sional images of the dentition can be produced
pothesis testing and treatment simulations of directly with an intraoral 3-D camera or indirectly
any number of treatment scenarios involving var- by converting an impression to a computer model.
ious tissues.2 The model and treatment ap- Images of the face can be produced using a num-
proaches can be analyzed by finite element ber of commercially available 3-D cameras. Addi-
methods and solved. Treatment alternatives may tionally, it is now possible to track mandibular
be simulated and assessed under a variety of position and facial movements by using ultrasonic
functional states. Esthetics, occlusion, and other or video-based methods.
outcomes may be assessed. The results are then Continued improvement in the quality of orth-
extracted and stored in a database. In addition, odontic services will require implementation of
materials such as arch wires, adhesives, appli- new diagnostic and treatment approaches. New
ances, or implants can be tested on such a technologies will provide improved precision, ac-
model. From here treatment options would be curacy, and perspective. This will provide the cli-
presented to the patient and clinical treatment nician with the means to optimize the way treat-
provided. Accurately predicting treatment out- ment is planned through computer modeling,
come has been shown to be an important aspect testing, and simulations. In this way, treatment
<<
Article
>> Home | TOC |
Index
Predictive Orthodontics
results can be predicted with greater accuracy, 2. Keeve E, Girod S, Kikinis R, et al. Deformable modeling
thereby reducing unfavorable outcomes and in- of facial tissue for craniofacial surgery simulation. Comp
Aid Surg 1998;3:228-238.
creasing efficiency. These approaches also can
3. Rai AK. Reflective choice in health care: Using informa-
serve as very powerful tools in orthodontic educa- tion technology to present allocation options. Am J Law
tion and patient communication. Med 1999;25:387-402.
4. Mah J, Sachdeva R. Computer-assisted orthodontic treat-
ment - the SureSmile_ process. Am J Orthod Dentofacial
References Orthop 2001;117:622-623.
1. Mah J, Bumann A. Technology to create the 3-dimen- 5. Schultz O, Sittinger M, Haeupl T, et al. Emerging strategies
sional patient record. Semin Orthod 2001;7:199-208. of bone and joint repair. Arthritis Res. 2000;2:433-436.
<<
Article
>> Home | TOC |
Index
The Life Cycle of Bone Cells entiation, whereas transgenic mice that lack the
OPG gene have severe osteoporosis caused by an
Differentiation of Osteoblasts
increased number of functional osteoclasts.10
Mesenchymal stern cells differentiate into osteo- OPG was shown to be as effective as bisphospho-
blasts when they are exposed to bone morpho- nates in protecting against bone loss in ovariec-
genic proteins (BMP). 5 BMPs are part of a large tomized estrogen-deficient rats.10 When injected
multigene family, the transforming growth fac- subcutaneously in postmenopausal women,
tor ß (TGFß) superfamily. Our knowledge of the OPG resulted in a decrease in the bone turnover
process involved in the development of an osteo- marker N-telopeptide.12 In a recently published
blast is very limited. A recent advancement has study, Kobayashi and colleagues localized the
been the identification of the core binding fac- OPG in alveolar bone surfaces during OTM.13
tor alpha 1 (Cbfal} gene during osteoblast de- They showed that in tension areas, as the bone
velopment and function. Cbfal is a transcription shifts from resorption to formation, there was a
factor that is expressed by cells of the osteoblas- simultaneous increase in TGF-ß and OPG mes-
tic lineage and is necessary for osteoblast differ- senger RNA (mRNA) and the disappearance of
entiation.6 Studies have shown the lack of osteo- osteoclasts through apoptosis.
blasts and consequently bone formation in mice Because OPG is a soluble molecule, it was
deficient in the Cbfal protein. Cleidocranial dys- predicted that its ligand would be a membrane-
plasia (CCD) is caused by mutations of the Cbfal bound molecule. Using molecular biology tech-
gene.7 It appears that Cbfal may also play a role niques, Suda and colleagues isolated a molecule
in osteoblast cell function because mice express- that they called osteoclast differentiation factor
ing a dominant-negative version of the gene (ODF).14 ODF was able to induce osteoclasto-
show decreased expression of genes that charac- genesis in bone organ cultures. Lacey et al also
terize the osteoblast phenotype (eg, collagen identified a molecule termed OPG-L that dem-
genes, osteopontin, and osteocalcin) .8
onstrated similar activities as ODF.15 When ODF
and OPG-L were cloned, it was revealed that the
Differentiation of Osteoclasts molecules were identical to an already cloned
member of the TNF ligand family, called
Osteoclasts differentiate from hematopoetic
cells. This pathway of differentiation is also TRANCE (TNF-related activation induced cyto-
shared by macrophages. It is not surprising then kine) or RANKL (receptor activator of NF-kB
that granulocyte-macrophage colony-stimulating ligand). RANKL or OPG-L binds to the receptor
factor (GM-CSF) and macrophage colony-stimu- activator of NF-kB ligand (RANK), which is
lating factor (M-CSF) are important in regulat- found on the cell membranes of osteoclasts. This
ing these shared stages of development.4 The binding can occur in two ways: one is cell bound,
recent discovery of osteopotegerin (OPG) and which requires cell-to-cell contact; the other is a
its binding ligand (OPG-L) as essential regula- soluble form. In this situation, OPG can act as
tors of osteoclasts formation and activity, how- decoy receptor, blocking the interaction be-
ever, has created a new paradigm in osteoclast tween RANK and OPG-L.16
biology.9 The ratio of OPG/OPG-L regulates the oste-
The molecule that inhibits osteoclastogenesis oclast's lifecycle from differentiation to fusion of
is known by two different names, OPG (osteopo- preocteoclasts to function to apoptosis. The hy-
tegerin) and OCIF (osteoclastogenesis inhibit- pothesis is that, at steady state, there is a bal-
ing factor), because it was identified simulta- anced level in the OPG/OPG-L ratio. Cytokines
neously by two different laboratories working on (TNF, interleukin-1 [IL-1], prostoglandin E2
different projects.10' n This molecule shares se- [PGE2] and growth factors (TGF-B, BMP) are
quence homology with the tumor necrosis factor upstream signals which regulate the OPG/
(TNF) receptor superfamily. OPG is secreted by OPG-L ratio. OPG/OPG-L serve as downstream
osteoblasts and functions to block the formation regulators of osteoclastogenesis. Thus, when the
of osteoclasts as well as bone resorption. Trans- balance favors OPG, there are fewer active oste-
genie mice overexpressing OPG develop osteo- oclasts; when the balance favors OPG-L, there is
petrosis caused by the lack of osteoclast differ- an increased number of active osteoclasts.
<<
Article
>> Home | TOC |
Index
How Bone Cells Detect Mechanical strain to osteoblasts has been shown to acti-
Strain vate stretch-activated calcium channels thereby
producing a large increase in intracellular cal-
After an orthodontic force is applied, the initial
cium.19 The mechanism by which a mechanical
step is the detection of a mechanical strain. The
stimulus modulates stretch-activated ion chan-
cells responsible for sensing mechanical strains
nel gating remains unknown. It is yet to be re-
in bone have been considered to be osteoblasts,
solved if channel gating is caused by direct me-
osteocytes, or both. Theories on how these cells
chanical perturbation of the channel or if the
sense strain in their environment include
channel is activated secondarily because of acti-
streaming potentials, strain-sensitive ion chan-
vation of stretch-sensitive phospolipase C or D. It
nels and cytoskeletal reorganization.
has been suggested that ion channels may be
linked to the cytoskeleton and are opened when
Strain Released Potentials their cytoplasmic tail is phosporylated.20 The
Mechanical coupling denotes the conversion of problem that arises from these studies is that it is
mechanical energy into a form that can be de- difficult to determine the amount of applied
tected by cells. Bone responds to an applied strain and consequently any in vivo relevance. In
strain. Strain (e) represents a change in length addition to the calcium channels, stretch-acti-
(1 fjis represents 1 jam of deformation per meter vated potassium channels have been identified
of length). Normal strains in humans range in osteoblast-like cells.21
from 400 JLLS to 3000 JJLS depending on the level
of activity. The highest peak strains have been Extracellular Matrix and the Cytoskeleton
measured in a galloping horse at 3200 juie.17
Bone begins to fail when strains reach 7000 JULS. The process of cellular adhesion has been impli-
Application of small bending forces to long cated in many biologic phenomena, including
bones result in compression on one side and cell migration, anchorage, proliferation, differ-
tension on the opposite side. This produces entiation, and mechanotransduction.22 There
a flow of interstitial fluid, through the canalic- are two types of cellular adhesion: cell-to-cell
ular network, generating streaming potentials adhesion and cell-to-extracellular matrix (ECM)
and/or fluid shear stress. Streaming potentials adhesion. Several families of adhesion receptors
occur when electrically charged fluid is forced have been identified, including integrins. Inte-
over a tissue (cell membrane) with a fixed grins are cell surface receptors that mediate cell-
charge. For example, bone is partially composed to-cell attachment or cell attachment to ECM
of proteoglycans, which are entrapped in a col- molecules, such as fibronectin, laminin, and col-
lagen network. Because of the negative charge of lagen. Integrins are a family of a/ß het-
proteoglycans, there is an excess of positive mo- erodimeric cell surface receptors composed of at
bile ions in the fluid. Charges are symmetrically least 14 distinct a subunits and 8 or more ß
arranged so that no net macroscopic electric subunits that can associate noncovalently in var-
field is present. Compression of bone produces ious combinations. Although over 100 combina-
streaming potentials by the displacement of mo- tions are possible, the actual diversity is more
bile ions relative to charged proteoglycans en- limited because many a subunits can associate
trapped by collagen. Fluid movement over the with only a single ß subunit. The ß subunit
cell surface may directly stimulate bone cells (80-90 kDa) is more conserved, and each ß sub-
because it generates shear stress.18 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
Activation of Ion Channels
different ligands. Osteoblasts have been shown
Mechanosensitive ion channels in the plasma to express the integrin a 2 ßi and OL5ßl9 the pro-
membrane have been observed with single chan- totype receptors for collagen type I and fi-
nel recordings in more than 30 cell types from bronectin, respectively. Integrins may be puta-
plants to animals. The stretch-activated ion tive mechanotransducers because they have
channels allow the passage of cations (calcium been found to regulate signaling pathways by
and potassium). Application of mechanical changing intracellular calcium, regulating inosi-
<<
Article
>> Home | TOC |
Index
tol lipid turnover and phosphorylation of intra- of osteoclasts, as well as a decrease in the num-
cellular proteins. Mechanical stretching of the ber of TRAP+ cells, was demonstrated when the
human osteosarcoma cell line TE-85 has been RGDS peptide was injected locally during
shown to induce ßl integrin expression and al- OTM.30 OTM and other parameters of bone
ter its distribution.23 turnover, however, were not measured.
The specific binding of individual integrins Previously, the osteocytes or osteoblasts were
has been elucidated with the use of cell adhesion considered to be the cells that were responsive
assays, monoclonal antibodies (Mabs), and affin- to mechanical stimulation. In a recently pub-
ity chromatography. The first receptor-binding lished study, however, Kurata and colleagues
site to be defined was the RGD (arginine-glycine- found mature osteoclasts to be responsive to
aspartic acid) sequence present in fibronectin.24 mechanical strain.31 In this in vitro study, ma-
Since then, other adhesive molecules, such as ture osteoclasts responded to mechanical strain
vitronectin and collagen, have been shown to by significantly increasing mRNA expression of
contain the RGD sequence. Proteins containing TRAP and cathepsin. By using a pit formation
the RGD sequence, or variations thereof, have assay, they also demonstrated an increase in the
been engineered to inhibit ECM binding and to number of pits as well as in the total pit area.
study cellular adhesion. One structural feature This suggests the up-regulation of bone resorp-
common to each of these peptides is the aspar- tion by osteoclasts directly from mechanical
tate residue. It is believed that this residue may stimulation. They attributed the mechanosens-
play a critical role in integrin binding because ing mechanism to a stretch-activated cation (SA-
alteration of the aspartic acid in the peptide cat) channel because when godolinium, a SA-cat
sequences abrogates adhesive activity. For exam- channel blocker was used, pit formation and
ple, when RGDS (arginine-glycine-aspartic acid- mRNA levels decreased.
serine) and RGES (arginine-glycine-glutamic
acid-serine) were investigated for their ability to
How Bone Cells Translate Mechanical
bind to osteoblasts and inhibit cell attachment
Strain
to fibronectin, it was found that although both
peptides bound to osteoblasts, only RGDS ef- Regardless of how a mechanical signal is re-
fected cell attachment.25 RGES had minimal ef- ceived, it must somehow influence the biochem-
fect on cell attachment, whereas RGDS partially ical machinery of the cell. In the signaling cas-
(55% to 60%) inhibited cell adhesion to fi- cade process, receptor activation is followed by
bronectin. second-messenger generation (adenosine 3',5'
The influence of ECM on the survival and cyclic monophosphate [cAMP] and inositol
function of bone cells cannot be overempha- trisphosphate [IPS]).32 These advance the signal
sized. Perturbing the interaction of osteoblasts to the nucleus through a series of kinases. In the
with its integrin receptors by using antibodies or nucleus, different second messengers account
RGD-like fragments suppresses the formation of for the differential pattern of immediate early
mineralized nodules in vitro and delays the ex- gene (IEG) expression. lEGs are among the ear-
pression of tissue-specific genes, including osteo- liest responses that can be measured at the tran-
calcin.26' 27 In addition, apoptosis of differenti- scription level. The transcription of the lEGs
ated osteoblastic cells can be triggered by (c-fos, c-jun, and egr-1) has been shown to in-
interfering with their interaction with fibronec- crease when cells are exposed to cytokines,
tin.28 Osteoclasts also express integrin receptors growth factors,33 or mechanical stimulation.34
including the vitronectin receptor (av/33), which Protein products from the c-fos and c-jun genes
plays an important role in the adhesion of oste- form a heterodimeric complex named activator
oclasts to bone surface. Peptides containing the protein-1 (AP-1). AP-1 is a transcription factor; it
RGD motif have been shown to inhibit oste- binds to the promoter region of a gene and then
oclast-mediated bone resorption in vitro and modulates its activity35. Depending on the state
prevent osteoporosis in vivo.29 These peptides of the cell or in the presence of various stimuli,
could potentially aid in the treatment of osteo- AP-1 can produce either cellular proliferation or
porosis and possibly in modulating OTM. In differentiation.
fact, in a recent study, a decrease in the number We are beginning to understand the pattern
<<
Article
>> Home | TOC |
Index
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
14d
of bone formation: Molecular clones and activities. Sci-
ence 1988;242:1528-1534.
6. Komori T, Kishimoto T. Cbfal in bone development.
Figure 2. Relative mRNA expression of c-fos as deter- Curr Opin Genet Dev 1998;8:494-499.
mined by densitometry. Densitometric analysis of gel 7. Otto F, Thornell AP, Crompton T, et al. Cbfal, a candi-
pictures by using a Hewlett-Packard desktop scanner date gene for cleidocranial dysplasia syndrome, is essen-
with NIH Image (v. 1.57) software provided a deter- tial for osteoblast differentiation and bone development.
mination of the intensity of the bands. For each sam- Cell 1997;89:765-771.
ple at each time point, the ratio of the integrated 8. Ducy P, Zhang R, Geoffrey V, et al. Osf2/Cbfal: A
optical densities (IOD) of the c-fos band to G3PDH transcriptional activator of osteoblast differentiation.
band was calculated. In the untreated rats at each Cell 1997;89:747-754.
time point, the ratio of o/os/GSPDH was set to 1, and 9. Aubin JE, Bonnelye E. Osteoprotegrin and its ligand: A
the expression of the c/os/GSPDH in treated rats was new paradigm for regulation of osteoclastogenesis and
expressed as a ratio of this value. Mean and standard bone resorption. Osteoporos Int 2000;! 1:905-913.
deviation of the c-/os/G3PDH ratio for each group
10. Simonet WS, Lacey DL, Dunstan CR, et al. Osteoprote-
and at each time point was calculated. For compari-
ge rin: A novel secreted protein involved in the regula-
son of two groups at the same time interval, the
tion of bone density.. Cell 1997;89:309-319.
unpaired Student's t test was used. P <.05 was consid-
11. Tsuda E, Goto M, Mochizuki S, et al. Isolation of a novel
ered significant.
cytokine from human fibroblasts that specifically inhibits
osteoclastogenesis. Biochem Biophys Res Commun
1997;234:137-142.
signal transduction because many pathways, 12. Bekker PJ, Holloway D, Nakanishi A, et al. The effect of
such as tyrosine kinase, p21 ras and MAP ki- a single dose of osteoprotegerin in postmenopausal
nases, induce its expression.42 In the later stages women. J Bone Miner Res 2001;16:348-360.
13. Kobayashi Y, Hashimoto F, Miyamoto H, et al. Force-
of the OTM cycle, as part of AP-1, it may regulate
induced osteoclast apoptosis in vivo is accompanied by
osteoblast differentiation. AP-1 binding sites are elevation in transforming growth factor beta and osteo-
found in promoter regions of gene that are in- protegerin expression. J Bone Miner Res 2000; 15:1924-
volved in mineralization, including collagen 1934.
type I, osteocalcin, and alkaline phosphatases. 14. Yasuda H, Shima N, Nakagawa N, et al. Identity of oste-
We cannot yet determine whether the induc- oclastogenesis inhibitory factor (OCIF) and osteoprote-
gerin (OPG): A mechanism by which OPG/OCIF inhib-
tion of c-fos described in this study occurs di- its osteoclastogenesis in vitro. Endocrinology 1998;139:
rectly or indirectly. Mechanical loading of bones 1329-1337.
in vivo has been correlated with increases in PGs, 15. Lacey DL, Timms E, Tan HL, et al. Osteoprotegerin
nitric oxide (NO), insulin-like growth factor, ligand is a cytokine that regulates osteoclast differentia-
and transforming growth factor-/3. When me- tion and activation. Cell 1998;93:165-176.
16. Hofbauer LC, Khosla S, Dunstan CR, et al. The roles of
chanical loading of bones was performed in the
osteoprotegerin and osteoprotegerin ligand in the para-
presence of inhibitors of PGs or NO, <>/asmRNA crine regulation of bone resorption. J Bone Miner Res
expression was partially inhibited.43 This sug- 2000;15:2-12.
gests that the application of an orthodontic 17. Rubin CT. Skeletal strain and the functional significance
force to the teeth may activate more than one of bone architecture. Calcif Tissue Int 1984;36:S11-S18.
signaling pathway. Ultimately, an understanding 18. Klein-Nulend J, Semeins CM, Burger EH. Prostaglandin
mediated modulation of transforming growth factor-
of these pathways may provide novel strategies beta metabolism in primary mouse osteoblastic cells in
on when and how forces should be applied to vitro. J Cell Physiol 1996;168:l-7.
teeth. 19. Duncan R, Misler S. Voltage-activated and stretch-acti-
<<
Article
>> Home | TOC |
Index
vated Ba2+ conducting channels in an osteoblast-like and their relevance to orthodontic theory and practice.
cell line (UMR 106). FEBS Lett 1989;251:17-21. Am J Orthod Dentofacial Orthop 1993;103:212-222.
20. Opsahl LR, Webb WW. Transduction of membrane ten- 33. Lau LF, Nathans D. Genes induced by serum growth
sion by the ion channel alamethicin. Biophys J 1994;66: factors, in Foulkes PCAJ (ed): The Hormonal Con-
71-74. trol Regulation of Gene Transcription. New York,
21. Davidson RM, Tatakis DW, Auerbach AL. Multiple forms Elsevier,1991, pp 257-293.
of mechanosensitive ion channels in osteoblast-like cells. 34. Dolce C, Kinniburgh AJ, Dziak R. Immediate early-gene
Pflugers Arch 1990;416:646-651. induction in rat osteoblastic cells after mechanical de-
22. Ingber DE. Integrins as mechanochemical transducers. formation. Arch Oral Biol 1996;41:1101-1108.
Curr Opin Cell Biol 1991 ;3: 841-848. 35. Lian JB, Stein GS, Bortell R, et al. Phenotype suppres-
23. Carvalho RS, Scott JE, Yen EH. The effects of mechanical sion: A postulated molecular mechanism for mediating
stimulation on the distribution of beta l integrin and the relationship of proliferation and differentiation by
expression of beta l-integrin mRNA in TE-85 human Fos/Jun interactions at AP-1 sites in steroid responsive
osteosarcoma cells. Arch Oral Biol 1995;40:257-264. promoter elements of tissue-specific genes. J Cell Bio-
24. Ruoslahti E, Pierschbacher MD. New perspectives in cell chem 1991;45:9-14.
adhesion: RGD and integrins. Science 1987;238:491-497. 36. Raab-Cullen DM, Thiede MA, Petersen DN, et al. Me-
25. Puleo DA, Bizios R. RGDS tetrapeptide binds to osteo- chanical loading stimulates rapid changes in periosteal
blasts and inhibits fibronectin-mediated adhesion. Bone gene expression. Calcif Tissue Int 1994;55:473-478.
37. King GJ, Keeling SD, McCoy EA, et al. Measuring dental
1991;12:27l-276.
drift and orthodontic tooth movement in response to
26. Akiyama SK, Olden K, Yamada KM. Fibronectin and
various initial forces in adult rats. Am J Orthod Dento-
integrins in invasion and metastasis. Cancer Metastasis
facial Orthop 1991;99:456-465.
Rev 1995;14:173-189.
38. SambrookJ, Fritsch EF, Maniatis T. Molecular Cloning:
27. Clover J, Dodds RA, Gowen M. Integrin subunit expres-
A Laboratory Manual, vols 1-3. Cold Spring Harbor, NY:
sion by human osteoblasts and osteoclasts in situ and in
Cold Spring Harbor Laboratory Press, 1989.
culture. J Cell Sei 1992;103:267-271. 39. Chambers TJ, Fox S, Jagger CJ, et al. The role of pros-
28. Frisch SM, Francis H. Disruption of epithelial cell-matrix taglandins and nitric oxide in the response of bone to
interactions induces apoptosis. J Cell Biol 1994;124:619- mechanical forces. Osteoarthritis Cartilage 1999;7:422-
626. 423.
29. Engleman VW, Nickols GA, Ross FP, et al. A peptidomi- 40. Herrlich P, Ponta H. 'Nuclear' oncogenes convert extra-
metic antagonist of the alpha(v)beta3 integrin inhibits cellular stimuli into changes in the genetic program.
bone resorption in vitro and prevents osteoporosis in Trends Genet 1989;5:112-115.
vivo. J Clin "invest 1997;99:2284-2292. 41. Roberts WE, Chase DC. Kinetics of cell proliferation and
30. Terai K, Takano-Yamamoto T, Ohba Y, et al. Role of migration associated with orthodontically-induced os-
osteopontin in bone remodeling caused by mechanical teogenesis. J Dent Res 1981;60:174-181.
stress. J Bone Miner Res 1999; 14:839-849. 42. Sadoshima J, Izumo S. Mechanical stretch rapidly acti-
31. Kurata K, Uemura T, Nemoto A, et al. Mechanical strain vates multiple signal transduction pathways in cardiac
effect on bone-resorbing activity and messenger RNA myocytes: Potential involvement of an autocrine/para-
expressions of marker enzymes in isolated osteoclast crine mechanism. EMBO J 1993;12:1681-1692.
culture. J Bone Miner Res 2001;16:722-730. 43. Chow JW, Fox SW, Lean JM, et al: Role of nitric oxide
32. Sandy JR, Farndale RW, Meikle MC. Recent advances in and prostaglandins in mechanically induced bone for-
understanding mechanically induced bone remodeling mation. J Bone Miner Res 1998;13:1039-1044.
<<
Article
>> Home | TOC |
Index
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 the 10 trillion cells composing each human be-
brought to center stage. In talking about ing. These 6 feet of DNA per cell, coiled and
technological advances in dentistry, one cannot tightly packed into 46 chromosomes, withstand
but gaze with awe on the tremendous accom- the strenuous task of duplication every time the
plishments of the Human Genome Project cell divides. Additionally, about 50,000 different
(HGP). kinds of proteins must be supplied "in the right
What a long and exciting trip it has been,1 quantities, at the right times, and to the right
from the time the father of the science of genet- places" for the well-being of the human mind
ics, Gregor Mendel (1822-1884), the man who and body.3 Health is truly a miracle considering
created a method to investigate heredity, an- the estimated de novo mutation rate (30 muta-
nounced his theory in a meeting in 1865, to tions per lifetime) and the hundreds of genetic
1901 when biologist William Bateson coined the mutations we all inherit from our parents. For-
word genetics, and finally to 2001 when biology's tunately, most of these misprints are safe be-
gigaproject, the so-called Human Genome Se- cause they do not affect the decision-making
quencing, came into focus at the Nature Web parts of a gene.3
site (http://www.nature.com). A century after the A range between 25,000 and 40,000 genes
first description of Mendelian human genetics, contains the recipe that defines us. This is sur-
Nature and Science contain the first reports of the prising because it represents only twice the num-
complete sequence of the human genome. "My ber needed to make a fruit fly, a worm, or a
time will come," Mendel said expressing his con- plant. We can blame our genome that the 5.5
viction of the significance of his theory, and he billion people on this planet are so similar and at
was right again. the same time so different. Eye color, psychiatric
The human genome sequence contains the disorders, prognathism, susceptibility to infec-
genetic code that resides at the core of every 1 of tion, 32 or fewer teeth, and any other differences
between two unrelated individuals could be con-
tributed to a surprisingly small number of dis-
From the Department of Orthodontics, Tufts University School of crete genetic instructions.2'3
Dental Medicine, Boston, MA, and the University of Athens Faculty
of Dentistry, Athens, Greece.
Where do teeth stand in this pursuit of the
Supported in part by Howard Hughes Medical Institute, Na- "code"? From descriptive dental histology and
tional Hellenic Scholarship Foundation, Alexander S. Onassis studies on tooth development, we have known
Foundation, and March of Dimes Birth Defects Foundation. that the interaction of oral epithelial cells and
Address correspondence to Heleni Vastardis, DDS, DMSc, 31 the underlying mesenchymal cells of neural
Agias Sophias Street, Neo Psychico 15451, Greece.
Copyright 2002, Elsevier Science (USA). All rights reserved. crest descent is required to form a tooth. The
1073-8746/02/0801-0003$35.00/0 fundamental process is similar for all teeth. Each
doi:10.1053/sodo.2002.287l4 tooth develops through successive bud, cap, and
14 Heleni Vastardis
bell stages. During these early stages, the tooth incisors is missing, and the other has a conical
germs grow and expand, and the cells differen- shape, the so-called peg lateral. Pegs form bilat-
tiate to form the hard tissues of the teeth. Sig- erally at a frequency of 1.6%; a similar percent-
nificant biological information has emerged age appears when a peg is combined with a
over the preceding 50 years in the field of odon- well-shaped, but small-sized lateral.9 The inci-
togenesis providing clues on the genetic net- dence drops dramatically in whites when tooth
works regulating tooth development. Not sur- agenesis affects central incisors, especially the
prisingly, every stage of dental development has maxillary, canines, and first permanent mo-
been shown to be under strict genetic control. lars.10'12 There are few cases reported in the
Genes do play an important role in orchestrat- literature with simultaneous presence of super-
ing dental development.4 However, the focus of numerary teeth and missing teeth.12
these studies has not been directed toward hu- In the general white population, tooth agen-
mans, and the molecular basis of human dental esis ranges from 1% to 9.6%, between 6.6% and
development has yet to be defined. 9.2% in the Japanese population, and 7.7% in
Human molecular genetics provides a new the black population.13'14 The woman to man
methodology to answer old questions. The same ratio ranges from 3:2 for whites to 2:1 for
methodology can equally be applied to dental blacks.15'16
and medical research. Having a dental back- In the cases of primary dentition agenesis,
ground, I have put the following questions to there is an increased chance for permanent
myself: (1) Why are teeth missing? (2) Why are teeth failing to develop.17 However, in the ma-
some of them missing more often than others? jority of permanent tooth agenesis, the primary
(3) What accounts for the specificity of tooth dentition is intact, suggesting different genetic
shape and position? and (4) Is tooth agenesis mechanisms for the two sets of teeth.
genetic in origin?
Answering the last question has become easy.
When Nobel laureate Paul Berg of Stanford Uni-
Dentistry in the HGP
versity in a cancer symposium a few years ago
said "every human affliction is genetic in origin" Back in 1992, the time when the HGP was show-
people thought he was exaggerating, but it has ing tremendous promise, it was decided to ap-
turned out that he was not. proach the mystery of odontogenesis with a hu-
Agenesis of one or more teeth is the most man molecular genetic approach and learn
common developmental anomaly of the human from a genetic error. Medical research had al-
dentition. Although trauma, radiation, and sys- ready set the example. The genes responsible for
tematic destruction of the developing tooth bud the adult form of muscular dystrophy and a form
are likely origins of this condition, clinical stud- of dwarfism were identified. Much of the
ies clearly show that a considerable portion of all progress in reading DNA was coming from anal-
cases is inherited. Familial tooth agenesis is yses of genetic defects.
transmitted as an autosomal dominant, reces- The clinical observation was that tooth agen-
sive, or X-linked condition. There are numerous esis is phenotypically variable. On the basis of
reports in the literature on the clinical genetics this phenotypic variability and the hereditary
of tooth agenesis.5 character of tooth agenesis, it was hypothesized
The primary dentition is less frequently af- that defects in each of the several molecules
fected (0.1%) than the permanent. The decidu- associated with the steps of odontogenesis result
ous tooth at stake is usually an incisor followed in the clinical variation of tooth agenesis. The
by the first upper molar.6'7 In the permanent objective became the identification of the dental
dentition, the incidence of tooth agenesis varies genes that become mutated in families with miss-
with different teeth; third molars are missing ing teeth.
most frequently (30%) followed by mandibular Application of the HGP methodology de-
second premolars (3.4%) and maxillary lateral pends on availability of human material and spe-
incisors (2.2%). 8 Incisal hypodontia, in particu- cifically on the willingness of the families with
lar, presents a marked variability. In approxi- the condition of interest to donate a small quan-
mately 1.9% of the population, one of the lateral tity of their blood. After the clinical information
<<
Article
>> Home | TOC |
Index
is gathered and secure diagnoses are made, the failed to associate MSX1 with familial or isolated
experimental part of the process begins. tooth agenesis.25'26 Their results, however, agree
Systematic genome-wide search through ge- with our genetic heterogeneity hypothesis that
netic linkage analyses has identified the location more than one gene is responsible for the phe-
of a defective dental gene in a family with an notypic variability of tooth agenesis.
autosomal dominant agenesis of second premo-
lars and third molars on chromosome 4p. The
critical locus included the MSX1 gene, and mu- Conclusions
tational analysis showed a missense mutation in We are used to describing human beings in an
the homeodomain, the most highly conserved anatomic manner. The time has come when a
region of the MSX1 gene.18 Further biochemical number of genes will describe us. Physical char-
and functional analyses showed that this specific acteristics (phenotype) may matter less than the
mutation renders the gene inactive, and, there- set of genes one carries (genotype). Classifica-
fore, the reduced amount of MSX1 causes the tion of characteristics will soon be based on
particular phenotype in this family.19 genetic analyses. Orthodontists may have a diffi-
Since 1996, when the MSX1 gene was first cult time changing their morphologic classifica-
associated with familial tooth agenesis, the ge- tions (eg, Angle). However, the move from phe-
netics of tooth agenesis has received so much notypic to genotypic classification is inevitable.
attention that it has become one of the most The HGP has made a promise. If we answer
favored research areas. Significant work is being the following three questions, we will most likely
carried out, and an increasing number of genes be able to prevent or treat any of the estimated
involved in the regulation of tooth morphogen- 4,000 human genetic diseases: (1) What is the
esis is being identified as summarized later. genetic defect responsible for the condition of
A family with autosomal recessive hypodontia interest? (2) What is the normal gene product
associated with various dental anomalies, such as (protein)? and (3) Can the defective gene/pro-
enamel hypoplasia and failure of eruption, is tein be fixed or replaced?
linked to a region on chromosome 16q.20 There Human genetics provides a powerful tool to
is no report on the specific genetic defect re- study dental patterning. Genetic models will
sponsible for this type of hypodontia, as yet. eventually explain clinical observations such as
A family with agenesis of permanent molars is the anteroposterior site-specificity of tooth agen-
associated with a frame-shift mutation of the esis and the occurrence of associated dental
paired domain PAX9 gene through genome- anomalies.27 Mutations of genes have already
wide linkage analyses and candidate gene se- showed some of the key genetic interactions that
quencing.21 determine tooth shape and position.28 Func-
Another MSX1 mutation, (MetGlLys) up- tional investigation of the defects could reveal
stream of the homeodomain of MSX1, is identi- the mechanisms underlying patterning dysmor-
fied in a family presenting hypodontia.22 phologies and could point to better understand-
MSX1 is also associated with syndromic forms ing of odontogenesis.
of tooth agenesis. A nonsense mutation in the Although the raw genetic information has
MSX1 gene in a family with autosomal dominant limitations, its availability offers new experimen-
tooth agenesis and combinations of cleft palate tal approaches for studying craniofacial malfor-
only and cleft lip and palate provides evidence mation. What biological insight can be gleaned
for the importance of this gene in craniofacial from the draft sequence? This question surely
development and offers an appreciation of the cannot be answered by staring at a genome.
associations between different organs in the de- However, it is certain that this work will have
veloping body.23 profound long-term consequences in diagnos-
A nonsense mutation in the homeodomain of tics and targeted therapeutics.
MSX1 has also been found to cosegregate with
the phenotype of Witkop syndrome, also called
tooth and nail syndrome, suggesting that MSX1 References
is critical for both tooth and nail development.24 1. Davies K. Cracking the genome: Inside the race to un-
In the last six years, two independent groups lock human DNA/The sequence: Inside the race for the
<<
Article
>> Home | TOC |
Index
16 Heleni Vastardis
human Genome. Weidenfeld and Nicolson 2001, pp hypodontia and hyperodontia. Odont Revy 197l;22:309-
288-320. 315.
2. Available at: http://www.nature.com.genomics/human/ 17. Daugaard-JensenJ, Nodal M, Skovgaard LT, et al. Com-
papers. parison of the pattern of agenesis in the primary and
3. Available at: http://www.hhmi.org/genetictrail/reading/ permanent dentitions in a population characterized by
read.htm. agenesis in the primary dentition. Int J Paediatr Dent
4. Peters H, Balling R. Teeth. Where and how to make 1997;7(3):143-148.
them. Trends Genet 1999;15(2):59-65. 18. Vastardis H, Karimbux N, Guthua SW, et al. A human
5. Vastardis H. The genetics of human tooth agenesis: New MSX1 homeodomain missense mutation causes selective
discoveries for understanding dental anomalies. Am J tooth agenesis. Nature Genet 1996;13:417-421.
Orthod Dentofac Orthop 2000;117(6):650-656. 19. Hu G, Vastardis H, Bendall AJ, et al. Haploinsufficiency
6. Bennett CG, Ronck SL. Congenitally missing primary of MSX1: A mechanism for selective tooth agenesis. Mol
teeth. Case report. J Dent Child 1980;47:346-349. Cell Biol 1998;18(10):6044-6051.
7. Daugaard-JensenJ, Nodal M, Kjaer I. Pattern of agenesis 20. Ahmad W, Brancolini V, Faiyaz ul Haque M, et al. A
in the primary dentition: A radiographic study of 193 locus for autosomal recessive hypodontia with associated
cases. Int J Paediatr Dent 1997;7(l):3-7. dental anomalies maps to chromosome 16ql2.1. Am J
8. Symons AL, Stritzel F, Stamatiou J. Anomalies associated Hum Genet 1998;62:987-991 (letter).
with hypodontia of the permanent lateral incisor and 21. Stockton DW, Das P, Goldenberg M, et al. Mutation of
PAX9 is associated with oligodontia. Nature Genet 2000;
second premolar. J Glin Pediat Dent 1993;17:109-111.
24:18-19.
9. Le Bot P, Salmon D. Congenital defects of the upper
22. Reising BC. The role of MSX1 mutation in the etiology
lateral incisors (ULI): Condition and measurements of
of hypodontia. Am J Orthod Dentofac Orthop 2001;
the other teeth, measurements of the superior arch,
119(5):564-565 (abstr).
head and face. Am J Phys Anthrop 1977;46:231-244.
23. van den Boorgaard M-JH, Dorland M, Beemer FA, et al.
10. Dermaut LR, Goeffers KR, De Smit AA. Prevalence of
MSX1 mutation is associated with orofacial clefting and
tooth agenesis correlated with jaw relationship and den- tooth agenesis in humans. Nature Genet 2000;24:342-
tal crowding. Am J Orthod Dentofac Orthop 1986;90: 343 (letter).
204-210. 24. Jumlongras D, Bei M, Stimson JM, et al. A nonsense
11. Lapeer GL. Congenitally missing maxillary first perma- mutation in MSX1 causes Witkop syndrome. Am J Hum
nent molars: A case report. J Can Dent Assoc 1990;56(6): Genet 2001;69(l):67-74.
535-536. 25. Nieminen P, Arte S, Pirinen S, et al. Gene defect in
12. Sharma A. A rare non-syndrome case of concomitant hypodontia: Exclusion of MSX1 and MSX2 as candidate
multiple supernumerary teeth and partial anodontia. genes. Hum Genet 1995;96(3):305-308.
J Glin Pediatr Dent 2001;25(2):167-169. 26. Scarel RM, Trevilatto PC, Di Hipolito O Jr, et al. Absence
13. Niswander JD, Sujaku C. Congenital anomalies of teeth of mutations in the homeodomain of the MSX1 gene in
in Japanese children. J Phys Anthropol 1963;21:569-574. patients with hypodontia. Am J Med Genet 2000;92(5):
14. Salinas CF, Jorgenson RJ. Dental anomalies in a black 346-349.
population. J Dent Res 1974;53:237 (abstr). 27. Peck S, Peck L, Kataja M. Site-specificity of tooth agen-
15. Jorgenson Rf. Clinician's view of hypodontia. J Am Dent esis in subjects with maxillary canine malpositions. Angle
Assoc 1980;101:283-286. Orthod 1996;66(6):473-466.
16. Egermark-Eriksson I, Lind V. Congenital numerical vari- 28. Thesleff L Genetic basis of tooth development and den-
ation in the permanent dentition: Sex distribution of tal defects. Acta Odontol Scand 2000;58(5):191-194.
<<
Article
>> Home | TOC |
Index
Different diagnostic methods, such as lateral onance (MR) tomography, now allow images to
and frontal skull radiographs and dental be obtained without overlying structures. Three-
casts, are integrated in the treatment of se- dimensional (3-D) images can be interactively
vere dentomaxillofacial deformities. In cephalo- reconstructed by data processing from axial CT
grams, the position of jaws and teeth in relation scans.1 3-D imaging allows topographic evalua-
to the base of the skull are measured on a two- tion of dentomaxillofacial deformities.2 Most
dimensional film plane. All conventional radio- 3-D reconstruction procedures require expen-
logic techniques are limited in their ability sive high-capacity computer hardware and soft-
to accurately measure the soft tissues and the ware; therefore, only radiologic centers or mod-
skull in all three dimensions. Also, conventional ern university clinics are able to integrate these
model surgery has a limited ability to assess max- new methods.3-5 The improved performance of
illofacial morphology because mounted dental usual personal computers enables CT recon-
casts mainly visualize the teeth and alveolar pro- struction.6
cesses. The data set of the CT scanner can also be
Progress in imaging techniques, especially used for the computer-aided manufacturing of
computed tomography (CT) and magnetic res- 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-
From the Department of Orthodontics Medical Faculty, Aachen, fer CT data in skull models is stereolithography.
Germany. This technique uses a photocurable monomer
Address correspondence to Professor Robert A. W. Fuhrmann, hardened by an ultraviolet laser.
Dr. med, Dr. med. dent., MD, DDS, PhD, Klinik für Kieferortho- Orthodontic treatment could be a part of
pädie, RWTH-Aachen, 52057 Aachen, Germany.
Copyright 2002, Elsevier Science (USA). All rights reserved. 3-D-model treatment simulation.9 In the last 10
1073-8746/02/0801-0004$35.00/0 years, we have assessed the integration of 3-D
doi:10.1053/sodo.2002.28167 cephalometry and individual 3-D skull models
18 Robert A. W. Fuhrmann
20 Robert A. W. Fuhrmann
22 Robert A. W. Fuhrmann
ment tools in the 3-D object on the monitor. The maxillofacial surgery. Amer J Orthod Dentofac Orthop
access of 3-D imaging with a readily available 1988;94:469-475.
2. Vannier MW, Marsh JL, Warren JO. Three-dimensional
computer system for 3-D cephalometry of the CT-reconstruction images for craniofacial surgical plan-
head will enable a larger number of clinicians to ning and evaluation. Radiology 1984;150:179-184.
use 3-D methods in diagnosis and treatment 3. Zonneveld FG, Normaan van der Dussen F. Three-di-
planning. The offline personal computer and mensional imaging and model fabrication in oral and
the user-specified software permit a higher de- maxillofacial surgery. Oral Maxillofac Surg Clin North
gree of flexibility independent of expensive Amer 1992;4:19-33.
4. Zonneveld FG, Fukuta K. A decade of clinical three-
workstations or radiologic centers. dimensional imaging: A review. Part II: Clinical Applica-
The plasticity of the visceral skeleton pro- tions. Investigat Radiol 1994;29:574-589.
vided by a life-size 3-D skull model is a precon- 5. Zonneveld FG. A decade of clinical three-dimensional
dition for realistic simulation of the complex imaging: A review. Part III: Image analysis and interac-
treatment sequence. 3-D skull model-based sur- tion, display options and physical models. Investigat Ra-
gery on milled or stereolithographic built mod- diol 1994;29:7l6-725.
6. Fuhrmann R, Schnappauf A, Diedrich P. Three-dimen-
els permit verification of the most suitable mo- sional imaging of craniomaxillofacial structures with a
bilization and placement of jaw segments. The standard personal computer. Dentomaxillofac Radiol
individual prediction could be copied during 1995;24:260-263.
the real clinical treatment. 3-D-model surgery 7. Lambrecht JT, Brix F. Individual skull model fabrica-
with integrated orthodontic setups represents a tion for craniofacial surgery. Cleft Palate J 1990;27:
382-387.
new dimension of treatment prediction. How-
8. Fuhrmann R, Frohberg U, Diedrich P. Treatment pre-
ever, because of the cost and time consumption, diction with three-dimensional computer tomographic
a 3-D working concept with 3-D cephalometry skull models. Amer J Orthod Dentofac Orthop 1994;106:
and 3-D skull models is presently limited to cases 156-160.
of severe asymmetric dentofacial disharmonies. 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
References Orofac Orthop 1996;57:32-45.
1. Moss JP, Grindrod SR, Linney AD, et al.A computer 10. Graber TM. Comment of the editor-in-chief. Amer
system for the interactive planning and prediction of J Orthod Dentofac Orthop 1995;107:360 (letter).
<<
Article
>> Home | TOC |
Index
24 Robert A. W. Fuhrmann
lying structures, and in identifying periodontal adapted when necessary. The contiguous axial
lesions.4'6'7 CT scans of the jaw ran from the cemento-
This review summarizes our experiences over enamel junction to the apices of the roots with a
the last 10 years integrating HR-CT scanning scan thickness and table feed of 1.0 mm. Accord-
during and after orthodontics. The aim was to ing to the curve of Spee and the root length, 15
investigate the incidence of bone dehiscences or to 30 scans per jaw were necessary. The CT data
fenestrations, root resorption, and osteoblastic were recorded in high-resolution mode with a
bone remodeling during and after orthodontic reconstruction matrix of512x512 pixels, a tube
treatment in adult patients. 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
Material and Methods
width of 4000 Hounsfield units (HU) and a cen-
Twenty-one adult patients with reduced peri- ter of 600 HU with a laser camera. To facilitate
odontal bone tissue (ie, narrow alveolar process, visualisation of the axial scan sequences and to
basally extended maxillary sinus, or advanced simplify qualitative assessment of the bone sur-
alveolar bone loss) were selected for this inves- face (dehiscence diagnosis), additional three-
tigation. In 15 of the patients, the first CT exam- dimensional reconstructions were computed
ination was performed before the start of orth- from the primary axial scans for the maxilla and
odontic treatment. The remaining 6 patients mandible of selected patients with the recon-
were scanned after the first phase of orthodontic struction software of the CT console.
treatment.
All patients were treated with fixed straight
Case Report
wire appliances (0.022-in slot width) and contin-
uous arch wires. Additional segment arch wires In an 18-year-old patient with a primarily narrow
were integrated during the levelling and guid- mandibular alveolar process and crowding of
ance phases. The second CT examination was the lower incisors, the first premolars were ex-
performed at an interval of 12 to 36 months after tracted after the pre-treatment CT-examination
the first CT examination according to the treat- (Fig 1). The axial CT scans revealed that the oval
ment time following removal of the fixed appli- section of the canine roots corresponded to the
ances. In 6 patients, a third CT examination was maximal width of the alveolar process. However,
performed after a retention period of 6 to 36 no bone dehiscences and fenestrations over the
months after debonding. buccal surface of the lower canines and incisors
Comparability of the patients was limited by could be evaluated in the axial CT scans or
the varying labiolingual width of the alveolar three-dimensional reconstructions (Fig 1A and
process, attachment level, and orthodontic treat- B). During the orthodontic treatment, the lower
ment concepts. For that reason, a descriptive canines were derotated and distalized with Sen-
evaluation of the CT findings was made with talloy springs. The further levelling of the fron-
reference to the individual morphologic situa- tal mandibular crowding was accomplished with
tion and the biomechanic treatment concept to the insertion of resilient continuous arch wires.
highlight different anatomic or therapeutic risk This resulted in the further protrusion of the
factors in the genesis of periodontal lesions. lower incisors.
After two years of orthodontic treatment with
straight wire appliances, a second CT examina-
CT Scanning Parameters tion showed extensive vestibular bone dehis-
The patients were placed horizontally on the cences above the root surfaces of the lower ca-
table of the CT scanner (Somatom plus, Siemens nines and the incisors in the three-dimensional
Ine, Erlangen, Germany). Each patient's head reconstruction (Fig 1C). Because of the narrow
was fixed with foamed plastic to avoid movement alveolar process, this uncontrolled protrusion
artifacts. After an initial lateral topogram, the led to bone dehiscences over the vestibular root
axial CT scanning direction was orientated ac- surfaces. The extent of primary bone covering
cording to the maxillary or mandibular occlusal and secondary subperiosteal apposition were not
plane. The gantry inclination was individually sufficient to prevent bone dehiscences. The cor-
<<
Article
>> Home | TOC |
Index
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 The dehiscences over the root surfaces of the
no periodontal recession. lower canines and the lateral incisors were corn-
After three years of retention with removable pletely filled with new alveolar bone. However,
appliances, a third CT examination showed par- the dehiscenses over the root surfaces of the
tial periodontal remodeling of the vestibular central incisors persisted partly. Only the apical
bone and dehiscences over the root surfaces. and lateral area of these bone defects was re-
<<
Article
>> Home | TOC |
Index
26 Robert A. W. Fuhrmann
paired 5 (FiglE). The clinical examination not observed over the palatal surfaces of the
showed no differences between the lower inci- upper incisors. The retrusion of the upper inci-
sors (Fig ID). sors, however, initiated root resorption.
Therapeutic risks can be observed with un-
controlled tipping tooth movements with con-
Results tinuous resilient arch wires and with a tooth
Comparison of the first, second, and third CT movement vertical to the alveolar process. Pro-
findings revealed substantial differences in the trusion, retraction, and intrusion of mandibular
positions and angulations of individual teeth as a or maxillary incisors were especially critical
result of orthodontic treatment, therefore, the tooth movements. Similarly, overloading of the
CT sectioning did not always correspond. There intraoral anchorage by using intermaxillary elas-
was no clinical evidence of marginal periodonti- tice or cortical root torque may result in substan-
tis or periodontal recession at the orthodonti- tial attachment loss.
cally moved teeth in any of the patients before, A complete osteoblastic periodontal remodel-
during, or after orthodontic treatment. ing of the orthodontically induced bone defects
The CT findings revealed substantial individ- was seen above the palatal root surfaces of the
ual variation, depending on the extension of the upper premolars and molars. The lowest level of
alveolar process and the therapeutic concept. In osteoblastic periodontal remodeling was seen in
cases of translatory tooth movements in a can- the lower frontal area above the buccal and lin-
cellous alveolar process, generally there was no gual root surfaces of the incisors and canines. In
notable root resorption or bone dehiscences re- some teeth, the dehiscences or fenestrations
vealed. were totally repaired, in others not. A spontane-
Expansion of the maxillary arch with a quad- ous reorientation or relapse of single teeth in
helix and continuous expanded arch wires initi- the direction of the initial tooth position initi-
ated a partial resorption of the covering bone ated a complete periodontal remodeling.
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
Discussion
posterior teeth. It was conspicuous that the mo- As the labiolingual diameter of the alveolar pro-
lars and premolars at the center of maximum cess decreases and the eccentric positioning of
expansion had the highest degree of bone re- the tooth increases, the facial/lingual bone plate
sorption and of subperiosteal bone apposition. above the root surfaces is reduced. The thera-
Despite the reactive osteoblastic remodeling of peutic risk inherent in the application of uncon-
the buccal bone plate and the primary bone trolled orthodontic force systems may lead to a
covering, dehiscences were seen in the second breakdown of the bone plate and the covering
CT examination. No corresponding lateral root soft tissue in terms of periodontal recession. The
resorption, nor clinically detectable gingival re- bone can be only slightly increased by periosteal
cession was found. apposition during orthodontic tooth movement.
CT examination of adult patients indicates On extending beyond the primary bone volume
various anatomic and/or topographic and ther- of the alveolar process and the secondary, newly
apeutic risk factors in the initiation of periodon- formed bone apposition, the periodontal liga-
tal lesions. Some of these were anatomic and/or ment fuses with the apposing periosteum, and a
morphologic risks arising from a disproportion bone dehiscence develops. Additional factors,
between tooth width and labiolingual extension such as traumatic lesions resulting from forceful
of the alveolar process or from an eccentric cleaning techniques or plaque-induced mar-
tooth position, a deep maxillary sinus, or ad- ginal gingivitis and periodontitis, may accelerate
vanced bone loss. Bone dehiscences or fenestra- the development of periodontal recessions.
tions were repeatedly found at mandibular CT scanning in orthodontically treated pa-
incisors. A small symphysis with reduced labio- tients allows single-tooth interpretation of re-
lingual bone width, frontal crowding, and thin modeling and hard-tissue lesions at the peri-
facial or lingual bone plates predispose to bone odontal tissue in the wake of various therapeutic
dehiscences. Fenestrations or dehiscences were techniques.8 The periradicular osteodynamics
<<
Article
>> Home | TOC |
Index
resulting from a therapeutically induced tooth bone apposition, there was no histologic evi-
movement can be evaluated three-dimensionally dence of cortical perforation.
by comparing two or three different CT exami- Marginal bone dehiscences were detected
nations during or after treatment. above the buccal root surfaces of teeth in the
Various anatomic, morphologic, and thera- expansion center of a maxillary arch. Clinically,
peutic risk factors are intensified reciprocally no localised periodontal recessions were found
because the side effects of orthodontic therapy in these patients. The empirical concept of cor-
increase with reduced attachment level. When tical anchorage preparation has the inherent
teeth with a critical ratio between the labiolin- risk of a massive attachment loss at the incisors
gual tooth diameter and the width of the alveo- and at the molars because the roots are being
lar process are derotated, marginal bone dehis- moved in the dense compact tissue.13 The
cences are induced if the bone apposition is present CT findings revealed extensive bone de-
inadequate. In histologic posttreatment exami- hiscences after cortical anchorage preparation.
nations of a human specimen dehiscences and In particular, orthodontic uprighting of the in-
fenestrations of various sizes were found.9'10 Ex- cisors to cephalometric norms may be question-
tensive fenestrations resulting from tooth move- able in cases of a narrow symphysis, thin bone
ment in an atrophied alveolar process or the plates, or advanced alveolar bone loss.
dentoalveolar maxillary sinus were detected his- Orthodontically induced bone dehiscences
tologically. Overall, the present CT findings con- were partly repaired by osteoblastic periodontal
firm these histologic findings. Clearly, there is a remodeling or spontaneous reorientation of the
predisposition to attachment loss at these sites. teeth in the retention period. Within the scope
of differential therapeutic assessment of tooth
The degree of bone apposition by the superim-
movements, a reasonable risk-benefit calculation
posed periosteum is not enough to avoid bone
is needed to assess the extent to which the initial
dehiscences in cases of a primarily narrow apical
periodontal situation, the adaptability of the cov-
base or a deep maxillary sinus and extensive
ering hard and soft tissue and the expected os-
tooth movements. With critical initial findings,
teoblastic remodeling permit extensive orth-
tooth movements should be confined to the an-
odontic tooth movement.
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 References
arch wires for levelling resulted partly in tooth 1. Engelking G, Zachrisson BU. Effects of incisor reposi-
tipping. Treatment with uncontrolled force sys- tioning on monkey periodontium after expansion
tems corresponding to the widespread ideal through the cortical plate. Amer J Orthodont 1983;82:
arch concept from straight-wire mechanics is 23-32.
2. Wainwright WM. Faciolingual tooth movement: Influ-
open to criticism with periodontal involution ence on the root and cortical plate. Amer J Orthodont
because the risk of iatrogenic lesions increases 1973;64:278-289.
with decreasing attachment level. Mulie and Ten 3. Thilander B, Nyman S, Rarring T, et al. Bone regenara-
Hoeve also demonstrated that bone dehiscences tion in alveolar bone dehiscences related to orthodontic
tooth movements. Europ J Orthodont 1983;27:105-114.
occur with a narrow frontal alveolar process.11'12 4. Fuhrmann R, Bücker A, Diedrich P. Assessment of alve-
With sagittal incisor movements in the maxilla, olar bone loss with high resolution computed tomogra-
low-grade fenestrations were found above the phy. J Periodont Res 1995;30:258-263.
facial root apices in cases of uncontrolled tip- 5. Graber TM. Comment of the editor-in-chief. Amer
J Orthod Dentofac Orthop 1995; 107: 360.
ping. Even after extensive palatal movement, a
6. Fuhrmann R. Three-dimensional interpretation of
continuous bone covering was revealed on the alveolar bone dehiscences. An anatomical-radiological
marginal and apical aspects over the palatal root study. J Orofac Orthoped 1996;57:62-74.
surface in the CT scans. Wehrbein et al recorded 7. Fuhrmann R. Three-dimensional interpretation of labio-
extensive osseous remodeling processes directly lingual bone width of the lower incisors. An anatomical-
radiological study. J Orofac Orthoped 1996;57:168-185.
above the root during histologic examination of 8. Fuhrmann R. Three-dimensional interpretation of peri-
the periradicular bone of a human specimen odontal lesions and remodeling during orthodontic
with retracted incisors.9 Despite thin palatal treatment. J Orofac Orthopedic 1996;57:224-237.
<<
Article
>> Home | TOC |
Index
28 Robert A. W. Fuhrmann
9. Wehrbein H, Fuhrmann R, Diedrich P. Periodontal con- and standardized occlusal films. J Clin Orthodont 1976;
ditions after facial root tipping and palatal root torque of 10:882-899.
incisors in man. Amer J Orthodont Dentofac Orthoped 12. Ten Hoeve A, Mulie RM. The effect of antero-postero
1994;106:455-462. incisor repositioning on the palatal cortex, as studied with
10. Wehrbein H, Fuhrmann R, Diedrich P. Human tissue re- laminography. J Clin Orthodont 1976;10:804-822.
sponse after long-term orthodontic tooth movement. 13. Wehrbein H, Bauer W, Diedrich P. Mandibular incisors,
Amer J Orthodont Dentofac Orthoped 1995;107:360-37l. alveolar bone, and symphysis after orthodontic treat-
11. Mulie RM, Ten Hoeve A. The limitations of tooth move- ment - A retrospective study. Amer J Orthodont Dento-
ment within the symphysis studied with laminography fac Orthoped 1996;110:239-249.
<<
Article
>> Home | TOC |
Index
publication of methods for fabricating headgear the temporal characteristics of headgear wear;
timers.10"12 These timers were bulky and often 2) estimate how much force is delivered by the
necessitated placing the electronics in the neck headgear; 3) detect patient attempts to falsify
strap, which could diminish patient comfort. headgear use; and 4) provide readily accessible
Some designs removed the safety disconnect fea- feedback to the patients, parents, and orthodon-
ture that should be a part of all headgear force tists about headgear use. Initial prototypes were
modules. Most designs offered little protection fabricated and evaluated in laboratory and clin-
against patient attempts to falsify headgear use ical tests.
data and provided only a cumulative measure of The electronics and battery of the prototype
wear time. Finally, these devices provided the monitor are contained entirely within the head-
orthodontist with intermittent headgear wear in- gear force module. The monitor uses a micro-
formation, and none gave the patients easily processor with nonvolatile memory, magnet sen-
accessible feedback about their degree of adher- sors, an infrared light-emitting diode (LED) and
ence to the headgear wear regimen.5'6 phototransistor for bi-directional communica-
Newer devices to measure headgear use have tion and a 4-digit liquid crystal display (LCD)
been developed. The Ortho Kinetics Corpora- screen to display output. A lithium battery pow-
tion (Vista, CA) has made a commercially avail- ers the monitor. A small cylindrical rare-earth
able headgear timer. An optical sensor, which magnet is embedded in the end of the adjust-
can detect when the spring in the module is ment strap. The basic unit of measurement is
compressed, activates the timer in this headgear the position of the magnet within the body of
force module. This device can record each pe- the force module, which changes as the strap is
riod of headgear use, and the record of head- extended and the internal spring is compressed.
gear wear can be reviewed during visits at the By measuring the position of the adjustment
orthodontic office. Unfortunately, it is possible strap within the force module, it is possible to
for patients to activate the timer without actually determine when the headgear is in use. In addi-
wearing the headgear. To create a long-term tion, the force resulting from a known amount
record of falsified wear data, however, a patient of spring compression can be calculated from
would need to adhere to this pattern of decep- calibration data stored in a look-up table in the
tive behavior over a long period (eg, remember monitor's memory.
to compress the spring before bed and allow the
spring to relax on awakening). Scientists at the
University of Florida have developed a micropro- Laboratory Testing
cessor-controlled headgear-timing device that Three laboratory tests are described. The first
can record the amount of spring compression test measured the accuracy of the clock crystal
that occurs during headgear use.13 This design that determines when headgear wear occurred.
has the advantage of measuring the dynamic The second test determined whether the moni-
spring compression that occurs during headgear tor could accurately measure the amount of
wear and so can provide information about spring compression. The third test evaluated
headgear use and the amount of force produced whether the amount of force generated by the
by the spring. The device has limited clinical spring during its compression could be esti-
utility because it is somewhat fragile and its de- mated by knowing the amount of spring com-
sign requires connecting the bulky electronics, pression.
located in the neckstrap, to the headgear force
module.
Clock Accuracy
A novel headgear monitor has been devel-
oped at the University of Washington (US Patent The microprocessor in the headgear monitor is
#5,980,026); the rationale underlying its design equipped with a standard 32,768-Hz crystal os-
is described elsewhere.5'6 In brief, the guiding cillator internal clock. This clock is used to reg-
design principles were to make the device small, ister the headgear use data in real time. To test
durable, comfortable, and to maintain the safety the accuracy of the timing unit, the clock was
breakaway feature so that it could be used clin- synchronized with the National Bureau of Stan-
ically. The device was designed to 1) measure dards' clock through internet access. The head-
<<
Article
>> Home | TOC |
Index
*\
Maximum
Spring 25 -i
compression. The second test described how
\
Compresson W
spring compression changes during headgear
wear.
Estimated
Magnet 15 -
20 -
\N
u
^
Position "ffitfftj
Effect of Head Posture on Spring JW.
Compression
(mm)
10 - \r l* ^
t
removed. Output from both magnet sensors from the University of Florida has suggested that
were collected at a rate of four samples per an absence of movement over an hour would be
minute, and these data were converted to an a good criterion to indicate no wear.14
estimate of adjustment strap position by using One of our design goals5 was for the monitor
the look-up table. The four episodes of headgear "to provide continuous, accurate, and easily in-
use were detected accurately by the monitor. terpretable visual feedback to patients, parents,
Figure 4 illustrates the position of the adjust- and clinicians about headgear use." This is an-
ment strap for both monitors during an eight- other unique aspect of our design because no
hour period when the subject was asleep. Al- headgear timers have provided feedback directly
though the subject was sleeping, there was to patients in an effort to enhance adherence via
considerable variation in the degree of spring improved self-regulation.6 The LCD on the mon-
compression over time, which is indicative of itor can be programmed to show the average
head movement. These data also support the daily wear time and/or the cumulative number
observation that the spring compression of the of hours of wear since the last visit. This feature
right and left sides are frequently related (eg, as is convenient for the orthodontist to assess the
one side compresses, the other relaxes). As patient's progress, and it provides the patient
could be expected, data collected while the sub- with ongoing feedback to regulate his or her
ject was awake exhibited greater variability than own behavior. Self-monitoring has been sug-
data collected during sleep. gested as a useful strategy to enhance patient
When the headgear is not being worn, there adherence.5'7
is very little variability in the position of the
adjustment strap over time. Similarly, when the
monitors were placed on a mannequin head to Conclusion
simulate (falsify) headgear use, the data showed These initial laboratory and clinical tests were
an absence of variability in the movement of the conducted to validate the functionality of this
adjustment strap. Thus, actual wear can be dis- novel approach to monitoring headgear wear.
tinguished from static attempts to falsify head- Based on the encouraging findings of these tests,
gear use by measuring the variability in the po- an improved version of the headgear monitor is
sition of the adjustment strap over time. As being developed, and these monitors will un-
shown in Figure 4, there is considerable variabil- dergo extensive evaluation in private orthodon-
ity in the data, although there was a 45-minute tic offices. The development of a sophisticated
period between hours 6 and 7 when there was headgear monitor will be valuable for both sci-
little variability in the position of the adjustment entific and clinical orthodontic applications.
strap even though it was being worn. Research Sufficient flexibility has been designed into this
headgear monitor so that it can meet the needs
20- of scientists as well as clinicians.
Estimated 15 -
Magnet
Position for 10 -
Left Monitor
(mm) Acknowledgment
We would like to acknowledge the contributions of Chris
20 -
Prall, Mani Soma, and Susan Herring to this article.
Estimated 15 -
Magnet
Position for ,
Right Monitor 10 "
(mm)
References
1. GhafariJ, King GJ, TullochJF. Early treatment of Class
Time (hours) II, division 1 malocclusion—Comparison of alternative
treatment modalities. Clin Orthod Res 1998;1:107-117.
Figure 4. Two headgear monitors were worn with a 2. Keeling SD, Wheeler TT, King GJ, et al. Anteroposterior
cervical headgear appliance to collect data from both skeletal and dental changes after early Class II treatment
the subject's left and right sides simultaneously. Al- with bionators and headgear. Am J Orthod Dentofacial
though the subject was asleep, there was considerable Orthop 1998; 113:40-50.
variation in the degree of spring compression over 3. TullochJF, Phillips C, Koch G, et al. The effect of early
time which is indicative of head movement. intervention on skeletal pattern in Class II malocclusion:
<<
Article
>> Home | TOC |
Index
A randomized clinical trial. Am J Orthod Dentofacial 8. Northcutt ME. The timing headgear. J Clin Orthod
Orthop 1997;111:391-400. 1974;8:321-324.
4. Tulloch JF, Proffit WR, Phillips C. Influences on the 9. Banks PA, Read MJF. An investigation into the reliability
outcome of early treatment for Class II malocclusion. of the timing headgear. BrJ Orthod 1987; 14:263-267.
Am J Orthod Dentofacial Orthop 1997;111:533-542. 10. Cureton SL, Regennitter FJ, Orbell MG. An accurate,
5. Ramsay DS, Soma M, Sarason IG. Enhancing patient inexpensive headgear timer. J Clin Orthod 1991;25:749-
adherence: The role of technology and its application to 754.
orthodontics, in McNamara JA Jr, Trotman CA (eds): 11. Guray E, Orhan M. Selcuk type headgear-timer (STHT).
Creating the Compliant Patient. Craniofacial Growth Am J Orthod Dentofacial Orthop 1997;111:87-92.
Series, Center for Human Growth and Development, 12. Kyriacou PA, Jones DP. Compliance monitor for use
University of Michigan, Ann Arbor, MI, 1997;33:141-165. with removable orthodontic headgear appliances. Med
6. Lyons EK, Ramsay DS. A self-regulation model of patient Biol Eng Comput 1997;35:57-60.
compliance in orthodontics: Implications for the design 13. Johnson PD, Bar-Zion Y, Taylor M, et al. Effects of head
of a headgear monitor. Semin Orthod 2000;6:224-230. posture on headgear force application. J Clin Orthod
7. Dunbar-Jacob J, Schlenk E. Patient adherence to treat- 1999;33:94-97.
ment regimen, in Baum A, Revenson TA, Singer JE 14. McGorray SP, Dean GS, Keeling SD, et al. Monitoring
(eds): Handbook of Health Psychology. Mahwah, NJ: orthodontic headgear wear with a micro-electronic data
Lawrence Erlbaum Associates, 2001, pp 571-580. acquisition system. J Dent Res 1997;76:127 (abstr 909).
<<
Article
>> Home | TOC |
Index
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.
36 Dolce et al
bony segments. With rigid fixation, jaw immobi- institution. Thirty days before the scheduled sur-
lization is eliminated, allowing jaw function gery date, notification of protocol assignment
within a few days after surgery. (either screw or wire) was made by the project
Many studies have compared both the short- statistician.
and long-term stability of rigid and wire fixation. The subjects were evaluated 7 to 14 days be-
Short-term relapse has been demonstrated with fore surgery (T2), within 1 week after surgery
wire2'4'5 and with rigid fixation.6'9 Long-term sag- (T3), and at 8 weeks (T4), 6 months (T5), 1 year
ittal relapse with wire fixation ranges from 23% (T6), and 2 years (T7) after surgery. Some sub-
to 45 %,10'12, whereas with rigid fixation it ranges jects entered the study before the start of orth-
from 8% to 18%.6' 8 ' 1S All the studies cited above odontic treatment (Tl); however, most were en-
are retrospective. It is not surprising, therefore, rolled after the start of orthodontic treatment,
that the practitioner is confused because con- largely because of the timing of the orthodon-
clusions based on case reports, case series, or tist's referral.
retrospective designs are often in conflict re- The soft tissue dissection for the sagittal split
garding the most basic outcomes of routine pro- was completed according to Bell et al.17 Splitting
cedures.14 of the mandible was achieved by a horizontal
In 1990, the National Institutes of Health corticotomy on the lingual surface of the ramus
funded a large-scale, multicenter, randomized above the mandibular foramen and a vertical
clinical trial comparing rigid fixation with wire corticotomy on the buccal surface near the sec-
fixation. The results of this trial have now clari- ond molar. A connecting corticotomy was ex-
fied many of these controversies. In an earlier tended between the previous two. Splitting was
publication,15 we compared relapse in patients accomplished by using small osteotomes and
receiving rigid or wire fixation, two years after gently prying. This was performed bilaterally.
surgery, by using the Johnston analysis. We Suspension was achieved by using both circum-
found that skeletal changes were essentially sta- ferential mandibular and intraosseous maxillary
ble in the rigid fixation group, whereas the wire 24-gauge wires. Before application of fixation,
fixation group exhibited a 26% average mandib- an acrylic splint was wired to the maxillary teeth,
ular symphysis relapse. Both groups experienced and the mandible was placed in its predeter-
dental changes. In another study,16 we examined mined occlusion and stabilized with maxillo-
skeletal changes between the two groups from a mandibular fixation (MMF). Patients random-
cranial base reference. At two years after surgery, ized to the rigid fixation group received three
the mandibular symphysis position was un- 2-mm bicortical position screws bilaterally and
changed in the rigid group, but relapsed 30% in used elastics for 5 to 7 days. Patients randomized
the wire group. The purpose of the present in- to wire fixation received bilateral inferior border
vestigation was to compare the skeletal and den- wires and were placed in skeletal MMF for 6
tal changes between rigid and wire fixation for weeks. All patients received postoperative phys-
mandibular advancement by using conventional iotherapy instruction after an established exer-
cephalometrics. cise protocol. During surgery, the surgeon re-
tained the right to change the method of
fixation if the situation demanded a change
Patients and Methods
from the assigned study protocol. All protocol
This study was a prospective, randomized clini- violations were followed-up and described.
cal trial that began in 1990 and was conducted at Lateral cephalometric radiographs were
The University of Texas Health Science Center obtained using a Wehmer calibration ruler
at San Antonio, Texas, Emory University in At- mounted on the cephalostat. Radiographs ob-
lanta, Georgia, and the University of Florida in tained immediately after surgery (T3) were the
Gainesville, Florida. Details of this study have only radiographs taken with an interocclusal
been published elsewhere.15' 16 In brief, subjects acrylic splint in place. The clearest radiograph
who satisfied all inclusion and exclusion criteria between T2-T4 was selected, and the details of
read and signed an informed consent form. Sub- the cranial base structures were drawn. An X-Y
jects were stratified on degree of neuroticism, cranial base coordinate system was constructed
presurgical over]et (<7 mm or >7 mm), and through Sella with the X-axis drawn parallel to
<<
Article
>> Home | TOC |
Index
Frankfort Horizontal. The Y-axis was drawn pass- was modeled after that used by Houston19 and
ing through Sella, perpendicular to the X-axis. has been reported.18 Between group differences
In addition, templates were constructed for the at each time point were examined using un-
maxilla, mandible, mandibular symphysis, and paired t tests; within group differences con-
the maxillary and mandibular incisors. Details trasted the change at surgery (T3-T2) with
on template construction for the cranial base, change at each of the succeeding intervals using
the maxilla, and the mandibular tooth-bearing paired t tests. All statistical tests were performed
segment have been previously described.18 Reg- at a = 0.05.
istration crosses were drawn on the templates.
The dental and skeletal templates were oriented
independently on each radiograph by best-fit
Results
visual approximation, and the registration There were 127 subjects who received a BSSO.
crosses of each structure transferred to a sepa- Of these, 29 subjects were treated at Emory Uni-
rate acetate sheet for each film. The X and Y versity, 22 at The University of Florida, and 77 at
coordinates for the landmarks were registered The University of Texas Health Science Center
by digitization on a Hipad TM 1200 series Digi- at San Antonio. Our earlier reports followed
tizer (Houston Instruments, Houston, TX), in- intent-to-treat protocol. In this report we analyze
terfaced with a 386 computer system (KD Tech- relapse based on treatment received. Among the
nologies, Gainesville, FL) using the software study subjects, 49 received wire fixation, whereas
program Ceph-Master (Trilobyte Software, Ann 78 received rigid fixation. Differences between
Arbor, MI). The skeletal and dental landmarks fixation groups with respect to surgical changes
are shown in Figure 1. All tracings were digitized are shown in Table 1. Both fixation groups
twice, and means of the two measurements were showed similar surgical change in all the mea-
used. sured variables.
Linear and angular measurements were used The surgical change (T3-T2) in overbite and
to compare skeletal and dental changes over the overjet was similar in both groups (Table 1).
following time intervals: surgical change (T3 mi- Changes in overbite and overjet represent stabil-
nus T2), eight-week change (T4-T3), six-month ity of the occlusion. Immediately after surgery
change (T5-T3), one-year change (T6-T3), and (T3) (Table 2), the overbite was greater in the
two-years change (T7-T3). The analysis of errors 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
Figure 1. Measuring points and reference lines used the wire fixation group. Analysis of variance
in the cephalometric analysis. (ANOVA) showed postsurgical changes oc-
<<
Article
>> Home | TOC |
Index
38 Dolce et al
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 mandibular and maxillary incisor, respectively.
shown in Table 4, the wire fixation subjects ex- In general, the position of the mandibular in-
perienced sagittal relapse as early as eight weeks cisal edges (Figure 3) for both groups mimics
after surgery (T4). At this time point, B-point that of the B-point, although not at the same
had moved posteriorly 0.95 mm or 18% (Figure rate. Unlike the B-point, there was no difference
2). This horizontal relapse continued, albeit not between groups in the horizontal position of the
at the same rate, and at two years after surgery mandibular incisor (P < .07, ANOVA). There
(T7), the B-point had moved 1.55 mm posteri- were within group differences. At the eight-week
orly. This was equivalent to a 29% sagittal re- period (T4), the sagittal position of the incisal
lapse. The rigid fixation subjects experienced a edge in the wire group moved only half as much
transient anterior movement in B-point (Table 4 as the B-point (0.34 mm). Two years after sur-
and Figure 2). The peak of this anterior move- gery (T7), however, the position of the mandib-
ment (9%) occurred at six months after surgery ular incisal edge had moved posteriorly (1.37
(T5). At two years after surgery (T7), B-point was mm), nearly the same amount as point B-point
unchanged from its immediate postsurgical po- (1.55 mm). This sagittal change was significantly
sition. The BSSO (T3-T2) produced an increase different (P < .01) than after surgery (T3) at all
in the vertical dimension in both wire and rigid time points, except at the eight-week period.
fixation groups (3.52 mm and 3.04 mm, respec- Also, in the rigid fixation group, the incisal edge
tively). The vertical changes in the B-point are initially moved anteriorly, and by two years after
shown in Table 5. Both groups experienced a surgery (T7), its sagittal position placed it 0.5
loss of vertical dimension (the superior move- mm more posteriorly than B-point. In the rigid
ment of the B-point) (P< .01,ANOVA) between group, the sagittal position was significantly dif-
T3 and T7. Although the values of the rigid ferent from after surgery at the eight-week and
group were higher than the wire group, the two-year time points.
values were not statistically significant between The vertical position of the mandibular inci-
the two groups. sor also followed the position of its respective
Figures 3 and 4 depict the incisal edge posi- B-point. There were no between groups differ-
tions, in both sagittal and vertical planes, of the ences (P = .66). Within group differences (P <
40 Dolce et al
on the treatment received. When the data was Figure 4. Horizontal and vertical position of the max-
analyzed as intent-to-treat, similar conclusions illary incisor. The area delineated around the time
were drawn. points includes the 95% CI in both the sagittal and
The skeletal and dental surgical movements horizontal direction. T3 (immediate postsurgery was
were similar in both groups (Table 1). Skeletal set at zero for both groups) = immediate postsurgery;
T4 = 8 weeks; T5 — 6 months; T6 = 1 year; and T7 =
stability of the mandible was assessed by follow- 2 years. To find the appropriate CI value, extrapolate
ing changes in B-point in both the sagittal and either horizontally or vertically to the nearest edge of
vertical position. The immediate surgical change the CI area. The CIs are not drawn symmetrically
resulted in B-point moving inferiorly 3.5 mm for around the plotted group means.
the rigid group and 3.0 mm for the wire group.
After surgery, in both groups, B-point moved moved at 6 weeks with splints and suspension
superiorly to approach the presurgical vertical wires removed at 7 weeks. That is one reason why
dimension. This was expected in both groups. both groups had radiographs at 8 weeks rather
The rigid group had the splint removed at two than the traditional 6 weeks after surgery. An
weeks, whereas the wire group had MMF re- 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.
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
>> Home | TOC |
Index
Table 7. Changes in Maxillary Incisor Angulation group partially (0.28 mm) followed the mandib-
(Ul-SN) ular incisors, there is an increase in overjet with
P time. In the wire fixation group, Watzke et al
Time Interval Rigid Wire (R v W) reported a -0.6-mm change in overjet; a value
8 wks (T4-T3) -1.65* (2.69) 0.59 (2.33) .088 that is about 1 mm less than what we report. In
6 mos (T5-T3) -2.17* (3.98) 0.04 (3.62) .074 the rigid fixation group, the change of the man-
1 yr (T6-T3) -1.58* (4.39) -0.14(3.95) .161
2 yrs (T7-T3) -1.04* (4.81) 0.60 (4.40) .124 dibular incisor was 0.1 mm; a value that is similar
to our findings. At the early time points, the
NOTE. ( — ) represents retroclination. ( + ) represents procli-
nation. All values expressed as mean (standard deviation). mandibular and maxillary incisors of the rigid
*Significantly different within group from postsurgery (T3) group also followed B-point and move anteri-
(P < .01).
orly. At the two-year time point, however, the
mandibular incisors are further posterior than
B-point. Because the maxillary incisors again
Thereafter, relapse continued, but at a much
partially follow the mandibular incisors, by two
slower rate. Finally, by two years, the sagittal
relapse had progressed to 28%. In the rigid years an increase in overjet occurs. Although the
group, at the two-year time point, B-point was final overjet is similar in both groups, the man-
stable or unchanged from its postsurgical posi- ner in which it occurs in the two groups of
tion, even though B-point did move anteriorly at patients is different. In the rigid fixation group,
the earlier time points. The results of this study it is produced from dental changes, whereas in
mimic what we reported in previous reports in the wire group, it is the result of skeletal relapse.
which relapse was examined using different A recently published study that compared rigid
analyses15 or points of reference.16 Our study, as fixation with wire fixation also found increases
well as those of other investigators,10 support the in overjet after surgery with no differences be-
idea that the skeletal stability is much greater tween the two groups.20
where rigid fixation is used. On the contrary, The results from this study have implications
Watzke et al21 reported no difference between for any orthodontist preparing his or her patient
rigid and wire fixation one year after surgery. for surgery. If the orthodontist is to maintain a
Yet, another study reported relapse with either stable occlusal result, then he or she may choose
type of fixation.20 Most, if not all studies, that to set up mandibular advancement cases differ-
have examined skeletal relapse, have been ret- ently, depending on the type of fixation to be
rospective. Because of the many limitations of used. For example, if rigid fixation will be used,
retrospective studies, their results should often the maxillary incisors may be left slightly pro-
be interpreted cautiously. clined or with some space distal to the lateral
Given the skeletal relapse in the wire group, it incisors. This space could then be used to ac-
is surprising to find no difference in overbite commodate any changes in mandibular incisor
and over]et between these two groups. We spec- position. In the rigid fixation group, during
ulate that in both groups the dental changes
treatment, the orthodontist should try not to
cannot compensate for the skeletal changes.
adjust the occlusion immediately after surgery to
The dental changes that take place after surgery,
accommodate the skeletal changes. For a patient
have been attributed to orthodontic detailing of
the occlusion and/or to compensatory tooth who is to receive wire fixation, the mandibular
movement caused by skeletal changes. Changes incisor may be placed in a retroclined position
in the sagittal and vertical position of the inci- before surgery. This may require extraction of
sors in this study can only be compared with the mandibular premolars. Then, as skeletal relapse
study by Watzke et al because other studies used occurred, the overjet could be maintained by
different selection criteria. In this study, we the proclination of the incisors. The previously
found changes in both the maxillary and man- mentioned study could also be used to deter-
dibular incisors with both fixation groups. The mine retention schemes. For example, perma-
sagittal and vertical position of the mandibular nent retention for the mandibular incisors could
incisor of the wire group followed B-point very be used with rigid fixation to prevent them from
closely. Because the maxillary incisor of the wire moving posteriorly.
<<
Article
>> Home | TOC |
Index
42 Dolce et al
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- patient. If the patient has spacing, you must
able interest in orthodontic treatment for decide if you are going to simply close space or
adult patients. A survey conducted by the AAO work with a restorative dentist to move teeth for
showed an increase in the percentage of patients composites or veneers. If the patient has crowd-
over 21 years of age from 4% 10 years ago to ing, you must decide if you are going to strip,
almost 7% today. This is expected to grow to expand, use distalization mechanics, or extract.
nearly 11% after another decade.1 Innovations This information must be put onto 3 treatment-
in appliances have caught the attention of many planning forms that are sent to Align Technol-
adults who would not otherwise seek orthodon- ogy. You can also treatment plan on the Web by
tic treatment.2 The concept behind using a se- using the Web site and Web treatment planning
ries of customized removable appliances to treat in your favorite Web browser. Align Technology
a malocclusion, first appeared in orthodontic also requires you to take polyvinylsiloxane (PVS)
literature in 1945 with an article by Kesling.3 The impressions4 and to send a copy of the pan-
purposes of this article are to present an over- oramic radiograph or full-mouth series of in-
view of the Invisalign System and then present a traoral radiographs and cephalograms when ap-
case treated with Invisalign. propriate to Align Technology.5
The basic steps in how the orthodontist inter- Align Technology receives the case submis-
acts with the Invisalign manufacturing process sion, and the case is checked for all necessary
are shown in Figure 1. information including PVS impressions, photo-
graphs, radiographs, bite, and treatment-plan-
Records, Polyvinylsiloxane Impressions, and ning forms. Once it is determined to be com-
Treatment Planning Forms plete, the impressions are then poured up in
dental die stone. The records are checked by a
As with all orthodontics, diagnosis and treat-
clinical team to make sure that the case is suit-
ment planning are critical. The clinician must
able for treatment and that the treatment plan is
have clear ideas about what will be done for the
clinically viable. If direct communication is re-
quired to clarify treatment goals, this will be
From Align Technology, Santa Clara, CA. done at this time.
Address correspondence to Ross J. Miller, DDS, MS, 881 Martin
Avenue, Santa Clara, CA 95050. Scanning, Cutting, and Treating
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0801-0008$35.00/0 The models are scanned with white light us-
doi: 10.1053/sodo. 2002.28182 ing the bite as a reference. Once this informa-
vealed a brachycephalic pattern and straight left molars and canines were in a Class I relation-
profile (Fig 2). His lips were competent, and ship (Fig 3). There was 7 mm of overjet and 5
there was no strain on closure. When smiling, mm of overbite with incisor contact. The maxil-
the patient did not show all of his incisors. In- lary and mandibular midlines were centered to
traoral photographs showed that the right and the facial midline. There was excess upper space
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
Figure 5. Progress intraoral photograph (A) and clin-
planned intrusion. We concluded that because check image (B).
intrusive forces were placed on the lower inci-
sors, there was not enough counter force in the
months. A total of 20 plus 4 finishing aligners
posterior region to retain the appliances. There-
were required in the upper and 20 plus 3 align-
fore, attachments were bonded on the lower
ers in the lower to complete the treatment.
canines and premolars to provide retention for
the aligners in the posterior as anterior intrusive
Treatment Results
forces were being delivered. The first 4 sets of
aligners were redelivered, and intrusion was suc- Posttreatment facial photos show minimal
cessfully accomplished. Figure 5 shows an in- change in the facial profile (Fig 6). The initial
traoral photograph and corresponding clin- Class I molar and canine relationships were
check image of intrusion of the lower central maintained with complete upper space closure
incisors as compared to the lateral incisors. (Fig 7). The overjet was reduced from 7 mm to
Treatment continued with 2-week intervals until 4 mm. The upper midline was slightly shifted to
all aligners had been delivered. At this point, the left because of the fact that there was more
there were residual spaces remaining in the up- space to be closed on the left than on the right.
per arch that needed to be closed, and approx- The upper and lower arches showed acceptable
imately 1 mm more intrusion of the lower lateral alignment. The curve of Spee was decreased,
incisors and left central incisor was required. resulting in a reduction of the overbite. The
New PVS impressions and finishing aligners posttreatment panoramic radiograph shows
were made including attachments bonded on good root approximation (Fig 8). There appears
the lower canines and lower right central inci- to be a widened periodontal ligament around
sor. Incisal plasty was performed on all laterals. the upper right lateral incisor, which was asymp-
The patient was then placed in retention by tomatic. The posttreatment lateral cephalogram
using upper and lower clear retainers to be worn shows reduction of the overjet with retraction
at nighttime. The total treatment time was 14 and retroclination of the upper incisors (Fig 9).
<<
Article
>> Home | TOC |
Index
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
>> Home | TOC |
Index
J J
10
12
13
14
16
17
li
Figure 11. Pre treatment (solid line)/posttreatment 19
(dotted line) maxillary and mandibular superimpo- 20
sitions. 21
22
objectives were to maintain the buccal segments, Figure 13. Planned tooth movements for the upper
it was decided by the orthodontist to limit move- arch.
ments to anterior teeth only. Therefore in Clin-
check, only the upper and lower canines and arch treatment was initiated with intrusion to
incisors were sequentially moved. Figures 12 and achieve clearance for retraction of the upper
13 show a schematic representation of the tooth incisors. Therefore, upper arch treatment was
movements of the lower and upper arches, re- delayed by 5 stages. Because of the amount of
spectively. Each column represents a tooth, and lower-incisor intrusion that was needed, it was
each row represents a stage with its correspond- decided to begin initially with intrusion of the
ing aligners. The black lines span the stages in lower central incisors and then proceed to the
which a tooth moves. It can be seen that lower 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
3(2?)| 2 C 2 6 ) | 1(25)j 1 ( 2 4 ) f 2 ( 2 3 ) I 3(221 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
Figure 12. Planned tooth movements for the lower finishing aligners were needed to satisfactorily
arch. complete the treatment. The Invisalign System
<<
Article
>> Home | TOC |
Index
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
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.
W.B. SAUNDERS
Elsevier Science
EHS 01 DM166»« * Elm*» SCTCHC, ZWä
<<
Article
>>
Seminars in Orthodontics Home | TOC |
Index
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 The appearance of the code at the bottom of the first page
quarterly by W.B. Saunders. Months of issue are March, June, of an article in this journal indicates the copyright owner's
September, and December. Corporate and Editorial Offices: consent that copies of the article may be made for personal or
The Curtis Center, Independence Square West, Philadelphia, internal use, or for the personal or internal use of specific clients,
PA 19106-3399. Accounting and Circulation Offices: 6277 Sea for those registered with the Copyright Clearance Center, Inc.
Harbor Drive, Orlando, FL 32887-4800. POSTMASTER: Send (222 Rosewood Drive, Danvers, MA 01923; (508) 750-8400;
change of address to: Seminars in Orthodontics, W.B. Saunders, www.copyright.com). This consent is given on the condition that
Customer Service, 6277 Sea Harbor Drive, Orlando, FL 32887- the copier pay the stated per-copy fee for that article through the
4800. 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
Editorial correspondence should be addressed to: as copying for general distribution, for advertising or promotional
P. Lionel Sadowsky, DMD, BDS, DipOrth, MDent, Editor, purposes, for creating new collective works, or for resale.
Seminars in Orthodontics, Professor and Chairman, Department Absence of the code indicates that the material may not be
of Orthodontics, University of Alabama, 1919 Seventh Avenue processed through the Copyright Clearance Center, Inc.
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. Reprint inquiries should be addressed to Ginny
Saunders, Periodicals Department, P.O. Box 628239, Orlando, Nicholls, Elsevier Science, The Curtis Center, Independence
FL 32862-8239 or e-mail hhspcs@harcourt.com. Square West, Philadelphia, PA 19106-3399. Telephone (215)
Change of address notices, including both the old and new 238-5534, fax (215) 238-6423; e-mail: gnicholls@elsevier.com.
addresses of the subscriber and the mailing label, should be
sent at least 1 month in advance. Customer Service: 1-800-654-
2452 Advertising representative: MJ. Mrvica Associates, Inc,
2 West Taunton Ave, Berlin, NJ 08009. Telephone (609) 768-
9360. Fax (609) 753-0064.
Yearly subscription rates: United States and possessions: Publication of an advertisement in Seminars in Orthodontics
individual, $133.00; institution, $164.00; student and resident, does not imply endorsement of its claims by the Editor (s) or
$67.00; single issue, $50.00. All other countries: individual Publisher of the journal.
$166.00; institution, $198.00; student and resident, $83.00; The contributors have checked generic and trade names and
single issue, $50.00. For all areas outside the United States and verified drug doses for accuracy according to the standards
possessions, there is no additional charge for surface delivery. accepted at the time of publication. The ultimate
For air mail delivery, add $16.00. To receive student/resident
rate, orders must be accompanied by name of affiliated responsibility, however, lies with the prescribing physician.
institution, date of term, and the signature of program/residency Please convey any errors to the Editor.
coordinator on institution letterhead. Orders will be billed at
individual rate until proof of status is received. The ideas and opinions expressed in Seminars in
Prices are subject to change without notice. Current prices Orthodontics do not necessarily reflect those of the Editor or the
are in effect for back volumes and back issues. Single issues, Publisher. Publication of an advertisement or other product
both current and back, exist in limited quantities and are mention in Seminars in Orthodontics should not be construed as
offered for sale subject to availability. Back issues sold in an endorsement of the product or the manufacturer's claims.
conjunction with a subscription are on a prorated basis. Checks Readers are encouraged to contact the manufacturer with any
should be made payable to W.B. Saunders and sent to Seminars questions about the features or limitations of the products
in Orthodontics, W.B. Saunders, Periodicals Department, 6277 mentioned. Neither the Editor or Publisher assume any
Sea Harbor Drive, Orlando, FL 32887-4800. responsibility for any injury and/or damage to persons or
property arising out of or related to any use of the material
Copyright 2002, Elsevier Science (USA). All rights contained in this periodical. The reader is advised to check the
reserved. No part of this publication may be reproduced or appropriate medical literature and the product information
transmitted in any form or by any means, electronic or currently provided by the manufacturer of each drug to be
mechanical, including photocopy, recording, or any information administered to verify the dosage, the method and duration of
storage and retrieval system, without permission in writing administration or contraindications. It is the responsibility of
from the Publisher. Printed in the United States of America. 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.
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, Seminars in Orthodontics is indexed in the Cumulative
6277 Sea Harbor Drive, Orlando, FL 32887-4800. Telephone Index to Nursing and Allied Health Literature® print index
number: 1-407-345-2500. and the Cinahl® database.
W.B. SAUNDERS
<<
Article
>> Home | TOC |
Index
Seminars in Orthodontics
EDITOR
P. Lionel Sadowsky, DMD, BDS, DipOrth, MDent
EDITORIAL BOARD
Richard G. Alexander, Arlington, TX James A. McNamara, Jr, Ann Arbor, MI
Rolf G. Behrents, Memphis, TN Robert N. Moore, Grand Island, NE
Samir E. Bishara, Iowa City, I A Ravindra Nanda, Farmington, CT
Robert Boyd, DDS, San Francisco, CA Perry M. Opin, Milford, CT
Larry M. Bramble, Cypress, C A Sheldon Peck, Newton, MA
John S. Casko, Iowa City, IA William R. Proffit, Chapel Hill, NC
Harry L. Dougherty, Van Nuys, C A Cyril Sadowsky, Chicago, IL
T.M. Graber, Evanston, IL David M. Sarver, Birmingham, AL
Robert J. Isaacson, Richmond, VA T. Michael Speidel, Minneapolis, MN
Alexander Jacobson, Birmingham, AL William J. Thompson, Bradenton, FL
Lysle E.Johnston, Jr., Ann Arbor, MI James L. Vaden, Cookeville, TN
Gregory J. King, Seattle, WA Robert L. Vanarsdall, Jr., Philadelphia, PA
Vincent G. Kokich, Tacoma, WA Katherine Vig, Columbus, OH
Steven J. Lindauer, Richmond, VA C.B. Preston, Buffalo, NY
INTERNATIONAL
Zeev Abraham, Herzliya, Israel Shinkichi Namura, Tokyo, Japan
W.G. Evans, Johannesburg, South Africa George Skinazi, Paris, France
Roberto Justus, Mexico City, Mexico Björn U. Zachrisson, Oslo, Norway
<<
Article
>> Home | TOC |
Index
Seminars in Orthodontics
VOL 8, NO 2 JUNE 2002
Editorial 51
P. Lionel Sadowsky
Introduction 52
Rose D. Sheats
Seminars in Orthodontics
Future Issues
Recent Issues
Editorial
When selecting topics for the journal, it is clinicians to be able to understand and appreci-
always a temptation to satisfy our desire ate scientific method and its application in the
for those areas that we are most comfortable field of orthodontics and at least to attempt to
with and in clinical orthodontics specifically is- base our diagnosis and treatment on sound re-
sues of how to. We want to improve our clinical peatable evidence. Statistics is a vital tool in this
skills and knowledge and get pearls of wisdom or endeavor and although we as clinicians cannot
a take-home message. Perhaps what is more im- expect to understand all of the intricate details
portant is whether the pearl or the message has of statistical method and application, we should
a scientific basis. When mentioning the subject at least be familiar with statistical applications
of statistics, the responses range from rolling and their significance. We should be wary of
one's eyes to quotes that include Disraeli's those who warn us "not to cloud their conclu-
"There are three kinds of lies—lies, damned lies, sions with the facts."
and statistics,"1 or Andrew Lang's "He uses sta- It is my hope that as you read this issue of
tistics as a drunken man uses lamp-posts—for Seminars in Orthodontics you will reinforce your
support rather than illumination."2 Indeed, sta- ideas about the essential elements of statistics
tistics is an area that is immensely complicated and statistical method and appreciate the excel-
and a subject that most clinically oriented stu- lent efforts of the contributors who have labored
dents do not relish. long and hard to make this complex area more
And yet how do we improve our clinical skills easily digestible. An understanding will help us
and knowledge and on what do we base our appreciate what is scientifically sound, what is
decisions? Whose directions are we to follow? still opinion, and what seems to be effective even
Whose teachings and advice should we heed? It though we do not yet have the data to corrobo-
would seem logical that if the specialty of orth- rate the reasons for the outcomes. We will be
odontics is to advance and improve, then we will more selective in our following unproven or
need to make great efforts to apply the scientific even partially proven advice. We will develop a
method in as many ways as possible. Most assur- healthy skepticism while still appreciating the
edly we provide an excellent service to our pa- contributions of those who as yet have no scien-
tients, and we can all list a number of advantages tific proof.
of orthodontic treatment. However, it is incum-
bent on us as orthodontic clinicians to con- P. Lionel Sadowsky
stantly assess, challenge, and debate diagnostic Editor
and treatment regimens advocated and to seek
validation, as far as is possible, for the informa-
tion provided. References
Hence, it becomes essential for orthodontic 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
Copyright 2002, Elsevier Science (USA). All rights reserved. Book of Quotations. New York: Columbia University
doi:10.1053/sodo.2002.125161 Press, 1996
Seminars in Orthodontics
VOL 8, NO 2 JUNE 2002
Introduction
I accepted the task of guest editing this issue Carolina's prospective study on early treatment
of Seminars in Orthodontics with trepidation. of class II malocclusions, she has been intimately
Dr Sadowsky asked me to assemble a cadre of involved with orthodontic studies and clinical
clinician-scholars who were not only knowledge- trials. Her knowledge and appreciation for clin-
able about statistics and could relate key princi- ical orthodontics make her a natural choice for
ples to clinical orthodontic practice but who contributing to this issue. Dr Phillips' careful
could also present this material in a clinician and detailed discussion in this issue of the im-
friendly manner. It was a daunting challenge. My pact of different considerations that influence
greatest fear was that another dry treatise on sample size calculations will help clarify some of
statistics would emerge from the endeavor, and I the mystery associated with determining sample
immediately sought to identify contributors who size.
could meet the aforementioned expectations. Dr Shane Pankratz (another bona fide statis-
All contributing authors are either orthodon- tician) and I review some basic statistics in two
tists with a measure of expertise in statistics, or articles, each with a different emphasis although
statisticians with a measure of expertise in clini- both address topics that are difficult to separate.
cal orthodontics. The intent of this issue of Sem- If you end up jumping back and forth between
inars in Orthodontics is not to teach you to be- the two articles, direct your complaints to me
come statisticians or to attempt to present a and not to my esteemed colleague.
comprehensive review of statistical concepts. Dr Johnston devotes a second article solely
Rather I encouraged the authors to address se- to regression analysis, and I was thankful the
lected topics that they and I hope will be most "muses moved him" as I had already felt fortu-
beneficial to you as a practicing clinician. nate that he had initially agreed to one contri-
This issue leads off with an article by Dr Lysle bution! I had been unsuccessful in persuading
Johnston who, in his eloquent and inimitable various talented colleagues to write this article,
manner, points us in the proper direction. His and as this issue was progressing, I attended a
discussion of basic statistical concepts is lucid, state society meeting where Dr Johnston was the
relevant, and, dare I say, even humorous at featured speaker. In presenting data on a clini-
times. His first article stresses the opportunities cal question of interest, he slipped in a discus-
to enhance one's clinical practice by gaining a sion of regression analysis so effortlessly that I
familiarity with statistical analysis to critically doubt that the practitioners even knew they were
evaluate clinical research. The issue closes with getting a tutorial on this important statistical
an essay by Dr Sheldon Baumrind that addresses procedure. His skill at this accomplishment so
some of the very same concepts with another impressed me that I prevailed on him at the
spin. His provocative presentation also focuses meeting luncheon to just convert all those slides
on the application of outcomes from patient- into a "quick little paper." Of course, neither he
oriented studies to the individual patient and nor I were fooled by my simplistic request, and
discusses some limitations of statistical analysis you can now appreciate why he answered me
of clinical research. with the "muse" quote above.
In between these two articles you will find Dr Sue McGorray presents an important con-
contributions that elaborate on many of the con- cept in her discussion of the sensitivity and spec-
cepts touched on by Drs Johnston and Baum- ificity of statistical tests. Like her counterpart at
rind. Dr Ceib Phillips tackles the question of the University of North Carolina, she has long-
how large a sample size is adequate. As the sta- standing experience in analyzing data for the
tistical consultant for the University of North University of Florida's randomized clinical trial
Introduction 53
on early treatment of class II malocclusions. It tributor brings a perspective that not only un-
always amazes me how well she can clarify the derscores the many facets of statistical analysis
clinical questions and reduce the mass of data but also may potentially resonate with different
collected in large studies to summary statistics readers. I firmly believe in the teaching concept
and explanations that are actually comprehensi- of "telling it three times" to make sure the point
ble to us clinicians. The concept of sensitivity is made (ie, tell them what you're going to tell
and specificity is important to appreciate, and I them, tell them, and tell them what you told
was pleased she chose to address this issue. them). I think our various essays complement
I suspect I speak for all the authors when I say each other well in this regard. My hope is that
that it was a challenge to decide how to limit the each of you will value the contributions of the
material each of us discussed and how best to authors and will come away with a greater appre-
show it with meaningful clinical examples. As Dr ciation for the role of statistics in your clinical
Baumrind points out, statistics is an enormously practice as you strive to separate orthodontic
complex field. In this issue, introductory statisti- fact from orthodontic fiction.
cal concepts are presented and often repeated
among the articles. I have chosen to accept the Rose D. Sheats, DMD, MPH
overlap in material because I believe each con- Guest Editor
<<
Article
>> Home | TOC |
Index
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- handicapped are the rightful prey of the entre-
fessional journals, to have entered an era preneur. It need not be so.
of evidence-based dentistry. Orthodontics, how- Some phenomena are perfect and invariant.
ever, does not seem to be part of the trend. In Given a temperature in Fahrenheit, there is but
the marketplace of orthodontic ideas, much that one corresponding value on the Celsius scale. If
is bought and sold is little more than wishful you know one, you know the other. In the world
thinking shaped by the biases of referring den- of clinical orthodontics, however, things are not
tists, many of whom have a profound distrust determined by simple functional relationships.
both of orthodontics and of the people who Instead, a given outcome is the result of many
practice it. How can this be? Orthodontics is a factors, most of which are unknown. The result
learned specialty peopled by the best in den- is variation. From patient to patient, a single
tistry. Clearly, something has gone wrong. treatment never produces the same result; when
At least part of the problem stems from the compared, the effects of two or more treatments
difficulty of keeping up with a literature that are never exactly the same. Our problem there-
often can neither be understood nor evaluated fore reduces to a reasoned analysis of this inev-
without reference to the methods of statistical itable variation: is it due to chance or does it
inference. Like it or not, statistics has become signal a real effect—a difference that might be
the second language of dental science. To be of clinical significance? Unfortunately, we must
unfamiliar with its concepts and techniques is make these decisions based on incomplete data,
to be functionally illiterate. Professionals thus 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.
From the Department of Orthodontics and Pediatric Dentistry,
If we could examine an entire population, deci-
School of Dentistry, The University of Michigan, Ann Arbor, MI.
Supported in part by NIDCR grant DE08716. sions would be a matter of simple, albeit infi-
Address correspondence to Lysle E. Johnston, Jr, DDS, PhD, FDS nitely cumbersome, arithmetic. Instead, we must
RCS(E), Department of Orthodontics and Pediatric Dentistry, School make do with a relatively small portion of the
of Dentistry, The University of Michigan, Ann Arbor, MI 48109- population, a representative sample from which
1078.
Copyright 2002, Elsevier Science (USA). All rights reserved.
we can take measurements. We then use the
1073-8746/02/0802-0003$35.00/0 resulting data, not only to describe the outcome
doi:10.1053/sodo.2002.32072 of the study, but also to determine the likelihood
that it reflects some real property of the popu- orthodontic specialist. If treatment is to be ratio-
lation, rather than just random variation. It is nal and evidence based, there is no other way.
this random noise that makes statistical infer- The purpose of this article is to provide a
ence a necessity. survey of the statistical tools and techniques on
Unless vetoed by the physical laws of the uni- which meaningful professional survival depends
verse, practically any result, no matter how im- including (1) descriptive statistics, mean, stan-
pressive, can occur purely by chance. There is a dard deviation, and so on; (2) inferential statis-
small but finite chance that one monkey of many tics, t tests and analysis of variance (ANOVA) to
sitting at infinite computers will type the Gettys- test hypotheses about between- and among-treat-
burg Address. To an American who is familiar with ment mean differences; (3) analysis of associa-
Lincoln's words, this monkey's writings might tion, correlation and regression to examine the
seem an impressive achievement. To the statisti- strength and form of the linear relationship be-
cian, however, it merely means that if you have tween two characteristics; and (4) nonparamet-
enough random events to choose from, an out- ric statistics to examine treatment effects when a
come like this is a certainty. Most of the time, normal (ie, symmetrical), bell-shaped, distribu-
even the smartest monkey's literary output usu- tion cannot be assumed.
ally will be something in the way of "hhs/?jslhpi Perhaps equally important, it is hoped that
v %guwgo nuxouo." Similarly, there is a small, this survey will permit someone to read the lit-
but finite, chance that a fair coin will come up erature without being intimidated by the termi-
heads 100 times in a row. Have we witnessed the nology. In the end, it is probably more impor-
equivalent of a miracle? Perhaps, but a simpler tant to know what a statistic does than how it
and much more likely explanation would be that does it.
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- Descriptive Statistics
tral issue of statistical inference: what is the prob-
The goal of most investigations is to learn about
ability that a given outcome could have hap-
some population (Class II malocclusions, high-
pened purely by chance? angle cases, growth modification treatments,
If the probability is high, we conclude that and so on) in the hope that this knowledge will
nothing much is going on; if the probability is aid in making individual treatment decisions.
small, we say that a statistically significant event Perhaps one of the first things one might want to
has occurred—one that, although possible, do with data from a sample is to describe it.
clearly is not probable. How then do we estimate Short of just listing all the numbers ("Here's the
the probability of an event? This is when statis- way my study came out: 14.3, 17.2, 16.8, . . . ."),
tics comes to the rescue. For our purposes, sta- statistics provides many ways of presenting a
tistics is merely a set of common-sense rules that shorthand description of the numerical out-
permit us to decide whether something out of come of an investigation. First, we can use some
the ordinary has occurred. As such, it is a key sort of pictorial representation, such as a pie
part of the process by which biomedical knowl- chart or a bar graph. Alternatively, we can use a
edge is generated. All should have at least an mathematical description that can also serve as a
intuitive grasp of the subject. first step in the process of statistical inference.
Admittedly, achieving a basic understanding Any measurable characteristic of a population
of the process of statistical inference is not a is called a parameter. In the real world, however,
trivial undertaking. On the other hand, statistics we can never study the presumably infinite mem-
is a good deal more useful and not nearly as bers of an entire population. Instead, we must
difficult, or as dry, as many of the subjects we all make do with a sample in the hope that it will
have had to master during the course of our serve as a representative surrogate. Things that
professional careers. Stated simply, an apprecia- we calculate from this sample are called statis-
tion of the role of statistics in scientific commu- tics, hence the name and scope of the subject
nication and a familiarity with its seemingly ar- and the topic of this article. For purposes of
cane terminology are within the grasp of every description, we are interested in the general
<<
Article
>> Home | TOC |
Index
56 Lysle E. Johnston, Jr
location (central tendency) and scatter (disper- suspect that each of these assertions is untrue,
sion) of the data along the number line. the neutral form of the null hypothesis (same,
Central tendency can be described by calcu- equally rapidly, and so on) emphasizes our will-
lating such well-known statistics/parameters as ingness to suspend judgment and to be guided
the mean (the arithmetic average), the median by data rather than by preconceived notions.
(the middle datum), or the mode (the most To test a null hypothesis, the first step is to
frequent score). Scatter, in turn, can be de- gather a representative sample that can be used
scribed by calculating the range or, more often, to estimate the parameters of interest. The sam-
by an average squared deviation from the mean ple mean (X) is a good estimate of the popula-
(the variance). The square root of the variance is tion mean; the sample standard deviation (S) is
called the standard deviation and is perhaps the a good estimate of the population standard de-
most common descriptor of dispersion. The for- viation. Good, however, does not mean perfect.
mulae for these statistics are beyond the scope of Each population features a certain level of in-
this article. Suffice it to say, they provide a mean- trinsic variability. Depending on sample size, any
ingful description of the data from a sample (or statistic we might calculate will tend to reflect
from an experiment); however, they also serve as this scatter and thus will differ somewhat from
estimates of the mean and standard deviation of the population parameter we are trying to esti-
the population from which the sample was mate. In the present context, good means
drawn. (among other things) unbiased: if we were to
A normally distributed population is defined repeat the study over and over and over, our
completely by its mean and standard deviation; statistics would vary from sample to sample, but
there is nothing else to know. Many effects that their average would tend to approach exactly the
are of clinical significance (eg, growth, cooper- parameter they estimate. (In contrast, a biased
ation, relapse, treatment time, and so on) are estimate would tend in the long run to over-
thought to have a multifactorial origin, which, in shoot or undershoot the parameter it purports
turn, commonly leads to an approximately nor- to estimate.) Therein lies the rub: sampling vari-
mal distribution. If our goal is to learn about ation makes any outcome at least remotely pos-
such a population, the mean and standard devi- sible; reality, however, makes some results so
ation are not only useful descriptors but also the improbable that we are obliged to seek alterna-
starting point in the process of statistical infer- tive explanations.
ence. The process of hypothesis testing therefore
involves estimating the probability of obtaining a
given outcome if the null hypothesis is abso-
Inferential Statistics lutely true. If, under these circumstances, an
outcome is unlikely, we reject (ie, disbelieve)
Tests on Means
our null hypothesis. A simple example of what
One or two populations (Student t tests). When commonly is called the Student t test can serve as
we examine a sample of treatment outcomes, we an example. To test the venerable assertion that
really are not interested in this particular group the mandibular incisor to mandibular plane an-
of patients; their treatment is over. Rather, we gle should be 90°, pretend that you gather a
study them to draw inferences about patients in presumably representative sample of patients.
general. Usually, our goal is to test a hypothesis, Given that we know the sample size (N) and can
which is some sort of guess, conjecture, or rea- calculate an estimate of the intrinsic variability
soned assumption about the mean of one or within the population (commonly, the sample
more populations. These conjectures take the standard deviation, S), it is possible to calculate
form of null hypotheses, which are simple de- a statistic, the standard error (S/N/7V), which
clarative statements (eg, extraction and nonex- describes the variability of means calculated
traction treatment have the same effect on the from such a sample. If the sample mean turns
profile, heavy forces and light forces cause the out to be 100°, we need to calculate the proba-
same amount of root resorption, sliding me- bility of such an outcome if the true population
chanics and sectional mechanics retract canines mean really is 90°. To do so, we standardize the
equally rapidly, and so on). Although we may deviation from hypothesis by expressing it in
<<
Article
>> Home | TOC |
Index
standard errors (10° divided by S/\/N). If our angle should be 130°; others argue that it should
sample's mean differs from the hypothesized be 135°. If we conduct an investigation and fail
mean by more than two or three standard errors to reject the null hypothesis that it equals 130°,
(the exact number is determined with reference we would like to be reasonably certain that we
to a table of percentiles for one of the so-called would have rejected 130° if, in fact, the true
t distributions), the probability of such an event population value is 135°. Ideally, a research pa-
is so low that we are forced to doubt our null per will provide an estimate of the study's
hypothesis. But how low does the probability power—commonly, the size of the deviation
have to be for us to draw such a conclusion? from hypothesis (in millimeters or degrees) that
Purely by convention, the cutoff is usually set a given test has a 90% chance of detecting. In
at less than 1 chance in 20 (significant) or less this context, detection is achieved when a false
than 1 chance in 100 (highly significant), hence null hypothesis is rejected.
the common expressions P < .05 or P < .01, The t test is appropriate to tests of hypotheses
which are used to characterize the strength of about either one population (that the mean
the rejection. On the other hand, if the proba- equals some specified value) or two (that the
bility is higher (commonly, P > .05), we fail to means of two populations are equal). Often,
reject. In effect, we reserve judgement—the hy- however, one may want to test the hypothesis
pothesis may be false (if you measure carefully that three, four, or even more populations have
enough, all null hypotheses are false) but not equal means. It would be possible to use nests to
false enough to be of concern, hence the com- compare sample means two at a time. Unfortu-
mon use of the term fail to reject rather than nately, as the number of populations increase,
accept. In this context, it is important to intro- the number of pair-wise t tests increases even
duce the concept of statistical power. faster. Because each individual test has a certain
The process of statistical inference can result fixed chance of wrongly rejecting a true null
in two kinds of error, both of which flow from hypothesis (the cutoff point chosen by the inves-
unavoidable sampling variation: we can reject a tigator, 1 in 20, 1 in 100, 1 in 1000), a prolifer-
null hypothesis that is true (a type I error) or we ation of tests makes it progressively more likely
can fail to reject a null hypothesis that is false (a that one or more tests will lead to rejection of
type II error). Some studies are so crude that no the null hypothesis purely by accident. The re-
matter how false the null hypothesis, the test sult is an uninterpretable smattering of signifi-
would be unable to detect it. Thus, whenever cant outcomes that are in reality just type I er-
there is a failure to reject (ie, a negative result), rors. Given this problem, it is more appropriate
one must always ask how much of a deviation to choose a method that can test for among-
from the stated hypothesis the test could have means equality in a single test and at a single,
detected. In other words, if some competing fixed level of significance. In the jargon of con-
alternative were true, the null hypothesis would temporary experimental design, the appropriate
be false. Under this circumstance, what is the method is some sort of analysis of variance
probability that the test would have led to the (ANOVA).
correct decision, namely to reject the hypothe- Three or more populations (ANOVA). Within
sis? any given population, there is an intrinsic level
The ability to reject a false null hypothesis is of apparently random variability. Not all edge-
called power and is analogous to the resolving wise patients turn out the same. Not all two-stage
power of a microscope. Power depends on the treatments progress at the same speed. Not all
inherent variability of the populations from Class II patients have small mandibles. Such vari-
which we have drawn our sample, the size of the ability commonly is described by the standard
sample, the level of significance at which the test deviation, which, for the purposes of tests on
was run (ie, the probability that we will reject multiple means, is assumed to be the same in
even if the null hypothesis is true), and the each of the various populations under compari-
discrepancy between the null hypothesis and son. Not only does this intrinsic variation serve
some competing alternative about which we are to ensure that sample means drawn from several
concerned. populations will differ but also to determine by
For example, some say that the interincisal roughly how much they should differ.
<<
Article
>> Home | TOC |
Index
58 Lysle E. Johnston, Jr
ANOVA is used to decide whether three or without justification, either empirical or theo-
more sample means vary more than is likely if retic.
the populations from which they were drawn In any consideration of prediction, two ques-
really do have equal means. Rejection of the null tions should come immediately to mind: (1) is
hypothesis of equal means implies that two or there really a relationship between the things we
more probably differ by more than can be ex- are trying to predict and the diagnostic charac-
plained easily by chance. In this event, it is usual teristics that we use to predict them, and (2) if
to use some sort of supplemental, after-the-fact there is, what is the form of the relationship? For
test to decide which of the means are responsi- answers to these questions, we look to correla-
ble for the significant outcome. Common exam- tion and regression.
ples are the methods of multiple comparison: Correlation. To test whether or not two vari-
the Fisher exact test (least significant differ- ables bear a linear relationship to each other (ie,
ence), Tukey test ("honestly significant differ- whether or not they vary together, either posi-
ence"), Scheffe test, Duncan multiple range test, tively or negatively), the technique of Pearson
Dunnett test, and others. For both ANOVA and product-moment linear correlation is commonly
the various post hoc comparisons, the reasoning used. Given a pair of characteristics suspected of
is much the same as for simple t tests. To a first being related, investigators commonly gather a
approximation, they differ only in terms of the sample and then calculate the correlation coef-
test statistic we calculate from the data and the ficient (r), a dimensionless index of the extent
table of probabilities (ie, the distribution) we to which the two characteristics vary together.
use to look up the likelihood that it came about Although many types of relationship (logarith-
by chance. mic, exponential, and so on) can be modeled, it
In its various forms (randomized block, facto- is usual in orthodontics to test for a significant
rial, repeated measures, Latin squares, and the linear correlation. The resulting correlation co-
like), ANOVA is the centerpiece of modern sta- efficient can range from +1, denoting a perfect
tistical inference and is a backbone of contem- positive relationship, to — 1, characteristic of a
porary experimental design. Not all questions, perfect negative relationship (if one variable is
however, involve differences between or among big, the other is small, and vice versa). Under
means; often we wish to examine the joint vari- the assumption of linearity, r = 0 would signify
ation of two or more characteristics. complete independence.
Obviously, it is unlikely that data obtained
from any population or from any sample that
Analysis of Association
might be drawn from it would show either a
Some treatments relapse, and some are stable. perfect correlation (r = +1 or —1) or complete
Some roots resorb, and some do not. Some pa- independence (r = 0). Instead, the correlation
tients grow well, and some do not. Much of what coefficients that we calculate almost always fall
passes for diagnosis and treatment planning in somewhere in between (0.6 or -0.3 or 0.1) and
orthodontics involves an attempt to account for thus must be subjected to the standard litmus
this variability by generating some sort of predic- test of inferential statistics: could a coefficient of
tion. Commonly, these predictions are based on this size come about easily by chance? As with
a presumed association between various pre- many other statistical tests, the decision is based
treatment dentofacial characteristics and the de- on the size of the deviation from hypothesis
tails of treatment on the one hand and impor- (usually, how much r differs from zero) ex-
tant outcome variables on the other. We assume, pressed in standard errors. Once again, the bor-
for example, that changes in intercanine width derline between chance and reality varies with
or in the angulation of the lower incisors are the size of the sample and the chosen level of
related to subsequent stability. Similarly, force significance (1 in 20, 1 in 100, and so on).
magnitude and the shape of the roots are Correlation, it should be noted, is easily abused
thought to predict subsequent root resorption, and often misinterpreted. Two cautionary notes
the shape of the mandible, and subsequent man- are in order.
dibular growth. The list of common-sense pre- First, it must be emphasized that statistical
dictors is seemingly endless; it is also largely significance is in part a function of sample size.
<<
Article
>> Home | TOC |
Index
Thus, given a large sample, a correlation can be simple linear equation, 10.5 plus half the width
statistically significant (ie, unlikely to have come of the permanent lower incisors. Linear regres-
about by chance) and yet be so small that knowl- sion commonly is the process by which this mod-
edge of the underlying relationship it depicts eling is accomplished.
would be of little or no use clinically. Second, no As with other statistical analyses, the first step
matter how high or how statistically significant is to gather a sample of bivariate data (eg, height
the correlation, one cannot infer that one vari- and weight, mandibular plane angle and change
able causes the other. Linear correlation merely in face height, and so on). Given these data, the
tests whether two characteristics vary together; next step is to infer the best possible prediction
causation must be inferred from our knowledge equation, which is usually, but not always, a lin-
of the biology of the system. The growth of the ear prediction equation. If the dependent vari-
mandible is correlated with the growth of the big able (Y, the characteristic you wish to predict) is
toe. One, however, does not cause the other. plotted on the vertical axis of a scattergram, and
Instead, both are byproducts of the general pro- the independent variable (X, the characteristic
cess of somatic growth. In such a situation, the on which the prediction will be based) is plotted
correlation would be strong but of little practical on the horizontal axis, the methods of regres-
or theoretic significance. sion are used to calculate a best-fit linear predic-
In contrast, there are situations in which the tion equation of the form, Y = A + $X, in which
mere existence of a relationship is all we care A is the Y-intercept, the point at which the line
about. For example, we might wish to examine crosses the ordinate, and B is its slope (rise/run,
the possibility of polygenic inheritance by testing increase in Y for a unit increase in X).
whether the correlation between facial dimen- If there is a significant linear relationship be-
sions of siblings is about 0.5. Usually, however, tween the dependent and independent vari-
our ultimate goal in studying correlation is to ables, the slope of the line will differ significantly
find a relationship (causal or otherwise) that is from zero. The prediction error, the scatter of Y
strong enough to be used clinically. Can we use around the regression line, will be less than the
a characteristic that we can see (the indepen- scatter around the overall mean (our best single
dent variable) to predict some important char- prediction in the absence of an equation). As
acteristic or outcome that we cannot (the depen- has been noted, a relationship can be statistically
dent variable)? significant (ie, unlikely to have occurred by
Some treatments relapse, and some are sta- chance) without being clinically significant. In
ble. Some roots resorb, and some do not. It other words, we always have to look beyond the
would be useful to know in advance which out- initial threshold of statistical significance to de-
come is more likely. Indeed, the process of di- cide whether or not a prediction equation is
agnosis and treatment planning is replete with actually worth using clinically.
references to relationships said to be of predic- Regression and correlation are closely related:
tive significance. "Them that has gets." "I looked one deals with the strength of a linear relationship
at the mother and I looked at the father and I and the other with its form. Indeed, the correla-
knew that she was going to be a bad grower." tion coefficient squared (r2) can be derived from
"He was a hyperdivergent patient with a poor the regression analysis and used as an estimate of
growth pattern." Talk, however, is cheap. A pre- the proportion of the variability in Y that can be
cise characterization of the strength and form of eliminated (accounted for) by using the predic-
the relationship (if any) is usually beyond the tion equation. Be warned, however, that r2 has to
intuition of even the most perceptive of clini- be quite large (eg, 0.5-0.6) before the equation
cians. Once again, statistics can be of assistance. would be of obvious clinical use. In orthodontics, I
Regression. If a linear relationship is signifi- can think of few, if any, prediction schemes for
cant statistically and is strong enough to be of which this criterion is met.
practical use, the next step is to model it math-
ematically in the form of a prediction equation
Nonparametric Statistics
so that it can be used clinically. For example, the
size in millimeters of the unerupted mandibular The statistical tests described thus far are com-
buccal segments can be estimated by way of a monly said to be parametric because their null
<<
Article
>> Home | TOC |
Index
60 Lysle E. Johnston, Jr
hypotheses deal with population parameters: the will be able to predict exactly how a given indi-
mean incisor to mandibular plane angle is 90°, vidual will grow or cooperate or respond to treat-
the mean growth rate is the same in Class I and ment. We can, however, hope to know how sim-
Class II populations, on average three treat- ilar patients would respond by examining the
ments produce the same result, and the correla- average response of progressively more precisely
tion between mandibular plane angle and the defined subpopulations (eg, boys, 8-year-old
pattern of growth is zero, and so forth. Their boys, 8-year-old boys with Class II malocclusions,
ability to generate a valid probability estimate 8-year-old boys with Class II malocclusions and
depends on the truth of a number of assump- long faces, 8-year-old boys with Class II maloc-
tions including normality, equal variance, and clusions and long faces who have small teeth,
ratio-scale measurements. Occasionally, these as- and so on). Please note, however, that once we
sumptions are not warranted, in which case it is run out of meaningful pigeon holes (sex, age,
necessary to use one of the so-called nonpara- malocclusion, face height, tooth size), our best
metric, distribution-free statistics. These x2 tests bet is an estimate of the mean of what is left, the
constitute a comprehensive parallel universe in- average result for a sample of 8-year-old boys
habited by nonparametric versions of most of with Class II malocclusions, long faces, and small
the common parametric analyses. Not only are teeth.
they insensitive to departures from the funda- At this stage, many workers posit clinical rules
mental assumptions of parametric statistical in- of thumb that purport to do better than just
ference, but also they are nearly as efficient. betting on the mean. Unfortunately, it is easy to
That is, for a given sample size, their power (the prove mathematically that, if clinical rules do
ability to detect a false null hypothesis) is almost not contribute additional information (ie, some-
as good as that of their parametric counterparts. thing that would place the patient in a smaller,
When there is doubt about the propriety of a t more exclusive pigeon hole) and if they lead to
test or ANOVA, the various nonparametric tests forecasts that differ from the Class mean, they
are safe, effective alternatives. 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
Envoi
mean of so select a subpopulation that it can be,
It is common to argue that the subject of statis- in effect, an efficient individual prediction. It
tics is artificial and irrelevant because it deals seems to me that this filtering process is a rea-
with large samples, whereas treatment, by defi- sonable goal for contemporary orthodontic clin-
nition, involves one patient at a time. Admit- ical research. An additional cautionary note is in
tedly, this assertion has the ring of common- order.
sense truth. But what are we to conclude? That Statistics, by definition, deals with samples.
nothing can ever be known? That a clinical in- We hope that the samples we read about are bias
vestigation becomes useless academic obfusca- free and thus a good approximation of the pop-
tion when its sample size goes beyond N = 1? ulation we wish to understand. Clinical studies,
The one-patient-at-a-time argument seems a rea- however, are notoriously difficult to conduct.
sonable substitute for careful thought until you For many of the most interesting questions, ran-
realize that the total sample in any given practice dom allocation of treatments is ethically ques-
adds up, one at a time, to thousands. Given tionable, whereas retrospective designs, even
thousands of repetitions, a few percentage those that seem to feature random sampling,
points can make a casino rich and its patrons may be irrevocably tainted by a host of biases.
poor. Why should it be any different in clinical Although the theory of sampling and experi-
orthodontics, an activity that often resembles a mental design is well beyond the scope of this
game of chance against nature? In the end, an article, not to mention the skills of its author, I
estimate of the mean response (or perhaps an would suggest that wary, skeptical clinicians ask a
interval—a confidence interval—that has a cer- few simple questions of the literature they read.
tain likelihood of containing the true mean) is If it is a case report, what are you supposed to
probably the best we can ever do. infer, that all patients will respond similarly, that
For a variety of reasons, we probably never well-trimmed models really look good, that the
<<
Article
>> Home | TOC |
Index
author is a better orthodontist than you are? of scientific investigation, however, the various
Often, the take-home message is not all that statistics listed here are just tools. There are
obvious. If, instead, you are reading a study that many more where these came from. At the risk
purports to be more than just a case report (ie, of overdoing the second language analogy, sta-
one in which N > 1), you always have to question tistical tests are similar to a vocabulary. Words
the sampling procedure. are important, but they do not automatically add
Of the thousands of patients who pass up to the great American novel. Similarly, in
through an orthodontist's practice, how did the science, the basic tools of data analysis are im-
few who figured in the study qualify for inclu- portant; however, the design of the experiment
sion? Often it is because they turned out excep- that generates the data is far more important.
tionally well. And why did the treatment work so Many studies are so poorly conceived that it is
well? Commonly, the author of the study would impossible to analyze the data they generate.
have you believe that his/her methods were re- Indeed, it is trivial projects and faulty experi-
sponsible; however, good results can come about mental design that so often combine to give
because the patients were, for some unknown clinical research and statistics a bad name. Given
reason, uniquely susceptible to the treatment the evangelical fervor of many who contribute to
under investigation. For example, some of the our literature, these defects often seem more a
most popular nonextraction philosophies are strategy than an accident. As noted by C.E. Ay-
just that: more-or-less conventional nonextrac- res, "A little inaccuracy saves a world of explana-
tion treatments that work well when applied to tion."1
nonextraction patients. If you want to go beyond Ultimately, the problem is not "liars who can
the data and apply them to obvious extraction figure" or even researchers who cannot or do
patients (as you often are tacitly encouraged to not, but rather a specialty that does not want to
do), that is your business and the patient's re- be bothered. Perhaps as a result, contemporary
gret. In the end, there may be a better chance of orthodontic progress is more or less limited to
winning the lottery than of being selected to be the materials sciences and is largely driven by
the subject of a case report or to be part of the the proprietary efforts of the manufacturers.
sample of a clinical study. Suffice it to say, faulty Our most interesting questions, however, go well
sampling is just one of a myriad of filters and beyond the choice of brackets, adhesives, and
biases that can affect the outcome of a study and archwires. What is to be the place of early treat-
distort its interpretation. ment, gnathology, extraction, expansion, molar
Statisticians commonly distinguish between distalization, surgery? The list is very long and
consumer's risk and producer's risk. With re- very old.
spect to the orthodontic literature, you, the con- In the next century, orthodontics will be de-
sumer, must be worried that the techniques and fined both by its ability and its determination to
ideas you read about (or listen to on the Chau- answer such questions. An honorable outcome
tauqua circuit of orthodontic continuing educa- will require both the generation of valid data
tion) are oversold and undertested. The produc- and an ability to interpret it properly. Without a
ers, in turn, commonly are just as worried about minimal knowledge of statistics and a respect for
devoting any more time and effort to research its importance in clinical research, the former is
than is needed to convince you of the superiority unlikely and the latter is impossible.
of whatever it is they are selling. They want the
benefit of the doubt; more often than we would Acknowledgment
like to admit, they get it.
The author thanks Dr Charles Kowalski, a real statistician, for
In the end, a passing acquaintance with sta- his many suggestions, both technical and stylistic. All were
tistical inference is a sword and shield for the helpful and all in some form or another made their way into
skeptical (ie, prudent) professional. In the world the final version of this manuscript.
<<
Article
>> Home | TOC |
Index
in nature. Angle's classification of malocclusion of the study to detect differences. There are,
is another example of nominal data, which do however, instances when it is beneficial to con-
not arrange themselves into a hierarchy of order vert continuous data to ordinal or nominal
of severity. Angle's Class III malocclusions may types. The study statistician will advise when it
not necessarily be worse than Angle's Class II, or may be advantageous to "collapse" the data.
indeed, some Angle's Class I malocclusions. Raw data that are continuous can be summa-
The classification of data type is important be- rized by calculating means and standard devia-
cause it is a factor in determining the statistical tions (SDs), statistics that characterize the sam-
tests that are appropriate for analyzing the data. ple and estimate parameters of the larger
Data types themselves are ordered: continuous population from which the sample was drawn.
data provide the most information, followed by Data that are categorical cannot be summarized
ordinal data, and finally by nominal data.3 Some- by these parameters. Be wary of converting cat-
times data of a higher level can be rescaled to a egorical data into numeric scores and then sub-
lower level such as when continuous data are cat- jecting them to arithmetic manipulations to ob-
egorized into ordinal or nominal levels.3 An exam- tain means and SDs. Such calculations may be
ple is to group mandibular plane angle cases into meaningless (eg, when trying to determine the
categories using cutoff values of the continuous mean sex or mean race of a sample). Percent-
measurement in degrees. Categories such as high, ages of observed response categories are useful
normal, or low angle cases may be created by summaries of nominal data.
grouping them according to their mandibular The importance of correctly identifying data
plane angle measured in degrees (Table 2). This type will be apparent from a subsequent article
continuous variable is rescaled to a categorical vari- in this issue. In brief, its significance lies in the
able in which each succeeding category from low potential ability to use parametric statistics if the
to high represents increasing mandibular plane sample can be summarized with parameters
angle. Because of this ordered progression, this is (means and standard deviations, percentages,
an ordinal variable. and so on). Nonparametric statistics are avail-
Higher orders of data can be rescaled to able to analyze data for which the appropriate
lower orders of data, but the reverse is not pos- parameters cannot be calculated or for those
sible. When data are rescaled to a lower order, data that do not meet other requisite assump-
information is lost, which influences the power tions for parametric analysis. These concepts will
be addressed further elsewhere in this issue.
Table 2. Rescaling Continuous Data to Categorical
Data
Continuous Categorical Distributions
MPA < 30° Low angle If you have a collection of data points, begin
30° < MPA < 38° Normal angle your initial analysis by plotting them on a graph
MPA > 38° High angle
to see how they are distributed. Often these
<<
Article
>> Home | TOC |
Index
o
2 o
CD
r\ very researcher who has been involved in a Sample size calculations are made using vari-
l J clinical study has an appreciation of the ous assumptions about the anticipated treat-
effort, cost, and often, inconvenience, to both ment effect or differences between treatments
investigators and study subjects. Even though together with realistic projections concerning
investigators often hope for positive findings, in patient accrual and follow-up. Calculating a sam-
clinical research true negative findings make ple size requires four things: (1) deciding on the
very valuable contributions to clinical practice. design of the study; (2) assessing the availability
However, false negative findings can occur ei- of resources; (3) specifying distribution assump-
ther by chance or because a study is underpow- tions; and (4) perhaps most challengingly, de-
ered, which means too few patients have been fining a clinically relevant effect. Table 1 pro-
studied to allow a clinically meaningful treat- vides a set of questions that can help organize
ment effect or difference to be detected.1 It is the information needed to calculate an appro-
important to realize that designing studies with priate sample size.
inadequate sample sizes may lead to erroneous The statistical methodology for calculating
results and false conclusions and subsequently to sample size has been extensively developed over
inappropriate treatment of patients. Such stud- the years. Although the sample size calculations
ies not only drain existing limited resources but are performed using mathematical methods, the
are also unfair to patients who have participated preparation for the calculation requires both
in them. statistical reasoning and clinical experience. De-
Underpowered studies cannot always be signing Clinical Research3 and Statistics4 are two
avoided. However, careful sample size calcula- introductory textbooks that provide a basic con-
tions can guide researchers as to what can and ceptual overview of sample size and power con-
cannot be accomplished in a study with a finite cepts. In this article, the basic principles under-
amount of resources. In fact, National Institute lying the sample size calculation are reviewed,
of Craniofacial Research guidelines for clinical and examples of sample sizes are provided.
trials2 now mandate the inclusion of such infor- Three kinds of data will be presented with a
mation on all clinical trial applications. discussion of how each kind influences sample
size calculations. The three kinds of data are: (1)
nominal data, information collected in studies
From the Department of Orthodontics, School of Dentistry, Uni- that can be classified by categories such as man/
versity of North Carolina, Chapel Hill, NC. woman or Class I/II/III (Angle's classification);
Supported in part by NIH DE NIH DE 10028 and DE05215. (2) continuous data, information collected in
Address correspondence to Ceib Phillips, MPH, PhD, Depart- studies that can be measured on a continuous
ment of Orthodontics, CB7450 UNC-CH, Chapel Hill, NC 27599.
Copyright 2002, Elsevier Science (USA). All rights reserved. scale, such as degrees (incisor inclination, man-
1073-8746/02/0802-0005$35.00/0 dibular plane angle), millimeters (overjet,
doi:10.1053/sodo.2002.32074 length of mandible), or grams of force (head-
68 Ceib Phillips
Table 1. Key Questions Before Sample Size tributed variable (bell-shaped curve) with a
Calculation mean ju^ and a standard deviation (SD) crl. The
1. What is the primary outcome? Clinicians frequently include mean refers to the average amount of mandib-
multiple outcomes in clinical studies. This creates a ular growth in 1 year and is represented by jUq
dilemma for calculating a sample size (n) because the n
required will vary from outcome to outcome. One or
because we do not know the true amount of
two outcomes should be designated as the primary basis growth. The SD tells us how much variation we
for sample size determinations. Other outcomes can expect in the amount of annual growth
typically are considered for secondary analyses.
2. What's the scale of measurement for the outcome? Outcomes
among all children age 7 to 10 years. The SD is
can be classified as binary or dichotomous (yes/no, represented by crl. Another special term is the
present/absent), categorical (mutually exclusive standard error of the mean (SEM). SEM is a
categories), ordinal (ranked or rated), or continuous
(measures on which arithmetic operations can be
measure of variability similar to the SD. SD indi-
performed). In general, binary and categorical cates the spread of the values for a measurement
measures provide less information and thus lower such as mandibular length changes. SEM tells us
power, requiring larger sample sizes.
3. What's the variability of the outcome? For binary measures,
about the spread of values of the means that
the variability is directly related to the proportion of would occur if we drew all of the possible sam-
positives (yes or present responses). For continuous ples of a given size from the population and
measures, the sample size required increases as the
standard deviation increases.
calculated the mean for each sample. It is im-
4. What is the desired level of significance and power? Typically, portant to understand the concept of means and
level of significance is set at 0.05 or 0.01, whereas power SDs because their estimates are used to calculate
is set at 80% or 90%. Increasing the power or
decreasing the level of significance increases the sample
sample sizes.
size required. If we draw a random sample of size (n x ) from
5. Are there special features in the study design? These include this population of 7- to 10-year-old children and
multigroup comparisons, clustered outcome data (such
as multiple observations of the same subject or measure the mandibular length changes (xj),
multicenter studies), long-term follow-up when dropout we would expect that the mean (average) ofthat
or attrition is expected, and inclusion of covariates sample would be scattered around the true (but
(other measures that are believed to somehow affect
the relationship between the treatment and the unknown) amount of mandibular growth (jLtj),
outcome). For many studies, software packages such as with a standard error given by o-j/Vn^ (Fig 1A).
nQuery Advisor or Epilnfo can be used to make Now assume that we simultaneously draw an-
preliminary sample size determinations. However,
special features often involve nonstandard calculations other random sample of size (n 2 ) from this same
and a statistician should be consulted. population, but this time the children in the
6. What defines a clinically relevant effect (effect size)? This is second sample have been treated with remov-
perhaps the most challenging question. The clinician
must define the clinically relevant effect in terms of the able appliances. If treatment with a removable
primary outcome. What difference expressed in the appliance does affect mandibular growth, then
values of the outcome measure would alter/change/ we would expect that the mandibular length
modify the way patients are treated) ?
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-
gear force, shear bond strength); and (3) time- pends on the variability (cr2) in the mandibular
to-event, information collected in studies that length changes in children treated with remov-
measures how long it takes for a specified event able appliances.
to occur, such as number of months to correct to We assume that both sample means are valid
Class I molars, and what proportion of subjects estimates of what we would have found if we
reach the event in a certain amount of time. could have measured both entire populations of
interest. One population would have consisted
of all untreated Class II children between 7 and
Hypothesis Testing
10 years. The other population would have con-
Clinical research is designed to determine if a sisted of all Class II children 7 to 10 years who
treatment has an effect or if different treatments were treated with a removable appliance. The
produce different outcomes. Suppose we believe greater the difference between x x and x2, the
that removable appliances increase mandibular sample means for the untreated and treated chil-
length. We might assume that in untreated Class dren, the more likely it is that the values of the
II children, age 7 to 10 years, mandibular length removable appliance population are not distrib-
growth during a 1-year period is a normally dis- uted about fa but around some other value
<<
Article
>> Home | TOC |
Index
70 Ceib Phillips
represents the risk the investigator is willing to (eg, P< .05) (Table 2): (1) there is a difference
take of incorrectly rejecting the null hypothesis or effect (true positive), or (2) you were unlucky
(and thus incorrectly concluding there is a dif- and there really is not a difference in the popu-
ference between the two groups). This is known lation, but your samples suggest there is (false
as a type I error (or false positive) (Table 2). positive by chance).
Investigators usually set a at 1% or at 5% (a = If the decision is made to reject the null hy-
.01 or a = .05) to describe the risk they are pothesis, then what? An alternative hypothesis
willing to take of a false-positive error rate. (H x ) becomes the fall-through decision. The al-
When the P value (ie, the probability associ- ternative hypothesis is generally established as
ated with finding a difference in growth of the two-tailed, which means that the direction of the
observed magnitude occurring, just by chance, if difference is not specified, (ie, the change in
there is no real difference in the population) is one group may be either larger than or smaller
greater than a, a result is said to be not statisti- than the other group). In other words, if the
cally significant. The shorthand is P > .05 (if a is means are not the same then they must be dif-
chosen by the investigator to be .05) or P > .01 ferent.
(if a is chosen by the investigator to be .01). When the P value fails to reach statistical sig-
A nonsignificant P value does not imply that nificance (ie, P > a) even though the underly-
the null hypothesis is true, that there is no dif- ing groups are truly different, then a type II
ference between the two treatments, or that the error (false negative) has occurred (Table 2).
two treatments are equivalent. A nonsignificant The probability of committing a type II error is
P value tells you merely that there is inadequate called ß. ß is the level of risk of accepting a false
weight of evidence against the null hypothesis. negative conclusion: the risk that an investigator
Nonsignificant results are inconclusive because is willing to take of declaring there is not a
the default position represented by the null hy- difference when there is one (false negative).
pothesis has been neither confirmed nor re- This is not the same as a, the risk an investigator
jected. is willing to take of declaring there is a differ-
The distinction between failure to reject the ence when there is not one (false positive).
null, which is the correct designation when P is The likelihood (or probability) of avoiding a
greater than a, and accepting the null is impor- false-negative error is termed the statistical
tant. The first implies a lack of evidence on power of the study. Power thus expresses the
which to reject the null, whereas the latter im- probability of detecting a true effect, a true pos-
plies the null is true. Three possibilities exist itive. Adequate power traditionally has been de-
when Pis greater than a (eg, P> .05) (Table 2): fined as 80% or 90% (ie, 0.8 or 0.9) and is equal
(1) there is no difference or no effect (true to (1 - ß). If power = 0.8 or 0.9 then ß - 0.2 or
negative); (2) you were unlucky and there really 0.1. Thus, the risk of committing a type II error
is a difference in the population, but your sam- (ß) is 0.2 or 0.1. (Remember that the risk of
ples do not reflect this population difference committing a type I error is a.)
(false negative by chance); or (3) the observed For specific sample sizes and values for a,
difference is real and clinically important, but power expresses how likely it is that a study will
the sample size was too small to reach statistical detect a difference of a certain magnitude when
significance (false negative because of poor that difference really exists in the population. If
study design). power is 90% then we expect 9 out of 10 trials
If the P value is less than or equal to a (P < (identical studies) to indicate a statistically sig-
a), the result is said to be statistically significant. nificant difference exists when in truth a differ-
This does not prove with 100% certainty that the ence does exist in the population. If the power is
null hypothesis is false nor does it mean that the 10%, then we expect that in only 1 out of 10
difference is clinically important. The null hy- trials would a statistically significant difference
pothesis is simply rejected because there is too occur even though a true difference exists.
much evidence against it— the likelihood of Another way to explain power is to say that
observing such a difference by chance, if there is when the null hypothesis is really false (there is
no effect or no difference, is less than the ac- truly a treatment effect), a study with a power of
ceptable risk level (a) chosen by the investiga- 10% has only a 10% chance of rejecting the null
tor. Two possibilities exist when P is less than a hypothesis and a 90% chance of being inconclu-
<<
Article
>> Home | TOC |
Index
sive. Thus, if there is a true treatment effect, you opinion regarding the effect that would be re-
have only a 10% chance of detecting this differ- quired to change their clinical practice, it may
ence. Your study is underpowered. be prudent to obtain input from a broader base
The specification of the alternative hypothesis than a single investigator or even a single aca-
as Hp fjil =£ JA?, (two-tailed alternative hypothe- demic care center's faculty.5 Detecting small dif-
sis) works in hypothesis testing only if the deci- ferences will generally require more patients
sion of interest is the risk we are willing to take than showing really strong treatment effects that
of accepting a false-positive decision (type I er- result in large differences.
ror). In other words, do we only care about Because it is often difficult to choose the ex-
whether or not we have evidence to reject the act magnitude of treatment effect that is of in-
null hypothesis? If, however, the risk of a false- terest, it is important to plan the study so that
negative decision (type II error) is important there is a high likelihood of rejecting the null
to us or if a sample size or power calculation hypothesis if the minimum treatment effect or
is desired, then the difference of interest must difference that would be of therapeutic impor-
be explicitly stated. The general inequality of tance exists. There are times when a 10% differ-
Hp fJL1 3= jLL2 must be replaced by the form ence is irrelevant; there also are times when a
H x : IjUq — /jL2| = 1.25 mm or whatever value the 10% difference is crucially important. If the rate
investigator considers clinically meaningful. In of apical resorption after orthodontic treatment
other words, the type II error rate or power were only 10% higher than in untreated subjects
cannot be calculated for a two-tailed alternative, with a P value of .06, it is likely that the investi-
which merely says that the two groups are differ- gator would not be disappointed that statistical
ent. significance was not attained. However, a statis-
Calculating the probability (ß) of a type II tically nonsignificant finding that early orth-
error, risk of accepting a false-negative decision, odontic treatment reduced incisor trauma by
requires that the size of the difference between only 10% would be disappointing and inconclu-
the two groups be explicitly stated. For example, sive. The goal of any clinical study should be to
in our study we may decide that a true absolute have sufficient numbers of subjects so that clin-
difference in the average amount of mandibular ically meaningful effects are also statistically sig-
growth in the treated and untreated children nificant.
would be clinically meaningful if the difference
were 1.25 mm (H^ [/ij — jut2| = 1.25). The size of
The Role of Variability
this difference should represent the minimum
effect that would be considered clinically rele- The results of a clinical trial are based on a
vant. limited sample of patients, and therefore the
As the value specified by the alternative hy- observed treatment difference is simply an esti-
pothesis changes (ie, the clinically relevant ef- mate of the true treatment difference. Even a
fect) , the sample size required to detect a spec- well-designed study can only give an idea of the
ified difference and the power of a study to answer sought because of random variation in
detect that difference also changes (if you do the sample studied and the variation that would
not change the a level and the variability in your occur between the samples if several samples
study stays the same). This is discussed further were drawn from the population. Thus, results
later. from a single sample are subject to statistical
uncertainty, which is inversely related to the size
of the sample. In other words, statistical uncer-
Clinically Relevant Effect
tainty decreases as sample size increases. For
The specification of the clinically relevant effect example, you are more likely to have confidence
or difference is based on clinical judgment not in an estimate of the rate of incisor trauma if this
statistical inference. The researcher should spec- estimate is based on a study of 200 rather than
ify beforehand what magnitude of treatment ef- on a study of 20. An estimate based on the larger
fect would be regarded as the minimum effect sample is likely to be more precise than the one
size of interest (ie, the minimum change or based on the smaller sample.
difference that would be clinically meaningful). In a study measuring continuous variables,
Because not all clinicians will share the same 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
(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
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
>> Home | TOC |
Index
Adjustments to Sample Size Calculation tients' primary care health professionals. Orr et al9
found, in a multicenter clinical trial, that the fac-
Sample size calculations represent the sample size
tors most strongly associated with incomplete fol-
per group needed at the end of the study and
low-up could not have been identified before en-
therefore underestimate the number of subjects
rollment. These included changes in marital or
who need to be enrolled if the study requires that
employment status, motives for enrolling in the
patients be evaluated on more than one occasion.
trial, and too little time spent with the study clini-
No matter how carefully a study is planned, some
cian.
patients are likely to be lost between enrollment
Sample size estimates need to be adjusted for
and completion (Fig 2). Patients may be dropped
any expected loss from the study. The number of
because of violations of the inclusion or exclusion
subjects to be enrolled in the study can be cal-
criteria or patient compliance, protocol errors of
culated as the sample size required per group
misdiagnosis or incorrect assignment, or patient
divided by the expected retention rate. For ex-
attrition resulting from loss to follow-up or missed
ample, if the sample size calculation indicates 37
examinations.7
patients per group are needed and if the ex-
Special attention should be paid during the
pected retention rate for the trial is 80% then
development of the protocol to mechanisms that
the number of enrolled subjects should be 47
will minimize loss of patients. The attitude of clinic
(37/.S = 46.25). As might be expected, the
staff toward patient complaints and the vigorous
number of subjects needed for enrollment in-
pursuit of patients who fail to keep study appoint-
creases as the projected retention rate decreases.
ments are important factors in reducing dropout.8
Strategies that have been used include reminding
patients of forthcoming appointments, assisting Post Hoc Analyses of Power
with transportation, minimizing waiting times,
On occasion, well-designed studies with prior
sending newsletters, providing monetary reim-
sample size and power calculations result in sta-
bursement, providing continuity of care, involving
tistically nonsignificant results. The concern of
family members, and maintaining contact with pa-
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
Do not meet study criteria
(n3) 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
Patient does not finish study Patient finishes study size, level of significance, and variability). How-
(n7) (n8; sample size calculation) ever, these calculations can lead to contradictory
evidence for and against the null hypothesis.10
Figure 2. Subsets of patients to be considered when An alternative approach to determining if
planning a study. Calculating the sample size to detect a population values (such as the true but un-
clinically meaningful effect is not the only consideration
in the number of patients to be screened and enrolled known difference) are supported by the data is
in a study. The investigator will also need to estimate the the calculation of confidence intervals. For some
proportion of patients who are likely to meet the study statisticians, this approach has more validity than
inclusion criteria, the proportion who will agree to par- calculating the power after the study has been
ticipate, and the proportion who are likely to complete completed. A 95% confidence interval is the
the study protocol (ie, the retention rate). (Modified
from Phillips C: Design principles in oral and maxillo- range of values that you are 95% certain con-
facial surgery clinical trials. Oral Maxillofac Surg Clin 13; tains the true but unknown value of the differ-
237-243, 2001) ence you are seeking. The interval is bounded by
<<
Article
>> Home | TOC |
Index
76 Ceib Phillips
a lower limit and an upper limit and is deter- symmetrical around zero (eg, —1.5, +1.5) then
mined by rigorous statistical methodology using the investigator has no evidence to support a di-
specific statistics derived from the sample in rectional effect in the population.
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- Conclusions
cause the difference between the treated group A study with negative results but adequate power
and untreated group may equal zero. to detect clinically meaningful differences may
For example, using our earlier example that be a valuable contribution to the literature. A
compared the average annual difference in man- negative study with inadequate power is incon-
dibular growth between 37 treated and 37 un- clusive at best. The question remains: is the sam-
treated children, let's assume you found that the ple size calculation worthwhile? The answer is
mean difference in growth was 1 mm, but the SD yes although sample size calculations are based
was 2.5 mm. The SD of your actual data turned on many assumptions and represent our best
out to be much higher than the value you as- guess. These calculations provide a guide to the
sumed for the sample size calculation. If an un- feasibility and practicality of undertaking a study
paired t test were used to make the comparison and focus attention on the consideration of what
between the treated and untreated groups using clinical effect, if it exists, would impact patient
these sample values (mean difference of 1 mm care.
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 References
mandibular growth. If we calculated the 95%
1. Cohen J: Statistical Power Analysis for the Behavioral
confidence interval for this comparison, we Sciences (ed 2). Hillsdale, NJ, Lawrence Erlbaum, 1988
would have found the interval to be defined by 2. NIDCR Policies and Procedures for Investigator Initi-
the limits of ( — 0.16 mm, +2.16 mm). ated Clinical Trials. Available at: www.nidcr.nih.gov/
The difference in mandibular growth was cal- research/ctp/clinical%5Ftrials.him. Accessed 19 Octo-
culated by subtracting the average amount of ber 2001
3. Hulley SB, Cummings SR: Designing Clinical Research:
growth in untreated children from the average An Epidemiologie Approach. Baltimore, MD, Williams
amount of growth in treated children. Thus, values and Wilkins, 1988
greater than 0 tell you that the treated children 4. Freedman D, Pisani R, Purves R, et al: Statistics (ed 2).
grew more than untreated children. Values less New York, NY, WW Norton, 1991
than zero tell you that the treated children grew 5. Fayers PM, Cuschieri A, Fielding J, et al: Sample size
calculation for clinical trials: The impact of clinician
less than untreated children. A value of zero beliefs. Br J Cancer 82:213-219, 2000
means there is no difference in the growth be- 6. Elashoff JD: nQuery Advisor Version 4.1 User's Guide.
tween treated and untreated children. Although Boston, MA, Statistical Solutions, 2000
we still do not know what the real difference in 7. Phillips C, Tulloch JFC: The randomized clinical trial
growth is, we can say that we are 95% confident (RCT) as a powerful means for understanding treatment
efficacy. Semin Orthod Dentofac Orthoped 1:128-138,
that, in the population, untreated children, on 1995
average, might grow as much as 0.16 mm more 8. Goldman JF, Holcomb R, Perry HM, et al: Can dropout
than treated children (this corresponds to —0.16 and other noncompliance be minimized in a clinical
mm) and that, on average, treated children will trial? Report from the Veterans Administration National
not grow more than 2.16 mm more than untreated Heart, Lung and Blood Institute cooperative study on
antihypertensive therapy: Mild hypertension. Control
children. The confidence interval includes zero, Clin Trials 3:75-89, 1982
but the asymmetry of the negative and positive 9. Orr PR, Blackhurst DW, Hawkins BS: Patient and clinic
boundaries around zero are in the direction of the factors predictive of missed visits and inactive status on a
hypothesized effect (ie, that treated children will multicenter clinical trial. Control Clin Trials 13:40-49, 1992
10. Hoenig JM, Heisey DM: The abuse of power: The per-
grow more than untreated children) and could be
vasive fallacy of power calculations for data analysis. Am
supportive evidence for a similar study with a sam- Statistician 55:19-24, 2001
ple size calculation based on larger variability esti- 11. Berry E, Coustere-Yakir, Grover NB: The significance of
mates. If, however, the confidence interval was non-significance. QJM 91:647-653, 1998
<<
Article
>> Home | TOC |
Index
possible to form conclusions about the entire Central Tendency and Dispersion
population based on specific measurements or
When a group of observations is examined
observations taken from this representative sam-
(humans, rats, teeth, cephalograms, and so on)
ple.
in an effort to answer a question, there is a range
Statistics used to summarize, or describe, the
of responses (ie, data points). A typical first step
characteristics of a sample are called descriptive
statistics. Typical summary statistics include the in understanding the data is to plot these data
mean, standard deviation, and variance of a points on a graph to get a feel for the distribu-
measurement of interest in a sample and will be tion of the data. The plot will probably show a
discussed in greater detail later. In theory, these scattering of the data points (hence, the term
same measures, or summary statistics, could be scatter plot), but some data values may occur
computed from the entire population. When more frequently than others. Central tendency
this is done, the computed population statistics refers to the center of the distribution of data
are often referred to as parameters. Because it is points (the tendency of data from biologic sys-
often impracticable to perform a study on an tems to cluster). This center point is a single
entire population of interest, samples from the value that is used to describe the set of observa-
population are drawn, and sample statistics of- tions without listing the values for every subject
ten form the basis for drawing conclusions con- (or tooth or cephalogram). Frequently, the scat-
cerning entire populations. Such conclusions ter plot, or distribution of the data, may look like
may simply describe the population, but they a bell-shaped curve. In statistics, this distribution
may also make comparisons among populations pattern is called the normal distribution (also
or even among subgroups within a specific pop- called the Gaussian distribution).
ulation. Data that can have an infinite number of
Statistical procedures that are used to make values between any two points (such as millime-
inferences (ie, draw conclusions) about the pop- ters overjet or interincisal angle) are called con-
ulation that the sample represents are called tinuous data because they are measured on a
inferential statistics. Because the sample does continuous scale. For continuous data, the most
not include all members of the population, it is commonly used measure of central tendency is
unlikely that the measurements on the sample the mean. It is calculated by summing the values
would be precisely the same as the measure- of a specific measurement from all observations
ments on the entire population, if it were possi- and dividing by the number of observations (ie,
ble to measure everyone in the population. calculating the average of that measurement for
Thus, the true value of the measurement for the the group).
entire population remains unknown. However, Sometimes data cannot be quantified on a
by using the statistics that describe representa- continuous scale. Examples of such data are sex
tive samples, statisticians can estimate with rea- or race. Data that are not continuous are cate-
sonable certainty the range of values (ie, the gorical because their values are assigned to cat-
interval) that is likely to contain the true mea- egories. The different values that the variable
surement of the population. Statisticians can can have are the levels of the category. For ex-
also make comparisons between or among two ample, the levels of the variable, sex, can be
or more groups to determine if the observed named male and female. Such categorical data
evidence supports the conclusion that a real dif- are called nominal (named) because the levels
ference is present between groups. of the category are named but have no mean-
We encounter the need for inferential statis- ingful order to the levels. Tooth type is another
tics throughout orthodontics. Without such example of a nominal variable. The values that
techniques, how would we decide if one treat- this variable could have would include named
ment renders a different result from another, if levels such as incisors, canines, and premolars.
one adhesive system is superior to another, if There is no order from lowest to highest value in
there is an association between a specific maloc- these tooth types, although there is an anatomic
clusion trait and relapse, or if certain cephalo- order in the mouth.
metric values predispose to good or bad treat- Another form of categorical data exists when
ment outcomes? there is a meaningful order to the levels. An
<<
Article
>> Home | TOC |
Index
example is the apical root resorption (ARR) this level than any other level. Thus, GI equals 3
scoring system of Malmgren et al.3 In this system, is the most frequent score, the mode. Another
the scoring levels of 0 to 4 reflect an increasing way to summarize the GI score of this group of
amount of ARR based on guidelines for visually patients is to find the score of the patient who
judging the amount of root resorption by radio- ranked right in the middle of this group (ie, the
graphic evaluation rather than actually measur- score of the 51st patient). From the table, it can
ing the amount of resorption with a ruler or set be seen that his score was 2 since patients 26
of calipers. Categorical data such as these, which through 55 have a GI = 2. Thus, the median
have a meaningful order, are ordinal (ordered). score is 2. From this example, it is evident that
They are similar to nominal data in that their the median and the mode are not always identi-
levels are also named. The difference between cal. Some statistical tests use the median in their
nominal and ordinal data is that the named computations, which will be discussed in more
levels of ordinal data have a natural order to detail later.
them. Note that the ARR scoring system uses The median is also used as a measure of
numbers as the names of the levels (ie, scores of central tendency for continuous data that are
0, 1,2, and so on), but the numbers do not refer skewed (Fig 1A, B) to minimize the effect of
to a measurement on a continuous scale. Instead extremely large or small values on the estimate
these levels refer to a qualitatively increasing of the center of the data. Data are skewed to the
amount of ARR. Malmgren et al3 could have left (Fig 1A) when extreme data values occur at
named the levels of root resorption none, slight, the low end of the scale. The result is that the
and moderate. Gingival inflammation is scored on mean is lower than the mode (the value under
an ordinal scale (none, mild, moderate, severe). the peak of the curve because this is the value
Orthodontists often classify crowding on a similar that occurs most frequently) or the median
ordinal scale (none, mild, moderate, or severe). (value of middle observation). In the normal
For ordinal data, the median or mode is used to distribution, an example of a symmetric distri-
represent the center of the data because a mean bution rather than a skewed distribution,
cannot be calculated as it can for continuous data. mean = median = mode (Fig 1C). In skewed
The median is the middle observation where half data, the relative positions of the mean, median,
the values are lower than the median and half are and mode depend on whether the distribution is
higher. The mode is the value that occurs most skewed to the left or right (Fig 1 A, B).
frequently in the group observations. Nominal data (data that are recorded by lev-
A hypothetic example using a modified gin- els such as sex or ethnic group) cannot be sum-
gival index score4 may help clarify the difference marized by means, medians, or modes. As dis-
between the median and the mode. Assume cussed later, this type of data is analyzed by
there are 101 patients with gingival index (GI) looking at the number of observations in each
scores as summarized in Table 1. From this ta- level of the variable (eg, number of men and
ble, it is noted that the most common GI score is women and the number of whites, blacks, His-
3 because more of the patients (40) scored at panics, and Asians). The data are summarized by
reporting the proportion or percentage of the
data that are classified in each level. These pro-
Table 1. Hypothetic Example Using GI Scores to
Compare Median and Mode (n = 101 Patients) portions can be analyzed by examining if the
proportions differ between groups to be com-
Gingival Index No. of Cumulative No.
Score Patients of Patients
pared. A common test for this type of analysis is
discussed later.
0 5 5
In addition to identifying a value to express
1 20 25
2 30 55 the central tendency of data (mean, median, or
3 40 95 mode), a summary measure of the amount of
4 6 101
dispersion or spread of data around the center is
NOTE. The mode — 3 because this is the level of GI score useful. For continuous data, the measure of
that occurs most frequently (40 times). The median = 2 spread or dispersion of the data around the
because this is the GI score of the middle patient (the 51st
patient) when the patients are ranked by GI score from low center of the distribution may be expressed in
to high. several ways. A commonly used technique is to
<<
Article
>> Home | TOC |
Index
51
0) O
G"
£
Q)
D"
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 =
Mean = median = mode mode. (C)
present the minimum and maximum values that points. If you did it this way, however, you would
the data span. This is known as the range. This quickly realize that the total of the differences
approach, however, is sensitive to extreme data would be zero because some points would be
values and may not be an accurate reflection of higher than the mean, and some would be lower
the spread of data points. An outlier, or extreme than the mean. To eliminate the effect of the
value on either end of the scale, could lead to an negative differences, each difference is squared,
inaccurate perception of how tightly data other- and an average of the squared differences is
wise cluster together. Another way to describe taken. This is called the variance of the data. The
dispersion is to present interquartile ranges, variance is therefore expressed in squared units.
such as the values for the 25th and 75th percen- If you wanted to get back to the average differ-
tile level, which are not as likely to be influenced ence, or deviation, from the mean expressed in
by the values at the extreme upper and lower the same units as the original measurement, you
end of the spread of data points. would need to take the square root of the vari-
When you use the mean (average of all the ance. This value is called the standard deviation.
data points) to represent the sample, you realize, Thus, the standard deviation is the average
of course, that most data points will be different amount by which each observation point varies
from the mean. The difference between the av- from the average (mean) of the points. Tables 2
erage of all the points and each individual point and 3 summarize these common measures of
is called the deviation of that point from the central tendency and dispersion.
mean (average). If you wanted to know the av-
erage size of the difference of each point from
Parametric Versus Nonparametric
the mean (ie, the average deviation from the
Statistics
mean), one way to determine this difference is
to total all the differences of each observation Inferential statistics are those statistical proce-
from the mean and divide by the number of dures that compare groups to see if the groups
<<
Article
>> Home | TOC |
Index
Median Value of data point that falls in the To summarize ordinal data
middle when data are arranged in To summarize continuous data that are
order; for even number of skewed
continuous data points, average of To minimize the effect of extreme values
the middle two values gives median
Mode The value that occurs most frequently Same as uses for median
when data points are arranged in To determine the most frequently
order occurring value
NOTE. For data that are normally distributed, mean = median = mode.
are significantly different from each other. of data for each variable must be the same in
There are two kinds of inferential statistics: para- each group being compared (the size of the
metric statistics and nonparametric statistics. variance or standard deviation of the variable is
The word parametric refers to the use of param- the same in each of the groups being com-
eters, those calculations of sample means, stan- pared). Common parametric tests include the
dard deviations, and variances that summarize Student t test and analysis of variance (ANOVA)
the sample and provide estimates of the popu- and will be discussed later.
lation parameters that are unknown. Thus, para- Data that are ordinal (eg, mild/moderate/
metric statistics refers to a group of statistical severe, ARR scores of 0,1,2,3,4) cannot be sum-
tests that uses means and a measure of variation marized by parameters (means and standard de-
(standard deviation, variance) to help deter- viations). Hence, such data cannot be analyzed
mine if groups are different from each other. by parametric tests, which rely on means, stan-
The mean and standard deviation completely dard deviations, and variances. Ordinal data are
describe the normal distribution (bell-shaped analyzed by nonparametric procedures. Non-
curve) in mathematical terms. Further clarifica- parametric statistics use the ranks of the data
tion about this important mathematical concept rather than means and standard deviations to
is beyond the scope of this article, but it explains make group comparisons.
why certain conditions regarding the data must Rank refers to the position in a hierarchic
be met before the simplest parametric tests, ordering of the data that each data value holds
based on means and standard deviations, may be (eg, from highest to lowest or vice versa). A
validly used. The following three conditions are familiar ranking of data among dental students
necessary: (1) the data must be continuous is the class rank, often based on grade point
(measured on a continuous scale, eg, millime- average (GPA). Each GPA is associated with a
ters, pounds, degrees), (2) a scatterplot of the rank after the GPAs are ordered from highest to
data must look like a normal distribution (bell- lowest value. The highest GPA is ranked number
shaped curve), and (3) the dispersion or spread 1, the next highest is ranked number 2, and so
Range Minimum and maximum value Wide ranges imply greater variability;
of data tight ranges imply lesser variability
Interquartile range Value of data at the 25th and Less sensitive to extreme high and low
75th percentile values
Variance Measure of the average squared Deviations are squared to eliminate the
deviation of each data point effect deviations that are below the
from the mean value of all mean (ie, negative in value)
the data points
Standard deviation Square root of the variance Variability expressed in same units as
original measurements (not squared)
<<
Article
>> Home | TOC |
Index
forth. The lowest GPA holds the rank equal to not significant, and investigators conclude that
the number of students in the class. their groups are not different from each other
From the ranks, a median value can be ob- (and thus treatment results are not different).
tained. Remember that the median is the middle Convention states that P < .05 is statistically
observation after data have been ordered from significant; however, this level (called the a
low to high or vice versa. Nonparametric tests level) may be changed if the investigator believes
compare medians, rather than means, for anal- a lower or higher probability level is more ap-
ysis. Common nonparametric tests based on propriate for his/her study. It is important, how-
ranks include the Mann-Whitney U test, the Wil- ever, for the investigator to state his/her a level
coxon signed rank test, and the Kruskal-Wallis at the beginning of the study before data analysis
test. These will be discussed later. takes place.
Nonparametric statistical tests are also used Commonly used statistical software packages
for continuous data that are not normally dis- provide the value of the test statistic or critical
tributed (bell-shaped curve). Although a mean ratio as well as the P value associated with it. The
can be calculated for such data, as discussed P value thus tells whether the evidence is strong
earlier, the median may be a more appropriate enough to conclude that there is a statistically
measure of the center of data that are not nor- significant difference between the groups.
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 Common Statistical Tests
groups, nonparametric tests are used.
Parametric Tests
Data that are nominal (eg, sex, tooth type)
cannot be summarized by means or ordered into Remember that to use a parametric test prop-
ranks. By summarizing the numbers of observa- erly, it must be verified that the data are contin-
tions in each level (eg, incisor, canine, premolar, uous, that the distribution of the data in each
and so on) of a category (eg, tooth type), one sample is normal (bell-shaped curve), and that
can determine the ratio or proportion of the the variance (or standard deviation) of the vari-
total number of observations that occur in each able to be examined is the same in each group to
level. Group comparisons of such data are made be compared. Two common parametric tests
by analyzing if the same proportion of tooth that are frequently found in the orthodontic
types (or men/women, Class I/II/III malocclu- literature are the Student t test and ANOVA.
sions) occur in all the groups. The most com- The Student t test is used when only two
mon test to analyze nominal data is the ^2 test, groups are being compared. The Student t test
which will be further detailed later. uses sample means and standard deviations to
calculate the probability or likelihood that the
groups are different. It helps you determine if
Test Statistics the means differ because the two groups repre-
Statistical procedures comparing samples sent two different populations or if the means
usually provide a test statistic or critical ratio that differ because the groups have different subjects
is associated with a probability level (P value). but each group represents the same population.
The probability level, called the P value for Figure 2A shows what might be expected if the
short, is the likelihood or chance that two groups represent two populations (I and II) that
groups, representative of the same population, are distinguished by different population means.
would be chosen, and that there would be a Figure 2B shows how two groups could be rep-
difference in the groups at least as big as the one resentative of the same population, (III) al-
detected. A P value < .05 means there is an though the groups have different values for their
equal or lower than a 5% chance (1 in 20) that group means.
the two groups could be samples from the same The Student t test actually exists in two forms,
population. By convention, when P ^ .05, inves- depending on whether the two groups under
tigators conclude that their groups are not from comparison are paired (matched) or indepen-
the same population and therefore are statisti- dent of each other. The paired t test (also known
cally, significantly different. A P value > .05 is as the matched-group t test), as the name im-
<<
Article
>> Home | TOC |
Index
so sometimes the test is called the F test. ANOVA ans of two groups, rather than the means, to see
makes a series of pair-wise comparisons for all if they are different. The median of each group
the groups in the comparison. For example, if is found by ranking the data in each group from
groups I, II, and III are compared, ANOVA will lowest to highest and identifying the middlemost
compare I to II, I to III, and II to III. A signifi- value (median). The test statistic for the Mann-
cant P value indicates that a difference exists Whitney U test is called the U statistic and, like
somewhere between any two comparisons, but all test statistics, is associated with a P value that
ANOVA does not identify which groups are dif- measures the likelihood that the difference be-
ferent. To determine which pairs differ (I-II, tween the groups could have occurred by chance
I-III, or II-III), any of several subsequent tests alone. The example of grade point averages and
(called post hoc or a posteriori tests) can be class rank may help clarify the use of the Mann-
used to examine the groups in detail and reveal Whitney £7 test.
which groups significantly differ from each What if the question was to determine if there
other. Note that these post-hoc tests are not used was a difference in the academic performance of
if the P value from the ANOVA is not significant. female students and male students? Because
Common post hoc tests are the Tukey-Kramer GPA is a continuous variable, one way to answer
honestly significant difference, Scheffé, Dun- this question would be to use an unpaired t test
nett, Duncan, and Newman-Keuls tests. The to compare the mean GPA of female students
choice of test depends on the specific research with male students. But what if one student had
design, and consultation with a statistician is an extremely low GPA? This would unfairly in-
recommended to identify the most appropriate fluence the mean GPA of that group. Eliminate
post hoc test for the study. 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
Nonparametric Tests
based on their GPA rank in the class is created.
The previous statistical techniques were ex- From these ranks, the median rank for the men
amples of common parametric tests. To use and for the women is determined. Because the
parametric procedures, remember that the con- medians of two independent groups (men and
tinuous data must follow a normal distribution women) are being compared, the Mann-Whit-
(bell-shaped curve) and that parametric tests ney U test is the appropriate nonparametric test
rely on means, standard deviations, and vari- to use. If the result of the analysis by the Mann
ances. Sometimes the continuous data are not Whitney U test returns a P value that is not
normally distributed, or there are ordinal data significant (P> .05), it is concluded that there is
(data that fall into levels that have a meaningful no difference in the median academic ranks
order). Ordinal data cannot be summarized by between men and women. But if this test is
means and standard deviations. In cases such as statistically significant (P < .05), it is concluded
these, parametric statistical tests are not suitable, that the two groups are actually different and
and nonparametric procedures are indicated. that there is a true difference in the academic
They are called nonparametric because they do ranks of men and women. Thus, if the median of
not use population parameters (ie, estimates of the female ranks is significantly lower than the
means, standard deviations, or variances) in median of the male ranks (the lower the rank
their computations. [no. 1], the better the academic performance),
A common nonparametric test for compari- it would be inferred that the female students
son of two unpaired samples is the Mann-Whit- perform better academically than the men. Of
ney Latest also known as the Wilcoxon rank sum course, it could be the other way around (medi-
test (not to be confused with the Wilcoxon an of male ranks lower than median of female
signed rank test, which is described later). It is ranks), and the opposite would be concluded.
similar to the unpaired t test described earlier, The comparable nonparametric test to the
which is a parametric procedure that compares paired t test is the Wilcoxon signed rank test. If
the means of two unpaired groups to see if they there are two matched or paired groups and
are different. The Mann-Whitney £7 test is a non- ordinal or continuous data are being examined,
parametric procedure that compares the medi- this is the appropriate nonparametric procedure
<<
Article
>> Home | TOC |
Index
to use. An example of the appropriate use of this the amount of adhesive remaining on the teeth
test can be found in a recent article comparing after the brackets were removed was made using
the oral hygiene in upper and lower jaws5 using the Adhesive Remnant Index (ARI).7 This index
several indices which are ordinal such as a mod- assigns a score from 1 to 5 to represent decreas-
ified gingival inflammation index (GI). 4 The ing amounts of adhesive remaining on the tooth
modified GI is ordinal because it scores gingival surface. Thus, the ARI is an ordinal scale and
inflammation in five levels ranging from absence must be analyzed by a nonparametric test. Be-
of inflammation to severe inflammation. In the cause an ordinal variable was being compared
study cited, the two comparison groups were among three groups (the three laser energy lev-
upper jaw oral hygiene scores and lower jaw oral els), a nonparametric alternative to ANOVA is
hygiene scores. The groups were paired rather necessary (ie, the Kruskal-Wallis test).
than independent because the upper jaw was Finally, how is nominal data analyzed? Nom-
being compared with the lower jaw in the same inal data cannot be summarized by means be-
patient. Nonparametric tests were required be- cause such data are not continuous. These data
cause the data were ordinal. If the Wilcoxon cannot be analyzed by medians because there is
signed rank test were statistically significant, it no natural order to the levels (ordinal). The
would have been concluded that there was a most common way to analyze nominal data is to
difference in oral hygiene between the upper use the x2 test. This is a very versatile test that has
and lower jaws. If the test were not significant, it many applications, but for the purposes of this
would have been concluded that patients article, discussion is limited only to its use in
cleaned their upper jaws as effectively as their comparing nominal data. It is used to compare
lower jaws. the proportion of the data that fall into each
What if there are more than two groups to level of the nominal variable. For example, we
compare and it is not appropriate to use para- might like to know if the number of Class I, II,
metric procedures such as ANOVA (which com- and III malocclusions is the same in different
pares three or more groups)? The nonparamet- ethnic groups. The levels of Angle class are nom-
ric test comparable to the ANOVA is the Kruskal- inal (Class I, Class II, Class III). The levels of
Wallis procedure. The ANOVA uses means and ethnic group are also nominal (white, black,
variances for its computations. The Kruskal-Wal- Asian, and so on). There is no inherent order to
lis test, like the previous nonparametric proce- the levels of either of these variables so they are
dures described, examines intergroup differ- not ordinal and therefore not amenable to anal-
ences based on ranks. ysis by tests that rely on medians or means. The
An example of the use of the Kruskal-Wallis X^ test compares the proportion of Class I, II,
test can be found in another recent study that and III malocclusions for each ethnic group to
examined three laser energy levels used to bond determine if these proportions differ by ethnic
brackets on extracted teeth.6 An assessment of group. A statistically significant ^2 test indicates
that there is a difference in the number of Class analytic procedures, although not comprehen-
I, II, and III malocclusions among the ethnic sive, may provide a useful review of the statistical
groups. A statistically nonsignificant \2 test indi- tests that one is likely to encounter while facing
cates that the number of Class I, II, and III the challenge of integrating patient-centered
malocclusions is the same among ethnic groups. clinical studies with an individual patient's
Table 4 tabulates the procedures discussed needs.
earlier and lists comparable parametric and non-
parametric procedures and their application. It
References
is by no means comprehensive but may be a
1. BeGole E: Interpreting the literature on the effectiveness
convenient tool to help clarify the indications of clinical procedures. Semin Orthod 6:67-75, 2000
for their use. Specific details of how each of 2. Sackett D, Straus S, Richardson W, et al: Evidence-based
these tests uses sample estimates, distributions, Medicine: How to Practice and Teach EBM (ed 2). New
and proportions will not be addressed in this York, NY: Churchill Livingstone, 2000
article but can be found in any standard statisti- 3. Malmgren O, Goldson L, Hill C, et al: Root resorption
after orthodontic treatment. Am J Orthod 82:487-491,
cal text.
1982
4. Lobene R, Weatherford T, Ross N, et al: A modified
gingival index for use in clinical trials. Clin Prev Dent
Conclusions/Summary 8:3-6, 1986
As our specialty strives to embrace the prac- 5. Thienpont V, Dermaut L, Van Maele G: Comparative
study of 2 electric and 2 manual toothbrushes in patients
tice of evidence-based dentistry to guide our with fixed orthodontic appliances. Am J Orthod Dentofac
clinical decisions, expectations will be raised re- Orthop 120:353-360, 2001
garding the use of appropriate statistical analysis 6. Talbot T, Blankenau R, Zobits M, et al: Effect of argon
when applicable. Dr BeGole's enlightening laser irradiation on shear bond strength of orthodontic
study, which was cited earlier, urges orthodontic brackets: An in vitro study. Am J Orthod Dentofac Or-
thoped 118:274-279, 2000
clinicians and graduate students to become fa- 7. Artun J, Bergland S: Clinical trials with crystal growth
miliar with common statistical methods. This conditioning as an alternative to acid-etch enamel pre-
presentation of statistical concepts and common treatment. Am J Orthod 85:333-340, 1984
<<
Article
>> Home | TOC |
Index
n the formulation and execution of an orth- If so, how much will the mandible be displaced? I
/ odontic treatment plan, the structure of the doubt that anyone really knows the answers to
decision process often is poorly characterized and these questions. Given this uncertainty, the best we
incompletely validated. At each step, therefore, the can do in many instances is to assume that every
clinician is faced with two key questions: does the patient will undergo the mean response as we have
decision rule have a real biological basis, and, if so, come to know it. For example, it is probably fair to
is its transfer function (the exact prediction equa- expect that the mandible will outgrow the maxilla,
tion) adequately characterized? Do high-angle pa- that molars will come forward if premolars are
tients always respond poorly to cervical traction? extracted, and that there will be clockwise mandib-
Or do they do so only occasionally? If there is some ular rotation during treatment. Although we know
sort of relationship, what is the formula by which that all patients do not respond in the same way,
we can estimate an increase in facial height from betting on the mean is probably better than noth-
initial mandibular plane angle? Does the form of ing at all and certainly better than a wild guess.
the mandible tell us how it is going to grow in the Some, however, argue that we must do more. Im-
future? If it does, how are we to combine the plicit in their argument is the assumption that we
various characteristics (symphysis width, condylar can do more.
angle, and so on) to calculate the expected growth Many clinicians think that the mean is, by
increment? Do premolar extractions always or oc- definition, inappropriate to the individual and
casionally lead to distal mandibular displacement? therefore not worth knowing. "I treat my cases
one at a time" is the usual refrain. Much the
From the Department of Orthodontics and Pediatric Dentistry,
same way that athletes, when pressed for some-
School of Dentistry, The University of Michigan, Ann Arbor, ML thing profound to say, deliver themselves of a
Address correspondence to Lysle E. Johnston, Jr., DDS, PhD, FDS sincere but ultimately trivial determination to
RCS(E), Department of Orthodontics and Pediatric Dentistry, School play their remaining games one at a time. But
of Dentistry, The University of Michigan, Ann Arbor, MI 48109-
how are we to render these individualized deci-
1078.
Copyright 2002, Elsevier Science (USA). All rights reserved.
sions and on what basis? As it turns out, theory,
1073-8746/02/0802-0007$35.00/0 intuition, memory, common sense, and the
doi:10.1053/sodo.2002.32189 opinion of helpful experts more often than not
88 Lysle E. Johnston, Jr
are inadequate to the task. If we are to do better, same, relying on expert opinion, and so forth).
we must return to the twin topics of correlation In the present "them that has, gets" example,
and regression introduced earlier in this issue. those who believe in common sense forecasting
A key event in the development of the science assume that there is an obvious and clinically
of biometrics was Sir Francis Galton's research into exploitable relationship between present facial
the inheritance of human physical characteristics. form and the way the face will grow in the future.
When he plotted children's heights as a function Good faces grow well; bad faces grow poorly.
of their parents' heights he found that, although When presented with such an assertion, the wary
the resulting scattergram depicted a strong rela- clinician should seek answers and evidence be-
tionship, it also featured what he termed a regres- fore betting the farm, not to mention the pa-
sion to the mean: children of tall parents also tient's well-being, on orthodontic old wives' tales
tended to be relatively tall but not as much above masquerading as diagnostic pearls.
average as were their parents. The same pattern of
inheritance—a regression to the mean—was seen
Is the Relationship Significant?
in the offspring of subjects who were shorter than
average. As a result of Galton's pioneering re- Many of the assumptions that fuel orthodontic
search, the process of modeling and analyzing the decision making, although popular and occa-
relationship between two characteristics has come sionally even logical, are of dubious or, at the
to be called regression. Although not often recog- very least, unknown significance. As a result, the
nized as such, much of what passes for orthodontic first step in evaluating a clinical rule of thumb is
diagnosis and treatment planning is, at bottom, a to ask whether it is based on a real (ie, statisti-
regression problem. cally significant) relationship between the thing
The common practice of predicting the size of we are trying to predict and the characteristic we
unerupted canines and premolars from the widths want to predict it with. For example, is there a
of the mandibular incisors, the likelihood of root relationship between what the patient has (man-
resorption from the form of the roots, or the pat- dibular plane angle, antegonial notching, face
tern of future growth from the form of the man- height, and the like) and whatever it is we would
dible are obvious examples. The venerable prob- like to know about the face in the future (what
lem of growth prediction—"them that has, gets"— the patient gets)? Correlation provides a means
can serve as a framework for a discussion of the of answering this sort of question.
various aspects of the topic of regression. The possible existence of a clinically useful
Clearly, it would be useful to know how an relationship commonly is explored by drawing a
individual patient is going to grow. Indeed, con- representative bivariate sample—a collection of
trary to the protestations of the one-patient-at-a- elements (in this case, patients) from which
time purists, it would even be useful to know both the thing to be predicted (Y, the depen-
how the average patient grows. If no significant dent variable) and the thing to be used to pre-
predictors can be found, the known changes in dict it (X, the independent variable) can be
facial form that accompany the usual (average, measured. In passing, it should be noted that a
mean, normal, expected) pattern of growth con- predictor variable can take many forms: it can be
stitute a useful alternative to no prediction at all, a single craniofacial characteristic, a change in
or worse, an arbitrary, idiosyncratic stab in the some characteristic measured over a period of
dark. My mother, for example, may not have observation (as with the so-called Tweed growth
known how much my feet were going to grow trends inferred from the pattern of change in
(quite a bit as it turns out); however, she knew A-Na-B), or even a weighted combination of sev-
they were going to get bigger. As a result, she eral variables. Given a potential predictor, it is
always bought me shoes that were a few sizes too necessary to test whether its variation matches
big. In so doing, she made a practical use of her that of the dependent variable we wish to fore-
knowledge of growth. Regression, however, tries cast. In other words, is there a correlation that is
for more: it attempts to render an individualized strong enough to be exploited clinically?
prediction whose error is significantly smaller A correlation between two variables can take on
than that of any alternative (betting on the many forms, including linear, exponential, power,
mean, assuming that all patients will be the and so on. Suffice it to say, a linear relationship is
<<
Article
>> Home | TOC |
Index
Regression 89
15-
What Is the Form of the Relationship?
10-
We know that Fahrenheit and Celsius tempera-
tures are related; however, we need to know the 5-
exact formula if we are to calculate one from the
other. Unfortunately, biological forecasts are dif- 0 —i
2 3 4 5 6 7 10
ferent; they are imperfect. We do not expect to
Independent variable (X)
see an exact mathematical relationship. Rather,
we hope merely to generate a prediction that has Figure 2. Scattergram depicting an apparently signif-
less error than some alternative approach, com- icant positive linear relationship between two charac-
monly the simple expectation that the average teristics, X and Y.
<<
Article
>> Home | TOC |
Index
90 Lysle E. Johnston, Jr
75-
10-
5-
O
3 4 5 6 7 8 10
Independent variable (X)
20 -i Thing to be predicted
20
1
15- 15-
Which line
is best?
10-
5-
0
3 4 5 6 7 10
Predictor variable
Figure 4. Linear regression. Error is measured as the
sum of the squared vertical scatter around the line. If Figure 6. Fitting a line to empirical data. For the
the regression is significant, this scatter will be less regression of Y on X, there are many apparently log-
than the scatter around the mean (see Fig 5). ical alternatives, only one of which is best.
<<
Article
>> Home | TOC |
Index
Regression 91
Screening and diagnostic tests play an impor- may be obtained by more elaborate tests or fur-
tant and ever-expanding role in medicine, ther subject follow-up.
dentistry, and orthodontics in helping us deter- To examine the usefulness of a screening test,
mine if a disease or condition is present. How- we are interested in the relationship between the
ever, the interpretation of test results can be test results and the true disease or condition
problematic. What exactly is a test result telling status. If we know both test result and true dis-
us? How do tests compare? A careful review of ease status for each subject, the data can be
the characteristics associated with test perfor- displayed as in Table 1. In this table, the total
mance will improve their use, understanding, number of subjects for whom we have both test
and interpretation. results and true status is represented by n. Let's
assume in our illustration that n = 100. Each of
Basic Definitions the 100 subjects is classified into one of the four
cross-classification cells of the table based on test
In the simplest case, consider a binary test, result and disease status. The four cells repre-
which is a type of test that has only two possible sent the numbers of subjects with the following
outcomes, such as positive (+) or negative ( — ) . combinations of test result and disease status:
A positive test indicates a high likelihood (but a = the number with test positive, disease
not certainty) that the disease or characteristic is present ( + ); b = the number with test positive,
present, whereas a negative test indicates a low disease absent (-); c = the number with test
likelihood (but not certainty) that the disease or negative, disease present ( + ) , and d = the num-
characteristic is present. Assume that we also ber with test negative, disease absent ( — ) .
know with certainty the true disease status or From this simple table, a surprising number
characteristic of the subjects. This true informa- of test characteristics can be estimated, and each
tion, sometimes known as the gold standard, characteristic shows a particular feature of the
relationship between test result and disease sta-
tus. To interpret properly a test result, it is im-
From the Department of Statistics, University of Florida, Gaines-
ville, FL.
portant to understand these characteristics.
Supported by the National Institutes of Health / National In- To begin, the sensitivity of a test is the prob-
stitute of Dental and Craniofacial Research (grant no. DE08715). ability that the test is positive for those subjects
Address correspondence to Susan P. McGorray, PhD, Depart- who actually have the disease. From Table 1, we
ment of Statistics, Box 100212, University of Florida, Gainesville, estimate sensitivity by the proportion (ratio)
FL 32610-0212.
Copyright 2002, Elsevier Science (USA). All rights reserved. a/(a + c) because a + c represents the total
1073-8746/02/0802-0008$35.00/0 number of patients with the disease (28) and a
doi: 10.1053/sodo. 2002.00000 represents the number in this group who tested
Table 1. General Case and an Example of negative rate is 20/28 (71%), and the false pos-
Classification of Subjects by Disease Status itive rate is 10/72 (14%).
and Test Result
A test with high sensitivity is important when
Test
True Disease Status or Characteristic the cost of missing a person who actually has the
Result Present (+ ) Absent (-) Total disease or condition is high. On the other hand,
high specificity is desirable when the goal is to
Positive
(+) a (8) b (10) a + b (18) rule out a disease on the basis of a test result.
Negative The most useful test has both high sensitivity
(-) c (20) d (62) c + d (82) and specificity, although unfortunately in the
Total a + c (28) b + d (72) n (100)
real world this is often not possible.
NOTE. Sample data are listed in parentheses. To estimate sensitivity and specificity of a new
test, one needs to identify subjects whose disease
status is known by some gold standard. Without
positive by the screening test (8). Sometimes this advance knowledge of disease status, the practi-
proportion is represented by the corresponding tioner then performs the new test on these sub-
percentage (proportion multiplied by 100). jects and compares test results to the known
From the example data in Table 1, the sensitivity disease status of the subjects. From these results,
of this test is 8/28 or 29%. This indicates that of the sensitivity and specificity estimates for the
those who truly have the disease, only 29% will new test can be calculated.
have a positive test result. A perfect test will have For example, one might wish to examine
a sensitivity of 100%. The sensitivity is also called methods used to determine temporomandibular
the true positive rate. joint (TMJ) disc displacement.1 We might be
The specificity of a test is the probability that interested in knowing the sensitivity and speci-
the test is negative for those in whom the disease ficity of a simple screening test, which relies on
is absent. From Table 1, this is estimated by clinical evaluation of joint sounds and symptoms
d/(b + d), or the corresponding percentage. to detect the presence of TMJ disc displacement.
The total number of patients without the disease We will assume that the true disease status (pres-
is b + d (72), and b is the number of patients in ence or absence of TMJ disc displacement) can
this group who test negative (62). This is also be ascertained by magnetic resonance imaging
called the true negative rate. A perfect test will (MRI) of the joint. Thus, MRI is considered the
have a specificity of 100%. From the example in gold standard. Subjects for this study would be
Table 1, the specificity estimate is 62/72 or 86%. selected based on their MRI results so that we
By knowing the specificity of this test, we would would know the true TMJ disc status. These sub-
expect that 86% of those who do not have the jects would then be evaluated by using the
disease would have a negative test result. screening test (clinical examination), and the
For this hypothetic example, the sensitivity of sensitivity and specificity of this screening test
our test is 29%, and the specificity is 86%. The would be calculated by using the MRI findings as
specificity of the test is much better than the the gold standard. If our clinical examination
sensitivity. Thus, the test performs better in sub- yielded high sensitivity and specificity, we might
jects who do not have the disease or condition conclude that the expense of MRI evaluation is
than it does in those subjects who do have the not justified to identify patients with the pres-
disease or condition by correctly testing negative ence or absence of TMJ disc displacement. On
for disease-free individuals in a higher percent- the other hand, if the sensitivity and specificity
age of cases than it correctly tests positive for were low, we might conclude that our clinical
disease-present individuals. examination methods (screening test) are not a
The false negative rate is defined as the pro- reasonable substitute for the MRI examination
portion of subjects who have the disease but test of the TMJ to detect disc displacement.
negative. This rate is given in Table 1 by c/(a + The true positive and true negative rates (sen-
c). The false positive rate is the proportion of sitivity and specificity) and the false positive and
subjects who do not have the disease and test false negative rates are calculated using as the
positive. This rate is given in Table 1 by b/ (b + denominator the numbers of patients who are
d). By using the Table 1 example data, the false known to have the disease or to be disease free.
<<
Article
>> Home | TOC |
Index
94 Susan P. McGorray
From a practical standpoint, however, we often the proportion of subjects that are correctly clas-
do not know the true disease status of a patient. sified, (a + d)/n, in which a is the number of
Instead the only information available to us may true positives and d is the number of true neg-
be a positive or negative test result. How confi- atives. In our example in Table 1, the accuracy is
dent can we be that the disease is present or 70% (true positives = 8, true negatives = 62, n =
absent based on the test results? Knowing the 100). In other words, the accuracy of the test is
sensitivity and specificity of a test may not be an overall summary of how well the test classifies
useful for us because these estimates use known those patients with and without the disease.
disease status for their denominators. Usually,
we are trying to determine the true disease status
Relationship Between Test
of our patients from our test. Two other test
Characteristics
characteristics, the positive and negative predic-
tive values (defined later), use as their denomi- We have defined several test characteristics by
nator the numbers of patients who test positive using the counts or frequencies based on the
or test negative and thus are more clinically four combinations of test result and disease sta-
relevant indices of test performance. The posi- tus as shown in Table 1. As might be suspected,
tive predictive value (PPV) is the probability of relationships exist between these test character-
the presence of the disease for those subjects for istics. In this section, we will look at how exactly
whom the test is positive. In Table 1, this can be they are related.
estimated by a/(a + b). Similarly, the negative Table 2 further shows applications of the con-
predictive value (NPV) is the probability of the cepts described in the preceding section with
absence of the disease in those subjects for another twist to add to the mix. We will compare
whom the test is negative, estimated by d/(c + two studies of TMJ dysfunction (artificial data)
d). In the Table 1 example, the positive predic- that examine the relationship between a screen-
tive value is 8/18 (44%), whereas the negative ing test (clinical evaluation) and the gold stan-
predictive value is 62/82 (76%). dard (MRI) assessment of disease status. As be-
Note that the denominators a + b and c + d fore, we will use MRI assessment to ascertain the
represent the number of positive tests and neg- true condition with respect to TMJ disc status.
ative tests, respectively. Compare this to the de- Our task is to evaluate the method of clinical
nominators a + c and b + d, which represent examination of the TMJ to determine how use-
the number of true cases with disease and true ful it can be as a screening test to identify disc
cases without disease. The accuracy of a test is displacement. Data presented from study 1
Table 2. Test Characteristics From Two Studies Evaluating the Relationship Between Clinical Assessment of
TMJ and MRI Assessment of TMJ
MRI Assessment of TMJ (True Disease Status)
Clinical Evaluation of TMJ (Test Result) Present (+ ) Absent (- ) Total
Study 1
Present ( + ) 40 20 60
Absent ( —) 10 30 40
Total 50 50 100
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 ( + ) 20 30 50
Absent ( —) 5 45 50
Total 25 75 100
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%
<<
Article
>> Home | TOC |
Index
might result from a study designed to estimate vary greatly depending on the prevalence of dis-
sensitivity and specificity of the clinical evalua- ease in the population that is under study. Prev-
tion because equal numbers of subjects with and alence of the same disease may vary among dif-
without the disease were chosen (50 in each ferent populations.
group). Having equal numbers of subjects with The importance of prevalence rate can be
and without TMJ disc displacement provides the seen from the two hypothetic studies previously
largest sample size of each and thus will provide presented in Table 2. Study 1 may have been
the best estimates of sensitivity and specificity. designed to assess the performance of the clini-
Data presented from study 2 might result from cal evaluation of the TMJ as a screening test for
examination of a group of subjects who are seek- TMJ disc displacement. The prevalence of the
ing general dental care. In the second case, we disease in that study was 50% by design (equal
must be content with what we get and do not numbers of subjects with and without TMJ disc
specify how many subjects we will evaluate with displacement were examined). The PPV and
and without TMJ disc displacement. We will as- NPV were found acceptable at 67% and 75%,
sume that this second group of subjects yielded respectively. Based on these encouraging results,
25 with disc displacement and 75 without. Table we will assume the clinical evaluation was then
2 lists the test characteristics for these two stud- used as a screening test for TMJ disc displace-
ies. Careful inspection of this table, however, will ment on a larger number of subjects, with lower
show that the test characteristics (sensitivity, prevalence of TMJ (25%) than in study 1. The
specificity, PPV, NPV, accuracy) are not all iden- drop of the PPV (percent of those with positive
tical for the two studies. How is this possible? clinical evaluation for TMJ dysfunction who ac-
There is one more key component that con- tually had TMJ dysfunction) from 67% to 40%
tributes to the relationship between test charac- may have been both surprising and disappoint-
teristics: the prevalence of disease. The preva- ing. In study 1, the prevalence of the condition,
lence of a disease or condition is the frequency disc displacement, is 50%, and the correspond-
with which it is found in the population under ing PPV is 67%. In study 2, the prevalence is
study. In study 1 in Table 2, TMJ disc displace- 25%, and the PPV is 40%. If the prevalence in
ment was present in 50% of the study subjects, another population of interest, perhaps a differ-
whereas in study 2, it was present in only 25%. ent ethnic or age group, is only 10% and screen-
Positive and negative predictive values depend ing is accomplished by using the clinical exami-
on the sensitivity, specificity, and the prevalence nation, the corresponding PPV would be only
of disease in the setting in which the test is being 19%. In this last example, 19% of those who test
used. If we know the sensitivity and specificity of positive for the condition would actually have
a test as well as the prevalence of the disease or the condition. Furthermore, 81% (approximately
condition, we can calculate the PPV and NPV. 8 of 10 subjects) who test positive based on the
The mathematical formulas that define the rela- clinical examination would not have the condi-
tionship between the prevalence, predictive val- tion confirmed by MRI. This is a disappointing
ues, and sensitivity and specificity are as follows: finding indeed and shows why new tests that
seem promising at first evaluation may not im-
(prevalence X sensitivity) prove diagnostic capability when put into com-
PPV = —
[(prevalence X sensitivity) mon use.
+ (1 — prevalence) X (1 — specificity)] The previously mentioned examples show the
importance of examining a test's performance
and
characteristics and the disease prevalence when
[(1 — prevalence) X specificity] planning a research study. To aid in understand-
NPV- ing the relationship between PPV and NPV and
[(1 — prevalence) X specificity
prevalence, a graphical approach may be help-
+ prevalence X (1 — sensitivity)] ful. Figure 1 shows such a display using data
Why is it important to know the disease prev- from our hypothetic example mentioned previ-
alence in your population of interest? The PPV ously. Let's assume that our screening test, the
and NPV are the most clinically useful charac- detection of TMJ disc displacement by clinical
teristics of a test's performance. But they can examination, has a sensitivity of 80% and speci-
<<
Article
>> Home | TOC |
Index
96 Susan P. McGorray
SENSITIVITY = 80 % SPECIFICITY = 60% specificity, the PPV decreases and the NPV in-
25% creases as the prevalence of a disease or condi-
8- 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%).
LLJ Note how the shape of the PPV and NPV curves
D
differs from Figure 1. For the test shown in
LJJ
Figure 1 (which has lower sensitivity and speci-
>
ficity), the prevalence of the disease or condi-
O
Q
tion has greater impact on the usefulness of the
CC
Q.
test than it does for the test depicted in Figure 2
(higher sensitivity and specificity). We can see in
s- PPV 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-
20 40 60 100
lar, but not as dramatic, in Figure 2. Thus, over
PREVALENCE (%)
a wide range of disease prevalence the test
NPV = 90 % and PPV = 40 % AT 25 % PREVALENCE shown in Figure 2 (higher sensitivity and speci-
NPV = 75 % and PPV = 67 % AT 50 % PREVALENCE ficity) would be more useful than the test in
Figure 1. PPV (dashed) and NPV (solid) as a func- Figure 1 because NPV and PPV values change
tion of prevalence for a test with 80% sensitivity and less over a range of prevalences. By using graphs
60% specificity. The squares correspond to the pre- of this sort, one can visually assess how the pre-
dictive values in Table 2, study 1, and the circles
correspond to the predictive values in Table 2,
SENSITIVITY = 95 % SPECIFICITY = 95%
study 2.
25%
8-
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 LU
=>
of both the NPV and PPV curves. The x-axis 8-
represents different levels of the disease preva- LU
lence (TMJ disc displacement). If the prevalence
of disc displacement in our population of inter- 1
Q
we see visually that the PPV and NPV change as NPV = 98.3% and PPV = 86.4 % AT 25% PREVALENCE
disease prevalence changes.
The PPV and NPV curves are unique for spe- Figure 2. PPV (dashed) and NPV (solid) as a func-
tion of prevalence for a test with 95% sensitivity and
cific sensitivity and specificity combinations (ie, 95% specificity. The predictive values that would be
sensitivity = 80%, specificity = 60%). In Figure obtained from a study with a 25% prevalence of dis-
1, one can see that for a given sensitivity and ease are indicated by diamonds.
<<
Article
>> Home | TOC |
Index
dictive values of a test will change if the test is funds could be allocated for those patients who
used as a screening test in a research study in meet the criteria for levels of recommended and
which the prevalence of the disease or condition above. Thus, the cut point for defining orth-
is lower or higher than that under which the test odontic treatment need in the first instance
was initially evaluated. If you know the test's would be necessary, and only those patients who
sensitivity and specificity, you can consult the meet those criteria would qualify for care. In the
graph with that sensitivity/specificity to estimate second instance, the cut point would be recom-
the PPV and NPV of the test for the level of mended, and all those patients who meet criteria
disease prevalence in your population. This will for both the recommended and the necessary
help you decide how useful a screening test levels would qualify for orthodontic care. When
would be in identifying disease in your popula- differing definitions of a disease or condition
tion of interest. are used, sensitivity, specificity, and prevalence
The oversimplification of the disease process estimates can vary, and studies that use different
may also affect estimates of sensitivity and spec- definitions cannot be compared in a meaningful
ificity. Often diseases or conditions are more way.
complex than simply being present or absent,
and some criteria or cut point must be specified
Accuracy and Agreement
to classify subjects as having or not having the
disease. Examples include hypertension, cancer, The use of one characteristic, such as accuracy,
and need for orthodontic treatment. Various to summarize the diagnostic capabilities of a test
sliding scales of need for orthodontic treatment is appealing but must be approached with cau-
are commonly used by public health agencies tion. Consider the two studies presented in Ta-
for prioritizing funds for orthodontic treatment. ble 3. We will assume that 1,000 subjects were
For example, if need for orthodontic treatment tested in each case. Although the sensitivity and
is classified as none, optional, recommended, or specificity are both higher for study 2, the accu-
necessary, based on some defined set of criteria, racy is higher for study 1. In study 1, the preva-
an agency may allocate funds for treatment to lence of the disease is low at 50/1,000 (number
only those patients who meet the criteria for the of patients with the disease present in popula-
necessary level of treatment need. Alternatively, 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
98 Susan P. McGorray
is low, many subjects will truly not have the first case and 78% in the second case, the kappa
disease and also test negative (true negatives). statistics are quite different, 0.70 and 0.32, re-
Because of this, the number of true negative spectively. It is also possible for the accuracy of
subjects is high and will account for a large one test to be higher than that of a second test,
proportion of the accuracy of the test. This can but the kappa statistic may be larger for the
obscure other test information. True negative second test. This is shown in Table 3, with accu-
subjects account for 97% of the accuracy (855/ racy of 88% and a kappa statistic of 0.24 for study
[855 +25]) in study 1, but only 77% of the 1 and accuracy of 86.5% and a kappa statistic of
accuracy (665/[665 + 200]) in study 2. Disease 0.66 for study 2. Because of the prevalence of the
prevalence must be similar in both populations disease and the proportion of positive test re-
when comparing the accuracy of tests from dif- sults, we would expect more agreement caused
ferent studies. Both components of accuracy, by chance in study 1, leading to higher accuracy.
the true negatives and the true positives, should However, much of the accuracy is caused by
be examined. When the disease prevalence dif- chance alone rather than the relationship be-
fers, it is not appropriate to compare the accu- tween test result and disease status. These para-
racy of two tests. doxes point out that both the accuracy and the
An alternative approach for assessing the chance-corrected agreement (ie, kappa statistic)
overall accuracy or agreement of a test involves should be considered when evaluating a test.
taking account of agreement that would occur One measure is not enough to provide a com-
purely by chance (ie, if the test results provided plete picture.
no information about the true disease status).
The statistic that captures this kind of agreement
is called the kappa statistic2 and its calculation
Diagnostic Likelihood Ratios
can be found in many general statistics books. The use of diagnostic likelihood ratios5 to de-
For the purposes of this article, it suffices to scribe test performance is becoming more com-
know that, in brief, the kappa statistic depends mon. This is a fairly complex concept to under-
on the observed agreement (between the screen- stand and is mentioned in this article only for
ing test and gold standard) and the expected the sake of completeness. The use of diagnostic
agreement that would occur by chance alone. If likelihood ratios is appealing because they pro-
the agreement between the screening test and vide information useful for ruling in or ruling
the true disease status (determined by the gold out disease, much like predictive values, yet un-
standard) is perfect, the kappa statistic is 1.00. If like the previous test characteristics we have dis-
the agreement we observe is equal to what would cussed, are not influenced by prevalence of the
be expected by chance alone (ie, not very good), disease. In addition to their usefulness as de-
the kappa statistic is 0. Negative kappa values scriptive characteristics of diagnostic or screen-
indicate a very bad situation when agreement is ing tests, more complex evaluation of screening
less than would be expected by chance alone. tests can be done by statistical modeling. This
Guidelines for interpretation of the kappa statis- level of statistical sophistication, however, is
tic have been proposed,3 with kappa values less likely beyond the grasp of most clinicians, and
than 0.40 indicating poor agreement beyond enlisting the expertise of a statistical consultant
chance, 0.40 to 0.75 indicating fair to good is highly encouraged.
agreement, and greater than 0.75 indicating ex-
cellent agreement.
Receiver Operating Characteristic
In our example depicted in Table 3, the
Curves
kappa statistic for study 1 is 0.24 and for study 2
is 0.66. Like the PPV and NPV, the kappa statistic The final section of this article discusses a
is also affected by prevalence.4 Kappa values can method that is becoming more widespread for
vary quite dramatically for a given level of accu- the interpretation of the sensitivity and specific-
racy, depending on the underlying expected ity characteristics of new diagnostic tests. The
agreement. For example, consider two cases in method is based on receiver operating charac-
which the observed agreement is relatively high teristic (ROC) curves.
at 85%. If the expected agreement is 50% in the In the discussion thus far, we have considered
<<
Article
>> Home | TOC |
Index
screening or diagnostic tests that result in only level (the screening test). A cut point is the
two outcomes (eg, positive or negative results). referral level where all patients on one side of
This is not always the case. Some tests may have and including the cut point level are referred
more than two outcomes. For example, the as- and all patients on the other side are not re-
sessment of orthodontic treatment need may ferred to an orthodontist. Each possible cut
yield ordered categories such as in our example point level results in a decision rule that clearly
previously mentioned in which treatment need identifies which children should be referred. For
may range from none, optional, recommended, example, if we select probably refer as the cut
or necessary (four possible outcomes). From this point, the decision rule is to refer all children in
information, a binary decision (only two choices the probably refer and definitely refer catego-
such as yes/no) may still be required. In our ries. Children in the other four categories are
treatment need example, the two choices would not referred. A different cut point would result
be either that the patient should receive treat- in a different decision rule. By using the gold
ment or the patient should not receive treat- standard to determine true need, one can deter-
ment. mine the sensitivity and specificity of each deci-
We will show this concept with a hypothetic sion rule (ie, how good is that decision rule in
study of dentists referring pediatric patients for distinguishing the patients who do and do not
orthodontic evaluation. In this example, each need treatment as compared to the gold stan-
child is classified into one of the following six dard [the orthodontists]).
levels of a referral scale: definitely refer, proba- In Table 4, we know that 104 children have
bly refer, possibly refer, possibly not refer, prob- true need based on the opinion of experienced
ably not refer, and definitely not refer. We also orthodontists (gold standard). We also see that
assume that a gold standard is available to iden- 96 children have no need according to our gold
tify which children actually do and do not need standard. If all children in the definitely refer
orthodontic treatment. Frequently, this determi- and probably refer categories are referred (43)
nation is based on the consensus opinion of and others are not (89), the sensitivity of this
experienced orthodontists who are considered strategy is 41% (43/104), whereas the specificity
the gold standard. The method of identifying is 93% (89/96). Another cut point at the possi-
treatment need based on pediatric referral pat- bly not refer level results in the following deci-
terns is thus considered the screening test. We sion rule: referring all those in the possibly not
will use the sensitivity and specificity of this refer or higher likelihood of referral categories.
screening test to generate ROC curves as de- That would mean that 98 patients would be re-
scribed later. ferred and 44 would not be referred. This yields
Data from this hypothetic study are presented a sensitivity of 94% (98/104) and a specificity of
in Table 4. From the dentists' referral scale of 46% (44/96). By calculating the sensitivity and
treatment need, different cut points are exam- specificity for each cut point level on the referral
ined to determine which children would receive scale, one obtains a series of points that can be
an orthodontic evaluation based on the referral graphically displayed as an ROC curve (Fig 3).
Points in the upper left corner represent the
desirable situation of simultaneous high sensitiv-
Table 4. Classification of Subjects by Orthodontic ity and specificity. The ROC curve is formed by
Referral Status (Rows) and True Orthodontic
Treatment Need (Columns) joining the points resulting from each decision
rule, as shown in Figure 3. Each point represents
Treatment Need (True Status)
a sensitivity/specificity combination for each of
Do Not Need Need the referral levels on the referral scale. The y-
Orthodontic Referral Status Treatment Treatment Total
axis is the scale of sensitivity values, and the
Definitely refer 3 17 20 x-axis is the scale of specificity values. If both
Probably refer 4 26 30 high sensitivity and high specificity are impor-
Possibly refer 10 40 50
Possibly not refer 35 15 50 tant (this is usually the case), the best decision
Probably not refer 25 5 30 rule corresponds to the point closest to the up-
Definitely not refer 19 1 20 per left-hand corner of the plot. In our example,
Total subjects 96 104 200
the best cut point that jointly maximizes sensi-
<<
Article
>> Home | TOC |
Index
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-
20 40 60 80 100 mental concepts will facilitate insight and crit-
[1 - specificity] (%) ical evaluation of research that relies on such
methodology.
Figure 3. The ROC curve for the data presented in The examination of characteristics of
Table 4. The starred point represents the most desir- screening or diagnostic tests can provide
able cut point on the referral scale indicating the much useful information but must be done
likelihood to refer for orthodontic evaluation. At this
cut point, patients in the referral groups designated as with care. The setting in which a test is used
definitely refer, probably refer, and possibly refer are needs to be considered and comparisons of
classified as needing orthodontic treatment. Patients characteristics made only when warranted.
in the referral groups designated as definitely not Studies to estimate test performance charac-
refer, probably not refer, and possibly not refer are teristics should be well designed. Aspects to
classified as not needing orthodontic treatment. The
sensitivity and specificity of this decision rule are 80% consider 12 include evaluation of potential bi-
and 82%, respectively. At this cut point, the sensitivity ases such as incomplete diagnostic work-up
and specificity of the decision rule are jointly maxi- and subjects not representative of the popula-
mized. tion on which the test will be used. Adequate
sample sizes are needed of those with and
without the disease or characteristic. Although
tivity (80%) and specificity (82%) classifies those
the question of does this patient have this
with referral status possibly or higher as needing
referral. This point can be easily identified from disease seems straightforward, the answer is
often complex, and better understanding and
the ROC curve displayed in Figure 3.
Other summaries of test performance can be use of diagnostic and screening tests will result
in an increased ability to answer this key ques-
calculated from ROC curves. One common mea-
tion.
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 References
and 100% specificity. The closer the point is to 1. Nebbe B, Major PW: Prevalence of TMJ disc displace-
this corner, the larger the area under the ROC ment in a pre-orthodontic adolescent sample. Angle
Orthod 70:454-463, 2000
curve. Larger areas correspond to better classi- 2. Fleiss JL. Statistical Methods for Rates and Proportions
fication scales or tests. The area can be inter- (ed 2). New York, NY, Wiley, 1981
preted as the proportion of correct decisions in 3. Landis JR, Koch GG: The measurement of observer
a two-alternative forced choice experiment (eg, agreement for categorical data. Biometrics 33:159-174,
treat/do not treat) .6 Further details about ROC 1979
4. Feinstein AR, Cicchetti DV: Higher agreement but low
curves can be found in articles exploiting this kappa: I. The problems of two paradoxes. J Glin Epide-
methodology.7'9 Two recent articles in the orth- miol 43:543-549, 1990
odontic literature that rely on ROC curve meth- 5. Boyko EJ: Ruling out or ruling in disease with the most
<<
Article
>> Home | TOC |
Index
sensitive or specific diagnostic test: short cut or wrong teristic analysis. Commun Dent Oral Epidemiol 24:
turn? Med Decis Making 14:175-179, 1994 303-306, 1996
6. Green DM, Swets J: Signal Detection Theory and Psycho- 10. McGorray SP, Wheeler TT, Keeling SK, et al: Evaluation
physics. New York, NY, Wiley, 1966 of orthodontists' perception of treatment need and the
7. Hanley JA: Receiver operating characteristic curves peer assessment rating (PAR) index. Angle Orthod 69:
(ROC), in Armitage P, Colton T (eds): Encyclopedia 325-333, 1999
of Biostatistics. New York, NY, Wiley, 1998, pp 3738- 11. Begun FM, Firestone AR, Vig KW, et al: A comparison of
3745 the reliability and validity of 3 occlusal indexes of orth-
8. Pepe MS: Three approaches to regression analysis of odontic treatment need. Am J Orthod Dentofacial Or-
receiver operating characteristic curves for continuous thop 120:240-246, 2001
test results. Biometrics 54:124-135, 1998 12. Gruggenmoos-Holzmann I, van Houwelingen HC: The
9. Beam D, Wright J, Kay E, et al: Perceptions of orth- (in)validity of sensitivity and specificity. Stat Med 19:
odontic treatment need: Receiver operating charac- 1783-1792, 2000
<<
Article
>> Home | TOC |
Index
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.
height or mandibular plane angle) we get a such as bias in sampling and control for mea-
collection of measurements, one for each sub- surement error always arise. And sometimes (as
ject in the group. Such a collection is called a in our last Presidential election), they assume
distribution. Because distributions for any sub- major proportions. But beyond these consider-
stantial group of subjects are necessarily diffuse, ations, enumerative statistics is conceptually
the need arises to develop a shorthand descrip- fairly straightforward.
tion of the group's properties with respect to There is, however, another general area of
each variable of interest. At minimum, we need statistics (of greater interest to us here) in which
for this purpose an estimate of averageness information gathered from some groups is used
(called a measure of central tendency) and an to infer present or future conditions in other
estimate of variability (called a measure of dis- groups. When, for example, we study the growth
persion). The most commonly used estimate of trajectory of a sample of subjects for whom metal
averageness is the arithmetic mean and the most implants were placed 40 years ago, we are not
common estimate of variability (ie, of the aver- primarily interested in those particular subjects.
age deviation from the average), is the standard Rather, we are interested in what the study of
deviation. Suppose, for example, we knew (from those earlier subjects can tell us about the likely
well-conducted prior studies) that the mean growth trajectory of the patients we are going to
value for increase in mandibular body length for treat from now on. The discipline that deals with
girls between 8.5 and 15.5 years is 6.3 mm with a valid extrapolation of findings between groups is
standard deviation of 5.2 mm. If we wished to called inferential statistics. It is a subject of major
apply this knowledge to the prediction of the concern to clinical practice and, as might be
growth of each of a series of individual female expected, its reasoning is considerably more
patients who successively occupy our chair, our complex than the reasoning of simple enumer-
best estimate for the growth of each patient ation. The remainder of this article will outline
would by 6.3 mm (the mean). But it is also the some aspects of the rationale of statistical infer-
case that if we used this mean value as a predic- ence.
tion of the expected size increase for each indi-
vidual, the average error would be approxi-
The Basic Principle of Inferential
mately 5.2 mm (ie, one standard deviation) per
Statistics
patient.
Thus, the size of the standard deviation gives In the simplest situations in which statistical in-
the wary clinician an important clue about the ferences need to be made, we are usually trying
confidence with which he can use the mean to to identify modal differences between two kinds
estimate expected events in growth or treat- of individuals. (For instance, we may be trying to
ment. The larger the standard deviation, the measure differences between treated individuals
greater the uncertainty is that accompanies the and untreated control individuals or to measure
use of the mean as a predictor in individual differences between two groups of individuals
cases. And as the size of the standard deviation that have received different kinds of treatment.)
approaches the size of the mean, the mean be- For these purposes, it is necessary to sample
comes progressively less useful for predicting without bias the collection (or population) of all
expected values in individual patients, although subjects who received each kind of treatment,
it may continue to be very useful as a ball-park and then to compare the samples. If all the
estimate. individuals within one sample measure the same
and all the individuals within the second sample
measure the same, then it is only necessary to
Enumeration Versus Inference
measure one member of each sample and then
One major use of statistics is to tally conditions to compare the two measurements. If the two
or relationships within one or more groups as an measurements do not differ (within our toler-
end in itself. Examples include bracket inven- ances), we may reasonably conclude that there is
tory, census taking, and vote counting. Such ex- no substantial difference between the two
ercises fall under the heading of enumerative groups. On the other hand, if the two measure-
statistics. Whenever tallies are conducted, issues ments differ substantially, then we can fairly rea-
<<
Article
>> Home | TOC |
Index
sonably conclude that the two samples were vided by an estimate of the pooled sample vari-
drawn from different populations. (Note that ability in the denominator. The larger the dif-
this is a fair description of what happens when ference is between the means as compared with
you take one central incisor bracket from each the measure of variability, the greater the likeli-
of two shipments and compare them to see if the hood is that two samples are truly from different
two shipments are consistent.) populations.
However, identifying differences between two This critical examination of the relationship
samples of patients who have received different between size of the average differences between
orthodontic treatments is not that simple, in groups and the amount of variability among the
large part because there is variability between individuals who comprise the groups is the hall-
the individual subjects within each treatment mark of all the tests that comprise inferential
group. Given that variability, it might seem the statistics. The variability observed may be from
most obvious thing to do would be to compare actual differences between individuals or from
the averages of the two samples. One might, for noise associated with the imprecision of the
example, measure all the members of each sam- measurement process. In either event, all infer-
ple, calculate the average (or arithmetic mean) ential tests, whether simple or complex, are de-
for each sample, and then compare the two signed to protect the user against the risk of
means. Unless our measuring instrument had confusing true between-group differences with
been quite crude, there would almost certainly chance occurrences during sampling and mea-
be a difference between the two means, and surement.
some inference could be drawn.
But let us consider further the risks of such an
The Relationship Between Sampling and
inference. Clearly, simply as a consequence of
Generalizability
individual variation, even different samples
drawn from the same population should be ex- Although it may not seem immediately evident,
pected to have somewhat different means. the manner in which samples for clinical studies
Hence, the mere fact of small differences in are gathered is of great importance to the clini-
mean does not establish that the populations cian seeking to apply findings from the studies
from which the two samples were drawn were to the treatment of the patient who actually
different. How then can we be at least reasonably occupies his or her chair. In evaluating findings,
sure that the observed difference between the the clinician inevitably must ask, "are the pa-
means of different samples in any particular ex- tients in this study similar to mine?" If the answer
periment is greater than might have occurred by to this question is either uncertain or in the
chance if there really was no difference in the negative, then the research findings, although
populations from which they were drawn? This is they may be of academic interest, are of greatly
equivalent to asking, given the variability in the reduced practical interest to the clinician as a
populations from which the different samples clinician. For this reason, articles reporting in-
were drawn, what is the likelihood that two sam- formation on clinical studies have the absolute
ples with means as different as the ones we ob- requirement of describing sample composition
serve could have occurred by chance alone? and sampling methods completely and precisely.
It seems intuitively obvious that the smaller Statisticians describe the property of a study's
the sample sizes and the greater the variability findings being directly applicable to the clinical
within the populations, the larger will be the situation as generalizability. In most medical tri-
chance that the means of two or more samples als, generalizability is insured by protocols that
will differ markedly by chance alone. In any specify the criteria for inclusion in the research
carefully conducted study, we know the sizes of sample. Such a protocol typically describes inclu-
the samples, and we can pretty well estimate the sion and exclusion criteria concerning age, gen-
variability of the populations from which they der, precise diagnosis, precise previous course of
are drawn by measuring the variability within the illness, and precise previous treatment. The cli-
samples themselves. We can therefore establish a nician can then directly compare the profile of
relationship in which the measured difference his current patient with the protocols of relevant
between the means (in the numerator) is di- clinical trials and determine which previous
<<
Article
>> Home | TOC |
Index
studies have findings that bear directly on the pendently (step 3, Fig 1). The group of patients
treatment of each particular patient in his or her for whom some judges advocated extraction
practice. Thus far, very few studies in our spe- treatment (whereas others advocated nonextrac-
cialty specify the nature of their samples well tion treatment) was then considered to consti-
enough to allow the clinician to estimate the tute a borderline sample.3
applicability of their findings to the treatment of Before presenting any cases to the clinician
specific patients with confidence. panel, we attempted to develop a set of objective
As an example, I would like to cite the limi- criteria that would allow us to segregate out all
tations of one of our own studies that was rela- or most of the straightforward cases for which
tively well received by the specialty a few years the judges' decision to extract or not to extract
ago.1'2 At that time, we sought to conduct a would have been unanimous, (steps 1 and 2, Fig
prospective randomized clinical trial to study the 1.) The construction of such a set of objective
differences in outcome of extraction versus non- criteria would have yielded two advantages. First,
extraction strategies in the treatment of border- it would have permitted us to reduce the work-
line Angle Class I and Class II patients. Figure 1 load of our clinician/judges (each of whom had
represents the design of our study. An ethical an active private practice). Second and more
and scientific precondition for such randomiz- important, it would have allowed us to codify the
ing of patients to either extraction or nonextrac- characteristics of the group of patients to which
tion therapy is the requirement that the investi- the results of our study could later be general-
gators establish for each individual patient that ized appropriately. Unfortunately, although our
there exists true uncertainty as to which treat- study produced a number of interesting and
ment is preferable for that particular patient. On useful findings, it was not successful in identify-
reflection, the reader will recognize that when- ing any unique set of objective measures that
ever a single clinician makes treatment deci- defines precisely the individual patients in any
sions, there are actually no borderline patients. clinician's practice to which the study's results
Rather, after due deliberation, the clinician al- appropriately generalize.
ways decides either to extract or not to extract. It should be noted that Johnston and col-
In the absence of any consensually accepted leagues confronting the same problem success-
protocol for making the extraction/nonextrac- fully used the method of discriminant analysis to
tion decision, our method for establishing true define the characteristics of a group of subjects
uncertainty was to have each of a number of who, on average, could be considered border-
experienced clinicians evaluate the pretreat- line in the sense that different clinicians would
ment records for each potential subject inde- 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.
Follow-up
Statistics About Groups Versus
Statistics About Individuals
Almost all clinical investigations in medicine and
dentistry involve the investigation of some indi-
Assign to clinician preferring Tx 1 Assign to clinician preferring Tx 2 viduals treated in the past or present for the
(with balancing) (with balancing)
purpose of accumulating information that can
*• Follow-up * be applied to the care of other individuals who
will be treated in the future. The individuals who
Figure 1. Design of a prospective randomized clinical
trial in which orthodontic patients were randomized
are subjects of our investigations are usually
to extraction or nonextraction treatment based on studied as members of groups called samples.
clinicians' disagreement as to which course of treat- Users of statistical findings differ in their profes-
ment was indicated.3 sional needs and orientations. Depending on
<<
Article
>> Home | TOC |
Index
the nature of the problems that concern them, chance differences in sampling. An example of
some users focus their interest primarily on the an analogous epidemiologic measurement in
characteristics of individual subjects, whereas orthodontics is the use of the peer assessment
other users are more interested in the charac- rating index in Britain to monitor the efficacy of
teristics of the groups to which the individuals the distribution of the limited funds allocated to
belong. Surprising as it may seem, the kinds of the treatment of malocclusion by the British
research design and statistical treatment that are national health service.9'12
appropriate to the collection and analysis of data As we have noted, research designs such as
differ substantially, depending on whether the the ones previously mentioned can be very im-
investigation is focused primarily on the perfor- portant and useful in public health dentistry, but
mance of individuals or on the performance of they are of less use in the understanding of
groups. individual variation or the differential planning
Allow me to expand on this surprising asser- of individual treatment, which is to say for an-
tion. By far the largest investments in clinical swering the kind of questions that are the imme-
research in dentistry, both in the United States diate focus of interest of clinicians who treat
and elsewhere, are and probably should be in patients one at a time. For such purposes, one
the domain of public health. Studies in this do- needs to study in depth the way in which clinical
main have as their main goal not the treatment treatment is actually delivered. Such studies in-
of individual patients but rather the establishing volve sampling of multiple records of different
of public health policy and the improvement of kinds for the same patient, just as the conscien-
the health of the public in general. For the tious clinician does when planning treatment.
purposes of such studies, the investigators are This is to say, we must sample a larger portion of
interested primarily in group trends rather than the total variance within each subject patient by
in individual outcomes. When, for example, the measuring head films and study casts and in-
US Public Health Service achieved its enor- traoral x-rays and facial photos. Additionally, we
mously successful test of the efficacy of fluoride must also measure many more parameters per
administration in the reduction of tooth de- individual, and we must accomplish these mea-
cay,6'8 major concern was not with the caries rate surements with much greater precision and ac-
for any particular individual but rather with the curacy. These tasks involve considerable labor
overall effect of fluoride administration on the and tend to be less attractive to public health
population as a whole. The goal of the investi- funding agencies than are more dramatic single-
gators was to achieve a measurable reduction in issue epidemiologic studies. For this reason and
tooth decay across the entire population, with- others, consequential support for clinical studies
out much concern about which particular indi- on the details of orthodontic treatment over the
viduals benefited and which did not. The exper- next few years will have to come from the orth-
imental strategy was to set up one of two similar odontic specialty itself, acting through agencies
communities with fluoridated drinking water like the American Association of Orthodontists
and another without and then to measure some Foundation.
simple measure of difference in decay rate in the
two communities through time. Thus, to estab-
Statistical Significance, Clinical
lish the usefulness of the treatment, it was only
Significance, and Effect Size
necessary to determine that on average the per
capita rate of decayed, missing, and filled teeth One of the most confusing areas in statistics
was lower in the fluoridated community than in involves the use of the term significance. This
the nonfluoridated community. common English word is defined in my dictio-
The general strategy of epidemiologic studies nary as "[q]uality of being important; . . . mo-
such as the one just cited is to make some sim- ment; . . . consequence."13 But that is not at all
ple, easily made, and relatively crude measure- what is meant when the term statistically signifi-
ments across a relatively large number of indi- cant is used. To examine what the word signifi-
viduals in the hope of showing that the average cant means when used in statistics, we must re-
differences between two groups being compared visit the issues of variability and inference that
are larger than can be accounted for merely by were discussed earlier. As has been noted (or at
<<
Article
>> Home | TOC |
Index
least implied), the essential task of inferential drawn from different populations (ie, that there
statistics is to determine what can reasonably be is a real difference between the two treatments
concluded about a population (eg, the popula- that the samples represent). In doing so, we
tion of all Class II, high-angle cases) on the basis accept the risk that there is 1 chance in 20 that
of having examined only a sample consisting of our assertion is in error. But there is an impor-
a limited number of individuals drawn from that tant caveat to be observed here. If we do many
population. The sample is thus a relatively small tests using this 5% yardstick, we are likely by
window through which the investigator hopes to definition to have one or more spurious findings
see the outlines of some larger, more inclusive of statistical significance merely by chance. For
reality. In some cases, such a glimpse may truly example, one might (as our laboratory did some
represent the larger reality, whereas in other years ago) test the differences associated with
cases it may misrepresent it, leading the investi- three different strategies for maxillary retraction
gator to erroneous conclusions. This is the case compared with an untreated control group.14'15
because whenever there is random variability If one then tests the differences between groups
inherent in the phenomena under investigation, two at a time, there are actually six tests for any
there is the possibility that what is observed re- single variable. In our case, for example, one
sults from nothing more than chance coinci- could test distalization of the upper first molar
dence—that if a different sample had been cusp in the following combinations: (1) control
drawn, examination might have led to markedly versus cervical, (2) control versus high pull, (3)
different conclusions. Until the possibility that control versus activator, (4) cervical versus high
the findings of a study reflect nothing more than pull, (5) cervical versus activator, and (6) high
coincidence has been addressed and rejected, pull versus activator. For this reason, one should
no conclusions one way or another can reason- try to limit the number of statistical tests to be
ably be drawn. performed and to preplan them by generating
By virtue of extensive study of patterns of testable hypotheses before data acquisition.
variability within groups of subjects, statisticians Note that the Pvalue, or risk of error we are
have devised methods of conservatively estimat- willing to accept, is merely a convention; we
ing the variability of populations from the vari- may choose alternatively to set the level of
ability within samples drawn from them. That significance at P < .01 or even P < .001 where
being the case, it is possible to estimate mathe- we consider appropriate although we are
matically the probability that two samples of honor bound to set this standard before we
given sizes and internal variability could by look at the data. Note also that the discussion
chance alone have differed in mean value as of statistical significance in this section has
much as they do, even if both samples had been really been a restatement, from an alternative
drawn from the same homogeneous population. perspective, of the discussion about the basic
If the numbers tell us that it is highly unlikely principle of inferential statistics earlier in this
that two such samples could reasonably have essay.
been drawn from the same population, then we Finally and most importantly, note that statis-
reject the possibility that there is no difference tical significance has nothing to do with impor-
between them and conclude that they were tance. It deals only with the problem of control-
drawn from truly different populations. Of ling for chance errors in sampling. Of course,
course, there is always the possibility that two one way to minimize chance errors in sample
considerably different samples could in fact be composition is to use larger samples. But this
drawn from the same population, in which case strategy can result in a paradox for the clinician
our assertion of a difference between them is in focused on individual case outcomes because as
error. But by convention, if the numbers show sample sizes get larger (which orthodontists
that the probability of such an error is less than have always thought of as an unmitigated boon),
5% (or 1 chance in 20), we assert that the dif- smaller and smaller differences between group
ference between two samples (or between one means come to be statistically significant.
sample with pre- and post-treatment difference Clearly, in addition to assurances that the find-
of zero) is statistically significant at the P < .05 ings we publish and read are not accidents of
level. Thus, we assert that the two samples were chance, we need information about how impor-
<<
Article
>> Home | TOC |
Index
tween condyle-pogonion distance and reading 3. Korn EL, Baumrind S: Randomized clinical trials with
comprehension in children between 6 and 16. It clinician-preferred treatment. Lancet 337:149-152, 1991
4. Paquette DE, Beattie JR, Johnston LE, Jr. A long-term
would be improper for an educator to infer from comparison of non-extraction and pre-molar extraction
such a finding that improved reading compre- edgewise therapy in "Borderline" Class II patients. Am J
hension causes jaw growth, just as it would be Orthod Dentofac Orthoped 102:1-14, 1992
improper for an orthodontist to conclude that 5. Cassidy DW, Herbosa EG, Rotskoff KS, et al: A compar-
mandibular growth improves reading compre- ison of surgery and orthodontics in "borderline" adults
with Class II, Division 1 malocclusions. Am J Orthod
hension. Rather, both measures increase sharply Dentofac Orthoped 104:455-70, 1993
through time as a consequence of physical and 6. Dean HT: The investigation of physiologic effects by the
mental growth. (To make the same point, a epidemiological method, in Moulton FR (ed): Fluorine
British biometrician pointed out in the 1920s and Dental Health. Washington, DC, American Associa-
that during the previous 100 years there had tion for the Advancement of Science, 1942, pp 23-31
7. Dean HT: Epidemiological studies in the United States,
been a highly significant association between the in Moulton FR (ed): Dental Caries and Fluorine. Wash-
rise in clergymen's wages and the consumption ington, DC, American Association for the Advancement
of beer. Clearly, it would have been inappropri- of Science, 1946, pp 5-31
ate without additional information to have con- 8. McClure FJ: Water Fluoridation: The Search and the
cluded that the vicars of the church had spent Victory. Bethesda, MD: US National Institute for Dental
Research 1970, pp 109-138
their salary increases on alcoholic beverages.) A 9. Brook PH, Shaw WC: The development of an index of
similar conceptual problem somewhat more dif- orthodontic treatment priority. Eur J Orthod 11:309-
ficult to grasp occurs when lecturers in orthodon- 320, 1989
tics speak of the dental compensations, which are 10. Shaw WC, Richmond S, O'Brien KD, et al: Indices of
presumed to occur as a consequence of certain treatment need and treatment standards. Br Dent J 170:
107-112, 1991
patterns of mandibular growth. Here it remains to 11. Richmond S, Shaw WC, Roberts CT, et al: The PAR
be established whether there does or does not Index (Peer Assessment Rating): Methods to determine
exist a causal relationship. Indeed, there may be outcome of orthodontic treatment in terms of improve-
such a relationship but whether or not there is ment and standards. Eur J Orthod 14:180-187, 1992
cannot be established by correlation alone. 12. Shaw WC, Richmond S, O'Brien K: The use of occlusal
indices: A European perspective. Am J Orthod Dentofac
This article has attempted to focus on some of Orthoped 107:1-10, 1995
the ideas involved in the use of statistics as a tool 13. Neilson, William Allan: Webster's New International Dic-
for increasing our understanding of the complex- tionary of the English Language. 2nd ed. Springfield,
ities of growth and treatment with regard to orth- MA, Merriam-Webster, 1953
odontics. Like orthodontics, statistic analysis has 14. Baumrind S, EL Korn, Isaacson Rf, et al: Quantitative
analysis of the orthodontic and orthopedic effects of
very real limitations. But properly used it does maxillary traction. Am J Orthod Dentofac Orthoped
allow us to identify, estimate, and to some extent 84:384-396, 1983
control for deficiencies in our capacity to predict 15. Baumrind S, EL Korn, Isaacson RJ, et al: Superimposi-
future events in an uncertain universe. This article tional assessment of treatment-associated changes in the
barely scratches the surface of the subject, but it temporo-mandibular joint and the mandibular symphy-
sis. Am J Orthod Dentofac Orthoped 84:443-465, 1983
does seek to point a direction. The clinician-ob- 16. Cohen J: Things I have learned (so far). Am Psychol
server needs to look very hard at the implicit and 45:1304-1312, 1990
explicit assumptions that underlie any statistical 17. Kazdin AE: The meanings and measurement of clinical
analysis. A basic understanding of the logic of sta- significance. J Consulting and Clinical Psychology 67:
tistical inference is much more important to the 332-339, 1999
18. Johnston LE Jr, Paquette DE, Beattie JR, et al: The
reader of the orthodontic literature than is the reduction of susceptibility bias in retrospective compar-
ability to manipulate or remember the numbers. 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
References Development, 1991, pp 155-77
1. Baumrind S, Korn EL, Boyd RL, et al: The decision to 19. Johnston LE Jr: Clinical studies in orthodontics: Taking
extract: Part 1—inter-clinican agreement. Am J Orthod the low road to Scotland, in Trotman C, McNamara JA Jr
Dentofac Orthoped 109:297-309, 1996 (eds): Orthodontic Treatment: Outcome and Effective-
2. Baumrind S, Korn EL, Boyd RL, et al: The decision to ness. Monograph 30, Craniofacial Growth Series, Center
extract: Part 2—analysis of clinicians' stated reasons. for Human Growth and Development. Ann Arbor, MI,
Am J Orthod Dentofac Orthoped 109:394-403, 1996 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
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
Available from
W.B. Saunders!
Churchill
Livingstone
Journal of Arthroplasty
Official Journal of the American Association
of Hip and Knee Surgeons
Journal of Electrocardiology
Official Journal of the International Society
for Computerized Electrocardiology and the
International Society of Electrocardiology
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
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.
W.B. SAUNDERS
Elsevier Science
EHS 01 DM166»« * Elm*» SCTCHC, ZWä
<<
Article
>> Home | TOC |
Index
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 The appearance of the code at the bottom of the first page
quarterly by W.B. Saunders. Months of issue are March, June, of an article in this journal indicates the copyright owner's
September, and December. Corporate and Editorial Offices: consent that copies of the article may be made for personal or
The Curtis Center, Independence Square West, Philadelphia, internal use, or for the personal or internal use of specific clients,
PA 19106-3399. Accounting and Circulation Offices: 6277 Sea for those registered with the Copyright Clearance Center, Inc.
Harbor Drive, Orlando, FL 32887-4800. POSTMASTER: Send (222 Rosewood Drive, Danvers, MA 01923; (508) 750-8400;
change of address to: Seminars in Orthodontics, W.B. Saunders, www.copyright.com). This consent is given on the condition that
Customer Service, 6277 Sea Harbor Drive, Orlando, FL 32887- the copier pay the stated per-copy fee for that article through the
4800. 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
Editorial correspondence should be addressed to: as copying for general distribution, for advertising or promotional
P. Lionel Sadowsky, DMD, BDS, DipOrth, MDent, Editor, purposes, for creating new collective works, or for resale.
Seminars in Orthodontics, Professor and Chairman, Department Absence of the code indicates that the material may not be
of Orthodontics, University of Alabama, 1919 Seventh Avenue processed through the Copyright Clearance Center, Inc.
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. Reprint inquiries should be addressed to Anne
Saunders, Periodicals Department, P.O. Box 628239, Orlando, Rosenthal, Elsevier Science, The Curtis Center, Independence
FL 32862-8239 or e-mail hhspcs@harcourt.com. Square West, Philadelphia, PA 19106-3399. Telephone (215)
Change of address notices, including both the old and new 238-5534, fax (215) 238-6423; e-mail: a.rosenthal@elsevier.com.
addresses of the subscriber and the mailing label, should be
sent at least 1 month in advance. Customer Service: 1-800-654-
2452 Advertising representative: M.J. Mrvica Associates, Inc,
2 West Taunton Ave, Berlin, NJ 08009. Telephone (609) 768-
9360. Fax (609) 753-0064.
Yearly subscription rates: United States and possessions: Publication of an advertisement in Seminars in Orthodontics
individual, $133.00; institution, $164.00; student and resident, does not imply endorsement of its claims by the Editor(s) or
$67.00; single issue, $50.00. All other countries: individual Publisher of the journal.
$166.00; institution, $198.00; student and resident, $83.00; The contributors have checked generic and trade names and
single issue, $50.00. For all areas outside the United States and verified drug doses for accuracy according to the standards
possessions, there is no additional charge for surface delivery. accepted at the time of publication. The ultimate
For air mail delivery, add $16.00. To receive student/resident responsibility, however, lies with the prescribing physician.
rate, orders must be accompanied by name of affiliated Please convey any errors to the Editor.
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. The ideas and opinions expressed in Seminars in
Prices are subject to change without notice. Current prices Orthodontics do not necessarily reflect those of the Editor or the
are in effect for back volumes and back issues. Single issues, Publisher. Publication of an advertisement or other product
both current and back, exist in limited quantities and are mention in Seminars in Orthodontics should not be construed as
offered for sale subject to availability. Back issues sold in an endorsement of the product or the manufacturer's claims.
conjunction with a subscription are on a prorated basis. Checks Readers are encouraged to contact the manufacturer with any
should be made payable to W.B. Saunders and sent to Seminars questions about the features or limitations of the products
in Orthodontics, W.B. Saunders, Periodicals Department, 6277 mentioned. Neither the Editor or Publisher assume any
Sea Harbor Drive, Orlando, FL 32887-4800. responsibility for any injury and/or damage to persons or
property arising out of or related to any use of the material
Copyright 2002, Elsevier Science (USA). All rights contained in this periodical. The reader is advised to check the
reserved. No part of this publication may be reproduced or appropriate medical literature and the product information
transmitted in any form or by any means, electronic or currently provided by the manufacturer of each drug to be
mechanical, including photocopy, recording, or any information administered to verify the dosage, the method and duration of
storage and retrieval system, without permission in writing administration or contraindications. It is the responsibility of
from the Publisher. Printed in the United States of America. 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.
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, Seminars in Orthodontics is indexed in the Cumulative
6277 Sea Harbor Drive, Orlando, FL 32887-4800. Telephone Index to Nursing and Allied Health Literature® print index
number: 1-407-345-2500. and the Cinahl® database.
W.B. SAUNDERS
<<
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
>> Home | TOC |
Index
Seminars in Orthodontics
EDITOR
P. Lionel Sadowsky, DMD, BBS, DipOrth, MDent
EDITORIAL BOARD
Richard G. Alexander, Arlington, TX James A. McNamara, Jr, Ann Arbor, MI
Rolf G. Behrents, Memphis, TN Robert N. Moore, Grand Island, NE
Samir E. Bishara, Iowa City, I A Ravindra Nanda, Farmington, CT
Robert Boyd, DBS, San Francisco, CA Perry M. Opin, Milford, CT
Larry M. Bramble, Cypress, CA Sheldon Peck, Newton, MA
John S. Casko, Iowa City, IA William R. Proffit, Chapel Hill, NC
Harry L. Dougherty, Van Nuys, CA Cyril Sadowsky, Chicago, IL
T.M. Graber, Evanston, IL David M. Sarver, Birmingham, AL
Robert J. Isaacson, Richmond, VA T. Michael Speidel, Minneapolis, MN
Alexander Jacobson, Birmingham, AL William J. Thompson, Bradenton, FL
Lysle E.Johnston, Jr., Ann Arbor, MI James L. Vaden, Cookeville, TN
Gregory J. King, Seattle, WA Robert L. Vanarsdall, Jr., Philadelphia, PA
Vincent G. Kokich, Tacoma, WA Katherine Vig, Columbus, OH
Steven J. Lindauer, Richmond, VA C.B. Preston, Buffalo, NY
INTERNATIONAL
Zeev Abraham, Herzliya, Israel Shinkichi Namura, Tokyo, Japan
W.G. Evans, Johannesburg, South Africa George Skinazi, Paris, France
Roberto Justus, Mexico City, Mexico 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%!
Mosby
A Division of Elsevier Science
© 2002 Mosby
<<
Article
>> Home | TOC |
Index
Seminars in Orthodontics
VOL 8, NO 3 SEPTEMBER 2002
Editorial 111
Timothy T. Wheeler
Seminars in Orthodontics
Future Issues
Recent Issues
Seminars in Orthodontics
VOL 8, NO 3 SEPTEMBER 2002
Introduction
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
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.
other individuals with a facial resemblance to ther specific genetic linkage studies (using DNA
Crouzon syndrome that occurred in the same markers) to determine areas of the genome that
family.4 The phenotype may be so variable that appear to be associated with the characteristics
an individual may appear to be clinically normal of a given trait.9
yet have the same gene mutation associated with A few points should be kept in mind when
Crouzon syndrome in three of his children and reviewing heritability estimates. First of all, they
two of his grandchildren. Only through cepha- refer to a specific sample and do not necessarily
lometry was a minimal expression of features pertain to a given individual even from within
suggestive of Crouzon syndrome evident.5 the sample. Thus, they do not allow one to tell to
The phenotypic variation present in these ex- what degree a particular trait was determined by
amples may be caused by modifying factors such genetic or environmental factors in a single in-
as environment and other (modifying) genes in dividual. In addition, heritability estimates are
the genome that interact with the effect of a descriptive of variances within a sample at a
specific mutation associated with a dominant given time, and they are not predictive.9 Herita-
trait. In fact, the concepts of variable expressivity bility estimates can change with age; for exam-
and reduced penetrance are applied to domi- ple, a longitudinal analysis of 30 sets of siblings
nant traits or conditions, acknowledging the po- that had not undergone orthodontic treatment
tentially variable phenotype that may not be ev- showed a significant increase in heritability esti-
ident at all in an individual with the gene mates between the ages of 4 and 14 years for 29
mutation.6 These examples give a clear message craniofacial skeletal variables, including in-
that even for a generally extreme autosomal creases for total anterior face height, upper an-
dominant phenotype, simply discovering the terior face height, total posterior face height,
gene mutation will very likely indicate that there and upper posterior face height. Despite the
will be an effect on craniofacial growth and de- general trend for all the craniofacial skeletal
velopment, but it does not give a precise picture variables to increase, there was a decrease for
of what that effect will be only what it may tend lower posterior face height. When a comparison
to be. 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
Estimating the Influence of Genetic and
traits. However, there was a decrease in the her-
Environmental Factors on Phenotype
itability estimate from the age of 14 for upper
A discussion of the methods and assumptions anterior face height and an increase for lower
made to estimate heritability, defined as the pro- posterior face height to that estimated at age 4
portion of the total phenotypic variance in a years.10
sample that is contributed by genetic variance,7 The heritability of a trait cannot necessarily
is beyond the scope of this article. For more be extrapolated from one sample and set of
information on these methods and assumptions, environmental conditions to another.7 An ad-
the reader may start with Genetics and Analysis of verse environment can alter the phenotypic ex-
Quantitative Traits by Lynch and Walsh.8 A trait pression that the genes would have promoted
with a heritability of 1 is said to be expressed under more favorable conditions. An extreme
without any environmental influence, whereas a example of this principle is the delayed growth
trait with a heritability of 0.5 would have half its seen from the effects of famine associated with
variability (from individual to individual) influ- war.11 Therefore, a high heritability does not
enced by environmental factors and half by ge- prevent a trait from being substantially influ-
notypic factors. Values over 1 may occur because enced by subsequent changes in environmental
the twin methodology provides an estimate of conditions in that sample.12
heritability, under several simplifying assump-
tions, that may be incorrect. Still, the estimation
Estimation Of Vertical Dimension
of heritability can provide an indication of the
Heritability
relative importance of genetic factors. Confirm-
ing that there is a certain degree of genetic Clinical consideration of the vertical dimension
influence on a trait is a preliminary step to fur- may include the evaluation of the ratio of the
<<
Article
>> Home | TOC |
Index
upper anterior face height to lower anterior face upper anterior face height (0.16) and posterior
height, as well as anterior to posterior face face height (0.26). These two relatively low her-
heights. Investigation of the anterior upper face itability estimates were probably artifacts caused
height to anterior lower face height in 30 by random variation in a limited sample. The
monozygotic twin pairs (ranging in age from likelihood of some random variation, expressed
12.0 to 18.8 years, with a mean of 15.9 years), in the path analysis used, was reinforced when
and 30 dizygotic like-sex twin pairs (ranging in markedly higher estimates of heritability for up-
age from 12.4 to 21.0 years, with a mean of 15.5 per anterior face height (0.81) and posterior
years) resulted in a heritability estimate of face height (0.88) were determined in the same
0.52.13 sample by using weighted means of monozygotic
Lateral cephalographs of 33 monozygotic and and dizygotic twin estimates instead of path anal-
46 dizygotic twins, who ranged in age from 9 to ysis.17
16 years (mean, 12.1 years) and had not under- Although attributed to random error in the
gone orthodontic treatment, were used in ge- path analysis method, the dichotomy of the her-
netic model fitting to determine the heritability itability between the upper anterior face height
of anteroposterior and vertical facial propor- and the lower anterior face height echoed the
tions.14 The analysis indicated that additive findings of an earlier study on 35 pairs of
genes and the specific environment influenced monozygotic twins and 21 pairs of like-sex dizy-
all the facial proportions. The heritability was gotic twins (ranging in age from 18-55 years,
0.71 for upper to lower anterior face height and with a median age of 24) in which there was a
0.66 for anterior to posterior face height. significant difference in the intrapair differences
The better-fitting model, with additive as op- (variance) between the monozygotic and dizy-
posed to dominant gene influence, indicates gotic twins for total anterior face height and
that genetic influence was the sum of some num- lower anterior face height but not for upper
ber of approximately equal gene effects. The anterior face height. The upper anterior face
specific environment aspect of the fitted model height was essentially the same between each
implies that the environmental influences were pair of twins, regardless of their zygosity. It was
of a more individual, as opposed to a more concluded that the dichotomy of the heritability
common environmental, nature. This is consis- between the upper anterior face height and the
tent with the prevailing concept that malocclu- lower anterior face height, along with the rela-
sion has a multifactorial origin (combination of tively high heritability of the total anterior face
a number of genetic and environmental factors) height, infers that it is the lower anterior face
and implies that specific environmental (treat- height that is primarily responsible for the her-
ment) factors might have some effect on the itability of the total anterior face height.18
traits. The dichotomy was also suggested when lat-
Analysis of the soft tissue associated with an- eral cephalometric measurements were made of
terior vertical height, as measured on the facial 67 monozygotic twin pairs and 29 dizygotic twin
profile of lateral photographs taken of 42 pairs pairs and investigated through factor analysis
of monozygotic twins and 37 pairs of dizygotic with subsequent estimation of heritability. The
twins, produced a heritability estimate of 0.66 factor with the largest heritability estimate
for anterior face height.15 Lateral cephalographs (0.76) included the lower anterior face height
were used in a path analysis study to compare and total anterior face height. The heritability
the heritability of horizontal and vertical dis- estimate for the factor that included upper an-
tances (as opposed to proportions or ratios) on terior face height was 0.48, whereas that for total
55 pairs of twins of the same gender, ranging posterior face height and lower posterior face
from 13 to 20 years of age who had not under- height was 0.31.19
gone orthodontic treatment.16 The mean age of If there are relatively high common inheri-
the monozygotic twins was 15.2 years, whereas tance estimates for the upper anterior face
the mean age for the dizygotic twins was 14.7 height and posterior face height as compared to
years. Although the heritability estimate for both the lower anterior face height and total anterior
the lower anterior face height and total anterior face height, this suggests that some common
face height was 0.86, it was markedly lower for (cultural) environmental factor(s) have greater
<<
Article
>> Home | TOC |
Index
influence on these traits.16 One conjecture subjects with an extreme malocclusion tend to
might be diet consistency20 on the posterior face be excluded.25 In a thought-provoking study of
height and the development of nasal airway pa- the heritability of cephalometric and occlusal
tency on the upper anterior face height; how- variables in siblings with overt malocclusions, it
ever, the presence of random variation may also was found that, in contrast to a series of similar
be an explanation for these relatively high com- subjects with naturally occurring good occlu-
mon inheritance estimates.16-17 In support of the sion, the heritability estimates for craniofacial
environmental effect on nasal development is skeletal variables in the subjects with overt mal-
the rejection of the null hypothesis that there is occlusions were significantly lower, and the her-
no common sibling effect on nasal height. These itability estimates for occlusal variations were sig-
results are based on path analysis of family re- nificantly higher.25 To quote King et al,25 "We
semblance using craniofacial anthropometric propose that the substantive measures of intersib
measurements of 1,763 individuals in 399 fami- similarity for occlusal traits reflect similar re-
lies from a rural community in Andhra Pradesh, sponses to environmental factors common to
India.21 both siblings. That is, given genetically influ-
Considering the effect of breathing on the enced facial types and growth patterns, siblings
vertical dimension, the most striking difference are likely to respond to environmental factors
was found in a comparison of cephalometric (eg, reduced masticatory stress, chronic mouth-
facial dimensions in 25 white children with pe- breathing) in similar fashions." Malocclusions
rennial allergic rhinitis. These apparent mouth appear to be acquired, but the fundamental ge-
breathers were compared with their 25 siblings netic control of craniofacial form predisposes
who apparently were not mouth breathers and siblings into comparable physiologic responses,
did not have perennial allergic rhinitis and 14 that often lead to development of similar maloc-
nasal breathing control subjects. The analysis clusions.25
revealed that the allergic children had a more
divergent facial pattern.22 This is consistent with
Does Knowing The Heritability Matter
the findings in a study of 100 11-year-old Finnish
In Treatment?
children in which there was an increase in the
vertical dimension in moderate and severely al- It has been stated that, "Variables with a lower
lergic subjects.23 An additional report with sim- genetic determination are more open to influ-
ilar findings was based on a study of 37 children ence by, for example, orthopedic correction
with perennial allergic rhinitis (ages 5-10 years) than are variables with a high genetic determi-
and matched controls.24 nation, which are not so easily changed by the
If an increase in total anterior face height and environment." 14 This implies that the genetic
lower anterior face height, in particular, are as- influence is a predetermining, unalterable fac-
sociated with perennial allergic rhinitis and tor. However, as has already been discussed, it
mouth breathing, why do some (although not may be altered under different environmental
all) studies indicate a dichotomy in the estimates (treatment) conditions. Certainly, treatment de-
of heritability of the upper anterior face height pends on the origin of a disorder if that cause is
and the lower posterior face height? One hy- known and specific. However, it has also been
pothesis is that the lower anterior face height pointed out that contrary to popular opinion the
may have a relatively greater heritability than the extent that a particular trait is influenced (or if
upper anterior face height in some individuals you wish even determined) by genetic factors
unless increased nasal obstruction, resulting in may have little if any effect on the success of
mouth breathing, becomes a predominating fac- environmental (treatment) intervention.26 What
tor. Again, heritability is a descriptive statistic for is important is the response of the individual to
a particular sample under defined environmen- the environmental (treatment) intervention,
tal conditions. which may be similar for comparable genotypes
Studies that estimate heritability of craniofa- (or at least the genes that are going to influence
cial structures may have a bias because they have the response to the particular intervention). Ge-
generally been performed with subjects who had netic factors that influence a trait may also in-
not undergone orthodontic treatment; thus, fluence the response to intervention designed to
<<
Article
>> Home | TOC |
Index
alter that trait, but that is not inherently known have had on a trait, which is a consideration with
in the estimation of the heritability of a trait. regard to the feasibility of a search for identify-
The question is, depending on the ability of the ing those factors. The search for DNA markers
individual to respond to a given environment linked with certain phenotypes may indicate ar-
(treatment), will the interaction of the new or eas of the genome that have a gene or genes that
altered environment, with the genetic factors influence the phenotype; however, this process
present, result in a change in the phenotype? 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-
Nature, Nurture, or Both? ever, the search for markers linked with certain
It has been stated that analyses of craniofacial phenotypes can indicate areas of the genome
structures have led to the conclusion that they that contain influential genes that were previ-
have moderate to high estimated heritabilities ously not known or even suspected to have an
and that they are primarily a consequence of influence on the phenotype. Once a particular
nature rather than nurture.25 In a sense, this is gene or genes in an area of the genome are
true for what estimates of the heritability repre- identified, they become candidate genes for spe-
sent, but it is often interpreted that genetic fac- cific analysis of their structure to pinpoint the
tors are influencing development independently relevant allele (s).
of the environment and that genetic factors have Study of the influence of particular genetic
controlled or determined the development. The factors on development may be performed by
basic interpretation of the estimation of herita- using a candidate gene chosen for the function
bility is that the genetic and environmental fac- of its associated protein. An example is a study of
tors are separate, thus they can be portioned and the association of the Pro561Thr (P56IT) vari-
do not interact. This has been typified by the ant in the growth hormone receptor gene
phrase nature versus nurture, which by its very (GHR), which is considered to be an important
construction defines a separation and even op- factor in craniofacial and skeletal growth. Out of
position. a normal Japanese sample of 50 men and 50
To say, for example, based on heritability es- women, those who did not have the GHR P56IT
timates of a particular sample that the anterior allele had a significantly greater mandibular ra-
face height is 70% genetic and 30% environ- mus length (condylion-gonion) than did those
mental gives a misleading dichotomy between with the G/ffiP56IT allele. The average mandib-
genetic and environmental factors and obscures ular ramus height, in those with the GHR P56IT
the fact that most if not all human disease (and allele, was 4.65 mm shorter than the average for
development) results from the interaction be- those without the GHR P56IT allele. This signif-
tween genetic susceptibility and environmental- icant correlation between the GHR P56IT allele
moderating factors.27 This is true even for con- and shorter mandibular ramus height was con-
ditions in which an environmental influence is firmed in an additional 80 women.29
known to be strong, such as in smoking and Interestingly, the association was with the
oropharyngeal cancer. Everyone who smokes mandibular ramus height but not mandibular
does not develop cancer, which indicates an in- body length, maxillary length, or anterior cra-
teraction of smoking with other factors, includ- nial base length. These data suggest an effect
ing genetic susceptibility.28 Essentially all aspects that is site, area, or region specific. Although it
of normal and abnormal development are in was concluded that the GHR P56IT allele may be
some way the result of the interaction of genetic associated with decreased growth of mandibular
and environmental factors; thus, there is no height and can be a genetic marker for it, it is
compelling reason to label a trait or condition as not clear if the effect is directly on the mandible
being either genetic or environmental.27 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
Searching For Genetic Factors and without the GHR P56IT allele, as a way of
Heritability estimates can indicate the relative looking at genetic and environmental interac-
contribution or influence that genetic factors tion.
<<
Article
>> Home | TOC |
Index
Summary and Conclusion of mice help estimate the number of genes that
influence a phenotype. The development of the
The relevance of analyzing the development of mouse genome project, not far behind the hu-
the vertical dimension to clinical practice is first man genome project, will increase the number
to determine if there is a vertical dimension of known DNA markers that may be used in the
component to the malocclusion and then ascer- study of putative relevant genetic factors and
tain what factors are having the greatest influ- genetic-environmental interactions, which may
ence on the vertical dimension problem in that then be tested for in the human population.
individual. Unfortunately, at this time, studies The human genome project resulted in not
on the genetic and environmental factors that only a single human genome sequence com-
influence the development of vertical dimension posed of overlapping parts from many humans
are representative of the samples studied and but also cataloged some 1.4 million sites of vari-
not necessarily of any particular individual. In ation in the human genome sequence. This
addition, the extent that a particular trait is in- increased number of variations (or polymor-
fluenced by genetic factors may have little if any phisms) may be used as markers to perform
effect on success of environmental (treatment) genetic (including genetic-environment interac-
intervention. It may be that genetic factors that tion) analysis in an outbred population such as
influenced a trait will also influence the re- human beings. Our genome varies from one
sponse to intervention to alter that trait, or other individual to the next, most often in terms of
genetic factors may be involved in the response. single-base changes of the DNA called single-
Therefore, the possibility for altering the envi- nucleotide polymorphisms. The main use of this
ronment to gain a more favorable dimension is human single-nucleotide polymorphism map
theoretically possible, even in individuals in will be to determine the contributions of genes
which the vertical dimension does have a rela- to diseases (or nondisease phenotypes) that
tively high genetic influence. However, the ques- have a complex, multifactorial basis. Although
tion of how environmental and genetic factors the scale of such studies could be daunting and
interact (a question that essentially cannot be there are still problems to solve, the potential for
answered in estimates of heritability) is most studying how natural variation leads to each one
relevant to clinical practice because it may ex- of our qualities is significant. This approach may
plain why a particular alteration of the environ- be the best opportunity yet to better understand
ment (treatment) in one compliant patient may the roles of nature and nurture rather than na-
be successful and not in another. Study of these ture versus nurture in development.32
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 Acknowledgment
genetic diversity from one patient to the next. The author thanks Dr W. Eugene Roberts for reviewing the
Animal studies using inbred strains compare manuscript and Ms. Robyn Tibbs, Ms. Madeline Hawkins,
and Ms. Claudette Maurer for retrieving and copying refer-
the different responses of an environmental fac- ences.
tor against consistent genotypes and the effect of
different background genotypes on the pheno-
typic expression of a specific gene mutation.30'31 References
Although mice, the most commonly used mam- 1. Baltimore D: Our genome unveiled. Nature 409:814-816,
malian species for genetic inbred strain studies, 2001
have been used and will continue to be used for 2. Mulvihill JJ: Craniofacial syndromes: no such thing as a
single gene disease. Nature Genet 9:101-103, 1995
the study of genes that cause disease and aber- 3. Park Wf, Bellus GA, Jabs EW: Mutations in fibroblast
rations in mammalian development, their differ- growth factor receptors: phenotypic consequences dur-
ent craniofacial morphology (ie, the presence of ing eukaryotic development. Am J Hum Genet 57:748-
snouts and single dentition with incisors and 754, 1995
molars only) may not be readily applicable to 4. Escobar V, Bixler D: On the classification of the acro-
cephalosyndactyly syndromes. Clin Genet 12:169-178,
some of the clinical questions orthodontists have 1977
regarding craniofacial growth in humans. Stud- 5. Everett ET, Britto DA, Ward RE, et al: A novel FGFR2
ies using mating crosses of various inbred strains gene mutation in Crouzon syndrome associated with
<<
Article
>> Home | TOC |
Index
apparent nonpenetrance. Cleft Palate-Craniofacial J 36: occlusion in world populations. Am J Orthod 86:419-
533-541, 1999 426, 1984
6. Jorde LB, Carey JC, Bamshad MJ, et al: Medical Genetics 21. Byard PJ, Poosha DV, Satyanarayana M, et al: Family
(ed 2). St. Louis, Mosby, 2000 resemblance for components of craniofacial size and
7. Goodenough U: Genetics (ed 3). Philadelphia, Saim- shape. J Craniofac Genet Dev Biol 5:229-238, 1985
ders, 1984 22. Trask GM, Shapiro GG, Shapiro PA: The effects of pe-
8. Lynch M, Walsh B: Genetics and Analysis of Quantitative rennial allergic rhinitis on dental and skeletal develop-
Traits. Sunderland, Sinauer, 1998 ment: a comparison of sibling pairs. Am J Orthod Dento-
9. LaBuda MC, Gottesman, II, Pauls DL: Usefulness of twin facial Orthop 92:286-293, 1987
studies for exploring the etiology of childhood and ad- 23. Hannuksela A: The effect of moderate and severe atopy
olescent psychiatric disorders. Am J Med Genet 48:47-59, on the facial skeleton. Eur J Orthod 3:187-193, 1981
1993 24. Sassouni V, Friday GA, Shnorhokian H, et al: The influ-
10. Harris EF, Johnson MG: Heritability of craniometric and ence of perennial allergic rhinitis on facial type and a
occlusal variables: a longitudinal sib analysis. Am J pilot study of the effect of allergy management on facial
Orthod Dentofacial Orthop 99:258-268, 1991 growth patterns. Ann Allergy 54:493-497, 1985
11. Tanner JM: Hormonal, genetic, and environmental fac- 25. King L, Harris EF, Tolley EA: Heritability of cephalomet-
tors controlling growth, in Harrison GA, Weiner JS, ric and occlusal variables as assessed from siblings with
Tanner JM, et al (eds): Human Biology (ed 2). Oxford, overt malocclusions. Am J Orthod Dentofacial Orthop
Oxford, University Press, 1977, pp 343 104:121-131, 1993
12. Vogel F, Motulsky AG: Human Genetics: Problems and 26. Smith RJ, Baut HL: Problems and methods in research
Approaches (ed 2). New York, NY, Springer-Verlag, 1986 on the genetics of dental occlusion. Angle Orthod 47:
13. Lobb WK: Craniofacial morphology and occlusal varia- 65-77, 1977
tion in monozygous and dizygous twins. Angle Orthod 27. Khoury MJ, Thrasher JF, Burke W, et al: Challenges in
57:219-233, 1987 communicating genetics: A public health approach.
14. Savoye I, Loos R, Carels C, et al: A genetic study of Genet Med 2:198-202, 2000
anteroposterior and vertical facial proportions using 28. Amador ÄG, Righi PD, Radpour S, et al: Polymorphisms
model-fitting. Angle Orthod 68:467-470, 1998 of xenobiotic metabolizing genes in oropharyngeal car-
15. Vanco C, Kasai K, Sergi R, et al: Genetic and environ- cinoma. Oral Surg Oral Med Oral Path 93:440-445, 2002
mental influences on facial profile. Aust Dent J 40:104- 29. Yamaguchi T, Maki K, Shibasaki Y: Growth hormone
109, 1995 receptor gene variant and mandibular height in the
16. Lundström A, McWilliam JS: A comparison of vertical normal Japanese population. Am J Orthod Dentofacial
and horizontal cephalometric variables with regard to Orthop 119:650-653, 2001
heritability. Eur J Orthod 9:104-108, 1987 30. Bourgeois P, Bolcato-Bellemin AL, Danse JM, et al: The
17. Lundström A, McWilliam J: Comparison of some ceph- variable expressivity and incomplete penetrance of
alometric distances and corresponding facial propor- the twist-null heterozygous mouse phenotype resemble
tions with regard to heritability. Eur J Orthod 10:27-29, those of human Saethre-Chotzen syndrome. Hum Mol
1988 Genet 7:945-957, 1998
18. Horowitz SL, Osborne RH, DeGeorge FV: A cephalomet- 31. Everett ET, Hartsfield JKJr: Mouse models for craniofa-
ric study of craniofacial variation in adult twins. Angle cial anomalies, in Davidovitch Z, MahJ (eds): Biological
Orthod 30:1-5, 1960 Mechanisms of Tooth Movement and Craniofacial Ad-
19. Nakata M, Yu PL, Davis B, et al: Genetic determinants of aptation. Boston, Harvard Society for the Advancement
cranio-facial morphology: A twin study. Ann Hum Genet of Orthodontics, 2000, pp 287-298
37:431-443, 1974 32. Chakravarti A: To a future of genetic medicine. Nature
20. Corruccini RS: An epidemiologic transition in dental 409:822-823, 2001
<<
Article
>> Home | TOC |
Index
tures—the long-face feature being more anaes- third of the face into an upper one third and a
thetic. Because of the challenge of the vertical lower two third. These divisions of the face can
dimension, the subject of this article is of ex- be used by the clinician to help diagnose vertical
treme importance to the orthodontic specialist. dimension problems. For example, does a pa-
Not only must the specialist recognize the prob- tient have a disproportionately long lower facial
lem, the specialist must understand the diagno- height because of vertical maxillary excess or to
sis of the problem so that all facets and compo- excessive chin height? Conversely, is a short fa-
nents of the vertical dimension enigma are cial height caused by vertical maxillary defi-
understood. The clinician must be able to rec- ciency or a short chin height?17 By using these
ognize the various components of a vertical di- accepted proportions as a guide, the patient
mension abnormality and understand the inter- shown in Figure 2 has an excessive lower ante-
relationship of all the elements of the problem. rior facial height, whereas the patient shown in
Before discussing the abnormal, it is prudent Figure 3 has diminished lower anterior facial
to understand the normal. Two of the most ac- height. Although it is evident that both have
cepted descriptions or publications of vertical vertical dimension abnormalities by looking at
facial proportions have been published by Fra- the face, measuring the facial proportions con-
kas14 and Frakas and Munro.15 In these, they firms this intuitive conclusion. A careful deter-
describe the ideal face as vertically divided into mination of the vertical proportions of the face
equal thirds by horizontal lines that approxi- is therefore the first step in the diagnosis of a
mate the hairline, the bridge of the nose, the ala vertical dimension problem.
of the nose, and menton (Fig I). 16 Figure 1 also
shows that in the ideal vertically proportioned
Role of Skeletal and Dental
face there is a further division of the lower one
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-
1/3 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,
1/3 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
1/3 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.
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 Condylar Growth
horizontal lines that approximate the hairline, the
bridge of the nose, and the ala of the nose and A common scenario affecting the skeletal prob-
menton. The lower third is further divided into an lem is mandibular growth and growth rotation,
upper third and a lower two third. which unfavorably impacts dentoalveolar devel-
<<
Article
>> Home | TOC |
Index
Figure 2. An example of a
patient with an excessive
lower facial height.
oprnent in both the maxilla and mandible. tients with long-face syndrome (Fig 6A and B)
Bjork18'21 and Bjork and Skieller23'24 have per- have a more posteriorly directed growth pattern
formed numerous studies that have shown that of the mandibular condyle (Fig 7). 26 > 27 These
the most common direction of condylar growth backward growth rotators have increased ante-
is vertical, with some anterior component. Pa- rior facial height, a more posterior position of
tients with a pronounced short lower anterior the chin, and in extreme cases, an anterior open
facial height (Fig 4A and B) generally exhibit bite may develop. Serial images of the patient
upward and forward condylar growth (Fig 5). taken to monitor the direction of condylar
These individuals generally have a deep vertical growth would be very useful for the diagnosis of
overbite with a deep mentolabial sulcus and a vertical growth. At the present time, serial imag-
strong overdosed appearance.25 In contrast, pa- ing poses certain concerns, most significantly
Figure 3. An example of a
patient with a diminished
lower facial height.
<<
Article
>> Home | TOC |
Index
radiation exposure. Advances in imaging tech- method of prediction of condylar growth rota-
nology may, in the future, permit the clinician to tion from a cephalogram offers the clinician
use these methods for diagnostic purposes with some guidelines. Bjork identified seven specific
greater safety. structural features that might develop as a result
An understanding of the maxillomandibular of remodeling during a particular type of growth
growth rotation of the patient would be most rotation. Bjork's suggestions for predicting con-
helpful in the diagnosis of vertical variations. dylar rotation have, however, not been widely
Bjork28 has contributed information that offers used by the specialty because (1) some of the
some guidelines for the clinician to assist in the indicators cannot be easily seen on the average
determination of the growth rotation of the cephalogram, (2) the use of the indicators is very
mandible so that the concomitant vertical time-consuming for the clinician, and (3) there
changes are more easily understood. Bjork's has been no scientific validation of the suggested
<<
Article
>> Home | TOC |
Index
Figure 5. An example showing upward and forward Anterior and Posterior Facial Height
condylar growth.
Vertical dimension skeletal abnormalities are
not solely caused by condylar growth direction.
indicators because of difficulties encountered in They are also caused by differences in anterior
study design. Some in the specialty also question facial height and posterior facial height develop-
whether several of the suggestions are valid in- ment. These differences in height development
dicators of a particular type of growth rotation. can lead to rotational growth or to changes in
However, when used for their intended purpose, mandibular position that greatly influence the
as guidelines only, the indicators have some use- position of the chin.30 Etiologies influencing un-
ful clinical applications in the diagnosis of the favorable differences in development of anterior
patient with vertical dysplasia (Table 1). and posterior facial height are multifactorial.
Using Bjork's guidelines, it is interesting to These factors can, for simplicity, be subdivided
study Figures 4B and 6B. Figure 4B, the forward into those caused by (1) dentoalveolar develop-
rotator, exhibits several of Bjork's indicators in- ment and (2) environmental factors.
cluding observations that (1) the condylar head
curves forward, (2) the mandular canal is
Dentoalveolar Development
curved, (3) the symphysis has a backward cant,
(4) the interincisal angle is obtuse and, (5) Issacson et al33 studied dentoalveolar develop-
lower anterior facial height is short. Figure 6B, ment in three groups of subjects—those with
Inclination of the condylar head Curves forward and back Straight or slopes up
Curvature of the mandibular canal Curved Straight
Shape of the mandibular lower border Curved downward Notched
Inclination of the symphysis (Anterior aspect just below "B" point) Slopes backward Slopes forward
Interincisal angle Vertical or obtuse Acute
Interpremolar or intermolar angles Vertical or obtuse Acute
Anterior lower face height Short Tall
<<
Article
>> Home | TOC |
Index
short anterior facial height, those with average short anterior facial height (low MP-SN angles).
anterior facial height, and those with excessive This difference of 5.1 mm of dentoalveolar de-
anterior facial height. The amount of maxillary velopment between the high angle and low an-
posterior alveolar development was found to de- gle groups is of significance.
crease as the MP-SN angle decreased. In patients Mandibular posterior alveolar development
with long anterior facial height (high MP-SN similarly decreased with decreases in the
angles), the mean distance from the occlusal MP-SN angle but much less dramatically than
plane to the inferior edge of the palate was 22.50 those found in the maxilla. Mandibular height
mm. This distance decreased to 19.6 mm for the showed a mean of 31.2 mm for the long ante-
average group and 17.1 mm for the group with rior face height group, 28.2 for the average
<<
Article
>> Home | TOC |
Index
14. Frakas LG: Anthropometry of the Head and Face in 33. Isaacson JR, Isaacson RJ, Speidel TM, et al: Extreme varia-
Medicine. New York, NY, Elsevier Science, 1981 tion in vertical facial growth and associated variation in
15. Frakas LG, Munro JR: Anthropometric Facial Propor- skeletal and dental relationships. Angle Orthod 41:219-
tions in Medicine. Springfield, IL, Charles C. Thomas, 228, 1971
1987 34. Janson G, Metaxas A, Woodside D: Variation in maxillary
16. Proffit WR: Diagnosis and Treatment Planning in Con- and mandibular molar and incisor vertical dimension in
temporary Orthodontics. St. Louis, MO: Mosby, 2000 12 year old subjects with excess, normal, and short lower
17. Sarver D, Proffit W, Ackerman J: Diagnosis and treat- anterior face height. Am J Orthod Dentofacial Orthop
ment planning in orthodontics from orthodontics, in 106:409-418, 1994
Graber TM, Vanarsdall RL (eds): Current Principles and 35. Moller E: The chewing apparatus. Acta Physiol 69:571-
Techniques (ed 3). St. Louis, MO, Mosby, 2000 574, 1966
18. Bjork A: Facial growth in man, studied with the aid of 36. Ingervall B, Thilander B: Relationship between facial
metallic implants. Acta Odontol Scand 13:9-34, 1955 morphology and activity of the mastecatory muscles.
19. Bjork A: Variations in the growth pattern of the human J Oral Rehap 1:131-147, 1974
mandible: longitudinal cephalometric study by the im- 37. Vig KWL: Nasal obstruction and facial growth: the
plant method. J Dent Res 400-411, 1963 strength of evidence for clinical assumptions. Am J
20. Bjork A: Sutural growth of the upper face studied by the Orthod Dentofacial Orthop 113:603-611, 1998
implant method. Acta Odontol Scand 24:109-129, 1966 38. Fields HW, Warren DW, Black K, et al: Relationship
21. Bjork A: The use of metallic implants in the study of between vertical dentofacial morphology and respira-
facial growth in children, method and application. Am J tion in adolescents. Am J Ortho Dentofacial Orthop
Phys Anthropol 29:243-254, 1968 99:147-154, 1991
22. Skieller V: Cephalometric analysis in the treatment of 39. Linder-Aronson S: Effects of adenoidectomy on the den-
overbite. Rep Congr Eur Orthod Soc 147-157, 1967 tition and facial skeleton over a period of five years, in
23. Bjork A, Skieller V: Facial development and tooth erup- Cook JT (ed): Transactions of the Third International
tion, an implant study at the age of puberty. Am J Orthodontic Congress. St Louis, MO, Mosby, 1975
Orthod 621:339-383, 1972 40. Woodside DG, Linder-Aronson S, Lundström A, et al:
24. Bjork A, Skieller V: Normal and abnormal growth of the Mandibular and maxillary growth after changed mode
mandible: A synthesis of longitudinal cephalometric im- of breathing. Am J Orthod Dentofacial Orthop 100:1-18,
plant studies over a period of twenty five years Eur 1991
J Orthod 5:1-46, 1983 4L Proffitt WR: The etiology of orthodontic problems. Con-
25. Neilsen I L: Vertical malocclusions: Etiology, develop- temporary Orthodontics (ed 3). St. Louis, MO: Mosby, p
ment, diagnosis and some aspects of treatment. Angle 136
Orthod 61:247-260, 1991 42. Schudy FF: The rotation of the mandible resulting from
26. Pearson LE: Vertical control in treatment of patients growth—Its implications in orthodontic treatment. The
having backward rotational growth tendencies. Angle Vertical Dimension of the Human Face. Houston, TX: D
Orthod 48:130-140, 1978 Armstrong & Co, 1992, pp 151-179
27. Schudy FF: Vertical growth vs anteroposterior growth as 43. Pearson LE, Pearson BL: Rapid maxillary expansion with
related to function and treatment. Angle Orthod 34:75- incisor intrusion: a study of vertical control. Am J
93, 1964 Orthod Dentofacial Orthop 115:576-582, 1999
28. Bjork A: Prediction of mandibular growth rotation. Am J 44. Klontz HA: Facial balance and harmony: an attainable
Orthod 55:585-599, 1969 objective for the patient with a high mandibular plane
29. Isaacson RJ: The geometry of facial growth and its effects angle. Am J Orthod Dentofacial Orthop 114:176-188,
on the dental occlusion and facial form. J Charles H. 1998
Tweed Int Found 9:21-38, 1981 45. Vaden JL: Alternative nonsurgical strategies to treat
30. Isaacson RJ, et al: Effects of rotational jaw growth on the complex orthodontic problems. Semin Orthod 2:90-113,
occlusion and profile Am J Orthod 3:276-286, 1977 1996
31. Schudy FF: The rotation of the mandible resulting from 46. Pearson LE: The management of vertical dimension
growth: Its implication in orthodontic treatment. Angle problems in growing patients, from the enigma of the
Orthod 35:36-50, 1965 vertical dimension. In McNamara JA Jr (ed): Craniofa-
32. Proffit WM: The development of orthodontic problems, cial Growth Series 36, Center for Human Growth and
in Proffitt WM, Fields HW (eds): Contemporary Orth- Development. Ann Arbor, MI, The University of Michi-
odontics. St. Louis, MO, Mosby, 2000, pp 102-106 gan, 2000
<<
Article
>> Home | TOC |
Index
Morphologic Characteristics
From the Department of Orthodontics, Baylor College of Den- The typical hyperdivergent open-bite (Hyp-OB)
tistry, The Texas A&M University System Health Science Center, patient presents with three-dimensional skeletal
Dallas, TX; Private Practice, Flower Mound, TX; and Department
of Orthodontics, University of Texas Health Science Center Houston, and dentoalveolar problems pertaining to both
Houston, TX. the maxilla and mandible. Variation in expres-
Address correspondence to Peter H. Buschang, Department of sion of traits among subjects should be ex-
Orthodontics, Baylor College of Dentistry, The Texas A&M Univer- pected. Within-subject differences clearly show
sity System Health Science Center, 3302 Gaston Avenue, Dallas, TX that most of the dysmorphology occurs in the
75246.
Copyright 2002, Elsevier Science (USA). All rights reserved. mandible.
1073-8746/02/0803-0004$35.00/0 Table 1 summarizes the problem list for the
doi:10.1053/sodo.2002.125432 maxilla. The most consistent findings across
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
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- report decreased posterior facial heights caused
toalveolar heights, which tend to be excessive. by smaller ramus heights. It has also been shown
When evaluated, the palatal plane angles have that the glenoid fossa is positioned more supe-
been reported to be flatter because of decreased rior relative to sella turcica. Mandibular dentoal-
anterior upper facial heights; posterior upper veolar heights are also generally reported to be
facial heights do not appear to be affected. excessive in Hyp-OB patients.
There is also a tendency for maxillas to be
shorter and slightly recessive. When transverse
Etiology of Hyp-OB Malocclusion
dimensions have been evaluated, the maxilla is
often reported to be narrow with an increased It is well established that abnormal muscles and
incidence of posterior cross-bites. habits are associated with Hyp-OB malocclu-
The most consistent mandibular characteris- sion.6 Finger, thumb, and tongue habits are per-
tics are increased lower anterior facial heights, haps the best known physical factors that pro-
steeper mandibular planes, and larger gonial duce open-bite malocclusions by lowering the
angles (Table 2). These traits combine to pro- mandible and preventing normal eruption.9 In-
duce increased lower to upper and lower to total dividuals with prolonged nonnutritive-sucking
anterior facial height ratios. Most studies also habits have been repeatedly shown to have a
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
>> Home | TOC |
Index
decrease overbite,10'14 increased overjet,10'11'13 features among individuals with genetically de-
decreased palatal width,11'15 and increased lip termined neuromuscular diseases that weaken
incompetence.12 masticatory muscles. Patients with myotonic dys-
There is considerable evidence showing trophy have 2 to 3 times less EMG activity of the
smaller, less active, muscles and weaker bite temporalis and masseter muscles during maxi-
forces among hyperdivergent subjects. Masseter mum clenching, lower maximum bite forces,
and medial pterygoid muscle volumes and cross- anterior open bite malocclusions, and hyperdi-
sectional areas have been positively correlated vergent growth patterns.32'33 Similarly, individu-
with posterior face and ramal heights and neg- als with spinal muscular atrophy have open-bite
atively correlated with the mandibular plane and malocclusions, increased vertical skeletal dimen-
gonial angles.16'21 Kiliaridis and Kalebo22 showed sions, steep mandibular plane angles, and hyper-
that women with thin masseter muscles had pro- divergent facial patterns.34 They also have
portionately longer faces. Bakke et al23 found weaker, less efficient, masticatory muscles that
that muscle thickness at the most voluminous, fatigue more quickly than those of matched con-
superficial portion of the masseter was signifi- trols.35
cantly correlated with bite force, anterior face Numerous experimental animal studies and
height, and mandibular inclination. Moreover, human clinical studies have also shown relation-
the amplitudes of masseter and temporalis elec- ships between the Hyp-OB phenotype and
tromyographic (EMG) activity at rest, during chronic upper, middle, and lower airway ob-
swallowing, chewing, and maximal biting have structions. Rhesus monkeys forced to become
also been negatively related with mandibular hy- mouth breathers showed greater than expected
perdivergence in children 9 to 11 years of age.24 increases in anterior face height associated with
Bakke and Michler25 showed that EMG activity
mandibular lowering, decreases in arch width,
during maximal voluntary contraction was neg-
and changes in tongue position and shape.36 In
atively correlated with anterior face height, man-
fact, both light and heavy nasal obstruction pro-
dibular inclination, vertical jaw relation, and go-
duces downward and backward mandibular ro-
nial angle.
tation, increased gonial angulation, and anterior
The smaller, less active muscles that charac-
open bite in primates37 and rodents.38
terize Hyp-OB subjects produce lower than nor-
mal bite forces.26 In adults, stronger molar and The relationship between airway obstruction
incisor bite forces are highly correlated with and skeletal hyperdivergence is well established
smaller gonial angles, lower mandibular plane in humans.39-40 Compared with their nose-
angles (MPA), and larger posterior face breathing counterparts, chronically allergic
heights.27'28 mouth-breathing children 6 to 12 years of age
The relationship between weaker bite forces have narrower maxillas; greater incidence of
and hyperdivergence is not as strong for chil- posterior cross-bites; longer anterior facial
dren as adults. Proffit and Fields29 originally heights; steeper palatal, occlusal, and mandibu-
showed no significant differences in occlusal lar planes; larger gonial angles; and more ret-
forces between long-face and normal children 6 rognathic mandibles.41 Mouth breathers with hy-
to 11 years of age. By using more refined analyt- pertrophied adenoids have narrow maxillas,
ical techniques, Garcia-Morales et al30 recently lower tongue positions, proclined incisors, and
found that children (age 9.3 ± 3.6 years) with increased lower anterior facial heights.42 Nor-
greater skeletal hyperdivergence had poorer me- malization of incisor inclination, maxillary arch
chanical advantage and lower maximum bite width, depth of the bony nasopharynx, and man-
forces. dibular plane inclination has been reported for
Although the association is well established, it children after adenoidectomies to correct severe
remains controversial whether lower occlusal nasopharyngeal obstruction.43'44 Children with
forces in long-face individuals produce hyperdi- enlarged tonsils also have more retrognathic
vergence26 or whether weaker occlusal forces are mandibles, larger lower anterior facial heights,
a biomechanical result of the long vertical facial and larger MPAs.45 The controversy concerning
proportions.31 This controversy is at least par- airway relates more to our ability to accurately
tially resolved by studies showing typical Hyp-OB diagnose airway problems for individuals rather
<<
Article
>> Home | TOC |
Index
than in the relationship between obstruction Hyp-OB patients because it has been shown to
and skeletal hyperdivergence. effectively hold maxillary sutural growth and ver-
The common denominator among Hyp-OB tical dentoalveolar development.48'53 Although
subjects with habits, weak muscles, and airway animal studies have reported absolute distal and
obstructions appears to be a lowered mandibu- superior displacement of metallic implants in
lar posture. An individual's response to airway the maxilla,54'55 human studies do not support
obstruction is immediate and predictable; the increased autorotation.49'56 In fact, Baumrind
activity of the neck and masticatory muscles, and coworkers51'57 found that high-pull patients
head posture, and mandibular posture change displayed relative increases in the MPA and re-
quickly.46'47 Habits and weak elevator muscles duced condylar growth.
might also produce changes in mandibular pos- Acrylic splints with HPHG create one large
ture that, during growth, necessitate develop- anchor unit that prevents unfavorable tipping of
mental adaptations leading to the Hyp-OB dys- the upper molars.54'55'58'61 Caldwell and collabo-
morphology. rators62 showed that this approach produced a
superior and distal displacement of the maxilla,
reductions of the Sella-Nasion-A (SNA) angle,
Treatment Modalities
clockwise rotation of the palatal plane, and rel-
Based on the forgoing, treatment of Hyp-OB ative intrusion of the upper molars. Importantly,
patients must address three-dimensional prob- they also reported increased lower molar erup-
lems pertaining to the dentoalveolar and skele- tion, decreased mandibular growth, and in-
tal structures of both jaws. Although hyperdiver- creased Sella-Nasion-B (SNB) angulation.
gent Class III open-bite cases are perhaps the
most difficult to treat nonsurgically, they occur Extractions
less frequently. In contrast, the Class II is the
Extraction therapy for hyperdivergent patients is
most common type of hyperdivergent patient. A
predicated on the belief that molars moved me-
severe hyperdivergent Class II patient requires
sially out of the occlusal wedge increase man-
reductions in dentoalveolar height throughout
dibular autorotation, decrease anterior facial
the maxilla and mandible (Table 3); reductions
height, and reduce open-bite malocclusions.
in gonial angulation; increased palatal plane an-
However, Yamaguchi and Nanda63 reported no
gulation; maxillary expansion; and true mandib-
differences in molar position or the A point-
ular autorotation to increase posterior mandib-
Nasion-B (ANB) angulation between extraction
ular height, redirect condylar growth, decrease
and nonextraction patients treated with high-
anterior lower facial height, and reposition the
pull face bows. Staggers64 showed no significant
chin forward (Fig 1).
differences between Class I nonextraction and
extraction cases in changes of anterior facial
High-Pull Headgear
height, upper to lower face height ratios, MPAs,
High-pull headgear (HPHG) has traditionally posterior to anterior facial height ratios, or the
been the appliance of choice for treating distances of the molars to the palatal and man-
<<
Article
>> Home | TOC |
Index
Dental
OPEN-BITE
MALOCLUSIONS Skeletal
Hyperdivergent
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- imal models67'73 and humans.74"81 However,
tients show decreases in the ANB angle, suprae- PBBs hinge the mandible open beyond its rest-
ruption of the lower molars, and increases in the ing position by varying amounts, which tends to
MPA.65 Pearson66 evaluating cases with moder- increase the gonial angle.69'72'79 McNamara79
ately steep MPAs treated with extractions and concluded that the maxillary complex was most
occipital headgears also reported significant su- affected by PBB, although changes have been
praeruption of the lower molars. Clearly, for- reported to occur throughout the craniofacial
ward mandibular rotation and loss or even main- complex.67
tenance of vertical facial and dentoalveolar Animal studies evaluating repelling magnets
heights does not occur in the extraction pa- embedded in bite-block appliances show superi-
tients. or-anterior maxillary displacement and molar
Combined HPHG and extraction treatment intrusion. However, they also show greater po-
produces results similar to extractions only. The tential for root resorption, deviated mandibular
primary differences pertain to the molars; the jaw posture that could produce skeletal asymme-
vertical movements of the upper molar are bet- tries, and lateral open bites.70'73'75'76 Kalra and
ter controlled, but the lower molar shows even Burstone,76 evaluating fixed magnetic bite
greater compensatory supraeruption.65'66 The blocks, reported increased mandibular length,
positive effects of combined HPHG and extrac- decreased facial convexity, intrusion of the lower
tion treatment are confined to the maxillary and upper molars, improved overjet and molar
dentoalveolar regions. If the goal is to improve relationships, and small decreases in the man-
the orientation, position, and shape of the man- dibular plane angle and y axis. After treatment,
dible, then other treatment approaches may be posterior tooth eruption closed the posterior
required. open bite, and the transverse jaw deviations self-
corrected. More pronounced treatment effects
Posterior Bite Blocks
have been reported for magnetic than spring-
Posterior bite blocks (PBB) have been shown to loaded bite blocks.80
effectively modify vertical skeletal patterns in an- Iscan and associates74 compared the effects of
<<
Article
>> Home | TOC |
Index
4. Parker JH: The interception of the open bite in the 25. Bakke M, Michler L: Temporaiis and masseter muscle
early growth period. Angle Orthod 41:24-44, 1971 activity in patients with anterior open bite and cranio-
5. Johnson ED, Larson BE: Thumb-sucking: Classification mandibular disorders. Scand J Dent Res 99:219-228,
and treatment. J Dent Child 60:392-398, 1993 1991
6. SubtelnyJD, Sakuda M: Open-bite: Diagnosis and treat- 26. Proffit W, Fields H, Nixon W: Occlusal forces in nor-
ment. Am J Orthod 50:337-358, 1964 mal- and long-face adults. J Dent Res 62:566-571, 1983
7. Sassouni V: A classification of skeletal facial types. Am J 27. Ringqvist M: Isometric bite force and its relation to
Orthod 55:109-123, 1969 dimensions of the facial skeleton. Acta Odont Scand
8. Schudy FF: Vertical growth versus anteroposterior 31:35-42, 1973
growth as related to function and treatment. Angle 28. Kiliaridis S, Johansson A, Haraldson T, et al: Craniofa-
Orthod 34:75-92, 1964 cial morphology, occlusal traits, and bite force in per-
9. Graber TM: Thumb and finger-sucking. Am J Orthod sons with advanced occlusal tooth wear. Am J Orthod
45:257-264, 1959 Dentofacial Orthop 107:286-292, 1995
10. Larrson E: Dummy- and finger-sucking habits with spe- 29. Proffit W, Fields H: Occlusal forces in normal and
cial attention to their significance for facial growth and long-face children. J Dent Res 62:571-574, 1983
occlusion. 4. Effects on facial growth and occlusion. 30. Garcia-Morales P, Buschang PH, Throckmorton GS, et
Sven Tandlak Tidskr 65:605-634, 1972 al: Maximum bite force, muscle efficiency and mechan-
11. Hawkins AC: A constructive approach to thumbsucking ical advantage in children with vertical growth pattern.
habit. J Glin Orthod 12:846-848, 1978 EurJ Orthod 2002
12. Bowden DB: A longitudinal study of the effects of digit- 31. Throckmorton G, Finn R, Bell WH: Biomechanics of
and dummy-sucking. Am J Orthod 52:887-901, 1966 differences. I. Lower face height. Am J Orthod 77:410-
13. Meisen B, Stensgaard K, PedersenJ: Sucking habits and 420, 1980
their influence on swallowing pattern and prevalence 32. Harvold EP, Vargervik K, Chierici G: Primate experi-
of malocclusion. EurJ Orthod 1:271-280, 1979 ments on oral sensation and dental malocclusions.
14. Nanda RS, Khan I, Anand R: Effect of oral habits on the Am J Orthod 63:494-508, 1973
occlusion in preschool children. J Dent Child 39:449- 32. Odman C, Kiliaridis S: Masticatory muscle activity in
452, 1972 myotonic dystrophy patients. J Oral Rehab 23:5-10,
15. Willmot DR: Thumb sucking habit and associated den- 1996
tal differences in monozygotic twins. Br J Orthod 11: 33. Kiliaridis S, Mejerso C, Thilander B: Muscle function
195-199, 1984 and craniofacial morphology: a clinical study in pa-
16. Weijs WA, Hillen B: Relationships between muscle tients with myotonic dystrophy. EurJ Orthod 11:131-
cross-section and skull shape. J Dent Res 63:1154-1165, 138, 1989
1984 34. Houston KD, Buschang PH, lannaccone ST, et al:
17. Gionhaku N, Lowe A: Relationship between jaw muscle Craniofacial morphological characteristics in patients
volume and craniofacial form. J Dent Res 68:805-809, with spinal muscular atrophy. Pediatric Res 36:265-269,
1989 1994
18. Hannam AG, Wood WW: Relationships between the 35. Granger M, Buschang P, Throckmorton G, et al: Mas-
size and spatial morphology of human masseter and ticatory muscle function in patients with spinal muscu-
medial pterygoid muscles, the craniofacial skeleton, lar atrophy. Am J Orthod 115:697-702, 1999
and jaw biomechanics. Amer J Phys Anthrop 80:429- 37. Yamada, T, Tanne K, Miyamoto K, et al: Influences of
445, 1989 nasal respiratory obstruction on craniofacial growth in
19. Van Spronsen PH, Weijs WA, ValkJ, et al: Relationships young Macaca fuscata monkeys. Am J Orthod Dentofa-
between jaw muscle cross-sections and craniofacial cial Orthop 111:38-43, 1997
morphology in normal adults, studied with magnetic 38. Scarano E, Ottaviani SE, DeGirolamo S, et al: Relation-
resonance imaging. Eur Orthod Soc 13:351-361, 1991 ship between chronic nasal obstruction and craniofa-
20. Van Spronsen PH, Weijs WA, ValkJ, et al: A Compari- cial growth: an experimental model. Int J Pediatr Oto-
sion of jaw muscle cross-sections of long-face and nor- rhinolaryngol 2:125-131, 1998
mal adults. J Dent Res 71:1279-1285, 1992 39. Subtelney JD: The significance of adenoid tissue in
21. Benington PC, Gardener JE, Hunt NP: Masseter muscle orthodontia. Ang Orthod 25:59-69, 1954
volume measured using ultrasonography and its rela- 40. McNamara JA: Influence of respiratory pattern on
tionship with facial morphology. EurJ Orthod 21:659- craniofacial growth. Angle Orthod 51:269-300, 1981
670, 1999 41. Bresolin D, Shapiro PA, Shapiro GG, et al: Mouth
22. Kiliaridis S, Kalebo P: Masseter muscle thickness mea- breathing in allergic children: its relationship to dento-
sured by ultrasonography and its relation to facial mor- facial development. Am J Orthod 83:334-340, 1983
phology. J Dent Res 70:1262-1265,1991 42. Linder-Aronson S: Adenoids: Their effect on mode of
23. Bakke M, Tuxen A, Vilmann P, et al: Ultrasound image breathing and nasal airflow and their relationship to
of human masseter muscle related to bite force, elec- characteristics of the facial skeleton and the denition.
tromyography, facial morphology, and occlusal factors. Acta Otolaryngologica Suppl 265:1-132, 1970
Scand J Dent Res 100:164-171, 1992 43. Linder-Aronson S: Respiratory function in relation to
24. Ingervall B, Thilander B: Relation between facial mor- facial morphology and the dentition. Brit J Orthod
phology and activity of the masticatory muscles. J Oral 6:59-71, 1979
Rehab 1:131-147, 1974 44. Woodside DG, Linder-Arsonson S, Lundström A, et al:
<<
Article
>> Home | TOC |
Index
Management of the Vertical Dimension 139
Mandibular and maxillary growth after changed mode nonextraction treatment on the mandibular position.
of breathing. Am J Orthod Dentofacial Orthop 100:1- Am J Orthod 100:443-452, 1991
18, 1991 64. Staggers JA: Vertical changes following first premolar
45. Behlfelt K, Linder-Aronson S, McWilliam J, et al: extractions. Am J Orthod Dentofacial Orthop 105:19-
Cranio-facial morphology in children with and without 24, 1994
enlarged tonsils. Eur J Orthod 12:233-243, 1990 65. Dougherty HL: The effect of mechanical forces upon
46. Hellsing E, Forsberg CM, Linder-Aronson S, et al: the mandibular buccal segments during orthodontic
Changes in postural EMG activity in the neck and treatment. Am J Orthod 54:29-49, 1968
masticatory muscles following obstruction of the nasal 66. Pearson LE: Vertical control through use of mandibu-
airway. Eur J Orthod 8:247-253, 1986 lar posterior intrusive forces. Angle Orthod 43:194-200,
47. Tourne LP, Schweiger J: Immediate postural responses 1973
to total nasal obstruction. Am J Orthod Dentofacial 67. Sergl HG, Farmand M: Experiments with unilateral bite
Orthop 110:606-11, 1996 planes in rabbits. Angle Orthod 45:108-114, 1975
48. Armstrong MM: Controlling the magnitude, direction, 68. Altuna G: The effect of excess occlusal force on the
and duration of extraoral force. Am J Orthod 59:217- eruption of the buccal segments and maxillary and
243, 1971 mandibular growth direction in the Macaca monkey.
49. Watson WG: A computerized appraisal of the high-pull Master's thesis, University of Toronto, Toronto, 1979
facebow. Am J Orthod 62:561-573, 1972 69. Altuna G, Woodside DG: Response of the midface to
50. Badell M: An evaluation of extraoral combined high- treatment with increased vertical occlusal forces. Treat-
pull traction and cervical traction to the maxilla. Am J ment and post-treatment effects in monkeys. Angle
Orthod 69:431-445, 1976 Orthod 55:251-263, 1985
51. Baumrind S, Korn, EL, Molthen, BS, et al: Changes in 70. Woods MG, Nanda RS: Intrusion of posterior teeth with
facial dimensions associated with the use of forces to magnets—an experiment in growing baboons. Angle
retract the maxilla. Am J Orthod 80:17-30, 1981 Orthod 58:136-150, 1988
52. Baumrind S, Korn EL, Isaacson RJ, et al: Quantitative 71. Woodside DG, Linder-Aronson S: Progressive increase
analysis of the orthodontic and orthopedic effects of in lower anterior facial height and the use of posterior
occlusal bite-block in its management, in Graber LW
maxillary traction. Am J Orthod 84:384-398, 1983
(ed): Orthodontics: State of the Art, Essence of the
53. Firouz M, Zernik J, Nanda R: Dental and orthopedic
Science, St Louis, MO, Mosby, 1986, pp 209-218
effects of high-pull headgear in treatment of Class II
72. Rowe, TK, Carlson, DS: The effect of bite-opening ap-
Division I malocclusion. Am J Orthod 102:197-205,
pliances on the mandibular rotational growth and re-
1992
modeling in the rhesus monkey (Macaca mulatta).
54. Elder JR, Tuenge RH: Cephalometric and histologic
Am J Orthod Dentofacial Orthop 98:544-549, 1990
changes produced by high-pull traction to the maxilla
73. Meisen B, McNamara JA, Hoenie DC: The effect of
in Macaca mulatta. Am J Orthod 66:599-617, 1974 bite-blocks with and without repelling magnets studied
55. Meldrum RJ: Alterations in the upper facial growth of histomorphologically in the rhesus monkey (Macaca
Macaca mulatta resulting from high-pull headgear. mulatta). Am J Orthod Dentofacial Orthop 108:500-
Am J Orthod 67:393-411, 1975 509, 1995
56. Worms FW, Isaacson RJ, Speidel TM: A concept and 74. Iscan HN, Akkaya S, Koralp E: The effects of the spring-
classification of centers of rotation and extraoral force loaded posterior bite block on the maxillo-facial mor-
systems. Angle Orthod 43:384-401, 1973 phology. Eur J Orthod 14:54-60, 1992
57. Baumrind S, Molthen R, West EE, et al: Mandibular 75. Kiliardis S, Egermark I, Thilander B: Anterior open bite
plane changes during maxillary retraction. Am J treatment with magnets. Eur J Orthod 13:447-457, 1990
Orthod 74:32-40, 1978 76. Kalra V, Burstone CJ: Effects of a fixed magnetic appli-
58. Thurow RC: Craniomaxillary orthopedic correction ance on the dentofacial complex. Am J Orthod Dento-
with en masse dental control. Am J Orthod 68:601-624, facial Orthop 95:467-478, 1989
1975 77. Dellinger EL: A clinical assessment of the active vertical
59. Joffe L, Jacobsen A: The maxillary orthopedic splint. corrector—A nonsurgical alternative for skeletal open-
Am J Orthod 75:54-69, 1979 bite. Am J Orthod Dentofacial Orthop 89:428-436, 1986
60. Fotis V, Meisen B, Williams S, et al: Vertical control as 78. Dellinger EL: Active vertical corrector treatment-Long-
an important ingredient in the treatment of severe term follow-up of anterior open bite treated by the
sagittal discrepancies. Am J Orthod 86:224-232, 1984 intrusion of posterior teeth. Am J Orthod Dentofacial
61. Orton HS, Slättery DA, Orton S: The treatment of Orthop 110:145-154, 1996
severe 'gummy' class II division I malocclusion using a 79. Lundström A, Woodside DG: Longitudinal changes in
maxillary intrusion splint. Eur J Orthod 14;216-223, facial type in cases with vertical and horizontal mandib-
1992 ular growth directions. Eur J Orthod 5:259-268, 1983
62. Caldwell SF, Hymas TA, Timm TA: Maxillary traction 79. McNamara JA: An experimental study of increased ver-
splint: A cephalometric evaluation. Am J Orthod 85: tical dimension in the growing face. Am J Orthod 71:
376-384, 1984 382-395, 1977
63. Cangialosi TJ: Skeletal features of anterior open bite. 80. Kuster R, Ingervall B: The effect of treatment of skeletal
Am J Orthod 85;28-36, 1984 open bite with two types of bite-blocks. Eur J Orthod
63. Yamaguchi K, Nanda RS: The effects of extraction and 14:489-499, 1992
<<
Article
>> Home | TOC |
Index
81. Barbre RE, Sinclair PM: A cephalometric evaluation of tation and angular remodeling during childhood and
anterior open bite correction with the magnetic active adolescence. Am J Human Biol 4:683-689, 1992
vertical corrector. Angle Orthod 61:93-109, 1991 100. Watanabe E, Demirjian A, Buschang PH: Longitudinal
82. Pearson LE: Vertical control in treatment of patients post-eruptive mandibular tooth movements of males
having backward-rotational growth tendencies. Angle and females. Eur J Orthod 21:459-468, 1999
Orthod 48:132-140, 1978 101. Naumann SA, Behrents RG, Buschang PH: Vertical
83. Pearson LE: Vertical control in fully-banded orthodon- components of overbite change: A mathematical
tic treatment. Angle Orthod 56:205-224, 1986 model. Am J Orthod Dentofacial Orthop 117:486-495,
84. Pearson LE: Treatment of a severe openbite excessive 2000
vertical pattern with an eclectic non-surgical approach. 102. Sankey WL, Buschang PH, English J, et al: Early treat-
Angle Orthod 61:71-76, 1991 ment of vertical skeletal dysplasia: The hyperdivergent
85. Haas AJ: A biological approach to diagnosis, mechanics phenotype. Am J Orthod Dentofacial Orthop 118:317-
and treatment of vertical dysplasia. Angle Orthod 50: 327, 2000
279-300, 1980 103. Tran MT, English JD, Throckmorton GS, et al: The
86. Majourau A, Nanda R: Biomechanical basis of vertical adjunctive treatment effects of light masticatory muscle
dimension control during rapid palatal expansion ther- training on hyperdivergent open-bite patients. A pilot
apy. Am J Orthod Dentofacial Orthop 106:322-328, study. 2002 (in press)
1996 104. Bell WH, Creekmore TD, Alexander RG: Surgical cor-
87. Alley TR: Social and applied aspects of face perception: rection of the long face syndrome. Am J Orthod 71:40-
An introduction, in Alley TR (ed): Social and Applied 67, 1977
Aspects of Perceiving Faces. Hillsdale, NJ, Lawrence 105. Isaacson JR, Isaacson RJ, Speidel TM, et al: Extreme
Erlbaum Assoc, 1988, pp 1-8 variation in vertical facial growth and associated varia-
88. Berschied E, Walster E, Bohrnstedt G: The happy tion in skeletal and dental relations. Angle Orthod
41:219-229, 1971
American body: A survey report. Pyschol Today 7:119-
106. Nahoum HI, Horowitz SL, Benedicto EA: Varieties of
131, 1973
anterior open-bite. Am J Orthod 61:486-492, 1972
89. Macgregor EC: Social and psychological implications of
107. Schendel SA, Eisenfeld J, Bell WH, et al: The long face
dentofacial disfigurement. Angle Orthod 40:231-233,
syndrome: Vertical maxillary excess. Am J Orthod 70:
1970
398-408, 1976
90. Dion KK, Berscheid E, Walster E: What is beautiful is
108. Lundström A, Woodside DG. A comparison of various
good. J Pers Soc Psychol 24:285-293, 1972
facial and occlusal characteristics in mature individuals
91. Shaw WC: The influence of children's dentofacial ap- with vertical and horizontal growth direction expressed
pearance on their social attractiveness as judged by at the chin. Eur J Orthod 3:227-235, 1981
peers and lay adults. Am J Orthod 79:399-415, 1981 109. Fields H, Proffit W, Nixon W, et al: Facial pattern
92. Kiyak HA, McNeill RW, West RA, et al: Personality differences in long-faced children and adults. Am J
characteristics as predictors and sequelae of surgical Orthod 85:217-223, 1984
and conventional orthodontics Am J Orthod Dentof 110. Haralabakis NB, Yiagtzis SC, Toutountzakis NM: Ceph-
Orthop 89:383-392, 1986 alometric characteristics of open bite in adults: A three
93. Bishara SE, Jakobsen JR: Longitudinal changes in three dimensional cephalometric evaluation. Int J Adult
normal facial types. Am J Orthod 88:466-502, 1985 Orthod Orthognath Surg 9:223-231, 1994
94. Nanda SK: Patterns of vertical growth in the face. Am J 111. Trouten JC, Enlow DH, Rabine M, et al: Morphologic
Orthod Dentofacial Orthop 93:103-116, 1988 factors in open bite and deep bite. Angle Orthod 53:
95. Nanda SK: Growth patterns in subjects with long and 192-211, 1983
short faces. Am J Orthod Dentofacial Orthop 98:247- 112. Lopez-Gavito G, Wallen TR, Little R, et al: Anterior
258, 1990 open-bite malocclusion: A longitudinal 10-year postre-
96. Lundström A, Woodside DG. Longitudinal changes in tention evaluation of orthodontically treated patients.
facial type cases with vertical and horizontal mandibu- Am J Orthod 87:175-186, 1985
lar growth directions. Eur J Orthod 5:259-268, 1983 113. Janson GR, Metaxas A, Woodside DG: Variation in
97. Garcia-Morales P, Buschang PH: Longitudinal stability maxillary and mandibular molar and incisor vertical
of divergent growth patterns. J Dent Res 81:388, 2002 dimension in 12-year-old subjects with excess, normal,
98. Karlsen AT: Craniofacial growth differences between and short lower anterior facial height. Am J Orthod
low and high MP-SN angle males: a longitudinal study. 106:409-418, 1994
Angle Orthod 65:341-350, 1995 114. Congialosi TJ: Skeletal morphologic features of ante-
99. Spady M, Buschang PH, Demirjian A: Mandibular ro- rior open-bite. Am J Orthod 85:28-36, 1984
<<
Article
>> Home | TOC |
Index
ontrolling the vertical dimension is critical creased lower anterior face height, a downward
C to functional appliance treatment because
clockwise rotation can have deleterious effects in
positioning of A-point, and an increase in the
total mandibular length for both the headgear
patients who have a high mandibular angle; like- and activator groups relative to the observation
wise counter clock rotation in those patients, group. There was also a significant increase in
who exhibit a low mandibular plane angle may the mandibular plane angle associated with ac-
be unstable. Numerous studies have reported on tivator treatment. Interestingly, this study did
the skeletal and dental vertical changes pro- "not support the hypothesis that activator treat-
duced by functional appliance treatment. ment can bring the mandible forward and affect
Jakobsson1 was among the first to point out condylar growth."1
that activator treatment altered the vertical di- Overall, the available literature is often times
mension, which was represented by an apparent the result of retrospective efforts that are com-
clockwise rotation of the palatal plane, an in- monly plagued by selection biases and insuffi-
ciently matched controls. It is also common to
From the Department of Orthodontics, College of Dentistry, Uni-
find relatively small experimental groups and a
versity of Florida, Gainesville, FL. narrow, if not shortsighted, selection of experi-
Supported by NIH/NICDR DE08715 mental parameters. In light of these and other
Address correspondence to Calogero Dolce, DDS, PhD, Depart- shortcomings, some findings seem to be fairly
ment of Orthodontics, Box 100444, JHMHC, Gainesville, FL constant among the reviewed investigations. Of
32610-0444.
Copyright 2002, Elsevier Science (USA). All rights reserved.
the studies that explored vertical ramus height
1073-8746/02/0803-0005$35.00/0 and/or vertical position of the lower molar (rel-
dot: 10.1053/sodo.2002.125433 ative to the mandible), there was a predominant
142 Dolce et al
agreement that both of these parameters in- to continue to wear the bionator or HG/BP
crease with early treatment involving either every other night for 6 months, followed by ob-
headgear or activator.2"8 With regard to anterior servation for 6 months. The nonretention
and posterior facial height, about half of the scheme lasted 1 year and began immediately
time the total anterior face height was studied after treatment. All subjects had the following
with no concurrent study of posterior facial data collection points (DC): (1) DC1, initial
height.1'5'6 When the proportion posterior lower records; (2) DC3, end of early Class II treatment
facial height to anterior lower facial height or observation; and (3) DC5, end of retention
was investigated, the ratio was greater than one period and observation period.
if a posttreatment observation period was in-
cluded.4'9'10 During treatment itself, a ratio less
Cephalometric Analysis, Calibration,
than one was seen. This means that posttreat-
and Digitization
ment changes in growth are occurring relative to
the observation group in the vertical dimension. All lateral cephalograms were taken on the same
This study evaluated the vertical treatment machine. The superimpositions of lateral ceph-
effects of the headgear/biteplane (HG/BP) and alometric tracings from DC 1 to DC5 were used
bionator during the early treatment of the Class to detect vertical changes in the sample popula-
II malocclusion. More specifically, the aim of tion. All skeletal measurements were performed
this work evaluated changes in (1) the posterior relative to an xy-axis superimposed on the struc-
and anterior facial height, (2) the alveolar ture of Sella (S) with the x component of the
height of the maxillary and mandibular molars, grid drawn parallel to the Frankfort horizontal
and (3) the orientation of the palatal and man- line. Additional templates were then con-
dibular plane as well as the molars that may structed for the maxilla, mandible, upper and
occur within the periods studied for all the lower molars, condyle, and gonial angle. The
groups. details on the template construction have previ-
ously been described.12 All points were regis-
tered with crosses on both the templates and the
Materials and Methods
final tracing for each film. Figure 1 depicts the
The details of this randomized clinical trial have angular measurements, and Figure 2 depicts the
been previously described by Keeling et al.11 linear measurements that were taken on the
Briefly, this was a prospective randomized clini- tracings.
cal trial designed to examine the timing of Class Before tracing, four crosshatches were estab-
II orthodontic treatment. The inclusion and ex- lished on each film, acetate paper was taped to
clusion criteria have already been described.11 A the left-hand margin, and the crosshatches were
stratified block randomization was used to assign transferred to the acetate. The dental and skel-
a subject into one of the following groups: etal templates were oriented independently on
bionator, HG/BP, or observation. Subjects were each radiograph by best-fit visual approxima-
stratified according to the following criteria: (1) tion, and the registration crosses of each struc-
severity of molar Class II malocclusion (mild, ture were transferred to the final tracing for
bilateral Vz cusp; moderate, one side 34 cusp; each film. In addition, points representing the
and severe, one side full cusp), (2) mandibular anterior and posterior extent of the functional
plane angle (low, <30°; normal, 30° - 40°; and occlusal plane (FOP) were marked on the ace-
high, >40°), (3) race, (4) gender, and (5) re- tate. The FOP was identified as the line of max-
tention. Each subject had complete orthodontic imum intercuspation of the first permanent mo-
records taken at every data collection point as lars, permanent first bicuspids, and/or primary
they progressed through the study. The end of first molars.
phase I treatment occurred when two project The following measurements were executed:
orthodontists independently agreed that a bilat- (1) mandibular plane (MP) angle, the angle
eral Class I molar relationship was achieved or 2 formed by the mandibular plane and the x-axis;
years had elapsed from the start of treatment. (2) palatal plane (PP) angle, the angle formed
Treated subjects were then randomized to reten- by the PP and the x-axis; (3) occlusal plane (OP)
tion or no retention. The retention scheme was angle, the angle formed by the FOP and the
<<
Article
>> Home | TOC |
Index
144 Dolce et al
cephalometric variables between DC l to DC5. A change was the eruption of the mandibular
P value of less than .05 was considered statisti- (L6MP) molars. Similar eruption of the mandib-
cally significant. ular molars occurred in both the HG/BP and
To standardize the data, the results presented bionator group when compared with the obser-
in Tables 1 through 4 are annualized, that is the vation group (Table 2). The observed skeletal
difference between time point 1 and time point changes were mainly in the HG/BP group and
2 is divided by the time between measurements included the increase in mandibular plane angle
to obtain a rate of change. This method was and lengthening of the UAFH. Skeletal changes,
chosen so as to correct for the time needed to which were observed in both the HG/BP and
achieve a given correction. In this manner, a bionator group, were increases in LAFH and
given amount of change is weighed more heavily TAFH. Again the HG/BG group showed the
if it occurred over a shorter period of time. greatest amount of change. TPFH increased
equally in the HG/BP and bionator groups
when compared with the observation group (Ta-
Results
ble 2).
Baseline characteristics for the 262 subjects are Tables 3 and 4 depicts the changes for the
presented in Table 5. With the exception of the total phase I time period (DC1-DC5). When the
alveolar height of the mandibular molar, all treatment phase and the 6 months of retention/
other measured variables were not significantly nonretention and 6 months of observation are
different at the beginning of treatment. The combined, many of the dental effects of treat-
mean active treatment (DC1-DC3) time, in ment for the headgear group have relapsed. For
years, for the bionator, HG/BP, and observation example, the change in the maxillary and man-
group was 2.2 ± 0.7, 2.1 ± 0.8, and 2.2 ± 0.4, dibular molar tip have disappeared. However, all
respectively. Although the total phase I (DC1- of the skeletal effects seen during treatment per-
DC5) treatment time, in years, for the bionator, sisted to the end of phase I for the HG/BP
HG/BP, and observation group was 3.4 ± 0.6, group with the exception of the anterior upper
3.2 ± 0.7, and 3.2 ± 0.4, respectively face height. All treatment effects for the biona-
The annualized skeletal and dental effects of tor group persisted to the end of the phase I
treatment (DC1-DC3) are shown in Tables 1 and time period.
2. Overall, the HG/BP group experienced sev- Because other factors besides treatment
eral statistically significant changes in the mea- group may play a role in cephalometric changes,
sured parameters that were not observed by in linear regression models were used to examine
the bionator group when compared with the the joint impact of treatment and gender, race,
observation group. The dental effects included mandibular plane angle, age, molar severity,
the distal tipping of the maxillary molars (U6 pre treatment, and retention during the DC1 to
tip) in the HG/BP group. Interestingly, the DC5 time period. Accounting for the additional
mandibular molar (L6 tip) tipped dis tally in all variables did affect the treatment group differ-
the three groups; however, the greatest amount ences by analysis of variance in three cases.
of change was seen in the HG/BP group (Table Change in mandibular plane angle did not dif-
1). Eruption of the maxillary (U6PP) molars fer significantly between the HG/BP group and
occurred in all three groups, the changes in the the other two groups after including the addi-
HG/BP group were statistically significant from tional variables. Tip back of the lower molar was
the other groups. The final observed dental affected by the following variables: retention
Table 1. Annualized Angular (°) Changes From the Treatment Phase (DC1-DC3)
Bionator HG/BP Control
Mean + SD Mean ± SD Mean ± SD P Value Difference
Palatal plane -0.02 -+- 0.6 0.03 ± 0.9 0.12 ± 0.7 .39
MP -0.23 -+- 0.8 0.50 ± 1.0 -0.39 ± 0.5 .0001 H/BC
Occlusal plane -0.31 •+- 1.3 0.01 ± 1.6 -0.51 ± 1.2 .06
U6 tip 0.35 -f- 1.9 -1.22 ± 2.9 0.84 ± 1.9 .0001 H/BC
L6 tip -0.42 + 2.3 -1.66 ± 2.3 -0.24 ± 1.4 .0001 H/BC
<<
Article
>> Home | TOC |
Index
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), upper and lower molar, the larger the baseline
pre treatment (P = .0413), and treatment group molar tip (the less it was tipped back), the more
with HG/BP differing from controls (P = it tipped back during phase I. The initial man-
.0177). UPFH change was affected by age at start dibular plane angle had a similar effect with the
of DC1 (P = .0245), and when additional vari- subjects with a moderate (between 30° and 40°)
ables were included, treatment group achieved mandibular plane angle and those with a high
only borderline significance (P = .0545). For all angle (greater than 40°) experiencing more tip-
other variables, the inclusion of additional vari- ping back of both the upper and lower molar
ables (covariates) did not change the treatment during the DC1 to DC5 time period. The larger
group differences we observed with univariate the baseline value for the UPFH, the smaller the
analysis. change was seen. In addition, male patients and
Regression models were developed to exam- those who had a longer span of time between
ine the impact of treatment group and other DC1 to DC5 experienced more of an increase in
variables (length of treatment, gender, race, the TPFH. Gender had an effect on the upper
mandibular plane angle, age, molar severity, anterior face height, with males experiencing
pretreatment, and retention) on the amount of more change. Also, the greater the amount of
change. The initial mandibular plane angle had time that elapsed between DC1 and DC5, the
an impact on the outcome for the change in the greater the amount of change seen for the an-
OP with those who initially had an average or terior upper, lower, and total face heights for all
high angle experiencing more change in the OP subjects
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
Discussion
upper and lower molar eruption. Those subjects The findings of this study show that HG/BP or
whose baseline lower molar height was larger bionator treatment produce vertical changes in
experienced more eruption, and those who had skeletal and dental structures, which are signifi-
retention experienced less mandibular molar cantly different from the observation group. Al-
eruption. Both the baseline value and initial though the resultant skeletal changes are stable
mandibular plane angle had a positive effect on 1 year after treatment irrespective of retention,
the amount of molar tip seen. For both the the dental changes have a propensity to relapse.
146 Dolce et al
Posterior Face Height tinuous, heavy distal forces to the upper molars
can alter resorption at craniofacial sutures in grow-
In this study, TPFH was depicted as the vertical
ing monkeys.13'14 Specifically, the zygomaticomax-
distance from the x-axis to condylion (UPDF) plus
illary, zygomaticotemporal, and the junction be-
the distance from condylion to gonion (LPFH).
tween the maxillary tuberosity, pterygoid plates of
Although statistical significance was not met for
UPFH and LPFH, the TPFH significantly in- the sphenoid, and the pyramidal process experi-
creased during active treatment with either ence more resorption with continual heavy head-
HG/BP or bionator. This skeletal change was last- gear force compared with an untreated observa-
ing because it was present at the end of the reten- tion group. Given these observations and the fact
tion phase. We speculate that the condyle de- that Jakobsson1 found a significant increase in the
scended relative to the x-axis to a greater extent UPFH for his headgear group, it is possible that
than the ramus increase in length for the following the glenoid fossa itself is affected by headgear
reasons. First, the bionator group nearly reached treatment in the human. Because the fossa itself is
significance for UPFH for both treatment (P — part of the temporal bone, it may be physically
.06) and overall differences that were detected moved or rotated by forces transmitted through
through the univariate analysis (P = .54) when the zygomatico temporal suture, thereby displacing
accounting for other variables. Second, the LPFH the condyle form the glenoid fossa enough to
changes did not compare with the UPFH. These cause an adaptive response. Alternatively, the tran-
results contradict other studies that have attrib- sitory bite opening that resulted from the molar
uted increases in TPFH to an increase in ramal tipping could trigger an adaptive response in the
length.2-4'15'16 However, they support Jakobsson's1 TMJ. Whatever the case, an increase in the TPFH
results showing the vertical descent of the condyle appears to be a lasting effect 1 year after headgear
during early treatment with the headgear or biona- treatment was discontinued.
tor. Similar, animal studies have shown that con- Whether this increase in the TPFH and pos-
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
>> Home | TOC |
Index
148 Dolce et al
altered occlusal vector from the upper molar. displacement secondary to the use of forces to retract
Alternatively, because the subjects who were in the maxilla. Am J Orthod 89:1-12, 1986
11. Keeling SD, Wheeler TT, King GJ, et al: Anteroposterior
the headgear group were instructed to wear the skeletal and dental changes after early Class II treatment
biteplane full time, it may be that the biteplane with bionators and headgear. Am J Orthod Dentofacial
itself has an effect on the lower molar tip. There Orthop 113:40-50, 1998
is no evidence in the literature that indicates 12. Keeling SD, Cabassa SR, King GJ: Systematic and ran-
that the biteplane wear can result in a tip back of dom errors associated with Johnston's cephalometric
analysis. Br J Orthod 20:101-107, 1993
the lower molar. The headgear's effect of tip- 13. Elder JR, Tuenge RH: Cephalometric and histological
ping back the molars may be responsible for the changes produced extra-oral high pull traction to the max-
increase of the LAFH, increase in the TAFH, illqa in Macaca Mulatta. Am J Orthod 66:599-617, 1974
and increase of the mandibular plane angle dur- 14. Sproule W: Dentofacial changes produced by extraoral
ing the active treatment phase (DC1-DC3). cervical traction to the maxilla of the Macaca Mulatta: A
histological and serial cephalometric study. Am J Orthod
56:532-545, 1969
Conclusions 15. McNamara JA, Bookstein FL, Shaughnessy TG: Skeletal
and dental changes following functional regulator ther-
Considerable variability is present in the vertical apy on class II patients. Am J Orthod 88:91-110, 1985
response to Class II orthodontic treatment with 16. Haynes S: A cephalometric study of mandibular changes
in modified function regulator (Frankel) treatment.
the bionator and HG/BP. The HG/BP groups Am J Orthod Dentofacial Orthop 90:308-320, 1986
experienced more changes in vertical parame- 17. Stockli PW, Willert HG: Tissue reactions in the temporo-
ters than the bionator or observation group. mandibular joint resulting from anterior displacement
Also, the observed skeletal changes remained of the mandible in the monkey. Am J Orthod 60:142-
stable until the end on phase I, whereas the 155, 1971
18. Hin ton RJ, McNamara JA: Temporal bone adaptations
dental changes seen at the end of active treat-
in response to protrusive function in juvenile and young
ment were lost by the end of phase I. adult rhesus monkeys (Macaca mulatta). Eur J Orthod
6:155-174, 1984
19. Woodside DG, Altuna G, Harvold E, et al: Primate ex-
References periments in malocclusion and bone induction. Am J
1. Jakobsson SO: Cephalometric evaluation of treatment Orthod 83:460-468, 1983
effect on Class II, division I malocclusions. Am J Orthod 20. Woodside DG, Metaxas A, Altuna G: The influence of
53:446-457, 1967 functional appliance therapy on glenoid fossa remodel-
2. Baumrind S, Korn EL, Molthen R, et al: Changes in ing. Am J Orthod Dentofacial Orthop 92:181-198, 1987
facial dimensions associated with the use of forces to 21. McNamara JA, Carlson DS: Quantitative analysis of tem-
retract the maxilla. Am J Orthod 80:17-30, 1981 poromandibular joint adaptations to protrusive func-
3. Bolmgren GA, Moshiri F: Bionator treatment in Class II, tion. Am J Orthod 76:593-611, 1979
division 1. Angle Orthod 56:255-262, 1986 22. Cook AH, Sellke TA, BeGole EA: Control of the vertical
4. Derringer K: A cephalometric study to compare the dimension in Class II correction using a cervical head-
effects of cervical traction and Andresen therapy in the gear and lower utility arch in growing patients. Part I.
treatment of Class II division 1 malocclusion. Part Am J Orthod Dentofacial Orthop 106:376-388, 1994
1—Skeletal changes. Br J Orthod 17:33-46, 1990 23. Baumrind S, Molthen R, West EE, et al: Mandibular
5. Mills JR: The effect of functional appliances on the plane changes during maxillary retraction. Am J Orthod
skeletal pattern. Br J Orthod 18:267-275, 1991 74:32-40, 1978
6. Chang HF, Wu KM, Chen KG, et al: Effects of activator 24. Creekmore TD: Inhibition or stimulation of the vertical
treatment on Class II, division 1 malocclusion. J Glin growth of the facial complex, its significance to treat-
Orthod 23:560-563, 1989 ment. Angle Orthod 37:285-297, 1967
7. Hultgren BW, Isaacson RJ, Erdman AG, et al: Mechanics, 25. Poulton DR: The influence of extraoral traction. Am J
growth, and class II corrections. Am J Orthod 74:388- Orthod 53:8-18, 1967
395, 1978 26. Kühn RJ: Control of anterior vertical dimension and
8. Derringer K: A cephalometric study to compare the proper selection of extraoral anchorage. Angle Orthod
effects of cervical traction and Andresen therapy in the 38:340-349, 1968
treatment of class II division 1 malocclusion. Part 27. Pearson LE: Vertical control in fully-banded orthodontic
2—Dentoalveolar changes. Br J Orthod 17:89-99, 1990 treatment. Angle Orthod 56:205-224, 1986
9. Vaden JL, Harris EF, Sinclair PM: Clinical ramifications 28. Schudy FF: Cant of the occlusal plane and axial inclina-
of posterior and anterior facial height changes between tion of teeth. Angle Orthod 14:69-82, 1963
treated and untreated Class II samples. Am J Orthod 29. Richardson ME: Mesial migration of lower molars in
Dentofacial Orthop 105:438-443, 1994 relation to facial growth and eruption. Aust Orthod J
10. Ben Bassat Y, Baumrind S, Korn EL: Mandibular molar 14:87-91, 1996
<<
Article
>> Home | TOC |
Index
Dentomaxillary
Center of Resistance
M p> M a
MP-Ma=Fd
Figure 3. Moments and vertical forces resulting from
Figure 2. High-pull headgear forces anterior or pos- placement of reverse curve of Spee archwires after
terior to the dentomaxillary center of resistance. extraction-site closure.
<<
Article
>> Home | TOC |
Index
when there is an inequality of uprighting mo- (Fig 4). Any constrained body (a tooth, a group of
ments, vertical forces occur, resulting in the poten- teeth, or an osseous structure joined to other os-
tial eruption of the posterior teeth. The spring rate seous structures through viable sutures) will react
primarily varies inversely as the cube of the inter- to the forces applied to it relative to its center of
bracket wire length across the extraction site(s) (R resistance.26'27 The locations of the centers of re-
a l/ L3). Thus, the rate is high, and the anterior sistance of the dentomaxillary complex have re-
and posterior moments easily become unequal be- cently been determined.12 They are located on a
cause the posterior and anterior tooth movement line drawn perpendicular to the functional occlu-
velocities are not necessarily equal. When the pos- sal plane (FOP) through the distal contacts of the
terior moment exceeds the anterior moment, an maxillary first molars as seen in a sagittal cephalo-
undesirable eruptive force will occur (Fig 3). This gram. They are further identified on the afore-
is undesirable in the patient already exhibiting an mentioned perpendicular line at one half the dis-
increased vertical dimension. The same problem tance from the FOP to a line drawn parallel to the
also exists with the placement of archwire V bends FOP through the inferior border of the orbit.
at the extraction site(s) related to closure. Again, There are two centers of resistance of the den-
the applied moments must be equal throughout tomaxillary complex when reviewed in the frontal
the range of activation, and because this approach aspect. This is because the dentomaxillary com-
invariably involves a high spring rate, it makes plex is essentially made up of two bones, a right
inequalities in the moments applied to the ante- and a left maxilla each containing one half the
rior and posterior teeth a likely occurrence as the dental arch. Each maxillary bone articulates with
anterior and posterior teeth approach each other. the other at the median palatine suture and rela-
To reduce this problem, frequent in-office appli- tively symmetrically on each side with the fron-
ance adjustments are required. One should also tomaxillary suture, the nasomaxillary suture, the
consider that the increased angulation of the arch- zygomaticomaxillary suture, and the transverse
wire at the brackets' interfaces might result in palatine suture. Because each of the protraction
significant undesirable frictional effects. forces (one on each side) results in a moment that
A controlled, efficient means of extraction- tends to stress the midpalatine suture greater at its
site closure has been outlined previously18"20 distal area than at its anterior area, each half of the
wherein moment alterations along with related dentomaxillary complex acts somewhat indepen-
vertical forces between the anterior and poste- dently. Thus, two centers of resistance are identi-
rior teeth are under control of the orthodontist fied. However, if protractive forces are applied in
throughout extraction-site closure. It should be the presence of a stiff 0.036-in stainless steel trans-
noted that in the most stringent requirement palatal arch or a sutural expander, the left and
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 <p = Dentomaxillary center of resistance
F = Protraction force applied to teeth
in the treatment of Class III malocclusions consists F' = F = Protraction force equivalent at center of resistance
of bilateral forces emanating from a face mask and M = F(Y) = Tipping moment at center of resistance
applied to the first molar or canine regions. The FOP = Functional occlusal plane
F(sin 6) = Eruptive component of protraction force
occlusogingival angle of the protraction forces in
the sagittal view is determined by the commissure Figure 4. The force system produced by a commonly
of the lips and the intraoral points of attachment used protraction device.
<<
Article
>> Home | TOC |
Index
class II traction, cause eruption of the mandib- extraoral appliances in light of recent research findings.
ular posterior teeth, thereby increasing the ver- Angle Orthod 69:81-84, 1999
14. Frankel R: A functional approach to orofacial orthope-
tical dimension. This is also accompanied by dics. Br J Orthod 7:41-51, 1980
rotation of the occlusal plane.33 Accordingly, 15. Harvold EP, Vargervik K: Morphogenic response to ac-
skeletal growth is critical to accommodate these tivator treatment. Am J Orthod 60:478-490, 1971
treatment changes. Use of Class II or Class III 16. Proffit WR, Fields HW Jr (eds): Contemporary Orth-
interarch elastics in patients with a vertical ex- odontics (ed 2). Philadelphia, PA, Mosby, 1992
cess is contraindicated. 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
Conclusion Orthod Dentofacial Orthop 97:495-504, 1990
18. Braun S: Diagnosis driven vs. appliance driven treatment
Before beginning active orthodontic therapy, it outcomes, in Sachdeva RC, Bantleon HC, White LW, et
is vital that the clinician clearly establish post- al (eds): Orthodontics for the Next Millenium. Glen-
treatment objectives for each patient. This in- dora, CA, Ormco Corp, 1997
cludes anterior facial height, mandibular plane 19. Marcotte MR: Biomechanics in Orthodontics. Philadel-
phia, PA, BC Decker, 1990
angle, location of the occlusal plane, and the an- 20. Burstone CJ: Modern Edgewise Mechanics and the Seg-
teroposterior position of the dentition, among mented Arch Technique. Glendora, CA, Ormco Corp,
others. Once this has been performed, a de- 1995
tailed individualized mechanical plan to support 21. Braun S, Sjursen RCJr, Legan HL: On the management
these objectives should be outlined. Various pro- of extraction sites. Am J Orthod Dentofacial Orthop
cedures and their biomechanical considerations 112:645-655, 1997
22. Tulley WJ: The role of extractions in orthodontic treat-
have been outlined for the clinician to consider ment. Br Dent J 107:199-205, 1959
to achieve these objectives. 23. Wyatt NE: Preventing adverse effects on the TMJ
through orthodontic treatment. Am J Orthod 91:493-
499, 1987
References 24. Kacaderali I: The effect of first premolar extraction on
1. Broadbent BH: The face of the normal child. Angle the vertical dimension. Am J Orthod Dentofacial Orthop
Orthod 7:183-207, 1937 116:47-50, 1999
2. Brodie AG: On the growth patterns of the human head 25. Casinano C, McLaughlin RP, Zernik JH: Effect of first
from the third month to the eighth year of life. Am J bicuspid extraction on facial height in high angle cases.
Anat 68:209-262, 1941 J Clin Orthod 27:594-598, 1993
3. Wylie W, Johnson E: Rapid evaluation of facial dysplasia 26. Nanda R, Bruce G: Biomechanical approaches to the
in the vertical plane. Angle Orthod 22:165-182, 1952 study of alterations of facial morphology. Am J Orthod
4. Tweed CH: A philosophy of orthodontic treatment. Am J 78:213-226, 1980
Oral Surg 31:74-103, 1945 27. Leone P: The effect of one type efface mask therapy on
5. Merrifield LL: Edgewise sequential directional force tech- the 'C' axis: The growth axis for the dentomaxillary
nology. J Charles H. Tweed Int Found 14:22-37, 1986 complex. Master's thesis, Saint Louis University, St
6. Shudy FF: Vertical growth vs. anteroposterior growth as Louis, MO, 2001
related to function and treatment. Angle Orthod 34:75-
28. Braun S, Hnat WP, Johnson BH: The curve of Spec
93, 1964
revisited. Am J Orthod Dentofacial Orthop 110:206-210,
7. Horn AJ: Facial height index. Am J Orthod 101:180-186,
1996
1992
29. Braun S: An examination of the effects of leveling with
8. Ellis E: The nature of vertical maxillary deformities.
nickel titanium archwires combined with torqued inci-
J Oral Maxillofac Surg 43:756-762, 1985
9. Fish LC, Wolford LM, Epker BM: Surgical-orthodontic sor brackets. Semin Orthod 7:215-220, 2001
correction of vertical maxillary excess. Am J Orthod 30. Bell WH, Proffit WR, Chase DL, et al: Mandibular defi-
73:241-257, 1978 ciency, in Bell WH, Proffit WR, White RP (eds): Surgical
10. Pancherz H, Fackel U: The skeletofacial growth pattern Correction of Dentofacial Deformities. Philadelphia, PA,
pre-and post-dentofacial orthopedics: A long term study Saunders, 1980, pp 684-843
of Class II malocclusions treated with the Herbst appli- 31. Guerroro CA, Bell WH, Meza LS: Intraoral distraction
ance. Eur J Orthod 12:209-218, 1990 osteogenesis. Atlas Oral Maxillofac Surg Clin North Am
11. Nanda SK: Patterns of vertical growth of the face. Am J 7:111-151, 1999
Orthod Dentofacial Orthop 93:103-116, 1998 32. Braun S, Rudman RT, Murdoch HJ, et al: The 'C' Axis:
12. Lee K, Ryu W, Park Y, et al: A study of holographic A growth vector for the maxilla. Angle Orthod 116:264-
interferometry on the initial reaction of the maxillofa- 270, 1999
cial complex during protraction. Am J Orthod Dentofa- 33. Braun S, Legan HL: Changes in occlusion related to the
cial Orthop 111:623-632, 1997 cant of the occlusal plane. Am J Orthod Dentofacial
13. Braun S, Lee K, Legan HL: A re-examination of various Orthop 111:184-188, 1997
<<
Article
>> Home | TOC |
Index
Vertical dimension of occlusion has been of height, tooth contacts, or other factors.1'2 When
interest for a long period of time in the the vertical dimension is increased, the stability
areas of prosthodontic reconstruction, orth- of this change is unclear and seems to be depen-
odontics and orthognathic surgery. It is interest- dent on whether the soft tissue adapts to this
ing to note that even though a number of spe- new position or the hard tissue adapts to the
cialties are involved in the manipulation of influences of the soft tissue (ie, relapse). The
vertical dimension, there have been very few purpose of this overview is to examine recent
studies on the impact of these changes on the literature to provide new insight into the
functional masticatory status of the patient. Most changes that may occur in the jaw muscles and
published studies provide empirical descriptions temporomandibular joint (TMJ) with changes in
of the clinical outcome of changing the vertical the vertical dimension.
dimension, but these studies lack the necessary
experimental design to adequately test whether
the effect was caused by changes in vertical What Is Vertical Dimension?
Vertical dimension by the simplest definition is
the vertical relationship between the maxilla and
From the Department of Orthodontics, College of Dentistry, Uni-
versity of Florida, Gainesville, FL. mandible. Terms such as vertical dimension of
Supported in part by grant No. DE 12207 from the National occlusion (VDO) and vertical dimension at rest
Institute of Dental and Craniofacial Research. (VDR) are prosthodontic terms that refer to the
Address correspondence to Charles G. Widmer, DDS, MS, Dept of vertical dimension measured with the maxillary
Orthodontics, Box 10044, JHMHSC, 1600 SW Archer Road, Uni-
and mandibular teeth in occlusion and at the
versity of Florida, Gainesville, FL 32610-0444.
Copyright 2002, Elsevier Science (USA). All rights reserved. postural rest position of the mandible respec-
1073-8746/02/0803-0007$35.00/0 tively. Vertical dimension can also be described
doi:10.1053/sodo.2002.125435 as lower facial height using the distance between
the anterior nasal spine (ANS) and gnathion. system associated with anxiety and stress, and by
These definitions indirectly describe the func- the central pattern generator for mastication
tional length of the jaw closing muscles either located in the reticular formation. It is common
when the teeth are in contact or in the rest to find that the mandibular postural position
position. It is important to emphasize the rela- can vary from day to day and even within the
tionship between the jaw muscles and maxillo- same day so that there is no specific measure
mandibular relationships because the jaw mus- that will accurately define this relationship.
culature acts as a primary determinant of vertical Many techniques have been used to estimate this
dimension or lower facial height.3 The defini- position particularly in edentulous patients and
tion of these measures must be modified, how- include functional assays such as speech patterns
ever, after mandibular surgical procedures such and swallowing techniques.7'8 Techniques that
as bilateral sagittal split osteotomy (BSSO) when use electrical stimulation over the right and left
the mandible is split into a proximal and distal sides of the face have been shown to only stim-
segment. The traditional measures of maxillo- ulate the superficial masseter muscle and do not
mandibular relationships do not reflect the mus- simulate the postural position of the mandible.9
cle length in this case because there is dissocia-
tion between the segment where the muscles are
VDO
attached (proximal segment) and the occlusal
or bony landmarks that are associated with the Muscular forces from the jaw muscles and, to a
distal segment. The relationship of the proximal lesser extent, tongue muscles, facial muscles,
segment to the cranial base would be a more and passive elastic properties of the soft tissues
accurate representation of the jaw closing mus- establish a vertical dimension of occlusion in a
cle functional length.4'5 The need for emphasiz- patient with a normal dentition. One example of
ing the relative position of the proximal segment this equilibrium can be seen when a slow attri-
has been recognized for a number of years, and tion of the dentition occurs secondary to
various methods have been proposed to position chronic bruxism and the vertical dimension of
this segment according to the treatment plan.5'6 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
VDR
occur with attrition of teeth in complete den-
The postural position of the mandible (VDR) tures because there cannot be a structural com-
while the patient is sitting or standing is main- pensation. In these cases, there is a true loss of
tained in an antigravity position by the jaw clos- vertical dimension that occurs over a number of
ing muscles including the masseter, temporalis, years and may result in masticatory musculoskel-
and medial pterygoid muscles. This postural po- etal pain in some patients.
sition or vertical dimension of rest is controlled The fact that vertical dimension of occlusion
by the activation of muscle spindles (stretch re- does not change significantly in the adult sug-
ceptors) that are located in parallel to the mus- gests that equilibrium is established between the
cle fibers. When the mandible is stretched open hard (teeth, alveolus, and compact bone) and
even by just a few tenths of a millimeter, these soft tissues (jaw muscles, tongue muscles, facial
exquisitely sensitive receptors will activate the muscles, and connective tissue). Any disruption
jaw-closing muscles to slightly close the jaw and of these components will affect the balance of
restore the mandibular postural position. These the forces, and change will occur in the form of
receptors also provide feedback to the central bony remodeling and/or soft-tissue adaptation.
nervous system (CNS) regarding muscle length To understand the role and adaptive capacity of
so that as the mandible is opened, the activation these components, it is necessary to first under-
of these receptors is more profound, and the stand normal form and function. Recent re-
CNS can receive information regarding the po- search has provided new insight into the normal
sition of the mandible. The sensitivity of the form and function of the masticatory muscles,
muscle spindle can be influenced by many fac- particularly the masseter muscle, and these new
tors such as peripheral pressure or pain in the findings may have a significant impact on the
oral region, by higher centers such as the limbic understanding of vertical dimension.
<<
Article
>> Home | TOC |
Index
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
>> Home | TOC |
Index
vertical dimension has a differential effect on ment of fibers is still unclear but may be related
stretching the jaw-closing muscle fibers such as to the different parts of the muscle that are more
the masseter. A recent study of different regions highly stretched (eg, the anterior masseter in
of the human masseter muscle has shown that which the fibers are generally slower in contrac-
the largest increases in muscle fiber length oc- tion speed and would maintain posture com-
curred in the medial part of the deep masseter pared with the posterior masseter in which the
followed by the anterior, superficial masseter, muscle is stretched the least and faster contract-
whereas the least change was found in the pos- ing fibers are located). Interestingly, the region
terior, superficial masseter.24 Similar differential in which the stretch is maximal is also the loca-
changes were also reported in the rabbit masse- tion of the muscle spindles in the masseter so
ter and, when sarcomere lengths were compared they are positioned ideally for detecting small
with the timing of activation of the muscle re- changes of mandibular posture.
gion, the optimal sarcomere length (2.4-2.8 jam) The MyHC phenotype profile within the mas-
coincided with the peak electromyographic ac- seter muscle has been found to be different
tivity.25 In summary, after an acute increase in between genders. The male masseter has a
vertical dimension, the stretched muscle fiber higher proportion of faster MyHC isoforms than
can generate higher forces because of optimiz- the female masseter.30 Although a logical func-
ing sarcomere length even without changing the tional explanation may be considered in the
activation pattern of the muscle and could con- larger mammals in which the male has larger
tribute to early adaptive changes of hard tissues canines and uses his teeth for fighting, it is not as
because of the increased load. 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
Myosin Composition of the Masseter
be dependent on the hormone testosterone.32
Muscle
For example, if a young male rabbit is castrated
Myosin heavy chain (MyHC) is one of the con- without testosterone supplementation, the
tractile proteins found in muscle fibers and is MyHC phenotype profile of his masseter muscle
associated with contraction speed as well as the when he reaches adult is no different from that
fatigue properties of the muscle fiber. Nine dif- observed in female rabbits. These differences in
ferent MyHC isoforms have been identified and male and female jaw-muscle contractile charac-
consist of developmental isoforms (embryonic teristics may affect the timing and amount of
and neonatal); slow, fatigue-resistant isoforms skeletal relapse observed after orthognathic sur-
(type I, a-cardiac); and fast isoforms (type Ila, gical experiences. Other hormones such as thy-
IIx, lib, Urn, Hextraocular). Type Urn and Hextraocular roid hormone also can change the phenotype of
are specialized isoforms that have been found in muscles, causing slower MyHC phenotypes to
masticatory muscles of cats and canines (Urn) switch to faster isoforms when given in large
and extraocular and laryngeal muscles.26'28 The doses.33
presence of different MyHC isoforms in a par- Another method that promotes muscle fiber
ticular muscle will vary depending on a variety of phenotype switching is changing the activity pat-
factors including the particular species, the func- tern of the muscle. For example, slow muscles
tional demand of the muscle, and hormonal have a prolonged firing pattern that extends
status of the animal. Generally, for jaw-closing over many hours usually for postural mainte-
muscles, small species such as mice and rats have nance, whereas fast twitch fibers fire in brief
fast MyHC type Ha, IIx, and lib isoforms that are bursts. When a fast muscle is activated for a
arranged in discrete locations within the mus- prolonged period similar to slow muscle activity,
cle.14 Larger species such as rabbits, pigs, and the phenotype of the fast muscle eventually
humans have slower, more fatigue-resistant fi- switches to a MyHC phenotype that is slower in
bers such as type I, a-cardiac, Ha, and IIx fibers contraction time and more fatigue resistant.34
that are generally concentrated with the slower Phenotype switching is one means by which mus-
fibers located more anteriorly and faster con- cles such as the masseter can adapt to the func-
tracting fibers located more posteriorly.29'31 The tional demands of the masticatory system. For
functional implication for this general arrange- example, placement of a bite-raising appliance
<<
Article
>> Home | TOC |
Index
in the pig increases the proportion of slow caution. All three animal models bruxed in an
MyHC expression and is probably because of the attempt to remove the appliance. This increased
increased grinding to attempt to remove the muscle activity in the jaw-closing muscles may
appliance (see later). have been a primary factor in the changes ob-
served in the MyHC isoform switch to a slow
phenotype. Prolonged activity levels over ex-
Early Effects of a VDO Increase on the tended periods of time such as with bruxism will
Jaw-Closing Muscles
influence phenotype switching as described ear-
Any appliance or prostheses that causes an in- lier. So it appears that an increase in vertical
crease in VDO has the potential to cause dimension along with an increase of jaw-closer
changes in the jaw-closing muscles. There are muscle function such as bruxism will promote
two lines of experimental evidence that are MyHC phenotype switching to a slower isoform.
markedly different and are dependent on
whether the vertical dimension was increased
Does Increasing the Vertical Dimension
but the muscle was splinted as in intermaxillary
Cause Bruxism?
fixation or whether the VDO was increased and
the jaw was freely moving, such as the use of an Placement of an interocclusal appliance will af-
interocclusal appliance. Immobilization of the fect many components including the vertical di-
muscle with an increase in muscle length results mension, tooth contacts, tongue posture, and
in the formation of new sarcomeres at the ends the patient's awareness.40 The response to any of
of the muscle fiber and atrophy of the muscle these factors may be a modification of the nor-
fibers. This atrophy is independent of the mus- mal muscle activity pattern either reflexively
cle fiber phenotype because both type I and type (such as to maintain mandibular posture and a
II fibers were affected.35 When functional length patent airway) or voluntarily. In the case of the
of the muscle was restored after the removal of animal studies, the intent to remove the foreign
the splint, the muscle fiber reduced the number material by bruxing was a primary behavioral
of sarcomeres to adapt the muscle fiber to the drive of the animal. Bruxing behavior has not
original length, and the atrophic fibers were been a significant response by patients that use
restored to their original size. Alternatively, al- an interocclusal appliance, and, in fact, this type
lowing the mandible to be freely movable with of appliance tends to initially reduce jaw-closing
function on the appliance seems to have the muscle activity with a gradual return to pretreat-
effect of adding sarcomeres to lengthen the ment levels.41
muscle fiber as well as resulting in the hypertro-
phy of the muscle fiber.
Vertical Dimension Increases and
Based on recent studies, changes at the mo-
the TMJ
lecular level also occur fairly rapidly in muscle
fibers after placement of a bite-raising appli- There is a paucity of studies that have examined
ance. For example, studies using these appli- the effects of increasing the vertical dimension
ances in rats and pigs have shown that an up- on the TMJ. Studies that have cemented interoc-
regulation in the RNA message for slower MyHC clusal appliances over the teeth of animals cause
isoforms occurs within a few days after appliance the animal to brux, and this increase in force
use.36'37 This is now known to be caused by the generation will be reflected by changes of load
release of mechano-growth factor (MGF), an iso- on the condyle.42 Indeed, changes were ob-
form of insulin growth factor-1, which is pro- served in the chondroblastic and prechondro-
duced locally in the muscle fibers and is signifi- blastic layers of the condylar cartilage support-
cantly upregulated in exercised and damaged ing a bony remodeling secondary to increased
muscle.38'39 Within 7 to 14 days, new MyHC pro- load. However, it is difficult to determine if the
tein is formed in muscle fibers, changing the condylar changes observed in these animals are
profile of the muscle toward a slower muscle. because specifically of the increase in vertical
However, interpretation of these results as they dimension or to the change in the loading pat-
apply to the human jaw-closing muscle response tern in the joint because of bruxing. In addition,
to a bite-raising appliance should be made with it is possible that biomechanical changes such as
<<
Article
>> Home | TOC |
Index
an increased jaw-closing torque because of an of increasing torques produced about the TMJ.
increase in mandibular height can produce an These factors contribute to the active mainte-
altered function in the TMJ (see earlier). Fur- nance of the vertical dimension. Therefore, the
ther studies are necessary to distinguish the ef- key to successful stabilization of facial skeletal
fect of vertical dimension on the temporoman- morphology after increasing the vertical dimen-
dibular joint. sion may lie in understanding the complex bio-
mechanical effects of the jaw muscles at multiple
levels and controlling the adaptive change of
Can Skeletal Stability be Maintained muscle and bone.
After Increasing the VDO?
Vertical relapse is observed after orthognathic
surgical procedures that increase the vertical di- References
mension and seems to occur at different rates 1. Sato S, Hotta TH, Pedrazzi V: Removable occlusal over-
depending on the type of fixation that was used. lay splint in the management of tooth wear: A clinical
report. J Prosthet Dent 83:392-395, 2000
Rigid fixation has a lower amount of relapse, at 2. Ormianer Z, Gross M: A 2-year follow-up of mandibular
least initially43 and may be because of the de- posture following an increase in occlusal vertical dimen-
mand for muscle fibers to function at a new, sion beyond the clinical rest position with fixed restora-
longer length, thus requiring additional sarco- tions. J Oral Rehabil 25:877-883, 1998
meres to be added as an adaptive response. The 3. Proffit WR, Turvey TA, Phillips C: Orthognathic surgery:
A hierarchy of stability. Int J Adult Orthodon Orthog-
lengthening of the muscle fibers will also nor- nath Surg 11:191-204, 1996
malize the output of the muscle spindles. Wire 4. Leonard M: Preventing rotation of the proximal frag-
fixation, however, allows the bony segments to ment in the sagittal ramus split operation. J Oral Surg
be flexed initially and allows muscle fibers to 34:942, 1976
function nearer their original length so that lit- 5. Epker BN, Wylie GA: Control of the condylar-proximal
mandibular segments after sagittal split osteotomies to
tle muscle adaptation would be required. In this advance the mandible. Oral Surg Oral Med Oral Pathol
case, the bony segments would adapt rather than 62:613-617, 1986
the muscle. After 5 years, however, it appears 6. Burye MT, Stella JP: An innovative method for accurate
that the vertical dimension stabilizes at the same positioning of the proximal segment in sagittal split
vertical height regardless of the type of fixa- osteotomies. Int J Adult Orthodon Orthognath Surg
15:59-63, 2000
tion.44 Therefore, the mechanism for long-term 7. Miralles R, Dodds C, Palazzi C, et al: Vertical dimension.
stabilization of the facial skeleton is still unclear Part 1: Comparison of clinical freeway space. Cranio
but is probably because of the equilibrium of 19:230-236, 2001
forces from a variety of muscles including the 8. Woda A, Piochon P, Palla S: Regulation of mandibular
jaw closers, the tongue, and the facial muscles. postures: Mechanisms and clinical implications. Grit Rev
Oral Biol Med 12:166-178, 2001
9. Dao TTT, Feine JS, LundJP: Can electrical stimulation
Summary be used to establish a physiologic occlusal position. J
Prosthet Dent 60:509-514, 1988
It has been hypothesized that increasing the 10. Schumacher GH: Funktionelle morphologic der kau-
vertical dimension will cause an increase in jaw- muskulatur. Jena: VEB Gustav Fischer, 1961
11. Weijs WA, Dantuma R: Functional anatomy of the mas-
closing force because of active and passive forces ticatory apparatus in the rabbit. Neth J Zool 31:99-147,
of the muscles and soft tissues. However, it has 1981
never been clear how this increased force is 12. Herring SW, Wineski LE, Anapol FC: Neural organiza-
generated. Based on our current knowledge, an tion of the masseter muscle in the pig. J Comp Neurol
increased vertical force is caused by the combi- 280:563-576, 1989
13. Widmer CG, Klugman D, English AW: Anatomical par-
nation of tonic low-level muscle activity caused titioning and nerve branching patterns in the adult rab-
by muscle spindle activation combined with the bit masseter. Acta Anat (Basel) 159:222-232, 1997
increased force produced by the stretched mus- 14. Eason JM, Schwartz GA, Pavlath GK, et al: Sexually di-
cle fiber. However, muscle fibers are not morphic expression of myosin heavy chains in the adult
stretched homogeneously within a jaw muscle as mouse masseter. J Appl Physiol 89:251-258, 2000
15. Widmer CG, Morris-Wiman JA, Nekula C: Spatial distri-
seen in the masseter muscle so the effect of bution of myosin heavy chain isoforms in mouse masse-
muscle stretch is regional. Finally, increasing the ter. J Dent Res 81:33-38, 2002
mandibular height has the biomechanical effect 16. English AW, Letbetter WD: Anatomy and innervation
<<
Article
>> Home | TOC |
Index
patterns of cat lateral gastrocnemius and plantaris mus- 32. Eason JM, Schwartz G, Shirley KA, et al: Investigation of
cles. Am J Anat 164:67-77, 1982 sexual dimorphism in the rabbit masseter muscle show-
17. English AW, Carrasco DI, Widmer CG: Torques pro- ing different effects of androgen deprivation in adult
duced by different compartments of the rabbit masseter and young adult animals. Arch Oral Biol 45:683-690,
muscle. J Appl Biomech 15:348-360, 1999 2000
18. Mosolov NN: On the anatomy of human masticatory 33. Baldwin KM, Haddad F: Effects of different activity and
musculature, in Schumacher GH (ed): Morphology of inactivity paradigms on myosin heavy chain gene expres-
the Maxillo-Mandibular Apparatus. Leipzig, VEB Georg sion in striated muscle. J Appl Physiol 90:345-357, 2001
Thieme, 1972, pp 65-69 34. Schuler M, Pette D: Fiber transformation and replace-
19. Bui AT, Widmer CG. Anatomical architecture of the ment in low-frequency stimulated rabbit fast-twitch mus-
human masseter muscle. J Dent Res 75:242, 1996 (abstr) cles. Cell Tissue Res 285:297-303, 1996
20. Bui AT, Widmer CG: Innervation patterns of anatomical 35. Mayo KH, Ellis E, III, Carlson DS: Histochemical char-
partitions within the human masseter muscle. J Dent Res acteristics of masseter and temporalis muscles after 5
76:29, 1997 (abstr) weeks of maxillomandibular fixation—An investigation
21. Wineski LE, Herring SW: Innervation and function in in Macaca mulatta. Oral Surg Oral Med Oral Pathol
the masseter complex of the pig. Fortschr Zool 30:285- 66:421-426, 1988
287, 1985 36. Ohnuki Y, Saeki Y, Yamane A, et al: Quantitative changes
22. Widmer CG, English AW, Carrasco DI, et al: Modeling in the mRNA for contractile proteins and metabolic
rabbit temporomandibular joint torques during a power enzymes in masseter muscle of bite-opened rats. Arch
stroke. Angle Orthod (in press) Oral Biol 45:1025-1032, 2000
23. Nordstrom SH, Yemm R: The relationship between jaw 37. Gedränge T, Luck O, Hesske G, et al: Differential ex-
position and isometric active tension produced by direct pression of myosin heavy-chain mRNA in muscles of
stimulation of the rat masseter muscle. Arch Oral Biol
mastication during functional advancement of the man-
19:353-359, 1974
dible in pigs. Arch Oral Biol 46:215-220, 2001
24. Goto TK, Langenbach GE, Hannam AG: Length
38. McKoy G, Ashley W, Mander J, et al: Expression of
changes in the human masseter muscle after jaw move-
insulin growth factor-1 splice variants and structural
ment. Anat Rec 262:293-300, 2001
genes in rabbit skeletal muscle induced by stretch and
25. Weijs WA, van der Wielen-Drent TK: Sarcomere length
and EMG activity in some jaw muscles of the rabbit. Acta stimulation. J Physiol 516:583-592, 1999
Anat (Basel) 113:178-188, 1982 39. Owino V, Yang SY, Goldspink G: Age-related loss of
26. Rowlerson A, Pope B, Murray J, et al: A novel myosin skeletal muscle function and the inability to express the
present in cat jaw-closing muscles. J Muse Res Cell Motil autocrine form of insulin-like growth factor-1 (MGF) in
2:415-438, 1981 response to mechanical overload. FEBS Lett 505:259-
27. Briggs MM, Schachat F: Early specialization of the super- 263, 2001
fast myosin in extraocular and laryngeal muscles. J Exp 40. Clark GT: A critical evaluation of orthopedic interocclu-
Biol 203:2485-2494, 2000 sal appliance therapy: Effectiveness for specific symp-
28. Shiotani A, Flint PW: Expression of extraocular-super- toms. J Am Dent Assoc 108:364-368, 1984
fast-myosin heavy chain in rat laryngeal muscles. Neuro- 41. Pierce CJ, Gale EN: A comparison of different treat-
report 9:3639-3642, 1998 ments for nocturnal bruxism. J Dent Res 67:597-601,
29. Thornell LE, Billeter R, Eriksson PO, et al: Heteroge- 1988
neous distribution of myosin in human masticatory mus- 42. Sim Y, Carlson DS, McNamara JA Jr: Condylar adapta-
cle fibers as shown by immunocytochemistry. Arch Oral tion after alteration of vertical dimension in adult rhesus
Biol 29:1-5, 1984 monkeys, Macaca mulatta. Cranio 13:182-187, 1995
30. English AW, Eason J, Schwartz G, et al: Sexual dimor- 43. Dolce C, Van Sickels JE, Bays RA, et al: Skeletal stability
phism in the rabbit masseter muscle: Myosin heavy chain after mandibular advancement with rigid versus wire
composition of neuromuscular compartments. Cells Tis- fixation. J Oral Maxillofac Surg 58:1219-1227, 2000
sues Organs 164:179-191, 1999 44. Dolce C, Hatch JP, Van Sickels JE, et al: Rigid versus wire
31. Anapol F, Herring SW: Ontogeny of histochemical fiber fixation for mandibular advancement: Five-year skeletal
types and muscle function in the masseter muscle of and dental changes. Am J Orthod Dentofac Orthop (in
miniature swine. AmJ Phys Anthropol 112:595-613, 2000 press)
<<
Article
>> Home | TOC |
Index
an increase in posterior maxillary facial height, subjects did not have incisal contact.4 Thus, the
resulting in downward rotation of the mandible. relapse rate varied considerably depending on
Fränkel and Fränkel2 investigated a func- the method chosen to measure overbite.
tional approach to the treatment of skeletal Another study of orthodontic interventions
open bite.2 (This study did not specifically ad- assessed the impact of tongue cribs.5 In this
dress stability, but because it followed subjects study, subjects were selected based on a negative
for approximately 8 years, it was included.) In a pretreatment overbite; the use of a tongue crib;
cohort study, 30 subjects treated with functional and the availability of records at the pretreat-
regulators and lip-seal exercises were compared ment, posttreatment, and at least 1-year post-
with 11 untreated subjects. Subjects were se- treatment times. This retrospective case series
lected based on a hyperdivergent skeletal pat- identified 26 adolescent (growing) patients and
tern, as well as clinical observation of lip incom- 7 adult (nongrowing) patients. The mean pre-
petence and postural weakness of the orofacial treatment overbite, which was measured relative
muscles. The treated subjects were followed to the nasion-menton line, was —2.8. The overall
from a mean age of 7 until a mean age of 15, success rate for achieving a positive overbite was
with the last follow-up at least 4 years postre- 88%, and of these successful subjects, none re-
tention. The corresponding means for the un- lapsed to negative overbite during the posttreat-
treated group were 8 and 16 years. The treated ment period. Although the stability of the suc-
subjects had significant closure of the mandibu- cessfully treated cases was impressive, a criticism
lar plane and gonial angles during the study of this study was the potential for selection bias.
period. However, the mean pretreatment char- In a retrospective case series, Katsaros and
acteristics of the two groups were not reported Berg6 assessed the stability of 20 subjects treated
so it is difficult to assess whether the treated and with fixed (n = 19) or functional (n = 1) appli-
untreated subjects were comparable before the ances at least one year after discontinuation of
start of the study. Additionally, although the retention. Selection of the sample was based on
normalization of the skeletal characteristics in the existence of a pretreatment open bite, as-
the treated subjects was impressive, no overbite sessed on plaster casts, and the availability of
values were reported, making it difficult to assess records at the pretreatment, posttreatment, and
whether the open bites were successfully closed postretention periods. Cephalometric measure-
and stable. ments were also taken, with overbite measured
Lopez-Gavito et al3 evaluated stability of orth- perpendicular to the nasion-menton line. This
odontic treatment (fixed appliances and head- sample exhibited typical open-bite characteris-
gear) in subjects at least 9 years and 6 months tics, with a mean pretreatment openbite of —1.9
after the discontinuation of retention devices. mm. Success, based on incisal contact measured
Subjects were selected from the pös tre ten tion on plaster casts in habitual occlusion or in a
files of the University of Washington based on forward position, was achieved in 15 of the 20
the availability of records. In this retrospective subjects (75%). None of the successfully treated
case series, Lopez-Gavito et al reported that 35% subjects had negative overbite at the postreten-
of their sample relapsed, as defined by a cepha- tion period. Again, the stability of successful
lometric distance >3 mm from the tip of the cases is impressive, but limitations include the
lower incisor to the nearest hard tissue (tooth or potential for selection bias, as well as a definition
palate) measured along the long axis of the of success that depends on the articulation of
lower incisor. Although these results suggest models rather than a cephalometric measure-
a high relapse rate, anteroposterior skeletal ment.
change may have influenced this distance in the Kim et al7 reported that the multiloop edge-
absence of vertical changes. In fact, a subsequent wise arch-wire technique is a stable method for
master's thesis, using this same sample but a orthodontic correction of open bites. In a recent
different method of measuring overbite (dis- study, 55 open-bite subjects treated with the mul-
tance between upper and lower incisor tips mea- tiloop edgewise arch-wire technique technique
sured relative to the nasion-menton line), found were identified, with 27 of these having final
that all of these subjects had positive overlap at records at the 2-year follow-up period. The in-
the postretention period, although 60% of these vestigators divided their sample into growing
<<
Article
>> Home | TOC |
Index
and nongrowing groups. Among the growing some believe skeletal open bites are less likely
subjects with complete records (n = 17), all than dental open bites to spontaneously im-
achieved a positive overbite at the end of treat- prove. Thus, given the lack of controls, pretreat-
ment, and one (6%) relapsed to negative over- ment characteristics are useful in assessing the
bite at the 2-year follow-up. Among the non- severity and nature of the open-bite subjects,
growing subjects with complete records (n = which, in turn, may provide information on the
10), all achieved a positive overbite at the end of likelihood of spontaneous correction. The open-
treatment, and one (10%) relapsed to negative bite subjects in the previous studies all had steep
overbite at the 2-year follow-up. Thus, the overall mandibular planes and moderate open bites,
relapse rate was 7% (2/27). These results are with the exception of Lopez-Gavito et al's sub-
encouraging but could be influenced by selec- jects. (Although the mean open bite they re-
tion bias and a considerable loss to follow-up. ported was —5 mm, the mean overbite was 0.3
We can learn several things from these studies mm if measured relative to the nasion-menton
(Table 1). First, there are relatively few studies line.) Additionally, all of the subjects were prob-
that evaluate the long-term stability of orthodon- ably in permanent dentition, except for the ad-
tic therapies. Second, the sample sizes for these olescent crib therapy group. Thus, the majority
studies are not large. Third, definitions for open of the subjects in the studies mentioned pre-
bite differ, as well as methods to assess success viously had permanent dentitions, significant
and stability. In an effort to compare these stud- open bites, and skeletal components, and would
ies, the following conventions were used. Sub- not be expected to improve without therapy.
jects were required to have an open bite (specif- The second issue that should be addressed is
ically, no incisal overlap) before treatment. sample selection. Although selecting subjects
Treatment success was defined as the achieve- based on the availability of records might seem
ment of positive overlap at the end of therapy, unbiased, we must keep in mind that having
and stability was defined as the maintenance of models and radiographs at various time points
positive overlap at the last follow-up period. The requires some degree of cooperation from pa-
overall success rate was the product of the per- tients and that willingness to participate in these
centage of treatment success and the percentage recall visits may differ among successfully and
of stable cases at the last follow-up. For example, unsuccessfully treated patients. Additionally, the
if a study reported 80% treatment success and likelihood of recalling successful and unsuccess-
90% stability, the overall success would be 72%. ful cases may differ as well as the effort placed
By using these conventions, the overall success into recalling these subjects. If successfully
of orthodontic therapies is between 75% and treated subjects are more likely to be recalled by
93%, which seems quite good (Table 2). How- orthodontists and they are also more likely to
ever, several other factors should also be consid- participate in recalls, an overly optimistic esti-
ered. mate of stability will be reported. Thus, every
First, were the open bites of these subjects effort should be made to identify either a ran-
likely to improve without therapy? Some studies dom sampling of subjects or all consecutively
have reported from 40% to 80% spontaneous treated subjects. In either case, it is also useful to
closure of untreated open-bite subjects.8'9 How- know how many subjects were treated with a
ever, their conclusions were based on cross-sec- particular therapy and what percentage was re-
tional data using subjects as young as 6 to 8 years called for the study.
old. At that age, open-bite malocclusions are In summary, the orthodontic therapies do not
difficult to define because of the normal transi- always result in positive overlap of the incisors at
tion of incisors. Most orthodontists would prob- the end of treatment. However, for subjects who
ably agree that open bites persisting into the late achieve a positive overlap, the stability of the
mixed dentition rarely close without interven- open-bite correction seems to be quite good.
tion. This view is supported by the National Although other nonsurgical therapies have been
Health and Nutrition Estimates Survey III data, prescribed, such as myofunctional therapy, pos-
which reports the prevalence of openbite to be terior bite blocks, magnets, and chin cups, no
3.6% in 8-11 year-olds, 3.5% in 12-17 year-olds, studies assessing the long-term stability of those
and 3.3% in 18-50 year-olds.10 Additionally, therapies could be located. As with many orth-
<<
Article
>> Home | TOC |
Index
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 ported that six of nine subjects had excellent
reviewed earlier are small sample sizes, selection stability (no appreciable dental or skeletal move-
bias, loss to follow-up, and a lack of untreated ment was detected), two had fair stability (de-
controls. tectable movement but no effect on the cor-
rected overbite), and one had poor stability (a
reopening of the bite). These results are difficult
Studies Assessing Combined to evaluate because the definitions of open-bite
Orthodontic-Surgical Therapies stability are not quantitatively defined.
Surprisingly, there are considerably more stud- Denison et al12 used a cohort design to com-
ies assessing stability of open bite with combined pare vertical stability in 66 subjects who had
orthodontic-surgical therapy. One of the earliest maxillary LeFort I osteotomies. These subjects
studies of stability after surgical correction of were selected from the practices of various fac-
open bites focused on tongue function and ulty members of the University of Washington
speech changes.11 This retrospective case series Department of Orthodontics, Seattle, WA. Over-
classified nine subjects into simple (canine to bite was measured relative to the nasion-menton
canine), compound (premolar to premolar), line, and records were available at pre treatment,
and infantile (extending posterior to premolars) posttreatment, and at least 1 year posttreatment.
open bites, and a variety of maxillary and man- The subjects were divided into three groups: (1)
dibular surgeries were performed. They re- an open-bite group (no incisal overlap, n = 28),
<<
Article
>> Home | TOC |
Index
(2) an incisal overlap group (positive overbite jects with surgically assisted or surgical expan-
but no contact of incisors, as assessed on the sion. From this data, they concluded that
cephalometric headfilm (ceph), n = 24), and surgical correction of openbite using rigid fixa-
(3) an incisal contact group (as assessed on the tion is relatively stable, and that orthodontic
ceph, n = 14). All subjects had maxillary supe- expansion may contribute to relapse in the ver-
rior repositioning surgery and were followed for tical dimension. As in many of the surgical stud-
a minimum of 1 year after the completion of ies, pretreatment cephalometric characteristics
orthodontic treatment. Although the incisal were not described, and the earliest follow-up
contact and incisal overlap subjects did not show period may have occurred before removal of
any significant changes in overbite during the orthodontic appliances.
posttreatment period, 43% of the open-bite sub- McCance et al15 reported on 21 high-angle
jects had significant increases in facial height, subjects who underwent fixed appliances and
eruption of maxillary molars, and decreases in bimaxillary surgery for open-bite correction.
overbite during the posttreatment phase. In Rigid fixation was used in the maxilla, with wire
21% of the open-bite subjects, the overbite re- fixation in the mandible, and intermaxillary fix-
lapsed to a negative value. ation. Records were available at the preopera-
Reitzik et al13 reported on a mandibular pro- tive, postoperative, and 1-year postoperative
cedure for open-bite correction—the reverse-L times. He reported variable postoperative move-
osteotomy with rigid fixation. Of 20 consecu- ments of the maxilla and mandible. In the Class
tively treated cases, the records of 16 were avail- III subjects (n = 11), overbite decreased form
able at the preoperative, postoperative, and >1- — 6 mm to 3.1 mm postoperatively but relapsed
year postoperative time periods. The surgical to 2.4 at the follow-up time. The overbite values
procedure resulted in a mean closure of 9° for for the Class II subjects (n — 10) were —4.6,
the gonial angle and 4.3° for the mandibular — 1.6, and —1.6, respectively. The results of this
plane angle. At 1 year postoperative, the gonial study are difficult to interpret because the
angle had relapsed 0.3° and the mandibular method for measuring overbite was not de-
plane angle 0.6°. The investigators also reported scribed nor was the actual number of subjects
that incisal contact had been maintained in all who achieved or maintained positive overlap re-
subjects, but overbite measurements were not ported. Additionally, the follow-up time was
reported. Based on these results, the investiga- short.
tors concluded that the reverse-L procedure Oliveira and Bloomquist16 investigated the
with rigid fixation is a stable method for open- stability of bilateral sagittal split osteotomy
bite correction. Their results are impressive, but (BSSO) with rigid fixation in 10 subjects who
the short follow-up time may have allowed some had records from the preoperative, postopera-
subjects to be just completing orthodontics. Ad- tive, and > 1-year postoperative time points (all
ditionally, cephalometric characteristics of the subjects completed orthodontic treatment be-
sample were not described at the pretreatment fore the last follow-up). The mandibular plane
time. angle was evaluated as an indicator of surgical
Haymond et al14 investigated the stability of stability, with a mean operative change of 3.9° of
open-bite correction using rigid fixation for closure followed by a mean increase of 1.3° dur-
maxillary procedures alone (n = 15), maxillary ing the postoperative period. Two subjects had
procedures in combination with mandibular ad- mandibular plane angles at the latest follow-up
vancement (n = 16) or set-back (n — 4), and that were greater than the preoperative value.
isolated mandibular set-back (n = 3). In this However, all 10 subjects had positive incisal over-
case series, records were available at the preop- lap at the latest follow-up. Although these results
erative, postoperative, and 1 to 5 year postoper- are encouraging for correcting open bites with
ative periods; 6 of the 38 subjects (16%) had isolated mandibular surgery, pretreatment char-
negative incisal overlap at the latest follow-up. acteristics were not reported, and the minimum
The investigators reported that dentoalveolar follow-up period was quite short.
changes were primarily responsible for relapse, Ermel et al17 reported on the Schuchardt
especially in 3 cases that were expanded orth- technique for open-bite correction, which is a
odontically. They did not note relapse in sub- segmented maxillary osteotomy involving impac-
<<
Article
>> Home | TOC |
Index
tion of the buccal segments. Of 44 subjects who with only 20% of the subjects in either group
underwent this procedure at the University of exhibiting good transverse stability. For multi-
Tubingen (Germany) from 1980 to 1994, 26 segment LeFort I procedures, the stability was
subjects agreed to participate, but only 20 had better with rigid fixation than with wires. Al-
complete records at the preoperative, postoper- though the transverse relapse was associated
ative, and long-term follow-up period (1-15 years with clockwise rotation of the mandible, it was
postretention). Fifteen of the 26 subjects also not well correlated with changes in overbite,
had BSSO, and in all subjects, rigid fixation was suggesting that compensatory eruption oc-
used. At the last follow-up visit, 3 of the 26 curred in some subjects.
subjects (12 %) exhibited a negative overbite, In a study investigating maxillary incisor ex-
and the investigators concluded that the Schu- trusion and vertical relapse in surgically treated
chardt technique is a relatively stable method to open bites, Lo and Shapiro20 collected records
address anterior open bite when the maxillary for 40 subjects who had maxillary LeFort I sur-
anterior teeth are in good relationship with the gery, with some also undergoing mandibular
upper lip. Pre treatment overbite was not re- surgical procedures. All subjects had records
ported, although the mean mandibular plane from the pretreatment, preoperative, postoper-
angle was 39°. ative, and >2-year posttreatment time periods.
In the largest study of surgically treated ante- These subjects were divided into a group that
rior open bite, Hoppenreijs et al18 reported on a underwent presurgical extrusion of the maxil-
total of 267 subjects in a multicenter study.18 lary incisors (n = 19) and a group that did not
Rigid fixation was used in 114 subjects, and wire (n = 21). The mean pretreatment open bite for
fixation was used in 153 subjects. In 144 subjects, these groups was —1.7 and —2.0, respectively.
only a LeFort I was performed, with the remain- The investigators did not find a significant dif-
ing 123 subjects receiving a combined LeFort I ference in the amount of relapse because both
with mandibular BSSO. Although records were groups had 5 subjects with no incisal overlap at
collected at the pretreatment (Tl), preoperative the final follow-up time. Thus, presurgical extru-
(T2), postoperative (T3), 6 to 19 months post- sion of the incisors did not appear to affect
operative (T4), and 20 to 210 months postoper- stability of open-bite correction in these subjects,
ative periods (T5), not all subjects had complete and the overall rate of relapse was 25%.
records from all time points. All subjects that Arpornmaeklong and Heggie21 compared 17
had Tl cephalometric characteristics had no in- subjects undergoing isolated maxillary superior
cisal overlap at that time, as measured relative to repositioning with 20 subjects who underwent
the occlusal plane, and at T5, all subjects were at bimaxillary surgery with rigid fixation. Records
least 1 year out of orthodontic appliances. In were available at the preoperative, postopera-
comparing the differing surgical procedures and tive, and > 1-year postoperative times. Good
methods of fixation, the investigators concluded maxillary stability was reported in both groups.
that the maxilla exhibited good vertical stability In the bimaxillary surgery group, 3 subjects
overall, and that the mandible exhibited better (15%) exhibited considerable anteroposterior
stability when rigid fixation was used. Their over- relapse of mandibular advancement. The mean
all relapse rate was 19% when incisal overlap was overbite for the LeFort I group was 0.9 at the
used as the criterion for success at T5. Again, final follow-up and 0.7 for the bimaxillary
pretreatment cephalometric characteristics were surgery group. Unfortunately, the number of
not reported. subjects with vertical relapse was not reported.
Hoppenreijs et al19 also reported on trans- Additionally, pretreatment cephalometric char-
verse stability after LeFort I surgery with and acteristics were not reported, and the final fol-
without BSSO by using methodology similar to low-up period was relatively short.
the previous study mentioned. They compared Proffit22 reported on the long-term stability of
77 subjects with orthodontic expansion and one all open-bite subjects in the University of North
piece LeFort I osteotomy with 53 subjects who Carolina database who had maxillary osteoto-
underwent multisegment LeFort I osteotomy mies (n = 28) or bimaxillary osteotomies (n =
and found no difference in the transverse stabil- 26), along with complete records. All subjects
ity of the two groups at the latest follow-up time, had at least 2-mm open bite at the pretreatment
<<
Article
>> Home | TOC |
Index
time, and records were available at the preoper- of orthodontic appliance removal. Therefore,
ative, postoperative, 1 year postoperative, and at the overbite measurements at that time may re-
least 3 years postoperative time periods. He flect orthodontic finishing rather than true sta-
found that 12% of the subjects had decreases of bility. From T4 to T5 (debanding to 1 year pos-
2 to 4 mm in overbite during the follow-up torthodontics), both the rotation group and
period. Although facial height increased in impaction group showed vertical relapse. From
more than 35% of the subjects, this was usually the T5 to T6 period (1 to 5 years after orthodon-
accompanied with incisor eruption, which main- tics), the overbite continued to decrease slightly.
tained the overbite relationship. The actual Thus, it would seem that the minimal period
overbite measurements at various time points required to monitor overbite stability is at least 1
are not reported, and therefore it is difficult to year after the removal of orthodontic appli-
compare this study with results from other stud- ances.
ies. Also, although relapse <2 mm was consid- Fischer et al24 reported on 58 consecutive
ered stable, it is difficult to know if those subjects subjects who were diagnosed with retrognathia
maintained positive incisal overlap. and open bite and underwent bimaxillary sur-
Moldez et al23 compared a random selection gery. Records were available at the pretreatment,
of 23 Class III open-bite subjects who underwent preoperative, postoperative, and >2-year post-
maxillary impaction and BSSO to 11 Class III operative periods. They found that 17 of the 58
subjects without open bite but similar surgical subjects (29%) had no incisal overlap at the final
correction. The open bites were further divided follow-up. Of these, 8 had multisegment LeFort
into two groups: one with maxillary impaction I osteotomies. They concluded that maxillary
but no rotation (n = 13) and one with clockwise stability was good but that the mandible tended
rotation of the maxilla (n = 10). In this well- to rotate downward. It is interesting that the
documented study, cephalometric characteris- relapse rate was so high, especially because the
tics were reported at the pretreatment (Tl), range of overbite at the pretreatment period was
preoperative (T2), postoperative (T3), postorth- -7.9 to 5.8, with a mean of -0.8.
odontic (T4), 1-year postorthodontic (T5), and In the most recent surgical study, Swinnen et
5-year postorthodontic (T6) time periods. The al25 evaluated 49 consecutively treated open-bite
pretreatment overbites for the impaction, rota- subjects.25 Of these, 38 had maxillary intrusion,
tion, and non-open-bite groups were —2.2, whereas 11 had maxillary extrusion. Records
— 3.8, and 2.0, respectively. Moldez et al re- were available at the pretreatment, preopera-
ported that 2 of 13 subjects (15%) in the max- tive, postoperative, 20-weeks postoperative, and
illary impaction group exhibited relapse of the 1-year postoperative time points, and all subjects
open bite, whereas no subjects from the rotation had rigid fixation. Both the intrusion and extru-
group or non-open-bite group displayed verti- sion groups exhibited good maxillary skeletal
cal relapse. Thus, they concluded that the Le- stability at the 1-year postoperative period, but
Fort I osteotomy with rotation, in which the 11% of the subjects had no incisal overlap at that
vertical dimension was maintained, was more time.
stable than LeFort I osteotomy with impaction, The surgical studies display considerable
in which the mandible rotated in a counter- variation in methodology. The comparison of
clockwise manner. The proposed rationale is the pretreatment characteristics is difficult because
temporary alteration of freeway space in the im- preoperative, rather than the pretreatment,
paction subjects. characteristics are more commonly reported
Perhaps the most interesting finding from the (Table 1). Although the mean mandibular plane
latter study is the change in overbite during the angles from these studies tended to be steeper
postoperative periods. From T3 to T4 (postop- than the nonsurgical studies, the mean open-
erative to debanding), both the rotation and bite measurements were not more severe (Table
impaction group displayed increases in the over- 1). Thus, the surgical subjects were typically
bite. This could be due to dental movements in adults with very steep mandibular planes and 1
the finishing stages of orthodontics. Many surgi- to 2 mm of open bite. Again, these are subjects
cal studies report stability at the 1-year postop- in which spontaneous improvement is unlikely.
erative time, which may be very close to the time Another issue that arises when assessing the
<<
Article
>> Home | TOC |
Index
surgical studies is the follow-up period. In the controlled trials, (2) cohort studies, (3) case/
orthodontic studies, all subjects were out of control studies, (4) case series, and (5) expert
treatment at least 1 year. In the surgical studies, opinion. The small number of nonsurgical stud-
more than half use the 1-year postoperative time ies is all case series, with relatively small samples
as the minimum follow-up. As previously men- and the potential for selection bias. Thus, we
tioned, this time point is likely to coincide with cannot be sure that the nonsurgical therapies
the removal of orthodontic appliances, and are truly as successful or as stable as the current
therefore the overbite would be expected to be literature suggests. The surgical studies consist
positive. The study by Moldez et al23 shows in- of case series and cohort studies that compare
creased overbite from the postoperative period different therapies or dental characteristics. Al-
to the end of orthodontic treatment and then a though most of the surgical studies also have
decrease during the first year after debanding. small samples and the potential for selection
Therefore, although it may be possible to evalu- bias, the large, multicenter study of Hoppenreijs
ate surgical stability 1 year postoperatively, over- et al18 suggests that the stability of surgical ther-
bite stability, which reflects both skeletal and apy is approximately 80%. The surgical studies
dental movements, cannot truly be assessed until provide a higher level of evidence than the orth-
a sufficient period of time has passed after the odontic studies based on their greater numbers,
removal of appliances. From the work by Moldez the case series that report on consecutively
et al, 12 months would seem to be an appropri- treated subjects, and the work of Hoppenreijs
ate period. et al.18
An attempt was made to measure overall suc- The lack of untreated controls in open-bite
cess for the surgical studies by using the conven- studies can be justified by the fact that these
tions from the nonsurgical studies. Surgical ther- subjects usually do not improve without therapy,
apy should allow positioning of the maxilla or especially after the eruption of the permanent
mandible to achieve a positive overlap of the incisors. Therefore, unless a study is assessing an
incisors so the treatment success was assumed to early intervention, it is probably more important
be 100%. With this assumption, the overall suc- to compare different types or combinations of
cess, which varied from 71% to 100%, only de- orthodontic and surgical therapies in cohort or
pended on the subsequent stability (Table 2). randomized designs. The use of incisal overlap
The four surgical studies that documented a as a criterion for measuring open bite should
negative overlap at the pretreatment or preop- also be discussed. This measurement has the
erative time and had a minimum follow-up of 1 advantage of being cephalometrically defined. It
year postorthodontics (Table 2 in bold) re- also is a good indicator of whether the lack of
ported a range of overall success from 75% to incisor contact is due to a true vertical problem
88%. This is very similar to the range of 75% to or an anteroposterior discrepancy. Additionally,
93% reported for the nonsurgical studies. How- two studies that have compared subjects with
ever, the nonsurgical studies tended to report incisal contact, incisal overlap, and no incisal
less treatment success and better stability. It is overlap found that the incisal overlap subjects
difficult to predict what the overall success of the reacted very similarly to the incisal contact
surgical studies would have been if incisor over- group, in which minimal vertical relapse oc-
lap were not achieved in all surgical subjects curred.4'12 Finally, from a clinical standpoint, a
because the treatment success would have de- lack of incisal overlap is usually associated with
creased, but the stability of subjects who were an inability to incise and therefore indicates an
successfully treated may have improved. Unfor- abnormal functional relationship of the incisors.
tunately, this assumption was necessary because Skeletal stability of the maxilla tended to be
most surgical studies did not report the percent- good with both wire and rigid fixation because it
age of subjects with incisal overlap postopera- typically was positioned superiorly. In bimaxil-
tively or posttreatment. lary surgeries, mandibular stability seemed to be
It is important to critically assess the level of more variable than maxillary stability, and Hop-
evidence provided by these studies. The gener- penreijs et al18 reported that rigid fixation
ally recognized hierarchy of clinical evidence, seemed to decrease relapse. Two studies re-
from strongest to weakest, is (1) randomized ported that poor transverse stability of the max-
<<
Article
>> Home | TOC |
Index
illa contributed to vertical relapse.14-19 All of 3. Lopez-Gavito G, Wallen TR, Little RM, et al: Anterior
these findings are generally consistent with the open-bite malocclusion: A longitudinal 10-year postre-
tention evaluation of orthodontically treated patients.
statements in an article by Proffit26 reviewing the Am J Orthod 871:175-186, 1985
stability of surgical movements. Relapse of open 4. Zuroff JP: Orthodontic treatment of anterior openbite
bites may also occur because of tongue size or malocclusion: Stability ten-years post-retention. Master's
posture, unfavorable growth patterns, orofacial thesis, University of Washington, Seattle, WA, 1990
musculature, respiratory problems, and dental 5. Huang GJ, Justus R, Kennedy DB, et al: Stability of
movements. Additionally, condylar resorption anterior openbite treated with crib therapy. Angle
Orthod 601:17-24, 1990; discussion 25-26
after orthognathic surgery has been reported to 6. Katsaros C, Berg R: Anterior open bite malocclusion: A
be a factor in relapse.27 follow-up study of orthodontic treatment effects. Eur
As stated at the beginning of this article, J Orthod 151:273-280, 1993
open-bite therapy is still a challenge to the orth- 7. Kim YH, Han UK, Lim DD, et al: Stability of anterior
odontic profession. The current literature sug- openbite correction with multiloop edgewise archwire
gests that 80% of anterior open-bite subjects therapy: A cephalometric follow-up study. Am J Orthod
Dentofacial Orthop 1181:43-54, 2000
will maintain positive overlap after treatment, 8. Kantorowicz A, Korhaus G: The self-correction of orth-
whether it is with orthodontics or combined odontic anomalies. Tr First Intl Orthod 40, 1926
orthodontics and surgery. However, these find- 9. Worms FW, Meskin LH, Isaacson RJ: Open-bite. Am J
ings are based on studies that usually have small Orthod 591:589-595, 1971
samples and the potential for selection bias. Re- 10. Proffit WR, Fields HW Jr, Moray LJ: Prevalence of mal-
occlusion and orthodontic treatment need in the United
viewing the existing studies allows us to improve
States: Estimates from the NHANES III survey. Int J
future investigations of anterior open bite. For Adult Orthodon Orthog Surg 131:97-106, 1998
instance, Hoppenreijs et al18 show the feasibility 11. Turvey TA, Journot V, Epker BN: Correction of anterior
of large, multicenter studies. If well designed, open bite deformity: A study of tongue function, speech
these types of studies can substantially increase changes, and stability. J Maxillofac Surg 41:93-101, 1976
sample size, which results in greater statistical 12. Denison TF, Kokich VG, Shapiro PA: Stability of maxil-
lary surgery in openbite versus nonopenbite malocclu-
power when comparing different therapies. Sev-
sions. Angle Orthod 591:5-10, 1989
eral of the surgical studies report on consecu- 13. Reitzik M, Barer PG, Wainwright WM, et al: The surgical
tively treated or randomly selected subjects, treatment of skeletal anterior open-bite deformities with
which minimizes selection bias. Also, the careful rigid internal fixation in the mandible. Am J Orthod
documentation by Moldez et al23 helps to show Dentofacial Orthop 971:52-57, 1990
the importance of an adequate follow-up period 14. Haymond CS, Stoelinga PJ, Blijdorp PA, et al: Surgical
orthodontic treatment of anterior skeletal open bite
after all treatment has been completed. To com- using small plate internal fixation. One to five year
pare the overall success of orthodontic and sur- follow-up. Int J Oral Maxillofac Surg 201:223-227, 1991
gical therapies, standardized information is nec- 15. McCance AM, Moss JP, James DR: Stability of surgical
essary. This includes complete records from the correction of patients with skeletal III and skeletal II
pretreatment period to describe the sample (in- anterior open bite, with increased maxillary mandibular
cluding age and gender); consistent and repro- planes angle. Eur J Orthod 141:198-206, 1992
16. Oliveira JA, Bloomquist DS: The stability of the use of
ducible methods to measure overbite; and care- bilateral sagittal split osteotomy in the closure of ante-
ful documentation of therapies with interim, final, rior open bite. Int J Adult Orthodon Orthognath Surg
and long-term records. Well-designed studies are 121:101-108, 1997
necessary to better understand the effectiveness 17. Ermel T, Hoffmann J, Alfter G, et al: Long-term stability
and stability of orthodontic and surgical thera- of treatment results after upper jaw segmented osteot-
omy according to Schuchardt for correction of anterior
pies as well as the mechanisms for relapse. This
open bite. J Orofac Orthop 601:236-245, 1999
will provide orthodontists with the evidence nec- 18. Hoppenreijs TJ, Freihofer HP, Stoelinga PJ, et al: Skel-
essary to recommend the most appropriate op- etal and den to-alveolar stability of Le Fort I intrusion
tions to their anterior open-bite patients. osteotomies and bimaxillary osteotomies in anterior
open bite deformities. A retrospective three-centre
study. Int J Oral Maxillofac Surg 261:161-175, 1997
References 19. Hoppenreijs TJ, van der Linden FP, Freihofer HP, et al:
1. Nemeth KB, Isaacson RJ: Vertical anterior relapse. Am J Stability of transverse maxillary dental arch dimensions
Orthod 651:565-585, 1974 following orthodontic-surgical correction of anterior
2. Frankel R, Frankel C: A functional approach to treat- open bites. Int J Adult Orthodon Orthog Surg 131:7-22,
ment of skeletal open bite. Am J Orthod 841:54-68, 1983 1998
<<
Article
>> Home | TOC |
Index
20. Lo FM, Shapiro PA: Effect of presurgical incisor extru- Stability after bimaxillary surgery—2-year treatment
sion on stability of anterior open bite malocclusion outcomes in 58 patients. Eur J Orthod 221:711-718,
treated with orthognathic surgery. Int J Adult Orthodon 2000
Orthog Surg 131:23-34, 1998 25. Swinnen K, Politis C, Willems G, et al: Skeletal and
21. Arpornmaeklong P, Heggie AA: Anterior open-bite mal- dento-alveolar stability after surgical-orthodontic treat-
occlusion: Stability of maxillary repositioning using rigid ment of anterior open bite: A retrospective study. Eur
internal fixation. Aust Orthod J 161:69-81, 2000 J Orthod 231:547-557, 2001
22. Proffit WR, Bailey LJ, Phillips C, et al: Long-term stability 26. Proffit WR, Turvey TA, Phillips C: Orthognathic surgery:
of surgical open-bite correction by Le Fort I osteotomy. A hierarchy of stability. Int J Adult Orthodon Orthog-
Angle Orthod 701:112-117, 2000 nath Surg 111:191-204, 1996
23. Moldez MA, Sugawara J, Umemori M, et al: Long-term 27. Hoppenreijs TJ, Freihofer HP, Stoelinga PJ, et al: Con-
dentofacial stability after bimaxillary surgery in skeletal dylar remodelling and resorption after Le Fort I and
Class III open bite patients. Int J Adult Orthodon Or- bimaxillary osteotomies in patients with anterior open
thog Surg 151:309-319, 2000 bite. A clinical and radiological study. Int J Oral Maxil-
24. Fischer K, von Konow L, Brattstrom V: Open bite: lofac Surg 271:81-91, 1998
<<
Article
>> Home | TOC |
Index
When the severity of vertical deformity is so ment would be greater for patients with partic-
great that reasonable correction cannot ular characteristics. A survey of some 1,000 pa-
be obtained by growth modification or camou- tients who sought evaluation in the Dentofacial
flage, the combination of orthodontics and or- Clinic at the University of North Carolina,
thognathic surgery may provide the only viable Chapel Hill, NC, in the 1980s concluded that
treatment option. Epidemiologie data suggest individuals with a long face or Class III problem
that there are approximately 1.8 million people were more likely to seek and receive surgical-
in the United States with dentofacial dispropor- orthodontic treatment than those who had a
tions too severe to be corrected by orthodontics Class II problem.3
alone.1'2 It seems likely that demand for treat- Long-face individuals make up approximately
one fourth of the dentofacial deformity group
who seek surgical treatment, almost 20% more
From the Department of Orthodontics and the Department of than would have been expected from the pre-
Oral and Maxillofacial Surgery, School of Dentistry, University of
North Carolina, Chapel Hill, NC; and a private practice, Birming-
sumed population incidence (Table 1). 4 Of the
ham, AL. 2,074 patients evaluated through the University
Address correspondence to L 'Tanya J. Bailey, DDS, MS, Depart- of North Carolina Dentofacial Clinic from 1979
ment of Orthodontics, School of Dentistry, University of North Caro- to 1998, 464 (22.4%) were judged to have exces-
lina, Chapel Hill, NC 27599-7450.
Copyright 2002, Elsevier Science (USA). All rights reserved.
sive face height. Anterior open bite >4mm was
1073-8746/02/0803-0009$35.00/0 found in about one in six of the long-face group.
dot: 10.1053/sodo.2002.125437 Just fewer than 25% of the patients who under-
174 Bailey et al
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
>> Home | TOC |
Index
Long face Problems 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-
79-80 81-82 Si-84 85-86 87-88 89-90 91-92 93-94 95-96 97-98
sors for long-face adolescents is very unattrac-
Year (2-year increments)
tive. Extraction of premolars to retract the upper
Figure 2. Graph showing distribution of dentofacial incisors will cause them to elongate even further
population for long-face problems in 2-year incre- and increase the nasolabial angle. Because this
ments. correction is often accomplished through the
use of Class II elastics, the mandible is likely to
further rotate down and back, accentuating an
live with than, for instance, mandibular defi-
already long-face pattern. Before the option of
ciency without excess face height.
vertical maxillary impaction became available,
The climate for surgical-orthodontic treat-
the negative facial aesthetics that resulted were
ment changed in the 1990s. This was for the
considered an inevitable consequence of im-
better in several ways because significant ad-
proving the dental occlusion. The continuation
vances in treatment occurred. For example,
of growth well into the late teens tends to worsen
computer software to generate surgical predic-
the deformity; however, the fact that there is
tions has greatly improved communication with
remaining growth provides an opportunity for
patients and their understanding of treatment
implementation of growth modification tech-
options, and the near-total replacement of inter-
niques. Growth modification after the adoles-
maxillary fixation with rigid internal fixation
cent growth spurt is theoretically possible but
now makes patients more comfortable during
actually improbable because very few adoles-
the postsurgical recovery period. It was for the
cents will wear a functional appliance with bite
worse for many patients, however, because
blocks or a headgear.
health insurance coverage for the surgery was
Anterior open bite in adolescents (or adults)
restricted in many plans and largely eliminated
often can be corrected with orthodontic treat-
by health maintenance organizations. Many
ment by intruding the posterior teeth, but that is
procedures are now being performed on an am-
almost impossible without surgery. A multiloop
bulatory care basis. This article discusses the
edgewise appliance, in conjunction with ante-
clinical management of patients with severe
rior vertical elastics, claims to produce posterior
long-face problems requiring surgical-orthodon-
intrusion and improvement of the skeletal prob-
tic treatment.
lem.5 Recent reports show that the open bite
correction occurs almost totally by elongating
Treatment Planning the incisor teeth.6 Successful camouflage of a
long-face problem is a function of both the pa-
Adolescents With Questionable Growth
tient's perception of the success of treatment
Potential
and the soft-tissue adequacy (the fuller the lips,
When evaluating orthodontic patients with pri- the better they can cover the teeth). If the chief
marily vertical problems, it is easy to dwell on complaint is excessive display of the teeth and a
anteroposterior problems because most long- gummy smile, elongating the incisors will not
face patients will have a receding chin and Class correct it. However, if the patient's major con-
II malocclusion. Often the chief complaint is cern is the open bite, an increased display of the
that the upper incisors are too prominent. An anterior teeth may be tolerable. The orthodon-
astute patient may be perceptive enough to de- tist must be sure the patient understands the
scribe the gummy smile as a problem that should aesthetic implication of the latter decision.
<<
Article
>> Home | TOC |
Index
176 Bailey et al
Patients who are treated with other extrusive, dylar axis to move up with it so that the chin
nonextraction orthodontic approaches may end moves upward and forward. Indirectly, the
up with their occlusion reasonably well cor- maxillary surgery repositions the mandible.
rected, but both facial aesthetics and long-term Often it is important not to elevate the an-
stability are questionable. In this particular situ- terior maxilla, and it may be necessary to
ation, the lower incisors will remain too protru- even rotate it downward so that an aesthetic
sive relative to the chin for good stability, the smile arc can be maintained.7
chin will still be deficient, and the lip incompe- 2. The second is mandibular surgery to bring
tence will still be present. If the nonextraction the lower jaw forward and upward, which
treatment is considered unsuccessful and the could be accomplished in an open-bite pa-
patient is retreated with extractions, esthetics tient by tilting the body of the mandible up
will be compromised even more. For borderline after a ramus osteotomy. This would correct
cases such as these, a lower border osteotomy of an anterior open bite, but the position of
the mandible to bring the chin upward and the maxilla would not be altered at all so this
forward can greatly improve both dental and treatment approach implies that the prob-
facial aesthetics because the lower lip relaxes lem is largely in the mandible, which rarely
and moves up as the chin is elevated. The lower is the case.
border osteotomy is not a complicated proce- 3. Superior repositioning of the chin by a man-
dure and may be accomplished in an ambulatory dibular lower-border osteotomy. This proce-
care setting. dure is a useful adjunct to either of the other
two surgical possibilities but is unlikely to be
Adults With Little or No Growth Potential adequate by itself in an adult.
Long-face patients with no prospect for success- The guideline for choosing between maxil-
ful growth modification have no real alternative lary and mandibular surgery is quite clear: in
to surgery for a successful and stable outcome. patients whose face height should be reduced,
Orthodontic camouflage is not a real option in maxillary surgery is the primary procedure. A
long-face problems. It may be better not to treat mandibular ramus osteotomy is recommended
a patient with a true vertical problem who re- only as a secondary procedure after the maxilla
fuses to consider surgical correction because aes- has been repositioned vertically. This is just the
thetics are likely to be severely compromised. opposite of the guideline for short-face patients:
for them, mandibular surgery is preferred when
Planning Surgical-Orthodontic face height must be increased, and a maxillary
Treatment osteotomy is a secondary procedure if required.
The maxilla is the focus of surgical treatment
As with mandibular deficiency and other skeletal in long-face patients for two major reasons. First,
problems, the surgical approach must be deter- the maxilla nearly always has excessive vertical
mined before the orthodontic approach can be development, whereas the mandible may not be
established. involved beyond the indirect rotation that max-
illary surgery corrects. Neither normal jaw and
Surgical Approach lip function nor good aesthetics can be achieved
Excessive lower-face height is the primary distin- without correcting the maxillary deformity for
guishing clinical characteristic of long-face pa- most patients. Second, moving the maxilla up
tients. It follows logically that a decrease in face produces a stable surgical correction. Rotating
height is needed. This can be accomplished in the mandible at the ramus osteotomy site in a
three ways: counterclockwise direction stretches soft tissues
posteriorly and is notoriously unstable. In pa-
1. The first is superior repositioning of the tients with a normal mandible that has been
maxilla, or at least the posterior part of the rotated downward and backward (ie, who could
maxilla, by total or segmental maxillary os- be characterized initially as Class I rotated to
teotomy. When the maxilla moves up, the Class II), superior repositioning of the maxilla to
mandible rotates around the horizontal con- correct the vertical discrepancy also corrects the
<<
Article
>> Home | TOC |
Index
anteroposterior problem because the mandible and expansion of the maxilla. The first stage is
rotates at the horizontal condylar axis. If the surgically assisted palatal expansion to widen the
mandible is both small and rotated, a ramus maxilla during the presurgical orthodontics,
osteotomy for further advancement is needed in and then a LeFort I osteotomy is performed to
addition to superior repositioning of the max- reposition it in one piece.8 In our view, better
illa. In patients with a large but rotated mandible clinical success with the two-stage surgery has
(who might have been characterized initially as not been shown, and the increased morbidity
Class III rotated to Class I), correcting the verti- and cost of two surgical procedures rather than
cal position of the maxilla causes the mandible one cannot be justified.9 We recommend one-
to rotate into a prognathic position, and a ramus stage surgery with segmentation when it is
osteotomy to shorten it is required. needed.
Some surgeons have advocated counterclock- In summary, the surgical approach to long-
wise rotation of the maxilla and mandible, which face patients almost always includes a LeFort I
theoretically could lead to neuromuscular adap- osteotomy to superiorly reposition the maxilla.
tation in a way that isolated mandibular surgery Maxillary segments, mandibular ramus osteot-
would not. This would allow the posterior max- omy to advance or set back the mandible, and
illa to be moved downward to improve smile lower-border osteotomy to reposition the chin
aesthetics and would improve the gonial angle. are added as the requirements of the individual
Currently, there are only anecdotal case reports case dictate.
to document this response. Long-term remodel-
ing at the gonial angle is a potential problem,
Orthodontic Approach
and long-term outcomes of this treatment ap-
proach have not been clearly established. As with any surgical-orthodontic case, the orth-
Many long-face patients have excessive erup- odontic approach is oriented toward positioning
tion of the lower incisors (ie, the distance from the teeth presurgically in all three planes of
the incisal edge to the chin is too great). In space so their position will facilitate the surgical
addition, the incisors tend to be flared forward, plan and the teeth will fit appropriately when
which produces poor chin-lip balance. Both of the surgery is completed. To accomplish this
these problems can be addressed with a mandib- goal, the orthodontist must know the general
ular lower-border osteotomy. The bony cuts are surgical plan and two things quite specifically:
angled up anteriorly, allowing the chin to be (1) whether the maxilla will be kept in one piece
moved up and forward, and a wedge of bone is or segmented and (2) whether chin-lip balance
removed above the chin if further vertical short- is to be achieved by orthodontically reposition-
ening is needed. ing the incisors or lower-border osteotomy.
During maxillary surgery, with the maxilla in Long-face patients rarely have a severely ex-
the down-fractured position, dentoalveolar seg- aggerated curve of Spee in the mandibular arch,
ments can be created readily. The usual indica- even if a deep overbite is present. In general, it
tion for two segments, created by a parasagittal is preferable to level the lower arch before sur-
osteotomy, is to allow the maxilla to be widened gery. Like the guideline for maxillary versus
as it is moved superiorly. Three segments, two mandibular surgery, this guideline is in sharp
posterior and one anterior, usually are used to contrast to the one for short-face patients, whose
correct a vertical step in the arch, typically by leveling often is done postsurgically. The reason
moving the posterior segments up more than is the same, however. Postsurgical leveling makes
the anterior. The canines can be in either the it easier to increase face height, whereas presur-
anterior or posterior segments depending on gical leveling is better when decreased face
their initial position. height is the goal.
A major reason for segmenting the maxilla, On the other hand, a long-face patient with
although not the only one, is to expand it trans- severe anterior open bite often has an extreme
versely. Problems with the stability of transverse curve of Spee in the upper arch to the point that
expansion have led some surgeons to recom- vertical steps exist in the arch. Usually, the steps
mend two-stage surgical treatment when a long- are distal to the canines, but they may occur
face patient needs both vertical repositioning between the lateral incisors and canines. The
<<
Article
>> Home | TOC |
Index
178 Bailey et al
more severe the steps, the more advantageous it gival stripping is most common around
is to segment the maxilla during the surgery and lower incisors after a vestibular incision, but
level the arch by repositioning the dentoalveolar loss of gingival attachment around the up-
segments rather than by moving the teeth orth- per incisors and canines can and does occur
odontically. The orthodontist's role when surgi- in patients having maxillary osteotomy, and
cal segments are planned is to level presurgically can be an aesthetic problem.
within the segments but not across the osteot- 2. In a patient with an anterior open bite
omy sites and to be sure that there is enough whose plan calls for a segmental maxillary
space between the roots of the involved teeth to osteotomy with anterior and posterior den-
allow interdental osteotomies. toalveolar segments, it is important not to
Similar thinking guides the decision as to level the upper arch during the presurgi-
whether to expand a narrow maxillary arch orth- cal orthodontics. Instead, the orthodontist
odontically, either with dental expansion (arch should level only within the segments. Al-
wires only) or orthopedic separation of the mid- though continuous arch wires with steps at
palatal suture (jackscrew appliance or equiva- the planned osteotomy sites can be used
lent), or to defer this for segmental osteotomy. until just before surgery if desired, usually it
The more severe the narrow maxilla and the is easier and better to use separate arch wire
older the patient, the better the decision to ex- segments throughout the presurgical treat-
pand surgically. If the patient is young enough ment. The size of the arch wire segments
to open the suture orthopedically, presurgical and their sequence would be the same as the
expansion with a jackscrew appliance is accept- continuous wires for any other patient, cul-
able. As noted previously, we do not recommend minating with full-dimension rectangular
surgically assisted palatal expansion to widen a stabilizing segments. An auxiliary arch wire
narrow maxilla in a patient who will be sched- fitted to the model surgery casts is placed in
uled for LeFort I osteotomy later nor is there any the headgear tubes at surgery. It is a mistake
reason to expand orthodontically if the dentoal- to level the upper arch presurgically in pa-
veolar segments are needed to correct vertical tients with severe open bite because this pro-
steps because the segments also can be ex- duces a relapse tendency. The leveling
panded at the time of surgery. would occur primarily by elongating the up-
per incisors. When the orthodontic appli-
ance is removed postsurgically, the incisors
Presurgical Orthodontics would tend to relapse apically to some ex-
There are three points of special interest for tent and, of course, that would lead to re-
long-face patients. opening of the bite anteriorly. As the time of
surgery approaches, if it is observed that the
1. A LeFort I osteotomy requires a long inci- maxillary incisors have been elongated dur-
sion in the maxillary vestibule. If a lower- ing the presurgical orthodontics, the contin-
border osteotomy is part of the surgical uous arch wire should be removed and re-
plan, an incision in the mandibular anterior placed by anterior and posterior segments.
vestibule will be necessary. These incisions This should be performed several weeks be-
tend to stress the gingival attachment of ad- fore the surgery to allow any relapse ten-
jacent teeth, presumably because scar con- dency to express itself before the final sur-
traction during healing pulls the attachment gical planning is performed.
apically. It is much easier to prevent strip- 3. Similar considerations guide the approach
ping of gingival tissues and unsightly expo- to maxillary width. If the arch will be ex-
sure of roots than it is to correct it afterward. panded orthodontically, this should be per-
When the gingival attachment is question- formed at the very beginning of the presur-
able, the attachment should be augmented gical orthodontics so the expansion can be
by placing gingival grafts in doubtful areas at maintained as long as possible before the
least 2 to 3 months before the orthognathic expansion appliance is eventually removed.
surgery (preferably, but not necessarily, be- If a LeFort I osteotomy with separate poste-
fore the orthodontic treatment begins). Gin- rior dentoalveolar segments is planned and
<<
Article
>> Home | TOC |
Index
the expansion will be accomplished surgi- separation in the younger patients, perhaps even
cally, the orthodontist should be careful not more in those over age 30. Incisor display is a
to produce any orthodontic expansion. In- youthful characteristic and decreasing it makes
deed, in these patients, any presurgical an individual look older.7 The associated soft-
transverse tooth movement should be con- tissue changes accentuate this effect. When the
traction rather than expansion of the arch maxilla is moved up, the soft tissues of the
so that orthodontic relapse will not contrib- cheeks are relaxed, which is good for stability
ute to any tendency to posttreatment nar- but bad for aesthetics because the wrinkles that
rowing of the arch. The surgical segments inevitably accompany aging become more prom-
inevitably will relapse somewhat toward the inent. Patients beyond their early twenties do
midline, and teeth tend to move back to- not appreciate suddenly looking older.
ward the midline after dental expansion. Exposure of 30% to 40% of the clinical crown
of the maxillary incisor beneath the lip is aes-
thetically pleasing, whereas completely covering
Final Presurgical Planning
it is not. Because the upper lip increasingly cov-
Complete records—panoramic and lateral ceph- ers the upper incisors with advancing age, de-
alometric radiographs, other radiographs if in- creasing the amount of exposure of the incisors
dicated (eg, periapical radiographs in areas in may be undesirable as the patient gets older
which interdental osteotomy for segmental sur- even if it looks good initially. Rotating the ante-
gery is planned), facial and intraoral photo- rior maxilla down as the posterior goes up to
graphs, and dental casts—are required immedi- establish the best exposure of the teeth may be
ately before surgery. Because these long-face needed for optimum aesthetics.
patients will have maxillary surgery, a facebow Long-face patients who have a moderate de-
transfer to a semiadjustable articulator is neces- gree of mandibular deficiency can be a treat-
sary. ment-planning dilemma because as the mandi-
The planning at this stage is largely a repeti- ble rotates upward, it also comes forward, which
tion of what was done before treatment started. means that if the maxilla were moved up
The goals are to verify the original plan, pre- enough, the mandible could come forward to a
cisely quantify the movements needed at sur- normal anteroposterior position. Or, if the max-
gery, and complete the model surgery so that illa were moved back as it moved up, the overjet
occlusal splints can be made. The first step is a could be corrected without so much vertical
cephalometric prediction. From this, the mea- movement. For many patients, moving the max-
surements that are necessary for model surgery illa just a little more could prevent having to
are taken, and the casts are repositioned on the operate on both jaws.
articulator. All other things being equal, one-jaw surgical
There are two critical elements in the plan- procedures have less morbidity, better stability,
ning at this stage: (1) how far the maxilla is and lower cost compared with two^jaw surgery.
moved up and (2) if there would be residual On the other hand, moving the maxilla up too
overjet with a straight vertical movement of the much is bad aesthetically, and moving it back is
maxilla, whether the maxilla is moved forward worse. Often the maxilla should be advanced
or back to correct overjet or the mandible somewhat to obtain the best lip support and
lengthened or shortened by ramus osteotomy. esthetics. Almost never should it be retracted. It
Moving the maxilla up too far is as harmful to is much better to accept the need to do two-jaw
facial aesthetics as leaving the long face uncor- surgery to obtain a good result than to signifi-
rected. A long-face patient should not be over- cantly compromise aesthetics to keep the surgery
treated because a few millimeters of lip incom- within one jaw. The planning is even more of a
petence is normal. It is a mistake to elevate the problem in a patient who habitually postures the
maxilla enough to bring the lips into contact at mandible forward so that there is more than a 1-
rest. Younger patients can tolerate more upward to 2-mm difference between intercuspal position
movement of the maxilla (aesthetically and psy- (centric occlusion) and retruded contact posi-
chologically) than can older patients. tion (centric relation). Articulator mountings
It is better to leave approximately 4 mm of lip and the rotation of the mandible around the
<<
Article
>> Home | TOC |
Index
180 Bailey et al
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
>> Home | TOC |
Index
182 Bailey et al
Case Report
Figure 7. JW, superimposition showing pretreatment JW, age 45.4, was seen on referral from her local
to posttreatment changes. 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
The most difficult part of postsurgical orth- (Fig 3A and B). On intraoral examination, the
odontics for long-face patients is maintaining teeth were well aligned in both arches, with a
transverse maxillary expansion, particularly sur- narrow and constricted maxillary arch. Ceph-
gical expansion. If the maxillary buccal seg- alometric analysis confirmed mandibular defi-
ments relapse medially after the surgery, not ciency primarily because of a mandible that was
only is there relapse toward crossbite but also rotated down and back and a skeletal open-bite
the bite tends to open anteriorly because of pattern because of vertical maxillary excess
cuspal interferences posteriorly. It takes at least (Fig 4).
6 months after surgery for the maxillary dentoal- Because of the mandibular deficiency and
veolar segments to stabilize transversely so they mandibular incisor protrusion, the treatment
must be held in their expanded position during plan was extraction of maxillary first premolars;
the finishing orthodontics. The easiest way to do LeFort I osteotomy to elevate the posterior but
this is to use a heavy labial auxiliary wire in the not the anterior maxilla would allow the mandi-
headgear tubes along with the light working ble to rotate up and forward for closure of the
arch wires. If segmental surgery was performed, anterior open bite without lengthening the ra-
this can be the same auxiliary wire that was used mus and stretching the pterygomaxillary sling.
at surgery. If dental expansion was performed Two presurgical predictions were presented
presurgically, it may be wise to make a new labial to the patient: a LeFort I and a LeFort I and a
auxiliary wire for use at the postsurgical stage. lower-border osteotomy. The patient elected the
An alternative for maintaining width is a LeFort I only because there was very little differ-
transpalatal lingual arch, which has the signifi- ence in the two images. At the completion of
cant advantage of excellent torque control. (Al- treatment, she had very good occlusion with
<<
Article
>> Home | TOC |
Index
ideal overjet and overbite (Fig 5A and B). Post- surgical orthodontic treatment-a current review. Int J
treatment cephalometric analysis showed im- Adult Orthod Orthognath Surg 16:1-13, 2001
5. Kim YH: Anterior open bite and its treatment with mul-
proved facial balance (Fig 6). The superimposi- tiloop edgewise archwire. Angle Orthod 57:290-321,
tion showing pretreatment to posttreatment 1987
change is shown in Figure 7. 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
References 7. Sarver DM: The importance of incisor positioning in the
1. Proffit WR, White RP: Who needs surgical-orthodontic esthetic smile: the smile arc. Am J Orthod Dentofac
treatment? Int J Adult Orthod Orthog Surg 5:81-89, Orthoped 120:98-111, 2001
1990 8. Silverstein K, Quinn PQ: Surgically-assisted rapid palatal
2. Bailey LJ, Proffit WR, White RP: Assessment of patients expansion for management of transverse maxillary defi-
for orthognathic surgery. Semin Orthod 5:209-222, ciency. J Oral Maxillofac Surg 55:725-727, 1997
1999 9. Bailey LJ, White RP, Proffit WR, et al: Segmental LeFort
3. Proffit WR, Phillips C, Dann C IV: Who seeks surgical- I osteotomy to effect palatal expansion. J Oral Maxillofac
orthodontic treatment? Int J Adult Orthod Orthog Surg Surg 55:728-731, 1997
5:153-160, 1990 10. Burstone CJ: The precision lingual arch: hinge cap at-
4. Bailey LJ, Haltiwanger LH, Blakey GH, et al: Who seeks tachment. J Clin Orthod 28:151-158, 1994
<<
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 Electrocardiology
Official Journal of the International Society
for Computerized Electrocardiology and the
International Society of Electrocardiology
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
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 The appearance of the code at the bottom of the first page
quarterly by W.B. Saunders. Months of issue are March, June, of an article in this journal indicates the copyright owner's
September, and December. Corporate and Editorial Offices: consent that copies of the article may be made for personal or
The Curtis Center, Independence Square West, Philadelphia, internal use, or for the personal or internal use of specific clients,
PA 19106-3399. Accounting and Circulation Offices: 6277 Sea for those registered with the Copyright Clearance Center, Inc.
Harbor Drive, Orlando, FL 32887-4800. POSTMASTER: Send (222 Rosewood Drive, Danvers, MA 01923; (508) 750-8400;
change of address to: Seminars in Orthodontics, W.B. Saunders, www.copyright.com). This consent is given on the condition that
Customer Service, 6277 Sea Harbor Drive, Orlando, FL 32887- the copier pay the stated per-copy fee for that article through the
4800. 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
Editorial correspondence should be addressed to: as copying for general distribution, for advertising or promotional
P. Lionel Sadowsky, DMD, BDS, DipOrth, MDent, Editor, purposes, for creating new collective works, or for resale.
Seminars in Orthodontics, Professor and Chairman, Department Absence of the code indicates that the material may not be
of Orthodontics, University of Alabama, 1919 Seventh Avenue processed through the Copyright Clearance Center, Inc.
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. Reprint inquiries should be addressed to Anne
Saunders, Periodicals Department, P.O. Box 628239, Orlando, Rosenthal, Elsevier Science, The Curtis Center, Independence
FL 32862-8239 or e-mail hhspcs@harcourt.com. Square West, Philadelphia, PA 19106-3399. Telephone (215)
Change of address notices, including both the old and new 238-5534, fax (215) 238-6423; e-mail: a.rosenthal@elsevier.com.
addresses of the subscriber and the mailing label, should be
sent at least 1 month in advance. Customer Service: 1-800-654-
2452 Advertising representative: MJ. Mrvica Associates, Inc,
2 West Taunton Ave, Berlin, NJ 08009. Telephone (609) 768-
9360. Fax (609) 753-0064.
Yearly subscription rates: United States and possessions: Publication of an advertisement in Seminars in Orthodontics
individual, $133.00; institution, $164.00; student and resident, does not imply endorsement of its claims by the Editor (s) or
$67.00; single issue, $50.00. All other countries: individual
$166.00; institution, $198.00; student and resident, $83.00; Publisher of the journal.
single issue, $50.00. For all areas outside the United States and The contributors have checked generic and trade names and
possessions, there is no additional charge for surface delivery. verified drug doses for accuracy according to the standards
For air mail delivery, add $16.00. To receive student/resident accepted at the time of publication. The ultimate
rate, orders must be accompanied by name of affiliated responsibility, however, lies with the prescribing physician.
institution, date of term, and the signature of program/residency Please convey any errors to the Editor.
coordinator on institution letterhead. Orders will be billed at
individual rate until proof of status is received. The ideas and opinions expressed in Seminars in
Prices are subject to change without notice. Current prices Orthodontics do not necessarily reflect those of the Editor or the
are in effect for back volumes and back issues. Single issues, Publisher. Publication of an advertisement or other product
both current and back, exist in limited quantities and are mention in Seminars in Orthodontics should not be construed as
offered for sale subject to availability. Back issues sold in an endorsement of the product or the manufacturer's claims.
conjunction with a subscription are on a prorated basis. Checks Readers are encouraged to contact the manufacturer with any
should be made payable to W.B. Saunders and sent to Seminars questions about the features or limitations of the products
in Orthodontics, W.B. Saunders, Periodicals Department, 6277 mentioned. Neither the Editor or Publisher assume any
Sea Harbor Drive, Orlando, FL 32887-4800. responsibility for any injury and/or damage to persons or
property arising out of or related to any use of the material
Copyright 2002, Elsevier Science (USA). All rights contained in this periodical. The reader is advised to check the
reserved. No part of this publication may be reproduced or appropriate medical literature and the product information
transmitted in any form or by any means, electronic or currently provided by the manufacturer of each drug to be
mechanical, including photocopy, recording, or any information administered to verify the dosage, the method and duration of
storage and retrieval system, without permission in writing administration or contraindications. It is the responsibility of
from the Publisher. Printed in the United States of America. 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.
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, : Seminars in Orthodontics is indexed in the Cumulative
6277 Sea Harbor Drive, Orlando, FL 32887-4800. Telephone Index to Nursing and Allied Health Literature® print index
number: 1-407-345-2500. and the Cinahl® database.
W.B. SAUNDERS
<<
Article
>> Home | TOC |
Index
Seminars in Orthodontics
EDITOR
P. Lionel Sadowsky, DMD, BDS, DipOrth, MDent
EDITORIAL BOARD
Richard G. Alexander, Arlington, TX James A. McNamara, Jr, Ann Arbor, MI
Rolf G. Behrents, Memphis, TN Robert N. Moore, Grand Island, NE
Samir E. Bishara, Iowa City, I A Ravindra Nanda, Farmington, CT
Robert Boyd, DDS, San Francisco, CA Perry M. Opin, Milford, CT
Larry M. Bramble, Cypress, CA Sheldon Peck, Newton, MA
John S. Casko, Iowa City, I A William R. Proffit, Chapel Hill, NC
Harry L. Dougherty, Van Nuys, CA Cyril Sadowsky, Chicago, IL
T.M. Graber, Evanston, IL David M. Sarver, Birmingham, AL
Robert J. Isaacson, Richmond, VA T. Michael Speidel, Minneapolis, MN
Alexander Jacobson, Birmingham, AL William J. Thompson, Bradenton, FL
Lysle E.Johnston, Jr., Ann Arbor, MI James L. Vaden, Cookeville, TN
Gregory J. King, Seattle, WA Robert L. Vanarsdall, Jr., Philadelphia, PA
Vincent G. Kokich, Tacoma, WA Katherine Vig, Columbus, OH
Steven J. Lindauer, Richmond, VA C.B. Preston, Buffalo, NY
INTERNATIONAL
Zeev Abraham, Herzliya, Israel Shinkichi Namura, Tokyo, Japan
Jack Dale, Toronto, Ontario, Canada George Skinazi, Paris, France
W.G. Evans, Johannesburg, South Africa William A. Wiltshire, Winnipeg, Manitoba, Canada
Roberto Justus, Mexico City, Mexico Björn U. Zachrisson, Oslo, Norway
<<
Article
>> Home | TOC |
Index
Seminars in Orthodontics
VOL 8, NO 4 DECEMBER 2002
CONTENTS
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
Seminars in Orthodontics
Future Issues
Recent Issues
Seminars in Orthodontics
VOL 8, NO 4 DECEMBER 2002
to this topic is quite fascinating for the uniniti- ciously provides the reader with the long view
ated. From this point on, this issue takes a turn regarding our ultimate evolution from clinical
and looks at orthodontic risk management from practitioner to professional citizen and the cor-
a different perspective. responding insight into the concerns we all
Dr Spiedel's essay on administrative liability share in transitioning one's practice from the
will knock your socks off. If you only think of risk perspective of both the buyer and the seller.
management and practice management as deal- I hope you enjoy this issue and keep it handy
ing with the prevention and defense of a mal- for reference purposes. Adhering to sound in-
practice suit, be forewarned; don't read this be- terpersonal skills, as well as practice and risk
fore going to bed. It will give you nightmares. management principles, is what enables us to
Astute readers will note that administrative lia- practice day in and day out without driving our-
bility is far more dangerous to us than are causes selves crazy. Every practitioner can attest to the
of action based in the civil arena. fact that it is not the orthodontics that does us
Randall Berning has been a major contribu- in. Instead, it is the daily business grind that
tor to the dental literature, particularly on topics ultimately wears on our professional souls.
dealing with practice valuations and transitions.
Looking at our careers along a time line through Laurance Jerrold DDS, JD
risk management glasses, his contribution gra- Guest Editor
<<
Article
>> Home | TOC |
Index
rthodontists often misunderstand the na- and all persons will be properly credentialed
O ture of the doctor-patient relationship. Es-
sentially, it is nothing more than a simple con- 2.
and licensed.
The dentist will not use experimental proce-
tract wherein on one hand, patients seek dures on patients and will not undertake pro-
treatment from doctors with the expectation cedures that he/she is unqualified to per-
that their professional needs will be addressed, form.
hopefully resulting in a cure of some type, 3. The dentist and staff will keep current with
whereas doctors, on the other hand, consensu- scientific and technologic advances within
ally agree to treat these patients with the dual their fields.
expectations of affecting such a cure and with 4. The dentist shall obtain the patient's in-
receiving compensation for the professional ser- formed consent before treatment, keep pa-
vices they render. As with most contracts, there tients apprised of their clinical progress, and
exist responsibilities and obligations on the part complete care in a timely manner.
of each party. The lion's share of these duties 5. The dentist will be reasonably available for
falls on the dentist. In essence, we owe our pa- emergencies and will not abandon the pa-
tients the following: tient.
1. The dentist will use competent personnel, 6. The dentist will charge a reasonable fee for
training and supervising them appropriately, 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-
From the Program Director, Orthodontic Residency Program, St. lowed by their license as defined by law.
Barnabas Hospital, Bronx, NY. 8. The dentist will maintain appropriate and
Address correspondence to LauranceJerrold, DDS,ß), 82 Laurel
accurate records of the treatment rendered
Drive, Massapequa Park, NY 11762.
Copyright 2002, Elsevier Science (USA). All rights reserved. and will maintain the confidentiality of the
1073-8746/02/0804-0002$35.00/0 patient.
doi:10.1053/sodo.2002.127864 9. The dentist and staff will comply with all reg-
ulatory rules and regulations pertaining to A big problem that practitioners face is not
the scope of his practice and will abide by the knowing under what circumstances it is legal to
Code of Ethics. terminate the dentist-patient relationship and
how to appropriately do so. If handled properly,
Contrast the duties we owe our patients, with the saying goodbye to a patient is a very effective
following duties that our patients owe us: practice and risk management tool.
1. All instructions will be followed (ie, wearing Legally, the ways to end a professional rela-
elastics, diet control, seeking recommended tionship are (1) both parties agree to end it, (2)
consultations, maintaining oral hygiene, and the patient is cured or a course of treatment is
so on). completed, (3) the dentist or the patient dies,
2. All scheduled appointments will be kept. (4) the patient decides to unilaterally terminate,
3. Fees for services rendered will be paid. or (5) the dentist decides to unilaterally termi-
4. Patients will conform to generally accepted nate.
modes of behavior. An example of the first scenario is the patient
5. Patients will be truthful regarding their whose spouse or parent gets transferred across
health history. country in the middle of treatment. The second
and third are fairly self-descriptive. In the fourth,
It is in the application of these contractual obli- the patient is usually dissatisfied with some as-
gations that the dentist-patient relationship pect, clinical or administrative, of the care they
takes on meaning. received, and there is often some acrimony
It is quite easy to establish a dentist-patient when they leave. It is the last one that vexes most
relationship. The bottom line is, if you consen- practitioners. To unilaterally terminate a patient
sually offer someone professional advice (the from your practice, the following procedure
physical setting is irrelevant) intending for them should be followed.
to rely on the advice, and they do so, you have First, the patient must be given sufficient no-
just established a doctor-patient relationship. tice of the intent to withdraw as their practitio-
On finding the existence of such a relationship, ner of record. A letter should be sent both by
any professional opinion offered or treatment certified mail with return receipt requested and
rendered related to the patient's concern will by regular mail by using a certificate of mailing.
now be judged according to the appropriate By using both methods, the dentist assures that
standard of care. It does not matter whether or the patient was either notified or that an attempt
not you formally examined the patient, where to do so was made.
you did it, or whether or not you charged a fee Second, in this letter, the patient should be
for your services. informed of the reason (s) you are terminating
Must you treat everyone who presents to your your professional relationship with them (Ap-
office for treatment as a patient? No! Our courts pendix A). The legally recognized reasons for
have routinely stated that dental practitioners allowing doctors to take this action occur when
are not akin to common carriers or innkeepers the patient breaches one or more of the contrac-
and therefore need not open their doors to all tual obligations they owed you as previously
who seek their services. You cannot, however, noted.
discriminate against or refuse services to any Next, if the patient is at a point in treatment
person based solely on the fact they fall into one such that continued care is still required, he/she
of many legally protected classifications such as should be strongly urged to seek it. State in this
race, religion, gender, sexual orientation, na- letter that you will provide adequate time for the
tional origin, handicapping condition, and so patient to seek substitute or alternative care.
on. You are permitted to legally discriminate on Give the patient or parent notice of a specific
the basis of such criteria as limiting your practice time frame during which they should seek out a
to a particular specialty eg, the inability of the new dentist (ie, 30 or 45 days). Inform the pa-
patient to assume the financial obligations asso- tient that during this period of time you will be
ciated with treatment, a patient's inability to available for emergency care only or to offer a
abide by reasonably constructed office protocol, referral if necessary. If you have placed active
and so on. appliances in the mouth such as retractive me-
<<
Article
>> Home | TOC |
Index
chanics, reverse curve arches, or distalizing ing treatment for nonclinical (financial) rea-
springs, you should inform the patient to come sons.
in immediately so these appliances can be re-
moved, citing the possibility of negative sequel- As long as all of these guidelines and the dis-
lae if this is not done. missal protocol are followed, it will be difficult
Finally, inform the patient that on request a for patients to sustain any claim that they were
copy of his/her records will be forwarded to abandoned.
them or to a subsequent treating practitioner. By virtue of the existence of the doctor-pa-
Although you may legally be entitled to charge tient relationship, once we agree to treat a pa-
the patient a reasonable fee for the duplication tient, we now have an obligation, a legally rec-
of their records, you cannot withhold the pa- ognized duty, to conform our ministrations to a
tient's records because they owe you money. Do minimally accepted standard of care. What is
not place a financial stumbling block in the this standard of care, and how is it determined?
patient's way while attempting to end the doctor- In its barest sense, the standard of care is mea-
patient relationship. sured by what a reasonable practitioner, practic-
It should now be evident why it is important ing in the same or in a similar community, would
to continually document all instances of (1) the have done or refrained from doing under the
patient not following instructions or cooperat- same or similar circumstances. Reasonable prac-
ing in his/her care; (2) the patient breaking, titioners are permitted latitude accounting for
missing, and/or canceling his/her appoint- the fact that although diagnostics have the po-
ments; (3) times when the patient was abusive or tential for some minimal differential variability,
there often exists even greater variability relative
disruptive to the staff or others; and (4) the
to a practitioner's mechanotherapeutic ap-
patient's failure to meet his/her financial obli-
proach to a given problem. How far can one go
gations to compensate you for the professional
and still be considered reasonable? Generally,
services you rendered.
you are safe if what you are doing is also being
Obviously, a copy of this correspondence,
done by a respectable minority of your peers. In
along with the proof of mailing, should be
other words, if you are the only one, or one of
placed into the patient's record. Does dismissing only a few, who is espousing a particular philos-
patients from one's practice constitute the tort ophy or form of treatment, you are skating on
of abandonment? The answer is a qualified no. thin ice from a standard-of-care perspective.
Abandonment may be defined as the following: What is this community thing? In essence, it
was never considered fair to compare the pro-
1. Not giving the patient any further appoint- fessional behavior of a practitioner who prac-
ments or refusing to treat him/her before ticed in a metropolis and had the benefits of
his/her course of treatment is completed ultrahigh-tech tools, a stable of super specialists
without having a legally recognized reason to to rely on, the benefits of major university and
do so. Thus, unless a patient of record is in a teaching hospitals' support and continuing edu-
life-threatening situation, is bleeding pro- cation programs, and patients who could afford
fusely, or is in excruciating pain, all rarities in to pay for state of the art cutting edge treatment
orthodontic practice, you have a legally rec- with that of the doctor who practiced in a small
ognized right to terminate the doctor-patient town in which the community may be several
relationship on proper notice. years behind in terms of the equipment available
2. Not being available to a patient who requires for use, no specialists for miles, a local commu-
follow-up therapy to a procedure performed nity hospital at best, less access to educational
is also a form of abandonment. Thus, not advancements, and patients who paid for ser-
being available and not providing for substi- vices by bartering odd jobs.
tute or emergency coverage should a prob- Historically, this disparity favored the devel-
lem materialize has been viewed as abandon- opment of the locality rule to be used in deter-
ing a patient. mining the standard of care to be applied. Doc-
3. Constructively abandoning the patient, which tors in small local communities were not held to
often takes the form of extending or prolong- the standards of their big city brethren. As times
<<
Article
>> Home | TOC |
Index
changed, so did the law. With more schools turn- lack of informed consent. Breaking down this
ing out more doctors, the numbers of specialists legal doctrine into its individual elements is crit-
increased. More sophisticated equipment was ical to understanding the risk management ap-
being made available geographically, coupled plication of this basic fundamental legal, ethical,
with communications and travel evolving to the and practice management principle.
point when continuing education was more ac- First, one has to start by realizing that there is
cessible; the community per se began to shrink. a difference between consent and informed con-
Although the locality rule still exists in some sent. To put it into simple terms, consent is a
states for general practitioners, for specialists in threshold issue. In other words, before you can
the United States today, the community to which obtain one's informed consent to render any
we are compared is now national is scope. specific treatment, you first must obtain his/her
Who determines this community/national general consent for you to be his/her doctor.
standard? We do. Every doctor, even nonorth- For a patient's consent for treatment to be valid,
odontists, by virtue of his/her skill, knowledge, it must first be given voluntarily. Second, the
education, expertise, and experience, can go person granting the consent must be of legal
into court and offer to be an expert witness. It is age. Finally, this person must be of sufficient
up to the court to qualify that person as an mental capacity. Let's look at these a little closer.
expert, and many states have different legislative The issue of one's mental capacity is usually
and precedential criteria for making this deter- restricted to the elderly and to those with easily
mination. Once the qualification has been recognizable mental deficiencies; fortunately,
made, the function of the expert witness is mul- we rarely encounter this issue in clinical orth-
tifaceted. odontics. However, consider, for example, the
The first task is to act as an interpreter, mak- patient or parent who shows up at the consulta-
ing sure all the facts of the case are understand- tion visit acting in a manner that would lead you
able to lay jurors. The remaining tasks are to to believe he/she was impaired secondary to
educate and inform the jury because jurors are drug or alcohol abuse. If he/she respond irra-
the ones who will determine whether the doctor tionally to the treatment plan you suggest, the
was liable or not. This education and informa- issue of his/her capacity to grant consent, for
tion dissemination will focus on whether or not himself/herself or for others, may come into
the standard of care was or was not breached play.
and if it was, in what way. The next task is to give The more common problems relating to con-
an opinion as to the extent and severity of any sent fall into the other two categories. The legal
injuries claimed by the plaintiff. Finally, the ex- age issue arises in two contexts. The first in-
pert must give an opinion as to whether or not stance occurs when the patient is a minor, some-
the defendant's breach of duty actually, directly, one under the legal age of majority (usually 18).
or proximately caused the injuries that the plain- Because infancy generally precludes the patient
tiff sustained. from being able to grant consent, you must ob-
These duties are important because they par- tain the legal guardian's consent to treat that
allel the burden of proof that the plaintiff must child. If grandma, an aunt, the housekeeper, or
sustain to prevail in a malpractice suit. The even the noncustodial parent accompanies the
plaintiff must prove all four of the elements that minor to your office, you do not technically have
constitute the requisites of a malpractice claim: legal consent to allow you to render treatment.
(1) the doctor failed to conform his/her profes- You may obtain consent by telephone, letter, or
sional actions to a certain standard of care; (2) health care proxy; just make sure the appropri-
the doctor breached this duty; (3) there was an ate party grants it to you.
injury sustained; (4) the breach of duty was the A second example arises when you are treat-
actual, direct, or proximate cause of the injury ing a minor and right in the middle of orth-
suffered. odontic therapy, the patient turns 18, the legal
Although it is relatively straightforward to ap- age of majority. The patient looks at you and says
ply this information to the clinical setting, one of "take my braces off." Mom retorts "over my dead
the biggest problems facing orthodontists in the body." What do you do? The patient, not mom,
legal arena today is that of defending a claim for has the final say. Leaving the braces on after the
<<
Article
>> Home | TOC |
Index
patient requests their removal is technically a that they have a retrognathic mandible, aplasia
battery, defined as unauthorized, harmful, or of tooth no. 7, and that the recommended treat-
offensive contact with another. ment is a sagittal split ramus osteotomy followed
Finally, whether the consent granted was vol- by an implant prosthesis in the maxilla may be
untary or not may turn on a patient's claim that technically correct. However, it is infinitely more
they were mentally coerced into accepting treat- comprehensible to say that the patient's bottom
ment. This can arise when patients are told jaw is too small, they are missing an upper front
something like "if you don't undergo orthodon- tooth, and that the best way to solve their prob-
tic therapy you will lose all of your teeth from lem is by surgically bringing the bottom jaw
periodontal disease, suffer TMD problems, etc." forward and replacing the missing tooth with an
These types of remarks should not be made if implanted fake one. Replacing English with
there is no clinical evidence to substantiate them Dentalese is a very common mistake. Many prac-
or if they cannot be supported by scientific evi- tice management experts have recommended
dence. Nor should you imply that a patient's self that your treatment coordinator, a nondentist,
image or facial esthetics is such that they will be take over the consultation duties because they
socially ostracized or scorned because of their are less apt to fall into the language trap. The
"crooked teeth." You cannot make untrue or information can be given by anyone in the office
unfounded statements that "bully or guilt" pa- and in almost any medium. Each of us should
tients into accepting orthodontic treatment. develop our own style of information transmis-
Once we have received a patient's consent, we sion. Some will go high tech by computer imag-
must then obtain his/her informed consent to ing and prepackaged interactive CD programs
our ministrations. Patients must be told in a or proprietarily produced videotapes. Others
language they can comprehend such things as will find that educational brochures or forms
what the problem is, what you are going to do, work best. For some, a good old-fashioned con-
how you are going to do it, how long treatment versation with the patient is the way to go. How
will take, the costs, and so on. Sounds simple it is done and by whom is irrelevant; the key is to
right? Maybe, maybe not. Let's take each item ensure that regardless of how it's done or who
separately. But before we do that, there's just does it, the patient or parent understands the
one more thing we have to know; which legal information being transmitted.
standard do we use for disclosing this informa- Whatever treatment you are proposing may
tion? There are basically two standards for dis- be clear cut in your own mind, but you must
closure, and which one you use depends on learn to put yourself in the patient's position. If
which state you practice in. The more traditional any doctor were to recommend a certain proce-
standard is professionally based. Under this ap- dure to be performed on you, I'm certain that
proach, the only information you need to dis- being the intelligent person you are, you would
close to your patients is the information usually want to know what your other treatment options
disclosed by other doctors under the same or or alternatives were. You would want to know
similar circumstances. In other words, the doc- what the risks are for each one, what the conse-
tor decides what and how much to tell the pa- quences are, and what limitations exist. Why
tient. The more modern approach is the patient should our patients be different from us? They
need to know standard. The idea here is to give are not, and just because we are talking about
the patient all of the information that a reason- teeth, does not make it any different. Patients
able person in the patient's position would should be apprised of all viable treatment alter-
deem material to make a decision to accept or natives, even if you do not perform them. Re-
reject the proposed treatment. From an ethical member, it's their bodies we are invading; only
perspective, the patient-based standard is the they can give us the right to trespass and in what
preferred alternative. way. Patients also have the right to know what
Looking further at the individual elements, will happen if they choose not to undergo treat-
one can see that patients who have a problem ment, for in many cases this is, at least from their
with English should be spoken to in their native perspective, another viable alternative.
tongue for them to understand what is being How long treatment will take and how much
said. Next, speak plainly. To tell your patient it will cost are also factors involved in obtaining
<<
Article
>> Home | TOC |
Index
one's informed consent. It is a common pre- what information you told the patient or parent,
sumptive mistake for the doctor to make unsub- again evidenced by their signature or initial.
stantiated judgements as to what patients can or This would again be followed by the same entry
should be able to afford. The amount of time without such acknowledgement. There is a le-
that treatment will take may be of major signifi- gally accepted presumption that if something
cance to some whereas far less crucial for others. was written in the chart it was said or done; the
Your estimate of how long treatment will take corollary also applies, if it was not written, it was
should not "guild the lily" in the hope that the not said or done. Toward the bottom of the risk
patient will not be turned off. We should not be management ladder, your chart might say that
engaging in paternalism by believing that we the risks and alternatives were discussed without
know that the treatment we are recommending actually elaborating on the discourse. The next
is the best thing for the patient. The reason for rung might note that informed consent was dis-
this is that often our best intentions are not the cussed. Finally, the last rung would depict a
best thing for that particular patient because of chart devoid of any evidence related to a consul-
financial or temporal concerns, whether real or tation discussion—your word against that of the
perceived. patient's.
Finally, the patient must have the opportunity Many doctors err by believing that if informed
to ask and have answered all of their questions. consent was obtained at the onset of treatment,
Remember, you do not give informed consent, their patients no longer have to be informed if
the patient grants you his/her informed consent negative sequellae start to occur. If, in the mid-
for the treatment being proposed and/or ren- dle of treatment, you discover incipient root
dered. In other words, it's a two-way street; doc- resorption, periodontal disease, or decalcifica-
tors must provide their patients with sufficient
tions, you know that at some point in time a
information that in turn allows them to accept
decision may have to be made to discontinue
or reject the course of therapy being recom-
active treatment prematurely, even if the treat-
mended.
ment goals have not been realized. Who makes
A frequently asked question is, what is the
that decision? If you answered the doctor, you
best way to document that you have obtained the
are being paternalistic. If you say the patient,
patient's informed consent? Although a few
states require a signed writing, think of docu- how are they to know how far to go before
mentation as a risk-management ladder. The calling it quits, and what are the ramifications of
higher rung on the ladder that you stand on, the doing so? It should be obvious that the process
greater the risk-management protection. Video- of informed consent is an ongoing give and take
taping the consultation would be the best de- between the doctor and the patient based on
scriptor, the highest rung on the risk manage- what is occurring at any given point along the
ment ladder. The next rung down on the risk treatment time line. Ideally, informed consent
management ladder is audiotaping what was should be continually obtained allowing you to
said. Both of these offer wonderful protection to either continue on with treatment, or one's con-
prove what was said, but they also prove what was sent can be withdrawn should the patient decide
not said. The third rung down is to have the to terminate treatment. Either way, this decision
patient write in his/her own hand what was rec- is solely dependent on the sufficiency of the
ommended and what the risks and alternatives information provided.
are. By doing this, the patient can never claim The following are two good ways to obtain
he/she did not understand what you told him/ informed consent. The first involves a sending a
her; after all, he/she wrote it. Next, the most preliminary letter followed up by a conventional
commonly used tool is to distribute a printed informed consent form that is signed and kept
pamphlet or form that is acknowledged via the with the chart (Appendix B). This is a good
patient or parent's signature. Following that method for poor verbal communicators. The
would be a notation in the patient's chart that second method (Appendix C) uses a checklist to
this form or booklet of information was given identify the factors to be discussed at the consul-
but without his/her signature or initials. The tation visit. It is for better communicators; how-
next rung down is documenting in the chart ever, it requires that the doctor memorialize the
<<
Article
>> Home | TOC |
Index
informed consent discussion with the patient in determine how sound they are and whether the
the patient's record. bony support for them is adequate, (2) a film of
If you are like most orthodontists, when a new the skull to see the relationship of the teeth to
patient calls for an appointment, you send them the jaws and the jaws to one another, (3) pho-
a welcome to our office packet. It probably in- tographs of the face to assess the profile and the
cludes a letter thanking the patient for selecting effect that any proposed tooth or jaw movement
your office as well as confirming the time of their may have on it, (4) models of your teeth for both
appointment. You might also include a brochure diagnostic purposes and to use as a baseline for
about the doctor, staff, mission, etc., a map assessing one's progress and the results
showing your location, and possibly you include achieved, (5) a clinical examination to check for
a health questionnaire to be filled out in ad- cavities and gum disease, and (6) a review of
vance. Let me suggest that you start obtaining your medical history as there are many underly-
informed consent before you even meet the pa- ing physical problems that can impact on the
tient. Along with the other data, use the follow- success or failure of orthodontic therapy. A con-
ing two letters, "Your Right to Know" and sultation between the doctor and the patient will
"Information Concerning Your Treatment" (Ap- then follow.
pendix B). These two communiques set the At the consultation visit, your doctor should
stage for an interactive relationship between the explain to you what the actual problem is in a
doctor and the patient. Send the following letter language you can understand. You should be
(Appendix B) with your "Welcome to Our Of- told why correction is advised and how it is to be
fice" packet of information. achieved. You should also be informed of all
reasonable alternative methods of resolving your
particular problem. Because no form of medical
Your Right to Know
treatment is without the potential for some risks,
It's been said "an informed consumer is our best limitations, or compromises, you should be
customer." This has never been truer than in the made aware of those that pertain to your specific
provision of health care services. What should situation. Next, you should be told what result
you expect when you first visit an orthodontist? you can expect to achieve, and, finally, what will
You should be told what the problem appears to occur if no treatment is undertaken.
be, what tests are necessary to properly diagnose At this time, the fee for the services to be
the problem and formulate a treatment plan; rendered should be discussed in full and suit-
and, in the case of a child, whether now is the able financial arrangements made between you
appropriate time to begin therapy or whether it and your doctor. Make sure that you have had
is best to wait until a future date to initiate the chance to ask and have answered all ques-
treatment. Let's look at these individually. tions regarding your treatment as well as how
When a patient either desires to have his/her long it will take and the financial responsibilities
teeth straightened or is informed that orthodon- you are assuming. It is now time to begin treat-
tic therapy is recommended, he/she should be ment.
told why it is in their best interest to undergo The appliances or braces will be specifically
such treatment. The most common reasons for tailored to your particular problem. You may
seeking treatment are (1) to improve cosmetics require permanent and/or removable appli-
(crooked teeth, overbite, and so on), (2) to fa- ances. Once they are attached to your teeth,
cilitate other necessary dental work (to properly make sure you have been given careful instruc-
position teeth for capping, bridges or implants), tions on how to care for them. Check to be sure
(3) to correct a jaw discrepancy or skeletal dis- that your doctor will either be available or will
harmony (one's bite is off), or (4) to help pa- provide for coverage if an emergency (break-
tients maintain their periodontal status (the age) arises. You also need to be informed of your
health of the supporting gums and bone). responsibilities if the best results possible results
The next step is the gathering of necessary are to be achieved. Some of these are maintain-
information by obtaining diagnostic records. ing good oral hygiene, wearing rubber bands or
The following diagnostic materials are the one's a headgear as instructed, and keeping regularly
usually obtained: (1) radiographs of the teeth to scheduled appointments. You must continue to
<<
Article
>> Home | TOC |
Index
see your general dentist at least twice a year that they become meaningless to any one indi-
unless your orthodontist recommends other- vidual. In other words, the risk factors should be
wise. case specific. Because it's not always possible to
At the completion of active treatment you will know in advance what changes will occur, it
undergo a period of retention care. This phase reenforces the notion of ongoing informed con-
of treatment is necessary to monitor and help sent. This checklist can then be placed in the
maintain the results achieved. An appropriate patient's chart indicating that the identified fac-
retainer will be fabricated to maximize the sta- tors were discussed.
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.
Summary
Evaluating these can only be done if you have Understanding the doctor patient relationship is
been given sufficient information on which to the key to practicing good medicine. It is from
base your decision to undergo care. You have a this relationship that the duty to conform to a
right to know; to be educated is to be able to standard of care arises. The standard of care is
choose wisely. merely a euphemism for exercising sound clini-
After the patient's diagnostic appointment, cal judgment in a reasonable manner. Obtaining
when they are rescheduled for the consultation a patient's informed consent, when done appro-
appointment, give them two copies of the next priately, can be a very rewarding practice, pa-
letter, "Information Concerning Your Treat- tient, and risk-management tool. It fosters a
ment" (Appendix B). Ask them to read it at their close doctor patient relationship as all parties
leisure and bring one signed copy with them to are now on the same track regarding the diag-
the consultation appointment. nosis, treatment plan, and prognosis. Your pa-
Finally, for those who do not like the formal- tients will understand and appreciate more of
ity and associated stigma of the signed form the complexities that are involved with their
route, Appendix C is a checklist of risk factors orthodontic treatment, thereby having a greater
that should be identified and individualized for respect for the fees charged. It gives the doctor
each patient. In other words, not all of these an aura of openness, honesty, and caring. All in
risks will happen to everyone. The doctor should all, if used as a sword and not a shield, appro-
review this list when the diagnosis and treatment priately obtaining informed consent benefits the
plan is established, thus any potential risks will patient, the doctor, and the profession through
not be transmitted in so over broad a fashion open communication.
<<
Article
>> Home | TOC |
Index
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.
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.
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
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.
ceptive, and unfair debt collection practices. Ex- or truth-in-lending form specifies that you
isting laws and procedures for redressing these will seek reimbursement for any fees incurred
injuries are inadequate to protect consumers." as a consequence of your collection efforts,
Following the strictures of the statute, even when you may not threaten to or add them on at a
they may not specifically apply to us, will provide later date.
some guidance on how to proceed with collect- Although the purpose of this statute is to
ing fees. Under the FDCPA and many state col- protect the consumer from unethical and overly
lection statutes, a dentist attempting to collect a aggressive collections tactics, we shall see that it
fee is prohibited from the following actions: is possible to comply assiduously with the letter
and spirit of the law and still pursue a delin-
1. A dentist may not communicate, or threaten quent account aggressively.
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 Statute of Limitations
so if the patient or the patient's employer asks All too often, suing a patient to collect a fee
you not to. triggers a retaliatory malpractice claim. By wait-
2. You may not threaten to take action against ing until they think the patient is no longer
the patient that is illegal, that you have no within the statutory time frame to sue for mal-
intention of doing, or that you do not ordi- practice, the savvy practitioner may think it is
narily take. For example, you may not now safe to initiate a suit to recover an unpaid
threaten to sue a patient for a fee if you have fee. It is only after misinformed dentists are
no intention of doing so and, as a matter of served with papers alleging malpractice that they
course, never sue to collect a fee. You may, realize that their attempts to collect a small sum
however, threaten to turn the patient over for have backfired. The following are a couple of
collection if you regularly do so. You may tolling provisions that extend the period of time
never threaten a patient with physical harm. that patients have to initiate a lawsuit.
3. You may not communicate with the patient or
any member of his/her family or household Toll for Infancy
with such frequency or at such unusual hours
or in such a manner as can reasonably be If the patient is an infant, in most jurisdictions
expected to abuse or harass the patient. You this means under the age of 18, the statute may
are only allowed to call a patient between the be tolled (suspended) until the child reaches
hours of 8:00 AM and 9:00 PM unless you know majority. In other words, a child does not lose
these hours to be disruptive; however, you his/her right to sue you for malpractice until
may call a patient at other hours if attempts to how ever long the statutory period is after his/her
reach him during the permitted hours are 18th birthday. This toll for infancy is very impor-
unsuccessful. Although there is no generally tant to us as orthodontists because we treat a large
accepted calling frequency, legal precedent number of minors. Often, we wrongly assume that
suggests that contacting the patient more because the statute of limitations for malpractice
than twice a week about an overdue account appears to have expired, aggressive fee collection
is excessive. efforts may be pursued against the responsible
4. You may not communicate with anyone other party, usually the parents without any malpractice
than the patient, the patient's spouse, parent repercussions. Any collection effort should be tem-
(if the patient is a minor), guardian, execu- pered with the knowledge that the parents can sue
tor, or administrator. You may not discuss on the child's behalf until the child reaches the
past due accounts with a patient's fiance, sib- legal age or majority. After that, the child has the
ling, child, coworker, or roommate. right to sue on his/her own behalf, often for sev-
5. You may not knowingly attempt to collect or eral more years.
assert a right to any collection fee, attorney's
Continuous Treatment
fee, court cost, or expense, unless such
charges are justly due and legally chargeable In many jurisdictions, another exception to the
to the patient. Unless your original contract basic statute of limitations governing malprac-
<<
Article
>> Home | TOC |
Index
tice is called the continuous treatment doctrine. give them an opportunity to cure the default. If
Simply put, any continual treatment pertaining a resolution is not forthcoming, send a certified
to the initial therapy prevents the clock from letter to the patient informing him/her of your
running for statute of limitations purposes. The decision to discontinue treatment. The letter
statute of limitations on a patient's claim does should include the following: (1) the reason you
not begin to tick down from the day you remove are discontinuing treatment, (2) an offer to treat
the braces; it begins to run from when you either the patient on an emergency basis for a reason-
formally dismiss a patient from your practice or able period of time until alternative care can be
they come in for that last retainer check. We all secured, (3) the importance of seeking contin-
have patients who, despite being chronically be- ued care with an alternative provider and the
hind on their account, shamelessly manage to risks involved in the failure to do so, and (4)
come in month after month. Be advised that notification that the patient's records are avail-
each time they return to your office, they may be able to either the patient or to a subsequent
resetting the clock on the statute of limitations, treating practitioner.
hence making it more difficult to pursue their The key element in a dismissal letter is notice.
delinquent account safely. The patient must have notice that he/she is
being dismissed, when the dismissal becomes
effective, and notice of the consequences for
Dismissal Versus Abandonment
failing to seek alternative care. By following
One of the thorniest problems we face is how to these legal requisites when dismissing a patient
dismiss a patient who fails to pay for treatment. for nonpayment of a fee, you will minimize the
Misconceptions abound regarding how best to risk of a claim of abandonment by the patient.
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 Practical Considerations: Fundamentals
effective collections tool. Following established of a Fee-Collection Protocol
legal guidelines will reduce the risk of being Having examined some of the laws relating to
accused of having abandoned the nonpaying collecting fees, it is now time to use this knowl-
dismissed patient. When a dentist agrees to treat edge to create a simple and effective protocol to
a patient, a legal contract is formed. Each party collect outstanding account balances. Before
to this contract agrees to conform to both the any attempt is made to collect a delinquent ac-
express and implied terms of that contract. For count, indeed before an account even becomes
example, the dentist agrees to competently per- delinquent, three background essentials need to
form certain procedures in exchange for remu- be in place to insure successful collections. They
neration from the patient. are (1) a clear understanding of who the patient
The doctor-patient contract may be termi- is and who the financially responsible party is,
nated at any time during treatment. Although (2) a clear understanding of when an account is
the patient may end treatment at any time for delinquent, and (3) a clear understanding of
any reason, a dentist must be able to point to a what the fees are.
specific reason why treatment is being termi-
nated. Acceptable reasons to terminate treat-
A Clear Understanding of Who the Patient
ment usually involve the failure of the patient to
and Who the Responsible Party is
fulfill any of the express or implied contractual
duties he/she owes the dentist. If treatment is Before we can collect a fee, we need to know
being provided for an agreed on fee, nonpay- who is responsible for paying it. If we need to
ment of that fee is a breach of a patient's duty undertake a collections effort, we need to know
and is a legally acceptable reason for a dentist to as much about the responsible party as we pos-
terminate the doctor-patient relationship. If a sibly can by the end of the initial visit. You will
dentist decides to dismiss a patient for failure to never have a better information-gathering op-
pay for dental treatment, he/she should take portunity than during the initial encounter.
steps to avoid a claim of abandonment by the Look over your patient information form. At a
patient. Inform the patient of the arrears and minimum, for collections purposes you should
<<
Article
>> Home | TOC |
Index
be asking the following: name of patient; age of A person is not required to reveal his/her
patient; address of patient; name of financially social security number, but when someone
responsible party (even for adults); address of chooses not to, you have another credit alert.
financially responsible party; telephone num- Usually people who refuse to provide a social
bers, home, work (if a minor, mother and father security number on an information form do so
work), emergency; name of person who referred because they are aware of the value of the num-
patient; name of dentist; name of physician; and ber in skip tracing and enforcing judgments.
social security number. The same goes for people who refuse to give a
Each of these has a specific collections pur- street address or a telephone number and pro-
pose. Obviously, you need to know the patient's vide you only a postoffice box. They may be
name. His/her age is helpful for several reasons. privacy freaks, but they may also be poor payers.
If the patient is a minor, he/she lacks the capac- With all of this in mind, review and revise
ity to form a legal contract with you. If you begin your patient history form. Make sure you are
treatment, a minor has no legal obligation to pay asking enough information to enable you to put
you, and you may be committing a battery by a collections effort into play if necessary.
doing so. A patient's age also alerts you to the
statute of limitations issues.
A Clear Understanding of When an Account
The address tells you where to send the bill
is Delinquent
and can be an instant credit check. Although
prejudging a patient's financial status is never a Have a clear office policy of what constitutes a
sure thing, a patient's address can be an indica- delinquent account and what steps you need to
tor of his/her ability to pay and can suggest that take to address one. The farther behind a pa-
a customized payment plan may be advisable. If tient's account gets, the less likely you are to
possible, inquire how long a patient has been at collect it. As orthodontists, we usually bill
their current address. monthly. An account that is two billing cycles
Even if the responsible party is the patient, behind should be considered delinquent and
asking who the responsible party is may rein- deserves immediate attention.
force the fact that somebody is responsible for You should personally look over your state-
paying you. Is the responsible party's address ments every month. It is your money that is not
different that the patient's? Is it a post office being collected. Your staff gets paid each week
box? If so, why? Is this a divorce case? If the latter regardless of who has not paid you. If you stay on
is the case, you need more information. Many top of your accounts and begin your collections
collections headaches arise from divorce issues, efforts promptly, the number of accounts you
and you do not want to be caught in the cross- have to deal with will shrink to next to nothing.
fire. Then, and only then, can you delegate the job of
Get as many different telephone numbers as reviewing your statements to a trusted staff mem-
possible. In the event that you need to pursue a ber.
fee, calling at work usually gets more of a re-
sponse than calling at home. When you call
A Clear Understanding of What the Fees
someone at work, you usually do not get an
Are
answering machine, and, more often than not,
you will be put through directly. When we discussed legal considerations, we re-
Having the name of the person who referred viewed the need for a truth-in-lending form.
the patient, as well as that of the patient's dentist Either you or a trained staff member should
and physician, will prove helpful later if you review the fee plan with the patient as carefully
cannot find a nonpaying patient. Chances are as you do the treatment plan. Very often, pa-
that even if a deadbeat patient is not taking your tients listen more attentively to your fee presen-
calls or answering your letters, he/she has not tation than your case presentation. My prefer-
severed all contact with their friends and other ence is to have the doctor discuss the fees with
health care professionals. In the event that you the patient. If you delegate this duty, train your
need to skiptrace a delinquent account, these staff to pick up the clues that may indicate that a
numbers will assist you. patient may be a slow payer. No matter who
<<
Article
>> Home | TOC |
Index
discusses the fee structure, make sure you leave you need to be a bit more decisive (actuarially,
nothing to the patient's imagination. 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
The Collections Effort how much they owe you and welcome them to
By using our knowledge of the legal issues in- call your office to discuss his/her account. Be
volved and building on the fundamentals we just sure to keep copies of all letters you send the
discussed, the actual collections efforts should patient. Keep in mind that no matter what your
go smoothly. You need to be systematic and fee, patients think orthodontic work is expen-
timely to be effective. The following is a sug- sive. Your statement will often be permanently
gested protocol. positioned at the bottom of the patient's pile of
bills, and unless you take some action to move
First Effort yourself to the top you may never get paid. Un-
like the utility, cable, luxury car, or private
Your first contact should be when a patient's
school bill, our bill does not have a must-pay
account is 30 days past due. Begin your collec-
imperative attached to it. We have not condi-
tions effort by jotting a note on their statement
tioned our patients to take it seriously, and there
that reads something like; "Please call if you have
is very little penalty attached to paying us late.
any questions about your account" or "Please
Remember, the truth-in-lending statute does
bring your account up to date." The doctor, not
not allow for retroactively adding a penalty for
a staff member, should write the note and either
paying late. You should inform the patient, in
initial or sign it. The response from those pa-
writing, at the initial fee presentation that there
tients who "just forgot" is dramatic. This will be
is a penalty for paying late, either through an
your single most effective collections tool.
interest charge or a late payment fee. You can
Often, the doctor is seen as distant from the
always waive the penalty, but you cannot add it.
financial concerns of the practice. Often, the
Virtually every other creditor charges late fees,
doctor either feels discussing money is unseemly
and people are conditioned to paying them.
(something I will address in the final section), or
There is no reason we should not keep the op-
he/she feels embarrassed discussing it. Patients
tion available to us and use it.
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 Third Effort
promptly, you should not expect to be paid
If your patient becomes more than 90 days plus
promptly. That is why it is important that you,
in arrears, you need to send a very decisive letter;
not a staff member, initial this first collections
one that spells out the potential penalty in store
effort. Try not to do this with a telephone call. A
for failure to pay. Like the last letter, this should
telephone call takes too much of your time, is
not say too much, but it should inform the pa-
usually delegated to a staff member who really
tient that you will not be able to continue treat-
does not relish the job and will not do it well,
ment unless the account is brought up to date.
and more often than not requires repeated ef-
Be careful not to give patients the idea that you
forts to successfully contact the debtor. All you
are abandoning them, but do let them know that
want is for the patient to bring the account up to
you reserve the right to dismiss them. What usu-
date. A letter, as opposed to a telephone call,
ally happens at this point is that the responsible
can be designed to invite only a positive re-
party calls to discuss developing a new payment
sponse.
schedule. You are not obligated to accept this
overture and often fairly but firmly insisting on
Second Effort
being paid will get you paid. If you do choose to
Although the response to your first effort will work something out with the patient, do not
bring in many of your easily collected overdue accept smaller payments over a greater number
accounts, you will still have a handful of ac- of months.
counts that require your attention. At this point, Legally, what is occurring is that you and the
you are in the 60- to 90-day past due range, and patient are forming a new contract. You should
<<
Article
>> Home | TOC |
Index
be prepared to dictate the new terms. Whatever previously, you will be perfectly within your
you agree to should be backed up with some rights to inform the patient you will not con-
guaranteed payment method such as a series of tinue his/her treatment. In the legal section on
post-dated checks, a nonrecourse credit card au- dismissal, we learned that notice is the key ele-
thorization, or permission to draw the money ment in patient dismissal. To be sure that the
directly from the patient's checking account patient gets fair notice, this letter must be sent
each month. A promise to pay that is not backed by certified mail, return receipt requested. The
up by one of these initiatives will put you right legal presumption is that the addressee received
back where you started when you send your next the letter even if it is returned to you. The
statement out. addressee is responsible for knowing what is in
I do not like collections agencies. Why should that letter and failure to pick it up does not
you give up 20% to 50% of your money to let absolve a responsible party of that burden. You
them collect for you? Collections agencies do should send another copy of the dismissal letter
not know or care about your relationship with at the same time you send the certified one, but
the patient. Their heavy handedness often pro- place this copy in a plain white envelope and
vokes patients, invoking a retaliatory lawsuit or handwrite their address. More often than not, it
resulting in a complaint to a disciplinary or reg- is this letter to which the patient responds.
ulatory agency. Following the simple protocol Virtually without exception, your office will
outlined earlier should be just as effective as get a call right after the letter is received. Most
using a collections agency; the difference being, often, the response is one of incredulity, "I never
you get to keep all of the money and still main- knew I owed you money," or "I never received a
tain a good rapport with your patient. bill." Take it in stride; insist that the account be
What if your patient promises to pay but in- brought up to date and continue with treatment.
stead of paying the $200 a month you agreed on, Dismissing the patient is a tactic of last resort,
begins to send you check for $20 a month? If you but it is one that is certain to get results as long
accept this check on a regular basis without pro- as you do it by the book.
test, you have reached what is legally known as a
new "accord and satisfaction." This means that
the patient has unilaterally renegotiated his/her Suing a Patient
contract with you and you have accepted these
terms. What if your patient owes you $1,000 and Suing a patient to collect your fee is not an
sends you a check for $500 marked "payment in option in my collections protocol. If you imple-
full." If you accept the check, you have indeed ment this collections protocol as previously de-
been paid in full. There is nothing you can write scribed, you should never have an account that is
on the check that will negate the words paid in delinquent enough to warrant filing a lawsuit,
full if clearly written. Your choices are to either even in small claims court. Begin your collection
cash the check accepting it as full payment or efforts early and cut loose the nonpayers early.
return it to the patient. Structure your fee schedule so that even if you
These efforts at collecting overdue accounts have to dismiss a patient, you are at least at a
will resolve most of your outstanding account break-even point. A nonpayer understands why
balances, but remember successful resolutions he/she is being dismissed and will call you to
decrease exponentially with the amount of time make arrangements to pay. If you sue that same
the account is in arrears. If you still have not person to collect your fee, you are needlessly
been successful in working something out by this antagonizing him/her, inviting him/her to re-
point, you either have to accept the fact that you spond with a complaint to the state dental board
are working for nothing or cut your losses and or to file a retaliatory lawsuit. Even if you win
dismiss the patient from your practice. your suit, you are often left with a judgment that
is difficult, if not impossible, to collect.
There is no reason that orthodontists, who
Dismissing the Patient
have the luxury of presenting clear fee arrange-
You have no obligation to continue to work for ments for an ongoing treatment plan, ever have
nothing. If you follow the guidelines discussed to resort to a lawsuit to collect our fee.
<<
Article
>> Home | TOC |
Index
he answer to the question asked in the title gages another to perform a service and who
T of this article, "Am I My Brother's Keeper,"
is yes but only when the brother has a specific
controls or has the right to control that person
regardless of whether or not that control is ac-
professional relationship with the brother's tually exercised. A servant or employee is the
keeper. Therefore, the purpose of this article person directed to perform the service. An In-
is to describe the relationships between the dependent contractor is one performing a ser-
brothers and when an innocent party (one vice for another but is not under the control of
brother) will be held liable to an injured patient the employer as to the time, manner, or meth-
as a result of the negligence of another (his ods used to complete the service.
brother). This transfer or substitution of liability Two cases decided in Minnesota provide us
is described in case law and is based on the with legal definitions of vicarious liability and
common law. It is not as a result of legislative respondeat superior. In the first case, decided in
acts or the rules and regulations of administra- 1961, the court stated,
tive agencies.
The discussion begins, as it must, with the Vicarious liability is based on a relationship between the
definition of terms used to describe the complex parties, irrespective of participation, either by act or omission,
legal doctrine of vicarious liability and a special of the one vicariously liable, under which it has been deter-
form of vicarious liability know as respondeat mined as a matter of policy that one person should be liable
superior that is limited to a master-servant or for the act of the other. Its true basis is largely one of public
employer-employee relationship between the or social policy under which it has been determined that,
parties. A master or employer is one who en- 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 Thus, vicarious liability is the transfer or sub-
York at Stony Brook, Stony Brook, NY. stitution of liability, making an innocent party
Address correspondence to Burton R. Pollack, DDS, JD, MPH, liable for the negligent act of another. For the
School of Dental Medicine, State University of New York at Stony
substitution to take place, there must be a rela-
Brook, Stony Brook, NY 11794.
Copyright 2002, Elsevier Science (USA). All rights reserved.
tionship between the parties. However, if the
1073-8746/02/0804-0004$35.00/0 relationship is one of master (employer) and
doi:10.1053/sodo.2002.127866 servant (employee), it becomes a subset of the
generic doctrine of vicarious liability and is the court will find the dentist liable for the injury
known as respondeat superior. to the patient despite his/her claim of inno-
In an earlier case, the court defined Respon- cence for the assistant's negligent act. The the-
deat Superior by stating, ory that courts apply to transfer liability is (1) the
dentist/employer hired the employee (assistant)
Under the doctrine of respondeat superior, according to
and thus placed him/her in a position to com-
the generally accepted view, vicarious liability to third per-
sons is imposed upon the master [employer] for his servant's
mit a negligent act, (2) the employer benefits
[employee's] torts, not because the master is at fault, or
(profits) from the services performed by the
because he authorized the particular act, or because the employee, (3) the employer has the right of
servant represents him, but because the servant is conducting control over all acts performed by the employee,
the master's business, and because the social interest in the and (4) the patient played no part in the selec-
general security is best maintained by holding those who tion of the employee and was not able to control
conduct enterprises in which others are employed to an the acts of the employee.
absolute liability for what their servants do in the course of In other words, the major factor is that of
the enterprise. Where the doctrine of respondeat superior is control, whether exercised or not.
relied on as a basis for recovery by a third person, the tortious
In our second hypothetical situation, the
act of the servant committed in the scope of his employment,
orthodontist/employer, when hiring the dental
and not the master's fault or the absence of it in hiring or
retaining the servant, is the basis of liability.2
assistant informs him/her of all the laws regulat-
ing the practice of dental assisting in the state in
In summary, vicarious liability is the transfer which his/her practice is conducted. However,
of liability to an innocent party for the negligent the dentist informs the assistant that he/she
act of a guilty party; it covers all transfers of does not want him/her to perform a specific
liability regardless of the relationship of the par- service even though the statute regulating the
ties; and it is a generic term. Respondeat supe- practice of dental assisting permits the task. Dur-
rior is the same transfer of liability but applies ing the dentist's absence, the assistant does what
solely to the employer-employee relationship. the dentist had prohibited, and the patient suf-
Let us first examine respondeat superior as it fers an injury. A lawsuit follows. The dentist is
applies to the orthodontic practice setting. In sued, not the assistant. The dentist's defense is
the typical orthodontic practice, there is likely to that he/she specifically prohibited the act the
be, in addition to the dentist/owner of the prac- assistant performed. Again, as in the prior situ-
tice, one or more dental assistants, and in some ation, the court will find the dentist liable, ap-
situations a dentist employee. These are the ser- plying the same legal logic.
vants (employees) of the master (employer) Now for the third hypothetical set of facts.
dentist/owner. The dentist, in hiring the assistant, gives the
The following are several hypothetical situa- assistant a copy of the law defining his/her per-
tions that show the transfer of liability under the mitted duties and tells the assistant not to per-
doctrine of respondeat superior. In the first, the form any duty not permitted by the law. During
orthodontist/employer on hiring a dental assis- the dentist's absence the assistant does, and the
tant instructs him/her to exercise care in the patient is injured. The dentist is sued, not the
performance of his/her assigned duties. While assistant. The dentist's defense is that the assis-
assisting the dentist in the placement of fixed tant was aware of the laws regulating his/her
appliances, and as a result of the dental assis- profession and violated the law despite the den-
tant's negligence, the patient's cheek is burned tist's admonitions; thus, the assistant should
from the acid etch material requiring repair by a have been named as the sole defendant. Follow-
plastic surgeon. The patient experienced consid- ing the same logic of the first set of facts de-
erable expense, and the injury is permanent, scribed earlier, the court will hold the dentist
leaving a slight scar. The patient then com- liable. In summary, under the doctrine of re-
mences a malpractice suit naming the dentist spondeat superior, the dentist hired, he/she
and not the assistant. In his/her defense, the profited, he/she controlled, which is why he/
dentist states that the injury to the patient was she should be responsible.
the fault of the assistant; the assistant was told to However, under common law, that is law
be careful, and he/she was not. In this situation, based on moral and ethical principles that
<<
Article
>> Home | TOC |
Index
Am I My Brother's Keeper? 207
guides courts in decision making, a party injured Under the doctrine of respondeat superior, the
by the negligent act of another should be com- employer/dentist may be held solely liable for
pensated for the injury, and further, the party the negligent acts of the employee. Likewise, the
guilty of causing the injury should be made to employee/dentist can be held solely liable to the
provide the compensation. In the hypothetical injured patient for his/her own negligence. The
situations presented previously, how is the com- joint suit option has some advantage for the
mon law satisfied? Clearly, the injured party has patient. In questionable situations, when it is not
been compensated, but the guilty party, the as- clear to the plaintiffs attorney who will be held
sistant, has not been held accountable. The ap- liable by the court, it may be prudent to sue both
parent defect can be corrected in that the em- and let the court decide who is to compensate
ployer is able to sue the negligent employee for his/her client for the injury. Another more com-
indemnification of his/her losses. This relief is pelling reason to enter suit against both is when
rarely, if ever, taken advantage of in the scenar- the plaintiffs attorney discovers that one of the
ios described, but it does occur in a hospital possible defendants does not posses enough in-
setting when the hospital is held liable for the surance coverage to compensate the patient for
negligent act of a nurse, and the hospital's in- the injury. It may be that to find the "deep
surance company then sues for indemnification pocket" both will have to be sued.
of its losses from the nurse's insurance company. There are other situations in which the doc-
The fact that the relief is available, even if rarely trine of vicarious liability, exclusive of respon-
exercised, satisfies the common law principle. deat superior, may apply. These are not related
What courts rarely admit is that they often to the employer-employee relationship. They oc-
look for the one with the "deepest pocket" in the cur regarding referrals to and from other prac-
transfer of liability. An employee is not likely to titioners.
have the financial resources available to com- First, let us examine liability issues in referrals
pensate an injured party; the employer has the made by an orthodontist to an orthognathic sur-
"deepest pocket." An easily understood situa- geon for treatment when the surgeon is negli-
tion, unrelated to dentistry, is that if your auto- gent in his/her care of the orthodontic patient,
mobile was struck by a Macy's truck that ran a and the patient, as a result of the negligence,
red light, who would you expect the court to suffers an injury. The attorney for the patient
hold liable, Macy's or the driver of the truck? discovers that the surgeon has insufficient insur-
Clearly, Macy's, despite that it can be shown that ance to fully compensate his/her client for the
all Macy's truck drivers were instructed to obey injury. So, the attorney joins the orthodontist in
the motor vehicle laws when making deliveries. the suit as a primary defendant. The jury may
The issue of fixing liability on the party having find the orthodontist not liable unless (1) the
the "deepest pocket" will be addressed in some orthodontist knew, or should have known, that
of the following examples. the surgeon to whom the referral was made was
The employer-employee relationship as it re- impaired or incompetent; (2) the orthodontist
lates to nondentist employees in the transfer of took an active part in the treatment provided by
liability has been covered. However, there are the referred to practitioner; or (3) the orth-
situations in dental practice, including orth- odontist benefited in some material way from
odontic practice, in which the employee is an- the referral.
other dentist. In another hypothetical set of Little need be said about the liability of a
facts, the employee/dentist performs a service referring practitioner when the referral is made
that results in an injury to one of the patients of to a practitioner who is known by the referring
the practice. The injured patient alleging negli- practitioner to be incompetent or impaired or if
gence on the part of the employee/dentist has the orthodontist took an active part in the treat-
the following options: (1) sue the employee ment provided by the referred to practitioner, as
without suing the employer, (2) sue the em- in both cases, liability is easily attached. How-
ployer without suing the employee, or (3) sue ever, the third basis for the transfer of liability
them jointly. requires some further examination. Clearly, if
In legal terms, the parties share joint and the referring practitioner receives a financial
several liability for each other's negligent act. benefit from the referral, such as receiving a
<<
Article
>> Home | TOC |
Index
208 Burton R. Pollack
percentage of the fee collected by the practitio- it looks like a partnership to the patient, it will
ner to whom the referral was made, commonly be treated as a partnership by the court, and
known as a split fee, the courts will transfer each dentist may be held jointly or severally
liability to the referring practitioner. Another liable.
situation to transfer liability is if the referring This transfer of liability does not take place if
practitioner receives a set amount of money for the practitioners are organized as a professional
each referral made, commonly called a kick- corporation or professional association (title de-
back. A more complex situation may take place pending on state law). Only the guilty party and
when a generalist owns the office building in the corporation or association is held liable. A
which a specialist rents space; the rental is not new legal entity, the professional limited, liabil-
comparable to that of professional space in the ity company, now often takes the place of the
community, and it is shown that the generalist professional corporation or association. It also
refers all his patients needing specialty care to protects individual members from the transfer
the specialist who rents space in the building. of liability and is available to members of the
Thus, the owner practitioner benefits materially dental, medical, legal, and other professions us-
(the increased rent that is tied to a percentage of ing the abbreviations PLLC or similar designa-
the specialist's earnings). Under these circum- tion as permitted under local law.
stances, the referring dentist may be held vicar- Some courts have held that when the relation-
iously liable for the negligent act of the special- ship is one between a principal and an indepen-
ist. dent contractor, the principal may be held liable
Gifts given at the end of the year from a for the negligent acts of the independent con-
specialist to his/her referral sources are not con- tractor. As defined previously, an independent
sidered such a sufficient material benefit to sup- contractor is one performing a service for an-
port a transfer of liability unless the gifts are other but who is not under the control of the
excessive in value. employer as to the time, manner, or method to
Associate practice relationships also may re- complete the service. However, courts look care-
sult in the application of vicarious liability. Part- fully into the relationship, often rejecting the
nerships are particularly risky in the transfer of independent contractor relationship in favor of
liability. Suppose, while partner A is absent from one of employer-employee, applying the doc-
the office and without his knowledge partner B, trine of respondeat superior in transferring lia-
because of his negligence, injures a patient. Can bility.
A be held liable to the patient? The answer the
courts have provided is yes. The legal logic is
Summary
that the partners are united in interest because
each partner profits from the services provided Under the legal doctrine of respondeat supe-
by the other partner. Therefore, under the doc- rior, an orthodontist will be held liable for the
trine of vicarious liability, they may be held negligent acts of his/her employees, including
jointly or severally liable as described earlier; the acts of an orthodontist employee (but not nec-
plaintiffs lawyer has the choice of whom to sue essarily an orthodontist who is an independent
or both. contractor); he is his brother's keeper.
In another practice relationship two dentists Under the legal doctrine of vicarious liability,
who are not partners share an office, and in it an orthodontist may be held liable for the neg-
they share a receptionist, the waiting room, the ligent acts of a practitioner whom the orthodon-
dark room, and at times use each other's treat- tist has referred his patient to (if he/she exer-
ment rooms. In addition, their names are on the cises a degree of participation or control over
common door to the office. As a result of one the referred to doctor's treatment or, if the re-
dentist's negligence a patient is injured. Under ferral was negligently made); he/she may be his
these set of facts, will the courts treat them as a brother's keeper.
partnership? Will the innocent dentist be held Under the same doctrine of vicarious liability,
either jointly or severally of liable? The innocent an orthodontist can be held liable for the negli-
dentist will claim they are not partners; in fact, gent acts of his/her partner orthodontist (or
they are not. However, courts have ruled that if those doctor's who the public might ostensibly
<<
Article
>> Home | TOC |
Index
believe are united in interest with him/her); he sional limited liability company; he is not his
may be his brother's keeper, at the option of the brother's keeper,
injured patient's attorney.
Orthodontists who practice as a professional
corporation, professional association, or profes-
sional limited liability company are immune References
from vicarious liability for the acts of other mem- 1. Nadeau V. Melin, 110 N.W. 2d 99, (Minn. 1961).
bers of the professional association or profes- 2. Porter V. Grennan, 16 N.W. 2d 906, (Minn. 1945).
<<
Article
>> Home | TOC |
Index
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- lawsuit. Next is the securing of detailed in-
tion of Orthodontists Insurance Company formed consent consultations with the patient
(a Risk Retention Group), which writes profes- and/or family, followed by placing a signed ac-
sional liability insurance, I have had a detailed knowledgment of this act in the patient's record.
view into the lawsuits that are filed against orth- Finally, being able to empathize with the patient
odontists. During the handling of those lawsuits, and focus on their treatment needs are also
I have been able to look into the practices of the critical in this analysis.
involved doctors and have made some general To illustrate these general observations, I will
observations about the causes and effects that summarize the details of several actual lawsuits
most frequently appear to have contributed to handled through our office, minus, of course,
the filing of the actions. Often, these patterns any identifying information to protect confiden-
repeat themselves, notwithstanding major varia-
tiality. The summaries will examine the nature
tions in a patient's diagnoses and treatment, the
of the patient's complaint. What did the doctor
locations of the practices, and the doctors in-
volved. say and/or do preceding the filing of a com-
What are some of the major issues that I see? plaint? Were office staff members or office pro-
Communication breakdown is an important cedures involved and in what way? What treat-
one. I am referring to listening and speaking ment was performed, and what was the
skills, providing critical information to the pa- outcome? Was the outcome a surprise to the
tient, and the coordination, or lack thereof, with patient or family; was it a surprise to the doctor?
other health care practitioners. Secondly, good What was the demeanor of the patient during
initial records are obviously needed to formulate the course of treatment? Were there signs of
appropriate diagnoses and treatment planning; dissention, such as arguments with the doctor or
they are also the front line in the defense of a staff, requests for records, or concerns about
fees? One doctor told me that orthodontists
love to work on teeth; the problem is, he said,
From the American Association of Orthodontists Insurance Com- the teeth have humans attached. It is the
pany, St Louis, MO. breakdown of the humanity inherent within
Address correspondence to Elizabeth Franklin, AAOIC, 401 the relationship that results in lawsuits. Let's
North Lindbergh Blvd, St Louis, MO 63141.
Copyright 2002, Elsevier Science (USA). All rights reserved.
look at some routine human problems to see if
1073-8746/02/0804-0005$35.00/0 they can be tempered enough to minimize
doi:10.1053/sodo.2002.127867 potential exposure.
defense in the event that a malpractice claim is tist and the general dentist who had seen the
made. patient only a few times during that 5-year pe-
Good communication with the patient or riod. The dental records did contain a signed
family is another effective way a doctor can pre- informed consent form.
clude a lawsuit. Listen to the patient. Is there The expert witness who examined the records
pain, bleeding, a compromise in function? Ad- on behalf of the doctor opined there was evidence
dress these issues satisfactorily and explain them of neglect and recommended the case be settled.
fully. A patient who understands what is taking He based this decision on several factors. First, the
place, the reason why specific treatment or re- depth and degree of the decay substantiated its
ferrals are recommended, and believes that the presence for a number of years. Secondly, because
doctor is concerned and sympathetic, is less the caries was located under the bands, he be-
likely to sue. lieved that washout was evident and no amount of
Always have an informed consent discussion adequate tooth brushing would have prevented
with the patient. Emphasize those risks that are the damage; and the significant number of re-
most important to his/her particular treatment placed or recemented bands also indicated inade-
and repeat these concerns during treatment, quacies with cementation. In addition, he also
redocumenting this discussion. Have the patient noted that the records were written in code with
sign a form memorializing the conversation and very little detail. Notations of "poor oral hygiene"
keep it with the records. This process manages a were not explained. The plaintiff also contended
patient's expectations and minimizes unpleasant that much of the treatment was performed by staff
surprises. A patient who realizes they have been rather than by the doctor.
clearly apprised of potential problems will have The expert also believed there was another
far more difficulty suing should that problem significant issue that encouraged early settle-
materialize. ment of this case. The patient had testified that
late in 1996 he began taking Acutane (Roche
Labs, Hoffman-LaRoche Inc., Nutley, NJ) for
Case Two treatment of severe acne. This drug is known to
cause xerostomia that can contribute to gingival
Facts
inflammation and tooth decay. The boy said he
The patient in this case was a 10-year-old child at apprised the orthodontist of this fact; however,
the start of treatment in 1992. The diagnosis was the orthodontist disputed this. The expert be-
an anterior protrusion. His treatment lasted 5 lieved that the presence of acne in a teenager
years, and over that period of time, he presented should alert a doctor to the possible use of this
with more than 60 instances of loose bands or drug. Although he opined that the orthodontist
broken appliances. He missed a significant num- should have made an inquiry, because the dam-
ber of appointments. The doctor documented age was under the bands, he did not believe the
poor oral hygiene several times in the records. Acutane was the primary cause of the decay.
In 1997, a general dentist examined the patient Until the time this case was settled, the pa-
and found extensive decay and demineralization tient's family had spent approximately $20,000
in 50% to 80% of the patient's teeth. Most of this on restorations. The initial demand had been
damage was visible to the naked eye. After the $350,000; the case was settled for $150,000. What
orthodontic appliances were removed, the areas went wrong here? The first issue appears to be
of decay and lines of decalcification were consis- one of office organization. The doctor reported
tent with the placement of the bands and brack- that he was frequently absent from his office
ets. The patient's current general dentist opined handling personal matters. He believed that it
that this condition should have been obvious to was his female partner who may have done some
any dentist, especially an orthodontist. He im- of the work; however, the plaintiff could not
mediately cleaned the teeth and applied a resin differentiate whether it was she or any of the
to provide temporary relief in the sensitive areas. dental assistants who was performing a signifi-
He anticipated that multiple units of restora- cant amount of the treatment. The patient tes-
tions would be necessary in the future and that tified that the doctor was not paying sufficient
some root canals would also likely be required. attention to his care. The records did not reveal
The family filed a lawsuit against the orthodon- who performed which procedures. In short, they
<<
Article
>> Home | TOC |
Index
contained little detail and were incapable of pro- perspective. It appears that information about
viding much support during the defense of the the effects of this drug on dental health is com-
case. This was especially true regarding the doc- monly available. Because acne is a frequent teen-
umentation of poor oral hygiene. Those entries age concern, a careful orthodontist needs to be
were not clarified, and the parents testified that proactive and make an appropriate inquiry, es-
they were never made aware of any problems in pecially if decay seems to be a problem. Any
that regard. Had there been details provided in change in the patient's complexion should be
the notes, perhaps even initials from the young readily visible. In this case, the doctor seemed to
man acknowledging these discussions, prefera- be unaware of many details regarding his pa-
bly with copies of the letters sent to the family tient, perhaps because of overdelegation. Better
apprising them of the problems, there might not focus and attention to detail might have pre-
have been a lawsuit. Better records would have at cluded the lawsuit.
least made the case easier to defend.
The matter of the excessive number of bro- Risk Management Recommendations
ken appliances was a shouting match between
Beware of overdelegation of duties to assistants
the patient and the doctor. The doctor believed
as small but important details can go unnoticed.
the boy was to blame for eating inappropriate
Regardless of the problem, if a patient is not
foods. The boy, of course, denied that he did so.
keeping appointments, has poor oral hygiene, is
The expert believed that the number of repairs
eating inappropriate foods, or has excessive
and replacements was far too extensive to be
damage to the appliances, the orthodonist
caused by anything other than poor bonding
should advise the patient and his parents of the
and banding technique. One very important
consequences. Communicate with the patient,
point to be considered from a review of this case
and in the case of minors, at least occasionally
is the possible need for early termination of
with the parents. In addition, notify the general
treatment secondary to noncooperation. If the
dentist of any problems with the patient's treat-
doctor believed there was poor oral hygiene and
ment. Document all of these communications in
that too many bands needed replacing, perhaps
the patient's chart. Ensure that the patient is
he should have stopped the treatment when the
regularly visiting his general dentist for cleaning
problems continued. This would have prevented
and checkups; document this fact, as well.
the exacerbation of the damages.
Orthodontists are charged with the responsi-
bility of recognizing and addressing (usually by Case Three
referral to other specialists or general dentists)
Facts
routine dental issues, such as caries or periodon-
tal disease. Merely having a sign on the office In this case, the patient was a young adult fe-
wall advising patients to routinely see their gen- male. Her diagnosis was a Class III malocclusion
eral dentist for cleaning and examination does secondary to a combined mandibular progna-
not insulate the orthodontist from responsibility thism and slight maxillary retrusion. The exam-
if the patient fails to comply. It is the orthodon- ination revealed a facial asymmetry, abnormally
tist's responsibility to determine compliance shaped lower anterior crowns, a midline devia-
when significant problems persist. Communi- tion, and partially impacted third molars. The
cate directly with referral doctors to ensure it is treatment time was projected to last 2 to 2V£
safe and appropriate to continue orthodontic years. The plan was to slenderize the abnormally
treatment. Refusal to continue treatment until shaped teeth. The doctor did not want to extract
there has been compliance must be seriously lower premolars because he believed that would
considered. That may interfere with your pro- flatten her profile and subject her to other den-
jected treatment schedule, but it will possibly tal problems such as root resorption or peri-
save you from having to defend a lawsuit later odontal disease. He anticipated extracting the
after the damage occurs and the patient forgets third molars at the end of treatment, which he
that you told them to have regular checkups. In believed would allow for settling of the upper
the end, you will be held accountable. and lower arches.
The issue involving Acutane provides another After 3 years of treatment, the braces were
interesting observation from a risk management removed, and the patient's maxillary anterior
<<
Article
>> Home | TOC |
Index
teeth were significantly flared and tipped labi- fact would have diminished his credibility. The
ally. An anterior open bite was present, and lip young plaintiff on the other hand was a sympa-
closure was difficult. The patient alleged her thetic witness. She testified that when she ex-
dentition was deformed, and she was facially pressed dissatisfaction with the outcome of her
disfigured. When she expressed consternation treatment, she was immediately dismissed from
to the doctor, she said she was abruptly dis- the practice. She felt abandoned and very em-
missed from the orthodontist's practice. At that barrassed by the condition of her teeth.
time, he gave her a small fee refund of less than The damages in this case were not significant;
$500. She immediately saw another orthodontist the plaintiff alleged expenses of $7,000, includ-
who extracted eight teeth, retreated for 2 more ing subsequent orthodontic care and the costs
years, and achieved an excellent final outcome. for the extractions. She was upset at having to be
She then sued the first orthodontist. in braces for a total of 6 years. She initially
The plaintiffs expert alleged that the doctor demanded $125,000 to compensate her for the
failed to conduct a tooth mass to arch length additional treatment, her embarrassment, pain,
analysis and that there was insufficient space and inconvenience. After investigation, defense
without resorting to extractions. He also opined counsel seriously recommended settlement of
that 3 years of treatment should have been suf- this case before trial. Although both parties were
ficient time to achieve success of this patient's good witnesses, the post-it note addendum to
problem. The expert witness who reviewed the the records had a negative effect. In addition to
records on behalf of the doctor opined he was that, once the expert witness testified that ex-
satisfied with the original treatment plan be- tractions were indicated, the fact that the teeth
cause he could see that the insured intended to were badly flared at the end of 3 years became
extract the third molars at the end of treatment more difficult to defend. The dismissal of the
to add to the space. He agreed with the propo- patient from the practice was also a problem.
sition that root resorption and periodontal dis- The case was settled for $37,000.
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 Risk Management Recommendations
severe flaring, he admitted that extractions did
appear to be indicated. Nevertheless, he had no From a risk management perspective, records
complaint with the length of treatment and be- and their quality are once again seen as critical
lieved that had the patient remained in the in- to the prevention of and subsequent defense of
sured's practice, she would have achieved a good a case. As we have said before, incomplete or
final result. inadequate records provide little support against
Why then, did this patient sue? Was the initial a plaintiffs allegations. Records that have an
diagnosis unsatisfactory? The allegation of the appearance of being altered are deadly. They
tooth mass/arch length discrepancy and the create such doubt in the minds of jurors that
lack of such an evaluation was challenging to they can erase any other good supportive factors
overcome. The records were devoid of such a in a case. Never engage in records alteration.
computation. The doctor eventually produced As we saw in the other two cases, communi-
one well after the records had been submitted to cation deficiencies appeared to influence this
the plaintiff. It was written on a small post-it case. Was the treatment plan carefully explained
note, and was not impressive evidence. It had so that the patient understood why the doctor
the appearance of having been created after the chose to treat without extraction? Was a good
fact. That negative impression no doubt pro- informed consent discussion held, and did the
vided an incentive to the plaintiff to sue. patient sign a form to that effect? When the
In response to that allegation, the insured treatment time extended past the original esti-
testified that he had always been aware of the mate, was an explanation provided? What aspect
tooth mass/arch length discrepancy. He charac- of communication broke down so unequivocally
terized his treatment plan as conservative and that the doctor felt compelled to dismiss the
appropriate, given the diagnosis. Even though patient? It seems easy to understand in this case
the insured made an adequate witness, the fact why the patient felt dissatisfied and abandoned
that he might have created evidence after the by the original orthodontist.
<<
Article
>> Home | TOC |
Index
Tying It All Together before, during, and after treatment so you know
what is happening physiologically. Record
We have now analyzed three very diverse cases.
One occurred on the West coast, another on the clearly and concisely all critical aspects of the
Eastern seaboard, and the third happened in the treatment, pertinent conversations, complaints,
South. One treatment period was extremely pro- and occurrences. Document all instructions and
tracted; the other two were normal or close to advice that you give and keep copies in the
normal. The patients were different—one was a patient record. Make it a habit to have an in-
middle-aged adult female, another a male child, formed consent consultation with every patient,
the third a young adult female. All of these tailored to their specific needs. Then, place a
patients sued their orthodontist. signed form in the records as proof of the con-
There are many factors that go into the actual sultation. Renew the informed consent if the
filing of litigation about which we cannot be treatment changes.
aware. It is commonly believed that a conversa- If there is one thing that I have learned over
tion with a lawyer at a cocktail party or one living the years of handling dental malpractice claims,
across the backyard fence may encourage some it is that no dentist enjoys being sued. Some are
litigious behavior. Television and other media more willing than others to go through the trial
forms also frequently emphasize success in mal- process, but none of them enjoy it. It is time
practice litigation and that motivates real-life lit- consuming at the least and antagonistic and of-
igation. In other words, some lawsuits will be fensive at its worst. It requires hours of answer-
filed, no matter what. Those of us involved in the ing written and oral questions about every detail
handling of professional liability claims and law- of the office procedures, the diagnosis and treat-
suits contend that knowing the bases for some ment, the records, and the communication or
problem patterns of behavior and interrupting lack thereof. Doctors who are sued spend a sig-
them can prevent some lawsuits from occurring. nificant amount of time working with an attor-
This ultimately provides some protection to ney to formulate and present a defense. They
orthodontists. spend hours in preparation, pretrial settlement
Let's review the most obvious forms of pro- forums, and listening to constant accusations of
tection. One is to improve communication skills. wrongdoing. The trial itself is more of the same,
Listen to the patient or their parents. Speak to only more intense, with one witness after an-
them, carefully explaining what is expected to other alleging practice below the standard of
happen, and what is actually happening. Explain care.
negative behavior or other problems that are If the emotional toll of litigation is not bad
impeding the success of the treatment. Contin-
enough, the financial cost is worse. Not only is
ually communicate with referral doctors to be
defense expensive in terms of legal charges and
sure that the overall dental health of the patient
expert witness support, but in addition, all of the
is the primary focus, not just the orthodontic
treatment. time a practitioner spends defending a lawsuit is
Focus on the treatment. If it is not progress- time away from a money-making practice. Most
ing satisfactorily, take note. Is it taking too long? doctors I know object to that.
Is the occlusion not being corrected? Make As I speak with dentists, I find that the moti-
changes if necessary. Focus on the patient. What vation to understand and make necessary
issues about them are occurring that might af- changes to preclude or minimize exposure to
fect the success of the treatment? Empathize lawsuits is generally high these days. Our society
with the patient. Realize that each person has a has become increasingly litigious in some areas
different perspective to pain, motivation for of the country, with some cities being worse than
treatment, and esthetics. Try to place yourself in others. Orthodontists must therefore be vigilant
their position and understand what they are ex- and recognize their own role in the process.
periencing. Let them know that you understand They must make appropriate changes in their
and are concerned. practice and in their relationships with patients
Take good records. Take enough radiographs to protect themselves from a malpractice lawsuit.
<<
Article
>> Home | TOC |
Index
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 encourage lawsuits to be filed in a timely man-
then your receptionist calls you out to the ner.
waiting room. The person standing there calls It is important to maintain treatment records
out your name and thrusts papers into your concurrent with when the treatment was ren-
hand stating, "Doctor, you have been served." At dered. Keep all entries intact and unchanged. If
this point, if you have not had any experience you need to correct an entry, the date of the
with the legal system, numbness and then panic correction should be noted and the entry made
sets in. First, take a deep breath then, call your at that time. It is not necessary to write that you
professional liability insurance carrier. Do not spoke to your insurer in the record. Remember,
call the patient or his/her attorney in an attempt a dental record is a recognized business record
to explain that there must be a mistake. Such a of the treatment provided a patient and all facts
call will do no good and may actually be detri- pertinent to that treatment. If there were any
mental to your interests. Rarely, if ever, is a communications with the patient, regarding dis-
defendant released from a lawsuit after talking satisfaction with treatment or any other com-
to the plaintiffs attorney. plaints, these should be noted at the time they
When you call your malpractice carrier, have were made, as well as any problems encountered
in front of you your policy number and the during the course of the treatment. Do not make
patient's records. You will be asked for the dates any changes or explanations on the record in
of treatment and a description of the treatment hopes of clarifying your notations at any time
you provided to the plaintiff. The dates are im- after the original entry was written. Any changes
portant to determine if the lawsuit has been in a record may be a criminal act, a felony in
brought in a timely manner; each state has a some states, and are surely actionable against
time limit pertaining to when the plaintiff can one who does so. If not actually a crime, a state's
institute a lawsuit. This is called the statute of Dental Practice Act usually prohibits records al-
limitations. These time periods been enacted to terations and doing so may result in censure or
even suspension of your license.
The course of a lawsuit is generally a long
From Medical Liability Mutual Insurance Company, East one. There are several stages to the litigation
Meadow, NY. process. Once you file a report with your carrier,
Address correspondence to Toni Reale, Esq., 23 Fern Street, you set in motion the mechanism for protecting
Floral Park, New York, NY 11001.
Copyright 2002, Elsevier Science (USA). All rights reserved.
your interests. Initially, you will be required to
1073-8746/02/0804-0006$35.00/0 send your carrier, along with the legal papers
doi: 10.1053/sodo. 2002.12 7868 you received, copies of your complete records.
This includes everything contained in the pa- followed. They may vary from state to state, but
tient's chart, including the dental and medical one thing is constant: failure of a plaintiff to
history forms, treatment plan(s), all consent properly serve a defendant can result in the
forms, prescriptions, any telephone messages, dismissal ofthat lawsuit. It is important to keep a
referral forms, reports to or from other dentists written record of how and when you were served
or physicians, all radiographs, models, corre- to help your attorney determine if the service of
spondences to or from the patient or anyone on the legal papers was proper.
their behalf, insurance forms, insurance approv- In addition, there are specific times under the
als for treatment, explanations of benefits, and law for defendants, through their attorneys, to
even the outside of the record jacket if it con- answer the allegations specified in the com-
tains notations. plaint. An answer will address each allegation
Once the insurer receives these documents, and will also set forth any affirmative defenses.
an attorney will be assigned to you. This attor- The burden of proof for these defenses rests
ney, although retained by the carrier, represents with the defendant. One defense usually raised
your interests. You are the client, and the attor- is the comparative or contributory negligence of
ney must not act in any way that is detrimental to the plaintiff. This means that the injuries alleged
you. The attorney will meet with you to discuss in were added to or caused by the plaintiff. Failure
detail the treatment you provided to the plain- to timely serve an answer may allow a plaintiff to
tiff. You can and should be candid and open at obtain a default judgment against you. In es-
this time. Any information that you give to your sence, this means that you will not have an op-
attorney is protected by the attorney-client priv- portunity to defend yourself against the allega-
ilege and cannot be divulged. Therefore, com- tions of malpractice made by the plaintiff.
plete honesty is essential in preparing your de- The course of a lawsuit is often long and
fense. protracted. Each stage may take several months
There have been an increasing number of or even years. Generally, the courts oversee this
suits against orthodontists, particularly with the process. The defense attorney will request autho-
increase in orthodontic treatment of adults. rizations to obtain the records of prior, concur-
These claims often involve failure to obtain the rent, and subsequent dentists and physicians
desired results, causation or exacerbation of who may have treated the plaintiff for the al-
temporomandibular joint problems, misapplica- leged injuries. Once a patient places his/her
tion of orthodontic appliances causing bite condition in issue, the defendants are entitled to
problems, failure to diagnose and/or treat decay obtain all records relating this. Your attorney
at the site of appliance placement, causing root may discuss with you some of the salient facts
resorption, and causing periodontal disease. ascertained from these treatment records.
The one claim that is in almost every lawsuit is Once the records are obtained, an examina-
lack of informed consent. This makes obtaining tion before trial, also known as a deposition, of
informed consent a very important defense tac- the plaintiff will be held. The defense attorney
tic. It is not sufficient to orally give the patient a will question the plaintiff under oath, with a
treatment plan. It should be done in writing and court reporter present, as to both their medical
include options as well as any potential compli- and dental history, the treatment that was pro-
cations for each viable treatment option. vided, what the plaintiff was told about the treat-
A lawsuit is usually initiated by service of a ment in issue, the benefits sustained, compli-
summons and a complaint. Some jurisdictions ance with the treatment plan, and any damages
may require that these documents be filed be- or injuries claimed as a result of the treatment.
fore service. The summons sets forth the names Because most plaintiffs swear that they did every-
of the parties and the venue, whereas the com- thing you advised them to do, it will be to your
plaint elaborates, usually in general terms, the benefit to have notations in your records indict-
nature of the allegations against you, the dam- ing each time that this did not occur.
ages allegedly suffered by the plaintiff, and may Often, your only defense in a lawsuit that
also contain a demand for monetary damages. occurs long after the treatment was provided is
To obtain jurisdiction over an individual, your records. Accurate, concise, complete
there are certain prescribed rules that must be records can sometimes be your only defense.
<<
Article
>> Home | TOC |
Index
Plaintiffs memories become sharper and clearer vided, it will be exchanged with the plaintiffs
the closer they get to trial. If your records, which attorney. You are also entitled to have the re-
were kept concurrent with the treatment, differ, ports from the plaintiffs examining doctors as
then the defense attorney can cast serious doubt well. In today's court system, there is an open
on the plaintiffs version of the facts. exchange of information. Long gone are the
After the defense attorney questions the days of the "Perry Mason" surprise witness. If the
plaintiff, the plaintiffs counsel then questions opposition does not furnish you with advance
the defendant(s). Before this, the defense attor- copies of reports or records, they may not be
ney will meet with you to go over your records admitted into evidence at the time of trial.
and explain the deposition process. Again, your After all parties have been deposed, and an
records are invaluable in this instance. They will independent physician or dentist has examined
provide backup for your recall. The plaintiffs the plaintiff, the discovery phase of the lawsuit is
attorney will most likely question you directly completed. At some point, after all treatment
from your records. You will be asked about each records have been obtained, your attorney will
notation. Plaintiffs attorneys pay particular at- have the file reviewed by an expert, or several
tention to what a patient was told regarding experts, if the claims or damages warrant it. One
treatment. If you do not recall the answer to any expert will be an orthodontist. The attorney's
question, just say so; do not make up an answer. purpose will be to determine if there were any
The opposing attorney may also ask you ques- departures from good and accepted practice re-
tions about your education and training. By do- garding your treatment. Additional experts may
ing this, he/she may be laying the foundation be retained to discuss the issue of damages.
for claiming a lack of education or experience in To prevail in a malpractice action, the plain-
performing a particular procedure. The most tiff must prove that the defendant had a duty of
important advice you need to remember while
care to the plaintiff, that there was a breach of
preparing for and attending your deposition is
this duty, and that this breach was the proximate
that your lawyer is there with you to protect your
cause of the plaintiffs damages. The experts
interests. Follow his/her instructions, and do
from both sides will consider all of the treatment
not volunteer any information. Keep your an-
swers short, and just answer the question asked. and damages documented to determine if the
Once your testimony is taken, you will receive plaintiff presented valid claims or not. To prove
a copy of the transcript to review for errors or the allegations of malpractice, the plaintiff must
inaccuracies. You may make corrections, with present expert testimony setting forth the stan-
the advice of your attorney on the forms pro- dard of care, the defendant's treatment did not
vided and are encouraged to do so. The plain- meet that standard, and that the breach of duty
tiffs attorney will use this transcript against you caused the damages alleged.
at trial if your trial testimony differs from the Standards of care may also be promulgated to
deposition testimony. Any deviations can under- some degree by professional societies or devel-
mine your case and your credibility. Your depo- oped through regulation. Some of these stan-
sition transcript is one of the best opportunities dards are mandatory; others, are merely guide-
to assist in your own defense. The attorney rep- lines. Individual practitioners should keep
resenting you knows the pitfalls of the examina- abreast of any changes in their own specialties.
tion before trial process and will be on the alert After the discovery phase, the case is placed
to protect you from them. on the trial calendar. The plaintiffs attorney
At some point during the course of litigation, may request a trial by jury or a bench trial in
a defendant is entitled to have the plaintiff ex- front of a judge without a jury. If he/she does
amined by a doctor of his/her (his/her attor- not request a jury trial, your attorney may do so.
ney's) choosing to determine the condition of Depending on the court calendar in a given
the plaintiffs mouth. This usually occurs after jurisdiction, this pretrial period can be several
the plaintiffs deposition. However, if the plain- weeks to many years. If the defense has been
tiff has had all of the dental work done by the able to find an expert to defend your treatment,
defendant replaced, the decision may be made the case will now proceed to trial. If not, you may
to waive this examination. Once a report is pro- be approached by your carrier to discuss settle-
<<
Article
>> Home | TOC |
Index
ment and to sign consent to settle, if one is plaintiffs' witnesses have testified, the plaintiff
required by your policy. rests his case.
Your attorney will be preparing for trial by The defense now puts on its case. You may
meeting with you, your experts, and getting any not be called to take the stand for a second time
exhibits made if necessary. In addition to your if you have already testified during the plaintiffs
records, you will also have your deposition tran- case. Your experts will now be called to refute
script for review. Don't be dismayed if there are the plaintiffs assertions. When all witnesses have
long periods of inactivity; this is a common oc- testified, the defense also rests. Both lawyers are
currence. then given the chance to persuade the jury for
The plaintiff has the burden to prove all his/ the final time when presenting their summation.
her claims. This must be done by a preponder- Here, the defense goes first; the final words the
ance of the evidence. This means that he/she jury will hear will be from the plaintiffs attorney.
must persuade the judge or jury that the facts After the summations, the judge will instruct
are more likely than not to support his/her the jury as to their role as jurors, he/she will
claim (s). The defendant has only to prove any explain the law, and their role to determine the
affirmative defenses refuting the plaintiffs true facts as the jury believes they were pre-
claims. The jury is the sole judge of the credibil- sented. He/she will then read the jury charge,
ity of the witnesses, the plaintiff, the defendant, which sets forth the alleged departures for ac-
and all of the experts. If contributory or com- cepted practice and the claimed damages. Based
parative negligence of the plaintiff is proven, the
on the evidence, they must determine if any
plaintiff may receive nothing or have any award
departures caused the injuries.
reduced by the degree of his/her negligence.
As a defendant, you may be expected to be in
Immediately before trial, you may be served
with a subpoena to testify at trial. This is the court each day or as specified by your attorney.
plaintiffs method of making sure that you testify While in the courtroom, you will be under the
during presentation of the case. A trial proceeds constant scrutiny of the jury. Your demeanor
in a proscribed manner. Both attorneys have the and cooperation as well as the testimony of the
opportunity to question and select jurors from a experts will be an integral factor in any success
panel to hear this case. After this, there are in the courtroom.
conferences with the assigned judge, after which In summary, the following are examples of
the attorneys present their opening argument to what to do and what not to do. You should (1)
the jury. The plaintiff puts on his/her case first. report service to your insurer, (2) provide com-
He/she will call the plaintiff to testify and any plete records to the carrier, (3) be accessible to
other witnesses, possibly family members. Then meet with your attorney and/or investigator for
he/she will call the defendant to testify, if the carrier, (4) cooperate with your attorney
he/she has served the defendant with a sub- during all phases of the action, and (5) feel free
poena (if not, the defendant will testify when the to contact your defense attorney with any ques-
defense presents its case). Finally, the plaintiff tions or concerns.
will have the testimony of his/her experts and On the other hand, you should not (1) call
examining or treating dentists or physicians. the patient or their attorney, (2) alter your
The defense attorney will have the opportunity records, (3) discuss the case with any other de-
to cross-examine all witnesses in an effort to fendants, or (4) do not offer gratuitous state-
disprove or discredit their testimony. Once all of ments or explanations when testifying.
<<
Article
>> Home | TOC |
Index
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 Best judgment is defined as that used by a rea-
O percentage of my dental malpractice case-
load because the injuries complained of are not
sonably prudent orthodontist practicing within
the community. This instruction is archaic, con-
usually severe, and orthodontic cases do not jus- voluted, confusing, and places a tremendous
tify the cost of bringing a case to trial. From burden on the plaintiff.
purely a business perspective, better medical There is further good news for orthodontists;
malpractice attorneys do not accept cases that national statistics reveal that the defendant pre-
have the potential to generate less then a certain vails in the overwhelming majority of cases that
value. Most orthodontic actions fall far below go to trial. In orthodontics, one reason for this
this threshold. It is also difficult to find orth- statistic is that the average personal injury attor-
odontic expert witnesses because the orthodon- ney representing a plaintiff is not technically
tic community still operates within an "old boy's conversant with orthodontic art and science. A
club" mindset. Orthodontists are less willing to second reason is that most orthodontic treat-
testify against their colleagues than are doctors ment occurs over several years, and the mechan-
in other dental and medical specialties. This ics are difficult for a lay jury to follow. The third
observation is truer outside of major metropoli- reason is that the defendant and the expert
tan cities but does not apply if the defendant is witnesses can and often do invent the medicine.
a general dentist or a pediatric dentist rendering Finally, as previously stated, the legal application
orthodontic services. of the law presents an additional hurdle.
Another additional obstacle to attorneys ac- Because I have been asked to tell this story
cepting orthodontic cases is, depending on the from the perspective of a plaintiffs attorney, I
jurisdiction, the instructions that the judge gives will begin with my introduction to orthodontic
to the jury to apply to the facts. In New York, for malpractice and end with a current recital of the
instance, juries are instructed that an orthodon- fact patterns from orthodontic cases in which I
tist is not a guarantor of a good result nor is have had personal involvement during the years
he/she liable for mere errors in judgment as 2000 and 2001 because these cases represent
long as he/she exercised his/her best judgment. what is currently happening.
Sometime in the late 1960s, approximately 12
years before my entering law school, I received a
Address correspondence to Joel Kotick, DMD, ]D, 501 East 79th telephone call from the head of the hospital in
Street, New York, NY 10021.
Copyright 2002, Elsevier Science (USA). All rights reserved.
which I was on the orthodontic staff. He asked
1073-8746/02/0804-0007$35.00/0 permission to give my name to an attorney who
doi:10.1053/sodo.2002.127869 needed someone to review an orthodontic mal-
practice case pending in a neighboring state Litigation Influence the Quality of Dental Care. I
because no one else would do it. My credentials reached the conclusion that dental malpractice
at the time were not overly impressive, and I did litigation does not influence the quality of care.
not know then what I know today: juries do not I do not agree with that conclusion today be-
differentiate between fair credentials and signif- cause I truly believe that the threat of malprac-
icant credentials. The plaintiffs attorney noted tice has made all of us better practitioners. The
that I was the codirector of the temporomandib- best prevention for malpractice suits is conscien-
ular joint clinic at a local dental school, I was an tious and caring practice.
attending on a local hospital staff where I was While in law school, I was referred a 19-year-
the chief of the section of orthodontics and old patient who was occluding only on his pos-
temporomandibular joint, I had published two terior teeth. Four first premolars had been ex-
scientific papers, I was past president of my den- tracted, and he presented with a prognathic
tal school alumni association, and at that time I mandible and an anterior open bite. It was a
was the president of the local dental society. To functional and esthetic nightmare. I sent him for
a lay jury, it was enough. a surgical consult, and, naturally, his parents
On examination of the patient and the were upset. They then asked me to examine his
records, I learned that this 30-year-old woman 16-year-old brother who was also prognathic, was
had four premolars extracted and was being also developing an open bite, and was in the
treated with the Begg appliance. While on vaca- middle of four bicuspid extraction therapy. I
tion with her husband one June, she thought she explained that the skeletal problem could only
felt her teeth with her tongue protruding into be corrected with a combination of surgery and
her palate. She asked her husband to look, and orthodontics. The parents then asked me to ex-
indeed her roots were palpable. Three months amine their 13-year-old daughter. She was ready
before, the orthodontist had taken a midtreat- to begin treatment with the same orthodontist
ment panoramic radiograph that revealed se- for essentially the same problem. He again had
vere root resorption. Instead of stopping treat- prescribed the extraction of four first premolars,
ment, the doctor initiated the third stage of and I explained that extraction would result in
Begg treatment and torqued the roots through the same poor ending. They accepted my treat-
the palate. In my opinion, there was no malprac- ment plan, and treatment was successfully com-
tice up until March. However, once the resorp- pleted in 18 months without any extractions.
tion was diagnosed, there was no excuse to begin Seeing the simple progress of their daughter,
the third stage of Begg. The panoramic radio- her parents asked me to refer them to a mal-
graph was the smoking gun. I agreed to testify practice attorney. I referred them to one of my
for the plaintiff. professors, who initially agreed to represent
I was young and very nervous; my credentials them and then changed his mind. The parents
were weak, and I had no clinical experience in asked me if I would represent them. By that
Begg mechnaotherapy. I had taken a 2-week time, I had recently graduated law school, taken
Begg course and had attended some Begg lec- the bar, and was awaiting the results. I explained
tures. My major anxiety was appearing stupid. I had no experience and the statute of limita-
Contrary to my fears, I discovered the orthodon- tions would toll if I did not pass the bar. They
tist's attorney had little understanding of the said they would gamble on me passing the bar
medicine. He could not impeach me, and there and my inexperience. I passed the bar and had
was no good defense for this case. The case my first two clients.
settled for $45,000 within minutes after I left the The orthodontist was board certified, had
witness stand. Even back then, $45,000 seemed complete records, and was able to impress the
too little for the loss of four anterior teeth, along jury with his knowledge. The expert for the de-
with the extraction of four first premolars to a fense was low keyed and simply told the jury that
30-year-old woman. I drove home proud that I growth was unpredictable, and the removal of
had the courage to stand up for justice. four premolars was the standard for the commu-
I had no involvement with malpractice until nity. I attempted to explain that the role of a
some 12 years later when I was in law school. I good diagnostician is to be able to predict the
authored a paper entitled Does Dental Malpractice potential for adverse growth. The cephalometric
<<
Article
>> Home | TOC |
Index
arguments were lost on the jury. I was not per- importance of convincing a jury that the very
mitted to tell the jury this orthodontist was about negligence complained of is the proximate
to make the same mistake for the third time and cause of the injury. At that time, the instructions
that the daughter had been treated successfully regarding proximate cause were archaic and
without extractions. The judge gave the usual confusing. The judge instructed the jury that an
New York State instructions that an orthodontist act is the proximate cause of an injury if a rea-
is not a guarantor of a good result, and the jury sonable person by a fair preponderance of the
came back quickly with a finding for the defen- evidence would believe that the act was a sub-
dant. I was personally disappointed. I believed stantial factor in bringing about the injury. No-
his misdiagnosis, not once but three times, was body speaks like that, and I am convinced that
malpractice. The orthodontist left court with an juries have no idea what the judge is talking
air of self-satisfaction, and the two young men about. Today, although still confusing, the prox-
each left court with a poor functional occlusion imate cause instruction is less onerous, and the
and an unsightly profile. jury is instructed it must be substantial factor in
The lessons that I learned was that you do not bringing about the injury. The final lesson for
have to be a good orthodontist to win; it is me was that a really bad orthodontist can get
important to have good credentials and make a away with providing a horrible result.
good personal appearance, and one must have Juxtapose these cases with the infamous Mich-
good records. I realized I was naive in expecting igan case of the mid-1980s. As I recall, the orth-
the jury to comprehend the nuances of growth odontist treated a Class II Division I malocclu-
prediction. I learned some good lessons. sion by extracting the upper first premolars. The
Compare this case with another case I lost. It orthodontist had excellent credentials, the
was also a four-premolar extraction case. The records were complete, and the treatment was
defendant orthodontist had weak credentials, reported to be excellent standard edgewise ex-
his records were poor, and his mechanics bad. I traction therapy. The patient developed tem-
argued the poor occlusion resulted in a tem- poromandibular joint problems requiring sur-
poromandibular joint problem. The plaintiff gery. The expert for the defendant was
had not yet undergone subsequent orthodontic nationally recognized; the expert for the plain-
therapy or paliative treatment for her joint prob- tiff was a general dentist who lectures around
lem. She had been to an oral surgeon who tes- the country on functional orthodontics and
tified that joint surgery was necessary. The de- against extraction. The negligence alleged was
fense attorney emphasized that because she did that the extraction of the upper premolars re-
not have the surgery nor any treatment to date, sulted in the patient's temporomandibular prob-
how bad could the problem be? The defendant lem. The verdict was over $1,000,000 for the
produced a very effective witness who used a plaintiff.
skull to explain the temporomandibular joint, The profession was up in arms. It was sug-
"lying" about the medicine. He convinced the gested in the dental literature that the case
jury that my client did not have a temporoman- would be reversed on appeal. It was not. The
dibular problem. The jury found that the defen- case prompted the first annual conference on
dant was negligent in his orthodontic treatment orthodontics and the temporomandibular joint.
but did not find that his negligence was the Many of our leaders drafted a position paper on
proximate cause of the temporomandibular orthodontics and the temporomandibular joint
joint injury. that was subsequently published. The position
From this case, I learned about venues. The paper was self-serving and concluded orthodon-
case was tried in a very conservative venue with a tic treatment does not cause temporomandibu-
bias toward the defendant. Some venues are lar joint problems.
much more favorable to the plaintiff than oth- Like many of you, I attended the conference.
ers. In other words, the same fact pattern in one While in Michigan, I visited with the plaintiffs
jurisdiction will prevail in one court and lose in attorney. He has a sign over his private office
another. The second lesson was that there are door "TMJ Specialist." He was generous with his
experts who will say anything and consider lying time and enjoyed discussing this case and other
to be part of their role. The third lesson was the temporomandiular joint cases he had won. He
<<
Article
>> Home | TOC |
Index
speaks to a jury like a used car salesman in a the orthodontist's office address when her child
manner I do not believe would prevail in other received treatment from him in 1985; she
jurisdictions. He attributed his win to the pom- agreed.
posity of the defendant. The orthodontist admit- In those days, I was still practicing orthodon-
ted he observed the patient's temperomandibu- tics full-time at the same time I was developing
lar joint symptoms during treatment but did not my legal practice. I was constantly running back
remove the active wires and stop treatment. I do and forth between the courthouse and my office.
not know the facts well enough to determine if it My client, incensed at the outright lies, took over
was a just result. The lesson I took away is that part of the lawyering while I treated patients.
juries are willing to ignore nationally recognized The plaintiff went to the doorman of the defen-
experts and sometimes as John Adams is quoted dant's building to secure his testimony as to what
as saying "those stubborn facts." year the defendant occupied the premises. She
A case deserving of significant discussion con- also found old telephone books listing his old
cerns a 35-year-old plaintiff who sought orth- office location for 1985. The judge would not
odontic treatment for correction of rotated in- allow the telephone listing to be admitted with-
cisors. The extraction of two upper premolars out securing a telephone company employee's
exacerbated the patient's undiagnosed and un- testimony as to their authenticity. I figured the
treated periodontal condition. To defend doorman's testimony was sufficient so I did not
against the claim of not recognizing or treating subpoena the employee. The trial was no longer
the patient's periodontal disease before initiat- about the medicine. I was sure that the jury
ing treatment, the defendant orthodontist pro- believed that the defendant was lying; we
duced a three page letter he said he wrote be- couldn't lose.
fore beginning treatment in September of 1985. After resting our case, the defendant pre-
The letter was allegedly sent to the patient's sented his case. The defense attorney intro-
general dentist and the patient advising her of duced the general dentist who testified that she
her periodontal disease and the need for treat- received the letter in 1985 and that she recalled
ment. The plaintiff was adamant about never the orthodontist having moved in 1984. The
having received this letter. The first two pages defendant then called one of his experts who
were dated 1985 but the third was dated 1986. coincidentally rented space to the defendant be-
The defendant had great credentials, board cer- fore his move. My client was now very con-
tified, had written a textbook, and so on. At trial, cerned. The defendant then testified about the
I asked him why he dated his letters 1986 during move and stated it occurred in 1984. When
1985. He fumbled, explaining it was a secretarial asked how he could be so sure about the date,
error. The jury understood that not many peo- the defendant replied that his wife had died in
ple date their letters 1986 in September of 1985. 1984, and his son had to be institutionalized that
I followed with why his letterhead bore a street same year. All of it was true, except the part
address that he was not practicing at in 1985. I about the move. The testimony was both power-
believed I had him on the ropes. He responded ful and devastating.
that he had moved in 1984. The jury was con- When questioned about whether or not he
fused. I challenged him to bring in telephone was able to provide documentation as to the
bills, a lease, or tax support documentation per- move, the defendant produced a subrider to his
taining to the move. lease indicating the move occurred in 1984. My
Coincidentally, only days before this trial, the client and I were the only people in the court-
mother of a child treated by this same orthodon- room who knew he was lying. I made a long
tist was referred to me regarding possible orth- record outside the presence of the jury telling
odontic malpractice. I determined that although the judge that the subrider was a fraud. I said if
there was malpractice, the damages were ex- it goes to the jury I will lose the case. I asked the
ceedingly small, and I told the mother I would judge for a representation from the defendant's
have to think about taking the case. Coming attorney that he did not know it was a fraud. The
back to the case at hand, after the orthodontist attorney said nothing. The judge said he would
lied about when he moved, I called this child's not ask an attorney for such a representation
mother and asked if she would be a witness as to and counseled that if it was a fraud, the attorney
<<
Article
>> Home | TOC |
Index
was well aware of the consequences. I had made ney, "Dear Mr. Attorney, enclosed are my
my record for an appeal. records, please destroy the original records I
My client was of foreign extraction and did sent." I thought his attorney would fall off her
not make a sympathetic witness. The defense chair. By accident, his attorneys sent me a copy
attorney told the jury she was bright, shrewd, of that letter with his treatment chart. The in-
cunning, and she made up this whole story surance company now settled a minor case for
about not receiving the letter and the different well over six figures. The defendant orthodontist
address to get tax-free money. The verdict was almost went to jail. His attorney was investigated.
for the defendant. One juror, an editor from a The offshoot of this debacle was a well-attended,
national magazine was the one dissenting juror. full-day seminar cosponsored by a component
In New York, you only need five out of the six orthodontic society and the insurance company.
jurors to win. He told me he was unsure about The format was a mock trial about an adult
the letter and address but instinctively believed orthodontic case with underlying periodontal
the defendant was lying. He said the other jurors disease. The most common orthodontic cases in
believed the lease and the testimony of defen- my legal practice are adults who are treated in
dant's expert. the presence of periodontal disease.
The following day, I personally served a sub- The lessons learned are obvious. First, I could
poena for the original lease on the defendant's no longer practice law part-time; it was all or
expert and served a subpoena for the original nothing. Second, a good attorney cannot be lazy;
leases from the dentist relating to his various you dot every "i" and cross every "t". As for the
practice locations; something I should have orthodontic lessons, we have all been taught
done at trial instead of returning to my office to never to change records. Here the orthodontist
treat my orthodontic patients. Neither dentist wrote a letter after the fact, he bungled the
complied. I ran to court and had the judge sign dates, used false letterhead, and then enrolled
an ordered subpoena and again, I personally other colleagues to lie for him. All could have
served them. I was under a time gun because I lost their licenses. I see altered records all the
had to make my posttrial motion within 2 weeks. time. When I convince a jury about the alter-
I learned later that the defendant's attorney, ation, it is devastating to the defendant. You
aware of the time restraints, instructed both den- have to be dumb and then dumber to alter
tists not to comply. I then went to court and records.
procured an order to show cause, again, served The occurrences stated previously are what
personally. Both dentists had to appear in court brought me to become a plaintiffs attorney.
with the leases or go to jail. I was convinced that Make no mistake about it, I am still an orthodon-
the judge believed I was correct. Neither dentist tist at heart, but I believe that occasionally we do
appeared. The judge was livid. Instead, they en- commit malpractice and on those occasions
gaged an attorney who came to court with the when a patient is injured, they deserve represen-
original leases. The judge read the leases and tation. I made a lot of mistakes at the beginning
gave me a copy that I attached as an exhibit with of my legal career, the same way we all made
my motion papers. The defendant had copied many mistakes at the beginning of our orth-
the subrider to the lease, changed the dates, and odontic careers. I am now much better at what I
had the expert dentist sign the changed sub- do, and, in the remainder of this article, I will
rider. discuss the fact patterns in orthodontic cases I
The judge reversed the verdict. The insur- have personally been involved with as plaintiffs
ance company settled. The defendant's expert attorney during the last 2 years. During this
frightened he might lose his license called me time, I have tried six orthodontic cases, five go-
apologizing for his stupidity in lying. I called the ing to verdict, and one ending in a nonsuit,
nice mother that had testified and said I would which means that the case has to be started all
take her daughter's case even though the dam- over again. I prevailed in all of the five cases that
ages were small. Truth being stranger than fic- went to verdict. Additionally, I have settled other
tion, I immediately initiated the lawsuit. At the orthodontic cases during this time frame before
deposition with this same defendant, I asked they went to trial. There is no shortage of orth-
him what he meant when he wrote to his attor- odontic malpractice.
<<
Article
>> Home | TOC |
Index
Case One verdict was for the plaintiff; the defendant ap-
pealed and then settled for $377,000 before we
The patient was a brachyfacial woman in her late
received a decision.
20s who presented with a Class II Division I
deep-bite malocclusion. Treatment involved the
Lesson
extraction of upper first premolars. The negli-
gence alleged was that the case required orthog- Skeletal problems in adults, particularly open
nathic surgery and that there was a failure to bites, require obtaining a surgical consultation
monitor for root resoprtion. The injuries as well as the necessity to take midtreatment
claimed were that the extraction spaces could radiographs. As an aside, this case was tried in 2
not be closed, there was no reduction in either days. We began trial at 9:00 AM on a Monday
the overjet or the overbite, and significant root morning. The judge told us that if your next
resorption had occurred. The defense claimed witness is not ready when your last witness com-
that root resorption happens all the time, the pletes testimony your case is dismissed. I asked
patient would not lose any teeth, and that surgi- what the court hours were as this was another
cal treatment was too extreme. The case was out-of-state case and unlike New York, each
tried in Virginia, and we received a verdict of judge is free to set his own schedule. The judge
$300,000. The verdict was appealed and af- said, "In my courtroom, we work from 8:00 AM
firmed. To my surprise, it was reported to be the and go through midnight if the jury is willing."
largest dental verdict to date in that state, which We worked until 9:30 PM Monday and had a
has a cap of $1,000,000 for all medical malprac- verdict by 10:00 PM on Tuesday. I tried the same
tice actions. type of case with the same number of witnesses
recently in New York. The case lasted for 4
weeks.
Lessons
You must consult with surgeons in skeletal cases
Case 3
and midtreatment radiographs are critical to
monitor for root resorption. In this case, the The patient was a 35-year-old woman with a Class
attorney for the defendant showed the orth- 1 bimaxillary protrusion who underwent four
odontic expert a panoramic radiograph and premolar extraction therapy. The venue (the
asked him if he saw any root resorption. The county in which the case was heard) was a "plain-
expert told the jury no. On the panoramic ra- tiff friendly" (the juries often side with the plain-
diograph, root resorption could not be seen. It tiff) locale in New York. The negligence claimed
was a poor-quality film with the spinal column was sloppy mechanics treated in the presence of
superimposed over the anterior teeth. This "de- periodontal disease, coupled with a prolonged
ception" was exposed. treatment time. The defense claimed that the
poor patient cooperation for months at a time
was not contributory, and the defendant as-
Case Two sumed that the general dentist was treating the
patient's periodontal condition. The injuries
The patient was a 35-year-old woman who pre-
claimed significant bone loss and future loss of
sented with an anterior open bite and spacing.
teeth. The plaintiff did not undergo any subse-
The treatment was the extraction of one upper
quent treatment. We received a verdict of
premolar because the patient was missing one
$225,000; the case was not appealed. Frankly, I
on the other side. The negligence alleged was
was disappointed with the award particularly
that the open bite required a surgical correc-
with a favorable venue.
tion, extractions were unnecessary given the an-
terior spacing, the mechanics were poor, and
Lesson
there was a failure to monitor the dentition with
midtreatment radiographs. The defense was that There is no excuse for treating cases in the
premolars are routinely extracted. The injuries presence of periodontal disease and certainly no
claimed were significant root resorption and defense for extracting teeth without first resolv-
that the open bite was made worse. The original ing the underlying periodontal disease.
<<
Article
>> Home | TOC |
Index
they showed evidence of periodontal destruc- prompting an insurance company to engage me.
tion. After 2 years, severe root resorption had An orthodontist, the head of a major teaching
occurred on approximately 18 teeth. No institution, had treated a young boy with a cleft
midtreatment radiographs were taken. The case palate for several years. It was exemplary cleft
settled for a large sum just before trial. palate orthodontic treatment. The problem was
In another case, a legal colleague asked if I that the patient, now a teenager, was found to
would look at an orthodontic case he was han- have severe root resoprtion and would lose his
dling. After our meeting, he asked if he could anterior teeth. The insurance company had the
pay me to take the deposition of the defendant. case reviewed by an orthodontic expert in Cali-
I agreed. I had successfully sued this orthodon- fornia who advised settling. The orthodontist
tist once many years earlier. He was not aware said he did nothing wrong and would not settle.
that I was involved until he walked into the room In reviewing the case, I read his deposition. He
the day of the deposition. Again, the injury was claimed although he has read about root resorp-
severe root resorption, the worst I have seen in tion he has never seen it. This comment trou-
my career. A young 18-year-old girl is going to bled me as well as the insurance company. It is
lose almost every tooth in her mouth. Because I hard to believe that a competent orthodontist, a
still maintain a limited orthodontic practice, I teacher and a leader in the field, has never seen
was upset when I saw this destruction. I can only root resorption, and, therefore, there was no
assume that this girl must have had some meta- reason to take any midtreatment radiographs to
bolic predisposition to resorption. monitor for this condition. I did not agree and
Unfortunately, the orthodontic treatment was said so to the insurance company representative.
very sloppy. A 2-year case took 4 years. I assume
The lesson is that if you personally have not
that it was round tripping mechanics that caused
experienced root resorption you are fortunate;
the extensive resorption. However, the smoking
you are also in the distinct minority. There is
gun was the defendant's lack of radiographs. For
some reason, he only took periodic bite-wing sufficient documentation in the literature for
radiographs. On these films, a careful diagnosti- every orthodontist to acquire in-depth knowl-
cian could see lateral root resorption of the edge concerning root resorption. As an aside, it
premolars. At the deposition, he was very honest is not the fact that root resorption has occurred
and admitted it was root resorption that he did that constitutes negligence. It is knowing its po-
not notice or record at the time. There is no tential for existence and not monitoring for its
question that during treatment he should have occurrence that constitutes negligent treatment.
taken anterior radiographs. The attorney for the Many orthodontists have the impression that
defendant has offered a very large settlement. I there are many personal injury attorneys ready,
have advised my legal colleague that the ten- willing, and able to sue orthodontists. Just the
dered amount is not enough, but I suspect that opposite is true. Very few attorneys accept these
this case will settle. types of cases. Good medical malpractice attor-
The last case was unusual in that it is fairly neys are willing to learn the medicine in medical
uncommon for me to consult on defense cases, malpractice cases because the damages are po-
although, as a very small part of my practice, I do tentially large. They are not willing in most den-
represent defendants who have no insurance, if tal cases, particularly orthodontic ones, because
I believe in their case. I also represent many the injuries are usually not severe and the den-
colleagues and friends as their personal attor- tistry too technical. However, as you can see,
neys by monitoring and helping their insurance orthodontic malpractice exists, and I have at-
company—appointed attorney with their defense. tempted to identify the more typical cases from
This case presented a special, sticky situation my own practice.
<<
Article
>> Home | TOC |
Index
skill is never more beneficial than when assess- to comprehend questions and provide answers
ing the case at its outset, knowing what informa- that are responsive and to the point and that will
tion to gather, how to weigh it, and how that be understood by lay people; the ability to testify
information fits into an overall feel for the case. consistent with a given theme or approach; the
What factors go into the initial assessment? ability to build word pictures from technical in-
formation; and, whether or not one has that
Defendant's Records elusive animal personality and likeability as a
A review of the defendant's records usually re- witness. A further part of this evaluation is
veals a great deal about the doctor and his/her whether any deficiencies as a witness are capable
practice. All of the records are pertinent in this of being cured.
review and include patient health-history forms; Assessing the defendant's knowledge of the
informed consent forms and related informed science and professional standards applicable to
consent booklets or other information; incom- the treatment rendered is an early indicator of
ing or outgoing referral slips and correspon- whether an error may have been made. Asking
dence; fee for service and insurance billing data; what the doctor believes may have gone wrong
pretreatment records such as models, photo- and how it could have been prevented may pro-
graphs, full-mouth or panoramic radiographs, vide the defense attorney with significant insight
cephalometric plates, tracings, and analysis; and, into what the problem's cause may have been.
finally, the patient's diagnosis and treatment The doctor's management of a problem at the
plan(s). These records indicate to the attorney, very outset of the case can be an indicator of
possibly aided by an expert consultant, how thor- trouble. This information can assist in knowing
ough was the doctor's information gathering which experts to seek review from and may serve
process; whether problems, actual or incipient, as an early indicator of a defendant who is well
were evident; and whether there are weaknesses behind current thinking in diagnosis, treatment
present, or conversely, strong defenses evident planning, treatment armamentarium, and some-
within the records. There may be problems on one who is weak in interpersonal and risk man-
issues such as informed consent, diagnosis, treat- agement skills. All of this information is taken
ment planning, poor responses to treatment, the into account in the micro and macro evaluation
development of negative sequellae, or evidence of the defendant doctor.
of contributory negligence. On a macro basis,
the overall organization of the records speaks
volumes about how well a practice is managed The Injury Involved
and in turn how well patients are managed.
The chronological treatment record is also The nature of the injury and its impact on the
examined carefully for consistency of entries, plaintiff carries significant weight in the overall
sanctity, missed, cancelled, or broken appoint- evaluation of the case. Clearly, a case involving
ments by the patient; level of oral hygiene com- an extensive or aggressive oral cancer or a case
pliance; indications of problems during treat- that involves the loss of a substantial number of
ment; length of treatment; progress of teeth is evaluated differently than a case that has
treatment; and the doctor's response (s) to any few life-altering effects on the plaintiff. This in-
problems. Consistency among records, and the formation can be learned from several sources.
ability to rely on them as accurate representa- The defendant doctor may have insight that the
tions of who did, or said what, and when by plaintiff shared with him/her before their rela-
recording all visits and events, have a high im- tionship ended. The plaintiffs attorney may
portance. have written a demand letter describing the in-
jury and its impact on the plaintiff. A telephone
Interview With the Defendant Orthodontist conversation with the plaintiffs attorney may
The evaluation of the defendant is multifaceted. also serve to elicit this information. Although
Certainly, identifying whether the defendant this information must be verified and con-
makes a good witness is a significant part of this firmed, it is nonetheless valuable and must be
evaluation. This evaluation measures: the ability taken into account in the initial assessment.
<<
Article
>> Home | TOC |
Index
from affirmative defenses, such as statute of lim- that the longer unrealistic positions are taken,
itations, contributory negligence, or lack of cau- the longer the case will last. This also leads to
sation? Did the discovery support or suggest that increasing each party's investment in the case,
the plaintiffs damages are overstated, over- sometimes dramatically. Unfortunately, these
blown, or related to other conditions that were factors make the party's positions more polar-
not associated with the defendant's treatment? ized and more difficult to bring to a compro-
As discovery progresses, good defense attor- mise. Contemporary thinking among most pro-
neys and well-advised insurance carriers are fessional liability carriers is that they are not
looking to see whether the facts adduced during afraid to try cases in which they believe the
discovery change the initial assessment of the plaintiffs demand is simply unrealistic or has a
case and, in turn, how the case should be han- low percentage for success. The plaintiff should
dled from that point. Changes may affect the keep in mind that this is not always a dollars and
timing of settlement overtures, the amounts au- cents game, and they cannot worry the carrier
thorized to obtain a settlement, and how the into settlement by aggressively working a case.
defense will be conducted. In other words, well- When the plaintiffs position is beyond the reach
handled cases are constantly reevaluated as new of compromise, the carrier may simply take the
information comes to light. It is part of the skill very reasonable position of making the plaintiff
set of the defense attorney to know how to work for his/her money by having to convince
search for any and all information that bears on the jury, not only as to the merits, but also the
every issue in the case. No one wants surprises at value of their case.
trial. The failure to cast a wide net creates a risk
that harmful information will be learned for the
Trial
first time in front of a jury. The plaintiff has the
advantage because it is generally information The end game of most orthodontic malpractice
within the plaintiffs knowledge that the defense cases is settlement. Probably something close to
seeks. 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
Pretrial Motion Practice
rare default option. Because trial can be enor-
The pre trial period is typically the last 90 to 120 mously expensive for both the plaintiff and de-
days before a set trial date. It is a time in which fendant and risky because not even the best
the plaintiff and the defendant generally know attorney can guess the outcome each and every
all the facts and circumstances surrounding a time, it should be reserved for those cases in
claim, what the witnesses are likely to say at trial, which no other method of disposition is possi-
whether the party witnesses are likely to be good ble. Indeed, some practitioners view a matter
or bad witnesses, what the likelihood of success going to trial as evidence of failure of the parties
or failure at trial may be, and the estimated to adequately assess their case.
verdict ranges. The parties may not agree on All trial attorneys know that trial is not a
these, but they have the same information and search for truth. It is a pre-scripted, well-re-
have formed their own opinions. This period is hearsed presentation to create an impression or
clearly the last best chance for both sides to emotional response that is favorable to their side
settle the case. A case that is incapable of settle- of the case. The critical players in orthodontic
ment during the pretrial stage usually means malpractice cases are the parties, followed by the
that one of the parties has not evaluated their attorneys, percipient witnesses, and finally the
case correctly, one of the clients is being unre- expert witnesses. In my view, that is the order of
alistic, or there are critical facts that cannot be importance in the jury's eyes regarding who con-
agreed on by the parties and must be decided by vinces them and whom they accept information
the trier of fact, the jury. Because most of the from. The defendant doctor is at the top of the
facts that will be presented at trial have been list of witness importance because it is usually
developed and documented before the pretrial based on what he/she says and what the records
stage, mediation can sometimes be a useful tool. show, from which the jury must make a number
It should be kept in mind, by both parties, of conclusions such as the level of attention that
<<
Article
>> Home | TOC |
Index
was paid to the patient, the degree of expertise standards of care and other detrimental testi-
the doctor had in recognizing, diagnosing and mony from these witnesses. Their testimony
treating the patient, and what the standard of should stick to what they observed, what they did
care was. Equally important, the jury must de- and why, rather than what they think the defen-
cide whether they like and respect the defen- dant may have done.
dant doctor on a personal as well as on an ob- Finally, expert witnesses are frequently seen
jective level. It is the defense attorney's job to as canceling each other out. However, when the
prepare the doctor for trial in order for his/her plaintiff has the burden of proof, particularly in
testimony to have a favorable impact on the jury, areas in which the science is not well developed
whether it is under direct or cross examination. and there are significant differences of opinion
Attorneys add or subtract to the case by bring- by well-qualified experts, having one expert can-
ing clarity to facts and terms that may not be cel the other out can be devastating to the plain-
familiar to the jury, by being speedy and efficient tiffs case. In fact, it is one of the instances when
in their presentation of the case, by treating the his/her case is most vulnerable. Selecting ex-
jury with respect for their intellectual capacity, perts who know their field and who can convey
by not being unduly argumentative, and by pre- technical information to lay people in a lan-
senting lucid facts and clear arguments. They guage that they can comprehend are very desir-
can severely prejudice their clients' cases by pre- able skills.
senting a sloppy case, treating parties or wit-
nesses with disrespect, being overly aggressive Conclusion
advocacy, failing to answer obvious questions It is likely that both parties view the workup of
that the jury must have based on the evidence their cases quite differently because they ap-
presented, constantly repeating the same evi- proach them differently, both from an economic
dence during direct or cross-examination of wit- and a management perspective. Nevertheless,
nesses or parties, and not giving the jury a rea- there are significant procedural similarities re-
son to side with their interpretation of the facts. garding the life cycle of a case and how the
Third-party witnesses can be swing witnesses attorneys and their clients have to assess their
in that they are frequently not invested in the position at each step along the way. After all,
case as parties or hired expert witnesses. Thus, they are dealing with the same law, generally the
they have a patina of objectivity or at least a lack same set of facts, and the same jury or trier of
of obvious bias. Depending on the anticipated fact. Good attorneys on either side of the case
testimony, defense counsel should be careful, usually reach similar conclusions, which proba-
particularly with subsequent treating doctors, in bly explains why nearly all malpractice cases are
avoiding introducing expert opinions regarding settled.
<<
Article
>> Home | TOC |
Index
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.
" . . . people believed to possess information be- and of itself usually confers expert witness status
yond the ordinary knowledge of judges and ju- in most courtrooms, the judicial process of qual-
ries."7 From these general characteristics, it is ifying medical experts is rightly seen as "a course
clear that courts have great discretion in quali- sieve for sorting knowledgeable witnesses from
fying medical expert witnesses. Anyone possess- poor ones."11 Unfortunately, in many instances,
ing a medical or dental degree may be granted the lack of medical knowledge by judges and
expert witness status by a court. This holds even juries often causes them to focus on the expert's
if the dentist or physician is not a specialist in the style of presentation or the elaboration of im-
field in question.8 It is possible, then, for an pressive sounding credentials instead of the pos-
orthodontist accused of negligence to have ex- session of true expertise.1
pert opinions on the standard of care and
breach thereof rendered by nonorthodontists.
Amendment to Federal Rule of Evidence
Types of Medical Witnesses 702: Establishing Expert Witness
Reliability
A fact witness may only testify about facts directly
observed, whereas the expert witness may offer The course sieve of expert witness courtroom
an opinion about facts not directly observed. A qualifying may become finer in coming years
treating physician, whether he/she treated the because of new instructions given to trial judges
patient in the past, presently, concurrently, or contained in recent amendments to Federal
subsequently to the defendant's treatment, and Rules of Evidence. Amended Rule 702 should
who is called as a witness, is legally obligated to help the courts fulfill their role as gatekeepers of
testify about the treatment rendered to a patient the expert witness on a more consistent basis.
of record. In this capacity, he/she may only serve Now, before the offering of any expert testi-
as a fact witness. However, the treating physician mony, trial judges are instructed to make a de-
may also qualify as an expert witness, and it is in termination of expert opinion's reliability. Be-
this capacity that he/she may then offer opin- fore the formalization of the gatekeeper role,
ions about the facts at hand.8 some trial courts had been remiss in carrying out
A nontreating medical expert witness is not this responsibility. Challenges by opposing par-
compelled to provide legal testimony. An exam- ties as to the reliability of expert opinions are
ple of such a practitioner is the physician hired also more likely to be heard by the trial judge.
by an insurance company to perform an inde- An in-depth investigation into the rationale of
pendent medical evaluation regarding an acci- an expert's opinion should be undertaken as
dent, a disability claim, an insurance applica- soon as possible in the litigation process. Coun-
tion, and so on. In this case, a patient of record sel must explore all information on which the
or treatment relationship is not formed, and expert relies, whether it is admissible or not. If
thereby a patient-doctor relationship is not es- the trial judge appears reluctant to conduct a
tablished except in unique circumstances. reliability assessment, counsel can suggest the
appointment of an independent expert to assist
in the evaluation. At pretrial conferences, the
Medical Expert Standards attorney should stress the need for the reliability
Honesty and objectivity are essential principles assessment of any anticipated expert testimony,
of ethical testimony.9 The American Dental As- as well as encouraging disclosure of all witnesses
sociation Principles of Ethics and Code of Pro- who might render expert testimony. The trial
fessional Conduct advises that "it is unethical for court should be given adequate time to make
a dentist to agree to a fee contingent upon the the determination of expert witness reliability
favorable outcome of the litigation in exchange but not so far before trial that an excluded ex-
for testifying as a dental expert."6 Rappeport10 pert witness can be replaced by another of the
notes the prime responsibility and most impor- same ilk. All too often the credentials, character,
tant legal service that medical experts serve is to and background of the expert are perhaps more
promote the search for the truth. important than his/her actual testimony. An ex-
Because possession of a health care degree in pert who possesses a distinguished record in
<<
Article
>> Home | TOC |
Index
these areas provides the best chance of securing their arguments. Attorneys perform direct exam-
honest and objective testimony.1 ination on experts they have hired, with oppos-
ing attorneys subjecting the expert to cross-ex-
Medical Expert Liability amination. This system is considered the best
route to truth and justice.
Although actions against privately hired medical Annas,14 Bayer,15 Jerrold,16 and Laskin17 re-
experts and court-appointed medical evaluators mind us that instead of building bridges to truth
are uncommon, even in cases of bizarre or bla- and justice, contradictory medical testimony can
tantly dishonest testimony, a trend is emerging create mountains of confusion between justice
toward holding expert witnesses liable for their and the judge and jury's understanding of the
professional errors.12 Eight states have ad- medical issues and facts. Although some legal
dressed the issues of expert witness immunity.
analysts believe that lay juries come to under-
Among them, the Supreme Court of New Jersey
stand complex medical and scientific issues,18
held that a forensic expert is not immune from
others argue that juries are easily swayed by non-
liability for deviating from accepted standards of
practice. A California appeals court stated that factual or scientifically weak testimony, known as
expert witness immunity "does not encourage junk science.19 In the ideal world, courts would
witnesses to testify truthfully; indeed, by shield- rule junk science as inadmissible, and juries
ing a negligent expert witness from liability, it would decide guilt or innocence based on truth-
has the opposite effect."12 Increased judicial rec- ful expert testimony and not be swayed by emo-
ognition of the relatively new tort of expert wit- tional appeals. Reality forces the acknowledge-
ness malpractice may discourage unethical testi- ment that decisions by judges and juries might
mony and elevate the quality of courtroom not rest on the truth but on the best perfor-
testimony. mance by expert witness or lawyer, along with
exploitations of courtroom emotions.
Skillful cross-examination by an opposing at-
Beware the Hired Gun
torney provides an important check on the use
Fletcher13 described the "hired gun" as an ex- of junk science. The cross-examiner has the op-
pert witness in a legal dispute whose opinions portunity not only to impugn an expert's credi-
are for sale to either party.13 Usually working bility but also to cast doubt on the reliability and
exclusively for the plaintiff or the defendant, significance of the medical expert's evidence.1
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 Neutral Expert
guns, the desire for financial gain supercedes Hope for improved quality of medical expert
professional ethics. Tsushima and Nakano8 testimony may be found in the independent or
wrote, "these physicians [or dentists] do great neutral scientific expert or panel. A neutral
harm to themselves and the medical [dental] medical expert can be appointed by a judge, as
profession, and fail to safeguard rights and free-
opposed to the plaintiff or defendant selecting
doms of plaintiffs or defendants." The opposing
his own. A neutral expert's work is separate from
attorney has the responsibility of alerting the
the adversary proceedings in the courtroom,
jury to contradictions presented by the hired
gun. Cross-examination is the best method of and his/her findings may override other court-
revealing hired guns on the witness stand and room expert testimony (Rule 706, Federal Rules
casting doubt on their credibility. 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 battle of the experts. The excitement over this
The adversarial nature of the American legal approach, however, is tempered by the inherent
system confers equal rights to plaintiffs and de- difficulty of changing an established and power-
fendants to present medical experts to enhance ful legal system.
<<
Article
>> Home | TOC |
Index
enerally, regulatory agencies are either agencies that do have the force of law behind
G federal or state. In addition to your State
Licensing Board, there is the State Department
them. Your response to a nongovernmental
body should be similar to your response to a
of Health, the Department of Human Services, state or federal regulatory agency.
the Pollution Control Agency or Environmental At the outset, one should understand that
Protection Agency, the Department of Com- regulatory agencies' interests are often in con-
merce, Occupational Safety and Health Admin- flict with the practitioner's interests. Regulatory
istration, Medical Privacy Rules, the State Attor- agencies' primary interests are to protect the
ney General's Office, and countless other well-being of the public, to set standards of care,
alphabet soup regulations and organizations. to affect quality improvement, to contain costs
There are also regulatory bodies that are neither and encourage the use of scarce resources
federal nor state but have considerable influ- wisely, to maximize cost/benefit ratios to the
ence on a dental practice such as your dental patient, to discourage behavior on the part of
society's peer review committee, a hospital's the practitioner that society deems inappropri-
peer review committee, or a managed care orga- ate, and to rehabilitate practitioners whose in-
nization's peer review committee. Even though terests in autonomy and profitability are in con-
their actions may not have the force of law, they flict with the regulatory body's interests.
can be as important in affecting your practice as Most dentists think of their State Licensing
Board as primarily responsible for the testing
and granting of licenses to practice. Although
From Hagglund, Weimer and Speidel, Minneapolis, MN.
Presented in part at the 1999 Annual Session Risk Management licensing is an important role, the State Licens-
Program in San Diego, CA. ing Board is also charged with disciplining vio-
Address correspondence to T. Michael Speidel, DDS, MSD, JD, lations of the state statutes related to dental
Hagglund, Weimer and Speidel, 4000 Water Park Place, 5101 practice. The grounds for discipline may be in-
Olson Memorial Highway, Minneapolis, MN 5.5422.
Copyright 2002, Ehevier Science (USA). All rights reserved.
competence, negligence, impairment, fraud and
1073-8746/02/0804-0010$35.00/0 dishonesty, mental incapacity, sexual impropri-
doi:10.1053/sodo. 2002.127872 eties, or miscellaneous infractions such as, in
some states, failure to complete mandatory con- Furthermore, you will not be able to compel the
tinuing education and failure to pay income appearance of witnesses, conduct pre trial discov-
taxes or child support. In many states, the ma- ery, have a public hearing, have a jury trial, or
jority of complaints are in the category of incom- invoke privilege against self-incrimination. You
petence, welfare fraud, or drug and alcohol also may not be entitled to a decision based on
abuse. the record with a statement of supporting rea-
The reason that the relationship between the sons.
practitioner and the regulatory agency is similar In addition to the absence of some of the
to David versus Goliath is based on the relative customary requirements of constitutional due
differences between the two in experience, re- process, there are some other limitations to the
sources, and the limited requirements for fair- administrative hearing. The rules of civil proce-
ness or due process. Most people think of due dure do not apply. The rules of evidence do not
process as something guaranteed under the apply; evidence that might be inadmissible in a
United States Constitution. That high standard courtroom because it is not reliable or relevant,
is only applicable when governmental action will may be admissible. The hearing body may bring
impair an individual's life, liberty, or property. up issues that were not in the original complaint,
When governmental or quasi-governmental ac- issues of which you had no notice. You may have
tions do not constitute a denial of an individual's to testify in a narrative manner, thus risking
life, liberty or property, the government or other self-incrimination. The decisions may be made
agency may not be required to give the person by individuals who lack expertise in all areas of
any hearing or process. Obviously, State Licens- dentistry, especially in a dental specialty area.
ing Board's actions do not deprive one of his/ The members of the hearing body may be pri-
her life or liberty, and it is unclear whether or marily political appointees. Finally, when judg-
not one has a property right in a dental license. ing you, the hearing body may subject you to
Furthermore, having a license and having a re- unrealistically high personal or aspirational stan-
stricted license are quite different things. dards of care, rather than the much lower legal
Regulatory agencies use a balancing test in standard of care.
determining how many due process require- In most states, the legislature has given the
ments must be provided to individuals who ap- licensing boards broad subpoena power. Stat-
pear before them. The regulatory agency bal- utes mandate cooperation with board investiga-
ances the worth of the procedure to the tions, and the prohibition against self-incrimina-
individual against the cost to society as a whole. tion attaches only in a criminal case. It is small
In most cases, the respondent (you, the lic- consolation that evidence and testimony consid-
ensee) will get limited due process. It is the ered in the disciplinary hearing usually cannot
belief that the government's interest in protect- be used in a criminal case, but, in a subsequent
ing the public would be hindered if every inves- proceeding, the evidence and prior testimony is
tigation had to be conducted with full due pro- often discovered. These factors place the dentist
cess protections. Therefore, the due process the at a considerable disadvantage in the hearing
respondent is accorded is usually limited to no- process.
tice, an opportunity to present your case, an It is important to consider the source of any
opportunity to present evidence, an opportunity complaints to the dental board. Complaints can
to have an attorney present, and a transcript of arise from patients, employees (more likely,
the proceeding. former employees), other dentists, or investiga-
In contrast to a formal judicial proceeding, tors employed by the Office of the Attorney
the respondent may or may not get a neutral General or the Board of Dentistry itself. Usually,
decision maker. Usually, the respondent will not complaints must be submitted in writing, and
be able to strike jurors or the judge. The respon- the board is required to investigate and respond
dent may not be able to present witnesses or to all complaints it receives. If the subject of the
expert witnesses, exclude irrelevant or immate- complaint is nonjurisdictional, such as an alle-
rial evidence, and confront and cross-examine gation of criminal conduct, the complaint will
witnesses. Often you do not know, and cannot be forwarded to the appropriate jurisdiction for
find out, who made the complaint against you. disposition.
<<
Article
>> Home | TOC |
Index
If the subject of the complaint is within the of the practitioner, any corrective action (s)
jurisdiction of the licensing board, the board taken by the practitioner, the potential for de-
may gather additional information from the terring future misconduct, the potential for re-
complainant or the respondent (licensee). Usu- habilitating the practitioner, the costs of investi-
ally the board will request from the respondent gations and hearings incurred by the board, and
a written response to the allegations in the com- the practitioner's ability to pay.
plaint. The board may send an investigator for The board has the burden of proof when
record reviews, interviews, or site inspections. taking adverse action against a licensee; how-
The investigators are no t your friends; be cordial ever, the standard of proof is usually the rela-
but no more helpful than is necessary. tively low preponderance of the evidence test. In
Some complaints are fact specific and present the case of an application for licensure, the bur-
little or no proof problem (eg, example the den is on the applicant, and the standard of
absence of a properly functioning sterilizer). proof is the same.
Others are more difficult to adjudicate such as Licensees should be aware that the board's
allegations of incompetence that are especially findings, other than dismissal, are usually public
problematic because bad results may stem from and may require voluntary or involuntary report-
one having actually rendered negligent treat- ing to other agencies. Adverse findings may re-
ment or substandard care, may occur through sult in bad publicity for you or your practice, a
no one's fault, may be satisfactory but fall below limitation on current or future employment,
the patient's or another complainant's expecta- prohibition from future employment or inde-
tions, or may simply be a result of poor commu- pendent practice, and a limitation to your pro-
nications between patient and practitioner. fessional liability insurance coverage or a change
After the board has finished its investigation, in insurance premium cost.
the board may dismiss the complaint, issue an When one receives a notice of complaint
advisory letter, or request a conference with the from the Board of Dentistry, a peer review com-
respondent. The conference may be informa- mittee, or even from a plaintiffs attorney, it is
tional or investigational and-either relatively in- very important to respond in a proactive man-
formal or quite formal. All conferences should ner. The following suggestions may be helpful in
be considered formal by the respondent and responding to the notice of complaint.
treated as such. After the hearing, the board will Do not delay; read the entire notice of com-
determine whether the complaint should be dis- plaint and respond in a timely manner. To do
missed or whether corrective action is war- otherwise may result in a default action against
ranted. you.
In addition to dismissal of the complaint, the Inform your professional liability insurance
Board of Dentistry may take a number of correc- carrier. You may or may not have insurance
tive actions. The most common are as follows: coverage; however, most policies require you to
(1) admonish or warn; (2) reprimand or cen- do so promptly. To do otherwise, may jeopardize
sure; (3) impose conditions to continue prac- any coverage you may have.
tice, such as remedial course work, practice un- Call an attorney who has experience in deal-
der supervision, or limitations that restrict the ing with your dental board and also possesses a
scope of one's practice; (4) required provision knowledge of dentistry, if possible. This is impor-
of nonremunerative service; (5) submission to tant because every licensing board is slightly dif-
periodic office inspection; (6) temporary sus- ferent. Make sure your attorney knows the
pension of one's license; or (7) permanent re- board's rules and the Dental Practice Act. Even
vocation of one's license. In the case of an ap- if you are asked to do no more than write an
plicant for licensure, licenses may be either explanatory letter to the board, make sure you
denied or conditioned. know and follow any required format. Let an
In determining a disciplinary response, the experienced attorney review your response as he
board will consider several factors including the may be able to minimize the charges and keep
existence of or clarity of a practice standard, the you from making any damaging admissions.
seriousness of the misconduct, the frequency or If you have coverage under your professional
duration of the misconduct, the perceived intent liability policy, the insurer will provide an attor-
<<
Article
>> Home | TOC |
Index
David Versus Goliath 241
ney for you. Usually, you will have several indi- tist's records, and you have kept the type of
viduals from which to choose. Interview them, records necessary for a good defense rather than
and make sure you are comfortable with the just the minimum records. If you cannot easily
attorney. If you are not, find your own, and ask obtain necessary records from a prior treating
the insurer to approve their representation of dentist, you will definitely need an attorney's
you. If you wish, you can have your personal help to obtain them.
attorney work with the attorney provided by the Do not alter any records! To do so, may result
insurance company. Often it is a good idea to in evidence being construed against you and
have someone who represents only your inter- may subject you to civil penalties. If your records
ests. The attorney provided by the insurance are difficult to read, include shorthand abbrevi-
company may have somewhat divided loyalties. ations or typewritten transcript, with explanatory
In the event that you are required to appear text in parenthesis, along with a copy of the
for a hearing, it is extremely important to be original.
represented by appropriate counsel. Many law- Appear before the board if necessary. The
yers expect a hearing to be like a formal judicial hearing will be transcribed by a court reporter,
proceeding and may irritate the hearing panel tape recorded, or both. The board will ask you
members by trying to impose their idea of ap- questions, not your attorney. Your attorney may
propriate procedure on a board composed of not answer for you. In answering questions, lis-
dentists and others who are totally unfamiliar ten carefully to the question. Make sure you
with formal judicial procedures. understand the question. If you do not under-
The aforementioned recommendations may stand, ask the questioner to repeat or clarify the
appear self-serving, but because of the relative question. If you answer, it will be presumed that
power differential between you and the regula- you understood the question. If you do not know
tory agency, to appear pro se (representing one- the answer to a question, say so. Do not guess. If
self), may jeopardize your defense and, in turn, you must guess, make sure the record reflects
your livelihood. that it was a guess.
Refrain from conversations with the board Tell the truth. Tell your side of the story in as
members and the complainant (if you know who succinct a manner as possible. Before speaking,
the complainant is). If you do not know who the take time to formulate the answers in your mind,
complainant is, do not discuss the complaint then answer and stop speaking. Do not volun-
with anyone until you consult with your attorney. teer any information not asked for as you may
If required to write a response to a board's in- make a damaging admission or inadvertently
quiry, make sure your attorney approves it, and, give the questioners additional potentially dam-
if necessary, your insurance company as well aging avenues to explore. One of the question-
before sending it to the board. er's favorite tricks is to sit back and let you
You must be absolutely candid with your at- ramble, giving you enough rope to hang your-
torney. To do otherwise may jeopardize your self. Do not volunteer information unless you
defense and subject your attorney to the possi- are sure it will help you. Do not try to be helpful
bility of being blindsided by damaging evidence or a nice guy. Use specially prepared exhibits,
unknown to him/her. Do not be concerned such as a time line chart, if it will help tell your
about making damaging admissions to your at- story.
torney; your admission will be protected by the Honest admissions of misconduct, along with
attorney/client privilege or the work-product good explanations and credible evidence of any
doctrine. The foregoing presumes that you also remedial action taken, may minimize or prevent
maintain confidentiality. If you disclose pro- adverse action by the board. In answering ques-
tected information to third parties, you may be tions, being vague or evasive damages your cred-
considered to have waived the attorney/client ibility. If you do not think a question is fair, often
privilege. you can rephrase it and then answer the ques-
Assemble all necessary records, including tion. Finally, do not try to answer compound
records from prior treating dentists if appropri- questions because the correct answer for one
ate. Hopefully, you did not accept a transfer part of the question may be incorrect for an-
patient without obtaining the prior treating den- other part of the question.
<<
Article
>> Home | TOC |
Index
242 T. Michael Speidel
Most lawyers have videotapes available on In sum, actions by regulatory agencies are
how to conduct yourself at a deposition; take truly David versus Goliath proceedings. This is
time to learn the principles. Watch the tapes a not a blanket condemnation, just a fact of life.
few times, or spend a few more dollars and ask You can give yourself an advantage by knowing
your attorney to prepare you for the hearing. It what they do and how they do it. You must be
will be one of the best investments you ever aware of the fact that you are involved in a
make. process that is somewhat informal and does not
Even formal hearings are somewhat informal, have the built-in protections for the respondent
like a deposition. Although the rules of evidence that formal judicial proceedings have. Even
may be relaxed, the proceedings will be re- though the proceeding is informal, you must
corded. Do not let the seeming informality of take it seriously and prepare adequately. The
the proceeding lull you into adopting a casual results are just as important as they would be if it
approach. Every word you utter is important. were a formal judicial proceeding. Finally, your
Be serious, humble, and cordial at all times. attorney, depending on the jurisdiction in which
Do not become indignant or defensive in your you practice, is not particularly welcome at ad-
manner. Remember, the board members are
ministrative proceedings and cannot protect you
responsible to the public, and they are just doing
as well as in a formal proceeding. Therefore, you
their job.
have to protect yourself by thorough prepara-
After the hearing is completed, you will re-
ceive a notice of the board's proposed action. If tion and obtaining and following good legal ad-
you disagree with the proposed action, your at- vice.
torney may be able to suggest to the board some This general overview of the basic functions
reinterpretation of the facts or a modification to of any regulatory agency and their relationship
the proposed action. The board may or may not with you, as a dental practitioner, cannot be
accept the counter-proposal. If the board does considered legal advice from the author, editor,
not accept your proposal, there are limited op- or publisher. The opinions expressed are those
portunities for formal appeal to an administra- of the author. The principals expressed may not
tive law judge or a lower court. Appeals are very be applicable to a specific situation. All agencies
tricky and should not be attempted without com- function slightly differently from one another.
petent legal help. Before appealing, try to ascer- The facts differ from situation to situation, and
tain the likelihood of a successful appeal in your the law is constantly changing. If you as the
jurisdiction because appeals are often unsuccess- reader have questions about your particular sit-
ful and simultaneously expose you to additional uation, you are encouraged to seek competent
expenses for the appeal process. and timely counsel.
<<
Article
>> Home | TOC |
Index
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 Concern for the Buyer or Not?
for fewer days per month. This can also be a
significant concern for the practitioner who has This is related to the previously mentioned issue
built a very large practice in either a single or but is different. In a professional practice trans-
multilocation format. The relevant consider- fer and sale, unlike a more routine commercial
ation is whether the practice will benefit from a business sale, one of the major differences is, or
long-term succession plan or a short-term tran- should be, the continuity of care for the patients
sition. If it is apparent that a short-term transi- who have trusted a chosen practitioner. Viewed
tion will impair the practice by not enabling the from this perspective, the question arises, how
next generation of practitioner(s) to gain ad- does the exiting practice founder, owner, or
ministration, finance, personnel, and clinical ca- partner assure continuity of a similar commit-
pability sufficient to effect a successful transfer, ment to treatment, both philosophically and
the doctor/owner should lean toward a succes- mechanotherapeutically? In other words, for
sion plan with the sale of incremental equity some doctors, it is very important to find a buyer
interests. By way of contrast, if it is apparent that or new partner who shares their philosophy of
either a smaller practice, or each of the offices of care and commitment to the mission of the prac-
a multilocation practice that will be split, cannot tice. Others are far less concerned about philos-
support two practitioners even with a reduced ophy and more concerned about technique and
schedule for the doctor/owner, or the available how cases will finish. If a seller is very concerned
space is cramped or the patient flow limited, the 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
Address correspondence to Randall K. Berning, JD, LLM, Pres- transition.
ident, Berning & Affiliates, Inc, National Administration Office,
3400 Tamiami Trail North, Suite 201, Naples, FL 34103-3717.
www. berningaffiliates. com Who's in Charge?
Copyright 2002, Elsevier Science (USA). All rights reserved.
1073-8746/02/0804-0011$35.00/0 In many routine commercial business sales, out-
doi:10.1053/sodo.2002.127873 side of professional practice transfers, the buyers
and their advisors are in charge. They make the Consequently, today many buyers and their
offer, they question or assert value, they do their advisors are very interested in pinning down the
due diligence, and they propose terms and often net and using an earnings method to arrive at
take control of document drafting. I have found value. For sellers to apply any other method for
that generally because of the inexperience of the arriving at value simply begs the question of how
new orthodontist in building a team of advisors can the buyer pay for the practice debt and meet
and the wide range of items to undertake cou- living expenses, which will have to be answered
pled with limited financial resources, orthodon- sooner or later and will ultimately result in using
tic practice buyers generally do not take control some form of net analysis to arrive at value. In
of the process. Therefore, sellers and their advi- short, it is my advice to sellers that it is much
sors very often put the transaction together. easier and more direct to simply use a profes-
However, as noted later, these transactions can sional earning-based valuation at the start and
fail if the deal's terms and documents are or be done with it.
appear to be unduly one-sided in favor of the At the same time that sellers grapple with how
seller and the buyer does not feel that an equi- to arrive at value, they need to bear in mind that
table transaction is occurring. A word to the wise value is not price. Value can be arrived at using
is to be solicitous in inviting comments on the a valuation methodology or a group of compa-
valuation methodology and the development of rables. But price is what is negotiated when the
the terms for the transaction. Direct your legal value is known and usually takes into account
counsel to drop unnecessary legalese and draft other considerations such as the tax-planning
documents in an evenhanded manner. This will strategies of the parties, the goals of each party,
go far in making the deal come together quicker and the terms of the deal.
and in a pleasant manner.
What Form of Transaction to Use?
Seller's Fundamental Considerations Form is a basic decision but one that must be
made and in the case of orthodontic practice
How to Get to Practice Value?
transfers often seems to be made by the sellers.
Of all of the many considerations leading up to What is being sold and how it is to be purchased
a consummated practice transfer or equity inter- leads to the time-honored debate that sellers
est purchase, valuation was once, and still can and their advisors enter into year in and year
be, the toughest. Thankfully, in contrast to 20 out. Namely, is the form of the transaction best
some years ago when I started in the area of set up as a sale of assets, in the case of an
practice transfer, there are now good resources incorporated practice a sale of the stock of the
to guide both the buyer and seller, first to un- professional corporation, or if the practice is a
derstand the theory of professional practice val- partnership, the sale of either the partnership's
uation and second how to apply it to formulate assets or the partnership interest of the part-
the value of the practice. In the case of orth- ner (s)? The various factors favoring an asset,
odontists, the American Association of Orth- stock, or partnership interest sale depend on
odontists has developed the Practice Transition who is making the decision, and the answer can
Alternatives binders, one for owners and one for be dramatically different for the seller and the
younger professionals. In each binder, there is a buyer engaged in the same transaction. The rea-
monograph entitled Guide to Orthodontic Practice son I find sellers generally making the call is that
Valuation.1 The guide discusses valuation options in orthodontic transfers, particularly of larger
but emphasizes the earnings based methods, practices that have a long-term buyout, immedi-
particularly the capitalized earnings method. ate asset sales often make little sense because the
The reason for this emphasis is that by using a longer transfer time frame usually pushes the
practice's determined net income to arrive at sale into either an equity (stock or partnership
value, a buyer can identify how much of that net interest) transfer. Further, sellers generally will
they can use to pay for the debt service and prefer to sell stock or a partnership interest be-
therefore be able to project how much income cause they receive capital gains treatment on any
they will have to live on. receipts from the transfer. However, the pur-
<<
Article
>> Home | TOC |
Index
chaser of an equity interest takes the practice is actually after you have secured your practice
subject to all liabilities including tax, contingent, valuation and started to weigh what the actual
and unknown. Purchaser's counsel will often purchase price will be.
seek to exclude certain liability categories be- If you are using a long-term transition, partic-
tween the buyer and seller via indemnity (it will ularly if it involves the sale of stock, significant
not apply to third-party creditors), and this is balancing of interests can be done to allocate a
generally a reasonable request. portion of the purchase price to the interest
Often overlooked, but of equal importance, is equity (allowing for capital gains) and another
what is not going to be sold. This is sometimes amount to management services (generally ordi-
thought of in the context of orthodontic cases nary income but currently deductible to the
that will be completed by the seller in a few payor). This type of allocation is not done as
months after closing or personal items that will fluff but in fact recognizes the time-consuming
be taken by the seller. Attorneys refer to what is mentoring that is necessary for a new practitio-
not to be sold as the liabilities and responsibili- ner to educate himself/herself to sufficiently
ties a purchaser retains after the sale. My expe- take over the management aspects associated
rience is that both areas must be addressed early with running a significant practice such as the
in the process of coming to terms for the parties' necessary orientation to practice administration
proposed transaction. Cases to be completed ties and clinical skill refinement. Comparable ap-
directly to risk-management concerns by the proaches can be fashioned for partnership trans-
purchaser. For example, it is far better for the fers. One should have a third-party arms length
purchaser to never touch a case in the last 4 to 6 negotiation of any allocation should any after
months of treatment than to have a purchaser closing taxing authority question the asset allo-
share portions of the patient's treatment. In the cation. Obviously, the ultimate allocation should
worst-case scenario, the parties get involved with be stated clearly in the transition agreement.
finger pointing regarding who did what, should
the patient be disgruntled or a professional mal-
Will You Get Paid?
practice action occur. Active and retention cases
to be transferred should be listed in an exhibit In most orthodontic practice transfers, other
to the transition agreement along with all as- than those that are immediate purchase and
sumed liabilities. sales, regardless of how they are to be financed,
Being mindful of tax considerations. Preliminar- debt-encumbered purchasers are often at a loss
ily, sellers need to bear in mind that although to come up with much of a downpayment. Tran-
many of the once common tax-planning tech- sitions that are immediate purchase and sales, if
niques are history because of legislative tax re- fairly valued and if structured properly, find the
form changes, some planning options still re- seller to be generally well protected using rela-
main. However, if either side pushes too hard tively standard agreement provisions. For those
for their best tax result, it can result in breaking that are not immediate transfers, the lack of a
the deal. Very different tax consequences can down payment throws out the old axiom that a
result from a purchase of assets or stock, but they purchaser should put 10% to 20% down. Fur-
are best left for discussion within the context of ther, although the amount of money a pur-
your personal situation with your accountant or chaser can put down should have a bearing on
tax advisor.2 the sale price, transition experiences over the
Rather than rehash general tax consider- last 5 years show it does not. This is because in
ations, it is better to focus on a couple of areas part of the fact that a virtual purchaser's market
that can make a significant impact. One has exists (due to fewer purchasers) where purchas-
been eluded to, specifically, admonish your tax ers and their advisors are driving harder bar-
advisor or transaction advisor to come up with a gains, despite the buyer's lack of resources.
balanced tax approach that does not severely An illustration of the sequence of steps to
disadvantage either seller or purchaser. Next, follow in practice transitions and how they might
pay attention to all allocations of the purchase progress follows. First, the seller should secure a
price but not late in the discussion, as many third-party professional valuation. This valuation
seem to do. Instead, do it early on. A good time must ultimately be agreed to by both parties.
<<
Article
>> Home | TOC |
Index
Second, the parties must negotiate terms for a terms. Drawing up the documents before agree-
multiyear buyin/out in which the cash flow is ing to the terms is time-consuming and expen-
tied to the valuation and the terms. Next, each sive, and starting with them followed by subse-
parties tax advisors should agree on the asset quent refinement after review by both parties
allocation vis a vis the purchase price. This is and their advisors is a far more torturous path.
followed by the document drafting. It is impor- Starting with a term sheet ensures that the buyer
tant that the placement of a downpayment be and seller have agreed on the key points before
deemphasized; for, in reality, it is nothing more incurring costs for attorneys and other advisors
than one more of the transition's terms. This to analyze and document the deal.
does not mean that a seller is not well advised to Advisors. The previous statements raise the
pay attention to a buyer's creditworthiness, char- critical issue of getting the right advisors in the
acter, personality (yes personality), and capabil- first place. It can be tempting for a seller to use
ity to take over the size of the practice under a long-time advisor, even if that person has no
consideration. These are areas that if the seller experience in this area. This can be a mistake
fails to investigate, he/she does so at his/her simply by virtue of using someone who has a
peril. I have seen transfers flounder with an lesser degree of expertise concerning what is
introvert purchaser taking over an extrovert sell- required to effect the transition. This will have a
er's practice because in part of the fact that the direct impact on the time and cost of getting the
referral sources and the patient base have sub- transition accomplished. It is often better to use
stantially different expectations of the purchaser the services of a transition consultant and/or a
than was in fact the case. Often, transitions in- corporate, business law, tax law, or health care
volve a structure that is detailed, well docu- law attorney. After consultation with experi-
mented, and carry a purchase period for an enced advisors, if desired, have the long-time
initial equity interest followed by a sale of the advisor review matters with you. Buyers some-
remaining equity interest. Likewise, any com- times are tempted to use family members or go
pensation and purchase payments must be prop- it alone until the last minute when they become
erly structured and documented to take advan- overwhelmed and finally seek counsel. The par-
tage of existing tax laws. 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-
What Is the Best Process?
selves rather than to wait too long. Both parties
For those sellers who are concerned about how should avoid strident my-way-only advisors. Deal
to make a deal or those wanting to avoid elemen- makers are usually better representatives than
tary mistakes relating to process, other commen- deal breakers. Finally, have a clear understand-
tary exists.3'4 Specific helpful pointers on good ing of the roles that your advisors will play right
process to follow in structuring a proposed prac- from the beginning. Namely, it should be the
tice transfer, whether it be a buyin/out or a seller's transition consultants, accountant, law-
more immediate variation of transferring or yer, and other advisors who prepare all financial,
transitioning from an associateship relationship tax, and valuation materials for the review by the
into an equity interest are provided below. buyer or prospective partner. A buyer's review
Terms not documents. In my view, doctors are should not include remaking the deal that par-
well advised to first get down what is important ties have articulated in principle. It can take
to them in the form of a terms listing. This could strong nerves and competent advisors on the
start with the seller identifying key items that seller's side to stay the course and not get side
make it worthwhile to go through all the cost tracked.
and time of developing the practice transition. It There is a difference. Should you seek an asso-
can also start with both parties using a facilitator ciateship first for a longer period of time or a
to guide them in raising key issues and items for more immediate buyin/out? Without a long ex-
their mutual consideration. Only after the terms planation, it is worthwhile to explore with a sell-
are fairly well established and agreed to in prin- er's advisors, given the particulars of a practice's
ciple should the seller proceed to have a set of situation and the desires of the seller, how best
documents prepared to reflect the parties' to approach the planned transition. There is no
<<
Article
>> Home | TOC |
Index
Transitioning the Orthodontic Practice 247
question that certain practice transition situa- can recommend a specific form based on his/
tions benefit from a more immediate transfer her experience.
approach. If this is the case, more needs to be Representations and warranties of the parties.
done in a shorter period of time. This includes Some sellers, and some advisors too, unfortu-
preparing a well-developed transition plan that nately do not give the weight necessary to tailor-
is tied to a timeline indicating to buyer and all of ing the representations and warranties made to
his advisors when valuation, cash flow, terms, specific transactions. In fact, it is these provisions
financing, document drafts, and closing dates that are the foundation on which each side relies
are expected or anticipated. 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
Seller's Additional Issues form the basis for the transition under consid-
eration. Particular care should be made, aside
When any seller of an equity interest in a prac- from more routine items, in representing the
tice or sale of the entire practice moves past financial status of the practice (what is true and
valuation and negotiation of preliminary terms, accurate as to any financial, patient, or treat-
there are still additional issues that need to be ment status for cases completed, in progress, or
addressed prior to drafting the written agree- in recall, in addition to whether or not the staff
ment. Some of the more important areas for a is staying with the practice, the referral base is
seller are as follows. stable, and so on). Sellers too should seek rep-
Covenant not to compete. Nothing other than resentations and warranties from the purchaser.
reasonableness dicates a geographic radius or One such significant item is that a buyer under-
other area and an amount of time for the cove- stands that an ownership change in the practice
nant. What considerations go into the terms of is a major event, and, particularly in the situation
such a covenant? In part, an appropriate geo- of the sale of an entire practice, there is no
graphic and temporal restriction should be suf- assurance that any projected result will occur.
ficient to protect the asset value of the practice. This is particularly the case in the specialty of
Consideration needs to be made if there is any orthodontics because most practices are heavily
area that a seller would want to practice in after dependent on referral relationships that have
the sale or a buyout. Should a covenant against a been either cultivated over time, age relative, or
seller still apply if the purchaser defaults on the those that are personality dependent. This as-
purchase? The answer is generally no. These pect must be recognized since loyalties often
issues illustrate the range of items that should be change when ownership changes occur.
specifically addressed in the transition docu- Orthodontic cases in treatment and a listing of
ments. Be aware that most buyers want the ben- patients. Sellers are well advised to protect their
efit of their bargain, and they will want a geo- records. There is seldom reason to leave all
graphic area and time period sufficient to records at the practice premises when leaving
protect the purchase of the goodwill they have the practice. Although it can be nice not to have
made. Be prepared to be reasonable. to store records, it has the potential to leave
Disputes and their resolution. In any long-term sellers particularly vulnerable. All active patients,
transition with a phased buyin/out, having a defined as those with an appliance in the mouth;
stated approach to resolving disputes is advanta- recall patients, defined as patients who have
geous. Rather than resorting to litigation, a bet- been examined but who have not started treat-
ter strategy might be meditation followed by ment yet and are subject to regularly scheduled
binding arbitration. My reasoning, in contrast to recall examinations; and retention patients, de-
some members of the litigation bar, is that the fined as those patients who are within 12 months
community of orthodontists is too small of a after the end of a full-treatment case, should be
professional population and patients too atten- listed on an exhibit to the appropriate transac-
tive to their doctors' circumstances to warrant tion document. All others patient records
allowing a dispute of this sort to become public should be removed to storage. Furthermore, if
via a trial. There are a variety of arbitration the purchaser ever deems it appropriate to dis-
forms available these days and your legal advisor pose of records that were active when the pur-
<<
Article
>> Home | TOC |
Index
248 Randall K. Berning
chase was made, notice to the seller should be tice, and can be down right picky about accept-
given coupled with a window of time during ing a stated value and negotiating terms than at
which the records can be secured before their any time in past. Believing otherwise tends to
destruction. protract the negotiation and document prepara-
tion period. In short, advance planning, having
a thoughtful approach to meeting both parties'
Conclusion needs, and the securing of competent advisors
It takes a good deal of advanced planning, con- can result in a wonderful and rewarding experi-
sidering all of the choices sellers have to make, ence for both sellers and buyers.
for the transition of one's orthodontic practice
to be well thought out and executed wisely.
Without taking adequate time to plan, transi- References
tioning a practice can be a harrowing experi-
1. Berning RK, Domer LR: "Guide to Orthodontic Practice
ence and a rough road to travel. Far too often Valuation for Residents and Practitioners", Practice Tran-
over my 22 years of facilitating transitions, I have sition Alternatives for Owners, American Association of
encountered doctors who think the practice will Orthodontists, St. Louis MO, 1999
transfer with little problem and that they will be 2. For a discussion and illustration of various tax aspects for
the C corporation see: Schiff AM, Jakubowitz LM, Janes
hired back as employees or consultants for as
PE: Tax Ramifications of Selling a Dental Practice: Sole Propri-
long as they desire. One common problem is the etorship, Partnership or Corporation, The Expert Series for
seller's perception that any buyer should be able Dentists™, Berning & Affiliates, Inc., Naples FL, pp 12-14,
to see and appreciate the value of his/her prac- 2002
tice and that any young professional will agree to 3. Berning RK: "Psychology of the Deal", Practice Transition
almost any terms just to be a part of the best Alternatives for Owners, American Association of Orthodon-
tists, St. Louis MO, 1999
practice around. It often comes as a rude awak- 4. Berning RK: "Common Mistakes to Avoid", Practice Tran-
ening that buyers today are better prepared, are sition Alternatives for Owners, American Association of
more capable of assessing the tenor of a prac- Orthodontists, St. Louis MO, 1999
<<
Article
>> Home | TOC |
Index
must carefully build on that personal interest to will be interviewing them to determine if this is
develop, as noted later, the practice's financial the type of person that you want to forge a
and valuation information along with the nego- personal and business relationship with or
tiated terms that will allow for a fully developed whether this person is a "my way or the highway"
business transaction relationship to be finalized. individual or group. Discuss treatment philoso-
Others may have an interest in a particular phy and the protocol for finishing cases. Discuss
style of practice, a large versus a small practice, management style and philosophy. Do you share
one based on a particular philosophy of care, or these common philosophies? Evaluate the per-
some other desired practice profile. To the de- sonality and character of the doctor or the
gree that a younger professional can elucidate group. Is this a mentoring kind of person or
the practice style or profile he/she seeks and group, or are you expected to be on your own
then proactively search for just that format, the immediately after the transaction? Evaluate your
more the likelihood that it will be found. Most personality and character. Do you perceive there
younger professionals fall somewhere in be- to be a good fit of personality and character
tween and at times seem unclear about exactly between the parties?
what they would like in a practice. In short, the Starting a relationship. If you are asked to be-
more specificity regarding the attributes of a gin with the practice as an associate, request a
practice and or practitioner that a younger pro- draft of an employment agreement. Do not start
fessional can identify ahead of time when seek- working in the practice until the agreement is
ing a practice opportunity, the better. 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
Following A Consistent Process understanding is seeing if you could explain the
contractual provisions to your friend or signifi-
Qualifying An Opportunity
cant other? Is the compensation and benefits
If a younger professional is actively searching for package clearly defined? Has the agreement
practice opportunities, there will be a number of been reviewed to ensure there are no ambiguous
interested practice owners open to discussing provisions? Has your risk exposure been evalu-
their practice and the expectations that they ated? Your advisors should be able to conclude
have in mind. However, unless the buyer uses a that the agreement is both fair and balanced,
consistent series of steps or a checklist to follow thus helping to ensure that the relationship gets
when qualifying various opportunities, he/she started on the right track.
can quickly miss key items in evaluating and Buyin or Buyout versus an outright sale. If the
comparing opportunities. The following brief associateship is likely to lead to a buyin/buyout
listing is prepared to assist prospective buyers. or an immediate sale with a potential hire back,
Buyers should feel free to tailor it to their own ask and get a commitment as to when you will
needs. It is important to bear in mind that in meet with the owner or partners to discuss
many cases a buyer is on the receiving end of a terms. Often, the terms for the transition will be
thought process that an owner or group practice agreed to in the initial documents requiring
has been considering for some time. Buyers of- only a triggering event such as a period of time,
ten have less time to consider each offer and achieving a certain gross or production, and so
therefore the task becomes one of evaluating the on to conclude the transaction. A contract's
opportunity in a timely manner and, as may be terms can be thought of as the core items relat-
needed, adding provisions that balance the ing to building an equity relationship. For exam-
terms between the parties resulting in meeting ple, discuss what will be the format for valuation,
both the buyer's and the seller's particular how will the price be paid, and what happens in
needs. the event of disability or death. In my view, the
Key questions. Start by qualifying the doctor/ process of developing the contract's terms shows
owner/partner. It can be helpful to have several compatibility, decision-making ability, and a
meetings with the practice owner or the doctors shared interest between the parties. It can be
comprising a group practice. Visit their offices helpful to secure the services of a facilitator to
and spend more than 1 or 2 hours. In effect, you raise key terms for both parties.
<<
Article
>> Home | TOC |
Index
For any proposed equity purchase-related how billings and collections are to be managed
transaction, get the practice valuation prepared and by whom; (4) the office policies and proce-
by a third-party professional valuation firm. One dures of the practice to be followed including
unbiased approach is to have the firm apply the the requirements concerning record keeping
capitalized earnings method as described in the must be made clear; (5) the confidential nature
Guide to Orthodontic Practice Valuation for Residents of information that an employee is exposed to
and Practitioners in the American Association of while working at the practice; (6) any prohibi-
Orthodontists Practice Alternatives Program tions on soliciting patients or referral sources
binder.2 and any covenant not to compete; (7) the exact
In certain situations, the offer may be for a compensation package, its format relating to tax
very short period of associateship or an immedi- status, an understanding of any benefits relating
ate purchase and sale, either containing provi- to payments for health or professional liability
sions for a hire back. In such a case, you will insurance, along with a clearly defined vacation
need to move quickly past vague statements of policy; and (8) the terms for termination.
how the deal will be structured to more solid All of these should be part of a well-prepared
terms and financial information. Ask for a associate-employee agreement.
binder or packet of materials that contain at
least the following information: first, a third- If a Buyin/Buyout or a Transitional or an
party practice valuation as per the Guide to Orth- Immediate Sale What Information Is
odontic Practice Valuation for Residents and Practi- Needed?
tioners as noted earlier or an indication of when
There are two groups of important information
it will be prepared. Second, a cash flow projec-
for a prospective partner or buyer to receive and
tion showing, without guaranteeing, that you
evaluate. The first group is nuts and bolts ori-
can earn a comfortable living and pay the debt
ented and often represents information deemed
load based on this valuation. Third, a list of the
important by the doctors. The second are struc-
terms or the name of the professional who will
tural such as legal and tax aspects and are im-
be helping you and the practice owner or group
portant to the advisor(s). The first grouping
develop the terms. Next, be sure to check the
varies widely by doctor and practice transition. A
background of your advisor(s). Finally, you must
minichecklist might include such items as the
have reference material on the practice such as
following.
financial and tax information. You may be asked
to sign a confidentiality agreement before being 1. What is the number of patients in treatment,
given this information; make sure that this in retention, and on observation; has this
agreement is also reviewed. number increased or decreased dramatically,
and what is the mechanism to transfer them
most effectively?
Buyer's Fundamental Considerations 2. Who are the practice's referrers, ranked from
most favorable to least, and how is each re-
If Entering Into an Associate Period, for
ferral group to be oriented to the new prac-
How Long, How Are You Paid, and Are
titioner?
There Any Restrictions?
3. Who on the staff is effective and likely to
Because many practice transitions occur with a continue with the practice, and what is the
preliminary associateship period, it can be help- best way to ensure a smooth transition of
ful to have a brief list of items to discuss between staff?
the parties. Generally, the parties should discuss 4. At the point of an eventual buyout or at the
or develop a clear statement of the duties of the sale of the equity interest of the current
prospective employee. These items should in- owner, what is an appropriate set of restric-
clude but are not limited to (1) what treatment tions (nonsolicitation, covenant not to com-
and patient management standards the practice pete, and so on) to ensure that the good will
maintains and what is expected from the em- of the practice transfers to the new owner
ployee in this regard; (2) a clear understanding without threat?
concerning the days and hours to be worked; (3) 5. How should the practice location identity be
<<
Article
>> Home | TOC |
Index
maintained, with regard to extending the cur- nated with all accounts written off for the 3
rent lease; if there is ownership of the real previous years.
estate ownership by the seller should there be 4. The practice valuation coupled with support-
an option to buy, right of first refusal, and ing patient, practice, and financial informa-
method to arrive at value of the real estate tion including clear title to all assets refer-
negotiated in tandem with the practice equity enced.
purchase? 5. Tax considerations relating to asset, stock, or
6. If there is to be a period of partnership be- partnership form of transaction.
fore a buyout, what details should be dis- 6. Liabilities being assumed, if any, and, pend-
cussed regarding a buy back of either party's ing litigation, if any.
interest in the event of either a permanent or 7. A proposed covenant not to compete and any
temporary disability or the death of either nonsolicitation provisions.
party? 8. The representations and warranties of the
7. If there is to be a period of partnership, what seller. This is one of the most important areas
provisions will provide for effective dispute on which buyers should focus. These are as-
resolution in the event the parties reach an surances that a seller has, in effect, told you
impasse on a critical practice, patient, staff, that you can rely on regarding all of the
financial, or tax issue? statements or information provided.
8. If there is to be a period of partnership be-
fore a buyout, what provisions will be negoti- Conclusion
ated now regarding decision making author-
ity relating to the ongoing affairs of the It takes a nimble professional to move quickly
practice during the partnership period? from an academic environment or an unsatisfac-
tory current associateship to a new practice re-
The second grouping is more standardized in lationship and to have met the challenge of
that various experienced advisors should be ca- doing so in a business like fashion, informed,
pable of guiding each party through the consid- well advised, and conversant with all available
erations relating to the issues noted later. This options. To do otherwise, however, is to risk a
checklist should be discussed and amended with transaction that is out of financial, legal, or prac-
the advisors that a prospective associate, partner, tice balance. Once in a binding relationship, it is
or buyer selects and should then be tailored to difficult to undo it. Just as for cautious sellers,
the particular transaction under consideration. advanced planning with an emphasis on looking
In all proposed equity interest or business trans- for a well thought-out approach with supporting
fers, it is important to thoroughly understand financial, valuation, and material terms will of-
the underlying practice. An in-depth investiga- ten make all the difference to the buyer. There
tion of the seller's practice and business prac- is too much at stake with a career and lifestyle in
tices must be completed. This is often referred the balance to do otherwise. Too often over my
to as a buyer doing his/her due diligence, and 22 years of facilitating transitions, I have encoun-
no young professional should feel reticent to ask tered younger professionals who rely on emo-
for and review any piece of financial, practice, tion and the "feel-good" impressions of a seller.
and tax information relating to the practice. It Today that is not enough. This is the time to do
can be helpful to secure the following items your homework, and, if you do, in the vast num-
among others. ber of instances, the transition will be fine.
1. The practice's financial statement along with
federal and state income tax returns for the 3 References
previous years. 1. Berning RK: Evaluating employment agreements, in A Guide for
2. The balance sheet and income statements for New Dental Practitioners. California Dental Association, Commit-
tee on the New Dental Professional, Sacramento, CA, 1997
the 3 previous years.
2. Berning RK, Domer LR: Guide to orthodontic practice
3. The schedule of production and collections valuation for residents and practitioners, in Practice Alter-
for the practice with contracts receivable and natives Program. American Association of Orthodontists,
accounts receivable broken out and coordi- 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
Posterior bite blocks, for hyperdivergent open Risk management and legal concerns
bite, 134-135 dentist-patient relationship and, 187-197
Potassium ion channels, activation of, in strain, 8 dismissal of patient, 188-189, 200, 203
Power fee collection, 198-204
in hypothesis testing, 70 governmental regulations, 238-242
post hoc analysis of, 75-76 informed consent. See Informed consent
to reject false null hypothesis, 57 lawsuits
Practice management, risk management in. See defendant's view of, 228-223
Risk management and legal concerns expert witness role in. See Expert witnesses
Predictive orthodontics, 2-5 for fee collection, 203
Price, of orthodontic practice, for transitioning, plaintiff’s view of, 220-233
244 procedure for, 216-219
Probability level, in statistics, 69-70, 82 reasons for, 210-215
Proportions, comparison of, sample size for, liability, 205-209
72-73 administrative, 238-242
Protraction device, maxillary, for anterior open standard of care. See Standard of care
bite in excessive vertical dimension, transitioning of practice
151-152 buyer's perspective of, 249-252
seller's perspective of, 243-248
Qualitative data, distribution of, 62
Quantitative data, distribution of, 62
Sagittal split osteotomy, bilateral
Range, in statistics, 80 for anterior open bite, orthodontic therapy
Rank, in nonparametric statistics, 81-82 with, 167-169
RANKL (receptor activator of nuclear factor-KB rigid vs. wire fixation of, skeletal and dental
ligand), in osteoclast differentiation, 7 changes in, 29-34
Raw data, distribution of, 63 Samples and sampling, in statistics, 55-56
Receiver operating characteristic curves, 98-100 bivariate, 89
Receptor activator of nuclear factor-KB ligand size of, 67-76
(RANKL), in osteoclast differentiation, 7 adjustments to, 75
Records. See Documentation calculation of, 67-68, 72-75
Referrals, liability in, 207-208 clinical relevance of, 71
Regression, in statistics, 59, 87-91 for comparing continuous measures, 73
Regulations, governmental, 238-242 for comparing proportions, 72-73
Relationships for comparing time-to-event measure, 73-74
dentist-patient. See Dentist-patient relation- hypothesis testing and, 68-71
ship post hoc analysis of, 75-76
in statistics variability and, 71-72
form of, 89-91 vs. generalizability, 104-105
significance of, 88-89 Scatter, in statistics, 56
Remodeling, periodontic, three-dimensional Schuchardt technique, for anterior open bite,
evaluation of, 23-28 orthodontic therapy with, 167-168
Representations, in practice transfer, 247 Screening tests, usefulness of, 92
Respondeat superior doctrine, 205-209 Selling, of orthodontic practice, 243-248
Rest, vertical dimension at, definition of, 155- Sensitivity and specificity, in statistics
156 accuracy and, 97-98
Reverse-L osteotomy, of mandible, for anterior agreement and, 97-98
open bite, orthodontic therapy with, diagnostic likelihood ratios and, 98
stability of, 167 receiver operating characteristic curves and,
Rigid fixation, of bilateral sagittal split osteot- 98-100
omy, skeletal and dental changes in, terminology of, 92-94
29-34 test characteristics and, 94-97
<<
Article
>> Home | TOC |
Index
215-238-8319
8. Complete Mailing Address of Headquarters or General Business Office of Publisher (Not printer)
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
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).
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