DIRECTORY: AAO OFFICERS AND ORGANIZATIONS - orto

DIRECTORY: AAO OFFICERS AND ORGANIZATIONS

American Association of OrthodontistsAnnual Session April 30-May 4, 2010,Washington, DC401 N Lindbergh Blvd, St Louis,MO 63141 (314) 993-1700President, Robert J. Bray, Sommers Point, NJPresident-Elect, Lee Graber, Kenilworth, ILSecretary-Treasurer, Michael B. Rogers,Augusta, GASpeaker of the House, Keith Levin, Winnipeg,MB, CanadaExecutive Director, Chris P. Vranas, 401 NLindbergh Blvd, St. Louis, MO 63141; tele-phone, 800-424-2841; e-mail, [email protected]

Great Lakes Association ofOrthodontistsNext meeting September 30-October 3,2010, Palm Beach, FLPresident, Ronald S. Good, Mount Lebanon,PAPresident-Elect, Robert F. Good II, Washing-ton, PAVice President, Jerry R. Hickman, Indianapo-lis, INSecretary-Treasurer, Michael W. Sherman,Thornhill, ON, CanadaTrustee, John F. Buzzatto, Allison Park, PAExecutive Director, Debbie Nunner, 17 S HighSt, Suite 200, Columbus, OH 43125- 3458;telephone, 877-274-6420; fax, 614-221- 1989;e-mail, [email protected]

Middle Atlantic Society of OrthodontistsNext meeting September 30-October 3,2010President, Constance Greeley, Wilmington,DEPresident-Elect, Natalie M. Parisi, Wyomiss-ing, PASecretary, Luis Toro, Jr., Fajardo, Puerto RicoTreasurer, Steven Siegel, Glen Burnie, MDTrustee, Nahid Maleki, Washington, DCExecutive Director, Anita Field, 17 S High St,Suite 200, Columbus OH 43215; telephone,866-748-6276; fax, 614-221-1989

Midwestern Society of OrthodontistsNext meeting September 22-26, 2010,Colorado Springs, COPresident, John S. Kanyusik, Mankato, MNPresident-Elect, Jacqueline M. Miller, Wash-ington, MOVice President, Babette Cohen, Winnipeg,Manitoba, CanadaSecretary-Treasurer, Ross L. Crist, SiouxFalls, SD

Trustee, Brent Larson, Minneapolis, MNExecutive Director, Kristi Burmeister, Bur-meister & Associates Inc, 3260 Upper BottomRd, St Charles, MO 63303; telephone, 636-922-5551; fax, 636-244-1650; e-mail,[email protected]

Northeastern Society of OrthodontistsNext meeting November 11-14, 2010,Montreal, Quebec, CanadaPresident, Lee P. Erickson, Bedford, NS,CanadaPresident-Elect, Philip M. Mansour, Goffs-town, NHSecretary, John Callahan, Fayetteville, NYTreasurer, Dennis C. Hiller, Littleton, NHTrustee, Hugh R. Phillis, Nashua, NHExecutive Director, Anita B. Craig, 410 NLindbergh Blvd, St. Louis, MO 63141; tele-phone, 800-424-2841; e-mail, [email protected]

Pacific Coast Society of OrthodontistsNext meeting October 9-13, 2010,Honolulu, HIPresident, Lili K. Horton, Honolulu, HIPresident-Elect, Lesley Williams, Delta, BC,CanadaSecretary-Treasurer, Robert M. Merrill, EastWenatchee, WATrustee, Robert E. Varner, Roseburg, ORExecutive Director, Jill Nowak, 401 LindberghBlvd, St Louis, MO 63141; telephone, 800-424- 2841; e-mail, [email protected]

Rocky Mountain Society ofOrthodontistsNext meeting November 4-7, 2010, IndianWells, CAPresident, Gary Kloberdanz, Greeley, COPresident-Elect, Douglas Bennion, Billings, MTVice President, Craig L. Coombs, Layton, UTSecretary-Treasurer, Bret E. Mooso, IdahoFalls, IDTrustee, Morris N. Poole, Logan, UTExecutive Secretary, Dorothy J. Shadrick, As-sociated Conferences & Exposition Mgmt,6000EEvans Ave, Suite 3-205, Denver, CO 80222;telephone 303-758-9611; fax 303-758-9616;e-mail, [email protected]

Southern Association of OrthodontistsNext meeting September 22-26, 2010Colorado Springs, COPresident, R. R. Reed Jr, Ocala, FLPresident-Elect, Robert D. Calcote, Charles-ton, SCSecretary-Treasurer, Rod Klima, Burke, VA

Trustee, Michael B. Rogers, Augusta, GATrustee Designate: DeWayne McCamish,Chattanooga, TNExecutive Director, Sharon Hunt, CAE, 32Lenox Pointe NE, Atlanta, GA 30324-3169;telephone, 800-261-5528; fax 404-261-6856;e-mail, [email protected]

Southwestern Society of OrthodontistsNext meeting September 28-31, 2010Austin, TXPresident, Stephen D. Robirds, Austin, TXPresident-Elect, Michael L. Mizell, Houston, TXVice President, Richard W. Boyd, Sugar Land,TXSecretary-Treasurer, J. Clifton Alexander,Dallas, TXTrustee, Gayle Glenn, Dallas, TXExecutive Director, Judith K. Salisbury, 10032Wind Hill Dr, Greenville, IN 47124; tele-phone, 812-923-2700; fax, 812-923-2900;e-mail, [email protected]

American Board of Orthodontics 410 NLindbergh Blvd, Suite 308, St Louis, MO63141 (314)432-6130President, Peter M. Greco, Philadelphia, PAPresident-Elect, Jeryl D. English, Houston,TXSecretary-Treasurer, Barry S. Briss, Boston,MADirector, Scott A. Jamieson, Marquette, MIDirector, Marvin C. Kastrop, Billings, MTDirector, Paul T. Castelein, Princeton, ILDirector, Eladio DeLeon, Jr., Augusta, GADirector, Steven Dugoni, So. San Francisco, CAImmediate Past-President, John E. Grubb,Chula Vista, CAExecutive Director, Christine L. Eisenmayer,St Louis, MO

American Association of OrthodontistsFoundation Board of DirectorsPresident, Stephen E. Hershey (GLAO)President-Elect, James H. Gallagher (RMSO)Executive Vice President, Robert W. Hazel,401 N Lindbergh Blvd, St Louis, MO 63141

College of Diplomates of the AmericanBoard of OrthodonticsNext meeting July 14-18, 2010,The Samoset Resort, Rockport, MEPresident, Phillip Markin, Columbia, MDPresident-Elect, Dorothy Whalen, Glen Cove,NYSecretary, Jim Morrow, McKinney, TXTreasurer, Rodney Hyduk, Troy, MIExecutive Director, Karen Seiler, 3260 UpperBottom Rd, St Charles, MO 63303; telephone,636-922-5551; fax, 636-244-1650; e-mail,[email protected]

ESTATE PLANNING & PLANNED GIVING

Estate Planning: The AAO Foundation offers information on estate planning to AAO members and their advisors on acomplimentary basis and at no obligation.

Planned giving: Persons who are contemplating a gift to the AAO Foundation through their estates are asked tocontact the AAOF before proceeding. Please call (800) 424-2481, extension 246.

Please remember the AAO Foundation in your estate planning.

150 American Journal of Orthodontics and Dentofacial Orthopedics/January 2010

mailto:[email protected]











Founded in 1915 Volume 137 Number 1 January 2010Copyright � 2010 by the American Association of Orthodontists

CONTENTS

COVER

On the cover: The 80th Annual Session of the American Association of Orthodontists took place inour nation’s capital in 1979. Our specialty was then transitioning from wrap-around metal bracesto bonded brackets. Few Annual Session attendees dreamed of owning a personal computer,much less using a computer to view digital imaging exams. When the AAO meets in Washington,D.C. for the 110th Annual Session the orthodontic specialty will have an amazing 30 years ofclinical advancement and worldwide growth behind us. One thing, however, has never changed:Our specialty’s unwavering commitment to providing the best possible treatment and service toour patients. That commitment is reflected in the theme of the 2010 Annual Session: Passion forExcellence. Come prepared to learn from outstanding presenters on clinical and business topics -and from other orthodontic team members from throughout the nation and the world.

EDITORIAL

Your copyright and the SPARC author addendum 1

David L. Turpin, Editor-in-Chief, Seattle, Wash

READERS’ FORUM

Accelerated Osteogenic Orthodontics 2

Neal C. Murphy, Los Angeles, Calif

Authors’ response 2

Seong-Hun Kim and Gerald Nelson, San Francisco, Calif

Appearances count when industry underwrites research 3

Morton I. Katz, Baltimore, Md

Editor’s response 4

David L. Turpin, Editor-in-Chief, Seattle, Wash

CONTINUING EDUCATION

Instructions 5

ONLINE ONLY

Comparison of prospectively and retrospectively selected American Board of Orthodontics cases 6

Blair H. Struble and Greg J. Huang, Bend, Ore, and Seattle, Wash

Editor’s Summary and Q & A 6

The American Journal of Orthodontics and Dentofacial Orthopedics (ISSN 0889-5406) is published monthly by Elsevier Inc., 360 ParkAvenue South, New York, NY 10010-1710. Periodicals postage paid at New York, NYand additional mailing offices. POSTMASTER: Send addresschanges to the American Journal of Orthodontics and Dentofacial Orthopedics, Elsevier Health Sciences Division, Subscription Customer Service,3251 Riverport Lane, Maryland Heights, MO 63043.

2A American Journal of Orthodontics and Dentofacial Orthopedics/January 2010

Class III camouflage treatment: What are the limits? 9

Nikia R. Burns, David R. Musich, Chris Martin, Thomas Razmus, Erdogan Gunel, and Peter Ngan, Pittsburgh,Pa, Schaumburg, Ill, and Morgantown, WV


Active or passive self-ligating brackets? A randomized controlled trial of comparative efficiencyin resolving maxillary anterior crowding in adolescents

12

Nikolaos Pandis, Argy Polychronopoulou, and Theodore Eliades, Corfu, Athens, and Thessaloniki, Greece


Tooth-wear patterns in subjects with Class II Division 1 malocclusion and normal occlusion 14

Guilherme Janson, Paula Vanessa Pedron Oltramari-Navarro, Renata Biella Salles de Oliveira,Camila Leite Quaglio, S�ılvia Helena de Carvalho Sales-Peres, and Bryan Tompson, Bauru, Brazil, andToronto, Ontario, Canada


Accuracy of linear measurements from cone-beam computed tomography-derived surfacemodels of different voxel sizes

16

Janalt Damstra, Zacharias Fourie, James J. R. Huddleston Slater, and Yijin Ren, Groningen, The Netherlands


ORIGINAL ARTICLES

Effectiveness of interceptive orthodontic treatment in reducing malocclusions 18

Gregory J. King and Pongsri Brudvik, Seattle, Wash, and Bergen, Norway

Young patients’ treatment motivation and satisfaction with orthognathic surgery outcomes:The role of ‘‘possible selves’’

26

Elizabeth A. Meade and Marita Rohr Inglehart, Rochester, Minn, and Ann Arbor, Mich

Psychosocial impact of hypodontia in children 35

Emma Laing, Susan J. Cunningham, Steven Jones, David Moles, and Daljit Gill, London, United Kingdom

Association of orthodontic treatment needs and oral health-related quality of life in young adults 42

Ali H. Hassan and Hatem El-Sayed Amin, Jeddah, Saudi Arabia, and Tanta, Egypt

Influence of palatal expanders on oral comfort, speech, and mastication 48

Nanci L. Oliveira De Felippe, Adriana C. Da Silveira, Grace Viana, and Bonnie Smith, Edina, Minn,Austin, Tex, and Chicago, Ill

Relationship between breastfeeding duration and prevalence of posterior crossbitein the deciduous dentition

54

Henri Menezes Kobayashi, Helio Scavone Jr, R�ıvea Ines Ferreira, and Daniela Gamba Garib,S~ao Paulo, Brazil

CONTENTS continued


Effectiveness of the cervical vertebral maturation method to predict postpeak circumpubertalgrowth of craniofacial structures

59

Piotr Fudalej and Anne-Marie Bollen, Warsaw, Poland, and Seattle, Wash

Midpalatal miniscrews for orthodontic anchorage: Factors affecting clinical success 66

Young Ho Kim, Seung-Min Yang, Seonwoo Kim, Joo Yong Lee, Kyu Eok Kim, Anthony A. Gianelly, andSeung-Hyun Kyung, Seoul, Korea, and Boston, Mass

Miniscrew stability evaluated with computerized tomography scanning 73

Jung-Yul Cha, Jae-Kyoung Kil, Tae-Min Yoon, and Chung-Ju Hwang, Seoul Korea

Sequential bone healing of immediately loaded mini-implants: histomorphometric andfluorescence analysis

80

Glaucio Serra, Liliane S. Morais, Carlos Nelson Elias, Marc A. Meyers, Leonardo Andrade, Carlos A. M€uller, andMarcelo M€uller, Rio de Janeiro, Brazil, and San Diego, Calif

Effects of miniscrew orientation on implant stability and resistance to failure 91

Michael B. Pickard, Paul Dechow, P. Emile Rossouw, and Peter H. Buschang, Moscow, Idaho, and Dallas, Tex

Pullout strength of miniscrews placed in anterior mandibles of adult and adolescent dogs:A microcomputed tomographic analysis

100

Zhiqiang Wang, Zhihe Zhao, Jing Xue, Jinlin Song, Feng Deng, and Pu Yang, Jinan, Chengdu, andChongqing, China

REVIEW ARTICLE

Miniscrews in orthodontic treatment: Review and analysis of published clinical trials 108

Adriano G. Crismani, Michael H. Bertl, Ales G. Celar, Hans-Peter Bantleon, and Charles J. Burstone,Vienna, Austria, and Farmington, Conn

CLINICIAN’S CORNER

Clinical observations and success rates of palatal implants 114

Karlien Asscherickx, Bart Vande Vannet, Peter Bottenberg, Heiner Wehrbein, and Mehran Moradi Sabzevar,Jette, Belgium, and Mainz, Germany

CASE REPORTS

Retreatment of a patient with Marfan syndrome and severe root resorption 123

John E. Bilodeau, Springfield, Va

Distalization of the mandibular dentition with mini-implants to correct a Class IIImalocclusion with a midline deviation

135

Kyu-Rhim Chung, Seong-Hun Kim, HyeRan Choo, Yoon-Ah Kook, and Jason B. Cope, Uijongbu andSeoul, Korea, Philadelphia, Pa, and Dallas, Tex

TECHNO BYTES

WiPics: Wireless and beyond 147

Ameet V. Revankar, Narayan H. Gandedkar, and Sanjay V. Ganeshkar, Dharwad, Karnataka, India

CONTENTS continued


DIRECTORY: AAO OFFICERS AND ORGANIZATIONS 150

CONTENTS

Editor’s choice 11A

David L. Turpin, Editor-in-Chief

READER’S SERVICE

Information for authors 19A

Change of address 12A

Information for readers 22A

Classified ad section 23A

CONTENTS continued


CONTENTS

Editor’s choice


Young patients’ treatment motivationand satisfaction with orthognathicsurgery outcomes: The role of ‘‘possibleselves’’

Elizabeth A. Meade and Marita R. Inglehart

Advances in orthognathic surgery over the past 25years, including computer imaging, rigid internal fixa-tion, and shorter hospital stays, have made it a viable op-tion for more patients. With an estimated 1.8 millionpeople in the United States in need of surgery to correctsevere jaw relationships, the success of treatment out-comes is of primary importance. Because many adoles-cents learn of their dentofacial deformity at an early age,there seems to be an increase in younger orthognathicsurgery patients. It has been reported that young patientsare more critical of their current appearance and less sat-isfied after surgery than patients who were older whenthe decision was made. Why is this? To increase knowl-edge in this area, the authors tested these hypotheses: (1)the more energized and enthusiastic the patients arewhen thinking about a future, postoperative possibleself, the more satisfied they will be with the outcomesof their surgery; and (2) the more clearly focused onthe future, postoperative possible self the patients are,the more satisfied they will be with the outcome.

This retrospective study called for surveying 318 pa-tients who had undergone orthognathic surgery betweenJanuary 1996 and December 2005. One hundred fifteenpatients and 117 parents responded to the questionnaire.The patients’ average ages were 16.89 years at the sur-gery and 21.84 years when responding to the investiga-tors. You must read the methodology of this study tofully understand how the survey was designed to mea-sure the various components of preoperative and post-operative possible selves. The findings suggest that themore excited and focused a patient is on the future sur-gery outcome, the more satisfied he or she will be withthe results. It could be helpful for both orthodontists andoral surgeons to ask patients about their motivations and

consider these motivations when finalizing a treatmentplan.

The information from these conversations could bevaluable for orthodontists and oral surgeons whendiscussing treatment options with a surgical candidate.The major limitation of this study was that it was retro-spective; the patients were asked to describe theirfeelings 4.8 years after they had actually undergonethe surgery. These authors suggest that giving your pa-tient vivid images of his or her future esthetic appear-ance or engaging the patient in an active comparisonof facial features before and after the surgery willincrease motivation considerably.

Psychosocial impact of hypodontia inchildren

Emma Laing, Susan J. Cunningham, Steven Jones, David

Moles, and Daljit Gill

Hypodontia is the term used to describe the develop-mental absence of at least 1 deciduous or permanenttooth. It is the most common dental developmentalanomaly and has occurred since at least Paleolithictimes. Although this might not seem unusual to you,how much do we know about the social, behavioral, ed-ucational, medical, and financial implications for an af-fected child and his or her family? These investigatorsquickly realized that there was much to learn. The aimof this study was to evaluate the psychological impactof hypodontia in children compared to those withouthypodontia.

Two groups of participants were recruited: a hypo-dontia group and an orthodontic group not affected byhypodontia. The mean number of missing teeth in thehypodontia group was 4.52. The criteria for inclusionin the study were age between 11 and 16 years andrather severe malocclusion as measured by the indexof orthodontic treatment need. A sample size calcula-tion determined that 61 patients would be required ineach of the 2 groups. The child perceptions question-naire completed by each child consisted of 37 questionsdivided into 4 health domains: oral symptoms, func-tional limitations, emotional well-being, and socialwell-being. In addition, the participants were asked tocomplete 2 visual analog scales related to the

Am J Orthod Dentofacial Orthop 2010;137:11A-12A

0889-5406/$36.00

Copyright � 2010 by the American Association of Orthodontists.

doi:10.1016/j.ajodo.2009.11.006

11A

appearance and function of their teeth. I was particularlyimpressed with this overall research design, the choiceof instruments, and the use of statistical power to deter-mine adequate sample size. What did they learn?

In the sample studied, there was no evidence of dif-ference in the psychosocial status of patients with hypo-dontia and those with other features of malocclusion butno hypodontia. Some statistical evidence suggested thatrelative hypodontia had an impact on the functional abil-ities of the hypodontia patients. This might mean thatretention of deciduous teeth when the permanent teethare missing is beneficial from a functional point of view.

Miniscrews in orthodontic treatment:Review and analysis of publishedclinical trials

Adriano G. Crismani, Michael H. Bertl, Ales G. Celar,

Hans-Peter Bantleon, and Charles J. Burstone

As of September 2007, a Medline search for the terms[screw orthodontic] and [implant orthodontic] returned

734 results. Of those, 14 articles matched the followingcriteria and were considered: studies of miniscrews pub-lished in English or German; human clinical trials; no‘‘case report’’ or ‘‘case series’’; no study with less than30 miniscrews; and additional data on the patients, minis-crews, surgery, or loading-related factors for correlationwith success rate.

When available, data were extracted that correlatedwith the miniscrews’ success rate: patient sex and age,screw length and diameter, method and location ofplacement, and time and amount of loading. The resultsshowed that 1519 miniscrews were used with a successrate of 83% in a total of 452 orthodontic patients. Bycomparison, palatal implants and miniplates reached90% to 95% success. Dental implants have less thana 7% to 9% risk of failure over 10 to 15 years. Minis-crews have the advantage of simple surgical application,and the orthodontist can perform the procedure. Screwswith diameters of 1.2 mm or greater were universallyused, with success rates above 70%.

In addition to this systematic review, this issue of theJournal includes 5 more articles reporting on bothhuman and animal investigations of miniscrew implants.

12A Editor’s choice American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

Official publication of the American Association of Orthodontists,its constituent societies, the American Board of Orthodontics, andthe College of Diplomates of the American Board of Orthodontics

EditorsDavid L. Turpin, DDS, MSD, Editor-in ChiefUniversity of Washington, Department of Orthodontics, D-569, HSC Box 357446, Seattle, WA 98195-7446

Telephone, (206) 221-5413; e-mail, [email protected]

Vincent G. Kokich, DDS, MSD, Case ReportsTacoma, Wash; [email protected]

Gerald Nelson, DDS, Clinician’s CornerOakland, Calif; [email protected]

Michael Rennert, DDS, MSc, Continuing EducationMontreal, Quebec, Canada; [email protected]

Theodore Eliades, DDS, MS, PhD, Dental MaterialsNea Ionia, Greece; [email protected]

Greg Huang, DMD, MSD, MPH, Evidence-based DentistrySeattle, Wash; [email protected]

Leslie A. Will, DMD, MSD, Growth and DevelopmentBoston, Mass; [email protected]

Demetrios J. Halazonetis, DMD, MS, ImagingKifissia, Greece; [email protected]

Laurance Jerrold, DDS, JD, Litigation, Legislation, and EthicsJacksonville, Fla; [email protected]

Alex Jacobson, DMD, MS, MDS, PhD, Reviews and AbstractsBirmingham, Ala

Robert P. Scholz, DDS, Techno BytesDiscovery Bay, Calif; [email protected]

John W. Stockstill, DDS, MS, TMD/FunctionAugusta, Ga; [email protected]

Peter M. Greco, DMD, ABO LiaisonPhiladelphia, Pa

Brent Larson, DDS, MS, EducationMinneapolis, Minn; [email protected]

StaffChris Burke, Managing EditorSeattle, Wash; [email protected]

Carol Haufler, Issue ManagerElsevier Inc. (215) 239-3397; fax, (215) 239-3388; e-mail, [email protected]

American Journal of Orthodontics and Dentofacial Orthopedics/January 2010 15A







http://[email protected]








Editorial Review BoardMaterialsKazuhito Arai, Tokyo, Japan

Michael Bagby, Morgantown, WV

David W. Berzins, Milwaukee, Wis

Samir Bishara, Iowa City, Iowa

Christoph Bourauel, Bonn, Germany

Stanley Braun, Indianapolis, Ind

Samuel Jack Burrow, Chapel Hill, NY

Charles Burstone, Farmington, Conn

Vittorio Cacciafesta, Pavia, Italy

Jie Chen, Indianapolis, Ind

Kwangchul Choy, Seoul, Republic of Korea

Dipak N. Chudasama, Jacksonville, Fla

Manville G. Duncanson, Oklahoma City, Okla

George Eliades, Athens, Greece

Andreas Faltermeier, Regensburg, Germany

Alvaro A. Figueroa, Chicago, Ill

Timothy Foley, London, Ontario, Canada

Maria Enrica Fracasso, Verona, Italy

Franklin Garcia-Godoy, Fort Lauderdale, Fla

Wolfgang Heiser, Innsbruck, Austria

Farzin Heravi, Mashad, Iran

Ross S. Hobson, Newcastle-upon-Tyne, UK

Tony Ireland, Bristol, United Kingdom

Haruo Ishikawa, Chou-Ku, Japan

Afrodite Kakaboura, Goudi, Greece

Thomas R. Katona, Indianapolis, Ind

Arndt Klocke, Hamburg, Germany

Michael Knoesel, Germany

Andrew J. Kuhlberg, Farmington, Conn

Shyh-yuan Lee, Taipei, Taiwan

Abraham B. Lifshitz, Fort Lauderdale, Fla

Eliakim Mizrahi, Essex, United Kingdom

Clarice Nishio, Rio de Janiero, Brazil

Larry Oesterle, Denver, Colo

J. Martin Palomo, Cleveland, Ohio

Nikolaos Pandis, Corfu, Greece

Neil Pender, Liverpool, UK

William Papaioannou, Athens, Greece

Argy Polychronopoulou, Athens, Greece

John M. Powers, Houston, Tex

Donald Raboud, Edmonton, Alberta, Canada

P. E. Rossouw, Dallas, Tex

Chiayi Shen, Gainesville, Fla

Nikolaos Silikas, Manchester, United Kingdom

Louis Taloumis, Colorado Springs, Colo

Lina P. Theodorakopoulou, Athens, Greece

Glenys A. Thorstenson, Chapel Hill, NC

Eser Tufekci, Richmond, Va

Ascension Vicente, Murcia, Spain

John Q. Whitley, Chapel Hill, NC

William A. Wiltshire, Winnipeg, Manitoba, Canada

Spiros Zinelis, Athens, Greece

Digital TechnologyAhmad Abdelkarim, Jacksonville, Fla

Yossi Bar-Zion, Los Angeles, Calif

Sebastian Baumgaertel, Cleveland, Ohio

Joseph M. Caruso, Loma Linda, Calif

Curtis S-K Chen, Seattle, Wash

Will Engilman, Middletown, Ky

William Edward Harrell, Alexander City, Ala

Philipp Gunz, Leipzig, Germany

Yukio Kojima, Nagoya, Japan

Jia-Kuang Liu, Tainan, Taiwan

John B. Ludlow, Chapel Hill, NC

Arthur J. Miller, San Francisco, Calif

Aaron Molen, Auburn, Wash

Wouter Mollemans, Leuven, Belgium

Frits Rangel, Nijmegen, The Netherlands

Meir Redlich, Jerusalem, Israel

William Charles Scarfe, Louisville, Ky

Peter M. Sinclair, Los Angeles, Calif

Richard G. Standerwick, Vancouver, BC

Diagnosis and Treatment PlanningHussam Abdel-Kader, Cairo, Egypt

Ahu Acar, Istanbul, Turkey

Z. Mirzen Arat, Ankara, Turkey

Tiziano Baccetti, Florence, Italy

Sheldon Baumrind, Berkeley, Calif

Robert E. Binder, Newark, NJ

Anna Bocchieri, Tricesimo, Italy

K. Hero Breuning, Tiel, Netherlands

Rodrigo Hermont Cancado, Bauru, SP, Brazil

Lucia H. S. Cevidanes, Chapel Hill, NC

Mohamad Hani Nouri Dalati, London, United Kingdom

Nanci Lara De Filippe, Chicago, Ill

Marinho Del Santo, S~ao Paulo, SP, Brazil

Calogero Dolce, Gainesville, Fla

Francisco Eraso, Indianapolis, Ind

Carla A. Evans, Chicago, Ill

Henry W. Fields, Columbus, Ohio

Allen R. Firestone, Columbus, Ohio

Erhan Gelgor, Kirikkale, Turkey

Mithran S. Goonewardene, Nedlands, Australia

Peter Greco, Philadelphia, Pa

Geoffrey M. Greenlee, Seattle, Wash

James Hartsfield, Lexington, Ky

Chung-Ju Hwang, Seoul, Republic of Korea

Guilherme Janson, S~ao Paulo, Brazil

Marcos Janson, Bauru, SP, Brazil

Seppo Jarvinen, Lahti, Finland

Earl Johnson, San Francisco, Calif

Sanjivan Kandasamy, Perth, Australia

Mustafa Kayalioglu, Adana, Turkey

Heidi Maria Kerosuo, Tromso, Norway

Tae Woo Kim, Seoul, South Korea

Yoon-Ah Kook, Seoul, South Korea

Lorne D. Koroluk, Chapel Hill, NC

Takayuki Kuroda, Yokohama, Japan

Budi Kusnoto, Chicago, Ill

Brent E. Larson, Minneapolis, Minn

C. L. B. Lavelle, Winnipeg, Manitoba, Canada

Shin-Jae Lee, Seoul, South Korea

Eric J.W. Liou, Taipei, Taiwan

Richard Masella, Fort Lauderdale, Fla

Christine Mills, Vancouver, British Columbia, Canada

Ravindra Nanda, Farmington, Conn

Leniana Santos Neves, Bauru, SP, Brazil

Husamettin Oktay, Erzurum, Turkey

Chukwudi Ochi Onyeaso, Ibadan, Nigeria

Ryo Otsuka, Tokyo, Japan

Moschos A. Papadopoulos, Thessaloniki, Greece

Leslie Pitner, Columbia, SC

Reint Albert Reynders, Milan, Italy

Eugene Roberts, Indianapolis, Ind

Margherita Santoro, New York, NY

Jen Soh, Singapore

Robert Staley, Iowa City, Iowa


M. Serdar Toroglu, Balcali, Adana, Turkey

Timothy Turvey, Chapel Hill, NC

Katherine W. L. Vig, Columbus, Ohio

Robert W. F. Vitral, Juiz de Fora, Minas, Gerais, Brazil

Bj€orn Zachrisson, Oslo, Norway

EstheticsMarc Ackerman, Bryn Mawr, Pa

Ezz Dean Azzeh, Boston, Mass

Dario Bertossi, Verona, Italy

S. J. Bowman, Portage, Mich

Luis Carriere, Barcelona, Spain

Jay Decker, Seattle, Wash

Donald Giddon, Wellesley, Mass

Roberto Justus, Mexico City, Mexico

Christopher Maulik, Solana Beach, Calif

Mario Polo, San Juan, Puerto Rico

David M. Sarver, Birmingham, Ala

Shiva Shanker, Columbus, Ohio

Growth and DevelopmentMuge Aksu, Ankara, Turkey

Gurkan Altuna, Toronto, Ontario, Canada

A.E. Athanasiou, Thessaloniki, Greece

Seung-Hak Baek, Seoul, South Korea

Hyoung Baik, Seoul, South Korea

Theodosia Bartzela, Nijmegen, The Netherlands

Adrian Becker, Jerusalem, Israel

Yocheved Ben-Bassat, Jerusalem, Israel

Alan J. Borislow, Philadelphia, Pa

Peter Buschang, Dallas, Tex

Thomas J. Cangialosi, New York, NY

Ismail Ceylan, Erzurum, Turkey

Stella Chaushu, Jerusalem, Israel

Yi-Jane Chen, Taipei, Taiwan

Martyn Cobourne, London, UK

Amy Lynn Counts, Jacksonville, Fla

Ilhan M. Dagsuyu, Erzurum, Turkey

Jose S. Dahan, Brussels, Belgium

John Daskalogiannakis, Toronto, Ontario, Canada

Tarek Hessin El-Bialy, Edmonton, Alberta, Canada

Eric Everett, Chapel Hill, NC

Mona M.S. Fayed, Cairo, Egypt

Virgilio F. Ferrario, Milano, Italy

Leonard S. Fishman, Skaneateles, NY

Carlos Flores-Mir, Edmonton, Alberta, Canada

Sylvia Frazier-Bowers, Chapel Hill, NC

Judah Garfinkle, New York, NY

Daniela G. Garib, Bauru, S~ao Paulo, Brazil

Michael Hairfield, Renton, Wash

Edward Harris, Memphis, Tenn

Bruce Haskell, Louisville, Ky

Serpil Hazar, Izmir, Turkey

Sue Herring, Seattle, Wash

Lysle Johnston, Ann Arbor, Mich

Alf Tor Karlsen, Stjordal, Norway

Christos Katsaros, Nijmegen, The Netherlands

Elias G. Katsavarias, Athens, Greece

Chung How Kau, Houston, Tex

Khaled Khalaf, Newcastle-upon-Tyne, United Kingdom

O. P. Kharbanda, New Delhi, India

Gregory King, Seattle, Wash

Anne Marie Kuijpers-Jagtman, Nijmegen, The Netherlands

Marie Marklund, Umea, Sweden

Steve Marshall, Iowa City, Iowa

Fernando Lima Martinelli, Porto Alegre, RS, Brazil

James A. McNamara, Ann Arbor, Mich

Luciane Macedo Menezes, Porto Alegre, RS, Brazil

Hideo Mitani, Sendai, Miyagi, Japan

Ram Nanda, Oklahoma City, Okla

Peter Ngan, Morgantown, WV

Ib Leth Nielsen, San Francisco, Calif

Sheldon Peck, Newton, Mass

Timo Peltomaki, Zurich, Switzerland

Bakr Rabie, Hong Kong, China

Chiarella Sforza, Milano, Italy

Anwar Ali Shah, Leeds, United Kingdom

Andrew L. Sonis, Boston, Mass

Julie Staggers, Winchester, Va

Tancan Uysal, Kayseri, Turkey

Atalia Wasserstein, Tel Aviv, Israel

A. E. Zaki, Chicago, Ill

BiologyVictor E. Arana-Chavez, S~ao Paulo, SP, Brazil

Efthimia K. Basdra, Athens, Greece

Anne-Marie Bollen, Seattle, Wash

Ilana Brin, Jerusalem, Israel

Tamar Brosh, Tel Aviv, Israel

Paolo Maria Cattaneo, Aarhus, Denmark

Daniele Cardaropoli, Torino, Italy

David A. Covell Jr, Portland, Ore

M. Ali Darendeliler, Sydney, Australia

Ze’ev Davidovitch, Cleveland, Ohio

L. R. Dermaut, Gent, Belgium

Joseph Ghafari, Philadelphia, Pa

Mark Hans, Cleveland, Ohio

Nan Hatch, Ann Arbor, Mich

Stuart D. Josell, Baltimore, MD

Dimitris Kletsas, Athens, Greece

Vinod Krishnan, Trivandrum, Kerala, India

Shingo Kuroda, Okayama, Japan

Kee-Joon Lee, Seoul, Korea

William Liljemark, Plymouth, Minn

Sean Shih-Yao Liu, Indianapolis, Ind

Dawei Liu, Milwaukee, Wisc

Zi-Jun Liu, Seattle, Wash

Kenshi Maki, Kitakyushu, Japan

Jaap C. Maltha, Nijmegen, Netherlands

Birte Melsen, Aarhus, Denmark

Neal C. Murphy, Los Angeles, Calif

Claudia Barbosa Quintella, Bauru, SP, Brazil

Yijin Ren, Groningen, The Netherlands

Edvaldo Antoni Ribeiro Rosa, Curitiba, Brazil

Glenn T. Sameshima, Los Angeles, Calif

Bhavna Shroff, Richmond, Va

Zongyang Sun, Columbus, Ohio

Lokesh Suri, Boston, Mass

Teruko Takano-Yamamoto, Sendai, Japan

Kazunori Yamaguchi, Kitakyushu, Japan

James John Zahrowski, Tustin, Calif

TMD/FunctionJose Luiz Villaca Avoglio, Sao Paulo, Brazil

Jose A. Bosio, Milwaukee, Wis

J.M.H. Dibbets, Marburg, Germany

Diane Johnson, Mooresville, Ind

Richard W. Katzberg, Sacramento, Calif

Stavros Kiliardis, Geneva, Switzerland

Rosalia Leonardi, Catania, Italy

Paul W. Major, Edmonton, Alberta, Canada

Karim. A. Mobarak, Oslo, Norway

Brian Nebbe, Sherwood Park, AB, Canada


Donald J. Rinchuse, Pittsburgh, Pa

Ross Tallents, Rochester, NY

Dayse Urias, Paran�a, Brazil

Treatment and BiomechanicsMarcio Rodrigues Almeida, Sao Paulo, Brazil

Sug-Joon Ahn, Seoul, Republic of Korea

Elie William Amm, Jbeil, Lebanon

Fernanda Angelieri, Bauru, S~ao Paulo, Brazil

Patrick S. Anhoury, Beirut, Lebanon

Jon Artun, Safat, Kuwait

Hasan Babacan, Sivas, Turkey

Faruk Ayhan Basciftci, Konya, Turkey

P. E. Benson, Sheffield, United Kingdom

Lars Bondemark, Malmo, Sweden

Naphtali Brezniak, Tel-Aviv, Israel

Friedrich K. Byloff, Graz, Austria

Sylvain Chamberland, Quebec, Canada

Jason B. Cope, Dallas, Tex

Susan J. Cunningham, London, United Kingdom

G. Frans Currier, Oklahoma City, Okla

Jaime De Jes�us-Vinas, San Juan, Puerto Rico

Toshio Deguchi, Nagano, Japan

Nagwa Helmy El-Mangoury, Cairo, Egypt

Toshiya Endo, Niigata, Japan

Giampietro Farronato, Milano, Italy

Lorenzo Franchi, Florence, Italy

Karina Maria Salvatore de Freitas, Bauru, SP, Brazil

Pawan Gautam, Chicago, Ill

Silvia Geron, Tel Aviv, Israel

Guoqiang Guan, Buffalo, NY

Chester Handelman, Chicago, Ill

Winfried Harzer, Dresden, Germany

John L. Hayes, Williamsport, Pa

Kristin Heimisdottir, Reykjavik, Iceland

Garland Hershey, Chapel Hill, NC

Sarandeep S. Huja, Columbus, Ohio

Hyeon-Shik Hwang, Gwangju, Republic of Korea

Hakan N. Isxcan, Ankara, Turkey

Teitur Jonsson, Reykjavik, Iceland

Onur Kadioglu, Oklahoma City, OK

Toru Kageyama, Shiojiri, Nagano, Japan

Seong-Hun Kim, Uijeongbu-shi, South Korea

Young Ho Kim, Seoul, South Korea

Gero S. M. Kinzinger, Aachen, Germany

Asuman Kiyak, Seattle, Wash

Neal D. Kravitz, Chicago, Ill

Hee-Moon Kyung, Daegu, Korea

Harry Legan, Nashville, Tenn

Liran Levin, Tel Aviv, Israel

Roberto M. A. Lima Filho, S~ao Paulo, Brazil

Steven J. Lindauer, Richmond, Va

Susan P. McGorray, Gainesville, Fla

Shouichi Miyawaki, Nara, Japan

Hong-Beom Moon, Los Angeles, Calif

Robert Moore, Grand Island, Neb

Yehya A. Mostafa, Cairo, Egypt

Hassan Noroozi, Tehran, Iran

Kevin O’Brien, Manchester, United Kingdom

Maria O’Reilly, Marlton, NH

Leandro Berni Osorio, Santa Maria, RS, Brazil

Maja Ovsenik, Llubljana, Slovenia

Hamid Reza Pakshir, Shiraz, Iran

Hans Pancherz, Giessen, Germany

Hyo-Sang Park, Daegu, Republic of Korea

Young Chel Park, Seoul, South Korea

Mohammad Razavi, Cleveland, Ohio

Straty Righellis, Oakland, Calif

Daniel J. Rinchuse, Greensburg, Pa

Michael Riolo, Ypsilanti, Mich

Roy Sabri, Beirut, Lebanon

Vikas Sehgal, Yamunanagar, India

Peter A. Shapiro, Seattle, Wash

Pramrod Sinha, Spokane, Wash

Thomas E. Southard, Iowa City, Iowa

Peter M. Spalding, Lincoln, Neb

Christine Bettina Staudt, Geneva, Switzerland

Arild Stenvik, Oslo, Norway

Junji Sugawara, Sendai, Japan

Sunjay Suri, Toronto, Canada

Tulin Taner, Ankara, Turkey

Kazuo Tanne, Hiroshima, Japan

Madhur Upadhyay, Farmington, Conn

Alexander Vardimon, Tel Aviv, Israel

Juha Eerik Varrela, Turku, Finland

Rodrigo Frizzo Viecilli, Indianapolis, Ind

Hans L. L. Wellens, Sint-Michiels, Brugge, Belgium

Timothy T. Wheeler, Gainesville, Fla

Gary Wolf, Norwalk, Ohio

Sumit Yadav, Indianapolis, Ind

Ibrahim Yavuz, Erzurum, Turkey

Noriaki Yoshida, Nagasaki, Japan

Xiao-guang Zhao, Shanghai

Giliana Zuccati, Firenze, Italy

BiostatisticsEllen A. Begole, Chicago, Ill

Tarisai C. Dandajena, Oklahoma City, Okla

Jorge Faber, Brasılia, Brazil

Jackie Badawi Fayad, Paris, France

Jayne Elizabeth Harrison, Liverpool, Merseyside, United

Kingdom

Conchita Martin, Madrid, Spain

Anestis Mavropoulos, Geneva, Switzerland

Terttu Pietila, Pori, Finland

Charles Brian Preston, Buffalo, NY

Yu-Kang Tu, Leeds, United Kingdom



Information for authorsElectronic manuscript submission and review

The American Journal of Orthodontics and Dento-facial Orthopedics uses the Elsevier Editorial System(EES), an online manuscript submission and reviewsystem.

To submit or review an article, pleasego to the AJO-DOElsevier Editorial System website: ees.elsevier.com/ajodo.

Send other correspondence to:Dr David L. Turpin, DDS, MSD, Editor-in-ChiefAmerican Journal of Orthodontics and DentofacialOrthopedicsUniversity of WashingtonDepartment of Orthodontics, D-569HSC Box 357446Seattle, WA 98195-7446telephone (206)221-5413e-mail, [email protected]

General Information

The American Journal of Orthodontics and Dento-facial Orthopedics publishes original research, reviews,case reports, clinical material, short communications,and other material related to orthodontics and dentofa-cial orthopedics.

Submitted manuscripts must be original, written inEnglish, and not published or under considerationelsewhere. Manuscripts will be reviewed by the editorand consultants and are subject to editorial revision.Authors should follow the guidelines below.

Statements and opinions expressed in the articlesand communications herein are those of the author(s)and not necessarily those of the editor(s) or publisher,and the editor(s) and publisher disclaim any responsi-bility or liability for such material. Neither the editor(s)nor the publisher guarantees, warrants, or endorses anyproduct or service advertised in this publication; neitherdo they guarantee any claim made by the manufacturerof any product or service. Each reader must determinewhether to act on the information in this publication,and neither the Journal nor its sponsoring organizationsshall be liable for any injury due to the publication oferroneous information.

Guidelines for Original Articles

Submit Original Articles via the online ElsevierEditorial System: ees.elsevier.com/ajodo. Organize yoursubmission as follows.

1. Title page. Put all information pertaining to theauthors in a separate document. Include the title ofthe article, full name(s) of the author(s), academic

degrees, and institutional affiliations and positions;identify the corresponding author and include anaddress, telephone and fax numbers, and an e-mailaddress. This information will not be available to thereviewers.

2. Abstract. Structured abstracts of 250 words or lessare preferred. A structured abstract contains thefollowing sections: Introduction, describing theproblem; Methods, describing how the study wasperformed; Results, describing the primary results;and Conclusions, reporting what the authors con-clude from the findings and any clinical implications.

3. Manuscript. The manuscript proper should be orga-nized in the following sections: Introduction andliterature review, Material and Methods, Results,Discussion, Conclusions, References, and figurecaptions. Express measurements in metric unitswhenever practical. Refer to teeth by their full nameor their FDI tooth number. For style questions, referto the AMA Manual of Style, 9th edition. Citereferences selectively, and number them in the ordercited. Make sure that all references have beenmentioned in the text. Follow the format for refer-ences in ‘‘Uniform Requirements for ManuscriptsSubmitted to Biomedical Journals’’ (Ann Intern Med1997;126:36-47); http://www.icmje.org. Include thelist of references with the manuscript proper. Submitfigures and tables separately (see below); do notembed figures in the word processing document.

4. Figures. Digital images should be in TIF or EPSformat, CMYK or grayscale, at least 5 inches wideand at least 300 pixels per inch (118 pixels per cm).Do not embed images in a word processing program.If published, images could be reduced to 1 columnwidth (about 3 inches), so authors should ensure thatfigures will remain legible at that scale. For bestresults, avoid screening, shading, and colored back-grounds; use the simplest patterns available to indi-cate differences in charts. If a figure has beenpreviously published, the legend (included in themanuscript proper) must give full credit to theoriginal source, and written permisson from theoriginal publisher must be included. Be sure youhave mentioned each figure, in order, in the text.

5. Tables. Tables should be self-explanatory andshould supplement, not duplicate, the text. Numberthem with Roman numerals, in the order they arementioned in the text. Provide a brief title for each.If a table has been previously published, include afootnote in the table giving full credit to the originalsource, and include written permission for its usefrom the copyright holder. Submit tables as text-


http://ees.elsevier.com/ajodo


http://ees.elsevier.com/ajodo

http://www.icmje.org

based files (Word or Excel, for example) and not asgraphic elements.

6. Model release and permission forms. Photographs ofidentifiable persons must be accompanied by a releasesigned by the person or both living parents or theguardian of minors. Illustrations or tables that haveappeared in copyrighted material must beaccompanied by written permission for their usefrom the copyright owner and original author, andthe legend must properly credit the source. Permissionalso must be obtained to use modified tables or figures.

7. Copyright release. In accordance with the CopyrightAct of 1976, which became effective February 1,1978, all manuscripts must be accompanied by thefollowing written statement, signed by all authors:‘‘The undersigned author(s) transfers all copyrightownership of the manuscript [insert title of articlehere] to the American Association of Orthodontistsin the event the work is published. The undersignedauthor(s) warrants that the article is original, doesnot infringe upon any copyright or other proprietaryright of any third party, is not under considerationby another journal, has not been previously pub-lished, and includes any product that may derivefrom the published journal, whether print or elec-tronic media. I (we) sign for and accept responsi-bility for releasing this material.’’ Scan the printedcopyright release and submit it via the ElsevierEditorial System, or submit it via fax or mail.

8. Conflict of interest statement. Report any commer-cial association that might pose a conflict of interest,such as ownership, stock holdings, equity interestsand consultant activities, or patent-licensing situa-tions. If the manuscript is accepted, the disclosedinformation will be published with the article. Theusual and customary listing of sources of supportand institutional affiliations on the title page isproper and does not imply a conflict of interest.Guest editorials, Letters, and Review articles may berejected if a conflict of interest exists.

Other Articles

Follow the guidelines above, with the followingexceptions, and submit via Elsevier Editorial System.

Case Reports will be evaluated for completenessand quality of records, quality of treatment, uniquenessof the case, and quality of the manuscript. A high-quality manuscript will include the following sections:introduction; diagnosis; etiology; treatment objectives,alternatives, progress, and results; and discussion. Thesubmitted figures should include extraoral and intraoralphotographs and dental models, panoramic radiographsand tracings from both pretreatment and posttreatment,and progress or retention figures as appropriate.

Short Communications should not exceed 2000words, including the bibliography, and should includea minimal number of figures or tables. Priority will be

given to communications relating to primary researchdata, preferably clinical but also basic. This sectionpermits time-sensitive material to be published within6 months of submission.

Techno Bytes items report on emerging technolog-ical developments and products for use by orthodontists.

Litigation, Legislation, and Ethics items reportlegal and ethical issues of interest to orthodontists.

Miscellaneous Submissions

Letters to the Editor and Ask Us questions andanswers appear in the Readers’ Forum section and areencouraged to stimulate healthy discourse concerningthe profession. Send letters or questions directly to theeditor, via e-mail: [email protected]. Submit a signedcopyright release with the letter, or fax or mail sepa-rately.

Brief, substantiated commentary on subjects ofinterest to the orthodontic profession is occasionallypublished as a Guest Editorial or Special Article.Submit Guest Editorials or Special Articles via theEES website. Include a signed copyright release anda conflict of interest statement.

Books and monographs (domestic and foreign) willbe reviewed, depending on their interest and value tosubscribers. Send books to the Editor of Reviews andAbstracts, Dr Alex Jacobson, University of AlabamaSchool of Dentistry, 1919 7th Ave S, Box 23, Birming-ham, AL 35294. They will not be returned.

Checklist for authors

— Title page, including full name, academicdegrees, and institutional affiliation and positionof each author, and author to whom correspon-dence and reprint requests are to be sent, includ-ing address, business and home phone numbers,fax numbers, and e-mail address

— Abstract— Article proper, including references and figure

legends— Figures, in TIF or EPS format— Tables— Copyright release statement, signed by all au-

thors— Photographic consent statement(s)— Conflict of interest statement— Permissions to reproduce previously published

material



American Journal of Orthodontics and Dentofacial Orthopedics

COPYRIGHT STATEMENT

Must be signed by ALL authors

‘‘The undersigned author(s) transfers all copyright ownership of the manuscript [title of article] to the AmericanAssociation of Orthodontists in the event the work is published. The undersigned author(s) warrants that thearticle is original, does not infringe upon any copyright or other proprietary right of any third party, is not underconsideration by another journal, has not been published previously, and includes any product that may derivefrom the published journal, whether print or electronic media. The author(s) confirm that they have reviewedand approved the final version of the manuscript. I (we) sign for and accept responsibility for releasing thismaterial.

The corresponding author must be named_______________________________________________________(Type name) Signature

Each author’s name must be typed underneath the signature.

_________________________________________________________________________________________

_________________________________________________________________________________________

_________________________________________________________________________________________

_________________________________________________________________________________________

Date:_______________________



Information for readersCORRESPONDENCE

The American Journal of Orthodontics & Dentofacial Orthopedics(ISSN: 0889-5406) is published monthly by Elsevier Inc., 360 ParkAvenue South, New York, NY 10010-1710. Periodicals postage paidat New York, NY and additional mailing offices.

POSTMASTER: Send address changes to American Journal ofOrthodontics & Dentofacial Orthopedics, Elsevier Health SciencesDivision, Subscription Customer Service, 3251 Riverport Lane,Maryland Heights, MO 63043.

Communications regarding original articles and editorial manage-ment should be addressed to David L. Turpin, DDS, MSD, Editor in-Chief, American Journal of Orthodontics and Dentofacial Ortho-pedics, University of Washington, Department of Orthodontics,D-569, HSC Box 357446, Seattle, WA 98195-7446; telephone(206)221-5413; e-mail, [email protected].

Information for authors appears in the January and July issues.Authors should consult these instructions before submitting manu-scripts to the Journal.

Customer Service (orders, claims, online, change of address).Elsevier Health Sciences Division, Subscription Customer Service,3251 Riverport Lane, Maryland Heights, MO 63043. Tel: (800) 654-2452 (U.S. and Canada); (314) 447-8871 (outside U.S. and Canada).Fax: (314) 447-8029. E-mail: [email protected] (for print support); [email protected](for online support).

Address changes must be submitted four weeks in advance.

Yearly Subscription Rates. United States and possessions: Individ-ual $262.00; Institution $520.00; Student/Resident: $139.00. Allother countries (prices include airspeed delivery): Individual$353.00; Institution $618.00; Student/Resident $174.00. Single Is-sues $52.00. To receive student/resident rate, orders must be accom-panied by name of affiliated institution, date of term, and thesignature of program/residency coordinator on institution letterhead.Orders will be billed at the individual rate until proof of status isreceived. Current prices are in effect for back volumes and backissues.

Further information on this journal is available from the Publisher orfrom this journal’s website (http://www.AJODO.org). Information onother Elsevier products is available through Elsevier’s website(http://www.elsevier.com).

Advertising information. Advertising orders and inquiries can be sentto: USA, Canada, and South America, Bruening Nagle Associates, 58Pine St, New Canaan, CT 06840; phone (203) 801-9696; fax (203)801-9618. Classified advertising orders and inquiries can be sent to JimMaloney, Elsevier Inc., 1600 John F. Kennedy Blvd, Suite 1800,Philadelphia, PA 19103; phone (215) 239-3130; fax (215) 239-3734;e-mail [email protected]. Europe and the rest of the world,Julie Toop; phone 144 (0) 1865 843016; fax 144 (0) 1865 843976;e-mail [email protected].

Author inquiries. For inquiries relating to the submission of articles(including electronic submission where available), please visit Elsevier’sAuthor Gateway at http://authors.elsevier.com. The Author Gatewayalso provides the facility to track accepted articles and set up e-mailalerts to inform you of when an article’s status has changed, as well asdetailed artwork guidelines, copyright information, frequently asked

questions, and more. Please see Information for Authors for individualjournal. Contact details for questions arising after acceptance of anarticle, especially those relating to proofs, are provided after registrationof an article for publication.sN The paper used in this publication meets the requirements ofANSI/NISO Z39.48-1992 (Permanence of Paper).

Reprints. For queries about author offprints, e-mail [email protected]. To order 100 or more reprints for educational, commercial,or promotional use, contact the Commercial Reprints Department, ElsevierInc., 360 Park Avenue South, New York, NY 10010-1710. Fax: (212)462-1935; e-mail [email protected]. Reprints of single articles availableonline may be obtained by purchasing Pay-Per-View access for $20 perarticle on the journal website, http://www.AJODO.org. Microform editionavailable from ProQuest Information and Learning, 300 N Zeeb Rd, AnnArbor, MI 48106-1346. � 2010 American Association of Orthodontists. Allrights reserved. This journal and the individual contributions contained in itareprotectedundercopyrightbytheAmericanAssociationofOrthodontists,and the following terms and conditions apply to their use: Photocopying.Single photocopies of single articles may be made for personal use asallowed by national copyright laws. Permission of the Publisher andpayment of a fee is required for all other photocopying, includingmultiple or systematic copying, copying for advertising or promotionalpurposes, resale, and all forms of document delivery. Special rates areavailable for educational institutions that wish to make photocopies fornon-profit educational classroom use. Permissions may be sought di-rectly from Elsevier’s Global Rights Department in Oxford, UK: phone(11) 215 239 3804 or 144 (0)1865 843830, fax 144 (0)1865 853333,e-mail [email protected]. Requests may also be com-pleted online via the Elsevier homepage (http://www.elsevier.com/permissions). In the USA, users may clear permissions and makepayments through the Copyright Clearance Center, Inc., 222 RosewoodDrive, Danvers, MA 01923, USA; phone: (11) (978) 7508400, fax:(11) (978) 7504744, and in the UK through the Copyright LicensingAgency Rapid Clearance Service (CLARCS), 90 Tottenham CourtRoad, London W1P 0LP, UK; phone: (144) 20 7631 5555; fax: (144)20 7631 5500. Other countries may have a local reprographic rightsagency for payments. Derivative Works. Subscribers may reproducetables of contents or prepare lists of articles including abstracts forinternal circulation within their institutions. Permission of the Publisheris required for resale or distribution outside the institution. Permission ofthe Publisher is required for all other derivative works, includingcompilations and translations. Electronic Storage or Usage. Permissionof the Publisher is required to store or use electronically any materialcontained in this journal, including any article or part of an article. Exceptas outlined above, no part of this publication may be reproduced,stored in a retrieval system or transmitted in any form or by any means,electronic, mechanical, photocopying, recording or otherwise, withoutprior written permission of the Publisher. Address permissions requeststo: Elsevier Rights Department, at the fax and e-mail addresses notedabove. Notice. No responsibility is assumed by the Publisher or theAmerican Association of Orthodontists for any injury and/or damage topersons or property as a matter of products liability, negligence orotherwise, or from any use or operation of any methods, products,instructions or ideas contained in the material herein. Because of rapidadvances in the medical sciences, in particular, independent verificationof diagnoses and drug dosages should be made. Although all advertisingmaterial is expected to conform to ethical (medical) standards, inclusionin this publication does not constitute a guarantee or endorsement of thequality or value of such product or of the claims made of it by itsmanufacturer.






http://www.AJODO.org

http://www.elsevier.com



http://authors.elsevier.com




http://www.AJODO.org


http://www.elsevier.com/permissions

http://www.elsevier.com/permissions

EDITORIAL

Your copyright and the SPARC author addendum


Seattle, Wash

What does it mean to hold the copyright toa manuscript or other printed material? Dur-ing a panel discussion at the University of

Washington entitled, ‘‘Publishing your work in the dig-ital age,’’ Clark Shores, Assistant Attorney General ofthe state of Washington, answered questions about au-thors’ rights. According to Shores, ‘‘You own copyrightto anything you write just as soon as it is created. Youdon’t have to do a thing to own this copyright.’’ Regis-tering your copyright gives you added protections if a le-gal dispute arises. But even with that, owninga copyright and enforcing it are 2 different things. AsI learned more about copyrights and authors’ rights, itbecame clear to me that it is a complex issue, especiallyin the digital age.

When authors submit a manuscript to the AmericanJournal of Orthodontics & Dentofacial Orthopedics(AJO-DO), we ask them to transfer the copyright owner-ship to the American Association of Orthodontists. Thisrequest seems to be acceptable to most authors, andsigning the copyright release is routine. The copyrighttransfer gives the AAO the exclusive right to publishthe article in the AJO-DO. Elsevier manages thecopyright process for us and automatically grants au-thors certain rights to their own work, including theright to post it on their own Web sites for exposure tostudents and those who listen to their scientific presen-tations. If you want to reuse a figure that has been pub-lished previously, Elsevier and most other publishersmake it easy with an online permissions process,[email protected].

But what if you want to retain more rights than Elsev-ier allows? For example, what if your funding requires‘‘open source’’ distribution of your article? This meansit must be sent to the National Institutes of Health(NIH) within 12 months after the original publicationdate so that it can be posted on the NIH Web site and bemade freely available to all readers. Doing this could re-quire a change in the original copyright agreement. Is thatpossible? If so, how can you ensure that it will happen?

The Scholarly Publishing and Academic ResourcesCoalition (SPARC) was formed with this purpose inmind. SPARC, an international alliance of over 800 aca-demic and research libraries, works toward a more opensystem of scholarly communication. It believes that, toachieve a more balanced approach to copyright manage-ment, authors should consider the following steps.

1. Read the publication’s copyright agreement withgreat care. It might capture more of your rightsthan are necessary to publish the work. Ensuringthat the agreement is balanced and clearly statesyour rights is up to you.

2. Negotiate for the agreement you want. Publishingagreements are negotiable, and publishers mightrequire only permission to publish an article, notwholesale transfer of copyright.

3. Value the copyright in your intellectual property. Ajournal article is often the culmination of years ofstudy, research, and hard work. The more the articleis read and cited, the greater its value. Before trans-ferring ownership of your intellectual output,understand the consequences and options.

If you plan to submit your article to the NIH open-source Web site after its publication in the AJO-DO orany other journal, consider using the SPARC AuthorAddendum, available at www.arl.org/sparc/author.This is a legal instrument that modifies the copyrightagreement and allows you to retain key rights to your ar-ticles. You can attach it to the standard copyright agree-ment that you include with your submission. The Website is also a good source of information on SPARC’seducation, advocacy, and publisher partner programsin North America, Europe, and Japan.

The issues of access to research findings and therights of authors will continue to be of concern to au-thors and publishers as they seek to find the proper bal-ance. There is every reason to believe that this balancecan be achieved when both sides communicate theirdeepest concerns and listen to one another as changescontinue in the publishing industry.

Note: nothing in this editorial is to be construed aslegal advice. Each situation is different, and you shouldseek legal counsel before acting on any of this informa-tion or these recommendations.

Am J Orthod Dentofacial Orthop 2010;137:1

0889-5406/$36.00


doi:10.1016/j.ajodo.2009.11.007

1


http://www.arl.org/sparc/author

READERS’ FORUM

Letters to the editor*

Accelerated Osteogenic Orthodontics

I was happy to see an article on surgically assisted ortho-dontic treatment in the September issue (Kim SH, Kook YA,Jeong DM, Lee W, Chung KR, Nelson G. Clinical applicationof accelerated osteogenic orthodontics and partially osseointe-grated mini-implants for minor tooth movement. Am J OrthodDentofacial Orthop 2009;136:431-9). Unfortunately, thearticle was a bit off the mark. It slipped through the review-process, as articles do from time to time, because too fewreviewers understand the science or the clinical rationale.

Specifically, these authors were disappointed in the lack-luster results of 7–9 months of treatment time. In my opinion,the authors deviated from a proven protocol and committederrors in their earnest but hapless sojourn into surgical ortho-pedics: (1) they did not need to graft the left molar site becausethere was sufficient bone in situ; (2) the corticotomy cuts (thepreferred term is ‘‘selective alveolar decortication’’ or SAD)were too shallow to elicit the necessary mesenchymal stemcells and too coronal and too timid to induce the degree oftherapeutic ‘‘trauma’’ necessary for intrusion; (3) the authors’comments that heavy forces are necessary conflicts with bothprior literature and their own words later in the article; and (4)their adjustments every 4 weeks were probably not frequentenough to perpetuate a therapeutic osteopenia or the so-calledregional acceleratory phenomenon of Frost and Jee thatderived principally from an engineered surgical trauma.1-3

Continued tensional stress altered in frequency and magnitudevia the roots every 1 to 2 weeks maintains the osteopenic stateand facilitates both accelerated tooth movement and a stablepost-treatment phenotype. Constant force allows bone toadapt to an inactive ‘‘steady state’’ equilibrium. Monthlyadjustments risk recalcification in midtreatment. In our office,we initiate biomechanics 5 minutes after the last suture isplaced and make biomechanical adjustments every 1 to 2weeks. This gives us tooth movement of 1 to 2 mm perweek. Case Western Reserve University and the Universityof Southern California departments of periodontics knowthis and teach the accelerated osteogenic orthodontics(AOO) and periodontally accelerated osteogenic orthodontics(PAOO) protocols in their standard periodontics curriculum(AOO and PAOO and the terms these acronyms representare trademarked by Wilckodontics Inc, Erie, Pa, for the sakeof intellectual integrity and patient protection because theresults are technique sensitive). However, they base their ped-agogy on a thorough review of the literature of this evidence-based technique and we clinicians should do likewise.

I have been very successful with the AOO protocol for 7years. Moreover, the underlying scientific principles uponwhich success is based are helpful in nearly all my patientsto some degree. I am comfortable with AOO and PAOO, but

I cringe at misapprehensions that earnest but inexperiencedneophytes might bring to good bone tissue engineeringscience.4 In my private practice, we have received rave re-views from patients to an almost evangelistic degree. Thishowever has been achieved only by assiduous research andcollegial dialogue with peers. We must be good scholars be-fore we can expect to be successful clinicians in surgical ortho-pedics. In sum, patients will perceive academic gravitas andappreciate progressive clinical innovation. However, that de-rived from being a good librarian before I was a successfulclinician.

Despite these sentiments, I rest assured that in orthodon-tics we still have the opportunity for free speech, meaningfuldialogue, and collaborative progress. For this I am gratefulto my ‘‘NewThink’’ colleagues Professors Lysle E. JohnstonJr (University of Michigan), Nabil F. Bissada (Case WesternReserve University), and Hessam Nowzari (Univeristy ofSouthern California), who contribute to the synthesis in thestyle of the Western dialectic rendering the hybridization ofdiversified thought a better alternative to all its components.

Neal C. MurphyLos Angeles, Calif


0889-5406/$36.00


doi:10.1016/j.ajodo.2009.11.003

REFERENCES

1. Frost HM, Charles C. Bone remodeling and its relation to metabolic

bone disease. Orthopedic Lectures, vol. III. Springfield: Thomas;

1973. p. 81.

2. Frost HM. The regionally acceleratory phenomenon: a review.

Henry Ford Hosp Med 1983;J31:3-9.

3. Jee WSS, Li XJ. Adaption of cancellous bone to overloading in the

adult rat: a single photon absorptiometry and histomorphometry

study. Anat Rec 1990;277:418-26.

4. Murphy NC. In vivo tissue engineering for orthodontists: a modest

first step. In: Davidovitch Z, Mah J, Suthanarak S, editors. Biolog-

ical Mechanisms of Tooth Eruption, Resorption and Movement.

Boston: Harvard Society for the Advancement of Orthodontics.

p. 357-64.

Authors’ response

We thank Dr Murphy for his interest in our recentarticle, ‘‘Clinical application of accelerated osteogenicorthodontics and partially osseointegrated mini-implantsfor minor tooth movement,’’ in the September 2009 issueof the AJO-DO.

Dr Murphy wrote that we commited 4 errors. First, he saidthat we ‘‘did not need to graft the left molar site because therewas sufficient bone in situ.’’ Although sufficient bone mighthave been present in the maxillary second molar region, it

* The viewpoints expressed are solely those of the author(s) and do not reflect

those of the editor(s), publisher(s), or Association.

2

was thin, and we thought it prudent to add bone graft materialto the indentation area.

Second, our corticotomy cuts ‘‘were too shallow to elicitthe necessary mesenchymal stem cells and too coronal andtoo timid to induce the degree of therapeutic ‘trauma’ neces-sary for intrusion.’’ Because the bone was thin between the buc-cal root surface and cortical bone, it was challenging to reachdeeply into medullary bone during decortication. The cortico-tomy was necessarily shallow to avoid damage to the rootsurfaces. Whether regional acceleratory phenomenon (RAP)was or was not induced is speculative, but the tooth movementresponse was a positive indication. It was our intent to inducerapid tooth movement without damaging the teeth. Two nota-ble options for individual molar intrusion were considered:intrusion by compression osteogenesis, in which corticalbone at the root apex is sufficiently removed, and the bone seg-ment and teeth are together intruded; and cortical activation,which depends on RAP.1-3 We chose the second, placed a buc-cal C-tube plate with a flap operation, and performed additionaldecortication to induce the RAP phenomenon.

Third, Dr Murphy said that our ‘‘comments that heavyforces are necessary conflicts with both prior literature’’ andour own words later in the article. Previous studies that RAPis also activated by normal orthodontic force are wellknown.4-6 However, studies are lacking regarding RAP activa-tion by intrusion or uprighting forces. The intrusion methodsused in several case reports on decortications typically appliedthe same amount of force on both buccal and palatal sides. Theintrusion force was 150 to 250 g in those cases. Mostafa et al7

applied a 400-g force for displacement of a premolar in dogexperiments. Whether the forces applied in our patients wouldalso be considered ‘‘heavy’’ is perhaps debatable. We wouldlike to see further studies assessing the optimal force to useafter decortication.

Finally, Dr Murphy thought that our adjustments every4 weeks were probably not frequent enough to perpetuatea therapeutic osteopenia or the so-called regional acceleratoryphenomenon of Frost and Jee.

Continued tensional stress altered in frequency andmagnitude via the roots every 1 to 2 weeks maintainsthe osteopenic state and facilitates both acceleratedtooth movement and a stable posttreatment pheno-type. So monthly adjustments risk recalcification inmidtreatment. In our office, we initiate biomechanics5 minutes after the last suture is placed and makebiomechanical adjustments every 1 to 2 weeks. Thisgives us tooth movement of 1 to 2 mm per week.

As shown by Iino et al6 and Wilcko et al,5 in space closureor when tooth uprighting is needed in a mesiodistal orientationin the long axis of the alveolus, bone reduction is accomplishedwith an ostectomy through the entire thickness of the alveolus toinclude the labial and lingual cortical plates and interspersedmedullary bone. The concept of corticotomy-facilitatedorthodontic treatment does not also define intrabony toothmovement. Our studies might help us to know whether decorti-cations alone have similar effects as osteoectomy.

We do appreciate these thoughtful questions by Dr Mur-phy. We look forward to not only future animal research andcase reports, but also meta-analyses with several relative ran-domized clinical trials for corticotomy-facilitated orthodontictreatment to be approved as an evidence-based technique.

Seong-Hun KimGerald Nelson

San Francisco, CalifAm J Orthod Dentofacial Orthop 2010;137:2-3

0889-5406/$36.00


doi:10.1016/j.ajodo.2009.11.005

REFERENCES

1. Kim SH, Lee KB, Chung KR, Nelson G, Kim TW. Severe bimax-

illary protrusion with adult periodontitis treated by corticotomy and

compression osteogenesis. Korean J Orthod 2009;39:54-64.

2. Kanno T, Mitsugi M, Furuki Y, Kozato S, Ayasaka N, Mori H. Cor-

ticotomy and compression osteogenesis in the posterior maxilla for

treating severe anterior open bite. Int J Oral Maxillofac Surg

2007;36:354-7.

3. Chung KR, Mitsugi M, Lee BS, Kanno T, Lee W, Kim SH. Speedy

surgical orthodontic treatment with skeletal anchorage in adults—

sagittal correction and open bite correction. J Oral Maxillofac Surg

2009;67:2130-48.

4. Park WK, Kim SS, Park SB, Son WS, Kim YD, Jun ES, et al. The

effect of cortical punching on the expression of OPG, RANK, and

RANKL in the periodontal tissue during tooth movement in rats.

Korean J Orthod 2008;38:159-74.

5. Wilcko MT, Wilcko WM, Pulver JJ, Bissada NF, Bouquot JE. Ac-

celerated osteogenic orthodontics technique: a 1-stage surgically

facilitated rapid orthodontic technique with alveolar augmentation.

J Oral Maxillofac Surg 2009;67:2149-59.

6. Iino S, Sakoda S, Ito G, Nishimori T, Ikeda T, Miyawaki S.

Acceleration of orthodontic tooth movement by alveolar cortico-

tomy in the dog. Am J Orthod Dentofacial Orthop 2007;131:

448.e1-448.e8.

7. Mostafa YA, Fayed MMS, Mehanni S, ElBokle NN, Heider AM.

Comparison of corticotomy-facilitated vs standard tooth-move-

ment techniques in dogs with miniscrews as anchor units. Am

J Orthod Dentofacial Orthop 2009;136:570-7.

Appearances count when industryunderwrites research

Leafing through the October 2009 issue of the AJO-DO, Irecognized the apparatus and illustrations in the article,‘‘Three-dimensional orthodontic force measurements’’(Badawi HM, Toogood RW, Carey JPR, Heo G, Major PW.Am J Orthod Dentofacial Orthop 2009;136:518-28). Theimages had been presented months ago to my orthodonticresidents by the representative of a popular supplier of ortho-dontic materials, with the representative’s claim that ‘‘Now wehave scientific proof that our bracket is superior.’’

Although I am not an engineer, the apparatus seems to beelegant and innovative in design and might provide much use-ful information of benefit to our specialty in the future. I ampositive that the equipment was quite costly, and I understand

American Journal of Orthodontics and Dentofacial Orthopedics Readers’ forum 3Volume 137, Number 1

was thin, and we thought it prudent to add bone graft materialto the indentation area.

Second, our corticotomy cuts ‘‘were too shallow to elicitthe necessary mesenchymal stem cells and too coronal andtoo timid to induce the degree of therapeutic ‘trauma’ neces-sary for intrusion.’’ Because the bone was thin between the buc-cal root surface and cortical bone, it was challenging to reachdeeply into medullary bone during decortication. The cortico-tomy was necessarily shallow to avoid damage to the rootsurfaces. Whether regional acceleratory phenomenon (RAP)was or was not induced is speculative, but the tooth movementresponse was a positive indication. It was our intent to inducerapid tooth movement without damaging the teeth. Two nota-ble options for individual molar intrusion were considered:intrusion by compression osteogenesis, in which corticalbone at the root apex is sufficiently removed, and the bone seg-ment and teeth are together intruded; and cortical activation,which depends on RAP.1-3 We chose the second, placed a buc-cal C-tube plate with a flap operation, and performed additionaldecortication to induce the RAP phenomenon.

Third, Dr Murphy said that our ‘‘comments that heavyforces are necessary conflicts with both prior literature’’ andour own words later in the article. Previous studies that RAPis also activated by normal orthodontic force are wellknown.4-6 However, studies are lacking regarding RAP activa-tion by intrusion or uprighting forces. The intrusion methodsused in several case reports on decortications typically appliedthe same amount of force on both buccal and palatal sides. Theintrusion force was 150 to 250 g in those cases. Mostafa et al7

applied a 400-g force for displacement of a premolar in dogexperiments. Whether the forces applied in our patients wouldalso be considered ‘‘heavy’’ is perhaps debatable. We wouldlike to see further studies assessing the optimal force to useafter decortication.

Finally, Dr Murphy thought that our adjustments every4 weeks were probably not frequent enough to perpetuatea therapeutic osteopenia or the so-called regional acceleratoryphenomenon of Frost and Jee.

Continued tensional stress altered in frequency andmagnitude via the roots every 1 to 2 weeks maintainsthe osteopenic state and facilitates both acceleratedtooth movement and a stable posttreatment pheno-type. So monthly adjustments risk recalcification inmidtreatment. In our office, we initiate biomechanics5 minutes after the last suture is placed and makebiomechanical adjustments every 1 to 2 weeks. Thisgives us tooth movement of 1 to 2 mm per week.

As shown by Iino et al6 and Wilcko et al,5 in space closureor when tooth uprighting is needed in a mesiodistal orientationin the long axis of the alveolus, bone reduction is accomplishedwith an ostectomy through the entire thickness of the alveolus toinclude the labial and lingual cortical plates and interspersedmedullary bone. The concept of corticotomy-facilitatedorthodontic treatment does not also define intrabony toothmovement. Our studies might help us to know whether decorti-cations alone have similar effects as osteoectomy.

We do appreciate these thoughtful questions by Dr Mur-phy. We look forward to not only future animal research andcase reports, but also meta-analyses with several relative ran-domized clinical trials for corticotomy-facilitated orthodontictreatment to be approved as an evidence-based technique.

Seong-Hun KimGerald Nelson

San Francisco, CalifAm J Orthod Dentofacial Orthop 2010;137:2-3

0889-5406/$36.00


doi:10.1016/j.ajodo.2009.12.006

REFERENCES

1. Kim SH, Lee KB, Chung KR, Nelson G, Kim TW. Severe bimax-

illary protrusion with adult periodontitis treated by corticotomy and

compression osteogenesis. Korean J Orthod 2009;39:54-64.

2. Kanno T, Mitsugi M, Furuki Y, Kozato S, Ayasaka N, Mori H. Cor-

ticotomy and compression osteogenesis in the posterior maxilla for

treating severe anterior open bite. Int J Oral Maxillofac Surg

2007;36:354-7.

3. Chung KR, Mitsugi M, Lee BS, Kanno T, Lee W, Kim SH. Speedy

surgical orthodontic treatment with skeletal anchorage in adults—

sagittal correction and open bite correction. J Oral Maxillofac Surg

2009;67:2130-48.

4. Park WK, Kim SS, Park SB, Son WS, Kim YD, Jun ES, et al. The

effect of cortical punching on the expression of OPG, RANK, and

RANKL in the periodontal tissue during tooth movement in rats.

Korean J Orthod 2008;38:159-74.

5. Wilcko MT, Wilcko WM, Pulver JJ, Bissada NF, Bouquot JE. Ac-

celerated osteogenic orthodontics technique: a 1-stage surgically

facilitated rapid orthodontic technique with alveolar augmentation.


6. Iino S, Sakoda S, Ito G, Nishimori T, Ikeda T, Miyawaki S.

Acceleration of orthodontic tooth movement by alveolar cortico-

tomy in the dog. Am J Orthod Dentofacial Orthop 2007;131:

448.e1-448.e8.

7. Mostafa YA, Fayed MMS, Mehanni S, ElBokle NN, Heider AM.

Comparison of corticotomy-facilitated vs standard tooth-move-

ment techniques in dogs with miniscrews as anchor units. Am


Appearances count when industryunderwrites research

Leafing through the October 2009 issue of the AJO-DO, Irecognized the apparatus and illustrations in the article,‘‘Three-dimensional orthodontic force measurements’’(Badawi HM, Toogood RW, Carey JPR, Heo G, Major PW.Am J Orthod Dentofacial Orthop 2009;136:518-28). Theimages had been presented months ago to my orthodonticresidents by the representative of a popular supplier of ortho-dontic materials, with the representative’s claim that ‘‘Now wehave scientific proof that our bracket is superior.’’

Although I am not an engineer, the apparatus seems to beelegant and innovative in design and might provide much use-ful information of benefit to our specialty in the future. I ampositive that the equipment was quite costly, and I understand

American Journal of Orthodontics and Dentofacial Orthopedics Readers’ forum 3Volume 137, Number 1

that financial support is always needed for research. What con-cerns me is the linkage between a supplier’s financial supportand the results, which just so happen to show that the bracketsystem—heavily marketed by that company—are ‘‘proven’’ tobe superior. The lead author is described as a consultant to thatvery company, and the orthodontic department that supportedthe project is noted at the end of the article as the recipient ofcontributions from the company for student scholarships forfuture research.

I am not impugning the professional competence of theorthodontists and engineers who participated in this projector the department’s motives. In an ideal world, moneywould play no part in experimental design or data gather-ing, but call me cynical—money does talk in the realworld. Readers of the article could justifiably wonderwhether financial support would be forthcoming from thecompany if the results had shown its product to be inferior.Unlike the physical sciences, where experiments are re-peated to verify claims (remember cold fusion), our exper-iments cannot be repeated because either the equipment isclosely held or the human or animal sample can never beexactly duplicated. In the biologic sciences, there isa greater need to work at arms length from industrybecause of the repeatability problem and to eliminate anyappearance of commercial influence.

Orthodontists are not alone in expressing concern for biasin research that is too tightly linked to business. Check out theOctober 23, 2009, New York Times article, ‘‘Research uproar ata cancer clinic’’ by Duff Wilson. Also, read the November 4,2009, New York Times article, ‘‘Health bills aim a light on doc-tors’ conflicts’’ by Natasha Singer. Or search out the numerousarticles about Medtronic and its ‘‘consultant’’ who submittedand had published a bogus article favorable to one ofMedtronic’s products in the medical literature.

Once refereed journals publish research, it becomes gos-pel. We must be ever alert that our precious specialty literaturenot be infused with advertising disguised as scientific inquiry.Claims of superiority for bracket and wire systems need to bedialed down, because there are many variables involved intooth movement.

Morton I. KatzBaltimore, Md

Am J Orthod Dentofacial Orthop 2010;137:3-4

0889-5406/$36.00


doi:10.1016/j.ajodo.2009.11.004

Editor’s response

I agree with Dr Katz that the sponsors of original researchmust be acknowledged with the dissemination of results. Atthe same time, we appreciate that commercial entities providebadly needed funding in the effort to answer today’s scientificquestions. These relationships can easily lead to bias in the re-porting of results, or at a minimum, the perception of bias.That is why journals like the AJO-DO are increasing their ef-forts to disclose them. As Dr Katz noted, we pointed out theauthors’ relationships with Ormco in two places in the article.

Regarding the conclusions reported in the article, it is allthe more important to verify in-vitro findings with the publica-tion of in-vivo studies. That is a challenge I welcome as editorof this journal.

David L. Turpin, Editor-in-ChiefSeattle, Wash


0889-5406/$36.00


doi:10.1016/j.ajodo.2009.12.006

Editors of the International Journal of Orthodontia (1915-1918),International Journal of Orthodontia & Oral Surgery (1919-1921),International Journal of Orthodontia, Oral Surgery and Radiography (1922-1932),International Journal of Orthodontia and Dentistry of Children (1933-1935),International Journal of Orthodontics and Oral Surgery (1936-1937),American Journal of Orthodontics and Oral Surgery (1938-1947),American Journal of Orthodontics (1948-1986), andAmerican Journal of Orthodontics and Dentofacial Orthopedics (1986-present)

1915 to 1932 Martin Dewey1931 to 1968 H. C. Pollock

1968 to 1978 B. F. Dewel

1978 to 1985 Wayne G. Watson

1985 to 2000 Thomas M. Graber2000 to present David L. Turpin

4 Readers’ forum American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

AAO Continuing Education

Earn 3 hours of CE credit CE Editor: Dr. Michael Rennert

Instructions: To submit your answers to this test and earn 3 hours ofCE credit, log on to www.AAOmembers.org. Select the link for theStore, where you can purchase the examination for $20 (a 20% dis-count off the regular $25 fee). You will receive e-mails confirmingyour order and instructing you how to access the test. Follow theinstructions to complete the examination (you will be directed to

a web page). Results are tabulated immediately. To earn 3 hours ofCE credit, you must answer 75% of the questions correctly. If youdo not receive a passing score the first time, you can take the test again,free of charge, until you pass. Upon successful completion of the test,your CE credits will be added to the AAO’s online CE Credit Manager,and you can use the CE Credit Manager to print a certificate.

Learning objectives

After completing this course, the participant will be able to:1. Discuss the effectiveness of interceptive orthodontic treatment in

reducing malocclusions.2. Discuss the influence of palatal expanders on oral comfort, speech,

and mastication.3. Understand the relationship between breastfeeding duration and

prevalence of posterior crossbite in the deciduous dentition.4. Discuss factors affecting the clinical success of midpalatal minis-

crews for orthodontic anchorage.5. Discuss the effects of miniscrew orientation on implant stability and

resistance to failure.

Article 1. Effectiveness of interceptive orthodontic treatmentin reducing malocclusions, by Gregory J. King and PongsriBrudvik1. The objective of this retrospective study was to study the effective-

ness of interceptive orthodontic treatment by comparing 1 groupreceiving interceptive orthodontic treatment with an observationgroup.a. This statement is true.b. This statement is false.

2. The results of this study showed that interceptive orthodontic treat-ment initially improves malocclusions with reductions in complex-ity and need compared with doing nothing.a. This statement is true.b. This statement is false.

3. The results also showed that interceptive orthodontic treatmentseldom requires follow-up treatment in the permanent dentition.a. This statement is true.b. This statement is false.

Article 2. Influence of palatal expanders on oral comfort, speech,and mastication, by Nanci L. Oliveira De Felippe et al4. The overall objective of this study was to compare and contrast the

hyrax expander with the Haas expander.a. This statement is true.b. This statement is false.

5. The results showed that the negative impacts of placing a palatalexpander were resolved in the first week of appliance wear.a. This statement is true.b. This statement is false.

Article 3. Relationship between breastfeeding duration and preva-lence of posterior crossbite in the deciduous dentition, by HenriMenezes Kobayashi et al6. One objective of this study was to analyze the relationship between

exclusive breastfeeding duration and the prevalence of posteriorcrossbite in the deciduous dentition.a. This statement is true.b. This statement is false.

7. Another objective was to compare the results with those from a sim-ilar group of children with a history of nonnutritive sucking habits.a. This statement is true.b. This statement is false.

8. The results showed no relationship between exclusive breastfeedingduration and the prevalence of posterior crossbite in the deciduousdentition.a. This statement is true.b. This statement is false.

Article 4. Midpalatal miniscrews for orthodontic anchorage:Factors affecting clinical success, by Young Ho Kim et al9. One objective of this study was to investigate the success rate of

midpalatal miniscrews used for orthodontic anchorage.a. This statement is true.b. This statement is false.

10. Another objective was to investigate the factors affecting clinicalsuccess.a. This statement is true.b. This statement is false.

11. The final objective was to determine the maximum load that can beimparted before failure of the miniscrew.a. This statement is true.b. This statement is false.

12. The results showed that splinting 2 miniscrews influenced theirclinical success.a. This statement is true.b. This statement is false.

13. The results also showed that the clinical success rate was higher inmales than in females.a. This statement is true.b. This statement is false.

5.e1

http://www.AAOmembers.org

Article 5. Effects of miniscrew orientation on implant stability andresistance to failure, by Michael B. Pickard et al14. The purpose of this study was to evaluate the effects of orthodontic

miniscrew implant orientation on stability and resistance to failureat the bone-implant interface.a. This statement is true.b. This statement is false.

15. Sixteen miniscrew implants were placed in the mandibles of 4male beagles.a. This statement is true.b. This statement is false.

16. The results showed that the more closely the long axis of the im-plant approximates the line of applied force, the lower the stabilityof the implants.a. This statement is true.b. This statement is false.

Answersa b c d a b c d

1 9

2 10

3 11

4 12

5 13

6 14

7 15

8 16

Program EvaluationAgree Neutral Disagree

1. The content was appropriate and timely.

2. The objectives were clearly stated and met.

3. The articles were well written.

4. Statistics were clearly explained and relevant.

5. Content is applicable to my daily practice.

6. I will apply most of the information

in my daily activities.

7. This method is valuable in helping me

complete my continuing education

requirements.

5.e2

ONLINE ONLY

Comparison of prospectively and retrospectivelyselected American Board of Orthodontics cases

Blair H. Strublea and Greg J. Huangb

Bend, Ore, and Seattle, Wash

Introduction: In this study, we compared the pretreatment conditions, treatment characteristics, and orthodon-tic outcomes of 3 groups of subjects selected for the American Board of Orthodontics (ABO) phase III clinicalexamination. One group was selected retrospectively by graduating residents just before their graduation.The 2 prospective groups were treated at separate institutions. The students at 1 institution were not awarethat these patients would be potential ABO cases (prospective, blinded), but the students at the second institu-tion were aware that these subjects would serve as their pool of potential patients for the ABO examination (pro-spective, unblinded). In addition to comparing the 3 groups, all cases were categorized as passing or failingbased on their total objective grading system (ABO-OGS) score to assess the ABO-OGS criteria that were themost challenging to meet. Methods: Chart histories and orthodontic dental casts (pretreatment and posttreat-ment) were collected for 133 subjects. Information regarding demographics, initial malocclusion type, treatmentmodality, treatment duration, appointment frequency, and missed appointments were collected from chart his-tories. Pretreatment dental casts were evaluated by using the discrepancy index; the index of complexity, out-come, and need; and the peer assessment rating. Posttreatment dental casts were evaluated with the peerassessment rating and the ABO-OGS. Results: The only significant pretreatment characteristic with predictivepower for favorable orthodontic outcome was Angle Class I (3.1 odds ratio for passing the ABO-OGS) comparedwith the Class II subjects. The prospective unblinded group received more extraction and headgear therapy thandid the other groups. The retrospective group had significantly lower total ABO-OGS posttreatment scores anda higher passing rate compared with the prospective groups. Conclusions: Angle Class I malocclusions appearto have some advantage for achieving passing ABO-OGS scores, as does the retrospective selection of cases.Successful board certification appears difficult to accomplish based on a prospective model for orthodonticgraduate residents. New graduate candidates might be at a disadvantage compared with traditional candidatesbecause they often cannot take advantage of the posttreatment settling phase. Alignment, marginal ridges, andocclusal contacts appear to be where most points are deducted in the evaluation of ABO-OGS certificationcases. (Am J Orthod Dentofacial Orthop 2010;137:6.e1-6.e8)

Since its inception in 1929, the American Board ofOrthodontics (ABO) has striven to certify asmany practicing orthodontists as possible and

elevate the standards of the practice of orthodontics.1,2

When the percentages of board-certified orthodontistswere 13% to 17% in the late 1970s, the board beganefforts to increase the numbers of board-certified ortho-dontists.1 In 2001, the ABO began actively pursuing theidea and the feasibility of certifying graduating ortho-dontic residents in the resident clinical outcomes study,or the pilot study (PS). This led to a 4-year collaborativeproject between the ABO and 16 American orthodontic

graduate programs accredited by the Commission onDental Accreditation.3,4 This project investigatedwhether orthodontic residents could provide start-to-finish treatment for 6 patients with ABO-quality results.

At its inception, this PS was designed so that caseswould be prospectively identified at the time of patientassignment, and residents, faculty, and patients at the par-ticipating programs would be aware of this PS designa-tion. Orthodontic program directors at the participatinginstitutions were asked to prospectively designate 12 pa-tients for each incoming 2002 resident, 6 of whom wouldbe presented to the board after treatment. Participatingorthodontic residents would treat the patients, from band-ing to debanding, and be eligible to present the cases tothe ABO to earn a 10-year time-limited certificate. Six-teen graduate orthodontic programs in the United Statesagreed to participate in the PS. Only 1 orthodontic pro-gram agreed to participate under the requirement thatall persons involved in the treatment would be blindedto the designation of the prospectively selected PSpatients. This method was chosen to prevent any

a Private practice, Bend, Ore.b Chair, Department of Orthodontics, University of Washington, Seattle.

The authors report no commercial, proprietary, or financial interest in the

products or companies described in this article.

Reprint requests to: Greg J. Huang, 1959 NE Pacific Street, D-569, Health

Sciences Building, Box 357446, Seattle, WA 98195-7446; e-mail, ghuang@u.

washington.edu.

Submitted, December 2008; revised and accepted, May 2009.

0889-5406/$36.00


doi:10.1016/j.ajodo.2009.05.016

6.e1



differential treatment of the PS subjects.The subjects atthe other 15 programs were treated in an unblinded fash-ion, so that residents, faculty, and patients knew theywere participating in the PS.

During the PS, the study protocol was altered. Theparticipant was allowed to present 6 cases for certifica-tion that included only 1 from the previously selected 12that were prospectively designated. An incentive ofa 15-year time-limited certificate was offered to resi-dents presenting all prospectively selected PS cases,and 12% of PS examinees successfully earned this15-year certificate.5

The PS concluded in February 2006, when 50participating orthodontic residents attended the ABOClinical Examination. Forty-five candidates successfullyobtained ABO certification, for a pass rate of 90%. Thiscompared with 33 traditional candidates passing theexamination at a rate of 85%. Therewas a mean differenceof 2.38 ABO-OGS points for passing cases between theresident and the traditional examinees in the PS. Theboard concluded that the cases presented had sufficientcomplexity, with an average discrepancy index (DI) scoreof 16.96 for the student cases, compared with an averageDI score of 21.84 for the regular examinees. The PS par-ticipants presented 422 cases, of which 58% were fromthe original prospectively selected PS group.5

The board believed that this result positivelyaffirmed that residents could treat to ABO standards dur-ing their orthodontic graduate programs. As a result, theboard has now instituted new certification guidelines forrecent graduates.4 The new certification process has pro-vided the impetus for graduate programs to self-evaluatetheir patient populations and the quality of orthodontictreatment provided in each residency program.

In this study, we aimed to determine, with the aid ofthe DI; the peer assessment rating (PAR); the index oftreatment complexity, outcome, and need (ICON); andthe ABO objective grading system (ABO-OGS),whether there were significant differences in pretreat-ment conditions, treatment characteristics, and ortho-dontic treatment outcomes between ABO casesselected by using the 3 methods. We determined whetherany pretreatment characteristics had predictive value indetermining the orthodontic treatment quality outcome.Additionally, we categorically compared the ABOpoints deducted for failing vs passing cases, to determinewhich intraoral locations were the most difficult for theentire study sample.

MATERIAL AND METHODS

All study procedures were approved by the institu-tional review board at the University of Washington.

The sample consisted of complete chart historiesand dental casts (pretreatment and posttreatment) ofall subjects. The sample comprised 3 groups. Groups1 (retrospectively selected) and 2 (prospectiveblinded) were collected from the retention archivesof a graduate orthodontic program participating inthe PS. Group 3 (prospective unblinded) was collectedfrom another graduate orthodontic program participat-ing in the PS.

No exclusion criteria were defined to preventpatients from participating in the study, as was thecase with the initial PS guidelines. Patients wereincluded irrespective of age, sex, race, or orthodonticproblem if they were nonsyndromic and comprehensiveorthodontic treatment was planned. The ABO stipulatedthat ‘‘cases should be representative of a cross section ofclinical problems and of adequate difficulty to representthe resident’s ability to diagnose and treat orthodonticpatients’’ in the original PS guidelines.5 When recordswere collected for this study, subjects were excludedif they were still in active orthodontic treatment, hadincomplete records, had transferred treatment outsidethe assigned graduate clinic, or had never startedtreatment after the prospective PS designation.

According to these criteria, 49 records were initiallycollected for group 1, which included subjects who wereretrospectively selected by the graduating residents(classes of 2003-2006) to be used in a simulated ABOexamination. Two subjects were excluded from thisgroup because of incomplete records, leaving 47subjects in group 1. Group 2, the prospective blindedgroup, was prospectively selected to be part of the PS,and their study participation was concealed from all per-sons involved in the orthodontic treatment. Group 2 ini-tially contained 57 subjects, but 16 were excluded basedon the exclusion criteria, leaving 41 subjects. Group 3records were gathered from another institution thattreated patients in a prospective unblinded fashion. Allpersons involved (faculty, residents, patients) wereaware of the PS designation. Group 3 started with 50subject records, but 5 were excluded, leaving 45prospective unblinded subjects. For the reasons outlinedabove, 23 subjects were excluded from the entire sub-ject sample. All subject materials were deidentifiedand labeled with an identification number to facilitateinvestigator blinding.

Subject records consisted of chart histories, andpretreatment and posttreatment dental casts. Chart his-tories were reviewed to gather information about demo-graphics, initial malocclusion, treatment type, treatmentduration, frequency of appointments, and missedappointments during active orthodontic treatment. Ina few cases, phase 1 treatment had previously been

6.e2 Struble and Huang American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

provided before the PS. In these circumstances, onlyinformation from the comprehensive phase of treatmentwas collected for these patients. This protocol is consis-tent with the ABO’s evaluation of 2-phase patients intraditional board examinations.

Pretreatment dental casts were scored by using theDI, PAR, and ICON by 2 calibrated, independent exam-iners. Posttreatment dental casts were scored by 2 exam-iners using the PAR, ICON, and ABO-OGS.6-10 Theradiographic component of the ABO-OGS index wasexcluded, because many patients had no posttreatmentpanoramic radiographs. Additionally, numerous studieshave questioned the usefulness of panoramic radiogra-phy to assess root parallelism because of inherent imagedistortion, especially in premolar extraction sites.11,12

The ABO-OGS scores were adjusted based on averagePS radiographic deductions to account for this exclu-sion. The pretreatment and posttreatment casts fromeach site were combined, deidentified, assigned identi-fication numbers, and measured in random order. Twoinvestigators measured all dental casts independently,and the mean score was used unless significant differ-ences were noted in the scores (weighted PAR, 5 points;weighted ICON, 9 points; ABO-OGS, 4 points). Whendifferences were greater than these values, the dentalcasts were rescored by consensus, and the consensusscore was used. Twenty-one (15.8%) cases had to berescored by consensus.

To determine intraexaminer error, 10 casts wererescored later by each examiner. Intraexaminer errorwas evaluated by using the intraclass correlation coeffi-cient for all examiners involved in the study (Table I).

Pretreatment conditions and treatment characteris-tics were assessed and compared both qualitativelyand quantitatively. These scores were compared withposttreatment conditions as assessed by the PAR,ICON, and ABO-OGS to determine treatment changesand the quality of orthodontic treatment between thegroups.

Descriptive statistics (means, standard deviations,and ranges) were calculated for pretreatment DI andICON, pretreatment and posttreatment PAR, and post-treatment ABO-OGS scores. Descriptive statisticswere also performed for patient demographics, initialmalocclusion (type and severity), treatment type, treat-ment duration, and number of orthodontic appoint-ments. Analysis of variance (ANOVA) was used totest for differences in continuous variables betweenthe groups. Pairwise between-group comparisons werecarried out when ANOVA indicated differencesbetween the groups. The Bonferroni adjustment to thesignificance level was used to correct for multiplecomparisons in post-hoc analyses. This correction was

applied to prevent inflation of the type 1 error ratecaused by multiple comparisons.13 When the samplewas divided into passing and failing cases, the averagescores were compared with t tests.

Logistic regression was used to determine whetherany pretreatment variables could be used as reliablepredictors of successful board-quality treatment. Astepwise model-building algorithm was used to identifya subset of available covariates that was highly predic-tive of successful board-quality treatment.

The statistician was blinded to treatment group iden-tification until the analyses were completed. For allanalyses, the levels of significance were set at P \0.05and P \0.017 when the Bonferroni adjustment wasperformed.

RESULTS

The 3 treatment groups were similar with respectto demographics and pretreatment characteristics(Table II). All groups had similar sex ratios, withmore females than males. There were similar percent-ages of white patients in the groups. Group 3 had noAsian or Hispanic subjects and more black subjects(26.7%) compared with groups 1 and 2. There weremore Class I subjects (53.2%) and fewer Class IIsubjects (38.3%) in group 1 (retrospectively selectedgroup) compared with the other groups.

There were no significant differences between thegroups regarding subject age at initial records, start oftreatment, or end of treatment (Table III). The group3 subjects had a younger average pretreatment age,but this was most likely because several older adultswere included in groups 1 and 2. When age mediansand ranges were examined, all 3 groups were similarat initial records. When average length of treatmentwas assessed, there was a significant difference betweenthe groups, P 5 0.004. With the Bonferroni adjustmentfor multiple comparisons, group 2 had a statisticallysignificant increase in average length of treatment(31.3 months) compared with both group 1 (25.0months, P 5 0.005) and group 3 (25.1 months,P 5 0.005). Likewise, there was a significant difference

Table I. Inraobserver error

Intraclass correlation coefficient

Examiner 1 (BHS) Examiner 2 (CJ) Examiner 3 (SH)

ICON 0.996 0.975

PAR 0.997 0.992

ABO-OGS 0.99 0.979

BHS, Blair H. Struble; CJ, Cameron Jolley; SH, Sara Haley.

American Journal of Orthodontics and Dentofacial Orthopedics Struble and Huang 6.e3Volume 137, Number 1

in the number of appointments, P 5 0.04. However,when adjustments were made for multiple comparisons,only group 2 (28.6 appointments) reached statisticalsignificance when compared with group 3 (24.4appointments, P 5 0.017). There was no significant dif-ference in the numbers of missed appointments betweenthe 3 groups. Group 3 had more subjects receivingextraction (77.8%) and headgear (40%) therapy duringorthodontic treatment than those in groups 1 and 2(Table II). Although it was difficult to quantify, it wasknown that some attending faculty in group 3 used treat-ment mechanics that included second-order tip-backbends. Attending faculty in groups 1 and 2 did not usethis type of treatment.

There was no significant difference between the3 groups for any pretreatment cast analyses (DI,ICON, or PAR, Table IV). There was, however, a statis-tically significant difference between the groups for theposttreatment analyses (PAR and ABO-OGS). Group 1(retrospective group) had statistically lower posttreat-ment PAR scores than did both prospectively selectedgroups (group 2, P 5 0.001; group 3, P 5 0.001).Likewise, for ABO-OGS scores, only the retrospectivegroup 1 (16.2) was significantly different comparedwith groups 2 (23.1, P 5 0.001) and 3 (28.4, P 5 0.001).

Because the radiographic root angulation compo-nent of the ABO-OGS was not scored in this study,the ABO’s passing score of 26 was reduced to 23 basedon the mean root-angulation point deduction of 2.6 fromthe PS rounded to the nearest whole number. Based onthis ABO-OGS pass-or-fail cutoff point, there were 4(8.5%) failures in group 1 (retrospective), 19 (46.3%)in group 2 (prospective blinded), and 28 (62.2%) ingroup 3 (prospective unblinded). There were 2 signifi-cant outliers: a subject in group 2 was debonded pre-emptively because of significant decay and restorativeneeds, and a subject in group 3 was debonded witha less than ideal result because of concerns about peri-odontal bone loss in the mandibular anterior region.These subjects were included in the analyses to followthe intent-to-treat principle.

When the entire sample was divided into subjectspassing (#23 points) vs those failing (.23 points) theABO-OGS, passing cases had significantly lower scoresfor all ABO-OGS categories except interproximalcontacts (Table V). Failing cases had the greatest aver-age point deductions in alignment (8.3), marginal ridges(5.4), and occlusal contacts (7.6). These areas added upto an average deduction of 21.3 points for failing cases.If these major point deductions had been avoided, the

Table II. Demographics, pretreatment conditions, and treatment

Group

Retrospective Prospective blinded Prospective unblinded1 2 3 Total

Ethnicity

White 35 (74.5%) 33 (80.5%) 32 (71.1%) 100 (75.2%)

Asian 8 (17.0%) 2 (4.9%) 0.0 10 (7.5%)

Hispanic 1 (2.1%) 4 (9.8%) 0.0 5 (3.8%)

Black 1 (2.1%) 1 (2.4%) 12 (26.7%) 14 (10.5%)

Other 2 (4.3%) 1 (2.4%) 1 (2.2%) 4 (3.0%)

Sex

Male 21 (44.7%) 17 (41.5%) 18 (40.0%) 56 (42.1%)

Female 26 (55.3%) 24 (58.5%) 27 (60.0%) 77 (57.9%)

Angle classification

Class I 25 (53.2%) 17 (41.5%) 16 (35.6%) 58 (43.6%)

Class II 18 (38.3%) 23 (56.1%) 21 (46.7%) 62 (46.6%)

Class III 4 (8.5%) 1 (2.4%) 8 (17.8%) 13 (9.8%)

Malocclusion type

Anterior crossbite 15 (32.0%) 5 (12.2%) 10 (22.2%) 30 (22.6%)

Posterior crossbite 6 (12.8%) 9 (22.0%) 4 (8.9%) 19 (14.3%)

Deepbite 6 (12.8%) 7 (17.1%) 8 (17.8%) 21 (15.8%)

Missing teeth 2 (4.3%) 5 (12.2%) 5 (11.1%) 12 (9.0%)

Impactions 2 (4.3%) 2 (4.9%) 4 (8.9%) 8 (6.0%)

Treatment modality

Extractions 20 (42.6%) 18 (43.9%) 35 (77.8%) 73 (54.9%)

Headgear 13 (27.7%) 12 (29.3%) 18 (40.0%) 43 (32.3%)

Orthognathic surgery 4 (8.5%) 4 (9.8%) 0.0 8 (6.0%)


January 2010

average failing total ABO-OGS score of 32.3 pointswould have been reduced to a passing score of 11.0points. These categories were explored in greater detailto determine where these points were lost.

A significant percentage of alignment deductionsoccurred for second molar-first molar, first molar-second premolar, and canine-lateral incisor contacts(67.1%). The most problematic areas for marginalridges were first molar-second premolar contact in themaxilla and second molar-first molar in the mandible.Overall, these areas accounted for 91.6% of thepoints lost in this category. Occlusal contact deductionsoccurred most commonly for second molar contacts andaccounted for 55.2% of the points lost (Table VI).

Marginal ridges and occlusal contacts are 2 areasthat have been shown to significantly improve duringa posttreatment settling period.14 When the ABOweighting formula developed by Nett and Huang14

was applied to the mean score for failing ABO casesin this study, the average failing score of 32.4 (excludingroot angulation) was reduced by more than 11 points to21.0 points. Most dental casts evaluated in this studywere taken at debanding. It is likely that, if a settlingperiod had been allowed, many failing scores wouldhave improved to passing ABO-OGS scores.

When the entire sample was examined by Angleclassification, there was nearly equal distribution ineach class for passing and failing cases, excep for ClassI subjects. The percentage of passing subjects in theAngle Class I category was much larger than for theClass II and Class III groups (Table VII).

Based on these findings, a stepwise model-buildingprocedure was used to construct a logistic regressionmodel for the probability of passing (adjustedABO-OGS #23) for the subjects in groups 2 and 3.Group 1 (retrospective) was not included in the model

Table III. Treatment timing

ANOVA

Group

1 2 3

Retrospective Prospective blinded Prospective unblinded

Mean Median Range SD n Mean Median Range SD n Mean Median Range SD n P

Patient ages (y)

Age at initial records 18.2 13.8 8.7-66.5 12.7 47 19.0 13.8 10.9-54.7 10.6 41 16.0 13.3 9.2-43.8 7.9 45 0.386

Age at start of treatment 19.2 14.5 10.1-67.0 12.4 47 19.7 14.5 11.1-55.5 10.9 41 16.2 13.5 10.4-43.9 7.9 45 0.256

Age at end of treatment 21.3 16.4 13.2-68.6 12.3 47 22.3 17.1 13.7-57.0 10.7 41 18.3 15.7 12.0-44.7 8.1 45 0.175

Treatment timing

Treatment length

(months)

25.0 23.0 13.0-51.0 9.1 47 31.3 28.0 15.0-64.0 11.2 41 25.1 25.0 8.0-58.0 8.9 45 0.004*

Number of

appointments

25.2 23.0 14.0-48.0 8.0 47 28.6 27.5 16.0-59.0 8.1 41 24.4 24.0 8.0-55.0 7.9 45 0.040*

Number of missed

appointments

1.4 1.0 0-7.0 2.0 47 2.9 1.0 0-15.0 4.4 41 2.6 1.0 0-10.0 3.2 45 0.061

*Denotes statistical significance.

Table IV. Dental cast analysis

ANOVA

1 2 3

Retrospective Prospective blinded Prospective unblinded

Mean Range SD n Mean Range SD n Mean Range SD n P

Pretreatment analyses

DI 16.1 3.0-64.0 10.6 47 17.6 5.0-36.0 8.9 41 18.2 4.0-44.0 9.3 45 0.564

ICON 59.6 18.0-100.0 19.3 47 66.3 32.0-108.0 19.7 41 64.7 21.0-107.0 18.5 45 0.227

Pretreatment PAR 27.9 7.5-52.0 12.1 47 29.5 8.5-70.0 14.2 41 27.7 9.5-47.0 9.5 45 0.765

Posttreatment analyses

Posttreatment PAR 2.1 0.0-8.0 1.9 47 5.1 1.0-23.0 4.2 41 4.8 1.0-19.5 4.0 45 0.001*

ABO-OGS 16.2 7.5-31.0 5.0 47 23.1 3.5-69.0 11.6 41 28.4 11.0-54.0 8.6 45 0.001*



building because inclusion in this group was sucha strong predictor for success that it tended to over-shadow any other possible predictors. Variables consid-ered in this procedure included group, ethnicity, sex, ageat start of treatment, length of treatment, number ofappointments, number of missed appointments, classifi-cation, extraction therapy, use of headgear, posteriorcrossbite, deep overbite, missing teeth, and impactedteeth. The only variable that showed a statisticallysignificant predictive power was a pretreatment ClassI malocclusion as compared with Class II. A Class Isubject had a 3.1 odds ratio for obtaining a passingABO-OGS score than a Class II subject (P 5 0.023;95% CI, 1.2-8.3). Pretreatment Class III malocclusionwas also identified by the model-building procedureas highly predictive of obtaining a passing score. How-ever, this association failed to achieve statistical signif-icance because of the small number of subjects in theClass III malocclusion group.

DISCUSSION

In comparing the 3 groups in this study, the subjectswere generally equal in sex, age, malocclusion type, andmalocclusion severity (measured by the DI, ICON, andPAR) before treatment. However, there were more ClassI patients in the retrospectively selected sample. Thismight indicate that, although there might be other mal-occlusion problems that contributed to higher severityscores, a proper anteroposterior relationship beforetreatment can be a positive predictor of a subject’s like-lihood of posttreatment orthodontic success as mea-sured by the ABO-OGS. The predictive model derivedfrom this study supported this idea in showing that ClassI subjects had an odd ratio of 3.1 for passing theABO-OGS compared with Class II subjects.

Because the subject groups were similar beforetreatment, perhaps the differences during orthodontic

treatment might help to explain the discrepancy inposttreatment results between them. The prospectiveblinded group had a significantly longer treatmenttime (31.3 months) than the retrospective (25.0 months)and prospective unblinded (25.1 months) groups,P 5 0.004. This is not surprising, since only theprospective blinded group providers were unaware ofthe potential ABO-OGS evaluation of their patients,and assigned students were under no pressure to finishtreatment. Most prospective blinded subjects weretreated by more than 1 resident. Also, several studentscould have treated the retrospectively selected patients.On the other hand, the students treating the prospectiveunblinded group knew that they must complete the PScases to present them to the ABO and were given a cer-tain date by which appliances must be removed. Thiswas done to allow settling before the student graduated,so that an additional set of posttreatment dental castscould be obtained for the ABO examination.

If treatment times between the retrospective (group 1)and prospective unblinded (group 3) groups were almostidentical, how do we explain the significant discrepancyin posttreatment ABO-OGS pass rates between thesegroups? The retrospective group had more Class 1 sub-jects, whereas the prospective unblinded group receivedmore headgear and extraction therapy. In addition to thesetreatment techniques, some prospective unblinded treat-ment providers incorporated second-order tip-back bendsfor their patients. This orthodontic technique resulted inmany marginal-ridge and occlusal-contact ABO-OGSpoint deductions if the cases were evaluated at debanding.

Table V. Pass vs fail ABO-OGS comparison

Pass Fail

Mean SD n Mean SD n P

ABO-OGS categories

Alignment 4.6 2.1 82 8.3 3.1 51 0.003*

Marginal ridges 3.3 1.7 82 5.4 2.5 51 0.008*

Buccolingual inclination 1.7 1.5 82 3.2 2.1 51 0.011*

Overjet 2.3 1.8 82 5.3 3.4 51 0.005*

Occlusal contacts 4.1 2.5 82 7.6 3.4 51 0.002*

Occlusal relationship 0.4 1.0 82 2.5 3.9 51 0.000*

Interproximal contacts 0.0 0.1 82 0.0 0.0 51 0.113

Total 16.3 4.7 82 32.3 8.2 51 0.008*


Table VI. Point deduction percentages by intraorallocation for alignment, marginal ridges, and occlusalcontacts

Alignment Marginal ridges

(% of totalpoints)

(% of totalpoints)

Second molar-first molar 25.0 45.1

First molar-second premolar 21.1 46.5

Second premolar-first premolar 13.4 8.4

First premolar-canine 11.1 0.0

Canine-lateral incisor 21.0 0.0

Lateral incisor-central incisor 7.6 0.0

Midline 1.7 0.0

Occlusal contacts

Buccal Lingual

Distal second molar 31.2 8.8

Mesial second molar 16.0 25.5

Distal first molar 15.4 25.5

Mesial first molar 19.0 14.2

Second premolar 12.0 26.0

First premolar 6.4 0.0


January 2010

Many group 3 subjects were debanded several months be-fore graduation to allow settling before impressions weretaken for posttreatment dental casts. This difference in or-thodontic treatment technique and philosophy might helpto explain the differences in the ABO-OGS mean scores,since most dental casts for the prospective unblindedgroup were evaluated at debanding rather than after post-treatment settling.14,15

The application of the ABO weighting formula toour samples showed that ABO-OGS scores tended toimprove after active orthodontic treatment.14 Most den-tal casts evaluated in this study were taken at debanding.It is likely that, if a settling period had been allowed forthese cases, many failing scores would have improvedto passing ABO-OGS scores. When this formula wasapplied to the average ABO-OGS scores for group 3(with the lowest ABO passing rate), there was an 8.6-point reduction in average ABO-OGS score from 28.4to 19.8 before the root-angulation category wasincluded. This represented a significant improvementin scoring and reduced an average failing score forthis group to an average passing score.

The topic of whether to retrospectively or prospec-tively select cases for ABO certification was alteredduring the development and evolution of the current cer-tification system.5 It seems likely that patients treatedprospectively by unblinded practitioners would receiveadditional attention compared with an average orthodon-tic patient. On the other hand, retrospectively selectedpatients might include better treatment cooperators andthose with more favorable growth and possibly less com-plex malocclusions. A retrospective system of selectionfor board certification treats the entire orthodontic patientpopulation as potential cases for board certification andtends to equalize the standard of care to all patients,compared with a prospectively unblinded system.

The ABO’s decision to alter the PS protocol froma prospective designation to a largely retrospectiveselection model seems justifiable based on our results.Collectively, about 50% of the subjects in the 2 prospec-tive groups of this study did not have passing ABO-OGSscores. Group 2 had a much longer average treatmenttime, indicating that these patients often had several stu-dents during their treatment. This would disqualify themas potential ABO patients because the board requires all

cases to be treated entirely by the ABO candidate. Ourresults might indicate that successful board certificationbased on a prospective model for orthodontic graduateresidents is difficult to accomplish.

Even though the providers were aware of thesubject’s PS designation in the prospective unblindedgroup, this group had a higher ABO-OGS failure ratethan did the prospective blinded group. The differencein failure rates between the 2 prospective groups couldbe explained in several ways. First, the blinded grouphad a significantly longer mean treatment time andmore appointments compared with the unblinded group.This period might have been important to accomplishfinal detailing of the occlusion. The prospectiveunblinded group had more extraction (77.8%) andheadgear (40%) therapy. They also might have hadsecond-order tip-back bends that often benefit froma period of settling after debanding for occlusal-contactand marginal-ridge problems to resolve. Most posttreat-ment dental casts for the entire sample were taken atdebanding, and this could have been a significant disad-vantage, especially to the prospective unblinded group.

Finally, it is interesting to see which categoriesappeared to present the greatest challenge to treatmentproviders in this study. Alignment, marginal ridges,and occlusal contacts were the 3 categories with thegreatest point deductions in all 3 groups. These mistakesoccurred in categories previously reported by the ABOto be among those with the most problems in prelimi-nary field tests. Second molars appeared to be a consis-tent problem in this sample, even though the medianages at start and end of treatment were approximately14 and 16 years, respectively. Even at this age, whensecond molars should be fully erupted, they still presenta significant challenge when evaluated with the ABO-OGS. Marginal ridges and occlusal contacts are bothsusceptible to improvement during the posttreatmentsettling phase.15 This settling advantage was not alwayspossible for resident examinees, who must completecertifcation cases in the time constraints of residencyprograms. Orthodontic residency programs interestedin providing the best possible opportunities for theirgraduates to accomplish orthodontic certificationshould pay special attention to these controllable factorsand ensure that patients are of an appropriate age witha fully erupted dentition.

This study defined 3 groups. Group 1, theretrospectively selected group, indicated that we treatsome patients to high standards, and we are good atidentifying excellent results after treatment. Group 2(prospective, blinded) represents a standard sample ofuniversity teaching cases during a specified time period,and, when there is no rush to complete treatment, it

Table VII. Pass vs. fail breakdown by Angle’s classifica-tion

Passing cases Failing cases

Class I 42 (72.4%) 16 (27.6%)

Class II 33 (53.2%) 29 (46.8%)

Class III 7 (53.8%) 6 (46.2%)


appears that about 50% of the cases could meet ABOstandards. Group 3 (prospective, unblinded) indicatedthat prospective identification of 12 ABO subjectsmight be an insufficient number to meet the board’sprior prospective pathway for new graduates, since thepassing rate in this group was less than 50%.

The major strengths of this study included blindedevaluation of a relatively large sample, several studysites, and patients who were treated during the sameperiod of time. Additionally, the groups were relativelysimilar with respect to pretreatment conditions. Nearlyall posttreatment dental casts were evaluated at de-banding. Because graduating students might need toobtain final records before they leave their institutions,dental casts from this time accurately reflect what thestudents can present to the board. Therefore, this couldbe considered a strength of this study. On the otherhand, posttreatment settling is likely to improve sev-eral ABO-OGS parameters, and the dental casts thatwere assessed in this study might underestimate theABO passing rate if settling had been allowed tooccur.15

There were some differences in ethnicity, Angleclassifications, and treatment modalities, and all ofthese might be limitations to this study. It was notpossible to assess growth and compliance from therecords we obtained, and these unknown factors couldhave influenced the outcomes. Finally, many factorscontribute to determining the outcome of a treatedpatient, and, although the predictive model used inthis study provides some insight into those factors,larger samples are needed to adequately address them.

CONCLUSIONS

1. Class I subjects seem to have a distinct advantage(odds ratio, 3.1) over Class II subjects for achievingpassing ABO-OGS scores.

2. It appears challenging to accomplish successfulboard certification with a prospective model for or-thodontic graduate residents based on these results.

3. Candidates should pay special attention to thealignment, marginal-ridge, and occlusal-contactcategories, especially regarding second molars,because these were the most problematic areas forall groups in this study.

4. New graduate examinees might have a disadvantagecompared with traditional examinees, since they

often do not have time to benefit from posttreatmentsettling.

REFERENCES

1. Cangialosi TJ, Riolo ML, Owens SE Jr, Dykhouse VJ,

Moffitt AH, Grubb JE, et al. The ABO’s 75th anniversary:

a retrospective glance at progress in the last quarter century. Am


2. Cangialosi TJ, Riolo ML, Owens SE Jr, Dykhouse VJ,

Moffitt AH, Grubb JE, et al. The American Board of Orthodontics

and specialty certification: the first 50 years. Am J Orthod

Dentofacial Orthop 2004;126:3-6.

3. Riolo ML, Owens SE Jr, Dykhouse VJ, Moffitt AH, Grubb JE,

Greco PM, et al. ABO resident clinical outcomes study: case

complexity as measured by the discrepancy index. Am J Orthod


4. Dykhouse VJ, Moffitt AH, Grubb JE, Greco PM, English JD,

Briss BS, et al. ABO initial certification examination: official

announcement of criteria. Am J Orthod Dentofacial Orthop

2006;130:662-5.

5. Dykhouse VJ, Moffitt AH, Grubb JE, Greco PM, English JD,

Briss BS, et al. A report of the ABO resident clinical outcome

study (the pilot study). Am J Orthod Dentofacial Orthop 2006;

130:656-61.

6. DeGuzman L, Bahiraei D, Vig KW, Vig PS, Weyant RJ,

O’Brien K. The validation of the peer assessment rating index

for malocclusion severity and treatment difficulty. Am J Orthod


7. Richmond S, Shaw WC, O’Brien KD, Buchanan IB, Jones R,

Stephens CD, et al. The development of the PAR index (peer

assessment rating): reliability and validity. Eur J Orthod 1992;

14:125-39.

8. Daniels C, Richmond S. The development of the index of

complexity and need (ICON). J Orthod 2000;27:149-62.

9. Savastano NJ, Firestone AR Jr, Beck FM, Vig KW. Validation of

the complexity and treatment outcome components of the index of

complexity, outcome and need (ICON). Am J Orthod Dentofacial

Orthop 2003;124:244-7.

10. Casko JS, Vaden JL, Kokich VG, Damone J, James RD,

Cangialosi TJ, et al. Objective grading system for dental

casts and panoramic radiographs, American Board of

Orthodontics. Am J Orthod Dentofacial Orthop 1998;114:589-99.

11. Mckee IW, Glover KE, Williamson PC, Lam EW, Heo G,

Major PW. The effect of vertical and horizontal head positioning

in panoramic radiography on mesiodistal tooth angulations. Angle

Orthod 2001;71:442-51.

12. Garcia-Figueroa MA, Raboud DW, Lam EW, Heo G, Major PW.

Effect of buccolingual root angulation on the mesiodistal angula-

tion shown on panoramic radiographs. Am J Orthod Dentofacial

Orthop 2008;134:93-9.

13. Glantz SA. Primer of biostatistics. New York: McGraw Hill; 1987.

14. Nett BC, Huang GJ. Long-term posttreatment changes measured

by the American Board of Orthodontics objective grading system.

Am J Orthod Dentofacial Orthop 2005;127:444-50.

15. Razdolsky Y, Sadowsky C, BeGole EA. Occlusal contacts follow-

ing orthodontic treatment: a follow-up study. Angle Orthod 1989;

59:181-5.


January 2010

ONLINE ONLY

Comparison of prospectively and retrospectivelyselected American Board of Orthodontics cases

Blair H. Struble and Greg J. Huang

Bend, Ore, and Seattle, Wash

Introduction: In this study, we compared the pre-treatment conditions, treatment characteristics, and or-thodontic outcomes of 3 groups of subjects selectedfor the American Board of Orthodontics (ABO) phaseIII clinical examination. One group was selected retro-spectively by graduating residents just before their grad-uation. The 2 prospective groups were treated atseparate institutions. The students at 1 institution werenot aware that these patients would be potential ABOcases (prospective, blinded), but the students at the sec-ond institution were aware that these subjects wouldserve as their pool of potential patients for the ABO ex-amination (prospective, unblinded). In addition to com-paring the 3 groups, all cases were categorized aspassing or failing based on their total objective gradingsystem (ABO-OGS) score to assess the ABO-OGScriteria that were the most challenging to meet.

Methods: Chart histories and orthodontic dentalcasts (pretreatment and posttreatment) were collectedfor 133 subjects. Information regarding demographics,initial malocclusion type, treatment modality, treatmentduration, appointment frequency, and missed appoint-ments were collected from chart histories. Pretreatmentdental casts were evaluated by using the discrepancy in-dex; the index of complexity, outcome, and need; andthe peer assessment rating. Posttreatment dental castswere evaluated with the peer assessment rating andthe ABO-OGS.

Results: The only significant pretreatment charac-teristic with predictive power for favorable orthodonticoutcome was Angle Class I (3.1 odds ratio for passingthe ABO-OGS) compared with the Class II subjects.The prospective unblinded group received more extrac-tion and headgear therapy than did the other groups. Theretrospective group had significantly lower total ABO-OGS posttreatment scores and a higher passing ratecompared with the prospective groups.

Conclusions: Angle Class I malocclusions appear tohave some advantage for achieving passing ABO-OGSscores, as does the retrospective selection of cases. Suc-

cessful board certification appears difficult to accomplishbased on a prospective model for orthodontic graduateresidents. New graduate candidates might be at a disad-vantage compared with traditional candidates becausethey often cannot take advantage of the posttreatmentsettling phase. Alignment, marginal ridges, and occlusalcontacts appear to be where most points are deducted inthe evaluation of ABO-OGS certification cases.

Read the full text online at: www.ajodo.org,pages 6.e1-6.e8

EDITOR’S SUMMARY

Since the inception of the American Board of Ortho-dontics (ABO) in 1929, its directors and many others inleadership positions of the American Association of Or-thodontists have striven to certify as many practicing or-thodontists as possible. The primary reason for thisemphasis has always been the belief that patient careimproves as more members of a specialty become boardcertified. The chief objective of the ABO is the pursuitof excellence in orthodontics.

Despite this, throughout the 1980s and 1990s, thepercentages of board-certified orthodontists hovered be-tween 20% and 25%. Something else had to be done toachieve certification of most orthodontists. In 2001, theABO considered the feasibility of certifying orthodonticresidents shortly after graduation and designed the res-ident clinical outcomes study, or the pilot study. This re-search project investigated whether residents couldprovide start-to-finish treatment for 6 patients withABO-quality results.

Researchers at the University of Washington wantedto determine whether there are significant differences inpretreatment conditions, treatment characteristics, andtreatment outcomes among ABO cases depending onhow the cases were selected for review. They used 3methods to amass samples. Group 1 was retrospectivelyselected from the retention archives of a graduate ortho-dontic program participating in the pilot study. Group 2was drawn from the same source but collected in a pro-spective, blinded manner. Group 3 was prospective butunblinded and was collected from another graduate or-thodontic program that participated in the pilot study.


0889-5406/$36.00


doi:10.1016/j.ajodo.2009.08.001

6

http://www.ajodo.org

For a detailed description of the methodology used tocompare these 3 groups, I encourage you to read the en-tire online article.

From my point of view, the results are also worth re-viewing. A sample retrospectively selected on the basisof the outcome from a large number of cases is bound tobe more successfully treated than a small prospectivelyselected sample chosen because the subjects might besuccessful in the end. It is also interesting to note whichcategories appeared to present the greatest challenges tothe treatment providers. Alignment, marginal ridges,and occlusal contacts were the 3 categories previouslyreported by the ABO to present the most problems inearlier field tests. The authors stated that the ABO’s de-cision to alter the pilot study protocol from a prospectivedesign to a largely retrospective selection model seemsjustifiable based on their results.

Another aim of the study was to determine whetherany pretreatment characteristics had a predictive valuein determining the quality of the outcome. As the au-thors point out, many factors contribute to treatmentoutcome, and it might not be possible to examine allfactors with the relatively small, highly selected, anddiversely treated sample from this study.

Take-home notes:

� Class I subjects seemed to have a distinct advantageover Class II subjects for achieving passing ABO-OGS scores.

� It appears challenging to accomplish successfulboard certification based on a prospective model fororthodontic graduate residents based on these results.

� Candidates should pay special attention to the align-ment, marginal-ridge, and occlusal-contact cate-gories, especially regarding the second molars.

� New graduate examinees might have a disadvantagecompared with traditional examinees because theyoften do not have time to benefit from posttreatmentsettling.

Table V. Pass vs fail ABO-OGS comparison

Pass Fail

PMean SD n Mean SD n

ABO-OGS categories

Alignment 4.6 2.1 82 8.3 3.1 51 0.003*

Marginal ridges 3.3 1.7 82 5.4 2.5 51 0.008*

Buccolingual inclination 1.7 1.5 82 3.2 2.1 51 0.011*

Overjet 2.3 1.8 82 5.3 3.4 51 0.005*

Occlusal contancts 4.1 2.5 82 7.6 3.4 51 0.002*

Occlusal relationship 0.4 1.0 82 2.5 3.9 51 0.000*

Interproximal contacts 0.0 0.1 82 0.0 0.0 51 0.113*

Total 16.3 4.7 82 32.3 8.2 51 0.008*


Table VI. Point deduction percentages by intraoral loca-tion for alignment, marginal ridges, and occlusalcontacts

Alignment(% of total

points)

Marginal ridges(% of total

points)

Second molar-first molar 25.0 45.1

First molar-second premolar 21.1 46.5

Second premolar-first premolar 13.4 8.4

First premolar-canine 11.1 0.0

Canine-lateral incisor 21.0 0.0

Lateral incisor-central incisor 7.6 0.0

Midline 1.7 0.0

Occlusal contacts

Buccal Lingual

Distal second molar 31.2 8.8

Mesial second molar 16.0 25.5

Distal first molar 15.4 25.5

Mesial first molar 19.0 14.2

Second premolar 12.0 26.0

First premolar 6.4 0.0

Table VII. Pass vs fail breakdown by Angle classification

Passing cases Failing cases

Class I 42 (72.4%) 16 (27.6%)

Class II 33 (53.2%) 29 (46.8%)

Class III 7 (53.8%) 6 (46.2%)

American Journal of Orthodontics and Dentofacial Orthopedics Struble and Huang 7Volume 137, Number 1

Q & A

Turpin: What motivated you initially to tackle thisproject?

Struble: When I gained access to the only sample ofsubjects in the country that participated in the pilotstudy under a prospective, blinded format, I knewthat I had a unique sample of ABO eligible casesto examine. I wanted to compare this group with 2other eligible ABO certification groups to determinewhether there were any significant differences in pre-treatment conditions, treatment characteristics, or or-thodontic outcomes. With the exception of the ABOin its pilot-study analysis, no one has explored theoutcomes or possible predictors of outcome success.In light of the recent changes in the ABO certifica-tion process, I thought this would be an interestingstudy.

Turpin: Did this study make it any easier for you foryour examination before the ABO directors aftergraduation?

Struble: I have no doubt that the experience fromthis project improved my familiarity with the ABO

certification process and facilitated my preparationfor the examination. I also found that it greatly im-proved my ability to evaluate my cases critically dur-ing finishing stages both during my residency and inprivate practice. This project greatly contributed tomy orthodontic education and clinical proficiency. Ithink I will be a much better orthodontist becauseof this investigation.

Turpin: Do you think that the recent changes in the-ABO’s testing procedure will have the desired ef-fect? Do you think that most residents who havecompleted certification follow through with showingadditional cases in 10 years?

Struble: I commend the ABO for attempting to im-prove our specialty and investigating ways to im-prove our certification processes. However, onlytime will tell how successful these changes will be.Orthodontic residents and new graduates must per-ceive a significant value in achieving certification.If educators, practicing orthodontists, patients, andpeers place no importance on certification in our spe-cialty, the percentage of board-certified orthodontistsmight not markedly improve in the long term.

8 Struble and Huang American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

ONLINE ONLY

Class III camouflage treatment: What are thelimits?

Nikia R. Burns,a David R. Musich,b Chris Martin,c Thomas Razmus,d Erdogan Gunel,e and Peter Nganf

Pittsburgh, Pa, Schaumburg, Ill, and Morgantown, WV

Introduction: The purpose of this study was to determine the skeletal, dental, and soft-tissue changes in re-sponse to camouflage Class III treatment. Methods: Thirty patients (average age, 12.4 6 1.0 years) with skel-etal Class III malocclusions who completed comprehensive nonextraction orthodontic treatment werestudied. Skeletal, dental, and soft-tissue changes were determined by using published cephalometric analy-ses. The quality of orthodontic treatment was standardized by registering the peer assessment rating index onthe pretreatment and posttreatment study models. The change in the level of gingival attachment with treat-ment was determined on the study casts. The results were compared with a group of untreated subjects. Datawere analyzed with repeated measures analysis and paired t tests. Results: The average change in the Witsappraisal was greater in the treated group (1.2 6 0.1 mm) than in the control group (–0.5 6 0.3 mm). The av-erage peer assessment rating index score improved from 33.5 to 4.1. No significant differences were found forthe level of gingival attachments between the treatment and control groups. The sagittal jaw relationship (ANBangle) did not improve with camouflage treatment. A wide range of tooth movements compensated for theskeletal changes in both groups. The upper and lower limits for incisal movement to compensate for ClassIII skeletal changes were 120� to the sella-nasion line and 80� to the mandibular plane, respectively. Greaterincreases in the angle of convexity were found in the treated group, indicating improved facial profiles. Greaterincreases in length of the upper lip were found in the treated group, corresponding to the changes in the hardtissues with treatment. Conclusions: Significant dental and soft-tissue changes can be expected in youngClass III patients treated with camouflage orthodontic tooth movement. A wide range of skeletal dysplasiascan be camouflaged with tooth movement without deleterious effects to the periodontium. However, properdiagnosis and realistic treatment objectives are necessary to prevent undesirable sequelae. (Am J OrthodDentofacial Orthop 2010;137:9.e1-9.e13)

Adeveloping skeletal Class III malocclusion isone of the most challenging problems confront-ing an orthodontist. The prevalence of Class III

malocclusion in the United States was approximately1%.1 However, approximately 16% of patients aged 4to 10 referred to an orthodontist have a diagnosis ofClass III malocclusion.2

Young patients who are diagnosed early with thisproblem can be treated orthopedically with a chincapor protraction facemask to normalize the underlyingskeletal discrepancy. Patients with no growth remainingmust be camouflaged by orthodontic tooth movement orfixed appliances. Camouflage treatment is the displace-ment of teeth relative to their supporting bone to com-pensate for an underlying jaw discrepancy.3 It impliesthat growth modification to overcome the basic problemis not feasible. The technique to camouflage a skeletalmalocclusion was developed as an extraction treatmentand introduced into orthodontics in the 1930s and1940s.3 During that era, extraction to camouflage a skel-etal malocclusion became popular because growth mod-ification had been largely rejected as ineffective, andsurgical correction had barely begun to develop. Thestrategy to camouflage a Class III malocclusion usuallyinvolves proclination of the maxillary incisors and ret-roclination of the mandibular incisors to improve thedental occlusion, but it might not correct the underlyingskeletal problem or facial profile. Studies have shown anincrease in the ANB angle, little or no change in the ver-tical dimension, and decreased concavity of the facial

aPrivate practice, Pittsburgh, Pa.bPrivate practice, Schaumburg, Ill.cAssociate professor, Department of Orthodontics, West Virginia University,

Morgantown.dProfessor and chair, Department of Diagnostic Service, West Virginia Univer-

sity, Morgantown.eProfessor, Department of Statistics, West Virginia University, Morgantown.fProfessor and chair, Department of Orthodontics, West Virginia University,

Morgantown.

The authors report no commercial, proprietary, or financial interest in the prod-

ucts or companies described in this article.

Reprint requests to: Peter Ngan, West Virginia University, Department of Ortho-

dontics, 1073 Health Science Center North, PO Box 9480, Morgantown, WV

26506; e-mail, [email protected].

Submitted, January 2009; revised and accepted, May 2009.

0889-5406/$36.00


doi:10.1016/j.ajodo.2009.05.017

9.e1


profile with Class III camouflage treatment.4-9 However,little information is available on possible tooth move-ments to camouflage this type of skeletal malocclusion.Our objective in this study was to determine the skeletal,dental, and soft-tissue changes in response to camou-flage Class III treatment. The null hypothesis was thatthere are no significant differences in the skeletal, den-tal, and soft-tissue changes between treated and controlClass III samples.


Forty-one patients, selected from the office files ofan author (D.R.M.), had completed Class III camouflagetreatment. The criteria for selection included (1) ClassIII molar relationship or mesial step in the mixed denti-

tion, (2) concave facial profile, (3) Wits appraisal\–1.5mm or ANB angle\1.0�, (4) a reduction in peer assess-ment rating (PAR) score .30%, (4) nonextraction com-prehensive orthodontic treatment, and (5) high-qualitypretreatment and posttreatment orthodontic records.Exclusion criteria included (1) dentofacial anomaliessuch as cleft lip and palate, (2) extracted or missingteeth, and (3) periodontal disease. Four patients wereeliminated because they had extraction treatment; 5were eliminated because of inadequate reductions inPAR scores; 2 were eliminated because there were nocontrol subjects of similar age, sex, and craniofacialmorphology to match them. No patient was eliminatedbecause of poor records. The final sample consisted of30 white patients (11 boys, 19 girls; average age, 12.4

Table I. Starting craniofacial morphology of treated and control samples

Skeletal and dental measurements

Treated Control

Variable Mean SD Mean SD Difference P value Sig

Sagittal (mm)

Skeletal Olp-A 70.67 5.83 67.8 4.14 2.87 0.03 *

Olp-B 75.69 7.87 72.72 5.83 2.97 0.1 NS

Olp-Pg 79.69 9.09 76.20 6.45 3.49 0.09 NS

Wits –7.16 2.81 –6.14 2.31 –1.02 0.12 NS

Co-ANS 89.46 7.05 86.69 4.8 2.77 0.08 NS

Co-Pg 113.34 9.37 109.45 6.46 3.89 0.06 NS

Dental Is/Olp 78.57 6.89 75.84 5.28 2.73 0.09 NS

Ii/Olp 76.4 6.48 73.90 5.64 2.5 0.11 NS

Overjet 2.11 2.12 1.92 1.91 0.19 0.71 NS

Ms/Olp 49.38 5.77 46.92 4.13 2.46 0.06 NS

Mi/Olp 53.06 5.9 51.73 4.43 1.33 0.33 NS

Molar relationship –3.7 2.01 –4.82 1.94 1.12 0.03 *

Vertical (mm)

Skeletal N-A 50.22 4.68 48.27 3.18 1.95 0.06 NS

ANS-Me 63.58 6.43 62.2 4.76 1.38 0.34 NS

Dental Is-NL 26.71 2.82 26.14 2.56 0.57 0.41 NS

Ii-ML 36.68 4.36 35.96 2.56 0.72 0.43 NS

Overbite 1.1 2.15 0.48 1.74 0.62 0.22 NS

Msc-NL 21.49 2.54 20.93 2.31 0.56 0.37 NS

Mic-ML 28.98 3.25 27.8 2.2 1.18 0.1 NS

ILG 4.3 5.09 0.93 1.65 3.37 0.001 †

Angular (�)Skeletal SNA 79.56 3.54 74.32 3.79 5.24 0.0001 †

SNB 80.1 4.11 75.17 4.49 4.93 0.0001 †

ANB –0.46 2.74 –0.85 2.19 0.39 0.55 NS

ANL-ML 33.68 6.16 32.97 5.74 0.71 0.64 NS

SNL-OL 17.36 4.82 17.54 5.52 –0.18 0.89 NS

SNL-NL 7.66 3.75 7.52 3.45 0.14 0.87 NS

Dental Is/SNL 107.36 6.93 103.32 5.9 4.04 0.01 *

Is-FH 118.03 6.77 114.36 4.53 3.67 0.01 *

Ii/ML 89.05 7.79 84.22 6.34 4.83 0.01 *

U1-NL 114.63 6.9 111.19 4.97 3.44 0.03 *

U1-L1 129.91 10.61 119.7 7.33 10.21 0.0001 †

NS, No significant difference between the means of the treatment and control groups at T1; Sig, significance.

*P \0.05; †P \0.001.

9.e2 Burns et al American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

6 1.0 years). The mean treatment time was 2 years 2months 6 7 months. Lateral cephalometric radiographswere taken before treatment (T1) and after treatment(T2). The cephalometric analyses used to evaluate skel-etal, dental, and soft-tissue changes were described inthe literature.10-13 The quality of orthodontic treatmentwas standardized by registering the PAR index on thepretreatment and posttreatment dental casts. CertifiedPAR calibration was obtained from Ohio State Univer-sity, Columbus, Ohio, before this project. The changein the level of gingival attachment with treatment wasmeasured on the study casts.

The control group consisted of serial radiographs of30 white subjects (11 boys, 19 girls) from the Bolton-Brush Study in Cleveland, Ohio, who were matched byage, sex, and craniofacial morphology to the experimen-tal sample. There were no significant differences in skel-etal age between the treated and the control groups. Aselection of cephalometric records describing the initialcraniofacial morphology of the control and treated sub-jects is shown in Table I. Significant differences werefound with variables OLp-A, molar relationship, SNA,SNB, ILG, Is/SNL, Is/FH, Ii/ML, U1-NL, and U1-L1.

The PAR index was used in this study to evaluate thequality of camouflage orthodontic treatment. The indexwas originally developed to assess how well orthodontictreatment reduces the severity of malocclusion.14 Ascore was assigned based on various occlusal traitsthat make up a malocclusion. The total score representsthe degree to which a person’s occlusion deviates fromnormal alignment. The difference in scores betweenpretreatment and posttreatment reflects the improve-ment or success of the treatment. According to Feghali

et al,15 a reduction in the PAR score of 22 or more pointsindicates ‘‘great improvement,’’ a reduction of 30% in-dicates an ‘‘improved condition,’’ and a reduction of lessthan 30% indicates ‘‘no improvement.’’ In this study,only patients who had a 30% or greater reduction intheir PAR scores were included.

To assess the periodontium with the study casts, wemeasured the change in the level of gingival attachmentwith an electronic caliper (Ultra-Cal Mark III, Fowler-Sylvec, Boston, Mass) and a cephalometric protractor(3M Unitek, Monrovia, Calif) for the 4 mandibular inci-sors. Crown height was measured from the deepest pointof the curvature of the facial vestibulogingival margin tothe incisal edge of the incisors. Measurements weremade to the nearest 0.1 mm with a Boley gauge. Apaired t test was used to evaluate treatment changes(T2-T1) in the treated group and growth changesbetween the serial radiographs (t2-t1) in the controlgroup. A repeated measures analysis was used to assessthe differences between the treated and the controlgroups (T2-T1)-(t2-t1).

Cephalometric changes during treatment were eval-uated on the lateral cephalometric radiographs taken atT1 and T2 for the treated sample and at t1 and t2 for thecontrol sample. All radiographs were analyzed by usinga combination of landmarks from various traditionalcephalometric analyses (Figs 1 and 2).10-13 Analysisof the sagittal skeletal and dental changes were recordedalong the occlusal plane (OL) and to the occlusal planeperpendicular (OLp) obtained from the radiographs at t1and T1. The OL and the OLp from the t1 and T1 tracings

Fig 1. Cephalometric landmarks used for hard-tissuemeasurements on lateral cephalograms.

Fig 2. Cephalometric landmarks used for soft-tissuemeasurements on lateral cephalograms.

American Journal of Orthodontics and Dentofacial Orthopedics Burns et al 9.e3Volume 137, Number 1

formed the reference grid for all sagittal measurementsbetween OLp and the cephalometric landmarks. Thegrid was then transferred to the radiographs at t1 andT2, superimposing the tracings on the sella-nasion line(SNL) and along the anterior cranial base structure.The distance between OLp and the cephalometric land-marks were measured. Overjet and molar relationshipwere then calculated by summing the skeletal and dentalcontributions. A matched-pairs test was used to evaluatethe significant treatment changes between T1 and T2 forthe treated sample and t1 and t2 for the control sample.A repeated measures analysis was performed to evalu-ate the changes between the 2 samples (T2-T1)-(t2-t1).

RESULTS

Errors in locating, superimposing, and measuringthe changes of the landmarks were calculated. Cron-bach’s correlation coefficient of reliability showed thatall sagittal, vertical, and angular measurements andtime periods were greater than 0.97; this indicateshigh reliability.

Sex differences were determined on all the testedvariables. Significant differences were found only forthe variable SN-NL and soft-tissue variables Ls-U1and Ns-Ls/FH (data not shown). Male and female sub-jects were then combined for subsequent analyses.Table II shows the sagittal, vertical, and angular

Table II. Sagittal, vertical, and angular skeletal and dental measurements at T1 and T2 for subjects in the treated group


T1 T2T2-T1

Variable Mean SD Min Max Mean SD Min Max Mean P value Sig

Sagittal (mm)

Skeletal Olp-A 70.67 5.83 59.9 83.5 72.48 6.24 61.3 88 1.81 0.0004 ‡

Olp-B 75.69 7.87 60.8 92.3 77.63 8.53 62.6 95.5 1.94 0.0006 ‡

Olp-Pg 79.69 9.09 65 96.9 82.89 10.1 64.9 102.3 3.2 0.0001 ‡

Wits –7.16 2.81 –12 –1.5 –5.98 2.92 –12 1 1.18 0.002 ‡

Co-ANS 89.46 7.05 69.9 110.6 93.66 7.91 75 113.8 4.2 0.0001 ‡

Co-Pg 113.34 9.37 91.6 133.6 119.46 11.2 96.9 119.46 6.12 0.0001 ‡

Dental Is/Olp 78.57 6.89 65.9 91.1 80.28 6.91 68.4 94.8 1.71 0.03 *

Ii/Olp 76.4 6.48 54.1 87.1 78.38 7.2 66.6 92.1 1.98 0.002 †

Overjet 2.11 2.12 –5.5 6 2.03 1.28 –2.6 3.8 –0.08 0.53 NS

Ms/Olp 49.38 5.77 37.1 60.2 53.02 5.31 43.4 63.8 3.64 0.0001 ‡

Mi/Olp 53.06 5.9 40.9 64 56.33 6.02 46.3 69.1 3.27 0.0001 ‡

Molar

relationship

–3.7 2.01 –12.5 –1.1 –3.3 1.57 –6.1 2.2 0.4 0.42 NS

Vertical (mm)

Skeletal N-A 50.22 4.68 42.9 63.1 52.57 4.61 43.6 66.2 2.35 0.0001 ‡

ANS-Me 63.58 6.43 53.1 74.1 67.4 6.9 55.4 77.8 3.82 0.0001 ‡

Dental Is-NL 26.71 2.82 20.6 31.6 27.66 3 22.7 32.8 0.95 0.002 †

Ii-ML 36.68 4.36 23.1 42.7 39.27 3.64 33 45.4 2.59 0.0001 ‡

Overbite 1.1 2.15 –5.2 5.3 1.04 0.87 –0.8 3 –0.06 0.85 NS

Msc-NL 21.49 2.54 15.4 26.1 23.5 2.45 18.9 28.8 2.01 0.0001 ‡

Mic-ML 28.98 3.25 22.7 36.4 30.88 2.74 25.5 36.6 1.9 0.001 ‡

ILG 3.61 3.67 0 13.9 0.41 1.47 0 7.4 –3.2 0.0003 ‡

Angular (�)Skeletal SNA 79.56 3.54 70 87 78.38 4.23 70 89.0 –1.18 0.32 NS

SNB 80.1 4.11 69.5 89 79.33 4.69 67.5 90 –0.77 0.06 NS

ANB –0.46 2.74 –7 4 –1.26 2.09 –7 2 –0.8 0.04 *

ANL-ML 33.68 6.16 21 53 33 7.55 20.5 58 –0.68 0.69 NS

SNL-OL 17.36 4.82 10 28 16.68 5.49 8 32 –0.68 0.3 NS

SNL-NL 7.66 3.75 0 14 8.08 4.47 –2 18 0.42 0.24 NS

Dental Is/SNL 107.36 6.93 91 123 108.91 6.77 96.5 127 1.55 0.14 NS

Is-FH 118.03 6.77 108 138 120.1 6.91 109 138 2.07 0.07 NS

Ii/ML 89.05 7.79 74 105 89.76 8.54 74 106 0.71 0.48 NS

U1-NL 114.63 6.9 99 134 116.31 6.76 107 134 1.68 0.15 NS

U1-L1 129.91 10.61 109 148 128.2 10.09 108 144 –1.71 0.28 NS

NS, No significant difference between the means at T1 and T2; Min, minimum; Max, maximum; Sig, significance.

*P \0.05; †P \0.01; ‡P \0.001.


January 2010

measurements at T1 and T2 for all subjects in the treatedgroup. Significant differences were found in 10 of 12variables in the sagittal measurements, 7 of 8 variablesin the vertical measurements, and 1 angular measure-ment. Table III shows the sagittal, vertical, and angularmeasurements at t1 and t2 for all subjects in the controlgroup. Significant differences were found in 10 of 12variables in the sagittal measurements, 5 of 8 variablesin the vertical measurements, and 6 angular measure-ments. Table IV compares the skeletal and dentalchanges between the treated and control groups. Forsagittal changes, significant differences were found forthe variables OLp-B, Wits, Is/OLp, and Ii/OLp. Greater

forward movement of the mandible was found in thecontrol group (P \0.01). The Wits appraisal wasdecreased in the treatment group (–7.16 to –5.98) butincreased in the control group (–6.14 to –6.67),P \0.002. The average maxillary incisor inclinationwas retroclined with treatment but proclined withgrowth in the control group (P \0.02). The averagemandibular incisor inclination was proclined with treat-ment but retroclined with growth in the control group(P \0.03).

No significant differences were found in overjet be-tween the treated and control groups (Fig 3). The aver-age changes in overjet in the treated and control groups

Table III. Sagittal, vertical, and angular skeletal and dental measurements at t1 and t2 for all subjects in the controlgroup


t1 t2t2-t1


Sagittal (mm)

Skeletal Olp-A 67.8 4.14 60.2 77.8 70.52 5.3 60.9 83.5 2.72 0.0001 ‡

Olp-B 72.72 5.83 62.2 85.2 76.7 6.92 61.7 89.8 3.98 0.0001 ‡

Olp-Pg 76.2 6.45 64.6 89.5 81.04 7.67 66.1 94.2 4.84 0.0001 ‡

Wits –6.14 2.31 –12.3 –2.8 –6.67 2.68 –12.7 –1.4 –0.53 0.2 NS

Co-ANS 86.69 4.8 77.3 95.4 90.99 4.86 83.9 102.3 4.3 0.0001 ‡

Co-Pg 109.45 6.46 97.5 123.8 116.38 6.43 106.3 131.7 6.93 0.0001 ‡

Dental Is/Olp 75.84 5.28 67.2 88.5 79.14 6.36 66.7 92.6 3.3 0.0001 ‡

Ii/Olp 73.9 5.64 62.6 86.7 77.72 6.56 66.7 93.3 3.82 0.0001 ‡

Overjet 1.92 1.91 –2.5 8.3 1.39 2.34 –3.8 8 –0.53 0.03 *

Ms/Olp 46.92 4.13 39.6 57.1 50.55 6.13 39.8 65.5 3.63 0.0001 ‡

Mi/Olp 51.73 4.43 42.9 60 55.65 6.27 46.4 70.7 3.92 0.0001 ‡

Molar

relationship

–4.82 1.94 –9.2 –2 –5.08 2.05 –10.2 –2.3 –0.26 0.49 NS

Vertical (mm)

Skeletal N-A 48.27 3.18 43.4 54.6 51.07 3.95 44.9 60.1 2.8 0.22 NS

ANS-Me 62.2 4.76 50.7 70.6 66.16 6.2 55.3 80.3 3.96 0.0001 ‡

Dental Is-NL 26.14 2.56 19.3 30.5 26.84 2.89 20.5 32.2 0.7 0.007 †

Ii-ML 35.96 2.56 28.2 42.1 38.35 3.28 32.2 46.9 2.39 0.0001 ‡

Overbite 0.48 1.74 –3.9 6.5 0.22 0.86 –1.3 1.9 –0.26 0.43 NS

Msc-NL 20.93 2.31 14.9 25.1 23.09 2.57 17.8 28.1 2.16 0.0001 ‡

Mic-ML 27.8 2.2 23.6 32.8 29.95 3.2 24.5 38.5 2.15 0.0001 ‡

ILG 0.93 1.65 0 6.7 0.62 1.2 0 4.6 –0.31 0.4 NS

Angular (�)Skeletal SNA 74.32 3.79 64.2 82.1 74.96 3.45 65.6 81.2 0.64 0.07 NS

SNB 75.17 4.49 63.2 82.1 76.53 4.28 64.2 85 1.36 0.0001 ‡

ANB –0.85 2.19 –8.5 3.8 –1.56 2.33 –7.6 2.8 –0.71 0.01 *

ANL-ML 32.97 5.74 24.5 44.4 31.72 6.58 20.8 48.1 –1.25 0.03 *

SNL-OL 17.54 5.52 8.5 34 15.2 4.27 8.5 22.7 –2.34 0.001 †

SNL-NL 7.52 3.45 1.4 14.2 7.87 3.79 0 16 0.35 0.33 NS

Dental Is/SNL 103.32 5.9 89.7 113.3 105.11 7.08 88.7 118 1.79 0.04 *

Is-FH 114.36 4.53 105.7 122.7 116 6.65 99.1 131.2 1.64 0.08 NS

Ii/ML 84.22 6.34 70.8 93.5 83.12 5.94 69.9 93.5 –1.1 0.1 NS

U1-NL 111.19 4.97 97.2 122.7 113.07 6.03 99.1 129.3 1.88 0.03 *

U1-L1 119.7 7.33 107.6 136.9 120.17 8.06 104.8 135.9 0.47 0.7 NS

NS, No significant difference between the means at t1 and t2; Min, minimum; Max, maximum; Sig, significance.

*P \0.05; †P \0.01; ‡P \0.001.


were –0.19 and –0.74 mm, respectively. However, largevariations in skeletal and dental changes were found inboth groups that contributed to the changes in overjet(Figs 4-11). The changes ranged from 2 to 8 mm inthe maxillary base and 3 to 9 mm in the mandibularbase in both groups. A similar distribution was foundin both groups. A wide range of incisal changes wasalso noted in both groups to compensate for the skeletalchanges. The changes in mandibular incisor inclinationsranged from –10� to 15� in the treated group and –10� to6� in the control group. The changes in maxillary incisorinclinations ranged from –6� to 12� in the treated groupand –3� to 12� in the control group. No significant dif-

ferences were found in molar relationships betweenthe 2 groups. The average changes in molar relationshipfor the treated and control groups were 0.37 and –0.27mm, respectively.

For vertical changes, significant differences werefound between the treated and the control groups for in-terlabial distance (ILG). A greater decrease in ILG wasfound in the treated group.

For angular changes, significant differences be-tween the treated and control groups were found forthe variable SNB. The forward movement of the mandi-bles (SNB) was less in the treated group compared withthe control group.

Table V shows the sagittal, vertical, and angularsoft-tissue measurements at T1 and T2 for all subjectsin the treated group. Significant differences were foundin 2 of 11 sagittal soft-tissue profile variables, 5 of 6 var-iables in vertical measurements, and 2 of 6 variables insoft-tissue thickness measurements. Table VI shows thesagittal, vertical, and angular soft-tissue measurementsat t1 and t2 for all subjects in the control group. Signif-icant differences were found in 4 of 11 variables in soft-tissue thickness measurements, 5 of 6 variables in verti-cal measurements, and 4 of 6 variables in soft-tissuethickness measurements. No significant differenceswere found in the lip-structure measurements. TableVII compares the soft-tissue changes between thetreated and control groups. For soft-tissue profilechanges, significant differences were found for the vari-ables Ns-Sls/SLs-Pos, Ls/Pn-Pos, Ns-St, Ns-Li, and Ns-Pog. A greater increase in facial convexity was found inthe treated group compared with the control group. Forvertical changes, significant differences were found forthe variable Sn-St. A greater increase in upper lip lengthwas found in the treated compared with the controlgroup.

Table VIII shows the changes in the level of gingivalattachment for the 4 incisors from T1 to T2 for all sub-jects between T1 and T2 in the treated group. Signifi-cant differences were found with the mandibular rightand left lateral incisors. Table IX shows the changesin the level of gingival attachment in the control group.Significant differences were found between t1 and t2 forall 4 incisors. Table X compares the changes in the levelof gingival attachment between the treated and controlgroups; no significant differences were found.

DISCUSSION

This study had several limitations. This was a retro-spective study with a sample from a private orthodonticpractice. The sample was not uniform in skeletal age,age at T1, and treatment period. An attempt was made

Table IV. Comparison of skeletal and dental changes be-tween the treated and control groups (T2-T1)-(t2-t1)


Variable P value Sig

Sagittal measurements

(mm)

Skeletal Olp-A 0.13 NS

Olp-B 0.01 †

Olp-Pg 0.06 NS

Wits 0.002 †

Co-ANS 0.92 NS

Co-Pg 0.57 NS

Dental Is/Olp 0.02 *

Ii/Olp 0.03 *

Overjet 0.29 NS

Ms/Olp 0.99 NS

Mi/Olp 0.48 NS

Molar relationship 0.28 NS

Vertical measurements

Skeletal N-A 0.29 NS

ANS-Me 0.86 NS

Is-NL 0.5 NS

Ii-ML 0.72 NS

Overbite 0.67 NS

Dental Msc-NL 0.74 NS

Mic-ML 0.69 NS

ILG 0.0009 ‡

Angular measurements

Skeletal SNA 0.32 NS

SNB 0.0001 ‡

ANB 0.85 NS

SNL-ML 0.07 NS

SNL-OL 0.08 NS

SNL-NL 0.88 NS

Dental Is/SNL 0.85 NS

Is-FH 0.76 NS

Ii/ML 0.14 NS

U1-NL 0.89 NS

U1-L1 0.27 NS

NS, No significant difference in the means changes over time between

the treatment and control groups; Sig, significance.

*P \0.05; †P \0.01; ‡P \0.001.


January 2010

to match the starting craniofacial morphology of thetreated and the control groups. The skeletal differences(ANB) in both groups were similar. However, the start-ing SNA and SNB angles were greater in the treatedsample. Maxillary and mandibular incisor proclinationswere greater in the treated group compared with thecontrol group. Long-term data on these patients werenot available to show whether camouflage tooth move-ment was stable after growth.

Camouflage treatment did not result in improvementin the sagittal jaw relationship. In both groups, the jawrelationships became worse with treatment because ofdisproportional growth of the maxilla and the mandible.Most patients in this study started treatment at the be-

ginning of the growth spurt (cervical vertebral matura-tion stage 2 or 3). It is therefore not surprising thatskeletal dysplasia became worse after camouflagetreatment.

Most patients who received camouflage orthodontictreatment were followed for several years after correc-tion of any centric occlusion-centric relation discrep-ancy to evaluate the changes in Wits appraisal withgrowth. Stellzig-Eisenhower et al16 reported that theWits appraisal was the most discriminative in determin-ing whether the developing Class III malocclusionshould be treated by camouflage treatment or surgery.The average Wits appraisal for patients who weresuccessfully treated with camouflage treatment was

Fig 3. Sagittal skeletal and dental changes in the treated and control groups.

Changes in Maxillary Base (A-OLp) in Treatment

Group (mm)

02468101214161820

2 4 6 8

Maxillary Base Changes (mm)

Nu

mb

er o

f S

ub

je

cts

Fig 4. Changes in the maxillary base in the treatedgroup (mm).

Changes in Maxillary Base (A-OLp) in Control

Group (mm)

02468101214161820

2 4 6 8

Maxillary Base Changes (mm)

Nu

mb

er o

f S

ub

je

cts

Fig 5. Changes in the maxillary base in the control group(mm).


�4.6 6 1.7. In our study, patients who had a Wits appraisalbetter than –5.0 had camouflage orthodontic treatment.

A greater improvement in the Wits appraisal wasfound in the treated group. This can be attributed toa decrease in the occlusal plane inclination with ClassIII treatment mechanics, resulting in a decreased SNBangle, extrusion of the posterior molars, and an in-creased mandibular plane angle. The average overjet

remained relatively unchanged in both groups, butmany skeletal and dental changes were observed inthese groups. In the control group, the maxillary inci-sors were proclined, and the mandibular incisors wereretroclined to compensate for the skeletal changes dur-ing the studied period. In the treated group, the maxil-lary incisors were retroclined, and the mandibularincisors were proclined with treatment, decreasing thedental compensation to skeletal discrepancies. The

Changes in Mandibular Base (Pg-OLp) in Treated

Group (mm)

0

5

10

15

3 6 9 12

Mandibular Base Changes (mm)

Nu

mb

er

of S

ub

je

cts

Fig 6. Changes in the mandibular base in the treatedgroup (mm).

Changes in Mandibular Base (Pg-OLp) in Control

Group (mm)

0

5

10

15

3 6 9 12

Mandibular Base Changes (mm)

Nu

mb

er

of S

ub

je

cts

Fig 7. Changes in the mandibular base in the controlgroup (mm).

0

1

2

3

4

5

6

10 7.5 5 2.5 0 2.5 5 7.5 10

Nu

mb

er o

f S

ub

je

cts

Lower Incisor Changes (°)

Changes of Lower Incisor Inclination

(Li/ML) in the Treatment Group (degrees)

Fig 8. Changes in mandibular incisor inclinations in thetreated group (�).


(Ii/ML) in the Control Group (degrees)

0

2

4

6

8

10

-10 -8. 5 -6 -4 -2 0 2 4 6Lower Incisor changes (º)

Nu

mb

er

of S

ub

jects

Fig 9. Changes in mandibular incisor inclinations in thecontrol group (�).

Changes in Upper Incisor Inclination

(Is/SNL) in Treatment Group (degrees)

0

5

10

15

-6 -3 0 3 6 9 12

Upper Inicosr changes (º)

Nu

mb

er

of S

ub

je

cts

Fig 10. Changes in maxillary incisor inclinations in thetreated group (�).


(Is/SNL) in Control Group (degrees)

0

5

10

15

-3 0 3 6 9 12

Upper Incisor changes (º)

Nu

mb

er

of S

ub

je

cts

Fig 11. Changes in maxillary incisor inclinations in thecontrol group (�).


January 2010

results for the control group agree with studies showingthat skeletal Class III compensation tends to worsenwith age without treatment.17,18 The results for thetreated group contrast with those of Troy et al.9 In theirstudy, the maxillary incisors after camouflage treatmentwere more proclined or compensated, and the mandib-ular incisors were more retroclined than at pretreat-ment. The main difference was that all subjects intheir study had already experienced their growth spurt(cervical vertebral maturation stage 4, 5, or 6),whereas, in our study, all subjects were experiencingtheir growth spurt (cervical vertebral maturation stage2 and 3). Long-term data from our study will confirmwhether tooth movements are similar when growth iscompleted.

Variability in growth and response to treatment werealso observed in this study. The average mandibular in-cisor angulation (Ii/ML) with treatment was 90.2�,which is close to the norm. However, variations in indi-vidual responses ranged from –10� to 10�, equivalent toa range of Ii/ML from 80� to 106�. The average maxil-lary incisor angulation (Is/SNL) with treatment was108�, which is close to the norm, but variations in indi-vidual responses ranged from –6� to 12�, equivalent toa range in Is/SNL of 102� to 120�. In a study of adultClass III patients (average age, 26.7 years) camouflagedwith orthodontic treatment, the mean maxillary incisalangulation (U1-SN) after treatment was 112.1� (range,95�-132�), and the mandibular incisal angulation(L1-MP) was 82.4� (range, 65�-100�).19 Casko and

Table V. Sagittal, vertical, and angular soft-tissue measurements at T1 and T2 for all subjects in the treated group

Total soft-tissue measurements

T1 T2 T2-T1

Mean SD Min Max Mean SD Min Max Mean P value Sig

Sagittal relationship of

soft tissue profile

Ns-Sls/ –4.89 8.05 –22 15.6 –7.12 6.3 –20.6 9 –2.23 0.09 NS

Sls-Pos(mm)

Ls/Pn-Pos (mm) 81.67 10.21 62.6 100.4 81.74 10.55 62.6 105.9 0.07 0.93 NS

Li/Pn-Pos (mm) 80.08 15.47 15.5 103.6 82.76 11 63.4 106.5 2.68 0.37 NS

Pn/Ns (mm) 26.56 4.77 17.8 38 28.08 5.24 20.3 39.7 1.52 0.0005 ‡

Ns-Sn (mm) –12.41 6.77 –25.5 11.7 –11.33 14.73 –24.5 65 1.08 0.54 NS

Ns/Sls (mm) –11.41 6.77 –25.7 11.9 –11.87 7.03 –24.2 15.7 –0.46 0.32 NS

Ns/Ls (mm) –16.04 8.4 –31.8 18.3 –15.44 6.84 –29.8 7.6 0.6 0.33 NS

Ns-St (mm) –11.06 6.53 –26.3 11.1 –10.07 6.09 –25.1 1 0.99 0.17 NS

Ns/Li (mm) –17.21 8.36 –35.3 15.8 –16 8.63 –33.7 18.8 1.21 0.02 *

Ns/Ils (mm) –12.27 7.58 –31.4 16 –12.03 6.55 –30.4 1.8 0.24 0.72 NS

Ns-Pog (mm) –17.31 8.01 –38.5 8.5 –17.35 8.61 –39.3 9.7 –0.04 0.93 NS

Vertical relationship of

soft tissue profile

Sn-Ms (mm) 69.68 6.54 60 82.2 72.89 6.43 61 84 3.21 0.0001 ‡

Sn-St (mm) 18.55 2.83 12.9 23.4 21.59 2.53 13.7 25.7 3.04 0.0001 ‡

St-Ms (mm) 47.73 5.28 37.8 58.2 50.52 5.33 39.8 60.9 2.79 0.0001 ‡

St-Ils (mm) 16.88 2.47 12.1 21.5 17.4 2.34 12.1 22.1 0.52 0.18 NS

Ns-Ms (mm) 120.23 10.11 103.2 142.4 125.03 10.31 105.4 145.3 4.8 0.0001 ‡

Ns-Sn (mm) 53.2 5.68 43.2 66.5 55.26 5.86 45.8 71.2 2.06 0.0001 ‡

Soft-tissue thickness

Sn-A (mm) 15.81 2.95 9.9 23 17.33 2.79 10.2 23 1.52 0.006 †

Ls-U1 (mm) 13.27 2.51 8.9 19 13.36 2 9.7 17.2 0.09 0.81 NS

Li-L1 (mm) 14.3 2.37 9.1 19.5 14 2.29 10.7 22.7 –0.3 0.41 NS

Pos-Pog (mm) 11.68 1.72 7.7 16.4 11.59 1.9 8.5 15.9 –0.09 0.73 NS

Sls-A (mm) 14.99 3.1 8.1 21.9 16.25 2.55 9.1 21.2 1.26 0.01 *

Ils-B (mm) 11.27 1.83 7.7 15.6 12.05 2.32 9 21.1 0.78 0.11 NS

Lip structure

Ns-Ls/FH (�) 100.51 4.89 93 115 99.34 5.57 77 108 –1.17 0.31 NS

Li-Ils/FH (�) 52.65 14.44 17 81 53.1 11.24 24 71 0.45 0.84 NS

Ils-Pos-Ls (�) –17.6 9.69 –32 17 –17.65 5.12 –26 –10 –0.05 0.97 NS

Li/Pos-Ls (�) 1.93 5.54 –7 12 0.75 4.77 –9 13 –1.18 0.18 NS

NS, No significant difference between the means at T1 and T2; Min, minimum; Max, maximum; Sig, significance.

*P \0.05; †P \0.01; ‡P \0.001.


Shepherd20 reported on the cephalometric values ofa sample of adults with normal occlusion and found var-iations in skeletal and dental parameters that were far be-yond mean values. Dietrich21 and Guyer et al22 alsoreported variability of skeletal Class III relationships us-ing cephalometeric analysis. Several Class III patients inthis study had a positive overjet because the underlyingskeletal malocclusion was compensated by retroclina-tion of the mandibular incisors. The objective of camou-flage treatment in these patients was to normalize theunderlying skeletal discrepancies and place the incisorsin the medullary trough to prevent bony dehiscence. Ithas also been shown that overjet is not a good predictorof sagittal relationship in Class III subjects.23

As for vertical changes, a decreased distance betweenthe upper and lower lip (ILG) was observed in the treated

group. The improvement in lip competency agrees withother studies that demonstrated that a decrease in facialconcavity improved the posture of the lips.6,24 Althoughthe following variables were not significant, our resultsshowed a trend that the maxillary and mandibular inci-sors, as well as the molars, had more extrusion in thetreated group than in the control group; this can be ex-plained by the use of Class III elastics. There was counter-clockwise rotation of the occlusal plane in the treatedgroup. Lin and Gu6 reported similar results and foundthat the relative extrusion of the mandibular incisors in re-lation to the maxillary molars during Class III traction ofelastics seemed to contribute to the counterclockwise ro-tation of the occlusal plane. The increase in mandibularplane angulation in the treated group could be attributedto the extrusion of the mandibular molars.

Table VI. Sagittal, vertical, and angular soft-tissue measurements at t1 and t2 for all subjects in the control groups

Total soft-tissue measurements

t1 t2 t2-t1

Mean SD Min Max Mean SD Min Max Mean P value Sig


soft-tissue profile

Ns-Sls/Sls-Pos(mm) –9.6 4.9 –22 2.5 –7.46 6.92 –25.2 15 2.14 0.06 NS

Ls/Pn-Pos (mm) 76.55 7.69 63 94 80.98 8.94 60.4 97.9 4.43 0.0001 †

Li/Pn-Pos (mm) 77.76 8.67 60.9 97.2 82.66 9.57 62.2 100.9 4.9 0.0002 †

Pn/Ns (mm) 27.55 8.33 15.3 66.6 28.09 4.52 19.4 36 0.54 0.75 NS

Ns-Sn (mm) –13.67 3.83 –25.6 –5.1 –13.98 4.77 –26.6 –6.8 –0.31 0.58 NS

Ns/Sls (mm) –12.66 3.77 –25.1 –4.3 –13.27 4.56 –25.3 –5.9 –0.61 0.31 NS

Ns/Ls (mm) –16.44 4.07 –30 –8.8 –17.01 4.94 –30.6 –9.8 –0.57 0.30 NS

Ns-St (mm) –11.15 4.2 –24.8 –2.4 –12.01 4.77 –23.9 –4.7 –0.86 0.1 NS

Ns/Li (mm) –17.52 4.96 –32.4 –7.6 –18.46 5.68 –32.1 –7.3 –0.94 0.1 NS

Ns/Ils (mm) –12.95 5.12 –26.3 –2.4 –14.42 5.78 –26.5 –2.9 –1.47 0.03 *

Ns-Pog (mm) –16.17 5.97 –28.7 –3.7 –18.61 6.77 –31.1 –4.9 –2.44 0.0003 †


soft-tissue profile

Sn-Ms (mm) 66.02 4.7 59.7 76 70.32 5.4 59.8 84.1 4.3 0.0001 †

Sn-St (mm) 18.14 2.19 12.8 22.8 18.71 2.82 14.8 25.2 0.57 0.32 NS

St-Ms (mm) 47.02 3.44 40.9 55.9 51.14 4.04 41.9 61.7 4.12 0.0001 †

St-Ils (mm) 16.77 2.57 11.9 22.5 18.37 2.7 13.9 24.4 11.6 0.02 *

Ns-Ms (mm) 120.94 12.53 107.2 178.2 126.58 7.52 114 143 5.64 0.03 *

Ns-Sn (mm) 56.17 4.09 47.7 64 59.29 4.5 49.1 66.7 3.12 0.0001 †


Sn-A (mm) 15.98 2.04 12 19.8 16.92 2.61 12.4 22.8 0.94 0.01 *

Ls-U1 (mm) 12.07 1.63 8.9 15 12.8 1.84 9.3 16.3 0.73 0.02 *

Li-L1 (mm) 12.93 1.93 8.9 15.6 13.07 2.06 8.9 18.9 0.14 0.69 NS

Pos-Pog (mm) 10.45 2.03 6.6 14.3 11.55 2.11 7.8 16 1.1 0.01 *

Sls-A (mm) 15.8 1.83 11.4 19.2 16.68 2.36 13.3 22.4 0.88 0.01 *

Ils-B (mm) 11.35 1.93 7.5 15.2 11.63 1.99 6.6 14.6 0.28 0.47 NS

Lips structure

Ns-Ls/FH (�) 95.39 4.8 83.1 105.7 95.81 4.94 86.8 107.6 0.42 0.55 NS

Li-Ils/FH (�) 53.33 7.57 36.8 67 55 10.44 32.1 79.3 1.67 0.39 NS

Ils-Pos-Ls (�) –13.02 3.77 –18.9 –4.7 –14.88 4.75 –22.7 –6.6 –1.86 0.07 NS

Li/Pos-Ls (�) 3.08 3.57 –5.7 10.4 1.98 4.38 –6.6 14.2 –1.1 0.16 NS


*P \0.05; †P \0.001.


January 2010

As for angular changes, patients in the control grouphad a more forward position of the mandibular base an-gle in relation to the cranial base (SNB). The differencemight be explained by the Class III treatment mechanics,which have a tendency to extrude the molars, increasethe mandibular plane angle, and decrease the SNB angle.These changes were not observed in the control group,which experienced a decrease in mandibular plane an-gle.24 Maxillary incisor position and inclination werefound to be greater in the control group. Proclinationof the maxillary incisors in response to either growthor treatment in Class III patients was reported by Bac-cetti et al,25 Lin and Gu,6 Moullas and Palomo,26 andDaher and Caron.27 No studies have compared treatedand untreated samples. In our study, the maxillary inci-

sors in the control group compensated for the skeletaldiscrepancy and thereby continued to procline withgrowth. The use of Class III elastics in the treated groupprovided forward movement of the maxillary molars,decreasing the incisors’ proclination. Mandibular inci-sor inclination in relation to the mandibular plane(Ii-ML) was greater in the treated sample. The treatedsample also had a greater interincisal angle (U1-L1).An increase in the interincisal angle usually means lessproclination of the incisors. In the present study, camou-flage treatment results in decompensation or more up-right maxillary and mandibular incisors than thecontrol group. Although proclination of the maxillary in-cisors may still takes place during treatment, it is lesspronounced than those resulted from growth alone. Stud-ies have shown that skeletal Class III discrepanciesworsen with age.18-20 Thus, the difficulty in treating a de-veloping Class III malocclusion successfully increaseswith time. If the skeletal discrepancy worsens withage, then the interincisal angle will decrease over time.With the advent of temporary anchorage devices, inter-maxillary elastics can be replaced by these devices andintra-arch mechanics. This will minimize extrusion ofthe molars and opening of the mandibular plane.

For sagittal soft-tissue changes, the angle of convex-ity (Ns-Sls/Sls-Pos) increased in the control group anddecreased in the treatment group; this indicates im-proved facial esthetics in the treated sample. The treatedgroup also had a higher mandibular plane angle (Sn-ML) than the control group; this can also contribute toan esthetic profile change. Decreases in convexitywere also found by Lin and Gu6 and Daher and Caron,27

who attributed the changes mostly to the changes inmandibular plane angulation.

For vertical soft-tissue changes, the length of theupper lip (Sn-St) in the treated group was longer thanthe control group. A study with facemask treatmentalso found an increase in the length of the upper lipin the treated group, but it was not significant.24 A pos-sible explanation could be that, since the treated grouphad less incisor proclination than the control group, thelength of the upper lip was greater, assuming that theposition of the teeth influences the position of the softtissues.

The average loss of attachment in the treated groupwas similar to the control group. This was even thoughthe average inclination of the mandibular incisors in thetreated group was more proclined than in the controlgroup. Individual variation in response to treatmentwas noted in both the range of incisor movement andthe response. These results suggest that camouflagetreatment can be successful in various tooth movementswithout deleterious effects to the periodontium.

Table VII. Comparison of soft-tissue changes betweenthe treated and control groups (T2-T1)-(t2-t1)

Soft-tissue measurements P value Sig


soft-tissue profile

Ns-Sls/Sls-Pos 0.01 †

Ls/Pn-Pos (mm) 0.0003 ‡

Li/Pn-Pos (mm) 0.49 NS

Pn/Ns (mm) 0.57 NS

Ns-Sn (mm) 0.46 NS

Ns/Sls (mm) 0.84 NS

Ns/Ls (mm) 0.15 NS

Ns-St (mm) 0.04 *

Ns/Li (mm) 0.006 †

Ns/Ils (mm) 0.07 NS

Ns-Pog (mm) 0.003 †


soft-tissue profile

Sn-Ms (mm) 0.19 NS

Sn-St (mm) 0.0012 †

St-Ms (mm) 0.14 NS

St-Ils (mm) 0.16 NS

Ns-Ms (mm) 0.73 NS

Ns-Sn (mm) 0.15 NS


Sn-A (mm) 0.36 NS

Ls-U1 (mm) 0.2 NS

Li-L1 (mm) 0.38 NS

Pos-Pog (mm) 0.01 NS

Sls-A (mm) 0.54 NS

Ils-B (mm) 0.41 NS

Lip structure

Ns-Ls/FH (�) 0.25 NS

Li-Ils/FH (�) 0.68 NS

Ils-Pos-Ls (�) 0.29 NS

Li/Pos-Ls (�) 0.94 NS

Repeated measurements analysis was used for testing the interaction

effect.

NS, No significant difference in the mean changes over time between

the treatment and control groups.

*P \0.05; †P \0.01; ‡P \0.001.


However, clinicians should be cautioned that, in thisstudy, both groups had significant growth changes dur-ing the study period. When treating a Class III patient,the clinician should monitor the patient so that he orshe does not grow out of the range of successful camou-flage treatment. If camouflage treatment is planned withthe irreversible step of extraction of premolars, verifica-tion that the goals of treatment can be achieved withnonsurgical treatment approach is essential. Frequently,

a nonextraction preliminary orthopedic stage—eg, 4 to6 months of therapeutic treatment with rapid palatalexpansion, Class III traction, and maxillary anteriorbraces—eliminates the mandibular functional shiftsthat are frequently present and make the Class III prob-lem look worse than it is.28 In addition, patients shouldbe followed for periodontal health after camouflagetreatment. Increased morbidity in long-term evaluationsmeasured by gingival recession has been reported inClass III patients camouflaged by greater dental com-pensations.19

CONCLUSIONS

The null hypothesis that there are no significant dif-ferences in skeletal, dental, and soft-tissue changes be-tween the treated and control groups was rejected. Mostdifferences were attributed to tooth movement to reducedental compensation of the skeletal malocclusion andimprove the facial profile. The range of skeletal anddental changes in response to orthodontic treatment sug-gests that a wide range of skeletal dysplasia can be suc-cessfully camouflaged with tooth movement withoutdeleterious effects to the periodontium. However,proper diagnosis and the establishment of realistic treat-ment objectives by the clinician and the patient are

Table VIII. Change in level of gingival attachments from T1 to T2 for all subjects in the treated groups

Periodontal measurements

T1 T2 T2-T1

Variable Mean SD Min Max Mean SD Min Max Mean P-value Sig

IL-ML 89.05 7.79 74 105 89.76 8.54 74 106 0.71 0.48 NS

LR2 8.11 0.88 6.6 9.8 8.73 0.88 6.8 9.8 0.62 0.0001 *

LR1 8.37 0.76 6.6 9.8 8.56 0.88 6.6 9.7 0.19 0.16 NS

LL1 8.44 0.88 6.6 10 8.64 0.88 6.9 10 0.2 0.15 NS

LL2 8.15 0.85 6 10 8.82 0.92 7.2 10.4 0.67 0.0001 *

NS, No significant difference between the means at T1 and T2.

*P \0.05.

Table IX. Change in level of gingival attachment from t1 to t2 for all subjects in the control group

t1 t2 t2-t1


IL-ML 84.22 6.34 70.8 93.5 83.12 5.94 69.9 93.5 –1.1 0.1 NS

LR2 8.41 1 6.7 10.4 8.74 0.98 6.8 10.6 0.33 0.1 *

LR1 8.55 1.05 6.1 10.7 8.85 1.06 6.4 10.6 0.3 0.04 *

LL1 8.54 1.05 6.2 10.7 8.89 1.12 6.4 11.2 0.35 0.01 *

LL2 8.36 0.9 6.6 9.8 8.84 0.92 6.7 10.5 0.48 0.0001 †


*P \0.05; †P \0.001.

Table X. Comparison of the periodontal changes be-tween the treated and control groups (T2-T1)-(t2-t1)

Angular measurements

Variable P value Sig

IL-ML 0.14 NS

ii/Olp 0.03 *

LR2 0.09 NS

LR1 0.58 NS

LL1 0.4 NS

LL2 0.28 NS

Repeated measurements analysis was used for testing the interaction

effect.

NS, No significant difference in the mean changes over time between

the treatment and control groups; Sig, significance.

*P \0.05.


January 2010

necessary to prevent undesirable sequelae in camouflag-ing a mild to moderate skeletal Class III malocclusion.

REFERENCES

1. Proffit WR, Fields HW, Moray LJ. Prevalence of malocclusion

and orthodontic treatment need in the United States: estimates

from the NHANES III survey. Int J Adult Orthod Orthognath

Surg 1998;13:97-106.

2. Musich DR, Busch MJ. Early orthodontic treatment: current clin-

ical perspectives. Alpha Omegan 2007;100:17-24.

3. Profitt WR, Fields HW, Sarver DM. Contemporary orthodontics.

4th ed. St Louis: Mosby; 2007. p. 6-14, 300-309.

4. Ngan P. Treatment of Class III malocclusion in the primary and

mixed dentitions. In: Bishara SE, editor. Textbook of orthodon-

tics. Philadelphia: W.B. Saunders; 2001. p. 375-6.

5. Costa Pinho T, Torrent J, Pinto J. Orthodontic camouflage in the case

of a skeletal Class III malocclsuion. World J Orthod 2004;5:213-23.

6. Lin J, Gu Y. Preliminary investigation of nonsurgical treatment of

severe skeletal Class III malocclusion in the permanent dentition.

Angle Orthod 2003;73:401-10.

7. Chang HF, Chen KC, Nanda R. Two-stage treatment of a severe

skeletal Class III deep bite malocclusion. Am J Orthod Dentofa-

cial Orthop 1997;111:481-6.

8. Rabie ABM, Wong RWK, Min GU. Treatment in borderline Class

III malocclusion: orthodontic camouflage (extraction) versus

orthognathic surgery. Open Dent J 2008;2:38-48.

9. Troy BA, Shanker S, Fields HW, Vig K, Johnston W. Comparison

of incisor inclination in patients with Class III malocclusion

treated with orthognathic surgery or orthodontic camouflage.

Am J Orthod Dentofacial Orthop 2009;135:146.e1–9.

10. Bjork A. The face in profile: an anthropological x-ray investiga-

tion of Swedish children and conscripts. Lund, Sweden: Berlin-

gska Boktrycheriet 1947;47:58.

11. Pancherz H. The mechanism of Class II correction in Herbst ap-

pliance treatment, a cephalometric investigation. Am J Orthod

1982;82:107-13.

12. Holdaway RA. Soft-tissue cephalometric analysis and its use in or-

thodontic treatment planning. Part I. Am J Orthod 1983;84:1-28.

13. Legan HL, Burstone CJ. Soft tissue analysis for orthognathic sur-

gery. J Oral Surg 1980;38:744-51.


Stephens CD, et al. The development of the PAR index (peer assess-

ment rating): reliability and validity. Eur J Orthod 1992;14:125-39.

15. Feghali R, Nelson S, Hans MG. Comparing orthodontic treatment

outcome of two geographically different clinic samples [abstract].

J Dent Res 1997;76:1181.

16. Stellzig-Eisenhauer A, Lux CJ, Schuster G. Treatment decision in

adult patients with Class III malocclusion: orthodontic therapy or

orthognathic surgery? Am J Orthod Dentofacial Orthop 2002;122:

27-38.

17. Enlow DH. Facial growth. 3rd ed. Philadelphia: W.B. Saunders;

1990. p. 200-21, 434-443.

18. Sugawara J, Mitani H. Facial growth of skeletal Class III

malocclusion and the effects, limitations, and long-term dentofacial

adaptations to chincap therapy. Semin Orthod 1997;3:244-54.

19. Sperry TP, Speidel TM, Isaacson RJ, Worms FW. The role of den-

tal compensations in the orthodontic treatment of mandibular

prognathism. Angle Orthod 1977;47:293-9.

20. Casko JS, Shepherd WB. Dental and skeletal variation within the

range of normal. Angle Orthod 1984;54:5-17.

21. Dietrich UC. Morphological variability of skeletal Class III rela-

tionships as veiled by cephalometric analysis. Cong Eur Orthod

Soc 1970:131-43.

22. Guyer EC, Ellis EE, McNamara JA, Behrents RG. Components of

Class III malocclusions in juveniles and adolescents. Angle

Orthod 1986;56:7-30.

23. Zupancic S, Pohar M, Farcnik F, Ovsenik M. Overjet as a predictor

of sagittal skeletal relationships. Eur J Orthod 2008;30:269-73.

24. Ngan P, Hagg U, Yiu C, Merwin D, Wei SH. Soft tissue and den-

toskeletal profile changes associated with maxillary expansion

and protraction headgear treatment. Am J Orthod Dentofacial

Orthop 1996;109:38-49.

25. Baccetti T, Franchi L, McNamara J. Growth in the untreated Class

III subject. Semin Orthod 2007;13:130-42.

26. Moullas AT, Palomo JM. Nonsurgical treatment of a patient with

a Class III malocclusion. Am J Orthod Dentofacial Orthop 2006;

129(Suppl 4):S111-8.

27. Daher W, Caron J. Nonsurgical treatment of an adult with a Class III

malocclusion. Am J Orthod Dentofacial Orthop 2007;132:243-51.

28. Musich DR, Busch MJ. Clinical management of the Class III

growing patient. Informationen aus Orthodontie & Kieferorthopa-

die 2008;40:13-26.


ONLINE ONLY

Class III camouflage treatment: What are thelimits?

Nikia R. Burns, David R. Musich, Chris Martin, Thomas Razmus, Erdogan Gunel, and Peter Ngan

Pittsburgh, Pa, Schaumburg, Ill, and Morgantown, WV

Introduction: The purpose of this study was todetermine the skeletal, dental, and soft-tissue changesin response to camouflage Class III treatment.

Methods: Thirty patients (average age, 12.4 6 1.0years) with skeletal Class III malocclusions who com-pleted comprehensive nonextraction orthodontic treat-ment were studied. Skeletal, dental, and soft-tissuechanges were determined by using published cephalomet-ric analyses. The quality of orthodontic treatment was stan-dardized by registering the peer assessment rating index onthe pretreatment and posttreatment study models. Thechange in the level of gingival attachment with treatmentwas determined on the study casts. The results were com-pared with a group of untreated subjects. Data were ana-lyzed with repeated measures analysis and paired t tests.

Results: The average change in the Wits appraisalwas greater in the treated group (1.2 6 0.1 mm) than inthe control group (–0.5 6 0.3 mm). The average peer as-sessment rating index score improved from 33.5 to 4.1.No significant differences were found for the level of gin-gival attachments between the treatment and controlgroups. The sagittal jaw relationship (ANB angle) didnot improve with camouflage treatment. A wide rangeof tooth movements compensated for the skeletal changesin both groups. The upper and lower limits for incisalmovement to compensate for Class III skeletal changeswere 120� to the sella-nasion line and 80� to the mandib-ular plane, respectively. Greater increases in the angle ofconvexity were found in the treated group, indicating im-proved facial profiles. Greater increases in length of theupper lip were found in the treated group, correspondingto the changes in the hard tissues with treatment.

Conclusions: Significant dental and soft-tissuechanges can be expected in young Class III patients treatedwith camouflage orthodontic tooth movement. A widerange of skeletal dysplasias can be camouflaged with toothmovement without deleterious effects to the periodon-tium. However, proper diagnosis and realistic treatmentobjectives are necessary to prevent undesirable sequelae.


EDITOR’S SUMMARY

Every time I examine a patient with complete ante-rior crossbite, I ask myself the same question. ‘‘Whatare the limitations to successfully treating this patientwithout surgery?’’ The earlier I focus on the limitationsof camouflage treatment, the more pleased I seem to bewith the final outcome. But what characteristics shouldwe look for when considering Class III nonsurgicaltreatment? The prevalence of Class III malocclusionin the United States is about 1%, but nearly 16% ofthe patients referred to orthodontists between the agesof 4 and 10 have Class III problems. The strategy forcamouflage treatment usually involves proclination ofthe maxillary incisors and retroclination of the mandib-ular incisors to improve the occlusion as well as to cor-rect the underlying skeletal problem. Although thesechanges can be accomplished with a chincap to the man-dible or a protraction facemask to the maxilla, few stud-ies describe the range of tooth movements that can alterthis type of malocclusion without leaving the occlusionin an unstable relationship. The objective of this inves-tigation was to determine the skeletal, dental, and soft-tissue changes that occur in response to camouflageClass III treatment.

The records of 30 patients (age, 12.4 years) were se-lected from the files of an author for study. All had com-pleted Class III camouflage treatment over an average of2 years 2 months. A control group of 30 children,matched by age, sex, and craniofacial morphology,was also identified. Read this entire article online to be-come familiar with the number of variables comparedfrom all available pretreatment and posttreatment ortho-dontic records. As a result of early treatment, thechanges were 2 to 8 mm in the maxillary base and 3to 9 mm in the mandibular base in both the treatedand control groups. Similar distributions were seen inboth groups. The changes in mandibular incisor inclina-tion were –10� to 15� in the treated group and –10� to 6�

in the control group. The changes in maxillary incisorinclination were –6� to 12� in the treated group and–3� to 12� in the control group. No significant differ-ences were found in molar relationships between thechanges in the treated group compared with the controlgroup.


0889-5406/$36.00


doi:10.1016/j.ajodo.2009.09.006

9


Camouflage treatment did not improve the sagittaljaw relationships. In fact, the jaw relationship becameworse with treatment because of the disproportionalgrowth of the maxilla and the mandible. In addition toother changes with additional growth, the average lossof gingival attachment in the treated group was similarto that in the control group. When treating patients ofthis type of problem, clinicians must monitor theirgrowth to determine whether they grow out of the rangeof successful camouflage treatment. In concluding re-marks, the authors noted, ‘‘Most differences were attrib-uted to tooth movement to reduce dental compensation

of the skeletal malocclusion and improve the facial pro-file. The range of skeletal and dental changes in re-sponse to orthodontic treatment suggests that muchskeletal dysplasia can be successfully camouflagedwith tooth movement without deleterious effects to theperiodontium.’’ Because long-term records are not yetavailable for this treatment group, caution is advisedin assuming the permanency of acceptable outcomesat this time. Proper diagnosis and the establishmentof realistic treatment objectives by the clinician and thepatient are necessary to prevent undesirable sequelaein Class III camouflage treatment.

Fig 3. Sagittal skeletal and dental changes in the treated and control groups.

Fig 8. Changes in mandibular incisor inclinations in thetreated group (�).


(Ii/ML) in the Control Group (degrees)

0

2

4

6

8

10

-10 0 2-8.5 -6 -4 -2 4 6Lower Incisor changes (º)

Nu

mb

er o

f S

ub

jects

Fig 9. Changes in mandibular incisor inclinations in thecontrol group (�).

10 Burns et al American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

Q & A

Turpin: Do you plan to gather more records on thesetreated patients to study long-term changes? If so,what do you expect to see?

Authors: Yes, we plan to obtain another series of thissame group 5 to 7 years after the posttreatment re-cords were taken. We might need to do some incisalequilibration to compensate for late-stage growthchanges; also, we expect some mandibular anteriorcrowding—common in most treated orthodontic pa-tients. We will attempt to reexamine the level of gin-gival attachment after treatment. The periodontiumadapts quite well in young patients. It will be inter-esting to see whether the level of gingival attachmentholds up with time.

Turpin: Do you expect any gingival recession in thelong term in some of these camouflage Class IIIpatients after treatment?

Authors: Some patients, especially boys who havethin gingival tissue at the start of treatment, willprobably have some recession. We think this wouldbe related to additional mandibular growth, demandsof incisal function at the camouflaged interincisal

angle, gingival biotype, or extraneous factors such assmoking and bruxism.

Turpin: Do you expect a change in the treatment offuture borderline Class III patients with increaseduse of temporary skeletal anchorage?

Authors: Yes, the range of camouflage skeletal ClassIII treatment should increase with temporary skeletalanchorage. The use of temporary skeletal anchoragedevices (TSADs) such as mini-implants and mini-plates can accelerate maxillary growth more predict-ably in young patients. We have been using TSADsfor camouflage in a late-stage growing boy, extractingthe mandibular second molars and using an implant toretract the mandibular molars and premolar and tobring the third molar into position to replace the sec-ond molar. The use of TSADs will bring new solutionsto old problems. As insurance companies reducecoverage for the treatment of skeletal malocclusions,orthodontists will be forced to treat more skeletalproblems without orthognathic surgery. Great caremust be taken to make the initial differential diagnosisand to plan treatment with awareness of the needfor periodic reassessment of growth changes that canundermine the conservative, nonsurgical approach.


(Is/SNL) in Treatment Group (degrees)

0

5

10

15

-6 -3 0 3 6 9 12

Upper Inicosr changes (º)

Nu

mb

er o

f S

ub

je

cts

Fig 10. Changes in maxillary incisor inclinations in thetreated group (�).


(Is/SNL) in Control Group (degrees)

0

5

10

15

3-3 0 6 9 12

Upper Incisor changes (º)

Nu

mb

er o

f S

ub

je

cts

Fig 11. Changes in maxillary incisor inclinations in thecontrol group (�).

American Journal of Orthodontics and Dentofacial Orthopedics Burns et al 11Volume 137, Number 1

ONLINE ONLY

Active or passive self-ligating brackets? Arandomized controlled trial of comparativeefficiency in resolving maxillary anteriorcrowding in adolescents

Nikolaos Pandis,a Argy Polychronopoulou,b and Theodore Eliadesc

Corfu, Athens, and Thessaloniki, Greece

Introduction: Our aim was to compare the time required to complete the alignment of crowded maxillaryanterior teeth (canine to canine) between Damon MX (Ormco, Glendora, Calif) and In-Ovation R (GAC, CentralIslip, NY) self-ligating brackets. Methods: Seventy patients from the first author’s office were included in thisrandomized controlled trial by using the following inclusion criteria: nonextraction treatment on both arches,eruption of all maxillary teeth, no spaces in the maxillary arch, no high canines, maxillary irregularity indexgreater than 4 mm, and no therapeutic intervention planned involving intermaxillary or other intraoral or extrao-ral appliances including elastics, maxillary expansion appliances, or headgear. The patients were randomizedinto 2 groups: the first received a Damon MX bracket; the second was bonded with an In-Ovation R appliance,both with a 0.022-in slot. The amount of crowding of the maxillary anterior dentition was assessed by using theirregularity index. The number of days required to completely alleviate the maxillary anterior crowding in the 2groups was investigated with statistical methods for survival analysis, and alignment rate ratios for appliancetype and crowding level were calculated with the Cox proportional hazard regression. An analysis of each pro-tocol was performed. Results: No difference in crowding alleviation was found between the 2 bracket sys-tems. Higher irregularity index values were associated with the increased probability of delayed resolving ofcrowding. Conclusions: The use of passive or active self-ligating brackets does not seem to affect treatmentduration for alleviating initial crowding. (Am J Orthod Dentofacial Orthop 2010;137:12.e1-12.e6)

The last decade has witnessed unprecedentedprogress in the development of new applianceswith alternative ligating features. Passive and

active self-ligating appliances with many ligating mech-anisms were introduced to presumably allow for effi-cient sliding mechanics. This feature was linked withseveral presumed effects including lower forces andmoments and higher rates of tooth movement, becauseof the reduced friction and absence of binding of liga-tures on wire. Nonetheless, for most of these appliances,there is little evidence about the characteristics and ca-pabilities claimed by the manufacturers, and, in some

cases, it seems that marketing-derived principles ratherthan scientific evidence is used to substantiate theirwell-advertised clinical performance. Thus, the resul-tant turmoil in this specialty has no support in the liter-ature, since most published trials do not show superiorefficiency of self-ligating brackets regardless of typeor ligating mechanism.1-8

The available evidence on the efficiency of self-ligating brackets derives from a few prospective andrandomized clinical trials, with most prospective andrandomized controlled trials (RCTs) demonstrating nodifference in treatment duration between conventionaland self-ligating brackets.3-7

Recently, the relatively low validity and reliabilityof retrospective studies as opposed to prospective andespecially RCTs have switched investigators’ intereststo the latter type of studies.9,10 RCTs are preferred be-cause of the elimination of selection and outcome biasesof retrospective studies.11 Currently, there is little evi-dence from this type of study on the potential differ-ences between passive and active self-ligatingbrackets in tooth movement rates.

The purpose of this study was to compare the timerequired to complete the alignment of crowded

aPrivate practice, Corfu, Greece.bAssistant professor, Department of Community and Preventive Dentistry,

School of Dentistry, University of Athens, Athens, Greece.cAssociate professor, Department of Orthodontics, School of Dentistry, Aristo-

tle University of Thessaloniki, Thessaloniki, Greece.

The authors report no commercial, financial, or proprietary interest in the


Reprints requests to: Theodore Eliades, 57 Agnoston Hiroon Str, Nea Ionia

GR-14231, Greece; e-mail, [email protected].

Submitted, June 2009; revised and accepted, August 2009.

0889-5406/$36.00


doi:10.1016/j.ajodo.2009.08.019

12.e1


maxillary anterior teeth (canine to canine) between pas-sive and active self-ligating brackets.


Seventy patients were included in this RCT, selectedfrom a large pool of patients from the first author’s of-fice. The following inclusion criteria were used: nonex-traction treatment in both arches, eruption of allmaxillary teeth, no spaces in the maxillary arch, nohigh canines, maxillary irregularity index greaterthan 4 mm, and no therapeutic intervention plannedinvolving intermaxillary or other intraoral or extraoralappliances including elastics, maxillary expansion ap-pliances, or headgear. The patients were selected andtreated between March 2007 and May 2009. Their de-mographics are shown in Table I.

The active self-ligating group was bonded with theRoth prescription In-Ovation R bracket (GAC, CentralIslip, NY), and the passive self-ligating group receivedthe high-torque version of the Damon MX (Ormco,Glendora, Calif), both with a 0.022-in slot. All first andsecond (when present) molars were bonded with Speedbondable tubes (Speed System Orthodontics, Cambridge,Ontario, Canada). Bracket bonding, archwire placement,and treatment were performed by the first author.

The amount of crowding of the maxillary anteriordentition was assessed by using the irregularity indexdescribed by Little.12 Measurements were made twiceon the initial casts by the first author with a digital cal-iper (Digimatic NTD12-6’’C, Mitutoyo, Tokyo, Japan).

Archwire sequence was the same for both treatmentgroups: 0.014-in Damon arch form copper-nickel-tita-nium 35�C (Ormco), followed by a 0.016 3 0.025-in Da-mon arch form copper-nickel-titanium 35�C (Ormco).

Seventy patients (mean age, 13.8 years) were ran-domized to either an active or a passive self-ligating

appliance. Randomization was accomplished by gener-ating random permuted blocks of variable size; thisensured equal patient distribution between the 2 trialarms. Numbered, opaque, sealed envelopes were pre-pared before the trial containing the treatment allocationcard. After patient selection, the secretary of the prac-tice was responsible for opening the next envelope insequence.

Based on previous research, it was assumed thata hazard ratio larger than 2 between the bracket groupswould be an important clinical finding.6 Sample sizewas calculated. Based on this assumption, the requiredsample size was calculated at 66 (a 5 0.05, power 5

80%), and it was decided to include 70 subjects incase of any losses.

The date (T1) that each patient was bonded was re-corded. All patients were followed monthly. Complete al-leviation of crowding was judged clinically by the firstauthor. On visual inspection of correction of proximalcontacts, the patient was considered complete, and thealignment date (T2) was determined and recorded onthe spreadsheet. Only the alignment of the 6 maxillary an-terior teeth was evaluated. In other words, we consideredthat a patient had reached the T2 stage if the 6 maxillaryanterior teeth were aligned, regardless of possible irregu-larities in posterior segments. The time to alignment (T2– T1) for each patient was calculated in days. Blinding ofoutcome assessment was not feasible for this study. To as-sess the reliability of the method, the irregularity indexwas remeasured a month later in 20 models, selected ran-domly, and good agreement was found between the firstand the second measurement (ICC .0.95).

Statistical analysis

Demographic and clinical characteristics were in-vestigated with conventional descriptive statistics.

Table I. Demographic and clinical characteristics of sample

Total (n 5 70)mean or % SD

Damon MX (n 5 35)mean or % SD

In-Ovation R (n 5 35)mean or % SD P value*

Demographic characteristics

Age (y) 13.8 1.8 13.8 1.8 13.8 1.7 NS

Sex (%)

Girls 58.6 60.0 57.0 NS

Boys 41.4 40.0 43.0

Clinical characteristics

Crowding (irregularity index, mm) 7.5 2.1 8.0 2.1 7.0 2.0 NS

Angle class (%)

I 48.6 51.4 45.7 NS

II 47.1 42.9 51.4

III 4.3 5.7 2.9

NS, Not significant.

*P value for comparison of group means by t test or differences in proportions by chi-square test and Fisher exact test.

12.e2 Pandis, Polychronopoulou, and Eliades American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

Comparisons of these between the 2 appliance groupswere conducted with t tests or chi-square tests, de-pending on the characteristic (parametric or nonpara-metric). Treatment duration—the time required toresolve crowding in both appliance groups—was in-vestigated with statistical methods for survival analy-sis; alignment rate ratios for appliance type andcrowding level were calculated with the Cox propor-tional-hazards regression. A 2-tailed P value of 0.05was considered statistically significant with a 95%confidence interval.

RESULTS

Figure 1 is the CONSORT flowchart. Table I showsthe distribution of demographic variables of the groupsincluding age, sex, irregularity index, and Angle classi-fication. There was no discrimination with respect to

these factors between the 2 groups, validating the ran-dom assignment of appliances to each group. Four pa-tients were excluded from the statistical analysisbecause of poor compliance, and the statistical analysiswas conducted per protocol, since loss to follow-up wasnot associated with type of treatment.

In Table II, the results of the treatment time to align-ment are shown for the 2 bracket groups; no statisticalsignificance was found.

The results of the Cox proportional-hazards modelare given in Table III. The In-Ovation R appliance hada 1.4 hazard ratio over the Damon MX bracket; this im-plied that the former had a 1.4 times higher probabilityof alleviating crowding earlier than the latter, but thiseffect did not reach statistical significance. On the con-trary, higher irregularity index values were associatedwith increased probability of delayed resolving ofcrowding (P \0.05).

Assessed for eligibility (n= 148)

Excluded:(n= 78)

Not meeting inclusion criteria (n=67)

Other reasons (n= 11)

Analyzed (n= 33)

Per protocol

Discontinued intervention (n= 2)

Poor compliance

Allocated to Damon group (n=35)

Received allocated intervention (n=35)

Discontinued intervention(n= 2)

Poor compliance

Allocated to In-Ovation R group (n=35)

Received allocated intervention (n=35)

Analyzed (n= 33)

Per protocol

Enrollment

Randomization

Follow-Up

Allocation

Analysis

Fig 1. CONSORT flowchart of the study.

American Journal of Orthodontics and Dentofacial Orthopedics Pandis, Polychronopoulou, and Eliades 12.e3Volume 137, Number 1

Figure 2 depicts the Kaplan-Meier survival curvesfor the 2 bracket groups; lack of separation implies nostatistically significant difference.

DISCUSSION

The results of this study emphasize the clinical irrel-evance of the typical in-vitro assessment of friction pro-tocols in many studies during the past decade. A numberof factors related to the oversimplicity of experimentalconfigurations and the overwhelming number of as-sumptions in the experimental design have deprivedex-vivo friction assessment of clinical relevance andscientific soundness. These briefly include the rate ofwire sliding onto slot walls, application of forces onwire, lack of intraoral aging of materials, and study ofvariables with little or no relevance to the actual clinicalsituation.13-19 A recent critical review of this topic clar-ified several misconceptions of the study of static andkinetic friction and their clinical applicability, suggest-ing that the importance of this parameter has been over-estimated in relevant research.20

Small differences in the torque prescriptions be-tween the 2 brackets were not expected to influencethe outcome because these were outweighed by thelarge free play that was more than 2 times higher thanthe torque differences in a conventional bracket.21

Registration of the irregularity index changes in thisstudy was performed at the end of treatment because therate of correction was unknown; therefore, changes reg-istered at a certain time during therapy might not holdfor the entire treatment. The use of monthly monitoring

intervals was not a bias because this method was appliedto both bracket groups.

The results of this RCTwith the body of evidence onthis issue suggest that the bracket-archwire free playmight not be the most critical factor in altering the toothmovement rate. The large clearance in self-ligating overconventional brackets and the presumed lower bindingof Damon MX relative to the In-Ovation R appliancemight be eliminated as archwires of larger cross sectionsare gradually placed in the bracket slot.22,23 The clini-cian might empirically appreciate the free play withself-ligating brackets, especially in patients with ex-treme tooth malalignment, when complete engagementof certain diameters of nickel-titanium wire might not

Table II. Treatment time to alignment by wire system and crowding severity

Total Mean time to alignment (d) Minimum (d) Median (d) Maximum (d) P value*

Wire system

Damon MX 33 107.1 56 99 175 NS

In-Ovation R 33 95.0 54 92 161

Total 66 101.0 54 95.5 175


*P value based on the log-rank test for equality of survivor functions.

Table III. Alignment rate ratios derived from the Cox proportional-hazards regression

Predictor Adjusted hazard ratio* 95% CI P value

Bracket system

Damon MX Baseline

In-Ovation R 1.46 0.85-2.51 NS

Crowding

Irregularity index (mm) 0.73 0.64-0.84 \0.05


*Hazard ratios adjusted for demographic characteristics and Angle class.0.

000.

250.

500.

751.

00

0 50 100 150 200analysis time

bracket = Damon bracket = In-Ovation R

Kaplan-Meier survival estimates

Fig 2. Kaplan-Meier survival curves for the 2 appliancesused in the study, indicating lack of separation; thisimplies no significant effect on treatment duration.


January 2010

be feasible with conventional brackets. This situation,however, changes drastically as treatment progressesand wires of higher stiffness are engaged in the bracket.Correction of rotations and achievement of proper buc-colingual crown inclination (torque), which are fre-quently required in mandibular and maxillary anteriorteeth, respectively, necessitate a couple of forces. Thisassumes the formation of contacts of wire inside thebracket slot walls, and thus the major advantage ofself-ligating brackets—free play—is eliminated as thecrowns gradually attain their proper spatial orientation.Especially for torque application, self-ligating bracketslose more torque compared with conventionalbrackets,24,25 whereas a clinical trial showed that thesebrackets can achieve comparable torque transmissiononly with reverse curve of Spee archwires.26 Alterna-tively, torquing auxiliaries, higher torque prescriptionbrackets, or pretorqued wires can be used to counteractthe greater torque loss from greater free play.

These results agree with several studies that foundno difference in treatment duration in canine retrac-tion27-29 and crowding alleviation between conventionaland various self-ligating brackets.4-8 Regardless of thetype of movement and ligation mechanism of self-ligat-ing brackets, the results of these trials are consistentwith a lack of treatment duration differences.

For the active self-ligating bracket, the clip shouldtheoretically prevent this by applying a force on thewire; however, the relaxation of clip for the In-OvationR bracket reported previously contributes to the lackof consistent engagement of the wire throughouttreatment.30

CONCLUSIONS

The results of this RCT suggest that active and pas-sive self-ligating brackets have no difference in treat-ment duration in the correction of maxillary anteriorcrowding, in contrast to the extent of crowding, whichhad an effect on the duration of treatment.

REFERENCES

1. Harradine NW. Self-ligating brackets and treatment efficiency.

Clin Orthod Res 2001;4:220-7.

2. Harradine NW. Self-ligating brackets: where are we now? J Or-

thod 2003;30:262-73.

3. Shivapuja PK, Berger J. A comparative study of conventional

ligation and self-ligation bracket systems. Am J Orthod Dentofa-

cial Orthop 1994;106:472-80.

4. Miles GP. Smartclip versus conventional twin brackets for initial

alignment: is there a difference. Aust Orthod J 2005;21:123-7.

5. Miles GP, Weyant RJ, Rustveld L. A clinical trial of Damon 2 vs

conventional twin brackets during initial alignment. Angle Orthod

2006;76:480-5.

6. Pandis N, Polychronopoulou A, Eliades T. Self-ligating vs con-

ventional brackets in the treatment of mandibular crowding: a pro-

spective clinical trial of treatment duration and dental effects. Am


7. Scott PT, DiBiase A, Sherriff M, Cobourne M. Alignment effi-

ciency of Damon3 self-ligating and conventional orthodontic

bracket systems: a randomized clinical trial. Am J Orthod Dento-

facial Orthop 2008;134:470. e1-8.

8. Fleming P, DiBiase AT, Sarri G, Lee RT. A comparison of the ef-

ficiency of mandibular arch alignment with 2 preadjusted edge-

wise appliances. Am J Orthod Dentofacial Orthop 2009;135:

597-602.

9. Eberting JJ, Straja SR, Tuncay OC. Treatment time, outcome, and

patient satisfaction comparisons of Damon and conventional

brackets. Clin Orthod Res 2001;4:228-34.

10. Hamilton R, Goonewardene MS, Murray K. Comparison of active

self-ligating brackets and conventional pre-adjusted brackets.

Aust Orthod J 2008;24:102-9.

11. Wang D, Bakhai A. Clinical trials. A practical guide to design,

analysis, and reporting. London, United Kingdom: Remedica;

2006. p. 15-21.

12. Little RM. The irregularity index: a quantitative score of mandib-

ular anterior alignment. Am J Orthod 1975;68:554-63.

13. Sims AP, Waters NE, Birnie DJ, Pethybridge RJ. A compari-

son of the forces required to produce tooth movement in vitro

using two self-ligating brackets and a pre-adjusted bracket

employing two types of ligation. Eur J Orthod 1993;15:

377-85.

14. Thomas S, Sherriff M, Birnie D. A comparative in vitro

study of the frictional characteristics of two types of self-li-

gating brackets and two types of pre-adjusted edgewise

brackets tied with elastomeric ligatures. Eur J Orthod 1998;

20:589-96.

15. Pizzoni L, Ravnholt G, Melsen B. Frictional forces related to self-

ligating brackets. Eur J Orthod 1998;20:283-91.

16. Hain M, Dhopatkar A, Rock P. The effect of ligation method on

friction in sliding mechanics. Am J Orthod Dentofacial Orthop

2003;123:416-22.

17. Cacciafesta V, Sfondrini MF, Ricciardi A, Scribante A, Klersy C,

Auricchio F. Evaluation of friction of stainless steel and esthetic

self-ligating brackets in various bracket-archwire combinations.


18. Khambay B, Millett D, McHugh S. Evaluation of methods of

archwire ligation on frictional resistance. Eur J Orthod 2004;26:

327-32.

19. Griffiths HS, Sherriff M, Ireland AJ. Resistance to sliding with 3

types of elastomeric modules. Am J Orthod Dentofacial Orthop

2005;127:670-5.

20. Burrow SJ. Friction and resistance to sliding in orthodontics:

a critical review. Am J Orthod Dentofacial Orthop 2009;135:

442-7.

21. Sebanc J, Brantley WA, Pincsak JJ, Conover JP. Variability of ef-

fective root torque as a function of edge bevel on orthodontic arch

wires. Am J Orthod 1984;86:43-51.

22. Kusy RP, Whitley JQ. Effects of sliding velocity on the coeffi-

cients of friction in a model orthodontic system. Dent Mater

1989;5:235-40.

23. Henao S, Kusy R. Frictional evaluations of dental typodont

models using four self-ligating designs and a conventional design.

Angle Orthod 2004;75:75-85.

24. Badawi H, Toogood RW, Carey JPR, Heo G, Major PW. Torque

delivery of self-ligating brackets. Am J Orthod Dentofacial Or-

thop 2008;133:721-8.

American Journal of Orthodontics and Dentofacial Orthopedics Pandis, Polychronopoulou, and Eliades 12.e5Volume 137, Number 1

25. Morina E, Eliades T, Pandis N, Jager A, Bourauel C. Torque ex-

pression of self ligating brackets compared to conventional metal-

lic, ceramic and plastic brackets. Eur J Orthod 2008;30:233-8.

26. Pandis N, Strigou S, Eliades T. Maxillary incisor torque with con-

ventional and self-ligating brackets: a prospective clinical trial.

Orthod Craniofac Res 2006;9:193-8.

27. Miles PG. Self-ligating vs conventional twin brackets during en-

masse space closure with sliding mechanics. Am J Orthod Dento-

facial Orthop 2007;132:223-5.

28. Bokas J, Woods M. A clinical comparison between nickel tita-

nium springs and elastomeric chains. Aust Orthod J 2006;22:

39-46.

29. Deguchi T, Imai M, Sugawara Y, Ando R, Kushima K, Takano-

Yamamoto T. Clinical evaluation of a low-friction attachment

device during canine retraction. Angle Orthod 2007;77:968-72.

30. Pandis N, Bourauel C, Eliades T. Changes in the stiffness of the

ligating mechanism in retrieved active self-ligating brackets.



January 2010

ONLINE ONLY

Active or passive self-ligating brackets? Arandomized controlled trial of comparativeefficiency in resolving maxillary anteriorcrowding in adolescents

Nikolaos Pandis, Argy Polychronopoulou, and Theodore Eliades

Corfu, Athens, and Thessaloniki, Greece

Introduction: Our aim was to compare the timerequired to complete the alignment of crowded max-illary anterior teeth (canine to canine) between Da-mon MX (Ormco, Glendora, Calif) and In-OvationR (GAC, Central Islip, NY) self-ligating brackets.

Methods: Seventy patients from the first author’soffice were included in this randomized controlledtrial by using the following inclusion criteria: nonex-traction treatment on both arches, eruption of all max-illary teeth, no spaces in the maxillary arch, no highcanines, maxillary irregularity index greater than 4mm, and no therapeutic intervention planned involv-ing intermaxillary or other intraoral or extraoral appli-ances including elastics, maxillary expansionappliances, or headgear. The patients were random-ized into 2 groups: the first received a Damon MXbracket; the second was bonded with an In-OvationR appliance, both with a 0.022-in slot. The amountof crowding of the maxillary anterior dentition wasassessed by using the irregularity index. The numberof days required to completely alleviate the maxillaryanterior crowding in the 2 groups was investigatedwith statistical methods for survival analysis, andalignment rate ratios for appliance type and crowdinglevel were calculated with the Cox proportional haz-ard regression. An analysis of each protocol wasperformed.

Results: No difference in crowding alleviation wasfound between the 2 bracket systems. Higher irregular-ity index values were associated with the increasedprobability of delayed resolving of crowding.

Conclusions: The use of passive or active self-ligat-ing brackets does not seem to affect treatment durationfor alleviating initial crowding.

Read the full text online at: www.ajodo.org,pages 12.e1-12.e6.

EDITOR’S SUMMARY

In the last decade, several new appliance systemshave been developed; among the more prominent arepassive and active self-ligating brackets. For most ofthese appliances, evidence supporting their marketedcharacteristics is lacking. Thus far, most prospectiveand randomized clinical trials demonstrated no differ-ence between conventional and self-ligating bracketsfor the rate of tooth movement. The purpose of this studywas to compare the time required to complete the align-ment of crowded maxillary anterior teeth (canine to ca-nine) between passive and active self-ligating brackets.

Seventy patients were included in this randomizedcontrolled trial (RCT), with their selection based on thefollowing criteria: nonextraction treatment plan, fullyerupted dentition, no spacing in maxillary arch, no highcanines, an irregularity index greater than 4 mm, no ex-traoral appliances or elastics, and no maxillary expansionappliances. The active self-ligating group was bondedwith the Roth prescription In-Ovation R bracket, andthe passive self-ligating group received the high-torqueversion of the Damon MX bracket. The amount ofcrowding was assessed with Little’s irregularity index.

Archwire sequence was the same for both treatmentgroups: 0.014-in Damon arch form copper-nickel-tita-nium followed by 0.016 3 0.025-in Damon archformcopper-nickel-titanium. The 70 patients who met the in-clusion criteria were randomized to receive either an ac-tive or a passive self-ligating appliance. The time foralignment of the maxillary 6 anterior teeth only foreach patient was calculated in days.

The conclusions of this RCT were consistent with theresults of other studies and provide an additional piece ofwhat is becoming a complex but useful picture. Active andpassive self-ligating brackets did not show a difference intreatment duration for correcting maxillary anteriorcrowding, although the degree of crowding did. The typeof movement and the ligation mechanism of self-ligatingbrackets does not seem to affect duration of treatment.


0889-5406/$36.00


doi:10.1016/j.ajodo.2009.08.018

12


Q & ATurpin: What was the most difficult aspect of thisstudy to manage?

Eliades: The first author, who handled all patientflow and associated treatment and logistics, had todeal with precise data collection and diligent fol-low-up of the patients in treatment. This task re-quired dedication and meticulous recording oftreatment sequences and details to ensure the reli-ability of the analysis.

Turpin: As a researcher, is it more rewarding to con-duct an in-vivo RCT than a lab-bench type of studywith no patient involvement?

Eliades: Laboratory investigations are far easier tohandle compared with clinical trials, particularlyRCTs. In-vitro studies offer the option of examiningvarious effects on properties of appliances and mate-rials as well as identifying the mechanisms underly-

ing these effects. Nonetheless, the actual clinical per-formance of materials and techniques cannot be as-sessed in vitro because of several issues pertinentto the simulation of clinical conditions and the in-traoral environment. Therefore, retrieval analysesand RCTs are the state-of-the-art methods ofresearch for material alterations in vivo and clinicalefficiencies, respectively.

Turpin: Recognizing signs of enthusiasm with thistype of study design, what is your next study likelyto examine?

Eliades: We are completing an RCT on the retentionof the treatment outcome and relapse with variousappliances and techniques. This follows up previousarticles on the topic and will complete the picture ofthe multi-faceted influence of various ligation typeson efficiency, bond strength, periodontal condition,root resorption, dental changes, and force levelswith self-ligation brackets.

Table I. Demographic and clinical characteristics of sample

Total (n 5 70)mean or % SD

Damon MX (n 5 35)mean or % SD

In-Ovation R (n 5 35)mean or % SD

Pvalue*

Demographic characteristics

Age (y) 13.8 1.8 13.8 1.8 13.8 1.7 NS

Sex (%)

Girls 58.6 60.0 57.0 NS

Boys 41.4 40.0 43.0

Clinical characteristics

Crowding

(irregularity index, mm)

7.5 2.1 8.0 2.1 7.0 2.0 NS

Angle class (%)

I 48.6 51.4 45.7 NS

II 47.1 42.9 51.4

III 4.3 5.7 2.9


*P value for comparison of group means by t test or differences in proportions by chi-square test and Fisher exact test.

Table II. Treatment time to alignment by wire system and crowding severity

Total Mean time to alignment (d) Minimum (d) Median (d) Maximum (d) P value*

Wire system

Damon MX 33 107.1 56 99 175 NS

In-Ovation R 33 95.0 54 92 161

Total 66 101.0 54 95.5 175


*P value based on the log-rank test for equality of survivor functions.

American Journal of Orthodontics and Dentofacial Orthopedics Pandis, Polychronopoulou, and Eliades 13Volume 137, Number 1

ONLINE ONLY

Tooth-wear patterns in subjects with Class IIDivision 1 malocclusion and normal occlusion

Guilherme Janson,a Paula Vanessa Pedron Oltramari-Navarro,b Renata Biella Salles de Oliveira,c

Camila Leite Quaglio,c S�ılvia Helena de Carvalho Sales-Peres,d and Bryan Tompsone

Bauru, Brazil, and Toronto, Ontario, Canada

Introduction: The aim of this study was to investigate the prevalence of tooth wear in adolescents with Class IImalocclusion, compared with those with normal occlusion. Methods: The sample consisted of dental castsobtained from 310 subjects, divided into 3 groups: group 1, 110 subjects with normal occlusion (mean age,13.51 years); group 2, 100 complete Class II Division 1 patients (mean age, 13.44 years); and group 3, 100half-cusp Class II Division 1 patients (mean age, 13.17 years). Dental wear was assessed by using a modifiedversion of the tooth-wear index. The 3 groups were compared by means of the Kruskal-Wallis and Dunn tests,considering the frequency and the severity of wear on each surface of each group of teeth. The level of sta-tistical significance was set at 5%. Results: The normal occlusion group had statistically greater toothwear on the palatal surfaces of the maxillary central incisors and the incisal surfaces of the maxillary caninesthan the corresponding surfaces in both Class II malocclusion groups. The complete and half-cusp Class IIDivision 1 malocclusion groups had statistically greater tooth wear on the occlusal surfaces of the maxillarysecond premolar and first molar, the occlusal surfaces of the mandibular premolars, and the buccal surfacesof the mandibular posterior teeth compared with the normal occlusion group. The half-cusp Class II Division 1malocclusion group had significantly greater tooth wear on the incisal surfaces of the mandibular incisorscompared with the complete Class II Division 1 malocclusion group. Conclusions: Subjects with normal oc-clusion and complete or half-cusp Class II Division 1 malocclusions have different tooth-wear patterns. Toothwear on the malocclusion subjects should not be considered pathologic but rather consequent to the differentinterocclusal tooth arrangement. (Am J Orthod Dentofacial Orthop 2010;137:14.e1-14.e7)

Because of decreasing dental caries in many soci-eties, more attention has been focused on toothwear from erosion, abrasion, and attrition.1,2

Tooth wear is the loss of mineralized tooth substancefrom the tooth surface as a result of physical or chemicalattack.3

Masticatory movements by cooperative interactionsamong various stomatognathic organs, their propriocep-

tors, and higher brain centers are closely related to thefunctional occlusal system. A change in any informa-tion related to occlusion, the temporomandibular joint,or the masticatory muscles affects the patterns of thechewing movements. Some researchers have arguedthat occlusion might influence the masticatory path4;this is supported by 1 investigation.5 Gradual attritionof the occlusal surfaces of the teeth appears to be a gen-eral physiologic phenomenon in all mammals, in everycivilization, and at all ages. There are relatively fewstudies of tooth wear in the literature. This lack of de-tailed research is partly because of problems involvedin measuring techniques.4

Measurement of tooth wear is difficult because somewear is normal throughout life, and no single index hasbeen universally accepted. Studies that aimed to analyzedental wear initially focused on clinical evaluation ofthe lesions, through estimation of their severity6 andthen began to report the distribution of the lesions.7-9

Smith and Knight9 introduced the tooth-wear index(TWI), which attempted to provide a solution to someproblems associated with measuring wear at the individ-ual and community levels. The TWI and modified ver-sions of it have been used in many studies; this suggestswidespread acceptance.7,10-22 However, it was described

aProfessor and head, Department of Orthodontics, Bauru Dental School, Univer-

sity of Sao Paulo, Bauru, Brazil.bAssociate professor, Department of Orthodontics, University of North Parana,

Londrina, Parana, Brazil; former graduate student, Department of Orthodontics,

Bauru Dental School, Unversity of Sao Paulo, Bauru, Brazil.cGraduate student, Department of Orthodontics, Bauru Dental School, Univer-

sity of Sao Paulo, Bauru, Brazil.dAssociate professor, Department of Community Dentistry, Bauru Dental

School, University of Sao Paulo, Bauru, Brazil.eAssociate professor and head, Discipline of Orthodontics, Faculty of Dentistry,

University of Toronto; Department of Orthodontics, Hospital for Sick Children,

Toronto, Ontario, Canada.



Reprint requests to: Guilherme Janson, Department of Orthodontics, Bauru

Dental School, University of Sao Paulo, Alameda Octavio Pinheiro Brisolla

9-75, Bauru, Sao Paulo, 17012-901, Brazil; e-mail, [email protected].

Submitted, April 2009; revised and accepted, August 2009.

0889-5406/$36.00


doi:10.1016/j.ajodo.2009.08.022

14.e1


as flawed when used in an aging population, because itdoes not take into account teeth that were restored dueto wear.23

Some studies indicate that masticatory forces andmalocclusion are the primary etiologic factors for non-carious lesion development24-29; others did not find thiscorrelation.30-33 Because of the high prevalence of mal-occlusions in children and the controversies in the liter-ature, it is relevant to verify the pattern of tooth wear indifferent occlusal relationships to help professionals todifferentiate between physiologic and pathologic pro-cesses. The absence of previous studies of tooth wearon specific malocclusions encouraged us to comparethe patterns of tooth wear in subjects with completeand half-cusp Class II Division 1 malocclusions, andcompare them with subjects with normal occlusion.


The study protocol was approved by the EthicsCommittee on Human Research of Bauru DentalSchool, University of Sao Paulo, Brazil.

The sample consisted of dental casts obtained from310 untreated subjects from the files of the Departmentof Orthodontics at Bauru Dental School, University ofSao Paulo, Bauru, Brazil, and from the BurlingtonGrowth Centre, Faculty of Dentistry, University ofToronto, Toronto, Canada, divided into 3 groups:group 1 included 110 subjects with normal occlu-sion34-36 (54 girls, 56 boys; mean age, 13.51 years;minimum, 10.31 years; maximum, 17.53 years); group2 included 100 complete Class II Division 1 patients

(51 girls, 49 boys; mean age, 13.44 years; minimum,11.08 years; maximum, 17.26 years); and group 3 in-cluded 100 half-cusp Class II Division 1 patients(56 girls, 44 boys; mean age, 13.17 years; minimum,10.54 years; maximum, 16.88 years). The dental castswere obtained of subjects with permanent maxillaryand mandibular teeth up to the first molars. Additionalinclusion criteria included no parafunctional habits,and no temporomandibular joint and airway problems,ascertained from the subjects’ charts. Patients withopen bite were not selected.

We used a modification of the TWI,9 described bySales-Peres et al.22 The modifications matched theWorld Health Organization standard, thus allowing ap-plication of the index in broad epidemiologic surveysfor both for deciduous and permanent dentitions.37

The modifications made calibration easier and resultedin greater reproducibility, because the modified TWIdoes not differentiate the depth of dentin involvement,as does the original TWI. In addition, the modified ver-sion includes a code for teeth that have been restoreddue to wear (code 4) and another code for teeth that can-not be assessed (code 9). The form sheet used to recordthe evaluations is shown in the Figure. The amount ofpermanent tooth wear was scored in numbers (Table I).A previously calibrated examiner (R.B.S.O.) performedthe dental-cast evaluations.

A benchmark dental examiner (gold standard)(S.H.C.S.), skilled in epidemiologic surveys, trainedand calibrated the examiner. The calibration processlasted 28 hours. Theoretical activities with discussionson diagnostic criteria of dental wear were performed.

Fig. Modified TWI (reprinted from Public Health 122, S.H. de Carvalho Sales-Peres et al, Prevalenceof dental wear among 12-year-old Brazilian adolescents using a modification of the tooth wear index,942-8, Copyright 2008, with permission from Elsevier.22

14.e2 Janson et al American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

Statistical analyses

To assess the reproducibility of the dental-cast anal-ysis, 30 days after the first evaluation, 10 casts werereevaluated and demonstrated an intraexaminer kappaof 0.80 (Table II).38

Tooth surfaces were excluded from the statisticalanalysis if they were missing, or had extensive caries,large restorations, or fractures (code 9). The amountof tooth wear in the groups was compared with the Krus-kal-Wallis test followed by the Dunn post-hoc test. The3 groups were compared by considering the frequencyand severity of wear on each surface of each group ofteeth (incisors, canines, premolars, and molars). Thelevel of statistical significance was set at 5%.

RESULTS

In total, 22,320 dental surfaces were evaluated;73.9% did not have dental wear (code 0), 20.5% had in-cipient lesions (code 1), 0.3% had moderate lesions(code 2), and 5.3% were excluded (code 9). No severelesions were found.

The normal occlusion group had statistically greatertooth wear on the palatal surfaces of the maxillary cen-tral incisors (Table III) and the incisal surfaces of themaxillary canines (Table IV) than the correspondingsurfaces in both Class II malocclusion groups.

The complete and half-cusp Class II Division 1 mal-occlusion groups showed statistically greater wear onthe occlusal surfaces of the maxillary second premolarand first molar, the occlusal surfaces of the mandibularpremolars, and the buccal surfaces of the mandibularposterior teeth compared with the normal occlusiongroup (Table V). In addition, there was a tendency ofgreater tooth wear on the palatal surfaces of the maxil-lary first molars (Table V).

The half-cusp Class II Division 1 malocclusion grouphad significantly greater wear on the incisal surfaces ofthe mandibular incisors compared with the completeClass II Division 1 malocclusion Group (Table III). Therewas also a tendency of more pronounced wear on the oc-clusal surfaces of the posterior teeth in this group whencompared with the normal occlusion patients (Table V).

DISCUSSION

We used a modified version of the universally usedTWI, with high intraexaminer agreement (kappa,0.80).8,9 The modified TWI does not differentiate thedepth of dentin involvement as does the originalTWI. Thus, the modified TWI has greater intra- andinterexaminer agreement, even when working in fieldconditions.22

Some reports in the literature have suggested an as-sociation between greater tooth wear and malocclu-sion,25,26,28 although others did not corroborate thispremise.30,39,40 Our results showed that the normal oc-clusion patients and those with complete or half-cuspClass II malocclusion had some tooth wear. However,the groups showed different wear patterns (Tables III-V).

In the normal occlusion group, tooth wear wasgreater on the palatal surfaces of the maxillary centralincisors and the incisal surfaces of the maxillary ca-nines, compared with the 2 malocclusion groups (TablesIII and IV). Greater tooth wear in the anterior region inthe normal occlusion group (Table III) probably

Table I. Criteria used for tooth-wear evaluation, according to the modified TWI

Degree

Criterion DescriptionDeciduousteeth

Permanentteeth

A 0 Normal, no evidence of wear No loss of surface features

B 1 Incipient, tooth wear into enamel Loss of enamel giving a smooth, glazed, shiny appearance; dentin

is not involved

C 2 Moderate, tooth wear into dentin Extensive loss of enamel with dentin involvement; exposure of dentin

D 3 Severe, tooth wear into pulp Extensive loss of enamel and dentin with secondary dentin or pulp exposure

E 4 Restored, tooth wear leading to

restoration

Tooth received restorative treatment because of wear

— 9 Could not be assessed Extensive caries, large restoration, fractured tooth, or missing tooth

Reprinted from Public Health 122, S.H. de Carvalho Sales-Peres et al, Prevalence of dental wear among 12-year-old Brazilian adolescents using

a modification of the tooth wear index, 942-8, Copyright 2008, with permission from Elsevier.22

Table II. Intraexaminer analysis (kappa statistics)

Toothwear

Percentage ofagreement (%)

Coefficientvalue

Strength ofagreement

Dental casts 92.85 0.80 Almost perfect

American Journal of Orthodontics and Dentofacial Orthopedics Janson et al 14.e3Volume 137, Number 1

Table III. Intergroup anterior tooth-wear comparisons (Kruskal-Wallis followed by Dunn tests): means, standard devi-ations, and P values

Tooth†

Normal occlusionComplete Class II

Division 1Half-cusp Class II

Division 1

P valueMean SD Mean SD Mean SD

Inicisal surfaces

Maxillary teeth

12 0.35 a 0.50 0.28 a 0.45 0.33 a 0.47 0.5472

11 0.54 a 0.50 0.49 a 0.52 0.59 a 0.53 0.4130

21 0.50 a 0.52 0.48 a 0.54 0.68 b 0.51 0.0108*

22 0.35 a 0.48 0.36 a 0.50 0.44 a 0.50 0.2770

Mandibular teeth

42 0.83 a 0.44 0.68 b 0.47 0.83 a 0.40 0.0333*

41 0.91 a 0.42 0.65 b 0.50 0.86 a 0.38 0.0003*

31 0.85 a 0.43 0.63 b 0.49 0.90 a 0.33 0.0000*

32 0.69 a 0.48 0.70 a 0.48 0.89 b 0.35 0.0018*

Palatal surfaces

Maxillary teeth

12 0.26 a 0.44 0.24 a 0.43 0.23 a 0.42 0.8595

11 0.37 a 0.49 0.19 b 0.39 0.24 ab 0.43 0.0375*

21 0.46 a 0.50 0.19 b 0.42 0.25 b 0.44 0.0000*

22 0.31 a 0.46 0.15 a 0.36 0.25 a 0.43 0.0501*

Labial surfaces

Mandibular teeth

42 0.05 a 0.21 0.02 a 0.14 0.01 a 0.10 0.2399

41 0.02 a 0.14 0.01 a 0.10 0.04 a 0.20 0.3454

31 0.04 a 0.19 0.01 a 0.10 0.01 a 0.10 0.2598

32 0.03 a 0.13 0.04 a 0.20 0.04 a 0.20 0.5657

Different letters represent statistically significant differences.

*Statistically significant at P \0.05; †FDI 2-digit system.

Table IV. Intergroup canine tooth-wear comparisons (Kruskal-Wallis followed by Dunn tests): means, standard devi-ations, and P values

Tooth†



Division 1


Inicisal surfaces

Maxillary teeth

13 0.58 a 0.53 0.24 b 0.43 0.39 b 0.49 0.0000*

23 0.67 a 0.67 0.20 b 0.40 0.37 b 0.49 0.0000*

Mandibular teeth

43 0.72 a 0.58 0.68 a 0.47 0.60 a 0.55 0.2455

33 0.66 a 0.56 0.70 a 0.48 0.68 a 0.53 0.8128

Palatal surfaces

Maxillary teeth

13 0.10 a 0.30 0.08 a 0.27 0.14 a 0.34 0.4906

23 0.14 a 0.35 0.12 a 0.32 0.19 b 0.40 0.3950

Labial surfaces

Mandibular teeth

43 0.08 a 0.28 0.13 a 0.34 0.13 a 0.34 0.4550

33 0.03 a 0.16 0.14 b 0.35 0.07 ab 0.26 0.0091*




January 2010

occurred because of normal vertical and horizontal an-terior tooth relationships, establishing immediate ante-rior and lateral guidance during protrusion and lateralmandibular excursions, respectively.41-43 Since theseteeth disclude the posterior teeth during mandibularfunctional movements, it seems logical that they wouldhave greater wear. However, this should be consideredphysiologic wear because of the amount observed. Tobe considered pathologic, the amount of wear shouldhave been much greater.44-46 Following the same ratio-nale, since Class II Division 1 malocclusion patientshave greater overjets, the amount of wear on the incisorswould also be expected to be less than in the normal oc-clusion group (Table III). During lateral excursions, be-cause of unfavorable anteroposterior positioning of thecanines, these teeth also do not frequently disclude theposterior teeth as in normal occlusion, due to interfer-

ences of the posterior teeth.41-43 Therefore, this is thereason for less wear on the incisal surfaces of the max-illary canines in the malocclusion groups (Table IV).

Complementing that explanation, the completeClass II Division 1 malocclusion group showed greatertooth wear on the posterior teeth; this can be explainedby 2 factors. One is the large overjet that increases thelikelihood of interferences of the posterior teeth duringprotrusion until the incisors make contact as the mandi-ble is advanced41-43 (Table V). The other is that, becausethe canines are not in a favorable position to disclude theposterior teeth, these take the role of the canines duringlateral mandibular excursions and are therefore sub-jected to greater wear.

On the other hand, half-cusp Class II Division 1 pa-tients had greater tooth wear on the incisal surfaces ofthe mandibular incisors than did the complete Class II

Table V. Intergroup posterior tooth-wear comparisons (Kruskal-Wallis followed by Dunn tests): means, standarddeviations, and P values

Tooth†



Division 1


Occlusal surfaces

Maxillary teeth

16 0.74 a 0.44 0.90 b 0.35 0.91 b 0.29 0.0001*

15 0.13 a 0.33 0.36 b 0.48 0.29 b 0.46 0.0003*

14 0.40 a 0.49 0.41 a 0.50 0.47 a 0.52 0.6091

24 0.42 a 0.50 0.50 a 0.50 0.45 a 0.50 0.5226

25 0.17 a 0.37 0.40 b 0.49 0.33 b 0.47 0.0006*

26 0.79 a 0.41 0.93 b 0.28 0.88 ab 0.33 0.0044*

Mandibular teeth

46 0.82 a 0.38 0.88 a 0.34 0.93 a 0.26 0.0986

45 0.13 a 0.33 0.55 b 0.50 0.30 c 0.46 0.0000*

44 0.30 a 0.46 0.63 b 0.49 0.48 b 0.50 0.0000*

34 0.29 a 0.46 0.54 b 0.50 0.45 b 0.50 0.0007*

35 0.14 a 0.35 0.55 b 0.52 0.15 a 0.35 0.0000*

36 0.89 a 0.32 0.87 a 0.37 0.92 a 0.28 0.6115

Palatal surfaces

Maxillary teeth

16 0.00 a 0.00 0.05 b 0.22 0.01 ab 0.10 0.0213*

15 0.00 a 0.00 0.00 a 0.00 0.02 a 0.14 0.1228

14 0.00 a 0.00 0.01 a 0.09 0.00 a 0.00 0.3499

24 0.00 a 0.00 0.03 b 0.17 0.00 a 0.00 0.0426*

25 0.00 a 0.00 0.00 a 0.00 0.00 a 0.00 1.0000

26 0.00 a 0.00 0.07 b 0.26 0.06 ab 0.24 0.0214*

Buccal surfaces

Mandibular teeth

46 0.07 a 0.26 0.42 b 0.50 0.24 c 0.43 0.0000*

45 0.00 a 0.00 0.14 b 0.35 0.10 b 0.30 0.0004*

44 0.02 a 0.13 0.07 ab 0.26 0.13 b 0.34 0.0072*

34 0.01 a 0.10 0.14 b 0.35 0.12 b 0.33 0.0014*

35 0.00 a 0.00 0.12 b 0.33 0.09 b 0.29 0.0013*

36 0.06 a 0.25 0.34 b 0.48 0.24 b 0.43 0.0000*




malocclusion patients, but it was similar to the normalocclusion group (Table III). This is probably becausethe overjet in these patients is not large enough and al-lows contact between the anterior teeth during protru-sion. Nevertheless, the amounts of wear on theocclusal posterior surfaces of the maxillary and mandib-ular teeth were also greater than in the normal occlusiongroup for the same reason as in the complete Class IIDivision 1 malocclusion, as explained above (Table V).

Considering these results, tooth wear in the maloc-clusion subjects cannot be considered pathologic.44-46

No severe lesions (into pulp or secondary dentin) werefound on the dental casts, and only 5.3% had moderatedental wear (exposure of dentin). The most tooth wearwas considered to be incipient lesions (20.5%), corrob-orating a previous study.22 There is growing evidencethat the major causes of severe wear in patients are sys-temic disease45 and intense parafunction.30 It seems thatthe difference between the studied groups is consequentto the different interocclusal tooth arrangement.

In addition, it is not always possible to differentiatebetween erosion, attrition, and abrasion, because theseconditions are frequently combined.47 However, if oc-clusal factors are involved in the cause of dental wear,they are probably related to attrition—tooth wearcaused by the rubbing together of opposing occlusal sur-faces.45 Attrition has specific characteristics. First, if at-trition is the only cause of tooth wear, all wear will belocated in areas of occlusal contact. There will be nowear on the buccal and lingual surfaces of teeth unlessmandibular movement can make the opposing toothtouch in these areas. Second, attrition creates wear fac-ets with a specific appearance: shiny, flat, and sharpedged. Third, attrition produces similar amounts ofwear on the opposing teeth. In other words, it is impos-sible for tooth grinding to produce significant tooth wearon the maxillary anterior teeth but not on the mandibularanterior teeth. Finally, if attrition is the cause, the wornteeth must have occlusal contact in some mandibularposition.45 The entire sample in this study was evaluatedaccording to this reasoning.

Some controversies in the literature regarding theamounts of wear in malocclusions are probably becauseof investigations in unspecific malocclusion groups.24-33

This study was specifically performed to investigate thedifferences in tooth wear between normal occlusion andcomplete and half-cusp Class II Division 1 malocclu-sions. Thus, it included many observations in eachgroup. In addition, the groups had similar ages; this is es-sential when analyzing tooth wear, which increases withage.48 It could be speculated that patients with Class IIsubdivision malocclusions would have similar wear pat-terns to those with normal occlusion and complete Class

II malocclusion on each side, but future studies must bedeveloped to confirm this. Tooth-wear patterns in pa-tients with Class II Division 2 malocclusion wouldalso be probably different. We intend to investigate toothwear in these malocclusion types in future studies.

Further clinical studies should be undertaken withsubjects of different age ranges to confirm these resultsand overcome the limitations of dental casts.49 The as-sessment of tooth wear at the epidemiologic level mightbe desirable in the near term, because of the increase inthe prevalence and severity of dental lesions.20,47,50,51

CONCLUSIONS

Subjects with normal occlusion and complete orhalf-cusp Class II Division 1 malocclusions have differ-ent tooth-wear patterns. Tooth wear in malocclusion pa-tients should not be considered pathologic but, rather,a consequence of the different interocclusal tootharrangement.

REFERENCES

1. Marthaler T. Changes in dental caries 1953-2003. Caries Res

2004;38:173-81.

2. Vehkalahti M, Tarkkonen L, Varsio S, Heikkila P. Decrease in and

polarization of dental caries occurrence among child and youth

populations, 1976-1993. Caries Res 1997;31:161-5.

3. Walls A. Prevention in the aging dentition. In: Murray JJ, editor.

Prevention of role of disease. New York: Oxford University Press;

1999. p. 173-88.

4. Kim SK, Kim KN, Chang IT, Heo SJ. A study of the effects of

chewing patterns on occlusal wear. J Oral Rehabil 2001;28:

1048-55.

5. Adams SH, Zander HA. Functional tooth contacts in lateral and in

centeric occlusion. J Am Dent Assoc 1964;69:465-73.

6. Eccles JD. Dental erosion of non-industrial origin. A clinical

survey and classification. J Prosthet Dent 1979;42:649-53.

7. O’Brien M. Children’s dental health in the United Kingdom 1993.

London: HMSO; 1994.

8. O’Sullivan EA. A new index for the measurement of erosion in

children. Eur J Paediatr Dent 2000;2:69-74.

9. Smith BG, Knight JK. An index for measuring the wear of teeth.

Br Dent J 1984;156:435-8.

10. Jones SG, Nunn JH. The dental health of 3-year-old children in

east Cumbria 1993. Community Dent Health 1995;12:161-6.

11. Millward A, Shaw L, Smith AJ. The distribution and severity of

tooth wear and the relationship between erosion and dietary con-

stituents in a group of children. Int J Paediatr Dent 1994;4:151-7.

12. Al-Dlaigan YH, Shaw L, Smith A. Dental erosion in a group of

British 14-year-old school children. Part II: influence of dietary

intake. Br Dent J 2001;190:258-61.

13. Al-Dlaigan YH, Shaw L, Smith A. Dental erosion in a group of

British 14-year-old, school children. Part I: prevalence and influ-

ence of differing socioeconomic backgrounds. Br Dent J 2001;

190:145-9.

14. Al-Malik MI, Holt RD, Bedi R, Speight PM. Investigation of an

index to measure tooth wear in primary teeth. J Dent 2001;29:

103-7.


January 2010

15. Asher C, Read MJF. Early enamel erosion in children associated

with the excessive consumption of citric acid. Br Dent J 1987;162:

384-7.

16. Bartlett DW, Coward PY, Nikkah C, Wilson RF. The prevalence of

tooth wear in a cluster sample of adolescent schoolchildren and its

relationship with potential explanatory factors. Br Dent J 1998;

184:125-9.

17. Hinds K, Gregory J. National diet and nutrition survey: children

aged 11⁄2 to 41⁄2 years. Report of dental survey. London:

HMSO; 1995.

18. Miolosevic A, Young PJ, Lennon MA. The prevalence of tooth

wear in 14-year-old children in Liverpool. Community Dent

Health 1994;11:83-6.

19. Shaw L, Weatherill EDT, Smith AJ. Tooth wear in children: an in-

vestigation of etiological factors in children with cerebral palsy

and gastro-esophageal reflux. J Dent Child 1999;65:484-6.

20. Smith BG, Robb ND. The prevalence of tooth wear in 1007 dental

patients. J Oral Rehabil 1996;23:232-91.

21. Williams D, Croucher R, Marcenes W, O’Farrell M. The preva-

lence of dental erosion in the maxillary incisors of 14-year-old

school children living in Tower Hamlets and Hackney, London.

UK. Int Dent J 1999;49:211-6.

22. Sales-Peres SHC, Goya S, de Araujo JJ, Sales-Peres A, Lauris JR,

Buzalaf MA. Prevalence of dental wear among 12-year-old Bra-

zilian adolescents using a modification of the tooth wear index.

Public Health 2008;122:942-8.

23. Donachie MA, Walls AW. The tooth wear index: a flawed epide-

miological tool in an ageing population group. Community Dent

Oral Epidemiol 1996;24:152-8.

24. Bryant SR. The rationale for management of morphologic varia-

tions and nonphysiologic occlusion in the young dentition. Int J

Prosthodont 2003;16:Suppl:75-7.

25. Carlsson GE, Egermark I, Magnusson T. Predictors of bruxism,

other oral parafunctions, and tooth wear over a 20-year follow-

up period. J Orofac Pain 2003;17:50-7.

26. Casanova-Rosado JF, Medina-Solis CE, Vallejos-Sanchez AA,

Casanova-Rosado AJ, Maupome G, Avila-Burgos L. Dental attri-

tion and associated factors in adolescents 14 to 19 years of age:

a pilot study. Int J Prosthodont 2005;18:516-9.

27. Egermark-Eriksson I, Carlsson GE, Ingervall B. Function and

dysfunction of the masticatory system in individuals with dual

bite. Eur J Orthod 1979;1:107-17.

28. Henrikson T, Ekberg EC, Nilner M. Symptoms and signs of tem-

poromandibular disorders in girls with normal occlusion and

Class II malocclusion. Acta Odontol Scand 1997;55:229-35.

29. Ritchard A, Welsh AH, Donnelly C. The association between oc-

clusion and attrition. Aust Orthod J 1992;12:138-42.

30. Bernhardt O, Gesch D, Splieth C, Schwahn C, Mack F, Kocher T,

et al. Risk factors for high occlusal wear scores in a population-

based sample: results of the Study of Health in Pomerania

(SHIP). Int J Prosthodont 2004;17:333-9.

31. Pullinger AG, Seligman DA. Overbite and overjet characteristics

of refined diagnostic groups of temporomandibular disorder

patients. Am J Orthod Dentofacial Orthop 1991;100:401-15.

32. Rugh JD, Barghi N, Drago CJ. Experimental occlusal discrep-

ancies and nocturnal bruxism. J Prosthet Dent 1984;51:548-53.

33. Seligman DA, Pullinger AG, Solberg WK. The prevalence of den-

tal attrition and its association with factors of age, gender, occlu-

sion, and TMJ symptomatology. J Dent Res 1988;67:1323-33.

34. Andrews LF. The six keys to normal occlusion. Am J Orthod

1972;62:296-309.

35. Wheeler TT, McGorray SP, Dolce C, Taylor MG, King GJ. Effec-

tiveness of early treatment of Class II malocclusion. Am J Orthod


36. Huth J, Staley RN, Jacobs R, Bigelow H, Jakobsen J. Arch widths

in class II-2 adults compared to adults with class II-1 and normal

occlusion. Angle Orthod 2007;77:837-44.

37. World Health Organization. Oral health surveys and basic

methods, 1997. Geneva: World Health Organization; 1997.

38. Fleiss JL. Statistical methods for rates and proportions. New York:

John Wiley & Sons; 1973.

39. Metha JD. Study of attrition and malocclusion in the dentition of

Shell Mound Indians of Alabama. Am J Orthod 1969;55:306-7.

40. Warren JJ, Yonezu T, Bishara SE. Tooth wear patterns in the decid-

uous dentition. Am J Orthod Dentofacial Orthop 2002;122:614-8.

41. Okeson JP. Orofacial pain. Guidelines for assessment, diagnosis

and management. Chicago: Quintessence; 1996.

42. Roth RH. Functional occlusion for the orthodontist. J Clin Orthod

1981;15:32-40, 44-51.

43. Roth RH, Rolfs DA. Functional occlusion for the orthodontist.

Part II. J Clin Orthod 1981;15:100-23.

44. Dawes C, Boroditsky CL. Rapid and severe tooth erosion from

swimming in an improperly chlorinated pool: case report. J Can

Dent Assoc 2008;74:359-61.

45. Spear F. A patient with severewear on the anterior teeth and minimal

wear on the posterior teeth. J Am Dent Assoc 2008;139:1399-403.

46. Guldag MU, Buyukkaplan US, Ay ZY, Katirci G. A multidisci-

plinary approach to dental erosion: a case report. Eur J Dent

2008;2:110-4.

47. Smith BG, Bartlett DW, Robb ND. The prevalence, etiology and

management of tooth wear in the United Kingdom. J Prosthet

Dent 1997;78:367-72.

48. Smith BG. Toothwear: aetiology and diagnosis. Dent Update

1989;16:204-12.

49. Tarawneh FM, Panos PG, Athanasiou AE. Three-dimensional as-

sessment of dental casts’ occlusal surfaces using two impression

materials. J Oral Rehabil 2008;35:821-6.

50. Bartlett D, Phillips K, Smith B. A difference in perspective—the

North American and European interpretations of tooth wear. Int J

Prosthodont 1999;12:401-8.

51. Shaw L. The epidemiology of tooth wear. Eur J Prosthodont

Restor Dent 1997;5:153-6.


ONLINE ONLY

Tooth-wear patterns in subjects with Class IIDivision 1 malocclusion and normal occlusion

Guilherme Janson, Paula Vanessa Pedron Oltramari-Navarro, Renata Biella Salles de Oliveira,

Camila Leite Quaglio, S�ılvia Helena de Carvalho Sales-Peres, and Bryan Tompson

Bauru, Brazil, and Toronto, Ontario, Canada

Introduction: The aim of this study was to investi-gate the prevalence of tooth wear in adolescents withClass II malocclusion, compared with those with normalocclusion.

Methods: The sample consisted of dental casts ob-tained from 310 subjects, divided into 3 groups: group 1,110 subjects with normal occlusion (mean age, 13.51years); group 2, 100 complete Class II Division 1 pa-tients (mean age, 13.44 years); and group 3, 100 half-cusp Class II Division 1 patients (mean age, 13.17years). Dental wear was assessed by using a modifiedversion of the tooth-wear index. The 3 groups werecompared by means of the Kruskal-Wallis and Dunntests, considering the frequency and the severity ofwear on each surface of each group of teeth. The levelof statistical significance was set at 5%.

Results: The normal occlusion group had statisti-cally greater tooth wear on the palatal surfaces ofthe maxillary central incisors and the incisal surfacesof the maxillary canines than the corresponding sur-faces in both Class II malocclusion groups. The com-plete and half-cusp Class II Division 1 malocclusiongroups had statistically greater tooth wear on the oc-clusal surfaces of the maxillary second premolar andfirst molar, the occlusal surfaces of the mandibularpremolars, and the buccal surfaces of the mandibularposterior teeth compared with the normal occlusiongroup. The half-cusp Class II Division 1 malocclusiongroup had significantly greater tooth wear on the inci-sal surfaces of the mandibular incisors compared withthe complete Class II Division 1 malocclusion group.

Conclusions: Subjects with normal occlusion andcomplete or half-cusp Class II Division 1 malocclu-sions have different tooth-wear patterns. Tooth wearon the malocclusion subjects should not be consideredpathologic but, rather, consequent to the differentinterocclusal tooth arrangement.


EDITOR’S SUMMARY

Tooth wear from erosion, abrasion, and attrition haslong been a concern to dentists. Some wear is normalthroughout life. Some studies indicate that masticatoryforces and malocclusion are the primary etiologic fac-tors for additional wear, but others did not find this cor-relation. The aims of this study, from universities in SaoPaulo, Brazil, and Toronto, Ontario, Canada, were tocompare patterns of tooth wear in subjects with com-plete and half-cusp Class II Division 1 malocclusionsand normal occlusion.

The sample consisted of dental casts from 310untreated adolescents in 3 groups: 110 with normal oc-clusion, 100 with complete Class II Division 1 maloc-clusion, and 100 with half-cusp Class II Division 1malocclusion. Additional inclusion criteria were no par-afunctional habits, no temporomandibular or airwayproblems, and no open bites. Read this entire article on-line for details of the techniques used to measure wearfacets by a calibrated examiner.

Overall, 22,320 dental surfaces were evaluated,and 73.9% showed no dental wear. Interestingly, thenormal occlusion group had statistically greater toothwear on the palatal surfaces of the maxillary centralincisors and the incisal edges of the maxillary ca-nines. Although some reports in the literature havesuggested an association between increased toothwear and malocclusion, others did not corroboratethis premise. It was concluded that subjects with nor-mal occlusion and complete or half-cusp Class II Di-vision 1 malocclusions have different tooth-wearpatterns. Wear in the malocclusion subjects shouldnot be considered pathologic but, rather, a conse-quence of the different interocclusal arrangement ofthe teeth.

These researchers plan to pursue future studies ofolder patients; it will be interesting to see the influenceof differing interocclusal relationships.


0889-5406/$36.00


doi:10.1016/j.ajodo.2009.09.004

14


Q & ATurpin: It is fairly obvious why you started with theadolescent age group for this study, but will you havea readily available sample when evaluating older agegroups as you plan to do in the future?

Janson and Oltramari-Navarro: Yes, we do havea readily available sample of older age groups, espe-cially for the normal occlusion sample. However, forthe Class II sample, we have fewer subjects, and wemust increase it.

Turpin: Do you expect to use this same evaluationtechnique to study the effects of parafunctionalhabits on selected groups of patients?

Janson and Oltramari-Navarro: Yes, this sameevaluation technique can be used in patientswith parafunctional habits, but we will have tocollect patients with these problems. We hada significant number of patients in this studybecause it was based on subjects who were con-tinuously collected during many years. To investi-gate the tooth-wear patterns in patients withparafunctional habits, we need to prospectivelycollect them.

Turpin: Will the availability of 3-dimensional imag-ing of the dentition make it easier to measure toothwear than it is now when measuring wear facets ondental casts?

Janson and Oltramari-Navarro: Yes, as long asthere is high imaging definition and proper hardwareand software tools, such as high-definition monitorsand digitizing tablets, and measuring, shadow,brightness, and contrast controls.

Table I. Criteria used for tooth-wear evaluation, according to the modified TWI

Degree

Deciduousteeth

Permanentteeth Criterion Description

A 0 Normal, no evidence of wear No loss of surface features

B 1 Incipient, tooth wear into enamel Loss of enamel giving a smooth, glazed, shiny appearance;

dentin is not involved

C 2 Moderate, tooth wear into dentin Extensive loss of enamel with dentin involvement; exposure of dentin

D 3 Severe, tooth wear into pulp Extensive loss of enamel and dentin with secondary dentin or pulp exposure

E 4 Restored, tooth wear

leading to restoration

Tooth received restorative treatment because of wear

— 9 Could not be assessed Extensive caries, large restoration, fractured tooth, or missing tooth

Reprinted from Public Health 122, S.H. de Carvalho Sales-Peres et al, Prevalence of dental wear among 12-year-old Brazilian adolescents using

a modification of the tooth wear index, 942-8, Copyright 2008, with permission from Elsevier.22

Table II. Intraexaminer analysis (kappa statistics)

Toothwear

Percentage ofagreement (%)

Coefficientvalue

Strength ofagreement

Dental casts 92.85 0.80 Almost perfect

American Journal of Orthodontics and Dentofacial Orthopedics Janson et al 15Volume 137, Number 1

ONLINE ONLY

Accuracy of linear measurements fromcone-beam computed tomography-derivedsurface models of different voxel sizes

Janalt Damstra,a Zacharias Fourie,a James J. R. Huddleston Slater,b and Yijin Renc

Groningen, The Netherlands

Introduction: The aims of this study were to determine the linear accuracy of 3-dimensional surface modelsderived from a commercially available cone-beam computed tomography (CBCT) dental imaging system andvolumetric rendering software and to investigate the influence of voxel resolution on the linear accuracy ofCBCT surface models. Methods: Glass sphere markers were fixed on 10 dry mandibles. The mandibleswere scanned with 0.40 and 0.25 voxel size resolutions in 3 sessions. Anatomic truth was established with6 direct digital caliper measurements. The surface models were rendered by a volumetric rendering program,and the CBCT measurements were established as the mean of the 3 measurements. Results: The intraclasscorrelation coefficients between the physical measurements and the measurements of the CBCT images of0.40 and 0.25 voxels were all more than 0.99. All CBCT measurements were accurate. There was no differencebetween the accuracy of the measurements between the 0.40 and 0.25 voxel size groups. The smallest de-tectable differences of the CBCT measurements were minimal, confirming the accuracy of the CBCT mea-surement procedure. Conclusions: The measurements on 3-dimensional surface models of 0.25 and 0.40voxel size data sets made with the 3D eXam CBCT scanner (KaVo Dental GmbH, Bismarckring, Germany)and SimPlant Ortho Pro software (version 2.00, Materialise Dental, Leuven, Belgium) are accurate comparedwith direct caliper measurements. An increased voxel resolution did not result in greater accuracy of thesurface model measurments. (Am J Orthod Dentofacial Orthop 2010;137:16.e1-16.e6)

Because of the high cost and relatively high radi-ation exposure of helical computed tomography(CT) imaging methods, cone-beam CT (CBCT)

is used more frequently for craniofacial assessment inorthodontics and oral maxillofacial surgery.1,2 CBCTcaptures the craniofacial structures with a single 360�

rotation of a tube-detector unit. This is contrary to clas-sic CT, in which imaging is performed in sections orlayers. During the rotational scanning, many single pro-jections are produced and these 2-dimensional imagesare churned by the reconstruction algorithm directlyinto a 3-dimensional (3D) or volumetric data set.

Drawing an object with a computer is called render-ing.3 The object is given some characteristics to make it

appear to be a real-world object with shadows and trans-parency. To draw a 3D image, the raw CT data are trans-formed to vector data by constructing a surface of manytriangles covering the object of interest.3 Volumetricrendering programs are used to construct the 3D surfacemodels from imported CBCT data sets by implementingan algorithm that is usually unique for each program.The 3D surface model allows for actions such as indicat-ing landmarks, making measurements, moving bonefragments, and performing virtual osteotomies. The ac-curacy of the derived surface model is therefore of ut-most importance, not only for diagnostic purposes butalso for treatment planning and outcome.

The accuracy of CBCT images has been con-firmed with various CBCT scanners.4-13 However, theaccuracy of surface models derived from CBCT seemsto vary.5-8,13 Some authors illustrated differences that,even though statistically significant, were not consideredclinically relevant.5,6,13 These studies used anatomiclandmarks on the surface models, which are subject toidentification errors in the segmentation process.5,6

These factors might influence the accuracy of the mea-surement procedure. Therefore, the accuracy of the mea-surement procedures should be calculated to fullydetermine whether there is a significant difference be-tween surface models and anatomic truth. To overcome

From the University Medical Center Groningen, University of Groningen,

Groningen, The Netherlands.aPostgraduate student, Department of Orthodontics.bAssistant professor, Department of Oral Health Care and Clinical Epidemiol-

ogy, Academic Center for Oral Health.cProfessor and chair, Department of Orthodontics.



Reprint requests to: Janalt Damstra, Department of Orthodontics, UMCG, PO

Box 30.001, Groningen, 9700 RB, The Netherlands; e-mail, j.damstra@dmo.

umcg.nl.

Submitted, March 2009; revised and accepted, June 2009.

0889-5406/$36.00


doi:10.1016/j.ajodo.2009.06.016

16.e1



the problem of landmark identification, Mischkowskiet al7 used gutta-percha markers and concluded thatthe CBCT device provides satisfactory informationabout linear distances. Lagravere et al8 used titaniummarkers with a hollow cone on a synthetic mandibleand concluded that volumetric renderings from theCBCT device produce a 1-to-1 image-to-reality ratio.

A factor that could possibly influence the accuracy ofthe surface models is the voxel resolution. Volume iscomposed of voxels, which can be considered tiny cubesarranged next to each other. Each voxel is a value (bright-ness or gray-scale color) that represents the x-ray densityof the corresponding structure. Reducing the voxel reso-lution can result in a lower-quality image, more noiseand artifacts, and less detailed anatomic information.2

Spatial resolution is lower at faster scanning times andlarger voxel sizes.4 Greater spatial and voxel resolutionresults in generally ‘‘smoother’’ images by increasingthe signal-to-noise ratio, with fewer artifacts from metal-lic restorations. However, greater voxel resolution is ac-complished with an increased scanning time, exposingthe patient to a higher radiation dosage, but there isalso an increased risk of patient movement. Therefore,the influence of voxel resolution on the linear accuracyof CBCT-rendered surface models needs further investi-gation, since the result might be clinically relevant.

Our aims in this study were to determine the linearaccuracy of CBCT-derived surface models, to investi-gate the influence of voxel resolution on the linear accu-

racy of CBCT-derived surface models, and to determinethe accuracy of the measuring procedures.


The sample included 10 dry anonymous partiallydentate adult mandibles, selected from the collectionof dry skulls at the Department of Orthodontics, Univer-sity Medical Center Groningen. Mandibles with teethcontaining metallic restorations were not used becauseof possible scattering and artifact formation. Twelveareas were prepared in the cortical bone of the mandi-bles with a round surgical bur. Spherical glass markerswith a diameter of 2.4 mm (KGM Kugelfabrik GebauerGmbH, Fulda, Germany) were fixed in the preparedareas with cyanoacrylate glue (Pattex, Uni-rapideGold, Henkel, Nieuwegein, The Netherlands). Thespherical glass markers were used to minimize inherentdifferences in landmark identification and to establishfiducial anatomic locations. Twenty-five linear dis-tances, representing all 3 planes of space, were mea-sured between the landmarks (Fig, A). The midpointof the outer-most part of the sphere from the direct fron-tal view, opposite where it was glued to the mandible,was the reference mark. The distances between the ref-erence marks were determined with an electronic digitalcaliper (GAC, Bohemia, NY) on 6 occasions, at least3 days apart, by 2 observers (J.D. and Z.F.). The meanof the measurements was designated as the referencevalue, or anatomic truth.

Fig. A and B, 25 linear distances measured between the 12 markers; C and D, 3D rendered surfacemodels of a mandible with glass markers used in the study. Surface models were made with 0.4 voxel(C) and 0.25 voxel (D) sizes.

16.e2 Damstra et al American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

To provide soft-tissue equivalent attenuation, a latexballoon filled with water was placed in the lingual areaof the mandible.5,6 Before imaging in the CBCT scan-ner, the mandible was adjusted with the mandibularplane parallel to the floor and the sagittal laser referencecoinciding with pogonion. The CBCT images were ac-quired with a 3D eXam scanner (KaVo Dental GmbH,Bismarckring, Germany). There were 3 scanning ses-sions of the mandibles at least a week apart. Each man-dible was scanned twice during each session: once witha 0.40 voxel resolution and once with a 0.25 voxel res-olution. Ultimately, each voxel size group consisted of30 3D images of scanned mandibles. The preset param-eters of the scanner are summarized in Table I. TheCBCT data were exported from the eXamVisionQ soft-ware (Imaging Sciences International LCC, Hatfield,Pa) in DICOM multi-file format and imported into Sim-Plant Ortho Pro software (version 2.00, MaterialiseDental, Leuven, Belgium) on an Aspire 7730G laptopcomputer (Acer, Hertogenbosch, The Netherlands)with a dedicated 512-mb video card (Geforce 9600M-GT, NVIDIA, Santa Clara, Calif). All measurementswere performed on the surface models on a 17-in Crys-talBrite LCD flat-panel color screen (Acer) with a max-imum resolution of 1440 3 900 pixels.

The 3D surface models of all mandibular imageswere generated by the preset threshold value for bone(250-3071 Hounsfield units) as specified by the render-ing software. The SimPlant Ortho Pro software providesvarious views by rotating and translating the renderedimage. The reference points were identified on thespherical glass markers by using a cursor-driven pointer.After landmark identification, a preprogrammed analy-sis provided the distances to the nearest 0.01 mm of the25 linear measurements described in the Figure, A. Thevalues were then exported and saved in Excel file format(Microsoft, Redmond, Wash). Each CBCT image wasrendered and measured on 3 occasions by 1 observer(J.D.). The mean of the 3 measurements was calledthe CBCT measurement value.


The accuracy of these measurements was expressedby means of the absolute error (AE) and absolute

percentage error (APE). Absolute error was defined asthe CBCT measurement value subtracted from the refer-ence value.7 Absolute percentage error was calculatedwith the following equation: APE 5 100 *(AE/refer-ence value).7 Means and standard deviations werecalculated.

As a measure of reliability, the intraclass correlationcoefficient (ICC) for absolute agreement based ona 2-way random-effects analysis of variance (ANOVA)was calculated between the 3 measurement techniques(digital caliper, 0.40 voxel size, and 0.25 voxel size)used in the study.

To determine the linear accuracy of the measure-ment procedures (direct caliper and CBCT measure-ments), the standard error of measurement (SEM) ofthe 3 consecutive CBCT sessions was calculated asthe variance of the random error (interaction betweenlocations and measurement session) from the 2-wayrandom-effects ANOVA. SEM values were calculatedfor each voxel size and the physical measurements.The smallest detectable difference (SDD) was then cal-culated as 1.96 * O2 * SEM2. All statistical analyseswere performed with a standard statistical softwarepackage (version 14, SPSS, Chicago, Ill).

RESULTS

Accuracy of the measurements was determined bythe AE and APE (Table II). The AE values were small:0.01 to 0.15 mm (0.05 6 0.04 mm) for the 0.4-voxelgroup and 0.00 to 0.16 mm (0.07 6 0.05 mm) for the0.25-voxel group. The APE values were 0.25% 6

0.37% and 0.33% 6 0.47% for the 0.40- and 0.25-voxelgroups, respectively. The ICC values between the phys-ical measurements and measurements of the CBCTimages of the 2 groups were all more than 0.99.

Measurements of the CBCT images and the digitalcaliper showed excellent intraoperator reliability, withICC values of 1.00. The SDD values calculated to deter-mine the accuracy of the CBCT measurement procedurewere 0.03 mm for the 0.40-voxel group and 0.02 mm forthe 0.25-voxel group. The SSD value for the directcaliper measurements was 0.03 mm.

The mean values and standard deviations of the ref-erence values (anatomic truth) and the CBCT measure-ments for the 0.40 and 0.25 voxel sizes are summarizedin Table II. The CBCT values had a tendency to under-estimate the reference values. This occurred in 61.3% ofthe measurements for the 0.40-voxel group (0.06 6

0.05 mm) and in 60% of the measurements for the0.25-voxel group (0.08 6 0.06 mm). However, the mea-surements were overestimated for 29.3% of the mea-surements for the 0.40-voxel group (0.03 6 0.02 mm)

Table I. Preset scanning parameters for a field of view of10 cm of the 3D eXam CBCT scanner

Voxel size Projections (n) kV mAs Scanning time (s)

0.40 236 120 18.54 8.9

0.25 400 120 37.07 26.9

kV, Kilovolts; mAs, milliampere-seconds.

American Journal of Orthodontics and Dentofacial Orthopedics Damstra et al 16.e3Volume 137, Number 1

and in 33.3% of the measurements for the 0.25-voxelgroup (0.06 6 0.03 mm).

DISCUSSION

This study was performed to establish the accuracyof the CBCT-derived surface models and to investigatethe possible influence of voxel resolution on the accuracythereof. We used the 3D eXam CBCT scanner and theSimPlant Ortho Pro software to produce the surfacemodels. Our results showed that linear measurementsmade on CBCT surface renderings of 0.40- and 0.25-voxel resolutions are accurate and confirmed the accu-racy of CBCT surface models reported in previousstudies.5-8,10 Our results justify the use of CBCT-derivedsurface models for orthodontic and craniofacial treat-ment planning. There was no difference between theCBCT measurements of the 0.4-mm and 0.25-mm voxelresolution groups compared with anatomic truth. Theseresults confirm the results of Ballrick et al4 and suggestthat 0.4-mm voxel resolution is adequate for measure-

ment of craniofacial structures. The increased voxel res-olution did not cause a difference in accuracy of thesurface models. Therefore, the benefits of a shorter scan-ning time (ie, lower radiation exposure and less patientmovement) might outweigh the poorer resolution. How-ever, care must be taken when interpreting this result.The diagnostic ability of CBCT images appears to beinfluenced by voxel size. Liedkte et al14 investigated sim-ulated external root resorption of tooth roots imaged withvoxel sizes of 0.40, 0.30, and 0.20 mm. They concludedthat, even though the results from the different voxelsizes were the same, diagnosis was easier at a smallervoxel size of 0.30 or 0.20 mm. Although the benefits ofa shorter scanning time satisfy the ‘‘as low as reasonablyachievable’’ principle, the risks of misdiagnosis andtreatment complications must also be weighed. There-fore, a scanning protocol with a 0.40-mm voxel sizemight not be suitable for every patient; voxel size shoulddepend on the patient’s problems and treatment plan.

The mean differences between the CBCT and thecaliper measurements were small: 0.05 6 0.04 mm for

Table II. Means, standard deviations, absolute errors (AE), and the mean percentage measurement errors

CBCT measurements 0.4 voxel CBCT measurements 0.25 voxel

Physicalmeasurement (mm) T1 T2 T3 AE T1 T2 T3 AE

Measurement Mean SD Mean SD Mean SD Mean SD Mean SD Mean SD Mean SD Mean SD Mean SD

1 26.14 5.43 26.09 5.32 26.09 5.33 26.08 5.35 0.09 0.02 26.17 5.43 26.13 5.42 26.13 5.41 0.02 0.01

2 24.52 2.96 24.45 3.02 24.43 3.00 24.42 2.97 0.05 0.05 24.44 2.92 24.39 2.85 24.44 2.88 0.10 0.03

3 23.73 2.84 23.73 2.84 23.63 2.70 23.68 2.82 0.14 0.03 23.72 2.81 23.73 2.77 23.73 2.81 0.00 0.01

4 10.27 0.94 10.15 0.93 10.10 0.96 10.13 0.93 0.06 0.04 10.07 0.96 10.07 0.96 10.07 0.99 0.20 0.00

5 27.41 2.64 27.36 2.59 27.51 2.57 27.43 2.52 0.03 0.04 27.50 2.53 27.50 2.52 27.49 2.49 0.09 0.01

6 10.65 0.64 10.63 0.66 10.58 0.71 10.65 0.71 0.05 0.01 10.62 0.68 10.63 0.71 10.57 0.67 0.04 0.03

7 28.69 2.19 28.64 2.11 28.63 2.09 28.64 2.09 0.13 0.04 28.74 2.05 28.72 2.12 28.71 2.06 0.03 0.02

8 30.61 2.31 30.45 2.13 30.46 2.10 30.52 2.15 0.01 0.01 30.56 2.10 30.51 2.11 30.52 2.09 0.08 0.03

9 29.16 2.51 29.17 2.43 29.15 2.33 29.18 2.38 0.01 0.02 29.25 2.32 29.29 2.35 29.24 2.34 0.10 0.03

10 36.51 3.94 36.55 3.86 36.51 3.84 36.51 3.82 0.01 0.00 36.50 3.85 36.50 3.87 36.47 3.89 0.02 0.02

11 11.10 0.29 11.09 0.30 11.11 0.28 11.09 0.28 0.02 0.01 11.02 0.27 11.09 0.30 11.07 0.31 0.04 0.04

12 37.85 4.06 37.88 4.14 37.86 4.10 37.84 4.16 0.08 0.01 37.73 4.18 37.72 4.10 37.72 4.15 0.13 0.01

13 39.47 3.59 39.39 3.69 39.38 3.66 39.39 3.67 0.05 0.02 39.33 3.66 39.33 3.64 39.32 3.67 0.14 0.01

14 38.04 3.73 38.08 3.69 37.99 3.71 37.97 3.73 0.07 0.03 37.93 3.75 37.94 3.69 37.90 3.74 0.12 0.02

15 28.49 2.38 28.44 2.32 28.42 2.39 28.39 2.42 0.15 0.02 28.45 2.42 28.47 2.40 28.50 2.37 0.02 0.02

16 10.12 1.38 9.98 1.34 9.99 1.36 9.95 1.31 0.06 0.02 9.94 1.36 9.96 1.36 9.97 1.36 0.16 0.02

17 29.45 2.71 29.53 2.62 29.51 2.61 29.50 2.66 0.01 0.01 29.49 2.59 29.53 2.70 29.57 2.64 0.08 0.04

18 31.88 2.43 31.88 2.33 31.86 2.35 31.86 2.37 0.03 0.02 31.88 2.37 31.95 2.40 31.94 2.36 0.04 0.04

19 29.67 2.33 29.71 2.26 29.70 2.28 29.66 2.32 0.04 0.02 29.64 2.23 29.66 2.28 29.68 2.27 0.02 0.01

20 20.67 3.88 20.64 3.90 20.61 3.89 20.65 3.90 0.01 0.01 20.64 3.87 20.56 3.86 20.56 3.87 0.08 0.05

21 22.38 4.56 22.39 4.52 22.40 4.48 22.39 4.51 0.01 0.02 22.48 4.54 22.42 4.54 22.37 4.56 0.05 0.05

22 25.57 4.72 25.60 4.69 25.57 4.73 25.58 4.71 0.04 0.02 25.59 4.77 25.56 4.78 25.60 4.75 0.02 0.01

23 94.53 2.75 94.50 2.76 94.51 2.79 94.47 2.79 0.02 0.01 94.58 2.72 94.56 2.73 94.53 2.73 0.03 0.03

24 86.49 4.88 86.52 4.82 86.50 4.85 86.50 4.88 0.04 0.04 86.56 4.82 86.52 4.80 86.49 4.82 0.03 0.04

25 72.83 3.76 72.79 3.75 72.75 3.69 72.83 3.77 0.09 0.02 72.78 3.74 72.76 3.75 72.74 3.74 0.07 0.02

Mean absolute measurement error 0.05 0.07 0.05

Mean percentage measurement error 0.25 0.33 0.46

T, Scanning session.


January 2010

the 0.4-voxel group and 0.07 6 0.05 mm for the 0.25-voxel group. These values were similar to those previ-ously reported in the literature for differences between3D CBCT renderings and direct caliper measurements.Stratemann et al10 used chromium balls of 2.4-mm diam-eter as markers and reported small mean differences(0.00 and 0.07 mm). However, the standard deviationsof the mean differences were significantly larger (0.41and 0.22 mm) compared with our results. Mischkowskiet al7 used prepared holes filled with gutta-perchamarkers and reported a mean absolute difference of0.26 mm. In a pilot study, we found that metal and chro-mium markers caused significant artifacts when render-ing the surface models because of scattering, whereasthe prepared gutta-perchs markers was not clearly visi-ble on our surface models. These markers were not con-sidered for this study. Hassan et al13 used anatomiclandmarks that resulted slightly larger differences of0.10 to 0.39 mm between the 3D renderings and the cal-iper measurements. The mean difference between therendered 3D surface models and the caliper measure-ments was less than the relevant error of 0.5 mm postu-lated by Marmulla et al15 and less than the voxel size ofthe image, and can therefore not be regarded as clinicallyrelevant for craniofacial measurements.

The results show that the CBCT valueshada tendencyto underestimate the referencevalues; however, it was notas severe as previously reported. The CBCT values wereunderestimated for 60.7% of the total measurements inthis study, but this is significantly less than the 94.4% re-ported by Ballrick et al.4 Lascala et al12 also reportedsmaller computer-based linear measurements than directdigital caliper measurements of dry skulls. However, theCBCT measurements in these studies were made on ax-ial, coronal, and sagittal cuts of the 3D image rather thanon the 3D surface renderings; this probably accounted forthe differences of underestimation.8

In this study, the SDD was used to determine the ac-curacy of the measurement procedures. The SDD wasproposed as an adequate measure for quantitative andstatistically significant difference between measure-ments.16,17 The SDD is expressed in the same unit asthe measurement device used and is generalizable toall included facets (observers, techniques, measurementtimes, repeated measures). The SDD values for theCBCT measurements were small: 0.03 mm for the0.40-voxel group and 0.02 mm for the 0.25-voxel group.The small SDD values confirmed the accuracy of themeasurement procedure we used. The SDD values indi-cate that the measurement procedure for the surfacemodels was just as accurate as direct caliper measure-ments. For a statistically significant difference between2 observations, the difference must be at least the SDD

of the measurement procedure. If this is not the case,and the SSD is larger than the reported difference, thedifference could be result of inaccuracies in measure-ment rather than a true difference between the observa-tions. In this study, the CBCT measurement procedurehad the power to detect differences of 0.03 mm. If themeasurement procedure is less accurate (ie, influencedby the segmentation process and landmark identifica-tion error), the SDD will be larger. A large SDD couldhave a significant effect on the interpretation of the dif-ferences between 2 observations, especially when thereported differences are small.

The CBCT measurements are accurate because thelandmark identification error was reduced by using opa-que glass spheres as fiducial markers. Additionally, thespherical glass markers are likely to be less affected bythe segmentation process because of their uniform den-sity. The glass spheres we used were produced fromsoda-lime-silica-glass, the most prevalent type of glassthat is commonly used for windows and glass containers(bottles and jars). The main advantage of glass vs metal-lic markers is that glass markers produce no scatteringand artifacts when rendered to surface models (Fig, Cand D). This is because bone and glass spheres havesimilar values on the Hounsfield scale.18,19

This study showed the technical limits of the CBCTscanner and rendering software, but these might not di-rectly apply to patient care. The mandibles we used didnot move and had fiducial markers for measurement;this is not the case with patients. We placed a latex bal-loon filled with water in the lingual area of the dry man-dible to simulate soft-tissue attenuation, a method alsoused by Brown et al5 and Periago et al.6 An alternativemethod used to simulate soft-tissue attenuation is a wa-ter bath, which might be problematic during positioningin the CBCT scanner and might damage the dryskulls.8,10,13 In addition, absorption of water by thedry mandibles could influence measurement accuracybecause of expansion of the bone. Neither the balloonfilled with water nor the water-bath method equates ineither quantity or distribution to patients’ soft tissues.Although the water-filled balloon in the lingual areaprovided some soft-tissue attenuation, the lack ofperipheral attenuation material might have allowed forincreased contrast of the landmarks.

CONCLUSIONS

Linear measurements on 3D surface models of 0.25-and 0.40-voxel CBCT data sets made with the 3D eXamCBCT scanner and SimPlant Ortho software are accu-rate when compared with direct caliper measurements.Increasing the voxel resolution from 0.40 to 0.25 mm

American Journal of Orthodontics and Dentofacial Orthopedics Damstra et al 16.e5Volume 137, Number 1

to construct a 3D surface model did not result inincreased accuracy of the CBCT measurements.

We thank Dr J. van der Meer of the Department ofOrthodontics of the University Medical Center Gronin-gen for his technical input and advice during thisproject.

REFERENCES

1. Harrell WE, Jacobson RL, Hatcher DC, Mah J. Cephalometric im-

aging in 3-D. In: Jacobson A, Jacobson RL, editors. Radiographic

cephalometry: from basics to 3-D imaging. 2nd ed. Hanover Park,

Ill: Quintessence; 2006. p. 233-48.

2. Schutyser F, van Cleynenbreugel J. From 3-D volumetric com-

puter tomography to 3-D cephalometry. In: Swennen GRJ,

Schutyser F, Hausamen JE, editors. Three-dimensional cephalom-

etry: a color atlas and manual. Heidelberg, Germany: Springer-

Verlag; 2006. p. 2-11.

3. Halazonetis DJ. From 2-dimensional cephalograms to 3-dimen-

sional computed tomography scans. Am J Orthod Dentofacial

Orthop 2005;127:627-37.

4. Ballrick JW, Palomo JM, Ruch E, Amberman BD, Hans MG. Im-

age distortion and spatial resolution of a commercially available

cone-beam computed tomography machine. Am J Orthod Dento-


5. Brown AA, Scarfe WC, Scheetz JP, Silveira AM, Farman AG.

Linear accuracy of cone beam CT 3D images. Angle Orthod

2009;79:150-7.

6. Periago DR, Scarfe WC, Moshiri M, Scheetz JP, Silveira AM,

Farman AG. Linear accuracy and reliability of cone beam CT de-

rived 3-dimensional images using an orthodontic volumetric ren-

dering program. Angle Orthod 2008;78:387-95.

7. Mischkowski RA, Pulsfort R, Ritter L, Neugebauer J,

Brochhagen HG, Keeve E, et al. Geometric accuracy of a newly

developed cone-beam device for maxillofacial imaging. Oral

Surg Oral Med Oral Pathol Oral Radiol Endod 2007;104:551-9.

8. Lagravere MO, Carey J, Toogood RW, Major PW. Three-dimen-

sional accuracy of measurements made with software on cone-

beam computed tomography images. Am J Orthod Dentofacial

Orthop 2008;134:112-6.

9. Pinsky HM, Dyda S, Pinsky RW, Misch KA, Sarment DP. Accu-

racy of three-dimensional measurements using cone-beam CT.

Dentomaxillofac Radiol 2006;35:410-6.

10. Stratemann SA, Huang JC, Maki K, Miller AJ, Hatcher DC. Com-

parison of cone beam computed tomography imaging with phys-

ical measures. Dentomaxillofac Radiol 2008;37:80-93.

11. Eggers G, Klein J, Welzel T, Muhling J. Geometric accuracy of

digital volume tomography and conventional computed tomogra-

phy. Br J Oral Maxillofac Surg 2008;46:639-44.

12. Lascala CA, Panella J, Marques MM. Analysis of the accuracy of

linear measurements obtained by cone beam computed tomogra-

phy (CBCT-NewTom). Dentomaxillofac Radiol 2004;33:291-4.

13. Hassan B, van der Stelt P, Sanderink G. Accuracy of three-dimen-

sional measurements obtained from cone beam computed tomog-

raphy surface-rendered images for cephalometric analysis:

influence of patient scanning position. Eur J Orthod 2009;31:

129-34.

14. Liedke GS, Dias da Silviera HE, Dias da Silviera HL, Duntra V,

Poli de Figueiredo A. Influence of voxel size in the diagnostic

ability of cone beam tomography to evaluate simulated external

root resorption. J Endod 2009;35:233-5.

15. Marmulla R, Wortche R, Muhling J, Hassfeld S. Geometric accu-

racy of the NewTom 9000 Cone Beam CT. Dentomaxillofac Ra-

diol 2005;34:28-31.

16. Kropmans THJB, Dijkstra PU, Stegenga B, Stewart R, De

Bont LGM. Smallest detectable difference in outcome variables

related to painful restriction of the temporomandibular joint. J

Dent Res 1999;78:784-9.

17. Kropmans THJB, Dijkstra PU, Stegenga B, Stewart R, De

Bont LGM. Smallest detectable difference of maximal mouth

opening in patients with painfully restricted temporomandibular

joint function. Eur J Oral Sci 2000;108:9-13.

18. Enomoto K, Nishimura H, Inohara H, Murata J, Horii A, Doi K,

et al. A rare case of a glass foreign body in the parapharyngeal

space: pre-operative assessment by contrast-enhanced CT and

three dimensional CT images. Dentomaxillofac Radiol 2009;38:

112-5.

19. Gor DM, Kirch CF, Leen J, Turbin R, Von Hagen S. Radiologic

differentiation of intraocular glass: evaluation of imaging tech-

niques, glass types, size, and effect on intraocular hemorrhage.

AJR Am J Ooentgenol 2001;177:1199-203.


January 2010

ONLINE ONLY

Accuracy of linear measurements fromcone-beam computed tomography-derivedsurface models of different voxel sizes

Janalt Damstra, Zacharias Fourie, James J. R. Huddleston Slater, and Yijin Ren

Groningen, The Netherlands

Introduction: The aims of this study were to deter-mine the linear accuracy of 3-dimensional surfacemodels derived from a commercially available cone-beam computed tomography (CBCT) dental imagingsystem and volumetric rendering software and to inves-tigate the influence of voxel resolution on the linearaccuracy of CBCT surface models.

Methods: Glass sphere markers were fixed on 10dry mandibles. The mandibles were scanned with 0.40and 0.25 voxel size resolutions in 3 sessions. Anatomictruth was established with 6 direct digital calipermeasurements. The surface models were rendered bya volumetric rendering program, and the CBCTmeasurements were established as the mean of the 3measurements.

Results: The intraclass correlation coefficients be-tween the physical measurements and the measure-ments of the CBCT images of 0.40 and 0.25 voxelswere all more than 0.99. All CBCT measurementswere accurate. There was no difference between the ac-curacy of the measurements between the 0.40 and 0.25voxel size groups. The smallest detectable differencesof the CBCT measurements were minimal, confirmingthe accuracy of the CBCT measurement procedure.

Conclusions: The measurements on 3-dimensionalsurface models of 0.25 and 0.40 voxel size data setsmade with the 3D eXam CBCT scanner (KaVo DentalGmbH, Bismarckring, Germany) and SimPlant OrthoPro software (version 2.00, Materialise Dental, Leuven,Belgium) are accurate compared with direct calipermeasurements. An increased voxel resolution did notresult in greater accuracy of the surface modelmeasurments.


EDITOR’S SUMMARY

We are receiving more submissions that use cone-beamcomputed tomography (CBCT). Many of them aim toassess the accuracy and reliability of CBCT with mea-surements on the axial, coronal, and sagittal slices, oron 3-dimensional (3D) iso-surface renderings. The re-sults generally show that CBCT accurately representsthe anatomy of our orthodontic patients to a clinicallyacceptable degree. However, another use of CBCT isto plan surgical treatment by simulating osteotomiesand performing virtual bone movements. These applica-tions require the construction of a surface model fromthe CBCT data. The surface model usually consists ofa dense triangular mesh that approximates the bone’ssurface and is automatically constructed from the voxeldata, requiring only the input of a threshold value thatspecifies what is bone and what is not. Once con-structed, the surface model can be used independentlyof the CBCT data. Because they require significantlyless memory storage and can take advantage of today’sgraphic processors’ capability to render trianglesquickly, surface models are well suited for surgical sim-ulations on personal computers.

These researchers aimed to assess the accuracy ofsurface models constructed from CBCT images. It isexpected that a surface model would be more faithfulto the original object when imaged with a small voxelsize (high resolution), but this increases the radiationdosage. These authors tested 2 voxel sizes to determinewhether the radiation dosage can be kept low withoutsacrificing accuracy.

The authors found that the larger voxel size resultedin equally accurate models. They used dry mandiblesand glass markers, which allowed establishment of the‘‘ground truth’’ by direct measurements. However, thismethodology does not accurately reflect clinical appli-cation, because no soft tissues are in contact withbone, thus making bone surface easier to identify.


0889-5406/$36.00


doi:10.1016/j.ajodo.2009.08.023

16


Q & ATurpin: You discussed the limitations of your studyregarding the absence of soft tissues around the drymandibles. Is this a major issue that might affect ac-curacy in the clinical setting, and are you planningfuture studies to address this problem?

Damstra: As we mentioned, the lack of soft tissues re-sulted in increased contrast of the landmarks; thisinfluenced the threshold and segmentation process.In a recent commentary to a related article, Halazone-tis1 explained how the midvalue between the densitiesof 2 materials is used as the threshold for segmenta-tion. This is known as the ‘‘full width at half maxi-mum’’ method. In our study, this method wouldindicate a threshold value equal to half the differencebetween the mandibles and air. However, in the clini-cal setting, mandibular hard tissues are in contact withthe soft tissues. The threshold should be set higher atthese levels because soft tissues are much denser thanair. To overcome this problem, the threshold must beadjusted at a local level. Therefore, in the clinical set-ting, the segmentation process becomes critical andtime-consuming when producing an accurate surfacemodel. We are currently investigating the influence ofsoft tissues and segmentation on the accuracy of sur-face models. Further research will also focus on im-proving the consistency of tissue densities of CBCTscanners to improve accuracy during segmentation.

Turpin: The largest voxel size you tested was 0.4mm. Could we accept even larger voxel sizes and fur-ther reduce the radiation dosage to patients?

Damstra: We tested the maximum preset voxel sizeof our CBCT scanner. Although a larger voxel sizemight reduce the radiation, it will also reduce thequality of the images and might influence the seg-mentation process. The risk of misdiagnosis of pos-sible pathology must be weighed. Therefore, thelargest voxel size we use now is 0.4 mm.

Turpin: Do you have any data on the related perfor-mance of other CBCT machines?

Damstra: A recent systematic review reported theradiation dose values and settings of various CBCTscanners used in imaging the oral and maxillofacialregion.2 Future research will no doubt examine thedifference in accuracy between various CBCT scan-ners and rendering software.

REFERENCES

1. Halazonetis DJ. Commentary. Am J Orthod Dentofacial

Orthop 2009;136:25-8.

2. De Vos W, Casselman J, Swennen GRJ. Cone-beam computer-

ized tomography (CBCT) imaging of the oral and maxillofacial

region: a systematic review of the literature. Int J Oral Maxillo-

fac Surg 2009;38:609-25.

Fig. A and B, 25 linear distances measured between the 12 markers; C and D, 3D rendered surfacemodels of a mandible with glass markers used in the study. Surface models were made with 0.40voxel (C) and 0.25 voxel (D) sizes.

American Journal of Orthodontics and Dentofacial Orthopedics Damstra et al 17Volume 137, Number 1

ORIGINAL ARTICLE

Effectiveness of interceptive orthodontictreatment in reducing malocclusions

Gregory J. Kinga and Pongsri Brudvikb

Seattle, Wash, and Bergen, Norway

Introduction: In this retrospective cohort study of the effectiveness of interceptive orthodontic treatment, wecompared patients receiving interceptive orthodontic treatment with untreated control subjects. Methods:Models were scored by using the index of complexity, outcome and need (ICON). Control models (n 5 113)were archival and were selected based on malocclusion in the early mixed dentition and no orthodontic treat-ment during the subsequent 2 years. The patients (n 5 133) were in the mixed dentition and consecutivelytreated in the University of Bergen orthodontic clinic. Initial ages were 9.4 years (6 1.4) for the treated groupand 9.3 years (6 0.8) for the control group. The treatment took a mean of 27.2 months (6 16.3) for the patients;the control group was observed for a mean of 24.4 months (6 3.6). Subject Groups were matched for age,need, complexity, duration, and all ICON components except spacing (P \0.006) and crossbite (P \0.000).Results: ICON scores decreased after treatment by 38.8% (P \0.0001) from 54.9 (6 16.6) to 33.6 (6 16.1).The controls were unchanged, with ICON scores of 54.0 (6 14.8) and 54.2 (6 16.9). Improvement gradeswere different (P \0.0001), with most controls categorized as ‘‘not improved or worse’’ (89.4%), whereasonly 36.1% of the treated group were in that category. However, there were increases in the ‘‘minimal,’’ ‘‘mod-erate,’’ and ‘‘substantial’’ improvement categories for the treated subjects (22.6%, 21.1%, and 17.3%, re-spectively). The controls did not change in any ICON component and worsened in crowding (P \0.007),whereas the patients improved in esthetics, crowding, crossbite, and overbite (P \0.007). Conclusions:These results indicate that interceptive orthodontic treatment is effective for improving malocclusion butdoes not produce finished-quality results. (Am J Orthod Dentofacial Orthop 2010;137:18-25)

Interceptive and preventive orthodontic proceduresare relatively simple and inexpensive treatment ap-proaches that target developing malocclusions dur-

ing the mixed dentition. Orthodontists perceive these asuseful ways to reduce the severity of malocclusions,1

improve a patient’s self-image, eliminate destructivehabits, facilitate normal tooth eruption, and improvesome growth patterns.2 Because of this, some have ad-vocated their wider use as public health measures aimedat reducing the burden of malocclusion in underservedpopulations3 and as a strategy for increasing access toorthodontic treatment when resources are limited.1

Available evidence suggests that patients at risk forsevere malocclusion can readily be identified in the

mixed dentition, and that the burden of these malocclu-sions in this age group is substantial (about 25%-30%).In 1 study, patients at risk for future orthodontic prob-lems were identified in 28% of those examined, andmost of the developing malocclusions were judged tobe suitable for interceptive orthodontic treatment.4 Asimilar study found that about 27% of the children ex-amined in a large Nigerian sample needed some formof interceptive orthodontic treatment.3 A third studyof children screened in a community dental clinic atages 9 and 11 years also found that one-third wouldbenefit from interceptive orthodontic treatment.5

Although interceptive orthodontic procedures oftendo not produce finished orthodontic results withouta second phase of treatment in the permanent dentition,several studies have suggested that systematicallyplanned interceptive treatment in the mixed dentitionmight contribute to a significant reduction in treatmentneed between the ages of 8 and 12 years, often produc-ing results so that further need can be categorized aselective. In a Finnish study, the need was reduced signif-icantly from ages 8 to 12 in a small group receiving in-terceptive treatment.6 In a similar study, 94% of thechildren receiving interceptive treatment in a commu-nity health clinic were judged to have completely suc-cessful results, with only 2% showing deterioration.5

aMoore Reidel Professor, Department of Orthodontics, University of

Washington, Seattle.bAssociate professor, Department of Orthodontics, University of Bergen,

Bergen, Norway.

Supported by NIDCR grant U54 DE14254.



Reprint requests to: Gregory J. King, University of Washington, Department of

Orthodontics, Box 357446, Seattle, WA 98195; e-mail, [email protected].

edu.

Submitted, December 2007; revised and accepted, February 2008.

0889-5406/$36.00


doi:10.1016/j.ajodo.2008.02.023

18



Using the peer assessment rating (PAR)7 and the indexof complexity, outcome and need (ICON),8 the authorsof another study reported significant reductions in mal-occlusion severity after early treatment in both Medic-aid and privately financed patients, with comparableresults in both groups.9 In addition, about two thirdsof those patients changed from a ‘‘medically necessary’’category, as judged by the handicapping labiolingualdeviation index10 to ‘‘elective’’ after mixed dentitionorthodontic intervention.11

Although the available data suggest that interceptiveorthodontic treatment can be effective, no randomizedclinical trials or large cohort studies have compared in-terceptive outcomes with no treatment in either the nearor long term. This has been primarily due to the lack ofsuitable cohorts of untreated patients with malocclu-sions to serve as control subjects. During the 1970s,the orthodontic faculty at the University of Bergen inNorway collected orthodontic study casts biannuallyfrom many local school children. A substantial numberof these children had orthodontic needs and were nottreated during the mixed dentition. Thus, it was possibleto collect an archival cohort of study casts of untreatedsubjects with excellent matching for malocclusionseverity to compare with a contemporary sample of pa-tients treated in an interceptive clinic. We hypothesizedthat interceptive orthodontic treatment would improvemalocclusions with reductions in their complexity andneed.


This was a retrospective cohort design. Power calcu-lations were based on a similar study with a 40% im-provement in ICON scores after interceptivetreatment.9 A sample size of 100 provided a power of90% and an alpha of 0.05 for this level of differencein ICON scores. For the treated group, 133 patientswith pre- and postinterceptive sets of dental casts wereselected from consecutively treated patients who metour inclusion criteria and were treated in the orthodonticclinic at the University of Bergen by dental students su-pervised by orthodontic faculty members. Most of thesepatients were treated for dental misalignment, crowdingor spacing, inversions, anterior open bite, and crossbitewith various removable appliances. The control group,consisting of 113 patients with 2 sets of dental caststhat also met the inclusion criteria, was randomly se-lected from departmental archives of biannual recordsof school children taken during the 1970s. Inclusionand exclusion criteria were initial casts in the earlymixed dentition, final casts after interceptive orthodon-tic treatment but no later than the late mixed dentition

(ie, final casts with a fully erupted permanent dentitionmesial to the first molars were excluded), final casts forthe control group who received no orthodontic treat-ment were taken 2 years after the first, initial casts show-ing a malocclusion suitable for funding under theNorwegian Social Security System as judged subjec-tively by an orthodontist (P.B.) experienced with thesecriteria, Scandinavian ethnicity, and no exclusion basedon sex.

All casts were scored with the ICON by a calibratedexaminer (G.J.K.) who was not blinded to group or timepoint. True blinding was not possible because the castsof the control group were easily recognizable because oftheir rough trimming, and the time points were obviousbased on the stage of tooth development evident on thecasts. The ICON scores overall occlusion and an es-thetic component of malocclusion on an interval scale,from 0 to 120 for the former and 0 to 10 for the latter.The higher the ICON score, the worse the malocclusion.The ICON has been validated based on the subjectivejudgments of 97 orthodontists from 9 countries on 240initial and 98 treated dental casts. Created as a singlemeasure of need, complexity, and outcome simulta-neously, the ICON has 2 advantages over the more com-monly used PAR as a dental outcome measure.7,12 It hasan esthetic component that is weighted highly by clini-cians and valued by patients, and it has clear and inter-nationally validated cut points for treatment need andoutcome with categories for complexity and improve-ment. Five weighted parameters are scored and com-prise the components of the ICON: dental esthetics,crossbite, anterior vertical relationship, maxillarycrowding or spacing, and buccal segment anteroposte-rior relationship. The components were individuallyscored from dental casts and multiplied by their respec-tive weights to yield a single summary ICON score. Thisfinal score was then used to determine initial need(ICON .43) and final outcome acceptability (ICON\31), and difference scores were used to determine im-provement. According to the convention recommendedby the developers of the ICON, improvement scoreswere calculated by subtracting 4 3 final scores fromthe initial scores.8 This permitted us to compare the im-provement in our samples with the categories validatedfor the ICON. Intrarater reliability of the examiner wasdetermined by using Dahlberg’s formula13 on 10 sets ofmodels remeasured 2 weeks apart and was considered tobe acceptable (4.1 ICON points).

Initial equivalence of the groups was assessed by us-ing age, sex, and malocclusion characteristics. The lat-ter consisted of weighted initial ICON scores andunweighted ICON component scores. Equivalence inICON components was assessed by using multiple

American Journal of Orthodontics and Dentofacial Orthopedics King and Brudvik 19Volume 137, Number 1

unpaired t tests with Bonferroni adjustments for multi-ple comparisons. Since 7 tests were performed, the levelof significance was set at P \0.007 (ie, P \0.05/7).ICON scores were compared between groups and atthe 2 time points with 2-way analysis of variance (AN-OVA) and post-hoc comparisons with the Kruskal-Wallis test if P \0.05. Initial need was determined byusing the weighted ICON score threshold of .43, andend-of-study acceptability was determined with the\31 threshold. The prevalences of subjects in the initialcomplexity grades and improvement categories werealso calculated, and these distributions were comparedbetween groups by using the chi-square statistic. Un-weighted ICON component scores were compared be-tween initial and final casts, and differences wereassessed with multiple t tests with Bonferroni adjust-ments and significance set at P \0.007.

RESULTS

Initially, the subjects in the interceptive and controlgroups had mean chronological ages of 9.4 years (61.4) and 9.3 years (6 0.8), respectively (Table I). Thesewere not different. However, although the treated groupwas about equally divided by sex (51.6% female), thecontrol group was predominantly male (31.8% female).These were statistically different (P 5 0.017). As as-sessed by the unweighted ICON component scores,the malocclusions in the 2 groups were largely wellmatched, with moderate esthetics scores, minimal max-illary crowding or spacing and openbite or overbite, andmoderate buccal anteroposterior occlusal relationships.Statistical differences were found only in maxillary an-terior spacing (P \0.006) and crossbite (P \0.000).

Table II shows that both groups had similar ICONscores when the first models were taken (54.9 and54.0, respectively). Based on the validated ICON as-sessment of need (.43), both groups of subjects neededtreatment. The intervals between the first and second

sets of casts averaged 27.2 months (6 16.3) in thetreated group and 24.4 months (6 3.6) in the control;these were not different. At the second time point, themean ICON scores were 33.6 (6 16.1) for the treatedgroup and 54.2 (6 16.9) for the untreated controls.This represented a 38.8% improvement for the former(P \0.0001) and no change for the latter. Based onthe ICON acceptability cut point of \31, both groupswould be judged unacceptable, but the treated cohortwas borderline.

The distributions of the initial complexity grades(Fig 1) were not different between the groups (chi-square: P 5 0.186). Figure 2 gives a comparison ofthe distributions of improvement grades between the 2groups. The difference between these was highly signif-icant (chi-square, P \0.0001). Whereas most subjectswere categorized as ‘‘not improved or worse’’ in the un-treated control group (89.4%), the treated group had36.1% in that category. This reduction was reflectedby roughly equivalent increases in the ‘‘minimal,’’‘‘moderate,’’ and ‘‘substantial’’ improvement categoriesfor the treated subjects (22.6%, 21.1%, and 17.3%, re-spectively) compared with the controls (7.9%, 1.8%,and 0.9%, respectively).

Improvements in the various ICON categories areshown in Figures 3 and 4. The untreated controls hadno improvement in any component and a statisticallysignificant worsening of maxillary crowding (P 5

0.007). In contrast, the treated sample showed statisti-cally highly significant improvements in the estheticand crossbite components (P\0.0001), with significantimprovements in the maxillary-crowding and open-bitecomponents (0.001 \ P \0.007).

DISCUSSION

This study supports the hypothesis that a systematicprogram of interceptive orthodontic treatment duringthe mixed dentition is more effective than doing nothingto improve malocclusions over the near term. Thisfinding supports other studies that have reported similarimprovements but has the added advantage of including

Table I. Initial comparison of groups

lnterceptive Control P value

Age (y) (SD) 9.4 (1.4) 9.3(0.8) 0.590

Female (%) 51.6 31.8 0.017

Mean ICON (SD) unweighted components

Esthetics (1-10) 5.3(1.8) 5.2 (1.7) 0.469

Crowding (0-5) 0.6 (1.3) 0.4 (1.6) 0.414

Spacing (0-5) 0.2 (0.5) 0.4 (0.7) 0.006

Crossbite (0-1) 0.6 (0.5) 0.2 (0.4) 0.000

Open bite (0-4) 0.3 (1.0) 0.2 (0.6) 0.277

Overbite (0-3) 0.6 (0.8) 1.0 (0.9) 0.008

Buccal AP (0-2) 1.1 (0.8) 1.2 (0.5) 0.500

AP, Anteroposterior relationship

Table II. Changes in ICON scores by group

lnterceptive(n 5 133)

Control(n 5 113)

ICON cut points(need or acceptability)

Initial mean ICON

(SD)

54.9 (16.6) 54.0 (14.8) .43

Final mean ICON

(SD)

33.6 (16.1)* 54.2 (16.9) \31

lmprovement % 38.8 �0.9

*P \0.0001 (initial vs final; interceptive vs control).

20 King and Brudvik American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

an untreated cohort with similar initial malocclusions toaddress spontaneous improvement or worsening duringthe transition from the deciduous to the permanent den-tition.5,6,9,11 The untreated control group in this studyshowed little change in overall ICON scores withmost subjects in the ICON ‘‘not improved or worse’’

category. This finding clearly supports the conclusionthat these malocclusions do not change in any signifi-cant ways during the mixed dentition without interven-tion. Conversely, the treated group had far fewerpatients in this ‘‘not improved or worse’’ category butincreases in the ‘‘minimal,’’ ‘‘moderate,’’ and

Fig 1. Percent of subjects in the ICON initial complexity categories: easy,\29; mild, 29-50; moder-ate, 51-63; difficult, 64-77; very difficult, .77. Interceptive and control distributions were not differentbased on the chi-square statistic.

Fig 2. Percent of subjects in the ICON improvement categories: improvement scores 5 initial ICONscore – 4 x final ICON score. Greatly improved, .–1; substantially improved, –25 to –1; moderatelyimproved, –53 to –26; minimally improved, –85 to –54; not improved or worse, \–85. Based on thechi-square statistic, the interceptive and control distributions were highly significantly different(P \0.0001).


‘‘substantial’’ improvement categories, suggesting thatinterceptive treatment improves malocclusions to vary-ing degrees. However, few patients in either group werein the ‘‘greatly improved’’ category, supporting the hy-

pothesis that interceptive orthodontic treatments oftenrequire further treatment in the permanent dentition.

When we considered the various components ofmalocclusion, the untreated group experienced no

Fig 3. Changes in ICON components in the untreated control group (unweighted). The initial scoreswere statistically different from the final scores based on the t test (*P \0.007).

Fig 4. Changes in ICON components in the interceptive treatment group (unweighted). The pretreat-ment scores were statistically different from the posttreatment scores based on the t test (*0.001 \P \0.007; **P \0.0001).


January 2010

improvement in any category and a statistically signifi-cant worsening of maxillary crowding, whereas thetreated group had improvements in esthetics, crossbite,maxillary crowding, and anterior open bite. The deteri-oration in maxillary crowding in the untreated sampleduring the mixed dentition might have been caused bya loss of arch perimeter caused by caries, since thisarchival sample subjectively appeared to have moreinterproximal caries, restorations, and early loss ofdeciduous teeth than did the more contemporary treatedsample. This suggestion is supported by Norwegianstudies reporting declines in caries incidence duringthe interval covered by the 2 groups in this study,14,15

with the extensive use of fluoride-based preventiveprograms cited as the major factor contributing to thedecline during the late 1960s and early 1970s.16

This comparison did not consider the long-term sta-bility of interceptive orthodontic corrections. Withoutsuch a comparison, it is impossible to know the extentto which these partial corrections have any lasting ben-efit, or how they compare over the long term with com-prehensive treatment. Preferably, a randomizedcontrolled trial designed to compare interceptive withcomprehensive treatment over the long term would berequired, but such a study has not yet been reported.Randomized controlled trials designed to examine1-phase vs 2-phase treatment of Class II malocclusionsindicate that early intervention has a short-term benefitover no treatment but offers no advantage over 1-phasetreatment at adolescence.17,18 However, these studiesdid not have groups receiving only phase-1 treatment,so they could not address the long-term stability of earlytreatment alone. Because interceptive orthodontic treat-ment has limited goals aimed at reducing, rather thaneliminating, features of malocclusion, it is consideredto be partial treatment by most orthodontists; our datasupport that conclusion. If producing an ideal occlusionlessens the risk of relapse as some contend, one mightpredict that comprehensive treatment would havegreater stability. However, the question of how the qual-ity of outcome influences posttreatment stability is de-batable, with some evidence suggesting that, over thelong term, excellent results tend to deteriorate, whereasaverage results tend to improve.19

The comparison groups were well matched with re-spect to the main outcomes for testing the hypothesis:age, need, complexity, and treatment duration. How-ever, there were some lack of comparability in less rel-evant features that might reflect clinician and patientbiases in favor of early treatment for some populationsubgroups and malocclusions as well as differences indental health between the years covered by the 2 groups.Comparison of initial complexity data suggests slightly

more mild-to-moderate malocclusions in the controlgroup and more difficult-to-very-difficult cases in thetreated group. This seems reasonable, since one expectsthat there would be parental pressure to begin treatmentearly for children with more significant impairments.20

Since mixed dentition treatment was an exclusion crite-rion for the control group in this study, some subjectswith the most severe problems requiring early treatmentmight have been excluded from the control sample. Thisalso might explain the curious finding of more boys inthe control sample, since demand for orthodontic treat-ment is known to be higher for girls.21

Matching also was generally good with respect tothe various components of malocclusion with differ-ences only in more maxillary spacing and fewer cross-bites in the controls. Many clinicians, including theorthodontic faculty at the University of Bergen, treatcrossbites during the mixed dentition because youngerpatients are thought to respond better to treatment; earlytreatment also prevents the risk of asymmetric facialgrowth and gingival damage.22-24 Despite recent reportsthat do not strongly support many of these beliefs, thispractice is common worldwide.25,26 This could explainwhy these patients were found less frequently in our un-treated sample. A greater amount of initial maxillaryspacing in the untreated sample might reflect greaterincisor flaring in this group. The tendency for higheroverbite scores in this group further supports this inter-pretation. Despite good matching on several importantcriteria for testing the main hypothesis, the failure toachieve perfect matching in all categories examined isa limitation of this study, since some of these mightbe confounders in data interpretation.

The wider use of interceptive orthodontic treatmenthas been proposed as a public health measure for reduc-ing the burden of malocclusions in developing coun-tries3 and for increasing access to orthodontic servicesfor underserved populations (low-income, ethnic minor-ities, and geographically isolated subgroups).1 Therationale for this derives from the hypothesis that ortho-dontists can more readily provide shorter, simpler inter-ceptive and preventive treatments to low-incomefamilies compared with the alternative of more expen-sive and longer comprehensive treatments. Cost-effec-tiveness analyses are necessary to demonstrate theeconomic value of this strategy compared with compre-hensive treatment in the permanent dentition. A Finnishstudy found that the cost was lower in 1-stage treatmentsstarted in the permanent dentition compared with2-stage treatments started in the mixed dentition.27

However, no studies have compared mixed dentition in-terceptive treatment alone vs comprehensive permanentdentition treatment alone. Nevertheless, data from


a decision analysis designed to evaluate potential sav-ings by reducing the proportion of children offeredfree orthodontic treatment through the National HealthService in Denmark suggest that this reduction actuallyresults in increased consumption of resources overall.28

This finding lends support to the idea that a comprehen-sive strategy designed to increase access to interceptiveorthodontic services might be more cost-effective over-all than the competing one of focusing primarily on com-prehensive treatment in the permanent dentition for themost difficult malocclusions.

The use of the ICON as the measure of the effective-ness of interceptive orthodontic treatment ignores possi-ble psychological and quality-of-life benefits that can bederived from these approaches. These factors are oftencited as justifications for orthodontic treatment. Thisstudy supported the effectiveness of interceptive treat-ments at improving esthetics but did not address qualityof life. According to several reports in the literature, thedecision to seek orthodontic care is based on the desireto improve dental esthetics, self-esteem, and social ac-ceptance, not necessarily on the degree of malocclusionseverity.29,30 In a study evaluating the psychosocial ef-fects of early treatment, self-concept improved, andnegative social experiences declined, suggesting thatimproved self-esteem might be an important benefit ofinterceptive orthodontic treatment.31

The unweighted ICON component data seem to sug-gest that some types of interceptive treatments are moreeffective than others. This is lost in the summary ICONdata, when the weightings are applied, and the contribu-tions from the individual components are masked. In theICON validation, certain component scores (eg, es-thetics) were weighted much more highly by the expertsthan others. Therefore, these components tend to drivethe summary results. Also, the ICON scores compo-nents of malocclusion, not the effectiveness of certaintreatment approaches. For the latter, it would be moreappropriate to actually measure the various features ofmalocclusion that are addressed by certain types of in-terceptive treatment (eg, overjet by headgear).

CONCLUSIONS

1. Interceptive orthodontic treatment initially im-proves malocclusions with reductions in complex-ity and need compared with doing nothing.

2. Interceptive orthodontic treatment often requiresfollow-up treatment in the permanent dentition.

We thank Kirsten Thunold for collecting the studymodels in the untreated control sample.

REFERENCES

1. King GJ, Hall CV, Milgrom P, Grembowski DE. Early orthodon-

tic treatment as a means to increase access for children enrolled

in Medicaid in Washington state. J Am Dent Assoc 2006;137:

86-94.

2. Kerosuo H. The role of prevention and simple interceptive mea-

sures in reducing the need for orthodontic treatment. Med Princ

Pract 2002;11 Suppl. 1:16-21.

3. Onyeaso CO. Need for preventive/interceptive orthodontic treat-

ment among 7-10-year-old children in Ibadan, Nigeria: an epide-

miological survey. Odontostomatol Trop 2004;27:15-9.

4. Karaiskos N, Wiltshire WA, Odlum O, Brothwell D, Hassard TH.

Preventive and interceptive orthodontic treatment needs of an in-

ner-city group of 6- and 9-year-old Canadian children. J Can Dent

Assoc 2005;71:649.

5. al Nimri K. Richardson A. Interceptive orthodontics in the real

world of community dentistry. Int J Pediatr Dent 2000;10:

99-108.

6. Vakiparta MK, Kerosuo HM, Nystrom ME, Heikinheimo KA.

Orthodontic treatment need from eight to 12 years of age in an

early treatment oriented public health care system: a prospective

study. Angle Orthod 2005;75:344-9.

7. Richmond S, Shaw WC, Roberts CT, Andrews M. The PAR index

(peer assessment rating): methods to determine outcome of ortho-

dontic treatment in terms of improvement and standards. Eur J Or-

thod 1992;14:180-7.

8. Daniels C, Richmond S. The development of the index of com-

plexity, outcome and need (ICON). J Orthod 2000;27:149-62.

9. Mirabelli JT, Huang GJ, Siu CH, King GJ, Omnell L. The effec-

tiveness of phase I orthodontic treatment in a Medicaid popula-

tion. Am J Orthod Dentofacial Orthop 2005;127:592-8.

10. Han H, Davidson WM. A useful insight into 2 occlusal indexes:

HLD(Md) and HLD(CalMod). Am J Orthod Dentofacial Orthop

2001;120:247-53.

11. Theis JE, Huang GJ, King GJ, Omnell ML. Eligibility for publicly

funded orthodontic treatment determined by the handicapping la-

biolingual deviation index. Am J Orthod Dentofacial Orthop

2005;128:708-15.


Stephens CD, et al. The development of the PAR index (peer as-

sessment rating): reliability and validity. Eur J Orthod 1992;14:

125-39.

13. Dahlberg G. Statistical methods for medical and biological stu-

dents. New York: Interscience Publications; 1940.

14. Gimmestad AL, Holst D, Fylkesnes K. Changes in restorative car-

ies treatment in 15-year-olds in Oslo, Norway, 1979-1996. Com-

munity Dent Oral Epidemiol 2003;31:246-51.

15. Birkeland JM, Haugejorden O. Caries decline before fluoride

toothpaste was available: earlier and greater decline in the rural

north than in southwestern Norway. Acta Odontol Scand 2001;

59:7-13.

16. von der Fehr FR, Haugejorden O. The start of caries decline and

related fluoride use in Norway. Eur J Oral Sci 1997;105:21-6.

17. Tulloch JF, Phillips C, Proffit WR. Benefit of early Class II treat-

ment: progress report of a two-phase randomized clinical trial.


18. King GJ, McGorray SP, Wheeler TT, Dolce C, Taylor M. Compar-

ison of peer assessment ratings (PAR) from 1-phase and 2-phase

treatment protocols for Class II malocclusions. Am J Orthod Den-

tofacial Orthop 2003;123:489-96.

19. Nett BC, Huang GJ. Long-term posttreatment changes measured

by the American Board of Orthodontics objective grading system.



January 2010

20. Tung AW, Kiyak HA. Psychological influences on the timing of

orthodontic treatment. Am J Orthod Dentofacial Orthop 1998;

113:29-39.

21. Wheeler TT, McGorray SP, Yurkiewicz L, Keeling SD, King GJ.

Orthodontic treatment demand and need in third and fourth grade

schoolchildren. Am J Orthod Dentofacial Orthop 1994;106:

22-33.

22. Thilander B, Wahlund S, Lennartsson B. The effect of early inter-

ceptive treatment in children with posterior cross-bite. Eur J Or-

thod 1984;6:25-34.

23. Ninou S, Stephens C. The early treatment of posterior crossbites:

a review of continuing controversies. Dent Update 1994;21:420-6.

24. Valencia RM. Treatment of unilateral buccal crossbites in the pri-

mary, early mixed, and permanent dentitions: case reports. J Clin

Pediatr Dent 2007;31:214-8.

25. Petren S, Bondemark L, Soderfeldt B. A systematic review con-

cerning early orthodontic treatment of unilateral posterior cross-

bite. Angle Orthod 2003;73:588-96.

26. Farella M, Michelotti A, Iodice G, Milani S, Martina R. Unilateral

posterior crossbite is not associated with TMJ clicking in young

adolescents. J Dent Res 2007;86:137-41.

27. Jarvinen S, Widstrom E. Determinants of costs of orthodontic

treatment in the Finnish public health service. Swed Dent J

2002;26:41-9.

28. Mavreas D, Melsen B. Financial consequences of reducing treat-

ment availability in a publicly-funded orthodontic service. A de-

cision analysis problem. Br J Orthod 1995;22:47-51.

29. Dann CT, Phillips C, Broder HL, Tulloch JF. Self-concept, Class II

malocclusion, and early treatment. Angle Orthod 1995;65:411-6.

30. Pratelli P, Gelbier S, Gibbons DE. Parental perceptions and atti-

tudes on orthodontic care. Br J Orthod 1998;25:41-6.

31. O’Brien K, Wright J, Conboy F, Chadwick S, Connolly I, Cook P,

et al. Effectiveness of early orthodontic treatment with the Twin-

block appliance: a multicenter, randomized, controlled trial. Part

2: psychosocial effects. Am J Orthod Dentofacial Orthop 2003;

124:488-94.


ORIGINAL ARTICLE

Young patients’ treatment motivation andsatisfaction with orthognathic surgery outcomes:The role of ‘‘possible selves’’

Elizabeth A. Meadea and Marita Rohr Inglehartb

Rochester, Minn, and Ann Arbor, Mich

Introduction: We investigated how young patients’ motivation for orthognathic surgery affected their satisfac-tion with treatment outcomes. The objective was to explore whether patients’ ‘‘possible selves’’ (ie, their ideasof what they might become in the future) and their parents’ proxy assessments of the patients’ possible selveswere significantly correlated with the patients’ treatment satisfaction. Methods: Questionnaire data were col-lected from 115 former patients (ages, 13-21 at time of surgery) and 117 parents (response rates, 41% and42%, respectively), with responses from 95 patient-parent pairs. The patients’ motivations before surgerywere assessed by determining how energized they were by thoughts about themselves after the surgery,and how much they had focused on the outcomes. The parents completed a parallel measure of their chil-dren’s motivation. Patient satisfaction was determined with the postsurgical patient satisfaction question-naire. Results: The more emotionally energized the patients had been before the surgery, the moresatisfied they were with the outcomes (Spearman rho 5 .54, P \0.001). Similarly, the more these patientshad focused on esthetic changes and improved functioning, the more satisfied they were with the outcomes(Spearman rho 5 .46, P \0.001; rho 5 .41, P \0.001, respectively). Parents’ recalls of their children’s moti-vation before the surgery were consistent with the children’s self-reports (all P \0.001) and correlated withthe children’s satisfaction (P \0.001 in the energized domain; P \0.01 for the esthetic changes domain).Conclusions: Young patients’ recalls of their possible self-based motivation for orthognathic surgery werehighly correlated with their treatment satisfaction. Oral surgeons and orthodontists should discuss with youngpatients and their parents the patient’s motivation during the consultation phase before treatment to assesshow energized and focused they are on future treatment outcomes. (Am J Orthod Dentofacial Orthop2010;137:26-34)

Based on the most recent data on occlusal charac-teristics and assuming that the most severe mal-occlusions are associated with an underlying

skeletal discrepancy, Bailey et al1 estimated that 1.8million people in the United States need surgery to cor-rect a severe malocclusion. Advances in orthognathicsurgery over the past 25 years—eg, computer imaging,2

rigid internal fixation,2 and shorter hospital stays3—made surgery a viable option for many patients. In addi-tion, the social acceptance of orthognathic surgery has

increased over the years.4 Many patients become awareof their dentofacial deformity in early adolescence,5 andit is not surprising that the demographic profile of or-thognathic patients is becoming increasingly younger.6

Because of these changes in the patient population, itseems important to understand how satisfied they arewith the outcomes of their orthognathic surgeries andwhich factors can predict their level of treatmentsatisfaction.

Patients generally reported high satisfaction withthe outcome of orthognathic surgery.7-13 For example,between 71% and 86.3% of patients reported that theywould reelect surgical correction,7-13 87% to 89%would recommend surgery to others,14-16 and between80% and 100% of patients reported satisfaction withthe surgery.8,9,11,13.15-17 Considering the increase inyounger orthognathic surgery patients, it is interestingto explore whether satisfaction with surgery outcomesdiffers depending on the patients’ age. Although someauthors found no differences in satisfaction levels be-tween age groups,8,11,13,18 2 studies showed clear evi-dence that the youngest patients were the least

aPrivate practice, Rochester, Minn.bAssociate professor, Department of Periodontics and Oral Medicine, School of

Dentistry; adjunct associate professor, Department of Psychology, College of

Literature, Science and Arts, University of Michigan, Ann Arbor.

The authors report no commerical, proprietary, or financial interest in the


Supported by a grant from the Delta Dental Fund to the first author.

Reprint requests to: Marita Rohr Inglehart, Department of Periodontics and Oral

Medicine, School of Dentistry, University of Michigan, Ann Arbor, MI 48109-

1078; e-mail, [email protected].

Submitted, October 2007; revised and accepted, March 2008.

0889-5406/$36.00


doi:10.1016/j.ajodo.2008.03.022

26


satisfied with the outcomes of orthognathic surgery. Ina long-term assessment, Lazaridou-Terzoudi et al19

found that the youngest age group (14-20 years at thesurgery) was most critical of their current appearanceand less satisfied after the surgery than patients in the2 other age groups (20-26 years and .26 years). In ad-dition, Scott et al20 reported that older patients weremore satisfied than younger patients at all postoperativetimes.

Because orthognathic surgery causes significantimprovement in general well-being,15,19 self-es-teem,9,10,15,16,20,21 self-concept,18,22 social inter-actions,9,15,23 overall body image,24-26 facial bodyimage,25 and profile body image,24 it is worthwhileto consider why younger patients are not more sat-isfied. It is important to explore how orthodontists andoral surgeons can predict—before the surgery—whichpatients will most likely be dissatisfied with the out-comes of their treatment. This study suggests that youngpatients’ treatment motivation before surgery affectstheir satisfaction with the treatment outcome.

What do we know about patients’ motivation and thefactors that influence their decisions to have surgery?Some studies analyzed the role of significant others onthe patients’ surgical decision-making process. Theyshowed that the opinions of others were an importantdecision-making factor for some patients. Flanaryet al27 found that 31.1% of the patients reported thatwhat others thought they should do was important fortheir own decision-making process. Garvill et al,12 ina longitudinal study with 27 patients, found that 52%had reached the decision on their own, but 48% wereinfluenced by family members or professionals. Jacob-son16 found that 22% of patients were urged by familymembers or friends. It seems important to considerthat these influences might be even stronger for youngerpatients.

Several studies analyzed the relationships betweenthese motivational factors and satisfaction with treat-ment outcomes. In a review based on other types of cos-metic surgery, Peterson and Topazian28 stressed thatpatients who had surgery to please parents or significantothers, and who had vague, nonspecific expectationswere more likely to be dissatisfied with the surgical re-sults. Other studies focusing on orthognathic surgeryshowed that lack of support from significant others,8,29

unrealistic expectations,8,30 emotional unprepared-ness,30 and pressure from others to undergo surgery8,30

led to dissatisfaction. In addition, Rispoli et al26 foundthat patients with more esthetic and functional concernsbefore surgery were more likely to be satisfied with theresults. Considering this rather eclectic set of findings,the question arises whether a unifying explanation can

be found for the relationship between patients’ motiva-tions before surgery and their treatment satisfactionafter surgery.

We suggest that considering patients’ ‘‘possibleselves’’ as a unifying concept could provide a connec-tion between patient motivation and satisfaction withthe outcome of orthognathic surgery.31 ‘‘Possibleselves’’ are the patients’ ideas of what they might be-come in the future. Possible selves can be positive andexpressed as hopes or dreams about positive future iden-tities. They can also be negative and take on the form offears of whom the person might become in the future.Research on the use of possible selves in connectionwith psychological phenomena is extensive and coversmany topics.32 Research based on this possible-self the-ory in health-related areas also addressed diverse phe-nomena. For example, some studies used the conceptof possible selves to explore lifestyle-related behaviorssuch as alcohol abuse,33,34 smoking,34-36 and exercis-ing.37 Other studies applied the concept of possibleselves to chronic pain,38 depression,39 borderline per-sonality disorder,40 and Alzheimer’s disease.41 How-ever, this is the first study that applies this widely usedconcept to orthognathic surgery patients.

When considering the relationship between possibleselves and orthognathic-surgery patients’ satisfactionwith treatment outcomes, it is important to understandthat these possible selves affect a person in 2 ways.42

First, the possible-selves concept energizes the personto work toward making positive possible selves becomea reality or to prevent negative possible selves from be-coming real. In this sense, possible selves affect the in-tensity or the strength of a patient’s motivation. Second,possible selves also provide a structure to a person’s mo-tivations by affecting how clearly he or she focuses ona specific positive or negative possible self. This secondmotivational component can be understood as affectingthe direction of the motivation.

Patients who consider undergoing combined ortho-dontic and orthognathic treatment are likely to differin how they engage in possible self-images of them-selves after the treatment. Some patients are excited tohave surgery; they might be looking forward to positivechanges in appearance or function. These patients areenergized. On the other hand, some might not be excitedabout change and might suppress even thinking aboutthe outcomes. Some patients might also be focused re-garding their future possible selves. For example, theymight imagine how they will look after the surgeryand specifically how esthetic their profile will be orhow their smile will change. Focused patients developvivid images about their future possible selves, whereasnonfocused patients do not have a clear picture of the

American Journal of Orthodontics and Dentofacial Orthopedics Meade and Inglehart 27Volume 137, Number 1

surgical outcomes. These differences clearly shape a pa-tient’s motivation to undergo orthognathic surgery and,according to the theory of possible selves of Markus andNurius,31 will determine their satisfaction when the pos-sible selves are realized in the future. An application ofthis theory to the situation of orthognathic surgery pa-tients therefore leads to the concrete hypotheses that(1) the more energized and enthusiastic patients arewhen thinking about a future, postoperative possibleself, and (2) the more patients are clearly focused ontheir future, postoperative possible self, the more satis-fied they will be with the outcomes of their surgery.


This study was approved by the Institutional ReviewBoard for the Medical Sciences at the University ofMichigan, Ann Arbor.

A total of 318 patients who had undergone orthog-nathic surgery at the Oral and Maxillofacial Surgery De-partment of the University of Michigan or in a privategroup practice of 3 oral surgeons in Ann Arbor, Mich,between January 1, 1996, and December 31, 2005,were contacted by their surgeons and informed aboutthe study. These recruitment letters were accompaniedby a survey for the patients and a survey for their parentswith stamped return envelopes. The inclusion criteriafor receiving this mailing were (1) patient age (13-21years at the surgery), (2) a developmental dentofacialdeformity that was corrected by the surgery, and (3)the ability to independently complete the questionnaire.Patients were excluded from the study if they were notin this age group, had surgery to correct secondary de-formities resulting from trauma or tumors, or had sur-gery not involving the tooth-bearing part of the jaws(eg, genioplasty alone).

Thirty-seven questionnaires were not deliverablebecause of invalid addresses. One hundred fifteen pa-tients (response rate, 41%) and 117 parents (responserate, 42%) returned the questionnaires. Of these sur-veys, 95 in each group came from a patient-parentpair. The patients’ average age at the time of surgerywas 16.89 years (SD, 1.920; range, 13-21 years), andthe average age when responding to the survey was21.84 years (SD, 3.054; range, 15-31 years). Sixty-nine percent of the responding patients were female,and 31% were male. Of the 105 patients who identifiedtheir ethnicity, 97 were white, 4 were black, 3 were His-panic, and 1 was Asian. Thirty-two patients (23%) hadmaxillary surgery, 62 (45%) had mandibular surgery,and 43 (31%) had surgery in both arches. Fourteen(12%) of the parent respondents were fathers, and 103(88%) were mothers.

A comparison of the respondents and nonrespondentsshowed no significant differences between the 2 groupsconcerning providers. These analyses showed that femalepatients were more likely to respond than male patients(45.9% vs 33%; P 5 0.02). The responding patients didnot differ significantly from the nonrespondents in ageat surgery (16.96 vs 16.90 years) or current age (21.80vs 22.28 years). However, the respondents had a tendencyto have had their surgery more recently than the nonre-spondents (4.84 vs 5.39 years ago; P 5 0.09).

The researchers gave the prepared mailings to the 4oral surgeons, who then attached address labels andmailed the surveys to the parents of the former patients.A second mailing was sent 6 weeks later to patients andparents who had not responded to the first mailing. Sixweeks after the second mailing, the nonresponding pa-tients from the University of Michigan clinic receiveda third mailing.

The recruitment cover letters for the parents and thepatients were written and signed by the patients’ pro-viders. Because the parents received the survey andhad to give it to their children, this ensured that the par-ents consented to have their children respond if theywere under 18 years of age. No written consent and as-sent were required by the Institutional Review Boardbecause the signatures on these forms would haverevealed the respondents’ names.

Because the measurements of possible selves is do-main specific and no prior research explored the role ofpossible selves for orthognathic surgery patients, it wasnecessary to develop the questions used to assess theseconcepts in this study. The reliability and validity ofthese questions will therefore be reported. In the patientsurvey, 12 questions concerning possible self and moti-vational issues were included (Fig). As can be seen fromthe wording of these questions, they have intuitive facevalidity. Four questions were designed to measure theenergizing component of the patients’ possible selves.Eight items measured how much the patients hadfocused on the postoperative possible self.

A factor analysis (extraction method: principal com-ponent analysis; rotation method: Varimax) was con-ducted to determine whether the 4 items concerningthe energizing component and the 8 items about the fo-cusing component consistently assessed these concepts.The 4 energizing items all loaded highly on a first factor(Cronbach a 5 0.89). A ‘‘possible-self energizing com-ponent’’ index was therefore computed by averaging theanswers to these 4 questions. The 8 focus questionsloaded on 2 factors, with 3 items focusing on postoper-ative esthetics (Fig, patient questions e-g) loading on 1factor and 5 items focusing on the surgery and postop-erative oral function (Fig, patient questions h-l) loading

28 Meade and Inglehart American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

on a second factor. The reliability of these 2 scales(Cronbach a 5 0.82 and 0.71, respectively) justifiedcomputing 2 separate indexes: ‘‘possible self—focus

I’’ and ‘‘possible self—focus II.’’ The results of thefactor analyses showed that these scales have constructvalidity. This fact allowed us to use the created

Fig. Possible-self assessment.


possible-self indices in the presentations of the results inTables I and II.

The patients’ satisfaction with the outcome of thesurgery was assessed with the postsurgical patient satis-faction questionnaire (PSPSQ) by Kiyak et al.24 Thisquestionnaire consisted of 3 questions (Table III) with7-point rating scales from 1, ‘‘not at all,’’ to 7, ‘‘very.’’The factor analysis confirmed that the 3 questionsloaded on 1 factor. The reliability of this scale wasexcellent (Cronbach a 5 0.90), and an index of overallpatient satisfaction was constructed by averagingthe responses to the 3 questions for each patient.

The parent survey contained 4 items to measure theparents’ perception of the energizing component of thepatients’ possible selves and 7 items assessing the par-ents’ perception of the focusing aspect of their chil-dren’s possible selves (Fig). The factor analysis of theparents’ responses showed that the 4 energizing itemsall loaded on 1 factor (Cronbach a 5 0.85). An indexof the parents’ perceptions of the ‘‘possible-self energiz-ing component’’ was therefore constructed by averagingthe answers to these questions. Evaluation of the 7 focusitems showed that the findings concerning the focus onpostoperative esthetics (possible self—focus I) could bereplicated (Cronbach a 5 0.76), whereas the 4 items re-lated to the focus on the surgery and postoperative oralfunction did not load on 1 factor. Only the focus I indexwas therefore computed. The possible-self scales hadface validity (Fig), construct validity as demonstratedby the results of the factor analyses, and external valid-ity. By assessing both patient and parent possible-selfmeasures independently, these measures serve as exter-nal criteria for each other. External validity is shown bythe results reported in Table II: that the independentlyassessed patient and parent possible-self conceptswere highly correlated.


The data were analyzed by using SPSS software(version 14.0, SPSS, Chicago, Ill). Factor analyseswere used to determine the construct validity of the pa-tients’ and the parents’ possible-self measures. The re-liability of the scales was determined by computingCronbach a reliability coefficients for each scale. De-scriptive statistics were used to provide an overviewof the distribution of the respondents’ answers concern-ing the concepts of interest (Tables I and III). Correla-tional analyses with Spearman rho coefficients wereperformed to determine whether the predicted relation-ships between the patients’ possible selves and the par-ents’ perceptions of the patients’ possible selvescorrelated as predicted with treatment satisfaction

(Tables II and IV). A P value of \0.05 was consideredstatistically significant.

RESULTS

Before testing the hypotheses, we consideredwhether the patients truly varied in the degrees to whichthey were motivated. Specifically, were there variationsin the energizing and focusing components of their pos-sible-self reflections? In addition, we consideredwhether parents’ perceptions of their children’s possi-ble-self components varied. Table I provides an over-view of the aggregated percentages of responses andthe mean values for these variables. On average, the pa-tients’ self-perceived and parent-perceived motivationswere high, with average scores between 3.41 and 3.8on a 5-point scale (5 was the highest level of motiva-tion). However, as expected, the patients differed intheir degrees of motivation. For example, in the pa-tients’ responses concerning how energized they werebefore the surgery, Table I shows that 24.3% of them

Table I. Percentages of patient and parent responses in-dicating low, medium, and high patient motivation fororthognathic surgery

Lowmotivation1 to 2.5*

Mediummotivation.2.5 to 3.5

Highmotivation.3.5 to 5 Mean

Patients’ self-

assessments of

possible selves

Energizing

component†n 5 28

24.3%

n 5 29

25.2%

n 5 58

50.4%

3.50

Focus on

postoperative

esthetics (focus I)‡

n 5 21

18.3%

n 5 19

16.5%

n 5 75

65.2%

3.80

Focus on

postoperative oral

function (focus II)§

n 5 15

13.4%

n 5 40

35.7%

n 5 57

50.9%

3.42

Parents’ assessments

of the children’s

possible selves

Energizing

componentkn 5 31

27.2%

n 5 28

24.6%

n 5 55

48.2%

3.41

Focus on

postoperative

esthetics (focus I){

n 5 20

17.5%

n 5 24

21.1%

n 5 70

61.4%

3.74

*The indexes were computed by averaging the responses to the items

in each scale. Answers were given on 5-point answer scales, with low

scores indicating a low degree of energizing and focusing motivation,

and higher scores indicating a higher degree of energizing and focus-

ing motivation.†This index was computed by averaging items 1, a to d, of the Figure.‡This index was computed by averaging items 1, e to g, of the Figure.§This index was computed by averaging items 1, h to l, of the Figure.kThis index was computed by averaging items 2, a to d, of the Figure.{This index was computed by averaging items 2, e to g, of the Figure.


January 2010

were not energized, 25.2% had an intermediate level ofmotivation, and 50.4% were clearly motivated. Basedon these findings, it seems justified to assume that thedegree of the patients’ motivations varied but wereskewed toward highly energized and highly focusedpossible selves. In addition, these data also show thatthe parents’ perceptions of the degree of their children’smotivation varied and were also skewed toward highlyenergized and highly focused possible selves.

One follow-up question was whether parents’ per-ceptions match their children’s self-perceptions. Ascan be seen in Table II, the patients’ self-assessed mo-tivation for orthognathic surgery and the parents’ as-sessments of their children’s motivation weresignificantly correlated. The more the patients re-ported that they engaged in energizing reflections,the more the parents perceived that their children en-gaged in energizing reflections (rho 5 0.60;P \0.001). Similarly, the more the patients reportedthat they focused on the esthetic outcomes of the sur-gery, the more the parents reported that their childrenhad focused on the esthetic outcomes (rho 5 0.58;P \0.001).

Table II also shows that energizing and focusing mo-tivational forces are not independent. The more the pa-tients were energized, the more they also focused onboth the esthetic and oral-function components of theirpostoperative possible selves (rho 5 .86, P \0.001;rho 5 .70; P \0.001, respectively).

The 2 hypotheses stated that (1) the more ener-gized patients were when thinking about a future,

postoperative possible self, and (2) the more patientswere clearly focused on the postoperative possibleself, the more satisfied they would be with the out-come of the surgery. Satisfaction with the treatmentoutcome was assessed with the PSPSQ of Kiyaket al.24 As can be seen in Table III, this questionnaireconsists of 3 questions with 7-point answer scales.Answers to the first 2 questions showed that

Table II. Correlations between patients’ self-assessed motivational factors and parents’ assessments of their children’smotivations

Patients’ self-assessments of possible selves

Patients’ self- assessmentsof possible selves

Energizingcomponent

Focus on postoperativeesthetics (focus I)

Focus on postoperativefunction (focus II)

Energizing component 1 .86

(P \0.001)

n 5 115

.70

(P \0.001)

n 5 112

Focus on postoperative esthetics (focus I) .86

(P \0.001)

n 5 115

1 .59

(P \0.001)

n 5 112

Focus on postoperative function (focus II) .70

(P \0.001)

n 5 112

.59

(P \0.001)

n 5 112

1

Parents’ assessments of the children’s

possible selves

Energizing component .60

(P \0.001)

n 5 93

.52

(P \0.001)

n 5 93

.48

(P \0.001)

n 5 90

Focus on postoperative esthetics (focus I) .54

(P \0.001)

n 5 93

.58

(P \0.001)

n 5 93

.43

(P \0.001)

n 5 91

Table III. Percentages of and average responses to thetreatment-satisfaction items

PSPSQ24 1 and 2* 3 to 5 6 and 7 Mean

If you had to make the

decision again, how likely

would you be to undergo

jaw surgery?

n 5 9

7.8%

n 5 25

21.7%

n 5 81

70.4%

5.83

Considering that

this was an elective

procedure, how likely

would you now be to

recommend this surgery

to others?

n 5 7

6.1%

n 5 28

24.6%

n 5 79

69.3%

5.72

Considering everything,

how satisfied are you now

with the results of the

surgery?

n 5 8

6.9%

n 5 17

14.8%

n 5 90

78.3%

6.05

Index† 1-2.5 .2.5-5.5 .5.5-7 Mean

Patients’ satisfaction

with the surgery

n 5 6

5.3%

n 5 24

21.0%

n 5 84

73.7%

5.86

*The answers were given on 7-point answer scales ranging from 1,

‘‘not at all,’’ to 7, ‘‘very.’’ †The index was computed by averaging

the responses to the 3 questions.


approximately 70% of the patients would likely re-elect surgery and recommend it to others; consideringeverything, 78.3% of the respondents reported clearsatisfaction with the results of the surgery. When anaverage score of the 3 responses was computed asan index of patient satisfaction, 73.7% of the respon-dents were highly satisfied, and 5.3% were not at allor not satisfied.

Table IV shows that the data clearly supported thehypotheses. The more energized the patients were beforesurgery, the higher their postsurgical satisfaction was, asmeasured with their responses to the 3 satisfaction items(rho 5 .46, P\0.001; rho 5 .52, P \0.001; rho 5 .43,P\0.001) and the overall satisfaction index (rho 5 .54,P \0.001). The patients’ focus on the esthetic compo-nent correlated significantly with each of the 3 items ofthe satisfaction scale (rho 5 .37, P \0.001; rho 5 .46,P\0.001; rho 5 .41, P\0.001) and the overall satisfac-tion index (rho 5 .46, P \0.001). In addition, the morethe patients had focused on the surgery, especially theoral-functioning component of their possible selves,the more satisfied they were. The patients’ possibleself—focus II index correlated significantly with eachof the 3 items of the satisfaction scale (rho 5 .35,P \0.001; rho 5 .41, P \0.001; rho 5 .35, P \0.001)and the overall satisfaction index (rho 5 .41,P \0.001) (Table IV).

Table IValso shows that the parents’ assessments ofthe patients’ possible selves correlated significantlywith their children’s treatment satisfaction. The morethe parents perceived their children as having beenenergized before the surgery, the more satisfied thepatients were with the treatment outcome (rho 5 .36,

P \0.001). The more the parents perceived theirchildren as having had a clear esthetic focus, the moresatisfied the patients were (rho 5 .29, P \0.01).

DISCUSSION

An analysis of the degree to which the respondents inthis study were satisfied with their surgery showed thatoverall 73.7% of the patients reported very high satisfac-tion. On a scale from 1, ‘‘not at all satisfied,’’ to 7, ‘‘verysatisfied,’’ the average satisfaction score was 5.86. Thesefindings are consistent with the results of other studiesthat used the PSPSQ to assess patient satisfaction withorthognathic surgery outcomes.24 The average satisfac-tion scores in these other studies ranged from 5.15 to6.11.18,24,26,29,43 The result of this study that approxi-mately 80% of the patients would reelect to have surgeryif they had to make the decision again also agreed withpast studies.7-12 However, 79% were likely to recom-mend the surgery to others; this is slightly lower thanthe range of 87% to 89% reported by other authors.14-16

Although these high percentages of satisfied pa-tients are impressive, the fact that some patients wereunhappy with their surgical results deserves attention.One interesting question in this context would be to de-termine how we can predict which patients will be dis-satisfied with the outcomes of orthognathic surgery. Wefocused on finding an answer to this question by usingthe theory of possible selves.31 Based on that, this studytested whether the degree to which patients were ener-gized by thinking about the surgical outcomes or fo-cused on the outcomes of the surgery would becorrelated with their postsurgical satisfaction. The

Table IV. Correlations between possible-self components and patient satisfaction with surgery

PSPSQ24

Patients’ self-assessmentsof possible selves

Parents’ assessmentsof children’s possible selves

Energizingcomponent

Focus onesthetics(focus I)

Focus onfunction(focus II)

Energizingcomponent

Focus onesthetics (focus I)

If you had to make the decision again, how

likely would you be to undergo jaw

surgery?

0.46

(P \0.001)

n 5 115

0.37

(P \0.001)

n 5 115

0.35

(P \0.001)

n 5 112

0.32

(P \0.01)

n 5 93

0.24

(P 5 0.02)

n 5 93

Considering that this was an elective

procedure, how likely would you now be

to recommend this surgery to others?

0.52

(P \0.001)

n 5 114

0.46

(P \0.001)

n 5 114

0.41

(P \0.001)

n 5 111

0.37

(P \0.001)

n 5 93

0.28

(P 5 0.01)

n 5 93

Considering everything, how satisfied are you

now with the results of the surgery?

0.43

(P \0.001)

n 5 115

0.41

(P \0.001)

n 5 115

0.35

(P \0.001)

n 5 112

0.26

(P 5 0.01)

n 5 93

0.28

(P 5 0.01)

n 5 93

Index

Patients’ satisfaction with the surgery 0.54

(P \0.001)

n 5 114

0.46

(P \0.001)

n 5 114

0.41

(P \0.001)

n 5 111

0.36

(P \0.001)

n 5 93

0.29

(P \0.01)

n 5 93


January 2010

results showed convincingly that both the energizingand the focusing aspects of the patients’ possible-selfevaluations were significantly correlated with theirpostsurgical satisfaction. In addition, the parents’ as-sessments of their children’s possible self componentswere also correlated with the children’s satisfaction.

These findings support the idea that the theory ofpossible selves can be used as a unifying concept tohelp predict patient satisfaction. As shown previously,unrealistic expectations,8,30 emotional unprepared-ness,30 and pressure from others to undergo surgery8,30

can cause patient dissatisfaction. These factors are re-lated to the concept of possible selves. All can be rein-terpreted as affecting the specificity, clarity, andemotional quality of a patient’s possible self. For exam-ple, emotionally unprepared patients might not be ex-cited about the potential outcomes of the surgery orclearly focused on a postoperative possible self. Thiscan create problems for the patient’s acceptance of thenew situation.

What are the practical implications of these find-ings? The answer to this question is twofold. First,the findings imply that it could be helpful for providersto understand the viewpoints of their younger patients ifthey want to predict whether a patient is likely to be sat-isfied or dissatisfied with the treatment outcome. Be-cause these findings suggest that the more excited andfocused a patient is on the future surgery outcomes,and the more satisfied he or she will be with the resultsof the surgery, it could be helpful for both orthodontistsand oral surgeons to ask their patients about their pos-sible-self motivations and consider these when finaliz-ing a treatment plan. In addition to asking about theirpossible-self motivations, it would also be helpful toask parents about their perceptions of their children’spossible selves. The information from these conversa-tions could be valuable for orthodontists and oral sur-geons when discussing treatment options witha surgical candidate.

A second practical consideration that can be basedon these findings is that motivations are not static butcan be shaped by communication with the patients. Iforthodontists and oral surgeons realize that patientsare not energized and positively focused on the surgeryoutcomes, and thus are in danger of ultimately beingdissatisfied with the treatment outcomes, active stepscan be taken to induce possible-self reflections in thesepatients. For example, providing a patient with vivid im-ages of his or her future esthetic appearance or engagingthe patient in an active comparison of facial features be-fore and after surgery could increase the patient’s moti-vation considerably and might ultimately improvetreatment satisfaction.

A clear limitation of this study was that it was retro-spective. Patients answered the questions about theirmotivations for surgery on average 4.8 years after theyhad actually undergone surgery. This limitation isclearly related to the fact that no prior research hasexplored the role of possible selves in orthognathic sur-gery patients’ treatment satisfaction. This study wasa first exploration of the usefulness and importance ofthe possible-self concept for this domain. With thesefindings, however, it is now important to conduct a pro-spective study to assess patients’ possible-self motiva-tions presurgically and then evaluate their satisfactionpostsurgically.

CONCLUSIONS

1. Patients differ in the degrees to which they are en-ergized by thinking about their postsurgical possi-ble selves and to which they focus on postsurgicalpossible selves. In addition, parents can make proxyassessments of their children’s possible-self consid-erations. These proxy assessments correlated sig-nificantly with the children’s own assessmentsregarding being energized and focused on esthetics.

2. The more energized and focused the patients re-called being before their orthognathic surgery, themore satisfied they were with the treatment out-comes.

3. The more the parents perceived that their childrenhad been energized and focused on their postsurgi-cal possible selves before surgery, the more satis-fied the patients were.

We thank Drs Airton Arruda, George Upton, andKatherine Kelly for their valuable feedback to this re-search as members of the first author’s masters thesiscommittee, and Drs Dalbert W. Fear, William D. Baxter,Roger B. Hitchcock, and George Upton for supportingthis study by allowing us to survey their patients andwriting a cover letter for the survey; without their gen-erous help, this study could not have been conducted.

REFERENCES

1. Bailey LJ, Proffit WR, White R Jr. Assessment of patients for

orthognathic surgery. Semin Orthod 1999;5:209-22.

2. Bailey LJ, Haltiwanger LH, Blakey GH, Proffit WR. Who seeks

surgical-orthodontic treatment: a current review. Int J Adult

Orthod Orthognath Surg 2001;16:280-92.

3. Tucker MR. Orthognathic surgery versus orthodontic camouflage

in the treatment of mandibular deficiency. J Oral Maxillofac Surg

1995;53:572-8.

4. Phillips C, Broder HL, Bennett ME. Dentofacial disharmony: mo-

tivations for seeking treatment. Int J Adult Orthod Orthognath

Surg 1997;12:7-15.


5. Broder HL, Phillips C, Kaminetzky S. Issues in decision making:

should I have orthognathic surgery? Semin Orthod 2000;6:

249-58.

6. Bennett ME, Phillips CL. Assessment of health-related quality of

life for patients with severe skeletal disharmony: a review of the

issues. Int J Adult Orthod Orthognath Surg 1999;14:65-75.

7. Athanasiou AE, Melsen B, Eriksen J. Concerns, motivation, and

experience of orthognathic surgery patients: a retrospective study

of 152 patients. Int J Adult Orthod Orthognath Surg 1989;4:47-55.

8. Chen B, Zhang ZK, Wang X. Factors influencing postoperative

satisfaction of orthognathic surgery patients. Int J Adult Orthod

Orthognath Surg 2002;17:217-22.

9. Cunningham SJ, Crean SJ, Hunt NP, Harris M. Preparation, per-

ceptions, and problems: a long-term follow-up study of orthog-

nathic surgery. Int J Adult Orthod Orthognath Surg 1996;11:41-7.

10. Derwent SK, Hunt NP, Cunningham SJ. A comparison of parents’

and patients’ views of orthognathic treatment. Int J Adult Orthod

Orthognath Surg 2001;16:171-8.

11. Finlay PM, Atkinson JM, Moos KF. Orthognathic surgery: patient

expectations; psychological profile and satisfaction with outcome.

Br J Oral Maxillofac Surg 1995;33:9-14.

12. Garvill J, Garvill H, Kahnberg KE, Lundgren S. Psychological fac-

tors in orthognathic surgery. J Craniomaxillofac Surg 1992;20:28-33.

13. Pahkala RH, Kellokoski JK. Surgical-orthodontic treatment and

patients’ functional and psychosocial well-being. Am J Orthod


14. Frost V, Peterson G. Psychological aspects of orthognathic sur-

gery: how people respond to facial change. Oral Surg Oral Med

Oral Pathol 1991;71:538-42.

15. Hugo B, Becker S, Witt E. Assessment of the combined orthodon-

tic-surgical treatment from the patients’ point of view. A longitu-

dinal study. J Orofac Orthop 1996;57:88-101.

16. Jacobson A. Psychological aspects of dentofacial esthetics and or-

thognathic surgery. Angle Orthod 1984;54:18-35.

17. Shalhoub SY. Scope of oral and maxillofacial surgery: the psychoso-

cial dimensions of orthognathic surgery. Aust Dent J 1994;39:181-3.

18. Ostler S, Kiyak HA. Treatment expectations versus outcomes

among orthognathic surgery patients. Int J Adult Orthod Orthog-

nath Surg 1991;6:247-55.

19. Lazaridou-Terzoudi T, Kiyak HA, Moore R, Athanasiou AE,

Melsen B. Long-term assessment of psychologic outcomes of or-

thognathic surgery. J Oral Maxillofac Surg 2003;61:545-52.

20. Scott AA, Hatch JP, Rugh JD, Hoffman TJ, Rivera SM, Dolce C,

et al. Psychosocial predictors of satisfaction among orthognathic

surgery patients. Int J Adult Orthod Orthognath Surg 2000;15:7-15.

21. Gerzanic L, Jagsch R, Watzke IM. Psychologic implications of or-

thognathic surgery in patients with skeletal Class II or Class III

malocclusion. Int J Adult Orthod Orthognath Surg 2002;17:75-81.

22. Flanary CM, Barnwell GM, VanSickels JE, Littlefield JH,

Rugh AL. Impact of orthognathic surgery on normal and abnor-

mal personality dimensions: a 2-year follow-up study of 61 pa-

tients. Am J Orthod Dentofacial Orthop 1990;98:313-22.

23. Modig M, Andersson L, Wardh I. Patients’ perception of improve-

ment after orthognathic surgery: pilot study. Br J Oral Maxillofac

Surg 2006;44:24-7.

24. Kiyak HA, Hohl T, West RA, McNeill RW. Psychologic changes

in orthognathic surgery patients: a 24-month follow up. J Oral

Maxillofac Surg 1984;42:506-12.

25. Kiyak HA, McNeill RW, West RA, Hohl T, Heaton PJ. Person-

ality characteristics as predictors and sequelae of surgical and

conventional orthodontics. Am J Orthod 1986;89:383-92.

26. Rispoli A, Acocella A, Pavone I, Tedesco A, Giacomelli E,

Ortiz L, et al. Psychoemotional assessment changes in patients

treated with orthognathic surgery: pre- and postsurgery report.

World J Orthod 2004;5:48-53.

27. Flanary CM, Barnwell GM Jr, Alexander JM. Patient perceptions

of orthognathic surgery. Am J Orthod 1985;88:137-45.

28. Peterson LJ, Topazian RG. Psychological considerations in

corrective maxillary and midfacial surgery. J Oral Surg 1976;

34:157-64.

29. Holman AR, Brumer S, Ware WH, Pasta DJ. The impact of inter-

personal support on patient satisfaction with orthognathic surgery.


30. Flanary CM, Alexander JM. Patient responses to the orthognathic

surgical experience: factors leading to dissatisfaction. J Oral Max-

illofac Surg 1983;41:770-4.

31. Markus H, Nurius P. Possible selves. Am Psychol 1986;41:

954-69.

32. King LA, Hicks JA. Whatever happened to ‘‘what might have

been’’? Regrets, happiness, and maturity. Am Psychol 2007;62:

625-36.

33. Quinlan SL, Jaccard J, Blanton H. A decision theoretic and proto-

type conceptualization of possible selves: implications for the pre-

diction of risk behavior. J Pers 2006;74:599-630.

34. Stein KF, Roeser R, Markus HR. Self-schemas and possible selves

as predictors and outcomes of risky behaviors in adolescents. Nurs

Res 1998;47:96-106.

35. Freeman MA, Hennessy EV, Marzullo DM. Defensive evaluation

of antismoking messages among college-age smokers: the role of

possible selves. Health Psychol 2001;20:424-33.

36. Shadel WG, Cervone D. Evaluating social-cognitive mechanisms

that regulate self-efficacy in response to provocative smoking

cues: an experimental investigation. Psychol Addict Behav

2006;20:91-6.

37. Ouellette JA, Hessling R, Gibbons FX, Reis-Bergan M,

Gerrard M. Using images to increase exercise behavior: proto-

types versus possible selves. Pers Soc Psychol Bull 2005;31:

610-20.

38. Morley S, Davies C, Barton S. Possible selves in chronic pain:

self-pain enmeshment, adjustment and acceptance. Pain 2005;

115:84-94.

39. Allen LA, Woolfolk RL, Gara MA, Apter JT. Possible selves in

major depression. J Nerv Ment Dis 1996;184:739-45.

40. Janis IB, Veague HB, Driver-Linn E. Possible selves and border-

line personality disorder. J Clin Psychol 2006;62:387-94.

41. Cotrell V, Hooker K. Possible selves of individuals with Alz-

heimer’s disease. Psychol Aging 2005;20:285-94.

42. Inglehart MR, Markus H, Brown DR. The effects of possible

selves on academic achievement—a panel study. In: Forgas JP,

Innes M, editors. Recent advances in social psychology: an

international perspective. Amsterdam, The Netherlands; 1989.

p. 469-77.

43. Motegi E, Hatch JP, Rugh JD, Yamaguchi H. Health-related

quality of life and psychosocial function 5 years after orthog-

nathic surgery. Am J Orthod Dentofacial Orthop 2003;124:

138-43.


January 2010

ORIGINAL ARTICLE

Psychosocial impact of hypodontia in children

Emma Laing,a Susan J. Cunningham,b Steven Jones,c David Moles,d and Daljit Gille

London, United Kingdom

Introduction: The purpose of this cross-sectional study was to determine the psychosocial impact of hypo-dontia (multiple dental agenesis) in children and the influence of factors such as severity of hypodontia, num-ber of retained deciduous teeth, age, and sex. The implications of hypodontia for affected patients have beenpoorly investigated; this, in part, relates to the lack of appropriate measurements to assess the impact of oralconditions on quality of life, particularly among children and adolescents. Methods: A total of 123 children(49.6% boys, 50.4% girls; mean age, 13.6 years; SD, 1.6 years) were recruited on the basis of predeterminedinclusion criteria to either a hypodontia group or a routine orthodontic group of similar treatment need (index oforthodontic treatment need, dental health component 4 or 5) but without hypodontia. Each patient completedthe child perceptions questionnaire and 2 visual analog scales to determine the global effects of hypodontia onesthetics and function. Results: The mean number of missing teeth in the hypodontia group was 4.52 (SD,3.33). There were no statistically significant differences in child perceptions questionnaire scores (overall orat domain level) or visual analog scores between the hypodontia and the routine orthodontic groups(P .0.05). Univariable linear regression analyses provided some evidence that difficulty with chewing was as-sociated with the severity of hypodontia (P 5 0.030). Conclusions: In this sample, hypodontia did not appearto affect the psychosocial status of patients any more than other features of a malocclusion measured with theindex of orthodontic treatment need, dental health component 4 or 5. Patients with hypodontia did, however,have more difficulty in chewing when the deciduous teeth associated with the missing permanent teeth hadbeen exfoliated. This highlights the possible importance of retaining deciduous teeth in patients with severehypodontia. (Am J Orthod Dentofacial Orthop 2010;137:35-41)

Hypodontia is the term used to describe the de-velopmental absence of at least 1 deciduousor permanent tooth, excluding the third mo-

lars.1 It is the most common dental developmentalanomaly2 and has occurred in humans since at least Pa-leolithic times.3 The lack of 1 or 2 permanent teeth, withno associated systemic disorders, is the mildest and themost common phenotype. Hypodontia is more commonamong persons who are genetically related than in thosewho are not.4,5

There is a wealth of research into the prevalence,probable etiology, and dentoskeletal effects of hypodon-tia, but the social, behavioral, educational, medical, andfinancial implications for an affected person and his or

her immediate family have been poorly investigated.The psychosocial impact of hypodontia has received lit-tle attention in the literature. Hobkirk et al,6 in a retro-spective study of 451 patients with hypodontia, foundthat the most common complaints were spacing be-tween the teeth and poor esthetics, and some patientswere aware that they had missing teeth. Functionalproblems because of the reduced surface area of theocclusal table comprised only 8.7% of patients’ com-plaints.6

Facial esthetics and esthetic dentistry have becomeprominent forces in today’s popular culture. Dentofacialappearance can affect interpersonal relationships andperceived qualities such as friendliness, social class, in-telligence, and popularity from infancy to adulthood.Attractive children are seen by others as more intelligentand having more positive social behavior, and they re-ceive more positive treatment than their less attractivecounterparts.7 Therefore, deviation from ideal dentofa-cial esthetics, particularly in children, might adverselyaffect self-esteem and self-confidence, and attractmockery from peers.8 It is reasonable to speculate thatdeviations from ‘‘ideal’’ or ‘‘normal’’ dentofacial es-thetics could be detrimental to a person’s psychosocialwell-being. The general assumption follows that less at-tractive children might experience some psychosocialdistress as a result of their condition.

From Eastman Dental Hospital, University College London Hospitals NHS

Trust, London, United Kingdom.aSpecialist registrar, Department of Orthodontics.bSenior lecturer, Department of Orthodontics.cConsultant and honorary senior lecturer, Department of Orthodontics.dSenior clinical lecturer, Health Services Research.eConsultant, Department of Orthodontics.



Reprint requests to: Emma Laing, Eastman Dental Hospital, University College

London Hospitals NHS Trust, 256 Gray’s Inn Rd, WC1X 8LD, London, United

Kingdom; e-mail, [email protected].

Submitted, November 2007; revised and accepted, January 2008.

0889-5406/$36.00


doi:10.1016/j.ajodo.2008.01.024

35


Orthodontic treatment need and effectiveness oftreatment are often assessed by morphologic changesin anatomic parameters such as occlusal indexes andcephalometric measurements. However, these types ofnormative assessments are not always relevant to pa-tients from a psychological, social, or functional aspectbecause ‘‘the demarcation between acceptable and un-acceptable occlusion is influenced by idiosyncraticjudgement.’’9 More recently, there has been an emphasison oral health-related quality of life (OHRQOL) re-search, which has evaluated the impact of oral condi-tions on patients’ lives and might enable a level ofcare that is appropriate for each patient.10 These ap-proaches provide insight into the potential conse-quences of the condition, beyond clinical parameters,on the day-to-day lives of affected patients.11 This facil-itates the understanding of the condition’s importance inthe provision of oral health care and can ensure that thebest practice guidelines are established.

Investigation of children’s OHRQOL has been ad-dressed only relatively recently in the literature, proba-bly because of the greater complexity of assessingOHRQOL in children than adults.12 The psychosocialimpact of hypodontia in children has received little at-tention in the literature to date, and it was our purposeto investigate this area. The null hypotheses for thestudy were (1) there is no difference in psychosocial sta-tus between subjects with hypodontia and those with nohypodontia but with a malocclusion of a similar treat-ment need as classified by index of orthodontic treat-ment need (IOTN), dental health component (DHC) 4or 5; and (2) age, sex, and extent of hypodontia haveno effect on psychosocial status.


The research proposal was approved by the JointUniversity College London/University College LondonHospitals Committee on the Ethics of Human ResearchCommittee. The subjects were recruited from new-patient orthodontic clinics in a teaching hospital by 1researcher (E.L) from July 2006 to January 2007, andconsecutive patients who satisfied the inclusion criteriawere invited to participate. Two groups of participantswere recruited: a hypodontia group and an orthodonticgroup not affected by hypodontia (the routine orthodon-tic group). The criteria for inclusion in the study werepatients between 11 and 16 years of age; IOTN, DHC4 or 5; and radiographically confirmed hypodontia(for the hypodontia group) or no hypodontia (for theroutine orthodontic group). Exclusion criteria were pre-vious orthodontic treatment, associated medical condi-

tions, not accompanied by a parent or guardian, ornon-English speaker.

Each patient and the parent or legal guardian weregiven written information outlining the details of thestudy, and its purpose was explained verbally. If the pa-tient and parent or legal guardian agreed to participate,they were taken to a quiet nonclinical area, wherewritten informed consent was obtained. The followingdemographic details were recorded: age and sex.

A sample size calculation was performed usingnQuery Advisor software (version 5.0, Statistical Solu-tions, Saugus, Mass) with data from the first 33 hypo-dontia patients and the first 10 routine orthodonticpatients to be recruited. It was found that, for a chi-square test (with a 0.05 level of significance) to have80% power to detect a 25% difference in questionnairescores, 61 subjects were required in each group.

Each child filled out a validated self-completed ques-tionnaire, the child perceptions questionnaire (11-14years) (CPQ) compiled by Jokovic et al13 at the Facultyof Dentistry, University of Toronto in Canada. The au-thors’ consent to use the questionnaire was obtained be-fore the study. The CPQ was developed withorthodontic and pediatric dentistry patients and performedwell as a discriminative measure between study group pa-tients and was therefore appropriate for our research ques-tion.13 It was shown to be a valid measure with excellentinternal consistency and test-retest reliability.13 Marsh-man et al14 further investigated these parameters and con-firmed that the validity and reliability were acceptable foruse in an orthodontic population in the United Kingdom.These aspects were therefore not investigated in this study.

The CPQ consists of 37 questions divided into 4health domains: oral symptoms, functional limitations,emotional well-being, and social well-being. The ques-tions cover the child’s views and perceived views ofpeers about his or her dental appearance, and behavioraldifficulties at home and at school. The response optionsand scores for the 37 CPQ questions were as follows: 0,never; 1, once or twice; 2, sometimes; 3, often; and 4,every day or almost every day.

After completion of the CPQ, the participants wereasked to complete 2 visual analog scales (VAS) relatedto the appearance and function of their teeth. Thesecomprised 2 horizontal lines, 100 mm long, anchoredby word descriptors at each end. The function VASwas anchored with ‘‘I find it easy to eat’’ and ‘‘I find ithard to eat’’; the appearance VAS had ‘‘I like the waymy teeth look’’ and ‘‘I hate the way my teeth look’’ ateither end. Each patient’s score was obtained by usinga ruler to measure the distance from the left anchor towhere he or she had marked the response. Possiblescores ranged from 0 to 100, corresponding to the

36 Laing et al American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

distance in millimeters to where the cross was placed.An example was provided to clarify the exercise. Sinceall questions in both tasks were closed, a box for com-ments was provided at the end of the questionnaire.

Participants were allowed as much time as theyneeded to complete the CPQ and the VAS, and it wasmade clear that these were for them to complete on theirown, without the assistance of a parent or guardian. Onaverage, the tasks took each child 10 minutes to com-plete. If the patient and the parent or guardian did nothave enough time that day, they were given a stamped,addressed envelope and asked to return the completedCPQ and VAS. It was made clear that the child shouldcomplete the exercise at home.


The analysis of the data was carried out by usingStatistical Package for the Social Sciences software(version 14, SPSS, Chicago, Ill). The data analysis com-prised 2 sections: analysis of both groups, by using de-scriptive statistics and 2-sample t tests to test fordifferences in CPQ scores (overall and at domain level)and Mann-Whitney U tests to test for differences in VASscores. In the second section, analysis was restricted tothe hypodontia group to investigate the specific effectsof the extent and location of hypodontia on the CPQand VAS scores. To investigate the effects of retentionof deciduous teeth, hypodontia was defined in this studyas ‘‘absolute’’ and ‘‘relative.’’ Absolute hypodontia wasthe number of missing permanent teeth, and relative hy-podontia was the number of missing permanent teethminus the number of retained deciduous teeth. These2 scoring methods enabled the effects of retention ofthe associated deciduous tooth to be considered. Uni-variable linear regression analyses were used to identifyvariables contributing to differences in the total and do-main CPQ and the VAS scores. In all regression analy-ses, the model assumptions were found to besatisfactory.

RESULTS

All 123 patients in the hypodontia and routine ortho-dontic groups who were approached agreed to participate;there were no withdrawals. Eleven participants took theirquestionnaires home to complete them, and they were allreturned by mail. The sample comprised 62 hypodontiapatients (35 boys, 27 girls) and 61 nonhypodontia patients(26 boys, 35 girls) (Table I). The chi-square test for inde-pendent samples showed no statistically significant differ-ence in sex distribution between the groups (P 5 0.176).The mean age of both groups was 13, with an age range of11 to 16 years (Table I).

An overall CPQ score was calculated for each childby adding all questionnaire item scores, with a maxi-mum possible score of 148. Scores ranged from 2 to80, and, for both groups, the scores were approximatelynormally distributed (Table II). The higher the CPQscore, the more often a child would be affected, and itwas inferred that the higher the score, the greater thepsychosocial impact. A 2-sample t test showed no statis-tically significant difference in total questionnairescores between the hypodontia and the routine ortho-dontic groups (P 5 0.566) (Table II). With respect toeach of the 4 CPQ domains, 2-sample t tests showedno evidence of any statistically significant differencesin mean scores between the 2 groups (Table II).

Mann-Whitney U tests for 2 independent sampleswere performed to investigate any potential differencesin median VAS scores for eating and appearance be-tween the groups. However, there was no evidence ofany intergroup differences in VAS scores for functionor appearance (P 5 0.766 and P 5 0.869, respectively)(Table III).

In the statistical analysis of the hypodontia group,the mean number of missing teeth per subject was4.52 (SD, 3.33). The numbers of developmentally miss-ing teeth per sextant and retained deciduous teeth persextant were recorded. From these data, it was possibleto calculate a relative hypodontia score. Thus, the meanabsolute hypodontia was 4.52, and the mean relative hy-podontia was 2.03 missing teeth per subject (SD, 1.76).There was an approximately even distribution of bothmean absolute and relative hypodontia by sextant, butthe mandibular middle sextant most commonly hadmissing teeth (Table IV). The maxillary labial segmenthad the fewest retained deciduous teeth and was the sex-tant with the greatest relative hypodontia (Table IV).

Univariable regression analyses were performed toestablish whether the overall questionnaire scores (de-pendent variable) of the hypodontia patients were influ-enced by any of the following factors (independentvariables): age, gender, total absolute hypodontia, totalrelative hypodontia, total absolute hypodontia in the

Table I. Characteristics of the sample

Group Hypodontia (n 5 62) Nonhypodontia (n 5 61)

Sex

Male 35.56% 26.43%

Female 27.44% 35.57%

Age (y)

Mean 13.72 13.46

SD 1.53 1.67

Minimum 11.00 11.05

Maximum 16.97 16.65

American Journal of Orthodontics and Dentofacial Orthopedics Laing et al 37Volume 137, Number 1

maxillary middle sextant, and total relative hypodontiain the maxillary middle sextant. The maxillary middlesextant was looked at in isolation, because it wasthought that hypodontia affecting the maxillary anteriorregion would lead to greater esthetic impairment andperhaps the most psychosocial impact. There was no ev-idence of any statistically significant relationships be-tween the total CPQ scores for the hypodontia groupand any independent variable described (Table V).

Univariable regression analyses were performed foreach CPQ domain to investigate the effects of the extentor the location of hypodontia on oral symptoms, func-tional limitations, emotional well-being, and socialwell-being (Table V). The only statistically significantrelationship was that the functional limitations domainwas significantly affected by total relative hypodontia(P 5 0.030) (Table V). Thus, as total relative hypodon-tia increased, so did the functional limitations domainscore; when more teeth were missing (without retentionof their deciduous predecessors), the worse the reportedeffects on function. The P values for the maxillary mid-dle sextant were not significant (P 5 0.399, P 5 0.154)(Table V); this suggests that any functional limitationswere limited to the teeth in the posterior sextants.

To further investigate these findings, post-hoc re-gression analyses were performed to determine the ef-fects of the extent of hypodontia on the VAS scores(Table V). Because of the findings of the regressionanalyses for the CPQ scores, it was decided to restrictthese analyses to 2 parts: the effect of total relative hy-podontia and the VAS score for eating; and the effect of

total relative hypodontia in the maxillary middle sextantand the VAS score for appearance. There was a statisti-cally significant relationship between the VAS scoresfor eating and total relative hypodontia (P 5 0.016),with the VAS score increasing (ie, worsening function)as the extent of relative hypodontia increased (Table V).

DISCUSSION

Few previous studies have investigated the psycho-social impact of a developmental defect of the dentitionin affected patients in terms of quality of life and self-image.10,15 There are no similar studies on which thesample size and selection for this study could be based;hence, the first 33 hypodontia patients and the first 10nonhypodontia patients recruited were used for the sam-ple size calculation. A similar study by Wong et al16

investigated the effects of severe hypodontia on OHR-QOL. However, there were no control group in theirstudy, no sample size calculation, and no rationalestated for their choice of 25 subjects.

An orthodontic sample was chosen as a controlgroup so that hypodontia patients could be comparedwith patients with malocclusions of a similar treatmentneed as classified by the IOTN, rather than a Class Iideal occlusion. Future studies could include an

Table II. Summary of the 2-sample t test to compare mean total and domain CPQ scores for the hypodontia and non-hypodontia groups

DomainMaximum

possible scoreHypodontia

meanNonhypodontia

meanMean differencein CPQ scores

95% CI ofdifference P value

Oral symptoms 24 5.03 5.74 0.705 –0.443, 1.855 0.227

Functional limitations 36 6.06 6.16 0.994 –1.526, 1.725 0.904

Emotional well-being 36 6.56 7.49 0.927 –1.647, 3.501 0.477

Social well-being 52 5.77 5.75 –0.020 –1.900, 1.860 0.983

Total CPQ score 148 26.82 28.52 1.70 –4.149, 7.553 0.566

Table III. Results of the Mann-Whitney U test comparingmedian VAS scores for the hypodontia and nonhypo-dontia groups

Question Group Median score P value

VAS eating Hypodontia 5.0 0.766

Nonhypodontia 7.0

VAS appearance Hypodontia 55.0 0.869

Nonhypodontia 60.0

Table IV. Mean absolute and relative hypodontia distri-butions by sextant in children with hypodontia

Absolute hypodontia* Relative hypodontia†

Sextant Mean 95% CI Mean 95% CI

Maxillary right 0.76 0.53, 0.99 0.26 0.11, 0.40

Maxillary middle 1.02 0.75, 1.29 0.76 0.55, 0.97

Maxillary left 0.65 0.42, 0.87 0.21 0.08, 0.34

Mandibular right 0.74 0.55, 0.94 0.19 0.08, 0.30

Mandibular middle 0.66 0.38, 0.94 0.40 0.21, 0.60

Mandibular left 0.69 0.50, 0.89 0.21 0.10, 0.32

Total 4.52 3.67, 5.36 2.03 1.58, 2.48

*Absolute hypodontia was the number of missing permanent teeth.†Relative hypodontia was the number of missing permanent teeth

minus the number of retained deciduous teeth.


January 2010

additional comparison group of nonorthodontic subjectswith no malocclusion, such as unaffected family mem-bers, children attending their general dental practitioneror hospital pedodontic clinics, or local schoolchildren,but this was not the purpose of this study.15

There is, as yet, no gold standard OHRQOL mea-sure for children or adults.10 The OHRQOL measuresavailable for children at the time of this study werethe child-oral impacts on daily performances and theCPQ.13,17 The CPQ was chosen for this study becauseit was specifically designed to investigate the impactof oral conditions during early adolescence, which ischaracterized by the increasingly important role ofpeer groups and preoccupation with others’ view ofself.18 Additionally, at the time of this study, the CPQhad been validated for use in a population in the UnitedKingdom, whereas the child-oral impacts on dailyperformances tool had not. The CPQ was originallyvalidated for children between 11 and 14 years ofage13; however, patients 11 to 16 years of age wererecruited for this study because it was thought that15- and 16-year-olds would have just as good, or better,comprehension of the questionnaire’s components,since cognitive development is age-dependent.18 In ad-dition, many hypodontia patients come to orthodonticclinics an these ages. It is accepted that this mightmake our study more difficult to compare with otherstudies using the CPQ that recruited 11- to 14-year-old patients, but the benefits of this were considered tooutweigh the drawbacks.

Measures such as the CPQ are useful for compari-sons across populations but have limited ability to cap-ture the effects of a particular condition because there isas yet no condition-specific measure for hypodontia.19

Hence, we used a general oral health measure in this

study. A shortcoming of the CPQ is its length, althougha 100% response rate was achieved in this study, and allquestionnaires were completed correctly.20 The short-ened versions of the CPQ might be more useful in futureresearch, once all psychometric properties of theseforms have been tested.20 VAS were used in this studyto supplement the questionnaire findings, by quantify-ing appearance and function (eating), which are subjec-tive parameters. By using continuous scales, VASvalues overcame some shortcomings of categoricalscales, and they are sensitive to small changes in theparameter investigated.21

There were no statistically significant differencesbetween the hypodontia and the routine orthodonticgroups for overall CPQ scores (P 5 0.566), any domainscore (P 5 0.227; P 5 0.904; P 5 0.477; P 5 0.983)(Table II), or VAS scores for eating and appearance(P 5 0.766 and P 5 0.869), respectively (Table III).One might therefore conclude that the 2 groups hadno differences in psychosocial impact. However, it isunknown whether the impact experienced by bothgroups was high or low relative to children without a sig-nificant malocclusion. A weakness of cross-sectionalstudies such as this is that associations are investigatedand not causality, so the reasons for the findings are notknown. Additionally, a low, medium, or high CPQ scorehas not been formally quantified to date, so it is un-known what total CPQ score would imply a considerableimpact. Jokovic et al,13 who compiled the CPQ, did notcategorize a particular total score as having a specificdegree of impact.

Similar studies suggest that hypodontia negativelyaffects patients. For example, Johal et al22 undertooka prospective cross-sectional study to assess the impactof increased overjets and spaced dentitions on the

Table V. Univariable linear regression analyses to establish the effect of the extent of hypodontia on the total CPQscores, functional limitations domain scores, and VAS scores

Dependent variable Independent variable (predictor) ß 95% CI P value

Total CPQ questionnaire score Age (per year) 1.518 –1.157, 4.193 0.261

Sex (female) 6.022 –2.113, 14.157 0.144

Total number of missing teeth (absolute hypodontia) 0.398 –0.840, 1.636 0.523

Total relative hypodontia 1.518 –0.795, 3.831 0.194

Total absolute hypodontia, maxillary middle sextant –1.606 –5.478, 2.265 0.410

Total relative hypodontia, maxillary middle sextant –2.167 –7.163, 2.829 0.389

Total functional limitations domain score of CPQ Total absolute hypodontia 0.332 –0.020, 0.684 0.064

Total relative hypodontia 0.731* 0.073, 1.389* 0.030*

Number of missing teeth, maxillary middle sextant –0.479 –1.608, 0.649 0.399

Total relative hypodontia, maxillary middle sextant –1.040 –2.482, 0.401 0.154

VAS score for function (eating) Total relative hypodontia 2.809* 0.533, 5.084* 0.016*

VAS score for appearance Total relative hypodontia, maxillary middle sextant –1.571 –9.233, 6.092 0.686

For this publication, only the regression for the functional limitations domain was included.

*Statistically significant findings.


quality of life of 180 children (ages, 13-15 years) andtheir parents, and found that both occlusal traits hada significant negative impact on the children’s and par-ents’ quality of life compared with the control subjects.Wong et al,16 in a recent cross-sectional study of OHR-QOL and severe hypodontia in Hong Kong, concludedthat severe hypodontia considerably impacts OHRQOL,although no control group was used. The ‘‘considerableimpact’’ was derived from a mean CPQ score for the se-vere hypodontia group of 29.0 (SD, 16.4) of a potentialmaximum score of 148. In the current study, the meanCPQ score for the mild, moderate, and severe hypodon-tia combined was 26.82 (SD, 16.03) of a potential max-imum score of 148. Therefore, in this study, if onlypatients with severe hypodontia had been included,this may have caused the mean CPQ score to havebeen higher. If Wong et al16 had used a control groupto compare the results of the hypodontia group, theymight have reached different conclusions.

A number of reasons might account for why thehypodontia group did not experience a greater psy-chosocial impact than did the routine orthodonticgroup. It may be that the routine orthodontic subjectswere equally psychosocially affected by their denti-tion, although for different reasons, and this is worthyof future research. Based on Brook’s 1974 classifica-tion,23 the ‘‘average’’ hypodontia found in this study(4.52; SD, 3.33) was moderate (3-5 missing teeth),and this severity of hypodontia could have been toomild for the patients to have been affected psychoso-cially by it. Also, adjacent teeth might have eruptedinto the space left by the missing tooth, giving the il-lusion that the dentition was intact; the effects of hy-podontia might have been camouflaged well andappeared as just mild spacing; and spacing of the den-tition is commonly found in the late mixed dentitionas the rest of the deciduous teeth—eg, the canines—are exfoliated. Thus, some subjects in the hypodontiagroup might not have been affected by visible spacingof the dentition and were unaware of permanentlymissing teeth.

In the analysis of the hypodontia group, univariablelinear regression analyses showed no statistical evi-dence of age, sex, and extent and location of hypodontiaaffecting overall CPQ scores (Table V). At the domainlevel, no effect was seen, except for the functional lim-itations domain that was significantly affected by the to-tal relative hypodontia (P 5 0.030) (Table V). Thisimplied some evidence that, as relative hypodontia in-creased, the subject’s functional impairment signifi-cantly worsened. In other words, as deciduous teethwere lost and not replaced by permanent teeth, the asso-ciated spacing increased, leading to functional limita-

tions with eating. This gives some tentative evidencethat retention of deciduous teeth would be beneficialfor hypodontia patients, at least in terms of function.The P values for the maxillary middle sextant werenot significant (P 5 0.399 and P 5 0.154), suggestingthat hypodontia in this region did not cause functionallimitations (Table V). This implies that functional limi-tations might be predominantly limited to the teeth inthe posterior sextants so that, for example, a missingpremolar would make chewing with the posterior teethmore difficult than if a maxillary lateral incisor weremissing.

The regression analyses for the VAS scores furtherinvestigated this and showed evidence of a statisticallysignificant relationship between the VAS scores foreating and total relative hypodontia (P 5 0.016)(Table V). This agrees with the findings of the CPQfunctional limitations domain regression analysis(Table V). There was no evidence of a relationshipbetween the VAS scores for appearance and relativehypodontia of the maxillary middle sextant (P 5

0.686). Either hypodontia in this area does not affectpatients psychosocially or the extent of hypodontiain this sample was too mild or too well camouflagedto affect patients psychosocially.

CONCLUSIONS

1. In this sample, there was no difference in the psy-chosocial status of patients with hypodontia com-pared with patients with other features of anIOTN, DHC 4 or 5 malocclusion, but not affectedby hypodontia.

2. There was also no evidence to suggest that age orsex has any influence on the psychosocial statusof either group of patients.

3. There was some statistical evidence to suggest thatrelative hypodontia has an impact on the functionalabilities of the hypodontia patients and that reten-tion of deciduous teeth is beneficial from a func-tional point of view.

4. This study has implications in understanding theimpact of hypodontia on quality of life. Measure-ment of treatment need and effectiveness should in-clude not only normative assessment but alsopsychological and social dimensions.

We thank Aleksandra Jokovic and his colleagues forpermitting us to use the CPQ, all patients and their par-ents or legal guardians who participated in the study,and the clinicians who allowed us to recruit patientsfrom their clinics.


January 2010

REFERENCES

1. Goodman JR, Jones SP, Hobkirk JA, King PA. Hypodontia: clin-

ical features and the management of mild to moderate hypodontia.

Dent Update 1994;21:381-4.

2. Mattheeuws N, Dermaut L, Martens G. Has hypodontia increased

in Caucasians during the 20th century? A meta-analysis. Eur J Or-

thod 2004;26:99-103.

3. Brabant H. Comparison of the characteristics and anomalies of the

deciduous and permanent dentition. J Dent Res 1967;46:897-902.

4. Arte S, Nieminen P, Apajalahti S, Haavikko F, Thesleff I,

Pirinen S. Characteristics of incisor-premolar hypodontia in fam-

ilies. J Dent Res 2001;80:1445-50.

5. Brook AH, Elcock C, Al-Sharood MH, McKeown HF, Khalaf K,

Smith RN. Further studies of a model for the aetiology of anom-

alies of tooth number and size in humans. Connect Tissue Res

2002;43:289-95.

6. Hobkirk JA, Goodman JR, Jones SP. Presenting complaints and

findings in a group of patients attending a hypodontia clinic. Br

Dent J 1994;177:337-9.

7. Langlois JH, Kalakanis LE, Rubenstein AJ, Larson AD,

Hallam MJ, Smoot MT. Maxims or myths of beauty: a meta-ana-

lytic and theoretical overview. Psychol Bull 2000;126:390-423.

8. Shaw WC, O’Brien KD, Richmond S, Brook P. Quality control in

orthodontics: risk/benefit considerations. Br Dent J 1991;170:

33-7.

9. O’Brien K, Wright JL, Conboy F, Macfarlane T, Mandall N. The

child perception questionnaire is valid for malocclusions in the

United Kingdom. Am J Orthod Dentofacial Orthop 2006;129:

536-40.

10. McGrath C, Broder H, Wilson-Genderson M. Assessing the im-

pact of oral health on the life quality of children: implications

for research and practice. Community Dent Oral Epidemiol

2004;32:81-5.

11. Cunningham SC, Hunt NP. Quality of life and its importance in

orthodontics. J Orthod 2001;28:152-8.

12. Eiser C, Morse R. The measurement of quality of life in children:

past and future perspectives. J Dev Behav Pediatr 2001;22:

248-56.

13. Jokovic A, Locker D, Stephens M, Kenny D, Tompson B, Guyatt G.

Validity and reliability of a questionnaire for measuring child oral-

health related quality of life. J Dent Res 2002;81:459-63.

14. Marshman Z, Rodd H, Stern M, Mitchell C, Locker D, Jokovic A,

et al. An evaluation of the child perceptions questionnaire in the

UK. Community Dent Health 2005;22:151-5.

15. Coffield K, Phillips C, Brady M, Roberts M, Strauss R, Wright T.

The psychosocial impact of developmental defects in people with

hereditary amelogenesis imperfecta. J Am Dent Assoc 2005;136:

620-30.

16. Wong AT, McMillan AS, McGrath C. Oral health-related quality

of life and severe hypodontia. J Oral Rehabil 2006;33:869-73.

17. Gherunpong S, Tsakos G, Sheiham A. Developing and evaluating

an oral health-related quality of life index for children; the Child

OIDP. Community Dent Health 2004;21:161-9.

18. Hetherington EM, Parke RD, Locke VO. Child psychology: a con-

temporary viewpoint. New York: McGraw-Hill; 1999.

19. Cunningham SJ, Garratt A, Hunt NP. Development of a condition-

specific quality of life measure for patients with dentofacial defor-

mity: I. Reliability of the instrument. Community Dent Oral Epi-

demiol 2000;28:195-201.

20. Jokovic A, Locker D, Guyatt G. Short forms of the child perceptions

questionnaire for 11-14-year-old children (CPQ11-14): develop-

ment and initial evaluation. Health Qual Life Outcomes 2006;4:4.

21. Le Resche L, Burgess J, Dworkin SF. Reliability of visual analog

and verbal descriptor scales for ‘‘objective’’ measurement of tem-

poromandibular disorder pain. J Dent Res 1988;67:33-6.

22. Johal A, Cheung MY, Marcene W. The impact of two different

malocclusion traits on quality of life. Br Dent J 2007;202:E6.

23. Brook AH. Dental anomalies of number, form and size: their prev-

alence in British schoolchildren. J Int Assoc Dent Child 1974;5:

37-53.


ORIGINAL ARTICLE

Association of orthodontic treatment needs andoral health-related quality of life in young adults

Ali H. Hassana and Hatem El-Sayed Aminb

Jeddah, Saudi Arabia, and Tanta, Egypt

Introduction: Our objective was to assess the effect of different orthodontic treatment needs on the oralhealth-related quality of life of young adults. Methods: The study sample comprised 366 young adult ortho-dontic patients (153 men, 213 women; age range, 21-25 years). Each participant was assessed for orthodontictreatment need and oral health-related quality of life by using the dental health component of orthodontic treat-ment need index and the shortened version of oral health impact profile questionnaire. Results: Orthodonticpatients who had little or no, borderline, and actual need for orthodontic treatment represented 14.8%, 56%,and 29.2% of the total sample, respectively. Orthodontic treatment need significantly affected mouth aching,self-consciousness, tension, embarrassment, irritability, and life satisfaction in both sexes. Also, orthodontictreatment need significantly affected taste and relaxation in both men and women. However, pronunciationand the ability to do jobs or function effectively were not significantly associated with orthodontic treatmentneeds in either sex. Conclusions: These findings emphasize the impact of malocclusion on oral health-relatedquality of life of young adults. (Am J Orthod Dentofacial Orthop 2010;137:42-7)

Traditionally, clinician-based outcome measureswere more important for dental researchersthan subjective patient-based measures such as

perceived functional status and psychological well-be-ing.1 However, patients and dentists differ in their eval-uation of oral health and the perception of oraldiseases.2,3 Recently, researchers and clinicians havefocused more on patients’ own perceptions of oralhealth status and oral health care systems to understandtheir needs, satisfaction with treatment, and ultimatelythe perceived overall quality of health systems.4,5

Oral diseases, including malocclusion, are highlyprevalent, and the consequences are physical, econom-ical, social, and psychological.4 They can impair thequality of life in many people and affect various aspectsof life, including function, appearance, and interper-sonal relationships.6 Therefore, considering the oralcavity as an autonomous landmark is now being ques-tioned and more emphasis is placed on how the oral

conditions affect health, well-being, and quality oflife.1 According to the concept of oral health-relatedquality of life (OHRQOL), good oral health is no longerseen as the mere absence of oral diseases and dysfunc-tion. OHRQOL encompasses the absence of negativeimpacts of oral conditions on social life and a positivesense of dentofacial self-confidence.7

Understanding the physical, social, and psychologi-cal impact of malocclusion on OHRQOL needs moreattention, since it sheds light on the effects of malocclu-sion on people’s lives and provides more understandingof the demand for orthodontic treatment beyond clini-cian parameters.8 In addition, since social and psy-chological effects are the key motives for seekingorthodontic treatment, OHRQOL can be consideredthe best measurement for orthodontic treatment needand outcome.9 Therefore, OHRQOL measurement isrecommended for orthodontists to supplement clinicalfindings, since OHRQOL outcome does not necessarilycorrelate with such objective findings.4

Several indexes were used to evaluate malocclusion.The index of orthodontic treatment need (IOTN) isa scoring system for malocclusion, developed by Brookand Shaw.10 It includes 2 independent components: thedental health component (DHC), a 5-grade index thatrecords the dental health need for orthodontic treatment,and the esthetic component that records the estheticneed for orthodontic treatment.2,7,8 The IOTN hasbeen used extensively in the literature to evaluate actualand perceptive orthodontic treatment needs.10-14 TheDHC grades patients’ treatment needs either as no

aAssociate professor, Department of Orthodontics, Faculty of Dentistry; direc-

tor, Saudi Board in Orthodontics Local Supervising Committee; supervisor,

Student Affairs, King Abdulaziz University, Jeddah, Saudi Arabia.bAssociate professor, Pedodontics and Dental Public Health Department, Tanta

University, Tanta, Egypt.



Reprint requests to: Ali H. Hassam, P.O. Box 80209, Jeddah 21589, Saudi Ara-

bia; e-mail, [email protected].

Submitted, October 2007; revised and accepted, February 2008.

0889-5406/$36.00


doi:10.1016/j.ajodo.2008.02.024

42


treatment need, little treatment need, borderline need, ortreatment required.

The oral health impact profile (OHIP) is an exten-sively used instrument for the assessment of OHR-QOL.15,16 The original version of the scale includes49 items divided into 7 domains. A short form of theOHIP containing only 14 items (OHIP-14) has been de-veloped.17 The OHIP is designed to determine the per-ception of the social impact of oral disorders and haswell-documented psychometric properties.15,18

The aim of this study was to assess the effect of dif-ferent orthodontic treatment needs on the OHRQOL ofyoung adults.


A cross-sectional study was conducted of orthodonticpatients to assess the relationship between orthodontictreatment needs assessed by the DHC of the IOTN andOHRQOL assessed by the OHIP-14 questionnaire.17

A consecutive sample of young adults seekingorthodontic treatment at the Faculty of Dentistry, KingAbdulaziz University, were recruited in the study ac-cording to the order of registration on the waiting list.Patients who had a perceived need for orthodontic treat-ment and who were about to undergo orthodontictherapy were included. Exclusion criteria were chronicmedical conditions, previous orthodontic treatment,craniofacial anomalies such as cleft lip and palate, un-treated dental caries, and poor periodontal health statusas indicated by a community periodontal index score of3 or more.19 This was to prevent possible confoundingeffects of these conditions on the participants’ qualityof life. After screening, the sample comprised 366orthodontic patients (153 men, 213 women) from 21to 25 years of age who were willing to participate inthe study.

Ethical approval was obtained at the beginning ofthe study. The participants were informed about the ex-amination procedures and were assured of the confiden-tiality of the collected information. Only those whogave consent were included in the research.

Each patient was examined for orthodontic treat-ment need with the DHC of the IOTN. Examinerswere calibrated to use it (kappa, 8.5). Treatment needsof the patients were categorized as (1) little or no treat-ment need, (2) borderline need, and (3) treatmentrequired. The DHC uses a simple ruler and anacronym—MOCDO (missing teeth, overjet, crossbite,displacements of contact points, overbite)—to identifythe most severe occlusal trait for each patient. The finaloverall score was given to the patient according to themost severe trait.10

The data collection instrument for assessment ofOHRQOL was the OHIP-14 questionnaire.17 The ques-tionnaires were administered by the examiners beforethe clinical examination. Each patient was asked aboutthe frequency that he or she experienced an impact on14 daily activities. Responses were made on a 5-pointLickert-type scale (never, hardly ever, occasionally,fairly often, and very often). A threshold of occasion-ally, fairly often, and very often was used to dichoto-mize responses, thereby indicating participants whohad experienced at least some oral health impact.

The daily activities were the following: had prob-lems pronouncing words, felt that the sense of tasteworsened, had painful aching in the mouth, found it un-comfortable to eat any food, have been self-conscious,felt tense, had an unsatisfactory diet, had to interruptmeals, found it difficult to relax, have been a bit embar-rassed, have been irritable with other people, had diffi-culty doing useful jobs, felt that life in general wasless satisfactory, and have been totally unable tofunction.


Data presentation and statistical analysis were per-formed with the SPSS statistical package (version 13,SPSS, Chicago, Ill). The chi-square test was used to an-alyze the qualitative data. The level of significance was0.05.

RESULTS

Table I shows that men and women were 41.8% and58.2% of our sample, respectively. The mean age of thetotal sample was 23 years. In this study, patients whohad little or no, borderline, and need for orthodontictreatment were 14.8%, 56%, and 29.2%, respectively.The corresponding percentages were 13.7%, 54.2%,and 32.1% in the men and 15.5%, 57.3%, and 27.2%in the women, respectively.

In Table II, the chi-square test shows that pronunci-ation was not significantly affected by the need fororthodontic treatment in either men (c2 5 2.6; P 5 0.2)or women (c2 5 1.11; P 5 0.5). Taste, however, wassignificantly affected by the level of orthodontic treat-ment need in men (c2 5 6.9; P 5 0.03) but not inwomen (c2 5 5.6; P 5 0.06).

Among the examined subjects, the proportions oforthodontic patients who found it uncomfortable to eatany food, had an unsatisfactory diet, and had to interrupttheir meals were significantly correlated with orthodon-tic treatment needs in both men (c2 5 11.9, 9.6, and 7.9;P 5 0.003, 0.008, and 0.01, respectively) and women(c2 5 7.8, 13.9, 11.3; P 5 0.02, 0.00, and 0.00,

American Journal of Orthodontics and Dentofacial Orthopedics Hassan and Amin 43Volume 137, Number 1

respectively). Also, in both male and female patients,the need for orthodontic treatment significantly affectedpainful mouth aching (c2 5 10.2 and 10.9; P 5 0.006 and0.00, respectively), self-consciousness (c2 5 16.4 and17.8; P 5 0.00), and feelings of tension (c2 5 12.8and 9.9; P 5 0.00). Relaxation was also significantlyassociated with the level of orthodontic treatmentneed in women (c2 5 6.8; P 5 0.03), but it did notreach the level of significance in men (c2 5 3.5; P 5 0.17).

Moreover, embarrassment, irritability with otherpeople, and the general feeling of less satisfaction inlife were significantly associated with higher orthodon-tic treatment needs in both men (c2 5 11.3, 16.7, and12.5; P 5 0.003, 0.00, and 0.00) and women (c2 5

10.1, 18.5, and 14.2; 0.00, 0.01, and 0.00). On the otherhand, orthodontic treatment needs did not significantlyaffect the ability of the patients to do their jobs or func-tion effectively (c2 5 3.8 and 2.07; P 5 0.15 and 0.35in men; c2 5 2.9 and 1.49; P 5 0.23 and 0.40 in women,respectively).

DISCUSSION

Clinicians are increasingly placing more emphasison patient-based evaluations of health-related qualityof life.20 This might be particularly important in cos-metic and elective treatments.21 Although it is generallyaccepted that malocclusion has physical and psycholog-ical consequences, there is still conflicting evidenceabout the extent of these effects. This could be due tothe different interpretations of what these impacts con-stitute and the lack of standardized approaches for as-sessment. Therefore, this study was conducted toassess the impact of orthodontic treatment needs onOHRQOL in orthodontic patients.

The OHIP questionnaire that we used has beenutilized in general populations and patients with certainoral disorders.22 More specifically, this questionnairewas used in assessing the impact of malocclusion onquality of life in several studies.16,17 The sensitivity

and specificity of this questionnaire have been evaluatedin both cross-sectional and longitudinal studies.18,23

Additionally, the Arabic version of OHIP was recentlytested in a convenience sample of Saudi people. Itsresponsiveness, reliability, and high internal consis-tency were confirmed.24 Similarly, the IOTN was previ-ously used in Saudi orthodontic patients.14

In this study, adolescent orthodontic patients werenot included because major life changes during adoles-cence affect their quality of life and make it difficult toidentify which daily activities are changed solely by theneed for orthodontic treatment.9 Thus, this study wasconfined to young adults, whose major life changeshave subsided.

Because quality of life is a relative rather than anabsolute measure, these results were expressed as a com-parison of the impacts on daily activities between ortho-dontic patients with different orthodontic treatmentneeds.25 Unexpectedly, women had nearly similar im-pacts of orthodontic treatment needs on their dailyactivities as did the male orthodontic patients. Thiswas in contradiction with the study of de Oliveria andSheiham,9 who reported that sex significantly affectsthe impact of malocclusion on OHRQOL, and womenwere 1.22 times more likely to have an impact than men.

Our results showed that orthodontic treatment needsdid not significantly affect speech and word pronuncia-tion. This confirmed the results of a previous study thatfound no association between speech problems and mal-occlusion.26 The nonassociation between orthodontictreatment need and pronunciation can be explained,since speech is a complex process that involves brain,teeth, lips, tongue, and muscles that can compensatemutually to ensure perceptually normal pronuncia-tion.16 Other researchers, however, observed a strongassociation between speech disorders and dentofacialabnormalities.26-28

However, in this study, taste, chewing ability, diet se-lection, and meal interruption were significantly affectedby the orthodontic treatment needs of the examined sub-jects. This confirmed other cross-sectional studies report-ing that subjects with malocclusion have less masticatoryefficiency compared with those with normal occlusion,suggesting that malocclusion can affect diet in terms oftaste and ability to chew.29-32 On the other hand, Danielsand Richmond33 reported that technical aspects of maloc-clusion such as dissatisfaction with the ability to chew areless likely to impact the quality of life among young adultsas more subjective aspects of dental esthetic and self-per-ception of dental appearance.

Also, the significant association between orthodon-tic treatment needs and oral pain observed in this studyagreed with previous studies reporting that

Table I. Characteristics and orthodontic treatment needof the study sample

Sample characteristics Men Women All combined

Number (%) 153 (41.8%) 213 (58.2%) 366 (100%)

Age (y)

Mean 24.08 22.27 23

SD 1.16 1.44 1.6

Treatment need

No or little

need (%)

21 (13.7%) 33 (15.5%) 54 (14.8%)

Borderline (%) 83 (54.2%) 122 (57.3%) 205 (56%)

Need (%) 49 (32.1%) 58 (27.2%) 107 (29.2%)

44 Hassan and Amin American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

malocclusion can cause pain indirectly by leading totemporomandibular disorders34,35 or increasing thelikelihood of trauma to proclined maxillary incisors.36

Additionally, retroclined maxillary incisors cause directtrauma to the labial gingiva of the mandibular incisorswith associated pain.37

In this study, embarrassment and self-conscious-ness, as a person’s intentional focus on his or her internal

sensations, were significantly correlated to orthodontictreatment needs. This agrees with Klages et al,38 whofound that young adults with more severe forms of mal-occlusion had higher self-consciousness scores. Also,these results paralleled the observation of Dion et al39

that self-consciousness is significantly affected by or-thodontic status. Moreover, our results coincided withseveral studies that reported that most patients who

Table II. Impacts on daily activities in relation to sex and orthodontic treatment needs

OHIP-14Daily activity

Orthodontic treatment need

No or littletreatment need Borderline treatment need

Treatmentneed

c2

P

Males21

Females33

Males83

Females122

Males49

Female58 Male Female

Had problem pronouncing words

Impact: n (%) 8 (38) 15 (45) 43 (52) 65 (53) 29 (59) 33 (57) 2.6 1.11

No impact: n (%) 13 (62) 18 (56) 40 (48) 57 (47) 20 (41) 25(43) 0.2 0.5

Felt sense of taste worsened

Impact: n (%) 9 (43) 11 (33) 52 (63) 69 (57) 37 (76) 31 (53) 6.9 5.6

No impact: n (%) 12 (57) 22 (67) 31 (37) 53 (43) 12 (24) 27 (47) .03* 0.06

Had painful aching in mouth

Impact: n (%) 8 (38) 9 (27) 55 (66) 70 (57) 38 (78) 35 (60) 10.2 10.9

No impact: n (%) 13 (62) 24 (73) 28 (34) 52 (43) 11 (22) 23 (40) .006* .00*

Found it uncomfortable to eat food

Impact: n (%) 5 (24) 14 (42) 50 (60) 80 (66) 33 (67) 41 (71) 11.9 7.8

No impact: n (%) 16 (76) 19 (58) 33 (40) 42 (34) 16 (33) 17 (29) .003* .02*

Have been self-conscious

Impact: n (%) 7 (33) 10 (30) 59 (71) 81 (66) 40 (82) 42 (72) 16.4 17.8

No impact: n (%) 14 (67) 23 (70) 24 (29) 41 (34) 9 (18) 16 (28) .00* .00*

Felt tense

Impact: n (%) 6 (29) 11 (33) 53 (64) 68 (56) 36 (74) 39 (67) 12.8 9.8

No impact: n (%) 15 (71) 22 (67) 30 (36) 54 (44) 13 (26) 19 (33) .00* .00*

Had an unsatisfactory diet

Impact: n (%) 8 (38) 10 (30) 57 (69) 79 (65) 37 (76) 38 (66) 9.6 13.9

No impact: n (%) 13 (62) 23 (70) 26 (31) 43 (35) 12 (24) 20 (34) .008* .00*

Had to interrupt meals

Impact: n (%) 9 (43) 12 (36) 53 (64) 71 (58) 38 (78) 42 (72) 7.9 11.3

No impact: n (%) 12 (57) 21 (64) 30 (36) 51 (42) 11 (22) 16(28) 0.01* .00*

Found it difficult to relax

Impact: n (%) 7 (33) 13 (41) 46 (55) 71 (58) 27 (55) 40 (69) 3.5 6.8

No impact: n (%) 14 (67) 20 (59) 37 (45) 51 (42) 22 (45) 18(31) 0.17 .03*

Have been a bit embarrassed

Impact: n (%) 8 (38) 12 (36) 53 (64) 70 (57) 39 (80) 41 (71) 11.3 10.1

No impact: n (%) 13 (62) 21 (64) 30 (36) 52 (43) 10 (20) 17 (29) .003* .00*

Have been irritable with people

Impact: n (%) 6 (29) 13 (39) 58 (70) 71 (58) 38 (78) 41 (71) 16.7 8.5

No impact: n (%) 15 (71) 20 (61) 25 (30) 51 (42) 11 (22) 17 (29) 0.00* .01*

Had difficulty doing useful jobs

Impact: n (%) 8 (38) 13 (39) 45 (54) 68 (56) 31 (63) 32 (55) 3.8 2.9

No impact: n (%) 13 (62) 20 (61) 38 (46) 54 (44) 18 (37) 26 (45) 0.15 0.23

Felt life in general less satisfactory

Impact: n (%) 6 (29) 10 (30) 55 (66) 72 (59) 35 (71) 41 (71) 12.5 14.2

No impact: n (%) 15 (71) 23 (70) 28 (34) 50 (41) 14 (29) 17 (29) 0.00* .00*

Have been unable to function

Impact: n (%) 9 (43) 18 (55) 49 (59) 78 (64) 25 (51) 39 (67) 2.07 1.49

No impact: n (%) 12 (57) 15 (45) 34 (41) 44 (36) 24 (49) 19 (33) 0.35 0.4

*Significant at 5% level; df 5 2.


need orthodontic therapy feel shameful and inferior, andthe higher the need for treatment, the greater the per-son’s embarrassment.40,41 Helm et al42 reported thatthe self-consciousness and embarrassment felt by ortho-dontic patients are not only displayed in adolescence,but also persist in adulthood. Other studies, however, re-ported nonsignificant associations between malocclu-sion and self-consciousness or embarrassment.43,44

Some cross-sectional and retrospective studies con-firmed our observation that young adults with highertreatment needs tended to be more socially deprivedthan those with lower treatment needs.40,44 Ironically,dental deformity elicits strong emotional reactions lead-ing to psychosocial problems including isolation anddepression.45 Additionally, facial appearance and thegeometric features of the face could, to a large extent,influence social activities and the success of interper-sonal relationships.46 Therefore, we found it not surpris-ing that orthodontic patients with clinically assessedgreater orthodontic needs reported more embarrassmentand irritability with other people than those with no orborderline orthodontic treatment needs.

These results support the assumption that orthodon-tic patients mainly suffer esthetic and social problemsrather than impairment of daily activties.9 This wasshown by the nonsignificant association between ortho-dontic treatment needs and orthodontic patients’ abilityto function and do their jobs. These results agree withthose of Albino et al,47 who reported that about 80%of orthodontic patients complain about esthetic ratherthan health and functional impacts.

In this study, the significant effect of orthodontictreatment needs on life satisfaction of orthodonticpatients agrees with the study of Kiyak et al,48 whofound that orthodontic patients thought that life ingeneral was less satisfying and viewed themselves lesspositively. This could be because orthodontic patientssuffer psychologically from dental and facial defor-mities with an associated decrease in self-confidencethat accompanies those changes.9

Some methodologic limitations must be consideredin the general relevance of these results. First, since ourobjective was to assess the impact of different orthodon-tic treatment needs on the quality of life of young adultorthodontic patients, our participants were orthodonticpatients with a perceived need for orthodontic treat-ment. However, they did not represent the entire youngadult population with varying levels of malocclusionand orthodontic treatment needs who might have differ-ent impacts on their daily activities. Second, the rele-vance of observations of young adults for olderpatients is limited, because the importance of physicalattractiveness in young adults appears to be greater.

Based on these results, it could be justified that anOHRQOL tool is recommended for assessing orthodon-tic treatment needs and consequently improving thequality of care. With the possibility that allocation ofresources in the future might be influenced by thesedata, our specialty can no longer afford to ignore theseconcepts.

CONCLUSIONS

These results highlight the impact of malocclusionon OHRQOL of young adults and emphasize the impor-tance of patient-based evaluation of oral health statusand oral health needs.

REFERENCES

1. Cunningham SJ, Hunt NP. Quality of life and its importance in

orthodontics. J Orthod 2001;28:152-8.

2. Ahmed B, Gilthorpe MS, Bedi R. Agreement between normative

and perceived orthodontic need amongst deprived multiethnic

school children in London. Clin Orthod Res 2001;4:65-71.

3. Hunt O, Hepper P, Johnston C, Stevenson M, Burden D. The aes-

thetic component of the index of orthodontic treatment need val-

idated against lay opinion. Eur J Orthod 2002;24:53-9.

4. Bedi R, Gulati N, McGrath C. A study of satisfaction with dental

services among adults in the United Kingdom. Br Dent J 2005;

198:433-7.

5. McGrath C, Bedi R. The value and use of ‘‘quality of life’’ mea-

sures in the primary dental care setting. Prim Dent Care 1999;6:

53-7.

6. Locker D. Concept of oral health, disease and the quality of life.

In: Slade GD, editor. Measuring oral health and quality of life.

Chapel Hill: University of North Carolina; 1997.

7. Inglehart M, Bagramiane R, editors. Oral health-related quality of

life. Chicago: Quintessence; 2002.

8. Zhang M, McGrath C, Hagg U. The impact of malocclusion and

its treatment on quality of life: a literature review. Int J Paediatr

Dent 2006;16:381-7.

9. de Oliveira CM, Sheiham A. Orthodontic treatment and its impact

on oral health-related quality of life in Brazilian adolescents. J

Orthod 2004;31:20-7.

10. Brook PH, Shaw WC. The development of an index of orthodontic

treatment priority. Eur J Orthod 1989;11:309-20.

11. Roberts EE, Beales JG, Dixon L, Willcocks AJ, Willmot DR. The

orthodontic condition and treatment status of a sample of 14-year-

old children in North Derbyshire. Community Dent Health 1989;

6:249-56.

12. Tulloch JF, Shaw WC, Underhill C, Smith A, Jones G, Jones M. A

comparison of attitudes toward orthodontic treatment in British

and American communities. Am J Orthod 1984;85:253-9.

13. Burden DJ, Pine CM, Burnside G. Modified IOTN: an orthodontic

treatment need index for use in oral health surveys. Community

Dent Oral Epidemiol 2001;29:220-5.

14. Hassan AH. Orthodontic treatment needs in the western region of

Saudi Arabia: A research report. Head Face Med 2006;2:2.

15. Slade GD, Spencer AJ. Development and evaluation of the oral

health impact profile. Community Dent Health 1994;11:3-11.

16. Slade GD. Assessing change in quality of life using the oral health

impact profile. Community Dent Oral Epidemiol 1998;26:52-61.

17. Slade GD. Derivation and validation of a short-form oral health

impact profile. Community Dent Oral Epidemiol 1997;25:284-90.

46 Hassan and Amin American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

18. John MT, Patrick DL, Slade GD. The German version of the oral

health impact profile—translation and psychometric properties.

Eur J Oral Sci 2002;110:425-33.

19. World Health Organization. WHO basic oral health survey. 4th ed.

Geneva, Switzerland: World Health Organization;1997.

20. Tsakos G, Gherunpong S, Sheiham A. Can oral health-related qual-

ity of life measures substitute for normative needs assessments in

11 to 12-year-old children? J Public Health Dent 2006;66:263-8.

21. Mandall NA, Vine S, Hulland R, Worthington HV. The impact of

fixed orthodontic appliances on daily life. Community Dent

Health 2006;23:69-74.

22. Murray H, Locker D, Mock D, Tenenbaum HC. Pain and the qual-

ity of life in patients referred to a craniofacial pain unit. J Orofa-

cial Pain 1996;10:316-23.

23. Allen PF, McMillan AS, Locker D. An assessment of sensitivity to

change of the oral health impact profile in a clinical trial. Commu-

nity Dent Oral Epidemiol 2001;29:175-82.

24. Ameer M, Szentpetery A. An Arabic version of oral health impact

profile: translation and psychometric properties. Int Dent J 2007;

57:84-92.

25. McGrath C, Bedi R. Population based norming of the UK oral

health related quality of life measure (OHQoL-UK). Br Dent J

2002;193:521-4.

26. Vallino LD, Tompson B. Perceptual characteristics of consonant

errors associated with malocclusion. J Oral Maxillofac Surg

1993;51:850-6.

27. Pahkala R, Laine T, Narhi M. Associations among different oro-

facial dysfunctions in 9-11-year-olds. Eur J Orthod 1995;17:

497-503.

28. Lee AS, Whitehill TL, Ciocca V, Samman N. Acoustic and per-

ceptual analysis of the sibilant sound /s/ before and after orthog-

nathic surgery. J Oral Maxillofac Surg 2002;60:364-72.

29. English JD, Buschang PH, Throckmorton GS. Does malocclusion

affect masticatory performance? Angle Orthod 2002;72:21-7.

30. Owens S, Buschang PH, Throckmorton GS, Palmer L, English J.

Masticatory performance and areas of occlusal contact and near

contact in subjects with normal occlusion and malocclusion.


31. Onyeaso CO, Aderinokun GA. The relationship between dental

aesthetic index (DAI) and perceptions of aesthetics, function

and speech amongst secondary school children in Ibadan. Nigeria.

Int J Paediatr Dent 2003;13:336-41.

32. Hollister MC, Weintraub JA. The association of oral status with

systemic health, quality of life, and economic productivity. J

Dent Educ 1993;57:901-12.

33. Daniels C, Richmond S. The development of the index of com-

plexity, outcome and need (ICON). J Orthod 2000;27:149-62.

Erratum in: J Orthod 2002;29:81.

34. Koroluk LD, Tulloch JF, Phillips C. Incisor trauma and early treat-

ment for Class II Division 1 malocclusion. Am J Orthod Dentofa-

cial Orthop 2003;123:117-25.

35. Shulman JD, Peterson J. The association between incisor trauma

and occlusal characteristics in individuals 8-50 years of age. Dent

Traumatol 2004;20:67-74.

36. Nicolau B, Marcenes W, Sheiham A. The relationship between

traumatic dental injuries and adolescents’ development along

the life course. Community Dent Oral Epidemiol 2003;31:

306-13.

37. Geiger AM. Malocclusion as an etiologic factor in periodontal

disease: a retrospective essay. Am J Orthod Dentofacial Orthop

2001;120:112-5.

38. Klages U, Bruckner A, Zentner A. Dental aesthetics, self-aware-

ness, and oral health-related quality of life in young adults. Eur J

Orthod 2004;26:507-14.

39. Dion KL, Dion KK, Keelan J. Appearance anxiety as a dimension

of social- evaluation anxiety: exploring the ugly duckling syn-

drome. Contemp Soc Psychol 1990;14:220-4.

40. Zhou YH, Hagg U, Rabie AB. Concerns and motivations of skel-

etal Class III patients receiving orthodontic-surgical correction.

Int J Adult Orthod Orthognath Surg 2001;16:7-17.

41. Zhou Y, Hagg U, Rabie AB. Severity of dentofacial deformity, the

motivations and the outcome of surgery in skeletal Class III

patients. Chin Med J (Engl) 2002;115:1031-4.

42. Helm S, Kreiborg S, Solow B. Psychosocial implications of mal-

occlusion: a 15-year follow-up study in 30-year-old Danes. Am J

Orthod 1985;87:110-8.

43. DiBiase AT, Sandler PJ. Malocclusion, orthodontics and bullying.

Dent Update 2001;28:464-6.

44. Lazaridou-Terzoudi T, Kiyak HA, Moore R, Athanasiou AE,

Melsen B. Long-term assessment of psychologic outcomes

of orthognathic surgery. J Oral Maxillofac Surg 2003;61:

545-52.

45. Kumpulainen K, Rasanen E. Children involved in bullying at el-

ementary school age: their psychiatric symptoms and deviance

in adolescence. An epidemiological sample. Child Abuse Negl

2000;24:1567-77.

46. Hawker DS, Boulton MJ. Twenty years’ research on peer victim-

ization and psychosocial maladjustment: a meta-analytic review

of cross-sectional studies. J Child Psychol Psychiatry 2000;41:

441-55.

47. Albino JE, Cunat JJ, Fox RN, Lewis EA, Slakter MJ, Tedesco LA.

Variables discriminating individuals who seek orthodontic treat-

ment. J Dent Res 1981;60:1661-7.

48. Kiyak HA, Hohl T, West RA, McNeill RW. Psychologic changes

in orthognathic surgery patients: a 24-month follow up. J Oral



ORIGINAL ARTICLE

Influence of palatal expanders on oral comfort,speech, and mastication

Nanci L. Oliveira De Felippe,a Adriana C. Da Silveira,b Grace Viana,c and Bonnie Smithd

Edina, Minn, Austin, Tex, and Chicago, Ill

Introduction: It is not known whether the design of the expander has an effect on initial adaptation, comfortlevel, speech, chewing, and swallowing, or whether age is a crucial aspect when dealing with speech adap-tations. The objectives of this study were to assess whether patients of different age groups undergoing palatalexpansion with various types of expanders experienced discomfort, speech impairment, chewing difficulty,and swallowing disturbances. Methods: A questionnaire was developed and distributed to patients whohad received palatal expanders in the preceding 3 to 12 months. Results: Regardless of the type of expander,most patients initially felt oral discomfort, and had problems with speech and mastication. However, these dis-turbances were confined to the first week after cementation of the device. Remarkable adaptation to thedevice in all aspects studied was observed by the end of the first week. In addition, age did not influencethe variables; younger patients and older teenagers responded similarly to the survey. In addition, the ques-tionnaire responses did not appear to be related to the respondents’ sex. Conclusions: Discomfort might notbe a deciding variable when choosing an appliance. Instead, clinicians should base their decision on factorssuch as its biomechanics. (Am J Orthod Dentofacial Orthop 2010;137:48-53)

Most adolescents consider orthodontic treat-ment a right of passage and something thatis socially desirable during this time in their

life. However, along with the benefits of achievinga beautiful smile, orthodontic patients can experiencemany problems. For example, Stewart et al1 accuratelystated that ‘‘orthodontic appliances must be interpretedas foreign bodies inserted in an important, and sensitive,area of the body.’’ Brackets and wires often cause pres-sure and ulceration of the mucosa; lingual appliancescause displacement of the tongue; palatal appliancescause a feeling of constraint; and fixed and removabledevices interfere with speech, swallowing, and chewing,not to mention the generalized dental soreness and paincaused by local inflammation that ultimately results intooth movement.

One might ask whether orthodontic patients fullyunderstand what to expect. It is well known that clinicalmanagement is more efficient when all explanations are

given, with anticipatory guidance provided before theappliances are placed.1-4 Moreover, information shouldbe tailored to suit the type of treatment planned, shouldbe accompanied by motivations, and should includea discussion of the effects of noncompliance.

In addition, few studies have monitored the long-term effects of orthodontic appliances on swallowing,1,4

comfort,1,5-10 and psychosocial aspects such as feelingsof embarrassment.1,4,11 Generally, current evidencesuggests that most negative effects of appliance wearbecome more tolerable with time, but most investiga-tions in this field have focused on pain and speech.

Speech production can be affected by any osseous,muscular, dental, or soft-tissue deformity or any deviceimpairing the movement or appearance of the speechsound articulators. For example, Johnson and Sandy12

evaluated malocclusion and abnormal tooth position inrelation to articulation problems and concluded that,because of the potential for mechanism compensationsand the ability of a motor act to adapt to changing land-marks, many patients achieve normal speech despiteabnormal tooth position. Similarly, dental appliancesmight cause articulatory production errors of linguoden-tal, labiodental, or linguoalveolar consonants, but thesedisruptions are minimized after a short period of wearbecause of functional adaptation.13

The effects of orthodontic appliances on speechhave been extensively studied during the pastdecades.14-16 Functional appliances have been ratedthe most deleterious to speech,4,15,17,18 followed bymaxillary retainers,14-16 lingual fixed appliances,8 and

aPrivate practice, Edina, Minn.bChief of orthodontics, Craniofacial Center, Dell Children’s Medical Center of

Central Texas, Austin, Tex.cBiostatistician, Department of Orthodontics, University of Illinois at Chicago.dProfessor Emerita of Speech Pathology, University of Illinois, Chicago, IL;

private practice, Murdock, Fla.



Reprint requests to: Nanci L. O. De Felippe, 5101 Vermon Ave South, Suite 502,

Edina, MN 55436; e-mail, [email protected].

Submitted, December 2007; revised and accepted, January 2008.

0889-5406/$36.00


doi:10.1016/j.ajodo.2008.01.023

48


lingual retainers.16 Usually, the impairment is tempo-rary, varying from a few days to a few weeks; however,Stewart et al1 reported speech problems up to 3 months.Strutton and Burkland15 tested the effects of various de-signs of maxillary retainers on the clarity of speech atinitial placement and concluded that Crozat type andmodified horseshoe type retainers are superior to the tra-ditional Hawley design.

With respect to speech problems, the mostfrequently affected sound is the /s/. In 1986, Laine19

studied palatal appliances and /s/ production and foundthat the narrower the palate, the greater the distortion ofthe /s/ sound because the appliance added a physicalbarrier to the palate and, therefore, diminished thetongue’s functional space. Haydar et al16 used audiotaperecordings of 15 patients to evaluate speech on the firstday of wearing the appliance, at 24 hours, and 1 weeklater. They concluded that, on the day of placement,there were statistically significant distortions of the/t,d/ and /k,g/ sounds with maxillary retainers, and the/t,n/, /k,g/, and /s,j/ sounds with maxillary and mandib-ular removable retainers. They found, at 24 hours ofappliance wear, improvement of the /d/ and /k,g/ soundswith the maxillary retainer and of /t,n/ and /s/ with bothretainers. At 1 week, they found that no significantsound problems remained. Hohoff et al8 evaluatedspeech and bonded lingual appliances. They concludedthat these appliances caused speech problems, espe-cially with the /s/ sound, and that smaller lingual appli-ances caused less-pronounced speech impairments.

Currently, there is little information about theeffects of palatal expanders on speech and masticatoryfunction, overall comfort of wear, chewing, and swal-lowing functions. Specifically, it is not known whetherthe design of the expander has an effect on initial adap-tation, comfort level, speech, chewing, and swallowing,or whether age is a crucial aspect in speech adaptations.Therefore, the objectives of this study were to assesswhether patients of different age groups undergoingpalatal expansion with various types of expandersexperienced discomfort, speech impairment, chewingdifficulty, and swallowing disturbances.

In addition, our goal was to gather sufficient infor-mation to create future educational guidelines to helpexpedite patients’ adaptation to palatal expanders.


A questionnaire was developed, comprising 23questions, to gather the data for this study. This ques-tionnaire was distributed to patients who had receivedpalatal expanders in the last 3 to 12 months from severalprivate practices and an orthodontic graduate clinic. Of

the 23 questions, 7 pertained to oral comfort (section 1),6 to speech (section 2), 5 to chewing (section 3), and 5 toswallowing (section 4). The subjects’ responses werefurther categorized according to their appliances(Fig 1).

The development of the questionnaire was based onprevious published research.1,4,7,8 The study’s protocolwas reviewed and approved by the institutional reviewboard of the University of Illinois at Chicago. Subjectand parental consents were obtained. Instructions weregiven to the subjects, and all their questions wereanswered before administration of the questionnaire.

The subjects were asked to read the questionnaireitems and circle the response that most closely agreedwith how they experienced this item. The questionsand responses, in a multiple-choice format, were keptto a minimum to ensure optimal compliance. Examplesof the questions are shown in Figure 2.

A total of 165 questionnaires were distributed andcompleted. Table I shows the demographics of thesample. Table II provides descriptive statistics for thesample. Cross-tabulation and appropriate statisticalanalysis with the Pearson chi-square test with the signif-icance set at the 0.05 level were carried out.

RESULTS

Not all of the 165 questionnaires contained all infor-mation requested. For example, only 146 questionnaireslisted sex. Of these, 60 were male (41.4%), and 86 werefemale (58.9%). Of the 132 questionnaires that listedage, 77 (58.3%) were in the age group between 11and 14 years; 28 (21.2%) were between 15 and 18 yearsold; 24 (18.2%) were between 7 and 10 years old; and 3(2.3%) were between 19 and 22 years old.

As noted earlier, section 1 of the questionnaire dealtwith the physical traits of oral discomfort. Of the 163respondents, 93.9% stated that their appliance generatedpain and discomfort, especially in the first few days oftreatment. There was no statistically significant correla-tion between the degree of discomfort and the type ofappliance. Figure 3 gives the percentages of discomfortreported for each appliance. These results also show nostatistically significant correlation between the demo-graphic data (sex and age) and the responses aboutdiscomfort. Both sexes of all ages reported pain in thebeginning of expansion regardless of the type ofexpander used.

Section 2 evaluated the effects of the expander onthe articulation of speech. Ninety-eight percent of therespondents completed this section of the questionnaire.The results showed that 89.4% of this group stated thatthe expander affected their speech. No statistically

American Journal of Orthodontics and Dentofacial Orthopedics De Felippe et al 49Volume 137, Number 1

significant correlation was found with sex, age, or typeof appliance. In addition, it was impossible from theirresponses to determine which phonemes were statisti-

cally significantly affected. However, the alveolarsounds—/t/, /d/, /s/, and /z/—were reported by 13.3%as problematic. Also, in general, speech problems

Fig 2. Examples of questions asked in each section of the survey.

Fig 1. Types of expanders included in the study: A, Hyrax; B, Haas; C, bonded; D, quad-helix.

50 De Felippe et al American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

resolved by the end of the first week after cementationof the expander.

Sections 3 and 4 dealt with chewing and swallow-ing, respectively. Almost all subjects completed thesesections. Most respondents (90.2%) stated that theexpander interfered with chewing, and 67.9% reportedthat it affected swallowing. These findings, however,were not correlated with sex, age, or type of appli-ance.

In our sample, the Haas expander was the mostfrequently used, accounting for 36.0% (Table I) of thesurveyed population. The most respondents were inthe age group of 11 to 14 years (58.3%). All types ofappliances studied had similar effects on discomfort,speech, and mastication (P .0.05).

DISCUSSION

Our questionnaire can be used by clinicians toreview patients’ adaptations to palatal expandersregarding oral discomfort, speech, and mastication ineach patient’s perception. For example, appliancedesign is often selected based on the clinician’s personalpreference and patient comfort.20 However, our studyresults showed that all investigated appliances causedsimilar levels of discomfort for the patients. Therefore,discomfort might not be a deciding variable whenchoosing an appliance but, rather, its biomechanics ina given situation.

Our sample showed that, regardless of the type ofexpander, most patients initially felt oral discomfort,and had problems with speech and mastication.However, these disturbances were confined to the first

week after cementation of the device. Remarkable adap-tation to the device in all aspects studied was observedby the end of the first week. In addition, age did notinfluence the variables; younger patients and older teen-agers responded similarly to our survey, and responsesdid not appear to be related to sex.

No prior studies have shown how patients perceiveoral comfort in relation to the kind of palatalexpanders. Sergl et al4 assessed feelings of oral con-straint and lack of confidence in relation to removable,functional, or fixed appliances. They concluded thatsignificantly fewer complaints can be expected be-tween 2 and 7 days after placement of the appliance.Pain and discomfort of moderate levels were reportedby 93.9% of our subjects. Fortunately, these disap-peared after the first week, in agreement with the studyof Stewart et al.1

It has been hypothesized that dentures and ortho-dontic appliances interfere with speech production. Inparticular, maxillary appliances impact linguodental,linguoalveolar, and linguopalatal articulatory contacts.Therefore, the sounds most likely to be affected wouldinclude ‘‘th,’’ /t/, /d/, /s/, /z/, /n/, /l/, ‘‘sh,’’ ‘‘zh,’’ /r/,‘‘ch,’’ and ‘‘j.’’ Our study could not determine whichsound was most affected by the expanders studied. Inregard to speech adaptation, Hamlet21 proposed thatyounger patients show more ability to adapt to a neworal environment. Our findings do not support this, sinceage was not significantly correlated to any of the vari-ables studied.

Our findings suggest that chewing is slightly moreaffected that swallowing. Nonetheless, both variableswere reported to return to normal after about a week.

Table II. Pain, speech, chewing, and swallowing findings

Category of discomfort Frequency Percentage

Pain

Yes 10 93.9

No 153 6.1

Total 163 100.0

Speech

Yes 144 89.4

No 17 10.6

Total 161 100.0

Chewing

Yes 148 90.2

No 16 9.8

Total 164 100.0

Swallowing

Yes 112 67.9

No 52 31.5

Total 165 100.0

Table I. Demographics of the sample

Frequency Percentage

Sex

Male 60 41.1

Female 86 58.9

Total 146 100.0

Age (y)

7-10 24 18.2

11-14 77 58.3

15-18 28 21.2

19-22 3 2.3

Total 132 100.0

Type of appliance

Hyrax 12 8.6

Haas 50 36.0

Bonded 37 26.6

Quad-helix 32 23.0

Did not know 8 5.8

Total 139 100.0

System not noted 26


CONCLUSIONS

We found that all potentially negative impactsof placing a palatal expander, including pain anddiscomfort, disruption of speech production, and chew-ing and swallowing problems were mild, transitory, andindependent of appliance design, sex, or age. All issueswere resolved within the first week of appliance wear.

Much of the information gathered in this study canbe readily incorporated into the patient educationprocess, so that patients can be counseled about thechanges to expect with relation to pain, discomfort,speech, chewing, and swallowing immediately after anappliance is placed and up to a week after wearing it.The patient’s psychological response to orthodontictreatment and ability to adapt to the appliance mightbe significantly improved with anticipatory guid-ance.1,3,4 Providing such information to patients canbe expected to expedite the adaptation process andimprove cooperation.

Based on the limitations of this study, we suggest thatfuture studies have stratified sample sizes (ie, the samenumbers of subjects in every age group and for eachtype of expander). Furthermore, it is advisable that clin-ical assessments by a speech pathologist and an ortho-dontist to evaluate the patients’ speech and the positionof the teeth and tongue, respectively, at different times,are included. Finally, it would be desirable to compare

the results of retrospective questionnaires, administeredafter the expander has been removed from patients’mouths, with those of prospective questionnaires con-taining their daily impressions of oral comfort, speech,and masticatory function.

We thank Monika Mahajan for help with the manu-script and Drs Andrew Hass, Cyril Sadowsky, AshokKothari, Terry Selke, Robyn Silberstein, and HeekYoung Jo for help with the data collection.

REFERENCES

1. Stewart FN, Kerr WJ, Taylor PJ. Appliance wear: the patient’s

point of view. Eur J Orthod 1997;19:377-82.

2. Tedesco LA, Keffer MA, Davis EL, Christersson LA. Effect of

a social cognitive intervention on oral health status, behavior re-

ports, and cognitions. J Periodontol 1992;63:567-75.

3. Sergl HG, Zentner A. A comparative assessment of acceptance of

different types of functional appliances. Eur J Orthod 1998;20:

517-24.

4. Sergl HG, Klages U, Zentner A. Functional and social discomfort

during orthodontic treatment—effects on compliance and predic-

tion of patients’ adaptation by personality variables. Eur J Orthod

2000;22:307-15.

5. Sinclair PM, Cannito MF, Goates LJ, Solomos LF, Alexander CM.

Patient responses to lingual appliances. J Clin Orthod 1986;20:

396-404.

6. Artun J. A post-treatment evaluation of multibonded ceramic

brackets in orthodontics. Eur J Orthod 1997;19:219-28.

7. Miyawaki S, Yasuhara M, Koh Y. Discomfort caused by bonded

lingual orthodontic appliances in adult patients as examined by

Fig 3. Percentages of the degree of discomfort by appliances included in the study.

52 De Felippe et al American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

retrospective questionnaire. Am J Orthod Dentofacial Orthop

1999;115:83-8.

8. Hohoff A, Stamm T, Goder G, Sauerland C, Ehmer U, Seifert E.

Comparison of 3 bonded lingual appliances by auditive analysis

and subjective assessment. Am J Orthod Dentofacial Orthop

2003;124:737-45.

9. Hohoff A, Fillion D, Stamm T, Goder G, Sauerland C, Ehmer U.

Oral comfort, function and hygiene in patients with lingual

brackets. A prospective longitudinal study. J Orofac Orthop

2003;64:359-71.

10. Hohoff A, Stamm T, Ehmer U. Comparison of the effect on oral

discomfort of two positioning techniques with lingual brackets.

Angle Orthod 2004;74:226-33.

11. Lewis HG, Brown WA. The attitude of patients to the wearing

of a removable orthodontic appliance. Br Dent J 1973;134:

87-90.

12. Johnson NC, Sandy JR. Tooth position and speech—is there a

relationship? Angle Orthod 1999;69:306-10.

13. Khalil AM. Short- and long-term effects of thumb-sucking habit

breaking appliance on speech in children. Egypt Dent J 1994;

40:827-32.

14. Erb DP. Speech effects of the maxillary retainer. Angle Orthod

1967;37:298-303.

15. Strutton CS, Burkland GA. The effect of maxillary retainers on

the clarity of speech. J Clin Orthod 1993;27:338-40.

16. Haydar B, Karabulut G, Ozkan S, Aksoy AU, Ciger S. Effects of

retainers on the articulation of speech. Am J Orthod Dentofacial

Orthop 1996;110:535-40.

17. Garber SR, Speidel TM, Siegel GM. The effects of noise and pal-

atal appliances on the speech of five-year-old children. J Speech

Hear Res 1980;23:853-62.

18. Garber SR, Speidel TM, Siegel GM, Miller E, Glass L. The effects

of presentation of noise and dental appliances on speech. J Speech

Hear Res 1980;23:838-52.

19. Laine T. Articulatory disorders in speech as related to size of the

alveolar arches. Eur J Orthod 1986;8:192-7.

20. Oliveira N, Da Silveira A, Kusnoto B, Viana G. Three-dimen-

sional assessment of morphologic changes of the maxilla: a com-

parison of 2 kinds of palatal expanders. Am J Orthod Dentofacial

Orthop 2004;126:354-62.

21. Hamlet SL. Speech adaptation to dental appliances: theoretical

considerations. J Baltimore Coll Dent Surg 1973;8:51-63.


ORIGINAL ARTICLE

Relationship between breastfeeding durationand prevalence of posterior crossbite in thedeciduous dentition

Henri Menezes Kobayashi,a Helio Scavone Jr,b R�ıvea Ines Ferreira,b and Daniela Gamba Garibb

Sao Paulo, Brazil

Introduction: This cross-sectional retrospective epidemiologic study assessed the relationship between ex-clusive breastfeeding duration and the prevalence of posterior crossbite in the deciduous dentition. Methods:Clinical examinations were performed in 1377 Brazilian children (690 boys, 687 girls), 3 to 6 years old, from 11public schools in Sao Paulo, Brazil. Based on questionnaires answered by the parents, the children were clas-sified into 4 groups according to the duration of exclusive breastfeeding: G1, never (119 subjects); G2, lessthan 6 months (720 subjects); G3, 6 to 12 months (312 subjects); and G4, more than 12 months (226 subjects).The statistical analyses included the chi-square test (P \0.05) and the odds ratio. Results: The posteriorcrossbite was observed in 31.1%, 22.4%, 8.3%, and 2.2% of the children, in groups G1, G2, G3, and G4,respectively. The results showed a statistically significant relationship between exclusive breastfeeding dura-tion and the prevalence of posterior crossbite. Conclusions: Children who were breastfed for more than 12months had a 20-fold lower risk for the development of posterior crossbite compared with children whowere never breastfed and a 5-fold lower risk compared with those breastfed between 6 and 12 months.(Am J Orthod Dentofacial Orthop 2010;137:54-8)

Mothers’ milk is a highly nutritious food thatdiminishes infant mortality, helps to preventdiseases, promotes immunologic and antial-

lergic protection, and reduces obesity and gastrointesti-nal problems; it is also directly linked to the baby’semotional and affective needs.1-3 From the oral-healthviewpoint, the method and duration of infant feedinghave been related to the development of severe earlychildhood caries.4-6 Furthermore, some authors havepointed out that breastfeeding provides the advantageof greater oral muscle exercise over bottle feeding.7-9

In 2002, based on a systematic review of the literature,the World Health Organization10 recommended a mini-mum of exclusive maternal breastfeeding up to the ageof 6 months. Moreover, in orthodontics, breastfeedingmight influence craniofacial growth and development,help to prevent nonnutritive sucking habits, and stimu-late the harmonious functional development of thestomatognathic system.11-14

Because alterations in occlusal development mightbe the result of genetic or environmental factors, variousauthors have studied the relationship between breast-feeding and malocclusion, but the literature is still con-troversial about this subject.15 Some authors found norelationship between breastfeeding and the develop-ment of malocclusions.16,17 Warren and Bishara,17 afterassessing 372 children, 4 to 5 years old, found no statis-tically significant associations between breastfeedingduration and the prevalence of anterior open bite, poste-rior crossbite, and increased overjet. However, otherstudies have pointed out that insufficient breastfeedingduration is related to malocclusions, particularly poste-rior crossbites.18-21 Because this type of malocclusiondevelops early and rarely self-corrects, the deciduousdentition is an excellent phase to promote preventiveor interceptive measures. Therefore, the purpose ofthis research was to analyze the relationship betweenexclusive breastfeeding duration and the prevalence ofposterior crossbite in the deciduous dentition.


This cross-sectional study was done according to theResolution Act 196/96 from the Brazilian NationalCommittee of Health.

The sample consisted of 1377 Brazilian children(690 boys, 687 girls) in the complete deciduous denti-tion phase, from 3 to 6 years of age, enrolled at 11 publicschools in eastern Sao Paulo, Brazil. Furthermore, other

From the Department of Orthodontics, University of Sao Paulo City, Universi-

dade Cidade de Sao Paulo, Sao Paulo, Brazil.a Postgraduate student.b Associate professor.

The authors report no commercial, proprietary or financial interest in the


Reprint requests to: Henri Menezes Kobayashi, R. Cesario Galeno, 432/448, Sao

Paulo-SP, 03071-000, Brazil; e-mail, [email protected].

Submitted, October 2007; revised and accepted, December 2007.

0889-5406/$36.00


doi:10.1016/j.ajodo.2007.12.033

54


inclusion criteria for sample selection were no extensivecarious lesions, missing teeth, dental anomalies ofshape, number, structure, and eruption, as well as no his-tory of orthodontic treatment, traumatic injuries to thecraniofacial complex, or oral surgeries. These criteriawere used to exclude changes in occlusal relationshipsthat could interfere with our results.

The clinical examinationss were performed by 3previously calibrated orthodontists (kappa: 0.89-1.00;r .0.90). The occlusal relationships were examinedby direct visual inspection with the teeth in centric oc-clusion. Posterior crossbite was diagnosed when an in-verted relationship of occlusion was observed betweenat least 1 posterior tooth (deciduous canine or molar)in the transverse plane.22,23 Posterior crossbite in the de-ciduous dentition was classified into 3 categories: bilat-eral, true unilateral, and unilateral with functionaldeviation of the mandible.22,24

Based on questionnaires answered by the mothers,a retrospective investigation was made concerning thelength of time that children were exclusively breastfed.Accordingly, children were classified into 4 groups:group 1 (G1), never breastfed (n 5 119); group 2(G2), breastfed for less than 6 months (n 5 720); group3 (G3), breastfed for 6 to 12 months (n 5 312); andgroup 4 (G4), breastfed for more than 12 months (n 5

226). Information on nonnutritive sucking habits wasalso requested in the questionnaires.

Statistical analyses were performed with Stata soft-ware (version 8.0, StataCorp, College Station, Tex). ThePearson chi-square test was used to verify the associa-tion between posterior crossbite prevalence and breast-feeding duration (P \0.05). In addition, the odds ratio(OR) was used to measure the strength of the associa-tion and the relative chances of developing theinvestigated malocclusion.

RESULTS

For the total sample, the results showed a posteriorcrossbite prevalence of 16.6%, with 2.8% of the chil-

dren having bilateral crossbite, 4.4% with true unilateralcrossbite, and 9.4% having functional unilateral cross-bite (Table I). Posterior crossbite was more prevalentin older than in younger children during the deciduousdentition (Table I).

Table II shows that 8.6% of the children were neverbreastfed (G1), 52.3% were exclusively breastfed forless than 6 months (G2), and 39.1% were exclusivelybreastfed for more than 6 months (G3 and G4). Further-more, the prevalence of posterior crossbite gradually de-creased as breastfeeding duration increased: 31.1% forG1 and only 2.2% for G4.

There was a statistically significant relationshipbetween exclusive breastfeeding duration and the preva-lence of posterior crossbite (Table III) in the 6 compari-sons in the 4 groups, particularly between groupsG1 and G3, G1 and G4, G2 and G3, and G2 and G4(P 5 0.0000). Therefore, children who had never beenbreastfed exhibited a higher prevalence of posteriorcrossbite compared with children who were exclusivelybreastfed between 6 and 12 months (OR 5 4.9) andalso compared with children who were breastfed formore than 12 months (OR 5 19.9). Children who werebreastfed for less than 6 months had a 3-fold higher riskcompared with children who were exclusively breastfedbetween 6 and 12 months, and a 12-fold higher risk

Table I. Prevalence of the types of posterior crossbite according to age in the total sample

Age (y)

Total sample3 4 5 6

Posterior crossbite n % n % n % n % n %

Absent 141 87.6 415 85.2 450 83.3 142 75.1 1,148 83.4

Bilateral 4 2.5 11 2.3 14 2.6 10 5.3 39 2.8

True unilateral 7 4.3 16 3.3 29 5.4 9 4.8 61 4.4

Functional unilateral 9 5.6 45 9.2 47 8.7 28 14.8 129 9.4

Total 161 100.0 487 100.0 540 100.0 189 100.0 1,377 100.0

Table II. Distribution of the sample and prevalence ofposterior crossbite in the 4 groups analysed, accordingto breastfeeding duration irrespective of gender

Sample Presence of posterior crossbite

Group n % n %

G1 119 8.6 37 31.1

G2 720 52.3 161 22.4

G3 312 22.7 26 8.3

G4 226 16.4 5 2.2

Total 1,377 100.0 229 16.6

G1, Never breastfed; G2, breastfed for \6 months; G3, breastfed for

6-12 months; G4, breastfed for .12 months.

American Journal of Orthodontics and Dentofacial Orthopedics Kobayashi et al 55Volume 137, Number 1

compared with children who were breastfed for morethan 12 months.

Table IV shows the distribution of posterior cross-bite prevalence according to the breastfeeding periodonly for children with no nonnutritive sucking habits(finger or pacifier). Again, a gradual decrease in theprevalence of this malocclusion was observed asbreastfeeding duration increased, particularly ingroups G3 and G4, comprising children breastfedfor more than 6 months. In these 2 groups, only 1child with posterior crossbite was found, indicatinga combined prevalence of 0.31%. When the chi-squaretest was applied in the group of children without non-nutritive sucking habits (Table V), statistically signif-icant relationships were seen between exclusivebreastfeeding duration and the prevalence of posteriorcrossbite between groups G1 and G3 (P \0.0000) andG2 and G3 (P \0.0003). Children who were neverbreastfed had a 29-fold higher risk for developing pos-terior crossbite compared with the children who wereexclusively breastfed between 6 and 12 months.Children breastfed for less than 6 months hada 16-fold higher risk compared with children whoseexclusive breastfeeding was interrupted between6 and 12 months. For the other paired comparisons

involving G4, it was not possible to estimate theOR because of a null prevalence of posterior crossbitein this group, making mathematical calculationsunfeasible.

DISCUSSION

Only 3 studies suggested a relationship betweenlonger breastfeeding and lower prevalence of posteriorcrossbite.18,19,21 Viggiano et al,18 with logistic regres-sion, compared 1099 children with nonnutritive suckinghabits who were breastfed with those with nonnutritivesucking habits who were bottlefed. They found that chil-dren with nonnutritive sucking habits who were bottlefedhad a higher risk of developing posterior crossbite com-pared with the children with similar sucking habits whowere exclusively breastfed. Karjalainen et al19 assessedonly 148 children (age, 3 years) and found that themean exclusive breastfeeding duration in the total sam-ple was 5.8 months, whereas, in the children with poste-rior crossbite, the mean duration was only 3.6 months.Furthermore, Peres et al21 examined 359 children (age,6 years) and verified that those who were breastfed forless than 9 months and also had nonnutritive suckinghabits between 1 and 4 years of age showed a 7.5-foldhigher risk compared with those who were breastfedfor more than 9 months and had no habits.

On the other hand, Ogaard et al16 and Warren andBishara17 found no significant relationship betweenbreastfeeding duration and prevalence of posteriorcrossbite. Nevertheless, these studies showed high per-centages of mothers who never breastfed their children;this could have made it difficult to make comparisonsamong the breastfed groups.

In relation to the previous studies, our investigationhad some particularities, since it was especially de-signed to evaluate the relationship between breastfeed-ing and a specific kind of malocclusion—posteriorcrossbite. Furthermore, analyses were carried out inboth the total sample and the group of children with

Table III. Intergroup comparisons for prevalence of pos-terior crossbite (total sample)

Comparison Chi-square P value OR

G1/G2 4.31 0.0378 1.57

G1/G3 35.67 0.0000 4.96

G1/G4 60.63 0.0000 19.94

G2/G3 28.84 0.0000 3.17

G2/G4 48.21 0.0000 12.73

G3/G4 9.03 0.0027 4.02



Table IV. Prevalence of posterior crossbite in the groups,excluding children with nonnutritive sucking habits

Group

SamplePresence of

posterior crossbite

n % n %

G1 22 4.4 4 18.2

G2 161 32.0 18 11.2

G3 132 26.2 1 0.8

G4 188 37.4 0 0

Total 503 100.0 23 4.6



Table V. Intergroup comparisons for prevalence of pos-terior crossbite, excluding children with nonnutritivesucking habits

Comparison Chi-square P value OR

G1/G2 0.89 0.3449 —

G1/G3 18.11 0.0000 29.11

G1/G4 — — —

G2/G3 12.95 0.0003 16.49

G2/G4 — — —

G3/G4 — — —



56 Kobayashi et al American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

no nonnutritive sucking habits. This last procedure ex-cluded the influence of this variable, considering thatmany studies have proved the relationship between per-sistent nonnutritive sucking habits and the developmentof posterior crossbite.7,8,12,13,17-19,21,24-28

In addition to the aspects discussed previously, ourresults also seem to suggest that the use of the feedingbottle could have a deleterious effect on the develop-ment of occlusion, perhaps as a predisposing factorfor posterior crossbite. This hypothesis can be raised be-cause children who were not breastfed were necessarilybottlefed. Some authors have argued that the feedingbottle is considered a deleterious habit, particularlyfor the development of the anterior segment of the den-tal arches.28-30 Our study suggests that this relationshipshould be better investigated, since various physiologicaspects of muscular mechanisms are involved in breast-feeding and bottlefeeding.7-9 The results demonstratedthat, in children who were never breastfed and had nononnutritive sucking habits (G1; Table IV), there wasa prevalence of 18.2% for posterior crossbite, whereasin group G1 of the total sample, which also includedchildren with nonnutritive sucking habits (Table II),the prevalence of this malocclusion was 31.1%. Thisdifference points to the fact that the absence of nonintu-itive sucking habits reduced the prevalence of posteriorcrossbite by almost 50%, but was not sufficient for thetotal prevention of this malocclusion. On the otherhand, in G3 and G4 of the total sample, their combinedprevalence was 5.76%, as opposed to only 0.31% in thechildren without nonnutritive sucking habits. Therefore,simply breastfeeding a child exclusively for more than 6months can sharply reduce the prevalence of posteriorcrossbite, compared with children who were neverbreastfed (31.1%), even without excluding the deleteri-ous influence of nonnutritive sucking habits. Moreover,when this latter factor was eliminated, the prevalencewas practically reduced to zero. These results seem topoint to an effect of breastfeeding that is, at least, doublybeneficial: reduction in nonnutritive sucking habits andprotection against posterior crossbite. This last effectwas mentioned by Viggiano et al18 and Karjalainenet al.19 Furthermore, exclusive breastfeeding reducesthe use of feeding bottles, which probably overstimu-lates buccinator muscle contraction activity, generatingnegative pressures inside the oral cavity and perhapspredisposing to a reduction in the maxillary dentalarch width. During breastfeeding, the muscular mecha-nisms involved are different, with repeated advance andwithdrawal of the tongue and mandible. Probably otherbeneficial effects of breastfeeding might be related,such as strengthening the immunologic system and theconsequent reduction in respiratory problems, which

can also interfere with the development of dental occlu-sion.31

Furthermore, Victora et al29 affirmed that introduc-tion of the feeding bottle could predispose the child toearly weaning because the milk is obtained more easily,causing the baby to gradually reject the breast. On theother hand, early weaning or complete absence ofbreastfeeding might be caused by other factors—eg, in-sufficient mother’s milk, unfavorable breast anatomy,mother’s lack of interest or emotional problems, oreven because maternity leave has ended. In these cases,the first habit to be introduced to feed the child is almostalways the feeding bottle, which satisfies only thebaby’s physiologic hunger but not its need to suck,which is generally compensated by introducing thepacifier.

Various factors could explain the origins of so manycontroversies with respect to the relationship betweenbreastfeeding duration and the development of maloc-clusions. Many studies suggested that breastfeedingseems to help reduce the acquisition of nonnutritivesucking habits.8,11,12,18,19,21,28-30,31,32 Because thesehabits are well-known etiologic factors of malocclu-sions, it could be expected that breastfeeding forprolonged periods would help to prevent the acquisitionof such habits and, consequently, the associated maloc-clusions.7,8,12,13,17-19,21,24-28,30,32,33 Nevertheless, thequestion appears to be more complex, since mostpublished studies could not clearly show a well-definedinterrelationship between exclusive breastfeedingduration and the development of malocclusions. Howcan it be explained? There are many possible hypothe-ses, ranging from factors related to the size of samples,inclusion and exclusion criteria, calibration of exam-iners, the method of dividing the sample groups, maloc-clusion assessments and classification modes,interference of nonnutritive sucking habits and feedingmethods, and many others. Therefore, it is not surpris-ing to find many controversies. This study seems tohave overcome some of these limitations, by workingwith a sample sufficiently large, combined with en-hanced selection criteria, careful division of the sub-groups, adequate assessment methods, and the use ofstatistical analyses compatible with the nature of thisstudy. These data indicated that prolonged breastfeed-ing duration can strongly reduce the prevalence of pos-terior crossbite during the deciduous dentition.

Our results agree with and provide additional sup-port for the World Health Organization’s recommenda-tion that children should be exclusively breastfed fora minimum of 6 months.10 Moreover, our results alsopoint out that lengthening this period can have addi-tional beneficial effects, since the group of children

American Journal of Orthodontics and Dentofacial Orthopedics Kobayashi et al 57Volume 137, Number 1

breastfed for more than 12 months had a prevalence ofposterior crossbite of only 2.2%, whereas the groupbreastfed between 6 and 12 months had a prevalenceof 8.3% for this malocclusion. In contrast, the groupof children who were never breastfed had a 31.1%prevalence of posterior crossbite.

CONCLUSIONS

These results show an association between exclu-sive breastfeeding duration and the prevalence of poste-rior crossbite in the deciduous dentition. Children whowere breastfed for more than 12 months had a 20-foldlower risk for the development of posterior crossbitecompared with children who were never breastfed andalso a 5-fold lower risk compared with those breastfedbetween 6 and 12 months.

REFERENCES

1. Clemens J, Elyazeed RA, Rao M, Savarino S, Morsy BZ, Kim Y,

et al. Early initiation of breastfeeding and the risk of infant diar-

rhea in rural Egypt. Pediatrics 1999;104:e3.

2. Oddy WH, Sherriff JL, Klerk NH, Kendall GE, Sly PD, Beilin LJ,

et al. The relation of breastfeeding and body mass index to asthma

and atopy in children: a prospective cohort study to age 6 years.

Am J Public Health 2004;94:1531-7.

3. Kramer MS, Matush L, Vanilovich I, Platt R, Bogdanovich N,

Sevkovska Z, et al. Effect of prolonged and exclusive breast feed-

ing on risk of allergy and asthma: cluster randomized trial. BMJ

2007;335:782-3.

4. Dini EL, Holt RD, Bedi R. Caries and its association with infant

feeding and oral health-related behaviours in 3-4-year-old Brazil-

ian children. Community Dent Oral Epidemiol 2000;28:241-8.

5. Nainar SM, Mohummed S. Diet counseling during the infant oral

health visit. Pediatr Dent 2004;26:459-62.

6. Azevedo TD, Bezerra AC, Toledo OA. Feeding habits and severe

early childhood caries in Brazilian preschool children. Pediatr

Dent 2005;27:28-33.

7. Westover KM, DiLoreto MK, Shearer TR. The relationship of

breastfeeding to oral development and dental concerns. ASDC J

Dent Child 1989;56:140-3.

8. Legovic M, Ostric L. The effects of feeding methods on the

growth of the jaws in infants. ASDC J Dent Child 1991;58:253-5.

9. Degano MP, Degano RA. Breastfeeding and oral health. N Y State

Dent J 1993;59:30-2.

10. Butte NF, Lopez-Alarcon MG, Gaza C. Nutrient adequacy of ex-

clusive breastfeeding for the term infant during the first six

months of life. Geneva: World Health Organization; 2002.

11. Finocchi LL. Breast feeding, bottle feeding and their impact on oral

habits. A review of the literature. Dent Hyg (Chic) 1982;56:21-5.

12. Farsi NMA, Salama FS, Pedo C. Sucking habits in Saudi children:

prevalence, contributing factors and effects on the primary denti-

tion. Pediatr Dent 1997;19:28-33.

13. Larsson E. Sucking, chewing, and feeding habits and the develop-

ment of crossbite:a longitudinal study of girls from birth to 3 years

of age. Angle Orthod 2001;71:116-9.

14. Aznar T, Galan AF, Marin I, Dom�ınguez A. Dental arch diameters

and relationships to oral habits. Angle Orthod 2006;76:441-5.

15. Mossey PA. The heritability of malocclusion: part 2. The influ-

ence of genetics in malocclusion. Br J Orthod 1999;26:195-203.

16. Ogaard B, Larsson E, Lindsten R. The effect of sucking habits, co-

hort, sex, intercanine arch widths, and breast or bottle feeding on

posterior crossbite in Norwegian and Swedish 3-year-old chil-

dren. Am J Orthod Dentofacial Orthop 1994;106:161-6.

17. Warren JJ, Bishara SE. Duration of nutritive and nonnutritive

sucking behaviors and their effects on the dental arches in the pri-

mary dentition. Am J Orthod Dentofacial Orthop 2002;121:

347-56.

18. Viggiano D, Fasano D, Monaco G, Strohmenger L. Breastfeeding,

bottle feeding, and non-nutritive sucking: effects on occlusion in

deciduous dentition. Arch Dis Child 2004;89:1121-3.

19. Karjalainen S, Ronning O, Lapinleimu H, Simell O. Association

between early weaning, non-nutritive sucking habits and occlusal

anomalies in 3-year-old Finnish children. Int J Paediatr Dent

1999;9:169-73.

20. Labbok MH, Hendershot GE. Does breast-feeding protect against

malocclusion? An analysis of the 1981 Child Health Supplement

to the National Health Interview Survey. Am J Prev Med 1987;3:

227-32.

21. Peres KG, Barros AJD, Peres MA, Victora CM. Effects of breast-

feeding and sucking habits on malocclusion in a birth cohort

study. Rev Saude Publica 2007;41:343-50.

22. Malandris M, Mahoney EK. Aetiology, diagnosis and treatment of

posterior crossbites in the primary dentition. Int J Paediatr Dent

2004;14:155-66.

23. Kisling E, Krebs G. Patterns of occlusion in 3-year-old Danish

children. Community Dent Oral Epidemiol 1976;4:152-9.

24. Scavone H Jr, Ferreira RI, Mendes TE, Ferreira FV. Prevalence of

posterior crossbite among pacifier users: a study in the deciduous

dentition. Braz Oral Res 2007;21:153-8.

25. Larsson E. Prevalence of crossbite among children with prolonged

dummy- and finger-sucking habit. Swed Dent J 1983;7:115-9.

26. Modeer T, Odenrick L, Lindner A. Sucking habits and their rela-

tion to posterior cross-bite in 4-year-old children. Scand J Dent

Res 1982;90:323-8.

27. Infante PF. An epidemiologic study of finger habits in preschool

children, as related to malocclusion, socioeconomic status, race,

sex, and size of community. ASDC J Dent Child 1976;43:33-8.

28. Turgeon-O’Brien H, Lachapelle D, Gagnon PF, Larocque I,

Maheu-Robert L. Nutritive and nonnutritive sucking habits:

a review. ASDC J Dent Child 1996;63:321-7.

29. Victora CG, Behague DP, Barros FC, Olinto TA, Weiderpass E.

Pacifier use and short breastfeeding duration: cause, consequence,

or coincidence? Pediatrics 1997;99:445-53.

30. Lopez del Valle LM, Singh D, Feliciano N, Machuca MC. Asso-

ciations between a history of breast feeding, malocclusion and

parafunctional habits in Puerto Rican children. P R Health Sci J

2006;25:31-4.

31. Trawitzki LVV, Anselmo-Lima WT, Melchior MO, Grechi TH,

Valera FCP. Breast-feeding and deleterious oral habits in

mouth and nose breathers. Rev Bras Otorrinolaringol 2005;

71:747-51.

32. Bishara SE, Warren JJ, Proffitt B, Levy SM. Changes in the prev-

alence of nonnuntritive sucking patterns in the first 8 years of life.


33. Bishara SE, Nowak AJ, Kohout FJ, Heckert A, Hogan MM. Influ-

ence of feeding and non-nutritive sucking methods on the devel-

opment of the dental arches: longitudinal study of the first 18

months of life. Pediatr Dent 1987;9:13-23.

58 Kobayashi et al American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

ORIGINAL ARTICLE

Effectiveness of the cervical vertebral maturationmethod to predict postpeak circumpubertalgrowth of craniofacial structures

Piotr Fudaleja and Anne-Marie Bollenb

Warsaw, Poland, and Seattle, Wash

Introduction: Our aim was to assess effectiveness of the cervical vertebral maturation (CVM) method to pre-dict circumpubertal craniofacial growth in the postpeak period. Methods: The CVM stage was determined in176 subjects (51 adolescent boys and 125 adolescent girls) on cephalograms taken at the end of treatment(T2; mean ages, 15.75 years [boys] and 15.23 years [girls]) in subjects from the postretention database atthe University of Washington in Seattle. Craniofacial growth was evaluated from the following measurementson cephalograms at T2 and end of follow-up (T3) (mean ages, 29.01 years [men] and 28.08 years [women]):condylion to gnathion, condylion to gonion, gonion to gnathion, sella to gnathion, nasion to menton, anteriornasal spine to menton, and sella to gonion. The change of each variable from T2 to T3 was assessed withpaired t tests. Parametric (t tests or analysis of variance [ANOVA]) or nonparametric (Mann-Whitney or Krus-kal-Wallis) tests were used to detect intergroup differences. Results: One hundred eight subjects (35 boys, 73girls) demonstrated CVM stage 3, 56 (16 boys, 40 girls) were in CVM stage 4, and 12 (all girls) were in CVMstage 5 at T2. Intrasex comparisons showed that boys in CVM stages 3 and 4 could be differentiated regardingchanges of all variables. In the girls, only those in CVM stages 3 and 4 could be differentiated based on theamount of changes of 2 measurements: condylion to gonion and sella to gonion. Intersex comparisonsshowed that boys in CVM stage 3 had significantly more changes than girls (P \0.01). Boys in CVM stage4 showed significant differences compared with girls in CVM stage 4 for only 2 variables (sella to gonionand condylion to gonion; P \0.001 and P 5 0.012, respectively). Conclusions: The CVM method was mod-estly effective in determining the amount of postpeak circumpubertal craniofacial growth. (Am J OrthodDentofacial Orthop 2010;137:59-65)

Data from randomized, controlled clinical studieson Class II treatment suggested that skeletal ef-fects with various protocols carried out in pre-

pubertal children are of minor importance in thecorrection of Class II molar relationships.1-3 However,the frequently observed improvement of a profile duringtreatment is directly related to the size or the position ofthe mandible. Findings of the investigations that aimedto elucidate this disparity implied that, when treatmentof Class II malocclusion starts just before the pubertalgrowth spurt, skeletal effects are larger and lasting com-pared with treatment completed before the circumpu-bertal growth peak.4,5 Cozza et al6 in their systematic

review found that the amount of supplementary mandib-ular growth appeared to be significantly larger if thefunctional treatment was performed at the pubertalpeak in skeletal growth. Baccetti et al5 compared man-dibular morphology in patients treated early vs late andconcluded that the optimal timing for Twin-block ther-apy for a Class II disharmony was during or slightlyafter the pubertal peak in growth velocity.

On the contrary, treatment of some orthodonticproblems necessitates little or no craniofacial growth.A planned camouflage in Class III subjects will besuccessful if facial growth is complete. If the mandibleoutgrows the maxilla during treatment, achievement ofstable correction is questionable. In children with con-genitally missing lateral incisors, an implant-based res-toration is often the method of choice. Since implantsbehave like ankylosed teeth,7,8 early implantation couldlead to submergence of the implant crown and cause anesthetic disaster.9 Coming of age is often recommendedas the suitable time to start treatment. Nevertheless,chronologic age was shown to be poorly correlatedwith development.10 Cessation of craniofacial growthin early maturers is probably complete a few years

a Assistant professor, Center for Craniofacial Disorders, Institute of Mother and

Child, Warsaw, Poland.b Professor, Department of Orthodontics, University of Washington, Seattle.



Reprint requests to: Piotr Fudalej, Center for Craniofacial Disorders, Institute of

Mother and Child, Kasprzaka Str. 17a, Warsaw, Poland; e-mail, pfudalej@

gmail.com.

Submitted, October 2007; revised and accepted, January 2008.

0889-5406/$36.00


doi:10.1016/j.ajodo.2008.01.018

59



earlier than in late maturers.11 Consequently, startingtreatment according to chronologic age irrespective ofthe patient’s maturity is not always appropriate.

Among methods for assessment of craniofacial ma-turity, the cervical vertebral maturation (CVM) methodhas recently gained popularity because of its ease andpostulated accuracy.12-14 The most frequently usedmodification of the CVM method15-18 was based onthe cephalometric analyses of longitudinal records of9 boys and 15 girls.14 Through discriminant analysis,several CVM stages were established that were intendedto correspond to skeletal maturity and, hence, future fa-cial growth. Although an objective of the CVM methodis to predict the peak velocity of growth, it should indi-rectly point to how much postpeak growth change is ex-pected. However, the small sample size used to devisethe CVM method and the lack of validation raise doubtsabout its effectiveness, especially in the later stages ofdevelopment when growth tapers off.14 These reserva-tions are supported by the findings of Baccetti et al,18

who examined growth changes in a cross-sectional sam-ple of 1091 Class III subjects and found that the CVMmethod detected between-stage growth alterationsonly in approximately 25% of the performed measure-ments. Moreover, some children go through a prolongedperiod of accelerated growth without a distinct growthpeak.11 It is unclear whether the CVM method can bean effective predictor in these subjects.

Therefore, our aim in this study was to assess theeffectiveness of the CVM method to predict circumpu-bertal craniofacial growth in the postpeak period.


A sample from a study on incisor stability was usedin this investigation;19 83.7% of the subjects from theoriginal sample comprising 301 subjects from the post-retention collection in the Department of Orthodonticsat the University of Washington in Seattle had theirCVM status established. Only those with a CVM statusthat met the following supplementary inclusion criteriawere chosen: good-quality lateral cephalograms madeat the end of orthodontic treatment (T2) and at least10 years out of retention (T3), no orthognathic surgery,and no additional orthodontic treatment between T2 andT3. The enlargement factor of most cephalograms couldnot be determined, so, to minimize the influence of en-largement, only subjects with cephalograms taken at T2and T3 in the same cephalostat were finally included.

The length of follow-up was calculated by subtract-ing the T2 date from the T3 date. Patient informationand treatment history for all subjects were obtained

from the database. Age at the end of treatment andsex were recorded.

Lateral cephalograms taken at T2 were used to de-termine the CVM stage. The assignment of the CVMstages is described elsewhere.19 To summarize, oneach cephalogram, the second, third, and fourth cervicalvertebrae (C2, C3, and C4) were identified. The land-marks and measurements we used are presented inFigure 1. The CVM stages are shown in Table I.

Craniofacial growth was evaluated on lateral cepha-lograms taken at T2 and T3. The landmarks identifiedand the measurements made according to the studythat described the CVM method are given in Figure 2.14


Descriptive statistics (means and standard deviations)were computed for age and length of follow-up in each

Fig 1. Schematic representation of landmarks identifiedand traced on the cervical vertebral body: LP, D, and LA,the most posterior, deepest, and most anterior points onthe lower border of the body of C2, C3, or C4; UP andUA, the most superior points of the posterior and ante-rior borders of the body of C3 or C4; Conc, distancefrom the line connecting and LA to D on the lower borderof C2, C3, or C4. Calculations: BAR, the ratio betweenthe length of the base (LP-LA) and the anterior height(UA-LA) of the body of C3 or C4; and PAR, the ratio be-tween the posterior (UP-LP) and anterior (UA-LA) heightsof the body of C3 or C4.

60 Fudalej and Bollen American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

CVM group. For the craniofacial measurements, means,standard deviations, and 95% confidence intervals werecalculated.

Paired t tests were used to assess craniofacial growthin each group. Shapiro-Wilks tests were used to evaluatenormality of distribution in each group. In case of normaldistribution, independent t tests or analysis of variance(ANOVA) were run. If distribution was not normal, non-parametric tests, Mann-Whitney for 2-group compari-sons or Kruskal-Wallis for 3-group comparisons, werecarried out. The Dunn multiple comparison procedurewith the Bonferroni adjustment was used for intergroupdifferences.

At P \0.05, the difference was considered signifi-cant. At P \0.1, the difference was considered margin-ally significant.

The reproducibility of the measurements was as-sessed by statistically analyzing the difference betweendouble measurements taken 1 week apart on 25 cepha-lograms selected at random. The error of the methodwas calculated from the equation:

SX 5

ffiffiffiffiffiffiffiffiffiffiffiffiPD2

2N

r

with D representing the difference between the corre-sponding first and second measurements and N thenumber of double determinations.

Intraobserver agreement for the CVM stage assign-ment was calculated as the Pearson correlation coeffi-cient and proportionally weighted kappa coefficientbased on 2 assignment sessions a week apart.

RESULTS

A total of 252 subjects were initially identified; 61were excluded for lack of good-quality cephalogramstaken in the same cephalostat at T2 and T3, and 15 ofthe remaining 191 subjects were in CVM stages 1 and2 (7 and 8 children, respectively), indicating prepubertaldevelopment, and were excluded. The final sample in-cluded 176 subjects: 29% male and 71% female. Thesample characteristics are presented in Table II.

Boys in the CVM stage 3 group were older than thegirls in the same stage at the beginning of follow-up by0.72 years; the difference was significant (P 5 0.017).No intersex difference was found regarding age in theCVM stage 4 group (P 5 0.412).

Age at the start of follow-up was similar in boys inCVM stages 3 and 4 (P 5 0.589). Analogous compari-sons in girls showed difference between CVM stages 3

Table I. Classifications of CVM stages according to theshape of the bodies of C2, C3, and C4

CVM stage Vertebral body shape

1 C3 and C4 flat

2 C3 concavity $1 mm; C4 flat

3 C2, C3, and C4 concavity $1 mm; C3 and/or C4

tapered or horizontal rectangular

4 C3 and/or C4 square; if C3 or C4 are not square,

then horizontal rectangular

5 C3 and/or C4 vertical rectangular

Flat: concavity \1 mm.

Tapered: C3 UP-C3 LP (or C4 UP-C4 LP) to C3 UA-C3 LA (or C4

UA-C4 LA) ratio (PAR) .1.20.

Square: C3 UP-C3 LP (or C4 UP-C4 LP) to C3 UA-C3 LA (or C4 UA-

C4 LA) ratio (PAR) of 0.80-1.20; C3 LA-C3 LP (or C4 LA-C4 LP) to

C3 UA-C3 LA (or C4 UA-C4 LA) ratio (BAR) of 0.85-1.15.

Horizontal rectangular: C3 LA-C3 LP (or C4 LA-C4 LP) to C3 UA-

C3 LA (or C4 UA-C4 LA) ratio (BAR) .1.15.

Vertical rectangular: C3 LA-C3 LP (or C4 LA-C4 LP) to C3 UA-C3

LA (or C4 UA-C4 LA) ratio (BAR) \0.85.

Fig 2. Craniofacial measurements: 1, mandibular length(Co-Gn); 2, posterior ramus height (Co-Go); 3, mandi-bular body length (Go-Gn); 4, sella to gnathion distance(S-Gn); 5, anterior facial height (N-Me); 6, lower anteriorfacial height (ANS-Me); 7, posterior facial height (S-Go).Gonion (Go) is located at the intersection of the mandib-ular (MP) and ramus (RP) planes.

American Journal of Orthodontics and Dentofacial Orthopedics Fudalej and Bollen 61Volume 137, Number 1

and 4 (P 5 0.016). No differences between the CVMstages 3 and 5 groups, and the CVM stages 4 and 5groups were found.

The length of follow-up in the various CVM groupsdid not differ in the sexes (P 5 0.924 and 0.970, respec-tively) and approximated 13.0 years.

Error of the method is shown in Table III. For mostvariables, errors did not exceed 0.75 mm. Condylion togonion and condylion to gnathion had errors in excess of1 mm (2.14 and 1.40 mm, respectively).

The Pearson correlation coefficient between the firstand second assignments of CVM stages in 25 subjectswas 0.67 (95% CI 5 0.39-0.84; r2 5 0.45), and theweighted kappa coefficient was 0.55.

During follow-up, most variables showed signifi-cant changes at P \0.01 (Table IV). However, mandib-ular length (Co-Gn) increased in the male CVM stage 4group by 2.25 mm (P 5 0.015), and 2 variables, anteriorfacial height (N-Me) and lower anterior facial height(ANS-Me) in males in CVM stage 4, had marginallysignificant differences (P 5 0.096 and 0.060, respec-tively). All measured structures in females showed sta-tistically significant changes (P \0.01) during thefollow-up.

Comparison between males in various CVM stagesshowed that craniofacial structures in the CVM stage3 group grew significantly more than in the CVM stage4 group. Length of the mandible (Co-Gn) increased by5 mm more in the CVM stage 3 compared with theCVM stage 4 group (P \0.001). Also, posterior ramusheight (Co-Go), sella-gnathion distance, and posteriorfacial height (S-Go) increased in the CVM stage 3boys by more than 4 mm than in the CVM stage 4 group(P\0.01). The least difference between male groups—1.7 mm—was observed for lower anterior facial height(ANS-Me) (P 5 0.038).

Females in various CVM stages demonstrated onlya few significant differences. Although most craniofa-cial variables increased more in the CVM stage 3 thanin the CVM stage 4 or 5 group, the only statistically sig-nificant differences between the CVM stages 3 and 4groups were posterior ramus height (Co-Go) and poste-rior facial height (S-Go) (P 5 0.016 and 0.021, respec-tively). No difference was found between females in theCVM stage 3 vs 5, or CVM stage 4 vs 5.

When males and females in the same CVM stagewere compared, a few significant distinctions werefound (Table V). All measured craniofacial structuresin boys from the CVM stage 3 group increased at leasttwice as much as in the girls in the same CVM stage(P 5 0.002 for lower anterior facial height [ANS-Me]and P\0.001 for the remaining variables). Fewer differ-ences were observed between the CVM stage 4 boys andgirls. Only posterior ramus height (Co-Go) and posteriorfacial height (S-Go) differed significantly between thesexes (P 5 0.012 and \0.001, respectively).

DISCUSSION

The observation that cervical vertebrae undergoconsistent morphologic changes during growth led tothe development of a method relating the vertebral alter-ations to craniofacial growth. Originally, the CVM

Table II. Age at the beginning and length of follow-up in groups in various CVM stages

Males Females Males 1 females

Age (y)Follow-uplength (y) Age (y)

Follow-uplength (y) Age (y)

Follow-uplength (y)

CVM stage n Mean SD Mean SD n Mean SD Mean SD n Mean SD Mean SD

3 35 15.68 1.29 13.31 4.60 73 14.96 1.52 12.92 4.06 108 Not pooled* 13.05 4.23

4 16 15.92 1.07 12.31 5.29 40 15.62 1.25 12.73 2.99 56 15.71 1.20 12.60 3.76

5 0 – – – – 12 15.55 1.20 12.83 4.28 12 15.55 1.20 12.83 4.28

51 P 5 0.589 P 5 0.924 125 P 5 0.016† P 5 0.970 176 P 5 0.996

*Ages for boys and girls in CVM stage 3 were significantly different (P 5 0.017); †Dunn multiple comparison tests with the Bonferroni adjustment

showed statistically significant differences in age between the CVM stages 3 vs 4 groups.

Table III. Error of measurements

Variable Error Variable Error

Co-Gn 1.40 C2 Conc 0.33

Co-Go 2.14 C3 Conc 0.30

Go-Gn 0.52 C4 Conc 0.28

S-Gn 0.67 C3 PAR 0.05

N-Me 0.69 C3 BAR 0.07

ANS-Me 0.74 C4 PAR 0.09

S-Go 0.71 C4 BAR 0.07

Conc, distance from the line connecting and LA to D on the lower bor-

der of C2, C3, or C4. Calculations: BAR, the ratio between the length of

the base (LP-LA) and the anterior height (UA-LA) of the body of C3 or

C4; PAR, the ratio between the posterior (UP-LP) and anterior (UA-

LA) heights of the body of C3 or C4.


January 2010

method distinguished 6 stages that corresponded to dif-ferent developmental phases. A recent modification lim-ited the number of stages to 5: from CVM stage 1indicating that a peak of craniofacial growth will occurnot earlier than 1 year after this stage, to CVM stage 5indicating that a growth spurt occurred at least 2 yearsbefore this stage.20 A growth peak, according to thismethod, should occur between the second and thirdstages in 95% of adolescents. Hassel and Farman13

stated that 65% to 85% of adolescent growth is expectedin a child in CVM stage 1, as opposed to 5% to 10% ofgrowth in a stage 4 person. A person in CVM stage 5should demonstrate little to no craniofacial growth.

Our findings suggest that the CVM method is onlymodestly effective in discriminating between subjectswith various amounts of circumpubertal craniofacialgrowth in the later stages of development. Although

a focus of the CVM method is prediction of the pubertalgrowth peak, the assumption that adolescents at a moreadvanced CVM stage will grow less seems to be valid.However, the CVM method could differentiate betweenboys in only CVM stages 3 and 4 of development. Thedetection ability of the CVM method in girls was muchpoorer. Of the 7 craniofacial variables measured, only 2(posterior ramus height and posterior facial height) werediscriminating for girls between CVM stages 3 and 4,and the levels of statistical significance (P 5 0.016and 0.021, respectively) were not high. Moreover, theCVM stages 4 and 5 did not predict different amountsof craniofacial growth. These findings generally agreewith the results of Franchi et al,14 who reported averagechanges of craniofacial structures during transition fromlower to higher CVM stages. Mandibular length (Co-Gn) increased accordingly from CVM stages 4 to 5

Table IV. Intrasex comparisons of the variables during follow-up in subjects in various CVM stages

Co-Gn Co-Go Go-Gn S-Gn

CVM stage Mean SD 95% CI Mean SD 95% CI Mean SD 95% CI Mean SD 95% CI

Males 3 7.26 6.02 5.19-9.32 7.89 4.89 6.21-9.56 4.54 3.85 3.22-5.87 7.34 5.48 5.46-9.27

4 2.25 3.30 0.49-4.01 3.75 2.67 2.33-5.17 2.38 2.13 1.24-3.51 2.94 3.47 1.09-4.79

P \0.001 P 5 0.003 P 5 0.041 P 5 0.005

Females 3 3.16 2.91 2.49-3.84 3.08 2.60 2.48-3.69 2.26 2.28 1.73-2.79 2.73 2.39 2.17-3.28

4 1.85 2.76 0.97-2.73 1.60 2.84 0.69-2.51 1.78 1.62 1.26-2.29 1.65 1.75 1.09-2.21

5 3.00 2.70 1.29-4.71 3.08 3.12 1.10-5.06 2.08 1.38 1.21-2.96 2.42 2.43 0.87-3.96

P 5 0.063 P 5 0.016; 3 vs 4 P 5 0.530 P 5 0.057

N-Me ANS-Me S-Go

CVM stage Mean SD 95% CI Mean SD 95% CI Mean SD 95% CI

Males 3 4.77 4.15 3.35-6.20 2.51 2.98 1.49-3.54 8.48 4.74 6.83-10.09

4 1.25 2.82 –0.25-2.75 0.81 1.60 –0.04-1.67 4.38 2.92 2.82-5.93

P 5 0.003 P 5 0.038 P 5 0.002

Females 3 2.41 2.60 1.80-3.02 1.07 1.81 0.65-1.49 2.84 2.38 2.28-3.39

4 1.90 1.97 1.27-2.53 1.08 1.53 0.59-1.56 1.55 1.87 0.30-0.95

5 2.50 2.20 1.11-3.89 2.17 2.33 0.69-3.65 2.50 2.68 0.80-4.20

P 5 0.503 P 5 0.174 P 5 0.021; 3 vs 4

Follow-up was the time from T2 (end-of-treatment records) to T3 (long-term out-of-retention records).

Table V. Intersex comparison of the measured variables during follow-up

CVM stage Co-Gn Co-Go Go-Gn S-Gn N-Me ANS-Me S-Go

3 Male 7.26 7.89 4.54 7.34 4.77 2.51 8.46

Female 3.16 3.08 2.26 2.73 2.41 1.07 2.84

Difference (male – female) 4.10‡ 4.81‡ 2.28‡ 4.61‡ 2.36‡ 1.44† 5.62‡

P 0.000 0.000 0.000 0.000 0.000 0.002 0.000

4 Male 2.25 3.75 2.38 2.94 1.25 0.81 4.38

Female 1.85 1.60 1.78 1.65 1.90 1.08 1.55

Difference (male – female) 0.40 2.15* 0.60 1.29 –0.65 –0.27 2.83‡

P 0.645 0.012 0.259 0.070 0.331 0.569 0.000

Follow-up was the time from T2 (end-of-treatment records) to T3 (long-term out-of-retention records).

*Statistically significant at P 5 0.05; †P 5 0.01; ‡P 5 0.001.


(equivalent to passage from CVM stages 3 to 4 in thisstudy, since they used the CVM stages 1 to 6 scale) byabout 2.9 mm in a pooled sample of 9 boys and 15 girls(P \0.05). Elongation amounts of the mandible (Co-Gn) in males and females between the correspondingstages in our study were 5.01 and 1.31 mm, respectively,and the change was statistically significant only inmales. The disparity of absolute values of mandibularlength increase between our findings and those of Fran-chi et al14 most likely results from a longer observationperiod in this investigation: 13.1 years rather than about1.5 years. Also, pooling the data for both sexes mighthave influenced the results.

Intersex comparisons demonstrated significant dif-ferences regarding all measured parameters solely inCVM stage 3. Mandibular length (Co-Gn), as well asother variables, increased at least twice as much inmales than females during follow-up. In CVM stage 4,2 variables—posterior ramus height (Co-Go) and poste-rior facial height (S-Go)—increased significantly morein male subjects compared with females. These findingssuggest that the data of the sexes should not be com-bined when the aim is to present mean changes of par-ticular variables over time. Perhaps, if the data formales and females had been separated by Franchiet al,14 a statistically significant change of mandibularlength between CVM stages 3 and 4 would have beenfound only for boys.

Surprisingly, chronologic ages in the CVM stages 3,4, and 5 groups were similar and equaled about 15 yearsfor both sexes. Only girls in CVM stage 3 and 4 showeda significant difference of approximately 7 months. Thissimilarity can be explained by large variations in chrono-logic ages for each CVM stage. Possibly, more subjectsin the groups would allow detection of more differences.Our results partially confirm the findings of Hassel andFarman,13 who attempted to correlate 2 methods of as-sessment of skeletal maturity: the hand-wrist evalua-tion21 and the CVM. They indirectly reported the agesof boys in CVM stages 3 and 4 to be 14.8 and 15.8 years,respectively. Girls in the analogous CVM stages, how-ever, were approximately 1.3 to 2 years younger whencompared with our group. The disagreement betweenthe findings of Hassel and Farman and ours might be be-cause they did not give directly the age of the subjectwith the same CVM stage, but reported what stage ofthe skeletal maturity index of Fishman21 correspondedto the appropriate CVM stage. Thus, the subjects’ agesmight be only indirectly obtained from Fishman’s study.However, correlation between the CVM and the skeletalmaturity index methods, although high (r2 5 0.89), wasnot perfect. Subjects in CVM stages 3 and 4 in the sam-ples of Franchi et al14 and Baccetti et al20 were 2 to 3

years younger than were our subjects. The possible ex-planation for the disagreement is the small sample sizesin those studies, especially when the distinctiveness ofthe sexes is considered.

The difficulty in assigning CVM stage 3, 4, or 5 toa subject might also explain the lack of substantial dif-ferences in age at the postpeak CVM stages. CVMstages 1 and 2 are not difficult to distinguish, since aneasily recognizable concavity at the lower border ofC2 or C3 differentiates them from more advancedstages. As Baccetti et al20 recommended, CVM stage3 is recognized when ‘‘concavities at the lower bordersof C2, C3, and C4 are present, and the bodies of both C3and C4 are rectangular horizontal in shape.’’ In case ofconflict between the CVM staging assumptions—eg,lack of concavity at the base of C4 and the rectangularhorizontal shape of C3 and C4 in some subjects in oursample—assigning the CVM stage is problematic. Sim-ilarly, CVM stage 4 should be characterized by thesquare shape of at least 1 body of C3 and C4. If notsquare, the body of the other cervical vertebra still oughtto be rectangular horizontal.20 By definition, a squareshape is equal length and height of the vertebral body.The application of such a strict criterion does notseem right for at least 2 reasons: error of the methodshould be considered, and positional changes of the cer-vical vertebral column are possible during cephalomet-ric x-ray exposure. Instead, an arbitrarily set border ofwhat is considered square must be applied. So, the dif-ference between horizontally rectangular, square, andvertically rectangular depend on the researcher’s arbi-trary decisions.

Ideally, to assess the ability of the CVM method todetect craniofacial growth alterations in CVM stages 3through 5, the longitudinal records taken annually untilthe age of 18 to 20 on large samples of both sexes shouldbe evaluated. The availability of such records is limited.Baccetti et al20 found only 18 boys and 12 girls withrecords 3 years before and 3 years after the pubertalgrowth peak of the 706 subjects with records at theUniversity of Michigan. Hassel and Farman13 and SanRoman et al22 used cross-sectional samples. We attemp-ted to use longitudinal posttreatment lateral headfilms tofind an association between craniofacial growth andCVM stage. Aweakness of this mixed longitudinal studydesign is the unknown amount of late postadolescentgrowth. According to many investigations, late growthchanges, however, are small.23-27 To additionally restrictthe influence of late growth alterations, groups with var-ious CVM stages had similar follow-up times.

When craniofacial growth is assessed on the postor-thodontic sample, the influence of posttreatment re-bound is difficult to measure. Driscoll-Gilliland et al28


January 2010

attempted to examine the difference between craniofa-cial growth in untreated and treated subjects. They eval-uated craniofacial growth in untreated (ages, 14.3-23.2years) and treated subjects (ages, 15.2-28.9 years) andfound that facial growth continued during follow-up.They demonstrated that both skeletal and dental changeswere similar in both groups, and the few statistically sig-nificant differences between groups were small.

A possible shortcoming of this study was that en-largement of each cephalogram could not be entirelycontrolled. Visual inspection and subsequent exclusionof the images taken in different cephalostats were prob-ably only partially effective. Also, including subjectswith various malocclusions might have affected theresults.

CONCLUSIONS

1. The CVM method was only modestly effective indetecting the amount of postpeak circumpubertalcraniofacial growth.

2. The CVM method discriminated only CVM stage 3from CVM stage 4 in boys for all measured variables.

3. The ages of the boys and girls in various CVMstages were about 15 years and 1 to 2 years olderthan the ages reported in other studies.

REFERENCES

1. O’Brien K, Wright J, Conboy F, Sanjie Y, Mandall N,

Chadwick S, et al. Effectiveness of early orthodontic treatment

with the Twin-block appliance: a multicenter, randomized, con-

trolled trial. Part 1: dental and skeletal effects. Am J Orthod Den-


2. Tulloch JF, Phillips C, Koch G, Proffit WR. The effect of early in-

terventiononskeletal pattern inClass IImalocclusion: a randomized

clinical trial. Am J Orthod Dentofacial Orthop 1997;111:391-400.

3. Keeling SD, Wheeler TT, King GJ, Garvan CW, Cohen DA,

Cabassa S, et al. Anteroposterior skeletal and dental changes after

early Class II treatment with bionators and headgear. Am J Orthod


4. Faltin KJ, Faltin RM, Baccetti T, Franchi L, Ghiozzi B,

McNamara JA Jr. Long-term effectiveness and treatment timing

for bionator therapy. Angle Orthod 2003;73:221-30.

5. Baccetti T, Franchi L, Toth LR, McNamara JA Jr. Treatment tim-

ing for Twin-block therapy. Am J Orthod Dentofacial Orthop

2000;118:159-70.

6. 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.e1-12.

7. Thilander B, Odman J, Grondahl K, Lekholm U. Aspects on os-

seointegrated implants inserted in growing jaws. A biometric and ra-

diographic study in the young pig. Eur J Orthod 1992;14:99-109.

8. Odman J, Grondahl K, Lekholm U, Thilander B. The effect of os-

seointegrated implants on the dento-alveolar development. A clin-

ical and radiographic study in growing pigs. Eur J Orthod 1991;

13:279-86.

9. Brugnolo E, Mazzocco C, Cordioll G, Majzoub Z. Clinical and ra-

diographic findings following placement of single-tooth implants

in young patients—case reports. Int J Periodontics Restorative

Dent 1996;16:421-33.

10. Hagg U, Taranger J. Maturation indicators and the pubertal

growth spurt. Am J Orthod 1982;82:299-309.

11. Gasser T, Sheehy A, Molinari L, Largo RH. Growth of early and

late maturers. Ann Hum Biol 2001;28:328-36.

12. Lamparski D. Skeletal age assessment utilizing cervical vertebrae

[thesis]. Pittsburgh: University of Pittsburgh; 1972.

13. Hassel B, Farman AG. Skeletal maturation evaluation using cer-

vical vertebrae. Am J Orthod Dentofacial Orthop 1995;107:

58-66.

14. Franchi L, Baccetti T, McNamara JA Jr. Mandibular growth as

related to cervical vertebral maturation and body height. Am J

Orthod Dentofacial Orthop 2000;118:335-40.

15. Grave K, Townsend G. Hand-wrist and cervical vertebral matura-

tion indicators: how can these events be used to time Class II treat-

ments? Aust Orthod J 2003;19:33-45.

16. Franchi L, Baccetti T, McNamara JA Jr. Treatment and posttreat-

ment effects of acrylic splint Herbst appliance therapy. Am J


17. 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 ap-

pliances. Am J Orthod Dentofacial Orthop 2003;123:306-20.

18. Baccetti T, Reyes BC, McNamara JA Jr. Craniofacial changes in

Class III malocclusion as related to skeletal and dental maturation.


19. Fudalej P, Rothe L, Bollen AM. Effects of posttreatment skeletal

maturity measured with the cervical vertebral maturation method

on incisor alignment relapse. Am J Orthod Dentofacial Orthop

2008;134:238-44.

20. Baccetti T, Franchi L, McNamara JA Jr. An improved version of

the cervical vertebral maturation (CVM) method for the assess-

ment of mandibular growth. Angle Orthod 2002;72:316-23.

21. Fishman LS. Maturational patterns and prediction during adoles-

cence. Angle Orthod 1987;57:178-93.

22. San Roman P, Palma JC, Oteo MD, Nevado E. Skeletal maturation

determined by cervical vertebrae development. Eur J Orthod

2002;24:303-11.

23. Forsberg CM. Facial morphology and ageing: a longitudinal ceph-

alometric investigation of young adults. Eur J Orthod 1979;1:

15-23.

24. Bhatia SN, Leighton BC. A manual of facial growth: a computer

analysis of longitudinal cephalometric growth data. Oxford:

Oxford University Press; 1993. p. 518-43.

25. Bondevik O. Growth changes in the cranial base and the face:

a longitudinal cephalometric study of linear and angular changes

in adult Norwegians. Eur J Orthod 1995;17:525-32.

26. Akgul AA, Toygar TU. Natural craniofacial changes in the third

decade of life: a longitudinal study. Am J Orthod Dentofacial

Orthop 2002;122:512-22.

27. Thilander B, Persson M, Adolfsson U. Roentgen-cephalometric

standards for a Swedish population. A longitudinal study between

the ages of 5 and 31 years. Eur J Orthod 2005;27:370-89.

28. Driscoll-Gilliland J, Buschang PH, Behrents RG. An evaluation of

growth and stability in untreated and treated subjects. Am J



ORIGINAL ARTICLE

Midpalatal miniscrews for orthodonticanchorage: Factors affecting clinical success

Young Ho Kim,a Seung-Min Yang,b Seonwoo Kim,c Joo Yong Lee,d Kyu Eok Kim,e Anthony A. Gianelly,f

and Seung-Hyun Kyungg

Seoul, Korea, and Boston, Mass

Introduction: The purpose of this study was to investigate the success rate of midpalatal miniscrews used fororthodontic anchorage and the factors affecting clinical success. Methods: One hundred twenty-eightconsecutive patients (101 female, 27 male; mean age, 23.4 years), who received a total of 210 miniscrewsin the midpalatal suture area, were examined. Success rates were determined according to 10 clinical vari-ables. Results: The overall success rates were 88.20% for the total number of patients and 90.80% for thetotal number of miniscrews. There were no significant associations among success rate and sex, total periodof treatment with miniscrews, diameter of miniscrews, types of tooth movements, and variables that representsagittal and vertical skeletal relationships (ANB, FMA, and Sn-GoGn). The operator’s learning curve, patient’sage, area (midpalatal or parapalatal), and splinting significantly influenced the success rates. After adjustingfor other variables, only 1— splinting—showed a significant effect on the success rate. Conclusions: The join-ing of 2 miniscrews by splinting, placement of the miniscrew in the midpalatal suture, patient’s age (especially.15 years), and operator’s skill were factors influencing the clinical success of orthodontic miniscrews in thepalate. (Am J Orthod Dentofacial Orthop 2010;137:66-72)

Since the introduction of implants as absoluteanchorage in orthodontic treatment, varioustypes of tooth movement without patient com-

pliance have become possible with newly developedminiscrews.1-3

Of the possible placement sites for miniscrews, themidpalatal area has been reported to be appropriate.4-8

The midpalatal suture is a highly dense structure withsufficient bone height up to the cresta nasalis,9,10 andvertical bone support is somewhat higher (at least 2mm) than is apparent on cephalograms.7 The midpalatalarea within 1 mm of the midsagittal suture is composedof the thickest bone available in the whole palate,11 andthe thickness of soft tissues in the midpalatal area isuniformly 1 mm posterior to the incisive papilla,10

ensuring biomechanical stability of the miniscrews.There are no roots, nerves, or blood vessels to compli-cate the placement of surgical miniscrews, and there isno need for additional surgery because of their easy re-moval.12 Miniscrews have been placed in the midpalatalsuture area of adults, and the parapalatal area in adoles-cents to prevent possible developmental disturbances ofthe midpalatal sutures.13 This is because the transversegrowth of the midpalatal suture continues up to the lateteens14 and is not fused completely even in adults.15,16

Various attempts to use implants as absolute anchor-age in the midpalatal suture area have been made.17-19

Now, midpalatal miniscrews are used for retraction ofmaxillary anterior teeth,20,21 intrusion,22-25 distaliza-tion,12 and protraction of maxillary posterior teeth,making it possible to produce movements that were, atbest, difficult with conventional orthodontic treatmentstrategies.

The purpose of this study was to investigate thesuccess rate of midpalatal miniscrews used as orthodon-tic anchorage for various types of tooth movements andfactors affecting clinical success.

aAssociate professor, Department of Orthodontics, Institute of Oral Health &

Science, Samsung Medical Center, Sungkyunkwan University School of

Medicine, Seoul, Korea.bAssociate professor, Department of Periodontics, Institute of Oral Health &

Science, Samsung Medical Center, Sungkyunkwan University School of

Medicine, Seoul, Korea.cSenior statistician, Biostatistics Unit, Samsung Biomedical Research Institute,

Samsung Medical Center, Sungkyunkwan University School of Medicine,

Seoul, Korea.dPrivate practice, Seoul, Korea.eResident, Department of Orthodontics, Institute of Oral Health & Science, Sam-

sung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea.fProfessor and chairman emeritus, Department of Orthodontics, Goldman School of

Dental Medicine, Boston University, Boston, Mass; now deceased.gAssociate professor, Department of Orthodontics, Institute of Oral Health & Sci-

ence, Samsung Medical Center, Sungkyunkwan University School of Medicine,

Seoul, Korea; visiting professor, Department of Orthodontics, Goldman School

of Dental Medicine, Boston University, Boston, Mass.

The authors report no commercial, financial, or proprietary interest in the prod-


Reprint requests to: Seung-Hyun Kyung, Department of Orthodontics, Institute

of Oral Health & Science, Samsung Medical Center, Sungkyunkwan University

School of Medicine, #50, Irwon-dong, Gangnam-Gu, Seoul 135-710, Korea;

e-mail, [email protected].

Submitted, August 2007; revised and accepted, November 2007.

0889-5406/$36.00


doi:10.1016/j.ajodo.2007.11.036

66



The subjects were 128 patients (101 female, 27male; age range, 8.1-56.2 years; mean age, 23.4 6 8.0years), who received miniscrews for orthodonticanchorage in the midpalatal area. All miniscrews wereplaced by 1 doctor (S.H.K.) at the Department of Ortho-dontics, Samsung Medical Center, Seoul, Korea,between 1999 and 2005. The patients were informedabout the possibilities of inflammation around andloosening of the miniscrews.

A total of 210 miniscrews were placed without flap el-evation under local anesthesia. They were placed in themidpalatal sutures and, sagittally, between the mesialand distal aspects of the maxillary first molar. In adoles-cents, miniscrews were placed in the parapalatal area in-stead of the midpalatal areas to prevent possible damageto the developing sutures. An orthodontic force was ap-plied immediately with elastomeric modules (powerchain), and they were replaced every 3 weeks.

Two types of self-drilled miniscrews of the samelength but different diameters were used (Fig 1). Onewas a surgical miniscrew (diameter, 1.5 mm; length,5.0 mm; KLS-Martin, Jacksonville, Fla) that oral sur-geons usually use for fixation of bone fragments. Theother was especially designed (diameter, 2.0 mm;length, 5.0 mm; Orthoplant, Biomaterials Korea, Seoul,Korea) and developed for orthodontic anchorage at themidpalatal area. Its head diameter was larger (4.0 mm)so that it provided a wide contact area betweenits head and a screw-supported bonded sheath(S-sheath), generating higher bonding strength suffi-cient to resist heavy orthodontic forces. The sheathwas custom-made with a normal lingual sheath weldedonto metal mesh.25

Although some patients were treated with 1 mini-screw as orthodontic anchorage, most patients, espe-cially adolescents and young adults, were treated with2 miniscrews splinted together to ensure stability. TheS-sheath was used because a heavy orthodontic forcewas needed to control several maxillary posterior teethsimultaneously. Splinting was done by simply bondingthe S-sheath on the top of 2 miniscrews with flowablecomposite resin. After placing the miniscrews, 500 to800 g of force was applied initially for various typesof tooth movement: distalization, mesialization, intru-sion, or retraction of anterior teeth, either singly or incombination. Elastomeric chains generally lose 50%to 70% of their initial force during the first day ofload application and, at 3 weeks, retain only 30% to40% of the original force.26-28 Therefore, we believedthat 250 to 400 g might be loaded over 3 to 7 teeth afterforce degradation of the elastomers.

Figure 2 shows the orthodontic mechanics for sometypes of tooth movements by single or splinted minis-crews with or without an extension arm: distalizationof maxillary molars, mesialization of maxillary molars,intrusion of maxillary molars, and retraction of anteriorteeth.

The procedure was regarded as a clinical successwhen a miniscrew remained without loosening until ithad accomplished its purpose. To examine the factorsaffecting the clinical success of midpalatal miniscrewsfor orthodontic anchorage, 10 clinical variables wereinvestigated: operator’s learning curve (determined bycalculating the success rate of miniscrews placed bythe operator over 4 time periods of 18 months each),sex, age, area (midpalatal or parapalatal), total treat-ment period with miniscrews, splinting (single screwvs joined screws), diameter of miniscrew (1.5 vs 2.0mm), types of tooth movements (distalization, mesiali-zation, intrusion, retraction of anterior teeth, and combi-nations), and variables representing sagittal and verticalskeletal relationships (ANB, FMA, and Sn-GoGn). Inprevious studies, there was no evaluation of the clini-cian’s skill as a factor influencing success rates. There-fore, this study included an evaluation over time ofwhether the operator’s increasing experience with theprocedure was a factor that affected the stability ofminiscrews.


Success rates related to the numbers of subjects andminiscrews were calculated. The success rate was also

Fig 1. Two types of miniscrews used in the study: left,cylindrical type (diameter, 1.5 mm; length, 5.0 mm;head diameter, 3.0 mm); right, tapered type (diameter,2.0 mm; length, 5.0 mm; head diameter, 4.0 mm).

American Journal of Orthodontics and Dentofacial Orthopedics Kim et al 67Volume 137, Number 1

presented for each category of clinical variable. Acontinuous clinical variable was categorized to providethe success rate. Logistic regression analysis was usedto examine the influence of each of 10 clinical variables(categorical variables and continuous variables withoutcategorization) on success. Multiple logistic regressionanalysis was also used to investigate the influence ofeach variable when the effects of other variables werecontrolled. The odds ratio (OR) for each factor wasalso calculated.

RESULTS

Fifteen of the 128 patients had at least 1 miniscrewfailure, for a success rate of 88.20%. Eighteen of the 197miniscrews failed, for a success rate of 90.80%. Theaverage time after placement for miniscrew failurewas 3.5 months.

Logistic regression analysis showed no significantassociation between the success rate and each of follow-ing variables: sex, total period of treatment with minis-crews, diameter of miniscrew, types of tooth movement,and variables representing sagittal and vertical skeletalrelationships (ANB, FMA, and Sn-GoGn).

Six of 51 miniscrews placed in 27 male patientswere recorded as failures, for a success rate of 88.2%.In 101 female patients, 12 of 146 miniscrews failed, rep-resenting a success rate of 91.8%. There was no statisti-

cally significant difference in the success rates betweenthe sexes (OR 5 0.70; P 5 0.4517).

The total period of treatment with miniscrews variedaccording to the purpose of orthodontic treatment.Although some patients were treated for less than6 months with miniscrews, other treatments lastedmore than 18 months. When a miniscrew failed beforeit achieved its purpose, it was replaced. In these cases,the total period of treatment with miniscrews was calcu-lated by the sum of the periods during which the 2 min-iscrews were loaded. In the group of patients withrelatively short treatments (\6 months), 3 of 11 minis-crews failed, for a success rate of 72.7%. In the groupwith longer treatments (.18 months), some patientsneeded more than 1 miniscrew because of the earlyloss of the first miniscrew, and 11 of 111 miniscrewsfailed, for a success rate of 90.1%. Even though the dif-ference in success rates between the groups with shorterand longer treatment periods (72.7% and 90.1%-95.1%,respectively) was high, there was no statistical signifi-cance between the treatment period and the successrate (OR 5 0.93; P 5 0.2386).

Similarly, the success rate for miniscrews witha 2.0-mm diameter (91.9%) was slightly higher thanfor those with a 1.5-mm diameter (89.0%), but the diam-eter of the miniscrews was not a significant factor.

With the midpalatal miniscrews, various kinds oftooth movements could be made: distalization, mesiali-zation, intrusion, retraction of anterior teeth, and

Fig 2. Midpalatal miniscrew placements: A, distalization of maxillary molars; B, mesialization of max-illary molars; C, intrusion of maxillary molars; D, retraction of anterior teeth.

68 Kim et al American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

combinations of these. Although these movementsrequire heavy forces of over 500 g, the success ratesvaried from 88.7% to 96.0%; there were no statisticallysignificant differences.

The success rate according to the ANB differencerepresenting sagittal skeletal relationships varied from89.1% (Class I) to 97.1% (Class III), but there was nosignificant association between the ANB value and thesuccess rate (OR 5 1.05). In patients whose FMA andSn-GoGn represented vertical skeletal relationships,those with high angles had high success rates of97.4% (FMA) and 97.1% (Sn-GoGn), but there werealso no significant associations (OR 5 1.03 [FMA)];1.02 [Sn-GoGn]).

There were significant associations between the suc-cess rate and the following variables: operator’s learn-ing curve, age, area, and splinting. The operator’slearning curve, which indicated his skill or experienceover time, had a significant association with the successrate; the longer his learning experience, the higher thesuccess rate (OR 5 1.60; P 5 0.0132). During the first18 months, when the operator was not accustomed to theprocedure for placing miniscrews in the palate, 9 of 36miniscrews failed, a success rate of 75%. This was muchlower than the rates of later periods (91.2%-97.9%); thesuccess rate increased to more than 95% after the thirdperiod of 18 months.

Age was also associated with the success rate, andthe logistic regression analysis showed that olderpatients had higher success rates (OR 5 1.01; P 5

0.0249). Notably, in the group of patients less than 15years of age, 9 of 31 miniscrews failed, for a successrate of 71.0%, which was much lower than rates forthe older groups (92.9%-100%).

In the parapalatal area, 5 of 24 miniscrews failed,whereas in the midpalatal area, only 13 of 173 failed.Thus, the miniscrews in the parapalatal area showeda significantly lower success rate (79.2%) than thosein the midpalatal area (92.5%) (OR 5 2.77;P 5 0.0426).

Splinting the 2 miniscrews produced a higher suc-cess rate (95.9%) than use of 1 miniscrew (82.4%)(OR 5 0.23; P 5 0.0033), and splinting was also theonly clinical variable that showed a significant associa-tion with the success rate (OR 5 0.09; P 5 0.013) aftercontrolling for the effects of the other variables (Table).

DISCUSSION

The criterion used to define the success rates forminiscrews in previous studies was how long they lastedunder loading—eg, 6 months,29 10 to 12 months,30 and1 year.31,32 However, the criterion in this study

depended on the achievement of purpose, not on theamount of time that the miniscrews lasted as anchorage.According to this criterion, 5 miniscrews were recordedas successes, although they lasted less than 6 months.On the other hand, 4 miniscrews were regarded as fail-ures even though they lasted more than 1 year becausethey did not complete their missions.

Most previous studies reported success rates on thebasis of the total number of miniscrews; however, thesuccess rate based on the total number of patients isalso meaningful. In this study, the success rates were88.20% for the total number of patients and 90.80%for the total number of miniscrews; these rates aresimilar to those reported for buccal miniscrews.29-33

The applied force was initially 500 to 800 g per mini-screw, although a miniscrew placed in the buccal bonecan withstand 200 to 250 g. Therefore, consideringthe amount of applied force and the success rates inthis study, we recommend the midpalatal area as anappropriate site to obtain a strong orthodontic anchor-age for group movements of maxillary teeth.

Many studies have reported the success rates forminiscrews placed during several years without consid-ering improvements of the operator’s skills. When theoperator in this study (S.H.K.) was a novice in placingthe midpalatal miniscrews, his success rate for the first18 months was 75%. Thereafter, his success rateincreased to over 90% and remained over 90% untilthe last period of 18 months. This result indicates thatthe operator’s skill or experience is critical to the suc-cess rate, and this finding might also be true for labialminiscrew applications.

Many studies have found no significant differencesbetween success rate and age, but, in this study, youngerpatients, especially those less than 15 years of age, hada higher failure rate than did older age groups.29-32 Thismight be attributed to a difference in bone densitybecause calcification of bone is not completed in adoles-cents, or a difference in area because miniscrews wereusually placed in the parapalatal area in adolescents.The midpalatal area has sufficient bone height for theplacement of miniscrews, although, even in adults, thereis low bony obliteration or fusion of the midpalatalsuture.15,16 However, in growing children and adoles-cents, the parapalatal area is recognized as an alterna-tive.34,35 Miniscrews placed in the parapalatal areahad a significantly higher failure rate, and there were3 failures in 1 adolescent patient. Thus, although cau-tion is required in the placement of miniscrews in theparapalatal area in adolescents, this procedure is notcontraindicated in patients younger than 15 years old.

The most important factor contributing to thesuccess rate of miniscrews in the midpalatal area was


Table. Success rates and number of loosened miniscrews according to 10 clinical variables

Logistic regression Multiple logistic regression

Clinical variable Miniscrews (n) Loosened miniscrews (n) Success rate (%) OR P value OR P value

Operator’s learning

curve (mo)

1.60 0.0132 1.33 0.2911

First 18 36 9 75.0

Second 18 68 6 91.2

Third 18 47 1 97.9

Fourth 18 46 2 95.7

Sex 0.4517 0.7164

Male 51 6 88.2 0.70 0.71

Female 146 12 91.8 1.00 1.00

Age (y) 1.01 0.0249 3.23 0.1269

\15 31 9 71.0

\20 32 0 100.0

\30 112 8 92.9

.30 22 1 95.5

Area 0.0426 0.2716

Midpalatal 173 13 92.5 2.77 3.23

Parapalatal 24 5 79.2 1.00 1.00

Total treatment period

using miniscrews (mo)

0.93 0.2386 0.94 0.4569

\6 11 3 72.7

6-12 41 2 95.1

12-18 34 2 94.1

.18 111 11 90.1

Splinting 0.23 0.0033 0.09 0.013

Single 74 13 82.4

Splint 123 5 95.9

Miniscrew diameter (mm) 0.74 0.4975 0.33 0.2674

1.5 73 8 89.0

2.0 124 10 91.9

Tooth movements 0.7875 0.6193

Distalization 55 4 92.7 0.55 0.56

Mesialization 10 1 90.0 0.40 0.28

Intrusion 36 4 88.9 0.36 0.39

Retraction of anterior

teeth

71 8 88.7 0.36 0.57

Combination 25 1 96.0 1.00 1.00

Sagittal skeletal

relationship

ANB (�) (mean) 1.05 0.5302 1.16 0.2195

Class I (3.28) 128 14 89.1

Class II (7.42) 35 3 91.4

Class III (–1.91) 34 1 97.1

Vertical skeletal

relationship

FMA (�) (mean) 1.03 0.3814 1.08 0.4444

Low angle (20.16) 36 4 88.9

Middle angle (29.5) 122 13 89.3

High angle (39.8) 39 1 97.4

Sn-GoGn (�) (mean) 1.02 0.5129 1.02 0.8381

Low angle (28.3) 30 3 90.0

Middle angle (38.0) 138 14 89.9

High angle (48.4) 34 1 97.1


January 2010

splinting. In the multiple regression analysis, splintingwas the only clinical variable that showed a significantdifference in the success rate, and this result stronglysuggests that the stability of midpalatal miniscrewscan be further enhanced by splinting 2 miniscrews.Although in some patients 1 of the 2 splinted minis-crews loosened, no patient had both splinted miniscrewsloosen simultaneously. It was difficult to detect loosen-ing of the splinted screws because the 2 miniscrewswere splinted firmly with composite resin and anS-sheath. The only sign indicating loosening was thecontinuous growth of inflammatory tissue around theS-sheath. We found that, unlike inflammation frompoor oral hygiene, inflammation caused by loose minis-crews was not controlled with improved oral hygiene.Therefore, we believe that inflammation or swellingaround a miniscrew might be a result of its looseningrather than a cause.

When screws were placed in the midpalatal area,there was no significant association between successrate and sex; this agrees with previous reports.29-32

The stability of a miniscrew increases as its diameterincreases; theoretically, this is because the applied forcecan be distributed over more bone area, resulting indecreased pressure. Petrie and Williams36 emphasizedthat the diameter of the implant was critical in decreas-ing the amount of crestal strain. In this study, no signif-icant difference was seen between miniscrews of 1.5and 2.0 mm in diameter; this might be because excellentbone quality and quantity could have masked the differ-ence in diameters. The total period of treatment withminiscrews was included as a variable in this study toinvestigate whether the duration of exposure to an ortho-dontic force influenced their success rates. If so, moreminiscrews might be expected to fail as the treatmentperiod increased. However, our results indicated thatthe length of the period under loading did not influencethe success rate of the miniscrews.

During treatment, alterations of force vectors wereneeded when the teeth did not move as expected.Force vectors could be adjusted easily by bending orrefabricating the mesially extended-TPA and distalarms, which were removable and placed in the S-sheath.25 The type of tooth movement determinedthe direction of force received by the miniscrews.Therefore, the type of tooth movement was selectedas a clinical variable to determine whether the specificdirections of the force vectors are associated with thestability of miniscrews. The results indicated that theywere not a factor related to the miniscrew success rate;this meant that the miniscrews were fixed evenly in 3dimensions and were not more resistant to any partic-ular direction of load.

There have been reports that the mandibular planeangle may be related31 or not related32 to the successrates of miniscrews. In this study, sagittal and verticalskeletal variables also did not show significant differ-ences in the success rate, supporting the midpalatalapproach for miniscrews used as absolute anchorage,regardless of a patient’s sagittal and vertical skeletalpattern.

CONCLUSIONS

The overall success rates of midpalatal miniscrewswere 88.20% for the total number of patients and90.80% for the total number of miniscrews under aninitial load of 500 to 800 g per miniscrew. The midpala-tal area is appropriate for miniscrews, and midpalatalminiscrews can serve as absolute orthodontic anchoragefor various types of tooth movements with high successrates. Factors that influenced the clinical successof miniscrews in the palate were splinting 2 miniscrews,placement of the miniscrews in the midpalatal suture,the patient’s age (especially .15 years), and the opera-tor’s skill.

REFERENCES

1. Creekmore TD, Eklund MK. The possibility of skeletal anchor-

age. J Clin Orthod 1983;17:266-9.

2. Kanomi R. Mini-implant for orthodontic anchorage. J Clin Orthod

1997;31:763-7.

3. Park HS. The skeletal cortical anchorage using titanium micro-

screw implants. Korean J Orthod 1999;29:699-706.

4. Kyung SH, Lim JK, Park YC. The use of miniscrew as an anchor-

age for the orthodontic tooth movement. Korean J Orthod 2001;

31:415-24.

5. Block MS, Hoffman DR. A new device for absolute anchorage

for orthodontics. Am J Orthod Dentofacial Orthop 1995;107:

251-8.

6. Wehrbein H, Merz BR, Diedrich P, Glatzmaier J. The use of

palatal implants for orthodontic anchorage. Design and clinical

application of the orthosystem. Clin Oral Implants Res 1996;7:

410-6.

7. Wehrbein H, Merz BR, Diedrich P. Palatal bone support for ortho-

dontic implant anchorage—a clinical and radiological study. Eur J

Orthod 1999;21:65-70.

8. Wehrbein H, Feifel H, Diedrich P. Palatal implant anchorage rein-

forcement of posterior teeth: a prospective study. Am J Orthod


9. Kyung SH. A study on the bone thickness of midpalatal suture

area for miniscrew insertion. Korean J Orthod 2004;34:63-70.

10. Kim HJ, Yun HS, Park HD, Kim DH, Park YC. Soft-tissue and

cortical-bone thickness at orthodontic implant sites. Am J Orthod


11. Kang S, Lee SJ, Ahn SJ, Heo MS, Kim TW. Bone thickness of the

palate for orthodontic mini-implant anchorage in adults. Am J

Orthod Dentofacial Orthop 2007;131(Suppl 4):S74-81.

12. Kyung SH, Hong SG, Park YC. Distalization of maxillary molars

with a midpalatal miniscrew. J Clin Orthod 2003;37:22-6.


13. Asscherickx K, Hanssens JL, Wehrbein H, Sabzevar MM. Ortho-

dontic anchorage implants inserted in the median palatal suture

and normal transverse maxillary growth in growing dogs: a bio-

metric and radiographic study. Angle Orthod 2005;75:826-31.

14. Melsen B. Palatal growth studied on human autopsy material.

A histologic microradiographic study. Am J Orthod 1975;68:

42-54.

15. Wehrbein H, Yildizhan F. The mid-palatal suture in young adults.

A radiological-histological investigation. Eur J Orthod 2001;23:

105-14.

16. Knaup B, Yildizhan F, Wehrbein H. Age-related changes in the

midpalatal suture. A histomorphometric study. J Orofac Orthop

2004;65:467-74.

17. Wehrbein H, Merz BR. Aspects of the use of endosseous palatal

implants in orthodontic therapy. J Esthet Dent 1998;10:315-24.

18. Byloff FK, Karcher H, Clar E, Stoff F. An implant to eliminate

anchorage loss during molar distalization: a case report involving

the Graz implant-supported pendulum. Int J Adult Orthod Orthog-

nath Surg 2000;15:129-37.

19. Karaman AI, Basciftci FA, Polat O. Unilateral distal molar move-

ment with an implant-supported distal jet appliance. Angle

Orthod 2002;72:167-74.

20. Park YC, Choi YJ, Choi NC, Lee JS. Esthetic segmental retraction

of maxillary anterior teeth with a palatal appliance and orthodon-

tic mini-implants. Am J Orthod Dentofacial Orthop 2007;131:

537-44.

21. Hong RK, Heo JM, Ha YK. Lever-arm and mini-implant system

for anterior torque control during retraction in lingual orthodontic

treatment. Angle Orthod 2005;75:129-41.

22. Lee JS, Kim DH, Park YC, Kyung SH, Kim TK. The efficient

use of midpalatal miniscrew implants. Angle Orthod 2004;74:

711-4.

23. Chang YJ, Lee HS, Chun YS. Microscrew anchorage for molar

intrusion. J Clin Orthod 2004;38:325-30.

24. Paik CH, Woo YJ, Boyd RL. Treatment of an adult patient with

vertical maxillary excess using miniscrew fixation. J Clin Orthod

2003;37:423-8.

25. Choi KJ, Choi JH, Lee SY, Ferguson DJ, Kyung SH. Facial

improvements after molar intrusion with miniscrew anchorage.

J Clin Orthod 2007;41:273-80.

26. Ash JL, Nikolai RJ. Relaxation of orthodontic elastomeric chains

and modules in vitro and in vivo. J Dent Res 1978;57:685-90.

27. Baty DL, Storie DJ, von Fraunhofer JA. Synthetic elastomeric

chains: a literature review. Am J Orthod Dentofacial Orthop

1994;105:536-42.

28. Kim KH, Chung CH, Choy K, Lee JS, Vanarsdall RL. Ef-

fects of prestretching on force degradation of synthetic elas-

tomeric chains. Am J Orthod Dentofacial Orthop 2005;128:

477-82.

29. Motoyoshi M, Hirabayashi M, Uemura M, Shimizu N. Recom-

mended placement torque when tightening an orthodontic mini-

implant. Clin Oral Implants Res 2006;17:109-14.

30. Park HS, Jeong SH, Kwon OW. Factors affecting the clinical

success of screw implants used as orthodontic anchorage. Am J


31. Miyawaki S, Koyama I, Inoue M, Mishima K, Sugahara T,

Takano-Yamamoto T. Factors associated with the stability

of titanium screws placed in the posterior region for ortho-

dontic anchorage. Am J Orthod Dentofacial Orthop 2003;

124:373-8.

32. Kuroda S, Sugawara Y, Deguchi T, Kyung HM, Takano-

Yamamoto T. Clinical use of miniscrew implants as orthodontic

anchorage: success rates and postoperative discomfort. Am J


33. Cheng SJ, Tseng IY, Lee JJ, Kok SH. A prospective study of the

risk factors associated with failure of mini-implants used for

orthodontic anchorage. Int J Oral Maxillofac Implants 2004;19:

100-6.

34. Bernhart T, Vollgruber A, Gahleitner A, Dortbudak O, Haas R.

Alternative to the median region of the palate for placement

of an orthodontic implant. Clin Oral Implants Res 2000;11:

595-601.

35. Bernhart T, Freudenthaler J, Dortbudak O, Bantleon HP,

Watzek G. Short epithetic implants for orthodontic anchorage in

the paramedian region of the palate. A clinical study. Clin Oral

Implants Res 2001;12:624-31.

36. Petrie CS, Williams JL. Comparative evaluation of implant

designs: influence of diameter, length, and taper on strains in

the alveolar crest. A three-dimensional finite-element analysis.

Clin Oral Implants Res 2005;16:486-94.


January 2010

ORIGINAL ARTICLE

Miniscrew stability evaluated with computerizedtomography scanning

Jung-Yul Cha,a Jae-Kyoung Kil,b Tae-Min Yoon,c and Chung-Ju Hwangd

Seoul, Korea

Introduction: In this study, we aimed to determine the effect of bone mineral density (BMD), cortical bonethickness (CBT), screw position, and screw design on the stability of miniscrews. Methods: Ninety-six mini-screws of both cylindrical and tapered types were placed in 6 beagle dogs. The BMD and CBT were measuredby computerized tomography and correlated with the placement and removal torque and mobility. A regres-sion equation to predict the placement torque was calculated based on BMD, CBT, screw type, and screwposition. Results: The placement torque showed a positive correlation in the order of removal torque(0.66), BMD of the cortical bone (0.58), and CBT (0.48). Placement and removal torque values were signifi-cantly higher in the mandible compared with the maxilla. Tapered miniscrews had higher placement torquethan did the cylindrical type (P \0.001). However, the removal torque was similar in both groups. Placementtorque was affected by screw position, screw type, and BMD of cortical bone, in that order. Conclusions:BMD of cortical bone, screw type, and screw position significantly influence the primary stability of mini-screws. (Am J Orthod Dentofacial Orthop 2010;137:73-9)

Easy placement is an advantage of orthodonticminiscrews. In addition, the tooth can be movedwithout the patient’s cooperation; this enhances

treatment efficiency. Some success has been reportedregarding the use of miniscrews as orthodontic anchor-ages. However, the clinical application of a miniscrewdoes not guarantee treatment success, and its stabilityis essential before it can be used as orthodonticanchorage.1-3

It is important to ensure the initial stability of anorthodontic miniscrew because most failures occurduring the initial stage. Furthermore, stability of theminiscrew in the initial stage reduces its micromove-ment, thus allowing for an appropriate environmentthat supports the healing process surrounding the bone.The success of an implant or miniscrew is determinedby the patient’s general condition, the biocompatibilityof the materials, the placement procedure, and bone

quantity and quality.4,5 The initial stability of minis-crews is considered essential in clinical use because ofimmediate or early loading in many patients.6

Two factors affect the initial stability of a screw: thescrew factor and the host factor. The screw factor isrelated to the characteristics of the screw design, includ-ing diameter and length.7 Various screw designs havebeen introduced to enhance initial stability.8 The goalis to increase initial fixation by inducing controlledcompressive forces in the cortical bone layer. In theprosthodontic field, it was reported that a tapered shapecan enhance stability for immediate loading by increas-ing the mechanical contact between the dental implantand the surrounding bone.9

The host factor is related to the quantity and qualityof the bone where the screw is placed. Cortical bonethickness (CBT) (quantity) can affect the initial stabilityof a screw.10 Finite element analysis showed that, whena lateral force is applied to the screw, most of the force isconcentrated on the cortical bone.11,12 For this reason,the previous studies focused on the anatomic back-ground related to CBT rather than bone quality, toimprove the stability of the miniscrew.

On the other hand, several studies have used com-puted tomography (CT) in prosthodontics and orthope-dics to study factors related to bone density.13-16 It wasreported that bone density differs according to sex, age,and physical condition. CT was recently suggested tosuccessfully measure bone mineral density (BMD) inorthopedics. BMD has been used as a parameter toestablish a treatment plan to ensure the stability ofimplants in dentistry.17

aAssistant professor, Department of Orthodontics, College of Dentistry, Yonsei

University, Seoul, Korea.bPrivate practice, Seoul, Korea.cPostgraduate student, Department of Orthodontics, College of Dentistry,

Yonsei University, Seoul, Korea.dProfessor, Department of Orthodontics, Dental Science Research Institute, The

Institute of Craniofacial Deformities, College of Dentistry, Yonsei University,

Seoul, Korea.



Reprint requests to: Chung-Ju Hwang, Department of Orthodontics, College of

Dentistry, Yonsei University, 134 Shinchon-dong, Seodaemun-gu, Seoul 120-

752, Korea:; e-mail, [email protected].

Submitted, August 2007; revised and accepted, March 2008.

0889-5406/$36.00


doi:10.1016/j.ajodo.2008.03.024

73


There are 2 methods for measuring BMD. One iswith Hounsfield units (HU), and the other is to measureBMD indirectly by using a hydroxyapatite block. SinceHounsfield units can be affected by the loaded voltageof x-rays and the protocol used, bone density analysiswith a BMD calibration standard (hydroxyapatiteblock) has greater advantages for evaluating bone qual-ity more accurately.18

One assessment method of initial stability is tomeasure the torque during placement.9,13,19 Cheng etal19 proposed that placement torque could be used asan indicator for initial stability, and, to obtain initial sta-bility, a certain amount of torque is necessary. Place-ment torque is the measurement of the resistance atthe screw-bone interface; it reflects the level of bonedeformational strain caused by the miniscrew.

In this study, we hypothesized that BMD values canbe used to predict the mechanical stability of minis-crews by evaluating placement torque. Our aim was todetermine the effect of BMD, CBT, screw position,and screw design on the stability of miniscrews.


For this study, miniscrews were placed in 6 beagledogs (age, 1 year; weight, 12 kg). Their purchase,selection, and management, and the experimental pro-cedures were carried out according to prescribed condi-tions of the institutional review board, the AnimalExperiment Committee of Yonsei Hospital, Seoul, Ko-rea. A nondrilling type of miniscrew, 1.4 mm in diam-eter and 7 mm in length, was used. Both the cylindricaltype (OAS-1507C) and tapered type (OAS-1507 T,Biomaterials Korea, Seoul, Korea) were selected; 96screws were used (Fig 1).

The location of the orthodontic miniscrew place-ment was determined after evaluating the reconstructedCT image, taken before placement, to determinewhether there was sufficient interdental space. Sites se-lected were between the roots of the second, third, andfourth premolars, and the first molar in the mandible,between the roots of the second and third premolars,the first molar, and between the second and third premo-lars in the maxilla (Fig 2).

The animals were injected subcutaneously with 0.05mg per kilogram of atropine followed by an intravenousinjection of rompun, 2 mg per kilogram, and ketamine,10 mg per kilogram, to induce general anesthesia. Theanesthesia was maintained with 2% enflurane, andeach animal’s temperature was maintained with a heat-ing pad and an electrocardiogram, and monitored.When placing the mini-implant, 2% hydrochloric acidlidocaine containing 1:100,000 epinephrine was infil-

trated into the placement area. Before placement,a 5-to-10 mm gingival incision was made under saline-solution irrigation, and complete placement of the screwinto the alveolar bone was confirmed.

Screw placement was performed manually with 70�

to 90� angulations to gingival surface in considerationof the buccolingual width of alveolar bone for eachexperiment. In all miniscrews, a force of 250 to 300 gwas applied with an elastomeric chain engaged recipro-cally from a cylindrical miniscrew to a tapered mini-screw immediately after the placement of miniscrews,and the elastic chain was exchanged every 3 weeks.

Screw mobility was measured twice on each mini-screw by using a periotest (Simens AG, Bensheim, Ger-many) before removing the screw. The sleeve of thehandpiece of the periotest was positioned with 1 to 2mm from the screw head perpendicularly after an elas-tomeric chain was removed. The average of 2 measure-ments for a miniscrew was recorded as the mobilityvalue. During the test period, a chlorhexidine solutionwas applied daily to maintain the animals’ oral hygiene(Fig 3).

The placement torque was the highest (in newtonsper square centimeter) when the miniscrew was placedcompletely into the bone. The highest removal torquewas measured during a quarter initial turn by using a tor-que sensor (MGT50, Mark-10 Co, New York, NY).

To measure the BMD of the implanted area, CTscanning was done 3 days before placing and removingthe miniscrews. Before scanning, each dog was placedunder intramuscular sedation. The gantry of the CTdevice was placed parallel to the occlusal plane, andthe dog’s head was fixed with a strap. The CT scanwas performed with a high-speed advantage CT scanner(GE Medical System, Milwaukee, Wis) by usingstandard CT protocol (high-resolution algorithm,512 3 512 matrix, 120 Kv, 200 mA) at a table feed of6 mm per second. The CT data were reconstructed to1-mm thick transaxial images.

A calibration standard (Dental Phantom, ImageAnalysis, Columbia, Ky) was applied to calibrate thelevel in Hounsfield units in the dogs during the CT scan-ning. The calibration standard containing 3 compart-ments with 0, 75, and 150 mg of hydroxyapatite percubic centimeter was attached to the phantom forBMD calibration. Calibration was performed by mea-suring the Hounsfield unit values in the 3 compartmentsof the dental phantom and relating these values to deter-mine the BMD with a linear equation. The calibrationswere performed for each dog.

The scanned images were analyzed by V-implant(CyberMed, Seoul, Korea). The CT data taken before re-moving the screws were reconstructed to a transaxial

74 Cha et al American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

image at 1-mm intervals for each occlusal plane of themaxilla and mandible (Fig 4, A). The vertical locationand inclination of each screw was localized by using a re-constructed image at the coronal plane from the postop-erative scans. This information was transferred to thepreoperative images. From the preoperative scanningdata, the average BMD of a cylindrical area (diameter,2 mm; depth, 5 mm) for each screw site was measuredas the BMD total by Hounsfield unit calibration. TheCBT and average bone mineral density in the corticalarea (BMD cortical) were also calculated (Fig 4, B).


To determine the relationship between Hounsfieldunits and BMD values, a calibration regression equationwas derived individually for all experimental dogs and

was based on a linear relationship: BMD 5 a HU 1

b, where a and b are calibration coefficients.Pearson correlation analysis was used to determine

the association between placement torque, removal tor-que, mobility, BMD, and CBT. The categorical vari-ables (screw type and position) were statisticallyevaluated by using independent t tests to determineany differences in placement torque. The results suggestthat the variables were linearly correlated. Thus, a mul-tiple regression model was used with placement torqueas the dependent variable, and CBT, BMD, screw type,and screw position as the independent variables.

RESULTS

A regression equation was obtained by using the av-erage Hounsfield value (Table I, Fig 5) for the 3 parts inthe standard calibration of each beagle. The parametera (inclination) was calculated within the range of0.813 to 0.935, and the intercept b ranged from 3.80to 24.47. The correlation coefficient of all regressionequations was .0.98.

A correlation was observed between placement andremoval torque, mobility, screw type, and BMD corti-cal. The mobility showed a negative correlation withplacement torque (–0.577) and a positive correlationwith BMD cortical (0.575) (Table II).

The values of placement and removal torque, CBT,and BMD were significantly higher in the mandiblecompared with the maxilla (P \0.05). The taperedscrew (14.57 N per square centimeter) had higher place-ment torque than the cylindrical screw (9.69 N persquare centimeter); this was statistically significant inthe mandible (P \0.001) (Table II). There were no sig-nificant differences in the removal torque between thecylindrical and tapered screws. The values of CBT andBMD were similar in the tapered and cylindrical screwgroups (Table III).

Fig 1. Features and measurements of the miniscrews tested (mm).

Fig 2. Schematic diagram indicating the approximatelocations of screws. Circles indicate screw heads, andorthodontic forces were loaded reciprocally betweena cylindrical miniscrew (CS) and a tapered miniscrew(TS).

American Journal of Orthodontics and Dentofacial Orthopedics Cha et al 75Volume 137, Number 1

Multiple regression analysis showed that, among allthe variables (CBT, BMD cortical, BMD total, screwtype, screw position), screw position, screw type, and

BMD cortical had significant influences on placementtorque (R2 5 0.801) (Table IV).

DISCUSSION

Recently in orthodontics, the use of miniscrews hasbeen generalized for the preparation of orthodonticanchorage in clinical patients. However, studies on thestability of orthodontic miniscrews are still in their earlystages compared with studies of prosthodontic implants.Factors that can affect a miniscrew’s stability aredesign,20 placement procedures,21,22 and quantity andquality of the bone related to the host factor.5,23 Thisstudy focused on the assumptions that the quantity(thickness) and quality (density) of the bone have signif-icant effects on a miniscrew’s stability.

Information on bone density can be obtained fromthe measurement of the Hounsfield units, but there areissues regarding the reliability of many CT scanners,since CT values are affected by changes in the effectiveenergy of the x-ray source, the so-called beam-harden-ing effect.18 To calibrate Hounsfield unit values, refer-ence calibration standards for these CT values areneeded for each subject. Calibration coefficients hada range of 0.813 to 0.935 in this study. The same1000-HU value for 2 subjects can have a significantdiscrepancy of about 120 BMD units. Even though 2subjects might have the same Hounsfield unit value,they can have different BMD values, and calibrationcoefficients must be used to calculate the BMD values.We therefore used a hydroxyapatite phantom to provideaccurate calibration.

Regarding the relationship between BMD andplacement torque, there was a positive correlation inthe cortical part of the alveolar bone but no similarcorrelation in the total bone. Previous studies15,17

of prosthodontic implants reported high correlation

Fig 3. Experimental protocol used in this study.

Fig 4. Procedure of BMD measurement in the region ofinterest: A, position of calibration phantom and recon-structed image according to the occlusal plane; B, mea-surement of Hounsfield units for the region of interest(red box).


January 2010

between BMD and placement torque14 and significantcorrelation between BMD and pull-out strength.15,22,24

Seebeck et al25 reported that in an axial pullout strengthand cantilever bending test to the implanted screw, thescrew’s stability was mostly related to the CBT and can-cellous bone density. It is also known that the strengthand stiffness of trabecular bone are in linear proportionto bone density on the axial and bending loads.26 Thisdemonstrates that the density of trabecular bone andcortical bone makes a significant contribution to the sta-bility of an implant. In a prosthodontic implant, most ofthe implant, either in the maxilla or the mandible, is

placed near cancellous bone, and the quality of cancel-lous bone is important in the stability of the implant.27

However, our results indicated a poorer correlation be-tween total bone density and placement torque com-pared with cortical bone density.

It was observed that placement torque is related toscrew position. This is due to thicker cortical boneand better bone quality in the mandible comparedwith the maxilla.27 Despite the implant position, the sta-bility of a prosthodontic implant is more influenced byits length and diameter in alveolar bone, rather thanthe thickness of cortical bone.27 However, the stabilityof orthodontics miniscrews, with their relatively smallerdiameter and length, is related to CBT. This studyshowed a significant correlation between placementtorque and cortical bone density.

Placement and removal torque also showed a posi-tive correlation, which supports previously reported

Table I. Calibration of BMD by using Hounsfield units (HU) in the 3 compartments of the dental phantom

Dog

Calcium hydroxyapatite

0 mg/cm3 75 mg/cm3 150 mg/cm3

HU HU HU Calibration coefficients

Mean SD Mean SD Mean SD Parameter a Parameter b Correlation coefficient

1 �4.2 0.8 81.6 0.7 168.7 1.1 0.868 3.80 0.99

2 �6.4 6.2 79.4 2.6 176.9 6.3 0.813 7.24 0.99

3 �3.9 1.8 82.3 3.0 178.8 3.4 0.819 4.76 0.99

4 �10.8 5.3 74.1 2.1 169.8 4.0 0.827 10.71 0.99

5 �7.6 2.0 53.0 4.9 149.8 1.4 0.935 14.16 0.99

6 �21.0 5.7 47.8 2.3 160.4 1.4 0.810 24.47 0.98

Fig 5. Graph depicting calibration of BMD values fromHounsfield units for dog 1. The linear regression equa-tion is shown on the graph, and the values are listed inTable I. mg HA/cm3 equal milligrams (0, 75, 150) ofcalcium hydroxyapatite per cubic centimeter.

Table II. Pearson correlation test and independent t testfor placement torque

Variable Correlation coefficient P value

Removal torque .661 0.001

Mobility �.577 0.001

CBT .476 0.001

BMD cortical .575 0.001

BMD total .078 0.504

Mean 6 SD

Screw type

Cylindrical 9.69 6 5.32 0.001

Tapered 14.57 6 8.13

Screw

Maxilla 6.06 6 3.03 0.001

position

Mandible 17.84 6 5.03

Pearson correlation test was done between removal torque, mobility,

CBT, and BMD, and the independent t test was performed for screw

type and screw position.


studies, and they also had a correlation with the BMD ofcortical bone.28 However, the BMD of the total bone didnot have a major effect on placement and removal torque.On the other hand, mobility was associated with the BMDof the total bone and CBT. Therefore, this suggests thatBMD and CBT could affect a miniscrew’s stability.

Tapered miniscrews showed 65% higher placementtorque than the cylindrical type. This coincides with theresult from a previous study that used artificial bone.8

As the tapered screw was placed, the torque along withthe deformation of the surrounding bone increased withincreasing diameter. In this study, a similar bone densitywas calculated at the placement area of each screwtype. This meant that screw type affected the initialstability of the miniscrew because it was placed intosome areas with low bone density or thinner cortical bone.

A significant difference in placement torque wasfound according to the screw type, but no difference

was found in removal torque. This suggested that thestrain generated between the screw and the bone afterplacement decreased during healing of the surroundingbone. Further research involving histologic analysis isrequired to confirm this.

A regression equation between placement torque,BMD, and screw type was derived. Screw type andposition were major variables for the prediction ofplacement torque. This results from the difference inbone density and structure in the cortical bones of themaxilla and the mandible. The maxilla consists ofmore trabecular bone than the mandible, and its corticalbone is relatively thinner. The BMD cortical was alsochosen as a significant variable, but the impact of thevariables on the regression equation was smaller thanthose of screw type and position. These results suggestthat it might be difficult to predict the primary stabilityof a miniscrew solely with BMD values, and screwdesign could play a major role in increasing the place-ment torque despite the quantity of bone.

Because of the high radiation dose during CTscanning, there are limitations in using it for clinicalpurposes. However, dental CT such as cone-beam CTcan be a useful technique for a noninvasive assessmentof BMD in estimating the stability of a miniscrew at thepreoperative stage.

CONCLUSIONS

The BMD of the cortical bone, screw position, andscrew type have a compound influence on the primarystability of a miniscrew. Cone-beam CT might be a use-ful and noninvasive way to assess the BMD for estimat-ing the stability of a miniscrew at the preoperative stage.

Table III. Comparison of variables according to screw location and screw type

Screw type Cylindrical Tapered

P valueVariable Position Mean SD Mean SD

Removal torque (Ncm) Maxilla 2.29 1.24 1.83 0.81 0.480

Mandible 4.83 2.07 5.89 4.59 0.337

P value 0.001 0.001

Mobility (PTV) Maxilla 3.21 4.41 1.61 3.94 0.203

Mandible 0.12 2.77 �2.94 4.03 0.012

P value 0.073 0.014

CBT (mm) Maxilla 1.38 0.6 1.42 0.51 0.510

Mandible 1.98 0.48 2.02 0.33 0.403

P value 0.001 0.001

BMD cortical (mg

of hydroxyapatite/cm3)

Maxilla 867.46 242.99 866.46 196.2 0.988

Mandible 1030.17 167.81 1049.51 202.3 0.722

P value 0.01 0.004

BMD total (mg

of hydroxyapatite/cm3)

Maxilla 579.41 197.73 583.26 172.38 0.945

Mandible 697.32 149.34 787.91 191.24 0.076

P value 0.025 0.001

Table IV. Regression equation model to predict place-ment torque with CBT, BMD cortical, BMD total, screwtype, and screw position

Unstandardizedcoefficient Standardized

coefficientVariable Beta SE Beta t P value

CBT 0.053 0.879 0.004 0.060 0.952

BMD cortical 0.005 0.002 0.159 2.781 0.007

BMD total 0.003 0.002 0.068 1.092 0.278

Screw type 4.692 0.680 0.326 6.897 0.000

Screw position 10.428 0.922 0.725 11.310 0.000

Corrected R2 0.801

Dependent value was placement torque (Ncm).


January 2010

REFERENCES

1. Lee SY, Cha JY, Yoon TM, Park YC. The effect of loading time on

the stability of mini-implant. Korean J Orthod 2008;38:149-58.

2. Park YC, Chu JH, Choi YJ, Choi NC. Extraction space closure

with vacuum-formed splints and miniscrew anchorage. J Clin

Orthod 2005;39:76-9.

3. Im DH, Kim YS, Cho MA, Kim KS, Yang SE. Interdisciplinary

treatment of Class III malocclusion using mini-implant: problem-

oriented orthodontic treatment. Korean J Orthod 2007;37:305-14.

4. Motoyoshi M, Matsuoka M, Shimizu N. Application of orthodon-

tic mini-implants in adolescents. Int J Oral Maxillofac Surg 2007;

36:695-9.




6. Melsen B, Costa A. Immediate loading of implants used for ortho-

dontic anchorage. Clin Orthod Res 2000;3:23-8.

7. Ivanoff CJ, Sennerby L, Johansson C, Rangert B, Lekholm U.

Influence of implant diameters on the integration of screw

implants. An experimental study in rabbits. Int J Oral Maxillofac

Surg 1997;26:141-8.

8. Song Y, Cha J, Hwang C. Evaluation of insertion toruqe and pull-

out strength of mini-screws according to different thickness of

artificial bone. Korean J Orthod 2007;37:5-15.




10. Motoyoshi M, Yano S, Tsuruoka T, Shimizu N. Biomechanical

effect of abutment on stability of orthodontic mini-implant. A

finite element analysis. Clin Oral Implants Res 2005;16:480-5.

11. Cattaneo PM, Dalstra M, Melsen B. Analysis of stress and strain

around orthodontically loaded implants: an animal study. Int J

Oral Maxillofac Implants 2007;22:213-25.

12. Cattaneo PM, Dalstra M, Melsen B. The finite element method:

a tool to study orthodontic tooth movement. J Dent Res 2005;

84:428-33.

13. Heidemann W, Gerlach KL, Grobel KH, Kollner HG. Influence of

different pilot hole sizes on torque measurements and pullout

analysis of osteosynthesis screws. J Craniomaxillofac Surg

1998;26:50-5.

14. Brown GA, McCarthy T, Bourgeault CA, Callahan DJ. Mechani-

cal performance of standard and cannulated 4.0-mm cancellous

bone screws. J Orthop Res 2000;18:307-12.

15. Homolka P, Beer A, Birkfellner W, Nowotny R, Gahleitner A,

Tschabitscher M, et al. Bone mineral density measurement with

dental quantitative CT prior to dental implant placement in

cadaver mandibles: pilot study. Radiology 2002;224:247-52.

16. Carano A, Melsen B. Implants in orthodontics. Interview. Prog

Orthod 2005;6:62-9.

17. Beer A, Gahleitner A, Holm A, Tschabitscher M, Homolka P. Cor-

relation of insertion torques with bone mineral density from dental

quantitative CT in the mandible. Clin Oral Implants Res 2003;14:

616-20.

18. Cann CE. Quantitative CT for determination of bone mineral

density: a review. Radiology 1988;166:509-22.


risk factors associated with failure of mini-implants used for or-

thodontic anchorage. Int J Oral Maxillofac Implants 2004;19:

100-6.

20. Kido H, Schulz EE, Kumar A, Lozada J, Saha S. Implant diameter

and bone density: effect on initial stability and pull-out resistance.

J Oral Implantol 1997;23:163-9.

21. Sowden D, Schmitz JP. AO self-drilling and self-tapping screws in

rat calvarial bone: an ultrastructural study of the implant interface.


22. Hitchon PW, Brenton MD, Coppes JK, From AM, Torner JC.

Factors affecting the pullout strength of self-drilling and self-

tapping anterior cervical screws. Spine 2003;28:9-13.




24. Koistinen A, Santavirta SS, Kroger H, Lappalainen R. Effect

of bone mineral density and amorphous diamond coatings on

insertion torque of bone screws. Biomaterials 2005;26:

5687-94.

25. Seebeck J, Goldhahn J, Stadele H, Messmer P, Morlock MM,

Schneider E. Effect of cortical thickness and cancellous bone den-

sity on the holding strength of internal fixator screws. J Orthop

Res 2004;22:1237-42.

26. Ciarelli MJ, Goldstein SA, Kuhn JL, Cody DD, Brown MB. Eval-

uation of orthogonal mechanical properties and density of human

trabecular bone from the major metaphyseal regions with mate-

rials testing and computed tomography. J Orthop Res 1991;9:

674-82.

27. Norton MR, Gamble C. Bone classification: an objective scale of

bone density using the computerized tomography scan. Clin Oral

Implants Res 2001;12:79-84.

28. Ueda M, Matsuki M, Jacobsson M, Tjellstrom A. Relationship

between insertion torque and removal torque analyzed in fresh

temporal bone. Int J Oral Maxillofac Implants 1991;6:442-7.


ORIGINAL ARTICLE

Sequential bone healing of immediately loadedmini-implants: histomorphometric andfluorescence analysis

Glaucio Serra,a Liliane S. Morais,a Carlos Nelson Elias,b Marc A. Meyers,c Leonardo Andrade,d Carlos A. Muller,e

and Marcelo Mullere

Rio de Janeiro, Brazil, and San Diego, Calif

Introduction: Mini-implants are often immediately loaded for orthodontic treatment; however, changes in in-terfacial tissues caused by early loading and its effects might compromise the mini-implant’s function. Thepurpose of this study was to compare the healing of interfacial tissues 1, 4, and 12 weeks after the placementof titanium-alloy mini-implants in New Zealand rabbits; some of the implants were loaded immediately andothers were left unloaded. Methods: Eighteen animals were used in the experiment. Each received 4 titaniumgrade 5 mini-implants (2.0 3 6.0 mm), 2 of which were immediately loaded with 1 N of force. Tissue healing wasverified at 1, 4, and 12 weeks after placement. Four different fluorescent molecules were injected into the rab-bits to label calcium deposition. After the rabbits were killed, mineralized bone samples with the mini-implantswere removed, fixed, cut, stained, and observed with bright-field, polarized, and fluorescence microscopy.Results: After 12 weeks of healing, higher bone contact and bone area were observed than after 1 or 4weeks, regardless of loading. Differences between the loaded and unloaded groups were not observed(P \0.05) at 1 and 4 weeks. The bone deposition rate was higher in the loaded group. Conclusions: The1-N immediate force application did not compromise bone formation around mini-implants. (Am J OrthodDentofacial Orthop 2010;137:80-90)

Rigid bone anchorage has been developed sincesuccessful use of conventional dental implantsin orthodontics.1-5 Changes in the design were

proposed to overcome the disadvantages of conven-tional dental implants such as limitations of placementsites, invasive surgical procedures, postoperative pain,and difficulty of attaching elastics or springs.6-10

First, changes were made to the dimensions of theimplant. Length reduction was proposed, and then thesize and design were completely changed, creating

specific systems.7,11 Miniplates, mini-implants, and on-plants were the most used rigid orthodontic bone an-chorage systems.6,12,13 Mini-implants became widelyused because they met orthodontic needs: eg, widerange of placement sites, simplified surgical proceduresand lower costs.8,10 However, these design changes re-sulted in higher failure rates than conventional dentalimplants.14-17 In this context, some factors must be con-sidered that might jeopardize the success of the mini-implants. Screw diameter of 1 mm or less, inflammationof peri-implant tissues, thin cortical bone at the place-ment site, micromovement during the first healingphase, and the timing and modulus of loading are possi-ble causes for failure of mini-implants.14,16,18,19

Some researches have analyzed the effects of earlyloading in the healing process, and the protocol formini-implant loading has been discussed. Wehrbeinand Diedrich5 proposed a healing time of 25 weeks be-fore load application. After that healing period, the im-plants were successfully loaded during 26 weeks with2 N of force without problems. Some researchers postu-lated that, when the load is placed prematurely, there isno uniform intimate bone-implant contact because ofthe interplay of fibrous tissue.20,21 On the other hand,a review by Szmukler-Moncler et al22 summarizedthat micromovement is more harmful than the load dur-ing the early healing phase. Saito et al11 thought that

a PhD student, Department of Mechanical and Aerospace Engineering, Univer-

sity of California-San Diego; Phd student, Engineering Military Institute, Rio de

Janeiro, Brazil.b Professor, Military Engineering Institute, Department of Mechanical Engi-

neering and Materials Science, Rio de Janeiro, Brazil.c Professor, Department of Mechanical and Aerospace Engineering, University

of California-San Diego.d Professor, Biomedical Science Institute, University Federal of Rio de Janeiro,

Rio de Janeiro, Brazil.e Researcher, Oswaldo Cruz Institute, Department of Animals Experimentation,

Rio de Janeiro, Brazil.

Supported by CAPES and CNPq, Ministry of Education and Culture, Brazil

(Process numbers 472449/2004-4, 400603/2004-7, and 500126/2003-6).

The authors report no commercial, proprietary, or finacial interest in the prod-


Reprint requests to: Glaucio Serra, Av. Nossa Senhora de Copacabana 1355, Apt

805, Copacabana, Rio de Janeiro. RJ, Brazil; e-mail, [email protected].

Submitted, July 2007; revised and accepted, December 2007.

0889-5406/$36.00


doi:10.1016/j.ajodo.2007.12.035

80


18 weeks was long enough for biologic fixation capableof resisting up to a load of 2 N for 24, 28, or 32 weeks.Aldikacti et al23 applied a 2-N load after 6 weeks ofhealing. They used a long loading period of 52 weeks,and the mini-implants maintained stability throughoutthe study. In 2005, some authors proposed an early load-ing protocol for mini-implants after just 1 or 2 weeks ofhealing.10,24 Yao et al24 suggested that 2 weeks shouldbe enough for soft-tissue healing. Kim et al10 loadedthe mini-implants after 1 week and had a 9% failurerate after 10 weeks of continuous loading.

Immediate loading has also been described in clini-cal studies.17,25 Park et al25 applied 2 N of force for 36weeks immediately after placing mini-implants in pos-terior sites. They obtained a success rate of 90%. Mo-toyoshi et al,17 using an immediate load in the sameregion and with the same force, had a success rate of85.5%. The consensus in the early load protocol isthat even experienced clinicians can have failures andthat the removal of the mini-implant after the treatmentis uneventful.6,9,26-28 Limited information is availableabout the effects of immediate loading on the biologicsequences of healing during the early phases of tissueintegration, when the primary mechanical stability ofmini-implants must be substituted for stability obtainedthrough biologic means.

Our aim in this study was, therefore, to evaluate theinterface reactions at early and late stages of osseointe-gration around mini-implants immediately loaded with1 N. We used bright-field, polarized, and fluorescencemicroscopy for this evaluation.


The protocol for ths animal study was approved bythe standing ethics committee on animal research of Os-waldo Cruz Foundation (Rio de Janeiro, Brazil), and allprocedures were conducted in accordance withCanadian Council of Animal Care guidelines.

Eighteen 6-month-old male New Zealand white rab-bits, weighing 3.0 to 3.5 kg, were used. The surgicalprocedures were common to all animals and consistedof placement of 4 mini-implants in the left tibial meta-physes of each animal. All surgeries were performedunder sterile conditions in a veterinary operating room.

The titanium-aluminum-vanadium alloy mini-im-plants had a cylindrical screw design and a hexagonalhead (6.0 3 Ø 2.0 mm; Conexao Sistemas de Proteses,Sao Paulo, Brazil). They were machined by turning,cleaned, passivated with nitric acid, and sterilized by25 Gy of cobalt radiation (Fig 1).

The rabbits were acclimatized for a month in a vivar-ium before the surgical procedure. Immediately beforethe surgery, the animals were anesthetized with intra-muscular injection of tiletamine (5 mg/kg) and zolaze-pan (5 mg/kg) followed by continuous delivery of 2%halothane and isofluthane throughout the surgery. Thehair on the medial surface of the upper portion of theleft leg was clipped and the skin was cleansed with io-dinate surgical soap. An incision approximately50 mm in length was made parallel to longitudinalaxis of the tibia, in the medial aspect. The periosteumwas elevated by using sterile surgical techniques, andthe bone was denuded. Four implantation holes about5 mm apart were drilled with a 1.6-mm drill under pro-fuse sterile saline-solution irrigation and at low rotaryspeed. The mini-implants were threaded at the first cor-tex of the tibia with their longitudinal axes parallel toeach other and perpendicular to the external corticaltibia. After placement, the 2 central mini-implantswere immediately loaded with reciprocal forces. Anickel-titanium closed-coil spring was attached to themini-implant’s head, providing 1 N of force. Then theperiosteum was closed with resorbable sutures, andthe skin was sutured.

After the surgical procedures, each animal had 4mini-implants, 2 loaded and 2 unloaded, for a total of72 mini-implants. Thirty mini-implants were used inthis study, and the other 42 were analyzed by removal

Fig 1. Experimental implant: cylindrical screw of tita-nium alloy. a, hexagonal head with 3.4 mm height; b, ac-tive area, 6.0 mm length and 2.0 diameter, with 0.51 mmbetween the top of the pitches; c, machined threadedsurface.

American Journal of Orthodontics and Dentofacial Orthopedics Serra et al 81Volume 137, Number 1

torque test and scanning electron microscope analysis inthe first part of this study (Table I).

The experimental design was delineated to analyze3 periods of healing: 1, 4, and 12 weeks. In each assess-ment period, there was 1 group loaded and another un-loaded in a total of 6 groups (6 samples per group). Inthis manner, after each proposed time, 6 rabbits werekilled by exsanguination.

Within 30 minutes after the rabbit’s killing, theplacement surgical procedure described above was per-formed on each animal’s right leg; a new mini-implantper rabbit was placed for a total of 18 mini-implants.The maximum placement torque was measured witha manual torque meter (Table I). Thus, 48 mini-implantswere used, from which 30 had the sequential interfacereactions analyzed and 18 had the maximum torqueinsertion measured.

Quadruple polychromatic fluorescence labeling wasperformed after the second postoperative day (Fig 2).The sequential administration of fluorescent dyes variedbetween the groups and followed the topographic local-ization and the rate of new bone formation. The animalsreceived intramuscular injections of oxytetracycline(15 mg/kg of body weight), calcein blue (30 mg/kg ofbody weight), alizarin-complexone (30 mg/kg of bodyweight), and xylenol orange (90 mg/kg of body weight)diluted in 2% sodium carbonate solution.

After sample preparation, the slides were analyzedby using a fluorescence microscope (BX 51WI, Olym-pus, Tokyo, Japan) with filters with wavelengths of450 to 490 nm (blue filter) and 340 to 380 nm (violet fil-ter). The distance between the label marks was mea-sured, and the mineral appositional rate (MAR) wascalculated in micrometers per day.27 The MAR wasmeasured in both sides of the unloaded samples and inthe tension and compression sides of the loaded samples(2 measurements in each area).

After the healing period of 1, 4, or 12 weeks, the an-imals were killed, and the left leg of each rabbit was care-fully sectioned and removed. The closed-coil springs

were cut, and the tibia block specimens containing themini-implants and at least 2 mm of surrounding tissuewere dissected. The specimens were fixed (4% parafor-moldehyde solution in 0.1 mol/L sodium phosphatebuffer, pH 7.2), rinsed in the same buffer, and dehydratedin a graded ethanol series until absolute. Thereafter, theblocks were embedded in methylmethacrylate (Techno-vit 7100, Heraeus Kulzer, Dormagen, Germany) andsectioned in the longitudinal plane with a microtome(Exakt Medical Instruments, Oklahoma City, Okla).The thick slides were ground and polished to about 50mm for fluorescent microscopic examination. Subse-quently, the slides were stained with 2% toluidine bluefor the bright-field microscopic examination and the his-tomorphometric measurements. Each tibia block samplewas prepared to result in 1 central slide.

Histologic analysis was carried out by bright-fieldpolarized light transmission microscopy. The micropho-tographic images were acquired by using 10, 20, and 40objectives with a digital camera (EOS D-30, Canon, To-kyo, Japan) interfaced with a computerized system andthe microscope. Before sample observation, a standardslide consisting of a linear gridline in micrometers wasphotographed for scaling the images. After obtainingthe images, the measurements were made on each sideof the bone within 1 mm2 of the mini-implant surface,which corresponded to the bone interface at the first 2screw threads. The Image J Launcher software (Java ver-sion 1.1.4. for Windows, Microsoft, Redmond, Wash)was used to measure and convert pixels to micrometers.

Two static variables were measured. The fractionalbone-to-implant contact (% BIC) consisting of the lin-ear bone-to-implant contact (mm) of the total of themini-implant linear surface (mm) in the analyzed areaand the fractional bone area (% BA) consisting of thetotal bone area (mm2) of the total area analyzed (mm).


Statistical analyses for reporting means and stan-dard deviations of data from MAR, % BIC, and % BAwere performed for all groups. To quantify the signifi-cance of the differences, the data were evaluated by2-way analysis of variance (ANOVA) followed by thepost-hoc Tukey test and the t test (P \0.05).

RESULTS

The 48 mini-implants were placed without macro-deformation or fracture. The rabbits had no complica-tions such as infection or leg fracture during the healingprocess.

All mini-implants monocortically placed into therabbits’ tibia bones were clinically immobile after the

Table I. Experimental design

New Zealand

rabbits (n)

18

Mini-implants (n) 48 2.0 3 6.0 mm

Load Immediate 1 N

Time of healing 1, 4, and 12 weeks

Placement torque 18 samples

Histologic analysis 30 samples

Groups (n 5 5) 1wUn, 1wLo, 4wUn,

4wLo, 12wUn, 12wLo

1wUn, 1 week unloaded; 1wLo, 1 week loaded; 4wUn, 4 weeks un-

loaded; 4wLo, 4 weeks loaded; 12wUn,12 weeks unloaded;

12wLo,12 weeks loaded.

82 Serra et al American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

surgical procedure. At sample preparation, 3 mini-im-plants exhibited excessive clinical mobility and wereconsidered lost samples, yielding an overall successrate of 90%. These 3 failed mini-implants were in the1-week loaded, 4-week unloaded, and 12-weekunloaded groups.

The maximum placement torque was between108.02 and 84.40 N mm (mean, 98.33 N per millimeter;SD, 6 9.52 N mm).

By polarized light, no micrographic signs of chronicinflammation at the interface at any healing intervalwere observed. The analysis showed monocortical fixa-tion for all mini-implants. Bone deformations and mi-crofractures were observed primarily in the 1-weekgroups; even so, the native bone around the mini-im-plants was preserved throughout the experiment (Fig 3).

Corticalization was observed in some samples in the4-week loaded group. The increase of cortical thicknesswas found only in the endosteal bone region, indicatingprimary bone formation in this area. Corticalization wasalso observed in the 12-week groups with periosteal andendosteal bone formation (Fig 4).

After 1 week of healing, the mini-implants were em-bedded into the bone with the top and the peripheral por-tions of the pitches of the screw in close contact with thenative bone. The portions between the pitches werefilled by wound tissue, and no histologic differencewas found between the loaded and unloaded samples(Fig 5, A and B). Thus, the native lamellar bone was pre-dominant, and it was not jeopardized by the immediateloading or the placement technique in this initial healingphase, providing mechanical stability for the mini-im-plants during the first healing period. In the 4-weekgroups, the relationship between the native bone, the

mini-implant, and the interfacial tissue was maintained.The wound-tissue aspect was distinct between theloaded and the unloaded samples. The latter seemedto be less dense than the former, indicating a modifiedhealing process (Fig 5, C and D). After 12 weeks ofhealing, there was significant bone formation betweenthe threads of the mini-implant in both the loaded andunloaded samples. Then most of the region initiallyfilled by wound tissue was replaced by newly formedbone. There was no difference in the histologic findingsbetween the compression and the tension sides of theloaded mini-implants; however, the tendency of differ-ences between the loaded and unloaded samples wasmaintained. The loaded samples exhibited moreorganized tissues (Fig 5, E and F).

Fig 2. Polychromatic fluorescence labeling: O, oxytetracycline; C, calcein blue; X, xylenol orange; A,alizarin-complexone. †Day of killing.

Fig 3. Bone deformation (BD) and microfractures (MF)present mainly in 1-week groups. Original magnification,10 times.


The amount of osseointegration quantified by directbone-to-implant contact (% BIC) and the area of boneobserved between the threads of the screw (% BA) arelisted in Tables II and III. The % BIC results ranged

from 37.03% to 39.73% in the 1-week and 4-weekgroups, increasing to 66.01% and 70.96% after 12weeks of healing (Table II). Similarly, the % BA resultsindicated significant increases in the 12-week groups,

Fig 4. Sequential corticalization: A, absence of bone deposition in endosteal or periosteal sides at 1week; B, first sign in the endosteal region in the 4-week groups (white arrows); C, at 12 weeks, theincrease of bone thickness was in both the endosteal and periosteal sides.

Fig 5. Histologic results. A and B, Interfacial tissue (IT) and preserved native bone (NB) in close con-tact with the mini-implant (M-i) surface; original magnification, 20 times. C and D, Slight difference inthe interfacial tissue of the loaded and unloaded sample; original magnification, 20 times. E and F,Bone penetrating in the internal pitch area. Newly formed bone is more organized with lamellar aspectin the loaded group; original magnification, 40 times.


January 2010

ranging from 64.45% to 70.35% after 1 and 4 weeks ofhealing, reaching 86.13% and 87.14% in the 12-weekgroups (Table III). The statistical analysis confirmedthat, although there was no significant bone growth inthe 1-week and 4-week groups, significant formationof new bone was indicated by both % BIC and % BA af-ter 12 weeks, regardless of loading (Tables II and III).The compression and tension areas did not show signif-icant differences at any healing time (Table IV).

A striking result came from the 2-way ANOVAanalysis, which proved that the load did not enhanceor impair the healing process in terms of the histomor-phometric parameters (% BIC, P 5 0.495; % BA,P 5 0.128). Also, the healing time had a significant in-fluence, in which the 12-week groups showed a relevantincrease in bone around the mini-implants (% BIC,P 5 0.0001; % BA, P 5 0.0001) (Tables III and IV).

Fluorescence microscopy images showed a gradualincrease of deposition of the dyes. In the 1-week groups,no specific marks could be seen in the bone around themini-implants (Fig 6). Bone deformation and micro-fractures were clearly identified in the slides of this pe-riod. Aweakly stained bone was observed after 4 weeks;however, some loaded samples had an unspecific aliza-rin-complexone dye, which was the last fluorescence la-

bel administered. This label was restricted to theendosteal region, indicating the beginning of the min-eral deposition just in that region of the loaded sample.The 12-week groups showed well-defined labels in allperi–mini-implant bone, in both the endosteal and peri-osteal regions, allowing for rate measurements (Fig 6).

The regions labeled by fluorochromes were easilyidentified in all 12-week specimens (Fig 7); however,the calcein blue dye was not visible when the blue filterwas used. It was visible with the violet filter. Neverthe-less, because of the need to have all labels in the sameview to proceed with the MAR calculations, thesedata were discarded. The statistical analyses showedsignificant differences between the unloaded and loadedgroups (P 5 0.024 and P 5 0.008, respectively),whereas the tension and compression sides of the loadedgroup had no significant difference (P 5 0.934)(Table V). This indicates accelerated healing in theloaded sample without a difference between the areasof tension and compression.

DISCUSSION

These findings indicate that immediate loading didnot compromise the healing process in the tissues

Table II. Histomorphometric values: % BIC rate

Histomorphometric values:% BIC rate (Mean 6 SD)

Healing time Unloaded Loaded

1 week 39.73 6 5.7a 37.68 6 6.1a

4 weeks 37.03 6 5.0a 38.76 6 3.4a

12 weeks 66.01 6 10.9b 70.96 6 7.1b

Term Two-way ANOVA interaction: 1 3 2

Load Loaded 3 unloaded / P 5 0.495

Time 1w 3 4w /P 5 0.927

1w 3 12w /P 5 0 .0001

4w 3 12w /P 5 0 .0001

Mean values with the same superscript letters are not significantly dif-

ferent (P .0.05). Mean values with different superscript letters indi-

cate statistically significant differences (P \0.05).

1w, 1-week groups; 4w, 4-week groups; 12w,12-week groups.

Table III. Histomorphometric values: % BA rate

Histomorphometric values: % BA rate (Mean 6 SD)

Healing time Unloaded Loaded

1 week 66.40 6 3.3a 69.57 6 2.6a

4 weeks 64.45 6 7.3a 70.35 6 3.7a

12 weeks 86.13 6 5.5b 87.14 6 4.2b

Term Two-way ANOVA interaction: 1 3 2

Load Loaded 3 unloaded / P 5 0.128

Time 1w 3 4w /P 5 0.998

1w 3 12w /P 5 0 .0001

4w 3 12w /P 5 0 .0001

Mean values with the same superscript letters are not significantly dif-

ferent (P .0.05). Mean values with different superscript letters indi-

cate statistically significant differences (P \0.05).

1w,1-week groups; 4w, 4-week groups; 12w,12-week groups.

Table IV. Comparisons of means of the tension and compression sides

t test for independent samples

Time Area % BIC mean 6 SD % BA mean 6 SD % BIC % BA

1 week Compression 37.10 6 6.78 70.32 6 3.25 P 5 0.8105 P 5 0.4576

Tension 38.27 6 5.11 68.89 6 2.29

4 weeks Compression 39.16 6 4.85 72.56 6 3.21 P 5 0.7330 P 5 0.0577

Tension 38.36 6 1.43 68.14 6 3.10

12 weeks Compression 70.52 6 6.78 87.18 6 2.23 P 5 0.8578 P 5 0.9780

Tension 71.40 6 8.18 87.10 6 5.98


around the mini-implants, although the quality and therate of bone formation had been altered. The analysisof the undecalcified sections provided an overview ofthe sequential healing evolution, since the primary me-chanical stability of the mini-implants obtained by thefriction and close contact with native bone until thelate biological attachment provided by the new boneformation at the bone–mini-implant interface. The fluo-rescence analysis allowed for comparison of the effectsof the immediate load application to the bone formationrate as well as to the topographic localization of theearly healing events.

The first prerequisite for the success of implants isminimal damage to the host tissues during the surgicalprocedure.1,2,5 Copious irrigation to prevent superheat-ing during drilling, preservation of the periosteal tissue,and maintenance of cleanliness are well-described pre-cautions. Furthermore, Motoyoshi et al17 described therecommended implant placement torque in the range

of 50 to 100 N mm. They concluded that low valuesare insufficient for establishing mechanical stability,and high values could generate excessively high stresslevels, resulting in degeneration of the interfacialbone. In our study, the mean placement torque was98.3 N mm, indicating a good relationship between sta-bility and interfacial stress. Additionally, all sampleswere clinically stable after placement, and the healingprocess resulted in osseointegration of most samples af-ter the maximum experimental interval. Thus, we as-sumed that the surgical technique and the postsurgicalconditions were suitable for the immediate loading se-quential analysis.

Within the framework of the time of loading, theproposed healing interval before orthodontic force ap-plication has been gradually decreased since the firstprotocol described.10,11,23,24 Now some experienced re-search groups have described a success rate of about90% with an immediate orthodontic loading

Fig 6. Fluorescence microscopy. A and B, Longitudinal section of mini-implant and surroundingbone after 1 week of healing without dye deposits. Bone deformation (BD) and microfractures (MF)are shown in both the load and unloaded groups. C and D, the 4-week groups. Unspecific aliza-rin-complexone dye marked in the endosteal region of the loaded sample (A). E and F, the12-week groups. Great level of fluorochrome labeling shown in all surrounding bone, regardlessthe loading. CCS, Closed-coil spring; M-i, mini-implant. Original magnification, 4 times.


January 2010

protocol.10,16,17,25 Even so, they suggested that immedi-ate loading could be used with no compromise in thestability of the mini-implants. In agreement, we founda 10% fail rate, and the lost samples were not relatedto immediate loading. On the contrary, Costa et al20

and Melsen and Costa21 postulated that prematureloading resulted in fibrous tissues in the interface.Szmukler-Moncler et al22 and Cope29 addressed thesecontroversial conclusions by suggesting that micromo-tion is more detrimental than premature loading, and,when these forces are acting simultaneously, the resultscould be misinterpretated. Orthodontic loading could beapplied up to threshold, above of which it would dam-age to the osseointegration process.16,22,29,30 The over-loading limit depends on the quality and quantity ofnative and newly formed bone and on the implant de-sign. Isidor,30 using finite-element analysis, concludedthat loads resulting in interfacial strains higher than6700 m are harmful to the healing process. The transferof stress at the bone–mini-implant interface after loadapplication is influenced by implant design and surfacegeometry.31 Load concentration tends to occur at thefirst threads of the screw-design implants after lateralforce application, possibly resulting in bone resorptionaround these areas.19,32 Buchter et al16 demonstratedthat immediate tip forces higher than 9 N resulted in cer-vical resorption around the mini-implant or failure ofthe mini-implant. Oyonarte et al31,33 compared thebone response in the proximity of machined-threadedimplants and porous-surface implants after orthodonticloading. They found significant cervical bone resorptionaround the machined-threaded implants and concludedthat implant geometry resulted in high stress concentra-tion and, consequently, bone loss. In our study, bone re-sorption around the first threads of the mini-implant

after 1, 4, or 12 weeks of loading was not found. Be-cause we did not apply a finite-element model to com-plement our in-vivo study, we could not correlate theloading areas with the histologic response, but it couldbe suggested that the 1 N force associated with themini-implant design used in this study did not reachthe overloading limit.

Hoshaw et al34 stated that the placement process ofscrew-shaped implants causes microdamage in the adja-cent bone, and the damage concentration decreases after4 weeks of healing. Despite the different dimensionsand loading protocol, in our study, many microfracturesand bone deformation in the 1-week groups were alsofound. After 4 weeks of healing, some were still present,but they were not found in the 12-week groups. Thebone modeling and remodeling processes not only filledthe interface with newly formed bone but also regener-ated bone in the regions of damage.

Regarding the histomorphometric findings, no sig-nificant statistical difference was found between the1-week and 4-week groups in terms of % BIC and %BA values, regardless of the immediate loading. Similarto our previous study, in which the removal torquevalues increased in the 12-week groups, the % BICand % BA means increased significantly after 12 weeksof healing; however, no statistical difference was foundbetween the loaded and unloaded groups.35 Moreover,the areas under tension and compression also had no sta-tistical difference, signifying that the immediate loaddid not influence the amount of osseointegration.

The increase of the static histomorphometric values(% BIC and % BA) after several weeks of healing is welldescribed, not only for the delayed loading protocol butalso for the immediate loading protocol.3,33,36,37 Never-theless, bone formation at the areas of tension and com-pression remains controversial. Consistent with ourfindings, Wehrbein and Diedrich5 observed no micro-morphologic differences between the compression and

Fig 7. Well-defined fluorochrome labels used in theMAR calculations: A, alizarin-complexone; X, xylenol or-ange; O, oxytetracycline. Original magnification, 40times.

Table V. MAR rate values compared with 1-wayANOVA and post-hoc Tukey test

GroupMineral appositional

rate (mean 1 SD)

12wUn (n 5 16) 2.96 1 0.8 mm/day

12wLo compression

(n 5 10)

3.68 1 0.5 mm/day

12wLo tension (n 5 10) 3.78 1 0.5 mm/day

ANOVA P 5 0.00035

Post-hoc Tukey test 12wUn vs 12wLo compression: P 5 0.024

12wUn vs 12wLo tension: P 5 0.008

12wLo compression vs 12wLo tension:

P 5 0.934

12wUn,12 weeks unloaded; 12wLo,12 weeks loaded.


tension side of the implants. The implants were loadedafter a healing period of 25 weeks. In addition, Melsenand Lang38 stated that the quantity of osseointegrationwas not influenced by the applied load (1-3 N) after12 weeks of healing. On the contrary, Buchter et al,37

in an immediate loading study, analyzed bone responsearound 2 geometrically different mini-implants loadedwith tip forces varying between 1 and 9 N. They con-cluded that forces of 5 or 6 N resulted in increasedbone–mini-implant contact. The force variation de-pended on the geometry of the mini-implant. Wehrbeinet al7 suggested that bone generation activity close toimplants could be influenced by the magnitude of forceand the quality of the bone. They concluded that the 1-Nforce did not cause an increase in bone deposition in al-veolar bone, but 2-N forces resulted in moderate boneapposition in the midsagittal palatal bone, especiallyin the areas under compression. Such bone formationdepends on several clinical factors including the magni-tude of the force, the quality of the supported bone, andthe geometry of the implant. The different experimentalprotocols used in these studies could explain the variedfindings.

The most interesting finding of our study was relatedto the bone deposition rate. The rate was significantlyhigher in the loaded implants than in the unloaded im-plants after 12 weeks (P 5 0.024 and P 5 0.008, respec-tively). The areas under compressive and tensile stresseswere not statistically different (P 5 0.934). The bone re-generation process could be accelerated by early biome-chanical stimulation. This was suggested by Sarmientoet al,39 who provided experimental radiographic, histo-logic, and mechanical evidence during the fracture heal-ing process. In the implant field, Rubin et al40 tested theearly biomechanical stimulation in porous titanium-al-loy implants and found that exposure to low-amplitudemechanical strain during the healing period enhancedprecocious biologic fixation. In addition, increasedbone deposition has been found in the bone aroundthe loaded implants when compared with bone sur-rounding unloaded implants.5,7,8,27,33 In this study, thebone deposition rate was higher in the loaded groups.Hence, it can be suggested that healing was acceleratedby early loading. Supporting this suggestion, the wovenbone produced along the endosteal surface is the pri-mary response to the cortical healing defects, and, inour study, some endosteal corticalization was seen inthe loaded samples after 4 weeks.2,41,42 Moreover, fluo-rescence alizarin-complexone dye was observed in theendosteal areas of these samples.

However, the accelerated healing process did not re-sult in greater osseointegration (% BIC or % BA) in theloaded group after 12 weeks. Huja and Roberts27 de-

scribed the continuous and accelerated remodeling pro-cess within 1 mm of the loaded implant surface asa possible mechanism whereby the loaded implant main-tains the integration with less mineralized bone. A highrate of bone remodeling produces incompletely mineral-ized lamellar tissue in contact with the implant surface.This mechanism has been suggested to be important toprevent microdamage and crack accumulation at the in-terfacial bone.22,27,33,43,44 Thus, it could be suggestedthat immediate loading influenced the quality of thenewly formed bone, quickly producing less mineralizedbone. This could be correlated with the findings of thefirst part of this study, in which a lower removal torquevalue was found in the 12-week loaded group comparedwith the unloaded 12-week group.35 The well-describedeasy removal of mini-implants in clinical studies couldbe also related to this suggestion.9,25,45 Quantificationof the bone mineral content in bone-tissue growth underimmediate loading conditions is necessary to confirmthis hypothesis.

Self-drilling mini-implants have recently been de-scribed.10 Self-tapping mini-implants were used in thisstudy. The choice between these 2 mini-implant typesshould be made carefully. The main differences betweenthem are the immediate postsurgical conditions: the self-drilling mini-implant has more bone–mini-implant con-tact and better primary stability. On the other hand, theimplant placement torque could be higher dependingon the characteristics of the host bone. It could generatea high stress level, resulting in local ischemia and necro-sis of the bone at the interface.17 Thus, the results of thisimmediate-loading study should not be extrapolated toself-drilling mini-implants.

Some important features—eg, soft-tissue inflamma-tion around the mini-implants, exposure to the oral en-vironment, and variations in cortical thickness—caninfluence the healing process but were not consideredin this animal study. So, the extrapolation of these re-sults to clinical applications should be done with cau-tion. Our data showed that osseointegration wasreached regardless of immediate loading. It can be sug-gested that the immediate loading protocol resulted inaccelerated healing and modified new bone formationwithout jeopardizing the stability of the mini-implantsduring the experimental period.

CONCLUSIONS

1. The titanium-aluminum-vanadium mini-implantswere appropriate as anchorage for 1 N of immediateloading.

2. The immediate 1-N load caused no significantchanges in the amount of osseointegration


January 2010

(histomorphometric parameters) after 1, 4, or 12weeks of healing (P \0.05).

3. The bone deposition rate was higher in the loadedgroups than in the unloaded groups, indicating ac-celerated healing.

We thank Lars Bjursten for helping with the manu-script and Jared Braden Goor for helping with theoptical microscope analysis (Department of Bioengi-neering, University of California-San Diego); theNational Science Foundation Ceramics Program(Linnette Madsen, Director) for travel and laboratorysupplies; and Conexao Sistemas de Proteses, Sao Paulo,Brazil, for supplying the mini-implants used in thisstudy.

REFERENCES

1. Turley PK, Kean C, Schur J, Stefanac J, Gray J, Hennes J, et al.

Orthodontic force application to titanium endosseous implants.


2. Roberts EW, Smith RK, Zilberman Y, Mozsary PG, Smith RS.

Osseous adaptation to continuous loading of rigid endosseous

implants. Am J Orthod 1984;86:95-111.

3. Roberts EW, Helm RF, Marshall JK, Gongloff RK. Rigid endo-

sseous implants for orthodontic and orthopedic anchorage. Angle

Orthod 1989;59:247-56.

4. Roberts EW, Marshall KJ, Mozsary PG. Rigid endosseous implant

utilized as anchorage to protract molars and close an atrophic

extraction site. Angle Orthod 1990;60:135-52.

5. Wehrbein H, Diedrich P. Endosseous titanium implants during and

after orthodontic load—an experimental study in dog. Clin Oral



1997;36:763-7.

7. Wehrbein H, Glatzmaier J, Yildirim M. Orthodontic anchorage

capacity of short titanium screw implants in the maxilla. An

experimental study in dog. Clin Oral Implants Res 1997;8:

131-41.

8. Ohmae M, Saito S, Morohashi T, Seki K, Qu H, Kanomi R, et al. A

clinical and histological evaluation of titanium mini-implants as

anchors for orthodontic intrusion in the beagle dog. Am J Orthod


9. Paik CH, Woo YJ, Kim J, Park JU. Use of miniscrews for inter-

maxillary fixation of lingual-orthodontic surgical patients. J

Clin Orthod 2002;36:132-6.

10. Kim JW, Ahn SJ, Chang YI. Histomorphometric and mechanical

analyses of the drill-free screw as orthodontic anchorage. Am J


11. Saito S, Sugimoto N, Morohashi T, Ozeki M, Kurabayashi H,

Shimizo H, et al. Endosseous titanium implants as anchors for me-

siodistal tooth movement in the beagle dog. Am J Orthod Dento-


12. Chung KR, Kim YS, Linton JL, Lee YJ. The miniplate with tube

for skeletal anchorage. J Clin Orthod 2002;36:407-12.

13. Janssens F, Swennen G, Dujardin T, Glineur R, Malevez C. Use of

an onplant as orthodontic anchorage. Am J Orthod Dentofacial

Orthop 2002;122:566-70.

14. Miyawaki S, Koyama I, Inoue M, Mashima K, Sugahara T, Ta-

kano-Yamamoto T. Factors associated with the stability of tita-

nium screws placed in the posterior region for orthodontic

anchorage. Am J Orthod Dentofacial Orthop 2003;124:373-8.

15. Choi BH, Zhu SJ, Kim YH. A clinical evaluation of titanium mini-

plates as anchors for orthodontic treatment. Am J Orthod Dento-


16. Buchter A, Wiechmann D, Koerdt S, Wiesmann HP, Piffko J,

Meyer U. Load-related implant reaction of mini-implants used

for orthodontic anchorage. Clin Oral Implants Res 2005;16:

473-9.

17. Motoyoshi M, Hirabayashi M, Shimizu N. Recommended place-

ment torque when tightening an orthodontic mini-implant. Clin

Oral Implants Res 2006;17:119-4.

18. Favero L, Brollo P, Bressan E. Orthodontic anchorage with spe-

cific fixtures: related study analysis. Am J Orthod Dentofacial

Orthop 2002;122:84-94.

19. Chen F, Terada K, Hanada K, Saito I. Anchorage effects of palatal

osseointegrated implant with different fixation: a finite element

study. Angle Orthod 2005;75:593-601.

20. Costa A, Raffaini M, Melsen B. Miniscrews as orthodontic an-

chorage: a preliminary report. Int J Adult Orthod Orthognath

Surg 1998;13:201-9.



22. Szmukler-Moncler S, Salama H, Reingewirtz Y, Dubruille JH.

Timing of loading and effect of micromotion on bone-implant in-

terface: review of experimental literature. J Biomed Mater Res

1998;43:193-203.

23. Aldikacti M, Acikgoz G, Turk T, Trisi P. Long-term evaluation of

sandblasted and acid-etched implants used as orthodontic anchors

in dogs. Am J Orthod Dentofacial Orthop 2004;125:139-47.

24. Yao CC, Lee JJ, Chen HY, Chang ZC, Chang HF, Chen IJ. Max-

illary molar intrusion with fixed appliances and mini-implant an-

chorage in three dimensions. Angle Orthod 2005;75:754-60.

25. Park HS, Lee SK, Knon OH. Group distal movement using im-

plant anchorage. Angle Orthod 2005;75:602-9.

26. Park YC, Lee SY, Kim DH, Jee SH. Intrusion of posterior teeth

using mini-screw implants. Am J Orthod Dentofacial Orthop

2003;123:690-4.

27. Huja SS, Roberts E. Mechanism of osseointegration: characteriza-

tion of supporting bone with indentation testing and backscattered

imaging. Semin Orthod 2004;10:162-73.

28. Eriksen FE, Axelrold WD, Melsen F. Bone histomorphometry. 1st

ed. New York: Raven Press; 1994.

29. Cope JB. Temporary anchorage devices in orthodontics: a para-

digm shift. Semin Orthod 2005;11:3-9.

30. Isidor F. Histological evaluation of peri-implant bone at implants

subjected to occlusal overload or plaque accumulation. Clin Oral


31. Oyonarte R, Pilliar R, Deporter D, Woodside DG. Peri-implant

bone response to orthodontic loading: part 1. A histomorphomet-

ric study of the effects of implant surface design. Am J Orthod


32. Vasquez M, Calao E, Becerra F, Ossa J, Enr�ıquez C, Fresneda E.

Initial stress differences between sliding and sectional mechanics

with an endosseous implant as anchorage: a 3-dimensional finite

element analysis. Angle Orthod 2001;71:247-56.

33. Oyonarte R, Pilliar R, Deporter D, Woodside DG. Peri-implant

bone response to orthodontic loading: part 2. Implant surface ge-

ometry and its effect on regional bone remodeling. Am J Orthod


34. Hoshaw SJ, Brunski JB, Cochran GVB. Mechanical loading of

Branemark implants affect interfacial bone modeling and remod-

eling. Int J Oral Maxillofac Implants 1994;9:345-60.


35. Serra GG, Morais LS, Elias CN, Meyers MA, Andrade L,

Muller C, et al. Sequential bone healing of immediate loaded

mini-implants. Am J Orthod Dentofacial Orthop 2009 (in press).

36. Deguchi T, Takano-Yamamoto T, Kanomi R, Hartsfield JK,

Roberts WE, Garetto LP. The use of small titanium screws for or-

thodontic anchorage. J Dent Res 2003;82:377-81.

37. Buchter A, Wiechmann D, Gaertner C, Hendrik M, Vogeler M,

Wiesmann HP, et al. Load-related bone modeling at the interface

of orthodontic micro-implants. Clin Oral Implants Res 2006;17:

714-22.

38. Melsen B, Lang NP. Biological reactions of alveolar bone to or-

thodontic loading of oral implants. Clin Oral Implants Res

2001;12:144-52.

39. Sarmiento A, Schaeffer JK, Beckerman L, Latta L, Enis JE. Frac-

ture healing in rat femora as affected by functional weight-bear-

ing. J Bone Joint Surg Am 1977;59:269-75.

40. Rubin CT, McLeod KJ. Promotion of bone ingrowth by fre-

quency-specific, low-amplitude mechanical strain. Clin Orthop

Relat Res 1994;298:165-74.

41. Lopes CC, Junior BK. Histological findings of bone remodeling

around smooth dental titanium implants inserted in rabbit’s tibia.

Ann Anat 2002;184:359-62.

42. Lange GD, Putter CD. Structure of the bone interface to dental im-

plants in vivo. J Oral Implantology 1993;19:123-35.

43. Huja SS, Katona TR, Burr DB, Garetto LP, Roberts WE. Micro-

damage adjacent to endosseous implants. Bone 1999;25:217-22.

44. Klokkevold PR, Johnson P, Dadgostari S, Caputo A, Davies JE,

Nishimura RD. Early endosseous integration enhanced by dual

acid etching of titanium: a torque removal study in the rabbit.


45. Melsen B, Petersen JK, Costa A. Zigoma ligatures: an alternative

form of maxillary anchorage. J Clin Orthod 1998;32:154-8.


January 2010

ORIGINAL ARTICLE

Effects of miniscrew orientation on implantstability and resistance to failure

Michael B. Pickard,a Paul Dechow,b P. Emile Rossouw,c and Peter H. Buschangd

Moscow, Idaho, and Dallas, Tex

Introduction: The purpose of this study was to determine the effect of miniscrew implant orientation on theresistance to failure at the implant-bone interface. Methods: Miniscrew implants (IMTEC, Ardmore, Okla)were placed in 9 human cadaver mandibles, oriented at either 90� or 45� to the bone surface, and tested tofailure in pull-out (tensile) and shear tests. The line of applied force and the orientation of the implants alignedat 45� were either parallel or perpendicular to the maximum axis of bone stiffness. In the shear tests, the im-plants aligned at 45� were angled toward and opposing the axis of shear force. Results: The implants alignedat 90� had the highest force at failure of all the groups (342 6 80.9 N; P\0.001). In the shear tests, the implantsthat were angled in the same direction as the line of force were the most stable and had the highest force atfailure (253 6 74.05 N; P\0.001). The implants angled away from the direction of force were the least stableand had the lowest force (87 6 27.2 N) at failure. Conclusions: The more closely the long axis of the implantapproximates the line of applied force, the greater the stability of the implant and the greater its resistance tofailure. (Am J Orthod Dentofacial Orthop 2010;137:91-9)

Despite the serendipitous discovery of the os-seointegrative properties of titanium and thesubsequent development of titanium dental im-

plants in the 1960s,1-3 they were not used in orthodon-tics until the 1980s.4-9 However, dental implants havelimited anatomic placement options and require a pre-cise 2-stage protocol and a 3 to 6 month healing pe-riod.3,4,6,10,11 Recently introduced miniscrew implants(MSIs) can be easily placed in almost any intraoral re-gion, have lower costs, are removed easily, and, conse-quently, have greater applications for orthodonticanchorage.12,13

The success of any implant in providing definitiveanchorage depends on its stability. Most clinical reportssuggest that MSIs are stable with applied forces rangingfrom 50 g (0.5 N)14 to 450 g (4.5 N).15 However, mini-screws should not be expected to remain absolutely sta-tionary during orthodontic loading;16 MSIs remain

stable as bone remodeling takes place in response to me-chanical stress.17-19 This is distinctly different frommovements associated with pathology; these result inloosened MSIs. Although it seems well establishedthat MSIs placed with appropriate surgical techniquescan withstand forces in the orthodontic range (1-3 N),there is only limited information available concerningthe maximum forces that can be applied to them.

Pull-out (tensile) tests are commonly used to evalu-ate the maximum forces that bone screws can withstand,and are considered an accurate method of comparing therelative strength or ‘‘holding power’’ of surgically placedbone screws.20-24 Tests have been conducted with vari-ous animal bones, including bovine femurs,20,24,25 por-cine ribs,26,27 dog femurs,20,21 and sheep parietalbones.28 There have been only limited pull-out tests onhuman mandibles.23 Importantly, pull-out tests aloneare not adequate for measuring the fixation potential ofbone screws, because they do not address shearingforces.21,29 Even though pull-out and shear tests produceforces that substantially exceed those typically used byorthodontists, these tests provide valuable informationpertaining to primary stability and material characteris-tics of MSIs.

There are presently no published data on the maxi-mum pull-out and shear forces that MSIs can withstand,and there has been only limited pull-out testing of bonescrews in actual anatomic bone sites. Furthermore, thereare no published data on the effect of MSI placementorientation relative to the bone surface and the axes ofmaximum and minimum bone stiffness, despite recentevidence demonstrating consistent patterns of material

a Private practice, Moscow, Idaho.b Professor and program director, Department of Biomedical Sciences, Baylor

College of Dentistry, Texas A&M University Health Science Center, Dallas.c Professor and chairman, Department of Orthodontics, Baylor College of Den-

tistry, Texas A&M University Health Science Center, Dallas.d Professor and director of orthodontic research, Department of Orthodontics,

Baylor College of Dentistry, Texas A&M University Health Science Center,

Dallas.



Reprint requests to: Peter H. Buschang, Department of Orthodontics, Baylor

College of Dentistry, 3302 Gaston Ave, Dallas, TX 75246; e-mail,

[email protected].

Submitted, September 2007; revised and accepted, December 2007.

0889-5406/$36.00


doi:10.1016/j.ajodo.2007.12.034

91


anisotropy in the cortical bone of various regions of thehuman craniofacial skeleton.30

The purpose of this study was to evaluate the effectsof orthodontic MSI orientation on stability and resis-tance to failure at the bone-implant interface. The studywas designed to answer the following questions. What isthe maximum amount of force that can be applied toMSIs in the human mandible? Does loading orientationaffect maximum force? Does the orientation of the longaxis of the miniscrew relative to the surface of the boneand the direction of the applied force affect theimplant’s stability and its resistance to failure?


Nine fresh-frozen, unembalmed, dentate, human ca-daver mandibles donated for anatomic research pur-poses were selected for implant placement and testing.The mandibles came from 3 female and 6 male white do-nors between 48 and 81 years of age. No donors wereknown to have suffered from primary bone diseases.

For mounting purposes, impressions were taken ofthe test mandibles, and models were poured. Up to 3custom acrylic (methylmethacrylate) bases were fabri-cated for each test mandible.28 To compensate for lin-gual surface variations, the acrylic base was adaptedto the cortical surface of the entire lingual corpus and in-ferior ramus. The opposite side of the acrylic base wasground to a level plane. The acrylic base allowed rigidfixation of the mandible to the test equipment whilemaintaining the test site surface orientation perpendicu-lar to the line of force in the pull-out tests and parallel tothe line of force in the shear tests. Then the custom fit toeach lingual surface allowed for uniform distribution ofreaction forces during the application of test loads. Itprevented the development of internal stress/strain in-duced by flexure secondary to fixation of the mandibleto the test apparatus. An anterior hole near the mentalforamen and a posterior hole in the inferior ramuswere drilled through the bone sample and the customacrylic support base to allow rigid fixation of themandible to the test equipment.

The tensile and shear testing was completed witha universal testing machine (model DDL 200RT,TestResources, Shakopee, Minn), outfitted with a 112-pound calibrated load cell used in tensile mode. Adjust-able x-axis and y-axis sliding tables (Sherline Products,Vista, Calif) were mounted to the base of the testing ma-chine. A 360� rotational table (Sherline Products) wasmounted on top of the sliding tables horizontally forpull-out tests and vertically for shear tests. A mountingplate was fabricated from a 0.5 3 3 3 9-in aluminumbar and attached to the 360� rotational table. Bolts

were used to secure the test mandible and its acrylicbase to the mounting plate.

A custom implant holder was attached to the load cellof the testing machine. It was designed to allow for rota-tional and x-y freedom during attachment of the test im-plants to the tensile machine. The bottom part of theimplant holder was fabricated from a 6.4-mm diameteraluminum cylinder with a 2-mm key way cut in itsbase to a depth of 4 mm. This space gave the head ofthe implant 0.4 mm of horizontal and 0.5 mm of verticalclearance. A 1.0-mm hole was drilled perpendicular tothe key way at a point 2.0 mm from the base. A 0.028-in piece of stainless steel orthodontic wire was runthrough the implant holder and the hole in the head ofthe implant to secure the implant to the implant holder.

MSIs (IMTEC, Ardmore, Okla), 6 mm long and 1.8mm in diameter, were placed in the buccal cortex of themandible, at angles of either 90� or 45� to the buccal sur-face. They were tested to failure by using pull-out andshear tests directed along the axes of maximum and min-imum bone stiffness. The design test matrix consisted of9 subgroups of 10 implants each (Fig 1). Three sub-groups were pull-out tests. The remaining 6 were sheartests. Three shear tests were sheared parallel to the max-imum axis of bone stiffness, and 3 were sheared parallelto the minimum axis of bone stiffness, which is perpen-dicular to the axis of maximum bone stiffness.

The pull-out tests consisted of MSIs in 1 of 3 orien-tations: placed orthogonally to the buccal surface, an-gled at 45� along the maximum axis of bone stiffness,and angled at 45� along the minimum axis of bone stiff-ness. All 3 implant orientations were tested to failure inpull-out tests in which the line of force was at 90� to thebone surface.

In both sets of the shear tests, along the axis of eithermaximum or minimum bone stiffness, the MSIs were in1 of 3 positions: orthogonal to the buccal surface, angledat 45� toward the same direction as the line of shearforce, and angled at 45� opposite the line of shear force.All MSI orientations were tested to failure in a shear testin which the line of shear force was along the bone sur-face and in a direction parallel to the maximum or min-imum axis of bone stiffness. The maximum force atfailure in the pull-out and shear tests and the corticalthickness were recorded for each sample.

Bone-implant test sites were selected on the buccalcortex of the corpus of the mandibles so that no proxi-mate defects could adversely affect the test results. A1.1-mm pilot drill was used to prepare the MSI site;the pilot holes were drilled only through the cortex. Aguide was fabricated and used with the pilot drill to en-sure the proper orientation for the MSIs oriented at 45�

to the bone surface. After the pilot holes were drilled,

92 Pickard et al American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

the MSIs were placed according to the manufacturer’sspecifications. Once the MSI was placed, the test man-dible was set in the custom acrylic base and fixed to themounting plate anteriorly and posteriorly. The mandiblewas then oriented to obtain the correct position of thetest site relative to the actuator, as previously described.

With the mandible secured to the test holder and ap-propriately positioned, a digital photograph of the im-plant was taken (DXC-390P digital video, Sony, NewYork, NY; macro lens, Konica Minolta, Wayne, NJ;and imaging software, SCION, Frederick, Md). The0.028-in stainless steel orthodontic wire was insertedthrough the holder and the MSI, and the load screwwas activated to eliminate any play between the MSIand the load cell. The tensile machine position wasthen zeroed, and the test began with the machine mov-ing at a rate of 2 mm per minute.24 Data collected dig-itally included time, load, and position during eachtest. At regular intervals during the testing, digital pho-tographs were recorded and identified with load or posi-tion information to document the changes of the

implant-bone interface during testing. At failure of theimplant-bone interface, the test was discontinued, thetest data were saved, and a final digital image of thepostfailure implant-bone interface was recorded. Foreach test, maximum force at failure was recorded.

During testing, the mandibles were kept moist. Alltests were conducted with the mandibles at room tem-perature (22�C). The mandibles were stored by wrap-ping them in wet paper towels, placing them insealable plastic bags, and freezing them at –5�C. Freez-ing does not adversely affect the elastic properties ofbone measured ultrasonically,31-33 although it mighthave some minor effects on mechanically determinedmaterial properties.34


Preparation of all mandibular specimens, testing,and data recording were performed by 1 tester(M.B.P.). Statistical analysis was completed with statis-tical software (version 14.0, Minitab, State College, Pa).

Fig 1. Test matrix with 9 subgroups.

American Journal of Orthodontics and Dentofacial Orthopedics Pickard et al 93Volume 137, Number 1

Although the variables’ distributions were normal, non-parametric Kruskal-Wallis tests were used for the com-parisons because of the small sample sizes of thesubgroups. The significance level of P \0.05 wasused for all tests.

RESULTS

The pull-out tests of the MSIs aligned at 90� to thecortical surface had a significantly higher maximumforce at failure (342 6 80.9 N; P \0.001) (mean 6

standard deviation) than all other test groups (Table,Fig 2). The loading curve for all 3 pull-out test groupswas largely linear until immediately before failure(Fig 3, A).

At failure, 1 of 2 bone-implant interface configura-tions was generally seen. In most cases, the bone re-mained in intimate contact with the MSI at failure.For MSI failures with intact bone, the 90� MSIs often

had an elliptical surface outline with a conical cross-sectional shape. Generally, the greatest dimension ofbone was in the direction of maximum bone stiffness(Fig 4, A). In other cases, the MSI separated from thebone, with cortical splinters of bone projecting upwardin the direction of the pull-out test.

The pull-out tests of the MSIs aligned at 45� to thebone surface—oriented in the direction of either theminimum or maximum bone stiffness—had a bone-fail-ure wedge in the 135� angle between the MSI and thesurface of the bone (Fig 4, B). Similar to the pull-outtests of the MSIs aligned at 90� to the cortical surface,the widths of the bone fragments were smaller for theMSIs oriented along the direction of maximum bonestiffness than for the MSIs oriented along the directionof minimum bone stiffness.

The maximum forces at failure in tests of MSIs an-gled 45� toward the line of shear force in the direction ofmaximum bone stiffness (253 6 74.1 N; P\0.001) and

Table. Descriptive statistics of maximum force at failure

Maximum force at failure

Test type Implant orientation Mean (n) SD (n) Min (n) Max (n)

Pull-out 90� 341.85* 81.0 257.0 493.2

45� maximum stiffness 107.9 32.1 71.5 160.1

45� minimum stiffness 141.4 57.0 92.8 250.8

Shear test (direction of maximum bone stiffness) 90� 123.8 26.5 85.3 179.3

45�-opposite force 102.3 25.4 74.7 163.0

45�-same as force 253.34* 74.1 152.5 355.6

Shear test (direction of minimum bone stiffness) 90� 138.1 34.6 88.5 174.0

45�-opposing force 87.5 27.2 62.2 123.0

45�-same as force 264.16* 21.0 230.2 278.3

*Maximum force at failure was significantly higher than other tests with P \0.001.

Min, Minimum; Max, Maximum.

Mean Maximum Force at Failure

(with 95% Confidence Interval)

342

108

141124

102

253

138

88

264

0

50

100

150

200

250

300

350

400

450

PO 90° PO 45°

Max

PO 45°

Min

SH Max

90°

SH Max

45° O

SH Max

45° S

SH Min

90°

SH Min

45° O

SH Min

45° S

Test Group

New

to

ns (N

)

Fig 2. Maximum force at failure with 95% CI.


January 2010

those in the direction of minimum bone stiffness (264 6

21.0 N; P \0.001) were significantly higher than theother 4 test groups (Table).

The 45� MSIs opposing the shear line of force mostoften failed in a bimodal fashion (Fig 3, B). The loadingcurve was initially similar to the other test subgroups

Fig 3. Representative loading curves of the 3 modes of implant failure: A, typical pull-out test or shearof 45� implants angled toward the shear force; B, typical shear test of 45� implants angled away fromthe shear force; C, typical shear test of 90� implants.

Fig 4. A, Postfailure bone-implant section of a 90� pull-out test; B, postfailure bone-implant sectionof 45� pull-out test; C, buccal surface response when implant contacts lingual cortex.


until primary failure, which occurred as the MSI startedto rotate from the original 45� placement. The loadingcurve generally flattened while the MSI ‘‘reoriented’’to approximately 90� relative to the cortical surface;as the MSI continued to rotate from 90� to 45� in thesame direction as the applied force, the loading curveusually increased until final failure of the MSI-boneinterface.

Shear tests of the MSIs aligned 90� to the corticalsurface also demonstrated a nonlinear loading curve(Fig 3, C). Similar to the MSIs aligned at 45�and oppos-ing the shear line of force, the loading curve fell after aninitial linear loading response. The slope of the increas-ing load response decreased but remained positive untilfailure. Unlike the MSIs aligned 45� to the cortical sur-face and opposing the shear line of force, a single pointof primary failure was not clearly identifiable.

DISCUSSION

This study was the first to demonstrate the effects ofMSI orientation on the stability and resistance to failureat the bone-implant interface. The stability and resis-tance to failure of bone screws used for rigid fixation,which are similar to MSIs, are known to depend onmany variables including MSI material,24,28,35 diame-ter,24 length, thread design,25 surgical placement proto-col,36 and the recipient bone qualities.20

Although the thread length of the MSIs was only6 mm, the lingual cortex was often reached duringplacement. When contact was made, the MSI was notplaced its full length into the bone. Contact with the lin-gual cortex was based on increased tactile placementtorque and superficial bone layers that ‘‘volcanoed’’ orpulled up along the thread of the screw (Fig 4, C).The rising of the superficial layers suggested axial load-ing. Some MSIs aligned at 45� to the bone surface alsoappeared to contact the lingual cortex. Because of theconsistency of the results, the potentially confoundingeffect of contact with the lingual cortex was limited.A clinical implication is that, when placing an MSI inthe mandible, even the short 6-mm thread length is po-tentially adequate to achieve bicortical achorage and ismore than adequate for monocortical placement. Saka23

concluded that it was unnecessary to have a bone screwmore than 7 mm in length and 2 mm in diameter formonocortical applications in the mandible.

MSIs loaded along their long axes showed the great-est stability and resistance to failure. This finding can beexplained by the fact that the screws that were orientedparallel to the line of force had threads that were perpen-dicular to the load and thus in an optimum position toresist the load. Since this principle applies regardless

of the force levels, the effects of orientation on MSI sta-bility and resistance to failure identified in this studymight apply to the clinical success of MSIs.

Although the physiologic thresholds of bone strainfor the mandible have not been determined,37 it is rea-sonable to suggest that, at levels above physiologicstrain limits, microdamage leading to bone resorptioncan occur.18,19,38 Strain is related to and proportionalto stress or force applied over an area. MSIs havea smaller area and, therefore, higher stress and strain as-sociated with an applied load than the larger dental im-plants. Clearly, the larger the area over which anorthodontic load is applied, the greater the force thatcan be applied without exceeding physiologic strainlimits. As this study confirms, the more closely theline of force and the long axis of the MSI are coincident,the greater the area of distributed load and the greaterthe MSI’s stability and resistance to failure.

Another consideration is that any discrepancy be-tween the MSI orientation and the line of applied forcetends to decrease the uniformity of load distribution onthe screw threads, resulting in disproportionate load dis-tribution at the implant-bone interface.37,39 The greaterthe proportion of the force that is not along the long axisof the MSI, the greater the amount of torque on thebone-implant interface.39 As the angle between thelong axis of the MSI and the line of force increases,stress concentrations increase in unfavorable areassuch as the acute angles of the bone-implant interface.In shear tests of the MSIs angled at 45� away (‘‘tent-pegged’’) from the line of force, the highest stress con-centration would be expected at the 45� degree angle ofbone near the cortical surface (Fig 5). These stress con-centrations resulted in reductions in MSI stability andresistance to failure.

The maximum pull-out force of the MSIs aligned at90� to the bone surface in this study compares wellwith the uniaxial pull-out tests of several miniscrew de-signs 1.5 to 2.0 mm in diameter and 5 to 7 mm long.23

Fig 5. Stress concentrations of 45� implants opposingthe line of shear force.


January 2010

Differences could be due to differences in screw de-sign, length, diameter, and placement location. Hujaet al,40 for example, showed significantly smaller max-imum pull-out forces for 6-mm MSIs placed in the an-terior (134.5 N) rather than in the posterior (388.3 N)mandible. Unfortunately, there are no comparable stud-ies in any bone that report pull-out test results of in-clined MSIs or bone screws. Shear tests have beenperformed on bio-absorbable implants loaded with50 N, but they were not tested to failure.29 Becauseorthodontic MSIs are primarily loaded in shear, morestudies are needed to explore the variables affectingMSI stability and resistance to failure when loadedobliquely or in shear.

A uniform load distribution in the peri-MSI bonemight explain the observed stability and high force atfailure of the uniaxially loaded orthogonal MSIs inpull-out tests. In this situation, the line of force andthe long axis of the MSI are colinear, and the force isdistributed to the surrounding bone uniformly.37,39,41

In most pull-out tests, failure did not occur at thebone-implant interface. Rather, it occurred in the sur-rounding bone. This suggests that the mechanical reten-tion of the implant-bone interface was greater than thecohesive strength of the surrounding bone matrix.

MSIs that have lost their primary stability and be-come displaced can still support an applied load. Thissupports findings of studies in which mobile MSIs couldstill resist orthodontic loading.16,42 In the shear tests, theMSIs aligned at 45� to the cortical surface and awayfrom the direction of loading (tent-peg orientation)lost stability and were displaced from their initial place-ment orientation, but they continued to support a signif-icant applied load. This produced a bimodal loadingcurve with 2 regions (points) of failure and 3 distinctphases of loading-failure behavior (Fig 3, B). Duringthe first phase, the implant-bone interface appears stableand resists the applied load with a linear load responsesimilar to the other pull-out tests (Fig 3, A). However,the MSI then goes through an identifiable ‘‘primary fail-ure’’ in which the implant-bone interface fails, but onlypartially, as indicated by a small decrease in the loadingcurve. During the second phase of loading, the MSI ro-tates from its initial placement position angled awayfrom the direction of loading to about 90� to the corticalsurface. During the third phase, the MSI continues itsrotation toward the line of force until additional load re-sults in ultimate failure. The increase of load observed issurprising, since the rotating MSI has left a trailing bonetrough in which reduced mechanical retention might beexpected. This bimodal pattern might be due to thescrew’s apex being forced into the lingual cortex, atwhich point it functions as a fulcrum or hinge point, re-

stricting the apex from swinging free and pulling thethreaded MSI into the bone on the same side as the ap-plied force. This effectively distributes the load into theleading edge of the resisting bone. When the MSI rea-ches about 45� or less toward the line of force, theapex and the entire MSI can pull out, leading to com-plete and ultimate implant-bone failure. Moreover, thedestructivenes of the failure process results in only par-tial resistance of the screw threads in direct pull-out, andthe force at failure is much less than that in screws thatwere originally angled at 45� in the direction of the load.In many cases, the ultimate load at failure slightly ex-ceeds that which occurred during primary failure, dem-onstrating the resistive effect of the apex wedging intothe lingual cortex during the rotation of the MSI. Thiswedging effect of the apex is supported by the observa-tion that some MSIs showed bending in their taperedapical portions.

It is likely that the apex of the MSI contacted the lin-gual cortex after it began rotating. The MSI acts asa Class II lever arm (load is between the fulcrum pointand the applied force). The apex acts as a fulcrum pointas it wedges into the inner surface of the lingual cortex,the buccal cortex acts as the resisting load, and the ap-plied load is the shear force at the head of the MSI.This effectively magnifies the effective load deliveredto the bone’s buccal cortex. The ratio of the distance be-tween the hole in the head of the MSI to the apex of theMSI (about 9.5 mm) and the distance between the loca-tion of the buccal cortex to the apex of the MSI (about4-5 mm) suggests that the effective load on the bonecan be increased by a factor of about 2. The geometryof the MSI relative to the buccal cortex and the innerportion of the lingual cortex is the likely explanationof the magnitude and mode of implant-bone failure. Inthese tests, the remaining self-tapping section of thescrew is predominantly in medullary bone, which pro-vides little primary mechanical retention.43 The primaryretention of the MSI is in the buccal cortex. The combi-nation of the mechanical lever arm effect and the geo-metric stress concentrations previously discussedresults in decreased stability and lower resistance to fail-ure in MSIs oriented at 45� opposing the line of shearforce, and enables significant movement of the MSIand increased cortical bone damage before ultimatefailure.

Similar to the MSIs aligned at 45� to the cortical sur-face and opposing the shear line of force, the MSIsaligned at 90� to the cortical surface and loaded in shearhad a nonlinear loading curve (Fig 3, C), but withouta clearly identifiable point of primary failure. Theirloading curve response, reduced stability, and lower re-sistance to failure appear to be due to the same


mechanical disadvantage as the MSIs aligned at 45� tothe cortical surface and opposing the shear line of force.

CONCLUSIONS

1. MSIs loaded along their long axis have the greateststability and resistance to failure. The more closelythe long axis of the MSI approximates the line ofapplied force, the greater the stability of the MSIand the greater its resistance to failure. Thus, itmight be important when placing orthodonticMSIs to avoid loading them in a direct shear mode.

2. MSIs angled in the same direction as the appliedload have greater stability and resistance to failurethan MSIs that are ‘‘tent-pegged’’ or orientedaway from the applied load. The mechanical andgeometric disadvantages of the latter orientationreduce MSI stability and resistance to failure.

3. MSIs originally loaded in shear that have lost theirprimary stability and become displaced can stillsupport an applied load, especially if the apex ofthe MSI is initially in contact with the deep surfaceof the lingual cortex. However, failure in this moderesults in greater damage to peri-implant bone thanfailure of MSIs loaded along their long axis.

4. MSI stability and resistance to failure is indepen-dent of MSI orientation along directions of maxi-mum and minimum bone stiffness. However,patterns of anisotropy in cortical bone do affectthe structure of the bone-MSI failure site.

REFERENCES

1. Branemark PI, Aspegren K, Breine U. Microcirculatory studies in

man by high resolution vital microscopy. Angiology 1964;15:

329-32.

2. Branemark PI, Adell R, Breine U, Hansson BO, Lindstrom J,

Ohlsson A. Intra-osseous anchorage of dental prostheses. I. exper-

imental studies. Scand J Plast Reconstr Surg 1969;3:81-100.

3. Adell R, Lekholm U, Rockler B, Branemark PI. A 15-year study

of osseointegrated implants in the treatment of the edentulous jaw.

Int J Oral Surg 1981;10:387-416.

4. Roberts WE, Smith RK, Zilbernam Y, Mozsary PG, Smith RS.

Osseous adaption to continuous loading of rigid endosseous im-

plants. Am J Orthod 1984;86:95-111.

5. Roberts WE, Helm FR, Marshall KJ, Gonglof RK. Rigid endo-

sseous implants for orthodontic and orthopedic anchorage. Angle

Orthod 1989;59:247-56.

6. Turley PK, Kean C, Schur J, Stefanac J, Gray J, Hennos J, et al.

Orthodontic force application to titanium endosseus implants. An-

gle Orthod 1988;58:151-62.

7. Douglass JB, Killiany DM. Dental implants used as orthodontic

anchorage. J Oral Implantol 1987;13:28-38.

8. Odman J, Lekholm U, Jemt T, Branemark PI, Thilander B. Os-

seointegrated titanium implants: a new approach in orthodontic

treatment. Eur J Orthod 1988;10:98-105.

9. Higuchi KW, Slack JM. The use of titanium fixtures for intraoral

anchorage to facilitate orthodontic tooth movement. Int J Oral

Maxillofac Implants 1991;6:338-44.

10. Smalley WM, Shapiro PA, Hohl TH. Osseointegrated titanium

implants for maxillofacial protraction in monkeys. Am J Orthod


11. Southard TE, Buckley MJ, Spivey JD, Krizan KE, Casko JS. In-

trusion anchorage potential of teeth versus rigid endosseous im-

plants: a clinical and radiographic evaluation. Am J Orthod



1997;31:763-7.



Surg 1998;13:201-9.

14. Park HS, Bae SM, Kyung HM, Sung JH. Micro-implant anchor-

age for treatment of skeletal Class I bialveolar protrusion. J Clin

Orthod 2001;35:417-22.

15. Kyung SH, Hong SG, Park YC. Distalization of maxillary molars

with a midpalatal miniscrew. J Clin Orthod 2003;37:22-6.

16. Liou EJW, Pai BCJ, Lin JCY. Do miniscrews remain stationary

under orthodontic forces? Am J Orthod Dentofacial Orthop

2004;126:42-7.

17. Gedrange T, Bourauel C, Kobel C, Harzer W. Three-dimensional

analysis of endosseous palatal implants and bones after vertical, hor-

izontal, and diagonal force application. Eur J Orthod 2003;25:109-15.

18. Frost HM. Wolff’s law and bone’s structural adaptations to me-

chanical usage: an overview for clinicians. Angle Orthod 1994;

64:175-88.

19. Frost HM. A 2003 update of bone physiology and Wolff’s law for

clinicians. Angle Orthod 2004;74:3-15.

20. Koranyi E, Bowman CE, Knecht CD, Janssen M. Holding power

of orthopedic screws in bone. Clin Orthop 1970;72:283-6.

21. Foley WL, Frost DE, Paulin WB, Tucker MR. Uniaxial pull-out

evaluation of internal screw fixation. J Oral Maxillofac Surg

1989;47:277-80.

22. Ellis JA, Laskin DM. Analysis of seating and fracturing torque of

bicortical screws. J Oral Maxillofac Surg 1994;52:483-6.

23. Saka B. Mechanical and biomechanical measurements of five cur-

rently available osteosynthesis systems of self-tapping screws. Br


24. Cheung LK, Zhang Q, Wong MCM, Wong LLS. Stability consid-

erations for internal maxillary distractors. J Cranio Maxillofac

Surg 2003;31:142-8.

25. You ZH, Bell WH, Schneiderman ED, Ashman RB. Biomechan-

ical properties of small bone screws. J Oral Maxillofac Surg 1994;

52:1293-302.

26. Boyle JM, Frost DE, Foley WL, Grady JJ. Torque and pullout

analysis of six currently available self-tapping and ‘‘emergency’’

screws. J Oral Maxillofac Surg 1993;51:45-50.

27. Boyle JM, Frost DE, Foley WL, Grady JJ. Comparison between

uniaxial pull-out tests and torque measurement of 2.0-mm self-tap-

ping screws. Int J Adult Orthod Orthognath Surg 1993;8:129-33.

28. Gosain AK, Song L, Carrao MA, Pintar FA. Biomechanical eval-

uation of titanium, biodegradable plate and screw, and cyanoacry-

late glue fixation systems in craniofacial surgery. Plast Reconstr

Surg 1998;101:582-91.

29. Glatzmaier J, Wehrbein H, Diedrich P. Biodegradable implants for

orthodontic anchorage. A preliminary biomechanical study. Eur J

Orthod 1996;18:465-9.

30. Schwartz-Dabney CL, Dechow PC. Variations in cortical material

properties throughout the human dentate mandible. Am J Phys

Anthropol 2003;120:252-77.


January 2010

31. Evans FG. Preservation effects. In: Evans FG, editor. Mechanical

properties of bone. Springfield, Ill: Charles C. Thomas; 1973. p.

56-60.

32. Dechow PC, Huynh T. Elastic properties and biomechanics of the

baboon mandible [abstract]. Am J Phys Anthropol 1994;22:94-5.

33. Zioupos P, Smith CW, An YH. Factors affecting mechanical proper-

ties of bone. In: An RA, Draughn RA, editors. Mechanical testing of

bone and the bone-implant interface. New York: CRC Press; 2000.

34. Martin RB, Sharkey NA. Mechanical effects of postmortem

changes, preservation, and allograft bone treatments. In: Corwin

SC, editor. Bone mechanics handbook. 2nd ed. Boca Raton, FL:

CRC Press; 2001. p. 1-24.

35. Tengvall P, Skoglund B, Askendal A, Aspenberg P. Surface immo-

bilized bisphosphonate improves stainless-steel screw fixation in

rats. Biomaterials 2004;25:2133-8.

36. Gantous A, Philips JH. The effects of varying pilot hole size on the

holding power of miniscrews and microscrews. Plast Reconstr

Surg 1995;95:1165-9.

37. Cehreli M, Duyck J, De Cooman M, Puers R, Naert I. Implant de-

sign and interface force transfer. A photoelastic and strain-gauge

analysis. Clin Oral Implants Res 2004;15:249-57.

38. Duyck J, Ronold HJ, Oosterwyck HV, Naert I, Sloten JV,

Ellingsen JE. The influence of static and dynamic loading on mar-

ginal bone reactions around osseointegrated implants: an animal

experimental study. Clin Oral Implants Res 2001;12:207-18.

39. Watanabe F, Hata Y, Komatsu S, Ramos TC, Fukuda H. Finite el-

ement analysis of the influence of implant inclination, loading po-

sition, and load direction on stress distribution. Odontology 2003;

91:31-6.

40. Huja SS, Litsky AS, Beck FM, Johnson KA, Larsen PE. Pull-out

strength of monocortical screws placed in the maxillae and mandi-

bles of dogs. Am J Orthod Dentofacial Orthop 2005;127:307-13.

41. Ueda C, Markarian RA, Sendyk CL, Lagana DC. Photoelastic

analysis of stress distribution on parallel and angled implants af-

ter installation of fixed prostheses. Braz Oral Res 2004;18:

45-52.



43. Freudenthaler JW, Haas R, Bantleon HP. Bicortical titanium

screws for critical orthodontic anchorage in the mandible: a pre-

liminary report on clinical applications. Clin Oral Implants Res

2001;12:358-63.


ORIGINAL ARTICLE

Pullout strength of miniscrews placed in anteriormandibles of adult and adolescent dogs: Amicrocomputed tomographic analysis

Zhiqiang Wang,a Zhihe Zhao,b Jing Xue,c Jinlin Song,d Feng Deng,d and Pu Yangb

Jinan, Chengdu, and Chongqing, China

Introduction: Miniscrews often loosen during orthodontic treatment, especially in teenage patients. The pur-poses of this study were to explore the differences of the pullout strengths of miniscrews placed in the anteriormandibles of adolescent and adult dogs and the structural parameters of peri-miniscrew bone, and to analyzethe correlation between the pullout strengths and the variables of the peri-miniscrew bone structure.Methods: Eight adult beagles and 8 young beagles with early permanent dentitions were used as experimen-tal subjects. Two miniscrews were symmetrically placed in the anterior mandible of each dog several minutesbefore death. The bone density, relative bone volume, and cortical bone thickness were evaluated by micro-computed tomography, and the pullout strength of the miniscrew was tested with a testing machine. Regres-sion analyses were used to study the relationship between pullout strength and bone density, relative bonevolume, and cortical bone thickness. Results: The values of bone density, relative bone volume, corticalbone thickness, and pullout strength were 781.94 6 21.46 mg of hydroxyapatite per cubic centimeter, 0.626 0.33, 1.14 6 0.11 mm, and 218.40 6 24.50 N for the adult dogs; and 713.61 6 13.08 mg of hydroxyapatiteper cubic centimeter, 0.57 6 0.20, 1.07 6 0.86 mm, and 130.82 6 2.20 N for the young dogs, respectively. Allpairs of pullout force and bone structural parameters had significant correlation coefficients. The pullout forceshowed the strongest correlation with bone density and the weakest with cortical bone thickness.Conclusions: The values of bone density, relative bone volume, cortical bone thickness, and the pulloutstrength of the adult group were higher than those of the young group. Furthermore, bone density is more sen-sitive in terms of showing pullout force compared with relative bone volume and cortical bone thickness. (Am JOrthod Dentofacial Orthop 2010;137:100-7)

Anchorage control is an important factor in suc-cessful orthodontic treatment. Various tech-niques have been devised and used in

orthodontic practices to reinforce the anchorage. Mini-screws, originally used for intermaxillary or bone fix-ation, have been used as absolute anchorage duringorthodontic treatment.1-5 With this skeletal anchorage,

a wider range of tooth movements—intrusion, retrac-tion, protraction, distalization, and uprighting—canbe achieved without patient compliance.4-10 Comparedwith other temporary anchorage devices, miniscrewshave the following advantages: minimal anatomic lim-itations for placement, lower medical costs, simplerplacement and removal surgeries, less discomfort afterplacement, and immediate or early loading.6,11,12 Theminiscrew has proven successful and is now usefulin clinical applications. Although the reported successrate of the miniscrew is approximately 90%,13-17 in theclinic, miniscrews can loosen during orthodontic treat-ment, often in teenagers.14,18,19 Miyawaki et al14 foundthat the success rate of mini-implants in patients lessthan 20 years old (80%) was slightly lower than inthose more than 20 years old (85%-88%). Park etal18 showed the importance of age for successfulplacement of mini-implants: implant failures were ob-served in 3 patients less than 14 years old, but therewere no failures in 8 patients aged 14 to 28 years.Also, the success rate of miniscrews in patients under16 years old was about 65% in a primary investigation(unpublished data). This is likely to be related to active

a Attending physician, Department of Orthodontics, Provincial Hospital affili-

ated with Shandong University, Jinan, China; graduate student, State Key

Laboratory of Oral Diseases, West China Stomatology Hospital, Sichuan

University, Chengdu, China.b Professor, Department of Orthodontics, State Key Laboratory of Oral Diseases,

West China Stomatology Hospital, Sichuan University, Chengdu, China.c Lector, State Key Laboratory of Oral Diseases, Sichuan University, Chengdu,

China.d Professor, Department of Orthodontics, College of Stomatology, Chongqing

University of Medical Sciences, Chongqing, China.



Supported by grants from the National Nature Science Foundation of China

(30470436 and 10572160).

Reprint requests to: Zhihe Zhao, West China College of Stomatology, Sichuan

University 14#, 3rd section, Renmin South Rd, Chengdu 610041, PR China;


Submitted, September 2007; revised and accepted, January 2008.

0889-5406/$36.00


doi:10.1016/j.ajodo.2008.01.025

100


bone metabolism and low maturation of the bone ingrowing patients. During childhood and adolescence,ossification occurs at a faster rate than bone resorption,so bones grow larger. During the early and middleyears of adulthood, bones neither grow nor shrink,but, after the age of 35 to 40 years, bone loss exceedsbone gain.20

When a mechanical evaluation of the stability ofscrew-shaped implantation including miniscrews ismade, placement and removal torques and pulloutstrengths used to evaluate the shear strength of thebone-implant interface are generally measured.21-23

An and Draughn24 showed that the experimental setupof a torque test is more elaborate than that requiredfor a pullout test. A pullout test is widely used becauseof the relative simplicity of its protocol, which usuallyrequires a uniaxial materials testing machine operatedunder displacement control with a simple support jigfor the push-out test or a hookup system for the pullouttest. Although more reports have shown that these im-mediately or early loaded miniscrews can providegood anchorage control, the stability of miniscrews inteenagers and young animals has seldom been reported.The purposes of this study were to record the pulloutstrengths of miniscrews placed in the anterior mandibles

of adult and adolescent dogs, analyze the peri-mini-screw bone structure by using microcomputed tomogra-phy (mCT), and explore the correlation between thepullout strengths and the variables of peri-miniscrewbone structure.


Sixteen male beagles were used as experimentalsubjects. They were divided into 2 groups accordingto age and dental age; each group had 8 beagles. Theadult dogs were 23 to 24 months old and weighed 13to 14 kg; their mandibular third molars had erupted(Fig 1, A). The dogs in young group were about 9months old and weighed 8 to 8.5 kg; their mandibularsecond and third molars had not erupted (Fig 1, B).The experiment was approved by the bioethics commit-tee of Sichuan University, Chengdu, China. The veteri-nary records indicated that the dogs were healthy.

Thirty-two Aarhus miniscrews (diameter, 1.6 mm;length, 6 mm) provided by the manufacturer (Medicon,Tuttlingen, Germany) were placed in anterior regions ofthe mandibles (Fig 2, A). Each dog received 2 mini-screws several minutes before death. The 2 miniscrewsplaced in each dog were placed on the left and right

Fig 1. A, The dentition of an adult dog (M3, third molar); B, the dentition of a young dog (M1, firstmolar).

Fig 2. A, The screw used in the experiment; B, the pilot drill; C, screws after placement.

American Journal of Orthodontics and Dentofacial Orthopedics Wang et al 101Volume 137, Number 1

sides symmetrically. The surgical implant procedurewas as follows. A mucoperiosteal flap between the 2mandibular canines was opened, and the alveolar bonewas denuded. The cortical bone was drilled witha 1.0-mm pilot drill (Fig 2, B) with water cooling forthe length of the cortical bone thickness. The miniscrewwas placed with a miniature screwdriver to the screwneck; we intended to place the screw perpendicular tothe cortical bone surface (Fig 2, C).

When the dogs were killed, their jaws were removedand sectioned into small blocks within 30 minutes, eachcontaining 1 screw, which was surrounded by at least 4mm of tissue with no soft tissue (Fig 3, A). The blockswere transferred to phosphate-buffered formalin for 48hours and kept in 70% ethanol. For a detailed qualitativeand quantitative 3-dimensional (3D) evaluation, theproximal 4 mm of bone around the miniscrew was ex-amined with an imaging system (mCT 80, Scanco Med-ical, Bassersdorf, Switzerland). For image acquisition,the specimens were mounted on a turntable shifted au-tomatically in an axial direction vertical to the longaxis of the miniscrew. The x-ray tube voltage was setto 70 kV to allow maximum x-ray transmission throughthe highly opaque titanium implant. To maximize thesignal-to-noise ratio, the system was operated at 114mA (maximum current for the 70-kV setting) and thelongest integration time (300 ms). The distance between2 adjacent slices was 20 mm. The titanium and cortical

bone were segmented from each other and from trabec-ular bone with a multi-level thresholding procedure.25,26

The peri-miniscrew bone region of interest included theentire bone compartment between the cross-sectionalplanes 2.8 mm proximally and 2.8 mm distally fromthe miniscrew longitudinal axis (Fig 4). In the analysesof bone specimens, 140 consecutive slices were recon-structed, excluding the teeth and the miniscrew; 3Dbone morphometric quantitative data were calculatedto explore micro-structural specifications of bone nearthe screw. The parameters obtained with mCT included:bone density (milligrams of hydroxyapatite per cubiccentimeter [mg HA/cm3]) of total volume (TV) andrelative bone volume (BV/TV). In the middle slice of

Fig 3. A, A block containing 1 screw; B, the custom-made cylinder with a hole; C, the bone-screwblock after it was embedded; D, AG-IS stretching machine.

Fig 4. The region of interest in white lines, extending 2.8mm proximally and distally to the axis of the miniscrew(black line).

102 Wang et al American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

the miniscrew (Fig 5, A and B), we measured corticalbone thickness.

After the mCT analysis, the specimens were pro-gressively rehydrated and kept in phosphate-bufferedsaline solution for 48 hours before the biomechanicaltesting. For pullout testing, the screw must be alignedwith the axis of the testing machine. This ensured thatno bending moment was created during the pullout testand that only axial pullout strengths were recorded.So, a custon-made cylinder of acrylic resin witha hole was designed (Fig 3, B). The bone-screw blockwas embedded in the hole with acrylic resin. The spec-imen was about 2 mm higher than the upper plane ofthe cylinder so that it was easy to clamp the screwhead to the jig. The screw was placed in the centerof the cylinder, and the long axis of the screw was ver-tical to the plane of the cylinder (Fig 3, C). Determina-

tion of the pullout strength is a standardized method oftesting the mechanical competency or holding powerof a screw. To observe the effects of test variables,standardized biomechanical tests must be conducted.The testing machine used for the pullout test is shownin Figure 3, D (AG-IS stretching machine, TrapeziumSoft, Shimadzu, Japan). The cylinder including thebone-screw block was fixed on the base of the ma-chine. The head of the miniscrew was clamped bya universal jig designed specifically for this studythat could be adjusted in all directions. To avoid pre-stress, we corrected the system after fixing the bone-screw block and clamping the screw head. A crossheadspeed of 2 mm per minute was applied to extract thescrew. The load-displacement data were recorded,and the peak load at failure was obtained from thedata file.

Pullout force, bone density, BV/TV, and corticalbone thickness of the adult and young dogs were ana-lyzed by using t tests. The results were considered sig-nificant at P \0.05. Regression analyses were used tostudy the relationship between pullout strength (depen-dent variable) and bone density, BV/TV, and corticalbone thickness (independent variable). These analyseswere carried out with statistical analysis software (ver-sion 12.0, SPSS, Chicago, Ill).

RESULTS

Twenty-two miniscrews were placed in the mandib-ular anterior regions; mCT examination of the bone-screw blocks showed that 3 miniscrews were placedthrough the roots of adjacent teeth (Fig 5, C), 1 in theleft mandible of an adult dog, and the others in theleft or right mandible of 2 young dogs; the other minis-crews were placed in bones. Although all miniscrewswere tested by the pullout method, we analyzed onlythe 29 blocks with the screws in the bones.

The values of the bone density were 781.94 6 21.46mg of HA per square centimeter for the adult dogs and713.61 6 13.08 mg of HA per square centimeter for theyoung dogs (Table I). There were statistically significantdifferences in bone density between the groups of dogs(P 5 0.000).

The values of BV/TV were 0.62 6 0.33 for the adultdogs and 0.57 6 0.20 for the young dogs (Table II).

Fig 5. mCT slices: A, the middle slice of a miniscrew ina young dog; B, the middle slice of a miniscrew in anadult dog; C, a screw through the roots of a tooth.

Table I. Bone density by age and dental age (mg ofHA/cm3)

Group n Maximum Minimum Mean SD P

Adult dogs 15 823.98 758.36 781.94 21.46 0.000

Young dogs 14 738.94 698.37 713.61 13.08


There were statistically significant differences inBV/TV between the groups (P 5 0.000).

The cortical bone thicknesses were 1.14 6 0.11 mmfor the adult dogs and 1.07 6 0.86 mm for the youngdogs (Table III). There were no significant differencesin cortical bone thickness between the groups (P 5

0.060).Statistically significant differences in the pullout

strengths were found between the groups (218.40 6

24.50 N for the adult dogs vs 130.82 6 2.20 N for theyoung dogs; P 5 0.000; Table IV). The pullout strengthsof miniscrews placed through the roots of teeth were436.37 N for the adult dogs, 328.69 N for 1 youngdog, and 342.17 N for the others. The pullout strengthwas greater than that of miniscrews placed in bonewhen compared with the respective group.

All pairs of pullout force and bone structural param-eters showed significant correlation coefficients. In gen-eral, the pullout force had the strongest correlation withbone density and the weakest with cortical bone thick-ness (Table V).

DISCUSSION

Miniscrews used for orthodontic anchorage can failfor various reasons. The stability of an orthodonticminiscrew throughout treatment depends on bonedensity, peri-implant soft tissues, miniscrew design,surgical technique, and force load.14,27,28 The key deter-minant for stationary anchorage is bone density.29,30

With new imaging technologies, dual-energy x-ray ab-sorptiometry, quantitative computed tomography, andmCT have markedly improved our ability to assessstructural parameters of the boned, including bonedensity.31-33 Yip et al34 thought that the expediency,nondestructive nature, and 3D imagery of the mCT tech-nique is changing the gold standard for assessing min-eral density patterns in bone. Therefore, in this study,

we measured bone density, BV/TV, and cortical bonethickness of dogs’ anterior mandibles using mCT. Ourresults suggested that the corresponding values of themeasured items of the adult group were higher thanthose of the young group, and there were statisticallysignificant differences in bone density and BV/TV be-tween the groups, whereas cortical bone thickness wasnot different. Nine months in a dog is similar to about11 years in a human, and 23 to 24 months in a dog issimilar to about 23 to 25 years in a human. Throughoutlife, bone formation (ossification) and bone destruction(resorption) proceed concurrently; ossification occurs ata faster rate than bone resorption until the early years ofadulthood, and bone quality and quantity increase withage.20 Pope et al35 stated that the rhesus monkey hasa natural pattern of change in bone mineralization thatparallels that in humans; bone density increases withage until about 30 years. We speculated that bonegrowth of dogs is similar to that of humans and rhesusmonkeys. From 9 to 24 months, a dog is in the growingstage. Bone gain outstrips bone loss. So bone densityand BV/TV increase with age in this stage. Corticalbone thickness does not significantly intensify, for thepossible reason that the dog’s bone volume mainlyforms in its adolescence; from adolescence to adult-hood, the main development is bone calcification.Huja et al23 indicated that the cortical bone thicknessof the mandibular anterior regions of adult dogs wasabout 1.3 mm; this is higher than our results (1.14mm). The cause may be that they measured it afterdrawing the screw, and the peri-bone of the screw wasdestroyed; this might have affected the results.

In our experiment, we examined the holding powerof the screws in simulated conditions when a screw isplaced in the jaws. No healing or adaptive responsecould occur. Thus, the holding power of these screws

Table II. BV/TV by age and dental age


Adult dogs 15 0.69 0.58 0.62 0.33 0.000

Young dogs 14 0.60 0.54 0.57 0.20

Table III. Cortical bone thickness by age and dental age(mm)


Adult dogs 15 1.34 0.98 1.14 0.11 0.060

Young dogs 14 1.16 0.89 1.07 0.86

Table IV. Pullout force by age and dental age (N)


Adult dogs 15 268.70 192.34 218.40 24.50 0.000

Young dogs 14 142.74 119.96 130.82 2.20

Table V. Correlation coefficients (r2 values) and P valuesbetween pullout force and the variables of peri-mini-screw bone structure

Bone density BV/TV Cortical bone thickness

Pullout force 0.92 0.81 0.263

P 0.000 0.000 0.004

All coefficients are significant at P \0.005.


January 2010

indicated their primary stability. Also, our resultsshowed that the mean pullout strength in the adult groupwas greater than that of the young group, with significantdifferences between them. It is known that the quality(bone density) of alveolar bone and bone-to-metalcontact are the important factors for the stability ofimplants. 29,30,36,37 In this experiment, we placed thesescrews using the same method, with screw neck-bonecontact, which ensured that the length of each screw inbone was nearly equal. So, the differences in pulloutstrengths were perhaps caused by different bone quality.Okuyama et al21 found a high correlation between place-ment torque and bone mineral density. Battula et al38

placed screws in normal and osteoporotic bones andmeasured the pullout strengths of the screws; they founda correlation between bone material and holding power;the pullout strengths of screws placed in normal boneswere greater than those of screws placed in osteoporoticbones. Hitchon et al39 also found a significant correlationbetween pullout strength and bone mineral density. Thepullout strengths of miniscrews placed through the rootsof teeth were 436.37 N for the adult dogs, 328.69 N for 1young dog, and 342.17 N for the others. The possiblereason was that the stiffness of dentin is higher thanthat of the mandible. This also showed that the quantityof peri-miniscrew tissue affected the pullout strength.Lim et al40 stated that an increase in screw diametercan efficiently reinforce the initial stability of the mini-screw. In an 8-week comparison study of pulloutstrength of machined vs dual-etched screws placed inthe rabbit tibiae, Baker et al41 found that, in short-termhealing in rabbit tibiae, the titanium implant witha dual-etched surface demonstrated a more rapid rateof pullout strength gain than that of the machined sur-face. Another study evaluated the holding power of tita-nium screws placed in 19 cadaver spines.39 In this study,a longer screw had a greater pullout force. Pulloutstrength was strongly related to screw length. A clinicalstudy19 showed that 3 months of healing before ortho-dontic loading significantly improved the success rateof mini-implants from 63% to 97% in adolescents.Therefore, in the clinic, mini-implants should be placedbefore starting orthodontic treatment in adolescents toprovide a latent period of more than 3 months. Thesestudies indicated that the design of the screw and thehealing time before orthodontic loading were relatedto the pullout strength and the success rate. When screwsare placed in low-density bone (jaws of osteoporotic oradolescent patients), diameter, length, and surface mate-rials of the miniscrew and healing time before orthodon-tic loading should be considered. Also, the proximity ofthe root at the implanted site should be considered. Hujaet al23 also showed that the pullout strengths of screws

placed in the mandibular anterior regions of adult dogswere about 134.5 N; this is lower than our results(218.40 N). For 1 reason, the screws used in these exper-iments were different. For another reason, the beagles inexperiments came from different regions.

We wanted to determine whether bone density, BV/TV, and cortical bone thickness could predict pulloutstrength. We found a weak but significant (r2 5 0.263;P 5 0.004) positive correlation between cortical bonethickness and pullout force, a significant (r2 5 0.81;P 5 0.000) positive correlation between BV/TVand pullout force, and the strongest (r2 5 0.92; P 5

0.000) positive correlation between bone density andpullout force. These findings suggest that bone densityis more sensitive in terms of showing pullout force com-pared with BV/TV or cortical bone thickness. That thecorrelation coefficient between cortical bone thicknessand pullout force in our study is higher than in the studyof Huja et al23 was attributable to standardization of im-plant placement, long axis of the screw vertical to theplane of cylinder, high precision of the 3D mCT imag-ing, quantitative evaluation, and the jig used for biome-chanical testing that permitted the application of anaxial vector.42

In clinical practice, the forces required for orthodon-tic tooth movement can range from approximately 0.3 to4 N.43 The direction of the orthodontic force was usuallytangential. A recent study suggested that screws testedin the axial mode have 34% higher pullout strengthsthan the same screws tested in the tangential (cantilever)mode.44 Correcting our results downward by 34% stillindicated sufficient strength to withstand orthodonticloads. But, in clinics, failures are still seen, especiallyin young patients. Park45 thought that failures mightbe caused by movable oral mucosa, excessive heat gen-erated during placement because of thick and dense cor-tical bone, or irritation from food. In our study, weconcluded that the values of bone density, BV/TV, cor-tical bone thickness, and pullout strength of the adultgroup were higher than those of the young group. In fur-ther studies, we will evaluate the pullout strengths ofminiscrews with different healing times (bone adapta-tion) to improve the success rate of screw anchoragein young orthodontic patients.

CONCLUSIONS

For the miniscrews placed in anterior mandibles toobtain orthodontic anchorage, we concluded the follow-ing.

1. The values of bone density, BV/TV, and corticalbone thickness of the adult dogs were higher thanthose of the young dogs. There were significant


differences in bone density and BV/TV betweenadult and adolescent dogs, whereas cortical bonethickness was not different.

2. There were significant differences in the pulloutstrengths between adult and adolescent dogs. Therewas a trend for greater pullout strengths in the adults.

3. All pairs of pullout force and bone structural pa-rameters had significant correlation coefficients.Pullout force had the strongest correlation withbone density and the weakest with cortical bonethickness.

4. Bone density is more sensitive for showing pulloutforce compared with BV/TVor cortical bone thick-ness.

5. To improve the success rate of screw anchorage inyoung orthodontic patients, the pullout strengthsof miniscrews with different healing times shouldbe studied.

The authors thank Beijing Jia Lian Cheng Ye Med-ical Instrument Company for providing the miniscrewsat a discount.

REFERENCES


1997;31:763-7.



Surg 1998;13:201-9.

3. Freudenthaler JW, Bantleon HP, Haas R. Bicortical titanium

screws for critical orthodontic anchorage in the mandible: a pre-

liminary report on clinical applications. Clin Oral Implants Res

2001;12:358-63.

4. Park HS, Kyung HM, Sung JH. A simple method of molar upright-

ing with microimplant anchorage. J Clin Orthod 2002;36:592-6.

5. Umemori M, Sugawara J, Mitani H, Nagasaka H, Kawamura H.

Skeletal anchorage system for open-bite correction. Am J Orthod


6. Park HS, Bae SM, Kyung HM, Sung JH. Micro-implant anchor-

age for treatment of skeletal Class I bialveolar protrusion. J Clin

Orthod 2001;35:417-22.

7. Kawakami M, Miyawaki S, Noguchi H, Kirita T. Screw-type im-

plants used as anchorage for lingual orthodontic mechanics: a case

of bimaxillary protrusion with second premolar extraction. Angle

Orthod 2004;74:715-9.

8. Byloff FK, Karcher H, Clar E, Stoff F. An implant to eliminate an-

chorage loss during molar distalization: a case report involving

the Graz implant-supported pendulum. Int J Adult Orthod Orthog-

nath Surg 2000;15:129-37.

9. Giancotti A, Arcuri C, Barlattani A. Treatment of ectopic mandib-

ular second molar with titanium miniscrews. Am J Orthod Dento-


10. Sugawara J, Daimaruya T, Umemori M, Nagasaka H, Takahashi I,

Kawamura H, et al. Distal movement of mandibular molars in

adult patients with the skeletal anchorage system. Am J Orthod


11. Kanomi R. Application of titanium mini-implant system for

orthodontic anchorage. In: Davidovitch Z, Mah J, editors. Biolog-

ical mechanisms of tooth movement and craniofacial adaptation.

Birmingham, Ala: EBSCO Media; 2000. p. 253-8.

12. Deguchi T, Takano-Yamamoto T, Kanomi R, Hartsfield JK Jr,

Roberts WE, Garetto LP. The use of small titanium screws for

orthodontic anchorage. J Dent Res 2003;82:377-81.


risk factors associated with failure of mini-implants used for ortho-

dontic anchorage. Int J Oral Maxillofac Implants 2004;19:100-6.

14. Miyawaki S, Koyama I, Inoue M, Mishima K, Sugahara T, Ta-




15. Park HS, Jeong SH, Kwon OW. Factors affecting the clinical suc-

cess of screw implants used as orthodontic anchorage. Am J


16. Tseng YC, Hsieh CH, Chen CH, Shen YS, Huang IY, Chen CM.

The application of mini-implants for orthodontic anchorage. Int J

Oral Maxillofac Surg 2006;35:704-7.



anchorage: success rates and postoperative discomfort. Am J Or-

thod Dentofacial Orthop 2007;131:9-15.

18. Park HS, Lee SK, Kwon OW. Group distal movement of teeth us-

ing microscrew implant anchorage. Angle Orthod 2005;75:602-9.



36:695-9.

20. Thibodeau GA, Patton KT, editors. Anthony’s textbook of anat-

omy and physiology. 18th ed. St Louis: Mosby; 2007. p. 229-54.

21. Okuyama K, Abe E, Suzuki T, Tamura Y, Chiba M, Sato K. Can

insertional torque predict screw loosening and related failures?

An in vivo study of pedicle screw fixation augmenting posterior

lumbar interbody fusion. Spine 2000;25:858-64.

22. Baker D, London RM, O’Neal R. Rate of pull-out strength gain of

dual-etched titanium implants: a comparative study in rabbits. Int

J Oral Maxillofac Implants 1999;14:722-8.

23. Huja SS, Litsky AS, Beck FM, Johnson KA, Larsen P. Pullout

strength of monocortical screws placed in the maxillae and man-

dibles of dogs. Am J Orthod Dentofacial Orthop 2005;127:

307-13.

24. An YH, Draughn RA. Implant pushout and pullout tests in

mechanical testing of bone and the bone-implant interface.

Boca Raton: CRC Press; 2000. p. 463-76.

25. Ruegsegger P, Koller B, Muller R. A microtomographic system

for the nondestructive evaluation of bone architecture. Calcif Tis-

sue Int 1996;58:24-9.

26. Buie HR, Campbell GM, Klinck RJ, MacNeil JA, Boyd SK. Au-

tomatic segmentation of cortical and trabecular compartments

based on a dual threshold technique for in vivo micro-CT bone

analysis. Bone 2007;41:505-15.

27. Buchter A, Kleinheinz J, Wiesmann HP, Kersken J,

Nienkemper M, Weyhrother H, et al. Biological and biomechan-

ical evaluation of bone remodeling and implant stability after us-

ing an osteotome technique. Clin Oral Implants Res 2005;16:1-8.

28. Sevimay M, Turhan F, Kilicarslan MA, Eskitascioglu G. Three-

dimensional finite element analysis of the effect of different

bone quality of stress distribution in an implant-supported crown.

J Prosthet Dent 2005;93:227-34.

29. Lee JS, Kim DH, Park YC, Kyung SH, Kim TK. The efficient use

of midpalatal miniscrew implants. Angle Orthod 2004;74:711-4.

30. Misch CE. Density of bone: effect on treatment plans, surgical ap-

proach, healing, and progressive bone loading. Int J Oral Implan-

tol 1990;6:23-31.


January 2010

31. Bagi CM, Hanson N, Andresen C, Pero R, Lariviere R,

Turner CH, et al. The use of micro-CT to evaluate cortical bone

geometry and strength in nude rats: correlation with mechanical

testing, pQCT and DXA. Bone 2006;38:136-44.

32. Park HS, Lee YJ, Jeong SH, Kwon TG. Density of the alveolar

and basal bones of the maxilla and the mandible. Am J Orthod


33. Dinea H, Sadikolu Y, Savci G, Demirci A, Tuncel E. Bone mineral

density measurement by quantitative computed tomography in

a normal Turkish population. Eur J Radiol 1995;21:79-83.

34. Yip G, Schneider P, Roberts EW. Micro-computed tomography:

high resolution imaging of bone and implants in three dimensions.

Semin Orthod 2004;10:174-87.

35. Pope NS, Gould KG, Anderson DC, Mann DR. Effects of age and

sex on bone density in the rhesus monkey. Bone 1989;10:109-12.

36. Albrektsson T, Branemark PI, Hansson HA, Lindstrom J.

Osseointegrated titanium implants. Requirements for ensuring

a long-lasting, direct bone-to-implant anchorage in man. Acta

Orthop Scand 1981;52:155-70.

37. Kim JW, Ahn SJ, Chang Y. Histomorphometric and mechanical

analyses of the drill-free screw as orthodontic anchorage. Am J


38. Battula S, Schoenfeld A, Vrabec G, Njus GO. Experimental eval-

uation of the holding power/stiffness of the self-tapping bone

screws in normal and osteoporotic bone material. Clin Biomech

(Bristol, Avon) 2006;21:533-7.

39. Hitchon PW, Brenton MD, Coppes JK, From AM, Torner JC. Fac-

tors affecting the pullout strength of self-drilling and self-tapping

anterior cervical screws. Spine 2003;28:9-13.

40. Lim SA, Cha JY, Hwang CJ. Insertion torque of orthodontic min-

iscrews according to changes in shape, diameter and length. Angle

Orthod 2008;78:234-40.

41. Baker D, London RM, O’Neal R. Rate of pull-out strength gain of

dual-etched titanium implants: a comparative study in rabbits. Int

J Oral Maxillofac Implants 1999;14:722-8.

42. Gabet Y, Muller R, Levy J, Dimarchi R, Chorev M, Bab I,

et al. Parathyroid hormone 1-34 enhances titanium implant an-

chorage in low-density trabecular bone: a correlative micro-

computed tomographic and biomechanical analysis. Bone

2006;39:276-82.

43. Ren Y, Maltha JC, Kuijpers-Jagtman AM. Optimum force magni-

tude for orthodontic tooth movement: a systematic literature

review. Angle Orthod 2003;73:86-92.

44. Pierce WA, Sucato DJ, Young S, Picetti G, Morgan DM. Ax-

ial and tangential pullout strength of uni-cortical and bi-corti-

cal anterior instrumentation screws. Proceedings of the 49th

Annual Meeting of the Orthopaedic Research Society; Febru-

ary 2-5, 2003; New Orleans, La. Rosemont, Ill: Orthopedic

Research Society; 2003.

45. Park HS. Clinical study on success rate of microscrew im-

plants for orthodontic anchorage. Korean J Orthod 2003;33:

151-6.


REVIEW ARTICLE

Miniscrews in orthodontic treatment: Review andanalysis of published clinical trials

Adriano G. Crismani,a Michael H. Bertl,b Ales G. Celar,a Hans-Peter Bantleon,a and Charles J. Burstonec

Vienna, Austria, and Farmington, Conn

Introduction: A systematic review of effects related to patient, screw, surgery, and loading on the stability ofminiscrews was conducted. Methods: Reports of clinical trials published before September 2007 with at least30 miniscrews were reviewed. Parameters examined were patient sex and age, location and method of screwplacement, screw length and diameter, time, and amount of loading. Results: Fourteen clinical trials included452 patients and 1519 screws. The mean overall success rate was 83.8% 6 7.4%. Patient sex showed no sig-nificant differences. In terms of age, 1 of 5 studies with patients over 30 years of age showed a significant dif-ference (P\0.05). Screw diameters of 1 to 1.1 mm yielded significantly lower success rates than those of 1.5to 2.3 mm. One study reported significantly lower success rates for 6-mm vs 8-mm long miniscrews (72% vs90%). Screw placement with or without a surgical flap showed contradictory results between studies. Threestudies showed significantly higher success rates for maxillary than for mandibular screws. Loading and heal-ing period were not significant in the miniscrews’ success rates. Conclusions: All 14 articles described suc-cess rates sufficient for orthodontic treatment. Placement protocols varied markedly. Screws under 8 mm inlength and 1.2 mm in diameter should be avoided. Immediate or early loading up to 200 cN was adequate andshowed no significant influence on screw stability. (Am J Orthod Dentofacial Orthop 2010;137:108-13)

In 1997, Kanomi1 first mentioned a temporarilyplaced miniscrew for orthodontic anchorage. Thefollowing years brought more refined screw de-

signs.2 Miniscrews have now become established ortho-dontic anchorage aids, with diameters of 1 to 2.3 mmand lengths of 4 to 21 mm.3-16 Nevertheless, manycase reports and only a few comprehensive studieshave been published on orthodontic miniscrews. The ar-ticles promised bright treatment prospects but oftenlacked evidence-based results.17,18 Hence, studies onscrew design as well as surgical and orthodontic treat-ment procedures are much needed. The aims of this re-view were to analyze the reported success rates ofminiscrews and to define guidelines for their selectionand application.


A Medline search was conducted with 2 search-termcombinations: [screw orthodontic] and [implant ortho-dontic]. The terms were chosen generally, since therehas been little conformity on the nomenclature.19 Aset of criteria was defined to subsequently filter the re-sulting articles: (1) studies on orthodontic mini-im-plants or miniscrews published in either English orGerman, (2) human clinical trials, (3) no case reportsor case series, (4) no studies with fewer than 30 mini-screws, and (5) additional data on factors related tothe patient, miniscrew, surgery, and loading availablefor correlation with the miniscrews’ success rates.

References in the criteria-matching results weresearched for additional articles.

When available, data were extracted that correlatedwith the miniscrews’ success rate: patient sex and age,screw length and diameter, method and location ofplacement, time, and amount of loading.

The statistical analyses of these uniformly formatteddata and their graphic representation were made by us-ing SPSS software (version 13 for Mac OS X, SPSS,Chicago, Ill).

RESULTS

As of September 2007, the MedLine search with theterms [screw orthodontic] and [implant orthodontic]returned 734 results. Of those, 14 articles matched allcriteria and were considered (Table I).

aProfessor, Department of Orthodontics, School of Dentistry, Medical Univer-

sity of Vienna, Vienna, Austria.bResident, Department of Orthodontics, School of Dentistry, Medical University

of Vienna, Vienna, Austria.cProfessor, Division of Orthodontics, School of Dental Medicine, University of

Connecticut, Farmington.

The authors report no commercial, financial, or propriety interest in the products

or companies described in this article.

Reprint requests to: Adriano G. Crismani, University Clinic of Orthodontics,

Department of Dentistry and Maxillofacial Surgery, Medical University of

Innsbruck, Innsbruck, Austria; e-mail, [email protected].

Submitted, December 2007; revised and accepted, January 2008.

0889-5406/$36.00


doi:10.1016/j.ajodo.2008.01.027

108


The overall success rates were available in all 14 ar-ticles and ranged from 59.4% to 100%. The meansuccess rate for all 14 studies averaged 83.6% 6

10.2%. Weighted by the number of screws in each study,the mean success rate was 83.8% 6 7.4%. The numberof treated patients ranged from 13 to 129, with the num-ber of miniscrews from 30 to 273. Overall, the analyzeddata comprised 452 patients treated with 1519 screws(Table II).

The success rate of the miniscrews was broken downby sex in 6 studies3,7-9,13,16 and by age in 4 studies.3,7,13,16

This did not include the findings of Liou et al,5 whotreated only female patients of 1 age group. No statisti-cally significant findings in terms of patient sex were ob-served in any studies, but Chen et al16 found significantlygreater success in patients over 30 years of age (Table III).

The data for screw length and diameter were givenin all included studies. Length ranged from 4 to 21mm and diameter from 1 to 2.3 mm. Nine studies re-ported success rates for screws of either differentlengths or different diameters.3,4,7,9-11,13,14,16 Three ofthem reported statistically significant findings(P \0.05). Miyawaki et al3 concluded that their 1-mmthick screw performed significantly worse than thosewith diameters of 1.5 and 2.3 mm. Similarly, Wiech-mann et al14 reported worse results for 1.1-mm thickscrews than for 1.6-mm ones. In the study of Chen etal,10 the success rate of 6-mm long screws was signifi-cantly lower than that of 8-mm long ones (P \0.05)(Table IV).

The method and the location of screw placementwere described in all studies. All authors but Miyawakiet al3 used pilot drilling before placing the screw. Bothflap and flapless surgery were performed. Fourstudies4,5,7,9 used a mucoperiosteal flap, and 7 stud-ies6,8,10,11,14-16 used a flapless method. In the remaining3 studies, both flap and flapless surgery were performed,

Table I. Articles reviewed

Authors Title Journal Year

Miyawaki et al3 Factors associated with the stability of titanium screws placed in the posterior

region for orthodontic anchorage

Am J Orthod Dentofacial Orthop 2003

Cheng et al 4 A prospective study of the risk factors associated with failure of mini-implants

used for orthodontic anchorage

Int J Maxillofac Implants 2004

Liou et al 5 Do miniscrews remain stationary under orthodontic forces? Am J Orthod Dentofacial Orthop 2004

Fritz et al 6 Clinical suitability of titanium microscrews for orthodontic anchorage—

preliminary experiences

J Orofac Orthop 2004

Park et al 7 Group distal movement of teeth using microscrew implant anchorage Angle Orthod 2005

Motoyoshi et al 8 Recommended placement torque when tightening an orthodontic mini-implant Clin Oral Implants Res 2006

Park et al 9 Factors affecting the clinical success of screw implants used as orthodontic

anchorage


Chen et al 10 The use of microimplants in orthodontic anchorage J Oral Maxillofac Surg 2006

Tseng et al 11 The application of mini-implants for orthodontic anchorage Int J Oral Maxillofac Surg 2006

Herman et al 12 Mini-implant anchorage for maxillary canine retraction: a pilot study Am J Orthod Dentofacial Orthop 2006

Kuroda et al 13 Clinical use of miniscrew implants as orthodontic anchorage: success rates and

postoperative discomfort


Wiechmann et al 14 Success rate of mini- and micro-implants used for orthodontic anchorage:

a prospective clinical study

Clin Oral Implants Res 2007

Motoyoshi et al 15 Application of orthodontic mini-implants in adolescents Int J Oral Maxillofac Surg 2007

Chen et al16 A retrospective analysis of the failure rate of three different orthodontic skeletal

anchorage systems

Clin Oral Implants Res 2007

Table II. Overall success rate

Authors Screws (n) Patients (n) Success rate (%)

Miyawaki et al3 134 44 77.8

Cheng et al4 92 44* 91.3

Liou et al5 32 16 100

Fritz et al6 36 17 70.0

Park et al7 (2005) 30 13 90.0

Motoyoshi et al8

(2006)

124 41 85.5

Park et al9 (2006) 227 87 91.6

Chen et al10 (2006) 59 29 84.7

Tseng et al11 45 25 91.1

Herman et al12 49 16 59.4

Kuroda et al13 116 58 86.2

Wiechmann et al14 133 49 76.7

Motoyoshi et al15

(2007)

169 57 85.2

Chen et al16 (2007) 273 129* 81.0

Total 1519 452 Mean: 83.6 6 10.2

Mean weighted by

number of screws:

83.8 6 7.4

*P \0.05.

American Journal of Orthodontics and Dentofacial Orthopedics Crismani et al 109Volume 137, Number 1

and the corresponding success rates were exam-ined.3,12,13 Although Miyawaki et al3 and Kuroda etal13 reported slightly higher success rates for flaplessplacement, the results of Herman et al12 were consider-ably better with a flap (Fig).

When information was available, 958 screws wereused in the maxilla with a mean success rate 89.0% 6

10.0%. In the mandible, 508 screws averaged a successrate of 79.6% 6 8.7%. Weighted by the number ofscrews in each location, the mean success rates were87.9% 6 7.6% for the maxilla and 80.4% 6 8.5% forthe mandible. In 3 studies, the differences between themaxilla and the mandible were significant (P \0.05)in favor of the maxilla (Table V).4,9,16

Three studies explored the differences betweenscrews placed in the patient’s left or right side.8,9,15

There was a significant finding (P \0.05) by Park etal9 in favor of the left side. The success rate for theleft side was also higher in the studies by Motoyoshiet al,8,15 although not significantly different.

The loading variables under evaluation were latencyperiod (time from placement to first loading), overalltime of loading, and force applied to the screws.

Data on the time until loading were available in 13studies and ranged from immediate loading (#48hours20)to 149 days. Most screws were thereby incorporated in theorthodontic mechanics, according to the definitions for im-mediate and early loading of implants.20 Four studies pro-vided success rates for different healing periods.3,13,15,16

Motoyoshi et al15 reported significant differences betweenthe adolescent early-load group and the late-load group butalso for the adult early-load group (P\0.05).

The amount of loading, available in 13 studies, rangedfrom 50 to 400 cN. Kuroda et al13 analyzed differentforces but found no significant differences (Table VI).

DISCUSSION

This review shows that miniscrews have been usedat success rates of 83% in orthodontic patients. By com-parison, palatal implants21 and miniplates3,16 reached90% to 95%. Dental implants have less than a 7% to9% risk of failure over 10 to 15 years.22,23 Miniscrewshave the advantage of simple surgical application, andan orthodontist can perform the procedure. Screwswith diameters of 1.2 mm or greater were universallyused at success rates above 70%.

Another significant factor has been length. Chen etal10 increased the success rate from 72% to 90% by us-ing 8-mm instead of 6-mm long screws. Four studiesalso showed higher success rates with additional lengthat the same diameter but without a statistically signifi-cant difference.7,10,11,13

Increasing screw diameter and length also raised therisk of root damage during placement.15 Miniscrews of1.2-mm diameter and at least 8-mm length have sufficientstability with a minimum risk of root damage. Carefulplanning and radiographic evaluation of the placementsite can further minimize this risk.24,25 Although it hasbeen shown in animal studies that minimal root damageduring placement can be repaired on removal of thescrew,26 root contact or even proximity is also considereda major factor for the failure of miniscrews.27,28

Considering the surgical protocols, a mucoperiostealflap is associated with more patient discomfort, swelling,and pain. The flapless method is less time-consumingbut might hinder a more precise placement of the screw.12

For both protocols, opposing success rates have beenpublished.3,12,13 Only 1 study reported higher successrates (100%) for screws placed with flap surgery.12

However, only 10 screws were used with this procedure.Further investigations are needed to clarify this issue.

Table III. Success rates (%) by sex and age

Authors Male Female \20 years 20-30 years .30 years Mean age

Miyawaki et al3 80 84.7 80.3 88.2 85.0 21.8

Cheng et al4 – – – – – 29.0

Liou et al5 – 100 – 100 – –

Fritz et al6 – – – – – 29.0

Park et al7 (2005) 91.7 88.9 83.3 100 – 17.9

Motoyoshi et al8 (2006) 90.0 85.1 – – – 24.9

Park et al9 (2006) 88.8 93.5 – – – 15.5

Chen et al10 (2006) – – – – – 29.8

Tseng et al11 – – – – – 29.9

Herman et al12 – – – – – 13.7

Kuroda et al13 85.7 88.9 92.5 82.4 100 21.8

Wiechmann et al14 – – – – – 26.9

Motoyoshi et al15 (2007) – – 78.3 91.9 – –

Chen et al16 (2007) 84.4 85.4 78.3* 84.1* 93.6* 24.5

*P \0.05.

110 Crismani et al American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

The angle between the miniscrew’s long axis and thecortical bone was evaluated only by Park et al.9 Withoutmajor differences in the success rates, it was argued thatplacing screws not perpendicular to the bone surface,but at an obtuse angle, lowered the risk of root damageand increased the screw’s contact with cortical bone.Liou et al29 analyzed the infrazygomatic crest in com-puted tomography scans and concluded that miniscrewsshould be placed at a steeper angle in this area.

Screw success rates were higher in the maxilla thanin the mandible in all but 1 study.15 Three studies signif-icantly supported the maxilla as a more suitable place-ment site for miniscrews.4,9,16 All 3 studies interpreted

the lower success in the mandible as a consequence ofoverheating the bone during placement. Particular caremust therefore be used for pilot drilling, screw tighten-ing, and rinsing. In addition, mandibular miniscrewsmight be more exposed to masticatory interferences.

Table IV. Success rate (%) by screw length and diameter(mm)

AuthorsScrews

(n) Length DiameterSuccess

rate

Miyawaki et al3 10 6 1 0.0*

101 11 1.5 83.9*

23 14 2.3 85*

Cheng et al4 48 5, 7 2 85.4

31 9 2 93.5

31 11 2 93.0

20 13 2 85.0

10 15 2 90.0

Liou et al5 32 17 2 100

Fritz et al6 36 6, 8, 10 1.4, 1.6, 2 70.0

Park et al7 (2005) 4 4 1.2 75.0

14 6 1.2 86.7

8 8 1.2 100

2 10 1.2 100

2 15 2 100

Motoyoshi et al8 (2006) 124 8 1.6 85.5

Park et al9 (2006) 19 5 1.2 84.2

157 6-8 1.2 93.6

46 4-10 1.2 89.1

5 10-15 2 80.0

Chen et al10 (2006) 18 6 1.2 72.2*

41 8 1.2 90.2*

Tseng et al11 15 8 2 80.0

10 10 2 90.0

12 12 2 100

8 14 2 100

Herman et al12 49 6-10 1.8 61.2

Kuroda et al13 37 7, 11 2, 2.3 81.1

13 6 1.3 69.2

6 7 1.3 83.3

45 8 1.3 93.3

12 10 1.3 91.7

3 12 1.3 100

Wiechmann et al14 79 5-10 1.1 59.6*

54 5-10 1.6 87.0*

Motoyoshi et al15 (2007) 169 8 1.6 91.9

Chen et al16 (2007) 72 4-10 1.2 76.4

201 5-21 2 82.6

*P \0.05.

Fig. Comparison of success rates for screws placedwith and without flap surgery.

Table V. Success rate (%) by location of screw place-ment

Maxilla Mandible

AuthorsScrews

(n)Success

rateScrews

(n)Success

rate

Miyawaki et al3 63 84.1 61 83.6

Cheng et al4 – 93.3* – 77.1*

Liou et al5 32 100 – –

Fritz et al6 18 – 18 –

Park et al7 (2005) 8 100 22 86.4

Motoyoshi et al8 (2006) 80 88.8 44 79.5

Park et al9 (2006) 124 96.0* 103 86.4*

Chen et al10 (2006) 43 86.0 16 81.3

Tseng et al11 27 96.3 18 83.3

Herman et al12 49 61.2 – –

Kuroda et al13 61 91.8 18 77.8

Wiechmann et al14 90 86.7 43 55.8

Motoyoshi et al15 (2007) 100 84.0 69 87.0

Chen et al16 (2007) 263 88.2* 96 77.1*

Total 958 Mean: 89.0

6 10.0

508 Mean: 79.6

6 8.7

Weighted by number

of screws

Mean weighted

by number

of screws:

87.9 6 7.6

Mean weighted

by number

of screws:

80.4 6 8.5

*P \0.05.


Significantly higher success on the left side might bea statistical outlier. Unilateral preference of the mastica-tion side and potentially better oral hygiene on the leftside because of the prevalence of right-handed patientsmight be further explanations.

A more critical issue is the determination ofa placement site. There are preferable areas that offersufficient bone and root distances. Poggio et al30 eval-uated tomographic images of mandibles and maxillaeto define ‘‘safe zones’’ for placing miniscrews. In themaxilla, they recommended interradicular spaces be-tween the canine and the second molar on the palatalside, and between the canine and the first molar onthe buccal side. In the mandible, they suggested inter-radicular spaces between the canine and the secondmolar.

Details of screw design might play another role inscrew stability but were not evaluated in any of thesestudies. Comprehensive research on thread width andpitch or different head or end configurations has yet tobe published.

Stability of dental implants relies on osseointegra-tion, which requires several weeks for the host bone toachieve intimate contact with the implant.31,32 Gener-ally, miniscrews are loaded sooner. No included studyrecommended longer healing periods for higher success

rates. Therefore, miniscrews can be loaded within daysafter placement with forces up to 200 cN.

Whereas miniscrews are sometimes believed toachieve their retention purely mechanically, the ideaof osseointegration should not be dismissed entirely.19

Miniscrews with an acid-etched surface for enhancedosseointegration have been used successfully.33 Partialosseointegration has since been shown in both in-vitroand in-vivo studies.34,35

CONCLUSIONS

In all articles analyzed, the authors described thesuccess rates of miniscrews as sufficient for orthodontictreatment. This review further showed that screws of1.2-mm diameter and at least 8-mm length are prefera-ble, because they are stable and minimize the risk ofroot damage. The maxilla was shown to be better suitedfor miniscrews. As for the placement protocol, the datawere inconclusive for definite recommendations. Atcomparable success rates, the flapless method shouldbe chosen because it is less invasive and causes lesspatient discomfort.

Immediate or early loading of miniscrews is possi-ble, since longer healing periods did not provide addi-tional stability at forces of up to 200 cN.

Table VI. Success rate (%) by latency period (days), duration (months), and amount of loading (cN)

Authors Screws (n) Time before loading Duration of loading Force Success rate

Miyawaki et al3 20 \30 12 \200 85.0

29 30-90 82.8

72 $90 87.5

Cheng et al4 92 14-28 – 100-200 91.3

Liou et al5 32 14 9 400 100

Fritz et al6 36 \28 5.3 6 3.2 – 70.0

Park et al7 (2005) 30 14 12.3 6 5.7 \200 90.0

Motoyoshi et al8 (2006) 124 Immediate 6 \200 85.5

Park et al9 (2006) 227 – 15 \200 91.6

Chen et al10 (2006) 59 14 – 100-200 84.7

Tseng et al11 45 14 – 100-200 91.1

Herman et al12 49 1-149 – 150 61.2

Kuroda et al13 59 Immediate 12 – 89.8

14 #30 85.7

6 .30 83.3

5 – 50 80.0

55 100 89.1

8 150 75.0

11 200 100

Wiechmann et al14 133 Immediate 6 100-200 76.7

Motoyoshi et al15 (2007) 133 \30 6 200 82.0

36 90 6 200 97.2

Chen et al16 (2007) 98 7-28 – 100-300 85.7

79 35-84 88.6

40 $91 92.5

112 Crismani et al American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

REFERENCES


1997;31:763-7.

2. Costa A, Raffainl M, Melsen B. Miniscrews as orthodontic an-


Surg 1998;13:201-9.

3. Miyawaki S, Koyama I, Inoue M, Mishima K, Sugahara T, Ta-




4. Cheng S, Tseng I, Lee J, Kok S. A prospective study of the risk

factors associated with failure of mini-implants used for ortho-

dontic anchorage. Int J Maxillofac Implants 2004;19:100-6.

5. Liou EJ, Pai BC, Lin JC. Do miniscrews remain stationary under or-

thodontic forces? Am J Orthod Dentofacial Orthop 2004;126:42-7.

6. Fritz U, Ehmer A, Diedrich P. Clinical suitability of titanium mi-

croscrews for orthodontic anchorage—preliminary experiences.

J Orofac Orthop 2004;65:410-8.

7. Park HS, Lee SK, Kwon OW. Group distal movement of teeth us-

ing microscrew implant anchorage. Angle Orthod 2005;75:602-9.




9. Park HS, Jeong SH, Kwon OW. Factors affecting the clinical suc-

cess of screw implants used as orthodontic anchorage. Am J


10. Chen CH, Chang CS, Hsieh CH, Tseng YC, Shen YS, Huang IY,

et al. The use of microimplants in orthodontic anchorage. J Oral


11. Tseng YC, Hsieh CH, Chen CH, Shen YS, Huang IY, Chen CM.

The application of mini-implants for orthodontic anchorage. Int J

Oral Maxillofac Surg 2006;35:704-7.

12. Herman RJ, Currier GF, Miyake A. Mini-implant anchorage for

maxillary canine retraction: a pilot study. Am J Orthod Dentofa-

cial Orthop 2006;130:228-35.





14. Wiechmann D, Meyer U, Buchter A. Success rate of mini- and mi-

cro-implants used for orthodontic anchorage: a prospective clini-

cal study. Clin Oral Implants Res 2007;18:263-7.



36:695-9.

16. Chen YJ, Chang HH, Huang CY, Hung HC, Lai EH, Yao CC. A ret-

rospective analysis of the failure rate of three different orthodontic

skeletal anchorage systems. Clin Oral Implants Res 2007;18:768-75.

17. Keim RG. Answering the questions about miniscrews. J Clin

Orthod 2005;39:7-8.

18. Papadopoulos MA, Gkiaouris I. A critical evaluation of meta-

analyses in orthodontics. Am J Orthod Dentofacial Orthop

2007;131:589-99.

19. Cope JB. Temporary anchorage devices in orthodontics: a para-

digm shift. Semin Orthod 2005;11:3-9.

20. Ganeles J, Wismeijer D. Early and immediately restored and

loaded dental implants for single-tooth and partial-arch applica-

tions. Int J Oral Maxillofac Implants 2004;19:92-102.

21. Crismani AG, Bernhart T, Schwarz K, Celar AG,

Bantleon HP, Watzek G. Ninety percent success in palatal

implants loaded 1 week after placement: a clinical evaluation

by resonance frequency analysis. Clin Oral Implants Res

2006;17:445-50.

22. Haas R, Polak C, Furhauser R, Mailath-Pokorny G, Dortbudak O,

Watzek G. A long-term follow-up of 76 Branemark single-tooth

implants. Clin Oral Implants Res 2002;13:38-43.

23. Jemt T, Johansson J. Implant treatment in the edentulous max-

illae: a 15-year follow-up study on 76 consecutive patients

provided with fixed prostheses. Clin Implant Dent Relat Res

2006;8:61-9.

24. Herman R, Cope JB. Miniscrew-implants: IMTEC mini ortho im-

plants. Semin Orthod 2005;11:32-9.

25. Wu JC, Huang JN, Zhao SF, Xu XJ, Xie ZJ. Radiographic and sur-

gical template for placement of orthodontic microimplants in in-

terradicular areas: a technical note. Int J Oral Maxillofac Implants

2006;21:629-34.

26. Asscherickx K, Vannet BV, Wehrbein H, Sabzevar MM. Root re-

pair after injury from mini-screw. Clin Oral Implants Res 2005;

16:575-8.

27. Chen YH, Chang HH, Chen YJ, Lee D, Chiang HH, Yao CC. Root

contact during insertion of miniscrews for orthodontic anchorage

increases the failure rate: an animal study. Clin Oral Implants Res

2008;19:99-106.

28. Kuroda S, Yamada K, Deguchi T, Hashimoto T, Kyung HM, Ta-

kano-Yamamoto T. Root proximity is a major factor for screw

failure in orthodontic anchorage. Am J Orthod Dentofacial Orthop

2007;131(Suppl 4):S68-73.

29. Liou EJ, Chen PH, Wang YC, Lin JC. A computed tomographic

image study on the thickness of the infrazygomatic crest of the

maxilla and its clinical implications for miniscrew insertion.


30. Poggio PM, Incorvati C, Velo S, Carano A. ‘‘Safe zones’’: a guide

for miniscrew positioning in the maxillary and mandibular arch.


31. Branemark PI. Osseointegration and its experimental background.

J Prosthet Dent 1983;50:399-410.

32. Schenk RK, Buser D. Osseointegration: a reality. Periodontol

2000;1998(17):22-35.

33. Chung KR, Kim SH, Kook YA. The C-orthodontic micro-implant.

J Clin Orthod 2004;38:478-86.

34. Vande Vannet B, Sabzevar MM, Wehrbein H, Asscherickx K. Os-

seointegration of miniscrews: a histomorphometric evaluation.

Eur J Orthod 2007;29:437-42.

35. Favero LG, Pisoni A, Paganelli C. Removal torque of osseointe-

grated mini-implants: an in vivo evaluation. Eur J Orthod 2007;

29:443-8.


CLINICIAN’S CORNER

Clinical observations and success ratesof palatal implants

Karlien Asscherickx,a Bart Vande Vannet,b Peter Bottenberg,c Heiner Wehrbein,d and Mehran Moradi Sabzevare

Jette, Belgium, and Mainz, Germany

Introduction: Anchorage control is a challenge in orthodontics. Implants can be used to provide absoluteanchorage.The aim of this study was to evaluate the success rates of palatal implants used for various anchor-age purposes. Methods: Thirty-four palatal implants were placed in 33 patients. In the adults (n 5 9), theimplants (n 5 9) were placed in the median palatal suture. In the adolescents (n 5 24), the implants (n 5 25)were placed in the paramedian region. The implants were used to support a transpalatal arch, a modified distaljet appliance, or a modified hyrax screw. An implant was considered successful if it could be used as plannedthroughout the orthodontic treatment. The patients were asked to evaluate their pain perception after place-ment and explantation procedures. Results: Three implants failed early (during the waiting period beforeorthodontic loading, within 3 months after placement). During the orthodontic loading period, no implantswere lost. No statistically significant correlations were found between success rate and sex, age, primarystability, placement site (median or paramedian), implant size, or palatal depth. Pain perception after surgerywas acceptable. The success rate of the palatal implants in this study was 91%. Conclusions: Palatalimplants are a reliable method of providing absolute anchorage control in a variety of patients for differentindications. They can be loaded both directly and indirectly. (Am J Orthod Dentofacial Orthop2010;137:114-22)

Anchorage control is a challenging problem inorthodontics. Several solutions have been pro-posed and tested. Grouping several teeth as

anchorage has been suggested.1 Burstone and Kuhl-berg2 tried to improve anchorage control by makinguse of the fact that tooth tipping is easier to achievethan axial or root movement. Extraoral anchorage hasbeen suggested when anchorage in the dental arch isinsufficient. None of these solutions has proven to beabsolutely successful. For this reason, implant-likedevices have been introduced in orthodontics for abso-lute anchorage. They are referred to as orthodonticimplant anchors or temporary anchorage devices.3

A placement site commonly used to provide anchor-age with temporary anchorage devices is the palate,which has been studied by several investigators.4,5

The midpalatal suture area seems especially to be anideal placement site, since it has both thin soft tissueand thick cortical bone.4 This is interesting when a screwis to be placed, since the main objective of an orthodon-tic screw is to gain maximum retention (good qualityand quantity of bone) and prevent soft-tissue inflamma-tion. Miniscrews can be placed nearly everywhere in themouth and also in the midpalatal suture area.

One temporary anchorage device designed specifi-cally to be used in this region is the palatal orthodonticimplant anchor of the Orthosystem (Straumann, Basel,Switzerland). Although not as popular as miniscrews, pal-atal implants are gaining acceptance as an important alter-native for achieving maximum intraoral orthodonticanchorage.6 In adult patients, the median palatal suturezone is the area of choice for placement of palatal im-plants. In adolescents, however, the paramedian regionis preferred to avoid possible growth impairment of themaxilla in a transverse direction by placing an implantin the median palatal suture.7 The paramedian regionhas been described as a suitable placement site forimplants.5,8 The site for palatal implants is standardized(median or paramedian palate); this is a major advantage,since a standard surgical procedure that is simple and eas-ily controlled in all stages is essential for the success of animplant.9 Various rates of success for implants have been

a Lecturer, Dental Clinic, Department of Orthodontics, Vrije Universiteit

Brussel, Jette, Belgium.b Professor, Dental Clinic, Department of Orthodontics, Vrije Universiteit

Brussel, Jette, Belgium.c Professor, Dental Clinic, Department of Operative Dentistry, Vrije Universiteit

Brussel, Jette, Belgium.d Professor, Klinik fur Kieferorthopadie, Universitatsklinik Mainz, Mainz,

Germany.e Professor, Dental Clinic, Department of Periodontology, Vrije Universiteit

Brussel, Jette, Belgium.



Reprint requests to: Karlien Asscherickx, Vrije Universiteit Brussel, Dental

Clinic, Department of Orthodontics, Laarbeeklaan 103, 1090 Jette, Belgium;


Submitted, September 2007; revised and accepted, February 2008.

0889-5406/$36.00


doi:10.1016/j.ajodo.2008.02.022

114


reported. For dental implants, 5-year cumulative successrates of 90% to 95% were reported10; success rates variedfrom 70% to 90% for miniscrews11-15 and from 84.8% to100% for palatal implants.16-19

The use of palatal implants is mainly indicated foradult patients. However, more adolescents are reluctantto wear extraoral appliances, with the result that compli-ance is greatly reduced with these treatment options. Toavoid wearing extraoral appliances, children as youngas 10 years now receive palatal implants.

The aims of this study were to evaluate (1) the successrate of palatal implants used for orthodontic purposes; (2)whether the success rate of palatal implants can be corre-lated to sex, age, primary stability, placement site (me-dian or paramedian), form of the palate (wide or deep),implant size, and type of suprastructure (loading type);and (3) the patients’ pain after placement and removal.

The working hypothesis of this study was that a pal-atal orthodontic anchorage implant is a safe and predict-able therapeutic technique, applicable to many clinicalsituations.


The study group consisted of 32 patients, each with1 implant, and 1 patient with 2 implants. They were con-secutive patients, treated at the Dental Clinic of theDepartment of Orthodontics at Vrije Universiteit Brus-

sel in Jette, Belgium, between October 1998 and May2003, who had finished their orthodontic treatment.

Table I shows the composition of the group by sexand age. A division was made between adolescent andadult patients. The adolescent group comprised 24patients, all under the age of 16 years. The adult groupcomprised 9 patients, all above the age of 20 years.

Most patients had an Angle Class II malocclusion.They were treated by extraction of the maxillary firstor second premolars or distalization of the maxillaryposterior segments. In the extraction patients, theimplants were loaded indirectly and provided anchoragereinforcement for the posterior teeth (n 5 18, Fig 1).When the aim of treatment was distalization of the max-illary posterior teeth, the implants were loaded directlyand used to anchor a modified distal jet appliance formaxillary molar distalization (n 5 13, Fig 2). In a patientwith multiple agenesis, the implant was used to anchora cantilever for aligning a palatally positioned canine. In1 patient, 2 palatal implants were used to anchor a mod-ified hyrax screw for skeletal palatal expansion (n 5 2,Fig 3).

All patients, or their parents, signed a consent formto take part in the study. This study was approved by theEthical Committee of Vrije Universiteit Brussel.

The palatal implant used was the Ortho-implant(Straumann). This is an endosseous orthodontic implantanchor system for palatal or retromolar anchorage. The

Table I. Composition of study group

Adolescent girls Adolescent boys Women Men

Patients (n) 11 13 8 1

Mean age 12 y 2 mo 13 y 7 mo 35 y 8 mo 47 y 5 mo

Age range 10 y 3 mo-14 y 6 mo 11 y 5 mo-15 y 6 mo 21 y 5 mo-53 y 2 mo 47 y 5 mo

Fig 1. A, Implant-anchored transpalatal arch to reinforce the maxillary first molars to distalize the an-terior segment after extraction of the maxillary second premolars; B, transformation of the transpa-latal arch, once a Class I occlusion is achieved in the canine region, to reinforce the anchoragepotential of the maxillary first premolars to mesialize the molars to establish a Class I molar relation-ship; C, extraction sites of the maxillary second premolars almost completely closed.

American Journal of Orthodontics and Dentofacial Orthopedics Asscherickx et al 115Volume 137, Number 1

fixture is designed for 1-stage application. The endo-sseous part of the implant is cylindrical and made ofpure titanium. It has a diameter of 3.3 or 4.0 mm anda length of 4.0 or 6.0 mm. The implant has a sandblastedand acid-etched surface. Above the polished transmu-cosal neck is an abutment on to which the desired supra-structure is soldered or laser-welded.

The implant was placed under local anesthesia. Theplacement procedure is easy and fast when performedby an experienced surgeon. First, the palate was anes-thetized. Then, a trephine bur was used to make a punchof the palatal mucosa. With a curette, the palatal mucosawas removed. By using a pilot drill and a standardizedprofile drill, the implant bed was prepared. The implantwas hand turned as far as possible, and, if necessary,a ratchet was used to tighte it into its final position. Afterplacement, some implants had a healing abutment; inthe others, only a screw was placed in the implant.

Implant selection for each patient was based on thevertical height of the anterior palate as determined onthe lateral cephalogram (Fig 4). All implants wereplaced by the same surgeon (M.M.S). In the adolescentpatients (\16 years, n 5 24), the implants were placed

in the paramedian region (first or second quadrant) toavoid possible growth impairment at the palatal suture7

(Fig 5). In the adults (.20 years, n 5 9), the implantswere placed in the median palatal suture (Fig 6).When primary stability was not achieved, the surgeondecided whether to place another (wider or longer)implant immediately or whether to leave the nonstableimplant in place.

After surgery, patients were instructed to usechlorhexidine digluconate 0.2% mouth rinse twicea day during the first 8 weeks after placement. After 8weeks, the patients were instructed to brush the implantwith an Interspace brush (Oral-B, P&G, Cincinnati,Ohio). In some cases, the patients were instructed towear a covering plate to prevent tongue pressure onthe implant (Fig 7). Whether or not to use such a platewas always evaluated before placement of the implantin consultation with the patient and the parents. A heal-ing period of at least 12 weeks was allowed beforeimpressions were taken to manufacture the appropriatesuprastructure.

Since no findings on ‘‘sleeping orthodonticimplants’’ (implants that are no longer used for

Fig 2. A, Implant-anchored distal jet to distalize the maxillary molars to create space for buccallyimpacted canines; B, end of distalization, with the maxillary molars passively stabilized in position,ready to start retraction of the premolars and align the maxillary arch.

Fig 3. A, Implant-anchored hyrax screw to skeletally expand the maxilla; B, fixation of the screw afterexpansion; C, occlusal radiograph with widening of the median palatal suture after expansion clearlyvisible.

116 Asscherickx et al American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

orthodontic or prosthetic purposes) have been publishedto date, the implants were retrieved after use. This pro-cedure was performed with a special drill (cylindricaland open), with which the implant and the surroundingbone were removed.

An implant was assigned to 1 of 3 categories definedas follows.

1. Success: an implant was considered successful if itcould be used for anchorage purposes throughoutthe orthodontic treatment as planned.

2. Early failure: an implant was classified as an earlyfailure if it loosened during the healing phase(before 12 weeks).

3. Late failure: an implant was classified as a late fail-ure if it loosened during the orthodontic loadingperiod.

Correlations were evaluated between failure rateand sex (male or female), age (adult or adolescent),primary stability (yes or no), placement site (medianor paramedian), form of the palate (wide or deep), andimplant size (4 or 6 mm length).

The form of the palate was categorized as eitherwide or deep. On the pretreatment study casts, the widthand the depth of the palate were measured with a digitalcaliper to the closest 0.1 mm, at the level of the secondpremolars. Width was measured between the palato-gingival borders of both maxillary second premolars.

Depth was measured between the palate and the occlu-sal plane at the level of the second premolars. When theratio of width to depth was less than 1.4, the palatal formwas classified as deep. When the ratio was more than1.4, the palatal form was classified as wide.

Patients were asked about the amount of painkillerstaken after surgery and during the following days. Adifferentiation was made between no painkillers, 1 pain-killer immediately after surgery, 1 painkiller immediatelyafter surgery and 1 in the evening, and more than 2 pain-killers in total.

The patients were instructed to rate their perceptionof pain 1 day, 1 week, and 1 month after surgery ona visual analog scale (VAS) from 0 (no pain at all) to10 (extremely painful).

At placement, length and diameter of the implant(endosseous part), duration of placement time, andspecial remarks, such as presence or lack of primarystability, were noted.

Satistical analysis

To examine correlations between the success rate andrespective classification of each variable, the Fisher exactprobability test was used. A probability of P \0.05 wasconsidered significant. These analyses were carried outwith statistical analysis software (version 12.0, SPSS,Chicago, Ill).

RESULTS

A total of 34 implants were placed in 33 patients.The overall success rate was 91%. In the adult group(median placement), the success rate was 88.8%. Inthe adolescents (paramedian placement), the successrate was 92%. Three implants failed early; they werelost during the healing phase (within 12 weeks afterplacement). No late failures (during loading phase)occurred. Eight implants did not have primary stability,and the surgeon (M.M.S.) decided to replace 2 of themwith longer ones (length, 6 mm; diameter, 3.3 mm) and2 with wider ones (length, 4 mm; diameter, 4 mm). The4 other implants without primary stability were left inplace. Three became stable after a week, and 1 waslost after 3 weeks. No implants were lost during the or-thodontic loading phase. Both direct (hyrax screw anddistal jet) and indirect (transpalatal arch) loading weresuccessful.

Table II shows the results of the correlation analysis.No statistically significant correlations were found be-tween failure and sex (P 5 0.409), age (P 5 0.616), pri-mary stability (P 5 0.389), placement site (P 5 0.616),palatal form (P 5 0.616), or implant size (P 5 0.662).

Fig 4. Lateral headfilm with palatal implant in situ.


In 2 subjects who had early failure, a specific reasonwas identified. One patient admitted to have played withthe tongue on the implant. Another patient startedbrushing the implant after 1 month. New implantswere placed in these patients; all were successfullyused but not included in this study.

The amounts of painkillers taken are shown in TableIII. Most patients took 1 painkiller immediately afterplacement. No patient needed to take more than 2 pain-killers. After removal, the amounts of painkillers weresimilar to those after placement. The amounts of pain-killers were similar for the adolescent and adult patients.In 1 adult, an abscess had developed at the removal site.After antibiotic therapy and curettage, wound healingwas uneventful. This patient needed to take painkillersfor a week.

The scores for pain on the VAS are shown in TableIV. The results were similar for both placement andremoval. One day after surgery, averages of 1.61 and1.48, respectively, were scored. The maximum score1 day after placement was 6.5. One month after place-ment, no patient felt pain. The maximum score 1 dayafter removal was 3.9. One month after removal, nopatient felt pain.

The implant sizes according to sex and age areshown in Table V. Although the 6-mm implantswere used more often in male patients (7 of 15)than in female patients (3 of 19), this differencewas not statistically significant (P 5 0.057). No rela-tionship was found between age and implant size(P 5 0.165).

Fig 5. Palatal implant placed in an adolescent patient in the paramedian region in the maxillary leftquadrant: A, clinical view; B, radiologic view.

Fig 6. Palatal implant placed in an adult patient in the median palatal suture zone: A, clinical view;B, radiologic view.

Fig 7. Covering plate can be used to cover the palatalimplant with the healing abutment after placement andprevent tongue pressure on the implant.


January 2010

Duration of the placement procedure was on aver-age 16 minutes (range, 12-23 minutes). The bone atthe placement site was spongy in 4 patients (3 parame-dian, 1 in the suture). In 3 of these patients, however,primary stability was obtained. In the fourth patient,the implant was removed after 3 weeks because it stillhad mobility.

In all patients, the original or the replaced implantwas used as planned for orthodontic anchorage purposesduring orthodontic treatment. Length of treatment wason average 1 year 10 months (6 7 months).

Wound healing after removal of the implants wasuneventful in all patients, except 1, who developed anabscess. Three months after implant removal, it washard to tell where the implant was (Fig 8).

DISCUSSION

In this study, we had a success rate of 91% forpalatal implants. This was similar to the results in otherstudies. Crismani et al16 reported a 90% success rate.They placed 20 palatal implants in 20 patients. Theimplants were loaded 1 week later. Wehrbein et al17

investigated prospectively 9 palatal implants placed inthe median palatal suture in adults and used for anchor-age reinforcement of the posterior teeth. The mean un-loaded implant healing period was 12 6 3 weeks. Allimplants showed primary stability directly after place-ment and were successfully used for anchorage pur-poses throughout the examination period, fora success rate of 100%. Bernhart et al18 placed 21 short

epithetic implants in 21 patients in the paramedianregion and found a time-related survival probability of84.8% after 22.9 months. Adequate primary stabilitywas achieved for all implants. Three implants werelost after the start of orthodontic loading (delayed for4 months after implant placement). This contrastswith our study, in which no implants were lost afterorthodontic loading began. Bantleon et al19 reporteda 92% success rate. Three of 40 implants were lostwithin 2 to 3 months after placement.

In our study, the failure or success of a palatalimplant could not be correlated to age or sex. It appearsthat palatal implants can be used in many patients with-out restrictions of age or sex.

We had 24 adolescents in this study; this was highcompared with the number of adults (n 5 9). Becauseso many were young, the surgeon often had to dealwith highly stressed patients. He noticed that whenthey were younger than 12 years (n 5 8), half of themwere highly stressed during placement. This high levelof anxiety resulted in more movement of the patient dur-ing the procedure and frequent closing of the mouth. Theuse of palatal implants in young patients is possible, butthe procedure should be carefully explained to them, sothat they will be relaxed before it starts. One implantwithout primary stability after placement was lost after3 weeks, and the patient (age, 12 years) admitted to play-ing with the tongue on the implant. The use of a coveringplate (Fig 7) might be indicated for young patients to pre-vent tongue pressure on the implant. These plates can bemade before placement of the implant, because the place-ment site (first or second quadrant in adolescents) is nor-mally decided at treatment planning. The use of theseplates should be discussed before placement of the im-plant with the patients and the parents, if appropriate.

No correlation was found between lack of primary sta-bility and age, sex, implant size, placement site, or form ofthe palate. When the surgeon noticed spongy or soft bone,this was not related to lack of primary stability. These find-ings suggest that primary stability is predominantly deter-mined by the experience and skill of the surgeon.

The ideal placement site for palatal implants can bediscussed. According to the manufacturer’s guidelines,

Table II. Success rates and number of implants accord-ing to clinical variables

Clinicalvariable

Successrate

Successful/total implants

Significance(Fisher exact test)*

Sex

Male 86.6% 13/15 0.409

Female 94.7% 18/19

Age group

Adult 88.8% 8/9 0.616

Adolescent 92% 23/25

Primary stability

Yes 93% 27/29 0.389

No 80% 4/5

Placement site

Median 88.8% 8/9 0.616

Paramedian 92% 23/25

Form of palate

Wide 92% 23/25 0.616

Deep 88.8% 8/9

Implant size

4.0 mm length 91.6% 22/24 0.662

6.0 mm length 90% 9/10

*P \0.05 is a significant difference.

Table III. Amount of painkillers taken

Adolescent Adult

Placement Removal Placement Removal

None 4 3 0 1

1 15 15 7 5

2 5 6 2 2

.2 0 0 0 1


palatal implants were originally designed to be placed inthe median palatal suture, since, in the broad median su-ture zone, the bone is relatively dense. In the adults inour study, the implants were placed in the median pala-tal suture, whereas a paramedian site was chosen in theadolescents to avoid possible growth impairment at thepalatal suture.7 No significant correlation was foundbetween placement site (median or paramedian) andfailure rate. Since the paramedian region was alreadydescribed as a good alternative for implant placement,this is suggested as the region of choice for palatalimplants in adolescent patients.5,8,18

The size of the implant used for each patient wasdetermined from the lateral cephalograms. It was dem-onstrated by Wehrbein et al20 that vertical bone supportin the midsagittal area of the palate is at least 2 mmhigher than is apparent on the lateral cephalogram.When primary stability is not achieved, the implantcan be replaced by a longer one (length, 6.0 mm)when sufficient bone height is apparent or by a widerone (diameter, 4.0 mm). The benefit of longer(6.0-mm length) implants can be questioned, sinceGedrange et al21 concluded from a study on human ca-davers that the quality of placement and bone structureis more important than the length of the orthodontic im-plant for implant stability. When there is doubt aboutavailable bone height, shorter implants can be used(4.0 mm). Patients have great variations in verticalbone volume.5 Several authors recommend the use ofa stent22-24 or a preoperative diagnostic evaluationwith dental computed tomography or computerizednavigation surgery for the safe placement of palatalimplants.6,25 In this study, no adverse effects of theimplants on surrounding anatomic structures wereobserved. Therefore, we assume that dental computedtomography (which inevitably leads to higher costs

and extra radiation to patients) should be limited topatients in whom there is doubt about sufficient boneheight for placement of 4.0-mm implants. This agreeswith Cousley,26 who questioned the justification fora computed tomography scan for surgical planning. Inthis study, the failure rate was higher in patients witha deep palate. However, this difference in failure ratesbetween deep and wide palates was not statisticallysignificant. This might be due to the few patients witha deep palate in this study. Placement of implants inpatients with a deep palate is possible, but, for them,the use of a stent to guide the pilot drill might be useful.

In our study, the patients’ perception of the implantwas evaluated only in terms of pain. The need to takepainkillers after placement was limited to a maximumof 2 painkillers, and, after 1 week, no patient had anypain. The same results were found after removal, exceptfor 1 patient, who developed an abscess and needed totake painkillers for a week. The scores on the VASwere comparable to those found by Feldman et al.27

They evaluated by scores on a VAS the perceived painintensity between the placement of an Orthosystempalatal implant and premolar extraction in adolescentpatients. The first evening after the intervention,patients who had undergone premolar extraction hadsignificantly more pain than those who had receiveda palatal implant. One week after the interventions,

Table IV. Scores on VAS for pain

1 dayafter placement

1 weekafter placement

1 monthafter placement

1 dayafter removal

1 weekafter removal

1 monthafter removal

Average 1.61 0.21 0 1.48 0.64 0

SD 1.23 0.04 0 1.05 0.24 0

Minimum 0 0 0 0 0 0

Maximum 6.5 1.2 0 3.9 1.7 0

Table V. Implant sizes

Length(mm)

Diameter(mm)

Adolescentgirls (n)

Adolescentboys (n)

Women(n)

Men(n)

4.0 3.3 8 6 7 1

6.0 3.3 2 7 1 0

4.0 4.0 1 1 0 0

Fig 8. Palatal mucosa 3 months after implant removal.


January 2010

pain was still significantly higher in patients withextractions, compared with the patients who hadreceived an implant. Gunduz et al28 evaluated the accep-tance rate of palatal implants in a questionnaire study;85 patients, whose orthodontic treatment included a pal-atal implant, answered the questionnaire. The resultsshowed that most patients became used to the implantin about 2 weeks, and 86% of the patients would recom-mend the treatment to others. Since the acceptance rateof palatal implants is high and the pain after placementand removal is acceptable, this minor surgery shouldencourage clinicians to include the Orthosystem palatalimplant in orthodontic treatment plans.

The palatal implants in this study were successfullyused for various anchorage purposes. They could beloaded indirectly (implant-anchored transpalatal arch)or directly (implant-anchored modified distal jet appli-ance or implant-anchored modified hyrax screw).They could successfully be used for both orthodonticand orthopedic anchorage purposes.

CONCLUSIONS

1. In this study, success rate of the palatal implantswas 91%.

2. Success of the implants was independent of age,sex, primary stability, placement site, palatalform, implant size, and type of suprastructure.

3. This success rate was comparable for palatalimplants placed in the median palatal suture zoneand the paramedian region.

Palatal implants are a reliable device for achievingmaximum anchorage control in orthodontic treatment,even in younger patients. They can be successfullyloaded both directly and indirectly.

REFERENCES

1. Quinn RS, Yoshikawa DK. A reassessment of force magnitude in

orthodontics. Am J Orthod 1985;88:252-60.

2. Burstone CJ, Kuhlberg AJ. T-loop position and anchorage control.


3. Mah J, Bergstrand F. Temporary anchorage devices: a status

report. J Clin Orthod 2005;39:132-6.




5. Bernhart T, Vollgruber A, Gahleitner A, Dortbudak O, Haas R.

Alternative to the median region of the palate for placement of

an orthodontic implant. Clin Oral Implants Res 2000;11:595-601.

6. Wexler A, Tzadok S, Casap N. Computerized navigation surgery

for the safe placement of palatal implants. Am J Orthod Dentofa-

cial Orthop 2007;131(Suppl):S100-5.

7. Asscherickx K, Hanssens JL, Wehrbein H, Sabzevar MM. Ortho-

dontic anchorage implants inserted in the median palatal suture

and normal transverse maxillary growth in growing dogs: a bio-

metric and radiographic study. Angle Orthod 2005;75:826-31.

8. Gahleitner A, Podesser B, Schick S, Watzek G, Imhof H. Dental

CT and orthodontic implants: imaging technique and assessment

of available bone volume in the hard palate. Eur J Radiol 2004;

51:257-62.

9. Favero L, Brollo P, Bressan E. Orthodontic anchorage with spe-

cific fixtures: related study analysis. Am J Orthod Dentofacial

Orthop 2002;122:84-94.

10. Wennstrom JL, Palmer RM. Consensus report on clinical trial

session. Implant dentistry. Proceedings of the 3rd European

Workshop on Periodontology; 1998 May 1-3; Geneva, Switzer-

land. Berlin, Germany: Quintessence; 1999. p. 256.

11. Miyawaki S, Koyama I, Inoue M, Mishima K, Suahara T, Takano-

Yamamoto T. Factors associated with the stability of titanium

screws placed in the posterior region for orthodontic anchorage.





13. Motoyoshi M, Hirabyashi M, Uemura M, Shimizu N. Recommen-

ded placement torque when tightening an orthodontic mini-


14. Wiechmann D, Meyer U, Buchter A. Success rate of mini- and mi-

cro-implants used for orthodontic anchorage: a prospective clini-

cal study. Clin Oral Implants Res 2007;18:263-7.





16. Crismani AG, Bernhart T, Schwarz K, Celar AG, Bantleon HP,

Watzek G. Ninety percent success in palatal implants loaded 1

week after placement: a clinical evaluation by resonance fre-

quency analysis. Clin Oral Implants Res 2006;17:445-50.

17. Wehrbein H, Feifel H, Diedrich P. Palatal implant anchorage rein-

forcement of posterior teeth: a prospective study. Am J Orthod


18. Bernhart T, Freudenthaler J, Dortbudak O, Bantleon HP,

Watzek G. Short epithetic implants for orthodontic anchorage in

the paramedian region of the palate. A clinical study. Clin Oral

Implants Res 2001;12:624-31.

19. Bantleon HP, Bernhart T, Crismani A, Zachrisson B. Stable ortho-

dontic anchorage with palatal osseointegrated implants. World J

Orthod 2002;3:109-16.

20. Wehrbein H, Merz BR, Diedrich P. Palatal bone support for ortho-

dontic implant anchorage—a clinical and radiological study. Eur J

Orthod 1999;21:65-70.

21. Gedrange T, Hietschold V, Mai R, Wolf P, Nicklisch M, Harzer W.

An evaluation of resonance frequency analysis for the determina-

tion of the primary stability of orthodontic palatal implants. A

study in human cadavers. Clin Oral Implants Res 2005;16:425-31.

22. Martin W, Hefferman M, Ruskin J. Template fabrication for a mid-

palatal orthodontic implant: a technical note. Int J Oral Maxillofac

Implants 2002;17:720-2.

23. Tosun T, Keles A, Erverdi N. Method for the placement of palatal

implants. Int J Oral Maxillofac Implants 2002;17:95-100.

24. Majumdar A, Tinsley D, O’Dwyer J, Doyle PT, Sandler J,

Benson P, et al. The ‘‘Chesterfield stent’’: an aid to the placement

of midpalatal implants. Br J Oral Maxillofac Surg 2005;43:36-9.

25. Kim SH, Choi YS, Hwang EH, Chung KR, Kook YA, Nelson G.

Surgical positioning of orthodontic mini-implants with guides fab-

ricated on models replicated with cone-beam computed tomogra-

phy. Am J Orthod Dentofacial Orthop 2007;131(Suppl):S82-9.


26. Cousley R. Critical aspects in the use of orthodontic pala-

tal implants. Am J Orthod Dentofacial Orthop 2005;127:

723-9.

27. Feldmann I, List T, Feldmann H, Bondemark L. Pain intensity and

discomfort following surgical placement of orthodontic anchoring

units and premolar extraction. A randomized controlled trial.

Angle Orthod 2007;77:578-85.

28. Gunduz E, Schneider-Del Savio TT, Kucher G, Schneider B,

Bantleon HP. Acceptance rate of palatal implants: a questionnaire

study. Am J Orthod Dentofacial Orthop 2004;126:623-6.


January 2010

CASE REPORT

Retreatment of a patient with Marfan syndromeand severe root resorption

John E. Bilodeau

Springfield, Va

This case report describes the retreatment of a patient with Marfan syndrome whose earlier orthodontic andsurgical treatment had been unsuccessful. Marfan syndrome is an inherited connective tissue disorder trans-mitted as an autosomal dominant trait. The disorder results from molecular defects in the fibrillin gene that areresponsible for the impaired structural integrity of the skeletal, ocular, and cardiovascular systems. When shesought retreatment, the patient had an open bite, mandibular anterior crowding, severe root resorption, andtemporomandibular joint derangement with some resorption of the condyles. The second treatment, which in-cluded extractions and surgery, resulted in balanced and harmonious facial proportions, and a Class I occlu-sion with normal overjet and overbite. There was no further loss of condylar tissue, and the temporomandibularjoints were asymptomatic. More root resorption on the mandibular left canine and the left second premolarwas evident after the second treatment. (Am J Orthod Dentofacial Orthop 2010;137:123-34)

Ayoung woman needed orthodontic retreatment

for an open bite, crowded mandibular incisors,and temporomandibular joint (TMJ) derange-

ment with some flattening (resorption) of the condyles.She had a history of previous orthodontic and orthog-nathic surgical treatment. She has Marfan syndromeand extensive root resorption. Did the earlier treatmentcause the flattening of the condyles and root resorptionor was there a genetic predisposition, or both? Why didthe first treatment fail? Was retreatment worth the risk?

HISTORY AND ETIOLOGY

The patient was a white woman, aged 28 years 5months, with a history of orthodontic and orthognathicsurgical treatment that began at age 13 and lasted for5 years, culminating with orthognathic surgery at age18. Her medical history confirmed that she had Marfansyndrome, a genetic disorder. She had arachnodactylywith positive wrist (Walker) and thumb (Steinberg)signs (Fig 1). She was taking a beta-adrenergic blockerto control blood pressure in hopes of preventing aorticdissection, because she had evidence of aortic dilatationand mitral valve prolapse. This disorder weakens theconnective tissue of the aorta as it enters the heart.She had dural ectasis, hypermobility of her joints, oste-

oarthritis of her knees, and scoliosis. She had a Class IImalocclusion complicated by a 5-mm open bite, 6 mmof mandibular anterior crowding, and severe root re-sorption. She had a long lower anterior facial height.Her chief concerns were her ‘‘crooked teeth, openbite, and facial appearance.’’ The discomfort she wasexperiencing in her TMJs had been somewhat relievedwith splint therapy.

DIAGNOSIS

Facial photographs showed malar hypoplasia, retro-gnathia, down-slanting palpebral fissures and a longlower anterior facial height. She was unable to closeher lips without mentalis strain (Fig 2).

Intraoral photographs and dental casts (Fig 3)showed missing maxillary first premolars and a 5-mmopen bite from the second premolars anteriorly to thecentral incisors. The molars were in an Angle Class IIrelationship, and the canines were in a Class III relation-ship. The maxillary arch was constricted with a high-arched palate. The mandibular dental midline was2 mm left of the facial midline, and there was 6 mmof mandibular anterior crowding.

The panoramic radiograph (Fig 4) showed that allthird molars were missing as were the maxillary firstpremolars. There was extensive root resorption. Mostteeth showed pulpal obliteration. The condyles wereworn (resorbed) and flattened. Surprisingly, there wasminimal mobility of the teeth.

The cephalometric head film and tracing (Fig 5)showed an ANB angle of 3�. The SNA angle of 72� re-flected a retropositioned maxilla, and the SNB angle of69� confirmed mandibular deficiency. The FMA was

Private practice, Springfield, Va.



Reprint requests to: John E. Bilodeau, 6116 Rolling Rd, Suite 201, Springfield,

VA 22152; e-mail, [email protected].

Submitted, February 2007; revised and accepted, May 2007.

0889-5406/$36.00


doi:10.1016/j.ajodo.2007.05.029

123


51�. The facial height index of Horn1 (posterior facialheight to anterior facial height) was .50, and it confirmeda skeletal open bite. The IMPA, a reflection of the rela-tionship of the mandibular incisor to the mandible, was96�. Because the FMA was high, 51�, the IMPA of 96�

confirmed a protrusive mandibular incisor position.

The Z-angle was 68�.2 The Wits measurement of0 mm was normal.3,4 The symphyseal area was gro-tesquely misshapen from the previous genioplasty. Theprevious surgical fixation devices were also evident.

When orthodontic retreatment is needed, it is impor-tant to review the prior diagnosis and treatment plan todetermine, if possible, why the treatment failed. Herprior treatment records were unavailable. However, af-ter studying her cephalometric and panoramic radio-graphs, it was apparent that orthodontic treatmentwith maxillary first premolar extractions with a maxil-lary surgical procedure and a genioplasty had beenperformed. It is reasonable to speculate that the primaryreason the treatment failed was that mandibular extrac-tions were not part of the initial treatment, and that thesurgical manipulations of the maxilla and the chin wereless than satisfactory.

TREATMENT OBJECTIVES

The following treatment objectives were deter-mined: (1) correct the open bite, (2) obtain a normal pro-file line to nose relationship and a normal Z-angle,3 (3)obtain normal canine and incisal guidance, (4) resolvethe crowding in both dental arches, (5) reduce the exces-sive lower anterior facial height, (6) reduce mentalisstrain, (7) eliminate TMJ dysfunction and discomfort, and(8) guard against further root and condylar resorption.

TREATMENT ALTERNATIVES

No treatment had to be considered an option,because of the Marfan syndrome, the failure of the firsttreatment, the amount of root resorption present, and thecondition of the condyles. Further root and condylarresorption were certainly possible.

The other alternative was extraction of the mandib-ular first premolars followed by an orthognathic surgicalprocedure to correct the vertical skeletal imbalance anda redo of the genioplasty to gain more of an esthetic pro-jection of her chin. This option would make it possibleto upright the mandibular incisors, reduce vertical facialheight, provide an esthetic change, and correct thedental malocclusion.

TREATMENT PLAN

After carefully considering the alternatives, theextraction and surgical option was chosen. The patientunderstood the risks of retreatment and was counseledthat teeth could be lost, and implant placement or splint-ing of her teeth would be necessary because of the com-plication of further root resorption. She was alsoinformed about the condition of her condyles and thatfurther loss of condylar tissue was possible. Despite

Fig 1. Signs of Marfan syndrome: A, arachnodactyly(spider fingers); B, positive Walker sign, with the distalphalanges of the first and fifth digits of 1 hand overlap-ping when placed around the opposite wrist; C, positiveSteinberg sign, with the thumb extending beyond the ul-nar border when completely opposed in the clenchedhand.

124 Bilodeau American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

the risks, she wanted to proceed; in fact, she was enthu-siastic about retreatment. She said, ‘‘anything would bebetter than what I have now, and I know I might losesome teeth.’’

When surgical intervention is part of the treatment,several analyses are used as guidelines to position thejaw bones both vertically and horizontally to providea pleasing and harmonious face.

Fig 2. Pretreatment facial and intraoral photographs.

Fig 3. Pretreatment dental casts.

American Journal of Orthodontics and Dentofacial Orthopedics Bilodeau 125Volume 137, Number 1

McNamara’s nasion Frankfort perpendicular wasused as a guideline to determine the placement of themaxilla and the maxillary incisors in the horizontalplane.5 The maxilla should be positioned so that PointA closely approximates this ‘‘line,’’ and the maxillaryincisor is 5 mm 6 2 mm anterior to nasion Frankfortperpendicular. The maxillary incisor should be placedabout 110� to the palatal plane.

The skeletal position of the mandible can bechecked with the analysis of Delaire et al,6 which usesa line from the frontal and maxillary bone intersectionconnected to the posterior clinoid process and fromthis point to menton. This angular measurement shouldbe 85� to 90�. Also, with a line drawn perpendicular tothis line, a vertical assessment can be made by measur-ing the upper facial height from nasion to anterior nasalspine (ANS) and the lower facial height from ANS tomenton. The distance from ANS to nasion should be45% of the total facial height, and the distance fromANS to menton along this line should be 55% of the to-tal facial height. To determine the ideal facial height, .45can be divided into the nasion-ANS distance. This is thetotal ‘‘hard-tissue’’ height that is ideal for a patient. Thismeasurement gives guidelines about whether to open orclose a patient vertically.

In any surgical treatment plan, a soft-tissue evalua-tion is necessary. Variations in the soft tissues that coverthe face can produce misleading conclusions if diagno-sis and treatment planning are based on skeletal mea-surements alone. By using the soft-tissue analysis ofLegan and Burstone,7 the clinician can evaluate the hor-izontal and vertical soft tissues of the mandible. Thisanalysis is used as follows: (1) the SN line is recon-structed 7� upward from its original position, and (2)a perpendicular is drawn from soft-tissue glabella tothis line. Soft-tissue pogonion should closely approxi-mate this line. Vertical soft-tissue proportions can be

checked by drawing a line perpendicular from glabellato soft-tissue nasal point and from subnasale to menton.The ratio of this distance should be 1:1. Another helpfulsoft-tissue evaluation is Merrifield’s Z-angle2 and theinterrelationship of the profile line to the middle of thenose. The profile line should intersect the nose at the an-terior aspect of the nares, and the Z-angle, when mea-sured to the Frankfort horizontal, should be between72� and 78�.

Merrifield’s total space analysis was used to deter-mine space requirements.8,9 The McNamara analysis5

confirmed that Point A was posterior to the nasionFrankfort perpendicular, and the maxilla would needto be surgically moved forward.

A decision was made to extract the mandibular firstpremolars. This extraction pattern would resolve the

Fig 4. Pretreatment panoramic radiograph showing rootresorption and flattening of the condyles.

Fig 5. Pretreatment cephalometric and headfilm tracing.


January 2010

dental crowding and allow the mandibular incisors to beuprighted to increase overjet to gain maximumadvancement of the mandible.

Both the soft-tissue analysis of Legan and Burstone7

and the analysis of Delaire et al6 showed that verticalheight needed to be reduced. Because of the poor condi-tion of the condyles, further study and a search of theliterature was done.

Arnett et al10 found that medial or lateral torque ofthe mandibular condyle associated with sagittal osteot-omy resulting in medial or lateral condylar compres-sion creates the possibility for late (9-18 months)condylar resorption and Point B relapse. It was rea-soned that any surgical manipulation of the proximal(condyle) segment would have an unpredictable out-come and that a sagittal split should be avoided.Therefore, after all space closure, only a LeFort Iosteotomy and a genioplasty were to be performed toreduce vertical facial height by autorotation of themandible. This type of surgical intervention wouldbe noninvasive to the mandibular condyles and wouldprovide a pleasing profile line to nose relationship anda favorable Z-angle. The genioplasty would also haveto be redone.

TREATMENT PROGRESS

Because of mitral valve prolapse, and on the adviceof her physician, this patient was premedicated with 2 gof amoxicillin to prevent bacterial endocarditis beforeall appointments. The mandibular teeth only werebanded or bonded sequentially with the 10-2 systemof Merrifield.9,11 A .022-in standard nontorqued, non-angulated edgewise appliance was used. The maxillary

Fig 6. Presurgical dental casts.

Fig 7. Presurgical panoramic radiograph.


teeth would remain without appliances until the man-

dibular incisors were uprighted and the mandibular

arch stabilized. It was reasoned that this approach

would protect the maxillary incisors from further root

resorption until they absolutely had to be aligned and

leveled. The patient was instructed to wear a high-

pull J-hook headgear directly against the mandibular

canine brackets to retract these teeth into the first pre-

molar extraction sites. After canine retraction, the man-

dibular anterior teeth were carefully and slowly

retracted with a .020 3 .025-in closing loop archwire.

The J-hook headgear was worn against the canine

brackets to support anterior retraction. After space clo-

sure in the mandibular arch, a .0213 .025-in stabilizing

archwire was placed.

At this juncture, the maxillary teeth were bandedand bonded. Reproximation of the maxillary anteriorteeth was necessary to create enough space to resolvethe maxillary crowding. Impressions were taken atevery appointment, and the dental casts were hand-articulated to assess the postoperative occlusion. Preop-erative records were taken to plan the orthognathicsurgical procedure (Figs 6-8).

PREOPERATIVE DIAGNOSIS

The presurgical cephalometric tracing (Fig 8)showed that the FMA remained at 51�. The facial heightindex of Horn1 remained the same at .50. The IMPAof 71� confirmed that the mandibular incisors had beenuprighted over basal bone. The Z-angle remained at 68�.

Point A and the maxillary incisors were 13 mmposterior to nasion Frankfort perpendicular.

The analysis of Delaire et al6 showed that themandible could come forward because the posteriorclinoid-FMA-menton angle was 80� and could bepositioned between 85� and 90�. The vertical analysisof Delaire et al showed an upper facial height of 57mm; therefore, the lower facial height should be 69mm. It was actually 88 mm or 19 mm more than whatit should be for a harmonious facial balance. Reducingthe vertical dimension with a LeFort maxillary impactionwould produce a large autorotation of the mandible (per-haps as much as 10-12 mm); this would cause forward po-sitioning of pogonion. Because mandibular surgery wasto be avoided, a large advancement of the maxilla wouldbe necessary to maintain a Class 1 molar relationship.

Fig 8. Presurgical cephalometric headfilm and tracing.

Fig 9. Computerized visual treatment objectives.


January 2010

Fig 10. Posttreatment facial and intraoral photographs.

Fig 11. Posttreatment dental casts.


The Legan-Burstone soft-tissue analysis confirmedthat the mandible needed to come forward, and verticalfacial height needed to be reduced 15 mm to achievea 1:1 ratio and a well-balanced facial profile. Dependingon the amount of autorotation achieved, the genioplastywould be redone to further project pogonion anteriorlyto more closely satisfy the soft-tissue projection of theanalysis, reposition the infrahyoid muscles, and achievea normal Z-angle.

A computerized visual treatment objective was cre-ated with the DFplus software (Dentofacial Planner,Toronto, Ontario, Canada) (Fig 9).

TREATMENT RESULTS

The posttreatment photographs (Fig 10) show thebalance and harmony of facial proportions that wasachieved with the orthodontic and surgical approach.The midline is in the center of the patient’s face. Shecan close her mouth without mentalis strain.

The posttreatment dental casts (Fig 11) show a Class Iocclusion with normal overjet and overbite. Theocclusion exhibits canine and incisal guidance. Theopen bite was corrected. The maxillary second molarsare still settling and will eventually come into occlusion.

The posttreatment panoramic radiograph (Fig 12)shows that the level of root length was maintained,except for the mandibular left canine and second premo-lar, which had decreases in root length. There was no fur-ther loss of condylar tissue. The TMJs wereasymptomatic. The mandibular incisors were uprightedover basal bone to an IMPA of 79�. Because of theLeFort impaction of the maxilla, the mandible wasautorotated 11 mm. This rotation allowed the FMA todecrease to 41�. The genioplasty projected pogonion far-ther anteriorly to approach the Legan-Burstone glabellaperpendicular. The Z-angle improved to a normal 75�.

All skeletal cephalometric measurements showedimprovement. With the analyses previously described,the McNamara analysis5 and that of Delaire et al6 illus-trate the postsurgical position of the teeth. The maxillaryincisor was positioned 4 mm closer to nasion Frankfortperpendicular at 110� to the palatal plane. The mandiblewas at 83� according to Delaire et al, and the verticalhard-tissue relationship was reduced by 10 mm. TheLegan-Burstone analysis7 showed that vertical soft-tissue glabella to soft-tissue subnasale and soft-tissuesubnasale to soft-tissue menton were in a 1:1 relation-ship. Soft-tissue pogonion was slightly behind soft-tissue glabella perpendicular. The composite cephalo-metric tracings (Fig 13) show mandibular incisor

Fig 12. Posttreatment panoramic radiograph.

Fig 13. Posttreament cephalometric headfilm andsuperimposed tracings.


January 2010

uprighting, maxillary anterior movement and impaction,mandibular autorotation and forward movement, andfacial profile improvement.

The periapical radiographs show the severe rootresorption (Fig 14).

Treatment time was 24 months. A .030-in mandibu-lar lingual retainer was bonded to each anterior tooth toproduce a splinted anterior segment. A removablemaxillary circumferential retainer was also placed.

DISCUSSION

There is no doubt that the retreatment of this patientwas clinically challenging and not without risk. Treat-ment was undertaken with much trepidation. Thisauthor had treated this patient’s adoptive mother withorthodontics and a mandibular advancement with a suc-cessful result several years earlier; this encouraged thepatient to seek retreatment. She had not consideredretreatment before because her first treatment wasdone at a dental school, and she accepted her first out-come as all that could be done. It was reasoned thateven if she lost teeth, prosthetic replacements wouldhave a better prognosis with the jaws in an optimalposition and the open bite corrected. As mentioned,the maxillary teeth were not banded until the mandibu-lar arch was stabilized to try to minimize further rootresorption. Patients with Marfan syndrome have anincreased risk of root resorption and pulpal necrosis

with orthodontic treatment.12 TMJ dysfunction and con-

dylar resorption can be important aspects of the disor-

der,13 as can obstructive sleep apnea and upper airway

resistance.14 Severe periodontitis has also been reported

by Straub.15

Marfan syndrome is an autosomal dominant hered-

itary connective-tissue disorder. The incidence is esti-

mated to be at least 1 case per 5000 to 10,000 people.

The syndrome is caused by gene coding for fibrillin-1,

an extracellular matrix glycol-protein. It was first

described by Dr Bernard Marfan in 1896 and was sub-sequently included among the hereditary disorders ofconnective tissues. The gene responsible for the muta-tion was identified in the region of chromosome15q21.1.16 This patient was adopted, and her familialhistory was unknown. Clinical features of the disordercan include tall stature, ectopia lentis, mitral valve pro-lapse, aortic-root dilation, aortic dissection, joint hyper-mobility, arachnodactyly, dural ectasis, highly archedpalate, dental crowding, down-slanting palpebralfissures, and retrognathia. This patient exhibited allthese features except aortic dissection and ectopialentis.

At the onset of treatment, some questions came tomind. Is the TMJ sensitive to changes in mechanicalloads? Can the amount of root resorption be controlled?What about retention and the need for future prosthetictreatment?

Fig 14. Periapical radiographs show root resorption.


Mongini17 showed condylar changes after occlusalequilibration. Peltola18 found radiographic changes inpatients treated with orthodontics when compared withcontrols. Arnett et al19,20 concluded that the TMJ isnot immutable and that changes in occlusion (lost teeth,orthodontic or orthognathic manipulations), excessiveparafunctional habits, and articular disc-condyle rela-tionships could contribute to remodeling of the articularstructures of the TMJ. They noted that 1 patient canexperience dysfunctional remodeling (ie, condylysis)whereas another subjected to a similar insult might adaptto the mechanical stress with functional remodeling.

Internal derangement can occur with21-25 and with-out23,24,26-28 remodeling. DeBont et al29 showed thatosteoarthrosis of the mandibular condyle can occur inthose with a normal articular disc-condyle relationship.

Furstman30 described the phenomenon that severeosteosclerotic changes of the mandibular condyle havebeen associated with the loss of occlusal stability. Gazitet al31 and Ehrlich et al32 found structural changes in theTMJ associated with unstable occlusion, including boneresorption and fibrocartilage calcification.

‘‘Posteriorization’’ of the mandibular condyle sec-ondary to occlusal changes might lead to postglenoidspine and posterior condylar resorption.10,27,33 Arnettet al,10 Arnett and Tamborillo,33 and Arnett34 observedcondylar resorption when the condyles were displacedposteriorly after orthognathic surgery. Wolford andCardenas35 described some characteristics that appearto make a patient most susceptible to idiopathic condy-lar resorption. These factors include (1) female sex(approximately 9:1 female to male ratio), (2) age rangeof 10 to 40 years with a strong predominance for teen-agers in their pubertal growth phase, (3) high occlusalplane angle and mandibular plane angle, and (4) ClassII skeletal pattern with or without open bite. They foundthat condylar resorption rarely occurs in patients withlow mandibular plane angle or those with a Class IIIskeletal relationship.

A number of systemic disease states can lead tocondylar resorption.36 These include rheumatoid orjuvenile rheumatoid arthritis, systemic lupus erythema-tosus, familial Mediterranean fever, Sjogren’s syn-drome, Marfan syndrome, psoriatic arthritis, andidiopathic condylysis. Arnett et al20 concluded that con-dylar resorption is multifactorial and based on the host’sadaptive capacity and mechanical stimuli. They stated,‘‘when predisposing host factors are not present, occlu-sal treatments normally result in functional remodeling.However, dysfunctional remodeling from low level me-chanical stress (orthodontics, orthognathic surgery,prosthetics) may occur subsequent to an inadequatehost adaptive capacity, coincidental internal derange-

ment of the joint, excessive parafunction, macrotrauma,or unstable occlusion. Dysfunctional remodeling pro-voked by the treatment of dentoskeletal deformities is,to some extent, dependent on the presence of these hid-den factors. However, it seems likely that excessivetreatment compression is capable of initiating substan-tial condylar resorption and resultant occlusal changeswithout contribution of other stimuli.’’ Arnett et al20

described a 3-fold treatment for condylar resorption:(1) control or eradicate the etiologic factors, (2) stabi-lize the unstable occlusion and the TMJs, and (3) correctthe resulting occlusal deformity.

Recommended treatment options for condylarresorption include (1) splint therapy to minimize jointloading, (2) arthroscopic lysis and lavage, (3) condylarreplacement with a costochondral graft if resorptionrecurs or cannot be controlled, and (4) maxillary surgeryto correct the occlusal deformity.19, 20,28,37,38

In this patient, maxillary surgery was chosen toreduce the load on the condyles. She was informedthat a costochondral graft was possible if the condylesresorbed completely. A case report showed that a patientwith virtually no condyles treated with the same regi-men of orthodontics and surgery experienced noadverse sequelae.39 The genioplasty was redone witha cortical osteotomy to suspend the mentalis muscle toachieve optimal facial balance and harmony.

External apical root resorption (EARR) is the loss ofroot structure in the apical region that can be seen onradiographs. It is an unpredictable phenomenon, andits etiology is unknown. Hartsfield et al40 found thatthe degree and severity of EARR are multifactorial,involving host and environmental factors. Geneticfactors account for at least 50% of the variation inEARR. Variation in the interleukin 1 beta gene in ortho-dontically treated patients accounts for 15% of thevariation in EARR. Those authors found historical andcontemporary evidence that the earliest event leadingto EARR is injury to the periodontal ligament (PDL)and supporting structures at the site of root compressionafter orthodontic force.40

Multinucleated cells called odontoclasts responsiblefor the resorption of the dental tissues’ cementum anddentin share many cytochemical and morphologic char-acteristics with osteoclasts that are responsible for boneresorption. Odontoclasts and osteoclast precursors orig-inate from hemopoietic cells in the bone marrow.41

Brezniak and Wasserstein41,42 reported that loss of api-cal root material is unpredictable, and, when it extendsinto the dentin, it is irreversible. Orthodontic force leadsto microtrauma of the PDL and activation of many cel-lular events associated with inflammation. Root resorp-tion begins adjacent to hyalinized zones and occurs


January 2010

during and after elimination of hyaline (necrotic)tissues. In their review of the literature, Brezniak andWasserstein41,42 found that EARR is classified into3 types: surface resorption, involving small areasfollowed by spontaneous repair from intact parts of thePDL; inflammatory resorption when resorption hasreached the dentinal tubules; and replacement resorptionwhen bone replaces the resorbed tooth material andleads to ankylosis. Brudvik and Rygh43 found that rootresorption continued in the area where hyalinized tissuepersisted even after active force had ended. They hy-pothesized that the determinants of resorption and repairgenerally seem to be associated with the persistence andremoval of necrotic tissue and a process of repair startedfrom the periphery in the resorbed lacunae where thePDL had been reestablished, whereas ongoing resorp-tion was observed beneath existing hyalinized tissue.

EARR is the bane of orthodontists and a commonsequela associated with orthodontic treatment.Although EARR is a frequent iatrogenic outcome asso-ciated with orthodontics, Harris and Butler44 and Harriset al45 found that it can also occur without orthodontictreatment, presumably as a function of occlusal forces.DeShields,46 Sharpe et al,47 and Parker and Harris48

reported that the amount of tooth movement is posi-tively associated with the extent of EARR. McNab etal49 found extraction patterns can influence EARRbecause of the increased tooth movement required toclose extraction spaces. Sameshima and Sinclair50

found that patients whose 4 first premolars wereextracted had more EARR than those treated withoutextractions or extractions of only the maxillary first pre-molars. Tainthongchai et al51 found that the amount oftime spent in orthodontic treatment can be a factor inEARR. Lee et al,52 in a clinical study, showed that expo-sure of the roots to 2 sequential orthodontic procedures,1 in adolescence and the other during adulthood,actually decreased the extent of EARR.

This patient’s EARR at the beginning of treatmentwas probably caused by many factors, including theduration of the first treatment (5 years), perhaps the useof vertical elastics to control the open bite, and certainlythe genetic influence of Marfan syndrome. EARR did notappreciably increase during the second treatment exceptfor the mandibular left canine and first premolar. This canbe attributed to careful and slow leveling and retractionof the mandibular incisors and by not having applianceson the maxillary teeth until the mandibular arch was sta-bilized. No elastics were used except for those used bythe surgeon during the surgical care phase of treatment.

Several anecdotal reports have demonstrated thestability of teeth with severe root resorption.53-55

Parker53 showed that severely resorbed maxillary

incisors after orthodontic treatment were still function-ing well after 33 years. Roberts56 suggested that retain-ing teeth with fixed appliances should be done withcaution because occlusal trauma to the fixed teeth orsegments might cause further EARR. A fixed mandibu-lar retainer was bonded from canine to canine, and thepatient’s occlusion was adjusted to provide optimalfunction in all excursive movements. She will be fol-lowed in the long term in retention.

Certainly, the long-term prognosis of her dentition isguarded. Technology and research are constantly evolv-ing. If she retains her teeth for another 10 to 15 years, thetissue and bone support can remain viable for futureesthetic implant placement that will maintain soft-tissuecontours and papillae forms. The early detection andmedical management of Marfan syndrome has signifi-cantly increased her life expectancy. Was the risk worththe reward? The patient thinks it was.

REFERENCES

1. Horn AJ. Facial height index. Am J Orthod 1992;101:180-6.

2. Merrifield LL. The profile line as an aid in critically evaluating

facial esthetics. Am J Orthod 1966;52:804-22.

3. Jacobson A. The ‘‘Wits’’ appraisal of jaw disharmony. Am

J Orthod 1975;67:125-38.

4. Jacobson A. Wits appraisal. In: Jacobson A, editor. Radiographic

cephalometry. Quintessence: Carol Stream, Ill; 1995. p. 97-112.

5. McNamara JA. A method of cephalometric evaluation. Am

J Orthod 1984;86:49-69.

6. Delaire J, Schendel SA, Tulasne JF. An architectural and struc-

tural craniofacial analysis: a new lateral cephalometric analysis.

J Oral Surg 1981;52:226-38.

7. Legan HL, Burstone CJ. Soft tissue cephalometric analysis for

orthognathic surgery. J Oral Surg 1980;38:744-51.

8. Merrifield LL. Differential diagnosis with total space analysis.

J Charles H. Tweed Int Found 1978;6:10-5.

9. Vaden JL, Dale JG, Klontz HA. The Tweed-Merrifield philoso-

phy. In: Graber TM, Vanarsdall RL, editors. Orthodontics: current

principles and techniques. St Louis: C.V. Mosby; 1994. p. 627-84.

10. Arnett GW, Tamborillo, Rathbone JH. Temporomandibular joint

ramifications of orthognathic surgery. In: Bell WH, editor. Mod-

ern practice in orthognathic and reconstructural surgery. Philadel-

phia: W.B. Saunders; 1992. p. 523-93.

11. Merrifield LL. Edgewise sequential directional force technology.

J Charles H. Tweed Int Found 1986;14:22-37.

12. Bauss O, Sadat-Khonsari R, Schwestka-Polly R. Dental hard tis-

sue abnormalities in patients with Marfan syndrome. Proceedings

of the European Orthodontic Society 80th Congress; 2004 June

7-11; Aarhus, Denmark. Available at: www.ejo.oupjournals.org.

13. Bauss O, Sadat-Khonsari R, Fenske C, Engelke W, Schwestka-

Polly R. Temporomandibular joint dysfunction in Marfan syn-

drome. Oral Surg Oral Med Oral Pathol Oral Radiol Endod

2004;97:592-8.

14. Cistulli PA, Richards GN, Palmisano RG, Unger G, Berthon-

Jones M, Sullivan CE. Influence of maxillary constriction and

nasal resistance and sleep apnea in patients with Marfan’s syn-

drome. Chest 1996;110:1184-8.

15. Straub AM. Severe periodontitis in Marfan’s syndrome: a case

report. J Periodontol 2002;73:823-6.


http://www.ejo.oupjournals.org

16. Incisivo V, Silvestri A. Skeletal and occlusal alterations in the

diagnosis of Marfan syndrome. Minerva Stomatol 2003;52:

457-66.

17. Mongini F. Condylar changes after occlusal therapy. J Prosthet

Dent 1980;43:568-77.

18. Peltola JS. Radiologic variations in mandibular condyles of Finn-

ish students, one group orthodontically treated and the other not.

Eur J Orthod 1993;15:223-7.

19. Arnett GW, Milam B, Gottesman L. Progressive mandibular

retrusion—idiopathic condylar resorption. Part I. Am J Orthod


20. Arnett GW, Milam B, Gottesman L. Progressive mandibular

retrusion—idiopathic condylar resorption. Part II. Am J Orthod


21. Link JJ, Nickerson JW Jr. Temporomandibular joint internal

derangements in an orthognathic surgery population. Int J Adult

Orthod Orthognath Surg 1992;7:161-9.

22. Nickerson JW Jr, Boering G. Natural cause of osteoarthritis as it

relates to internal derangement of the TMJ. In: Merrill RG, editor.

Oral maxillofacial clinics of North America. Philadelphia: W.B.

Saunders; 1989. p. 27-45.

23. Westesson PL, Eriksson L, Kurita K. Reliability of a negative

clinical temporomandibular joint examination: prevalence of

disk displacement in temporomandibular asymptomatic joints.

Oral Surg Oral Med Oral Pathol 1989;68:551-4.

24. Westesson PL. Structural hard-tissue changes in temporomandib-

ular joints with internal derangement. Oral Surg Oral Med Oral

Pathol 1985;59:220-4.

25. Eriksson L, Westesson PL. Clinical and radiological study of pa-

tients with anterior disk displacement of the temporomandibular

joint. Swed Dent J 1983;7:55-64.

26. Kircos LT, Ortendahl DH, Mark AS, Arakawa M. Magnetic reso-

nance imaging of the TMJ disc in asymptomatic volunteers. J Oral


27. Kaplan PH, Tu HK, Sleder P, Lydiatt DD, Laney TJ. Inferior

joint space arthography of normal temporomandibular joints:

reassessment of diagnostic criteria. Radiology 1986;159:585-9.

28. Talents RH, Hatala M, Katzberg RW, Westesson PL. Temporo-

mandibular joint sounds in asymptomatic volunteers. J Prosthet

Dent 1993;69:298-304.

29. de Bont LGM, Stecenga B. Pathology of the temporomandibular

joint internal derangement and osteoarthrosis. Int J Oral Maxillo-

fac Surg 1993;22:71-4.

30. Furstman L. The effect of loss of occlusion upon the mandibular

joint. Am J Orthod 1965;51:245-61.

31. Gazit D, Erlich J, Kohen J, Bab I. Effect of occlusal (mechanical)

stimulus on bone remodeling in rat mandibular condyle. J Oral

Pathol 1987;18:395-8.

32. Ehrlich J, Bab I, Jaffee A, Sela J. Calcification patterns of rat con-

dyle cartilage after induced unilateral malocclusion. J Oral Pathol

1982;11:366-73.

33. Arnett GW, Tamborillo JA. Progressive Class II development—

female idiopathic condylar resorbtion. In: West RA, editor. Oral

maxillofacial clinics of North America. Philadelphia: W.B. Saun-

ders; 1990. p. 669-716.

34. Arnett GW. A redefinition of bilateral sagittal osteoomy (BSO)

advancement replapse. Am J Orthod Dentofacial Orthop 1993;

104:506-15.

35. Wolford LM, Cardenas L. Idiopathic condylar resorption: diagno-

sis, treatment protocol, and outcomes. Am J Orthod Dentofacial

Orthop 1999;116:667-77.

36. Sarver D, Proffit W, Ackerman J. Diagnosis and treatment plan-

ning in orthodontics. In: Graber TM, Vanarsdall RL, editors. Or-

thodontics: current principles and practices. 3rd ed. St Louis: C.V.

Mosby; 2000. p. 61-64.

37. Hwang SJ, Haers PE, Zimmermann A, Oechslin C, Seifert B,

Sailer HF. Surgical risk factors for condylar resorption after or-

thognathic surgery. Oral Surg Oral Med Oral Pathol Oral Radiol

Endod 2000;8:542-52.

38. Huang YL, Pogrel MA, Kaban LB. Diagnosis and management of

condylar resorption. J Oral Maxillofac Surg 1997;55:114-9.

39. Bilodeau JE. Retreatment of a patient who presented with

condylar resorption. Am J Orthod Dentofacial Orthop 2007;131:

89-97.

40. Hartsfield JK, Everett ET, Al-Qawasmi RA. Genetic factors in ex-

ternal apical root resorption and orthodontic treatment. Crit Rev

Oral Biol Med 2004;15:115-22.

41. Brezniak N, Wasserstein A. Root resorption after orthodontic

treatment: part 1. Literature review Am J Orthod Dentofacial

Orthop 1993;103:62-6.

42. Brezniak N, Wasserstein A. Orthodontically induced inflamma-

tory root resorption. Part I: the basic science aspects. Angle

Orthod 2002;72:175-9.

43. Brudvik P, Rygh P. Transition and determinants of orthodontic

root resorption-repair sequence. Eur J Orthod 1995;17:177-88.

44. Harris EF, Butler ML. Patterns of incisor root resorption before

and after correction in cases with anterior open bites. Am J Orthod


45. Harris EF, Robinson QC, Woods MA. An analysis of causes of

apical root resorption. Quintessence Int 1993;24:417-28.

46. DeShields RW. A study of root resorption in treated Class II

Division 1 malocclusions. Angle Orthod 1969;39:231-45.

47. Sharpe W, Reed B, Subtelny JD, Polson A. Orthodontic relapse,

apical root resorption, and crestal alveolar bone levels. Am


48. Parker RJ, Harris EF. Directions of orthodontic tooth movements

associated with external apical roor resorption of the maxillary

central incisor. Am J Orthod Dentofacial Orthop 1998;114:

677-83.

49. McNab S, Battistutta D, Taverne A, Symons A. External root re-

sorption following orthodontic treatment. Angle Orthod 2000;70:

227-32.

50. Sameshima GT, Sinclair PM. Predicting and preventing root

resorption: part II. Treatment factors. Am J Orthod Dentofacial

Orthop 2001;119:511-5.

51. Taithongchai R, Sookkorn K, Killiany DM. Facial and dentoal-

veolar structure and the prediction of apical root shortening.


52. Lee RY, Artun J, Alonzo TA. Are dental anomalies risk factors for

root resorption in orthodontic patients? Am J Orthod Dentofacial

Orthop 1999;116:187-94.

53. Parker WS. Root resorption—long-term outcome. Am J Orthod


54. Desai HM. Root resorption: another long-term outcome. Am J


55. Savage RR. Restorative treatment options for patients with severe

orthodontic resorption. Compend Contin Educ Dent 2006;27:

302-6.

56. Roberts WE. Bone physiology, metabolism, and biomechanics in

orthodontic practice. In: Graber TM, Vanarsdall RL, editors. Or-

thodontics: current principles and techniques. 3rd ed. St Louis:

C.V. Mosby; 2000. p. 231-4.


January 2010

CASE REPORT

Distalization of the mandibular dentition withmini-implants to correct a Class III malocclusionwith a midline deviation

Kyu-Rhim Chung,a Seong-Hun Kim,b HyeRan Choo,c Yoon-Ah Kook,d and Jason B. Copee

Uijongbu and Seoul, Korea, Philadelphia, Pa, and Dallas, Tex

This article describes the orthodontic treatment for a young woman, aged 23 years 5 months, with a Class IIImalocclusion and a deviated midline. Two orthodontic mini-implants (C-implants, CIMPLANT Company,Seoul, Korea) were placed in the interdental spaces between the mandibular second premolars and first mo-lars. The treatment plan consisted of distalizing the mandibular dentition asymmetrically and creating spacefor en-masse retraction of the mandibular anterior teeth. C-implants were placed to provide anchorage forClass I intra-arch elastics. The head design of the C-implant minimizes gingival irritation during orthodontictreatment. Sliding jigs were applied buccally for distalization of the mandibular posterior teeth. The activetreatment period was 18 months. Normal overbite and overjet were obtained, and facial balance wasimproved. (Am J Orthod Dentofacial Orthop 2010;137:135-46)

Every orthodontic tooth movement is accompa-nied by a reaction. This can make it difficult tocorrect a malocclusion by using intraoral appli-

ances alone, especially when complete distal movementof the mandibular dentition is planned in nonsurgicalClass III malocclusion treatment. Traditionally, fixed ap-pliances and intermaxillary elastics have been used tomove mandibular molars distally, often resulting in un-desirable proclination of the maxillary incisors and ex-trusion of the maxillary molars as reciprocal sideeffects.1 This can cause an esthetic problem and instabil-ity, especially in long-faced adults. Also, because inter-maxillary elastic wear requires patient compliance, it isdifficult to predict the final result in uncooperative

patients. Therefore, several authors have attempted totreat this type of malocclusion by distal tooth movementalone. For example, animal studies and clinical investi-gations have used conventional implants as absoluteanchorage2-4 and miniplates for intrusion or distalizationof the mandibular posterior teeth.5,6 Because all portionsof the anchor plates and screws were placed outside thedentition in these studies, it was possible to move themandibular molars without disturbing tooth movement.

Recently, the mechanics of group distal movementof teeth with microscrew implant anchorage was intro-duced.7 A distalizing force is applied to the caninesthrough a nickel-titanium (NiTi) coil spring connectingthe miniscrew to hooks on the archwire. The primarytreatment effect in the mandible is distal tipping move-ment of the posterior teeth concurrent with uprightingand distal movement of the anterior teeth. Because thereis no force to move the maxillary anterior teeth forward(via Class III elastics), there are no side effects on themaxillary anterior teeth in microscrew implant-aidedmechanics. Therefore, distal movement assisted bya rigid orthodontic implant can be a good alternativefor treatment when intermaxillary elastics are not indi-cated or the patient is uncooperative.

The stability of temporary skeletal anchoragedevices is achieved from primary mechanical retentionbetween the implant surface and the cortical bone, andsecondary stability is provided by the healing processof the surrounding tissue. Primary stability is importantto minimize the potential for failure from micromotion.Secondary stability is related to the microstructure ofthe implant surface.8-11 In conventional dental

a President, Korean Society of Speedy Orthodontics, Seoul, Korea.b Assistant professor, Department of Orthodontics, Catholic University of

Korea, Uijongbu St. Mary’s Hospital, Uijongbu, Korea.c Attending orthodontist, The Children’s Hospital of Philadelphia; clinical asso-

ciate, University of Pennsylvania School of Dental Medicine, Philadelphia, Pa.d Professor and chairman, Department of Orthodontics, Catholic University of

Korea, Seoul St. Mary’s Hospital, Seoul, Korea.e Adjunct clinical assistant professor, Department of Orthodontics; adjunct

assistant professor, Department of Oral and Maxillofacial Surgery, Baylor

College of Dentistry, Dallas, Tex.

Partly supported by the Korean Society of Speedy Orthodontics, the alumni fund

of the Department of Dentistry, and the Graduate School of Clinical Dental Sci-

ence, Catholic University of Korea.



Reprint requests to: Seong-Hun Kim, Catholic University of Korea, Uijongbu

St. Mary’s Hospital, 65-1 Geumo-dong, Uijeongbu, Gyeonggi-do, 480-717,

South Korea; e-mail, [email protected] or [email protected].

Submitted, April 2007; revised and accepted, June 2007.

0889-5406/$36.00


doi:10.1016/j.ajodo.2007.06.023

135



prosthetic implants, implants with a rough surfaceshowed better stability and tissue reactions than didthose with a smooth surface. In orthodontic implants,porous-surfaced implants show higher marginal bonelevels and less relative implant displacement thanthreaded implants.12,13

The C-implant (CIMPLANT Company, Seoul, Ko-rea) was developed to use osseointegration as themain stabilizing mechanism.14-17 This mini-implanthas an upper abutment head component and a lowerthreaded body or screw-type component (Fig 1). Theunique head design makes it possible to apply multipleelastics while simultaneously preventing the elasticsfrom slipping off.15,16 The body or retentive componentof the C-implant is better able to resist the rotational ten-dency of heavy dynamic loads and control 3-dimen-sional tooth movement as a result of its higherosseointegration potential.

When distalizing the mandibular dentition witha mandibular C-implant, the most important consider-ation is its position. The placement site should be asclose as possible to the mesial surface of the mandibularfirst molar because this will help achieve optimal distal-ization of the mandibular dentition. The initial toothmovement in distalization is posterior movement ofthe second molar by using a sliding jig that is connectedto the main archwire, followed by moving the otherteeth posteriorly (Fig 2). While the second molar is dis-talizing, the first molar also moves distally as a result ofdrifting. When molar distalization is complete, the pre-molars will also begin to move with the sliding jig.While the premolars are distalizing, spaces might de-

velop between the anterior teeth. To retract the anteriorteeth with en-masse retraction, closing loops are placedbetween the lateral incisors and canines, and connectedto the C-implants by elastics. Because intermaxillaryelastics are not applied to the maxillary dentition,mesial movement of the maxillary arch and extrusionof the maxillary molars are avoided, and the incisorsare not flared. This case report describes the distaliza-tion of the mandibular dentition to treat a dentalClass III malocclusion with a deviated midline by usingC-implants.

DIAGNOSIS

The patient was a woman, aged 23 years 5 months,whose chief concern was protruding mandibular teeth.Her medical history was noncontributory, and occa-sional clicking of her temporomandibular joints (TMJ)was noted in her dental history.

The pretreatment facial photographs (Fig 3) show anacceptable facial profile, despite mild midface defi-ciency and slight mandibular prognathism. No facialasymmetry was noticeable in the frontal view. The clin-ical examination (Figs 3 and 4) showed a Class III molarand canine relationship that was more significant on theright side. Other findings included an anterior edge-to-edge relationship, a midline discrepancy, mild man-dibular anterior crowding, and mesial angulation ofthe mandibular posterior teeth. The lower midline wasnot coincident with the facial midline and was shiftedto the left by 2.5 mm. The maxillary third molars andthe mandibular right third molar were missing

Fig 1. Two-part design of the C-implant. Fig 2. Schematic illustration of the C-implant depen-dent on mandibular distalization mechanics.

136 Chung et al American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

(Fig 5A). There was only slight contact between themaxillary right second molar and the opposing tooth be-cause of the Class III molar relationship.

The cephalometric analysis (Fig 5B; Table) showeda skeletal Class III relationship with a high mandibularplane angle and a slightly retrognathic maxilla. The an-terior facial height was slightly long relative to the pos-terior facial height. The incisor position and interincisalrelationship were within normal limits except for theretroclined maxillary incisor. The patient was diagnosedwith a skeletal Class I malocclusion with mild maxillarydeficiency and a dental Class III relationship.

TREATMENT OBJECTIVES

A mandibular premolar extraction plan would bea relatively simple and stable way to resolve the anteriorcrossbite. Complex treatment mechanics and manytooth movements would not be needed. However, the

patient did not want extractions (except for the thirdmolars) or changes to her facial appearance; she wantedonly to correct the incisor relationship. Although themaxillary incisors were slightly upright, the patient re-quested that they not be allowed to move forward.Therefore, we rejected the premolar-extraction treat-ment option.

Based on the initial records and the patient’s desires,the treatment objectives were to distalize all mandibularteeth, improve the interincisal relationship to have nor-mal overjet and overbite, shift the mandibular midline tocoincide with the facial and maxillary midlines, andachieve Class I canine and molar intercuspal relation-ships. A conventional fixed appliance was prescribed.

TREATMENT ALTERNATIVES

Maxillary advancement surgery was not a viabletreatment option because the skeletal deficiency was

Fig 3. Pretreatment extraoral and intraoral photographs.

American Journal of Orthodontics and Dentofacial Orthopedics Chung et al 137Volume 137, Number 1

not significant, and the patient was pleased with herfacial appearance. Maximum anchorage and interarchelastics were discussed for en-masse movement ofthe mandibular dentition. She refused the interarchelastics because of their visibility. Her occasionalclicking was also a matter of concern because it mightlead to TMJ dysfunction symptoms during orthodontictreatment. Therefore, mandibular distalization witha C-implant in the posterior dentition and intra-archelastics was the treatment of choice. After distalmovement of the mandibular dentition, a full fixedappliance would be used in the maxillary dentitionfor finishing.

TREATMENT PROGRESS

Two C-implants, 1.8 mm in diameter and 8.5 mmlong, were placed in the interdental spaces betweenthe mandibular second premolars and first molars.Bone quality in the mandible was good, and the im-plants were loaded immediately. A 0.016-in NiTi initialarchwire was used for leveling and distalization of themandibular posterior dentition. Intra-arch elastics(1/4-in, 3.5 oz) were applied from the 0.7-mm-diameterstainless steel sliding jig to the neck of the C-implant fordistalization of the mandibular second molar and ante-rior decrowding (Fig 6A).

Fig 4. Pretreatment dental casts.

Fig 5. Pretreatment radiographs: A, panoramic; B,cephalometric.


January 2010

The maxillary dentition was not bonded initially be-cause the dental and facial midlines were coincident,and no forward movement of the anterior teeth was de-sired. While the mandibular dentition was distalizing,drifting occurred. Therefore, a power chain was appliedto correct the midline (Fig 6B). This applied an intrusivemovement to the mandibular incisors because the NiTiarchwire was not stiff, and elastics were applied from

the incisors to the apically placed C-implant. To correctthe anterior open-bite tendency from the force directionof the elastics, the mandibular archwire was changed toa 0.016 3 0.022-in NiTi archwire and then a 0.016 3

0.022-in stainless steel archwire with closing loops(Fig 7).

As the mandibular molars moved distally, the max-illary molars were extruded. To correct the extrusion,

Table. Cephalometric measurements

Measurement Average (women)* Pretreatment Posttreatment

SNA (�) 81.6 77 76.5

SNB (�) 79.2 76.5 76

ANB difference (�) 2.4 0.5 0.5

PFH/AFH (%) 85.1/127.4 (66.8) 88.5/137 (64.6) 91/140.5 (64.8)

SN-OP (�) 17.9 20.6 18

FH-U1 (�) 116.0 109.5 110

FMA (�) 24.3 33 33.5

IMPA (�) 95.9 91 80

FMIA (�) 59.8 56 66.5

UL-E plane (mm) �0.9 �1.5 �1

LL-E plane (mm) 0.6 1 0.5

Interincisal angle 123.8 127.5 136.5

Mx 1-NA (mm) 7.3 7 7.5

Mx 1-NA (�) 25.3 24 26

Mn 1-NB (mm) 7.9 8 4.5

Mn 1-NB (�) 28.4 27.3 17

SN-PP (�) 10.2 9.5 9.5

*For Korean women, data from, Korean Association of Orthodontists.31

Fig 6. Progress photographs: A, mandibular second molar distalization with 0.7-mm diameter stain-less steel sliding jig; B, power chain between C-implants and mandibular anterior teeth for retractionand midline correction.


the maxillary dentition was bonded for intrusion andleveling of the maxillary molars. A Class I molar rela-tionship of the mandibular left dentition was achievedby using a sliding jig. The mandibular right dentitionwas distalized 6 mm but still required further move-ment. The sliding jig was continuously applied to themandibular first molar. The mandibular premolars

continued to move separately (Fig 8). The closingloop of the 0.016 3 0.022-in stainless steel archwirewas used as a hook for mandibular en-masse retraction.Distalization of the mandibular dentition and midlinecorrection took 18 months. The fixed appliances wereremoved, and retention was provided by maxillary andmandibular fixed retainers.

Fig 7. Progress photographs of 0.016 3 0.022-in stainless steel archwire with closing loop appliedto the mandibular dentition with a sliding jig to the mandibular right first molar activated for molarrelationship improvement.

Fig 8. Progress photographs of the mandibular right second premolar distalized by the sliding jig andmandibular en-masse retraction with C-implants.


January 2010

TREATMENT RESULTS

The active treatment period was 18 months. Thepatient’s facial profile was mostly unchanged (Fig 9).A Class I canine and molar relationship and normaltooth alignment with better midline coincidence,and normal overjet and overbite were achieved(Figs 9 and 10). The maxillary incisors moved for-ward slightly. The mandibular incisors were retractedconsiderably and extruded. The upper and lower lipsmoved very little. The interincisal angle increased asthe mandibular incisors uprighted and the ANB angleremained unchanged. The posterior facial height-anterior facial height ratio and the FMA were onlyslightly changed in spite of the significant mandibu-lar molar distalization as seen in the superimposition(Fig 11).

The patient was pleased with the treatment results.An ideal incisor relationship and Class I canine andmolar relationship were obtained. All radiographic(Fig 12) and clinical measurements were within accept-able limits. Lingual bonded retainers and wrap-aroundretainers were placed. Intraoral photographs after 8months of retention (Fig 13) showed substantial relapseon the right side back to Class III molar and caninerelationships and a shallow overbite. We asked the pa-tient to wear the wraparound retainer more, and after26 months of retention, the occlusal relationship wasstable (Fig 14).

DISCUSSION

The entire mandibular dentition was distalized withintra-arch elastics and a supporting C-implant between

Fig 9. Posttreatment extraoral and intraoral photographs.


the second premolars and first molars bilaterally andwithout extrusion or forward movement of the maxil-lary dentition. Mandibular posterior distalization beganwith a NiTi wire for anterior decrowding and midlinecorrection. After distalization and decrowding, anterior

spacing was closed rapidly by using elastomeric chainfrom the C-implant. The implant site was based on cor-tical bone thickness, anatomic structures, and soft-tissuefunctional movements. Most reports suggest that thepreferred site for arch distalization with skeletal

Fig 10. Posttreatment dental casts.

Fig 11. Cephalometric superimposition. Black, pretreatment; gray, posttreatment.


January 2010

anchorage is the terminal molar.18,19 The retromolararea has been reported as an optimal placement site, of-fering a relatively thick cortical bone layer in the man-dible.18,19 However, soft-tissue problems can occuraround the screw implants because the soft tissue is usu-ally thicker and more movable in the retromolar areathan in other areas.7 This can result in inflammation,patient discomfort, and difficulty applying elastics orNiTi coil springs.

Alternative sites for posterior anchorage are theedentulous areas of the alveolar process and posterioralveolar bone.19,20 Therefore, the alveolar bone aroundthe posterior teeth can be the site of choice in patientswithout edentulous areas. Miniscrews can be placedbetween the roots of the posterior teeth without damag-ing the roots, because the cortical bone in these areas isnot very thin.20

Orthodontic mini-implants can be placed eitherbetween the first and second molars or between the sec-ond premolar and the first molar in the mandibular arch.The thickness of the cortical bone between the firstand second molars is enough to provide primary

stability,19,20 but this site is not recommended becauseof tissue irritation during mastication.21 Thus, thealveolar bone between the second premolar and the firstmolar might be a good choice for minimum discomfortand maximum stability. The mental foramen and man-dibular canal can be avoided if the implant is placednot too far from the apex of the adjacent teeth.21

We reviewed computed-tomography studies ofinterradicular space to prevent root damage.20,22

Park20 evaluated the computed tomography imagesand reported bone thicknesses and distances betweenroots 5 to 7 mm apical to the alveolar crest. The averagedistance between the roots of the mandibular secondpremolar and first molar was 3.47 mm (range, 2.0-4.8mm). The distance between the roots of the mandibularfirst and second molars was 4.57 mm (range, 2.7-6.5mm). Park et al21 found smaller distances of 2.4 to3.3 mm between the second premolar and the first mo-lar, and 2.8 to 3.7 mm between the first and second mo-lars. The distances from the cortical bone surface to theinterradicular space however, were relatively larger(3.7-4.2 and 5.3-7.0 mm, respectively). It follows thatorthodontic implants 1.8 mm in diameter can be placedat a slight angle of inclination relative to the buccal cor-tical bone to overcome the limitation of minimum inter-radicular space.

The amounts of distal tooth movement have beenreported previously. Saito et al23 reported 1.8 to10.7 mm of tooth movement in a dog study. In anotherreport, the average amounts of distalization of the man-dibular first molars were 3.5 mm at the crown level and1.8 mm at the root level. The average amount of relapsewas 0.3 mm at both the crown and root apex levels.6 Ina recent case report, the mandibular dentition was distal-ized 5 and 2 mm on the left and right sides, respec-tively.16 In another case report, the mandibularposterior teeth were distalized 6 and 4 mm on the rightand left sides, respectively.24

As seen in the treatment results reported here, distal-ization of the entire mandibular dentition was accompa-nied by a slight mandibular posterior extrusion, which isnot problematic in brachyfacial patients. However, insevere dolichofacial patients, intrusion of posterior teethwould be desirable, since counterclockwise mandibularrotation leads to more esthetic results. Maxillary poste-rior C-implants and interarch elastics might be goodchoices in such cases.16 Interarch elastics induce man-dibular posterior intrusion by tip-back mechanics,instead of extrusion. Maxillary C-implants can also beused as anchorage for additional maxillary posteriorintrusion. This allows selection of specific treatmentmechanics for distalization based on the skeletal andfacial pattern of patients.

Fig 12. Posttreatment radiographs: A, panoramic; B,cephalometric.


Another issue of concern is the relationship betweeninterarch elastics and temporomandibular disorders(TMD). According to a previous report, Class IIinterarch elastics were not related to TMD.25 However,Class III elastic usage in patients with subclinical TMDproblems needs more careful consideration. Someauthors suggested that a posteriorly positioned condyleis a common predisposing factor in anterior TMJdisc displacement.26,27 It has also been reported thatinterarch force might be an etiologic factor of TMD in

animals.28 Although the exact cause of TMD is notcompletely understood, loading is at least considereda possible etiologic causes. Therefore, it might be pru-dent to create a treatment plan that minimizes condylarloading in patients with potential TMD problems.

A final issue of concern is the relationship betweenthe long-term retention results and the entire dentitiondistalization method. We recommended the fixedretainer and wraparound retainer combined retentionmethod to the patient and also encouraged her to chew

Fig 13. Retention intraoral photographs at 8 months.

Fig 14. Retention intraoral photographs at 26 months.


January 2010

on both sides after treatment. However, she acceptedonly the fixed retainers. Eight months after debonding,significant relapse on the right side was observed. Therelapse tendency was assumed to be due to severe tip-ping of the mandibular molar distally, as can be seenin the cephalometric radiograph, and also to insufficientretention and habitual mastication on left side. Merri-field29 defined directional forces as those that use direc-tional control to precisely position the teeth, and Limaand Lima30 showed 4 years of stable retention aftermandibular dentition distalization treatment in a casereport.

If the mandibular molars in our patient werecontrolled bodily by additional loop mechanics aftertip-back movement, a more stable result was possible.The sliding jig application on a stiffer archwire wouldbe better than only tip-back treatment methods forbodily distalization of molars. On the contrary, thedistalized mandibular left dentition showed more stableretention compared with the right dentition. It wasassumed that second molar distalizing began a weekafter the mandibular left third molar extraction, andthe bodily tooth movement was possible because of rel-atively low cortical resistance on that side. Maxillaryand mandibular wraparound retainers were used inthis patient in that period, and she had a recall check 3months later. After 26 months of retention, the patientdid not show a significant relapse tendency as she hadafter 8 months of retention (Fig 14).

Mandibular premolar extraction without movingmolars distally would perhaps have led to a more stableresult. Despite the patient’s request, we should havepersuaded her to select this simpler treatment option.However, the tip-back tooth movement and long-termretention results of this patient will be helpful for clini-cians who consider similar treatment mechanics forpatients with Class III anterior crossbite.

In cases of whole-dentition distalization, we recom-mend both a wraparound removable retainer and a fixedretainer, and that chewing be done on both sides. ClassIII intermaxillary elastics applied to a splint-typeretainer with hooks for night wear can be a good alter-native to conventional retention methods.

CONCLUSIONS

The C-implant can withstand heavier loads thanother skeletal anchorage systems. It also has the advan-tage that its abutment head design can be used for elasticapplications. In this case, the C-implant and multipleintra-arch elastics distalized the entire mandibulardentition independently, without extrusion or flaringof the maxillary dentition.

We thank Jae-Hee Cho for assisting with manuscriptpreparation.

REFERENCES

1. Proffit WR. Interarch elastics: their place in modern orthodontics.

In: Hosl E, Baldauf A, editors. Mechanical and biological basics

in orthodontic therapy. Huthig Buchverlag GmbH, Heidelberg,

Germany: 1991. p. 173-8.

2. Roberts WE, Smith RK, Zilberman Y, Mozsary PG, Smith RS.

Osseous adaptation to continuous loading of endosseous implants.


3. Roberts WE, Helm FR, Marchall KJ, Gongloff RK. Rigid endo-

sseous implants for orthodontic anchorage and orthopedic anchor-

age. Angle Orthod 1989;59:247-56.

4. Higuchi KW, Slack JM. The use of titanium fixtures for intraoral

anchorage to facilitate orthodontic tooth movement. Int J Oral

Maxillofac Implants 1991;6:338-44.

5. Umemori M, Sugawara J, Mitani H, Nagasaka H, Kawamura H.

Skeletal anchorage system for open-bite correction. Am J Orthod


6. Sugawara J, Daimaruya T, Umemori M, Nagasaka H, Takahashi I,

Kawamura H, et al. Distal movement of mandibular molars in

adult patients with the skeletal anchorage system. Am J Orthod


7. Park HS, Lee SK, Kwon OW. Group distal movement of teeth using

microscrew implant anchorage. Angle Orthod 2005;75:602-9.

8. Wehrbein H, Glatzmaier J, Yildirim M. Orthodontic anchorage

capacity of short titanium screw implants in the maxilla. An experi-

mental study in the dog. Clin Oral Implants Res 1997;8:131-41.

9. De Pauw GA, Dermaut L, De Bruyn H, Johansson C. Stability of

implants as anchorage for orthopedic traction. Angle Orthod

1999;69:401-7.

10. Gotfredsen K, Berglundh T, Lindhe J. Bone reactions adjacent to

titanium implants subjected to static load. A study in the dog (I).


11. Gotfredsen K, Berglundh T, Lindhe J. Bone reactions adjacent to

titanium implants with different surface characteristics subjected

to static load. A study in the dog (II). Clin Oral Implants Res 2001;

12:196-201.

12. Oyonarte R, Pillar RM, Deporter D, Woodside DG. Peri-implant

bone response to orthodontic loading: part 1. A histomorphomet-

ric study of the effects of implant surface design. Am J Orthod


13. Oyonarte R, Pillar RM, Deporter D, Woodside DG. Peri-implant

bone response to orthodontic loading: part 2. Implant surface

geometry and its effect on regional bone remodeling. Am J Orthod


14. Chung KR, Kim SH, Kook YA. The C-orthodontic micro-implant.

J Clin Orthod 2004;38:478-86.

15. Chung KR, Nelson G, Kim SH, Kook YA. Severe bidentoalveolar

protrusion treated with orthodontic microimplant-dependent en-

masse retraction. Am J Orthod Dentofacial Orthop 2007;132:

105-15.

16. Chung KR, Kim SH, Kook YA. C-orthodontic microimplant for

distalization of mandibular dentition in Class III correction. Angle

Orthod 2005;75:119-28.

17. Jeon MS, Kang YG, Mo SS, Lee KH, Kook YA, Kim SH. Effects

of surface treatment on the osseointegration potential of orthodon-

tic mini-implant. Korean J Orthod 2008;38:328-36.

18. Kim JH, Joo JY, Park YW, Cha BK, Kim SM. Study of maxillary

cortical bone thickness for skeletal anchorage system in Korean.

J Kor Oral Maxillofac Surg 2002;28:249-55.


19. Costa A, Raffaini M, Melsen B. Miniscrews as orthodontic

anchorage: a preliminary report. Int J Adult Orthod Orthognath

Surg 1998;13:201-9.

20. Park HS. An anatomical study using CT images for the im-

plantation of micro-implants. Korean J Orthod 2002;32:

435-41.

21. Park YC, Kim JK, Lee JS. Atlas of contemporary orthodontics.

Vol. III. Seoul, Korea: Sinheung; 2005. p. 178-93.

22. Carano A, Velo S, Incorvati C, Poggio P. Clinical applications of

the mini-screw anchorage system (M.A.S) in the maxillary alve-

olar bone. Prog Orthod 2004;5:212-30.

23. Saito S, Sugimoto N, Morohashi T, Ozeki M, Kurabayashi H,

Shimizu H, et al. Endosseous titanium implants as anchors for

mesiodistal tooth movement in the beagle dog. Am J Orthod Den-


24. Park HS, Kwon TG, Sung JH. Nonextraction treatment with

microscrew implants. Angle Orthod 2004;74:539-49.

25. O’Reilly MT, Rinchuse DJ, Close J. Class II elastics and

extractions and temporomandibular disorders: a longitudinal

prospective study. Am J Orthod Dentofacial Orthop 1993;103:

459-63.

26. Farrar WB, McCarty WL. A clinical outline of temporomandibu-

lar joint diagnosis and treatment. Montgomery, Ala: Walker Print-

ing; 1983. p. 84-5.

27. Wyatt WE. Preventing adverse effects on the temporomandibular

joint through orthodontic treatment. Am J Orthod Dentofacial

Orthop 1987;91:493-9.

28. Fujimura K, Kobayashi S, Suzuki T, Segami N. Histologic evalu-

ation of temporomandibular arthritis induced by mild mechanical

loading in rabbits. J Oral Pathod Med 2005;34:157-63.

29. Merrifield LL. Differential diagnosis. Semin Orthod 1996;2:

241-53.

30. Lima CEO, Lima MTO. Directional force treatment for an adult

with Class III malocclusion and open bite. Am J Orthod Dentofa-

cial Orthop 2006;129:817-24.

31. Korean Association of Orthodontists. Cephalometric norm of

Korean adults with normal occlusion. Korea: Ji-Sung Publishing

Co.; 1998. p. 589-95.


January 2010

TECHNO BYTES

WiPics: Wireless and beyond

Ameet V. Revankar,a Narayan H. Gandedkar,a and Sanjay V. Ganeshkarb

Dharwad, Karnataka, India

A WiPics transmitter allows wireless tranfer of images from the camera to the computer as the photos are be-ing shot. This article describes this novel technological approach and its application in orthodontic imaging.(Am J Orthod Dentofacial Orthop 2010;137:147-9)

With instant results becoming the norm, the oldadage ‘‘time is money’’ never sounded moretrue. Innovations that can simplify and orga-

nize our increasingly hectic workplaces are valuable.One such modern invention is wireless computer network-ing technology, or WiFi. How does WiFi blend with digitalcameras? Awireless digital camera allows you to connectto a network without using cables. This makes it possibleto download and save photos directly to a computer as youshoot them, or even connect to a cellular network andshare photos, just as you would with a camera phone.1

Migrating to a wireless camera is not just a matter ofspending money on new equipment; only a handful ofwireless cameras are available. One cannot sacrifice theneed for essential features—eg, a macro lens—for thesake of freedom from wires. It is more rational to usea wireless transmitter that is compatible with your exist-ing image acquisition armamentarium. This objective iswell accomplished by using a WiPics transmitter (WiPics,United Imaging Solutions, Medina, NY). Other wirelessimage transmitters are also available, including modelsby Canon and Nikon, but they are compatible with onlyspecific camera models from the same manufacturer.

The WiPics transmitter was the brainchild of DaveRea, a student at Rochester Institute of Technology, Ro-chester, New York; he devised it as a part of his thesisproject in 2003.2 It consists of a dummy compact flashcard (camera interface card) connected to a belt-mounted unit (Fig 1). The belt-mounted unit has addi-tional slots for a conventional memory card and a hard

disk drive. The dummy flash card is inserted directlyinto the memory slot of the camera, which directs theshot images to the belt-mounted unit. The images arethen saved on the memory card or the hard drive, ortransmitted wirelessly to a computer via 802.11 g orits Ethernet port.3 For the different configuration possi-bilities of WiPics, see the Table.

Any digital camera that can accept type II compactflash cards will work with the WiPics transmitter. How-ever, to best use the hard drive option, the camera shouldhave FAT32 (a type of computer file system architecture:file allocation table) support. Otherwise, you will not beable to access all of the available hard-drive capacity.4

The WiPics transmitter easily configures with existingor new wireless networks. Two configurations are possible:(1) via the small liquid crystal display on its top plate and(2) via an integrated web server. When the integrated webserver is used, the transmitter is connected to the user’s net-work (usually via Ethernet), and settings can be dialed in bythe agency of commonly used web browsers such as Inter-net Explorer and Mozilla Firefox. Alternately, productssuch as Image Transfer Protocol, a software program(Pixagent Workflow Systems, Kleinburg, Ontario,Canada), can dial in these settings automatically.4

Fig 1. The WiPics transmitter: 1, digital camera;2,WiPics belt-mounted unit; 3, camera interface cardor dummy card (reproduced with permission fromUnited Imaging Solutions, Medina, NY).

From the Department of Orthodontics and Dentofacial Orthopedics, Sri

Dharmasthala Manjunatheshwara College of Dental Sciences and Hospital,

Dharwad, Karnataka, India.a Assistant professor.b Professor and chair.



Reprint requests to: Ameet V. Revankar, Department of Orthodontics and Den-

tofacial Orthopedics, Sri Dharmasthala Manjunatheshwara College of Dental

Sciences and Hospital, Sattur, Dharwad, India 580009; e-mail, drameet@

orthodontist.net.

Submitted, September 2008; revised and accepted, December 2008.

0889-5406/$36.00


doi:10.1016/j.ajodo.2008.12.019

147



The power supply to the belt-mounted unit isthrough a lithium ion rechargeable battery that haspower backup of up to 5 hours. The unit can also berun on a 110 to 240 v wall power adapter providedwith the unit; it also acts as a charger for the battery.Because the WiPics does not derive power from thecamera, there is no deterioration of camera function.

WiPics has the following advantages: (1) imagescan be sent securely and wirelessly; (2) time is saved;(3) there are no issues with ‘‘memory card full’’ mes-sages; (4) there are no risks of photographs being de-leted, formatted, or lost from the memory card; (5) ithas live image review and analysis capability: real-time visualization of the images on a larger computerscreen for blemishes that cannot be seen on the smallercamera display unit; (6) it can back up images at thetime of capture; and (7) it has a batch-renaming feature:it can scan unique bar codes to identify your images andrename them accordingly.3

Batch renaming allows you to change cryptic filenames generated by the camera into a more legible for-mat in a simple and transparent manner. You can scana bar code containing information about the patient,such as name and stage of the photographs. The imagesshot after a bar code scan are renamed automatically ac-cording to the bar code guideline (called ‘‘preshot dataassociation’’). The images are stored in separate directo-ries generated on the basis of the bar code, thus allowingautomatic organization as soon as the images are re-ceived by the computer. This makes future searchesa matter of few clicks. Patient data association canalso be accomplished by performing a bar code scan af-ter a set of images is taken (called ‘‘postshot data asso-ciation’’). Also, bar code scanning makes WiPicsa candidate for standard DICOM applications.

WiPics has the following disadvantages: (1) it re-quires an additional belt-mounted unit that the operatormust carry; (2) it requires an additional configuredwireless network; and (3) transmission depends on anuninterrupted power supply to the wireless network.

However, if the wireless network fails, the imagesare not lost because a backup image is always storedin the compact flash card or additional hard drive inthe belt-mounted unit.

Newer developments in wireless image transmissionare the Eye-Fi (Mountain View, Calif) secure digital(SD) wireless card5 and WiPics mobile. Eye-Fi is a wire-less-enabled SD card that can be inserted into the memoryslot of the camera. It directly transmits images from thecamera to the computer via any configured wireless net-work. It works without additional wires and cables. How-ever, it draws power from the camera and might result indeteriorated camera performance.6 Because the wirelesssignals are generated from the card inside the camera,the image transfer range is limited.6 And there is noway to automatically rename images according to patientdata with a bar code scan.

The WiPics Mobile is a miniaturized version of theWiPics that is being developed now. It will be smallerand faster, mount under the camera (without a belt mount-ing), connect to the camera via the USB port, and haveeven more options to apply data association to all images.

CONCLUSIONS

Armed with wireless technology, digital imaging isbreaking new ground. WiPics and other wireless devicescan eliminate the step of transferring images from a cam-era-based storage card to the computer by direct wirelesstransfer, thus eliminating camera-to-computer tethers,such as network cables. With WiPics, you can havemore than just wireless, as already discussed. Thus, wecan say that this technology strides truly beyond wireless.

Note: We want to clarify that all information men-tioned in this article was taken from the websites men-tioned in the references. We do not have personal,firsthand experience with any of these devices.

REFERENCES

1. Wireless digital camera guide. Available at: http://www.photography

review.com/wirelesscameracrx.aspx. Accessed on September 1, 2008.

Table. Configuration possibilities of WiPics

WiPics configuration type Components Function

1. WiPics base unit 2 flash card slots, maximum 64 MB

compatible

Flash card 1, storage of original camera file; Flash

card 2, storage of data associated file*

2. WiPics with hard drive 1 hard disk drive, 20 GB, partitioned into

2 drives of 10 GB each

Drive 1, storage of original camera file; Drive 2,

storage of data associated file*

3. WiPics with bar-code scanner Same as base unit and integrated

bar-code scanner

Both flash card slots used to enable

data association*

4. WiPics with bar-code scanner

and hard drive

Same as WiPics with hard drive and

integrated bar-code scanner

Data association enabled by partitioning the

hard drive into 2 drives of 10 GB each

*Data association means enabling the association of the data from the bar-code scan so that the images are renamed automatically and meaningfully.

In the first 2 configurations, an external bar-code scanner must be used; in the WiPics with bar-code scanner, the scanner is built into the base unit.

148 Revankar, Gandedkar, and Ganeshkar American Journal of Orthodontics and Dentofacial Orthopedics

January 2010

http://www.photographyreview.com/wirelesscameracrx.aspx

http://www.photographyreview.com/wirelesscameracrx.aspx

2. WiPics. Available at: http://www.prophotohome.com/forum/

pro-photo-wiki-select-articles/73321-wipics.html. Accessed on

September 1, 2008.

3. Wipics—more than just wireless. Available at: http://www.

wi-pics.com/applications.aspx. Accessed on September 1,

2008.

4. WiPics. Available at: http://www.photographicworkflow.com/wiki/

WiPics. Accessed on September 1, 2008.

5. Eye-Fi. Available at: http://www.eye.fi. Accessed on November 15,

2008.

6. Eye-Fi forums. Available at: http://forums.eye.fi/index.php.

Accessed on November 15, 2008.

American Journal of Orthodontics and Dentofacial Orthopedics Revankar, Gandedkar, and Ganeshkar 149Volume 137, Number 1

http://www.prophotohome.com/forum/pro-photo-wiki-select-articles/73321-wipics.html

http://www.prophotohome.com/forum/pro-photo-wiki-select-articles/73321-wipics.html

http://www.wi-pics.com/applications.aspx

http://www.wi-pics.com/applications.aspx

http://www.photographicworkflow.com/wiki/WiPics

http://www.photographicworkflow.com/wiki/WiPics

http://www.eye.fi

http://forums.eye.fi/index.php

DIRECTORY: AAO OFFICERS AND ORGANIZATIONS - orto

Documents

Transcript of DIRECTORY: AAO OFFICERS AND ORGANIZATIONS - orto