( ( 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 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 -