（
（
ºffl
H
-º-º: --------
##########################
iii:-
##############
ºffli
---
-- -
（
#:
*:::::::::::::::::::::::::
º
--- º #
ºffl ---
ºffli
------------- ---
---
#
-
-
（
º:
- #: ---
（ （
# º
#:
ºffl
#
ºffl
----------
º:
º
- ---
º:
------------
-º-º:
-
º
---
-------
ºffl
ºffl
ºffli
---
ºffl
º
- - ------
ºfflº -
-------- º

EARLY ORTHODONTIC TREATMENT:
IS THE BENEFIT WORTH THE BURDEN2
This volume includes the proceedings of the
Thirty-third Annual Moyers Symposium
February 25–26, 2006
Ann Arbor, Michigan
Editor
James A. McNamara, Jr.
Editorial Associate
Katherine A. Ribbens
Volume 44
Craniofacial Growth Series
Department of Orthodontics and Pediatric Dentistry
School of Dentistry
and
Center for Human Growth and Development
The University of Michigan
Ann Arbor, Michigan
C)2007 by the Department of Orthodontics and Pediatric Dentistry,
School of Dentistry and sº
Center for Human Growth and Development
The University of Michigan, Ann Arbor, MI 48109
Publisher’s Cataloguing in Publication Data
Department of Orthodontics and Pediatric Dentistry and
Center for Human Growth and Development
Craniofacial Growth Series
Early Orthodontic Treatment:
Is the Benefit Worth the Burden?
Volume 44
ISSN 0.162 7279
ISBN 0-929921-00-3
ISBN 0-929921–40–2
No part of this publication may be reproduced, stored in a retrieval system, or
transmitted in any form by any means, electronic, mechanical, photocopying, re-
cording, or otherwise, without the prior written permission of the Editor-in-Chief
of the Craniofacial Growth Series or designate.
CONTRIBUTORS
TIZIANO BACCETTI, Assistant Professor, Department of Orthodontics,
The University of Florence, Florence, Italy; Thomas M. Graber Visiting
Scholar, Department of Orthodontics and Pediatric Dentistry, School of
Dentistry, The University of Michigan, Ann Arbor, Michigan.
SHELDON BAUMRIND, Professor, Department of Orthodontics; Direc-
tor, Craniofacial Research Instrumentation Laboratory, Arthur A. Dugoni
School of Dentistry, University of the Pacific, San Francisco, California.
ROLF G. BEHRENTS, Executive Director and Program Director of Or-
thodontics, Center for Advanced Dental Education, St. Louis University,
St. Louis, Missouri.
LUCIA H.S. CEVIDANES, Department of Orthodontics, University of
North Carolina School of Dentistry, Chapel Hill, North Carolina.
PAOLA COZZA, Professor and Head, Department of Orthodontics, The
University of Rome “Tor Vergata,” Rome, Italy.
KURT FALTIN, Jr., Professor and Chair, Department of Orthodontics,
University Paulista, Sao Paulo, Brazil; Guest Professor, Facial Orthopedic
Department, University of Ulm, Germany.
LORENZO FRANCHI, Assistant Professor, Department of Orthodontics,
The University of Florence, Florence, Italy; Thomas M. Graber Visiting
Scholar, Department of Orthodontics and Pediatric Dentistry, School of
Dentistry, The University of Michigan, Ann Arbor, Michigan.
JAYNE HARRISON, Consultant Orthodontist, Liverpool University Den-
tal Hospital, Liverpool, UK.
HYEON-SHIK HWANG, Associate Professor and Chair, Department
of Orthodontics, College of Dentistry, Chonnam National University,
Kwangju, Korea.
LYSLE E. JOHNSTON, Jr., Professor Emeritus and former Chair, Depart-
ment of Orthodontics and Pediatric Dentistry, School of Dentistry, The
University of Michigan, Ann Arbor, Michigan.
CHUNG HOW KAU, Associate Professor, Department of Orthodontics,
University of Texas Health Science Center Dental Branch at Houston; Vis-
iting Senior Lecturer, Department of Dental Health and Biological Sci-
ences, School of Dentistry, Cardiff University, Wales, UK.
KEVIN KIEU, Private Practice of Orthodontics, Fountain Valley, Califor-
nia.
ETSUKO KONDO, Adjunct Associate Professor, Department of Ortho-
dontics, Nippon Dental University, Tokyo, and Tsurumi University, Kana-
gawa, Japan; Guest Professor, Beijing Stomatological Hospital, Capital
University, Beijing, Peoples Republic of China.
JAMES A. McNAMARA, Jr., Thomas M. and Doris Graber Endowed
Professor of Dentistry, Department of Orthodontics and Pediatric Den-
tistry, School of Dentistry; Professor of Cell and Developmental Biology,
School of Medicine; and Research Professor, Center for Human Growth
and Development, The University of Michigan, Ann Arbor, Michigan.
ALEXANDRE MOTTA, Professor, Department of Orthodontics, Univer-
sity of North Carolina, Chapel Hill, North Carolina.
PETER NGAN, Professor and Chair, Department of Orthodontics, West
Virginia University School of Dentistry, Morgantown, West Virginia.
KEVIN O’BRIEN, Dean, School of Dentistry; Professor, Department of
Orthodontics, School of Dentistry, University of Manchester, Manchester,
UK.
CEIB PHILLIPS, Professor, Department of Orthodontics, University of
North Carolina, Chapel Hill, North Carolina.
STEPHEN RICHMOND, Professor and Head, Department of Orthodon-
tics, Dental Health and Biological Sciences, School of Dentistry, Cardiff
University, Wales, UK.
FRANK STAHL, Research Associate, Department of Orthodontics, The
University of Rostock, Rostock, Germany; Post-doctoral Scholar; Depart-
ment of Orthodontics and Pediatric Dentistry, School of Dentistry, The
University of Michigan, Ann Arbor, Michigan.
MARTIN STYNER, Research Assistant Professor, Departments of Com-
puter Science and Psychiatry, Neurodevelopmental Disorders Research
Center, University of North Carolina at Chapel Hill, North Carolina.
PATRICK K. TURLEY, Private Practice of Orthodontics, Santa Monica,
California.
JAMES L. VADEN, Professor and Director, Department of Orthodon-
tics, College of Dentistry, University of Tennessee Health Science Center,
Memphis, Tennessee.
JUHA VARRELA, Professor and Head, Department of Oral Development
and Orthodontics, Institute of Dentistry, University of Turku, Turku, Fin-
land.
KATHERINE W.L. VIG, Emeritus Professor and Interim Head, Depart-
ment of Orthodontics, The Ohio State University College of Dentistry,
Columbus, Ohio.
PREFACE
The focus of the 2006 Moyers Symposium was early orthodontic
therapy, i.e., treatment provided to a patient during the mixed dentition
and occasionally during the late deciduous dentition. The presumed goal
of such treatment is to correct existing or developing skeletal, dentoalveo-
lar, and muscular imbalances in order to improve the environment before
the eruption of the permanent dentition is complete. Proponents of early
treatment maintain that by initiating orthodontic and orthopedic therapy
at this young age, the need for complex orthodontic treatment involving
permanent tooth extraction or orthognathic surgery is minimized. Is this
statement accurate? In an era of evidence-based orthodontics, what clini-
cal studies support or refute these claims?
In recent years, early orthodontic and orthopedic treatment has
become increasingly controversial, not only among orthodontists, but also
with the lay public. Articles on this subject have appeared in such promi-
nent lay publications as the New York Times, The Wall Street Journal, and
US News and World Report. Is two-phase treatment any better than treat-
ment in one phase? Is it better to manage Class II and Class III patients in
the early mixed dentition or later after the deciduous teeth are lost? Are
there any markers that can be used to determine the “best time” for treating
Specific types of malocclusion? What scientific evidence supports these
claims?
These and other such questions were addressed during the 33*An-
nual Moyers Symposium, which was held at The University of Michigan
on Saturday, February 25, and Sunday, February 26, 2006. The Moyers
Symposium has had a long tradition of dealing with the most current top-
ics in Orthodontics and craniofacial biology, and this meeting was no ex-
ception. We invited eight internationally known clinicians and scientists
to provide data on this controversial subject.
In addition, the 32” Annual International Conference on Cranio-
facial Research (the so-called “Presymposium”) was held on the Friday
before the Symposium (February 24") in the Ballroom of the Michigan
League. The presymposium conference on Friday featured papers rele-
Vant to orthodontics and craniofacial biology presented by an international
group of investigators. Many of these papers were relevant to the main
topic of the 2006 Symposium on early treatment and are included in this
Volume.
As in previous years, the Symposium honors Dr. Robert E. Moy-
ers, Professor Emeritus of Dentistry and Fellow Emeritus and Founding
Director of the Center for Human Growth and Development. The meet-
ing is co-sponsored by the School of Dentistry and the Center for Human
Growth and Development.
As in past years, Katherine A. Ribbens has facilitated the publi-
cation of this book through her tireless editing, manipulating a variety of
figure formats in Adobe Photoshop(R), interacting with the authors, and
formatting the layout of the book using Adobe InDesign CS2(R) software.
Kathy has provided the orthodontic specialty with a fine product again this
year through the publication of this volume.
We also would like to thank Debbie Montague, Michelle Jones,
and Karel Barton from the Office of Continuing Dental Education for or-
ganizing and running the Presymposium and Symposium in such an ef-
ficient fashion. Further, we thank Dr. Sunil Kapila, who, as Chair of the
Department of Orthodontics and Pediatric Dentistry, has provided the fi-
nancial resources to underwrite partially the publication of this book.
Finally, we thank the participants of the Symposium and Presym-
posium and those that buy the volumes of the Craniofacial Growth Series,
without whose support none of this would exist. All volumes still in print
are available at www.needhampress.com.
James A. McNamara, Jr.
Ann Arbor, Michigan
November, 2006
FRIENDS OF THE SYMPOSIUM
Dr. Edward D. Bayleran
Dr. Chester S. Handelman
Dr. Robert J. Isaacson
Dr. Etsuko Kondo
Dr. J.H. Kotze
Dr. William A. Patchak
Dr. Dean B. Sommerfield
Dr. Terry Tippin
Dr. Katherine W.L. Vig
TABLE OF CONTENTS
Contributors
Preface
Friends of the Symposium
Foreword: The Decision of When to Intervene: The Nature
of the Question in Terms of Faith, Passion and Evidence
Rolf G. Behrents
Early Orthodontic and Orthopedic Treatment
The Search for Evidence: Will It Influence Clinical Practice?
Katherine W.L. Vig, Kevin O'Brien, Jayne Harrison
If Wishes Were Horses
Lysle E. Johnston, Jr.
Another Perspective on Early Orthodontic Treatment
Sheldon Baumrind
Effectiveness and Timing of Mandibular Changes Produced by
Functional Jaw Orthopedics: A Systematic Review
Lorenzo Franchi, Tiziano Baccetti, Frank Stahl,
Paola Cozza
A Collateral Note on the Clinical Applicability of the Outcomes
Of RCTs in Orthodontics
Tiziano Baccetti, Lorenzo Franchi
Clinical Evidence in Treatment of Class II, Division 1 Maloc-
clusion with the Balters Bionator: Long-term Effective-
ness, Treatment Timing and Post-treatment Changes
Kurt Faltin, Jr.
The Long-term Perspective on Orthopedic Treatment of Class III
Malocclusion -
Tiziano Baccetti, Lorenzo Franchi
Facemask Therapy for Class III Malocclusion: Short-term and
Long-term Outcomes
Patrick K. Turley
Early Class III Treatment: Is the Benefit Worth the Burden?
Peter Ngan
13
39
53
67
81
87
105
117
135
Short-term and Long-term Stability of Changes in the Trans-
verse Dimension: Is Early Expansion Worth the Effort?
James A. McNamara, Jr.
The Tooth/Arch Discrepancy Problem: Is the Benefit of Early
Orthodontic Treatment Worth the Burden?
James L. Vaden
Responses of the Craniofacial Complex to Forced Mandibular
Protrusion with the Herbst Appliance: A Cephalometric Study
in Macaca Mulatta
Kevin N. Kieu, James A. McNamara, Jr.
Follow-up Study: Long-term Occlusal Stability in 120 Cases—
Proposed Treatment Goals
Etsuko Kondo
Orthodontics in the Early Mixed Dentition: Effectiveness and
Efficiency of the Eruption Guidance Appliance
Juha Varrela
A New Classification of Facial Asymmetry
Hyeon-Shik Hwang
Evaluation of Facial Growth: A 3D Soft Tissue Perspective
Chung How Kau, Stephen Richmond
3D Imaging for Early Diagnosis and Assessment of Treatment
Response
Lucia H.S. Cevidanes, Alexandre Motta,
Martin Styner, Ceib Phillips
147
| 63
189
223
257
269
295
305
FOREWORD
THE DECISION OF WHEN TO INTERVENE:
THE NATURE OF THE QUESTION IN TERMS OF
FAITH, PASSION AND EVIDENCE
Rolf G. Behrents
As in the past, the 33rd Annual Moyers Symposium addressed a contem-
porary issue by creating an interplay of thoughts, experiences, and knowl-
edge of scientists and clinicians. This year, the balanced consideration
of art and science focused on early orthodontic treatment and the refined
question of whether the benefits of early treatment are worth the effort
involved.
This is a reasonable and useful question in terms of scientific en-
deavor and clinical practice. The answer, however, is not determined eas-
ily. While the underpinnings of science involve the orderly search for and
discovery of truth, the clinical practice of orthodontics is not so simply
conceived and structured. Practitioners must decide what the appropriate
treatment protocol is when presented with patients; they cannot wait until
all that should be known about a given condition is known. Consequently,
practitioners often make decisions based on a mixture of knowledge, train-
ing, experience, belief and ethics. Such decisions, therefore, usually are
made on the basis of some form of well-meaning, reasoned uncertainty.
This is referred to as clinical judgment and, as such, often is the mother of
stimulus for invention. -
When adopting a new device or approach, practitioners do not
know what exact outcome to expect; they even may not know why or
how a specific outcome was achieved. If treatment is successful in terms
of both doctor and patient satisfaction, practitioners may adopt the new
treatment strategy and employ it in similar situations with great enthusi-
asm and confidence. When questioned as to the validity of a new treat-
ment approach, practitioners appear to be offended. Some believe that
there is no need to evaluate alternatives in the face of strong beliefs about
the relative effectiveness of a specific form of treatment. If questioned
further, proponents are quick to defend their decisions and approaches.
If, however, they fail to provide substantive and convincing evidence for
their decisions and actions, they often claim, with great passion, that their
innovation represents invention or “art” and that those who cannot du-
plicate their results have treated poorly (i.e., the treatment is “technique
Foreword
sensitive”) or simply performed treatment at the wrong time and/or on
the wrong patient. If the new device or strategy is adopted by others, this
“innovation” often is accepted on the basis of faith rather than on facts
derived from scientific analysis. Unfortunately, strongly held beliefs are
an extremely important determinant in the selection of treatments, even in
the absence of any supporting evidence.
In this context, the “art of orthodontics” refers to differences in the
clinical judgments of individual orthodontists concerning individual pa-
tients. That these differences in the way orthodontics is practiced exist is
widely acknowledged, but such variation often is not considered much of
a problem. Some of this practice variation stems from the way orthodon-
tists are trained, i.e., many different clinical approaches are taught to treat
the same condition, and some from the way that orthodontics is practiced,
i.e., independently and in isolation which does not foster comparison of
treatments and outcomes. However, most of this acceptance of and lack of
concern for the differences in clinical judgments stems from the “techni-
cal orientation” of orthodontics. This technical orientation is an extension
of a craftsmanship mentality in which the “quality” of service is defined
more by its technical perfection than by its success in resolving patient
problems. Simply stated, those who have “better hands” and who can
place teeth in strictly defined (and measured) positions are considered the
best practitioners. The patients themselves are not so strictly considered
and measured; they often are characterized by their treatment rather than
their condition. The outcome of treatment usually is assumed or casually
judged on the basis of patient “happiness” on the day the teeth are granted
freedom by the doctor.
So, is the presence and acceptance of the “art of orthodontics” a
problem? Should variation among practices be accepted or be considered
a menace? The current symposium speaks to this issue and rightly so, for
while one view might be that these differences in clinical judgments are
reasonable, another view might be that these differences identify aspects
of orthodontic practice for which there is uncertainty or disagreement con-
cerning the most effective approach to treatment. For example, if two
orthodontists consistently choose a different set of treatment procedures
for patients with similar conditions, then one orthodontist must be provid-
ing less effective or efficient care than the other, unless the care leads to
equivalent results for their patients. Similarly, if two orthodontists attain
equivalent results, but one practitioner’s treatment takes longer and costs
more, the more efficient and less costly method should be employed by
both. However, if a procedure costs more money and time but produces
greater benefits, then it should be used.
Behrents
Given that there are substantial differences among practitioners
in treating the same orthodontic conditions and given that the differenc-
es are due to variations in the identification of conditions, in decisions
to treat these conditions, and in treatment approaches, the acceptance of
these variations among practitioners is not always productive. To that end,
scientific inquiry can provide valuable information, and this symposium
presents us with an opportunity to move forward in the search for better
science for our specialty, better treatment choices for our doctors, and bet-
ter care of our patients.
Given the biases of contemporary practice, numerous appliances,
protocols, and strategies have been developed. Some “early treatment”
ideas are fairly well accepted (e.g., treatment to correct anterior or poste-
rior crossbites, habit therapy, etc.), but other treatment protocols remain
controversial. Of central concern are those treatments that alter the jaws
or dental arches and “tip the balance” so that “borderline” cases (and even
those across the border) might be treated successfully without the extrac-
tion of teeth or surgery. To be sure, some treatment approaches are more
efficient and effective than others. In this regard, each speaker at this sym-
posium told us what they know about efficiencies of various treatments in
an attempt to address the issue of which treatment, if any, is the “correct”
treatment for a given clinical situation.
Dr. Kate Vig provided an elegant and informative beginning. Fol-
lowing a tribute to Dr. Robert Moyers, she raised a number of important
clinical and research questions and spoke of the supposed benefits and
drawbacks of early orthodontic and orthopedic treatment. She also re-
viewed some of the findings of The Institute of Medicine Report of 1996
that calls for evidence-based and patient-centered care. Eliminating un-
necessary and ineffective treatment interventions will be at the heart of the
future of orthodontic practice.
Dr. Vig also discussed the need for and value of various forms
of evidence and their strengths. This included the ascending hierarchy
of case reports, case control studies, cohort studies, randomized clini-
cal trials, systematic reviews, and meta analysis. She further suggested
how the search for evidence should be conducted and valued. She also
explained the concepts and methodology of the Cochrane Collabora-
tion who use systematic reviews and meta-analysis techniques to “plow
the literature” to determine whether sufficient evidence exists to accept
or deny the truth of specific research questions. She then thoroughly
reviewed the classic randomized clinical trials (i.e., such as those from
the University of North Carolina, the University of Pennsylvania and the
University of Florida) and other studies that have formed much of the
Foreword
basis for our present understanding of one- and two-phase treatments.
Most importantly, she indicated that the combined message of the quality
studies presently available in the literature suggests that the evidence sup-
porting the modification of jaw growth in Class II treatments is weak.
Dr. Lysle Johnston, one of the most influential critics and analysts
of his generation, offered a summary of research findings and insightful
commentary on controversies that surround the subject of early orthodon-
tic and orthopedic treatments. He focused on the arguments for and against
one- and two-phase treatments. He pointed out the importance of looking
at both the patient and the doctor when considering the “benefits” of any
particular treatment. He also provided an informative lesson on decision
theory in terms of the value placed on straight teeth by patients, the risks
associated with treatment, pressures that are placed on the orthodontist by
referring dentists, and the assumed advantages of various appliances. He
pointed out that a disconnect between utility and need can cause a conflict
of interest and values.
Regarding rapid maxillary expansion, Dr. Johnston suggested that
a large amount of basal expansion only creates a relatively small increase
in arch perimeter resulting in the need for a great deal of arch expansion
(perhaps much more than is generally thought) to solve significant crowd-
ing problems. He pointed out that it also would take a great deal of ex-
pansion of the dental arches to solve significant crowding problems in the
mandible. Dr. Johnston believes that in either case there is little, if any,
evidence to support the conclusion that large increases in arch perimeter
can be gained or that these treatment effects are stable over the long term.
In commenting on brackets that “grow bone,” Dr. Johnston be-
lieves that there is little theoretical foundation or actual evidence to sug-
gest that one bracket produces more growth than any other. In support of
extraction therapy, he pointed out that when teeth are removed in patients
who present with significant crowding and/or protrusion, the face is not
adversely affected to any significant degree.
With regard to the ability of so-called functional appliances to
“grow” mandibles in an amount that is useful, Dr. Johnston maintains that
there is little reason to believe that this occurs. He suggests that the classi-
cal randomized clinical trials that looked at this question have been unfair-
ly criticized perhaps because they show that the change that is produced by
a functional appliance is so meager as to be inconsequential. This being
the case, the question of effectiveness is replaced by that of efficiency.
Behrents
Dr. Lorenzo Franchi described an extensive, systematic review
of the literature on the changes produced by functional appliances in the
treatment of Class II malocclusions (Cozza et al., 2006). In all, he and
colleagues reviewed 725 articles looking for evidence that the mandible
grows more during treatment of Class II malocclusion with functional ap-
pliances than in untreated Class II controls. Of the group of 725 articles,
they concluded that only 22 were informative.
The amount of supplementary mandibular growth seen in the treat-
ed cases varied widely (0.5 mm to 6.5 mm) compared to that of untreated
Class II controls. However, 20 of the 32 subjects described in the 22 stud-
ies did have outcomes that exceeded 2.0 mm of supplementary mandibular
growth at the completion of active treatment. In the final analysis, Dr.
Franchi suggested that most of the available strategies for treating Class
II malocclusions (i.e., functional appliances, Class II elastics, facebows,
molar distalizing appliances) do achieve the correction desired, albeit by
differing routes and amounts."
While this survey is interesting, it is clear that the criteria for in-
clusion of the specific 22 studies could be challenged with alternate crite-
ria producing different results and interpretations. For example, none of
the classical randomized clinical trials reported an effective change (i.e.,
greater than 2.0 mm) in mandibular length induced by functional appli-
ances. Similarly, comparisons with other untreated controls (e.g., Class I)
also might be informative.
Dr. Franchi also looked at supplementary mandibular growth rela-
tive to the purbertal growth spurt using biological markers, specifically
those of the maturing morphology of the cervical vertebrae. Citing his
and his colleagues studies and the work of others, Dr. Franchi indicated
that treatment performed during the peak of the growth spurt appears to
produce the most supplemental growth. In noting this relationship, Dr.
Franchi suggested that by timing treatment appropriately, two-phase treat-
ment could be collapsed into one-phase treatment by removing the in-
tervening retention phase. It should be noted, however, that while treat-
ment that is supposed to affect growth could be anticipated to show its
greatest effect when growth is greatest (i.e., at the peak), this response
might best be considered more normal than supplementary, unless clear-
ly demonstrated that it is otherwise.” Moreover, it must be recalled that
"This perhaps illustrates that Mother Nature has a sense of humor when it comes to orth-
odontic treatment or that we really do not understand how our appliances work.
*If this caution is not exercised, a practitioner might fantasize that his “appliance” not only
makes the jaws bigger, but it also makes the patient taller.
Foreword
growth is really represented by the accumulated area under the growth
curve rather than by the line of curve itself, which represents amount per
unit time. Thus, growth/appliance efficiency are perhaps greatest at or
near the peak; at other times the efficiency should be less thus requiring
longer treatment time. Not to belabor the point, but it is theoretically pos-
sible to fabricate a construct involving a fixed amount of expected growth
(as might occur when the mandible lengthens) in tandem with roughly
equivalent amounts of time (as say might occur over approximately a year
of treatment) both pre- and post-peak. As a result, the question of treating
too early or too late is not as consequential as one might think.
Dr. Patrick Turley spoke on the short- and long-term outcomes of
Class III treatments, a subject that he has been studying for decades. His
excellent presentation described his ongoing work that considers impor-
tant clinical issues and asks and answers important questions. Dr. Turley
has reviewed various treatment protocols, treatment timing, treatment ef-
fects, the effects of post-treatment growth, and the long-term stability of
facemask therapy. The results of his analyses indicate that while Class III
patients with maxillary deficiencies would benefit from early treatment,
Class III patients with true mandibular prognathism might not.
Dr. Turley described an important addition to the literature: a pro-
spective, randomized clinical trial in which rapid palatal expansion was
combined with facemask therapy in an effort to effect greater orthopedic
change (Vaughn et al., 2005). He suggested that while this protocol might
be useful for a patient in treatment who needs rapid palatal expansion to
address transverse discrepancies or crowding, it has not been demonstrat-
ed yet that the “combination” treatment would produce additional anterior
orthopedic improvement. In terms of the timing of treatment, both early
and later (but still “early”) treatments were reasonably effective, and sta-
bility of treatment results appeared to be long term. The conclusion drawn,
therefore, is that facemask therapy is an effective method for correction of
Class III malocclusion in growing children.
Dr. Tiziano Baccetti considered Class III treatments as well
(Baccetti et al., 2006), but he also focused on the treatment of openbite
malocclusions. He is to be congratulated for doing so, because so lit-
tle substantive and clinically helpful information exists concerning the
correction of vertical discrepancies. With regard to openbite malocclu-
sion, Dr. Baccetti and colleagues systematically reviewed the literature
on openbite malocclusion and identified seven articles out of a possible
1,049 that were instructive (Cozza et al., 2005). After a detailed study,
they concluded that headgear and some functional appliances applied
Behrents
during the mixed dentition period showed some degree of success. Un-
fortunately, the evidence upon which such conclusions were based was
meager (i.e., two controlled clinical trials).
Drawing from a number of studies, Dr. Baccetti pointed out that
puberty affects the quality of response to treatment in patients with verti-
cal problems. As expected, Class III treatments are most effective when
carried out before or at the peak of the growth spurt, while treatment that
occurs later (i.e., during the late mixed dentition) is less effective (Fran-
chi et al., 2004). In such treatments, it is felt that the Cervical Vertebral
Maturation (CVM) method is helpful in timing treatment to produce the
maximum amount of growth alteration. Dr. Baccetti related these issues to
cases that were successfully and unsuccessfully treated with a rapid palatal
expander and facemask. Patients who were treated early were compared
to patients who were treated later. While some of the interpretations Dr.
Baccetti presented might step beyond the inferential testing described, it is
clear that the biological timing of Class III treatment (as opposed to timing
by age or basing the decision to treat merely on the basis of when the pa-
tient showed up at the office) is an important determinant of the efficiency
and success of treatment.
Based on his experience and a vast amount of clinical research,
Dr. James McNamara provided an extensive treatise on his approach to
the treatment of patients that require increases in the transverse dimen-
sion and the long-term stability of such changes. It is important to stress
that Dr. McNamara does not believe that all patients require these kinds
of increases and that treatment for some patients necessitates extractions.
Dr. McNamara outlined his protocols for comprehensive treatment us-
ing rapid maxillary expansion, the Schwarz appliance, and several other
forms of therapy that are designed to effect arch length preservation or
augmentation during the mixed dentition. He suggested that RME is par-
ticularly useful in correcting posterior crossbites, increasing arch length,
facilitating Class II and III correction, facilitating canine eruption, cor-
recting posterior cuspal inclinations, improving nasal breathing, and
“broadening the smile.” He estimated that a typical treatment produces
approximately 6 mm of increase in arch perimeter in the maxilla and 4
mm in the mandible. He pointed out that expansion occurring before
the growth peak produced more skeletal changes and that expansion oc-
curring after the peak produced more dental and alveolar changes. As a
caution, Dr. McNamara suggested that the stability of the attained arch
perimeter increase is still in question, but that current research is encour-
aging (Geran et al., 2006; O’Grady et al., 2006). Questions regarding the
Foreword
relationship between perimeter and crowding, “spontaneous” sagittal cor-
rection of Class II malocclusions, and the affect of various retention pro-
tocols also are still under scrutiny (Baccetti et al., 2001; McNamara et al.,
2003).
Dr. James Vaden concluded the formal presentations by describ-
ing his clinical approach to the management of tooth-size/arch size prob-
lems in the mixed and early permanent dentitions. Based on his extensive
clinical experience, he suggested that a good diagnosis usually will lead
to the most efficient treatment plan for the both the orthodontist and the
patient. Regarding stability, he believes that the dental arches should not
be expanded as relapse will occur if they are expanded. He suggested
that early treatment for Class III malocclusions can be useful in that early
treatment makes the patient more “treatable” later. He also suggested that
serial extraction is a useful tool in the treatment of tooth-size/arch length
problems. He indicated, however, that he was skeptical as to the value
of early Class II treatments. His guiding principles include adopting ap-
proaches that result in the shortest time in treatment, a reasonable fee, and
a lasting benefit for the patient.
Many issues were explored during this symposium, but to what
end? To a certain extent it feels as though we have been here before. It
appears that we still are arguing over whether functional appliances actu-
ally “work,” whether the dental arches should be expanded, and whether
one- or two-phase treatment is better. In a way this meeting was cast
as a debate, finding no resolution. Few seem willing to admit that their
strongly held beliefs might be wrong. Given that the topic at hand — early
orthodontic treatment — has been the focus of many recent publications
and meetings including a previous Moyers Symposium conducted in 2001,
one could ask why we should address this issue again. Fortunately there
is an answer.
In essence this symposium did not provide the definitive answers
that we desire (meetings like this seldom do), but it did provide a subtle
new direction. For a typical question such as whether a certain appli-
ance works or does not, the answer is typically yes, but this gathering
showed that we have moved beyond the basic level of inquiry to one
of refinement of our questions, our treatment approaches, and our treat-
ments. For example, to the simple question of do functional appliances
work, the simple answer is yes. Biological evidence does confirm that the
growth of the skeletal tissues of the craniofacial region can be modified
by amount (although often a small amount) and direction. Refinement
of the question, however, centers on the determination of whether these
appliances should be used, when and for how long they should be used,
8
Behrents
under what conditions they should be used, and, finally, what are reason-
able expectations for outcomes and stability.
These new directions require broader inquiry and new forms of
evidence. For example, the concept of the “appropriateness” of treatment
is a new concept and was a part of all the presentations at the current sym-
posium (i.e., is the benefit worth the burden?). This term suggests that
the expected benefit of any treatment exceeds the negative consequences
of the treatment by a sufficiently wide margin that the treatment is worth
doing. A recent study by Gust (2005) makes this point. He studied the
records of a talented practitioner who has followed an early treatment pro-
tocol for many years. This practitioner believes that there are clearly three
types of tooth size/arch length cases: those for which treatment should be
performed in one phase and require extraction, those for which treatment
should be performed in one phase but with no extractions, and those that
are “borderline” extraction cases. The practitioner developed a diagnostic
scheme to identify the type of patient and adopted a two-phase treatment
approach involving RME and lip bumper therapy that was designed to
move the “borderline” case during Phase I into a second phase of non-
extraction treatment.
Over time, however, the practitioner determined that the expected
result was successful in some patients, but not in others. The additional
early treatment phase did not reduce the need for treatment involving ex-
tractions, it increased treatment time for the patient, and it increased the
overall cost of care. Thus, for some patients, Phase I treatment was a
waste of time and money. As to the possible explanations for the failed
cases, there are several possibilities. The diagnostic scheme that was used
might have been incomplete or flawed in terms of its power to discriminate
between patient types, the patients may have responded poorly to treat-
ment, or the patients may have changed during treatment in ways that had
nothing to do with the treatment.
Gust studied 47 borderline pretreatment, treatment, and post-
treatment cases (i.e., approximately four years post-Phase I and before
Phase II) records. Statistical analysis identified the most important flaw
in this treatment strategy. The patients who needed extractions at the be-
ginning of Phase II had much more crowding at the beginning of Phase I
(7.32 mm) than did the patients who needed no extractions at the begin-
ning of Phase II (4.84 mm). What was surprising was that the crowd-
ing in question was maxillary crowding. The early treatment protocol
clearly was successful in altering the dental and skeletal patterns of all
patients, but the patients were different with regard to maxillary crowd-
ing at the beginning of Phase I. As might be expected, patients who had
Foreword
significantly different amounts of maxillary crowding but who were treat-
ed the same way in Phase I needed different Phase II treatments. At the
beginning of Phase II, the unsuccessful patients (those who needed extrac-
tions) had greater proclination and procumbency of the maxillary incisors,
a more constricted maxilla, and greater tooth-size/arch length discrepan-
cies in the mandible than they had at the beginning of Phase I treatment.
In addition, maxillary crowding increased during early treatment.
If success for the patient is receiving the correct treatment and
achieving the expected result, lack of success would be the patient under-
going an inappropriate treatment, wasting time and money, and still not
achieving the expected result. What then is the lesson of Gust’s study? In
the end, the early treatment approach worked, but the diagnostic scheme
needed some adjustment to achieve the most success for both practitioner
and patient. For example, in Gust’s sample if the borderline for maxillary
crowding was set at 6 mm, 21% of “borderline” patients would receive
unwarranted extractions and 15% of “borderline” patients would receive
unwarranted Phase I therapy. If it was set at 7 mm, 8.5% of “borderline”
patients would receive unwarranted extractions and 23% of “borderline”
patients would receive unwarranted Phase I therapy. Finally, if it was set
at 8 mm of maxillary TSALD, 4% of “borderline” patients would receive
unwarranted extractions and 32% of “borderline” patients would receive
unwarranted Phase I therapy. While such refinements are possible and
useful, it is also clear that further advancements are needed if the correct
treatment always is to be matched to the correct patient.
CONCLUSIONS
A story is told about Steve Jobs, the inventor and President of
Apple Computers, who introduced a new computer to the media during a
press conference. After a glorious visual and verbal description of the fea-
tures and capabilities of the new computer, a reporter posed the following
question: “Mr. Jobs, do you consider this new computer a breakthrough?”
After a slight moment of thought, Jobs responded, “No, this is a great
computer: it has a lot of new features and is better than anything currently
on the market, but it’s not a breakthrough. If I walked into my office one
day and my computer said, ‘Good morning Mr. Jobs, what would you like
to do today?' – that would be a breakthrough.”
Have we witnessed any breakthroughs with regard to early treat-
ment during the past few years? I would argue not. By imagination
10
Behrents
or accident we have not constructed an appliance that is so in tune with our
patients and their biology that a breakthrough appliance has been produced.
Such an appliance placed in the mouth of a Class II patient or in that of an
extremely crowded patient, who then misses many appointments but does
wear the appliance and who eventually returns for an appointment, would
be seen to have such an exuberant response that there would be immediate
need for aggressive Class III treatment or the closing of excess spacing
between the teeth. As far as I know, this has not happened yet.
Refinements of our present treatment methodologies are what is
presently available to us; these are needed and greatly desired. Through
consideration of clinical experience, clinical research, and structured anal-
yses of the literature, this symposium, as it reflects the present understand-
ing and direction of our specialty, will thus contribute to our advance-
ment through further refinements of our treatment protocols. Let us not
continue to quibble about a millimeter or two, for to do so is to miss the
point. Breakthroughs in treatment will not come from any form of care-
ful reasoning, reading, re-analysis of the present literature, or group think.
To believe that such mechanisms will produce the breakthroughs desired
is to wager that someone (even you or I) has overlooked the “answers”
to the important questions already contained in our literature. The real
breakthroughs will come from a better understanding of biology in the
form of predicable cell responses and controlled facial growth, and these
breakthroughs will be the result of thoughtful research that, hopefully, will
be presented at a future Moyers symposium.
REFERENCES
Baccetti T, Franchi L, Cameron CG, McNamara JA Jr. Treatment timing
for rapid palatal expansion. Angle Orthod 2001;71:343-350.
Baccetti T, Franchi L, Schulz SO, McNamara JA Jr. Treatment timing for
an orthopedic approach to patients with increased vertical dimension.
Am J Orthod Dentofacial Orthop 2006, in press.
Cozza P. Muceduro M, Baccetti T, Franchi L. Early orthodontic treatment
of skeletal open-bite malocclusion: A systematic review. Angle Or-
thod 2005;75:707-713.
Cozza P. Baccetti T, Franchi L., De Toffol L, McNamara JA Jr. Mandibular
changes produced by functional appliances in Class II malocclusion:
A systemic review. Am J Orthod Dentofacial Orthop 2006; 129:599.
e1-599.e12. Commentary by Johnston, LE Jr, 2006;129:e1-e4 and
author’s response 129:e4-e6.
11
Foreword
Franchi L, Baccetti T. McNamara JA Jr. Postpubertal assessment of treat-
ment timing for maxillary expansion and protraction therapy fol-
lowed by fixed appliances. Am J Orthod Dentofacial Orthop 2004;
126:555–568.
Geran RE, McNamara JA Jr, Baccetti T, Franchi L., Shapiro LM. Long-
term stability of rapid maxillary expansion in the mixed dentition. Am
J Orthod Dentofacial Orthop 2006, in press.
Gust JE. A comparative analysis of borderline extraction cases treated in
two phases utilizing rapid palatal expansion and lip bumper therapy.
Unpublished thesis, Saint Louis University, 2005.
IOM report, National Academy of Science 2003.
McNamara JA Jr, Baccetti T, Franchi L, Herberger TA. Rapid maxillary
expansion followed by fixed appliances: A long-term evaluation of
changes in arch dimensions. Angle Orthod 2003;73:344–353.
O’Grady PW, McNamara JA Jr, Franchi L, Baccetti T. Long-term sta-
bility of rapid maxillary expansion concurrent with Schwarz appli-
ance therapy in the mixed dentition. Am J Orthod Dentofacial Orthop
2006, in press.
Vaughn GA, Mason B, Moon H, Turley PK. The effects of maxillary
protraction therapy with or without rapid palatal expansion: A pro-
spective, randomized clinical trial. Am J Orthod Dentofacial Orthop
2005;121:299-309.
12
EARLY ORTHODONTIC AND
ORTHOPEDIC TREATMENT
The Search for Evidence:
Will It Influence Clinical Practice?
Katherine W. L. Vig
Kevin O'Brien
Jayne Harrison
“Nothing is known in our profession by guess; I do not
believe that from the first dawn of medical science to the
present moment, a single correct idea has ever emanated
from conjecture: it is right therefore that those who are
studying their profession should be aware that there is no
short road to knowledge.”
Sir Astley Paston Cooper*
The 33" Annual Moyers Symposium honors Dr. Robert E. Moyers, Pro-
fessor Emeritus of Dentistry and Fellow Emeritus and Founding Direc-
tor of the Center for Human Growth and Development. This 11" Moyers
Memorial Lecture recognizes the many contributions made by Dr. Moyers
to the orthodontic profession. He was the recipient of numerous awards
and was recognized at a young age for serving his country with distinc-
tion during World War II. He was appointed by the Office of Strategic
Services as the Chief Medical Liaison Officer in Greece during WW II
and became the most highly decorated dentist in the US Army having been
awarded the Bronze Star, the Legion of Merit, the Purple Heart, the Order
of the British Empire, and the Order of the Phoenix. In peacetime he was
the recipient of an honorary degree from the University of Thessaloniki,
Greece. His legacy to orthodontics lives on in the many students he taught
and mentored. His leadership in advancing our understanding of cranio-
facial growth and development was established in 1966 when he founded
the Center for Human Growth and Development. His many publications
are widely recognized and his textbook, Handbook of Orthodontics, has
reached its 4" edition and, after 40 years, still is being assigned as required
reading for students.
*Preface to Chapter 1, Handbook of Orthodontics, 4th ed. RE Moyers.
13
The Search for Evidence
The topic of the Symposium in 2006 was “Early Orthodontic
and Orthopedic Treatment: Is the Benefit Worth the Burden?” This was a
timely subject as early orthodontic and orthopedic treatment has become
increasingly controversial, not only among orthodontists, but also with the
lay public. The belief that early orthodontic or interceptive orthodontic
treatment will modify and redirect craniofacial growth has been debated
for more than a century, as have the benefits of such treatment.
The purpose of this chapter is to consider the evidence available
in the published scientific literature on the effectiveness of early orth-
odontic treatment, with particular attention paid to evidence derived from
systematic reviews. This methodology became popular in the 1990s and
is used widely in medicine, not only for the design of clinical trials, but
also for the evaluation of scientific literature by systematic reviews and
meta-analysis. This approach is becoming an established methodology in
dentistry, including orthodontics, for providing a basis for determining the
efficiency and effectiveness of early orthodontic treatment.
There still is considerable debate on the timing of orthodontic
treatment, and “blame” is apportioned equally to those who practice early
orthodontic treatment and those who do not: therefore, the available evi-
dence needs to be evaluated systematically to determine whether or not
the evidence exists to resolve this contentious issue. For example, current
clinical opinion informs us that there is little doubt that there are some
malocclusions such as anterior and posterior crossbites that would benefit
from early treatment. Nevertheless, concrete evidence is lacking, even for
this conclusion. Prior probability estimates of the outcome of a treatment
intervention would be an efficient and quantitative method of informing
patients of the likelihood of the risks and benefits of a procedure. Current-
ly, however, there is little high quality data available in orthodontics, and
clinical trials are expensive to conduct. Although the randomized clinical
trial (RCT) is considered the gold standard for gathering reliable evidence,
the proper use of clinical trials and their rigorous experimental designs
depends on the question(s) being asked. The need to generate evidence
to support our clinical decisions and beliefs is always a high priority for
clinicians (Greenberg, 1993).
THE CONCEPT OF EVIDENCE-BASED
CLINICAL PRACTICE
The concept of evidence-based clinical practice evolved from the
medical model, which was espoused by Sackett and colleagues (2000)
and which is based on the vision of Archie Cochrane (1972, 1989), who
14
Vig et al.
undertook some of the first ever clinical trials during WW II when he was
a prisoner of war (POW). He was assigned to look after 22,000 POWs
as their medical officer in Europe. His ability to speak fluent German and
Spanish was an advantage both in communicating with the prison guards
and with the multilingual groups of prisoners for whom he was providing
medical care. With the challenges of poor conditions, limited medical sup-
plies and severely sick and wounded men to care for, Cochrane observed
that some individuals would get better and others would not, irrespec-
tive of the medical treatment he provided. This was the basis of his early
clinical trials. Cochrane later applied the concept of randomization in
experimental design to his clinical trials, a concept that was developed
by the statistician Sir Austin Bradford Hill (1937, 1991) and introduced
to Cochrane when he was a medical student at Cambridge University at a
time when statistics still was in its infancy as a discipline.
EARLY ORTHODONTIC TREATMENT
CONTROVERSIES
Interceptive orthodontic treatment has been practiced for more
than a century with reported success. It has been taught and practiced by
clinicians who hold strong beliefs that they were providing the best treat-
ment for their patients. This is a classic finding in health care as most
controversies are characterized by opposing viewpoints based on strong
convictions, weak evidence and questionable theory. With the emerg-
ing 21" century, controversy concerning early orthodontic treatment has
escalated and resulted in two major symposia in North America. The
first was the International Symposium on Early Orthodontic Treatment
held in Phoenix, Arizona, in February of 2002. The interest was so in-
tense that a second international symposium on early orthodontic treat-
ment was held in January, 2005, in Las Vegas. A third symposium was
held at the University of Michigan in February, 2006, entitled Early
Orthodontic Treatment: Is the Benefit Worth the Burden? It is this sym-
posium that is the basis for this volume. The controversy surrounding
this aspect of orthodontic care has attracted the attention of funding agen-
cies including the National Institute of Dental and Craniofacial Research
(NIDCR) in the USA and the Medical Research Council (MRC) in Brit-
ain. These agencies have funded clinical trials the aims of which were
to provide evidence that could be incorporated into treatment guidelines.
This work was meant to resolve the dilemma of when to provide treat-
ment for a child with a malocclusion. Ultimately, the evidence from
such clinical trials should enable orthodontists and parents of children
15
The Search for Evidence
with malocclusions to decide whether (1) treatment should begin during
the mixed dentition stage followed by a second course of treatment when
the permanent dentition has erupted (two-phase treatment) or (2) treat-
ment should wait until the permanent dentition has erupted (one-phase).
The current controversy has been fueled by the emphasis on evi-
dence-based clinical practice and the effectiveness of early orthodontic
intervention. Will the early phase of treatment be effective and eliminate
the need for later treatment in the permanent dentition? Is it an efficient
method of treatment? What are the risks, costs and benefits of early orth-
odontic treatment compared to those incurred when waiting until the per-
manent teeth erupt and treating the malocclusion in one phase? Semantic
confusion has arisen concerning the term “early orthodontic treatment.”
Does it mean:
• Timing of early treatment interventions?
• Class 1 crowding in the mixed dentition?
• Crossbites with/without CR/CO shift?
• The results from the Class II Randomized Clinical Trials?
• Class III modification and redirection of growth with a protrac-
tion face mask?
• The assumption that early intervention will be more efficient
and effective?
The AAO website suggests that children should have an orthodon-
tic consultation by the age of seven years and indicates multiple benefits
for early orthodontic treatment which include:
1. Guiding the growth of the jaw
2. Regulating the width of the upper and lower dental arches
3. Lowering the risk of trauma (accidents) to protruded upper
incisors (front teeth)
4. Improving personal appearance and self-esteem
5. Potentially simplifying and/or shortening treatment time for
later corrective orthodontics.
Is there any evidence to support these contentious beliefs? If
evidence exists to support the effectiveness and efficiency of early orth-
odontic treatment, we need to integrate this evidence with our clinical
expertise and patient values need to be identified (Sackett et al., 1991,
2000). One method of achieving this goal is to perform systematic re-
views that incorporate a meta-analysis of the available data. This ap-
16
Vig et al.
proach has been used for several years in medicine with the benefit of
enabling patients and physicians to make informed decisions on care. A
systematic review provides a method of identifying all of the available
literature on a topic and synthesizing it into easily accessible knowledge.
We will illustrate this approach in this chapter as we consider data derived
from systematic reviews regarding treatment for posterior crossbites and
early treatment for Class II, division 1 malocclusions.
THE SEARCH FOR EVIDENCE
The first step in carrying out this critique of the literature was to
perform an initial search covering the general field of early orthodontic
treatment, as opposed to one covering a specific intervention such as serial
extraction or crossbites in the mixed dentition. Questions to be addressed
included:
1. Is it better to manage skeletal Class II and Class III patients in
the mixed dentition?
2. In the Class I crowded dentition, is the concept of serial ex-
traction rational?
. Is there a “best time” to treat specific malocclusions?
Is two-phase treatment better than one-phase treatment?
5. What scientific evidence exists to answer these questions?
i
STRUCTURING THE RESEARCH QUESTION
The clinical question “What is the effectiveness of early orth-
odontic treatment?” or “Is early orthodontic treatment efficient?” may not
be constructed in a format that is answerable by conducting a systematic
search of the scientific literature. The essential components of developing
an answerable question are considered in the PICO (T) format:
1. Patient/population – children in the mixed dentition
2. Intervention — early orthodontic treatment
3. Comparison – no early orthodontic treatment
4. Outcome – reduce severity of malocclusion/prevent a second
phase of orthodontic treatment
5. Time of intervention
Finding the best available evidence from sources of published
and unpublished studies requires a systematic approach that should
be standardized to avoid bias (Eggar et al., 2001). The quality of data
17
The Search for Evidence
retrieved from a careful, systematic and standardized review of the scien-
tific literature may be quantitative and/or qualitative in nature (Glasziou et
al., 2001). Therefore, discrete steps are required to find the relevant studies
when searching computer data bases to retrieve a body of literature that
then requires careful selection and appraisal.
For example, an initial search of the question “Is early orthodon-
tic treatment effective?” produced:
1. Medline – free text
a. Orthodontic treatment: 4,800 hits
b. Mixed dentition: 1,593 hits
c. Treatment effectiveness: 1,089 hits
d. Early orthodontic treatment: 109 hits
2. Combining terms: Boolean AND/OR
a. Mixed dentition AND Orthodontic treatment: 169 hits
b. Early orthodontic treatment AND mixed dentition: 16 hits
c. Treatment effectiveness AND early orthodontic treatment:
6 hits
A search criterion based on combining “orthodontic treatment”
AND “mixed dentition”. AND “effectiveness” and limiting the scope of
the search to include only scientific papers written in English found 55
abstracts that subsequently were reviewed for content areas. This initial
search resulted in restructuring the research question: in children with
malocclusions [clinical condition], is early orthodontic treatment [in-
tervention] more effective than later treatment [comparison] in promot-
ing normal occlusion [outcome]?
SEARCHING METHODS
Population Intervention Comparison Outcome
Search terms | Search terms Search terms Search terms
º Early orthodontic | Control group/ Normal/improved
Children e
treatment no early treatment occlusion

A focused search using Ovid MEDLINE/PUBMED databases and
using Free text and MeSH (medical subject headings) terms produced:
Early orthodontic treatment: 843 hits
Searching by the title early orthodontic treatment: 100 hits
Imposing limits child, English and human: 31 hits
18
Vig et al.
MeSH terms “orthodontics, interceptive”: 642 hits (MeSH terms
are automatically exploded)
MeSH with limits child, English, humans and RCTs: 16 hits
MeSH “dentition, mixed” AND “orthodontics”: 10 hits (addition
of Boolean operators AND/OR)
Truncation by searching effective”: 616,448 hits
Adding # AND # with clinical queries focused the search on 17
RCTS.
Inclusion criteria were limited to English language, child, human
and randomized clinical trials. The criteria craniofacial anomalies, cleft
lip/palate, adults and qualitative studies were excluded. Using both MeSH
and free text terms was the most productive method for finding relevant
studies.

Database Number of Relevant Studies
Hits
PUBMED/MEDLINE 17 7 (1 was discarded: it
had no control group)
Cochrane Database of systematic 19 3 registered protocols
reviews. 1 systematic review
Hand Search AJO-DO 1
CRITICAL APPRAISAL
First Author Journal Year Study Database
º PAR in 2-phase
King AJO-DO 2003 Tx. for Class ll MEDLINE
Wheeler AJo-Do | 2002 | Prºtiyºnºs. | PUBMED
early Tx Class ll
Ghafari AJO-DO 1999 Early Tx:Class 11 PUBMED
malocclusion
© e Effectiveness of
Mirabelli AJO-DO 2005 Hand search
Phase 1 TX
Kau J Orthod | 2004 | Early treatment PUBMED
for crowding
Tulloch AJo-Do | 2004 |9|ºs" "y" | PUBMED
Outcomes
O’Brien AJO-Do | 2003 || Parly Ortho TX PUBMED
in Class ll

19
The Search for Evidence
Of the 17 Studies identified in PUBMED/MEDLINE based on the
criteria outlined above, seven publications addressed the focused question.
Of those seven, one (Ghafari et al., 1999) later was excluded as no well-
defined control group could be identified. The remaining 10 RCTs were
either not relevant or addressed qualitative aspects of early orthodontic
treatment such as self-esteem, teasing and bullying.
The RCTs relating to Class II malocclusions were funded by NIH
in the USA (Wheeler et al., 2002; King et al., 2003; Tulloch et al., 2004)
and the MRC in the UK (O’Brien et al., 2003). As these RCTs were well
designed, they have been combined in a meta-analysis that will be pre-
sented with the relevant forest plots later in this chapter.
The RCT conducted by Kau and colleagues (2004) was for in-
terceptive early orthodontic treatment involving extractions performed
during the mixed dentition stage to relieve crowding and straighten the
lower teeth. The randomization of the treatment and control groups is well
described.
Mirabelli and colleagues (2005) compared early orthodontic treat-
ment in patients who paid for their own treatment to that of patients whose
treatment was paid for by Medicaid (a program that subsidizes treatment
for socio-economically disadvantaged children). A significant reduction
in malocclusion severity was reported with early treatment, but no control
population was identified.
MAKING RATIONAL DECISIONS
WHEN UNCERTAINTY EXISTS
In the practice of orthodontics, clinical experience suggests to us
that some conditions are best treated early for biological, social or practi-
cal reasons, whereas treatment for other conditions should be deferred.
However, the answer to the question of when to treat is not always clear.
For example, when anterior crossbites exist in the early mixed dentition
due to Class I crowding or a mild, developing Class III skeletal pattern,
should we treat the problem to avoid perpetuating the malocclusion with
possible labial gingival recession on the mandibular incisor from the trau-
matic incisor relationship or wait until the permanent teeth have erupted
in the late mixed dentition stage to correct the malocclusion (Fig. 1)? How
do we reconcile such conflicts?
When using a protraction face mask in an attempt to move the
nasomaxillary complex forward, our knowledge of craniofacial growth
and development indicates that early intervention when the circum-
maxillary suture system will be responsive is appropriate. In correcting
20
Vig et al.
Figure 1. Anterior view of an anterior crossbite in the early
mixed dentition with a traumatic relationship of the right per-
manent central incisors.
Figure 2. Anterior view of permanent central incisors that have
erupted into a crossbite in the early mixed dentition.
the anterior crossbite early (Fig. 2), the permanent lateral incisors erupt
into a normal relationship, which supports the concept of effective and
efficient early treatment intervention. With further growth, however, the
Class III tendency may result in the anterior crossbite re-establishing


21
The Search for Evidence
Figure 2. Continued. Anterior crossbite corrected with a pro-
traction facemask with the erupted permanent lateral incisors in
a normal occlusal relationship.
itself. If pathology exists in the mixed dentition such as submerging teeth
(Fig. 3), it probably should be treated early to avoid total submergence.
There are problems with using an evidence-based approach to
clinical decision making if the scientific literature, when systematically
reviewed, does not have an abundance of prospective, randomized clinical
trials relating to the specific treatment in question. However, the RCT is
still the gold standard for the hierarchical model of the strength of evi-
dence. So, what is the rational for clinical judgment and decision making
when it cannot be based solely on scientific evidence at the highest level
and must rely heavily on clinical experience?
SEARCHING AND APPLYING THE EVIDENCE
One of the most common early orthodontic treatment interven-
tions in the mixed dentition is the correction of posterior crossbites, with
or without a CR/CO shift, with maxillary expansion. This may be con-
sidered a well accepted clinical practice but what evidence exists in the
Scientific literature?
Harrison and Ashby (2001) published a systematic review of
orthodontic treatment for posterior crossbites in the Cochrane Database
of Systematic Reviews. It is a comprehensive review of randomized and
controlled clinical trials in the Scientific literature that reported quantita-
tive data on the outcomes of crossbite correction. At first glance it would
appear that there is an extensive number of publications available on this

22
Wig et al.
topic; however, when a systematic review is done to identify, by quality,
the studies that should be used to make stronger inferences, the number
of publications is small. The search criteria, defined by a priori inclu-
sion criteria, identified only seven RCTs and five controlled clinical trials
(CCTs) that were relevant studies of orthodontic treatment of posterior
crossbites.
Figure 3. Occlusal view (top) of a submerging man-
dibular left primary molar in the mixed dentition. Lat-
eral view (bottom) showing mesial drift and tipping of
the permanent first molar with the submerging primary
Second molar.

23
The Search for Evidence
Figure 3. Continued. Panoramic radiograph of submerging
primary molar and associated radio-opaque mass – possibly a
malformed, unerupted second premolar.
Cochrane reviews have the advantage of being regularly updated
as new information becomes available. The abstract, updated by Harrison,
included studies finished since 2001 which added 113 abstracts to be as-
sessed for potential inclusion. Of these, 38 were obtained and assessed for
eligibility. An additional five reports to three RCTs and one CCT, together
with another report concerning a previously included CCT, satisfied the
inclusion criteria.
To make sense of appraising a systematic review, critical apprais-
al tools have been developed by the Critical Appraisal Skills Programme
(CASP 2004). Ten questions were constructed that would consider three
broad issues when appraising a systematic review: validation of the re-
view, definition of the results, and application of the results. The first
two questions are screening questions. If they cannot be answered in the
affirmative, it probably is a waste of time to proceed. As an example of the
use of this critical appraisal tool, the ten-question checklist was applied
to Harrison’s and Ashby's systematic review of orthodontic treatment for
posterior crossbites.
1. Did the review ask a clearly-focused question? The Cochrane
review focused on a specific dental condition in children for which alter-
native clinical interventions could be compared. There were six types of
outcomes measures. The primary outcome was elimination of a posterior
crossbite as a result of effective early orthodontic treatment and occlusal

24
Vig et al.
adjustments. Secondary outcomes included a reduction in signs, symp-
toms or a patient’s risk of developing jaw, joint or respiratory problems;
stability of crossbite correction; and quality of life.
2. Did the review include the right types of studies? The review
included appropriate studies reported in the scientific literature. The study
designs were inconsistent, but they were the best that was available.
3. Were all relevant studies identified? The two reviewers (JH
and DA) each performed a rigorous database search independently and
then cross-checked each other for articles to be included in the review.
The Cochrane Oral Health Group (OHG) search strategy was used and
the British, European and American Journals of Orthodontics were hand
searched for papers and abstracts published between the years 1970-1999.
The OHG Database of Clinical Trials used the Boolean terms AND/OR,
and a MEDLINE search was performed for the years 1966 through 1999
for MeSH terms with truncation. First named authors of RCTs and CCTs
were contacted to identify any unpublished studies. The databases were
searched for all studies irrespective of language, although the review was
limited to clinical trials published in English.
4. Was the quality of each study assessed? The quality of the in-
cluded studies was predetermined by the two authors, Harrison and Ashby.
The Cochrane Handbook principles and the Jadad 1996 scoring system
were applied and resulted in three trials scoring 1/5 and three scoring 2/5.
Although seven RCTs and five CCTs were identified, correspondence with
the authors resulted in a reclassification with five RCTs and seven CCTs
being included in the review. Of these, four RCTs and five CCTs were
analyzed, but the remaining three studies (one RCT and two CCTs) were
excluded because they did not address the “focused” question. The meth-
ods of randomization were not described in the trials, and four types of
bias were identified: selection, performance, attrition and detection.
5. Was it reasonable to combine the results of the studies? To an-
swer this, the results are illustrated with tables that compare the charac-
teristics of the Methods, Participants, Interventions, Outcomes and Notes
and include RCT/CCT types of studies. Allocation concealment was re-
ported as “B” for five studies, “C” for three and “D” for one of the studies.
Graphs and Forest plots compared the outcomes in the mixed dentition
and the permanent dentition for grinding versus no treatment, bonded ver-
sus banded expansion appliances and subsequent relapse with removable
appliances versus Quadhelix fixed appliances.
25
The Search for Evidence
6. How are the results presented, and what is the main result?
Odds ratios and 95% confidence intervals were reported. The overall ef-
fect was reported as a Z-score with p-value in the 12 figures. Chi-square,
used to test for heterogeneity, was reported for the two trials with adequate
sample sizes for a statistically significant result. The outcome for crossbite
correction compared grinding (occlusal adjustment) versus no treatment in
the primary dentition. The result favored grinding, with an overall effect
of z =7.48 and p < 0.00001. The combined results from these two trials
supported early correction of posterior crossbites by occlusal grinding in
3- to 6-year-old children to prevent the perpetuation of the crossbite in the
mixed or permanent dentition.
7. How precise were the results? The results were not precise in
seven of the nine clinical trials because they were inadequately powered
to make clinical recommendations. None of the trials reported odds ratio,
relative risks (RR), absolute RR and RR reduction. The authors calculated
these for event data and weighted mean differences. Associated 95% con-
fidence intervals were calculated for continuous data.
8. Can the results be applied to the local population? In countries
with a national health service, all children are eligible for dental care. The
clinical trials in this review were carried out predominately in Europe and
Scandinavia where there are national health services. In the United States,
health care coverage for children is dependent on eligibility for public
funding, self pay or health insurance. For children who have dental care
coverage, early intervention of occlusal grinding or an expansion appliance
would be appropriate. Children in low socio-economic environments often
have rampant caries; occlusal contacts often are lost because of dental de-
cay. Removing premature occlusal contacts to correct a posterior crossbite
would have a low priority in these children for whom caries control and
relief of pain are the primary goals.
The answer to the PICO question (“in children with malocclusion
(posterior crossbite) is early orthodontic treatment [grinding/maxillary
expansion] more effective than later treatment [untreated group] in pro-
moting a normal occlusion [recurrence of the posterior crossbite]”) is that
this study supports the effectiveness of the clinical practice of early orth-
odontic intervention to prevent the perpetuation of posterior crossbites.
9. Were all outcomes considered? The outcomes produced by
correcting posterior crossbites in the primary dentition by occlusal grind-
ing with/without using expansion appliances during the mixed dentition
were compared to the outcomes produced by no treatment. Different
26
Vig et al.
types of maxillary expansion did affect treatment outcome. The results of
two of the trials suggested that grinding the baby teeth in the primary den-
tition to eliminate premature occlusal contacts is effective in preventing
the recurrence of posterior crossbite. As the cause of posterior crossbite is
not understood well, any evidence that supports early correction of poste-
rior crossbite because of adverse growth disturbances is speculative. Cor-
recting posterior crossbites is a widely accepted clinical practice, although
the timing and type of treatment is controversial. Health care policy deci-
sions based on cost/benefit support a one-time intervention as opposed to
interceptive treatment in the developing dentition. Outcomes related to the
burden of care were not included nor the community at large.
10. Should clinical practice change as a result of the evidence de-
rived from this review? The purpose of this systematic review was to eval-
uate early orthodontic intervention to expand the upper dental arch and
correct posterior crossbites. The evidence was inconclusive, because the
samples were too small and inadequately powered to support policy deci-
sions. The early correction of posterior crossbites is considered a benefit in
the United States, and the evidence from this review is unlikely to change
this opinion. In those countries with public health services, the evidence
from this review does not identify the early treatment of posterior cross-
bites as a major advantage.
It becomes clear when trying to quantitate the evidence using
systematic reviews and meta-analysis that a definition of evidence-based
clinical practice requires the careful and considered use of mathematics
and statistics; that is, “Evidence-based medicine is the enhancement of
a clinician’s traditional skills in diagnosis, treatment, prevention and the
related areas through the systematic framing of relevant and answerable
questions and the use of mathematical estimates of probability and risk”
(Donald and Greenhalgh, 2001). The advantage of a systematic review is
that it will limit bias by using a methodological approach to develop strict
inclusion criteria for articles, and this makes the conclusions more reliable
and accurate (Greenhalgh, 2001).
ONE PHASE VERSUS TWO PHASE TREATMENT IN
CLASS II MALOCCLUSION – META-ANALYSIS
The publication of the results of the RCTs on Class II maloc-
clusion provides some of the strongest evidence available on whether to
treat early or later, but the debate continues. For example, criticism has
27
The Search for Evidence
been leveled against the choice of functional appliance used in the stud-
ies, because the appliances were not considered to be contemporary or
appropriate. It has been suggested, furthermore, that the inclusion criteria
of the studies was not relevant to the “real world” of orthodontic prac-
tice. Nevertheless, these investigations do provide strong evidence on the
effectiveness of early treatment, and those who simply dismiss them as
not being relevant to their practice may be entering a future medico-legal
minefield. In order to provide a summary of the current state of evidence
on the effectiveness of early Class II treatment, we (KO’B and JH) carried
out a systematic review of this subject.
The purpose of the review was to assess the effectiveness of orth-
odontic treatment for prominent upper front teeth, when treatment is pro-
vided either when the patient is seven-to-nine years old or during early
adolescence. The null hypothesis tested in the meta-analysis was that early
Class II treatment, when compared to treatment provided in one phase,
did not result in any difference in skeletal pattern, overjet or dental occlu-
Sion.
The search strategy used electronic databases including Cochrane
Oral Health Group’s trial register, CENTRAL MEDLINE, MEDLINE
pre-indexed, and EMBASE.
#1 MALOCCLUSION-ANGLE-CLASS-II (ME)
#2 (“Class II” AND ((Angle OR Angle's) OR malocclusion OR
bite) -
#3 (Explode) ORTHODONTIC-APPLIANCES-FUNCTIONAL
(ME)
#4 (Explode) ORTHODONTIC-APPLIANCES-REMOVABLE
(ME)
#5 “Frankel” OR “Twin*block” OR “Fixed appliance”
#6 ((Extraoral OR “extra oral” OR extra-oral) AND appliance”)
#7 (“growth modif" AND (jaw OR maxilla” OR mandible)
#8 (“head gear” OR headgear)
#9 ((two-phase (treatment OR therapy)) AND (orthodontic.” OR
malocclusion))
#10 ((orthopedic” OR orthopaedic”) AND (dental OR orthodon-
tic” OR facial))
#11 #1 OR #2
#12 #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9 OR #10
#13 #11 AND #12
Electronic searches identified 174 titles and abstracts. From the
102 full reports, only 28 studies were considered to be eligible based on
28
Vig et al.
the defined inclusion criteria for trial design, participants, interventions
and outcomes. Fifteen trials were excluded, and those selected met the
following criteria:
Design – random allocation of participants
Participants – children receiving treatment for prominent front
teeth
Intervention – any orthodontic treatment
Primary outcomes – reduction of prominent front teeth, post-treat-
ment facial skeletal pattern with data recorded on the second-
ary outcomes from information regarding methods, partici-
pants, interventions, outcome measures.
Results were extracted independently in duplicate by two review-
ers (KO’B and JH). The Cochrane Oral Health Group statistical guidelines
were followed, and weighted mean differences were calculated using ran-
dom effects models. Potential sources of heterogeneity were examined in
random effect meta-regression analysis.
Methodological quality of the studies was reviewed for allocation
concealment, which was adequate for four of the trials but unclear for the
remaining trials. The outcome assessor was blinded in three trials, but it
was not clear if this was the case for the other trials. Withdrawals were ad-
equately reported in eight of the nine trials. The kappa scores between the
two reviewers were 1.0 for allocation concealment, 1.0 for blinding of out-
come assessment and 0.9 for clear information on withdrawals. Five trials
were assessed as having a low risk of bias (O’Brien 2003a,b,c; Banks,
2004), and four were assessed as having a moderate risk of bias.
In the seven trials included in the review, there were 826 patients
who presented with Class II, division 1 malocclusions. The range of pa-
tients was from 13 to 72 per treatment/control group. The studies were
divided into two groups. The first group included the studies that reported
the effects of early treatment (Phase I treatment) at either the end of Phase
I treatment or at the end of Phase II follow-up treatment. The second group
included the studies that reported the effects of one-phase treatment pro-
vided during adolescence.
BENEFITS OF EARLY INTERVENTION
We found strong evidence that when orthodontic treatment was
provided for children eight-to-ten years old who had prominent front
teeth (early or Phase I treatment), there was a clinically and statistically
29
The Search for Evidence
significant reduction in incisor prominence at the end of Phase I treatment.
This effect occurred whether the patient received treatment with a func-
tional appliance or headgear. We also found some differences in skeletal
pattern, but we did not considered them to be clinically significant. This
information can be illustrated in the form of Forest plots, which allows us
to represent the findings of the review as graphics and data. The vertical
or “no effect” line favors intervention when the results are on the left and
does not favor intervention when the results are on the right. The results of
the combined studies were pooled and are represented by the diamond and
its spread and relationship to the “no effect” vertical line.
When the final outcome of treatment was reviewed at the end of
a second phase of treatment when the patients were in their early ado-
lescence, we found that incisor prominence had been reduced, but there
were no differences in treatment outcome for overjet or ANB between the
groups of children who had received one or two phases of treatment. As
a result, two-phase treatment does not appear to have any advantages over
one-phase treatment. Again, this is illustrated well by the Forest plots seen
in Figures 6 and 7.
IMPLICATIONS FOR FUTURE CLINICAL PRACTICE
AND CLASS II MALOCCLUSION
Early orthodontic treatment (Phase I) followed by a later phase
of treatment (Phase II) during early adolescence, has no advantages over
treatment that is provided in one phase in early adolescence aside from a
transitory boost in self esteem (O’Brien et al., 2003b). When functional
appliance treatment is provided in early adolescence, it appears that there
are minor beneficial changes in skeletal pattern, but these changes are not
clinically significant and the differences were not maintained following
Phase II treatment. Similarly, the choice of functional appliance does not
seem to confer any advantage. We can conclude, therefore, that provid-
ing early orthodontic treatment for prominent upper front teeth is no more
effective than providing one course of orthodontic treatment during early
adolescence. Furthermore, as this evidence is derived from a systematic
review, it is compelling and should be part of the information that we give
to our patients when obtaining consent for treatment.
30
Review: Orthodontic treatment for prominent upper front teeth in children (version 03)
Comparison. 01 Early treatment at the end of Phasel. Functional vs control
Outcome: 01 Final overjet
Study Functional Control WMD (random) Weight WMD (random)
or sub-category N Mean (SD) N Mean (SD) 95% CI º 95% CI
O'Brien 2003a 39 3.70:2.27% 34 10. 70 :2.40% --- 33-47 -7.00 I-7-70, -6.301
Florida 35 3. 88 fl. 90) 79 S. 42 ºz. 67) --- 33.45 -1.54 I-2-25, -0.831
North Carolina 4l 5.38:2-67) 54 8- 94.1. 842 --- 33. J3 -3.56 I-4.51, -2. 611
Total (95% CI) 215 2.17 -º-, -º- 100.00 -4-04 I-7-47, -0. 601
Test for heterogeneity: Chi – 11702, df = 2 (P<0.00001), F = 98.3%
Test for overall effect Z=2.30 (P=0.02)
-10 -5 d 5 10
Favours Functional Favours Control
Review: Orthodontic treatment for prominent upper from teethin children (Version 03)
Comparison: 01 Early treatment at the end of Phase Functional vs control
Outcome: 02 Final A-N-B
Study Functional Control WMD (random) Weight WMD (random)
or sub-category N Mean (SD) N Mean (SD) 95% CI º 95% CI
O'Brien 2003a e? 3- 85 fl. 80% 74 7-35 - 7 - 80% +-- 22.47 -3. So I-S-32, -1- 681
Florida 35 3-96 - 1 - 95% 78 4- 49.2 - 19) ---. 40. 32 -0. 53 I-l. 17, 0-ll]
North Carolina 41 4- 82 (2- 0.8% 54 S. 77 ± 2.08% --- 37 - 21 -0-95 I-l. 79, -0-1ll
Total (95% CI) 2.13 206 -º- 100.00 -1-35. I-2-57, -0-141
Test for heterogeneity: Chº - 9:17, df = 2 (P=0.01), F = 78.2%
Test for overall effect: Z = 2.19 (P = 0.03)
-4 -2 D 2 4.
Favours Functional Favours Control
Figure 4A (upper) and B (lower). Forest plots for the effect of early treatment with functional appliances on overjet and ANB at the
end of Phase I treatment.

º

Florida 95. 3. 8s ºl. 8s) 78 4. 49 (2.19) - 54. 98 -0.60 I-l. 21, 0.011
North Carolina sº 4.33 ± 1. SO) 54 S-70 -2.00) - 45.02 -0-87 I-l. 55, -0.191
Total (95% CI) 145 132 {} 100.00 -0. 72 I-l. 18, -0.271
Test for heterogeneity: Chº - 0.34, df = 1 (P=0.56), F = 0%
Test for overal effect: Z = 3.12 (P=0.002)
Figure 5A (upper) and B (lower). Forest plots for the effect of early treatment with headgear appliances on overjet and ANB at the
Review: Orthodontic treatment for prominent upper front teeth in children (Version 03)
Comparison. 02Early treatment at the end of Phasel, Headgear vs control
Outcome: 01 Final overjet
Study Headgear Control WMD (random) Weight WMD (random)
or sub-category N Mean (SD) N Mean (SD) 95% CI º 95% CI
Florida 35 3. 8s (2-33) 7s S. do ſº. 67) --- 55.75 -1.01 I-l. 76, -0.261
North Carolina so 7 - 80 ºz. 48) S4 8- 94 ºl. 84) --- 44.25 -l. 14 I-1-98, -0.301
Total (95% CI) 145 133 º 100.00 -1.07 [-l. 63, -0.5.1.1
Test for heterogeneity: Chº - 0.05, df = 1 (P = 0.82), F = 0%
Test for overall effect: Z = 3.72 (P=0.0002)
-10 –5 D 5 10
Favours treatment Favours control
Review: Orthodontic treatment for prominent upper front teeth in children (Version 03)
Comparison. D2 Early treatment at the end of Phase I Headgear ws control
Outcorne: U2 Final A-N-B
Study Headgear Control WMD (random) Weight WMD (random)
or sub-category N Mean (SD) N Mean (SD) 95% C. º 95% CI
-10 -5 D 5 10
Favours treatment Favours control
end of Phase I treatment.
§
Review: Orthodontic treatment for prominent upper front teeth in children (Version 03)
Comparison: 04 Early treatment at the end of Phase Iſ Functional vs control
Outcome: 01 Final overjet
Study Treatment Control WMD (random) Weight WMD (random)
or sub-category N Mean (SD) N Mean (SD) 95% CI º 95% CI
Florida s? 2. 60 fl. 12) 58 2- 49 : 1 - 08.) 43.84 0.11 I-0-26, 0.481
North Carolina 33 3. 72 :2-04) 51 3.99 fl. 75) 25-77 -0-27. I-1.07, 0- 531
O'Brien 2008 56 4- 30 - 2. 151 52 3. 44 (1.49) --- 30-40 0.86 [0. 19, l. 531
Total (95% CI) 152 181 100.00 0-24. I-0-32, 0-801
Test for heterogeneity: Chº - 523, df = 2 (P=0.07), F = 61.8%
Test for overall effect: Z = 0.84 (P=0.40)
-10 -5 d 5 10
Favours treatment Favours control
Review: Orthodontic treatment for prominent upper front teeth in children (Version 03)
Comparison. 04 Early treatment at the end of Phase I, Functional vs control
Outcorne: 02 Final A-N-B
Study Treatinent Control WMD (random) Weight WMD (random)
or sub-category N Mean (SD) N Mean (SD) 95% CI º 95% CI
Florida 55 3 - 70 ºl. 90) 62 3. 49 (2-35) T. 35. 55 0.21 [-0. 54, 0.361
North Carolina 39 3-72 t 2 - 12) Sl 4 - 36 (2.06) + 27. 89 -0.64 I-l. 51, 0.231
O'Brien 2008 52 4.- 00 ºl. 99) 63 3. 81 ºz. 28) - 36. 48 0.19 I-0. 54, 0.92]
Total (95% CI) 168 18.1 100.00 -0.03 I-0- 55, 0.48]
Test for heterogeneity: Chº - 2.52, df = 2 (P=0.27), F = 23.6%
Test for overall effect: Z = 0.13 (P=0.90)
-10 -5 º 5 10
Favours treatment Favours control
Figure 6A (top) and B (bottom). Forest plots for the effect of early treatment with functional appliances on overjet and ANB at the end
of Phase II treatment.

3

Study Headgear Control WMD (random) Weight WMD (random)
or sub-category N Mean (SD) N Mean (SD) 95% CI º 95% CI
Florida 72 2 - 40 ºl. 38) 58 2 - 49 : 1 - 08.) 54. 70 -0-09 I-0.50, 0.321
North Carolina 47 3 - 48 fl. 29) Sl 3- 99 : 1 - 75) 35. 30 -0. 51 I-l. 12, 0-10]
Total (95% CI) 113 L13 100.00 -0-24 I-0-63, 0- 161
Test for heterogeneity: Chi" = 1.27, df = 1 (P = 0.26), F = 21.2%
Test for overall effect: Z = 1.19 (P = 0.24)
Review: Orthodontic treatment for prominent upper front teeth in children (Version 03)
Comparison. 05 Early treatment at the end of Phase Iſ Headgear ws control
Outcome: U2 Final A-N-E
Study Headgear Control WMD (random) Weight WMD (random)
or sub-category N Mean (SD) N Mean (SD) 95% CI º 95% C.
Florica 71. 3- 30 ºl. 80) 52 3. 49 : 2-35) 54-42 -0. 19 I-0. 91, 0.531
North Carolina 47 4.- 0:0 (l- 91) Sl 4-36 (2-06) 45.58 -0.36 I-l. 15, 0.431
Total (95% CI) 118 113 100.00 -0-27 [-0-80, 0.261
Test for heterogeneity: Chº - 0.10, df = 1 (P=0.75), F = 0%
Test for overall effect Z = 0.99 (P = 0.32)
Figure 7A (top) and B (bottom). Forest plots for the effect of early treatment with headgear on overjet and ANB at the end of Phase II
Review: Orthodontic treatment for prominent upper front teeth in children (Version 03)
Comparison. 05 Early treatment at the end of Phase I Headgear vs control
Outcorne: 01 Final overjet
-10 –5 d 5 10
Favours treatment Favours control
-10 –5 d 5 10
Favours treatment Favours control
treatment.
º
Vig et al.
SUMMARY
The best evidence from searching the question, “In children with
malocclusion is early orthodontic treatment more effective than later
treatment in promoting normal occlusion?” had the highest quality of
articles identified in a systematic review on children with Class II maloc-
clusion characterized by prominent upper front teeth. Orthodontic beliefs
derived from clinical experience determines orthodontic decision making.
However, those responsible for informing patients of alternative treat-
ments may be influenced by the results of this meta-analysis as patient
preferences for treatment options are based on informed decisions.
Historically, medical regimens have remained unchanged even
when well designed clinical trials have provided evidence that a change
is warranted. Treatment decisions based on clinical experience and beliefs
have been shown to take 17 years to change after results of clinical trials
have been reported (IOM report, 2003). In medicine, for example, throm-
bolytic therapy and the administration of streptokinase underwent clinical
trials in 1960 resulting in findings in favor of the therapy. By 1975, 40
RCTs had been conducted and by 1985, 50,000 patients were enrolled, all
of which provided evidence confirming that thrombolytic therapy was ef-
fective. When a systematic review and meta-analysis conclusively showed
the effectiveness of thrombolytic therapy, it finally was accepted as the
standard of care in 1990. If the contemporary methodological approach
to evidence-based practice had been established 30 years ago, lives could
have been saved.
Even when evidence is available, clinicians still may be unable
to relinquish the beliefs that are based on their clinical experience. In the
practice of clinical orthodontics, treatment decisions often are made based
on the belief that early intervention is beneficial even when evidentiary
data shows that this practice is not beneficial in terms of effectiveness, ef-
ficiency, risk, cost and benefit of alternatives.
Orthodontics, which is the oldest specialty in dentistry, recog-
nizes that strong scientific evidence is an important goal for the future of
the profession. Patients are waiting to be treated even though we cannot
provide prior probability estimates for the outcome of alternative treat-
ment options at the time of the initial consultation. It still is necessary
for patients to have their preferences considered and this should be re-
flected in their consent for treatment. Advances often are first brought
to our attention through anecdotal case reports and observation, as was
the discovery of penicillin. Although low on the strength of evidence
pyramid, these initial reports have value as do retrospective studies, case
35
The Search for Evidence
series and clinical experience. Currently the profession needs to be realis-
tic and, to quote Ackerman and colleagues (2006), “the soundest mode of
practice today is reflective, experienced evidence-bolstered orthodontics.”
This does not mean that we should accept the present state of orthodontics
as a science, but that we should become more rigorous in designing clini-
cal trials to determine what works, what doesn’t and what is just inspired,
albeit attractive, rhetoric with little scientific support or substance.
REFERENCES
AAO website www.braces.org.
Ackerman JL, Kean MR, Ackerman MB. Evidence-bolstered orthodon-
tics. Australian Orthod J 2006:22:69–70.
Banks P. Wright J, O'Brien K. Incremental versus maximum bite advance-
ment during twin-block therapy: a randomized controlled clinical trial.
Am J Orthod Dentofacial Orthop 2004;126:583-588.
Cochrane AL. Effectiveness and Efficiency. Random Reflections on Health
Services. London, Nuffield Provincial Hospital Trust, 1972.
Cochrane AL, Blythe M. One Mans Medicine. An Autobiography of Pro-
fessor Archie Cochrane. Cambridge University Press, 1989.
Critical Appraisal Tools 2004. www.phru.nhs.uk/casp/critical appraisal.
Donald A, Greenhalgh T. A Hands-on Guide to Evidence Based Health-
care: Practice and Implementation. Oxford, Blackwell Science 2001.
Early Treatment Symposium. Am J Orthod Dentofacial Orthop 2006; 129:
S45-S70.
Eggar M, Smith JD, Altman DG, eds. Systematic Reviews in Health Care:
Meta-Analysis in Context. 2" ed., BMJ 2001.
Ghafari J, Shofer FS, Jacobsen-Hunt U, Markowitz DL, Laster LL. Head-
gear versus functional regulator in the early treatment of Class ll, divi-
sion 1 malocclusion: A randomized clinical trial. Am J Orthod Dento-
facial Orthop 1999; 113:51-61.
Glasziou P, Irwig L, Bain C, Colditz G. Systematic Reviews in Health
Care. Cambridge University Press 2001.
Greenberg H. On the proper use of clinical trials. In: Warren K, Mosteller
F, eds, Doing More Good Than Harm – The Evaluation of Health
Care Interventions. New York, New York Academy of Sciences
1993;703:41-43.
36
Vig et al.
Greenhalgh T. Papers that summarize other papers (systematic reviews
and meta-analysis). In: How to Read a Paper: The Basics of Evidence-
Based Medicine. London, BMJ 2001;8:120-138.
Harrison JE, Ashby D. Orthodontic treatment for posterior crossbites. The
Cochrane Database of Systematic Reviews. Hoboken, NJ John Wiley
& Sons Ltd. 2001; 1:Art. No. CD000979.
Hill AB Sir, Hill ID. Principles of Medical Statistics, 1" ed. 1937; 12" ed.
1991 London, Edward Arnold – Hodder and Stoughton.
IOM report, National Academy of Science 2003.
Kau CH, Durning P. Richmond S, Miotti FA, Harzer W. Extraction as a
form of interception in the developing dentition: A randomized con-
trolled trial. J Orthod 2004:31:107-114.
King GJ, McGorray SP. Wheeler TT, Dolce C, Taylor MG. Comparison
of peer assessment rating (PAR) from 1-phase and 2-phase treatment
protocols for Class II malocclusions. Am J Orthod Dentofacial Orthop
2003;123:489-496.
Mirabelli JT, Huang GJ, Siu CH, King GJ, Omnell L. The effectiveness of
Phase 1 orthodontic treatment in a Medicaid population. Am J Orthod
Dentofacial Orthop 2005;127:592–598.
O’Brien K, Wright J, Conboy F, Chadwick S, Connolly I, Cook P. Bir-
nie D, Hammond M, Harradine N. Lewis D, McDade C, Mitchell L,
Murray A, O’Neill J, Read M, Robinson S, Roberts-Harry D, Sandler
J, Shaw I, Berk NW. Effectiveness of early orthodontic treatment
with the Twin-block appliance: A multicenter, randomized clinical
trial. Part 2: Psychosocial effects. Am J Orthod Dentofacial Orthop
2003a; 124:488–494.
O’Brien K, Wright J, Conboy F, Sanjie Y, Mandall N, Chadwick S, Con-
nolly I, Cook P. Birnie D, Hammond N, Lewis D, McDade C, Mitch-
ell L, Murray A, O’Neil J, Read M, Robinson S, Roberts-Harry D,
Sandler J, Shaw I. Effectiveness of early orthodontic treatment with
the twin block appliance: A multicenter, randomized, controlled trial.
Part 1: Dental and skeletal effects. Am J Orthod Dentofacial Orthop
2003b;124:234-243. -
O’Brien K, Wright J, Conboy F. Sanjie Y, Mandall N, Chadwick S, Con-
nolly I, Cook P. Birnie D, Hammond M, Harradine N, Lewis D,
McDade C, Mitchell L, Murray A, O’Neill J, Read M, Robinson S,
Roberts-Harry D, Sandler J, Shaw I. Effectiveness of treatment for
Class II malocclusion with the Herbst or twin-block appliances: A
37
The Search for Evidence
randomized, controlled trial. Am J Orthod Dentofacial Orthop
2003c, 124:128-137.
Sackett DL, Haynes RB, Guyatt GH, Tugwell P. (1991) Clinical Epide-
miology: A Basic Science for Clinical Medicine. 2" ed. Toronto and
Boston, Little Brown and Co. 1991.
Sackett D, Strauss SE, Richardson WS, Rosenberg W. Haynes RB. Evi-
dence-Based Medicine: How to Practice and Teach EBM 2" ed., Lon-
don, Churchill-Livingston 2000.
Tulloch JF, Proffit WR, Phillips C. Outcomes of a 2-phase randomized
clinical trial of early Class II treatment. Am J Orthod Dentofacial Or-
thop 2004;125:657-667.
Wheeler TT, McGorray SP, Dolce C, Taylor MG, King GJ. Effectiveness
of early treatment of Class II malocclusion. Am J Orthod Dentofacial
Orthop 2002;121:9-17.
38
IF WISHES WERE HORSES
Lysle E. Johnston, Jr.
Another year, another “early treatment” conference. This time it is to
be “Early orthodontic treatment: Is the benefit worth the burden?” The
enduring popularity of this theme, both with attendees and organizers,
implies controversy. In turn, controversy implies uncertainty, which, in
orthodontics, is license. Our disagreements, therefore, tend to be immor-
tal. So it is with “early treatment.” Accordingly, I suspect that a garden-
variety review designed to marshal evidence in support of one side or the
other would resonate only with the converted, who, in any event, have
heard it all before. I think it more appropriate to look briefly at the prob-
lem confronting a patient facing the decision we have been charged with
discussing.
According to the most recent AAO early treatment conference
in Las Vegas, there are many condition that can, and perhaps should, be
treated “early”: avulsed and malformed teeth, congenital absence, anky-
losis, agenesis, autotransplantation, canine eruption problems, crossbites,
extreme overjet, extreme crowding, to mention but a few (see AJO-DO,
2006 in references). I would argue, however, that to the average or-
thodontist “early treatment” most often implies a two-phase treatment in
which the first phase is prosecuted by some sort of “functional” appliance
designed to modify growth. Two-phase treatment has become popular,
not only with referring dentists (who sometimes expect malocclusions to
be treated early, often, and without extraction), but also with orthodontists
who may believe that they are adding something “extra” to their treatment
outcome. Implicit in our charge to balance “benefit” and “burden” is the
acknowledgement that two-phase treatment commonly takes longer and
costs more. Unfortunately, as with apples and oranges, benefit and burden
are incommensurable. I would argue, however, that the science of deci-
Sion theory constitutes a balance on which they can be weighed and com-
pared. Unfortunately, its application to orthodontics is remarkably tricky.
Pneumonia is a disease. Squamous-cell carcinoma is a dis-
ease. But what about Class II malocclusion? Although many may dis-
agree strongly, I would argue that orthodontic “need” only rarely can be
inferred from the pathophysiology of any particular malocclusion. Ab-
sent an obvious medical justification, on what basis can we presume to
39
If Wishes Were Horses
characterize the benefit of treatment to a given patient? In other words,
how are we to answer the primal question, “Doctor, do I need treatment?”
At the outset, it is axiomatic that we will see our treatments as
beneficial. We all had a similar educational experience and, perhaps as a
result, we all tend to hew to a common “party line” that can be codified
in the form of indices of “need,” “outcome,” and the like. But what is
the gold standard against which these various indices are to be validated?
Professional opinion would be circular: the indices, after all, were gener-
ated by this same opinion. Ultimately, a patient’s personal value system
must constitute some sort of gold standard for need. Unfortunately, the
personal desirability of any given outcome, its “utility,” will differ from
patient to patient. What is a trivial effect to one patient may be almost a
matter of life or death to another. How are we to know?
In medicine, utilities can be estimated by determining what the
patient would risk to achieve a given outcome. This approach — the so-
called “standard gamble” (von Neumann and Morgenstern, 1944) — esti-
mates utilities by means of a graded series of lotteries in which the patient
is asked to choose between no treatment and a treatment that has a prob-
ability p of achieving the desired result and a probability q (where p + q
= 1) of a failure leading to the worst possible result (say, death in the case
of surgery; loss of teeth in orthodontics). The values of p and q are varied
systematically, and, when the patient cannot chose between the present
malocclusion and the lottery, that particular value of q gives an estimate
of the utility of the correction. A similar result can be achieved by way of
the so-called “time trade off,” a technique in which the patient is asked to
choose between, say, 50 years of life with the malocclusion or some lesser
life span with a perfect occlusion. As with the standard gamble, the point
at which the patient cannot decide whether it would be better to sacrifice
N years of life or forego treatment gives an estimate of the value of a suc-
cessful treatment to that particular patient.
It may seem far fetched to assume that a patient would risk death
or tooth loss for “straight teeth”; however, in practice they say they would,
and it can be assumed that patients who profess a willingness to risk much
or to sacrifice much place a greater value on treatment than those who do
not. The techniques for administering standard-gamble or time-trade-off
tests are reasonably straightforward and commonly are used in medicine
to assist patients in making tough decisions. For example, in choosing
a treatment for prostate cancer, individual patients will differ in their at-
40
Johnston
titudes toward impotence, incontinence, and even life, itself. These at-
titudes ultimately will determine the appropriate therapeutic strategy.
Fortunately, sequelae are of lesser import in the world of orthodontics.
In these less trying circumstances, can we instead substitute our studied
evaluation of a given malocclusion and, based on our assessment of its
seriousness, determine how much a patient ought to want a correction?
Recent research at the University of Michigan bears on this question.
To test whether our professional opinions are consistent with pa-
tient utilities, Straffon (2001) employed the pre-treatment, post-treatment,
and recall records of 48 patients from the Graduate Orthodontic Clinic.
These records were used to generate Peer Assessment Rating (PAR), In-
dex of Complexity, Outcome, and Need (ICON), and “American Board
Grading System” (ABO) scores. To minimize technical error (or at least
to minimize concern about error), the PAR and ICON scores were gener-
ated by Steve Richmond; the ABO scores, by Jim Vaden. In addition, the
records were evaluated by 20 members of our clinical faculty, who were
asked to use the so-called “Q-sort” procedure (Stephenson, 1953) to rank
the 48 sets of models – pre-treatment to assess need; post-treatment to
evaluate outcome; and both sets, together, to evaluate the overall goodness
of the treatment.
In this implementation of the Q-sort process, the 20 observers
each were asked to examine the records as a group and then pick the two
subjects at each extreme of the distribution (say, the two with the greatest
treatment “need” and the two with the least; the two with the best outcome;
the two with the worst; etc.). These models were set aside, and, from the
remaining 44 records, the clinicians were asked to select and set aside the
four best and four worst. The winnowing continued with the selection of
five and then eight from each tail of the remaining records. The 10 sets
of casts that remained formed a “neutral” pile in the middle of a quasi-
normal, 9-group distribution. The groups were numbered (0–8), and each
subject’s Q-sort score was determined by its group membership. For each
of the 48 subjects, the 20 scores were averaged to produce a single overall
Q-sort score to represent the opinion of experienced clinicians.
There was a moderately high correlation (re 0.5-0.7) among the
orthodontic indices and the clinicians’ Q-sort scores. In other words, we
tend to agree with each other and with the indices we invent. But what do
these various versions of the “party line” mean? Do our opinions in their
Various forms match the values of our patients? In other words, can we
know when the personal benefit is worth the burden?
41
If Wishes Were Horses
To examine this question, Skinner (2000) recruited 48 prospective
patients for the purpose of examining the relationship among PAR scores,
both weighted and unweighted, clinicians’ opinions (again by way of Q-
sort scores), and individual patient utilities estimated by way of the stan-
dard gamble technique. There was agreement among orthodontic opinion
and the various PAR scores; however, there was a poor correlation with pa-
tient utilities, the standard-gamble scores. Indeed, the correlation between
the American-weighted PAR scores and the standard-gamble utilities was
statistically non-significant. It gets worse.
Although patients might not be able to evaluate properly the value
of a good occlusion prior to treatment, it was hypothesized that the utilities
of ex-patients should be captured more precisely by received professional
opinion. In other words, we would expect ex-patients to look in the mirror
and place a higher value on results that we would see as good than they
would on results we would classify as deficient. Gupta Iyer (2003) studied
96 patients who had already completed treatment, who were 18 years or
older, and who had been financially responsible for their own treatment.
For these patients, there was no uncertainty: they presumably knew all
there was for a patient to know about treatment, including its burdens and
the goodness or badness of its outcome. These ex-patients were asked to
imagine that they were back the way they were before treatment. They
were shown their pre-treatment study models and, by way of the standard-
gamble technique, the utility of their actual treatment outcome was esti-
mated. Professional assessments of initial severity, quality of outcome,
and degree of improvement were generated by way of Q-sort scores (17
clinicians) and the various weightings of the PAR and ICON. Once again,
there was good agreement (r - 0.4–0.9) among the various professional
indices; however, these measures were almost completely unrelated to the
patients’ standard-gamble scores. This disconnect between professional
opinion and patient utilities argues that in the end the patient who needs
treatment is the patient who wants treatment. But for the patient who
wants treatment, what is the strategy with the greatest benefit? It all de-
pends . . . .
In assessing “benefit,” the final complication is the uncertainty of
treatment. Treatment is a gamble. Will the patient happen to consult a cli-
nician whose methods and treatment “philosophy” are appropriate to the
malocclusion at hand? Further, although most treatments tend to “work”
most of the time, sometimes they do not. Will a given patient’s outcome
be one of the successful ones? Indeed, some treatments can be quite risky.
In truth, patients contract not for improvement, but rather for the prob-
42
Johnston
ability of improvement, a complication that brings us to the problem of
decision theory and the strategy of orthodontic treatment.
Orthodontic decisions can be made under conditions of certainty,
risk, or uncertainty. Certainty implies that each treatment leads to a single
outcome. Decisions then would involve only a choice among consequenc-
es. Although no orthodontic outcome is a certainty, each treatment has a
restricted range of outcomes, some of which do not overlap (Johnston,
1986). Thus a patient, if given the opportunity, might be able to make a
rational choice among moderately distinct outcomes. In this instance, the
myths of the marketplace (e.g., “patients like full profiles”) would have
to be set aside in deference to the actual preferences of the individual pa-
tient. This approach, although a step in the right direction, probably is
too great a simplification, given that there still is considerable variation
among outcomes. Perhaps it would be more realistic to acknowledge the
fact that orthodontic treatment plans must be made either in the face of risk
(known probabilities) or uncertainty (probabilities unknown). Although
an informed discussion of decision theory is beyond both the scope of this
communication and the skills of the author, a few basic comments perhaps
are in order.
We cannot guarantee to deliver any specific outcome, let alone
one that the patient values. Instead, we can talk about “expectations.” If
we were to conduct a lottery in which ten tickets would be sold leading
to a prize of $100, then the value of each ticket prior to the draw (its “ex-
pected value”) would be the prize times its probability: $100 x 0.1 = $10.
In health care, decisions tend to be analyzed on the basis of “expected
utility” — the payoff expressed not in dollars, but in some estimate of the
value of an outcome to the patient times the probability that it will occur.
Not all outcomes are good, so we also can talk of disutilities, undesirable
outcomes. Armed with a utility estimate (determined, say, by time-trade-
off or standard-gamble techniques) and some idea of the relevant prob-
ability, it is possible to calculate the expected utility of an outcome from
the standpoint of the patient. If a patient is faced with a choice between
two treatments, each with good and bad outcomes, the sum of all the ex-
pected utilities (positive and negative) gives the “value” of each option.
The simplest strategy then would be for the patient to choose the treatment
with the greater expected payoff. There are, however, other strategies that
merit consideration.
A patient, for example, may place a very high utility on a relative-
ly unlikely outcome. For this patient, the strategy that might lead to this
particular outcome might take precedence over a strategy (i.e., a treatment
43
If Wishes Were Horses
plan) with a higher expected value. One might encounter this sort of maxi-
max strategy in a patient faced with a decision between orthodontics and
surgery. Alternatively, a patient might be risk averse and seek some sort of
minimax strategy designed to minimize the maximum loss. This particular
strategy might lead another patient to opt for orthodontics, regardless of
how “comfortable” the surgeon might be with the surgical option. After all,
nobody dies of anchorage loss. Perhaps at the current stage of orthodontic
evolution, this approach to patient education and informed consent might
seem an impractical elaboration. If, however, you were faced with choosing
among variably morbid treatments for a life-threatening disease in you, de-
cision theory and expected gain might seem considerably more “relevant.”
Whatever your view of the day-to-day practicality of decision theory, its
heuristic significance to this symposium is real and thus merits comment.
Consider two treatments, each with several outcomes. For pur-
poses of an orthodontic discussion (and in line with the charge of this
Symposium), two such strategies can be given names: 1) two-stage non-
extraction featuring “growth modification” as a substitute for extraction,
and 2) one-stage late mixed-dentition fixed-appliance therapy featuring,
say, 25-30% extraction to deal with the protrusive/crowded end of the con-
tinuum. In this admitted over-simplification, the two responses/states of
nature are “enough arch perimeter” and “crowding and protrusion.” How
do the two treatments fare?
Given that bone cannot grow interstitially, it is difficult to see
how “growth” — modified or otherwise — can be called upon to generate
space to ameliorate crowding and protrusion. Rapid maxillary expansion
has been around for well over a century and commonly is used to facilitate
non-extraction treatment. Despite its popularity, it apparently has little
long-term effect on perimeter. For example, Geran and associates (2006)
conducted a prospective study of the long-term effect of acrylic splint rap-
id maxillary expansion (RME). They reported that RME produces a lesser
reduction in perimeter than was seen in untreated controls (characterized
as a “relative increase” or an “additional maintenance”); however, in ab-
solute terms, there was no “increase” – on average, there was no change
in perimeter in the maxillary arch and a 3.6 mm loss in the mandible. In
terms of space, it apparently was not a retreat, but rather an advance to the
Iſeaſ.
Further, the literature, including the recent spate of randomized
clinical trials (RCT), seems to be converging on the conclusion that two-
phase “growth-modification” treatments have little effect above and be-
44
Johnston
yond what would be expected of more conventional treatments (Ghafari
et al., 1998; Wheeler et al., 2002; Tulloch et al., 2003; O’Brien et al.,
2003a,b). For a variety of reasons, these studies have generated heated de-
bate and have been subjected to sharp criticism. In my opinion, however,
many of the arguments demonstrate only that the critics have missed the
point of randomization. Admittedly, the RCTs can be criticized for their
definition of “Class II,” for their lack of blinding, for what must be a truly
remarkable Hawthorne effect, for costing too much, even for arrogance
(Johnston, 1998a, 2002). Moreover, because of differential morbidity,
many of our most important treatment comparisons (say, surgery versus
orthodontics) will prove difficult, if not impossible, to address by way of
fully-informed randomization. In the present context, however, random-
ization confers advantages that cannot be ignored. Whether or not you or
someone else could have assigned the treatments more appropriately in
isolated instances is beside the point: the RCTs were designed to test for
an effect. Moreover, it is probable that in the conduct of these studies, all
appliances were portrayed in the best possible light. Had one treatment or
another been able, on average, to do something that the others could not,
these prospective studies certainly would have had the power to detect it.
In this event, the resulting knowledge could then be applied clinically, not
randomly as in the RCT, but rather by “wet gloved” clinicians according
their own clinical expertise and experience. In the end, the RCTs’ real sin
is not randomization; it is their failure to find any major impact of func-
tional appliances on growth. Shoot the messenger.
The structured review by Cozza and associates (2006) and the ac-
companying comments provide something of a contemporary point/coun-
terpoint on the related subject of mandibular “growth modification.” On
balance, I think it fair to suggest that, if functional appliances have any
effect at all, it must be quite small (see the averaged cephalometric su-
perimpositions in Livieratos and Johnston, 1995). No matter: it really
is not unnecessary to invoke growth modification to explain the action of
functional appliances; the normal pattern of growth and the elimination of
dentoalveolar compensation is enough (Johnston, 1998b).
At this juncture, it would perhaps be well to recall the charge of
this symposium: “Is the benefit worth the burden?” If single-stage, non-
extraction treatments seem to produce about the same effect as “early,”
two-phase, non-extraction alternatives (Johnston et al., 1991), and given
that there is no proven means of increasing basal arch perimeter as a sub-
stitute for extraction, it can be argued that two-phase growth modification
45
If Wishes Were Horses
is a “weakly dominated” strategy, that is, a strategy that is at least tied and
occasionally beaten by some alternative. In decision theory, dominated
strategies generally are inadmissible. This brief invocation of expected
utilities and the various common treatment strategies assumed that the
question of “benefit” is to be assessed from the standpoint of the patient.
Although this interpretation seems to go without saying, things are not
always so straightforward.
If a hypothetical extraction/non-extraction payoff matrix is viewed
from the standpoint of an embattled contemporary orthodontist, an entirely
different strategy might emerge inadvertently. Given the occasionally mis-
guided demands of referring dentists and the fear of litigation, the clinician
might seek to minimize the maximum personal loss. If the maximum loss
is thought to be a lawsuit whose theory is that extractions caused TMD
or a “dished-in” profile, then the minimax strategy might be to employ a
non-extraction treatment. Under these circumstances, the problem is that
an optimal strategy for the orthodontist might not be particularly beneficial
for the patient. For example, given a fear of extraction, there are many sal-
able substitutes; however, it remains to be seen whether any of them are,
from the standpoint of the patient, an effective substitute. For example, at
0.7 mm of perimeter gain for each millimeter of expansion, it would take
more than 10 mm of stable expansion to duplicate the more-or-less routine
effect of first-bicuspid extraction. As a result, for the protrusive, crowded
patient, a decision to employ two-stage, non-extraction, growth-modifica-
tion therapy may well be of greater benefit to the orthodontist than to the
patient. If it is to be avoided, this potential conflict of interest first must
be recognized.
I have not always been a skeptic. Indeed, my father occasion-
ally found it necessary to temper my enthusiasm for obviously improbable
enterprises by saying, “If wishes were horses, beggars would ride.” So
it is with much that is sold and bought in contemporary orthodontics. If
wishes were horses, we would be able to grow jaws, abandon extraction,
and make everyone look like Julia Roberts or Tiger Woods. In ortho-
dontics, however, the wish is not tantamount to the deed. For example,
people actually seem to prefer profiles that are much less protrusive than
those that are encountered in the outer reaches of the non-extraction-at-
any-cost spectrum (see, for example, Scott and Johnston, 1999; Bowman
and Johnston, 2000, 2002). The remainder of the argument, namely that a
decision to extract bicuspids is a wanton act of malpractice that guarantees
temporomandibular dysfunction (TMD) and overly flat profiles, although
46
Johnston
specious, completes the popular justification for a detour down the path of
therapeutic least resistance.
With respect to the profile, normal growth has more of an im-
pact than bicuspid extraction (Paquette et al., 1992; Luppanapornlarp and
Johnston, 1993). For the average adolescent white patient, growth will
increase the lower lip to E-plane distance by about 1 mm. Post-treatment,
growth will flatten the profile an additional 2.5 mm. A decision to extract
first bicuspids will, on average, produce an additional 2 mm of flattening
during treatment. Thus, when the patient reaches adulthood, a non-extrac-
tion profile will have changed a total of about 3.5 mm compared to 5.5
mm for the first-bicuspid-extraction patient. It is perhaps appropriate to
emphasize here that the difference on average is not two inches or two
feet; it is 2 mm; in black patients, it may be even less (Hagler et al., 1998).
Clearly, there are many patients for whom 2 mm would be the straw that
breaks the camel's back; however, for the protrusive, crowded patient, a
little straightening of the profile (and with it a chance to achieve a long-
term improvement in irregularity) may well be a godsend. Moreover, from
the standpoint of PAR scores, “clear-cut” extraction patients (that is, those
with considerable crowding and protrusion) seem to show the greatest im-
provement relative to other categories of Class II patients (Hannapel and
Johnston, 2002). Finally, contrary to untutored opinion, these benefits are
not counterbalanced by obvious functional pathology: a long-term recall
of carefully-matched extraction and non-extraction patients, both black
and white, revealed no significant differences in TMD prevalence (Beattie
et al., 1994; Hagler et al., 1998).
From the time of Angle to the present, we occasionally have flirted
with universal strategies. Thirty years ago, some sort of bicuspid extrac-
tion was for many the one treatment that fits all; today, “early” two-phase,
non-extraction treatments enjoy great popularity. In the end, however, we
have to obey the laws of physics and biology. I can see no way to avoid
the conclusion that some patients should be treated early and some late;
some with extraction and some without. Accordingly, patients who are
Susceptible to one strategy but who are treated by clinicians who have
pledged their lives and practices more or less exclusively to another (either
extraction or non-extraction) will pay twice: once in dollars and again in
regret — the difference between what they got and what they could have
gotten, given the most appropriate treatment. It is a Sobering thought.
47
If Wishes Were Horses
ENVOI
It has been the purpose of this short communication to argue that
a reliance on convenient, occasionally self-serving, mythology rather than
hard data distorts the decision-making process. The title of this sympo-
sium seems to demand that we discuss the benefits and burden of “early
treatment” and somehow reach conclusions that are of general benefit to
the patient. Unfortunately, this charge is beyond us. It is the patient who
must weigh burden and benefit; our job is to help in the deliberations and
ultimately to prosecute the strategy that, for the patient, has the optimal
expected gain. It is for this reason that I have approached the subject from
the standpoint of formal decision theory, both from the perspective of the
patient and the orthodontist.
Treatment decisions are fueled by data; however, our research —
at least the research that people pay attention to — often seems designed
to give the appearance of progress without at the same time leading to
any firm conclusions. When everything works well enough to support a
practice, conclusive evidence can be an inconvenient, potentially embar-
rassing distraction. Perhaps it is for this reason that the “evidence-based
dentistry” movement is so controversial. But in the absence of data, how
can there ever be truly informed consent?
I would suggest that the first step in helping a patient to make an
informed decision is to be informed ourselves. This is our burden for the
benefit of the patient.
REFERENCES
Am J Orthod Dentofacial Orthop. No authors, 2006;129 (4 Supply:S45-
S90.
Beattie JR, Paquette DE, Johnston LE Jr. The functional impact of ex-
traction and non-extraction treatments: A long-term comparison in
“borderline,” equally-susceptible Class II patients. Am J Orthod Den-
tofacial Orthop 1994;105:444-449.
Bowman SJ, Johnston LE Jr. The esthetic impact of extraction and non-
extraction treatment on Caucasian patients. Angle Orthod 2000;70.
3–10.
Bowman SJ, Johnston LE Jr. Much ado about facial esthetics. In: Mc-
Namara JA Jr, Kelly KA, eds, Treatment Timing: Orthodontics in
Four Dimensions. Craniofacial Growth Series, Department of Or-
48
Johnston
thodontics and Pediatric Dentistry, School of Dentistry, and Center
for Human Growth and Development, The University of Michigan,
Ann Arbor, 2002:39: 199-215.
Cozza P. Baccetti T. Franchi L., De Toffol L. McNamara JA Jr. Mandibu-
lar changes produced by functional appliances in Class II malocclu-
sion: A systematic review. Commentary by Johnston LE Jr. Authors’
response. Am J Orthod Dentofacial Orthop 2006; 129:599.e1-e12,
el-e6.
Geran RG, McNamara JA Jr, Baccetti T, Franchi L, Shapiro LM. A pro-
spective long-term study on the effects of rapid maxillary expansion in
the early mixed dentition. Am J Orthod Dentofacial Orthop 2006; 129:
631-640.
Ghafari J, Shofer FS, Jacobsen-Hunt U, Markowitz DL, Laster LL. Head-
gear versus functional regulator in the early treatment of Class II, Di-
vision 1 malocclusion. Am J Orthod Dentofacial Orthop 1998; 113:
51-61.
Gupta IM. The costs and benefits of orthodontic treatment: Patients’ val-
ues compared to professional judgments. Thesis. School of Dentistry,
The University of Michigan, Ann Arbor, 2003.
Hagler BL, Lupini J, Johnston LE Jr. Long-term comparison of extraction
and non-extraction alternatives in matched samples of African Ameri-
can patients. Am J Orthod Dentofacial Orthop 1998; 114:393-403.
Hannapel ED Johnston LE Jr. Extraction vs. Non-extraction: PAR-score
reduction as a function of initial susceptibility. Prog Orthod 2002:3:
1–5.
Johnston LE Jr. A comparative analysis of Class II treatments. In: Vig
PS, Ribbens KA, eds, Science and Clinical Judgment in Orthodontics.
Craniofacial Growth Series, Center for Human Growth and Develop-
ment, The University of Michigan, Ann Arbor 1986; 19:103-148.
Johnston LE Jr. Let them eat cake: The struggle between form and
substance in orthodontic clinical investigation. Clin Orthod Res
1998a;1:88-93. .
Johnston LE Jr. Growing jaws for fun and profit: A modest proposal. In:
McNamara JA, Jr. ed., What Works, What Doesn't, and Why. Mono-
graph 35, Craniofacial Growth Series, Center for Human Growth and
Development, The University of Michigan, Ann Arbor, 1998b;35:
63–86.
Johnston LE Jr. Moving forward by looking back: “Retrospective” clini-
cal studies. J Orthod 2002:29:221-226.
49
If Wishes Were Horses
Johnston LE Jr. If wishes were horses: Functional appliances and growth
modification. Prog Orthod 2005;6:36–43.
Johnston LE Jr, Paquette DE, Beattie JR, Cassidy DW Jr, McCray JF,
Killiany DM. The reduction of susceptibility bias in retrospective
comparisons of alternative treatment strategies. In: Vig KD, Vig PS
eds, Clinical Research as the Basis of Clinical Practice. Craniofa-
cial Growth Series, Center for Human Growth and Development, The
University of Michigan, Ann Arbor 1991:25:155-177.
Livieratos FA, Johnston LE Jr. A comparison of one- and two-stage non-
extraction alternatives in matched Class II samples. Am J Orthod
Dentofacial Orthop 1995; 108:118-131.
Luppanapornlarp S, Johnston LE Jr. The effects of premolar-extraction
treatment: A long-term comparison of outcomes in “clear-cut” ex-
traction and non-extraction Class II patients. Angle Orthod 1993;63:
257-272. &
Neumann J. von, Morgenstern O. Theory of Games and Economic Behav-
ior. Princeton, NJ: Princeton University Press, 1944.
O’Brien K, Wright J, Conboy F. Sanjie Y, Mandall N, Chadwick S, Connolly
I, Cook P. Birnie D, Hammond M, Harradine N. Lewis D. McDade C,
Mitchell L, Murray A, O’Neill J, Read M, Robinson S, Roberts-Harry
D, Sandler J, Shaw I. Effectiveness of treatment for Class II maloc-
clusion with the Herbst or twin-block appliances: A randomized, con-
trolled trial. Am J Orthod Dentofacial Orthop 2003a; 124:128-137.
O’Brien K, Wright J, Conboy F. Sanjie Y, Mandall N, Chadwick S, Con-
nolly I, Cook P. Birnie D, Hammond M, Harradine N, Lewis D, Mc-
Dade C, Mitchell L, Murray A, O’Neill J, Read M, Robinson S, Rob-
erts-Harry D, Sandler J, Shaw I. Effectiveness of early orthodontic
treatment with the twin-block appliance: A multicenter, randomized,
controlled trial. Part I: Dental and skeletal effects. Am J Orthod Den-
tofacial Orthop 2003b;124:234-243.
Paquette DE, Beattie JR, Johnston LE Jr. A long-term comparison of non-
extraction and bicuspid-extraction edgewise therapy in “borderline”
Class II patients. Am J Orthod Dentofacial Orthop 1992;102:1-14.
Skinner B. Treatment need: A comparison of personal and professional
judgments. Thesis, The University of Michigan, Ann Arbor, 2000.
Straffon D. The American Board of Orthodontics grading system: A vali-
dation against contemporary orthodontic opinion. Thesis, The Uni-
versity of Michigan, Ann Arbor, 2001.
50
Johnston
Stephenson W. The Study of Behavior. Chicago, The University of Chi-
cago Press, 1953.
Tulloch JF, Proffit WR, Phillips C. Outcomes in a 2-phase randomized
clinical trial of early Class II treatment. Am J Orthod Dentofacial
Orthop 2002; 121:9-17.
Wheeler TT, McGorray SP, Dolce C, Taylor MG, King GJ. Effectiveness
of early treatment of Class II malocclusions. Am J Orthod Dentofacial
Orthop 2004; 125:657-667.
51
52
ANOTHER PERSPECTIVE ON EARLY
ORTHODONTIC TREATMENT
Sheldon Baumrind
There is a long-standing controversy as to which of several approaches
is best for the correction of malocclusions among patients who present
for orthodontic evaluation in the early mixed dentition stage. In the past
several years, it has been inferred by some observers, on the basis of ran-
domized clinical trials conducted at the University of North Carolina and
the University of Florida (NC and FL), that early intervention is without
consequential value in the treatment of mixed dentition malocclusions
(Wheeler et al., 2002; Tulloch et al., 2004). Indeed, one leading orth-
odontic educator recently expressed the opinion to one of our orthodontic
residents that mixed dentition treatment per se is tantamount to malprac-
tice. And, we have heard through the grapevine that this was not a unique
event. It brings to mind the statement attributed to H.L. Mencken: “For
any complex problem, there is an answer that is clear, simple, straight-
forward, and wrong!”
One of the abiding characteristics of every orthodontic clinician
is the strong and aggressively held conviction that the method he or she
uses for the correction of any patient’s malocclusion is the best currently
available for the treatment of that particular patient. Indeed, that strength
of conviction is not an undesirable trait in a clinician. Any orthodontist
who does not believe that the treatment he or she is delivering is the best
available for the patient being treated should not be treating that patient!
In addition to the obvious ethical considerations, the clinician’s conviction
of the appropriateness of his or her treatment plan is one of the strongest
motivators for patient compliance — perhaps the greatest uncontrollable
variable in orthodontic treatment.
Indeed, it does seem self-evident that as long as true uncertainty
exists as to which treatment approach is really best for the treatment of a
particular patient, any thoughtful and ethical clinician will think long and
hard before saying that another orthodontist’s clinical approach to treat-
ment is improper. Certainly the authors of the NC and FL studies have
not made such global judgments. Both groups designed carefully crafted
studies with specific, limited goals, and each study was conducted un-
der carefully controlled conditions. Many valuable conclusions have been
drawn from both studies, but the question is to what patient populations
53
Another Perspective
and treatment conditions are these results generalizable. For our purposes,
“generalizability” can be defined as the extent to which the results ob-
served under the specific conditions of an experimental study are equiva-
lent to those encountered in the less structured environment of every day
clinical experience.
As consumers of research information, clinicians are entitled to
ask to what extent the design of any study is such that findings from it are
directly applicable to the conditions of their own practices. In the present
debate, one may ask more specifically to what extent the designs of the NC
and FL studies are such that conclusions drawn from them are applicable
to the usual conditions of clinical treatment in the offices of experienced
orthodontists who advocate and deliver mixed dentition treatment for an
appropriate subset of their patients.
The designs of the prospective studies at the Universities of North
Carolina and Florida were remarkably similar in intent. Both were care-
fully targeted and restricted to the study of subjects with moderately se-
Vere Class II malocclusions. Both were focused essentially on determining
which of two kinds of Phase 1 mechanics (headgear vs. intraoral appli-
ances of the bionator type) was more effective in producing “growth modi-
fication,” a somewhat under-defined term that appears, in this case, to refer
primarily to a possible acceleration of mandibular growth.
In each of these studies, there was an initial intervention of fixed
duration (Phase 1) during which each subject was treated with one of two
appliances. This initial intervention was followed by a hiatus in which lit-
tle or no retention was employed. A second phase of full bonded treatment
then was delivered to both groups of Phase 1 patients (the early treatment
groups) and to a third group of patients whose original malocclusions were
of equivalent severity but who had not received early treatment. Treatment
was highly standardized to reduce the variability in outcomes.
The two main observations at each site appear to be 1) that the
corrections of the malocclusions that occurred during Phase 1 were sub-
stantially lost during the uncontrolled, between-stages hiatus, and 2) that
at the end of Phase 2, which was the definitive course of full treatment,
there were, on average, no substantial differences between the patients in
the two early treatment groups and the patients in the third group who had
received only the later course of full treatment.
There is every reason to believe that these studies were con-
ducted extremely carefully and that much data of normative value can be
garnered from them. There also is reason to believe, however, that their
54
Baumrind
designs and goals were not such that the findings can be generalized glob-
ally to the treatment delivered by experienced mixed dentition clinicians.
In the North Carolina study, for example, the subjects were enrolled solely
on the basis of a preliminary clinical exam that determined that overjet
exceeded 7 mm and that the subjects were “developmentally at least a
year before their peak pubertal growth.” Initial records were taken later,
after each child had been randomized to one of the three treatment groups.
In the Florida study, in order to control for differences in operator skill,
several clinicians alternated treatment of each patient at successive visits.
These and other deviations from standard clinical practice may have aided
in controlling bias and reducing outcome variability, but they reduced the
generalizability of the findings because they departed substantially from
the usual treatment regimen of most experienced advocates of mixed den-
tition treatment.
In the graduate program in orthodontics at the Arthur A. Dugoni
School of Dentistry of the University of the Pacific (UOP), all residents
participate in didactic and clinical programs specifically dedicated to the
study and delivery of early mixed dentition treatment. These programs
were developed over a period of more than thirty years and were based
originally on the thinking of Nance, Hahn, Dewel and other pioneers of
early intervention in the treatment of malocclusion. The courses originally
were developed under the supervision of Dr. Arthur Dugoni and currently
are taught by a team headed by Dr. Steven Dugoni. Treatment strategies in
this program (and, we believe, in the practices of many experienced clini-
cians who practice early treatment) differ from the models of the NC and
FL studies in at least the following ways:
• Mixed dentition treatment is not considered appropriate for
all patients who present with early mixed dentition malocclusions.
Instead, treatment is deferred until the adult dentition period for
approximately half of such patients on the basis of specific, proto-
col-defined characteristics. Failure to screen patients prior to ran-
domization in order to identify only those for whom early treat-
ment is considered appropriate by its advocates obviously would
distort experimental samples by introducing additional noise into
both the early and deferred treatment sub-samples.
• Early treatment is not restricted to patients with Class II
malocclusions. Consequential numbers of patients with Class I
malocclusions and smaller numbers of patients with Class III mal-
occlusions also are treated in the mixed dentition.
55
Another Perspective
• Patients are not dismissed between the end of the ini-
tial phase of treatment and the beginning of the final phase
of treatment. Rather, they enter a period of active supervi-
sion during which the gains of the initial phase of treatment
are maintained and extended by periodic visits and the use
of removable appliances such as headgear, adjustable lingual
arches, and removable retainers with auxiliary attachments.
• The essential role of early treatment in our therapeutic
model is not “growth modification” to the extent that that term
is taken to mean the encouragement of mandibular growth. The
goal is the complete correction of the existing malocclusion and
the achievement and maintenance of “ideal occlusion” as defined
in traditional Angle terms. Treatment continues until that goal is
achieved or until it becomes obvious that it cannot be achieved.
The usual duration of Phase 1 treatment is 12 to 18 months.
• Patients who receive early treatment are not assigned auto-
matically to a second phase of treatment. Their status is reevalu-
ated with a full, new set of records taken at or near the time of
eruption of the second molars. It is reported that at that time, be-
tween 30% and 50% of the early treatment patients are not in need
of further orthodontic treatment. Treatment for these patients is
considered complete without a second fixed appliance phase, al-
though they return for periodic re-examination during the remain-
ing period of active skeletal growth.
Based on a comparison of the above points with the available pro-
tocols of the NC and FL studies, we have concluded that the treatment
regimens of these studies were sufficiently different from those taught at
our school to make it inappropriate to generalize their results to the treat-
ment that our patients receive. Accordingly and motivated in large mea-
sure by the findings of our colleagues at NC and FL, we decided several
years ago to initiate a study at UOP of the usual course and outcome of the
full range of treatment delivered to patients who presented for treatment in
the early mixed dentition phase in the practices of members of the clinical
faculty at our school. The remainder of this paper documents the rationale
and some very early findings from a study of patients treated in the clinical
practice of Dr. Steven Dugoni who is head of the Mixed Dentition Clinic
at UOP. Other early findings from this study have been reported elsewhere
(Dugoni et al., 2006).
56
Baumrind
RESEARCH APPROACH
The design of the UOP study is markedly different from those of
the NC and FL studies. The NC and FL studies are prospective and experi-
mental in design, while the UOP study is retrospective and observational
(naturalistic) in design (Baumrind, 1993). The NC and FL studies were
designed essentially to find out which of two therapeutic devices yielded
superior results when equivalent patients were assigned to each treatment.
The UOP study is intended to measure the usual course and outcome of
treatment when an experienced clinician treats each patient with the thera-
peutic modality that the clinician considers best for that patient. &
To amplify briefly, the experimental designs of the NC and FL
studies standardized the therapy of all patients receiving treatment with
each type of appliance in order to reduce variability of outcome within
each treatment. Random sampling was used in these studies to assure that
the equivalent proportions of subjects with each different kind of Class
II malocclusion would be assigned to each treatment. Much effort was
focused on ensuring that all of the conditions of treatment (and especially
treatment duration) were the same for all early treatment patients. How-
ever, the possibility that different types of Class II malocclusion might
fare better with different kinds of treatment was not considered in the de-
sign. Thus, the object of the studies became, in effect, the determination of
which of the two narrowly structured treatments being evaluated was best
for the average Class II patient.
The observational approach of the UOP design started very dif-
ferently from the observation that real world clinicians assign patients
with different facial characteristics and different types of malocclusion
to different types of treatment. In our experience, this is true even within
the same Angle class (Moyers et al., 1980). Further, we recognized that
real world clinicians modify their treatment plans frequently during the
course of treatment as a function of differences between individual pa-
tients that only become apparent as the treatment process unfolds. For
these reasons, our goals at UOP are more modest than those of our col-
leagues at NC and FL. We do not seek the one treatment that is “best on
average” for a broad and non-homogeneous group of patients (e.g., the
entire universe of Class II patients “in general”). Rather, at least for the
present, we seek only to identify and describe with a minimum of bias
the usual course and outcome of orthodontic treatment when experienced
57
Another Perspective
orthodontists apply the treatment strategy that they consider appropriate
for each individual patient.
We, too, employ randomization in case selection, but with a dif-
ferent aim from that of the NC and FL studies. We use randomization to
select the cases in our sample: 1) to ensure that all patients in the practice
under consideration are sampled without conscious bias as to outcome,
and 2) to insure that the different types of malocclusion and of treatment
are represented in the sample in proportion to their actual incidence in
the practice. This research strategy explicitly involves acceptance of the
full range of variability among patients in diagnosis, treatment planning,
treatment progress, and treatment outcome, thus accepting the greater
variability that is characteristic of real world orthodontic treatment. Such
an approach increases sample variability, making it much more difficult
to identify differences between different treatments. By the same token,
however, it increases the generalizability of those observations of differ-
ences between groups that are found to be real beyond chance, i.e., “sta-
tistically significant.”
In observational studies such as the one we are conducting at UOP,
treatment is not modified or perturbed experimentally in order to facilitate
data acquisition. Instead, the aim is to acquire data that describe the usual
course and outcome of routine treatment across the range of experienced
clinicians. To that end, as our studies continue, we hope to add representa-
tive samples from the practices of other experienced clinicians who use
the same or different protocols for mixed dentition treatment.
For each patient at each time point, data are acquired and evalu-
ated from each of the primary physical records employed in orthodontic
treatment — lateral cephalograms, intraoral x-rays, study casts, and facial
photographs (see Fig. 1). Because experienced clinicians use records of
all four types to diagnose, plan, and evaluate treatment, we believe that
data from all four kinds of record are needed to detect and describe ap-
propriately the variability among individual patients.
We acknowledge the limitation imposed by this requirement of
full records for each patient at each critical time point. We recognize that
in practice it introduces some biases into our sampling procedure, be-
cause many patients do not have full records available in retrospective
samples. This is indeed a classical limitation of retrospective studies, but
since we cannot gather data unless we have records, it is a limitation we
believe we must accept for the present. In fact, even our “full records”
are not as complete as we would wish them to be, because Dr. Dugoni,
58
Baumrind
Figure 1. Basic types of physical records.
like most other private practitioners, typically does not take complete re-
cords at initial visits of those patients for whom treatment is being de-
ferred. The long-term remedy is to conduct carefully designed and con-
trolled prospective studies of observational design. For that reason, we
have now started such a study in the graduate orthodontic clinic at UOP
Description of the UOP Retrospective Study
The first step in the UOP retrospective study was to identify all
patients who originally had presented for evaluation in the early mixed
dentition stage in Dr. Dugoni's practice during the period between 1986
and 1998. The charts of all of the patients who satisfied these criteria were
Sorted in random order and the first 500 were examined for completeness
of records. At presentation, each patient had received a full examination,
and a decision had been made as to whether he or she could benefit from
early orthodontic intervention. The charts of patients with incomplete re-
Cords and those with other than Class I or Class II malocclusions were
excluded. The remaining subjects were classified into two groups.
Group 1 was comprised of patients for whom early intervention
Was not considered to be optimal. These patients were monitored peri-
odically until the eruption of the adult dentition, at which time conven-
tional, full bonded orthodontic treatment was initiated. Group 2 included

59
Another Perspective
patients for whom early intervention was considered to be beneficial.
These patients received an immediate phase of treatment of limited dura-
tion, the intent of which was to correct the existing malocclusion in such
a way as to produce optimal conditions for subsequent occlusal develop-
ment. Group 2 patients then were monitored periodically until the eruption
of the second molars. At that time, they were re-examined and a new, full
set of records were taken. A decision then was made as to whether further
treatment was indicated. A number of the early treatment patients were
deemed not to require additional treatment and were designated Group
2.1. The remaining early treatment patients received a second phase of full
bonded treatment and were designated Group 2.2.
Data acquisition currently is in progress for the first sample of 81
Class I and Class II patients with complete records. Preliminary statistical
analyses have been performed on quantitative data from the cephalograms
of all patients and from the intraoral images and study casts of approxi-
mately half of this sample.
(We now believe it was a mistake to restrict our original sample to
Class I and Class II Angle cases. To expand our information base, we are
therefore screening another group of 500 randomly sampled patients who
first presented at an early age in Dr. Dugoni’s office for completeness of
records. In this second sample, we are including all patients with complete
records regardless of Angle class or other special characteristics. In addi-
tion, we are developing our statistics on prevalence of different classifica-
tion categories on the basis of all 500 patients being screened regardless of
completeness of records.)
SOME EARLY FINDINGS AND THEIR
POSSIBLE IMPORT
Data analysis in the UOP study is at a very early stage, but some
of our preliminary findings are interesting. Of the 81 patients in our first
sample, 52 patients received early treatment and treatment was deferred
for 29 patients. Among the 52 for whom early treatment was recommend-
ed, the age at first visit (mean + SD in years) was 8.20 + 0.74 and the age
at start of treatment was 8.57 = 0.92. Among the 29 patients for whom it
was suggested that treatment be delayed, the age at first visit was 8.70 +
0.83, and the age at start of treatment was 11.36 + 1.51.
Information on the distribution of subjects within the sample by
sex, extraction/non-extraction status and Angle class is supplied in Table
1 in the form of a series of small contingency tables that are labeled 1A, B
and C and 2A, B and C.
60
Baumrind
Table 1. Composition of the sample by sex, extraction status and Angle
class. 1A-C: Full sample. 2A-C: Early treatment cases only.
1A - Distribution by Sex

Male | Female | Total
Group 1 - Deferred Treatment 16 13 29
Group 2 - Early Treatment| 16 36 52
Totall 32 49 81
1B - Distribution by Extraction/Non-extraction
Non-Ext| Ext Total
Group 1 - Deferred Treatment| 21 8 29
Group 2 - Early Treatment| 41 11 52
Total | 62 19 81
1C - Distribution by Angle Class
Class || Class || |Class || || Total
Group 1 - Deferred Treatment 15 14 O 29
Group 2 - Early Treatment| 20 31 1 52
Total || 35 45 1 81
2A - Distribution by Sex
Male | Female | Total
No Phase 2 Treatment Needed 8 12 20
2 - Further Treatment Indicated 8 24 32
Total | 16 36 52
2B - Distribution by Extraction/Non-extraction
- Non-Ext| Ext Total
No Phase 2 Treatment Needed| 20 O 20
2 - Further Treatment Indicated | 21 11 32
Total || 41 11 52
2C - Distribution by Angle Class
Class || Class || |Class || || Total
No Phase 2 Treatment Needed 8 12 O 20
2 - Further Treatment Indicated 12 19 1 32
Totall 20 31 1 52





The three subtables 1A-C compare the early treatment group
with the deferred treatment group. Note that in 1A, there are significantly
more girls than boys in the early treatment group. The values in 1B in-
dicate that the ratio of extraction to non-extraction cases in the deferred
group (8/29 = 28%) is slightly higher than that in the early treatment
group (11/52 = 21%), but this difference is not statistically significant. In
61
Another Perspective
1C, perhaps the most important information is that 20 of the 52 early treat-
ment patients (~40%) had Class I malocclusions. Note that these patients
would not have qualified for inclusion in the NC or FL studies, because
those studies were limited to Class II subjects. It also is mildly interesting
that one patient who originally was classified as Angle Class I later was
determined by measurement to fall into the supposedly excluded Class III
Category.
Part 2 of Table 1 stratifies the 52 early treatment cases. It distin-
guishes between the 20 patients in Group 2.1, for whom further treatment
was considered unnecessary at the review following the eruption of the
premolars and second molars, and the 32 patients in Group 2.2, for whom
a second phase of full bonded treatment was recommended. One has the
impression (subtable 2A) that girls were more likely to receive a second
phase of treatment than boys, but the observed difference does not reach
statistical significance. In subtable 2B, it is dramatically clear that more
extractions were performed in the two-phase patients than in the single
phase, early treatment patients (a ratio of 11:0), although on reflection this
is hardly surprising. And in subtable 2C, it appears that Angle Class I cases
seem to have been as likely to be treated successfully in a single phase as
Class II patients (about 40% for each Angle class).
We reported above that the NC and FL studies stated that the ef-
fects of early treatment were transient and had dissipated by the time of
review for second phase treatment. Table 2 addresses this issue of stability
of treatment results following early treatment. For a subset of early treat-
ment patients in the UOP study for whom both cephalometric and study
cast data were available for analysis, Table 2 summarizes changes in a
number of representative variables that occurred during the time between
initial record taking and record taking at the evaluation for second phase
treatment. Significance values of t-tests for the statistical significance of
change between initial and second record-taking time points are shown for
each variable. To facilitate interpretation of these findings, variables with
statistically significant changes are highlighted. Means and standard de-
viations that perhaps give a more meaningful estimate of the actual mag-
nitude of change are also supplied.
Table 2. Changes that occurred during early treatment sorted by
Angle class, with t-tests showing the statistical significance of the
changes between “initial record taking” and record taking at the
evaluation for second phase treatment. —P
62
Baumrind
Angle Class I Cases Angle Class Il Div 1 Cases
Lateral Ceph
Variables
Lateral Ceph
Mean Std Dev t Value Prs. It Variables
Mean Std Dev t Value Pr: it
Study Cast
Variables
Study Cast
Mean Sto Dev tValue Prº. It Variables
Mean Std Dev twafue Pr: |t|
For Angle Class I Single Phase Cases: For Angle Class I Two Phase Cases:
Lateral Ceph
Variables
Lateral Ceph
Variables Mean Std Dev t Value Prº ſt
Mean Std Dev t Value Prº ſt
Study Cast
Variables
Study Cast
Variables Mean Std Dev t Value Pr: |t|
Mean Sto Dev twalue Pr: it
For Angle Class Il Single Phase Cases: For Angle Class Il Two Phase Cases:
Lateral Ceph
º N Mean Std Dev t Value Pr: |t|
Variables
Study Cast
Variables
Study Cast
Mean Std Dev t Value Pr: |ti Variabies
Mean Std Dev t Value Prº t






63
Another Perspective
Even from the small number of variables and the small size of
this initial sample, it can be seen that there are consequential differences
between the pretreatment records and the evaluation-for-second-phase re-
cords. Among the variables reported here, the cephalometric changes for
the Class II patients are more dramatic than those for the Class I patients,
which should not come as a surprise.
In evaluating findings such as these, one needs to remember that
the observed changes are not the product of growth alone. This truism is
particularly important with respect to the study cast data, since the early
treatment period is one of active growth and dramatic spontaneous chang-
es in the dentition. Yet these caveats notwithstanding, these very early
findings seem sufficient to demonstrate that there are real differences in
the observed effects of early treatment between this study and those of FL
and NC.
It may be argued that the UOP sample was retained and that the
patients received further optimizing care between the end of Phase 1 and
the review at the beginning of Phase 2. But that is precisely the point. The
treatment of the UOP patients was consequentially different from that of
the patients in either the NC or the FL study. Hence, the NC and FL find-
ings cannot be readily generalized to the treatment of the UOP patients.
Nor, we strongly suspect, can they be appropriately generalized to the
treatment of patients in the practices of most other experienced advocates
of early treatment.
Indeed, the fact that few meaningful differences were detected be.
tween the pretreatment records and those taken at the beginning-of-Phase
2 review in the NC and FL studies may have further consequences to our
understanding of the implications of these studies. Such a finding would
seem to imply that at the beginning of the final phase of treatment in these
two studies, there were no substantial differences between the early treat-
ment and deferred treatment patients. To the extent that this was the case,
one might expect there to be few differences in outcome between the two
groups at the end of the final phase of treatment.
This suggestion is not meant to derogate in any way the impor-
tance of the contributions of our colleagues at Florida and North Caro-
lina. Against considerable odds, they and Ghafari and coworkers at the
University of Pennsylvania (Ghafari et al., 1998) have conducted the
first successful large-scale prospective clinical studies of orthodontic
treatment in the United States. It should be the fervent hope of all or-
thodontists that other prospective studies will follow before too long. We
believe that the prospective study of mixed dentition treatment currently
64
Baumrind
being performed in the Mixed Dentition Clinic at UOP will be one contri-
bution to that end.
CONCLUSION
The available evidence indicates that definitive judgments con-
cerning the efficacy of different strategies for the treatment of early mixed
dentition malocclusions are vastly premature. Indeed, we do not have even
yet a strong consensus as to what the proper outcome variables should be.
That being the case, continued investigation of many alternative treatment
strategies is in order. The growth mechanisms and the responses to thera-
peutic intervention that we are trying to investigate are indeed complex
— far more complex, for example, than rocket science. It is proper that
we seek parsimonious answers, but as we do so, we need to remember
the injunction attributed to Einstein: “Things should be made as simple as
possible, but not simpler.”
To achieve a greater and more sophisticated understanding of cra-
niofacial development and the response to treatment, our specialty clearly
needs to expand its collective information base. We can do so by conduct-
ing new clinical studies that test alternative hypotheses and by sharing
the information we already have among all interested professionals. To
achieve the first of these ends, we have implemented a controlled, longi-
tudinal, observational study of mixed dentition treatment in the graduate
orthodontic clinic at UOP. To achieve the second, we have been posting
the raw data and individual case findings from all of our laboratory’s stud-
ies on our web site to the extent that resources are available. Early materi-
als from our retrospective mixed dentition study can currently be accessed
at www.CRIL.org, Study 1030.
REFERENCES
Baumrind S. The role of clinical research in orthodontics. Angle Orthod
1993;63:235-240. r
Dugoni S, Aubert M, Baumrind S. Differential diagnosis and treatment
planning for early mixed dentition malocclusions. Am J Orthod Den-
tofacial Orthop 2006;129:S80-S81.
Ghafari J, Shofer FS, Jacobsson-Hunt U, Markowitz DL, Laster LL.
Headgear versus function regulator in the early treatment of Class II,
division 1 malocclusion: A randomized clinical trial. Am J Orthod
Dentofacial Orthop 1998; 113:51-61.
65
Another Perspective
Moyers RE, Riolo ML, Guire K, Wainwright B. Differential diagnosis of
Class II malocclusion. Am J Orthod 1980:78:477-495.
Tulloch C, Proffit WR, Phillips C. Outcomes in a 2-phase randomized
clinical trial of early Class II treatment. Am J Orthod Dentofacial Or-
thop 2004;125:657-667.
Wheeler TT, McGorray SP, Dolce C, Taylor MG, King GJ. Effectiveness
of early treatment of Class II malocclusion. Am J Orthod Dentofacial
Orthop 2002; 121:9-17.
66
EFFECTIVENESS AND TIMING OF MANDIBULAR
CHANGES PRODUCED BY FUNCTIONAL JAW
ORTHOPEDICS: A SYSTEMATIC REVIEW
Lorenzo Franchi
Tiziano Baccetti
Franka Stahl
Paola Cozza
Time itself is of little value.
It is what you make of it that counts.
Common Italian saying
Class II malocclusion occurs in about one third of the population of the
United States (Kelly and Harvey, 1977; Proffit et al., 1998), with man-
dibular skeletal retrusion being diagnosed most consistently. Functional
jaw orthopedics (FJO) is the treatment of choice to stimulate mandibular
growth to correct this type of skeletal and occlusal disharmony (McNama-
ra and Brudon, 2001). Although many animal studies have demonstrated
that skeletal mandibular changes can be produced by posturing the man-
dible forward (McNamara, 1973; McNamara and Bryan, 1987; Xiong et
al., 2004), the effects on humans are more equivocal and controversial. A
wide range of treatment protocols, sample sizes, and research approaches
has led to disparate outcomes in studies on human subjects. There have
been very few comprehensive and systematic reviews of the effectiveness
of FJO in the scientific dental journals. One such attempt was carried out
by Chen and co-workers (2002) who analyzed the relevant literature for
the years 1966 to 1999 using a Medline search strategy. They limited the
study to data taken from randomized clinical trials (RCTs) only. The re-
sults were inconclusive. The main difficulty when analyzing the RCTs was
the inconsistencies in measuring treatment outcome variables. In addition,
treatment duration varied among studies, and treatment groups were com-
pared either to untreated control groups or to subjects undergoing other
forms of treatment.
A randomized clinical trial is the standard for comparing alterna-
tive treatment approaches. However, it is not easy to do this in orthodon-
tics for several reasons. It is difficult to gather a large number of patients
67
Effectiveness and Timing of Mandibular Changes
with a specific occlusal deviation, and there is the ethical issue of leav-
ing a group of patients untreated for a rather long time. In addition, there
are several requirements for such a trial (Antczak et al., 1986; Jadad et
al., 1996) that are obviously not applicable to orthodontics, e.g., neither
patient nor clinician can be “blinded” as to who has or has not received
treatment. These considerations suggest that a rational, systematic review
should include longitudinal prospective and retrospective controlled clini-
cal trials (CCTs) in order to increase the amount of scientific information
available pertaining to treatment effects induced by orthodontic appliances
(Petrén et al., 2003). Moreover, all of the recent investigations of treatment
outcomes of functional appliances not included in the Chen study should
be re-examined to supplement the data analyzed by Chen and associates
(2002). A systematic review of this kind should be limited to those clinical
trials that compare treated Class II subjects to matched untreated Class II
subjects. For example, there are studies that demonstrate that mandibular
growth in Class II subjects differs significantly from that of subjects with
normal occlusions (Harris, 1962; Buschang et al., 1986, 1988).
Another recent study (Stahl et al., 2006) found that total mandibu-
lar length (Co-Gn) in Class II subjects increased about 3 mm less than
in subjects with normal occlusions throughout the circumpubertal period.
Two of the 3 mm of growth difference occurred during the pubertal growth
spurt. It is important to note that the selection of studies that included
untreated Class II controls allowed for the integration of the outcomes
of CCTs and RCTs, because the untreated controls had, by definition, the
same type of malocclusion as the treated subjects.
We report here the results of a systematic review (Cozza et al.,
2006) that was undertaken to answer these specific questions:
1. Does the mandible grow more in Class II subjects
who have been treated with functional appliances
than it does in untreated Class II subjects?
2. Does timing of intervention make a difference in
terms of mandibular changes produced by FJO'?
The original findings of this systematic review (Cozza et al., 2006) have
been re-analyzed on the basis of the outcomes of a longitudinal study on
growth in untreated Class II subjects compared to subjects with normal
occlusions (Stahl et al., 2006).
68
Franchi et al.
DOES THE MANDIBLE GROW MORE IN CLASS II
SUBJECTS WHO HAVE BEEN TREATED WITH
FUNCTIONAL APPLIANCES THAN IT DOES IN
UNTREATED CLASS II SUBJECTS?
Search Strategies
The strategy for performing the present systematic review was
influenced mainly by the National Health Service (NHS), Center for Re-
views and Dissemination (Green and Higgins, 2005). A Medline (Entrez
PubMed, ncbi.nim.nih.gov) literature search was carried out to identify all
of the studies that examined mandibular growth that were written in Eng-
lish during the period from January 1966 to December 2005. We used the
Medical Subject Headings (MeSH) term “malocclusion, Angle Class II”
cross-referenced with the MeSH term “functional appliances” to identify
the appropriate studies. A search in the Cochrane Clinical Trials Register
(cochrane.org/reviews) also was performed.
Selection Criteria
Studies that reported data on mandibular growth were included if
they were randomized clinical trials, meta-analyses, controlled clinical tri-
als, or longitudinal studies (both prospective and retrospective) (Table 1).
These studies analyzed cephalometrically the effects of functional therapy
on mandibular dimensions (including mandibular length as measured by
using the anatomical point Condylion) with respect to untreated Class II
controls. No restrictions were set for sample size. Abstracts, laboratory
studies, descriptive studies, case reports, case series, reviews and opinion
articles were excluded. -
Analysis of Reported Outcomes and Definition of “Effectiveness”
With the aim of providing the reader with a quantitative appraisal
of modifications in mandibular dimensions and Sagittal position in Class
II patients treated with functional appliances compared to untreated Class
II controls, the following data were evaluated for each of the retrieved
studies: mandibular sagittal position (SNB), total mandibular length (Co-
Gn or Co-Pg), mandibular ramus height (Co- Go), and mandibular body
length (Go-Gn, Go-Me, or Go-Pg). Studies that used point Ar for the
measurement of either mandibular length or ramus height were excluded,
as point Ar is not an anatomical landmark that pertains to the mandible
69
Effectiveness and Timing of Mandibular Changes
exclusively (Chen et al., 2002). Because a majority of the cases investi-
gated in the retrieved studies reported mandibular changes as annualized
(expressed as annualized mean differences between treated and untreated
groups), annualization was applied to the data of the studies in which this
was not the case (with the exception of the data in the studies with a treat-
ment duration of less than nine months). The amount of supplementary
elongation in total mandibular length at the completion of active treatment
with functional appliances was analyzed as well.
Table 1.
INCLUSION CRITERIA EXCLUSION CRITERIA
Meta-analyses, randomized controlled Case reports, case series and descriptive
trials, prospective and retrospective studies, review articles, opinion articles
controlled clinical trials and abstracts
Articles in English Laboratory studies
Articles published from January 1966 to Studies performed on adults
December 2005
Studies performed on magnetic
Studies performed on growing patients resonance imaging
Studies conducted on lateral cephalograms | Measurements of total mandibular length
including measurements of total with the use of point Articulare
mandibular length (with the use of point º
Condylion) Treatment combined with extractions
Untreated Class II control subjects Treatment combined with fixed
appliances
Surgical treatments
Success of therapy (both at the occlusal
and skeletal levels) as a criterion for case
selection
The review included an evaluation of the ability of different types
of functional appliances to induce supplementary elongation of the man-
dible. Supplementary elongation was defined as clinically significant
elongation of the mandible with respect to the controls at the end of the
overall treatment period. In accordance with both the power of the re-
trieved studies and the findings of a recent longitudinal study on growth
in subjects with untreated Class II malocclusions (Stahl et al., 2006),
clinical significance was defined as the presence of at least 3.0 mm of
supplementary mandibular length in treated subjects when compared to
untreated subjects at the end of treatment. In fact, the average amount of
70
Franchi et al.
deficiency in growth expressed by total mandibular length in Class II sub-
jects during the circumpubertal period (from CS1 through CS6; Baccetti
et al., 2005) is about 3 mm (2.9 mm as derived from lateral cephalograms
with a standard magnification factor of 8%). Therefore, besides the statis-
tical significance of the retrieved investigations reported here, an effective
result of FJO should be, on average, the correction of this deficiency in
mandibular growth in order to overcome average differences in mandibu-
lar skeletal growth between subjects with Class II malocclusion and sub-
jects with normal occlusion.
Results
Using the search strategy, we found 725 articles. After applying
the inclusion and exclusion criteria seen in Table 1, 22 articles qualified
for the final review analysis (Jakobsson, 1967; Pancherz, 1982; McNa-
mara et al., 1985, 1990; Jakobsson and Paulin, 1990; Windmiller, 1993;
Nelson et al., 1993; Tulloch et al., 1997; Illing et al., 1998; Franchi et al.,
1999; Toth and McNamara, 1999; Mills and McCulloch, 2000; Baccetti et
al., 2000; Chadwick et al., 2001; de Almeida et al., 2002; Basciftci et al.,
2003; Pangrazio-Kulbersh et al., 2003; Faltin et al., 2003; Janson et al.,
2003; O’Brien et al., 2003; Cozza et al., 2004; de Almeida et al., 2005).
The study design of the 22 articles is shown in Table 2. The 22 articles in-
cluded four RCTs, two prospective CCTs, and 16 retrospective CCTs. No
meta-analyses were found. The analysis of reported outcomes is described
in Table 3.
Functional appliances produced a statistically significant annual-
ized supplementary elongation in 21 of 32 samples for total mandibular
length, in 12 of 17 samples for mandibular ramus height, and in 8 of 21
samples for mandibular body length. Outcomes in terms of changes in
mandibular position relative to the cranial base (SNB angle) were statis-
tically significant in 12 of 25 articles. The actual amount of difference
in SNB was limited (ranging from 0.1° to 1.9°). When overall treatment
duration was considered, only 10 of the 32 samples in the 22 studies de-
scribed effective FJO, as defined by the production of supplementary man-
dibular growth greater than or equal to 3.0 mm in the treated group when
compared with the untreated group at the end of treatment.
Effectiveness of Functional Appliances on Mandibular Growth
The aim of the literature search was to select all randomized and
controlled clinical trials with untreated Class II controls that evaluated
treatment outcomes of FJO in Class II patients. All studies agreed that
71
Effectiveness and Timing of Mandibular Changes
Table 2. P is a prospective study, R is a retrospective study, L is a longitudinal
study, RCT is a randomized clinical trial and CCT is a controlled clinical trial.

Articles Study design
Jakobsson (1967) RCT, L
Pancherz (1982) P, L, CCT
McNamara et al. (1985) R, L, CCT
Jakobsson and Paulin (1990) R, L, CCT
McNamara at al. (1990) R, L, CCT
Windmiller (1993) R, L, CCT
Nelson et al. (1993) CT, L
Tulloch et al. (1997) RCT, L
Illing et al. (1998) P, L, CCT
Franchi et al. (1999) R, L, CCT
Toth and McNamara (1999) R, L, CCT
Mills and McCulloch (2000) R, L, CCT
Baccetti et al. (2000) , R, L,CCT
Chadwicket al. (2001) R, L, CCT
de Almeida et al. (2002) R, L, CCT
Basciftci et al. (2003) R, L, CCT
Pangrazio-Kulbersh et al. (2003) R, L, CCT
Faltin et al. (2003) R, L, CCT
Janson et al. (2003) R, L, CCT
O'Brien et al. (2003) RCT, L
Cozza et al. (2004) R, L, CCT
de Almeida et al. (2005) R, L, CCT
mandibular position, relative to the cranial base as measured by the SNB
angle, was not impacted in a clinically significant way by FJO. It should be
noted, however, that the SNB angle is a poor indicator of the effectiveness
of functional jaw orthopedics.
The amount of mandibular supplementary growth when compared
to untreated Class II controls varied widely among studies. With regard to
the changes in total mandibular length, as measured by Co-Gn or Co-Pg,
only one-third of the samples reported induced supplementary mandibu-
lar growth at the completion of active treatment when compared with the
untreated controls (Table 3).
Interestingly, none of the four RCTs reported a clinically sig-
nificant change in mandibular length induced by FJO. To further ex-
plain this, it was necessary to analyze the influence of treatment tim-
ing (individual skeletal maturity at the start of FJO) on treatment re-
sults, which leads us to the second question of this systematic review.
72
Franchi et al.
Table 3.
Annualized changes
ſº Go-Gn Actual
Active Co-Gn
Articles Appliance treatment SNB (or Co-Pg) Co-Go (or Go-Me; Co-Gn
d * In II* or Go-Pg) (or Co-Pg)
uration In In
In In In In
Jakobsson (1967) act 18m - 0.5 (ns) - 0.7
Pancherz (1982) Herbst 6m* 1.4f 2.2 (s)** - - 2.2++
FR-2 E 24m 0.3 (s) 1.2 (s) 1.0 (s) 0.0 (ns) 2.4
McN t al. (1985
cNamara et al. (1985) FR-2 L 24m 0.4 (s) 1.8 (s) 1.5 (s) 0.1 (ns) 3.6
- act M 32m 0.6 (s) 1.2 (s) - - 3.2
Jakobsson and Paulin (1990) act F 30m 0.1 (ns) 0.2 (ns) -- - ().5
Herbst 12m 1.6 (s) 2.7 (s) 2.1 (s) 0.2 (ns) 2.7
McNamara et al. (1990) FR-2 2lm 0.5 (ns) 2.2 (s) 1.8 (s) 0.3 (ns) 3.8
Windmiller (1993) Herbst 12m 1.3 (s) 3.5 (s) 2.9 (s) 0.3 (ns) 3.5
FR-2 18m 0.2 (ns) 0.5 (ns) 0.0 (ns) 0.7 (ns) 0.7
Nelson et al. (1993) act 18m 0.2 (ns) 0.9 (ns) -0.7 (ns) 1.2 (s) 1.3
Tulloch et al. (1997) Bio 15m 0.6 (s) 1.3 (s) - - 1.6
Bass 9m 1.5 (ns) 0.5 (ns) - 0.4
Illing et al. (1998) Bio 9m 1.1 (ns) 3.7 (s) * 2.8
TB 9m 1.3 (ns) 3.4 (s) - - 2.6
Franchi et al. (1999) Herbst 12m - 3.0 (s) 1.3 (s) 1.2 (ns) 3.0
TB 16m 1.0 (s) 2.2 (s) 1.3 (s) 0.7 (s) 3.0
Toth and McNamara (1999) FR-2 24m 0.3 (ns) 1.4 (s) 1.0 (s) 0.1 (ns) 2.8
Mills and McCulloch (2000) TB 14m 1.9 (s) 3.6 (s) 2.5 (s) 1.1 (s) 4.2
* TB E 14m - 1.9 (s) 0.3 (ns) 1.0 (ns) 2.2
Baccetti et al. (2000) TB L 17m * 4.7 (s) 2.7 (s) 1.7 (s) 6, 7
Chadwicket al. (2001) FR-2 20m 0.4 (s) 0.3 (ns) - - 0.6
- Bio | 6m 1.3 (s) 1.7 (s) 1.1 (s) 2.3
de Almeida et al. (2002) FR-2 17m 0.4 (ns) 0.8 (s) 0.8 (s) 1.2
Basciftci et al. (2003) act 16m 0.7 (ns) 3.9 (s) - 2.2 (s) 5.2
Pangrazio-Kulbersh et al. (2003) MARA 11 m 1.0 (s) 2.7 (s) 2.7 (s) 0.1 (ns) 2.7
e Bio E 22m - 0.4 (ns) 0.0 (ns) 0.2 (ns) 1.9
Faltin et al. (2003) Bio L 28m - 2.1 (s) 2.1 (s) 0.0 (ns) 4.3
Janson et al. (2003) FR-2 28m 1.4 (ns) 0.0 (ns) 0.1 (ns) 0.7 (s) 0.5
O'Brien et al. (2003) TB 15m - 12++ + - * 1.5
Cozza et al. (2004) act 21m 0.8 (s) 1.5 (ns) 0.1 (ns) 2.7
De Almeida et al. (2005) Herbst 12 m 0.9 (ns) 1.7 (ns) & * 1.7

Act = activator, Bass = Bass appliance, Bio = bionator appliance, FR-2 = Functional regulator of Fränkel, MARA = mandibu-
lar anterior repositioning appliance, TB = twin-block appliance, Contra controls, E = early, L = late, M =: male, F = female, m
= months, s = statistically significant, ns = not significant according to authors, “Effective" differences (at least 3 mm) are re-
ported in bold, and italics.
*= outcomes of this study were not annualized
“= measured as Pg/OLp-ColoLp
t= not evaluated statistically by the author
DOES TIMING OF INTERVENTION MAKE A
DIFFERENCE IN TERMS OF FJO-INDUCED
MANDIBULAR CHANGES.”
It has been demonstrated widely in the past that effectiveness of
functional treatment of mandibular growth deficiencies strongly depends
on the biological responsiveness of the condylar cartilage, which in turn
is dependent on the growth rate of the mandible (expressed as pre-peak,
peak, and post-peak growth rate relative to the pubertal growth spurt;
(Malmgren et al., 1987; Hägg and Pancherz, 1988; Petrovic et al., 1990).
Only seven of the 22 studies (Pancherz, 1982; Tulloch et al., 1997; Fran-
chi et al., 1999; Baccetti et al., 2000; Faltin et al., 2003; O’Brien et al.,
2003; de Almeida et al., 2005) included in this review described actual
73
Effectiveness and Timing of Mandibular Changes
skeletal maturity based on a biological indicator (hand and wrist analysis,
cervical vertebral maturation method, etc.). In these seven studies, nine
samples of patients treated with functional appliances were investigated:
six samples received treatment before the pubertal peak in skeletal growth
(Pancherz, 1982; Tulloch et al., 1997; Baccetti et al., 2000; Faltin et al.,
2003; O’Brien et al., 2003; de Almeida et al., 2005) and treatment time in
three samples included the pubertal peak (Franchi et al., 1999; Baccetti et
al., 2000; Faltin et al., 2003). The amount of actual supplementary man-
dibular growth induced by treatment (as measured by Co-Gn or Co-Pg)
was equal to or greater than 3.0 mm in all of the "peak” samples. None of
the samples treated in the pre-peak period exhibited an effective amount of
supplementary mandibular growth. The inclusion of the pubertal growth
spurt in the treatment period appears to be a key factor in the effectiveness
of FJO on mandibular growth.
With regard to treatment timing as reported for the RCTs, it must
be stressed that two of the studies did not include an adequate appraisal
of skeletal maturity (Jakobsson, 1967; Nelson et al., 1993), and two of
the studies described the results of using functional appliances during the
pre-peak period (Tulloch et al., 1997; O’Brien et al., 2003). The lack of
clinical significance in the outcomes of these last two RCTs may correlate
with the pre-pubertal treatment timing of the reported samples. Similarly,
Jakobsson (1967) reported outcomes of activator treatment performed at
an average age of 8.5 years (a very early age with regard to the peak in
mandibular growth).
FINAL REMARKS
The present study re-analyzed the reported outcomes of a system-
atic review of the literature (Cozza et al., 2006) by taking into account
recent data on mandibular growth deficiency in untreated Class II maloc-
clusions as a reference measure for the assessment of the effectiveness of
FJO. On the basis of the analysis we conducted on the 22 studies that we
re-examined, we concluded that:
• One-third of the patients in the 22 studies reported an effec-
tive supplementary elongation in total mandibular length as
a result of active treatment with functional appliances.
• The effectiveness of FJO on mandibular growth is signifi-
cantly greater if the functional treatment is performed dur-
ing the adolescent growth spurt.
74
Franchi et al.
• None of the subjects in the four RCTs that we re-examined
showed clinically significant supplementary growth of the
mandible due to functional appliances. This possibly is a
result of treatment timing in that treatment occurred during
a pre-pubertal stage of skeletal maturity for subjects in three
Of the four RCTS.
In other words, there is good evidence that functional jaw orthope-
dics is not an effective stimulant of mandibular growth when treatment is
performed at a pre-pubertal stage of development (so-called “early” treat-
ment). However, there is good evidence that functional jaw orthopedics is
effective in producing supplementary mandibular elongation with respect
to untreated controls when the treatment period includes the period during
which the pubertal growth spurt occurs.
When the effect of FJO on mandibular growth is a consideration,
the use of functional appliances is not recommended as a “first phase”
treatment in a two-phase treatment regimen that includes an interim (some-
times long) period between the two phases of treatment. This is because
treatment would occur pre-pubertally. Functional appliances should be
worn during the pubertal peak, which corresponds either to the late mixed
dentition or to the early permanent dentition (Franchi et al., 2006). Fixed
appliances then can be used to refine the occlusion and stabilize the results
of FJO and can follow functional therapy immediately. This results in the
practical advantage of a relatively short overall treatment duration (about
28 months on average) for treatment protocols that include the twin block
or the Herbst appliance followed by fixed appliances (Burkhardt et al.,
2003; Schaefer et al., 2004). This treatment time is comparable to that
incurred when using either headgear combined with fixed appliances or
fixed appliances combined with Class II elastics (27 months on average;
Tulloch et al., 2004). In this way, functional appliances may be seen as the
initial component of one-phase treatment modalities aimed at the correc-
tion of Class II malocclusions through intensive stimulation of mandibular
growth.
Finally, it is worth remembering that the present systematic re-
view analyzed a specific component of skeletal change, i.e., mandibular
growth change as a result of FJO. It was not a comprehensive analysis
of the various components of treatment for Class II malocclusions and
the resulting dentoskeletal effects. The validity of this kind of systematic
review (which is not different from that of a scientific trial) only applies
to the specific topic under investigation and is dependent upon the sound-
ness of the initial criteria of inclusion and/or exclusion (both in the case of
75
Effectiveness and Timing of Mandibular Changes
scientific articles reviewed and in the case of patients/treatment modalities
in clinical trials).
REFERENCES
Antczak AA, Tang J, Chalmers TC. Quality assessment of randomized
control trials in dental research. I. Methods. J Periodont Res 1986:
21:305-314.
Baccetti T, Franchi L, Toth LR, McNamara JA Jr. Treatment timing for
twin-block therapy. Am J Orthod Dentofacial Orthop 2000; 118:159-
170.
Baccetti T, Franchi, L, McNamara JA Jr. The cervical vertebral matura-
tion (CVM) method for the assessment of optimal treatment timing in
dentofacial orthopedics. Semin Orthod 2005; 11:119-129.
Basciftci FA, Uysal TU, Büyükerkmen A, Sari Z. The effects of activator
treatment on the craniofacial structures of Class II, division 1 patients.
Eur J Orthod 2003:25:87–93.
Burkhardt DR, McNamara JA Jr, Baccetti T. Maxillary molar distalization
or mandibular enhancement: A cephalometric comparison of compre-
hensive orthodontic treatment including the pendulum and the Herbst
appliances. Am J Orthod Dentofacial Orthop 2003;123:108-116.
Buschang PH, Tanguay R, Turkewicz J, Demirjian A, La Palme L. A poly-
nomial approach to craniofacial growth: Description and comparison
of adolescent males with normal occlusion and those with untreated
Class II malocclusion. Am J Orthod Dentofacial Orthop 1986;90:437-
442.
Buschang PH, Tanguay R, Demirjian A, LaPalme L, Turkewicz J. Math-
ematical models of longitudinal mandibular growth for children with
normal and untreated Class II, division 1 malocclusion. Eur J Orthod
1988; 10:227–234.
Chadwick SM, Aird JC, Taylor PJS, Bearn DR. Functional regulator
treatment of Class II, division 1 malocclusions. Eur J Orthod 2001;
23:495-505.
Chen JY, Will LA, Niederman R. Analysis of efficacy of functional ap-
pliances on mandibular growth. Am J Orthod Dentofacial Orthop
2002; 122:470-476.
Cozza P. De Toffol L, Colagrossi S. Dentoskeletal effects and facial profile
changes during activator therapy. Eur J Orthod 2004:26:293–302.
Cozza P. Baccetti T. Franchi L., De Toffol L, McNamara JA Jr. Mandib-
ular changes produced by functional appliances in Class II maloc-
76
Franchi et al.
clusion: A systematic review. Am J Orthod Dentofacial Orthop
2006; 129:599.e1-599.e12.
de Almeida MR, Henriques JFC, Ursi W. Comparative study of Fränkel
(FR-2) and bionator appliances in the treatment of Class II malocclu-
sion. Am J Orthod Dentofacial Orthop 2002; 121:458–466.
de Almeida MR, Henriques JF, de Almeida RR, Weber U, McNamara JA
Jr. Short-term treatment effects produced by the Herbst appliance in
the mixed dentition. Angle Orthod 2005;75:540-547.
Faltin K Jr., Faltin RM, Baccetti T, Franchi L. Ghiozzi B, McNamara JA
Jr. Long-term effectiveness and treatment timing for bionator therapy.
Angle Orthod 2003;73:221-230.
Franchi L, Baccetti T, De Toffol L, Polimeni A, Cozza P. The phases of the
dentition for the assessment of skeletal maturity: A diagnostic perfor-
mance study. Am J Orthod Dentofacial Orthop 2006;in press.
Franchi L, Baccetti T. McNamara JA Jr. Treatment and post-treatment ef-
fects of acrylic splint Herbst appliance therapy. Am J Orthod Dentofa-
cial Orthop 1999; 115:429–438.
Green S, Higgins JPT, Eds. Cochrane Handbook for Systematic Reviews
of Interventions. http://www.cochrane.org/resources/handbook, Ver-
sion 4.2.5, May 2005.
Hägg U, Pancherz H. Dentofacial orthopaedics in relation to chronologi-
cal age, growth period and skeletal development. An analysis of 72
male patients with Class II, division 1 malocclusion treated with the
Herbst appliance. Eur J Orthod 1988; 10:169-176.
Harris E. A cephalometric analysis of mandibular growth rate. Am J Or-
thod 1962:48:161-173.
Illing HM, Morris DO, Lee RT. A prospective evaluation of Bass, bion-
ator and twin block appliances. Part I: The hard tissues. Eur J Orthod
1998:20:501-516.
Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan
DJ, McQuay H.J. Assessing the quality of reports of randomized clin-
ical trials: Is blinding necessary? Control Clin Trials 1996; 17:1-12.
Jakobsson SO. Cephalometric evaluation of treatment effect on Class II,
division 1 malocclusion. Am J Orthod 1967:53:446-456.
Jakobsson SO, Paulin G. The influence of activator on skeletal growth in
Angle Class II:1 cases. A roentgenocephalometric study. Eur J Orthod
1990; 12:174-184.
Janson GRP, Toruno JLA, Martins DR, Henriques JFC, de Freitas MR.
Class II treatment effects of the Frankel appliance. Eur J Orthod
2003:25:301-309.
77
Effectiveness and Timing of Mandibular Changes
Kelly JE, Harvey C. An assessment of the teeth of youths 12–17 years,
DHEW Publication No (HRA) 77-1644. Washington, DC: National
Center for Health Statistics, 1977.
Malmgren O, Ömblus J, Hägg U, Pancherz H. Treatment with an appli-
ance system in relation to treatment intensity and growth periods. Am
J Orthod Dentofacial Orthop 1987;91: 143-151.
McNamara JA Jr. Neuromuscular and skeletal adaptations to altered func-
tion in the orofacial region. Am J Orthod 1973;64:578-606.
McNamara JA Jr. Components of a Class II malocclusion in children 8-10
years of age. Angle Orthod. 1981:51:177–202.
McNamara JA Jr, Bryan FA. Long-term mandibular adaptations to protru-
sive function in the rhesus monkey (Macaca mulatta). Am J Orthod
Dentofacial Orthop 1987;92:98-108.
McNamara JA Jr, Bookstein FL, Shaughnessy TG. Skeletal and dental
changes following functional regulator therapy on Class II patients.
Am J Orthod 1985;88:91-110.
McNamara JA Jr, Howe RP, Dischinger TG. A comparison of the Herbst
and Frankel appliances in the treatment of Class II malocclusion. Am
J Orthod Dentofacial Orthop 1990;98:134-144.
McNamara JA Jr, Brudon WL. Orthodontics and Dentofacial Orthope-
dics. Ann Arbor, Needham Press, 2001:67–80. g
Mills CM, McCulloch K.J. Post-treatment changes after successful cor-
rection of Class II malocclusions with the twin-block appliance. Am J
Orthod Dentofacial Orthop 2000;118:24-33.
Nelson C, Harkness M, Herbison P. Mandibular changes during functional
appliance treatment. Am J Orthod Dentofacial Orthop 1993;104:153-
161.
O’Brien K, Wright J, Conboy F, Sanjie Y, Mandall N, Chadwick S, Con-
nolly I, Cook P, Birnie D, Hammond M, Harradine N, Lewis D, Mc-
Dade C, Mitchell L, Murray A, O’Neill J, Read M, Robinson S, Rob-
erts-Harry D, Sandler J, Shaw I. Effectiveness of early orthodontic
treatment with the twin-block appliance: A multicenter, randomized,
controlled trial. Part I: Dental and skeletal effects. Am J Orthod
Dentofacial Orthop 2003;124:234-243.
Pancherz H. The mechanism of Class II correction in Herbst appliance
treatment: A cephalometric investigation. Am J Orthod 1982;82:104-
113.
Pangrazio-Kulbersh V, Berger JL, Chermak DS, Kaczynski R, Simon
ES, Haerian A. Treatment effects of the mandibular anterior reposi-
78
Franchi et al.
tioning appliance on patients with Class II malocclusion. Am J Orthod
Dentofacial Orthop 2003;123:286-295.
Petrén S, Bondemark L, Söderfeldt B. A systematic review concerning
early orthodontic treatment of unilateral posterior crossbite. Angle Or-
thod 2003;73:588-596.
Petrovic A, Stutzmann J, Lavergne J. Mechanism of craniofacial growth
and modus operandi of functional appliances: A cell-level and cyber-
netic approach to orthodontic decision making. In: Carlson DS, ed,
Craniofacial Growth Theory and Orthodontic Treatment. Craniofa-
cial Growth Series, Center for Human Growth and Development, The
University of Michigan, Ann Arbor 1990:23:13-74.
Proffit WR, Fields HW, Moray LJ. Prevalence of malocclusion and orth-
odontic treatment need in the United States: Estimates from the N-
HANES III survey. Int J Adult Orthod Orthognath Surg 1998;13:97-
106.
Schaefer AT, McNamara JA Jr, Franchi L, Baccetti T. A cephalometric
comparison of treatment with the twin-block and stainless steel crown
Herbst appliances followed by fixed appliance therapy. Am J Orthod
Dentofacial Orthop 2004;126:7-15.
Stahl F. Baccetti T, Franchi L, McNamara JA Jr. Longitudinal growth
changes in untreated subjects with Class II and Class I malocclusion.
Am J Orthod Dentofacial Orthop, in press.
Toth LR, McNamara JA Jr. Treatment effects produced by the twin-
block appliance and the FR-2 appliance of Fränkel compared with
an untreated Class II sample. Am J Orthod Dentofacial Orthop 1999;
116:597–609.
Tulloch JF, Proffit WR, Phillips C. Outcomes in a 2-phase randomized
clinical trial of early Class II treatment. Am J Orthod Dentofacial Or-
thop 2004;125:657-667.
Tulloch JFC, Phillips C, Koch G, Proffit WR. The effect of early interven-
tion on skeletal pattern in Class II malocclusion: A randomized clini-
cal trial. Am J Orthod Dentofacial Orthop 1997;111:391-400.
Windmiller EC. The acrylic-splint Herbst appliance: A cephalometric
evaluation. Am J Orthod Dentofacial Orthop 1993;104:73-84.
Xiong H, Hägg U, Tang GH, Rabie ABM, Robinson W. The effect of con-
tinuous bite-jumping in adult rats: A morphological study. Angle Or-
thod 2004;74:86–92.
79
80
A COLLATERAL NOTE ON THE CLINICAL
APPLICABILITY OF THE OUTCOMES OF RCTS
IN ORTHODONTICS
Tiziano Baccetti
Lorenzo Franchi
When looking back at the results of the systematic review of the effective-
ness of functional jaw orthopedics (FJO) that we put forward in our earlier
articles in this volume, we realized that some comments may be misunder-
stood with regard to the interpretation of the outcomes of the randomized
clinical trials (RCTs) that are currently available on Class II treatment.
The intent was not to underestimate the quality of this type of research,
but, at the same time, not to overrate it when clinical recommendations are
derived from it.
RCTs have been used rarely in orthodontics and with good reason:
it is difficult to gather a large number of patients who have a specific oc-
clusal deviation; there is the ethical issue of leaving a group of patients
untreated; and several design requirements of quality reviews such as pa-
tients or observers blinded to treatment clearly do not apply (Antczak et
al., 1986; Jadad et al., 1996). Moreover, from the analysis of the results of
the systematic review presented in this volume, the reader can recognize
easily the following problem.
The RCTs of one-phase treatment versus two-phase treatment in
orthodontics (Tulloch et al., 1997; Keeling et al., 1998; Ghafari et al.,
1998) argue that the effects of FJO followed by fixed appliance therapy
are not significantly different from the effects of treatment with headgear
combined with fixed appliances or from the effects of single-phase, fixed
appliance therapy. The only major difference is treatment duration, which
is longer when FJO is part of two-phase treatment. A longer treatment time
is a practical disadvantage for this type of treatment protocol. With regard
to the changes produced by FJO in mandibular growth, the systematic re-
view of the RCT results presented in this volume shows that there was an
insignificant amount of growth modification (less than 2 mm).
Accepting all of these results for what they appear to be, that is
the best possible distilled outcome of the best possible type of research
investigation, it would seem that a consensus should form in the orth-
odontic community that not only is FJO not the first treatment of choice
81
Applicability of RCT Outcomes
for the “average” Class II patient, but that one-phase treatment with fixed
appliances is the treatment of choice for the “average” Class II patient.
Now, when we reinterpret the mandibular effects of FJO as reported in the
RCTs, we see that while one-phase treatment is performed at puberty, FJO
was performed in three out of four patients during the pre-pubertal stage
of development, which is the least favorable time for growth modification
of the mandible. The decision of the RCTs to perform FJO prepubertally
was made in spite of many well-documented studies on optimal treatment
timing for FJO that clearly demonstrated that the effects of FJO are sig-
nificantly greater when the growth spurt is included in the active treatment
period (Petrovic, 1984; McNamara et al., 1985; Hägg and Pancherz, 1988;
Malmgren et al., 1987; Petrovic et al., 1990; Baccetti et al., 2000; Bac-
cetti and Franchi, 2001; Faltin et al., 2003). Should we just assume that
researchers in charge of the RCTs on Class II treatment were not aware of
this information, or that their focus on the comparison between one-phase
and two-phase treatment caused them to overlook the issue of effective-
ness of FJO inadvertently?
Most of the problems arise from poor patient selection criteria at
the start of an RCT. Criteria for patient selection must be rigorous, be-
cause the results of the RCT will be applicable only to the specific clinical
condition that is reflected by the patient selection criteria. The clinical rec-
ommendation that derives from the outcomes of an RCT should be able to
isolate the impact of intervention for that particular condition in one given
patient. In the words of Sackett and co-workers (1996), evidence-based
medicine must be “the conscientious, explicit, and judicious use of current
best evidence in making decisions about the care of individual patients.”
As described by Darendeliler (2006), patient selection criteria
in orthodontic RCTs have been too aspecific, often including a variety
of “clinical conditions” under the umbrella of a vague definition. For in-
stance, in the University of North Carolina trial, Tulloch and colleagues
(1997) defined a Class II malocclusion as an overjet > 7 mm. There are
many types of clinical conditions that can be associated with an overjet
> 7 mm, many of which involve incisor inclination rather than true Class
II occlusal or skeletal relationships. On the other hand, the outcomes of
the RCT that started with the analysis of subjects with a generic “overjet
> 7 mm” are proposed as a guideline for “Class II treatment.” It appears,
therefore, that the RCTs on Class II treatment were evaluating symptoms
rather than clinical conditions. It would have no meaning to perform an
82
Baccetti and Franchi
RCT on the treatment of “headaches,” because we know very well that
headaches as a symptom can be secondary to many different conditions
such as vascular problems, muscular tension, and even tumors (Darendeli-
ler, 2006). The outcomes of that RCT would have minimal, if any, clinical
applicability. The use of “overjet > 7 mm” as a patient selection criterion
for an orthodontic RCT on the treatment of Class II malocclusion greatly
reduces the clinical significance of the RCT outcomes in a similar way.
Insufficient criteria for the definition of the clinical condition
analyzed and the complete lack of attention to individual signs or fea-
tures within “Class II malocclusion” severely limit the usefulness of the
outcomes of orthodontic RCTs in treatment planning for the “individual
patient.” In addition to the issues of treatment timing and the role of in-
dividual skeletal maturity at the start of treatment, factors involving indi-
vidual patient responsiveness to treatment deserve a greater consideration,
as they can lead to a better definition of expectation of treatment outcome
in the “individual patient.” The effects of FJO, for instance, are linked to
cell-level and tissue-level characteristics of the condylar cartilage of the
individual subject (Petrovic et al., 1990).
Rabie and co-workers (2003) already have identified some genetic
markers that correlate with these tissue-level characteristics in an animal
model, and this may help identify which patients are potential the best
responders to dentofacial orthopedics. Our research group has recently in-
dicated the role of mandibular morphology (namely the Co-Go-Me angle)
as a predictor for the outcomes of FJO in the pre-treatment evaluation of
Class II patients (Franchi and Baccetti, 2006). A generic, blind application
of the outcomes of RCTs to treatment planning in Class II malocclusion
would wipe out all these indicators in favor of a standardized, “average”
treatment for an “average” patient, who really does not exist. In fact, the
consequence would be more problematic than that. The unreasoned appli-
cation of the statements reported by the RCTs would create an additional,
maybe more severe, detrimental effect for the patient. Insurance compa-
nies define it as “loss of chance.” It is the suppression of the chance of a
benefit to individual patients who are good candidates for that benefit. A
Class II patient who would present with features that were very favorable
to a positive response to FJO (ideal timing according to individual skeletal
maturity, characteristics of craniofacial morphology, etc.) would not have
much of a chance of receiving the specific type of treatment needed to cor-
rect his/her precise clinical condition.
83
Applicability of RCT Outcomes
If there are no explicit patient selection criteria based on the spe-
cific components of the malocclusion, no taking into account previous re-
search findings, no attention paid to factors of individual responsiveness,
no information available as to the proficiency of the orthodontists who
delivered specific types of treatment, do the outcomes of RCTs make sense
from a clinical point of view? Or, rather, do they prevent those individuals
who could really benefit from a specific treatment from the probability of
actually benefiting from it?
On the other hand, we should remember that there are alterna-
tive research approaches to RCTs that can strengthen the scientific validity
of controlled clinical trials, even if not randomized. A good example of
this is the research paper by Paquette and co-workers (1992). They used
discriminant analysis on the initial sample of patients treated with extrac-
tion and non-extraction protocols to identify “borderline cases” that could
have been treated either way by the clinicians. The use of patients treated
consecutively with a given treatment modality regardless of treatment out-
comes is another option to improve the quality of retrospective clinical
trials. If nothing else, these approaches are much less expensive, and they
make many more samples available for investigation.
This discussion is not intended to downscale the validity of RCTs
as an investigation method, nor is it aimed at retrieving credit from the
RCTs on one-phase versus two-phase treatment in Class II malocclusion
performed in the past. The purposes here were to highlight some funda-
mental risks in the possible interpretation of the findings of those RCTs,
and to point out potential pitfalls in the clinical application of these out-
comes to the individual patient. The ultimate consequence of an indis-
criminate use of the results of the RCTs would be the reduction in the pos-
sibility of a therapeutic benefit for patients who should be good candidates
for it. In political terms, the attempt here was to prevent the dictatorship of
a rigid clinical application of the outcomes derived from RCTs, in favor of
the democracy in which the choice for the best possible treatment for the
patient as an individual is available to all.
ACKNOWLEDGEMENT
The authors wish to thank Mr. Michael Powell for his editorial
assistance.
84
Baccetti and Franchi
REFERENCES
Antczak AA, Tang J, Chalmers TC. Quality assessment of randomized
control trials in dental research I. Methods. J Periodont Res 1986:21:
305-314.
Baccetti T, Franchi L, Toth LR, McNamara JA Jr. Treatment timing for
Twin-Block therapy. Am J Orthod Dentofacial Orthop 2000;118:159-
170.
Baccetti T, Franchi L. Maximizing esthetic and functional changes in Class
II treatment by appropriate treatment timing. In: McNamara JA Jr and
Kelly KA, eds, Frontiers of Dental and Facial Esthetics. Craniofacial
Growth Series, Department of Orthodontics and Pediatric Dentistry
and Center for Human Growth and Development, The University of
Michigan 2001:38:237-251.
Darendeliler MA. Validity of randomized clinical trials in evaluating the
outcome of Class II treatment. Seminars Orthod 2006: 12:67-79.
Faltin K Jr, Faltin RM, Baccetti T, Franchi L. Ghiozzi B, McNamara JA
Jr. Long-term effectiveness and treatment timing for bionator therapy.
Angle Orthod 2003;73:221-230.
Franchi L, Baccetti T. Prediction of individual mandibular changes in-
duced by functional jaw orthopedics followed by fixed appliances in
Class II patients. Angle Orthod, 2006, in press.
Ghafari J, Shofer FS, Jacobsson-Hunt U, Markowitz DL, Laster LL. Head-
gear versus function regulator in early treatment of Class II, Division
I malocclusion: A randomized clinical trial. Am J Orthod Dentofacial
Orthop 1998; 113:51-61.
Hägg U, Pancherz H. Dentofacial orthopaedics in relation to chronologi-
cal age, growth period and skeletal development. An analysis of 72
male patients with Class II, division 1 malocclusion treated with the
Herbst appliance. Eur J Orthod 1988; 10:169-176.
Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan
DJ, McQuay H.J. Assessing the quality of reports of randomized clin-
ical trials: Is blinding necessary? Control Clin Trials 1996; 17:1-12.
Malmgren O, Ömblus J, Hägg U, Pancherz H. Treatment with an appli-
ance system in relation to treatment intensity and growth periods. Am
J Orthod Dentofacial Orthop 1987;91:143-151.
Keeling SD, Wheeler TT, King GJ, Garvan CW, Cohen DA, Cabassa
S, McGorray SP. Taylor MG. Anteroposterior skeletal and dental
changes after early Class II treatment with bionators and headgear.
Am J Orthod Dentofacial Orthop 1998; 113:40-50.
85
Applicability of RCT Outcomes
McNamara JA Jr, Bookstein FL, Shaughnessy TG. Skeletal and den-
tal adaptation following functional regulator therapy. Am J Orthod
1985;88:91-110.
Paquette DE, Beattie JR, Johnston LE Jr. A long-term comparison of non-
extraction and premolar extraction edgewise therapy “borderline”
Class II patients. Am J Orthod Dentofacial Orthop 1992; 102:1-14.
Petrovic A, Stutzmann J, Lavergne J. Mechanism of craniofacial growth
and modus operandi of functional appliances: A cell-level and cyber-
netic approach to orthodontic decision making. In: Carlson DS, ed,
Craniofacial Growth Theory and Orthodontic Treatment. Craniofa-
cial Growth Series, Center for Human Growth and Development, The
University of Michigan 1990:23:13-74.
Petrovic AG. Experimental and cybernetic approaches to the mechanism
of action of functional appliances on mandibular growth. In: McNa-
mara JA Jr, Ribbens KA eds, Malocclusion and the Periodontium.
Craniofacial Growth Series, Center for Human Growth and Develop-
ment, The University of Michigan 1984; 15:213-268.
Rabie ABM, She TT, Hägg U. Functional appliance therapy accelerates
and enhances condylar growth. Am J Orthod Dentofacial Orthop
2003;123:40-48.
Sackett DL, Rosenberg WMC, Gray JAM, Richardson WS. Evidence
based medicine. What it is and what it isn’t. Br Med J 1996;312:71-
72.
Tulloch JFC, Phillips C, Koch G, Proffit WR. The effect of early interven-
tion on skeletal pattern in Class II malocclusion: A randomized clini-
cal trial. Am J Orthod Dentofacial Orthop 1997;111:391-400.
86
CLINICAL EVIDENCE IN TREATMENT OF CLASS II,
DIVISION 1 MALOCCLUSION WITH THE BALTERS
BIONATOR: LONG-TERM EFFECTIVENESS, TREAT-
MENT TIMING AND POST-TREATMENT CHANGES
Kurt Falfin, Jr.
The Class II, division 1 malocclusion with mandibular retrognathism is
a distinct category of orthopedic, Class II morphology (Moyers, 1988)
that can be identified by a differential diagnostic protocol and which is
One of the most prevalent dentofacial anomalies. Therapy for this kind of
facial disharmony must be efficient with regard to effecting the vertical
and forward growth of the mandible. The most pure advancement of the
mandible with rotational control is the method of choice, the aim of which
is to restore facial harmony, normalize function and produce a balanced
Occlusion.
After its introduction in 1964 (Balters, 1964, 1973), the Balters bi-
Onator (Fig. 1) became the object of several investigations aimed at identi-
fying the dentoalveolar and skeletal effects of this appliance (Lange et al.,
1995; Morris et al. 1998). Most of these studies investigated the short-term
Outcomes of bionator therapy using varied types of control groups, treat-
ment methods and appliance modifications (Janson, 1977, 1983; Bolmgren
and Moshiri, 1986; Mamandras and Allen, 1990; Illing et al., 1998).
|
Figure 1. The original Balters biomator. Note the lack of anterior tooth
COVerage.

87
Balters Bionator
To our knowledge, the only long-term study on the craniofacial
changes induced by the biomator is the study by Rudzki-Janson and No-
achtar (1998), in which a group of patients were studied after they had
undergone biomator therapy and had been out of retention for five years.
These patients showed an increase in the size of the mandible and a de-
crease in both the distal skeletal jaw relationship and the gonial angle.
The Balters bionator is one of the appliances most used to correct
this category of Class II, division 1 malocclusion; therefore, the clinical
results of treatment need to be investigated in an evidence-based study. To
evaluate what the long-term effects of Balters bionator therapy might be,
we focused on the following three aspects in our investigation:
1. Long term effectiveness: are the short-term results stable?
2. The importance of treatment timing: does the timing of the
start of treatment make any difference?
3. What can be expected long term after active treatment is fin-
ished?
AIM OF THE FIRST PART OF THIS REPORT
The aim of the first portion of this chapter is to provide infor-
mation about the long-term effects and the optimal treatment timing for
patients with Class II, division 1 mandibular retrognathism taken from a
previous study of ours (Faltin et al., 2003).
Subjects and Methods
We analyzed the lateral cephalograms of 23 Class-II, division 1
patients treated with the bionator and those of 21 controls with untreated
Class II malocclusions (Table 1). The cephalograms were taken at the
start of treatment and after the completion of growth for long-term (LT)
observation. Treatment included a period of bionator therapy followed by
a period of treatment with fixed appliances.
The treated sample was divided into two groups based on skeletal
maturity as determined by the cervical vertebral maturation stage (CVMS)
proposed by Franchi and colleagues (Franchi et al., 2000; Baccetti et al.,
2002; Fig. 2). The early-treated group (13 subjects) began treatment be-
fore the peak in mandibular growth (CVMS I and II), which occurred af.
ter completion of bionator therapy. The late-treated group (10 subjects)
received bionator treatment during the peak of the growth period corre-
sponding to CVMS III to IV.
88
Faltin
Table 1. Descriptive statistics for age and observation periods.

Mean Observation
Mean Age Period
Initial LT Diff. (LT - Initial)
Early-treated group.
[n = 13]
Early-control group.
[n = 11]
Late-treated group.
[n = 10]
Late-control group.
[n = 10]
(\{\|\\ |\ |\\ |\\
D D […] [I] . [...] []
9y/8mo + 1y/3mo 17y/5 mo + 2y 7y/8mo + 2y/5mo
9y/5mo + 1y/3mo | 16y/4mo + 1y/9mo || 6y/10mo + 1y/6mo
10y/9mo + 1y/8mo | 19y/2mo + 2y/2mo || 8y/4mo + 1y/8mo
11y/2mo + 1y/6mo 17y/2mo + 1y/lmo || 6y/Omo + ly/10mo


D U | U
[SIS | St | [ ] | I][] . [...] [...]
CVMS I CVMS II |CVMS || || CVMS IV CVMS V
Figure 2. The CVMS method of determining skeletal maturity.
The treated sample was compared with the untreated control sam-
ple selected from the University of Michigan Elementary and Secondary
School Growth Study (Riolo et al., 1974). The control sample also was
divided into two groups based on CVMS. The early control group (ECG)
consisted of 11 subjects (five females and six males) and the late control
group (LCG) consisted of 10 subjects (five females and five males). The
control groups matched the corresponding treated groups vis-a-vis CVMS
at each observation period. The early and late control groups had similar
characteristics to those of the treated groups
The initial and LT lateral cephalograms of both treated and control
groups were standardized with regard to magnification and then digitized
using a digitizing tablet (Numonics, Lansdale, Pennsylvania, PA) and
digitizing software (Viewbox, version 2.6, Kifissia, Greece) developed by
Halazonetis (1994). Changes in the treated groups were compared with
changes in the control groups by nonparametric statistical analysis (P<
.05).
89
Balters Bionator
We performed a cephalometric analysis derived from Pancherz's
(1982) and Johnston’s (1986) original analyses that included a modified
reference system for the superimposition procedure. The definitions for
the landmarks used in the analysis have been described previously (Fran-
chi et al., 1999; Figs. 3 and 4).
Figure 4. Additional cephalometric measures.

90
Faltin
The occlusal line (OL) and the occlusal line perpendicular (OLp)
from the cephalogram at T1 were used as a reference grid. The grid was
transferred from the initial tracing to subsequent tracing at T3 by super-
imposing the tracings on the T-FMN line, with T point as the registration
point. All linear measurements were performed parallel to OL and perpen-
dicular to OLp. The following factors were measured:
• Incisal overjet
• Molar relation
• Maxillary base
• Mandibular base
• Condylar head
• Mandibular supplementary elongation = Co - Pg
• Mandibular ramus height = Co - Go
Comparing the long-term outcome values of the early treated
group (ETG) and the early control group (ECG) with those of the late
treated group LTG and late control group LCG (Tables 2 and 3), we are
able to analyze the following corresponding factors.
• The overjet correction in ETG patients was 1.6 mm and 4.2 mm
in LTG patients relative to controls. This is a significant difference.
• The molar relationship values were the same for ETG and LTG
patients relative to controls, i.e., 2.2 mm in ETG patients and 2.1 mm in
LTG patients.
• A significant mesial advancement of the mandibular dentition
occurred at both the molar and incisor regions in LTG patients.
• There was a small difference in the maxillary base in ETG (-.1
mm) and LTG (-2.2 mm) patients compared to controls.
• There was a difference in mandibular base values of +2.2 mm
in ETG patients and, surprisingly, no difference in LTG patients.
• The condylar head increase showed a difference of 1.2 mm in
ETG patients and a significant 2.4 mm in LTG patients.
• A comparison of the long-term outcomes of the patients whose
treatment began and ended before the pubertal peak (ETG) with corre-
sponding outcomes in the control group showed that the amount of man-
dibular elongation was 1.9 mm more in the ETG patients than in the con-
trols.
• The amount of significant supplementary elongation of the
mandible in the LTG patients was 5.1 mm more than that in controls.
91
Balters Bionator
• Comparing the mandibular ramus height values, we found that
there was a significant increase of 4.8 mm in LTG patients and only 0.5
mm increase in ETG patients.
• At the skeletal level, significant supplementary increases in to-
tal mandibular length and in ramus height were found in LTG patients
when compared with ETG patients.
Table 2. Changes long term-initial in the early groups.

rigo-10 | Ecco-in wº wº,
Factors Mean SD Mean SD * ETG/ECG
Incisal overjet –3.2 | 2.7 |-1.6 2.4 NS -1.6
Molar relation –3.6 | 1.5 || -1.4 1.5 NS –2.2
Maxillary base +4.3 2.5 +5.4 2.7 NS -1.1
Mandibular base . +11.3 || 5.1 +9.1 4.3 NS +2.2
Condylar head –0.9 | 1.7 || -2.1 1.9 NS +1.2
Mandibular supplementary | 115.4 || 4.9 || || 13.5 3.2 | NS | 11.9
elongation
Mandibular ramus height +9.2 || 3.0 +8.7 2.5 NS +0.5
S = significant comparison (P< 0.5); NS = not significant comparison
Table 3. Changes long term-initial in the late groups.

Mann- º
LTG (n = 10) || LCG (n = 10 ſº erence
(n = 10) n = 10) |Whitney LTG / LCG
Factors Mean | SD | Mean | SD U-Test
Incisal overjet –4.6 || 2.5 | -0.4 1.2 S –4.2
Molar relation –2.9 | 1.9 || -0.8 1.2 S –2.1
Maxillary base +1.6 || 3.4 || +3.8 2.9 NS –2.2
Mandibular base . +7.0 || 7.2 +7.0 || 4.6 NS 0.0
Condylar head –3.4 || 2.5 -1.0 2.2 S –2.4
Mandibular supplementary +14.4 || 5.4 | +9.3 5.3 S +5.1
elongation g e { } e tº
Mandibular ramus height +10.0 || 3.4 | +5.2 3.7 S +4.8
. S = significant comparison (P<.0.5); NS = not significant comparison
Recently, the issue of optimal treatment timing for functional
jaw orthopedics has gained the attention of both researchers and clini-
92

Falfin
cians. Cephalometric studies have shown that the therapeutic effectiveness
of functional appliances is greater when these appliances are used during
the ascending portion, closer to the peak, of the individual pubertal growth
spurt (McNamara et al., 1985; Malmgren et el., 1987; Hägg and Pancherz,
1988: Petrovic et el., 1990, 1991; Baccetti et al., 2000; Baccetti and Fran-
chi, 2001; Tulloch et el., 1997). The findings of the present study concur
with this finding, bionator therapy followed by fixed appliance therapy is
more effective and stable when it is performed during the pubertal growth
spurt (LTG; Fig. 5). Optimal timing to begin treatment with the bionator is
When a concavity appears at the lower borders of the second and the third
cervical vertebrae (CVMS II or III).
ETG: LTG:
treatment treatment
before the peak at the peak
co-pg mm : +1.9 co-pg mm : +5.1 *
Co-go mm : +0.5 co-go mm : +4.8 °
Figure 5. Differences between ETG and LTG measurements.
CONCLUSIONS
In the long term, the amount of significant supplementary elonga-
tion of the mandible in patients with Class II, division 1 malocclusions
who were treated with the bionator during the pubertal growth spurt
(LTG patients) was 5.1 mm more than that in untreated subjects (LCG).
Similarly favorable findings were found in LTG patients regarding incre-
ments in mandibular ramus height and the more backward direction of
condylar growth, both of which were significant when compared to the
controls (LCG). This study also indicated that this treatment protocol is
most effective and stable when it occurs during the pubertal growth spurt.
Optimal timing for the start of bionator treatment is when a concavity
is evident at the lower borders of both the second and the third cervical
Vertebrae (CVMS II to III); therefore, we need to consider that the treated
8TOup named “late” is misnamed in that the timing for the beginning of
treatment for this group is ideal for producing the best orthopedic results
(Figs, 6-13).
93
Balters Bionator
Figure 6. Initial intraoral photos.
Figure 8. Long-term (LT) follow-up intraoral photographs.
Figure 9. Initial facial photos.




Falfin
Figure 12. Initial (left), end-of-treatment (middle) and long-term follow-
up (right) profile facial photographs.

95
Balters Bionator
Figure 13. Initial (left), end-of-treatment (middle) and long-term follow-up (right)
lateral x-rays and cephalograms.
AIM OF THE SECOND PART OF THIS REPORT
The aim of the second part of this chapter is to provide informa-
tion about the long-term effects of bionator treatment followed by fixed
appliance therapy. To this end, we did a retrospective study of 33 Class
II, division 1 patients who had finished treatment. We analyzed lateral
cephalograms for each patient at the start of treatment (T1), at the end
of active treatment (T2) and seven years after the T2 point in time (T3).
These patients were not wearing any kind of appliances and were included
without regard to treatment timing (Table 4).
Table 4. Time points for age and observation periods.
SAMPLE:
33 Patients T1 T2 T3 T2 - T1* T3 – T2 T3 – T1
13 males 10y/2mo =|13y/1mo +
20 females 1.5 2.2 20y + 2.4 2y/9mo 6y/9mo 9y/8mo
(*) Mean biomator treatment time = 2.7 years
The cephalograms of 13 male and 20 female Class II, division
1 malocclusion patients who were consecutively treated with the same
protocol were collected from the same orthodontist. TI corresponds to a
mean age of 10.2 years, T2 corresponds to a mean age of 13.1 years and
T3 corresponds to a mean age of 20 years (7 years after the point in time
at which treatment was finished). The mean time for treatment with the
Balters biomator was 2.7 years.
- The cephalograms were standardized with regard to magnification
and then digitized using a digitizing tablet (Numonics, Lansdale, Pennsyl-
vania, PA) and digitizing software (JOE, RMO, Denver, Colorado). The
cephalograms were analyzed using Ricketts’ and McNamara’s orthopedic
evaluation methods (Ricketts et al., 1982; McNamara, 1984; Fig. 14).

96
Falfin
Figure 14. Cephalometric tracing based on the methods of
Ricketts and McNamara.
The following variables were measured:
• Facial axis angle = Ba-Na with Pt - Gn
• Mandibular plane angle = Me – Go with Frankfort Horizontal
• Se-Na-Point A = SNA angle
• Se-Na-Point B = SNB angle
• A-Na-B = ANB angle
• Effective maxillary length = Co - Point A
• Effective mandibular length = Co - Gn
Treatment effects at the end of treatment (T2-T1), at the end of the growth
period (T3–T2) and long-term, overall treatment effects (T3–T1) are seen
in Table 5.
During active treatment (T2-T1), the facial axis angle decreased
05°,+17° and after treatment (T3–T2), it increased 13° E1.3°. This result
indicates clearly that after active treatment, the facial axis increased more
than it decreased during treatment, resulting in a normal pattern of growth
(+0.7°). The same observation can be made relative to the mandibular
plane angle: during active treatment, there was an increase of 0.2°,+1.7°:
after active treatment, there was a decrease of 2.9° E2.0° and an overall
2.7°,+2.1° decrease in counterclockwise rotation. These values show that
treatment with the Balters bionator does not produce a vertical opening of
the mandibular plane angle; on the contrary, it avoids clockwise rotation
of the mandible.

97
Balters Bionator
Table 5. Treatment outcomes.

Time Points Differences in Time Points
Factors T1 T2 T3 T2 - T1 T3 - T2 T3 - T1
º * | 88.7°4. 4.0 | 88.1° E 3.9 | 89.4° 44.1 -0.5° E 1.7 | 1.3° E 1.3 0.7° E 2.0
* |25.1°152 |25.3°, 5.4|224, 16.2 || 02" | L7 |-29' 12.0|-2.7°12.
plane angle
*** | 81.1° 13.5 | 81.1° 13.9 || 81.1° a 3.7 || 00 0.0 0.0
A angle
***" |75.0°43.5 |764° 136||780° 13.3 | 1.4° E 1.8 | 1.5 + 1.4 |29' 12.
B angle
ANB angle | 6.1°E 2.0 || 4.6° E 1.9 || 3.2°E 2.1 | -1.4° E 1.7 |-1.5°E 1.1 |-2.9° E 1.9
Effective
maxillary 88.6 + 4.6 || 92.2 + 5.6 || 95.3 + 5.4 || 3.6 + 3.8 || 3.1 + 3.0 | 6.7 -- 4.3
length (mm) *
Effective
mandibular | 110.4 + 6.2 | 119.8 + 8.2| 128.3 + 8.7 | 9.5 + 7.0 | 8.5 + 6.1 | 18.0 + 7.7
length (mm)
There were no differences in the values for SNA angle throughout
the entire observation period, and there was a normal increase in the ef-
fective midface length when compared to norms published by McNamara
(McNamara and Brudon, 2001). These results demonstrate that the treat-
ment protocol has no effect on maxillary growth.
There was an increase of 1.4° =1.8° in SNB angle during active
treatment and another increase of 1.5°-E1.4° by the end of treatment, mak-
ing the overall increase 2.9° E2.1°. The ANB angle exhibited similar be-
havior to that of the SNB angle; it decreased 1.4° +1.7° during the active
phase of treatment and another 1.5° E1, 1% in the post-treatment period,
producing an overall decrease of 2.9° E1.9°. This corresponds to an effec-
tive factor enhancing facial harmony in the sagittal plane.
Total mandibular length increased 9.5 +7.0 mm during the active
treatment phase and another 8.5 +6.1 mm in the post-treatment period for
a total increase of 18.0 +7.7 mm. This amount of mandibular growth is
significantly greater than expected for the pre- to post-pubertal period of
development in subjects with untreated Class II, division 1 malocclusions
(14.3 +2.2 mm). It is comparable to the expected mandibular growth of
subjects with untreated, Class I occlusions (17.2 +2.3 mm; Stahl et al.,
in press). In conclusion, we found no relapse after active treatment was
finished (Figs. 15-19).
98
Falfin
Figure 16. End-of-treatment intraoral photos.
Figure 17. Long-term (LT) intraoral follow-up photos.
- -
Figure 18. Initial (left), end-of-treatment (middle) and long-term follow-
up (right) profile facial photographs.
Figure 19. Initial (left), end-of-treatment (middle) and long-term follow
up (right) lateral x-rays and cephalograms.








99
Balters Bionator
FINAL CONSIDERATIONS
The results of the investigations discussed in this chapter demon-
strate, based on clinical evidence, that the Class II, division 1 malocclu-
sion with mandibular retrognathism can be corrected with a significant
component of enhancement of mandibular growth that occurs during ac-
tive treatment time by means of the original Balters bionator appliance and
a step-by-step advancement construction protocol.
The best timing for the start of treatment is near the pubertal
growth peak, with the active treatment period including the period of
peak growth. In order to optimize orthopedic results, the treatment period
should encompass acceleration and deceleration portions of individual pu-
bertal growth curves.
During the long-term post-treatment follow-up, complementary
growth adaptations occurred and facial harmony was enhanced. This con-
firms that there was no orthopedic relapse. Functional facial orthopedics,
as described by Balters (1964, 1973) and Fränkel (1989), works in an ab-
solute and physiological way, restoring the natural mechanisms of normal
craniofacial development and function.
ACKNOWLEDGMENTS
The author would like to express his gratitude to the following
orthodontists Drs. Márcia Aparecida Alves de Almeida, Rolf Marçon Fal-
tin, Lorenzo Franchi, Tiziano Baccetti and James A. McNamara, Jr. for
their collaboration on this manuscript and scientific research. This paper
is dedicated to the passion and enormous intellectual contributions of Wil-
hem Balters, Rolf Fränkel and Alexandre Petrovic.
REFERENCES
Baccetti T, Franchi L, McNamara JA Jr. An improved version of the cer-
vical vertebral maturation (CVM) method for the assessment of man-
dibular growth. Angle Orthod 2002;72:316–323.
Baccetti T, Franchi L, Toth LR, McNamara JA Jr. Treatment timing for
twin block therapy. Am J Orthod Dentofacial Orthop 2000;118:159-
170.
Baccetti T, Franchi L. Maximizing esthetic and functional changes in
Class II treatment by means of appropriate treatment timing. In: Mc-
Namara JA Jr, Kelly K, eds. Frontiers in Facial and Dental Esthetics.
100
Faltin
Craniofacial Growth Series, Department of Orthodontics and Pediat-
ric Dentistry and Center for Human Growth and Development, The
University of Michigan, Ann Arbor 2001:38:237-251.
Balters W. Die Technik und Übung der allgemeinen und speziellen Biona-
tor-Therapie. Quintessenz 1964;Heft 5.77, IV.
Balters W. Einführung in die Bionator Heilmethode Ausgewöhlte, Schrif-
ten und Vortrange. Heidelberg: Drückerei Holzer; 1973.
Bolmgren GA, Moshiri F. Bionator treatment in Class II, division 1. Angle
Orthod 1986:56:255-262.
Faltin Jr K, Faltin RM, Baccetti T. Franchi L. Ghiozzi B, McNamara JA
Jr. Long-term effectiveness and treatment timing for biomator therapy.
Angle Orthod 2003;73:221-230.
Franchi L, Baccetti T, McNamara JA Jr. Mandibular growth as related to
cervical vertebral maturation and body height. Am J Orthod Dentofa-
cial Orthop 2000;118:335-340.
Franchi L, Baccetti T, McNamara JA Jr. Treatment and post-treatment ef-
fects of acrylic splint Herbst appliance therapy. Am J Orthod Dentofa-
cial Orthop 1999; 115:429-438.
Fränkel R. Orofacial Orthopedics with Function Regulator. Berlin: Kar-
ger Gmbh, 1989.
Hägg U, Pancherz H. Dentofacial orthopaedics in relation to chronologi-
cal age, growth period and skeletal development. An analysis of 72
male patients with Class II, division 1 malocclusion treated with the
Herbst appliance. Eur J Orthod 1988; 10:169-176.
Halazonetis DJ. Computer-assisted cephalometric analysis. Am J Orthod
Dentofacial Orthop 1994;105:517-521.
Illing HM, Morris DO, Lee RT. A prospective evaluation of Bass, bionator,
and twin block appliances. Part I — The hard tissues. Eur J Orthod
1998:20:501-516.
Janson I. A cephalometric study of the efficiency of the bionator. Trans Eur
Orthod Soc 1977:53:283-293.
Janson I. Skeletal and dentoalveolar changes in patients treated with a bi-
onator during puberty and prepuberty. In: McNamara JA Jr., Ribbens
KA, Howe RP, eds. Clinical Alterations of the Growing Face. Cra-
niofacial Growth Series, Center for Human Growth and Development,
The University of Michigan, Ann Arbor 1983;16:131-154.
Johnston LE Jr. A comparative analysis of Class II treatments. In: Vig
PS, Ribbens KA, eds. Science and Clinical Judgment in Orthodon-
tics. The Center for Human Growth and Development and the
101
Balters Bionator
Department of Orthodontics and Pediatric Dentistry, The University
of Michigan, Ann Arbor 1986; 19:103-198.
Lange DW, Kalra V. Broadbent HB, Powers M, Nelson S. Changes in
soft tissue profile following treatment with the bionator. Angle Orthod
1995;65:423-430.
Malmgren O, Ömblus J, Hägg U, Pancherz H. Treatment with an appli-
ance system in relation to treatment intensity and growth periods. Am
J Orthod Dentofacial Orthop 1987;91: 143-151.
Mamandras AH, Allen LP. Mandibular response to orthodontic treatment
with the bionator appliance. Am J Orthod Dentofacial Orthop 1990;
97: 113–120.
McNamara JA Jr, Bookstein FL, Shaughnessy TG. Skeletal and dental
changes following functional regulator therapy on Class II patients.
Am J Orthod 1985;88:91-110.
McNamara JA Jr, Brudon WL. Orthodontics and Dentofacial Orthope-
dics. Ann Arbor, Needham Press 2001:319-332.
McNamara JA Jr. A method of cephalometric evaluation. Am J Orthod
1984;86:449-469.
Morris DO, Illing HM, Lee RT. A prospective evaluation of Bass, bionator
and twin block appliances. Part II: The soft tissues. Eur J Orthod
1998:20:663-684. &
Moyers, RE. Handbook of Orthodontics. Chicago. Year Book Medical
Publishers, 1988.
Pancherz H. The mechanism of Class II correction in Herbst appliance
treatment. A cephalometric investigation. Am J Orthod 1982;82:104-
113.
Petrovic A, Stutzmann J, Lavergne J, Shaye R. Is it possible to modulate
the growth of the human mandible with a functional appliance? Int J
Orthod 1991:29:3-8.
Petrovic A, Stutzmann J, Lavergne J. Mechanism of craniofacial growth
and modus operandi of functional appliances: A cell-level and cyber-
netic approach to orthodontic decision making. In: Carlson DS, ed,
Craniofacial Growth Theory and Orthodontic Treatment. Craniofa-
cial Growth Series, Center for Human Growth and Development, The
University of Michigan, Ann Arbor 1990:23:13-74.
Ricketts, RM, Roth RH, Chaconas SJ, Schulhof, RJ, Engel GA. Orthodon-
tic Diagnosis and Planning. St. Louis, Rocky Mountain Orthodontics
1982:127–136.
Riolo ML, Moyers RE, McNamara JA Jr, Hunter WS. An Atlas of Cra-
niofacial Growth: Cephalometric Standards from the University
102
Faltin
School Growth Study. Craniofacial Growth Series, Center for Human
Growth and Development, The University of Michigan, Ann Arbor,
1974.
Rudzki-Janson I, Noachtar R. Functional appliance therapy with the bion-
ator. Seminar Orthod 1998:4:33–45.
Stahl F. Baccetti T, Franchi L, McNamara JA Jr. Longitudinal growth
changes in untreated subjects with Class II and Class I malocclusion.
Am J Orthod Dentofacial Orthop, in press.
Tulloch JFC, Phillips C, Koch G, Proffit WR. The effect of early interven-
tion on skeletal pattern in Class II malocclusion: a randomized clinical
trial. Am J Orthod Dentofacial Orthop 1997;111:391-400.
103
104
THE LONG-TERM PERSPECTIVE ON ORTHOPEDIC
TREATMENT OF CLASS III MALOCCLUSION
Tiziano Baccetti
Lorenzo Franchi
Attack is the best defense.
There is ample evidence to prove that the Class III malocclusion is a dis-
harmony that demands a great deal from the orthodontist. Recent data have
shown that the result of early orthopedic treatment in untreated subjects
with Class III malocclusions is unstable (Reyes et al., 2006). Excessive
mandibular growth throughout the circumpubertal period, longer duration
of the pubertal growth spurt with respect to subjects with normal occlusion,
and continued growth during the post-pubertal stages are all elements that
challenge the possibility of maintaining a Class III correction achieved
by orthopedic treatment throughout the developmental ages. With these
concerns in mind, the purpose of this paper is to attempt to answer three
fundamental questions:
1. What is the average long-term effectiveness of orthopedic treat-
ment, i.e., rapid maxillary expansion (RME) and facial mask (FM)
therapy followed by fixed appliance therapy, of Class III maloc-
clusions?
2. What are the main factors contributing to the long-term success of
orthopedic Class III treatment?
3. Is it worth treating every patient who has a Class III malocclusion
orthopedically?
WHAT IS THE AVERAGE LONG-TERM
EFFECTIVENESS OF ORTHOPEDIC
TREATMENT OF CLASS III MALOCCLUSION?
Our research group looked at the long-term effects of RME and
FM therapy on Class III malocclusions. This study (Westwood et al.,
2003) looked at patients who underwent an initial phase of rapid maxil-
lary expansion and protraction facemask therapy followed by a second
phase of pre-adjusted edgewise therapy. Specific outcomes were as-
sessed for all patients following the pubertal growth spurt as determined
by the cervical vertebral maturation (CVM) method (Baccetti et al.,
2005). Craniofacial changes in treated Class III subjects were compared
105
Long-Term Perspective
to the growth changes in untreated Class III controls during the treatment
and post-RME/FM phases.
Sample Selection
The parent sample consisted of the cephalometric records of 102
Class III subjects who were treated with rapid maxillary expansion and
protraction facemask therapy followed by comprehensive pre-adjusted
edgewise therapy. Twenty female and 14 male patients were selected for
the treatment group (TG) from this sample based on the following crite-
I 131.
1. All patients were of European-American ancestry (Caucasian).
2. Each patient presented with a Class III malocclusion when first
examined (T,) that was characterized by an anterior crossbite or
edge-to-edge incisal relationship and a Wits appraisal (Jacobson,
1975) of -1.5 mm or less.
3. No permanent teeth were congenitally missing or extracted before
or during treatment.
4. All patients had good quality cephalograms taken prior to the start
of RME/FM therapy (T) and again at the beginning of the long-
term observation period following the end of two-phase treatment
(T,). -
5. At the final examination (T,), skeletal maturation for all patients
was post-pubertal based on the CVM method of developmental
staging, i.e., CVM stages four through six (Baccetti et al., 2005).
Twenty-two untreated Class III control subjects (13 female and 9
male) were obtained from the Department of Orthodontics at the Univer-
sity of Florence, from The University of Michigan Elementary and Sec-
ondary School Growth Study (Riolo et al., 1974) and from three private
orthodontic practices in Michigan. Magnification was corrected to an 8%
enlargement for all radiographs included in both the treated and the control
samples. The matched control sample allowed for a direct comparison of
treatment effects on the differences between the values at different time
periods without the need for annualizing the data.
Lateral cephalograms taken at all three observation periods for the
treated patients were analyzed. The treated subjects then were compared
to the control group in order to evaluate the long-term effectiveness of
RME/FM therapy followed by fixed appliance therapy. The mean age of
the groups at T and T, and the mean duration of observation intervals are
reported in Table 1.
106
Baccetti and Franchi
Table 1.
Treatment Group (TG)
Observation
Period/Interval N Mean SD Minimum Maximum
T 34 8y 3m ly 10m 5y 5m 12y 0m
T2 34 14y 8m ly 9m 11y 4m 20y 8m
T1-T2 34 6y 4m 2y 3m 2y lm 12y 11m

Control Group (TG)

N Mean SD Minimum Maximum
T 22 8y 8m 2y 5m 4y 2m 14y 2m
T2 22 15y 2m ly 1 lm 12y Om 18y 8m
T1-T2 22 6y 5m 2y 2m 3y 5m 1 ly 3m
Treatment Protocol
The three components of orthopedic facemask therapy used in
this study were a maxillary expansion appliance, a facemask, and heavy
elastics (McNamara and Brudon, 2001). Treatment was initiated with the
placement of a bonded or banded maxillary expander to which were at-
tached vestibular hooks extending in a superior and anterior direction. Pa-
tients were instructed to activate the expander 1-2 times per day until the
desired transverse width was achieved.
Patients were given facemasks with pads fitted to the chin and
forehead for support either during or immediately after expansion. Elas-
tics were attached from the soldered hooks on the expander to the sup-
port bar of the facemask in a downward and forward vector, producing
orthopedic force levels up to 300 to 600 grams per side. Patients were
instructed to wear the facemask for a minimum of 14 hours per day. All
patients were treated to a positive dental overjet before treatment was dis-
continued. Most patients were overcorrected toward a Class II occlusal
relationship. As occurs in any study involving removable appliances, pa-
tient compliance with the instructions of the orthodontist and staff varied.
After an interim period, most patients underwent a second phase
of pre-adjusted edgewise therapy, during which a removable maxillary
stabilization plate typically was worn. In a few instances, this second
107
Long-Term Perspective
phase of therapy was begun immediately following the rapid maxillary
expansion and protraction treatment phase. On average, fixed appliance
therapy lasted 27 months. Class III elastics (and in a few instances Class
II elastics) were used when appropriate to eliminate minor occlusal dis-
crepancies.
Cervical Vertebral Maturation Staging
Developmental staging of the cervical vertebrae serves as a bio-
logical indicator of skeletal maturity for males and females independent of
their chronological age (Baccetti et al., 2005). It was the aim of this study,
therefore, to evaluate the long-term effects of RME/FM followed by fixed
appliance therapy for individuals based on post-pubertal skeletal maturity
rather than on chronological age.
The pubertal peak in skeletal growth rate occurs between cervical
vertebral maturation stages three and four. It is during this interval that the
greatest increments of mandibular growth rate are observed. After stage
four, there is a significant deceleration in craniofacial growth that contin-
ues through stage six. For this reason, the long-term effects of two-phase
Class III correction (T,) were assessed when individuals had attained a
skeletal maturity stage of CS4, CS5 or CS6 as these stages correlate to
post-pubertal growth spurt maturity.
Cephalometric Analysis
A customized digitization regimen and analysis procedure de-
signed using Dentofacial Planner" software (Dentofacial Software, To-
ronto, Ontario, Canada) was used for all of the cephalograms that were ex-
amined in this study. The cephalometric analysis required the digitization
of 77 landmarks and 4 fiducial markers. The customized cephalometric
analysis containing measurements from the analyses of Jacobson (1975),
McNamara (1984), Ricketts (1960), and Steiner (1953) generated 36 vari-
ables (13 angular and 23 linear) for each tracing.
Statistical Analysis
Multivariate analysis did not reveal significant differences between
the treated group (TG) and the control group (CG) at T. In order to assess
the overall effects of the Class III treatment, the craniofacial changes that
occurred from T to T, in the treated group (TG) were contrasted to those
that occurred in the control group (CG) by means of independent-sample
t–teStS.
108
Baccetti and Franchi
Method Error
Linear measurement accuracy ranged from 0.1 mm to 0.3 mm,
with a standard deviation of approximately 0.8 mm. Angular measure-
ments varied 0.1 degrees, with a standard deviation ranging from 0.4 to
0.6 degrees.
Results (Table 2 and Fig. 1)
In the long term, the skeletal changes in the treated group included
a slight (1.2 mm) increase in the sagittal position of the maxilla at Point
A relative to the nasion perpendicular when compared to the untreated
Class III control group. In the mandible, there was a significantly smaller
increase in both the SNB angle (-2.6°, p < 0.01) and the sagittal position of
pogonion (-3.0 mm) when compared to untreated Class III controls. The
combined effects of the maxillary and mandibular skeletal changes result-
ed in significant increases in both the Wits appraisal (6.1 mm) and closure
of the ANB angle (2.9°). In addition, significantly smaller increments in
the maxillomandibular differential of the treated group (-4.1 mm, p < 0.01)
were noted. With the exception of the inclination of the occlusal plane to
Frankfort horizontal displaying a greater decrease in the treated group, no
significant changes in the vertical dimension were noted throughout the
long term.
Overall, the overjet increased significantly in the treated group
relative to controls (4.4 mm), whereas the molar relationship decreased
significantly (-3.9 mm). In addition, the anteroposterior position of the
lower incisor to the Point A-Pogonion line decreased significantly in the
treated group (p<0.01).
Orthopedic and orthodontic intervention in Class III patients by
means of a two-phase therapeutic approach consisting of rapid maxil-
lary expansion and protraction followed by fixed appliance therapy leads,
on average, to significant skeletal and dentoalveolar improvements in
the long term. The correction was due mainly to favorable changes in
the mandibular component of the skeletal discrepancy during RME/FM
therapy; improvements in maxillary position played a lesser role. Al-
though the reduction in mandibular length (approximately -2.5 mm),
closing of the mandibular plane angle (-1.2°) and increases in the mid-
facial length (1.6 mm) and upper face height (1.0 mm) were not signifi-
cant “independently” over the long term, the cumulative effects of these
craniofacial modifications cannot be overlooked as contributing factors
to the overall maintenance of positive dental relationships. The actual
amount of improvement in the treated group during the observation period
109
Long-Term Perspective
Table 2.
Treated Group (TG) Control Group (CG) TG vs CG Ti-T2
N=34 N=22
Cephalometric Measures Mean SD Mean SD Net p value
Difference
Cranial Base
Cranial flexure (°) 0.8 2.1 -0.6 3.0 1.4 ,037 *
Maxillary Skeletal
Co-Pt A (mm) 9.2 3.9 7.6 2.9 1.6 ... 105
SNA (°) 1.8 2.0 1.5 2.7 0.3 .628
Pt A to Nasion perp (mm) 1.7 1.8 0.5 2.1 1.2 .028 +
Mandibular Skeletal
Co-Gn (mm) 16.9 7.5 19.3 6.8 -2.4 .222
SNB (°) 0.7 2.0 3.3 2.9 -2.6 ,000 ***
Pg to Nasion perp (mm) 3.3 4.0 6.3 4.5 -3.0 .012 *
Gonial angle (°) -4.1 3.7 -2.4 3.3 -1.7 .094
Maxillary/Mandibular
Wits appraisal (mm) 3.4 2.3 -2.7 2.3 6.1 ,000 ***
Max/Mand differential (mm) 7.6 4.5 11.7 4.7 -4.1 .002 **
ANB (°) 1.0 1.5 - 1.9 1.9 2.9 ,000 ***
Pertical Skeletal
FH to occlusal plane (°) , -4.4 *3.5 -2.0 3.1 -2.4 .01.1 °
FH to palatal plane (°) -0.9 1.9 -0.6 2.3 -0.3 .612
MPA (*) -0.9 2.7 -2.1 3.2 1.2 .146
Nasion to ANS (mm) 7.7 3.6 6.7 4.0 1.0 .339
ANS to Me (mm) 8.9 3.8 9.8 4.9 -0.9 .449
Interdental
Overbite (mm) 0.7 1.4 0.7 2.0 0.0 .981
Overjet (mm) 3.2 2.3 -1.2 2.7 4.4 ,000 ***
Interincisal angle (°) -5.8 10.2 -2.8 12.9 -3.0 .345
Molar relationship (mm) -0.5 2.2 3.4 2.9 -3.9 ,000 ***
Maxillary Dentoalveolar
U1 to Pt A vert (mm) 2.9 2.1 2.9 2.1 0.0 .992
U1 horizontal (mm) 3.4 2.3 3.6 2.6 -0.2 .706
U1 vertical (mm) 2.1 1.9 2.7 2.2 -0.6 .250
U6 horizontal (mm) 3.9 2.0 3.8 3.0 0.1 .920
U6 vertical (mm) 5.0 2.4 4.5 1.9 0.5 .439
U1 to Frankfort (*) 8.4 7.2 6.4 6.6 2.0 .295
Mandibular Dentoalveolar
L1 to Pt A Pg (mm) -1.2 2.1 1.6 2.1 -2.8 ,000 ***
Ll horizontal (mm) -1.4 1.9 -1.1 1.6 -0.3 .545
Ll vertical (mm) 6.5 2.5 6.7 3.3 -0.2 .757
L6 horizontal (mm) 2.2 1.4 1.5 2.4 0.7 .168
L6 vertical (mm) 5.1 2.3 5.9 2.6 -0.8 .205
L1 to MPA (°) -1.8 6.6 -1.6 7.0 -0.2 .899
Soft Tissue
UL to E plane (mm) -2.0 2.2 -3.2 2.4 1.2 .054
LL to E plane (mm) –2.2 1.9 -2.2 2.5 0.0 .947
Nasolabial angle (°) -2.9 10.6 –2.3 13.9 -0.6 .868
Cant of upper lip (*) -0.2 8.1 1.9 9.2 -2.1 .371
* p-0.05; **p-0.01 ; *** p-0.001
matched the average amount of correction needed for the molar relation-
ship. The improvement needed for the overjet at the start of treatment
(approximately 4 mm for both measurements) was achieved also. Class
III therapy effectively produced a normal dental relationship in the long
term following the pubertal growth spurt, although the pattern of Class III
growth was re-established in the post-treatment period (see Westwood et
al., 2003 for details).
Was orthopedic treatment successful in each and every patient?
No! One out offive patients who were treated by means of the RME/FM
110
Baccetti and Franchi
A. Treated Group
(Ti to T2) = 6y 4m
T.
T2
B. Control Group
(Ti to T.) = 6y 5m
Ti — +
T2
Figure 1. Craniofacial changes derived from superimpositions of the T-T inter-
Val A. Treated group. B. Control group.
Protocol did not have acceptable occlusal/skeletal outcomes in the long
ºrm. This brings us to the next question.


| ||
Long-Term Perspective
WHAT ARE THE MAIN FACTORS CONTRIBUTING TO
THE LONG-TERM SUCCESS OF
ORTHOPEDIC CLASS III TREATMENT2
The findings of Westwood and co-workers (2003) indicate that the
establishment of a positive occlusal relationship during RME/FM therapy
helps maintain a normal dental relationship five and one-half years after
the initial phase of therapy. In fact, 26 of the 34 treated individuals (76%)
had a positive overjet relationship after attaining post-pubertal skeletal
maturation, and three additional patients (9%) had edge-to-edge incisal re-
lationships. Ngan and co-workers (1997) reported similar outcomes with
respect to overjet in 75% of the 20 Chinese subjects followed for four
years after maxillary expansion and protraction treatment.
Evaluation of the overall treatment and post-treatment changes
clearly indicates that the beneficial changes on the skeletal relationship can
be obtained only during active rapid maxillary expansion and protraction
therapy, and that the later phase of fixed-appliance therapy can maintain
the dental components. About 93% of the skeletal changes that occurred
prior to the pubertal growth spurt withstood the subsequent growth that
occurred during the remaining stages of skeletal maturation.
Overcorrection of Class III skeletal discrepancies with orthopedic
appliances always is advisable. In our study, patients corrected to overjets
of 4.5 mm or greater during the initial phase of RME/FM treatment all
sustained favourable outcomes over the long term. The eight individuals
who were unable to maintain a positive overjet throughout the pubertal
growth spurt had attained smaller increments of overjet change compared
to the other patients. Aggressive overcorrection of the Class III skeletal
malocclusion to an overjet of perhaps as much as 5 to 8 mm and a Class II
molar relationship is recommended.
There is evidence that suggests that some specific pre-treat-
ment cephalometric characteristics such as cranial base angulation, size
of the mandibular ramus, and inclination of the mandibular plane to the
cranial base may provide prognostic indications of the long-term out-
come of RME/FM therapy in Class III patients (Baccetti et al., 2004).
The narrower the cranial base angle, the longer the mandibular ramus; the
steeper the mandibular plane angle, the more unfavorable the prognosis
for Class III orthopedic treatment. Still, 15% to 25% of patients treated
with orthopedic appliances during developmental ages experience incom-
plete success/failure of therapy in the long term. The next relevant ques-
tion, therefore, is should all Class III patients be treated orthopedically?
112
Baccetti and Franchi
SHOULD EVERY PATIENT WITH A CLASS III
MALOCCLUSION BE TREATED ORTHOPEDICALLY2
Too often patients with severe Class III malocclusions are classi-
fied as surgical patients because it is believed that patients who undergo
orthopedic treatment during the pre-pubertal growth phase will have un-
favorable treatment outcomes. They are left untreated orthopedically, and
are referred to orthognathic surgeons for treatment at the completion of
their growth spurt. This belief implies that surgery would be able poten-
tially to correct a Class III malocclusion regardless of the severity of the
dentoskeletal disharmony. Unfortunately, even orthognathic surgery has
some limitations, often in the form of relapse after surgery.
A systematic review of the literature based on the terms “orthogna-
thic surgery,” “Class III malocclusion” and “relapse” was able to detect at
least one major factor in the rebound of Class III dentoskeletal character-
istics after surgery: a combined maxillomandibular approach. The review
retrieved three scientific articles (Franco et al., 1989; Proffit et al., 1991;
Politi et al., 2004) that deal with the identification of predictive factors for
relapse following Class III surgery. All three articles agree that when the
Sagittal intermaxillary discrepancy in Class III patients is greater than 7
to 8 mm before surgery, orthognathic surgical outcomes are less stable
when re-evaluated three years post-treatment.
In order to investigate this topic further and to correlate the findings
with the role of orthopedic treatment in Class III patients, we re-analyzed
the long-term data of the study by Westwood and co-workers (2003). We
chose the Wits appraisal as a measurement for sagittal intermaxillary dis-
crepancy in both treated and untreated Class III patients at the long-term
(post-pubertal) observation. The results showed that 100% of the Class III
patients who received RME/FM treatment presented with a sagittal inter-
maxillary discrepancy smaller than 7 mm, which was true for only 60% of
the Class III controls who were left orthopedically untreated. This means
that 100% of the Class III patients who were treated with orthopedic ap-
pliances would be candidates for stable results of orthognathic surgery in
case of long-term failure of orthopedic treatment. On the contrary, stable
outcomes of orthognathic surgery could be anticipated in only 60% of the
Class III subjects who had not experienced orthopedic treatment during
growth on the basis of the large intermaxillary discrepancy that those sub-
jects exhibited at the completion of growth, i.e., before surgery.
These findings are clinically significant because they shed new
light on the relationship between orthodontics and orthognathic surgery
113
Long-Term Perspective
for Class III malocclusions. Instead of the dichotomy of the past, ortho-
pedic treatment can be a first treatment choice, with orthognathic surgery
the treatment of choice for those patients whose Class III malocclusion
re-establishes itself after puberty.
Is it worth it to orthopedically treat every patient with a Class III
malocclusion? The answer is yes! Orthopedic therapy will correct Class
III malocclusions in 75% to 85% of Class III patients. The remaining
15% to 25% of Class III patients, those whose orthopedic corrections re-
lapse, will experience a more stable outcome with orthognathic surgery
performed after completion of growth.
REFERENCES
Baccetti T, Franchi L, McNamara JA Jr. Cephalometric variables pre-
dicting the long-term success or failure of combined rapid maxillary
expansion and facial mask therapy. Am J Orthod Dentofacial Orthop
2004; 126:16-22.
Baccetti T, Franchi L, McNamara JA Jr. The cervical vertebral maturation
(CVM) method for the assessment of optimal treatment timing in den-
tofacial orthopedics. Semin Orthod 2005; 11:119-129.
Franco JE, Van Sickels JE, Thrash WJ. Factors contributing to relapse in
rigidly fixed mandibular setbacks. J Oral Maxillofac Surg 1989;47:
451–456.
Jacobson A. The “Wits” appraisal of jaw disharmony. Am J Orthod
1975;67:125-138.
McNamara JA Jr. A method of cephalometric evaluation. Am J Orthod
1984;86:449-469.
McNamara JA Jr, Brudon WL. Orthodontics and Dentofacial Orthope-
dics. Ann Arbor, Needham Press 2001:67–80.
Ngan P, Hägg U, Yiu C, Wei H. Treatment response and long-term dento-
facial adaptations to maxillary expansion and protraction. Semin Or-
thod 1997;3:255-264.
Politi M, Costa F, Cian R, Polini F, Robiony M. Stability of skeletal class
III malocclusion after combined maxillary and mandibular procedures:
Rigid internal fixation versus wire osteosynthesis of the mandible. J
Oral Maxillofac Surg 2004;62:169-181.
Proffit WR, Phillips C, Turvey TA. Stability after surgical-orthodontic
corrective of skeletal Class III malocclusion. 3. Combined maxillary
and mandibular procedures. Int J Adult Orthodon Orthognath Surg
1991;6:211-225.
114
Baccetti and Franchi
Reyes BC, Baccetti T. McNamara JA Jr. An estimate of craniofacial growth
in Class III malocclusion. Angle Orthod 2006;76:577–584.
Ricketts RM. The influence of orthodontic treatment on facial growth and
development. Angle Orthod 1960:30:103-133.
Riolo ML, Moyers RE, McNamara JA Jr, Hunter W.S. An Atlas of Cranio-
facial Growth: Cephalometric Standards from The University School
Growth Study, Craniofacial Growth Series, Monograph, Center for
Human Growth and Development, The University of Michigan, Ann
Arbor, 1974.
Steiner CC. Cephalometrics for you and me. Am J Orthod 1953:39:729-
755.
Westwood PV. McNamara JA Jr, Baccetti T, Franchi L, Sarver DM. Long-
term effects of Class III treatment with rapid maxillary expansion and
facemask therapy followed by fixed appliances. Am J Orthod Dento-
fac Orthop 2003;123:306–320.
115
116
FACEMASK THERAPY FOR CLASS III
MALOCCLUSION: SHORT-TERMAND
LONG-TERM OUTCOMES
Patrick K. Turley
In the last two decades, facemask therapy has become a common approach
for the treatment of growing children with Class III malocclusions. With
increasing use, however, questions concerning the effects and long-term
stability of facemask treatment have become more important. For ex-
ample, although we commonly refer to facemask therapy as maxillary
protraction, is that really the main effect of treatment? How much of the
correction is maxillary versus mandibular, and how much is skeletal ver-
sus dental? Does palatal expansion accentuate the amount of maxillary
protraction obtained? What is the optimal time to initiate treatment? Is
there a small window of opportunity, or can children of various ages ben-
efit from treatment? What happens after facemask therapy has been com-
pleted? Does the maxilla relapse backward? Does the maxilla continue to
grow forward? Does the maxilla grow like Class I or Class III? Does the
mandible rotate/grow forward to the pretreatment condition? What are the
long-term effects of facemask therapy? In this paper, we will attempt to
provide answers to these questions from an analysis of published studies
from the literature. -
TREATMENT PROTOCOL
Most facemasks incorporate a forehead, chin support and a labial
bow for the attachment of elastics. Various types of palatal expansion
appliances are used, however, such as the bonded occlusal coverage ap-
pliance, the Haas, the Hyrax and the quad helix appliance. Our treatment
protocol involves the use of a banded, soldered, and acrylic palatal expan-
sion appliance as described by Haas (1965, 1970a,b). A custom facemask,
fabricated from a facial impression and plaster model, is worn three to four
months full-time until positive overjet is achieved. This is followed by
three to four months of daytime wear for overcorrection if necessary, and
three to four months of bedtime wear for retention (Fig. 1; Turley, 1988).
117
Facemask Therapy for Class III Malocclusion
Figure 1. Palatal expansion appliance and custom facemask.

118
Turley
SHORT TERM EFFECTS
To determine the effects of facemask (FM) therapy, we examined
the records of 21 patients whose ages ranged from 3.8 years to 10.8 years
and averaged 7.2 years. Treatment time averaged 11.1 months with a
range of 5 months to 29.8 months (Nartallo-Turley, 1998). We performed
an extensive cephalometric analysis involving traditional cephalometric
measures, an X-Y coordinate system, and occlusal plane analysis as de-
scribed by Johnston (1986). Phase 1 treatment changes were examined by
comparing the differences between the pretreatment (T1) and post-treat-
ment (T2) lateral cephalograms.
Facemask therapy resulted in statistically significant anterior
movement of the maxilla, with increases in SNA (2.4°), maxillary depth
(2.2°) and ANB (3.7°) angles. Anterior movement of point A (3.3 mm) and
ANS (3.2 mm) also were observed. The maxilla rotated counterclockwise,
with PNS moving down more than ANS (2.2 and 0.8 mm respectively).
Mandibular position also changed, with mild decreases in SNB (-1.3°) and
facial depth (1.1°), and significant downward movement at menton (-4.3
mm). The occlusal plane analysis demonstrated that the correction was
due more to the maxilla than mandible (2.4 and -1.9 mm respectively).
Dentally, the upper molars and incisors moved forward (1.7 mm) in the
maxilla. Molar and overjet correction measured 5.6 mm and 5.2 mm re-
spectively. The skeletal and dental changes that occurred resulted in the
profile becoming more convex. The nose and upper lip moved forward
3.4 mm and 3.7 mm respectively, whereas the soft tissue chin (menton)
moved down 3.5 mm, increasing lower face height. The lower lip and the
chin remained about the same anteroposteriorly. These results show that
facemask/palatal expansion therapy improved the Class III malocclusion
through a combination of skeletal and dental changes that occurred both
anteroposteriorly and vertically (Fig. 2).
How do these results compare to those reported by other au-
thors? In the last ten years, several studies have reported downward and
backward movement of the mandible combined with retroclination of
the mandibular incisors (Chong et al., 1996; Bacetti et al., 1998; Degu-
chi, 1999; Westwood et al., 2003, Franchi et al., 2004). Chong and col-
leagues (1996) stated, for example, that “the major reason for negative
overjet correction in our sample was downward and backward movement
of the mandible combined with retroclination of the mandibular inci-
sors.” How much of the correction actually is due to the maxilla and how
much is due to the mandible? The amount of maxillary protraction ob-
tained may be due to several factors including appliance design, elastic
119
Facemask Therapy for Class III Malocclusion
force levels and direction of pull, the mobility or “disarticulation” of the
maxilla caused by palatal expansion, the number of hours the appliance
is worn per day, or the overall treatment time. Williams and colleagues
(1997) and Chong and colleagues (1996) both reported less than one degree
of improvement in the SNA value with treatment; however, their overall
treatment time was only 4.7 and 7.3 months respectively. Studies report-
ing greater than two degrees of improvement in SNA had a much longer
treatment time of approximately one year (Kapust et al., 1998; Kilicoglu,
1998: Nartallo-Turley, 1998).
º
ſ
t
|
ſ
|
Figure 2. Superimposition of pretreatment and post-treatment cephalometric trac-
ings showing a typical response to facemask therapy.
How much of the correction is due to dental movement and how
much is due to skeletal movement? We found the majority of the cor-
rection to be skeletal with an average mesial movement of the maxillary
molars and incisors of 1.7 mm and no retroclination of the lower incisors.
The mandibular incisors erupted normally, possibly due to the relatively

120
Turley
quick correction that occurs with full-time facemask wear and aggressive
overcorrection. Other studies have shown retrusion of the lower incisor.
Decreased hours and shorter treatment times may contribute to an incline
plane effect that retroclines the lower incisor (Williams, 1997).
In summary, the short-term effects of early Class III correction
with facemask therapy may include downward and forward maxillary
movement, downward and backward mandibular movement, dentoalveo-
lar movement and improvement in the facial profile and smiling esthet-
ics. The longer the treatment time, the greater the amount of maxillary
advancement, the greater the overcorrection, and the lesser the amount
of lower incisor retrusion. Inhibition of mandibular growth has been re-
ported by some authors (Franchi et al., 2004).
THE EFFECTS OF PALATAL EXPANSION
The documented effects of palatal expansion (PE) are numerous
and may include downward and forward maxillary movement, correction
of posterior crossbites, an increase in anterior arch length, and clockwise
mandibular rotation (Haas 1965, 1970a,b; Wertz, 1970, 1977). Few stud-
ies however, have examined the effects of palatal expansion in conjunc-
tion with facemask therapy. Baik (1995) examined this question in 60
Korean children 8 to 13 years of age. Forty-seven of the 60 children were
treated with a palatal expansion appliance and the remaining 13 children
were treated with a labial-lingual appliance. Baik found greater maxillary
protraction with PE (2.0 mm with PE vs. 0.9 mm without PE). -
Vaughn and colleagues (2005) examined the effects of maxillary
protraction therapy with and without PE in a prospective, randomized
clinical trial. Using a block randomization table, the subjects were as-
signed to one of three groups:
Group A: Facemask therapy and palatal expansion (N = 15+6).
Group B: Facemask therapy and a passive palatal expansion ap-
pliance (N = 14+8).
Group C; Observation only for at least 12 months (N = 17) fol-
lowed by random assignment to treatment group A (N = 6) or
B (N = 8).
There were no statistically significant differences between the
expansion and non-expansion groups in any measured variable. There
also were no statistical differences in overall treatment time or in the
121
Facemask Therapy for Class III Malocclusion
time it took to achieve anterior crossbite correction. The results suggest
that, in the absence of objective reasons for expansion such as maxillary
width or space deficiencies, expansion did not aid the correction of Class
III malocclusions with facemask therapy. It is possible that a repetitive
weekly protocol of alternate rapid maxillary expansion and constriction
for the purpose of disarticulating the maxilla may produce a more benefi-
cial effect (Liou and Chen, 2003).
OPTIMAL TREATMENT TIME
Several authors have suggested that there is a narrow window
of opportunity with regards to the timing of facemask therapy (Proffitt,
1993). Is there a best time to treat and what are the differences between
those children treated early versus those treated late?
Baik (1995) examined this question in 47 Korean subjects divided
into three groups: less than 10 years old (N=11), 10 to 12 years old (N=21)
and older than 12 years (N=15). The study concluded that younger chil-
dren were not significantly different from older children in their response
to facemask therapy. Sung and Baik (1998) examined a larger sample
of 129 patients subdivided by age. The results showed that the amount
of skeletal change among the different age groups was not statistically
significant.
Merwin and colleagues (1997) attempted to determine whether
there was a greater orthopedic effect and less dental movement when Class
III malocclusions were treated with maxillary expansion and protraction
before and after the age of eight years in the mixed dentition. They found
a striking similarity in therapeutic response between the younger and older
age grOupS.
Williams and colleagues (1997) examined 28 children in similar
age groups (younger and older than eight years of age) treated with maxil-
lary protraction and expansion. In contrast to the previous studies, they
found a trend toward a slightly greater treatment effect in the older age
group statistically. -
With these conflicting results, Kapust and colleagues (1998)
undertook a study of 63 children divided into three age groups: 4 to 7
years of age (N=15), 7 to 10 years of age (N=32) and 10 to 14 years of
age (N=16). As in previous studies, an extensive cephalometric analysis
was performed, which included traditional measures, an X-Y coordinate
axis, and occlusal plane analysis. They found trends that favored earlier
122
Turley
treatment. The younger children showed twice the amount of maxillary
advancement than did the children in the older age group. Children in the
older age groups did demonstrate significant treatment effects however,
suggesting that there might be a much larger window of opportunity than
previously thought. Similarly, Baccetti, Franchi and colleagues (Baccetti
et al., 1998; Franchi et al., 1998) examined 46 patients divided into early
and late mixed dentition groups and also concluded that facemask/palatal
expansion therapy is more effective in the early mixed dentition than in
the late mixed dentition.
Cha (2003) compared three groups based on Fishman’s skeletal
maturity indicator using hand-wrist radiographs. Eighty-five Class III pa-
tients were divided into pre-pubertal (SM 1–3; N=34), pubertal (SM 4-7;
N=32), and post pubertal (SM 8–11; N=19). Significant changes were
seen in nearly all parameters in all three age groups. Similar maxillary
advancement was observed in the pre-pubertal and pubertal growth peak
groups. The post-pubertal group showed less maxillary advancement but
greater proclination of maxillary incisors.
Most recently, Franchi and colleagues (2004) performed a post-
pubertal assessment of treatment timing for maxillary expansion and pro-
traction followed by fixed appliances. Thirty-three children in the late pri-
mary or early mixed dentition (average age was 7.4 years) were compared
to 17 children in the late mixed dentition (average age was 10.8 years).
The radiographs of 24 untreated Class III patients were used as controls.
The final radiographs were taken at the end of comprehensive fixed ap-
pliance therapy and were post-pubertal (CVM IV or V). They found that
early treatment resulted in greater maxillary advancement (2 mm vs. 0.7
mm) that was maintained long term. Of interest was the finding that later
treatment resulted in a greater inhibition of mandibular growth. In con-
trast to the patients receiving treatment, the untreated Class III patients
showed a worsening of the maxillomandibular relationship by 1 mm per
year, pointing to the importance of starting treatment as early as possible.
GROWTH AFTER FACEMASK THERAPY
But what happens following therapy? Does the maxilla relapse
backward? Does the maxilla continue to grow forward? Does the maxilla
grow like a Class I occlusion or Class III malocclusion? Does the man-
dible rotate/grow forward to the pre-treatment position?
123
Facemask Therapy for Class III Malocclusion
Wisth and co-workers (1987) examined the post-treatment growth
of 22 children treated with a facemask and quad helix appliance. Treat-
ment times varied from 3 to 12 months; four children did not achieve posi-
tive overjet. The patients were followed for 6 to 48 months after treatment
and their growth was compared to that of 40 patients 4 to 9 years of age
who had normal occlusions. The changes in the treated and control groups
were very similar (only two of the 24 measures used showed a statistical
difference).
Chong and colleagues (1996) examined changes following the
use of protraction headgear for early correction of Class III malocclusion.
Sixteen consecutive patients treated with a Delaire facemask with either
a soldered labiolingual wire to the upper molar or a bonded RPE were
compared to 13 Class III controls (average ANB was #2.6°) from the Bur-
lington and Bolton growth centers. The post-treatment observation period
averaged 3.6 years. No differences were found between treated patients
and controls during the post-treatment follow-up period.
In contrast to these studies, Gallagher and colleagues (1998) found
a “relative relapse” of the maxilla in a sample of 22 children treated with
palatal expansion and facemask therapy. During the 17-month post-treat-
ment observation period, maxillary growth was less in the treated subjects
than in the Class I controls, whereas mandibular growth was similar to that
of the controls.
To further investigate post-treatment growth, Macdonald and col-
leagues (1999) examined the records of 24 Class III patients treated with
custom facemasks and banded palatal expanders. Post-treatment growth
of 2.3 years was compared to that of two untreated control samples (24
Class I and 27 Class III subjects). When the annual growth of Class I and
Class III untreated patients was compared, the Class III patients showed
less maxillary growth measured at point A (0.4 mm vs. 0.9 mm) and great-
er forward movement of the mandible (1.7 mm vs. 0.6 mm). After face-
mask therapy, the maxilla grew the same as it did in the untreated Class III
patients, but less than it did in the Class I patients (0.4 mm vs. 1.0 mm).
Mandibular growth was similar in all age groups (1.7 mm/year). Although
growth reverted to a Class III pattern, all patients showed positive overjet
when examined 2.3 years after treatment, which probably was due to the
overcorrection achieved with treatment. The results showed that although
facemask therapy may normalize jaw position, it does not normalize post-
treatment growth. Treated Class III patients retained a Class III growth
pattern characterized by deficient maxillary growth and normal to exces-
sive mandibular growth.
124
Turley
If we know that a Class III growth pattern reasserts itself after
facemask therapy, should we attempt to prevent relapse by retaining the
correction with some type of appliance? Numerous appliances have been
used as orthopedic retainers including the FR-3 of Fränkel (Petit, 1983;
Pangrazio-Kulbersh, 1998), mandibular retractor or Class III activator
(Ngan et al., 1997), removable maxillary stabilizing plate (Westwood et
al., 2003) or a Class III bionator (Turley, 1988; Cozza, 2004). But is this
retention worthwhile if not used long-term? What about the problems of
long-term retention such as compliance and the patient viewing it as a
never ending continuation of active treatment? In lieu of long-term re-
tention, I prefer aggressive overcorrection to provide a number of extra
years of maxillary growth, and then observation of the patient at six-month
intervals to evaluate occlusal stability. Post-treatment observation of a
patient who is not in retention provides valuable information regarding
the patient’s growth pattern, which will be helpful when planning Phase 2
treatment. This approach works well for most patients. A small number
of patients may need a later period of facemask therapy, which often can
be accomplished just prior to or in conjunction with comprehensive fixed
appliance therapy.
LONG TERM EFFECTS
What do we know about the long-term effects of facemask thera-
py? To date the available information is limited and the few “long-term”
studies published include patients who were only 11 or 12 years of age at
the “long-term” evaluation. Some of these studies have attempted to limit
the examination to patients who were post-pubertal based on either the
hand wrist radiograph (Fishman’s skeletal maturity indicator; Cha, 2003)
or the cervical vertebra maturation (CVM) stage on the lateral cephalo-
metric radiograph (Franchi et al., 2004).
To assess post-treatment stability, Ngan and colleagues (1997) ex-
amined 20 patients two and four years post-treatment. At four years post-
treatment, “half the subjects were in the pubertal growth period.” Treat-
ment produced a positive overjet in all patients, and the results remained
stable two years after the removal of appliances. Four years post-treat-
ment, however, “15 of the 20 subjects maintained either a positive overjet
or an end-to-end relationship.” During the latter two years of observation,
growth was comparable to that of the Class III control group. Patients who
reverted to a negative overjet displayed excess mandibular growth that
was not compensated for by proclination of the maxillary incisors.
125
Facemask Therapy for Class III Malocclusion
Westwood and colleagues (2003) examined patients 5% years af.
ter facemask therapy and immediately following fixed appliance therapy.
Patients averaged 14.7 years of age (a range of 11.3 years to 20.7 years).
Growth was compared to Class III controls. Treated Class III patients im-
proved 3.9 mm to 4.5 mm (OJ/molar relation), while the Class III controls
worsened by 3 mm. Correction mainly was due to the mandible (-2.5 mm)
and less due to the maxilla (+1.6 mm). At the final observation, 26 patients
of the 34 total patients (76%) had a positive overjet and three (9%) had an
edge-to-edge incisal relationship. Their findings of “long-term” mandibu-
lar growth restraint are interesting in relation to long-term studies on chin
cup therapy, in which catch up or excess mandibular growth occurred after
cessation of treatment. For example, Mitani and colleagues (1986) found
that there was no significant mean difference between chin-cup patients
and controls at age 14 years. Sugawara and co-workers (1990) concluded
that “the final skeletal profile of each group (age 17) shows no significant
difference between the anteroposterior positions of the mandible when
compared with those of the control sample.”
At what age or maturity level will our evaluations be adequate to
determine clinical success or failure? In females, by the time comprehen-
sive fixed appliance therapy is complete, growth also has slowed, so post-
treatment stability is generally good (Fig. 3). Males, on the other hand,
seem to be disproportionately affected by Class III relapse, so their long-
term post-treatment stability is harder to predict (Fig. 4). Males continue
to grow during and post-Phase II treatment. Often in treatment for longer
periods of time than girls, burn out is a problem for males. They generally
are older than girls when in Phase II treatment, and may not cooperate as
well with regard to elastics or further facemask wear. Before and after pu-
berty, males exhibit more condylar growth, and the direction of mandibu-
lar growth is more horizontal in males than in females (Behrents, 1985).
Are there any anatomical features that would delineate patients
who would respond favorably or poorly to facemask therapy? Ghiz and
co-workers (2005) examined 64 patients treated with a facemask three
years after treatment and found that Class III growing patients with a
forward position of the mandible, small ramal length, large mandibular
length, and obtuse gonial angle are strongly associated with unsatisfactory
treatment outcomes after pubertal growth.
126
Turley
Figure 3. Top. Female patient with a Class III malocclusion treated with
facemask and palatal expansion. Pretreatment at age 10.6 years (left). Post
Phase 1 treatment at age 13.5 years (middle). Post, comprehensive fixed
appliance therapy at age 13 years, 6 months (right). Bottom. Four years
post-treatment at age 22.08 years (left). Six years post-treatment at age
24.3 years (middle). Twelve years post-treatment at age 30.08 years (right).
Long-term stability is excellent.



127
Facemask Therapy for Class III Malocclusion

Turley
Figure 4. Continued. Post-treatment growth resulted in a significant
Class III malocclusion.

129
Facemask Therapy for Class III Malocclusion
Taking all of these factors into account, the long-term stability
of Class III treatment results depends on the amount of overcorrection
originally achieved, the growth discrepancy between the maxilla and the
mandible, and the number of years of disproportionate growth remaining.
SUMMARY
Facemask therapy is an effective method for correcting Class III
malocclusions in growing children. Correction occurs due to a combina-
tion of skeletal and dental changes that improve the maxillomandibular re-
lationship and the soft tissue profile. The earlier the correction is instituted
and the longer the treatment continues, the greater the amount of maxillary
protraction, which appears to be maintained long term.
A Class III growth pattern appears to reassert itself following
facemask therapy characterized by deficient maxillary growth and nor-
mal to excessive mandibular growth. Hence, overcorrection is routinely
advised. The long-term studies of facemask therapy are not really long
term; therefore, continued vigilance with regards to post-treatment growth
is recommended. º
Are the benefits of early facemask therapy worth the burden? The
burdens of early facemask therapy are the economic cost and the time to
treat; the “pain and suffering” aspects are minor. The benefits of early
facemask therapy, however, are many. Clinical experience has shown
that early facemask therapy is one of the easiest orthodontic procedures
to accomplish, both for the orthodontist and the patient. Children treated
early generally are very cooperative, treatment time is short, and the suc-
cess rate is high. Early treatment provides an improved occlusion and
improved dental and facial aesthetics. It also produces greater maxillary
advancement that is maintained long term, and usually allows the patient
to avoid orthognathic surgery. In a small number of patients who experi-
ence a return to a Class III occlusion, additional treatment still can produce
positive changes. For the rare patient who needs orthognathic surgery,
the complexity of treatment usually is less, which contributes to a greater
predictability and stability of the surgical outcome.
REFERENCES
Baccetti T. McGill JS, Franchi L, McNamara JA Jr, Tollaro I. Skeletal
effects of early treatment of Class III malocclusion with maxillary
130
Turley
expansion and facemask therapy. Am J Orthod Dentofacial Orthop
1998; 113:333-343.
Baik HS. Clinical results of maxillary protraction in Korean children. Am
J Orthod Dentofacial Orthop 1995; 108:583–592.
Behrents RG. Growth in the Aging Craniofacial Skeleton. Monograph 17,
Craniofacial Growth Series, Center for Human Growth and Develop-
ment, The University of Michigan, Ann Arbor, 1985.
Cha K. Skeletal changes of maxillary protraction in patients exhibiting
skeletal Class III malocclusion: A comparison of three skeletal matu-
ration groups. Angle Orthod 2003;73:26-35.
Chong YH, Ive C, Artun J. Changes following the use of protraction head-
gear for early correction of Class III malocclusion. Angle Orthod
1996;66:351–362.
Cozza P. Marino A, Mucedero M. An orthopaedic approach to the treat-
ment of Class III malocclusions in the early mixed dentition. Eur J
Orthod 2004:26:191-199.
Deguchi T, Kanomi R, Ashizawa Y, Rosenstein SW. Very early facemask
therapy in Class III children. Angle Orthod 1999;69:349-355.
Franchi L, Baccetti T. McNamara JA Jr. Shape-coordinate analysis of
skeletal changes induced by rapid maxillary expansion and facial
mask therapy. Am J Orthod Dentofacial Orthop 1998; 114:418-426.
Franchi L, Baccetti T, McNamara JA Jr. Postpubertal assessment of treat-
ment timing for maxillary expansion and protraction therapy followed
by fixed appliances. Am J Orthod Dentofacial Orthop 2004;126:555-
568. -
Gallagher W, Miranda F, Buschang PH. Maxillary protraction: Treat-
ment of post-treatment effects. Am J Orthod Dentofacial Orthop
1998; 113:453-462.
Ghiz MA, Ngan P, Gunel E. Cephalometric variables to predict future
success of early orthopedic Class III treatment. Am J Orthod Dento-
facial Orthop 2005;127:301-306.
Haas AJ. The treatment of maxillary deficiency by opening the midpalatal
suture. Angle Orthod 1965:35:200–217.
Haas A.J. Palatal expansion: Just the beginning of dentofacial orthope-
dics. Am J Orthod 1970a:57:219–255.
Haas A.J. Rapid expansion of the maxillary dental arch and nasal cavity by
opening the mid-palatal suture. Angle Orthod 1970b;58:41-66.
Johnston LE Jr. A comparative analysis of Class II treatments. In: Vig
PS, Ribbens KA, eds. Science and Clinical Judgment in Orthodon-
tics. Craniofacial Growth Series, Center for Human Growth and
131
Facemask Therapy for Class III Malocclusion
Development, The University of Michigan, Ann Arbor 1986; 19:103-
148.
Kapust AJ, Sinclair PM, Turley PK. Cephalometric effects of facemask/
expansion therapy in Class III children. A comparison of three age
groups. Am J Orthod Dentofacial Orthop 1998; 113:204-212.
Kilicoglu G, Kirlic Y. Profile changes in patients with Class III maloc-
clusions after Delaire facemask therapy. Am J Orthod Dentofacial
Orthop 1998; 113:453-462.
Liou EJW, Chen PKT. New orthodontics and orthopedic movements on
the premaxillary deformities in patients with bilateral cleft before al-
veolar bone grafting. Ann Coll Surg HK 2003;7:73–82.
Macdonald KE, Kapust AJ, Turley PK. Cephalometric changes after the
correction of Class III malocclusion with maxillary expansion/face-
mask therapy. Am J Orthod Dentofacial Orthop 1999; 116:13-24.
Merwin D, Ngan P. Hagg U, Yiu C, Wei SH. Timing for effective applica-
tion of anteriorly directed orthopedic force to maxilla. Am J Orthod
Dentofacial Orthop 1997; 112:292-299.
Mitani H, Fukuzawa H. Effects of chincap force on the timing and amount
of mandibular growth associated with anterior reversed occlusion
(Class III malocclusion) during puberty. Am J Orthod Dentofacial
Orthop 1986;90:454-463. -
Nartallo-Turley PE, Turley PK. Cephalometric effects of combined palatal
expansion and facemask therapy on Class III malocclusion. Angle
Orthod 1998;68:217-222.
Ngan PW, Hagg U, Yiu C, Wei SH. Treatment response and long-term
dentofacial adaptations to maxillary expansion and protraction. Sem-
in Orthod 1997;3:255-264.
Pangrazio-Kulbersh V. Effects of protraction mechanics on the midface.
Am J. Orthod Dentofacial Orthop 1998;114:484-491.
Petit HP. Adaptation following accelerated facial mask therapy. In: Mc-
Namara JA Jr, Ribbens KA, Howe RP, eds. Clinical Alterations of
the Growing Face. Craniofacial Growth Series, Center for Human
Growth and Development, The University of Michigan, Ann Arbor
1983;14.
Proffit WR. Contemporary Orthodontics 2" ed., St. Louis, Mosby, 1993.
Sugawara J, Asano T, Endo N, Mitani H. Long-term effects of chincap
therapy on skeletal profile in mandibular prognathism. Am J Orthod
1990:90:127-133.
132
Turley
Sung SJ, Baik HS. Assessment of skeletal and dental changes by maxillary
protraction. Am J Orthod Dentofacial Orthop. 1998; 114:492-502.
Turley PK. Orthopedic correction of Class III malocclusion with palatal
expansion and custom protraction headgear. J. Clin Orthod 1988:22:
3 || 4–325.
Vaughn G, Mason B, Moon HB, Turley PK. The effects of maxillary
protraction therapy with or without rapid palatal expansion: A pro-
spective, randomized clinical trial. Am J Orthod Dentofacial Orthop
2005; 128:299-309.
Wertz RA. Skeletal and dental changes accompanying rapid midpalatal
suture opening. Am J Orthod 1970:58:41-66.
Wertz RA. Midpalatal suture opening: A normative study. Am J Orthod
1977;71:367-381.
Westwood PV, McNamara JA Jr, Baccetti T, Franchi L, Sarver DM. Long-
term effects of Class III treatment with rapid maxillary expansion and
facemask therapy followed by fixed appliances. Am J Orthod Dento-
facial Orthop 2003;123:306–320.
Williams MD, Sarver DM, Sadowsky PL, Bradley E. Combined rapid
maxillary expansion and protraction facemask in the treatment of
Class III malocclusions in growing children: A prospective long-term
study. Semin Orthod 1997;3:265-274.
Wisth PJ, Tritapunt A, Rygh P. Boe OE, Nordeval K. The effect of max-
illary protraction on front occlusion and facial morphology. Acta
Odontol Scand 1987;45:227-237.
133
134
EARLY CLASS III TREATMENT:
IS THE BENEFIT WORTH THE BURDEN2
Peter Ngan
The objective of early orthodontic treatment is to create a more favorable
environment for future dentofacial development. Early Class III treatment,
such as correction of anterior crossbite or normalization of the jaw posi-
tions, can prevent progressive and irreversible soft-tissue or bony changes,
reduce the amount of dental compensation to skeletal discrepancy that of
ten is associated with a more severe malocclusion in late adolescence,
prevent abnormal incisal wear, improve lower lip posture and generally
improve the psychosocial well-being of the child (Joondeph, 1993). How-
ever, success in early Class III treatment depends on accurate diagnosis
of the underlying skeletal and dentoalveolar problems and on anticipat-
ing the probable growth changes. Class III malocclusion may be made
up of different combinations of skeletal and dentoalveolar components.
Consideration of the various components is essential to understanding the
underlying causes of the discrepancy which, in turn, is essential to choos-
ing the appropriate treatment.
According to Guyer and colleagues (1986), the most frequent
Class III pattern is a normal maxilla and prognathic mandible. Approxi-
mately 25% of Class III patients in their study had a deficiency in the
maxilla and 57% of patients with either a normal or prognathic mandible
had a deficiency in the maxilla. Protraction facemask therapy has been
successful in the early treatment of Class III patients with maxillary defi-
ciencies (McNamara, 1987; Turley, 1988; Ngan et al., 1992). However,
25% to 33% of the treated patients reverted to an anterior crossbite when
they underwent their pubertal growth spurt (Ngan et al., 1997; Hägg et
al., 2003; Westwood et al., 2003). These studies also found very few
morphological differences at the start of treatment between patients who
were successfully camouflaged by orthodontic treatment and those who
eventually needed surgical intervention. Turpin (1981) suggests that early
treatment will be more successful in patients who have a convergent facial
type with an anteroposterior functional shift, symmetrical condyle growth,
mild skeletal disharmony, and who are pre-pubescent. There is a need for
a diagnostic scheme that will differentiate patients whose malocclusions
can be successfully camouflaged with orthodontic treatment from those
who eventually will need surgical intervention.
135
Early Class III Treatment
Clinicians often are reluctant to engage in early Orthopedic treat-
ment in Class III patients because it is difficult to predict excessive mandib-
ular growth (Ngan, 2002). Thompson (1994) commented on the individu-
ality of the facial skeleton and concluded that growth cannot be predicted
because patient growth spurts can occur anytime before, during or after
orthodontic treatment. Patients who receive early orthodontic or orthope-
dic treatment can end up needing surgery at the end of the growth period.
The ability to predict mandibular growth early in life can help clinicians
decide whether early orthodontic treatment or post-pubertal surgery is the
most efficacious treatment. The objective of this chapter is to outline the
strategy for differentiating pseudo and true Class III malocclusions and to
discuss tools that can be used to predict future growth changes.
DIF FERENTIAL DIAGNOSIS OF
CLASS III MALOCCLUSION
Familial History of Class III Malocclusion
The few studies of human inheritance and its role in the Class
III malocclusion support the belief that the growth and size of the man-
dible are affected by heredity. The most famous study of inheritance, as
described by McGuigan (1966), is that of the Hapsburg family. Of the
40 family members for whom records were available, 33 had prognathic
mandibles. Litton studied the families of 51 individuals who had Class
III anomalies and concluded that Class III characteristics were related to
genetic inheritance for both offspring and siblings (Litton et al., 1970). It
is recommended, therefore, that a patient’s record should include informa-
tion on family members who have Class III malocclusions. In addition, a
lateral cephalogram should be taken of such family members, if possible,
to determine whether their Class III malocclusions have a maxillary defi-
ciency component, a mandibular excess component or a combination of
both.
Dental Assessment
In the mixed dentition, patients with Class III malocclusions may
present with a flush terminal plane or mesial terminal plane together with
retroclined mandibular incisors. Patients with a mesial step of 2 mm or
more can have a 19% chance of developing a Class III permanent molar
relationship (Bishara et al., 1988). In the permanent dentition, a full-step
Class III molar relationship usually has a worse prognosis than a half-step
136
Ngan
Class III molar relationship. The Class III molar relationship also can
be accompanied by a negative overjet. If a positive overjet or end-to-
end incisal relationship is present together with retroclined mandibular
incisors, one should suspect a compensated Class III malocclusion. If a
negative overjet is present, a functional assessment should be undertaken.
Functional Assessment
To determine whether a centric relation/centric occlusion (CO/
CR) discrepancy exists, the relationship of the maxilla to the mandible
must be evaluated. Anterior positioning of the mandible may result from
abnormal tooth contact that forces the mandible forward. Patients who
present with a forward shift of the mandible on closure may have a Class
I skeletal pattern, a normal facial profile, and Class I molar relationship in
centric relation. This condition is referred to as a pseudo Class III mal-
occlusion. Elimination of the CO/CR discrepancy or correction of the
anterior crossbite should reveal whether the condition is a simple Class I
malocclusion or a true Class III malocclusion.
Profile Assessment
The lateral profile of patients in repose and smiling can provide
valuable information in diagnosing a Class III malocclusion. A straight
or concave facial profile and mandibular prognathism in a young patient
indicate that there is an underlying skeletal malocclusion. When examin-
ing the lateral profile of a patient in a repose position, an imaginary line is
dropped down from the middle of the forehead to evaluate the midface and
the chin position. In young children, the anatomical landmark ‘subnasale’
usually is 1 to 2 mm ahead of this imaginary line, and the chin position
usually is behind this line. When examining the lateral profile of a young,
Smiling patient, the maxillary central incisors should be close to or touch-
ing the imaginary line from the forehead. The presence of a midface defi-
ciency would be indicated if the incisors are behind the imaginary line and
the patient’s tissue contour is straight or concave.
Cephalometric Assessment
When compared to subjects with normal skeletal patterns, Class
III patients have a shorter anterior cranial base, a more obtuse gonial angle,
the glenoid fossa is positioned further forward, the maxillary incisors are
more proclined and the mandibular incisors are more retroclined (Ngan et
al., 1996). However, there are very few cephalometric studies that compare
pseudo Class III patients with true skeletal Class III patients. One study
137
Early Class III Treatment
comparing the cephalometric measurements of pseudo Class III patients
and true skeletal Class III patients with Class I patients revealed that most
of the measurements for pseudo Class III patients fell between those for
true skeletal Class III and Class I individuals (Li and Lin, 1987). The only
measurement with diagnostic value for differentiating pseudo Class III pa-
tients from true skeletal Class III patients was the gonial angle. The aver-
age gonial angle of pseudo Class III patients was lower (120°) than that for
both Class I (122°) and true skeletal Class III (124°) patients. Discriminant
analysis found the WITS analysis to be most decisive in distinguishing the
need for camouflage treatment from the need for surgical treatment (Stell-
zig-Eisenhauer et al., 2002). The average WITS for the non-surgery group
was -4.6 mm + 1.7 mm and -12.2 mm + 4.3 mm for the surgery group.
Growth Prediction of Class III Malocclusion
No two persons are identical, and each person has a unique fa-
cial growth pattern. Thompson (1994) commented on the uniqueness of
individual facial growth and concluded that growth cannot be predicted
because maximum growth can occur before, during or after orthodontic
treatment. However, poor skeletal patterns tend to stay the same or be-
come worse, and excellent skeletal patterns tend to stay the same or get
better, and grow over a longer time span. This observation was corrob-
orated by studies that followed patients into their pubertal growth spurt
(Ngan et al., 1997; Hägg et al., 2003; Westwood et al., 2003). Patients
with excess mandibular growth continued to get worse during the pubertal
growth period and patients with mild to moderate Class III growth were
eventually able to camouflage it with orthodontic treatment.
Several investigations have attempted to predict the progres-
sion of Class III malocclusion. Björk and Skieller (1983) used a single
cephalogram to identify seven structural signs of extreme mandibular
growth rotation occurring during growth. The seven signs are related
to the inclination of the condylar head, the curvature of the mandibular
canal, the shape of the lower border of the mandible, the width of the
symphysis, the interincisal angle, the intermolar angle, and the anterior
lower face height. Discriminant analysis of long-term results of early
treatment identified several variables that had predictive value. Franchi
and colleagues (1997) found that the inclination of the condylar head,
the maxillomandibular vertical relationship together with the width of
the mandibular arch could predict success or failure for early Class III
treatment. Ghiz and Ngan (2001) found that the position of the mandi-
138
Ngan
ble, the ramal length, the corpus length, and the gonial angle could be
used to predict successful outcomes with 95% accuracy. However, this
method can predict unsuccessful outcomes with only 70% accuracy.
The use of serial radiographs and a growth treatment response
vector (GTRV) analysis to predict excessive mandibular growth is more
specific to the individual’s growth rate and direction. Using this method,
the horizontal growth changes of the maxilla and the mandible, which
occur during the time period between when the pre-treatment radiograph
is taken and when the follow-up radiograph is taken, can be determined
by locating the Point A and Point B on the first radiograph. The occlusal
plane (O) is constructed by using the mesial buccal cusp of the maxillary
molars and the incisal tip of the maxillary incisors as landmarks. The
lines AO and BO then are constructed by connecting Point A and Point B
perpendicular to the occlusal plane.
The first tracing is superimposed on the follow-up radiograph us–
ing the stable landmarks on the midsagittal cranial structure. Point A and
Point B on the follow-up radiograph are located and the lines AO and
BO the are constructed by connecting Points A and B of the follow-up
radiograph to the occlusal plane of the first tracing. The distance between
Point A of the two tracings along the occlusal plane represents the growth
changes of the maxilla, and the distance on the occlusal plane of Point B
represents the growth changes of the mandible. The GTRV ratio is calcu-
lated by using the following formula:
GTRV = Horizontal growth changes of the maxilla
Horizontal growth changes of the mandible
In a study in which patients were treated with a protraction face-
mask, the patients were divided into two groups of 20 each (Ngan and
Wei, 2004). The first group was comprised of patients who were treated
Successfully with the protraction headgear based on the follow-up radio-
graph and the second group was comprised of patients who were treated
unsuccessfully with the protraction headgear. Significant differences were
found between the GTRV ratios for the two treatment groups (p < 0.05).
The mean GTRV ratio for the successful group was 0.49 + 0.14 with a
range of 0.33 to 0.88. The mean GTRV ratio for the unsuccessful group
was 0.22 + 0.10 with a range of 0.06 to 0.38.
CASE REPORT
A typical case is illustrated here to demonstrate the use of GTRV
ratio to predict excessive mandibular growth. A 10-year-old Chinese girl
139
Early Class III Treatment
presented with a skeletal Class III malocclusion and a flat facial profile
(Fig. 1). Clinical examination revealed an anterior crossbite with Class III
molar and canine relationships. The inclination of her maxillary incisors
was within the normal range, but her mandibular incisors were retroclined.
The cephalometric radiograph revealed a deficient maxilla and a normal
mandible with a WITS appraisal of -3.0 mm (Fig. 2).
Figure 1. A 10-year-old Chinese girl with a flat facial profile, an anterior
crossbite and crowding of the upper and lower dentitions.
The patient was treated with a maxillary expansion appliance to:
gether with a protraction facemask for eight months. A positive overjet
was established after eight months of treatment (Fig. 3). Superimposition
of the pre- and post-treatment radiographs revealed primarily a downward
growth of the maxilla (Fig. 4). The patient was followed for three years
until the age of 13 years at which time a decision regarding treatment
choices was necessary. Should the patient begin Phase II orthodontic
treatment to camouflage the malocclusion or should she wait until growth
was completed and undergo surgery to correct the jaw discrepancy? Fig-
ure 5 shows the profile and intraoral photos of the patient at the age of 13
yearS.



140
Ngan
Figure 2. Cephalometric radiograph showing a skeletal Class
III malocclusion with increased lower face height.
Figure 3. Patient was treated with
maxillary expansion and protrac-
tion facemask for eight months.
Note the positive overjet after
treatment.



141
Early Class III Treatment
Figure 4. Cephalometric radiograph taken after facemask treatment
(left). Superimposition showing that correction of the anterior crossbite
was due primarily to downward growth of the maxilla and downward
and backward rotation of the mandible (right).
Figure 5. Two-year post-treatment follow-up record showing changes in
the soft tissue profile and molar relationship.



142
Ngan
A GTRV analysis was performed using the pre-treatment and fol-
low-up radiographs (Fig. 6). The ratio of 0.8 (norm = 0.77) indicated
parallel growth of the maxilla and mandible, and the vectors indicated a
forward and downward growth of the maxilla and mandible after face-
mask treatment (Fig. 7). The decision was made to camouflage the mal-
Occlusion with orthodontic treatment. Figure 8 shows the post-treatment
facial and intraoral photos of the 15-year-old patient who has a satisfactory
Occlusion at the end of treatment.
Figure 6. Cephalometric radiograph taken immediately prior
to the beginning of Phase II, comprehensive orthodontic treat-
ment to camouflage the skeletal malocclusion.

143
Early Class III Treatment
Figure 7. Post-treatment records show improvement in occlusion and
facial profile after phase II orthodontic treatment.
.
Figure 8. Superimposition showing the growth changes
during the two phases of treatment.



144
Ngan
CONCLUSIONS
1. Success in early Class III treatment depends on accurate diag-
nosis of the underlying skeletal and dentoalveolar problems and on antici-
pating the probable growth changes.
2. Early Class III treatment allows for favorable skeletal re-
sponse with facemask or chin-cup treatment and improvement in facial
profile and self-esteem. In addition, early Class III treatment provides the
clinician with an opportunity to observe individual growth rates and direc-
tion.
3. The use of serial radiographs and GTRV analysis helps the
clinician to decide whether to begin phase II orthodontic treatment during
the pubertal growth period or to wait until growth is completed and inter-
Vene Surgically.
REFERENCES
Bishara SE, Hoppens BJ, Jakobsen Jr, Kohout F.J. Changes in the molar
relationship between the primary and permanent dentitions: A longitu-
dinal study. Am J Orthod Dentofacial Orthop 1988;93:19-28.
Björk A, Skieller V. Normal and abnormal growth of the mandible. A
synthesis of longitudinal cephalometric implant studies over a period
of 25 years. Eur J Orthod 1983;5:1-46.
Franchi L, Baccetti T, Tollaro L. Predictive variables for the outcome of
early functional treatment of Class III malocclusion. Am J Orthod
Dentofacial Orthop 1997;112:80-86.
Ghiz M, Ngan P. Cephalometric variables to predict future success of ear-
ly orthopedic Class III treatment. Am J Orthod Dentofacial Orthop
2005;127:301-306.
Guyer EC, Ellis EE, McNamara JA Jr, Behrents RG. Components of
Class III malocclusion in juveniles and adolescents. Angle Orthod
1986:56:7-30.
Hägg U, Tse A, Bendeus M, Rabie AB. Long-term follow-up of early
treatment with reverse headgear. Eur J Orthod 2003:25:95-102.
Joondeph DR. Early orthodontic treatment. Am J Orthod Dentofacial Or-
thop 1993;104:199-200
Li KW, Lin JJ. Comparison between pseudo and true Class III malocclu-
sion by Veterans General Hospital cephalometric analysis. Clin Dent
1987;7:69-78.
Litton SF, Ackermann LV, Isaacson RJ, Shapiro BL. A genetic study of
Class III malocclusion. Am J Orthod 1970:58:565-577.
145
Early Class III Treatment
McGuigan DG. The Hapsburgs. London, WH Allen, 1966.
McNamara JA Jr. An orthopedic approach to the treatment of Class III
malocclusion in young patients. J Clin Orthod 1987:21:598-608.
Ngan P. Biomechanics of maxillary expansion and protraction in Class III
patients. Am J Orthod Dentofacial Orthop 121:582-583;2002.
Ngan P, Hägg U, Yiu C, Merwin D, Wei SHY, Dentoskeletal and soft tis-
Sue profile changes in response to protraction headgear treatment. Am
J Orthod Dentofacial Orthop 1996; 109:38-49.
Ngan P, Hägg U, Yiu C, Wei SHY, Treatment response and long-term
dentofacial adaptations to maxillary expansion and protraction. Sem-
in Orthod 1997;3:255-264.
Ngan P. Wei SHY. Early treatment of Class III patients to improve facial
aesthetics and predict future growth. Hong Kong Dent J 2004;1:24-
30.
Ngan P. Wei SHY, Hägg U, Yiu CK, Merwin D, Stickel B. Effect of
protraction headgear on Class III malocclusion. Quintessence Int
1992:23:197–207.
Stellzig–Eisenhauer A, Lux CJ, Schuster G. Treatment decision in adult
patients with Class III malocclusion: Orthodontic therapy or orthogna-
thic surgery? Am J Orthod Dentofacial Orthop 2002;122:27-38.
Thompson JR. The individuality of facial skeletal growth. Part 2 Am J
Orthod Dentofacial Orthop 1994; 105:224-240.
Turley PK. Orthopedic correction of Class III malocclusion with palatal
expansion and custom protraction headgear. J Clin Orthod 1988:22:
314–325.
Turpin DL. Early Class III treatment. Unpublished thesis presented at 81*
session, Am Assoc Orthod, San Francisco, 1981.
Westwood PV. McNamara JA Jr, Baccetti T, Franchi L, Sarver DM. Long-
term effects of Class III treatment with rapid maxillary expansion and
facemask therapy followed by fixed appliances. Am J Orthod Dento-
facial Orthop 2003;123:266-278.
146
SHORT-TERMAND LONG-TERM STABILITY OF
CHANGES IN THE TRANSVERSE DIMENSION:
IS EARLY EXPANSION WORTH THE EFFORT2
James A. McNamara, Jr.
This chapter is an overview of the last 30 plus years of research and clinical
practice that has focused, in part, on short-term and long-term adaptations
to changes in the transverse dimension in growing patients. The material
in this chapter concerning our ongoing prospective and retrospective clini-
cal investigations has been presented in both oral and written form over
the years, most recently at the 2006 Moyers Symposium. Other such pre-
sentations include those at the international symposia on early orthodontic
treatment sponsored by the American Association of Orthodontists in 2002
in Phoenix and 2005 in Las Vegas (McNamara, 2002, 2006).
The underlying question for consideration is whether the trans-
verse dimensions of the dental arches can be widened in such a way as to
remain stable over the long term. Most orthodontists cite the correction
of crossbite as the primary reason for altering the transverse dimension
by way of rapid maxillary expansion (RME). A less obvious but more
common orthodontic problem, the etiology of which is related, in part, to
imbalances in the transverse dimension, is a discrepancy between tooth
size and arch size. The most frequently observed type of malocclusion in
routine orthodontic practice is dental crowding, an underlying imbalance
between tooth size and available arch perimeter. This relationship also
may be expressed clinically as protrusion and flaring of the teeth relative
to underlying basal bone. Proposed solutions to this imbalance include
extraction, interproximal reduction, and orthodontic and orthopedic ex-
pansion of the dental arches.
PREVIOUS LONG-TERM CLINICAL STUDIES
It has been known for nearly 150 years that the maxilla can be
expanded (Angell, 1860). Even though Andrew Haas made RME routine
in many orthodontic practices beginning in the 1960s (Haas, 1961, 1965),
it is surprising how few of the long-term RME studies are sound meth-
odologically when judged by today’s standards, and most of these latter
studies have been published only during the last decade.
147
Long-term Transverse Adaptations
Many of the earlier investigations are based solely on case reports.
For example, Haas (1980) presented long-term data from 10 subjects. He
reported that immediately after expansion, the average initial increase was
9 mm in apical base width and 4.5 mm in nasal cavity width. None of his
subjects were said to have undergone a loss in either dimension at the time
of re-evaluation 6 to 14 years post-retention. A slight loss of maxillary
dental arch width was noted in two subjects following retention, and a
slight increase in dental arch width was found in two other subjects.
Stockfish (1969), in one of the most extensive investigations of
RME (N=150), presented little in the way of statistical analysis, save for a
summary graph showing percentage relapse related to initial arch expan-
sion. In contrast to the Stockfish study, the remaining long-term investi-
gations of RME are compromised by small samples. Haas (1961) exam-
ined only 10 patients, Krebs (1964) and Linder-Aronson (1979) examined
23 patients, Timms (1968) looked at 19 patients, and Wertz and Dreskin
(1977) analyzed 56 patients of whom only 27 were available for follow-
up. Indeed, in the 1976 study of Timms (1976), 81 of 100 patients were
excluded due to the availability of insufficient records.
Herberger (1987) conducted one of the few long-term RME stud-
ies involving a relatively large sample of patients. He evaluated 55 RME
subjects with starting ages of 9 to 14 years who were treated with a Haas-
design palatal expander. Nine years after RME, the residual dental ex-
pansion averaged 90% of that gained initially. The maxillary first molars
uprighted, on average, almost 12° during the retention period to approxi-
mately pre-treatment values.
Moussa and co-workers (1995) also investigated the long-term
stability of RME treatment and edgewise therapy. Their sample consist-
ed of dental casts from 55 patients who had been out of retention for 8
to 10 years and who had a mean age of 30 years. Measurements were
made directly on dental casts that were obtained pretreatment, imme-
diately post-treatment, and post-retention for each patient. Differences
between post-treatment and post-retention measurements were statisti-
cally significant for all dimensions except lower intermolar width; only
for lower and upper arch lengths and perimeters, however, were the dif-
ferences greater than 2.0 mm. Treatment with the rapid maxillary ex-
pander produced good stability for upper intercanine width, upper and
lower intermolar widths and incisor irregularity; however, they observed
148
McNamara
poorer stability in lower intercanine width, mandibular arch length and
arch perimeter.
The studies of Herberger (1987) and Moussa and co-workers
(1995) are representative of most clinical studies of the long-term effects
of RME in that there was no untreated control group. This is important
because in the absence of treatment, most dental arch dimensions typically
decrease during the late teen and early adult years. For example, it is a
common clinical observation that lower incisor crowding increases with
age in both males and females, regardless of whether or not orthodon-
tic treatment was undertaken (Shapiro, 1974; Bishara et al., 1989; Little,
1990, 1994; Carter and McNamara, 1998). Such normally occurring dec-
rements in arch dimensions must be taken into consideration in any analy-
sis of the long-term changes in the dental arches of patients treated with
RME. In fact, with the exception of the cephalometric study of Wertz
and Dreskin (1977), no previous investigation has evaluated the change in
arch dimensions relative to serial dental casts from a matched, untreated
population. The long-term effect of RME on the dental arches must be
evaluated in light of the “developmental crowding” that often occurs in
untreated individuals in the late teen and early adult years.
PREVIOUS MICHIGAN EXPANSION STUDIES
As part of an ongoing clinical trial, our research group has been
gathering data on all patients undergoing rapid maxillary expansion in the
early mixed dentition in the private faculty group practice of the author in
Ann Arbor (N=1,100+). Subsamples of this patient population have been
used previously in a variety of clinical studies. In addition, we have been
assembling longitudinal dental cast and cephalometric records on samples
treated elsewhere in private orthodontic practices as well as in the Gradu-
ate Orthodontic Clinic at the University of Michigan.
The control groups for the studies described in this chapter were
derived from the University of Michigan Growth Study (UMGS; Riolo et
al., 1974; Moyers et al., 1976) and in the Netherlands from the University
of Groningen Growth Study (Dibbets and Nolte, 2002). In all investiga-
tions involving the analysis of dental casts, arch widths, arch depth, arch
perimeter, and molar angulation were assessed in all examined individuals
at all observation times. The cephalometric controls were taken from the
UMGS.
149
Long-term Transverse Adaptations
ADOLESCENT DENTITION
Dental Cast Analysis
The first RME sample studied in detail was obtained from the pri-
vate practice of Drs. Robert and Thomas Herberger of Elyria, Ohio. An
attempt was made to recall all patients treated during a specific time inter-
val (1972–1985) who had undergone RME followed by standard edgewise
orthodontics and who were five or more years post-treatment. Treatment
outcome was not a selection criterion. Longitudinal dental casts from 112
patients and 41 untreated subjects were analyzed using a digital imag-
ing system (Brust, 1992; Brust and McNamara, 1995). A Haas-type rapid
maxillary expansion appliance (Fig. 1) was used; treatment followed a
standardized protocol (Haas, 1961). Three sets of serial dental casts were
available for all subjects. The casts were taken pretreatment (T,), after
expansion and fixed appliance therapy (T,), and at the end of a long-term
observation period (T,). The mean duration of the T-T, and T-T, periods
for the treatment group was 3 years and 2 months = 5 months, and 6 years
and 1 month +1 year and 2 months respectively, with the last observation
interval occurring at 20 years of age.
Figure 1. The Haas-type expander, which is used primarily for
patients who have attained their permanent dentitions.

150
McNamara
RME treatment followed by fixed appliance therapy produced sig-
nificantly favorable long-term changes in almost all of the maxillary and
mandibular arch measurements. When compared to controls, the treated
subjects evidenced a net gain of 6.0 mm in the maxillary arch perimeter and
a net gain of 4.5 mm in the mandibular arch perimeter long-term (20 years
of age or older). The duration of retention with a fixed lower appliance in
the post-treatment period did not appear to affect the long-term outcomes
of the treatment protocol significantly (McNamara et al., 2003).
Fenderson and co-workers (2004) also conducted a study of the
long-term treatment effects of RME treatment. The purpose of their study
was to compare the long-term stability of maxillary expansion achieved
by either RME or facebow expansion followed by fixed edgewise treat-
ment. The original parent sample included 154 non-extraction patients
who started their orthodontic treatment between the years 1973 and 1981
in the private orthodontic practice of Dr. Charles Veith of Dover, Dela-
ware. The exclusion criteria subsequently reduced the number of patients
to 61 in the cervical pull facebow group (CFB) and 41 in the cervical pull
facebow/rapid maxillary expansion group (CFB/RME). All subjects were
in the late-mixed to early permanent dentition stage at the start of treat-
ment. The maxillary and mandibular dental casts of these 102 patients
were measured with the aid of our digital imaging system at the start of
treatment (T), at the end of active treatment (T,), at the end of retention
(T,), and during a post-retention follow up visit (T,).
The results of the study by Fenderson and co-workers (2004) in-
dicated that the CFB/RME protocol produced a greater increase in max-
illary arch width (6.1 mm) than did the CFB protocol alone (4.0 mm).
The CFB/RME protocol also provided more net maxillary arch perim-
eter increase than did expansion with an inner bow of a cervical face-
bow. The CFB/RME group had 3.0 mm more arch perimeter 10 years
post-treatment than did the CFB group. The stability of expansion
achieved with an inner bow of a facebow is equal to that achieved with
a Haas-type rapid maxillary expansion appliance. Both expansion pro-
tocols retained 90% (5.5 mm for the CFB/RME group and 3.6 mm for
the CFB group) of the initial intermolar expansion 15 years post-expan-
sion therapy. Maxillary expansion by either method, however, had only
a modest effect on mandibular arch perimeter. Neither method produced
a net post-retention increase. Mandibular arch perimeters, however, were
greater than they would have been had these patients not been treat-
ed. Thus, both methods of maxillary expansion evaluated in this study
151
Long-term Transverse Adaptations
produced long-term effects on both the maxillary and mandibular denti-
tions, with the expansion produced appearing to be orthopedic rather than
orthodontic in nature.
Cephalometric Analysis
The Herberger sample was analyzed further in two cephalomet-
ric studies, one using lateral cephalograms and the second posteroanterior
headfilms. The lateral cephalometric study (Chang et al., 1997) of the
long-term skeletal and dentoalveolar adaptations and side-effects of a sub-
group of patients treated with the same protocol revealed that that RME
therapy used did not have a significant long-term effect on either the verti-
cal or the anteroposterior skeletal dimensions of the face when compared
to a matched group of patients treated with fixed appliances alone or to
untreated controls. For example, there was no opening of the mandibular
plane angle and no forward or backward movement of Point A over the
long term.
The posteroanterior cephalometric study on a subgroup of the
same patient sample (Cameron et al., 2002) demonstrated long-term sta-
bility of the skeletal correction in the transverse dimension. The aim of
this study was to investigate long-term effects induced by rapid maxillary
expansion in a sample of 42 patients compared to normal growth changes
in a sample of 20 subjects. Posteroanterior cephalograms were analyzed
for each subject in both the study and control groups at T (pre-treatment)
and at T. (long-term observation). The mean age at T was 11 years and 10
months for both the treated and the control groups. The mean chronologi-
cal ages at T, also were comparable (20 years, 6 months for the treated
group and 17 years, 8 months for the control group). The study included
transverse measurements on dentoalveolar structures, maxillary and man-
dibular bony bases, and other craniofacial regions (nasal, zygomatic, or-
bital, and cranial). The results of the Cameron study indicated that RME
therapy appears to be an effective procedure to increase transverse facial
dimensions in the long term, both at the skeletal and dentoalveolar level.
Significant pretreatment deficiencies in maxillary width, maxillary incisor
apex width, and maxillary first molar width were corrected with therapy at
a mean age of about 20 years.
152
McNamara
EARLY MIXED DENTITION
Dental Cast Analysis
The first preliminary study of patients in the longitudinal sample
reported the effects seen in 162 subjects who had undergone rapid maxil-
lary expansion with an acrylic splint expander (Fig. 2) during the mixed
dentition before the eruption of the maxillary premolars and canines (Spill-
ane, 1990; Spillane and McNamara, 1995). The expansion procedure was
followed by a retention protocol that included wearing a removable palatal
plate (Fig. 3) for at least one year, after which time a maintenance plate
typically was worn on a full-time or part-time basis for at least one addi-
tional year. The average increase in intermolar width was about 6 mm.
º
º
-
Figure 2. The acrylic splint expander used in mixed dentition
patients. The appliance is activated once per day until the de-
sired amount of expansion is reached.


153
Long-term Transverse Adaptations
Figure 3. A removable maintenance plate that is worn full-
time for at least a year following removal of the maxillary
expansion appliance to sustain the achieved result.
During the post-retention period, the expanded maxillary dental
arches were stable. For example, more than 90% of the original expansion
was maintained after the first year, and more than 80% remained at the end
of the observation period, which was 2.4 year post-expansion. At this time
and before the initiation of the final phase of comprehensive orthodontic
treatment, Phase II records were taken. Similar findings were observed
during the transition from the primary molars to the premolars and from
the primary canines to the permanent canines.
Brust (Brust, 1992; Brust and McNamara, 1995) studied a larger
sample that included 376 consecutively treated mixed-dentition patients
who had undergone rapid maxillary expansion singly or in combination
with prior lower expansion with a Schwarz appliance (Fig. 4). After ap-
plying multiple inclusion criteria, the remaining subsample of 146 patients
who had received RME treatment and 36 patients who had received RME
treatment following Schwarz appliance therapy was studied. Serial dental
casts of 50 untreated controls from the University of Michigan Growth
Study were used as a matched control group. Longitudinal changes were
analyzed with a digital imaging system specifically tailored to the analysis
of dental casts. Changes in arch width, arch perimeter, and molar an:
gulation were evaluated pre-expansion, immediately post-expansion, and
when Phase II records were taken.

154
McNamara
Figure 4. The removable lower Schwarz expansion appliance
used in patients in the mixed dentition. The appliance is acti-
Vated once per week until the desired amount of expansion is
reached.
At the end of the observation period, residual expansion was
greatest in the RME/Schwarz group (7.2 mm); it was only 4.9 mm in the
RME group and 0.9 mm in the control group. This increase in width was
mirrored by an increase in arch perimeter. Maxillary arch perimeter also
increased following rapid maxillary expansion, with greater increases ob-
served in mandibular expansion when a Schwarz appliance was used prior
10 RME. Indeed, clinically relevant increases in mandibular arch perimeter
Wºre noted only in the RME/Schwarz group. When the net arch length de-
**ase in the control group was compared to the net arch length increase in
the RME/Schwarz group, the net gain was about 3.3 mm. At first glance,
this amount may seem like a small increase, but the gain is fully half that
Seen with premolar extraction after anchorage loss from space closure is
taken into consideration (Zablocki et al., 2007).
Cephalometric Analysis
Thus far, the primary focus of data analysis from this sample has
been the development and implementation of the digital imaging system

155
Long-term Transverse Adaptations
for dental cast analysis. In addition, Wendling and co-workers (2005) con-
ducted a study of the short-term changes produced by an acrylic splint
expander composed of a wire framework and 3 mm thick splint Biocryl"M
during an initial six-month treatment interval. Pre- and post-treatment
cephalograms of 25 patients treated solely with the acrylic splint expander
and 19 patients treated with the removable lower Schwarz appliance fol-
lowed by RME have been analyzed. The results of this analysis indicate
that the acrylic splint expander appears to prevent the bite opening effect
of RME noted in previous short-term studies (Haas, 1961, 1965; Davis
and Kronman, 1969; Wertz and Dreskin, 1977), presumably due to the
posterior bite block effect produced by the acrylic covering of the maxil-
lary posterior teeth. Some intrusion of the maxillary buccal segments was
noted as well. The anteroposterior position of the maxilla and mandible
remained unaffected by the expansion procedures and an opening of the
bite was not observed.
No significant between-group differences were observed with
regard to Vertical and horizontal changes during treatment. Further, the
use of the Schwarz appliance to widen the mandibular dental arch did not
produce proclination of the lower incisors, as might be expected with a re-
movable expansion appliance. Thus, the lower incisors were not proclined
by the Schwarz appliance treatment, a type of movement that has been
shown to be unstable over the long term (Little, 1990).
RECENT MICHIGAN CLINICAL STUDIES
RME Alone
We have completed two prospective clinical investigations of
the long-term effects of the expansion protocol used in mixed dentition
patients that clarify the long-term treatment outcomes. The first study
(Geran et al., 2006) examined the treatment effects of the acrylic splint
RME appliance (Fig. 2) when used alone in the early mixed dentition,
i.e., no prior lower expansion. Dental casts of 51 consecutively treated
patients were compared to those of 41 untreated controls at three different
intervals: pre-treatment (T), after expansion and fixed appliance therapy
(T,), and at the end of the long-term observation period (T,). The mean
ages for the treatment group were 8 years and 10 months at Tº. 13 years
and 10 months at T, and 19 years and 9 months at T.
Treatment with an acrylic splint expander followed by fixed ap-
pliance therapy produced significantly favorable long-term changes
in almost all of the maxillary and mandibular arch measurements. The
156
McNamara
amount of change in both the maxillary and mandibular intermolar and
intercanine widths fully corrected the initial discrepancies. Approximately
4 mm of long-term relative increase in maxillary arch perimeter and 2.5
mm additional maintenance of mandibular arch perimeter were observed
in patients when compared to untreated subjects. It should be remembered
that these patients were judged not to need active expansion of the lower
dental arch at the beginning of treatment. These results suggest that this
treatment protocol is effective and that the results are stable for the treat-
ment of constricted maxillary arches, while it also is able to relieve modest
deficiencies in arch perimeter.
Schwarz/RME Combination
In about 40% of patients undergoing early orthopedic expansion,
we choose to expand the lower arch orthodontically prior to RME be-
cause of anterior crowding and/or lingually-inclined lower posterior teeth
(RME-Sz group). The lower Schwarz appliance (Fig. 4) was activated
once per week in an effort to create a modest amount of arch length anteri-
orly and to upright the lower posterior dentition prior to RME (McNamara
and Brudon, 2001).
The final study in this series to date (O’Grady et al., 2006) consid-
ered two groups of patients, one treated with RME alone and one treated
with the Schwarz/RME sequence. Both groups were matched longitu-
dinally to untreated controls. The dental casts of 27 RME-only patients
were compared with those of 23 RME-Sz patients and with those of 16
untreated controls with constricted maxillary arches. The casts were taken
at four different intervals: pre-treatment (T,), after expansion/before fixed
appliance therapy (T,), after fixed appliance therapy (T,), and at the end
of the long-term observation period (T,). The mean ages for the treated
groups were approximately 9 years at Tº. 12 years at T., 14 years at T.,
and 20 years at T. .
Treatment with an acrylic splint RME alone and in combination
with a lower Schwarz appliance followed by fixed appliance therapy pro-
duced significant long-term increases in maxillary arch widths relative to
controls. The use of the lower Schwarz expander prior to RME led to
significantly more favorable results when compared to those produced
by the RME-only protocol: there was a significantly greater increase in
the transverse width of the lower arch and the mandibular arch perimeter
as well as an uprighting of the lower posterior teeth buccally, which al-
lowed for an amount of maxillary expansion that was clinically effective
in the correction of moderate tooth-size/arch-size discrepancies. During
157
Long-term Transverse Adaptations
the overall observation interval, the significant increases in maxillary and
mandibular arch perimeters in the RME-Sz group were 3.8 mm and 3.7
mm respectively when compared with the matched control group. The
RME-only protocol produced modest long-term increases in maxillary
arch perimeter (2.6 mm), but the average long-term increase in mandibu-
lar arch perimeter (2.0 mm) in the RME-only group was not statistically
significant.
SUMMARY AND CONCLUSIONS
The results of clinical studies conducted by our group and by oth-
ers indicate that the maxillary complex is responsive to orthopedic ex-
pansion over both the short-term the long-term. It appears that the key
to successful widening of the dental arches is the use of rapid maxillary
expansion as an initial treatment modality. Mandibular expansion is not
stable if attempted without concomitant orthopedic widening of the upper
arch. The occlusal interdigitation of the teeth in the widened maxilla with
the uprighted mandibular dentition may account for the stability of the
treated results. Thus, in patients with mild to moderate tooth-size/arch-
size problems, particularly in the mandibular dental arch, RME appears to
be a viable treatment option over the long-term. -
ACKNOWLEDGEMENTS
Most of the studies described in this chapter have been conducted
in collaboration with Dr. Lorenzo Franchi and Dr. Tiziano Baccetti of the
Department of Orthodontics at the University of Florence. I also would
like to acknowledge the contributions of the many orthodontic residents
from the University of Michigan who participated in these projects and
who appear as authors on the publications cited in this chapter.
Funding for many of these projects has been made available
through the Thomas M. and Doris Graber Endowed Professorship at the
University of Michigan and through the generosity of Dr. Vincent Finazzo.
Additional support was available through two grants from the American
Association of Orthodontists Foundation.
158
McNamara
REFERENCES
Angell EC. Treatment of irregularities of the permanent or adult teeth.
Dent Cosmos 1860; 1:541-44, 599–600.
Bishara SE, Jakobsen JR, Treder JE, Stasi M.J. Changes in the maxillary
and mandibular tooth size-arch length relationship from early adoles-
cence to early adulthood. A longitudinal study. Am J Orthod Dentofa-
cial Orthop 1989;95:46-59.
Brust EW. Arch dimensional changes concurrent with expansion in the
mixed dentition. Unpublished Masters thesis, Department of Ortho-
dontics and Pediatric Dentistry, The University of Michigan Dental
School, Ann Arbor, 1992.
Brust EW, McNamara JA Jr. Arch dimensional changes concurrent with
expansion in mixed dentition patients. In: Trotman CA, McNamara JA
Jr, eds. Orthodontic Treatment: Outcome and Effectiveness. Craniofa-
cial Growth Series, Center for Human Growth and Development, The
University of Michigan, 1995:30:193-225.
Cameron CA, Franchi L, Baccetti T. McNamara JA Jr. Long-term effects
of rapid maxillary expansion: A posteroanterior cephalometric study.
Am J Orthod Dentofacial Orthop 2002;121:129-135.
Carter GA, McNamara JA Jr. Longitudinal dental arch changes in adults.
Am J Orthod Dentofacial Orthop 1998; 114:88-99.
Chang JY, McNamara JA Jr, Herberger TA. A longitudinal study of skel-
etal side-effects induced by rapid maxillary expansion. Am J Orthod
Dentofacial Orthop 1997; 112:330–337. -
Davis WM, Kronman JH. Anatomical changes induced by splitting of the
midpalatal suture. Angle Orthod 1969:39:126-132.
Dibbets JMH, Nolte K. Effect of magnification on lateral cephalometric
studies. Am J Orthod Dentofacial Orthop 2002;122:196-201.
Fenderson FA, McNamara JA Jr, Baccetti T, Veith C.J. A long-term study
on the expansion effects of the cervical-pull facebow with and without
rapid maxillary expansion. Angle Orthod 2004;74:439-449.
Geran RG, McNamara JA Jr, Baccetti T, Franchi L., Shapiro LM. A pro-
spective long-term study on the effects of rapid maxillary expan-
sion in the early mixed dentition. Am J Orthod Dentofacial Orthop
2006;129:631-640.
159
Long-term Transverse Adaptations
Haas A.J. Rapid expansion of the maxillary dental arch and nasal cavity by
opening the mid-palatal suture. Angle Orthod 1961:31:73-90.
Haas AJ. The treatment of maxillary deficiency by opening the mid-palatal
suture. Angle Orthod 1965:35:200–217.
Haas AJ. Long-term post-treatment evaluation of rapid palatal expansion.
Angle Orthod 1980:50: 189-217.
Herberger TA. Rapid palatal expansion: Long term stability and periodon-
tal implications. Unpublished Master’s thesis, Department of Ortho-
dontics, University of Pennsylvania, Philadelphia, 1987.
Krebs A. Midpalatal suture expansion studies by the implant method over
a seven year period. Trans Eur Orthod Soc 1964;40: 131-142.
Linder-Aronson S, Lindgren J. The skeletal and dental effects of rapid
maxillary expansion. Br J Orthod 1979;6:25-29.
Little RM. Stability and relapse of dental arch alignment. Br J Orthod
1990; 17:235-241.
McNamara JA Jr. Early intervention in the transverse dimension: Is it
worth the effort? Am J Orthod Dentofacial Orthop 2002; 121:572-
574.
McNamara JA Jr. Long-term adaptations to changes in the transverse di-
mension in children and adolescents: An overview. Am J Orthod Den-
tofacial Orthop 2006;129:S71-S74. -
McNamara JA Jr, Baccetti T, Franchi L, Herberger TA. Rapid maxillary
expansion followed by fixed appliances: A long-term evaluation of
changes in arch dimensions. Angle Orthod 2003;73:344–353.
McNamara JA Jr, Brudon WL. Orthodontics and Dentofacial Orthope-
dics. Needham Press, Ann Arbor, 2001.
Moussa R, O’Reilly MT, Close JM. Long-term stability of rapid palatal
expander treatment and edgewise mechanotherapy. Am J Orthod Den-
tofacial Orthop 1995;108:478-488.
Moyers RE, van der Linden FPGM, Riolo ML, McNamara JA Jr. Stan-
dards of Human Occlusal Development. Monograph 5, Craniofacial
Growth Series, Center for Human Growth and Development, The
University of Michigan, Ann Arbor, 1976.
O’Grady PW, McNamara JA Jr, Baccetti T, Franchi L. A Long-term evalu-
ation of the mandibular Schwarz appliance and the acrylic splint ex-
pander in early mixed dentition patients. Am J Orthod Dentofacial
Orthop 2006;130:202-213.
Riolo ML, Moyers RE, McNamara JA Jr, Hunter WS. An Atlas of Cranio-
facial Growth: Cephalometric Standards from the University School
160
McNamara
Growth Study, Monograph 2, Craniofacial Growth Series, Center for
Human Growth and Development, The University of Michigan, Ann
Arbor, 1974.
Shapiro PA. Mandibular dental arch form and dimension. Treatment and
postretention changes. Am J Orthod 1974;66:58-70.
Spillane LM. Arch dimensional changes in patients treated with maxillary
expansion during the mixed dentition. Unpublished Master’s thesis,
Department of Orthodontics and Pediatric Dentistry, The University
of Michigan, Ann Arbor, 1990.
Spillane LM, McNamara JA Jr. Maxillary adaptations following expan-
Sion in the mixed dentition. Semin Orthod 1995; 1:176-187.
Stockfish H. Rapid expansion of the maxilla - Success and relapse. Trans
Eur Orthod Soc 1969;45:469-481.
Timms DJ. An occlusal analysis of lateral maxillary expansion with mid-
palatal suture opening. Dent Pract 1968;18:435-440.
Timms DJ. Long-term follow-up of cases treated by rapid maxillary ex-
pansion. Trans Eur Orthod Soc 1976;52:211-215.
Wendling LK, McNamara JA Jr, Franchi L, Baccetti T. Short-term skeletal
and dental effects of the acrylic splint rapid maxillary expansion ap-
pliance. Angle Orthod 2005;75:7-14.
Wertz R, Dreskin M. Midpalatal suture opening: A normative study. Am J
Orthod 1977;71:367–381.
Zablocki HL, McNamara JA Jr, Baccetti T, Franchi L. The effect of the
transpalatal arch during extraction treatment. Am J Orthod Dentofa-
cial Orthop 2007;in press.
161
162
THE TOOTH/ARCH DISCREPANCY PROBLEM:
IS THE BENEFIT OF EARLY ORTHODONTIC
TREATMENT WORTH THE BURDEN2
James L. Vaden
One has to ask if the benefits of so called “early treatment” outweigh the
burdens for both the public and the orthodontic specialty. The burden sus-
tained by the public for two regimens of treatment involves time, money,
and “aggravation” – time for the patient and the patient’s family; money
from the patient’s family paid to the orthodontist; “aggravation” to both
the patient and the parent, because orthodontic appliances often are un-
comfortable.
Purported benefits of early treatment are 1) a shorter time in “full”
appliances, 2) the conversion of a severe problem into a less severe prob-
lem, 3) a cosmetic benefit in that the patient “always” has straight teeth,
and 4) a psychological benefit that is a result of straighter teeth in the
mixed dentition. Are these benefits supported by the literature? How does
one balance these benefits with the burdens? The answers to these ques-
tions probably depend upon the problem with which the patient presents,
the goal for the ultimate resolution of that problem, and the desires of the
patient and parents.
Most mixed dentition patients who present for orthodontic treat-
ment either have crowding or a bad bite or a combination of both. In order
to eliminate the crowding, the orthodontist must expand the arch (or, to use
a popular term, “develop the arch”) or extract teeth. Prior to carrying out
arch development procedures, it might be a good idea to study the refereed
orthodontic literature on the subject. In 1949, Strang talked about the fal-
lacy of arch expansion. Others also have studied arch expansion and ex-
traction (Little et al., 1981, 1990). A review of 103 retention and stability
articles led Blake and Bibby (1998) to the following conclusions:
1. A patient’s pretreatment lower arch form should be maintained.
Original lower intercanine width should be maintained because
expansion of lower intercanine width is the most predictable orth-
Odontic relapse.
3. Mandibular arch length decreases with time.
163
Tooth/Arch Discrepancy
4. The most stable position of the mandibular incisor is its pretreat-
ment position. Advancing lower incisors is unstable.
A cursory review of the literature would seem to indicate that arch
development generally is not a good alternative for the long-term health of
the dentition. In other words, early treatment with a goal of changing an
extraction problem into a non-extraction problem is suspect and not sup-
ported by current scientific evidence. For simplification of this discussion,
early treatment will be broken down into the Class III problem, the Class
I problem, and the Class II problem.
THE CLASS III PROBLEM
Much has been written about Class III intervention (McNamara,
1987; Ngan, 1992, 1998, 2002, 2005; Ngan et al., 1996; Baccetti et al.,
1998; Nartallo-Turley and Turley, 1998; McGill and McNamara Jr. et al.,
1999; Baccetti et al., 2000; McNamara and Brudon, 2001; Mozayeni-Azar
and Wolf, 2003; Franchi et al., 2004). It is no secret that many clini-
cians successfully use rapid palatal expanders (RPE) with a reverse pull
headgear, chin-cups, orthodontic appliances with Class III elastics, or a
combination of any of these things to impact the developing Class III mal-
occlusion favorably. .
One of the nicest ways to predict whether or not a patient is ame-
nable to early Class III treatment is to use Ngan’s (2005) growth treatment
response vector (GTRV). Simply stated, the GTRV formula can be used to
predict a growth pattern ratio for each patient. The ratio for an individual
with a normal growth pattern from 8 to 16 years of age was calculated to
be 0.77. In a study of 40 patients, 20 of whom were treated successfully
and 20 of whom were treated unsuccessfully, the successful group had a
GTRV range of 0.33 to 0.88. The unsuccessfully treated sample had a
GTRV range of 0.06 to 0.38.
The patient in Figure 1 has a ratio of 100%. Though this patient
presented with an anterior crossbite, her problem was amenable to early
Class III intervention. An appliance was placed that moved the maxillary
incisors forward. Subsequent radiographs (Fig. 2) showed an arch length
discrepancy. Maxillary and mandibular premolars were removed, and ap-
pliances placed in both the mandible and the maxilla. In addition, her
airway improved during early Class III intervention because she was sent
to the otolaryngologist for a tonsillectomy and adenoidectomy procedure.
It has been suggested that the resolution of occluded airways helps with
the resolution of these types of problems.
164
Vaden
Figure 2. After maxillary “2x4” and chin cup treatment, the cross
bite was resolved, but the tooth arch discrepancy was not.

1.65
Tooth/Arch Discrepancy
The bottom line on early Class III intervention is that many, many
patients need and should receive some type of interceptive guidance if
they present with a Class III malocclusion at a very young age (Woodside
et al., 1986). The GTRV ratio can be a very helpful guide in determining
whether a patient can be treated successfully with orthodontics alone or
whether a patient should be left untreated until cessation of growth, at
which time surgical orthodontic treatment can be planned.
THE CLASS I MALOCCLUSION
Some Class I malocclusions are amenable to early occlusal guid-
ance with either early extraction or space preservation procedures. One of
the most comprehensive reviews of this type of treatment has been written
by Jack and Hali Dale (Dale and Dale, 2005). The authors go into great
detail and provide wonderful drawings of eruption guidance procedures.
Their review is based upon clinical experience, the literature, and the work
of Conrad F. Moorrees (Moorrees, 1957, 1958; Moorrees et al., 1957).
The young patient whose x-rays are seen in Figure 3 presented
with a crowded dentition, an orthognathic skeletal pattern, and a low
Frankfort mandibular plane angle. Maxillary and mandibular first pre-
molars were removed at nine years of age to treat the severe crowding.
The patient presented two years later for the initiation offixed orthodontic
treatment. Figure 4 shows the pretreatment face and cephalogram along
with the cephalometric values. Note that after the canines erupted (Fig.
5), there was very little extraction space remaining to be closed with orth-
odontic appliances. The pretreatment and post-treatment facial photo-
graphs (Fig. 6) illustrate the maintenance of facial esthetics with a slight
decrease in the outward curl of the lower lip. The pretreatment and post-
treatment casts (Fig. 7) confirm that the Class I occlusion was maintained
and that appropriate interdigitation of the teeth and acceptable overjet and
overbite were obtained. The pretreatment and post-treatment panoramic
radiographs (Fig. 8) confirm the uprighting of the roots into the extraction
spaces. The pretreatment and post-treatment cephalograms (Fig. 9) and
their respective tracings confirm vertical control and space closure with
no retraction of the maxillary anterior teeth. The pretreatment and post-
treatment superimpositions (Fig. 10) illustrate the downward and forward
movement of the mandible.
166
Vaden
First Premolars
Were Extracted
FMIA FMA | IMPA SNA SNB ANB AOBOOCC Z | UL TC | PFH | AFH INDEX
70 || 20 90 82 77 5 || 2 mm 6 || 70 || 13 mm 9 mm 45 mm. 59 mm .76
Figure 3. Patient presented with a crowded dentition. First premolars
Were extracted.
FMIAIFMAMPASNA SNBANEAO-BOOcc z UL TC PFH AFH INDEx
72 || 17 | 91 81 78 || 3 || 1 mm || 5 || 75 12 mm 10mm/45mm 39mm 0.76
Figure 4. Pretreatment photographs, cephalogram and cephalometric
Values.
The benefits of this “guidance of occlusion” type of treatment for
a Class I patient definitely outweigh the burdens. These benefits include
less treatment time and thus less cost. However, it is not appropriate for all
“crowded” Class I patients to undergo this type of serial extraction proce-
dure. Many patients need leeway space preservation with lingual arches.




167
Tooth/Arch Discrepancy
Gianelly (1995) has documented the successful resolution of many, minor
Class I crowding problems using this approach.
For patients who have minor crowding and for whom the lingual
arch does not resolve the problem, occlusions should be allowed to de-
velop without interference. Treatment should be initiated only after all
permanent teeth have erupted. Most of these patients must undergo ei-
ther arch expansion with interproximal reduction or second premolar ex-
Figure 6. Pretreatment/post-treatment facial photographs.


168
Vaden
traction. Because arch expansion has been proven to be unstable, second
premolar extraction is the treatment of choice.
º,
N º º
Figure 7. Pretreatment/post-treatment casts.
Figure 8. Pretreatment/post-treatment panoramic radiographs.




169
Tooth/Arch Discrepancy
FMIA IFMA | IMPA SNA SNB I ANB IAOBO OCC Z | UL TC PFH | AFH |INDEX
72 17 91 81 78 3. 1 mm. 5 75 12 mm. 10 mm 45 mm. 59 mm 76
71 16 93 82 80 2 1 mm 2 78 || 16 mm. 15 mm. 53 mm 65 mm .80
Early Treatment
Pretreatment
Posttreatment
The following records illustrate a patient whose second premo-
lars were extracted. This is a relatively easy orthodontic treatment that
has very few unfavorable sequelae. The patient’s facial photographs, the
cephalogram and its tracings (Fig. 11) illustrate a relatively pleasing face
Figure 9. Pretreatment/post-treatment cephalograms.
Figure 10. Pretreatment/post-treatment superimpositions.


170
Vaden
Pretreatment
FMIA
IMPA
SNA
SNB
ANB
AO-BO
OCC
Z
TC
PFH
INDEX
64
14
102
82
81
1
0 mm
5
83
14 mm
16mm
53 mm
57mm
0.93
Figure 11. Pretreatment facial photographs, cephalogram and tracing.
Figure 12. Pretreatment casts.
With no facial distortion, a low mandibular plane angle, and somewhat
protrusive mandibular incisors.
The casts (Fig. 12) confirm a Class I
malocclusion with minor maxillary anterior crowding but more severe
mandibular anterior crowding. The pretreatment/post-treatment facial


171
Tooth/Arch Discrepancy
Figure 14. Pretreatment/post-treatment casts.
photographs (Fig. 13) and the pretreatment/post-treatment casts (Fig. 14)
illustrate correction of the crowding, and maintenance of the Class I oc-
clusion. The pretreatment/post-treatment cephalograms (Fig. 15) and the
Superimpositions (Fig. 16) confirm maintenance of the mandibular incisor
position even though extractions were part of the treatment plan.
CLASS II TREATMENT
In today's world, many young patients are subjected to all types
of Class II “correctors” in the mixed dentition. These “magic appliances
are sold in every booth at every orthodontic meeting. Though there are
substantial benefits to early Class III and Class I intervention, a review
of the literature on early Class II treatment indicates that patients receive
no lasting benefit from early intervention (Pancherz, 1979; Pancherz and
Hägg, 1985; Lai and McNamara Jr., 1998; Von Bremen and Pancherz.

172
Vaden
FMLA FMA | IMPA SNA SNB ANB |AOBO OCC Z | UL TC PFH AFH INDEX
64 14 102 82 81 1 0 mm 5 || 83 |14mm. 16mm/53 mm.57mm .93
65 13 102 81 81 0 || 0 mm 5 85 |13mm, 15 mm.59mm.59mm 1.00
Figure 15. Pretreatment/post-treatment cephalograms.
Pretreatment
Postreatment
Figure 16. Pretreatment/post-treatment tracings.
2002; Gianelly, 2003). There just does not seem to be a “magic bullet”
that will help clinicians routinely treat mixed dentition Class II problems.
Clinicians must be cognizant of the fact that a bell curve exists for almost
ºverything. This means that there are exceptions to every rule and that
clinicians who perform a significant amount of Class II early treatment
intervention can show some very favorable results in selected patients.


173
Tooth/Arch Discrepancy
Generally, these patients fall at an extreme end of the bell curve. Howev-
er, as Lysle Johnston states, many early Class II non-extraction treatment
decisions are practice management decisions rather than biological ones
(Livieratos and Johnston, 1995).
Studies conducted at the University of Florida found that the
skeletal effects of Phase I treatment disappear by the time the patient is
old enough for fixed appliance treatment (King et al., 2003; Dolce et al.,
2005). Research conducted at the University of North Carolina concludes
that “the differences created between the treated children and the untreated
control group by Phase I treatment before adolescence disappeared when
both groups received comprehensive fixed appliance treatment during ad-
olescence” (Tulloch et al., 2004). In other words, no study that has been
published to date has conclusively demonstrated that there are lasting ben-
efits of early Class II treatment.
The question then becomes what benefit should we really expect
from early treatment of the Class II malocclusion (Johnston, 1986, 1998;
McNamara et al., 1996)? We might expect some retraction of the maxil-
lary anterior teeth. This benefit probably would enhance the patient’s self-
esteem and please the parents. The opposite question to that of benefits
is what kind of burdens are there to early Class II treatment? First, the
patient must wear appliances, there are trips to the orthodontist’s office,
and there is a financial cost. Second, there is the burden to the orthodontist
in time required to see the patient. Third and very important, the ortho-
dontist and the patient’s family must understand that the problem is not
going to be resolved and probably will not be made less severe by early
intervention. If the patient presented with a tooth/arch discrepancy before
early intervention, it still will be present after treatment or will have been
pushed to the posterior part of the maxillary arch if maxillary molars are
distalized. Therefore, early Class II treatment should be attempted with
diagnostic care and with informed consent obtained from all parties. No
Class II patient should be treated early with either the patient or the parents
expecting that early intervention will eliminate the need for later treat-
ment, or shorten treatment time or eliminate the original diagnosed need
for extractions. The records of the following patient illustrate this point.
The facial photographs, cephalogram, tracings and cephalometric
values seen in Figure 17 show a patient with a significant Class II maloc-
clusion. The casts (Fig. 18) confirm that the patient has a full step Class
II malocclusion with impinging overbite and protruded incisors. This
174
Vaden
Early Treatment
FMIA) FMA | IMPASNA SNB ANB |AO-BOL OCC Z | UL | TC | PFH | AFH [Index
56 25 99 || 83 || 74 9 || 10 mm 10 || 65||11 mm 13 mm 60 mm.40 mm 0.66
Figure 17. Pretreatment facial photographs, cephalogram, tracings and
cephalometric values for patient with a Class II malocclusion.
Figure 18. Pretreatment casts for Class II patient.


175
Tooth/Arch Discrepancy
Figure 19. Pretreatment photographs of Class II patient (left) and pho-
tographs taken after six months of treatment and six months of growth
plus “relapse” (right).
ent
Figure 20. Pretreatment casts taken of Class II pati
(left) and casts taken after six months of treatment and
six months of growth plus “relapse” (right).
FMIAIFMA|IMPA|SNA SNBANBAOBO |OCC Z | UL | TC | PFH | AFH |INDEX
56 25 99 || 83 || 74 || 9 || 10 mm. 10 || 65||11mm 13mm, 60 40 66
58 || 23 99 || 84 || 75 || 9 || 10 mm 9 || 62 |11 mm 12mm 45 60 .75
Figure 21. Pretreatment and interim cephalograms of
Class II patient.



176
Vaden
patient was treated for six months with maxillary appliances and maxillary
headgear. The photographs and casts seen in Figures 19 and 20 were taken
after six months of treatment and six months of growth plus “relapse”
(pretreatment records are on the left and early treatment records are on
the right). The photographs show an improvement in facial esthetics, but
the casts show only minor retraction of the maxillary anterior teeth. The
pretreatment and six-month cephalograms (Fig. 21) confirm that there has
been some retraction of the maxillary anterior teeth, but there has not been
a lot of change in the skeletal numbers.
The patient was placed in retainers, and after her permanent teeth
had erupted, her maxillary first premolars and mandibular second pre-
molars were extracted to enable the retraction of the maxillary incisors
and the protraction of the mandibular molars. The interim/post-treatment
photographs (Fig. 22) show marked improvement in facial esthetics for
both profile and frontal views. The interim/post-treatment casts (Fig. 23)
confirm correction of the Class II occlusion to a Class I canine-protected
Occlusion. The interim/post-treatment cephalograms, tracings and cepha-
lometric values (Fig. 24) show a reduction of the ANB angle, maintenance
of mandibular incisor position, and significant retraction of the maxillary
anterior teeth. The interim/post-treatment superimpositions (Fig. 25) show
significant downward and forward growth of the mandible, retraction and
intrusion of the maxillary incisors, and protraction of the mandibular first
molars to facilitate a Class II occlusion correction. The recovery records
(Figs. 26-28) confirm continued improvement in facial esthetics, occlu-
Sion and in mandibular response. The pretreatment/recovery smiles (Fig.
29) illustrate the treatment results. Total treatment time was 38 months.
Figure 22. Interim/post-treatment photographs of Class II patient.
Was this patient helped with early Class II intervention? Yes and
no. While there was improvement in this patient’s facial esthetics during
the mixed dentition phase, the fundamental occlusal problem remained.
Treatment still required extraction of premolars and significant additional
treatment time. Treatment of the underlying malocclusion was not made
ºasier with early treatment. This patient's records illustrate the point that
there is no “magic bullet” for early Class II treatment.

177
Tooth/Arch Discrepancy
Figure 23. Interim/post-treatment casts of Class II
patient. -
FMIA FMA | IMPA SNA SNB | ANB AOBOOCC Z | UL TC | PFH | AFH INDEX
58 23 99 84 75 9 |10mm 9 || 62 | 11 mm. 12 mm 45 mm 60 mm .75
62 23 95 79 75 4 || 5 mm 9 || 75 | 16 mm 13 mm. 50 mm 63 mm .79
Figure 24. Interim/post-treatment cephalograms, tracings and cephalo-
metric values for Class II patient.




178
Vaden
Pretreatment
Posttreatment ---
Figure 25. Interim/post-treatment superimpositions for Class II patient.
FMIA|FMA IMPA snailsne ANB | Ao-Bo occ z | UL Tc PFH | AFH |Index
62 22 96 || 79 || 75 4. 3 mm 9 || 76 | 16 mm. 14 mm. 51 mm 63 mm 0.8
Figure 26. Patient's photographs, cephalogram, tracings and cephalo-
metric values taken at one-year recall.



179
Tooth/Arch Discrepancy
Pretreatment
Posttreatment
Retention
Figure 28. Patient’s cephalogram taken at one-year recall.

180
Vaden
Figure 29. Pretreatment/interim/post-treatment/recall smiles.
On the other hand, the patient seen in Figures 30–35 was not treat-
ed until the late mixed/early permanent dentition phase. The pretreatment
photographs, cephalogram, tracing, and cephalometric values (Fig. 30)
confirm that the patient had a skeletal and dental Class II occlusion with
poor facial esthetics. The pretreatment casts (Fig. 31) show a Class II
Occlusion, a deep bite, and excessive overjet. Granted, the dental maloc-
clusion is not as severe as that of the previous patient, but many clinicians
who routinely perform early treatment would have put this child in appli-
ances at the age of nine years.
FMAIFMA
IMPA
SNA
SNB
ANB
AOBO || OCC
Z
UL
TC
PFH
65 17
98
82
75
7 mm 5°
70°
15 mm.
8 mm
43 mm
Figure 30. Pretreatment photographs, cephalogram, tracing, and cepha-
lometric values of Class II patient who had no early intervention.



181
Tooth/Arch Discrepancy
Figure 31. Pretreatment casts of Class II patient who
had no early intervention.
Figure 32. Pretreatment/post-treatment photographs of Class II patient
who had no early intervention.
The pretreatment/post-treatment photographs (Fig. 32) show a
marked improvement in facial esthetics; the mandible has a much bet-
ter relationship to the upper part of the face. The pre- and post-treatment
casts (Fig. 33) confirm that the dental malocclusion has been corrected.
The patient’s maxillary first premolars and mandibular second premolars
were extracted exactly as was done for the patient previously described.
Treatment and force systems used also were exactly the same. The pre-


182
Vaden
Figure 33. Pretreatment/post-treatment casts of Class
II patient who had no early intervention.
|
FMIA IFMA | IMPA SNA I sne ANB |Aobol occ z | UL TC | PFH | AFH | INDEX
65 17 98 82 75 7 7 mm 5° 70 | 15 mm 8 mm 43 mm 60 mm .71
66 17 97 78 75 3 2 mm. 5° 76 || 13 mm. 10 mm 46 mm, 61 mm . .76
Figure 34. Pretreatment/post-treatment cephalograms, tracings and
Cephalometric values for Class II patient who had no early intervention.
treatment/post-treatment cephalograms, tracings and cephalometric val-
*S confirm that the FMA and the mandibular incisor position were main-
tained and that both the SNA and ANB were decreased (Fig. 34). The su-
Perimpositions (Fig. 35) show a nice, forward mandibular response along
With control of the posterior dentition.


183
Tooth/Arch Discrepancy
Figure 35. Superimpositions for Class II patient who had no early in-
tervention.
This patient was in treatment for 22 months. Compare that with
the previous patient who was in treatment for 38 months. Was the en-
tire 38 months of treatment necessary? Both patients were treated in
precisely the same way with the same force systems. While it is true
that the second patient did not have as severe a problem as the first par
tient, the first patient could have been treated in less time and with
less visits to the orthodontist and still have received the same result.
SUMMARY
“Interception” of a Class III malocclusion is definitely warranted
for many patients. A careful analysis of a patient’s malocclusion, breath-
ing, etc., is essential if one is to make the correct decision as to whether or
not the patient’s problem is amenable to early Class III intervention.
In most instances, the Class I patient can benefit from occlusal
guidance if the crowding is severe. This will shorten treatment time and
lessen the burden for both patient and family. Space preservation is also
a viable procedure if the Class I patient has minor mandibular anterior
crowding.
Early Class II intervention should be attempted only if all parties
concerned know the literature and know what to expect. Of the three types
of malocclusions, the Class II malocclusion presents with the most prob:
lems. Early treatment must be viewed with a more pessimistic outlook.

184
Vaden
REFERENCES
Baccetti T, Franchi L, McNamara JA Jr. Treatment and post-treatment
craniofacial changes after rapid maxillary expansion and facemask
therapy. Am J Orthod Dentofacial Orthop 2000; 118:404-413.
Baccetti T. McGill JS, Franchi L. McNamara JA Jr, Tollaro I. Skeletal
effects of early treatment of Class III malocclusion with maxillary
expansion and facemask therapy. Am J Orthod Dentofacial Orthop
1998; 113:333-343.
Blake M, Bibby K. Retention and stability: A review of the literature. Am
J Orthod Dentofacial Orthop 1998; 114:299-306.
Dale JG, Dale HC. Interceptive guidance of occlusion with emphasis on
diagnosis. In: Graber TM, Vanarsdall RL Jr, Vig KWL, eds, Orthodon-
tics: Current Principles and Techniques, 4th ed, St. Louis: Elsevier
2005:375–469.
Dolce C, Schader RE, McGorray SP, Wheeler TT. Centrographic analysis
of 1-phase versus 2-phase treatment for Class II malocclusion. Am J
Orthod Dentofacial Orthop 2005; 128:195-200.
Franchi L, Baccetti T. McNamara JA Jr. Postpubertal assessment of treat-
ment timing for maxillary expansion and protraction therapy followed
by fixed appliances. Am J Orthod Dentofacial Orthop 2004; 126:565-
568.
Gianelly A. Leeway space and the resolution of crowding in the mixed
dentition. Semin Orthod 1995; 1:188-189.
Gianelly A. Rapid palatal expansion in the absence of crossbites: Added
value? Am J Orthod Dentofacial Orthop 2003;124:362-365.
Jensen E, Kai-Jen Yen, Moorrees CF, Thomsen SO. Mesiodistal crown
diameters of the deciduous and permanent teeth in individuals. J Dent
Res 1957:36:39–47.
Johnston LE Jr. A comparative analysis of Class II treatment. In: Vig PS,
Ribbens KA, eds, Science and Clinical Judgment in Orthodontics,
Craniofacial Growth Series, Center for Human Growth and Develop-
ment, The University of Michigan, Ann Arbor 1986; 19:103-148.
Johnston LE Jr. Growth and the Class II patient: Rendering unto Caesar.
Semin Orthod 1998:4:59-62.
King GJ, McGorray SP, Wheeler TT, Dolce C, Taylor M. Comparison of
peer assessment ratings (PAR) from 1-phase and 2-phase treatment
protocols for Class II malocclusions. Am J Orthod Dentofacial Orthop
2003:3:489–496.
185
Tooth/Arch Discrepancy
Lai M, McNamara JA Jr. An evaluation of two-phase treatment with the
Herbst appliance and preadjusted edgewise therapy. Semin Or-
thod 1998; 1:46-58.
Little RM, Riedel RA, Stein A. Mandibular arch length increase during the
mixed dentition: Postretention evaluation of stability and relapse. Am
J Orthod Dentofacial Orthop 1990:97:393-404.
Little RM, Wallen TR, Riedel RA. Stability and relapse of mandibular an-
terior alignment: First premolar extraction cases treated by traditional
edgewise orthodontics. Am J Orthod Dentofacial Orthop 1981:80:349-
365.
Livieratos FA, Johnston LE Jr. A comparison of one-stage and two-stage
non-extraction alternatives in matched Class II samples. Am J Orthod
Dentofacial Orthop 1995; 108:118-131.
McGill JS, McNamara JA Jr. Treatment and post-treatment effects of rapid
maxillary expansion and facial mask therapy. In: McNamara JA Jr.
ed, Growth Modification: What Works, What Doesn't And Why, Cra-
niofacial Growth Series, Center for Human Growth and Development,
The University of Michigan, Ann Arbor 1999:35:123-152.
McNamara JA Jr. An orthopedic approach to the treatment of Class III
malocclusion in young patients. J Clin Orthod 1987;9:598-608.
McNamara JA Jr, Brudon WL. Orthodontics and Dentofacial Orthope-
dics. Ann Arbor, Needham Press, 2001.
McNamara JA Jr, Peterson JE Jr, Alexander RG. Three-dimensional diag-
nosis and management of Class II malocclusion in the mixed denti-
tion. Semin Orthod 1996:2:114-137.
Moorrees CF. Growth changes of the dental arches: A longitudinal study. J
Canadian Dent Assoc 1958:24:449.
Moorrees CF, Chadha JM. Available space for the incisors during dental
development: A growth study based on physiological age. Angle Or-
thod 1965:35:12–22.
Mozayeni-Azar M, Wolf G. Post-treatment and postretention evaluation of
protraction facemask therapy using traditional cephalometrics, Mas-
ters Thesis, Case Western Reserve University, 2003.
Nartallo-Turley PE, Turley PK. Cephalometric effects of combined palatal
expansion and facemask therapy on Class III malocclusion. Angle
Orthod 1998;68:217-224.
Ngan P. Biomechanics of maxillary expansion and protraction in Class III
patients. Am J Orthod Dentofacial Orthop 2002;121:582-583.
186
Vaden
Ngan P. Early timely treatment of Class III malocclusion. Semin Orthod
2005; 11:140-145.
Ngan P, Hägg U, Yui C, Merwin D, Wei SH. Soft tissue and dentoskeletal
profile changes associated with maxillary expansion and protrac-
tion headgear treatment. Am J Orthod Dentofacial Orthop 1996; 109:
38-49.
Ngan P. Wei SH, Hägg U, Merwin D, Stickel B. Effects of protraction
headgear on Class III malocclusion. Quintessence Int 1992:23:197-
207.
Ngan P, Yiu C, Hu A, Hägg U, Gunel E. Cephalometric and occlusal
changes following maxillary expansion and protraction. Eur J Orthod
1998:20:237-254.
Pancherz H. Treatment of Class II malocclusions by jumping the bite with
the Herbst appliance. A cephalometric investigation. Am J Orthod
1979;76:423–442.
Pancherz H, Hägg U. Dentofacial orthopedics in relation to somatic matu-
ration. An analysis of 70 consecutive cases treated with the Herbst
appliance. Am J Orthod 1985;88:273-287.
Strang RHW. The fallacy of denture expansion as a treatment procedure.
Angle Orthod 1949;12-17.
Tulloch JF, Proffit WR, Phillips C. Outcomes in a 2-phase randomized
clinical trial of early Class II Treatment. Am J Orthod Dentofacial Or-
thop 2004;6:657-667.
Turley P. Orthopedic correction of Class III malocclusion with palatal ex-
pansion and custom protraction headgear. J Clin Orthod 1988;5: 314-
325.
Woodside DG, Linder-Aronson S, Lundström A. Mandibular growth di-
rection following adenoidectomy. Am J Orthod 1986;89:273-284.
Von Bremen J, Pancherz H. Efficiency of early and late Class II, Division
1 treatment. Am J Orthod Dentofacial Orthop 2002; 121:31-37.
187
188
RESPONSES OF THE CRANIOFACIAL COMPLEX TO
FORCED MANDIBULAR PROTRUSION WITH THE
HERBST APPLIANCE: A CEPHALOMETRIC STUDY
IN MACACA MULATTA
Kevin N. Kieu
James A. McNamara, Jr.
The Herbst appliance, developed by Emil Herbst a century ago (Herbst,
1910), currently is the most commonly used functional appliance in the
United States (McNamara and Brudon, 2001). This appliance, reintro-
duced in the late 1970s by Pancherz (1979, 1981, 1982), has gained steadi-
ly in popularity for nearly three decades, in part because patient coopera-
tion is not essential for success.
Since the publication of Pancherz's early cephalometric studies,
the treatment effects of this type of appliance have been evaluated in nu-
merous clinical investigations (Pancherz, 1985, 1997; McNamara et al.,
1990; Wieslander, 1993; Windmiller, 1993; Burkhardt et al., 2003; Schae-
fer et al., 2004). In the adolescent, this tooth-borne appliance produces
about a 50% skeletal change (mostly in the mandible) and a 50% dentoal-
Veolar adaptation (i.e., upward and backward movement of the maxillary
posterior teeth; lower incisor proclination).
TIMING OF INTERVENTION
The Herbst appliance has been recommended for use primarily
in the early permanent dentition patient, although some clinicians have
advocated earlier Herbst therapy such as initiating treatment in preschool
children (Wieslander, 1984; Hamilton, 2001) or in children in the mixed
dentition (Dischinger, 1989). Considerable controversy remains concern-
ing the most advantageous time to begin orthopedic treatment of Class II
malocclusions, with some clinical investigators recommending a deferral
of functional jaw orthopedics until the time of the pubertal growth spurt.
If the maturation of the cervical vertebrae (CVM) is used as a marker,
increased growth responses have been observed with functional appliance
therapy at about the time of the pubertal growth spurt (CVM stages 3-4)
than at earlier stages of maturation (CVM stages 1–2; Baccetti et al., 2000,
2005; Baccetti and Franchi, 2001).
189
Experimental Herbst
The developmental time frame has been extended by Pancherz
and colleagues (Ruf and Pancherz 1999a; 1999b, 2003; Pancherz and von
Bremen, 2002) who have broadened the discussion of the age of inter-
vention to include orthopedic intervention in young adults by way of the
Herbst appliance. They summarize the contemporary concept of routine
Class II treatment as growth modification in juveniles and adolescents,
camouflage orthodontics in post-adolescents, and corrective jaw surgery
in adults. These clinical investigators, however, have raised the possibility
of orthopedic correction of Class II malocclusion in young adults, recom-
mending such intervention in females 18-24 years of age and in males
20–25 years of age. Camouflage orthodontics and orthognathic surgery
remain treatment options in older adults.
Pancherz and Ruf based their recommendations on the results of
cephalometric and magnetic resonance imaging (MRI) studies of the tem-
poromandibular joint performed in adolescent and young adult Herbst pa-
tients (Ruf and Pancherz, 1998; 1999a; Pancherz and Ruf, 2000). These
studies showed that condylar growth in young adults could be reactivated.
The increase in mandibular length accomplished by Herbst therapy in both
adolescents and young adults seemed to be a result of condylar and gle-
noid fossa remodeling (Ruf and Pancherz, 1998; 1999a). Additionally, the
mandibular skeletal contribution to Class II molar and overjet correction
in young adult Herbst patients amounted to 25% on average (Pancherz and
Ruf, 2000).
Although the Herbst appliance has been evaluated extensively
in various patient populations, there have been very few comprehensive
studies of the treatment effects produced by the Herbst appliance carried
out in a controlled experimental environment. The purpose of the current
research effort was to do just this by investigating the effects of the Herbst
appliance on the dentofacial complex of juvenile and adult rhesus mon-
keys (Macaca mulatta), a species of primates that are morphologically
similar to humans. Using rhesus monkeys as the research model allowed
the analysis of the results to be conducted at the tissue level. The juvenile
and adult groups were treated for different lengths of time, and by compar-
ing the findings from these groups, the interrelationship between treatment
duration, the maturation level of the animal, and treatment effects could
be examined.
190
Kieu and McNamara
PREVIOUS EXPERIMENTAL STUDIES
There have been many experimental studies investigating the ef-
fects of functional appliances on animals. Such studies can be divided
arbitrarily into two groups: studies of functional protrusion and studies of
forced mandibular protrusion.
Functional Protrusion
In the search for a biologic and physiologic explanation of the
findings of clinical studies involving functional appliances, numerous ex-
perimental studies have investigated the dentofacial responses to anterior
displacement of the mandible in research animals. In his classic study,
Breitner (1940, 1941) found that the use of incisor and molar bite planes
to create Class III dental relationships in young rhesus monkeys resulted
in the resorption of the angle of the mandible and in bone deposition along
both the posterior wall of the glenoid fossa and the posterior of the con-
dyle. Breitner’s histological findings suggested there was an increase in
mandibular growth as well as anterior repositioning of the joint. He felt
that the practical implications of such manipulations were that Class II
malocclusions could be treated by forward movement of the mandible.
Baume and Derichsweiler (1961) studied the anterior displace-
ment of the mandible by bite planes in two young rhesus monkeys. They
found that the alterations in the morphology of the mandibular condyles
were of sufficient magnitude to be obvious in cephalometric radiographs.
They concluded from their histological findings that the condylar cartilage
responded to the functional displacement by becoming twice as wide in
the two experimental animals as the condylar cartilage in the one normal
control animal. They also believed that there was “concrete evidence”
showing that the condylar growth center responds to functional therapy.
Hiniker and Ramfjord (1966) carried out an experimental study
on rhesus monkeys using a technique similar to that of Baume and De-
richsweiler except that they selected mature animals in order to elimi-
nate the factor of growth in their study. They reported that the temporo-
mandibular joints of all five experimental monkeys essentially were
191
Experimental Herbst
unaffected. They found a narrow zone of newly formed bone on the an-
terior aspect of the postglenoid spine and the posterior border of the neck
of the condyle. The new bone formation was no greater in the two-week
animal than in the 40-week animal. They concluded that the changes were
only adaptive in nature and that there was no mandibular growth. It ap-
peared that the teeth were moving to compensate for the anterior displace-
ment of the condyle in the joint cavity. The maxillary dental arch moved
distally, while the mandibular dental arch moved mesially. This move-
ment of the splinted arches of the teeth and the corresponding lack of sig-
nificant change in the TMJ was considered proof by Hiniker and Ramfjord
that in adult rhesus monkeys, the accommodation to anterior displacement
of the mandible took place in the dentition rather than in the TMJ. They
also found that the changes, seen in the temporomandibular joints were
insignificant, nonprogressive and possibly reversible.
Ramfjord and Enlow (1971) inserted bite planes in three adult
monkeys to induce the animals to function in a forward mandibular posi-
tion. At intervals of one and a half, two and three years, the monkeys were
sacrificed and studied. Ramfjord and Enlow concluded that there was no
alteration in condylar growth, only dentoalveolar compensation.
Joho (1968) treated four young rhesus monkeys for a period of
150 to 260 days with fixed gold splints in order to initiate a forward func-
tioning position of the mandible during closure. Two test animals were
observed for 60 and 90 days respectively after removal of the appliance
in order to note any relapse. Three similarly aged monkeys were used as
controls. All of the experimental animals developed a Class III relation-
ship. The mandibles of the treated monkeys exhibited an increase in length
that was twice that of the control animals. Joho attributed this increased
mandibular length to an enlargement of the condylar-ramus angle rather
than to a stimulation of mandibular growth. During the post-experimental
phase, there was no change in the condylar-ramus angle, but the resulting
dentoalveolar changes relapsed markedly.
Stöckli and Willert (1971) used eight growing Macaca irus
monkeys with full gold coverage splints in order to displace the mandi-
ble anteriorly for periods ranging from 5 to 210 days; one animal had
the splints removed for 90 days after 180 days of anterior displacement
during closure. In their histological assessment of the TMJ, Stöckli and
Willert found substantial compensatory tissue response in the condylar
cartilage and in the articulating portion of the temporal bone. The most
distinct histological reaction was observed after 25 days of anterior func-
192
Kieu and McNamara
tional displacement during closure. The cartilaginous covering of the con-
dylar head increased posteriorly to become five times as thick as the ante-
rior portion. An anterior crossbite was established in the 120-, 180- and
210-day animals. Stöckli and Willert concluded that the condylar carti-
lage as well as the articular portion of the temporal bone could be induced
to adapt by mechanical anterior displacement of the mandible in young
animals. In addition, they concluded that the tissue response was of a
“transitional character” and that the results were permanent if the appli-
ance was worn more than 180 days.
In the initial nonhuman primate study at the University of Michi-
gan, Elgoyhen and colleagues (1972) used 16 juvenile rhesus monkeys
(six experimental animals and 10 control animals) in studies of functional
protrusion. Unimaxillary cast gold appliances were used on three mon-
keys and bimaxillary appliances were used on three monkeys, so that the
mandible was prompted forward during function and closure. All of the
experimental monkeys had Class III occlusions after five months of treat-
ment. The investigators found an average degree of incremental condylar
growth that was 58% to 66% greater than that seen in the control group.
They concluded that the appliance increased both the rate and the amount
of condylar growth. In this study there was an inhibited anterior and verti-
cal eruption of the maxillary buccal segments as well as a mesial migration
of the mandibular buccal segments.
The first major experimental study of treatment timing in func-
tional protrusion therapy was carried out by McNamara (1972, 1973) as
a follow-up to the Elgoyhen investigation. The effects of anterior func-
tional mandibular displacement were studied in a large sample of rhesus
monkeys (N=64). McNamara divided the 28 experimental animals into
four groups (infant, juvenile, adolescent and adult); 36 additional rhesus
monkeys served as controls. Gold castings were inserted that caused the
animals to bite forward 2 mm and open vertically 2 mm. Radiographic,
histologic, and electromyographic studies were performed for both the
control and experimental groups. Increases in growth at the condyle
were the greatest in the infant and juvenile groups. The average growth
increment for the experimental juvenile animals measured at condylion
during the 13th week of the experimental period was 51% greater than
the average increment of the same animals during the 13-week control
period. In the adolescent and adult animals, most of the changes in the
mandible occurred only in the dentition. McNamara suggested that the
alteration of condylar growth direction seen in the infant and juvenile
193
Experimental Herbst
monkeys may be as important for mandibular position as the increase in
growth increment.
Petrovic and co-workers (1975) used growing rats as their experi-
mental model. The rats wore removable functional devices for 8 to 12 hours
per day for variable periods of time. The appliance caused the mandible
to occlude in a mesial position. There was an increase in condylar growth
that occurred through stimulation of the prechondroblastic zone (interme-
diate) of the condylar head. The increase in condylar growth resulted in a
slightly longer mandible in the experimental group when measured from
the condyle to the mental foramen when compared to controls.
In another experiment using nonhuman primates, McNamara and
Carlson (1979) investigated the effects of anterior mandibular displace-
ment in 28 Macaca mulatta monkeys ranging in age from 18 to 24 months
(14 experimental animals and 14 control animals). Cast ticonium appli-
ances were placed on the maxillary and mandibular dentition so that the
mandible functioned in a 4 mm anterior and a 3 mm inferior position.
The animals were sacrificed at 2, 4, 6, 8, 10, 12, and 24 weeks after appli-
ance placement and studied histologically. Significant adaptive changes
were found in the prechondroblastic and chondroblastic zones in the 2-,
4-, 6-, 8- and 10-week animals, with the maximum response occurring in
the 6-week animals. There also was an increased rate of bone resorption
anteriorly at the insertion of the lateral pterygoid muscle. McNamara and
Carlson concluded that adaptive responses can occur in the mandibular
condyle of the juvenile rhesus monkey following alteration of the func-
tional position of the mandible. They hypothesized that the localized in-
crease in thickness of the condylar cartilage probably indicates an increase
in both the rate and amount of proliferation of chondrocytes, which leads
to an overall increase in length of the mandible.
In order to determine whether or not these adaptive changes were
transient or permanent, McNamara and Bryan (1987) followed longitu-
dinally a group of 11 experimental juvenile male rhesus monkeys wear-
ing a protrusive appliance from the mixed dentition period to adulthood.
Twelve animals were used as controls. The results of their study indicated
that the final mandibular length of the experimental animals was signifi-
cantly greater, i.e., 6.7 mm, than that of the control animals.
194
Kieu and McNamara
Forced Mandibular Protrusion
The Herbst appliance differs significantly from the appliances
used in the previously discussed studies in that the mandible is held in a
fixed forward position rather than induced into a forward position during
closure. There have been relatively few experimental studies reported in
the literature in which nonhuman primates were treated with the Herbst
appliance. Woodside and co-workers (1983) reported that after placement
of a Herbst appliance in a juvenile cynomolgus monkey for a 13-week
experimental period, a small amount of remodeling was noted in the Supe-
rior aspect of the condyle. Histological findings showed bone formation
on the surface of and within the condyle. The glenoid fossa also showed
extensive bone formation in the posterior region after 13 weeks. Based
on the results of this experiment, Woodside and colleagues suggested that
chronic or continuous alteration in mandibular position within the neu-
romuscular environment produces extensive condylar remodeling and
change in mandibular size.
In a later study, Woodside and co-workers (1987) investigated the
remodeling changes in the condyle and glenoid fossa of nonhuman pri-
mates following a period of Herbst appliance therapy. Seven cynomol-
gus monkeys (Macaca fascicularis) were used in this study: five of the
animals wore Herbst appliances for 6 to 13 weeks and two animals wore
inactivated appliances as sham controls. The appliances were bonded to
the dentition and activated 2 mm initially. Thereafter, the mandible was
advanced progressively 1 to 2 mm every two weeks. The total activation
was 7 to 10 mm. -
Decalcified sections from the mandibular condyles and the gle-
noid fossa were examined histologically. The anterior wall of the post-
glenoid spine showed considerable new bone deposition and dense fi-
brovascular tissue proliferated in the posterior part of the articulated disk
to fill in the space created by the condylar displacement. These effects
were demonstrable in both the juvenile and adult animals. Woodside
and co-workers postulated that forward remodeling of the glenoid fossa
also may have contributed to an anterior repositioning of the mandible.
The maxillary superimpositions showed that three of four experimental
animals had distal movement of the maxillary first molars, but minimal
lingual and inferior displacement of the incisors. Dentoalveolar chang-
es in the mandible of the experimental animals consisted of a labial and
195
Experimental Herbst
inferior movement of the incisors and a forward and slightly superior
movement of the first molars.
In Summary, animal studies generally have shown significant and
consistent treatment responses to functional appliances; orthopedic results
in human patients have not been as predictable. On the other hand, both
clinical and experimental studies involving the Herbst appliance dem-
onstrated definite patterns of dentofacial changes in the treated groups.
These changes included:
1. An increase in mandibular length
Distal movement of the maxillary buccal segment
Lingual tipping of the maxillary incisors
Labial tipping of the mandibular incisors
Forward and vertical movement of the mandibular buccal seg-
ment
:
In retrospect, the experimental studies of Herbst appliance treat-
ment have significant limitations. For example, the studies conducted by
Woodside and co-workers (1983, 1987) had small samples. The earlier of
the two studies had a sample size of only one. In their later study, the sam-
ple consisted of only seven monkeys, two of which were sham controls.
In addition, the responses of the experimental animals were not evaluated
according to their age, even though the investigators identified one animal
as a juvenile, one animal as an adult and three animals as adolescents. The
two sham control animals were adolescents. Moreover, a total of 7 to 10
mm of mandibular protrusion activated on these experimental animals was
extreme, given the overall length of the mandibles of the monkeys in the
studies.
THE PURPOSE OF THE CURRENT STUDY
Additional experimental research with a more comprehensive
study design encompassing both growing and adult animals was needed
to provide more information about the treatment effect of the Herbst ap-
pliance on the craniofacial complex. In response to this need, a federally-
funded investigation was carried out to determine:
1. the effects of the bonded Herbst appliance on different parts
of the dentofacial complex of the Macaca mulatta monkey;
2. the relationship of maturational level to the magnitude of
dentoskeletal changes produced by the Herbst appliance; and
196
Kieu and McNamara
3. which parts of the craniofacial complex would vary their re-
sponses as treatment duration was lengthened.
MATERIALS AND METHODS
Sample
Sixty-four rhesus monkeys (Macaca mulatta) from the primate
colony of the University of Michigan's Center for Human Growth and
Development were used. These animals were separated into juvenile and
adult groups according to their stage of dental eruption. The juvenile ani-
mals had complete deciduous dentitions and fully erupted first permanent
molars. According to the tooth eruption scales of Hurme and Van Wa-
genen (1953, 1961), these animals were approximately 18 to 24 months
of age. The adult monkeys had complete permanent dentitions, including
fully erupted third molars indicating that they were about six to seven
years of age. The experimental animals were followed for periods of 3,
6, 12, and 24 weeks after appliance placement. The temporomandibular
joint regions of each experimental animal were prepared for histologic
analysis at the conclusion of the study. The results of these analyses are
reported in detail elsewhere (McNamara et al., 2003; Peterson and McNa-
mara, 2003). The histologic and cephalometric parts of the overall study
are compared later in this chapter.
The juvenile group consisted of 35 animals, 19 of which were
experimental subjects. Three animals wore a Herbst appliance for three
weeks, six animals for six weeks, six animals for 12 weeks, and four ani-
mals for 24 weeks. The remaining 16 monkeys served as controls. The
three-week Herbst animals were not monitored cephalometrically; they
were only analyzed histologically.
There were 29 animals in the adult group, 14 of which were ex-
perimental subjects. The duration of Herbst appliance wear also was vari-
able in the adult group. Three animals wore the appliance for three weeks,
five animals wore the appliance for six weeks, three animals for 12 weeks,
and three animals for 24 weeks. The remaining 15 adult animals served
as controls. In addition, the 24-week juvenile and adult groups also were
radiographed at the 6- and 12-week intervals to assess maxillary and man-
dibular skeletal changes. The three-week adult animals were not radio-
graphed, so they were not included in the cephalometric analysis portion
of the study.
197
Experimental Herbst
Appliance Design
The appliance was constructed to mimic as closely as possible
the acrylic splint design of McNamara (1988) used for patients (Fig. 1).
In order to decrease the likelihood that appliance might disengage from
the dentition during the experiment, the maxillary splint was modified so
that it covered the entire palate and the maxillary dentition. The plunger
direction also was reversed in order to avoid possible impingement of the
plunger on the mandible ramus. The mandible was advanced once, ap-
proximately 3 mm. Unlike the study of Woodside and co-workers (1987),
the appliance was not reactivated during the study period.
Metallic Implants
Tantalum implants (Björk, 1955, 1968) have been used as reliable
reference markers in many experimental studies (e.g., McNamara, 1973:
McNamara and Graber, 1975; McNamara et al., 1976). When positioned
correctly in areas in which there is no resorptive modeling, these implants
provide an optimal means for superimposing head films taken in a stan-
dardized fashion. In this study, tantalum implants were inserted into the
mandible, maxilla, frontal bone and the cranial base region of each animal
prior to the beginning of the experimental period.
- - -
Figure 1. The acrylic splint Herbst appliance modified for use in the rhesus mon-
key. The direction of the Herbst mechanism was reversed to avoid impingement
of the mandibular ramus.
Four implants were placed extraorally in the right side of the man-
dible: two in the lower border of the mandible, one in the chin and one in
the lower third of the ramus (Fig. 2). Four pairs of implants were placed
bilaterally in the maxillary area; one in the bone between the central and
lateral incisors, one above the second deciduous molar or the first per-
manent molar, one in the palatal region above the canine, and one near

198
Kieu and McNamara
the zygomatic arch. Four implants were placed in the cranial base in the
area above and below the spheno-occipital synchrondrosis through a small
incision in the soft palate (Fig. 3). Finally, three implant pins were placed
in the frontal bone (Fig. 2).
Cephalometric Analysis
After being anesthetized, each animal was radiographed imme-
diately before and after appliance placement and at the end of the study
periods. Kodak Type M Industrial film was used to enhance visualization
of detail. Two lateral films were taken on each occasion, one with the teeth
in occlusion and one with the mouth wide open. The wide-open position
permitted a more accurate tracing of the condyle.
Changes in the craniofacial complex of the experimental animals
Were analyzed by modifying the technique used in previous studies (Mc-
Namara, 1972; McNamara and Bryan, 1987). Two reference coordinate
Systems were constructed in the initial radiographs—one in the maxilla
and one in the mandible.
Figure 2. Drawing of monkey skull indicating the position
of the tantalum implants in the maxilla, mandible and frontal
bone.

199
Experimental Herbst
Figure 3. Schematic drawing of the placement of tantalum im-
plants in the cranial base of the rhesus monkey.
Maxilla. The maxillary horizontal reference line (X, ) was con-
structed by drawing a line through sella parallel to the original occlusal
plane. The vertical reference line (Y) was constructed perpendicular to
the X axis at sella (Fig. 4). In order to determine the overall effects
in the maxillary complex, the lateral tracings were superimposed on the
implants in the anterior portion of the cranial base and along the inferior
portion of the endocranial surface of the orbital roof. The movement of
each implant was measured in both the horizontal and vertical dimensions
(Fig. 4). Consequentially, the net effect of the appliance on the process
of displacement and migration of the maxilla was determined by compar-
ing the measurements obtained from treatment and control groups. By
Superimposing on the maxillary implants (Fig. 5), the effects of maxillary
displacement were cancelled. As a result, the response of the maxillary
dentition to the appliance was determined by comparing experimental data
to control data.
Mandible. The mandibular horizontal reference line (X, ) was
drawn through the symphyseal point parallel to the X axis (Fig. 6).
These templates provided a means of measuring skeletal and dentoalveo-
lar changes relative to these coordinate systems in successive tracings.
The position of the symphyseal point was determined by bisecting a line
drawn between infradentale and the intersection point of the lingual out-
line of the mandibular symphysis with the lower border of the mandible

200
Kieu and McNamara
(McNamara and Bryan, 1987). The relative angulation of the condyle and
ramus to the body of the mandible was determined first by constructing a
line tangent to the posterior border of the condyle and ramus. The condy-
lar-ramus angle was determined by measuring the angle of intersection of
this line with the X axis.
Statistical Analysis
Student's t-test was used to compare the changes in the different
Variables between the control and experimental groups. Summary statis-
tics in two-way ANOVA also were used to test for treatment differences,
age differences, and any interaction between these factors.
Error of the Method Study
In order to determine the method error in locating and measuring
the changes in different landmarks, pre- and post-treatment radiographs
of 10 randomly selected subjects were traced and superimposed with
ANTERIORIMPHOR
->
-
Nº. ...ANTERIORIMPVERT
Figure 4. Points and lines used to determine adaptations within the max-
illa. The positions of the maxillary and mandibular implants are shown.
The black line indicates the tracing of the initial cephalogram and the red
line indicates the tracing of the subsequent cephalogram.




201
Experimental Herbst
Xm.
ZO
s
Figure 5. Points and lines used to determine the movement of the maxil-
lary dentition. The black line indicates the tracing of initial cephalogram
and the red line indicates the tracing of the subsequent cephalogram. M:
molar; DM: deciduous molar; C canine; I: incisor.
CONDYLAR-
RAMUS ANGLE
Figure 6. Points and lines used in measuring mandibular dimensions. SC: Su-
perior condylion, PC; posterior condylion; C. condylion. The condylar-ramus
angle is formed by the intersection of a line drawn along the posterior aspect Of
the ramus and the X, line. The symphysis point is determined at the midpoint
of a line drawn from infradentale to the inferior curvature of the symphysis. The
black line indicates the tracing of initial cephalogram and the red line indicates the
tracing of the subsequent cephalogram.





202
Kieu and McNamara
measurements recorded on two different occasions. Intraclass correlation
coefficients were calculated; the correlation coefficients for 18 variables
were in the range of 0.91 to 0.98. A correlation coefficient greater than
0.90 was deemed acceptable.
RESULTS
Comparison of Starting Forms
In that one of the aims of this investigation was to determine the
response of different parts of the craniofacial complex, particularly the
mandible, to the Herbst appliance, it was important to verify that the con-
trol and experimental groups were not significantly different with regard
to their initial mandibular measurements.
In the juvenile groups, there were no statistically significant dif-
ferences between the experimental and control groups at the start of the
study. There was an average difference of only 0.1 mm in the pretreatment
mandibular lengths and 0.4° average difference in initial condylar-ramus
angle. These average differences were not statistically significant.
The adult groups exhibited slightly greater differences in the pre-
treatment mandibular lengths and condylar-ramus angles. The average
differences between the control and experimental groups were about 1.5
mm and 2°, neither of which was statistically significant.
Treatment Effects: Juvenile Animals
Remembering that the overall study was both cephalometric and
histologic in nature, the number of animals in each experimental group
decreased as animals were euthanized for histologic analysis at each study
interval. Also, the teeth were obscured in the cephalograms when the ap-
pliance was bonded in place. As mentioned earlier, the three-week ani-
mals were studied at the histological level only.
Maxillary Skeletal Changes (Table 1)
Six weeks. The horizontal displacement of the maxilla in the ex-
perimental group was restricted slightly by the appliance at the end of
six weeks. There was significantly less anterior movement of the maxil-
lary implants in the experimental group as compared to the controls. The
Vertical component of maxillary displacement, represented by the inferior
movement of the implants, was not affected by the appliance. The ante-
rior and posterior maxillary implants in the control and the experimental
groups move inferiorly about the same amount.
203
Experimental Herbst
Twelve weeks. By the end of 12 weeks, the forward movement
of the maxillary implants was inhibited slightly by the appliance. Both
implants had less forward displacement of 0.3 mm in comparison to con-
trols. The posterior implant moved downward an average amount of 0.1
mm in the experimental group compared to 0.5 mm in the control group,
whereas there was no difference in the movement of the anterior maxillary
implant.
Twenty-four weeks. At the end of 24 weeks, the horizontal move-
ment of the maxilla was reduced by about 1 mm in the experimental group,
a statistically significant difference. There also was an inhibition of infe-
rior movement of the maxilla in the Herbst group by 0.5 mm.
Table 1. Maxillary skeletal changes in juvenile animals.

Ave SD Ave SD Sig
Six Weeks Herbst (N=6) Control (N=15)
Anterior Implant
Horizontal (mm) 0.20 0.15 0.60 0.30 IlS
Vertical (mm) 0.20 0.12 0.20 0.30 11S
Posterior Implant
Horizontal (mm) 0.30 0.27 0.60 0.30 *
Vertical (mm) 0.20 0.30 0.20 0.25 IlS
Twelve Weeks Herbst (N=10) Control (N=15)
Anterior Implant
Horizontal (mm) 1.00 0.11 1.20 0.18 sk
Vertical (mm) 0.20 0.04 0.50 0.45 11S
Posterior Implant
Horizontal (mm) 0.90 0.18 1.20 0.17 >k:k
Vertical (mm) 0.10 0.07 0.50 0.35 >k:k
Twenty-four Weeks Herbst (N=10) Control (N=15)
Anterior Implant
Horizontal (mm) 1.10 0.15 2.10 0.43 :k:k
Vertical (mm) –0.10 0.30 0.80 0.41 sk xk
Posterior Implant
Horizontal (mm) 1.10 0.19 2.10 0.35 >k sk
Vertical (mm) 0.50 0.60 1.00 0.56 11S
* p < 0.05; ** p < 0.01; ns = not significant
204
Kieu and McNamara
Mandibular Skeletal Changes (Table 2)
Six weeks. Significant changes in mandibular length, represented
by the measurement from condylion to the symphysis, could be observed
at the end of six weeks. The average increase of mandibular length in the
experimental group was 2.0 mm compared to 1.0 mm in the controls. The
posterior aspect of the condyle in the experimental group expressed an av-
erage incremental increase twice greater than that of the controls (1.3 mm
versus 0.6 mm). Significant incremental growth in the posterosuperior
direction also was noted in the experimental animals. The experimental
group exhibited an average increase of 1.6 mm compared to 0.9 mm in the
control animals; however, this difference was not statistically significant.
There also was a relative opening of the condylar-ramus angle of 2.4° in
the experimental group.
Twelve weeks. The overall increase in mandibular length of the
experimental group was 1.4 mm greater in the Herbst group at the end of
12 weeks, and the condylar-ramus angle opened by 4.4° in contrast to the
controls whose condylar-ramus angle closed by 1.5°. The condyle of the
12-week animals exhibited a significant larger posterior and posterosu-
perior growth than that of the control animals. There was no significant
change in the superior growth between the two groups.
Table 2. Mandibular skeletal changes in juvenile animals.

Ave SD Ave SD Sig.
Six Weeks Herbst (N = 6) Control (N = 15)
Mand Length (mm) 2.00 0.80 1.00 0.60 sk 2:
Condylion (mm) 1.60 0.88 0.90 0.53 IIS
Sup Cond (mm) 1.10 1.72 0.80 0.47 11S
Post Cond (mm) 1.30 0.46 0.60 0.43 *::
Cond Ramus (deg) 1.50 0.50 -0.90 0.50 sk sk
Twelve Weeks Herbst (N = 10) Control (N = 15)
Mand Length (mm) 3.60 0.78 2.20 0.52 sk *
Condylion (mm) 3.20 0.72 2.20 0.50 * †
Sup Cond (mm) 1.60 0.68 1.70 0.55 IIS
Post Cond (mm) 2.80 1.70 1.70 0.56 sk ºr
Cond Ramus (deg) 4.40 0.39 -1.50 0.56 skil:
Twenty-four Weeks Herbst (N = 4) Control (N = 15)
Mand Length (mm) 6.50 1.10 4.40 0.87 2: 3:
Condylion (mm) 5,70 0.90 4.30 0.98 it ºr
Sup Cond (mm) 2.10 1.16 3.60 0.74 sk 2.
Post Cond (mm) 5.70 0.67 3.10 1.07 × Sº
Cond Ramus (deg) 6.10 0.50 –2.20 0.67 sk it
* p <0.05; **p-0.01; ns = not significant
205
Experimental Herbst
Twenty-four weeks. A comparison of the overall mandibular length
between the Herbst and control animals revealed that at the end of the
six-month period, the experimental animals exhibited greater mandibular
growth than the controls; an average difference of 2.1 mm, a significant
value, was seen between the two groups. At this point, the incremental
growth of the mandibular condyle in the experimental group also under-
went significant changes. Condylar growth in the posterior and posterosu-
perior directions was increased approximately 2 mm and 1.4 mm, respec-
tively, as compared to the controls. However, the superior aspect of the
condyle grew less than that of the controls by an average of 1.5 mm.
In addition to influences on condylar growth, the appliance signifi-
cantly altered the condylar-ramus angle in the Herbst group in comparison
to the control group. Instead of decreasing in value, the condylar-ramus
angle in the Herbst group increased by 6.0°, a difference of 8.3° compared
to the controls.
Maxillary Dentoalveolar Changes
As mentioned previously, the outlines of the teeth were obscured
when the acrylic splint appliance was bonded in place. Thus, dentoalveo-
lar structures in the animals were measured only before appliance place-
ment and after appliance removal. For the purpose of this chapter, only
data from the 24-week Herbst animals are reported (Table 3).
Six weeks. Significant changes in the horizontal movement of the
maxillary dentition were noted at the end of the six-week period. In con-
trast to the slight forward movement in the control group, the maxillary
dentition of the experimental animals exhibited a posterior displacement.
In addition, the canines and the incisors showed a larger backward move-
ment than did the molars and the deciduous molars. In the experimental
group, the anterior segment of the maxillary dentition experienced a sig-
nificant amount of vertical eruption. There were statistically significant
increases in the vertical development of the incisors, the canines and the
deciduous molars of the experimental animals.
Twelve weeks. Significant changes in the horizontal movement
of the maxillary dentition were observed at the end of 12 weeks. In the
control group, there was an average amount of 0.4 mm forward move-
ment observed in the canine and molar regions, and the incisors moved
anteriorly 0.5 mm. The opposite movement was seen in the Herbst
group, in which the entire maxillary dentition moved posteriorly. For
instance, a lingual tipping (-0.3 mm) of the anterior segment was ob-
206
Kieu and McNamara
served in the experimental group resulting in a net treatment effect of 0.8
mm and 0.7 mm in the incisor and canine positions, respectively. Similar
findings also were seen in the posterior segment in which backward move-
ment of the molar (-0.4 mm) was observed.
Differences in the vertical positions of the molar between the con-
trol and experimental groups were not noticeable. However, significant
differences were evident when the vertical positions of the canine and inci-
sor of the two groups were compared. These teeth were extruded an aver-
age amount of 1.2 mm in the animals treated with the Herbst appliance,
which was 0.7 mm greater than that of the controls.
Twenty-four weeks. Although the entire maxillary dentition of the
experimental group exhibited an inhibition of anterior movement, the only
significant difference (0.8 mm) between the two groups at the end of the
24 weeks was in the incisor region (Table 3). More importantly, a forward
movement of the maxillary dentition was apparent at the end of this ex-
perimental period in contrast to the backward movement observed in the
animals in the six- and twelve-week groups. There also was an indication
of an intrusive effect of the appliance on the molar region with an intrusion
of 0.7 mm in the Herbst group compared to an eruption of 1.0 in the con-
trol group. On the other hand, in the experimental group the incisor and
canine exhibited a significant amount of extrusion (0.9 mm and 0.7 mm,
respectively) relative to that of control group (Table 3).
Table 3. Maxillary dentoalveolar changes in 24-week juvenile animals.

Herbst (N = 4) Control (N = 15)
Ave SD Ave SD Sig
Incisor
Horizontal (mm) 0.20 0.10 1.00 0.68 Sk
Vertical (mm) 1.80 0.35 0.90 0.53 >k:k
Canine
Horizontal (mm) 0.20 0.16 0.70 0.50 ITS
Vertical (mm) 1.70 0.34 1.00 0.44 >k
Molar
Horizontal (mm) 0.20 0.16 0.60 0.54 11S
Vertical (mm) -0.70 0.11 1.00 0.30 sk
* p <0.05; **p-0.01; ns = not significant
Mandibular Dentoalveolar Changes
Six weeks. A comparison of the anterior movement in the man-
dibular dentition between the control and the experimental groups at the
end of six weeks revealed that the appliance produced a significantly
207
Experimental Herbst
greater forward movement in the experimental animals than it did in the
control animals. For instance, the mandibular first molar in the experi-
mental group moved forward approximately 0.5 mm compared to 0.1 mm
in the control group. The differences between the control and experimen-
tal groups became greater when the anterior movement of the incisor and
canine were considered. In the experimental group, the incisor and canine
moved forward 0.9 mm and 0.7 mm, respectively, in contrast to an anterior
movement of 0.2 mm and 0.1 mm in the controls.
The vertical dentoalveolar changes seen in the experimental group
were variable, depending on the region in the dental arch. For example,
the appliance did not have a significant effect in the vertical movement of
the anterior segment and the molar region, but the deciduous molars of the
experimental group showed a statistically significant amount of extrusion
as compared to the controls.
Twelve weeks. At the end of the three-month experimental period,
there was a significant change in the anterior movement of the mandibular
dentition between the control and the experimental animals. Average dif-
ferences ranging from 0.7 mm to 0.8 mm, depending on which tooth in
the arch was measured, indicated that the forward movement of the lower
dentition in the Herbst group was enhanced when compared to that of the
controls. -
A comparison of the vertical movement of the molars in the two
groups revealed that the experimental animals had a statistically signifi-
cant greater movement (p<0.01) in this region; however, the appliance
had no significant effect on the vertical movement of the rest of the lower
dentition.
Twenty-four weeks. Similar to what happened at the end of the
six- and twelve-week periods, the horizontal movement of the mandibular
dentition in the Herbst group increased significantly at the end of 24 weeks
(Table 4). For example, an increase of 1.2 mm was seen in the anterior
region in the Herbst group compared to control group. Similar findings
were seen in the buccal segment. The effect of the appliance on the verti-
cal movement of the mandibular dentition of the experimental animals
varied depending on the region in the arch. In comparison to the control
group, the posterior region showed a significant amount of extrusion. The
anterior region, on the other hand, exhibited a significant inferior move-
ment (0.3 mm) compared to that of the controls.
208
Kieu and McNamara
Table 4. Mandibular dentoalveolar changes in 24-week juvenile animals.

Herbst (N = 4) Control (N = 15)
Ave SD Ave SD Sig.
Incisor
Horizontal (mm) 2.00 0.40 0.80 0.47 >k:k
Vertical (mm) 0.40 0.13 0.70 0.19 >k:k
Canine
Horizontal (mm) 1.80 ().28 0.60 0.30 >k >k
Vertical (mm) 0.50 0.19 0.80 0.20 >k >k
Molar
Horizontal (mm) 1.70 0.33 0.60 0.38 >k:k
Vertical (mm) 1.90 0.28 1.00 0.26 >k:k
* p <0.05; **p-0.01; ns = not significant
Treatment Effects: Young Adult Animals
It also should be remembered that the adult experimental group
consisted of 29 animals. Five of the animals were radiographed after six
weeks of wearing the appliance, three after 12 weeks and three after 24
weeks. The 24-week animals also were radiographed at the 12-week peri-
od to determine maxillary and mandibular skeletal responses to the Herbst
treatment. In short, six monkeys were radiographed at the end of the 12-
week period. The three-week animals were not radiographed.
Maxillary Skeletal Changes (Table 5)
Six weeks. At the end of the six-week interval, there was no max-
illary response of the adult group to forced protrusive function. Although
the implants in the experimental group displayed greater vertical and
horizontal movement, the average difference (0.02 mm) between the two
groups was beyond the level of accuracy of the measurement.
Twelve weeks. The inferior movement of the posterior maxillary
implant (0.2 mm) indicated that the appliance had a slight but statistically
significant intrusive effect on the experimental animals after the three-
month period. In contrast to what was observed in the vertical dimension,
there was no significant difference in the anterior movement between the
two groups. The horizontal displacement of the maxilla in the adult group
did not appear to be significantly influenced by the Herbst appliance at this
interval.
209
Experimental Herbst
Twenty-four weeks. There was a significant change in the vertical
growth of the maxilla in the experimental group at the end of 24 weeks.
There was an intrusive effect seen in the premolar and molar regions with
superior movements of 0.4 mm and 0.2 mm, respectively, compared to the
control group. There was no significant difference, however, in horizontal
displacement of the maxillary implants.
Mandibular Skeletal Changes (Table 6)
Six weeks. At the end of six weeks, there was a hint of increased
condylar growth in the experimental animals compared to the controls, but
this increase was only 0.3 mm greater than that seen in the controls. In
addition, there were slight changes in the incremental condylar growth in
all directions: posterior, Superior and posterosuperior. In both groups, the
condylar-ramus angle remained relatively unchanged.
Twelve weeks. Mandibular length increased an average of 0.8 mm
in the Herbst group compared to an increase of 0.4 mm in the control
group at the end of 12 weeks; however, this difference was not statisti-
cally significant. In the Herbst group, there were statistically significant
increases in Superior and posterior condylar growth. There also was a sta-
tistically significant difference in the change of the condylar-ramus angle
between the experimental group and the control group; however, these
changes were not clinically significant.
Twenty-four weeks. The only direction of condylar growth signifi-
cantly changed at the end of 24 weeks was posterior; however, the aver-
age difference of 0.4 mm in posterior condylar growth was not clinically
significant.
Maxillary Dentoalveolar Changes
Six weeks. The analysis of the maxillary dentoalveolar changes in
the animals treated with the Herbst appliance for six weeks demonstrated
that the maxillary dentition was distalized instead of moving anteriorly
as seen in the control animals. For example, the maxillary first molar in
the Herbst group demonstrated a significant greater posterior movement
(0.3 mm) than that in the control group. A similar finding was seen in the
movement of the anterior segment in the experimental group. In addition
to backward movement, there also was a statistically significant increase
in the vertical movement in the upper incisor region. Nevertheless, the
changes observed in the downward movement of the buccal segment were
clinically insignificant.
210
Kieu and McNamara
Table 5. Maxillary skeletal changes in adult animals.

Ave SD Ave SD Sig
Six Weeks Herbst (N = 5) Control (N = 15)
Anterior Implant
Horizontal (mm) 0.04 0.09 0.02 0.07 11S
Vertical (mm) 0.04 0.09 0.01 0.05 IlS
Posterior Implant
Horizontal (mm) 0.04 0.09 0.02 0.07 11S
Vertical (mm) 0.04 0.09 0.01 0.05 11S
Twelve Weeks Herbst (N = 6) Control (N = 15)
Anterior Implant
Horizontal (mm) 0.10 (). 12 0.10 0.18 11S
Vertical (mm) 0.20 0.16 0.01 0.05 11S
Posterior Implant
Horizontal (mm) 0.10 0.12 0.10 0.18 11S
Vertical (mm) –0.10 0.12 0.04 0.08 >k sk
Twenty-four Weeks Herbst (N = 3) Control (N = 15)
Anterior Implant
Horizontal (mm) 0.10 0.11 0.30 0.35 11S
Vertical (mm) -0.40 0.40 0.03 0.07 >k >k
Posterior Implant
Horizontal (mm) 0.10 0.11 0.30 0.36 11S
Vertical (mm) –0.10 0.15 0.10 0.15 :k
* p <0.05; **p-0.01; ns = not significant
Twelve weeks. In comparison to the control animals, the entire
maxillary dentition of the experimental animals was relocated posteriorly
by the end of 12 weeks. The molar region in the Herbst group had an av-
erage difference of 0.4 mm in posterior movement compared to the molar
region in the control group. A lesser amount of distal movement (0.2 mm)
Was seen in the canine region. In addition, 0.6 mm of lingual tipping in
the incisor region was observed in the Herbst group. Further, in a manner
Similar to the six-week experimental animals, these 12-week experimental
animals exhibited a significant upward movement in the molar region and
a significant downward movement in the incisor region when compared to
the control animals.
211
Experimental Herbst
Table 6. Mandibular skeletal changes in adult animals.

A ve SD Ave SD Sig
Six Weeks Herbst (N = 6) Control (N = 15)
Mand Length (mm) 0.50 0.1 ! 0.20 0.20 x *
Condylion (mm) 0.50 0.07 0.20 0.18 × sk
Sup Cond (mm) 0.50 0.07 0.10 0.17 × 5],
Post Cond (mm) 0.10 0.22 0.03 0.10 ×
Cond Ramus (deg) 0.10 0.22 -0.03 0.13 *
Twelve Weeks Herbst (N = 10) Control (N = 15)
Mand Length (mm) 0.80 0.08 0.40 0.42 x: xk
Condylion (mm) 0.80 0.09 0.50 0.47 × 3%
Sup Cond (mm) 0.50 0.20 0.20 0.23 InS
Post Cond (mm) 0.80 0.40 0.30 0.37 × k
Cond Ramus (deg) 0.90 0.45 0.30 0.40 xir sk
Twenty-four Weeks Herbst (N = 4) Control (N = 15)
Mand Length (mm) 1.00 0.06 0.90 0.29 InS
Condylion (mm) 1.00 0.06 0.90 0.26 InS
Sup Cond (mm) 0.40 0.60 0.40 0.20 IIS
Post Cond (mm) 1.00 0.12 0.60 0.34 sk
Cond Ramus (deg) 1.40 0.15 0.70 0.36 × 5],
* p <0.05; **p-0.01; ns = not significant
Twenty-four weeks. At the end of 24 weeks, analysis of the hori-
zontal movement of the maxillary dentition indicated that the experi-
mental group experienced a distalizing force from the appliance (Table
7). The anterior and posterior dentoalveolar regions both exhibited a pos-
terior movement significantly differently from the anterior movement of
the teeth seen in the controls. There also were significant changes in the
vertical movement of the maxillary dentition. In comparison to the con-
trol animals, the experimental animals exhibited a significant amount of
extrusion in the anterior region; the posterior region appeared to move
inferiorly (Table 7).
Table 7. Maxillary dentoalveolar changes in adult animals at 24 weeks.

Herbst (N = 3) Control (N = 15)
Ave SD Ave SD Sig
Incisor
Horizontal (mm) 1.10 0.47 1.00 0.20 *::
Vertical (mm) 1.80 0.35 0.90 0.53 Sk sk
Canine
Horizontal (mm) 0.90 0.40 0.20 0.26 >k sk
Vertical (mm) 0.70 0.10 0.20 0.17 sk x),
First Molar
Horizontal (mm) 0.80 0.26 0.20 0.16 sk 2:
Vertical (mm)-0.70 0.30 0.23 0.20 0.15 *::
* p <0.05; **p-0.01; ns = not significant
212
Kieu and McNamara
Mandibular Dentoalveolar Changes
Six weeks. At the end of six weeks, differences in the horizon-
tal movement of the mandibular dentition between the control and the
Herbst animals were statistically significant. In general, a greater amount
of forward movement was observed in the Herbst group than in the control
group. In terms of vertical development, the response of the mandibular
dentition was variable, depending on the region in the arch. In the experi-
mental group, the anterior segment exhibited an intrusion amount of 0.2
mm. On the other hand, the posterior segment demonstrated a superior
movement of 0.3 mm.
Twelve weeks. At the end of 12 weeks, a significantly greater for-
ward movement occurred in the entire mandibular dentition of the Herbst
animals compared to controls. There was an average of 0.9 mm greater
anterior movement of the lower incisor in the Herbst group than in the
control group. In the molar region, there was a lesser difference (0.4 mm)
in the forward movement between the two groups.
There were two opposite vertical movements observed in the
mandibular arch of the experimental animals. The anterior region showed
intrusive movements of 0.3 mm and 0.7 mm at the incisor and canine,
respectively, as compared to those of the control animals. There was an
average of 0.4 mm greater movement in the superior direction in the first
molar region of the Herbst group, indicating an extrusion of the posterior
dentition by the appliance.
Table 8. Mandibular dentoalveolar changes in adult animals at 24 weeks.

Herbst (N = 4) Control (N = 15)
Ave SD Ave SD Sig
Incisor
Horizontal (mm) 1.30 0.17 0.10 0.20 × 3k
Vertical (mm) 0.30 O.12 0.10 0.17 Sk:k
Canine
Horizontal (mm) 1.30 0.25 0.10 0.13 × Sk
Vertical (mm) 0.20 0.29 0.02 0.14 InS
First Molar
Horizontal (mm) 1.50 0.30 0.20 0.16 Skºk
Vertical (mm)-0.70 0.50 O.35 0.20 0.23 sk 2:
* p <0.05; **p-0.01; ns = not significant
Twenty-four weeks. After six months of wearing the appliance,
anterior movement of lower dentition in the Herbst group was signifi-
cantly increased, averaging 1.3 mm greater than that of the control group
213
Experimental Herbst
(Table 8). There also were indications of changes in the vertical move-
ment of the mandibular dentition. Similar to what occurred during the 6-
and 12-week periods, differential extrusive/intrusive movements of teeth
were observed in the mandibular dentition of the experimental animals.
Extrusion was seen in the molar region and intrusion was seen in the an-
terior segment.
DISCUSSION
This study was a cephalometric appraisal of the skeletal and dentoalveolar
adaptations produced through forced mandibular protrusion by way of the
acrylic splint Herbst appliance (McNamara, 1988). As one might expect,
greater skeletal treatment effects were observed in the juvenile animals,
with significant dentoalveolar adaptations observed in both groups.
In the juvenile Herbst animals, there was a slight inhibition of
the forward and downward movement of the maxillary skeletal complex
compared to those of the control animals, as indicated by the movement of
the anterior and posterior maxillary tantalum implants. The greatest man-
dibular response to the Herbst appliance was seen in the juvenile animals,
with an increase in mandibular length of slightly more than 2 mm seen at
the 12-week and 24-week intervals (Table 2). In the adult animals, the
difference in all increases in mandibular length were less than 0.5 mm be-
tween groups at all study intervals (Table 5), indicating that there were no
clinically significant increases in mandibular length in the adult animals at
any treatment interval.
Histological Analysis of Current Animals
The overall design of this federally-funded project described in
detail above involved both histological and cephalometric analyses. Be-
cause of the protocol used, the sample size decreased as the study con-
tinued. Fortunately, however, it was possible to look at the temporoman-
dibular joints of the animals histologically, something that normally is not
possible in a cephalometric study. The results of these companion inves-
tigations are described below.
Juvenile Animals. Peterson and McNamara (2003) evaluated the
temporomandibular joint changes in the juvenile animals considered in
this investigation. Structural adaptations were noted in the mandibular
condyle, infratemporal fossa, and the posterior ramal border of the ex-
perimental animals compared to the control animals. Specifically, in-
214
Kieu and McNamara
crease proliferation of the condylar cartilage was noted. These adaptations
occurred primarily in the posterior and posterosuperior condylar regions.
Bone deposition was noted along the anterior border of the postglenoid
spine; bone resorption occurred along its posterior border. Significant
bony apposition also was evident along the posterior border of the man-
dibular ramus during early experimental periods. No gross or microscopic
pathological changes were noted in the temporomandibular joint in any of
the animals.
The results of this study in growing animals indicated that the
response of the temporomandibular joint region in the juvenile animals
undergoing forced mandibular protrusion with the Herbst bite-jumping
mechanism was similar to previous studies investigating the response to
functional protrusion with cast occlusal splints (McNamara, 1972, 1973;
McNamara and Carlson, 1979). It appeared that all of the young animals
in the current study were actively growing and responding to the func-
tional stimuli provided by the Herbst appliance.
Young Adult Animals. The adaptations within the temporomandib-
ular joint region of the adult animals from the present investigation were
studied by McNamara and co-workers (2003). Significant variation was
observed in the temporomandibular joint regions of the control animals,
with some animals having identifiable prechondroblastic and chondroblas-
tic layers of condylar cartilage and others with more fibrous coverings of
the mandibular condyle. In the experimental animals, adaptive changes
in the condylar cartilage were evident as early as three weeks, with the
dimensions of the condylar cartilage increasing gradually throughout the
experimental period. Only minor changes were noted in the articular tis-
sue. All adult control animals had a bony cap, and the bony cap persisted
in the experimental animals. Significant bone deposition occurred along
the anterior surface of the postglenoid spine only in the 6- and 12-week
experimental groups. No significant areas of bone deposition and resorp-
tion were noted along the posterior border of the ramus. In comparison to
the juvenile animals, however, the response to the Herbst appliance was
relatively modest.
Comparison of Histological and Cephalometric Findings
The morphological differences in treatment response between
“growing” (juvenile) and “nongrowing” young (adult) animals are shown
graphically in Figure 7. The baseline data from the juvenile and adult
animals are shown on the left side of the bar graphs. It is obvious that
responses at the tissue level were observed in both juvenile and adult
215
Experimental Herbst
animals, with the greater response occurring in the more rapidly growing
animals. The condylar cartilage in all regions at baseline (i.e., control val-
ues) was two or more times as thick in the juvenile animals as it was in the
adult animals. The juvenile animals responded to the forced mandibular
protrusion more rapidly than did the adult animals (Fig. 7), although re-
sponses at the tissue level were observed at all study intervals.
500 - - 500 ---
ºum Posterior Cartilage 500 Post. Sup, Caſtilage um Superior Cartilage
400 I 400 400
300 300 300
200 200 200
too too 100
d o º
Control 3 wº. 6 wº. 12 wº. 24 wº Control 3 wº. 6 WK 12Wº 24W: Contro. 3 WK 6 wº. 12Wº 24W:
- Condylar Cartilage - Juvenile
um Posterior Cartilage 500 um Post. Sup, Cartilage um Superior Cartilage
400 400 400
300 300 300
200 F l 200 200
100 too 100
-
0. o 0.
Control 3 WK 6 WK 12W, 24W, Control. 3 WK 6 WK 12Wº 24W, Control 3 WK 6 wº 12Wº 24W:
Condylar Cartilage - Adult
Figure 7. Graphic representation of the relative thicknesses of the condylar car-
tilages of the mandibular condyle using the Bioquant" measuring system. For
each condyle, the thicknesses were determined at three regions: superior, pos-
terosuperior, and posterior. The results for the juvenile and young adult animals
are shown (derived from Peterson and McNamara, 2003, and McNamara et al.
2003).
From a cephalometric standpoint, the juvenile Herbst animals re-
sponded in a similar manner to animals in previous experimental stud-
ies of functional mandibular protrusion described earlier in this chapter,
showing at least a 2 mm increase in mandibular length compared to con-
trol animals. In contrast, the adult Herbst animals, which demonstrated an
obvious response at the tissue level, did not show indications of changes in
mandibular length greater that 0.5 mm, a clinically insignificant amount.
Both age groups, however, demonstrated dentoalveolar responses in both
arches cephalometrically, indicating that the appliance produced signifi:
cant dentoalveolar adaptations in both dental arches.
Clinical studies generally have shown that the Herbst appliance
used in growing Class II patients results in about a 50% skeletal (primar-


216
Kieu and McNamara
ily in the mandible) and a 50% dentoalveolar adaptation throughout the
craniofacial complex. In general, the same ratio of skeletal-to-dentoal-
Veolar change was observed in the growing animals evaluated in the cur-
rent study. In the relatively few clinical studies of the Herbst appliance
used in young adult Class II patients, however, the percentage reported
was 20% skeletal adaptation and 80% dentoalveolar adaptation. Although
responses were seen at the tissue level within the mandibular condyle in
the young adult animals, no clinically meaningful increase in mandibular
length was produced. Thus, the responses to Herbst appliance therapy in
the young adult may be overwhelmingly dentoalveolar in nature.
SUMMARY AND CONCLUSIONS
This study investigated the effects of the acrylic splint Herbst ap-
pliance on the dentofacial complex of rhesus monkeys. The experimental
group monitored cephalometrically consisted of 20 juvenile and 14 adult
animals. These animals wore the appliance for periods ranging from 6 to
24 weeks. An additional seven animals (four juveniles and three adults)
were sacrificed at three weeks and only studied histologically. Fifteen
juvenile and 15 adult animals were used as controls. Specific skeletal and
dental adaptations were studied using serial cephalometric radiographs
and metallic implants. The following results were obtained:
1. Absolute mandibular growth was significantly increased in
the juvenile experimental group. Negligible changes were
observed in the young adult group.
2. Posterior and posterosuperior incremental growth in the juve-
nile group was increased significantly, leading to a change in
the direction of condylar growth. No similar results were seen
in the young adult group.
3. The forward growth of the maxilla in the juvenile group was
slightly but significantly inhibited.
4. Both juvenile and young adult groups exhibited similar dento-
alveolar adaptations:
a. restriction of horizontal movement of the maxillary buc-
cal segment;
b. forward movement of the mandibular dentition;
217
Experimental Herbst
c. intrusion on the maxillary posterior segment and mandib-
ular anterior region; and
d. extrusion of the maxillary anterior region and lower pos-
terior region.
It is inappropriate, of course, to make a direct extrapolation from
these experimental findings to clinical applications. However, from the
consistent and significant patterns of responses found in this study plus
the supporting findings from related clinical studies, the Herbst appliance
appears to be suitable for the treatment of growing individuals with Class
II malocclusions presenting with a retrognathic mandible and retroclined
mandibular incisors. In young adult animals, however, the responses ob-
served cephalometrically were limited primarily to the maxillary and man-
dibular dentoalveolar regions; there was no evidence to support the idea
that the Herbst appliance can produce a “clinically significant” change in
mandibular length in the young adult.
ACKNOWLEDGMENTS
The authors wish to thank Dr. Barbara Nesbitt Emerick for her
invaluable technical support during the experimental phase of this project.
This research was supported in part by United States Public Health Ser-
vice Grant DE-03610 and by funds made available through the Thomas M.
and Doris Graber Endowed Professorship at the University of Michigan
School of Dentistry. Illustrations are by William L. Brudon.
REFERENCES
Baccetti T, Franchi L. Maximizing esthetic and functional changes in Class
II treatment by appropriate treatment timing. In: McNamara JA Jr,
Kelly KA, eds. Frontiers of Dental and Facial Esthetics. Craniofacial
Growth Series, Department of Orthodontics and Pediatric Dentistry
and Center for Human Growth and Development, The University of
Michigan, Ann Arbor 2001:38.
Baccetti T, Franchi L, McNamara JA Jr. The cervical vertebral maturation
(CVM) method for the assessment of optimal treatment timing in den-
tofacial orthopedics. Semin Orthod 2005;11:119-129.
Baccetti T, Franchi L, Toth LR, McNamara JA Jr. Treatment timing for
twin block therapy. Am J Orthod Dentofacial Orthop 2000;118:159-
170.
218
Kieu and McNamara
Baume L, Derichsweiller H. Is the condylar growth center responsive to
orthodontic therapy? An experimental study in Macaca mulatta. Oral
Surg 1961; 14:347-362.
Björk A. Facial growth in man, studied with the aid of metallic implants.
Acta Odontol Scand 1955; 13:9–34.
Björk A. The use of metallic implants in the study of facial growth in chil-
dren: Method and application. Am J Phys Anthrop 1968:29:243-254.
Breitner C. Bone changes resulting from experimental orthodontic treat-
ment. Am J Orthod Oral Surg 1940:26:521-547.
Breitner C. Further investigations of bone changes resulting from experi-
mental Orthodontic treatment. Am J Orthod 1941:27:605-632.
Burkhardt DR, McNamara JA Jr, Baccetti T. Maxillary molar distaliza-
tion or mandibular enhancement: A cephalometric comparison of the
Pendulum and Herbst appliances. Am J Orthod Dentofacial Orthop
2003;123:108-116.
Dischinger TG. Edgewise bioprogressive Herbst appliance. J Clin Orthod
1989:23:608–617.
Elgoyhen JC, Moyers RE, McNamara JA Jr, Riolo ML. Craniofacial adap-
tation of protrusive function in young rhesus monkeys. Am J Orthod
1972;62:469-480.
Hamilton DC. Personal communication. 2001.
Herbst E. Atlas und Grundriss der Zahnārztlichen Orthopādie. Munich,
JF Lehmann Verlag, 1910.
Hiniker J, Ramfjord S. Anterior displacement of the mandible in adult
rhesus monkeys. J Pros Dent 1966; 16:503–512.
Hurme V, Van Wagenen G. Basic data on the emergence of deciduous
teeth in the monkey (Macaca mulatta). Proc Am Philo Soc 1953;97:
29.1-315.
Hurme V, Van Wagenen G. Basic data on the emergence of permanent
teeth in the rhesus monkey (Macaca mulatta). Proc Am Philo Soc
1961;105:446-457.
Joho J-P Changes in form and size of the mandible in the orthopaedically
treated Macacus Irus: An experimental study. Trans Eur Orthod Soc
1968;44:161-173.
McNamara JA Jr. Neuromuscular and Skeletal Adaptations to Altered
Orofacial Function. Craniofacial Growth Series, Center for Human
Growth and Development, The University of Michigan, Ann Arbor
1972; 1:1-184.
219
Experimental Herbst
McNamara JA Jr. Neuromuscular and skeletal adaptations to altered func-
tion in the orofacial region. Am J Orthod 1973;64:578-606.
McNamara JA Jr. Fabrication of the acrylic splint Herbst appliance. Am J
Orthod Dentofacial Orthop 1988;94:10-18.
McNamara JA Jr, Brudon WL. Orthodontics and Dentofacial Orthope-
dics. Ann Arbor: Needham Press, 2001.
McNamara JA Jr, Bryan FA. Long-term mandibular adaptations to protru-
sive function: An experimental study in Macaca mulatta. Am J Orthod
Dentofacial Orthop 1987;92:98-108.
McNamara JA Jr, Carlson DS. Quantitative analysis of temporomandib-
ular joint adaptations to protrusive function. Am J Orthod 1979;76:
593-611.
McNamara JA Jr, Graber LW. Mandibular growth in the rhesus monkey
(Macaca mulatta). Am J Phys Anthrop 1975;42:15-24.
McNamara JA Jr, Howe RP, Dischinger TG. A comparison of the Herbst
and Fränkel appliances in the treatment of Class II malocclusion. Am
J Orthod Dentofacial Orthop 1990;98:134-144.
McNamara JA Jr, Peterson JE Jr, Pancherz H. Histologic changes asso-
ciated with the Herbst appliance in adult rhesus monkeys (Macaca
mulatta). Semin Orthod 2003;9:26-40.
McNamara JA Jr, Riolo ML, Enlow DH. Growth of the maxillary com-
plex in the rhesus monkey (Macaca mulatta). Am J Phys Anthrop
1976:44:15-26.
Pancherz H. Treatment of Class II malocclusions by jumping the bite with
the Herbst appliance: A cephalometric investigation. Am J Orthod
1979;76:423-442.
Pancherz H. The effect of continuous bite jumping on the dentofacial com-
plex: A follow-up study after Herbst appliance treatment of Class II
malocclusions. Eur J Orthod 1981;3:49–60.
Pancherz H. The mechanism of Class II correction in Herbst appliance
treatment: A cephalometric investigation. Am J Orthod 1982;82:104-
113.
Pancherz H. The Herbst appliance: Its biologic effects and clinical use.
Am J Orthod 1985;87:1-20.
Pancherz H. The effect, limitations, and long-term dentofacial adaptations
to treatment with the Herbst appliance. Semin Orthod 1997;3:232-
243.
Pancherz H. Dentofacial orthopedics or orthognathic surgery: Is it a matter
of age? Am J Orthod Dentofacial Orthop 2000;117:571-574.
220
Kieu and McNamara
Pancherz H, Ruf S. The Herbst appliance: Research-based updated clinical
possibilities. World J Orthod 2000; 1:17–31.
Pancherz H, von Bremen J. Efficacy of early and late Class II treatment.
In: McNamara JA Jr, Kelly KA, eds. Treatment Timing: Orthodon-
tics in Four Dimensions. Craniofacial Growth Series, Department of
Orthodontics and Pediatric Dentistry and Center for Human Growth
and Development, The University of Michigan, Ann Arbor, 2002;
39:25–53.
Peterson JE Jr, McNamara JA Jr. Temporomandibular joint adaptations as-
sociated with Herbst appliance treatment in juvenile rhesus monkeys
(Macaca mulatta). Semin Orthod 2003;9:12-25.
Petrovic A, Stutzmann JJ, Oudet C. Control process in the postnatal growth
of the condylar cartilage. In: McNamara JA Jr, ed. Determinants of
Mandibular Form and Growth. Craniofacial Growth Series, Center
for Human Growth and Development, The University of Michigan,
Ann Arbor 1975;4:101-153.
Ramfjord S, Enlow R. Anterior displacement of the mandible in adult rhe-
sus monkeys: Long term observations. J Pros Dent 1971:26:517-531.
Ruf S, Pancherz H. Temporomandibular joint growth adaptation in Herbst
treatment: A prospective magnetic resonance imaging and cephalo-
metric roentgenographic study. Eur J Orthod 1998:20:375-388.
Ruf S, Pancherz H. Temporomandibular joint remodeling in adolescents
and young adults during Herbst treatment: A prospective longitudinal
magnetic resonance imaging and cephalometric radiographic investi-
gation. Am J Orthod Dentofacial Orthop 1999a;115:607-618.
RufS, Pancherz H. Dentoskeletal effects and facial profile changes in young
adults treated with the Herbst appliance. Angle Orthod 1999b;69:239-
246.
Ruf S, Pancherz H. When is the ideal period for Herbst therapy - early or
late? Semin Orthod 2003;9:47-56.
Schaefer AT, McNamara JA Jr, Franchi L, Baccetti T. A cephalometric
comparison of two-phase treatment with the Twin Block and stainless
steel crown Herbst appliances followed by fixed appliance therapy.
Am J Orthod Dentofacial Orthop 2004;126:7-15.
Stöckli PW, Willert HG. Tissue reactions in the temporomandibular joint
resulting from anterior displacement of the mandible in the monkey.
Am J Orthod 1971;60: 142-155.
221
Experimental Herbst
Wieslander L. Intensive treatment of severe Class II malocclusions with a
headgear-Herbst appliance in the early mixed dentition. Am J Orthod
1984;86:1–13.
Wieslander L. Long-term effect of treatment with the headgear-Herbst ap-
pliance in the early mixed dentition. Stability or relapse? Am J Orthod
Dentofacial Orthop 1993; 104:319-329.
Windmiller EC. The acrylic-splint Herbst appliance: A cephalometric
evaluation. Am J Orthod Dentofacial Orthop 1993; 104:73-84.
Woodside DG, Altuna G, Harvold E. Herbert M, Metaxas A. Primate
experiments in malocclusion and bone induction. Am J Orthod
1983;83:460-468.
Woodside DG, Metaxas A, Altuna G. The influence of functional appli-
ance therapy on glenoid fossa remodeling. Am J Orthod Dentofacial
Orthop 1987;92:181-98.
222
FOLLOW-UP STUDY;
LONG-TERM OCCLUSAL STABILITY IN 120 CASES
Proposed Treatment Goals
Etsuko Kondo
Post-orthodontic occlusal stability clearly is one of the most important
factors to consider when treating malocclusions and maintaining a good
orthodontist-patient relationship. It is difficult to predict, however, if the
occlusion provided at the end of active treatment will be harmonious with
the patient’s biologic system and conducive to long-term occlusal stability.
It is an important challenge for orthodontists to achieve long-term occlusal
stability with smooth mandibular movement in addition to successful cor-
rection of malocclusions. What type of occlusion needs to be established
at the end of active treatment for this purpose? How does the occlusion
change with post-treatment maxillomandibular growth and jaw movement
to get to the point where the occlusion remains stable in the long term?
One possible way to answer these questions is to search for occlusal fea-
tures common to functionally and esthetically stable cases. A study thus
was conducted of 120 subjects whose active treatment was finished during
the period of maxillomandibular growth or mandibular growth and who
maintained occlusal stability for more than five years post-treatment. In
all subjects, the occlusion achieved at the end of active treatment contin-
ued to improve with post-treatment growth and jaw movement and ap-
proached the occlusal status of control subjects with normal occlusions.
MATERIALS AND METHODS
Subjects
This study included 120 post-treatment orthodontic patients
(31 male, 89 female) whose active treatment finished during the period
of maxillomandibular growth or mandibular growth and whose stable
occlusion was maintained for five years or longer (up to 38 years). All
of the patients were treated with orthodontics and myofunctional train-
ing (MFT), had undergone no surgical procedures, and had worn retain-
ers and a tooth positioner for two to three years. Mandibular growth was
completed post-retention in all subjects. The study group consisted of 42
Class I cases (32 extraction, 10 nonextraction), 35 Class II, division 1
223
Long-term Occlusal Stability
cases (28 extraction, 7 nonextraction), 5 Class II, division 2 cases (3 ex-
traction, 2 nonextraction) and 38 Class III cases (20 extraction, 18 nonex-
traction; Table 1).
The study also included a control group of 22 non-growing indi-
viduals (10 male, 12 female) with normal occlusions: a Class I molar and
canine relationship, normal overjet and overbite, no crowding in either the
maxillary or mandibular arch, and no temporomandibular condition (Table
1). Table 2 shows mean ages of the study and control groups pretreatment,
at the end of active treatment and post-retention.
Table 1. Angle classification of 120 subjects with long-term occlusal stability
and 22 control subjects with normal occlusions.

tºy Sex Extraction Non-extraction Total
Subjects distinction &
º: Total Male Female | Total Male Female | Total Male | Female
I 32 9 23 10 3 7 42 12 30
Study
subjects | II divl 28 6 22 7 2 5 35 8 27
with
long-term
occlusal II divº 3 I 2 2 0 2 5 l 4
stability
III 20 5 15 18 5 13 38 10 28
total || "I'div; 33 21 62 37 10 27 120 31 89
II div2+III
.
subjects
normal 22 10 12
occlusion
There was no significant gender difference (shrunk test and general linear mixed model), and there
was no significant differences between extraction and nonextraction (Welch's test) in the study
Sample.
Methods
The study evaluated lateral, PA and axial cephalograms, panoramic
and joint radiographs, lateral and frontal photographs of the upper quarter
of the body, intraoral photographs, Sagittal cross sections of dental casts
and electromyographic (EMG) recordings taken pretreatment, at the end
of treatment, and 5 to 38 years out-of-retention; post-retention CT and
3D-CT images also were taken. These records were used to make the fol-
lowing measurements:
224
Kondo
Table 2. The mean value of age in the control subjects and study subjects at pre-
treatment, post-treatment and out-of-retention.
extraction statistic
Normal
Angle Class I
Angle Class Il div. 1
MEAN YM
Angle Class ll div. 2
Angle Class

225
Long-term Occlusal Stability
A. Angular measurements in degrees (Fig. 1)
1.
2.
6.
SNA, SNB and ANB
Angles that the functional occlusal plane (F.Occp) makes
with the following:
(a) AB plane (F.Occp-to-AB plane)
(b) Axis of the maxillary first molar (F.Occp-to-UM.)
(c) Axis of the mandibular first molar (F.Occp-to-LM)
Angles that the mandibular incisor axis (1) makes with
the following:
(a) DC-Lli line (Kondo, 1998) (a line connecting the
midpoint of the condylar (DC) and the lower incisal
edge (Lli) (1 to DC-Lli line angle)
(b) GoMe plane (1 to GoMe plane angle)
Axis of the maxillary incisor (1) to SN plane (1 to SN
plane angle)
SN to GoMe plane
Palatal plane (Pal.P) to GoMe plane
Angles that SN plane makes with the following:
(a) F line (relationship of the connecting the point of F2
and Bp that represents the lingual surface of the upper
incisors; SN to F line angle) -
(b) CDM line (this line is used to represent the inclina-
tion of the anterior wall of the glenoid fossa—used for
condylar guidance; SN to CDM line)
(c) Parallelism between F line and CDM line
B. Linear measurements in millimeters
1.
Overjet and overbite
2. The proportion of study subjects with extensions of the
functional occlusal plane passing through the mandibular
foramen (F.Occp on mandibular foramen)
Position of the apex of the dens to McGregor line
(McGregor, 1948; Kondo, 1991, 2002). A positive
value was assigned when the apex of the dens was
226
Kondo
above the McGregor line, and a negative value was as-
signed when the apex was below the line
C. Others
1. Evaluation of the morphology of the condyles and rami
on panoramic and standardized radiographs and 3D-CT
images of the joints post-retention (Figs. 1 and 2)
2. EMG of masticatory muscles
3. Mean values of the study subjects compared with those of
control subjects (95% confidence limits)
1 to SN
• position of the measurement
1. Functional occlusal plane angle (FOccp to AB)
2-1 to SN Plane angle (1 to SN)
3. Fline to SN plane angle (Fline to SN)
4. CDM line to SN plane angle (CDM line to SN)
5. to DC-Lii line angle (1 to DC-Lii line)
6. Functional occlusal plane to the tooth axis of upper and lower posterior teeth
7. Position of the apex of the dens to the McGregor line
8. The extension of FOcclusal plane pass on the mandibular foramen
9. Overjet and Overbite
Figure 1. The values measured on the lateral cephalogram for the patients and
the control subjects. Patient measurements were taken pre- and post-treat-
ment and again when the patient was out-of-retention.

227
Long-term Occlusal Stability
Sagittal
R
CDM line
Oblique
Figure 2. CT and 3D images and joint radiographs taken 16 years post-treatment
for a Class III open-bite patient treated without extraction or surgical therapy. The
patient was 33 years and 3 months old.
D. Statistical analysis
1. Welch's test, general linear mixed model and Dunnett's
test were performed to determine the significance level
of the differences observed between the study and control
groups pretreatment, at the end of treatment and post-re-
tention (Marcello and Kimberlee, 2003).

228
Kondo
3D image
In Occlusion In Opening
42,0mm opening
CT image of curved MPR
from posterior
Joint radiograph
Figure 2. Continued.
2. Gender differences were analyzed with shrunk test. (Figs.
3 and 4).
3. Pearson’s correlation coefficients between F line and
CDM line were calculated (Matsubara et al., 1991; Ya-
jima et al., 1992; Jason, 1996).

229
Long-term Occlusal Stability
A. Pre-treatment (12Y1M) ºn ºf ºne |
Figure 3. This figure illustrates treatment of Class III subjects with posterior
occlusal vertical dimension problems. A (top): Skeletal Class III malocclu-
sion with deep bite. These patients present with short posterior occlusal verti-
cal dimension and excessive chewing muscle activity. B (bottom): Skeletal
Class III malocclusion with open-bite. These patients present with a exces-
sive posterior occlusal vertical dimension and less chewing muscle activity
pretreatment.
T
Pre-treatment 16 years 3 mºnº pºst-treatment
(1477M) - º º


230
Pre-treatment 5 years 5 months pºst-treatment
12 YIM º
Figure 3. Continued. C. Skeletal Class III malocclusion with lateral devia-
tion. These patients present with an asymmetrical posterior occlusal vertical
dimension and asymmetrical chewing muscle activity.
- 10 years -
Pre-treatment Dºost-treatment ºn. pºst ºn
(*Y 9M) ( 12Y 4M) ( 22W 4M sº º
Figure 4. Occlusal change from pretreatment to post-retention for a Class II,
division patient whose treatment included extraction and whose long-term
occlusal stability had lasted for more than 38 years.


231
Long-term Occlusal Stability
FINDINGS
In all study subjects with long-term occlusal stability, the occlu-
sion established at the end of active treatment continued to improve with
post-treatment growth and jaw movement and came infinitely closer to
the goals of normal occlusion by the end of growth, leading to long-term
occlusal stability. Evaluations were made to determine the occlusal and
facial features obtained at the end of active treatment and the occlusal
changes observed post-treatment.
Facial Findings (Figs. 5-6)
Facial profiles became well-balanced in all study subjects by the
end of active treatment, as abnormal behaviors of the tongue and of the
perioral and chewing muscles were eliminated. In addition, all study sub-
jects acquired the ability to breathe through their nose with lips closed
eliminating the need to mouth-breathe, which contributed to further im-
provement in facial profile during the period of post-treatment maxillo-
mandibular growth. Frontally, the face appeared symmetrical in all study
subjects in the post-retention period.
EMG Findings (Fig. 3)
• In subjects with excessive pretreatment chewing muscle activ-
ity and short posterior occlusal vertical height, muscle activity decreased
and occlusal vertical height increased by the end of active treatment (Fig.
3A).
• In subjects with less pretreatment chewing muscle activity and
excessive posterior occlusal vertical height, muscle activity increased and
occlusal vertical height decreased by the end of active treatment (Fig.
3B).
• In subjects with asymmetrical pretreatment chewing muscle
activity and bilateral posterior occlusal vertical height, symmetry was re-
stored by the end of active treatment (Fig. 3C).
• For all study subjects, nose breathing and harmonious activ-
ity of the tongue and of the perioral, chewing and cervical muscles were
achieved by the end of treatment regardless of pretreatment muscle sta-
tuS.
• In all study subjects, the tooth supporting dentoalveolar struc-
tures had great adaptive capacity and could be expanded and/or reshaped
to restore harmony to the function of the tongue and perioral muscles
and provide adequate airway for nose breathing (Figs. 2, 6-8). All teeth
232
Kondo
ſaeſ
|-|
|× ſ）|（…）-ſ)|- ſkiſ,
|-ſaeſ（T)| （）ſ ſ |ſºſ|
traction cases
None
whose occlusion was stable long term.
division 1 and division 2; and Class III
Class I nonextraction cases (left side) and extraction cases (right side).
º
º
Class II
Patients, taken from the 120 study subjects
º
Figure 5. Representative Class I;


233
Long-term Occlusal Stability
Nonextraction
-| ſae
| （）
|× … ：-）
|--1. Mae
| __|
---- ___ .№ſ),
（） （）- ſ.
（） （） （ſ.ſºlſ).
|-|-- |-|-|-|-
|-_ _
--. ）
-| ſ ||
|-- （ ）
|×ſae） --
|№ | k ||
|-.|--
|- （ ）
№ſiſ （).№ſſae|-
№.
- .| ºſ：
ſ.
----
|-
№raeſ) |
|№ſºſ,
Nonextraction
(left side); extract
and Class II, division 2 nonextraction cases (right side).
division 1 nonextraction cases
º
II
º
Figure 5. Continued. Class
tion cases (middle);

234
Kondo
№raeſ）
№
ſ ſ
№.
№ ：|- .
|
|----- │ │-
ſ.№ſſae|-
-| （）ſae
， ！！!!！！！！！!!！
-ſ ºſſ （; ſ. №
（
ſ. ）,
.
|×
|3
ſae
Extraction cases
.… ：（. .
-： №
（ſ.|-
-- ：）| | ||×|
№， №ſº （ ）ſae ae
） _） ：-） º laeſ （~|--）） ）- |- |-│ │ │
·|× （ ）№ſae| （， ， ， ，
|- ſae|-||- | ſ || （! ！！ !！
： ~ ¡ ¿ ：| ±（+)，
|--
|
：
º
ſae
|-
E-
|-
№
： ~~~~
|-
|-
№
， !
·
E
|-
， !
|-
·
ſiſ|×
|-№.
|-- -
|- |.… ：）-
ºſſ ||№ſºſ||ºſſ||
-- |-. - |-|-| ||
Figure 5. Continued. Class III nonextraction and extraction cases.

235
Long-term Occlusal Stability
Pretreatment Post-treatment 2y Post-treatment 10y Post-treatment 38y Post-treatment
(9) 9m) (12y 4m) (14y 4m) (22y 4m) (50y 4m)
Figure 6. A representative Class II, division 1 patient whose occlusion remained
extremely stable for more than 38 years post-retention.
were well-aligned adequate labiolingual and buccolingual torque within
the alveolar cancellous bone, conforming to the shape of a relaxed tongue
without broken contacts (Figs. 2 and 8).
• As chewing muscle activity was normalized, adequate posterior
occlusal vertical height was established resulting in stable intercuspation
of the canines and posterior teeth by the end of treatment (Figs. 3, 6-7).



236
Kondo
-Lii line angle
nº-ºst."
— Pre - treatment (979M) .
- Post - treatment (12Y4M)
Post-treatment (12Y4M)
3 months after retention (12Y7M)
2 years after retention (14Y4M)
10 years after treatment (22Y4M)
38 years after treatment (50Y4M)
Figure 6. Continued.



237
Long-term Occlusal Stability
Pretreatment 6m into active - Post-treatment 10y Tom Post- 16y 3m Post-
(14y 7m) treatment (15y lm) (16) 10m) treatment (27y 8m) treatment (33y 3m)
Figure 7. A representative Class III open-bite patient whose occlusion remained
extremely stable for more than 16 years post-retention.
• As chewing muscle activity was normalized, the maxillary inci-
Sors continued to tip labially until the edge of the mandibular incisor ap-
proached the inflection point (Bp) on the lingual surface of the maxillary
incisor. The overjet and overbite thus were increased (Figs, 3-7).
• As chewing muscle activity was normalized, maxillary poste-
rior teeth were tipped buccally at the end of treatment (Fig. 4).


238
Kondo
57mm
70 mm
DC-Lii line to T
N 110.0 – 102.0"
- Pre-treatment (14Y 7M)
– 6 months into active treatment (15Y IM)
Post-treatment (16Y10M)
2)
K
DC-Lii line to 1
1020” – 940 – 940
gonial angle
1306 – 1240"
- AB to FOcp 90'
• SN to GoMe angle
36.5" -> 330°
* I to DC-Liangle
102.0° - 940
* T to Gome angle
81.8" – 890"
— post-treatment (16X 10M)
– 3 years post-treatment (19)(10M)
– 16 years 3 months post-treatment (33X 3M)
Figure 7. Continued.







239
Long-term Occlusal Stability
Oblique
Figure 8. CT and 3D images and joint radiographs of a Class II, division
| extraction case taken 38 years post-retention. The patient was 50 years
and 4 months old.
Lateral cephalometric findings
ANB Angle (Fig. 9, Table 3)
• In the all of the study subjects except for the Class III study sub-
jects, the ANB angle was within one standard deviation (SD) of the mean
of the control subjects at the end of active treatment. There was further
improvement with growth post-treatment, which brought the ANB angle
to within the 95% confidence limits of the mean of the control subjects.

240
Kondo
- -
3D image
CT image of curved MPR Joint radiograph
from posterior
Figure 8. Continued.
• No significant differences were found in the mean ANB angle
between the study and control subjects except for Class II, division 1 ex-
traction subjects either at the at the end of active treatment or post-reten-
tion (Dunnett's test).
• The post-retention mean value of Class II, division 1 extraction
Cases (4.0°) was significantly different from that of the control group at
0.01 < P = 0.05, but it was within one SD of the mean of the control sub-
jects (2.7°,+ 1.54°).
• The maxillomandibular relationship was harmonized in the all
study subjects by the end of active treatment and continued to improve
With post-treatment maxillomandibular growth (Figs. 5 and 6).

241
Long-term Occlusal Stability
ANB Angle Class I ANB Angle Class II, division 1
130
120
110
100
90
80
º
Tºp 70
-
ºf 60
: so
º
-
ºn-
º -
< 130 130
3 120- 120-
C.
3 *- 110-
© 100- - 100-
º
90- D 90-
80- 80-
70- Age Mean 70- - age Mean
toyom Il pre 100m
13x3m 60- post 13yim
20x9m Normal L. R. 23y2m
-T-I-T- 50 i I F-
-10 -5 0. 5 10 15 -10 -5 0. s 10 15
ANB Angle
Figure 9. Change of ANB angle and F.Occp-to-AB plane angle from pretreat-
ment to post-retention. Top: Nonextraction cases; Bottom: Extraction cases.
Pre: pretreatment; Post: post-treatment; Re: retention; CLM confidence lim-
its; 95% LCLM: 95% lower confidence limits of mean value of control subjects;
95% UCLM: 95% upper confidence limits of mean value of control subjects.
Functional Occlusal Plane to AB Plane Angle (FOccp-to-AB
Plane Angle, Fig. 3, Table 3)
• At the end of active treatment, the mean F.Occp-to-AB plane
angles were 87.5° and 86.7° for Class I nonextraction and extraction
cases, respectively, 88.4° and 88.9° for Class II, division 1 nonextraction
and extraction cases, respectively: 87.0° and 89.3° for Class II, division
2 nonextraction and extraction cases, respectively; and 83.8° and 84.3
for Class III nonextraction and extraction cases, respectively. The mean




242
Kondo
ANB Angle Class II, division 2 ANB Angle Class III
130
120
110
100
90
80
70
60 |- Normal
meanssºci.
50
-10 -5 0. 5 10 15
130 130
120 120
110 110
100 100
90 90
80 80
70 70
60 60
| Normal
50 50
15 -
ANB Angle
Figure 9. Continued. Top: Nonextraction cases; Bottom: Extraction cases.
-10 -5 0. s 10 10 -5 0. 5 10 15
FOccp-to-AB plane angle in the control subjects was 91.3°, with a 95%
confidence interval of 89.3° to 93.3° and a SD of +4.4°. Thus, the F.Occp-
to-AB plane angle was corrected to within one SD of that of the control
Subjects (91.3°4-44) by the end of active treatment except for the F.Occp-
to-AB plane angle in Class III subjects.
• In all of the study subjects, the F.Occp-to-AB plane angle gradu-
ally moved towards 90° with post-treatment growth. The proportion of
Study subjects with the F.Occp-to-AB plane angle lying within the 95%
Confidence interval of 89.3° to 93.3° increased to more than 70% and
78% for Class I nonextraction and extraction subjects, respectively; 100%
and 89% for Class II, division 1 nonextraction and extraction subjects,
respectively; more than 50% and 100% for Class II, division 2 nonex-
traction and extraction subjects, respectively; and 66.7% and 45.0%



243
Long-term Occlusal Stability
Table 3: Mean value of angular measurement on the lateral cephalograms at pre-
treatment, post-treatment and out-of-retention in the study subjects and control
subjects and incidence of the Angle of F.Occp-to-AB, T-to-DC-Lli, 1 to GoMe
became 90° within 95% CLM mean value of the control subjects. Results were
analyzed statistically with Welch's test, general linear mixed model, Dunnett's test
and Pearson's correlation coefficient. P:AdjP (Dunnett); ***:Ps 0.001; **:0.001
< P × 0.01; *:0.01 < P × 0.05.

Angle - - Control
Item subjects §: Class I Class II div. 1 Class ll div.2 Class III wºn
t.
Ext.
Mean value Time Non Ext. Ext. Non Ext. Ext. Non Ext. Ext. Non Ext. Ext. Non Ext.
Pre
treatment 8Y 10M | 10YOM | 10Y4M 10YOM | 13Y7M 10Y3M 9Y8M 9Y9M
Post 22Y8M
Age Hºlment | 12Y7M 13Y3M 13Y4M 13Y4M 15Y8M 12Y8M | 13Y4M 13Y7M i.
Relention 20Y&M 20Y9M 1971M. 23x2M 21x1M 18Y6M. 21xoM 22Y6M
Pre
--- * .7:E 1.
ANB angle treatment 2.2 39 . 69 ... 7.0...| 1.5 62 ..] -3.2...] -12...] 27+134
(degree) Post &
treatment 2.8 3.2 3.7 46 .. 1.0 4.2 0.7 . i.2 º
AdjP (Dunnett) . . º.º.
Retention 3.1 2.9 3.2 40 . 1.0 4.5 1.0 f.4
Pre
Foccp to AB L treatment 827... 817...] 89.1 86.1...] 84.5 88.5 749...] 74.5...|913+442
angle (degree) | Post
treatment 87.5 867...] 88.4 88.9 87.0 89.3 838...| 843...| 95%CLU
AdjP (Dunnett) 89.3-93.3
Retention 88.6 88.7 89.7 89.6 87.5 89.7 87.7 85.6.
Incident Rate : Pre 0.0 1.0 0.0 14.3 0 0 0 0
FOccp to AB angle treatment 20.0 9.4 71.4 28.6 0 33.3 5.6 0 22.7
was 90 Post 10.0 25.0 57.1 42.9 0 66.7 16.7 10.0
1. treatment 70.0 50.0 71.4 7 H.4 50 100 16.7 25.0
%9E | Retention |300 59.4 85.7 71.4 0 66.7 38.9 20.0 59,1
(89.3°- 93.3°) 70,0 78.1 100 89.3 50 100 66.7 45.0
Pre
I to DC-Lli line treatment 84.4 822...] 83.6 .. 77.1...| 1080. 95.3 89.9 87.2 88.9-E3.67
*** | *... sº 90.3 86.6 89.3 87.5 91.0 92.7 95.4 ;
AdjP (Dunnett) ...jºu.
J Retention 91.0 91.0 88.4 89.8 88.0 92.3 92.6 95.8 .
Incident Rate Pre 0.0 9.4 0.0 7.1 0 33.3 0 5.0
I to DC-Lli line treatment 28.6 15.6 28.6 7.1 33.3 22.2 25.0 || 13.6
angle was 90 Post 42.9 53.1 28.6 57.1 50.0 66.7 44.4 25.0
1. treatment 60.0 65.6 57.1 60.7 50,0 66.7 50.0 25.0
%9 | Retention |574 75.0 57.1 T1.4 50.0 66.7 61.1 35.0 54.5
(87.3°-90.5°) 71.4 81.3 85.7 82. I 50.0 100 66.7 35,0
Pre
1 to GoMe treatment 93.7 9 92.3 1008. 785... 820...] 88.1 .. 889 |939+3.87
*** | *. 90.0 86.2 93.4 90.7 99.0 86.7 87.4 80.9 || 95%CLM
ent # * * :: *::: ***| 92.2-95.6
AdjP (Dunnett) e e
Retention 91.3 89.0 92.4 91.8 98.0 90.0 88.6 84.6
# * * * * * *
Pre
1 to SN treatment | 104.4 106.5 107.4 1116,. 87.5... 845... 102.1 105.8 || 106,4:E5.08
*** | *... loss 997.| 1028 975...] 109.0 | 101.5 1120 ... 105.1 || 95%CIM
104.2-106.4
Adip (Dunnett
** | Retention | 110s 106.3 106.1 104.5 111.5 102.3 113.8 ... 1102
Pre
SN to F line treatment | 132.1 , || 1305...] 129.1 1363...] 115.5 ... 105.3...] 124.1 1292. 1234+3.50
*** | ..., | 129s 122.5 126.1 119.1. 1300 120.8 1303. 1285 || 95%CLM
121.9-125,0
AdjP (Dunnett
etention & 128. * & I 32.5 120. s -
P(Punnet) seems 1315. 1283. 1259 126.5 32 20.0 133.5., 1326...
Post-treatment: end of active treatment; Correlation: Pearson’s; Differences: Pearson’s; CLM:
confidence limits; 95% LCLM: 95% lower confidence limits of mean value of control subjects;
95% UCLM: 95% upper confidence limits of mean value of control subjects; UM1: Axis of maxil-
lary first molar; LM1: Axis of mandibular first molar.
244
Kondo
Table 3. Continued.

Angle - - Control
Item sample §§ Class I Class II div. 1 Class II div.2 Class III Mºšn
Ext.
º: Time | Non Ext. Ext. Non Ext. Ext. Non Ext. || Ext. | Non Ext. || Ext. | Non Ext.
Pre
e & s & e - & 120.7+3.23
SN to CDM line treatment 1296. 1298...] 1160 1278. 1145 | 16.7 1387...] 1322...
angle (degree) || Post 130.8 128.6 123.3 124.9 119.5 122.2 133.2 132.6 95%CLM
treatment * * # * * * * * # * *
- 118.4-123.0
AdjP (Dunnett) -
Retention 131.2 128.8 125.0 127.2 120.0 120.3 132.9 133.5
4 × # * * $: *::::ſ: ::::::::
Pre -
Incident Rate treatment | Ns, 99 ss 94 |ss 00 |ss 14.3 ss. 0.0 ss 33.3 Ns 5.6 Ns 50 Correlation
F line P --- --- F line and
OSt CDM line
0.0 .3 28.6 10.7 0.0 33.3 27.8 10.0
ãº. treatment P=0.005 Pºſ),001 P-0.002 P-0.001 || N.S. N.S. P=0.010 | N.S. parallel
% Retenti - - e p g º º e (Pearson')
(%) €tent lon ...80 P-0001 ** ..."ºw .."ºw. ss 00 ss 66.7 ...” .."ºw P-0.001
Pre -j-
sno come treatment | * 39.2 . 352 390 . 330 38.5 35.9 38.7 . .349+3.33
angle (degree) Post
treatment 34.2 39.6 . 36.9 38.6 27.0..] 380 33.8 38.8 . 95%CLM
- 33.4-36.4
AdjP (Dunnett) s
Retention 32.7 37.9 37.2 36.9 29.0 39.0 32.9 37.3
Pre 26.4+4.44
Palp to GoMe treatment 26.2 29.2 26.8 29.4 16.5 . 25.3 25.7 297.
angle (degree
gle (degree) *mem 26.7 29.1 25.9 29.5 17.0 24.0 26.1 30.4 95%CLM
:: x:
AdjP (Dunnett) 24.4-28.3
Retention 24.2 27.2 27.5 28.5 170 . 28.8 25.9 29.3
Incident Rate Pre 30.0 53.1 14.3 50.0 0.0 33.3 50.0 35.0
UM! to treatment 10.0 6.3 0.0 12.5 0.0 33.3 22.2 10.0
Poºpangle Rate Post 100.0 100.0 85.7 87.5 100.0 100.0 100.0 100.0
was 90%/ LM1 to treatment 100.0 100.0 85.7 87.5 100.0 100.0 4 90.0
(%) º: Retention |1999 100.0 85.7 100.0 100.0 100.0 100.0 100.0
angle was 90 (%) 100.0 85.7 93.8 100.0 100.0 100.0 95.0
Incident Rate *mem 10.0 62.5 0.0 21.4 0.0 33.3 44,4 55.0
FOccp Post
on Mand *ment 40.0 71.9 42.9 51.9 50.0 66.7 44.4 35.0 63.6
Foramen -
(%) Retention | 100.0 98.4 100.0 98.0 100.0 100.0 61.1 45.0
ºforming Pre 60.0 60.0 0 61.5 100.0 0 40.0 26.3
through the Mand. treatment 100.0 100.0 85.7 100.0 100.0 66.7 75.0 55.0
Foram 100
Yºu. Post 80.0 60.0 0 65.4 100.0 0 40.0 25.0
wn. Hºt 100.0 100.0 85.7 100.0 100.0 66.7 85.0 70.0
a &.
"/www.merrine Retention |800 66.7 0 73.1 100.0 0 35.0 30.0 100
100.0 100.0 85.7 100.0 100.0 66.7 90.0 80.0 -
Pre
Overjet treatment 0.5 :::it 2.3 9.9 :::::: 6.6 :::::: 0.0 *::: 0.0 ::::::: 3.2 sº it it 1.7 *:::: 2.6+0.97
(mm) *iment 20 2.0 2.4 1.8 1.8 i.7 2.0 1.9 º
AdjP(Dunnett) - º ºg º' tº
Retention 2.6 2.7 2.9 2.7 2.5 2.8 2.6 2.3
Pre 2.3 2.3 6.7 3.6 7.5 6.7 2.5 1.7 2.6+0.91
Overbite treatment x: x <xt *:::::: :::::::
(mm) Post
treatment 2.6 2.6 2.6 2.3 2.3 2.8 2.6 2.7 95%CLM
& 2.2-3.0
AdjP(Dunnett) *
Retention 3.1 2.9 2.8 2.9 2.8 3.3 2,8 2.8
for Class III nonextraction and extraction subjects, respectively, in the
post-retention period.
• As a result, there was no significant difference in the mean
F.Occp-to-AB plane angle between the study and control subjects in the
post-retention period except in Class III extraction cases (Dunnett's test:
0.001 < P × 0.01).
245

Long-term Occlusal Stability
Angle Class I Angle Class II, division 1
130- 130-
120- 120-
110 - 110-
D
100- D 100- BH.
L D
90- º 90-
80- D ſº 80-
70- 70-
Age Mean Age Mean
60- D Pre 8y10m 60- D. Pre 10yºm
D. Post | 12Yºm L Post 13y4m
50- - Re 20y8m 50- - re 19yim
º i I i i I i i i t i i i I I i i i
> 50 60 70 80 90 100 110 120 130 50 60 70 80 90 100 110 120 130
5
->
* 130- 130-
| wº
120- D 120- PE
I 10- 110- {
D
100- 100- D
90- 90- D D
D
80- 80-
D D EE
70- D 70-
Age Mican Age Mean
60- C Pre 10x0m 60- D. Pre 10x0m
| | post 1353m L. Post 13yam
50- - Re 20yºm 50- - Re 23x2m
i I I i I I i i i i I i i i i I
50 60 70 80 90 100 110 120 130 50 60 70 80 90 100 110 120 130
1 to DC L1i
Figure 10. Change of i-to-DC-Lli line angle and i-to GoMe plane angle from
pretreatment to post-retention. I-to-DC-Lli line approaches 90° more clearly than
1-to Golve line. Top: Nonextraction cases; Bottom: Extraction cases. Pre: pre-
treatment; Post: post-treatment; Re; retention; CLM confidence limits; 95%
LCLM: 95% lower confidence limits of mean value of control subjects; 95%
UCLM 95% upper confidence limits of mean value of control subjects.
1-to-DC-Lli Line Angle (Fig. 10, Table 3)
• The mean value of i-to-DC-Lli line angle for the study sub-
jects at the end of active treatment was 89.9° and 90.3° for Class I non-
extraction and extraction subjects, respectively; 86.6° and 89.3° for Class
II, division 1 nonextraction and extraction subjects, respectively; 87.5°
and 91.0° for Class II, division 2 nonextraction and extraction subjects,
respectively; and 92.7° and 95.4° for Class III nonextraction and extrac-
246
Kondo
Angle Class II, division II Angle Class III
130- 130-
120- 120-
110- 110-
D
100- ſºlº 100- D
90- 90-
D
80- 80- D
70- D | 70-
60- L Pre 60- º Prº-
I L
50- - im 50-
º i i i i i i i i i i i i I I i i i i
> 50 60 70 80 90 100 110 120 130 50 60 70 80 90 100 110 120 130
§
º
- 130- Age Mican
| wº D Pre 9y.9m
120- D. Post 13yºm
- Re 22yöm
º - *
D 70- [t]] #F
D 60- D
Fº 50-
I i i I i i i i
50 60 70 80 90 100 110 120 130 50 60 70 80 90 100 110 120 130
1 to DC L1 i
Figure 10. Continued. Top: Nonextraction cases; Bottom: Extraction cases.
tion subjects, respectively. The mean value for control subjects was 88.9°
+ SD of 3.679 with a 95% confidence interval of 87.3° to 90.5°.
• Thus, T-to-DC- Lli line angle was within the 95% upper confi-
dence limit (90.5°) for the control subjects at the end of active treatment
ºxcept for Class III extraction subjects.
• The proportion of the study subjects with the 1-to-DC- Lli line
angle at 90° or within 95% UCLM of 90.5° at the end of active treatment
Was 42% (60%) and 53% (65%) for Class I nonextraction and extraction
subjects, respectively; 28% (57%) and 57% (60%) for Class II, division 1
"OneXtraction and extraction subjects, respectively; 50% (50%) and 66.7%
(66.7%) for Class II, division 2 nonextraction and extraction subjects, re-
Spectively; and 44% (50%) and 25% (25%) for Class III nonextraction and
*traction subjects, respectively.







247
Long-term Occlusal Stability
• At the end of active treatment, the mean value of the 1-to-DC-
Lli line angle remained almost unchanged throughout the follow-up pe-
riod in all of the study subjects regardless of growth status. Dunnett's test
revealed no significant difference in the mean T-to-DC-Lli angle between
the study and control subjects at the end of active treatment or post-reten-
t1On.
I-to-Go Me (Fig. 10, Table 3)
Significant differences were found in the mean value of T-to-
GoMe angle between the control subjects and Class I extraction; Class II,
division 1 extraction; and Class III nonextraction and extraction subjects
at the end of active treatment and post-retention.
1-to-SNAngle (Figs. 4, 6–7, 11; Table 3)
• 1-to-SN angle was not within the 95% confidence limits for the
mean value of the control subjects at the end of active treatment except
for Class I and Class III extraction subjects. However, all study subjects
showed labial tipping of the maxillary incisor, which brought the lower in-
cisor edge into contact with or extremely close to the inflection point (Bp)
on the lingual surface of the maxillary incisor by the end of mandibular
growth. 1-to-SN plane angle continued to increase as the maxillary inci-
sors tipped labially with post-treatment mandibular growth, approaching
the 95% confidence limits for the mean value of the control subjects post-
retention in Class II, division 1 nonextraction and extraction subjects.
• There were no significant differences in the post-retention 1-to-
SN plane angle between the study and control subjects except in Class III
nonextraction subjects (Dunnett’s test).
F Line and CDM Line (Fig. 12, Table 3)
• In all of the groups, the proportion of the study subjects with
F line and CDM line parallel to each other was very low at the end of
active treatment except in Class III extraction subjects. However, the la-
bial inclination of the maxillary incisor increased significantly by the end
of post-treatment growth, possibly due to protrusive mandibular move-
ment. This caused gradual changes in F line, and CDM and F line be-
came parallel in 80% and 84% of Class I nonextraction and extraction
subjects, respectively; 71% and 71% of Class II, division 1 nonextraction
and extraction subjects, respectively; and 38% and 70% of Class III non-
248
Kondo
1 to SN angle
cºlas I_Nonextractiºn
e class 1 Extraction
eclass it div. Nonextractiºn
e class II divºl Extraction
T
- - - - Normal 95% CL
Pre-treatment Post-treatment Retention
Figure 11. Change of 1-to-SN angle from pretreatment to post-retention. ***: P
s 0.001; **: < P × 0.01; *: 0.01 < P × 0.05.
extraction and extraction subjects, respectively. This parallelism was not
achieved in Class II, division 2 nonextraction subjects.
• In all study subjects except Class II, division 2, Pearson’s cor-
relation coefficient test revealed a significant correlation between F line
and CDM line post-retention at P × 0.001.
• It is interesting to note that the anterior wall of the glenoid fossa
underwent morphologic changes along with the labial tipping of the max-
illary incisor so that anterior guidance and condylar guidance were harmo-
nized to allow smooth jaw movement.
Functional Occlusal Plane to Axes of Maxillary and Mandibular
First Molars (FOccp-to-UM1 and FOccp-to-LM1, Table 3)
• In both nonextraction and extraction subjects, the axes of the
maxillary and mandibular first molars were perpendicular to the functional
Occlusal plane.
• These relationships were maintained throughout the follow-up
Period regardless of post-treatment growth status.
• The maxillary posterior teeth (except the maxillary second mo-
lar) tipped mesially in concert with the functional occlusal plane changes
during the period of post-treatment mandibular growth to stay perpendicu-
lar to the functional occlusal plane (Figs. 6-7, Table 3).

249
Long-term Occlusal Stability
Angle Class I Angle Class II, division 1
160- ...” 160-
150- º 150-
140- ...” 140- º
D D ºf a
130- DD 130- º
jº, P º PD
120- ...” 120- º D
...” D E. D
110- -- I 10- -
- - Correlation - ..” Age Mean Correlation
100 N.S. 100 .." |D Pre 10x4m N.S.
º - 0.80 (n=10, p=0.005) _j ..." Post 13v.in 0.94 (n-7. p-0.002)
E 90-1. 1.00 (n=10.p:0.001) 90- . º Re 1951m_0.97 (n=7, p<0.001)
º i T-T-T-I-T-T- r-r-ţ-t-t-t-t-t-
> 90 100 110 120 130 140 150 160 90 100 110 120 130 140 150 100
º
Q
3 160- .." 160-
Z --
Qº 150- 150-
140- 140-
130- 130-
120- 120-
D --
110- .." 110-
100- ... Age Mean Correlation 100- ..” A. H correlation
.." H Pre # N.S. .." L Pre 10x0m N.S.
_ _ ! Post|13y3m |0.67 (n=32. p-0.001). _ ..” L. yºn ºn-2s, pººl.00.
90-1. TR Bººm ſnººn–33, pºonſ, 90-1. HºH:
I i I i I i I i I *-i-
90 100 110 120 130 140 150 160 90 100 110 120 130 140 150 160
SN to F Line
Figure 12. Relationships between F line and CDM line. SN-to-F line runs paral-
lel with SN-to-CDM line. Top: Nonextraction cases; Bottom: Extraction cases.
Pre: pretreatment; Post: post-treatment; Re: retention; CLM confidence lim-
its; 95% LCLM: 95% lower confidence limits of mean value of control subjects;
95% UCLM: 95% upper confidence limits of mean value of control subjects.
Overbite (Figs. 4, 6–7, Table 3)
• In all study subjects, the overbite was very close to the 95%
LCLM (2.2 mm) of the control subjects at the end of active treatment.
• The overbite continued to increase post-treatment until maxil.
lary growth stopped except in Class III study subjects. The mean overbite
of the study subjects thus approached the 95% UCLM (3.0 mm) of the
control subjects post-retention.


250
Kondo
Angle Class II, division 2 Angle Class III
160- ...” 160- D
150- º 150-
140- º 140-
130- ...” 130-
..' D-
120- P 120-
110 .* tº 110
... 'D -*
100- ..” Age Mean Correlation 100- ... Age Meanicorrelation
..” D. Pre 13yºm N.S. .." |D Pre 1988m N.s.
º _j ..." T Post 15y&m N.S. _j ..." L. Post 13 y-Im IO-59 (n=18, p=0.010
.E 90-1. - Re （2iyim N.S. 90-1. - Re Elyūm_0.90 (n=1 .001
- i i i I I i i i t -T-I-T-
> 90 100 110 120 130 140 150 160 90 100 110 120 130 140 150 ió0
º
Q
->
* 160- 160-
% -
150- - - .."
..” 150 III]
.." B - D
140- ." 140-
...” D D D
130- --" 130-
gº FI’ Iſſ
120- 120- D ..."
D [… .g.
110 - D. .." 110- .."
100- ...” Age Mean Correlation 100- ...” Age Mean Correlation
.." D Pre 10y3m N.S. ...' Dºre ºm N.S.
90- ..” In post 12ysm N.S. 90- ...” D. Post 1357m N.S.
.." - Re 18wom N.S. - -- Re 22-6m U.92 ſn-20, p-0.001
i -T-I-T-I-T- i i *-******
90 100 110 120 130 140 150 160 90 100 110 120 130 140 150 160
SN to F Line
Figure 12. Continued. Top: Nonextraction cases; Bottom: Extraction cases.
• There was no significant difference in overbite between the
study subjects and the control subjects at the end of active treatment or
p0St-retention.
Overjet (Figs. 4, 6–7, Table 3)
• In all study subjects, the overjet was within the 95% LCLM (2.2
mm) of the control subjects at the end of active treatment.
• During the post-treatment period, the maxillary incisor contin-
ued to tip labially with mandibular growth until the edge of the mandib-
ular incisor came into contact with or extremely close to the inflection
point (Bp) on the lingual surface of the maxillary incisor. The overjet thus
Increased to provide good incisal guidance and freedom of mandibular
In OVement in all directions.

251
Long-term Occlusal Stability
• Consequently, the mean overjet of the study subjects came very
close to the 95% UCLM (3.0 mm) of the control subjects.
• There was no significant difference in overjet between the study
subjects and control subjects at the end of the active treatment or post-re-
tention periods.
Relations in the Extensions of the Functional Occlusal Plane and
the Mandibular Foramen (Fig. 13, Table 3)
• The proportions of study subjects who had an extension of the
functional occlusal plane passing through the mandibular foramen (F.Occp
on the mandibular foramen) at the end of active treatment were 40% and
71% for Class I nonextraction and extraction subjects, respectively; 42%
and 51% for Class II, division 1 nonextraction and extraction subjects, re-
spectively; 50% and 66% for Class II, division 2 nonextraction and extrac-
tion subject, respectively; and 40% and 35% for Class III nonextraction
and extraction subjects, respectively.
• These proportions increased significantly with post-treatment
mandibular growth to 70% and 84% for Class I nonextraction and extrac-
tion subjects, respectively; 84% and 100% for Class II, division 1 nonex-
traction and extraction subjects, respectively; 100% and 100% for Class
II, division 2 nonextraction and extraction subjects, respectively; and 6.1%
and 45% for Class III nonextraction and extraction subjects, respectively.
The proportion of the control subjects with this relationship was 63.6%.
Rate of the Mand. Foramen F. Occp
Angle class I Angle class II div. 1 Angle class II div. 2 Angle class III
i
i
Pre- Post re
Nonextraction
Extraction
Figure 13. The proportion of study subjects with the extension of FOccp passing
through the mandibular foramen. In more than 90% of the study subjects, except
Class III subjects, the extension of F.Occp passed through the mandibular fora-
men at the end of the growth period.

252
Kondo
Position of the Apex of the Dens of the Axis to the McGregor Line
(Table 3)
• All control subjects had the apex of the dens positioned less than
4.0 mm above McGregor line and as did more than 80% of the study sub-
jects at the end of active treatment, except Class II, division 2 and Class III
extraction subjects.
• During the post-treatment period, this proportion increased to
90% and 80% in Class III nonextraction and extraction subjects, respec-
tively.
• No subjects had the apex of the dens more than 6.0 mm above
McGregor line.
• The post-retention PA cephalograms of all of the study subjects
showed no morphologic abnormalities of the cervical spine, condyle or
rami or abnormal inclination of the dens.
DISCUSSION AND CONCLUSIONS
Long-term occlusal stability was achieved in the 120 cases in
which abnormal activities of the tongue; perioral, masticatory, and cervi-
cal muscles; and mouth breathing were corrected with orthodontics and
MFT at an early stage of maxillomandibular growth. The form of occlu-
sion established at the end of active treatment continued to improve with
post-treatment growth and jaw movement and came infinitely closer to
the goals of normal occlusion over time. These findings support the im-
portance of creating an optimal environment in which undisturbed maxil-
lomandibular growth and smooth mandibular movement can take place.
Factors contributing to this objective include good anterior guidance with
proper labial inclination of the maxillary incisors, stable intercuspation
in the canine and posterior area, adequate overjet and overbite, and good
functional balance between the tongue and perioral muscles. Van der Lin-
den (1988) stated that stable intercuspation of posterior teeth favorably
affects condylar growth and anteroposterior growth of the maxilla and
mandible, which was supported by the results of this study. Whenever
an environment conducive to smooth mandibular movement was created
with orthodontic occlusal correction and MFT, subsequent growth and oc-
clusal changes occurred in a harmonious way so that stable intercuspation
of the posterior teeth and smooth anterior guidance were established with
adaptive changes in the inclination of the maxillary incisors.
253
Long-term Occlusal Stability
Graber stresses that it is important to normalize the activities of
the tongue and the perioral, masticatory and cervical muscles because the
growth, morphology, function and action of these muscles have signifi-
cant effects on the overall skeletal development as well as the position
and growth direction of the teeth and jaws (Graber, 1966, 1997a,b,c). The
results of this study also clearly indicate that orthodontic treatment sup-
ported by the restoration of harmony to the tongue, perioral, chewing and
cervical muscle activities, and establishment of correct breathing through
the nose is the most effective way to maintain long-term occlusal stability
(Ohno, 1981; Kubein-Meesenburg, 1993; Kondo, 1998). As the great anat-
omist Harry Sicher said: “Whenever there is a struggle between muscle
and bone, muscle wins!”
The findings of this study suggest that the following features may
serve as specific criteria for setting orthodontic goals for long-term oc-
clusal stability:
1. Sufficient tongue space to secure the airway and allow nose
breathing with lips closed.
2. Good balance between the maxillofacial skeletal structures
and Surrounding muscles (tongue, perioral, masticatory and cervical mus-
cles).
3. Alignment of all teeth within the cancellous bone of the al-
Veoli where tongue pressure and forces of the perioral and masticatory
muscles are in equilibrium and no broken contacts in an arch form that
confirms to the shape of a relaxed tongue. This will provide a balanced lip
profile.
4. Normal symmetrical cervical muscle activity for normal cer-
vical posture and free rotational movement of the head.
5. The apex of the dens does not exceed 4.0 mm above the Mc-
Gregor line.
6. The axes of the maxillary and mandibular posterior teeth are
perpendicular to the functional occlusal plane, so that posterior occlusal
support is provided in order to unload the joints for symmetrical growth.
7. The AB plane is perpendicular to the functional occlusal plane
for optimal occlusal relationship between the maxilla and mandible.
8. The axis of the mandibular incisor is perpendicular or nearly
perpendicular to DC-Lli line (a line connecting the center of the condyle
and the mandibular incisal edge), except in Class III subjects.
9. Overjet and overbite of approximately 2.2 mm each (95%
LCLM for the mean value of normal occlusion subjects) when active
254
Kondo
treatment is finished before maxillomandibular growth has been complet-
ed, and approximately 3.0 mm each (95% UCLM of the control subjects)
when active treatment is finished after mandibular growth has been com-
pleted.
10. Proper inclination of the maxillary incisors so that F line be-
comes almost parallel to CDM line by the end of growth for harmonized
anterior and condylar guidance in order to allow smooth jaw movement
without overloading the joints.
11. The mandibular incisal edge positioned close to the inflec-
tion point (Bp) on the lingual surface of the maxillary incisor to create
adequate incisal guidance in harmony with condylar guidance for smooth
jaw movement.
12. An extension of the functional occlusal plane passes through
the mandibular foramen by the end of mandibular growth, except in Class
III cases.
When active treatment is finished before the end of maxilloman-
dibular growth has been completed, orthodontists should be fully aware of
the occlusal changes that may take place during the remaining period of
growth and should inform their patients of the possible changes. The study
results suggest that occlusal improvement at an early stage of growth will
facilitate recovery of jaw function and establishment of nose breathing as
well as esthetic improvement, all of which contributes greatly to sound
development of the maxillofacial skeleton and symmetrical joint morphol-
ogy. Thus, occlusal stability can be maintained for many years. It is also
important to maintain good rapport with patients after treatment is finished
So that appropriate actions can be taken should any occlusal abnormalities
occur. Lifetime management of oral hygiene is also essential for mainte-
nance of occlusal stability.
ACKNOWLEDGEMENTS
We would like to thank Dr. James McNamara and Ms. Katherine
Ribbens for their help in editing. We also would like to thank our office
staff. Statistics were analyzed by Ms. Keiko Miyakoda. The excellent
translation of the manuscript into English was done by Ms. Sanae Iwaka-
mi. This article never would have been possible without her help.
REFERENCES
Graber TM. Orthodontics: Principles and Practice. Philadelphia, WB
Saunders 1966:249-325.
255
Long-term Occlusal Stability
Graber TM. Orthodontics; Principles and Practice, 3rd Ed. Philadelphia,
WB SaunderS 1997a: 129-179.
Graber TM. Physiologic Principles of Functional Appliances. St Louis,
Mosby 1997b;2-12.
Graber TM. The unique nature of temporomandibular joint metabolism.
In: Rabie AM, Urst MR, Bone Formation and Repair. Amsterdam,
Elsevier Science 1997c.
Jason CH. Multiple Comparisons Theory and Methods. Boca Raton, Chap-
man and Hall 1996.
Kondo E. Occlusal stability in Class II, division 1 deep bite cases followed
up for many years after orthodontic treatment. Am J Orthod Dentofa-
cial Orthop 1998; 114:611-630.
Kondo E. Case report of malocclusion with abnormal head posture and
TMJ symptoms. Am J Orthod Dentofacial Orthop 1999; 116:481-493.
Kondo E, Nakahara R, Ono M, Arai S, Kuboniwa K, Kanematsu E, Toyo-
mura Y, Graber TM. Cervical spine problems in patients with tem-
poromandibular disorder symptoms: An investigation of the orthodon-
tic treatment effects for growing and nongrowing patients. World J
Orthod 2002:3:295-312.
Kubein-Meesenburg D, Nagerl H. Biomechanical Aspects of Stability of
Occlusion: Retention and Stability in Orthodontics. Philadelphia, WB
Saunders 1993:17.1-202.
Marcello P. Kimberlee G. Principles of Biostatics, 2nd Ed. Pacific Grove,
Duxbury Press, 2003.
Matsubara N, Nawata K, Nakai N. Introduction to statistics. College of
Arts and Sciences, The University of Tokyo 1991:4–114, 189-244,
267–268.
McGregor M. The significance of certain measurements of the skull in the
diagnosis of basilar impression. Br J Radiol 1948;21:171-181.
Ohno T, Yogosawa F, Nakamura K. An approach to openbite with tongue
thrusting habits with reference to habit appliances and myofunctional
therapy as viewed from an orthodontic standpoint. Int J Orofacial My-
ology 1981;7:3-10.
Van der Linden FPGM (translated by Miura F, Kuroda T). Facial Growth
and Facial Orthopedics. Tokyo, Quintessence 1988:58-215.
Yajima Y, Hirotsu C. Statistics. College of Arts and Sciences, The Univer-
sity of Tokyo 1992:83, 60–213, 139, 141, 196, 221, 229.
256
ORTHODONTICS IN THE EARLY MIXED DENTITION:
EFFECTIVENESS AND EFFICIENCY OF THE
ERUPTION GUIDANCE APPLIANCE
Juha Varrela
The Finnish health care system requires that the municipal health centers
provide free dental care to all children. Orthodontics is included, but be-
cause of limitations in manpower and funds, most clinics have to restrict
access to orthodontic treatment. Widespread adoption of early treatment
among the Finnish orthodontists reflects the view that early timing offers
the most efficient use of limited resources, and thereby increases access to
treatment. According to a recent nationwide survey (Pietilä et al., 2004),
the majority of Finnish orthodontists prefer to initiate treatment no later
than the middle of the mixed dentition. In most cases, early treatment is
part of a multi-stage therapy protocol in which the growth modification
phase of treatment is normally followed by a second treatment phase in
the late mixed or early permanent dentition (Väkiparta et al., 2005). How-
ever, there are attempts to develop treatment modalities where an efficient
early treatment phase would eliminate, as much as possible, the need for
Subsequent treatment phases.
This chapter will outline the preliminary results of an ongoing
clinical trial the purpose of which is to evaluate the effectiveness of an
orthodontic intervention carried out in the early mixed dentition using
eruption guidance appliances. The trial assesses the therapeutic effects of
the appliance that is used to guide occlusal development during the erup-
tion of the permanent incisors and first molars.
The study sample was collected from two municipal dental clin-
ics in the western part of Finland. The treatment protocol followed in
these clinics is a comprehensive early treatment protocol that attempts
to address all types of malocclusion in the early stages of development.
Eruption guidance therapy is applied to all children who show signs of
Class II relationship, crowding, excess overbite, excess overjet, gingival
incisors relationship, openbite, anterior crossbite, and/or buccal crossbite
(scissors bite). Treatment is initiated at the onset of the mixed dentition
period and is continued until all permanent incisors and first molars are
fully erupted. This treatment protocol was designed by Dr. Katri Keski-
Nisula, who began using eruption guidance appliances in the mid-1990s.
She first used the appliance as a replacement for fixed appliances during
257
Eruption Guidance Appliance
the second treatment phase, but after its capabilities both in growth modi-
fication and tooth alignment became evident, Dr. Keski-Nisula and her
team began using the appliance during the early phase of treatment with
the intention of reducing the need for a second treatment phase.
This was a novel idea, and it called for a study to evaluate sci-
entifically the potential of this treatment modality. A clinical investiga-
tion was started, therefore, to study the short- and long-term effects of the
treatment. The purpose of the ongoing trial, designed as a prospective,
controlled cohort study, is to evaluate the effects of the eruption guidance
appliance in a “real world” situation, i.e., in dental clinics where clinicians
follow a pre-established treatment protocol (O’Brien 2002, O’Brien et al.,
2003). Only minor adjustments were made to the normal daily routines,
and those adjustments mainly were to ensure that data collection would be
timely and controlled. At the same time, however, the established protocol
prevented the execution of a true randomized controlled trial. For ethical
and practical reasons, it was not possible to randomly assign patients into
treatment and control groups within the two municipalities. Therefore, a
control sample had to be collected from another municipality – one in
which orthodontic treatment normally was started later in the mixed denti-
tion stage.
This paper analyzes and summarizes changes that took place be-
tween the early and middle mixed dentition stages in 167 children in the
treatment group and 104 children in the control group. Once all children
have entered the early permanent dentition stage, more information will
become available as to the long-term effects of the treatment. The long-
term goal of the trial is to follow the children until they are young adults.
STUDY SUBJECTS
The children who were treated with the eruption guidance appli-
ance were selected from all of the children born in 1992 and 1993 in Jalas-
järvi (population 9,000) and in 1992 in Kurikka (population 11,000). All
children were screened during the late deciduous dentition period, and the
children who were diagnosed as needing treatment received a full clini-
cal examination (see Keski-Nisula et al., 2003, for a more comprehensive
description of the screening and clinical examination protocols). Children
showing one or more of the following occlusal characteristics were in-
cluded in the treatment group:
• distal step of > 1 mm
• Class II canine relationship of > 1 mm
• crowding
258
Warrela
• overjet of > 3 mm and lack of tooth-to-tooth contact between the
incisors
• Overbite of > 3 mm and lack of tooth-to-tooth contact between
the incisors
• anterior crossbite
• buccal crossbite (scissors bite).
The number of children fulfilling these criteria was 315. Of these,
33 children were diagnosed as having a moderate to severe maxillary
constriction. Their treatment protocol required that they be first treated
with an expansive arch, so they were excluded from the study sample.
In 27 cases, the child or the family refused orthodontic treatment before
any treatment was started. Eruption guidance therapy was started in 255
children. During treatment, 12 children moved to another municipality
and their records were excluded from the analyses. Of the remaining 243
children, 167 completed the treatment successfully. Seventy-six children
(31%) were excluded from the study because they refused to wear the
appliance. Difficulties in cooperation appeared to arise mostly from psy-
chosocial problems. Three children became seriously ill and could not
complete the treatment. In these 76 cases, treatment and participation in
the study was terminated when it became clear that the children could not
or would not cooperate. This usually occurred within a few months after
starting treatment. No further records were collected for these children and
they were excluded from the analyses.
A random sample of 104 children from Seinäjoki (population
30,000) who fulfilled the same criteria, i.e., they were screened during
the late deciduous dentition period and had similar occlusal deviations,
formed the control group. All of the children in both the treatment and
control samples were Finnish, all were healthy, and none had undergone
orthodontic treatment previously. All children and their parents were free
to decline participation in the study at any time.
The timing of the examinations and interventions was based on
the stage of dental development of each child and not on chronological
age. Treatment was initiated at the beginning of the mixed dentition period
(T1) (Keski-Nisula et al., 2003). The occlusal changes in both the treat-
ment and control groups were evaluated after full eruption of all perma-
nent incisors and first molars (T2). The present results are based on the
analyses of the occlusal and skeletal changes that occurred from T1 to T2
in 167 children in the treated group (85 boys and 82 girls) and 104 children
in the control group (52 boys and 52 girls). The mean age for both groups
was 5.1 years (SD 0.5) at T1 and 8.4 years (SD 0.5) at T2.
259
Eruption Guidance Appliance
TREATMENT PROTOCOL
During treatment, each child wore two or three prefabricated
eruption guidance appliances of different sizes (Nite-Guide" or Occlus-o-
Guide", Ortho-Tain Inc.) The appliance was worn only during the night.
Daytime wear of one hour was recommended if problems with night-wear
were encountered. The mean duration of active treatment, i.e., from the
time that the first deciduous tooth was exfoliated (T1) until all permanent
incisors and first molars were fully erupted (T2), was 3.3 years. At point
T2, all treated children entered the retention period (see Keski-Nisula et
al., 2007a for further details of treatment).
METHODS OF ANALYSIS
At points T1 and T2, both the treatment and control children un-
derwent a full clinical examination. The occlusal and skeletal characteris-
tics were assessed (as described in Keski-Nisula et al., 2003, 2006). The
statistical assessment included use of the chi-square test and the t-test.
OCCLUSAL AND SKELETAL EFFECTS
The eruption guidance appliance had several corrective effects on
the incisor segment (Keski-Nisula et al., 2007a). As indicated by Figures
1 and 2, both overjet and overbite decreased during treatment, the former
by 1.2 mm from 3.1 mm (SD 1.4) to 1.9 mm (SD 0.7) and the latter by 1.1
mm, from 3.2 mm (SD 1.7) to 2.1 mm (SD 0.9). Normally, both overjet
and overbite increase during the transition from deciduous to permanent
incisors. In the control group, overjet increased by 1.2 mm from 2.9 mm
(SD 1.8) to 4.1 mm (SD 1.9) and overbite by 0.9 mm from 3.3 mm (SD
1.9) to 4.1 mm (SD 1.3). The differences in the changes in overjet and
overbite between the control and treated groups at T2 were significant (P
< .001).
At T1, a tooth-to-tooth contact in CR was found in a minority
of the children in both groups – 18% in the treatment group and 22%
in the control group. All of the other children either had an open bite or
the lower incisors were in contact with the upper gingival/palatal muco-
sa. The contact relationship of the incisors remained fairly unchanged in
the control group. From T1 to T2, the frequency of tooth-to-tooth con-
tact increased to 25%, while the prevalence of an openbite increased from
30% to 36% and that of gingival contact decreased from 48% to 40%.
In the treatment group, on the other hand, 99% of the children showed a
260
Warrela
mm mm
15 - 10 —





•
©
|
–5 I I I —5 I I i
T1 T2 T1 T2 T1 T2 T1 T2
Control group Treatment group Control group Treatment group
Figure 1 (Left). Overbite at T1 and T2, as measured from the deciduous incisors
at T1 and the permanent incisors at T2. There was no significant difference at T1.
At T2, the difference was highly significant.
Figure 2 (Right). Overbite at T1 and T2, as measured from the deciduous incisors
at T1 and the permanent incisors at T2. There was no significant difference at T1.
At T2, the difference was highly significant.
tooth-to-tooth contact at T2. A mild openbite persisted in two children,
but no gingival contacts remained. A similar effect was seen in incisor
crowding. The prevalence of crowding was approximately 10% in the up-
per arch and approximately 45% in the lower arch in both groups at T1. In
the control group, upper crowding increased from 9% at T1 to 32% at T2,
and lower crowding from 44% to 47%. At T2, 98% of the children treated
with the eruption guidance appliance had well-aligned upper incisors and
99% had well-aligned lower incisors.
The eruption guidance appliance used in this clinical trial is de-
signed to position the lower jaw into a Class I relationship (Bergersen,
1984). This brings the mandible forward in patients who have a Class II
tendency. In the treatment group, correction of the sagittal relationship
in the posterior segments was evident (Keski-Nisula et al., 2007a). The
improvement in the canine relationship was 1.4 mm, and at T2, it was
off only 0.2 mm toward a Class II relationship (Fig. 3). Terminal plane
relationship changed 2 mm, from a distal step of 0.7 mm to a mesial step
of 1.3 mm (Fig. 4). In the control group, both the canine and the terminal
261
Eruption Guidance Appliance
mm.
mm 10 —
7,5-






5,0-


2,5-
—O.
©
©
-O
| | | | | | ||
—5,0 I TH I –5 I I i
T1 T2 T1 T2 T1 T2 T1 T2
Control group Treatment group Control group Treatment group
Figure 3 (Left). Sagittal relationship of the deciduous canines at T1 and T2. The
zero point indicates a full Class I relationship where the tip of the upper canine
coincides with the contact point of the lower canine and first molar. A positive
value indicates a shift towards a Class II relationship. There was no significant
difference at T1. At T2, the difference was highly significant.
Figure 4 (Right). Sagittal relationship of the deciduous second molars at T1 and
T2. The zero point indicates a flush terminal plane, a positive value indicates a
distal step and a negative value indicates a mesial step. There was no significant
difference at T1. At T2, the difference was highly significant.
plane relationship remained virtually unchanged during the observation
period. The differences between the treatment and control groups at T2 of
1.2 mm in the canine relationship and 1.7 mm in terminal plane relation-
ship are clinically significant.
For patients in whom crowding is present or predicted, the ap-
pliance given to them should be larger than the actual size of the dental
arches. The expansive force of the appliance, however, is limited, and oc-
casionally the appliance must be replaced several times during the course
of the treatment to achieve a good alignment of the permanent teeth.
For this reason, eruption guidance appliances are recommended only if
mild to moderate expansion is required (Bergersen, 1985). In more se-
Vere cases, a quad-helix or a rapid maxillary expander are better choices.
Analysis of the arch dimensions indicates that arch widths were larg-
er in the treatment group at T2 (Keski-Nisula et al., 2007b). The extra
262
Warrela
Widening brought about by the appliance was 1 mm in the maxillary arch
and 1.7 mm in the mandibular arch.
A cephalometric analysis of the treatment and control groups
revealed few significant skeletal effects of eruption guidance appliance
(Keski-Nisula et al., 2007b). It appears that the effects of therapy were
largely restricted to dentoalveolar structures. It is of interest, however,
that the length of the mandible increased with treatment. The length of
the mandible, measured from condylion to gnathion, increased 11.2 mm
in the treatment group and only 7.2 mm in the control group during the
average period of 3.3 years from T1 to T2. A difference of 4 mm is large
enough to be clinically significant. It seems likely that the increase in man-
dibular length was a reaction to the forward positioning of the lower jaw
caused by the eruption guidance appliance. This suggests that the growth
of mandible responds well to external stimuli at this stage of development.
Whether or not this difference between the treatment and control children
will persist as they grow older is not known at this point.
The effects of eruption guidance therapy on occlusion are summa-
rized in Table 1 (Keski-Nisula et al., 2007a). The table gives the percent-
ages of deviating occlusal characteristics at T2 including overjet > 5 mm,
overbite × 5 mm, open bite, gingival contact of the lower incisors, crowd-
ing and Class II relationship. Thirteen percent of the treated children and
88% of the control children showed one or more deviations in the middle
mixed dentition stage. The difference between the treatment and control
groups is striking and indicates that a treatment with eruption guidance ap-
pliance in the early mixed dentition stage can reduce the need for further
treatment significantly.
Table 1. Frequency of deviating occlusal characteristics in the middle mixed
dentition (T2). The difference in distribution of the characteristics between the
groups is highly significant (P< .001).

Treatment Group Control Group
N = 167 N = 104
Overjet > 5 mm 0% 30%
Openbite × 5 mm 19% 38%
Openbite 1% 36%
Gingival contact of the lower incisors 0% 40%
Upper crowding 2% 32%
Lower crowding 1% 47%
Unilateral Class II 7% 17%
Bilateral Class II 3% 35%
263
Eruption Guidance Appliance
DISCUSSION
The short-term results of the clinical trial seem to confirm the
clinical experience that the early intervention with eruption guidance ap-
pliance is an effective treatment modality. The children who completed the
treatment needed very little additional treatment. Most of them had good
tooth alignment and favorable intermaxillary relationships, with overbite
and overjet close to 2 mm, the incisors in a tooth-to-tooth contact, a mesial
step in the molar region, and the canines in almost full Class I relation-
ship. In the control group, on the other hand, very few positive changes
took place. This indicates that during the transition from the deciduous to
the middle mixed dentition stage, normal occlusal development seldom
leads to a self-correction of malocclusions. It should be emphasized that
the findings outlined here are only short-term results. As yet, no data are
available on the occlusal development beyond the middle mixed dentition
period.
It seems, in addition, that this treatment modality can be applied
to a large variety of malocclusions. Signs of malocclusion such as a Class
II relationship, crowding, excess overjet, openbite and/or deepbite are
common in children, and it is these children who form the majority of
cases treated in an average orthodontic office. All of these occlusal devia-
tions can be treated with eruption guidance appliances. The manufacturer
recommends that the appliance be used mainly for correction of mild to
moderate malocclusions. Severe malpositions of the teeth or intermaxil-
lary discrepancies that are present after the eruption of the permanent inci-
sors and first molars may prevent proper use of the appliance. It seems,
however, that if treatment is begun very early in the mixed dentition, the
severity of the malocclusion is seldom a contraindication to treatment with
eruption guidance appliances.
It is interesting to note that all of the children having a Class II
relationship, or a tendency to develop a Class II relationship, were in-
cluded in the early treatment protocol. In his discussion on treatment
timing during the 28" Moyers Symposium, Proffit (2002) concluded that
when the overall effectiveness and efficacy of different treatment modali-
ties are analyzed on the basis of the existing data, early Class II treat-
ment is indicated only for a select group of children. It would be prema-
ture to argue against this assertion on the basis of the present findings.
Only after a detailed analysis of the long-term results of the use of early
treatment with eruption guidance appliances can the effectiveness and ef-
ficacy of such treatment be determined. Nevertheless, clinical data sug-
gest that the effectiveness and efficacy of early intervention with erup-
264
Warrela
tion guidance appliances may be better than the current literature Sug-
geStS.
The finding that the growth of the mandible was enhanced in the
treatment group raises another controversial issue; whether or not the
length of the mandible can be increased through orthodontic treatment
seems to be a subject of never-ending debate. The 4 mm difference in the
length of the mandible between the treatment and control groups indicates
that the mandibular growth can be influenced to a considerable degree
during the early mixed dentition. Whether this effect is temporary or per-
manent remains to be seen. The entire question seems irrelevant, however,
as long as this growth enhancement contributes to the correction of the
Sagittal discrepancy.
The greatest problem with the present treatment modality seems
to be compliance. Although 69% of the children who originally agreed
to start the treatment completed it successfully, treatment had to be ter-
minated in 31% of the children because these children did not wear the
appliance properly. The reasons for this lack of cooperation seemed to
be largely psychosocial, indicating that family background and the readi-
ness of parents to participate are paramount to the success of treatment.
The young age of the patients per se clearly was a much lesser problem,
although it must be taken into account, particularly during the initial phase
of the therapy. Treatment of a young child with a removable appliance
Seldom can succeed if the parents are not willing to support and guide
the child. The level of non-compliance in the present investigation was
Somewhat higher than was reported in a clinical trial for the Twin-block
appliance (O’Brien et al., 2003) and about the same as that for the Fränkel
appliance (Gafari et al., 1998).
In line with the topic of the 2006 Moyers Symposium, “Early
Orthodontic Treatment: Is the Benefit Worth the Burden,” I will briefly
discuss what might be the pros and cons of an intervention that is car-
ried out in the early mixed dentition with eruption guidance appliances.
This discussion is based not only the findings of the clinical trial discussed
in this chapter, but also on the clinical experience that the team has ac-
cumulated regarding this treatment modality. As emphasized by Proffit
(2002), all treatment must be evaluated for its effectiveness and efficacy,
i.e., how well it works and what are the costs and risks relative to the
outcome. Albeit preliminary, the findings of the trial suggests that this
treatment produces a good outcome in a high percentage of patients, i.e.,
it is effective. As to the costs, eruption guidance appliances allow long
intervals between check-ups and short visits consuming relatively little
265
Eruption Guidance Appliance
chair-side time. In addition, there are very few emergency visits. All of
these are signs of a treatment modality that will load the patient, parents
and clinician relatively lightly. Length of chair-side time is a major factor
affecting total cost of treatment. In a publicly funded health case system as
in Finland, savings in chair-side time often are used to improve access to
treatment. Furthermore, risks of decalsification or root resorption seem to
be minimal with eruption guidance appliance.
The length of treatment is defined by the speed of eruption, be-
cause eruption guidance appliances are used to guide the eruption of the
permanent teeth. In the treatment sample, it took 3.3 years on average
from the onset of the mixed dentition for all of the permanent incisors and
molars to erupt fully. This can be considered a burden to the patient and
parents, particularly when the young age of patient is taken into account.
The high percentage of non-compliant patients can also be considered a
burden to all parties in terms of wasted time and effort.
SUMMARY AND CONCLUSIONS
The preliminary findings of a clinical trial evaluating occlusal
changes in 167 children treated with eruption guidance appliances during
the early mixed dentition stage and 104 control children led to the follow-
ing conclusions: -
1) Treatment with eruption guidance appliances in the early
mixed dentition stage can effectively correct the occlusion and decrease
the further need of treatment.
2) The effects of the appliance are mainly dentoalveolar. How-
ever, the length of the mandible is increased by the treatment.
3) The success of the treatment depends on parental involvement.
Thirty-one percent of the children who began treatment did not complete
treatment because they did not wear the appliance properly.
The participants in this clinical trial will be followed until they
have entered the early permanent dentition stage, and further data will be
collected. This will allow for further analysis of the effectiveness and ef-
ficacy of the use of eruption guidance appliances.
REFERENCES
Bergersen EO. The eruption guidance myofunctional appliance: Case se-
lection, timing, motivation, indications and contraindications in its
use. Funct Orthod 1985:2:17–33.
266
Warrela
Bergersen EO. The eruption guidance myofunctional appliance: How it
works, how to use it. Funct Orthod 1984; 1:28-35.
Gafari J, Shofer FS, Jacobsson-Hunt U, Markowitz DL, Laster LL. Head-
gear versus function regulator in the early treatment of Class II, Divi-
sion 1 malocclusion: A randomized clinical trial. Am J Orthod Dento-
facial Orthop 1998; 113:51–61.
Keski-Nisula K, Lehto R, Lusa V. Keski-Nisula L., Varrela J. Occurrence
of malocclusion and need of orthodontic treatment in early mixed den-
tition. Am J Orthod Dentofacial Orthop 2003;124:631-638.
Keski-Nisula K, Keski-Nisula L., Mäkelä P. Mäki-Torkko T. Varrela J.
Dentofacial features of children with distal occlusions, large overjets,
and deepbites in the early mixed dentition. Am J Orthod Dentofacial
Orthop 2006, in press.
Keski-Nisula K, Hernesniemi R, Heiskanen M, Keski-Nisula L., Varrela J.
Orthodontic intervention in the early mixed dentition: A prospective
controlled study on the effects of eruption guidance appliance. Am J
Orthod Dentofacial Orthop 2007a, in press.
Keski-Nisula K, Keski-Nisula L., Varrela J. Skeletal effects of orthodontic
treatment with eruption guidance appliance in the early mixed denti-
tion. Eur J Orthod 2007b, in press.
O’Brien K. Treatment timing for Class II malocclusion: The results of two
UK-based multi-center randomized controlled trials. In: McNamara
JA Jr, Kelly KA eds. Treatment Timing: Orthodontics in Four Dimen-
sions. Craniofacial Growth Series, Department of Orthodontics and
Pediatric Dentistry and Center for Human Growth and Development,
The University of Michigan, Ann Arbor 2002,39:75-94.
O’Brien K, Wright J, Conboy F. Sanjie Y, Mandall N, Chadwick S, Con-
nolly I, Cook P. Birnie D, Hammond M, Harradine N, Lewis D, Mc-
Dade C, Mitchell L, Murray A, O’Neill J, Read M, Robinson S, Rob-
erts-Harry D, Sandler J, Shaw I. Effectiveness of early orthodontic
treatment with the twin-block appliance: A multi-center, randomized,
controlled trial. Part 1: Dental and skeletal effects. Am J Orthod Den-
tofacial Orthop 2003;124:488-494.
Pietilä T, Alanen P, Nordblad A, Kotilainen J, Pietilä I, Pirttiniemi P. Var-
rela J. Hampaiden oikomishoito terveyskeskuksissa. Sosiaali-ja terve-
ysalan tutkimus-ja kehittāmiskeskus Stakes, Raportteja 279, Helsinki
2004.
Proffit WR. Treatment timing: Effectiveness and efficacy. In: McNama-
ra JA Jr, Kelly KA eds. Treatment Timing: Orthodontics in Four Di-
mensions. Craniofacial Growth Series, Department of Orthodontics
267
Eruption Guidance Appliance
and Pediatric Dentistry and Center for Human Growth and Develop-
ment, The University of Michigan, Ann Arbor 2002,39:13–24.
Väkiparta MK, Kerosuo HM, Nyström ME, Heikinheimo KA. Orthodon-
tic treatment need from eight to 12 years of age in an early treatment
oriented public health care system: A prospective study. Angle Orthod
2005;75:344–349.
– 268
A NEW CLASSIFICATION OF FACIAL ASYMMETRY
Hyeon-Shik Hwang
With greater importance being attached to facial appearance by society in
general, more patients are unhappy with their facial asymmetries. Some
patients insist that asymmetry develops or worsens during or after orth-
odontic treatment. When facial asymmetries develop and exhibit diverse
characteristics due to various etiologic factors, a systematic and accu-
rate diagnostic classification system is essential for the proper manage-
ment of this problem in an orthodontic practice (Bishara et al., 1994).
Based on clinical experience, it is clear that facial asymme-
try patients can be classified into four types according to their mor-
phologic characteristics. In fact, all facial asymmetry patients can
be managed with little or no difficulty if their asymmetry is classi-
fied correctly. This chapter presents a new classification system for fa-
cial asymmetry that includes (1) an examination of how the system
was developed, (2) an analysis of how asymmetry patients are diag-
nosed and (3) a discussion of the proper management of such patients.
DEVELOPMENT OF A NEW
CLASSIFICATION SYSTEM
How do facial asymmetries occur? Why are there differences
between the right and left sides of the face? Many questions have been
raised and a numerous investigations have been conducted to answer
these and related questions. As a result of clinical experience, four types
of asymmetry have been defined based on morphologic characteristics.
Figure 1 illustrates the most common type of facial asymmetry.
Although the patient presented with anterior crowding and crossbites of
Several teeth, her chief concern was her facial asymmetry. Analysis of the
posteroanterior (PA) cephalogram revealed ramal length differences be-
tween the right and left side, occlusal plane canting, and menton deviation.
What was the reason for all of these features? The panoramic radiograph
showed a significant difference in condylar length between the right and
left side. This indicates that all asymmetric features of this patient were
due to asymmetric condylar growth.
269
A New Classification
Figure 1. A 20-year-old patient complaining of facial asymmetry. Although she
had a crossbite and prognathic mandible (top row), her chief concern was her
facial asymmetry (2° row). The PA cephalogram showed a difference in ramal
length between the right and left sides, the occlusal plane canting, and the men-
ton deviation. Panoramic radiographs showed that the asymmetry was caused by
asymmetric condylar growth.

270
Hwang
Figure 2 illustrates a different type of asymmetry. While the pa-
tient did not present with a chin deviation, he did have a swelling of his
right cheek. Radiographs revealed that the second premolar and the first
and second molars were missing on the upper left side. He also had a
unilateral mastication habit. Considering that a strong masticating muscle
tone results in a low angle and that a weak masticating muscle tone results
in a high angle, the patient’s chewing side developed to a low angle, while
his non-chewing side developed to a high angle. In addition, the unilateral
muscle hypertrophy due to unilateral mastication resulted in soft tissue
asymmetry. In this patient, both the skeletal and soft tissue asymmetries
Were a result of the patient’s unilateral mastication habit.
Figure 2. A 10-year-old male patient with a swelling on one side of his face. His
dental occlusion showed no midline discrepancy (top row), but his mother was
Concerned about his asymmetric face. The PA cephalogram showed no significant
menton deviation, but there was a difference in left and right ramal length. The
*Symmetry was caused by unilateral mastication due to the congenital absence of
the second premolar and the first and second molars on the upper left side (bottom
row).

271
A New Classification
The patients seen in Figures 1 and 2 had very different facial
asymmetry characteristics. The causes of these asymmetries also were
quite different; therefore, a classification system was needed to enable a
differential diagnosis. The first patient was classified as type RM (Ramus
Menton) because she had a difference in ramal length from one side to the
other and because she had a menton deviation. The second patient was
classified as RA (Ramus Angle) because he did not have a menton devia-
tion but did present with a difference in ramal angle shape between the
right and left sides.
Figure 3 illustrates a third type of asymmetry. The mother of this
patient was concerned about her daughter's facial asymmetry. The patient
presented with a posterior crossbite on the right side only. The PA cepha-
logram revealed a distinct menton deviation, but no difference in ramal
lengths. This type of facial asymmetry is different from that of both previ-
ous patients. The cause of this patient’s facial asymmetry was a functional
shift of the mandible. This patient was classified as type M (Menton) be-
cause the only problem was a deviation in menton.
Figure 3. A seven-year-old female patient who complained about her chin de-
viation. She presented with a posterior crossbite on the right side (top row and
frontal photograph). The PA cephalogram showed a menton deviation, but no
ramal length difference and no occlusal plane canting. Her asymmetry was due
to a functional shift of the mandible, which probably was caused by a narrow
maxilla.


272
Hwang
Figure 4 illustrates a unique type of facial asymmetry. Although
the patient had a good posterior occlusion on both sides and did not show
significant facial asymmetry, she complained of an asymmetric face. While
she did have a dental midline discrepancy, it was the result of a congeni-
tally missing lower incisor and not a skeletal asymmetry. The PA cephalo-
gram revealed no ramal length difference, no occlusal plane canting, and
no menton deviation. If there was any difference between the right and left
side of this patient, it was a bulkiness of the mandible. This patient was
classified as type B (bulkiness) because the only problem is in the bone or
Soft tissue contour of the mandible.
| \
Figure 4. A 21-year-old female patient who complained of facial asymmetry. She
had an upper and lower dental midline discrepancy with good posterior occlusion
on both sides (top row). Occlusal views revealed that the midline discrepancy
Was due to a missing lower incisor and was not a skeletal problem (2” row).
The patient complained that her face looked asymmetric, but the PA cephalogram
showed no ramal length difference, no occlusal plane canting and no menton de-
Viation. The only difference between the right and left sides was in the bulkiness
of mandible.

273
A New Classification
RM Type Facial Asymmetry
This type of asymmetry results from asymmetric condylar growth,
i.e., overgrowth or undergrowth of one side of the condyle. The chin devi-
ates in the opposite direction of the side of the longer ramus. Compensa-
tory asymmetric Vertical growth of the maxilla is common, particularly
when the difference in right and left mandibular growth is great. This is
the most common type of facial asymmetry and is the result of a variety of
factors such as unilateral hyperplasia or hypoplasia of the condyle, anky-
losis and various forms of pathology and trauma of the temporomandibu-
lar joint. Most patients who have this type of facial asymmetry are skeletal
Class III or Class II patients.
RA Type Facial Asymmetry .
The RA type of facial asymmetry results from unilateral mastica-
tion. The chewing side of the jaw develops at a low angle while the non-
chewing side develops at a high angle. While there is a difference in ramal
lengths, there is no chin deviation. In addition to the different angle shape
that is skeletal, there also is soft tissue asymmetry resulting from unilat-
eral muscle hypertrophy on the chewing side. This type of asymmetry can
result from a missing tooth or group of teeth, a total Scissors bite, occlusal
interference or poor dental restoration(s) on one side.
M Type Facial Asymmetry
The M type of facial asymmetry results from a functional shift of
the mandible. While this type of asymmetry exhibits a deviation of men-
ton, there is no significant difference in ramal length. The only difference
between the right and left sides is the menton deviation. The functional
shift of the mandible usually is due to various types of occlusal interfer-
ence such as a malposed tooth, a crossbite of one or several teeth, or a
constricted maxillary arch. The abnormal initial tooth contact in centric
relations results in the subsequent mandibular displacement in centric oc-
clusion. This can be called a displacement asymmetry, while RA and RM
facial asymmetries are structural asymmetries (Schmid et al., 1991).
B Type Facial Asymmetry
If a patient exhibits a difference in bulkiness of the mandible
between the right and left sides, this is classified as B type asymmetry.
The only difference between the right and left sides is bone or soft tissue
contour with no significant ramal length difference or menton deviation.
Considering that there is no such thing as perfect symmetry in the human
274
Hwang
body (Shah and Joshi, 1978; Peck et al., 1991), all individuals might be
candidates for this type of facial asymmetry. While most individuals are
not aware of the nature of this type of asymmetry, there are those who
think that their faces look asymmetric. This type of asymmetry is diag-
nosed by the patient’s own subjective perception and also is classified as
B type asymmetry.
DIAGNOSTIC PROCEDURES
Facial asymmetry types can be identified easily by taking a PA cephalo-
gram. As shown in Figure 5, RM, M, RA, or B type can be defined using
only two measurements in the PA cephalogram: menton deviation and the
difference between the right and left ramal length. While the selection of
a reliable reference line is important, we prefer to use crista galli (Cg) and
anterior nasal spine (ANS) for the construction of the midsagittal refer-
ence lines. Menton deviation is defined by the angle between the reference
line and the ANS-menton line, and ramal length difference is measured by
locating the antegonion (Ag) on the both sides of the PA cephalogram. Fa-
cial asymmetry types can be diagnosed simply using only a few landmarks
On the PA cephalogram: Cp, ANS, Me and right and left Ag (Fig. 6).
Menton Deviation
yes / \n.
Ramus Length Difference (Rt / Lt)
yeSſ \ In O
ngle type
\""
| ETTUS
Figure 5. A diagnostic procedure for facial asymmetry classification, RM, M, RA
or B facial asymmetry types can be defined using only two measurements from
the PA cephalogram: the menton deviation and the difference in ramal length dif-
ference between the right and left sides.
º º
uliness type

275
A New Classification
Figure 6. A frontal cephalometric tracing illustrates the measurements that define
facial asymmetry types. Menton deviation (1) is defined by the angle between
ANS-Me line and the midsagittal reference line (Cg-ANS). Ramal length differ-
ence is defined by the difference between the right and left vertical position of the
antegonion (Ag).
To determine whether or not a patient has facial asymmetry, nor-
mative data is needed for each measurement. Hwang and colleagues (2004)
recently developed normative data for the diagnosis of facial asymmetry
using sixty Class I individuals with well-balanced faces. They established
that mean and standard deviations were 0.84° -- 0.77° and 1.90 mm +
1.72 mm for menton deviation and ramal length difference respectively.
Based on these results, mean and standard deviation, the number of 2*
and 3 mm can be used as a guideline to define the presence of asymmetry.
However, these numbers should be used as a reference only on the PA
cephalogram. The presence of asymmetry can be defined comparatively
rather than quantitatively.

276
Hwang
3D images were introduced into orthodontic diagnosis with the
development of computed tomography (CT). Using CT technology, we
have developed a 3D image analysis for the diagnosis of facial asymmetry
(Hwang, 2004). This 3D CT imaging technique can be most helpful in
cases that cannot be diagnosed using PA cephalometry alone.
TREATMENT STRATEGY
Once the facial asymmetry type has been determined, a treatment
plan can be formulated according to the type.
B Type Facial Asymmetry
Type B asymmetry can be corrected with reduction or augmenta-
tion plastic surgery. The decision to undergo surgery, however, must be
made by the patient. Most patients do not choose a surgical option even
when they are diagnosed as having an asymmetry. On the other hand, pa-
tients worry about the possibility of asymmetry if they do not understand
the exact nature of their diagnosis. Many people who have occlusal asym-
metries think that the problem is related to their face. For example, patients
who have a dental midline discrepancy due to simple crowding or a miss-
ing lower incisor often think that their face is asymmetric. Some of them
demand a correction of their facial appearance; however, most patients un-
derstand that this is not necessary after learning that the asymmetry is not
related to their occlusion and that their face has no other problem except a
small asymmetry in mandibular skeletal or soft tissue bulkiness. For this
reason, the nature of asymmetry should be explained carefully to patients.
Once a patient becomes concerned about the appearance of his or her face,
even a small asymmetry should not be disregarded. This applies to all
patients, even those who have good-looking faces. Verbal explanation is
not enough for these patients. All relevant examinations, including taking
radiographs, should be performed in order to alleviate patients’ concerns
and explain the exact nature of their asymmetry. This is a very important
component of patient management for these patients (Fig. 7); however, a
patient should be referred to a surgeon for plastic surgery if the patient has
a significant asymmetry or if the patient remains adamant about having
his or her facial appearance corrected. Treatment planning for this type
of asymmetry should include the patient’s expressed need as well as the
clinical diagnosis.
277
A New Classification
Figure 7. A 27-year-old male patient with type B asymmetry. He complained of
chin deviation and underwent orthognathic surgery (asymmetric set-back of 8.5
mm on the right side and 2.5 mm on the left side). A comparison of pre- (left page)
and post-surgery (right page) photographs, cephalograms and tracings shows that
a good result was attained. The patient complained that he still had an asymmetry.
which was due to a difference in mandibular contour between the right and left
sides of his face. Six months after orthognathic surgery, he underwent reduction
Surgery (bottom row, right page). He then was satisfied with the results, even
though the pre- and post-operative change was not significant.

278


279
A New Classification
-
Figure 7. Continued. End of treatment photographs.
RA Type Facial Asymmetry
If a patient is diagnosed as having an RA type of facial asymme-
try, treatment must result in the restoration of bilateral chewing. In most
cases, scissors bite is the cause of the unilateral mastication that results
in asymmetry. This can be corrected easily with a precision lingual arch
appliance (Burstone, 1989). A lingual arch is fabricated using 032 x 032
TMA wire, and activated in a manner of unilateral expansion. The lower
molar on the scissors-bite side can be expanded buccally without causing
any significant movement of opposite side of the molar (Fig. 8). While
bilateral mastication can be restored with orthodontic correction of a scis-
sors bite, a perfectly symmetrical facial appearance cannot be guaranteed
if the patient is an adult. Further asymmetry, however, can be prevented
with this type of orthodontic intervention.
—).
Figure 8. A 53-year-old male patient with type RA asymmetry, whose original
concern was restoration of missing teeth in the upper right posterior area (A). The
patient’s prosthodontist wanted the scissors bite corrected to reduce overload on
future implant restoration on the right side and to obtain proper function on the
left side (B-D). A precision lingual arch was fabricated using 032 x 032 TMA
wire and activated in a manner of unilateral expansion (E). The lower left side
of the first molar was expanded without significant movement of the right side
molar (F–G). Using the molar expanded with the lingual arch as an anchorage, the
second molar was expanded with a TMA box loop (H-I). The upper second molar
was moved palatally with the use of a transpalatal arch with distal extension wire
(J-K). After correcting the scissors bite (L), the upper anterior teeth were aligned
and redistributed for esthetic restoration (M-O).

280
Hwang

281
A New Classification
Figure 8. Continued. The before and after treatment photographs (P-R) showed
no significant improvement in the patient’s facial appearance, even after comple:
tion of successful orthodontic treatment (S); however, further asymmetry WaS
prevented from occurring, and a healthy and properly functioning occlusion Was
achieved.
M Type Facial Asymmetry
If a patient is diagnosed as having an M type of facial asymme-
try, the cause is probably a functional shift of the mandible. Any dental
prematurity that may result in a functional shift of the mandible should be
removed. Usual treatment to correct this condition includes alignment and


282
Hwang
expansion of the maxilla using plate appliances or rapid maxillary expand-
ers (Cameron et al., 2002). Occlusal splints may help to eliminate habitual
posturing and deprogram the musculature. If the problem involves the pri-
mary teeth, it can be corrected simply with occlusal adjustments. Gener-
ally, treatment of this type of asymmetry is easy and the prognosis is good,
particularly in growing individuals. It is important to note, however, that
this type of asymmetry can develop into an RM or RA type of asymmetry
if it is not treated at an early stage (Fig. 9).
- -
Figure 9. A 12-year-old boy with a type M asymmetry who complained of a chin
deviation. He presented with a functional shift of the mandible to the right side
due to scissors bite on the left side (A-D). A precision lingual arch was used in a
manner of unilateral expansion on the left side (E-F) with a concomitant edgewise
appliance (G-L). This treatment resulted in a bilateral balanced occlusion.


283
A New Classification
Figure 9. Continued. Comparison of pre- (M-N) and post-treatment (O-P) pho-
tographs show a dramatic change in facial appearance. Treatment was managed
simply, but this type of asymmetry can develop easily into to an RM or RA type
of asymmetry if not treated at an early stage.

284
Hwang
RM Type Facial Asymmetry
The prognosis for a patient with an RM type asymmetry is ques-
tionable because the asymmetry is due to differential condylar growth.
Forecasting and controlling condylar growth is not always possible. Func-
tional jaw orthopedic treatment is indicated if the patient still is grow-
ing (Subtelny, 2000). Hybrid appliances can be used to modify growth
(Vig and Vig, 1986). Treatment outcomes generally are good if the patient
wears the appliances properly, but perfect correction of the asymmetry is
not possible, particularly if condylar growth is unfavorable (Fig. 10).
Figure 10. A nine-year-old boy who presented with type RM facial asymmetry
and a unilateral posterior crossbite. A dental midline discrepancy (A-C) was cor-
rected with a rapid maxillary expander (D-F) and a hybrid appliance (G-L). A
perfect correction of the skeletal asymmetry could not be obtained, however, even
With good patient cooperation. This is evident in a comparison of pretreatment
and post-treatment photographs (M-T). Facial asymmetry usually becomes worse
With growth in this type of asymmetry.


285
A New Classification
IV.
Figure 10. Continued. Pretreatment photographs, PA cephalogram and tracing.

286
Hwang
Figure 10. Continued. Post-treatment photographs, PA cephalogram and tracing.

287
A New Classification
If the RM type of patient is an adult, camouflage treatment or Sur-
gical correction is the treatment of choice. Camouflage treatment can be
performed successfully without correcting the facial asymmetry. This is a
good option for patients who are happy with their facial appearances (Fig.
11); however if a patient insists that the asymmetry be corrected, the only
treatment choice is surgical correction of the asymmetry. Mild asymme-
tries are corrected with mandibular surgery with or without genioplasty.
More severe deviations such as those with maxillary canting need both
mandibular and maxillary surgery, including a maxillary Le Fort I oste-
otomy. Augmentation or reduction surgery can be added, depending on the
severity of the asymmetry or the patient’s desire (Fig. 12).

288
Hwang
Figure 11. Continued. Post-treatment photograph and PA cephalogram.
<-
Figure 11. A 25-year-old female patient with a type RM asymmetry who com-
plained of upper anterior crowding and protrusion. While she presented with sig-
nificant facial asymmetry, she was not concerned about the asymmetry. Camou-
flage treatment was performed successfully with asymmetric extraction mechan-
ics. Although pre- and post-treatment comparisons did not show any improve-
ment in the degree of skeletal asymmetry, she was very satisfied with the results
of treatment.


289
A New Classification
JHJ (20Y 7M)
Mn: SSRO
asymmetric set-back 7.5 mm (Rt)
3.0 mm (Lt)
HSW (23X 1M)
Mn: SSRO
asymmetric set-back 8.0 mm (Rt)
2.0 mm (Lt)
Genioplasty, Vertical reduction 3.0 mm
Figure 12. Surgical correction of RM type of asymmetry. Mild asymmetry Ca" be
corrected with mandibular surgery alone (patient JHJ), Genioplasty can be added
if it is necessary to correct the chin contour (patient HSW).





290
Hwang
SJN (23X 6M)
Mx: Le Fort Osteotomy
Rt side 3.0 mm impaction
Mn: SSRO
asymmetric set-back 11.5 mm (Rt.)
5.0 mm (Lt)
Mn angle shaving (Lt side)
BGY (25Y 7M)
Mx: Le Fort | Osteotomy
Rt side, Vertical reduction 3.0 mm
Lt side, Vertical elongation 3.0 mm
Mn: SSRO
asymmetric set-back 9.0 mm (Rt),
(no set-back on the left side)
Medpore augmentation
(Rt. Zygoma area and Lt. Mn angle)
Figure 12. Continued. If the patient has an occlusal plane cant visible from
the frontal view, two-jaw surgery would be necessary. Reduction surgery (patient
SJN) or augmentation surgery (patient BGY) can be added to the treatment plan
depending on the severity of the asymmetry or the patient's wishes.




291
A New Classification
CONCLUSION
Facial asymmetry can be classified into four groups: RM (Ra-
mus Menton), RA (Ramus Angle), M (Menton) and B (Bulkiness). These
groups are based on menton deviation, as defined by the angle between
the reference line and the ANS-menton line, and ramal length differences
as measured by locating the antegonion (Ag) on the both sides of the PA
cephalogram. This new system of classifying facial asymmetries allows
the clinician to determine the cause of a given asymmetry and to formulate
a proper treatment strategy for those patients who are concerned about
their facial asymmetry (Table 1).
Table 1. Summary of a new classification system for facial asymmetry.
RM RA M B
Main Ramal length Ramal length No ramal Difference of
characteristics difference difference length bulkiness
Menton Difference in difference
deviation angle shape Menton
deviation
Dental Asymmetric Related or not Asymmetric Not related
characteristics related
Prevalence' 74.5% 6.5% 4.5% 5.5%
Causes Asymmetric Unilateral Functional sº
condylar growth mastication shift of
mandible
Treatment Growth Restoration of Removal of Explanation
modification bilateral prematurity Plastic
Camouflage maStication Surgery
treatment
Orthognathic
Surgery
"The subjects considered above were Korean patients who visited a university hospital and had PA
cephalograms taken for diagnostic purposes. The sum of all types presented in the table was 91%;
the remaining patients (9%) were diagnosed as being within normal limits.
292
Hwang
REFERENCES
Bishara SE, Burkey PS, Kharouf JG. Dental and facial asymmetries: A
review. Angle Orthod 1994;64:89–98.
Burstone CJ. Precision lingual arches. Active applications. J Clin Orthod
1989:23: 101-109.
Cameron CG, Franchi L, Baccetti T. McNamara JA Jr. Long-term effects
of rapid maxillary expansion: A posteroanterior cephalometric evalu-
ation. Am J Orthod Dentofacial Orthop 2002; 121:129-135.
Hwang HS. Maxillofacial 3D image analysis for the diagnosis of facial
asymmetry. J Korean Dent Assoc 2004:42:76–83.
Hwang HS, Lee KH, Park JY, Kang BC, Park JW, Lee JS. Development
of posteroanterior cephalometric analysis for the diagnosis of facial
asymmetry. J Korean Dent Assoc 2004:42:219–231.
Peck S, Peck L, Kataja M. Skeletal asymmetry in esthetically pleasing
faces. Angle Orthod 1991;61:43–48.
Schmid W. Mongini F, Felisio A. A computer-based assessment of struc-
tural and displacement asymmetries of the mandible. Am J Orthod
Dentofacial Orthop 1991; 100:19-34.
Shah SM, Joshi MR. An assessment of asymmetry in the normal craniofa-
cial complex. Angle Orthod 1978;48:141-148.
Subtelny JD. Early Orthodontic Treatment. Chicago, Quintessence 2000:
139–153.
Vig PS, Vig KW. Hybrid appliances: A component approach to dentofacial
orthopedics. Am J Orthod Dentofacial Orthop 1986;90:273-285.
293
294
EVALUATION OF FACIAL GROWTH:
A 3D SOFT TISSUE PERSPECTIVE
Chung How Kau
Stephen Richmond
Orthodontics has earned its place as dentistry’s oldest specialty. For more
than a century, the practice of orthodontics has continued to evolve and
change. From Angle's and Andrew’s classifications of dental occlusions to
the great extraction/non-extraction debates (Case, 1964) with their focus
on dental structures to the incorporation of facial balance indicators (facial
esthetic keys) into treatment planning goals (Arnett, 1993; Sarver, 1998;
Ackerman, 1999), orthodontics is a profession that is reinventing and re-
newing itself on a daily basis.
Modern technology has enabled the development of new methods
for diagnosis such as Bolton-Broadbent’s cephalometer (Broadbent et al.,
1975) in the early 1900s and the modern-day three-dimensional imaging
systems (Papadopolous et al., 2002; Hajeer et al., 2004; Mah and Hatcher,
2004; Kau et al., 2005a). With continuing innovations in technology, the
last decade has seen an introduction of both hard and soft tissue imag-
ing devices, the latter being non-invasive, easy to use and radiation free.
These “surface imaging” or “soft tissue” imaging systems, which initially
were expensive, bulky and technique sensitive, have gone through several
modifications. Today they are easily portable, and soft tissue images can
be captured quickly and efficiently (Harrison et al., 2004; Aldridge et al.,
2005; Kau et al., 2005c, 2006a). However, the development of three-di-
mensional systems is still in its infancy when compared to the equivalent
counterparts (Blais, 2004), as realistic soft tissue simulation is a complex
task (Mah and Bumann, 2001; Mah, 2002; Mah and Enciso, 2003).
The challenges, therefore, lie not in the impressive three-dimen-
Sional images themselves, but in the reliability, reproducibility and clinical
application of the images. This chapter is part of the process by which we
hope to establish the rightful place of three-dimensional imaging systems
in contemporary orthodontics. Without a real understanding of the prob-
lems associated with three-dimensional imaging and the development of
useful tools, these fancy images will remain nothing more than beautiful
displays for clinicians to admire.
295
3D Soft Tissue Perspective
To date, no study in the literature has included a true follow-up
look at the soft tissue facial changes and growth in a longitudinal sample;
and yet, the focus of modern orthodontics has been on the external facial
profile and facial balance! This study, therefore, aims to show the various
tools that can be used in three-dimensional image analysis of soft tissue
growth.
SUBJECTS AND METHODS
The laser scanning system that was used consists of two high-
resolution Minolta Vivid VI900 3D cameras, with a reported manufactur-
ing accuracy of 0.3 mm, operating as a stereo pair. Each of these cameras
emits an eye-safe Class I laser (FDA) \, = 690 nm at 30 mW with an
object-to-scanner distance of 600 to 2500 mm and a fast-mode scan time
of 0.3 seconds. The system uses a one-half-frame transfer charged couple
device (CCD) and can acquire 307,000 data points. The scanner's output
data is 640 x 480 pixels for 3D and red, green and blue (RGB) color data.
Data was recorded on a desktop workstation with a 2 GHz Pentium 4 pro-
cessor. For surface registration, a Minolta medium range lens with a focal
length of 14.5 mm was used. The scanners were placed at a distance of
1350 mm from the head frame and were controlled with Multi-scan" soft-
ware (Cebas Computer GmBH, Eppelheim, Germany). Data coordinates
were saved in a vivid file format (vvd).
Information was transferred to RF4, a reverse modeling software
package (Rapidform" 2004, INUS Technology Inc, Seoul, Korea) for
analysis. This software provides nine different three-dimensional work ac-
tivities which, when used together, allow the creation of high quality poly-
gon meshes, accurate freeform non-uniform rationale b-spline (NURBS)
surfaces and geometrically perfect solid models. RF4 generates data as
absolute mean shell deviations, standard deviations of the errors during
shell overlaps, maximum and minimum range maps, histogram plots and,
finally, color maps. All linear measurements were made in millimeters.
This imaging system has been previously reported and validated (Kau et
al., 2004, 2005a, 2005b, 2006b).
The subjects in the study were Caucasian children drawn from
the 7th grade of two large comprehensive schools in the South Wales
Valleys area. Subjects with craniofacial anomalies were excluded. Ethi-
cal approval was obtained from the Director of Education, head teachers,
school committees and the relevant ethics committees. An introductory
letter was sent to the parents by the schools' head teachers, inviting the
296
Kau and Richmond
children to participate. Written informed consent was obtained prior to ob-
taining the 3-dimensional laser scans. Height and weight parameters were
collected and facial scans were done at six-month intervals.
This chapter contains selected examples of subjects evaluated in
this two-and-a-half year longitudinal study. Shell-to-shell differences and
volumetric changes are discussed.
Shell-to-Shell Differences
Each subject’s face was analyzed at every time interval. Shell-to-
shell difference parameters, representing a sum total of the absolute dif-
ference of the match of two faces, were recorded using T1 as a baseline.
A tolerance of 0.85 mm, representing clinical reproducibility error, was
applied at each comparison to reflect the results of previous validation
research (Kau et al., 2005b).
In general, there was a decrease in percentage (%) readings in all
groups as the study progressed over time. This finding was mirrored by
an increase in the absolute mean shell-to-shell differences over time. This
meant that as the subjects grew older, there was a corresponding decrease
in the matching of two faces. The magnitude of change between shells
measured in mm also increased in line with the percentage decrease in
shell-to-shell match. The results were analyzed further on the basis of the
shell-to-shell deviation scores in the following manner:
• Little-to-no change N > 85%
• Reasonable change 65% - N < 85%
• Great change 65% - N
The scores for little-to-no change between composite faces were
based on the previous validation findings that 90% of the maps should be
well aligned before a reproducible facial morphology was acceptable (Kau
et al., 2005, 2005b). A 5% leeway was incorporated; hence the value of
N > 85%. Reasonable change was defined as change in shell-to-shell dif-
ference of 1/3 of the face. Great change was defined as change in shell-to-
shell difference of more than 65%. Examples for both reasonable change
and for great change are presented below.
Reasonable Change — Female. The composite face CNO023 was
chosen at random to illustrate a female subject whose surface changes be-
tween all four shell-to-shell comparisons were reasonable (Fig. 1).
The shell-to-shell differences in percentages were 84.87%,
76.55%, 75.48% and 70.50% for C1, C2, C3 and C4 respectively. The
shell-to-shell differences in linear measurements represented as mean
297
3D Soft Tissue Perspective
i
Figure 1. Reasonable Changes. This figure depicts surface changes to subject
CN0023’s composite faces over time. The composite scan at T1 was used as a
baseline for all four comparisons. The time interval was two years and the subject
was 12 years old at the start of the study. Positive changes are seen as red areas
and negative changes are seen as blue areas. Height, weight and BMI scores also
were included.
scores were 0.41 mm, 0.61 mm, 0.60 mm and 0.70 mm for C1, C2, C3 and
C4 respectively.
• Forehead and Eyes. There were no apparent surface changes to
the forehead region or eyes in this subject over the study period.
• Nose. The nose showed considerable changes over time. There
was a general straightening of the nasal bridge and elongation of the nose in
an anterior and downward direction. These changes ranged from 2.91mm
to 3.53 mm at the most prominent areas.
• Cheeks. There were visible and negative changes in the cheek
areas across the time range, and the changes ranged from negligible chang-
es to -2.33 mm at its deepest points.
• Lips. There were some initial changes in the lips, but the overall
change was negligible by the end of the study period. The changes Were *
direct result of the translation of the lips in a downward direction in co-
respondence with the nose. The changes ranged from 1.36 mm to 2.06 mm
for the upper lips. The lower lip changes ranged from negligible readings
to 2.06 mm.

298
Kau and Richmond
Figure 2. Great Changes. This figure depicts surface changes to subject CNO0.46’s
composite faces over time. The composite scan at T1 was used as a baseline for
all four comparisons. The time interval was two years, and the subject was 11.7
years old at the start of the study. Positive changes are seen as red areas and nega-
tive changes are seen as blue areas. Height, weight and BMI scores also were
included.
• Chin Soft Tissue. There was a small area of elongation in the
Vertical dimension of the face ranging from 1.98 mm to 3.53 mm.
Great Change — Male. The composite face CNO0.46 was chosen
at random to depict a male subject whose surface changes between all
four shell-to-shell comparisons were great when compared to the baseline
Values (Fig. 2).
The shell-to-shell differences in percentages were 96.50%, 71.97%.
61.44% and 59.69% for C1, C2, C3 and C4 respectively. The shell-to-shell
differences in linear measurements represented as mean scores were 0.28
mm, 0.86 mm, 0.94 mm and 1.00 mm for C1, C2, C3 and C4 respectively.
The changes at C1 were not clinically significant.
• Forehead and Eyes. There was little change to most of the cen-
tral portion of the forehead region with the exception of C3. The eyes
deepened as the face changed over time, the change ranging from -0.30
Imm to -1.36 mm.

299
3D Soft Tissue Perspective
• Nose. There were changes in the nose over time. There was a
general straightening of the nasal bridge and elongation of the nose in an
anterior and downward direction. There also was a general broadening of
the nose in a triangular fashion. These changes had a maximum difference
of 2.35 mm.
• Cheeks. There were considerable, visible negative changes in
the cheek areas. These values ranged from -1.42 mm to -2.58 mm at its
deepest points.
• Lips. The lip changes appeared to follow the translation of the
lips in a downward direction in correspondence with the nose. There was
a greater translation of the upper lip, with the changes ranging from mm to
1.95 mm. The lower lip changes were approximately 0.90 mm.
• Chin Soft Tissue. There was a projection of the chin in an inte-
rior direction of approximately 5.05 mm resulting in considerable elonga-
tion of the vertical dimension of the face.
Volumetric Changes
A procedure for measuring three-dimensional volume changes be-
tween two selected facial shells was developed in-house and followed the
steps outlined below.
• Two facial shells were selected, checked and appropriately
aligned.
• Areas in which the two shells diverged from each other were as-
certained, i.e., they diverged by more than a threshold distance, which can
be set by the user. Each of the shells was treated as a collection of vertices.
The software scanned each facial shell vertex by vertex, looking for and
removing the pairs of vertices that were closer to each other than the se-
lected threshold. This resulted in two facial shells each of which contained
only the essentially divergent areas.
• Small areas of no or little importance were removed. This was
set manually at 200 vertices and often required further manual selection to
remove such areas.
• The resulting divergence maps were separated by connecting
the regions mathematically for further analysis.
• Finally, volumes denoting three-dimensional shape changes
were created by manually selecting two matching regions and invoking
the appropriate software function in the macro.
300
Kau and Richmond
• The procedure for creating a volume-difference shell begins by
selecting a surface region, which refers to one of the two facial shells be-
ing compared, and extruding it toward the second reference shell by a dis-
tance of 1.2d., where d is the maximum distance between the selected
region and the reference shell. This creates an extruded volume shell. The
reference shell then is extruded towards the selected region by 1.2d., to
produce a second volume shell. The Boolean difference between the two
extruded volume shells is determined further, resulting in the desired vol-
ume difference shell.
Facial volumes were calculated and separated into regional areas
(Fig. 3) of the face for different groups, i.e., gender differences and treat-
ment types. The accuracy of the measurement method was independently
Verified and found to be within 1.5%.
ºne
3D Facial: Volume & Area Changes ºb ºx
Find differences Tolerance (mm): | 0.5 #
Selectsmall areas small area (faces): 200 =
Select manually Unselect all
- Shell name: T5vs.T1Nose
Number of vertices: 4121 Area: 4341.26 mm.º.
Number of faces: 82.38 Volume: 2220.94 mmº-3
Figure 3. Regional areas of change isolated by boundary boxes. These tools were
developed in our craniofacial laboratory by Dr. Alexei Zhurov. Information is
obtained by pressing the area of interest. For example, if the area of the nose is
chosen, the analysis shows that the volume is 2220.94 mm', with a surface area
of 4341.26 mm2.
In general, the nose exhibited the most consistent increase in vol-
ume. There was a gradual increase in size in a triangular pattern, with






301
3D Soft Tissue Perspective
the base becoming broader and the apex becoming thicker. The nose pro-
jected at the nasal tip anteriorly as well (Fig. 4). Males had a larger mag-
nitude of volumetric growth.
t2ustí
SERIES A
T3WST1
SERIES B
| - º T4 vs.T1
- SERIES C.
Figure 4. Surface areas and shape changes to the face as it develops. This series
of slides shows the composite average face of males (n = 33) who received no
orthodontic treatment. Blue areas indicate negative changes and red areas indicate
positive changes. No studies to-date have shown such surface changes.
SUMMARY
The procedures for using 3D technology to illustrate surface
changes that represent facial growth and change are developing rapidly
and are likely to become routine tools for assessing individuals over the






302
Kau and Richmond
next ten years. This technology combined with cone-beam computerized
technology can construct a fully three-dimensional soft tissue and skel-
etal picture that provides clinicians with the opportunity to improve their
knowledge of facial changes due to growth and orthodontic treatment.
REFERENCES
Ackerman JL, Proffit WR, Sarver DM. The emerging soft tissue paradigm
in orthodontic diagnosis and treatment planning. Clin Orthod Res
1999; 2:49-52.
Aldridge K, Boyadjiev SA, Capone GT, DeLeon VB, Richtsmeier JT.
Precision and error of three-dimensional phenotypic measures ac-
quired from 3dMD photogrammetric images. Am J Med Genet A
2005; 138:247-253.
Arnett G, Bergman RT. Facial keys to orthodontic diagnosis and treat-
ment planning. Part 1. Am J Orthod Dentofacial Orthop 1993;103:
299-312.
Blais F. Review of 20 years of range sensor development. J Electron Imag
2004; 13:231-240.
Broadbent BHS, Broadbent BHJ, Golden WH. Bolton Standards of Dento-
facial Developmental Growth. St Loius, CV Mosby, 1975.
Case CS. The questions of extractions in orthodontics. Am J Orthod
1964:50:658–691.
Hajeer MY, Millett DT, Ayoub AF, Siebert JP Applications of 3D imaging
in Orthodontics: Part I. J Orthod 2004:31:62-70. -
Harrison JA, Nixon MA, Fright WR, Snape L. Use of hand held laser
scanning in the assessment of facial swelling: A preliminary study. Br
J Oral Maxillofac Surg 2004:42:8-17.
Kau CH, Richmond S, Palomo JM, Hans MG. Three-dimensional cone
beam computerized tomography in orthodontics. J Orthod 2005a:32:
282–93. :
Kau CH, Richmond S, Savio C, Mallorie C. Measuring adult facial mor-
phology in three dimensions. Angle Orthod 2006a;76:771-776.
Kau CH, Richmond S, Zhurov AI. Application of 3-dimensional soft tis-
Sue imaging in orthodontics. In: McNamara JA Jr., Kapila S, eds.
Digital Radiography and Three-Dimensional Imaging. Craniofacial
Growth Series. Department of Orthodontics and Pediatric Dentistry
and the Center for Human Growth and Development, The University
of Michigan, Ann Arbor, 2006b;43:159-179.
Kau CH, Richmond S, Zhurov AI, Knox J, Chestnutt I, Hartles FR,
Playle R. Reliability of measuring facial morphology using a 3-
303
dimensional laser scanning system. Am J Orthod Dentofacial Orthop
2005b; 128: 424–430.
Kau CH, Zhurov AI, Bibb R, Hunter ML, Richmond S. The investiga-
tion of the changing facial appearance of identical twins employing
a three-dimensional laser imaging system. Orthod Craniofacial Res
2005c,8:85–90.
Kau CH, Zhurov AI, Scheer R, Bouwman S, Richmond S. The feasibility
of measuring three-dimensional facial morphology in children. Or-
thod Craniofac Res 2004;7:198-204.
Mah J. 3D imaging in private practice. Am J Orthod Dentofacial Orthop
2002;121:14A.
Mah J, Bumann A. Technology to create the three-dimensional patient re-
cord. Semin Orthod 2001;7:251-257.
Mah J, Enciso R. The virtual craniofacial patient. In: McNamara JA Jr,
ed, Information Technology and Orthodontic Treatment. Craniofacial
Growth Series, Center for Human Growth and Development, The
University of Michigan, Ann Arbor 2003:40: 167-184.
Mah J, Hatcher D. Three-dimensional craniofacial imaging. Am J Orthod
Dentofacial Orthop 2004;126:308-309.
Papadopoulos MA, Christou PK, Athanasiou AE, Boettcher P. Zeilhofer
HF, Sader R, Papadopoulos NA. Three-dimensional craniofacial re-
construction imaging. Oral Surg Oral Med Oral Path Oral Radiol En-
dod 2002;93: 382–393.
Sarver DM. Esthethic Orthodontics and Orthognathic Surgery. St Louis,
Mosby, 1998.
304
3D IMAGING FOR EARLY DIAGNOSIS AND
ASSESSMENT OF TREATMENT RESPONSE
Lucia H. S. Cevidanes
Alexandre Motta
Martin Styner
Ceib Phillips
Three-dimensional image analysis from cone-beam computed tomog-
raphy (CBCT) scans offers improved diagnostic information and, more
importantly, it provides a better way to evaluate the changes created by
treatment and the adaptive displacement and remodeling that occur (Cevi-
danes, 2005a). As new technologies are being applied to orthodontic clini-
cal practice, the previous emphasis on 2D cephalometric analysis of hard
tissue and dental occlusion relationships has shifted to using 3D technol-
ogy to assess both soft and hard tissue facial relationships.
The new emphasis on soft tissues as the limiting factors in treat-
ment and soft tissue relationships in establishing the goals of treatment
combine to produce major changes in diagnosis and treatment planning.
The shift away from the earlier emphasis on dental occlusion and hard
tissue relationships establishes a new paradigm for our understanding of
response to treatment in which cephalometric analysis plays a smaller
role and clinical examination of facial proportions a more important one.
Randomized clinical trials are beginning to provide data that will allow
clinicians to consider the efficiency and the effectiveness of early (pre-
adolescent) treatment versus later (adolescent) treatment. The challenge
of 3D analysis of soft tissue growth and response to treatment remains, as
soft tissue changes are determined by the muscles attached to the skeletal
structures. The soft tissue surfaces from spiral CT, cone-beam CT, laser
Scanning or 3D photography only contain information about the skin and
facial appearance; they do not allow us to assess the underlying muscles.
Although the growth of the face and jaws can be measured in
two dimensions, structural changes at specific locations are not reflected
sufficiently in cephalometric measurements (Tulloch et al., 1990; Gha-
fari et al., 1998; Harrell et al., 2002; Togashi et al., 2002; Turpin, 2002).
The SNB angle has been called a poor indicator of the effectiveness of
functional jaw orthopedics, because the increase in lower anterior fa-
cial height due to correction of a Class II relationship camouflages the
305
3D Imaging for Early Diagnosis
concurrent increase in mandibular length. Nevertheless, in a recent sys-
tematic review of the literature, studies of mandibular changes produced
by functional appliance in Class II malocclusion patients still report that
the mandibular position to the cranial base as measured by the SNB angle
was not impacted in a clinically significant way by functional jaw ortho-
pedics (Cozza et al., 2006).
During growth and/or response to orthopedic treatment, the man-
dibular rami and condyles develop in many directions relative to all pos-
sible individual variations in the nasomaxilla and middle cranial fossae
anatomic patterns (Björk and Skieller, 1983, 1985; Coben, 1998; Wilhelm
et al., 2001). However, identifying the rami’s role relative to skeletal com-
pensations in maxillomandibular discrepancy corrections cannot be ac-
complished by analyzing population norms, angles or interlandmark dis-
tances (Lewis et al., 1985; Dibbets, 1996).
The radiation dose in spiral CT (Mozzo et al., 1998; Mah and
Hatcher, 2004), the high cost of CT and magnetic resonance imaging (MRI;
Kawamata et al., 1998; Ludlow et al., 2003; Cevidanes et al., 2005a,c),
and the lack of simple 3D image analysis tools for clinical use have limited
their usefulness in the study of post-treatment changes. Recently, CBCT
has been described as the 3D method of choice for maxillofacial imaging
in dentistry because of the reduction in equipment needs, operating costs,
radiation and acquisition time; the ability to obtain high-resolution imag-
ing for facial bones and teeth; and the possibility to obtain from only one
exposure the usual set of orthodontic lateral, frontal, panoramic or periapi-
cal radiographs (Mozzo et al., 1998; Ludlow et al., 2003; Mah and Hatch-
er, 2004). CBCT allows 3D reconstruction with complete visualization of
the facial structures and the capacity to make measurements accurately
within 0.36 mm in any direction (Cevidanes et al., 2005a). Importantly,
a quantitative assessment of condylar rotation/displacement that was not
feasible previously with 2D technology now can be accomplished using
CBCT Scans.
Evidence that the rami and the condyles might play compensa-
tory roles in skeletal growth and in response to orthodontic treatment
comes from the findings of implant and histological studies (Björk and
Skieller, 1983, 1985; Petrovic et al., 1990). Quantitative descriptors of
3D rami skeletal compensations now can assess the manner of assembly
of structural components involved in facial morphogenesis. The quantifi-
cation of changes in skeletal morphology includes two significant devel-
opmental processes during bone growth: primary or secondary displace-
ment and bone surface remodeling (Petrovic et al., 1990). The alteration
306
Cevidanes et al.
of landmark position during growth and treatment involves the simultane-
ous processes of bone surface remodeling, primary displacement by indi-
vidual bone growth, and secondary displacement by the growth of adja-
cent structures. In this chapter we will present diagnostic findings, growth
assessment and Fränkel Functional Regulator II treatment outcomes based
on 3D imaging techniques.
VISUALIZATION AND MEASURING OF
SUPERIMPOSED 3D MODELS
Imsel software, which was developed by the University of North
Carolina (UNC) Department of Neurosurgery, is used to convert the New-
Tom DICOM files to a format (GIPL) that is readable by other public soft-
ware imaging processing tools. Voxels are reformatted for an isotropic
of 0.5 x 0.5 x 0.5 mm, avoiding increases in image file size that would
require greater computational power and user interaction time. In addi-
tion, Imagine software (developed by UNC Computer Sciences) is used
to compact the size of GIPL format files created with Imsel. From a set of
more than 300 axial, lateral and anteroposterior cross-sectional slices for
each image acquisition, the segmentation of the cranial base and mandible
is performed with InsightSNAP (developed by UNC Computer Scienc-
es). Segmentation is the process of constructing 3D models by examining
cross-sections of a Volumetric data set to outline the shape of structures
(Cevidanes, 2005a). A key feature of the CBCT images is the ability to
navigate through the volumetric data set in any of the orthogonal slice
windows. After segmentation, a 3D graphical rendering of the volumet-
ric object allows navigation between voxels in the volumetric image by
Zooming, rotating and panning.
To evaluate growth and treatment changes of a specific patient,
images taken at different ages are superimposed using a fully automat-
ed method of voxel-wise registration to avoid observer-dependent lo-
cation of points identified from overlap of anatomic landmarks. In that
the growth of the anterior cranial fossae and the ethmoid bone is com-
pleted in early infancy, its surfaces are used in the registration procedure
where the software compares the grey level intensity of each voxel be-
tween two CBCT images. In this way, the anterior cranial base of the
CBCT images are used as reference for superimposing different time
points (Fig. 1). As Baumrind and colleagues (1983) pointed out, these
Superimposition methods determine that the 3D changes described are
relative to the individual cranial base, and not absolute displacement.
307
3D Imaging for Early Diagnosis
Figure 1. The superimpositions of 3D models of growing children dif-
fer from adult patients for which the whole surface of the cranial base is
used for registration. For growth assessment we use the anterior cranial
fossa surfaces for registration, as the superimpositions describe growth
relative to the individual cranial base.
The next step in the analysis involves overlaying the 3D models
surfaces with another tool, i.e., Fltk/3DSOV, which allows different de-
grees of transparency to visually assess the boundaries of the mandibular
rami and condyles between superimposed models of two different time
points. The visualization of the superimpositions clearly identifies the lo-
cation, magnitude and direction of mandibular displacements, and allows
quantification of vertical, transverse, and anteroposterior bone displace-
ments and the remodeling that accompanies both growth and response to
treatment.
Measurements of the surface distances between different phases
can be computed in color-coded maps using the Mesh Valmet tool. This

308
Cevidanes et al.
procedure computes thousands of distances in millimeters between sur-
face's triangles, and quantifies how much, on average, the two surfaces
differ from each other. After combining all 3D models at various time
points, specific regions of interest such as the chin prominence, condyles
and the posterior border of the rami can be selected and analyzed.
The color maps can be used to indicate inward (blue) or outward
(red) displacement between overlaid structures. Green indicates the ab-
sence of surgical displacement. For example, mandibular advancement
forward displacement would be shown in red in the anterior surfaces and
blue in the posterior surfaces. A medial displacement of the condyles and
rami will display red medial surfaces and blue lateral surfaces. No post-
Surgical change would be green. This method has been published and used
since 2004 (Bailey et al., 2004; Cevidanes et al., 2005a).
EARLY 3D DIAGNOSIS AND TREATMENT PLANNING
Applications of early 3D diagnosis in orthodontics include den-
tal root inclination and torque, impacted and supernumerary tooth posi-
tions (Fig. 2), thickness and morphology of bone at sites of mini-implants
for anchorage, and osteotomy sites in surgical planning. Findings such as
resorption, hyperplasic growth, displacement, shape anomalies of man-
dibular condyles and morphological differences between the right and
Figure 2. The close relationship of upper canines and lateral incisors
clearly can be seen with 3D reconstruction.

309
3D Imaging for Early Diagnosis
left sides emphasize the diagnostic value of computed tomography acqui-
sitions. Furthermore, relationships between soft tissues and the airway can
be assessed in three dimensions.
For cases of severe asymmetry, e.g., hemifacial microsomia,
knowing the degree of mandibular involvement is crucial in planning the
timing and choice of treatment such as distraction osteogenesis, costocon-
dral grafting, functional orthopedics or observation of growth until post-
pubertal surgical reconstruction (Fig. 3). Treating these cases with early
distraction osteogenesis has been controversial as it was not clear whether
or not treating early makes a difference or whether facial asymmetry pro-
gresses with growth (Posnick, 1998; Kau et al., 2006; Maeda et al., 2006).
In Figure 4, the growth superimpositions from a period of 1.5 years dur-
ing the ascendant part of the pubertal growth spurt curve of a child who
had previous costocondral grafting show changes in facial deformity that
indicate surgical correction prior to completion of growth. Mirror images
Figure 3. A four-year-old patient with hemifacial microsomia type 3 for whom
important distraction osteogenesis treatment planning decisions need to be
made. While a costocondral graft might establish a joint on the affected side.
its future growth is unpredictable. For distraction osteogenesis, the challenge of

310
Cevidanes et al.
Surface distance cºlº map ºn mºn
Figure 4. Growth superimpositions of a period of 1.5 years during the ascendant
part of the pubertal growth spurt curve of a hemifacial microsomia patient who
received a costocondral graft at 7.5 years of age. Registered on the anterior cra-
nial fossa, the 3D models on the left show vertical and lateral growth through the
Semi-transparent overlay. The models in white were taken at age 9.5 years of age
and in red at 11 years of age. The 3D models on the right show the 3D color maps
of the surface distances between the two time points displaying the lateral growth
of the costocondral graft.
of 3D models now can aid surgical planning and assessment of the severity
of the asymmetry (Figs. 5-7; Chapuis et al., 2004). -
The radiographs conventionally used for orthodontic diagnosis,
e.g., lateral cephalograms and panoramic X-rays, might fail to early di-
agnose conditions such as supranumerary teeth (Fig. 8), Odontomas (Fig.
9), and aid in the differential diagnosis of failure of eruption as shown in
Figures 10 and 11.
* - Figure 3. Continued.
stimulating growth in 3D still remains. For functional orthopedics the question is
how much can the deficient/absent muscles on the affected side aid mandibular
growth. A fourth option is to observe growth until post-pubertal surgical recons-
truction.

3 ||
3D Imaging for Early Diagnosis
Figure 5. Mirror images of the 3D models. A-D: Left and right sides of the
original model were inverted through a constructed midsagittal plane, building
a mirror model that was superimposed on the actual patient model. The white
models are the original images, and the mirror images are shown in semi-trans-
parent red. The images on the left were taken at 9.5 years of age and on the

312
Cevidanes et al.
Figure 6. Visualization of right and left differences of soft tissue structures. The
model in red shows the original patient’s morphology, and the model in green
shows the surface distances between the superimposition of the original relative
to its mirrored image. Soft tissue asymmetry can be seen especially in the nasal
and orbital regions.
Figure 7. Surgical simulation with CMF tool to plan displacement of each of the
colored segments (Chapuis et al., 2004).
* - Figure 5. Continued.
right at 11 years of age. E and F. The right side of the mandible was mirrored to
the left using a CMF tool (Chapuis et al., 2004). The lateral position of the costo-
Condral graft and the absence of an articular fossa can be seen.


3.13
3D Imaging for Early Diagnosis
Figure 8. 3D imaging used for early diagnosis of a supranumerary tooth that
could not be seen in the panoramic X-ray.
The peak adolescent growth spurt and early permanent denti-
tion have been described as the gold standard for the timing of orthodon-
tic treatment. This is because there is some growth (especially vertical
growth) remaining to assist treatment, permanent teeth are available for
final positioning, treatment usually ends as the adolescent growth spurt
ends, and the shorter treatment time lowers the burden of treatment.
—).
Figure 10. Frontal and lateral views of a severe case of the failure of the posterior
teeth to erupt that show the lingual inclination of premolars and molars.

3.14
Cevidanes et al.
Figure 9. CBCT visualization with Dolphin 3D (Dolphin Imaging and Manage-
ment, California) shows the presence of odontoma tooth apical to the upper left
canine.


3.15
3D Imaging for Early Diagnosis
Figure 11. The orthodontist waited for the eruption of the permanent teeth until
the patient was 16 years old, and then referred the patient as a failure of eruption
case. Extraction of premolars and the upper left canine opened space for traction
of the impacted teeth.
The question of whether it is advantageous to treat before or after
the gold standard time now can be addressed using 3D imaging. We have
recently used 3D imaging to investigate changes, relapse and stability after
surgical correction for both late teen and adult treatment and relapse and
stability after treatment with the Fränkel Regulator II during the ascendant
part of the growth spurt curve (Fig. 12).

316
Cevidanes et al.
Figure 12. Pretreatment profile changes (upper left), after treatment with the
Fränkel Regulator II appliance (upper middle) and two-years post-treatment
(upper right). The color maps of the superimposition of pre- and post-treatment
MRIs show the vertical, lateral and anterior growth of the mandibular rami.
3D ASSESSMENT OF MANDIBULAR GROWTH AND
RESPONSE TO ORTHOPEDIC TREATMENT
We have applied generalizable methods for 3D landmark data,
focusing on the morphogenic basis of Class II malocclusion. We ad-
dressed the biologic question of whether mandibular rami growth is al-
tered relative to its equivalents by using 3D imaging to measure the
changes in the mandibular rami relative to how far the middle cranial
fossae places the nasomaxillary complex anteriorly and how widely it
places the two condyles bilaterally (Björk and Skieller, 1983; Bhat and
Enlow, 1985; Cevidanes et al., 2003). We used 3D magnetic resonance

317
3D Imaging for Early Diagnosis
imaging to study relationships among craniofacial components during the
pubertal growth spurt and in response to Fränkel appliance therapy (Cevi-
danes et al., 2005b).
The subjects in this study were 78 Brazilian children, 28 of whom
were treated for Class II malocclusions, 25 of whom had untreated Class
II malocclusions and 25 children who had untreated normal occlusions.
Two high-resolution magnetic resonance images with 1 mm isotropic
voxel resolution were taken for each subject for a total of 156 images;
one image was taken at the beginning of treatment (T1) and the second 18
months post-treatment or after an 18-month observation period (T2). De-
velopmentally, all subjects were at the beginning of their pubertal growth
spurt as diagnosed by hand and wrist x-rays.
We used a Procrustes geometric transformation of 3D skeletal
landmarks to assess growth or treatment alterations from T1 to T2. The
standard biometric approach of Procrustes, more than a description rel-
ative to stable intraosseous reference points like implants, analyzes the
relative displacement of key counterpart components during growth and
response to treatment. Therefore, the Procrustes fit of each subjects’ land-
mark coordinates showed the displacement of landmarks relative to all
landmarks included in the 3D models, controlling for the variations in
rotation, translation, and scale (Slice, 2001).
Not only were the principal dimensions of skeletal alterations dif-
ferent for treated and untreated Class II and normal-occlusion subjects,
but the PCA scattergrams also showed considerable individual variabil-
ity in growth and response to treatment alterations (Chen et al., 2002).
A remarkable product of this study was the visualization using deforma-
tion grids of the underlying 3D patterns of relative skeletal alterations of
the mandibular rami. The non-significant configuration changes from T1
to T2 in the deformation grids for both the untreated Class II controls
and the normal-occlusion subjects showed maintenance of the landmark
configuration with growth, with slight relative mandibular advancement
characteristic of the differential maxillary/mandibular growth at the begin-
ning of the pubertal growth spurt. For the treated group, the skeletal altera-
tions visualized in the T1-T2 deformation grids were significantly differ-
ent: P .001 for treated Class II subjects vs. untreated Class II controls
and for treated Class II subjects vs. subjects with normal occlusions. The
differential anteroposterior location of 3D landmarks observed in the de-
formation of coronal/axial gridlines showed more forward (anterior) rami
alignment relative to their counterparts in the posterior nasomaxilla and
3.18
Cevidanes et al.
the middle cranial fossae. The deformation grids also showed differential
vertical location of landmarks with increased relative mandibular rami
vertical dimension. The striking statistically significant 3D skeletal altera-
tions shown graphically by the deformation grids are considered a favor-
able skeletal response to treatment with the Fränkel Regulator II aimed at
increasing relative mandibular growth.
REFERENCES
Bailey LJ, Cevidanes LHS, Proffit WR. Stability and predictability of or-
thognathic surgery. Am J Orthod Dentofacial Orthop 2004; 126:273-
277.
Baumrind S, Moffitt FH, Curry S. The geometry of three-dimension-
al measurement from paired coplanar x-ray images. Am J Orthod
1983;84:313–22.
Bhat M, Enlow DH. Facial variations related to headform type. Angle Or-
thod 1985:55:269-280.
Björk A, Skieller V. Normal and abnormal growth of the mandible. A syn-
thesis of longitudinal cephalometric implant studies over a period of
25 years. Eur J Orthod 1983;5:1-46.
Björk A, Skieller V. Contrasting mandibular growth and facial develop-
ment in long face syndrome, juvenile rheumatoid polyarthritis, and
mandibulofacial dysostosis. J Craniofacial Genet Dev Biol (Supply
1985; 1:127-138.
Cevidanes LHS, Bailey LJ, Tucker GR, Styner MA, Mol A, Phillips CL,
Proffit WR, Turvey T. Superimposition of 3D cone-beam CT models
of orthognathic surgery patients. Dentomaxillofac Radiol 2005a:34:
369-375.
Cevidanes LHS, Franco AA, Gerig G, Proffit WR, Slice DE, Enlow DH,
Yamashita HK, Kim Y, Scanavini MA, Vigorito JW. Assessment of
mandibular growth and response to orthopedic treatment with 3-di-
mensional magnetic resonance images. Am J Orthod Dentofacial Or-
thop 2005b;128:16-26.
Cevidanes LHS, Franco AA, Gerig G, Proffit WR, Slice DE, Enlow DH,
Lederman HM, Amorim L, Scanavini M, Vigorito JW. Comparison of
relative mandibular growth vectors with high-resolution 3-dimensio-
nal imaging. Am J Orthod Dentofacial Orthop 2005c;128: 27–34.
Cevidanes LHS, Franco AA, Scanavini MA, Enlow DH, Vigorito JW,
Proffit WR. Clinical outcomes of Fränkel appliance therapy assessed
3.19
3D Imaging for Early Diagnosis
with counterpart analysis. Am J Orthod Dentofacial Orthop 2003;
123:379-87.
Chapuis J, Rudolph T. Borgesson B, De Momi E, Pappas IP. Hallermann
W. Schramm A, Caversaccio M. 3D surgical planning and navigation
for CMF surgery. Proc SPIE Med Imaging 2004:403-410.
Chen JY, Will LA, Niederman R. Analysis of efficacy of functional ap-
pliances on mandibular growth. Am J Orthod Dentofacial Orthop
2002; 122:470-476.
Coben SE. The spheno-occipital synchondrosis: The missing link between
the profession’s concept of craniofacial growth and orthodontic treat-
ment. Am J Orthod Dentofacial Orthop 1998; 114:709–712.
Cozza P. Baccetti T. Franchi L., De Toffol L. McNamara JA Jr. Mandibular
changes produced by functional appliances in Class II malocclusion:
A systematic review. Am J Orthod Dentofacial Orthop 2006;129:599.
e1-599.e12.
Dibbets JM. Morphological associations between the Angle classes. Eur J
Orthod 1996; 18:111-118.
Ghafari J, Baumrind S, Efstratiadis SS. Misinterpreting growth and treat-
ment outcome from serial cephalographs. Clin Orthod Res 1998;
1:102-106.
Harrell WE Jr, Hatcher DC, Bolt RL. In search of anatomic truth: Three-
dimensional digital modeling and the future of orthodontics. Am J Or-
thod Dentofacial Orthop 2002;122:325-330.
Kau CH, Zhurov A, Richmond S, Bibb R, Sugar A, Knox J, Hartles F.
The 3-dimensional construction of the average 11-year-old child
face: A clinical evaluation and application. J Oral Maxillofac Surg
2006;64:1086–1092.
Kawamata A, Fujishita M, Kuniteru N, Kanematu N, Niwa K, Langlais
R. Three-dimensional computed tomography evaluation of postsurgi-
cal condylar displacement after mandibular osteotomy. Oral Surg Oral
Med Oral Pathol Oral Radiol Endod 1998:85:371-376.
Lewis AB, Roche AF, Wagner B. Pubertal spurts in cranial base and man-
dible: Comparisons between individuals. Angle Orthod 1985;55:17-
30.
Ludlow JB, Davies-Ludlow LE, Brooks SL. Dosimetry of two extraoral
direct digital imaging devices: NewTom cone beam CT and Orthophos
Plus DS panoramic unit. Dentomaxillofac Radiol 2003:32:229–234.
Maeda M, Katsumata A, Ariji Y, Muramatsu A, Yoshida K, Goto S, Ku-
rita K, Ariji E. 3D-CT evaluation of facial asymmetry in patients with
320
Cevidanes et al.
maxillofacial deformities. Oral Surg Oral Med Oral Pathol Oral Ra-
diol Endod 2006; 102:382–390.
Mah J. Hatcher D. Three-dimensional craniofacial imaging. Am J Orthod
Dentofacial Orthop 2004; 126:308-309.
Mozzo P. Procacci C, Tacconi A, Martini PT, Andreis IA. A new volumet-
ric CT machine for dental imaging based on the cone-beam technique:
Preliminary results. Eur Radiol 1998;8:1558-1564.
Petrovic A, Stutzmann J, LaVergne J. Mechanism of craniofacial growth
and modus operandi of functional appliances: A cell-level and cyber-
netic approach to orthodontic decision making. In: Carlson DS, ed.
Craniofacial Growth Theory and Orthodontic Treatment. Craniofa-
cial Growth Series, Center for Human Growth and Development, The
University of Michigan, Ann Arbor 1990:23:13-74.
Posnick JC. Surgical correction of mandibular hypoplasia in hemifacial
microsomia: A personal perspective. J Oral Maxillofac Surg 1998;
56:639–50.
Slice DE. Landmark coordinates aligned by Procrustes analysis do not lie
in Kendall’s shape space. Syst Biol 2001:50:141-149.
Togashi K, Kitaura H, Yonetsu K, Yoshida N, Nakamura T. Three-dimen-
sional cephalometry using helical computer tomography: Measure-
ment error caused by head inclination. Angle Orthod 2002;72:513–
520.
Tulloch JFC, Medland W, Tuncay OC. Methods used to evaluate growth
modification in Class II malocclusion. Am J Orthod Dentofacial Or-
thop 1990;98:340-347.
Turpin DL. Research topics: Past, present, and future. Am J Orthod Den-
tofacial Orthop 2002;122:341.
Wilhelm BM, Beck FM, Lidral AC, Vig KWL. A comparison of cranial
base growth in Class I and Class II skeletal patterns. Am J Orthod
Dentofacial Orthop 2001; 119:401–405.
321
RSITY OF MICHIGA
iiiiI.
1
OO64 1797
Illi
3.90
15



- ºffli H º º º --- --- --- -
################### - - - ############################## *
H H
- Hº:
Hº º
º H º
- --------- º ##################### # - H.
#: ######### H - ##############（ ---
# Hä ############ - （ - （
################# ####################### º --------- º - ºffli
ºffli （ --- （
#: ############ - # iii. - HºHº H -
- ----------- H - ºffl
- - --- # # H Hº
ºffl Hº
- - - --- - （ # - ###################
ºffl Hº: ######## # #º
######################### ###########
E-º-º-º-º-º: º: º --- -- ---
i ºffli i- ºffl ºffl -
--- - ºffli - --- - - º
----------------- --- H
####### #################
---
-
ºffl
-
#: - H º ######################
ºffl ---------- ºffli ºffl ###############
--~~ # --- H
º: H
Hº: H. （ º
----------- -- （
# º º
（
Hiß
- H - （
#. ºffli ####################
（ -- Hº:
ºffl º # -
---
º
---
- -------- ---
Häß
|-- --~~~
- ºffli ####### #
############ ºffli
ºffl ºffl -
ºffl -- - ºffli º
ºffl - --- --- - -
-------
--
---
Eº
ºfflº
ºffli -
- -------------
ºffl
---
Eºi
---
--~-
--------
ºfflº
H
º
#
#
º
H--
---
ºfflº
---
º
-
º º
########## -- - -
ºffl --- - ºffli
- - - ---
ºffl -
- --- º
ºffli ºffl ºffli -
ºffli ºffl ---
ºfflº ºffl --- ºfflº
ºffl - - - -
--- - º --- - º
------ º - -
º # ºffl -
--- --- --- º - - --- - - --------
ºffli ºffl ##################.
º: Eº ºffl ºffl - Häß º:
º: - ºffl º - º --- - - º º ºffli º --- º
---
--~~~~
ºffl
- --
ºffli

-