Angle

Volume 80, Issue 4 (July 2010)
What's New in Dentistry
611 Latest research reports from the dental literature
Vincent Kokich
Citation | Full Text | PDF (40 KB)
Original Articles
613 Stability of surgically assisted rapid maxillary expansion and orthopedic

maxillary expansion after 3 years' follow-up
Gökmen Kurt, Ayşe Tuba Altug-Ataç, Mustafa Sancar Ataç, and Hakan Alpay Karasu
Abstract | Full Text | PDF (256 KB)
620 Pattern and amount of change after orthodontic correction of upper front
teeth 7 years postretention
Anders Andrén, Sasan Naraghi, Bengt Olof Mohlin, and Heidrun Kjellberg
626 Canine retraction rate with self-ligating brackets vs conventional

edgewise brackets
S. Jack Burrow
634 Temporary anchorage device insertion variables: effects on retention
Joseph S. Petrey, Marnie M. Saunders, G. Thomas Kluemper, Larry L. Cunningham, and

Cynthia S. Beeman
642 Early headgear effect on the eruption pattern of maxillary second molars
Yossi Abed and Ilana Brin

649 Suitability of orthodontic brackets for rebonding and reworking following

removal by air pressure pulses and conventional debracketing
techniques
Michael Knösel, Simone Mattysek, Klaus Jung, Dietmar Kubein-Meesenburg, Reza Sadat-
Khonsari, and Dirk Ziebolz
656 Patient attitudes toward retention and perceptions of treatment success
Nikolay D. Mollov, Steven J. Lindauer, Al M. Best, Bhavna Shroff, and Eser Tufekci
662 Digital models vs plaster models using alginate and alginate substitute
materials
Gilda Torassian, Chung How Kau, Jeryl D. English, John Powers, Harry I. Bussa, Anna Marie
Salas-Lopez, and John A. Corbett
670 Reliability of reference distances used in photogrammetry
Muge Aksu, Demet Kaya, and Ilken Kocadereli

678 Clinical photography vs digital video clips for the assessment of smile
esthetics
Brian J. Schabel, Tiziano Baccetti, Lorenzo Franchi, and James A. McNamara Jr

685 Accuracy and reliability of palatal superimposition of three-dimensional

digital models
Dong-Soon Choi, Young-Mok Jeong, Insan Jang, Paul George Jost-Brinkmann, and Bong-Kuen
Cha
692 Skeletal changes of maxillary protraction without rapid maxillary

expansion
Dong-Yul Lee, Eun-Soo Kim, Yong-Kyu Lim, and Sug-Joon Ahn

699 A cephalometric study to investigate the skeletal relationships in patients

with increasing severity of hypodontia
Priti N. Acharya, Steven P. Jones, David Moles, Daljit Gill, and Nigel P. Hunt
707 3-D analysis of facial asymmetry in children with hip dysplasia
Shane Rex Tolleson, Chung How Kau, Robert P. Lee, Jeryl D. English, Virpi Harila, Pertti
Pirttiniemi, and Marita Valkama
713 Quantity and quality assessment of randomized controlled trials on

orthodontic practice in PubMed
Tatsuo Shimada, Hisako Takayama, and Yoshiki Nakamura

719 Cervical vertebra morphology in different skeletal classes
Miyuki Watanabe, Tetsutaro Yamaguchi, and Koutaro Maki

725 Longitudinal growth changes of the cranial base from puberty to

adulthood
Zuleyha Mirzen Arat, Hakan Türkkahraman, Jeryl D. English, Ronald L. Gallerano, and Jim C.
Boley
733 Craniofacial growth in ectodermal dysplasia
Claudia Dellavia, Francesca Catti, Chiarella Sforza, Davide G. Tommasi, and Virgilio Ferruccio
Ferrario
740 The role of heme oxygenase-1 in mechanical stress- and

lipopolysaccharide-induced osteogenic differentiation in human
periodontal ligament cells
Jin-Hyoung Cho, Sun-Kyung Lee, Jin-Woo Lee, and Eun-Cheol Kim

748 Feeding and nonnutritive sucking habits and prevalence of open bite and
crossbite in children/adolescents with Down syndrome
Ana Cristina Oliveira, Isabela Almeida Pordeus, Cintia Silva Torres, Milene Torres Martins, and
Saul Martins Paiva
754 Quantification of three-dimensional orthodontic force systems of T-loop

archwires
Jie Chen, Serkis C. Isikbay, and Edward J. Brizendine

759 Cytotoxic effects of orthodontic composites
Siddik Malkoc, Bayram Corekci, Hayriye Esra Ulker, Muhammet Yalçın, and Abdülkadir Şengün
765 Influence of ceramic (feldspathic) surface treatments on the micro-shear

bond strength of composite resin
Sumit Yadav, Madhur Upadhyay, Gilberto Antonio Borges, and W. Eugene Roberts
Case Reports
771 Absolute anchorage with universal t-loop mechanics for severe deepbite
and maxillary anterior protrusion and its 10-year stability
Yoon Jeong Choi, Chooryung Judi Chung, Kwangchul Choy, and Kyung-Ho Kim
783 New treatment modality for maxillary hypoplasia in cleft patients
Seung-Hak Baek, Keun-Woo Kim, and Jin-Young Choi

Editorial
792 Why read this journal?
Robert J. Isaacson
Citation | Full Text | PDF (44 KB)

What’s New in Dentistry
Vincent Kokich, DDS, MSD
Success rates for immediate placement and countries is routine attenders, with rates being lower
immediate loading of molar implants are encour- among men and in particular social, ethnic, or age
aging. When implants were re-introduced into dentist- groups. A study published in the Journal of Dental
ry in the early 1980’s, the standard protocol for placing Research (2010;89:307–311), evaluated whether long-
molar implants into extraction sites was to wait for term routine dental attenders had better self-rated oral
several months to allow the socket to heal. In addition, health and lower experience of dental caries and
after placement of an implant in a healed edentulous missing teeth by age 32. This was a longitudinal study
site, the time of loading of a molar implant was that assessed a group of 1037 individuals who were
traditionally four to six months. But today, clinicians part of a health and development study. These
and researchers are suggesting that molar implants individuals had an oral examination performed at three
can be placed into fresh extraction sockets or loaded years of age, and then at 15, 18, 26 and 32 years of
immediately in healed sites. Is there sufficient ev- age. Information on use of dental services was
idence to suggest that earlier placement and loading collected at ages 15, 18, 26 and 32 years. At each
protocols are successful? A systematic review and age, dental examinations for caries and missing teeth
meta-analysis published in the International Journal of were conducted. Individuals were also asked to self-
Oral and Maxillofacial Implants (2010;25:401–415) rate their oral health. The participation rate of the study
assessed the survival of immediately placed single was high, with 96% of the participants taking part in all
implants in fresh molar extraction sites and immedi- five assessment periods. When the participants were
ately restored/loaded single molar implants in healed asked whether they were routine or problem-based
molar sites. These researchers searched the main attenders, the authors found that routine attending
electronic databases. The authors identified nine prevalence fell from 82% at age 15 to 28% at age 32.
studies describing 1013 immediately placed implants However, at any given age, routine attenders had
and seven studies with 188 immediate or delayed better-than-average oral health. In fact, by age 32
loaded implants. The authors discovered that the routine attenders had better self-reported oral health
survival rate of immediately placed molar implants along with less tooth loss and fewer caries. The
was 99.0%, while the immediately restored implants authors conclude that routine dental attendance is
had a survival rate of 97.9%, with no difference associated with better oral health.
between immediate and delayed loading. However,
the authors acknowledge that their study has several Discectomy of the TMJ reduces pain and
limitations. These include the questionable quality of improves function. It is estimated that temporoman-
the existing literature, with most of the included studies dibular joint dysfunction may exist in 10% to 30% of the
classified as fair or average, as well as the great general population. Most of these individuals can be
variability in study designs of previous studies. As a managed successfully with nonsurgical methods such
result, the authors state that their review shows as physical therapy, bite splints, moist heat, arthro-
encouraging results for immediately placed or imme- centesis, intra-articular injections, or pharmacothera-
diately loaded implants placed in molar sites. py. But about 5% of patients whose nonsurgical
therapy fails, require open joint surgery. Discectomy
Routine dental visits associated with better oral is the most common surgery performed for the painful
health. Adult users of dental care can be divided into TMJ. The degree of success for this procedure was
two categories: those who are routine attenders and reported in an article that was published in the Journal
those who seek care because of an acute or chronic of Oral and Maxillofacial Surgery (2010;68:782–789).
problem. While promoting regular dental visits is one of The purpose of this study was to evaluate outcomes of
the cornerstones of preventive dentistry, there is only patients who underwent temporomandibular joint
cross-sectional evidence in the literature that regular discectomy without replacement as the primary treat-
dental visits lead to better dental health. Typically, only ment for internal derangement after failure of nonsur-
about half of the adult population in most Western gical therapy. A cohort of thirty consecutive patients
611 Angle Orthodontist, Vol 80, No 4, 2010

612
with TMJ internal derangement was treated with Furthermore, this research model showed that only the
discectomy. Although six patients were lost to follow- disease score was significantly associated with the
up, 24 patients were recalled and evaluated to probability of patients losing a specific number of teeth.
determine their mandibular mobility and joint junction, The authors conclude that classifying a patient by
as well as the degree of reduction in TMJ and severity of periodontal disease may be beneficial in the
muscular facial pain. Prior to the discectomy, all management of the periodontal patient. The disease
patients had moderate to severe pain in the TMJ and score provides an objective means to quickly deter-
masticatory muscles, and/or locking of the joint. mine disease severity, which leads to establishing the
Postoperatively, 20 of 24 patients had a TMJ that patient’s future risk for tooth loss.
was in a clinically symptom-free state or with a small,
minor dysfunction. TMJ pain, muscle pain, and pain Visible presence of third molars in young adults
with mobility scored low, indicating a subjectively is associated with periodontal inflammation of
successful outcome. The authors conclude that disc- non-third molars—Several years ago, researchers
ectomy of the TMJ as a primary surgical option clearly documented an association between the
significantly reduces pain and improves function in presence of third molars and periodontal pathology
patients with moderate to severe internal derangement that affected adjacent second molars. But is there any
of the temporomandibular joint. relationship between the presence of visible third
molars and periodontal inflammation in other areas of
Periodontal disease and risk scores can be used the mouth? A study published in the Journal of Oral
to predict tooth loss. Periodontitis is a variably and Maxillofacial Surgery (2010;68:325–329), sought
progressive and dynamic pathologic process that to determine whether a relationship could be confirmed
causes attachment loss, destroys alveolar supporting between four visible asymptomatic third molars and
bone and can terminate with tooth loss. A goal of the presence of periodontal inflammation in other parts
periodontal therapy is to stop the loss of bone and of the dentition. Two groups of subjects were
thereby preserve the natural dentition. Tooth loss due identified. The visible group of 342 subjects had at
to periodontal disease varies among subjects and has least one third molar exposed and visible. The not
been shown to be related to the severity of the visible group consisted of 69 subjects who had all four
disease. In addition to severity, the risk for future third molars not exposed. The authors did a thorough
periodontal deterioration is a factor of tooth loss by periodontal assessment to determine the presence of
its effect on the rate of disease progression. Is it periodontal disease. Based upon their assessment,
possible to predict future tooth loss during periodontal the authors found that the subjects in the visible group
treatment, when patients are categorized at the were significantly more likely to have at least one
inception of treatment by disease severity and risk pocket depth of 4 mm or greater on non-third molars
level? A study published in the Journal of Periodonto- than those in the not visible group. In both groups, first
logy (2010;81:244–250) evaluated that research ques- and second molars were more likely to be affected
tion. In order to determine the answer, each of nine than non-molars. The authors found that the severity of
periodontists evaluated 100 consecutive periodontal the disease was low in the adolescent and young adult
maintenance patients. The disease severity and risk subjects with only beginning stages of periodontal
level were determined from data at the initial exami- disease detected. However, the number of first and
nation. The number of teeth lost was determined from second molar pocket depths of 4 mm or greater tended
data at the initial and maintenance visits. Based upon to be higher for the visible group than for the non-
the evaluation of this cohort of patients, the authors visible group. The authors conclude that the visible
showed that disease scores (severity) and risk (level) presence of third molars in adolescents and young
scores could predict the mean tooth loss rate. The adults was significantly associated with periodontal
adjusted correlation coefficient (r2) was high at 88%. inflammatory disease of non-third molars.
Angle Orthodontist, Vol 80, No 4, 2010

Original Article
Clinical photography vs digital video clips for the

assessment of smile esthetics
Brian J. Schabela; Tiziano Baccettib; Lorenzo Franchib; James A. McNamara, Jr.c
ABSTRACT
Objective: To test the null hypothesis that there is no clinically significant difference between the
post–orthodontic treatment images of smiles of subjects captured by clinical photography and the
smiles of the same subjects obtained from digital video clips.
Materials and Methods: Clinical photographs and digital video captures were obtained from 48
orthodontically treated patients. An updated version of the Smile MeshTM program was used to
quantify and compare smile characteristics obtained with the two methods. A paired-samples t-test
was performed to test for mean differences in Smile Mesh measurements generated from both
smile images. The relationship between the various Smile Mesh measurements obtained from both
smile images was examined by way of Pearson product-moment correlation.
Results: A significant difference was found between 7 of the 14 mean Smile Mesh measurements.
The absolute values of all these differences, however, were smaller than 1 mm and therefore were
not clinically significant. With the exception of lower lip to maxillary incisor, all measurements
showed a moderate to strong relation with each other (P values ranging from .47 to .82; P , .001).
Conclusions: The hypothesis cannot be rejected. A significant positive correlation was noted
between Smile Mesh measurements obtained from smiles captured by clinical photography and
those captured with digital video clips. This supports the conclusion that a standard digital
photograph appears to be a valid tool for analysis of the posttreatment smile. (Angle Orthod.
2010;80:678–684.)
KEY WORDS: Smile; Digital photograph; Video clip; Orthodontic treatment; Esthetics
INTRODUCTION been prolific in recent years with articles that have

examined various aspects of dentofacial esthetics.
In an attempt to meet the ever-increasing esthetic
Unfortunately, these reports often are contradictory
demands of patients, orthodontic researchers have
and misleading, in part because of the subjective
nature of beauty and the lack of a standardized scale
Graduate Orthodontic Program, The University of Michigan,
a
by which to measure it. In addition, the reliability of
Ann Arbor, Mich. Private practice, Santa Cruz, Calif. static photographs for evaluating the smile has been
b
Assistant Professor, Department of Orthodontics, The
University of Florence, Florence, Italy; Thomas M. Graber
questioned, and digital videography has been advo-
Visiting Scholar, Department of Orthodontics and Pediatric cated for use in capturing the dynamic nature of
Dentistry, School of Dentistry, The University of Michigan, Ann facial animation with special emphasis on the
Arbor, Mich. smile.1,2
c
Thomas M. and Doris Graber Endowed Professor of Ackerman et al.1 among others have defined two
Dentistry, Department of Orthodontics and Pediatric Dentistry,
main types of smiles: social smiles and enjoyment
School of Dentistry; Professor of Cell and Developmental
Biology, School of Medicine; and Research Professor, Center smiles. A social smile is ‘‘the voluntary smile a person
for Human Growth and Development, The University of uses in social settings or when posing for a photo-
Michigan, Ann Arbor, Mich. Private practice of orthodontics, graph.’’ The social smile is ‘‘posed,’’ which means that
Ann Arbor, Mich. it is not elicited or accompanied by emotion. This type
Corresponding author: Dr Tiziano Baccetti, Department of
of smile can be sustained as a static facial expression
Orthodontics, Università degli Studi di Firenze, Via del Ponte di
Mezzo, 46-48, 50127, Firenze, Italy and does not appear strained.2 On the other hand,
(e-mail: t.baccetti@odonto.unifi.it) enjoyment smiles are involuntary and are elicited by
Accepted: June 2007. Submitted: May 2007. laughter. The enjoyment smile is unposed and reflects
G 2010 by The EH Angle Education and Research Foundation, the emotion that one is experiencing at that moment.
Inc. This smile appears strained because the mouth bursts
Angle Orthodontist, Vol 80, No 4, 2010 678 DOI: 10.2319/052207-243.1

PHOTOGRAPHS VS VIDEO CLIPS OF THE SMILE 679
forward to reveal the maximal expansion of the lips, from digital video clips. These smiles were quantified
but it cannot be sustained. with the Smile Mesh program to determine whether
The unstrained social smile has been referred to as these two methods of smile capture differed signifi-
a reliable reference for measurement and character- cantly.
ization of the smile.3 Orthodontic records play an
essential role in capturing the unstrained social smile
MATERIALS AND METHODS
to be used for objective analysis. These records must
allow us to observe each patient frontally, vertically, Patient Selection
obliquely, and from profile, both statically and dynam-
Subjects enrolled in this study were recruited from
ically, to obtain a true smile representation.4,5
the University of Michigan Graduate Orthodontic Clinic
Static records used to capture the smile include
during a routine posttreatment appointment (ie, final
study models, radiographs, and film or digital photo-
records or retention check). Potential subjects were
graphs.4 The American Academy of Cosmetic Dentist-
given a brief introduction to the study and were asked if
ry Photographic Accreditation Review in 1995 recom-
they would be willing to participate. None of the
mended that facial photographs for esthetic treatment
planning should include full face smiling, full face with subjects received compensation for their participation.
lips relaxed, profile full smile, and right and left lateral Each adult subject (ie, 18 years of age or older)
views of full smile.4 It is interesting to note that this reviewed and signed a consent form created in
proposed sequence is advocated for appropriate accordance with the rules and regulations of the
visualization of even a single restorative unit (tooth), University of Michigan Health Sciences Institutional
yet the universal orthodontic standard for facial images Review Board. Each subject younger than 18 years of
includes frontal at rest, frontal smile,6 and profile at age reviewed and signed a child’s assent form, and a
rest. legal guardian reviewed and signed a consent form, in
Digital videography has become an adjunct tool for accordance with the Institutional Review Board. Each
orthodontic and orthognathic surgery evaluation.3,5,7 subject also reviewed and completed a consent form
Video clips taken before, during, and after treatment created by the University of Michigan in accordance
enable the clinician to observe the dynamic display with the Health Insurance Portability and Accountabil-
zone in the frontal view during facial animation; such ity Act for the use and disclosure of protected health
clips can be used as a means of comparison to assess information.
the effects of treatment and facial change over time. In To be included in the study, subjects had to present
addition to diagnostic information acquired from with the following characteristics: (1) age ranging from
dynamic visualization of the smile, video imaging has 12 to 20 years; (2) white ancestry; (3) orthodontic
the potential to affect communication at consultations treatment completed within the last 6 months; (4)
and at staff meetings, as well as interactions with other absence of missing or malformed teeth; and (5) a
offices, and in other areas not yet realized.7 complete set of diagnostic posttreatment records,
Tarantili et al.8 have described a progression of the including intraoral/extraoral photographic series and a
smile using digital video that consists of an initial attack good quality video clip of the smile. The protocol
period, a sustaining period, and a fade-out or decay proposed for the study required that 48 subjects be
period. If a clinical photograph is taken during the recruited to satisfy the design of the Q-sort method. A
attack or the decay phase, the resulting smile will not test was performed to determine the power of this
be a reliable reference. For this reason, it is postulated sample size with respect to correlation tests (Type I
that video may have a distinct advantage over clinical error 5 .05). For a bivariate normal distribution and a
photographs for accurately capturing a true represen- sample size of 48, a test of H0:P 5 0 (ie, the correlation
tation of the smile.3,8 coefficient under the null hypothesis) was found to
To quantify the reliability and reproducibility of the have a power of 0.80 to detect a linear correlation of r
posed smile, Ackerman et al.1 developed the Smile 5 0.38. Thus, the default sample size for the Q-sort
MeshTM (TDG Computing, Philadelphia, Pa) program. procedure was deemed adequate for purposes of
They reported high interrater and intrarater reliability of testing for correlation.
the Smile Mesh program and a high correlation
coefficient (r 5 0.78 to 0.99) between repeated
Image Capture
measures. They also found smiles in their study to
be reproducible. Clinical photography. The extraoral photographic
The aim of the present study was to compare the series included photographs of the subject in repose,
smiles of subjects after orthodontic treatment when during smiling, and in profile. For the purpose of the
captured by clinical photography vs smiles obtained current study, only the extraoral smiling photographs

680 SCHABEL, BACCETTI, FRANCHI, MCNAMARA
were used. A CanonH EF 35 mm SLR camera

(Canon U.S.A., Inc., Lake Success, NY) was mount-
ed to a frame set at a fixed distance of 36 inches
between the lens and the subject. The camera was
connected to a two-strobe lighting source that
illuminated the subject indirectly from a flash that
reflected off of a photographic umbrella. All photo-
graphs were taken by one of two dental school staff
photographers.
Before taking the smiling image, the photographer
instructed the subject to ‘‘smile.’’ The reproducibility of
the posed smile derived from the static photograph has
been demonstrated by Ackerman et al.1 Each image Figure 1. A standardized smile image obtained using the 30 3
was captured on KodakH EV-100 slide film (Eastman 50 template.
Kodak Company, Rochester, NY). The film was
developed, and the 20 3 20 slides were used. The Image Editing
slides were scanned using the NikonH Super Coolscan
4000 ED (Nikon Inc., Melville, NY) and were imported A 30 3 50 template was created to standardize the
directly into a commercially available image editing size and location of each image. Images were opened
software program (AdobeH PhotoshopH 7.0, Microsoft in PhotoshopH (Microsoft Corporation), and the tem-
Corporation, Redmond, Wash). Each slide was plate was superimposed on top of the image. Smile
scanned at maximum dpi (dots-per-inch) to enhance images were enlarged until the outer commissures of
image quality. the lips matched the vertical tickmarks inset three-
Digital videography. A digital video camera was quarters of an inch from the border of the template.
used to record the dynamic range of each subject’s The smiling images then were positioned so that the
smile, with slight modifications to the protocol report- maxillary incisal edges coincided with the horizontal
ed by Ackerman and Ackerman.3 To standardize the line of the template (Figure 1).
technique, a PanasonicH PV-GS200 digital video After the images were enlarged and positioned
camera (Panasonic, Secaucus, NJ) was used in the correctly, the portion of the image outside of the
same location under standard fluorescent lighting. template was cropped. Resulting images were edited
The camcorder was mounted on an adjustable further in Photoshop by using the healing brush tool to
microphone stand and was set at a fixed distance of remove blemishes, skin irregularities, and other
60 inches from the subject. Each subject was seated extraneous marks that could influence the rater when
and had his or her head positioned such that an evaluating the image. Images were labeled with a four-
imaginary line between the top of the ear and the digit number unique to each subject that was obtained
midpoint between the upper eyelash and eyebrow from a random number generator. Following the
paralleled the floor. The video camera was adjusted number, photos obtained from still photography were
vertically to be directly in line with the subject’s mouth, denoted with a ‘‘p’’ and photos obtained from digital
and the zoom feature was used to focus only on the video clips were denoted with a ‘‘v.’’ Once the editing
mouth and adjacent soft tissues to protect the was complete, each image was compressed to
anonymity of the subject. Each video clip was approximately 150 kb and was saved as a JPEG file.
obtained by the senior author.
Before the video clip was recorded, subjects were Video Editing
given the following instructions: Raw digital video clips of each subject were
1. You will be asked to smile and then relax three transferred to a computer using a commercially
separate times. available video editing software package (AdobeH
2. When you are asked to relax, please touch your lips PremiereH 6.0, Microsoft Corporation). This program
lightly together. allowed the streaming video to be converted into
individual photographic frames at the rate of approx-
3. When you are asked to smile, please smile until you
are told to relax again. imately 30 frames per second. Thus, a 10 second
video resulted in roughly 300 individual frames. The
Once the instructions were understood, the record- frame representing the subject’s posed unstrained
ing began. Each video clip lasted approximately 10 to social smile was selected, as advocated by Ackerman
15 seconds. et al.1,3 This frame, identified by the examiner as the

Figure 2. The Smile Mesh program used to measure various lip-tooth relationships associated with anterior tooth display.
‘‘held smile,’’ was one of 15 consecutive frames in ity performed on the data revealed that these variables
which the smile did not change. This unedited image were distributed normally. Therefore, parametric sta-
was saved as a JPEG file. tistics were used for inferential tests.
To test the hypothesis that an individual’s smile
Smile Mesh Assessment captured by clinical photography is the same as that
obtained from a digital video clip, a paired-samples t-
An updated version of the Smile Mesh program was test was performed to test for mean differences in
used in the current study to quantify and compare the Smile Mesh measurements generated from both smile
characteristics of anterior tooth display found in images. The relationship between the various Smile
‘‘attractive’’ vs ‘‘unattractive’’ smiles. Edited smile Mesh measurements obtained from smiles captured
images captured by clinical photography and obtained by clinical photography and from smiles obtained from
from digital clips of each of the 48 subjects used in this digital video clips was examined by way of Pearson
study were scanned into the Smile Mesh program. The product-moment correlation. The correlation coeffi-
height and width of the right maxillary central incisor for cient estimated the strength of the relationship
each corresponding image were entered into the between these two methods of smile capture.
program before starting. Two adjustable vertical lines, The type I error rate for all statistical tests was set at
superimposed on the smile image, were moved to .05. All statistical tests were performed with the aid of a
correspond with the mesial and distal border of the statistical software program (Statistical Package for
right central incisor. This enabled a computer-gener- the Social Sciences for Windows, version 12.0,
ated algorithm to calibrate the smile measurements to Chicago, Ill).
actual life size.3,5 The Smile Mesh consisted of an
adjustable grid system that comprised seven vertical RESULTS
lines and five horizontal lines superimposed on the
smile image. These grid lines were adjusted to Standard descriptive statistics were calculated for
correspond with specific hard and soft tissue land- Smile Mesh measurements taken from smile images
marks (Figure 2). The Smile Mesh then generated 15 obtained from clinical photographs and digital video
lip-tooth characteristics associated with anterior tooth clips. The significance levels (P values) of the paired
differences between all measurements are summa-
display (Table 1).
rized in Table 2. A significant difference was found
between 7 of the 14 mean Smile Mesh measurements.
Statistical Analysis
Pearson correlation coefficients were calculated to
Standard descriptive statistics (means, standard examine the relationship between Smile Mesh mea-
deviations, and ranges) were calculated for the Smile surements of individual subjects obtained by the two
Mesh measurements. A Shapiro-Wilks test for normal- methods of smile capture (Table 3). Other than lower

Table 1. Characteristics of Anterior Tooth Display Obtained from the Smile Mesh Program
Smile Attribute Description
Maximum incisor exposure Amount of vertical display of the maxillary central incisors
Upper lip drape Amount of vertical coverage of the maxillary central incisors by the upper lip (or amount of gingival display)
Lower lip to upper incisor Vertical distance from the incisal edge of the maxillary right central incisor to the deepest midline point on
the superior margin of the lower lip
Interlabial gap Distance between the most inferior portion of the tubercle of the upper lip and the deepest midline point on
the superior margin of the lower lip
Visible posterior teeth width Distance from the most lateral aspect of the most visible maxillary posterior tooth on the right and left sides
Smile width Distance from the right outer commissure to the left outer commissure
Smile index Ratio of smile width divided by interlabial gap
Commissure corridor left Horizontal distance from the left inner commissure to the left outer commissure
Commissure corridor right Horizontal distance from the right inner commissure to the right outer commissure
Buccal corridor left Horizontal distance from the most lateral aspect of the left most posterior visible tooth to the left inner
commissure
Buccal corridor right Horizontal distance from the most lateral aspect of the right most posterior visible tooth to the right inner
commissure
Buccal corridor ratio Distance between the most visible maxillary right and left teeth/Distance between the right and left inner
commissures
Smile arc Curvature of the incisal edges of the maxillary incisors, canines, and first premolar relative to the curvature
of the lower lip
Upper lip thickness Vertical distance from the most superior margin of the upper lip to the most inferior portion of the tubercle
of the upper lip
Lower lip thickness Vertical distance from the deepest midline portion of the superior margin of the lower lip to the most
inferior portion of the lower lip
lip to maxillary incisor, all measurements showed a fundamental question arises: Are standard static
moderate to strong relation with each other (P values records obtained routinely by orthodontists capable
ranging from .47 to .82; P , .001). of capturing the smile accurately?
Ackerman et al. 1 introduced the Smile Mesh
program to quantify characteristics of anterior tooth
DISCUSSION
display from photographs. They reported that this
The aim of the present study that focused on the morphometric tool could measure lip-tooth relation-
esthetics of the smile was to evaluate the relationship ships of the posed social smile accurately and reliably
between smiles captured by clinical photography and in a clinical setting. The Smile Mesh program was used
smile images obtained from digital video clips. Be- in the present study to quantify and compare 14
cause esthetics concerns have become more critical in characteristics of smiles captured by clinical photog-
orthodontic diagnosis and treatment planning, a raphy and digital videography.
Table 2. Descriptive Statistics and Paired-Samples t-Test of Smile Mesh Measurements Obtained from Images of Smiles Captured by Clinical
Photographs and Digital Video Clips
Smiles Captured by Clinical Photography Smiles Obtained from Digital Video Clips
Smile Mesh Measurements Mean SD Mean SD P Value
Maximum incisor exposure, mm 8.5 1.5 8.9 1.3 .02*
Upper lip drape, mm 0.9 1.9 0.1 2.3 ,.01**
Lower lip to maxillary incisor, mm 3.0 1.9 2.8 1.5 .60
Interlabial gap, mm 11.9 2.9 12.5 3.2 .12
Visible posterior teeth width, mm 48.5 4.2 47.7 4.4 .89
Smile width, mm 59.0 5.0 59.1 5.3 .88
Smile index 5.3 1.6 5.0 1.1 .12
Commissure corridor left, mm 5.3 1.8 4.8 1.6 .06
Commissure corridor right, mm 5.0 1.6 4.6 1.7 .07
Buccal corridor left, mm 5.0 1.6 5.9 1.6 ,.01**
Buccal corridor right, mm 4.5 1.3 5.5 1.7 ,.01**
Buccal corridor ratio 0.8 0.1 0.8 0.1 ,.01**
Upper lip height, mm 7.4 1.8 8.0 1.8 ,.01**
Lower lip height, mm 10.1 1.8 11.1 1.4 ,.01**
* P , .05
* P , .01

Table 3. Correlation Between Smile Mesh Measurements Pearson product-moment correlation was used to
Captured by Clinical Photography and Smile Mesh Measurements examine the relationship between individual Smile
Obtained from Digital Video Clips
Mesh measurements among smiles captured by
Smile Mesh Measurements of Smiles photographs and digital video clips. Each Smile Mesh
Smile Mesh Measurements
Obtained from Digital Video Clips
of Smiles Captured by measurement of the 48 subjects was correlated
Clinical Photography Correlation, r P Value significantly (correlation coefficients ranging from
Maximum incisor exposure 0.78 ,.001 0.47 to 0.82; P , .001), with the exception of the
Upper lip drape 0.80 ,.001 measurement of lower lip to maxillary incisor (P , .01).
Lower lip to maxillary incisor 0.36 ,.01 Of particular interest, correlations between the statis-
Interlabial gap 0.56 ,.001
Visible posterior teeth width 0.71 ,.001
tically significant differences measured with the paired
Smile width 0.81 ,.001 samples t-test (other than those associated with the
Smile index 0.47 ,.001 buccal corridor) ranged from 0.74 to 0.82. The strength
Commissure corridor left 0.65 ,.001 of these correlation coefficients suggests that anterior
Commissure corridor right 0.60 ,.001
tooth display is similar in a smile captured by clinical
Buccal corridor left 0.64 ,.001
Buccal corridor right 0.51 ,.001 photography and a digital video clip.
Buccal corridor ratio 0.53 ,.001 As a technical aside, selecting the specific frame
Upper lip height 0.82 ,.001 that represented the posed social smile from the video
Lower lip height 0.74 ,.001 clip, as advocated by Ackerman and Ackerman,3
seemed as arbitrary as capturing the smile at a single
time point with clinical photography. As mentioned
A paired samples t-test was conducted to evaluate
previously, Tarantili et al.8 noted a progression of the
mean differences between Smile Mesh measurements
smile that consisted of an initial attack period, a
obtained from clinical photographs and digital video
sustaining period, and a fade-out or decay period, when
clips of the 48 participants. Significant differences (P ,
the smile is captured by digital video. This progression
.001) were found between 7 of the 14 mean Smile
also was observed in the present study; however, these
Mesh measurements. However, examination of the
differences were slight, especially when still images of
descriptive statistics, namely, the mean measurement
the smile captured at 30 frames per second were
values, revealed some interesting trends. Smiles
evaluated. Undeniably, error was associated with
obtained from digital video clips had larger mean
selecting the appropriate still frame that represented
Smile Mesh measurements with respect to three direct
the posed social smile; similarly, a photograph taken of
measurements of the buccal corridor (buccal corridor
the smile has error associated with it.
right, buccal corridor left, and buccal corridor ratio).
Results of the present investigation suggest that a
These three measurements could have varied be-
clinical photograph is adequate for analyzing the smile
cause of methodologic differences in smile capture (ie,
of subjects after orthodontic treatment. The accessi-
use of ambient lighting when obtaining smiles from
bility of digital photography, in particular, should allow
digital video clips, as opposed to use of a supplemental
us to capture the posed social smile more accurately
flash when capturing smiles with clinical photography)
and reliably because we have instant access to the
rather than anatomic differences in the smiles.
image. Regardless of whether static or dynamic
More to the point, capturing a smile with ambient
records are used to capture the smile, the resultant
light could have created an illusion of increased buccal
image is only as good as the clinician’s ability to
corridor space and decreased visible posterior teeth
capture it accurately.
width as seen in smiles obtained from digital video clips.
It should be noted that these results in no way
Other investigators have reported that the buccal
discount the use of digital video as a diagnostic tool for
corridor (which also affects the width of visible posterior
treatment planning. Streaming video allows the clini-
teeth) appears more pronounced when no supplemen-
cian to observe the dynamic character of the smile that
tal light is added, and that these dark spaces can be
cannot be seen with a static photograph. Reemphasis
eliminated simply by using a flash on the camera.2,9,10
on the clinical examination of the patient supplemented
An important consideration with regard to the
by static and dynamic records simply enhances our
remaining statistically significant paired Smile Mesh
ability to define specific esthetic goals before providing
measurements (eg, upper lip drape, upper lip height,
treatment.
lower lip height) is clinical significance. Mean differ-
ences of 1 mm or less generally are regarded as
CONCLUSIONS
clinically insignificant. Therefore, it should be pointed
out that none of these average measurements differed N A significant positive correlation was noted between
by more than 1 mm. Smile Mesh measurements obtained from smiles

captured by clinical photography and digital video Arbor, Mich: Monograph 40, Craniofacial Growth Series,
clips. Department of Orthodontics and Pediatric Dentistry and
Center for Human Growth and Development, The University
N Digital video clips offer a tremendous amount of of Michigan; 2003:195–203.
information for analyzing the dynamic character of 4. Sarver DM. The face as the determinant of treatment choice.
the smile, but a standard digital photograph allows In: McNamara JA Jr, ed. Frontiers of Dental and Facial
for immediate viewing, and is a valid tool for analysis Esthetics. Ann Arbor, Mich: Monograph 38, Craniofacial
of the posttreatment smile. Growth Series, Department of Orthodontics and Pediatric
Dentistry and Center for Human Growth and Development,
The University of Michigan; 2001:19–54.
ACKNOWLEDGMENTS 5. Sarver DM, Ackerman MB. Dynamic smile visualization and
Funds for this research were derived in part from the LeGro quantification: Part 1. Evolution of the concept and dynamic
Fund, as well as from sources made available through the records for smile capture. Am J Orthod Dentofacial Orthop.
Thomas M. and Doris Graber Endowed Professorship of the 2003;124:4–12.
University of Michigan. 6. Janzen EK. A balanced smile—a most important treatment
objective. Am J Orthod. 1977;72:359–372.
7. Sarver DM. Videoimaging: the pros and cons. Angle Orthod.
REFERENCES
1993;63:167–170.
1. Ackerman JL, Ackerman MB, Brensinger CM, Landis JR. A 8. Tarantili VV, Halazonetis DJ, Spyropoulous MN. The
morphometric analysis of the posed smile. Clin Orthod Res. spontaneous smile in dynamic motion. Am J Orthod Den-
1998;1:2–11. tofacial Orthop. 2005;128:8–15.
2. Ackerman JL, Ackerman MB. Smile analysis and design in 9. Ackerman MB, Bresinger C, Landis JR. An evaluation of
the digital era. J Clin Orthod. 2002;36:221–236. dynamic lip-tooth characteristics during speech and smile in
3. Ackerman JL, Ackerman MB. Digital video as a clinical tool adolescents. Angle Orthod. 2004;74:43–50.
in orthodontics: dynamic smile analysis and design in 10. Dougherty HL. Clubs, quips, phrases, and hype: musing for
diagnosis and treatment planning. In: McNamara JA Jr, the new millennium. Am J Orthod Dentofacial Orthop. 2000;
ed. Information Technology and Orthodontic Treatment. Ann 117:586–588.

Case Report
Absolute anchorage with universal T-loop mechanics for severe deepbite

and maxillary anterior protrusion and its 10-year stability
Yoon Jeong Choia; Chooryung Judi Chungb; Kwangchul Choyc; Kyung-Ho Kimd
ABSTRACT An adult patient with severe maxillary protrusion and deepbite who was congenitally
missing two mandibular incisors was treated successfully by maximum retraction of the maxillary
anterior teeth after extraction of the maxillary first premolars using a moment differential between
the anterior and posterior segments created by a universal T-loop. Anterior teeth were moved with
controlled tipping, and little anchorage loss of the posterior segments was experienced using the
universal T-loop spring. Reduction of overbite was performed by absolute intrusion of both
maxillary and mandibular anterior teeth. With retraction of the maxillary anterior teeth and
recontouring of the mandibular canines, proper overjet and overbite were achieved. This report
shows the 10-year stability of the case treated with the universal T-loop for the first time. (Angle
Orthod. 2010;80:771–782.)
KEY WORDS: T-loop mechanics; Deepbite; Protrusion; Stability; Absolute anchorage; Congenital
missing
INTRODUCTION convex profile and also increase the relapse potential

for adults.3–8 Because the segmented arch can
A deep overbite can be corrected by intrusion of
minimize extrusion of the posterior teeth,8 it may be a
anterior teeth, extrusion of posterior teeth, or a
more stable approach for deepbite correction than a
combination of both.1 When a continuous archwire is
continuous archwire technique when arch leveling by
used to correct deepbite, extrusion in the molar area
incisor intrusion is indicated.2
with subsequent posterior rotation of the mandible may
For the correction of protruded anterior teeth, premo-
occur.2 Clockwise rotation of the mandible can provide
lar extraction followed by retraction of the anterior teeth
relative improvement of a deep overbite problem in the
is essential, and anchorage control is important. To
anterior region; however, it may worsen the Class II
control anchorage loss, headgears, intermaxillary elas-
tics, transpalatal arches, tipback springs, and, more
a
Clinical and Research Assistant Professor, Department of
Orthodontics, Gangnam Severance Dental Hospital, College of recently, temporary anchorage devices (TADs) have
Dentistry, Institute of Craniofacial Deformity, Oral Science been used. In many cases, the patient’s compliance and
Research Center, Yonsei University, Seoul, South Korea. discomfort from appliance use may interfere with
b
Assistant Professor, Department of Orthodontics, Gangnam treatment effectiveness. Although TADs overcome
Severance Dental Hospital, College of Dentistry, Institute of
Craniofacial Deformity, Oral Science Research Center, Yonsei
these limitations and offer absolute anchorage,9–11 a
University, Seoul, South Korea. surgical procedure is unavoidable, and the location of
c
Private practice, Adjunctive Professor, College of Dentistry, the TAD is limited by the amount and quality of alveolar
Yonsei University, Seoul, South Korea. bone.12 Additionally, in cases of TAD failure, alternative
d
Professor and Department Chair, Department of Orthodon- treatment options are needed.
tics, Gangnam Severance Dental Hospital, College of Dentistry,
The T-loop has been suggested as a mechanism to
Institute of Craniofacial Deformity, Oral Science Research
Center, Yonsei University, Seoul, South Korea. control anchorage movement during extraction space
Corresponding author: Dr Kyung-Ho Kim, Professor and closure; the T-loop operates by producing differential
Department Chair, Department of Orthodontics, Gangnam moments between the anterior and posterior seg-
Severance Dental Hospital, College of Dentistry, Institute of ments.13–16 Desired tooth movement can be achieved
Craniofacial Deformity, Oral Science Research Center, Yonsei
by changing the angulation of the preactivation
University, 712 Eonjuro, Gangnam-Gu, Seoul, 135-720 South
Korea bends,13 by altering the dimensions of the spring,14,15,17
(e-mail: khkim@yuhs.ac) or by changing the position of the T-loop,18,19 with no
Accepted: September 2009. Submitted: May 2009. need for additional procedures such as TAD implan-
G 2010 by The EH Angle Education and Research Foundation, tation. The universal T-loop was developed for all
Inc. types of anchorage through a process by which the
DOI: 10.2319/052809-304.1 771 Angle Orthodontist, Vol 80, No 4, 2010

772 CHOI, CHUNG, CHOY, KIM
Figure 1. Pretreatment facial (A) and intraoral (B) photographs.
loop was located differently, with a uniform design. no significant medical or dental history. Pretreatment
This was a departure from the pattern in which the records showed a convex profile and mentalis strain
conventional T-loop had different shapes according to with protrusive upper anterior teeth (Figure 1). She
the intertube distance or type of anchorage.19 In this had an excessive overjet (10.0 mm) and overbite
report we present the case of an adult with maxillary (6.0 mm), with the two mandibular incisors congeni-
protrusion and deepbite who was congenitally missing tally missing. The mandibular incisors impinged on the
two mandibular incisors; successful correction was maxillary palatal gingiva because of a deep curve of
accomplished by the segmented arch technique using Spee (COS). The maxillary incisors were severely
the universal T-loop. We present this patient’s records proclined, and mild arch length discrepancies were
10 years after treatment to demonstrate the stability of present in both arches. The molars and canines were
the treatment. in a Class I relationship (Figure 2).
The panoramic radiograph showed that all teeth were
present except for two mandibular lateral incisors and
CASE REPORT
the mandibular left third molar (Figure 3). Cephalometric
A female, aged 19 years and 1 month, sought analysis showed a Class I skeletal pattern with a low
treatment for upper anterior dental protrusion. She had mandibular plane angle (SN to MP angle, 27.5u). The U1

TREATMENT OF PROTRUSION AND DEEPBITE WITH T-LOOP 773
Figure 2. Pretreatment dental casts. The mandibular incisors were impinged on the palatal gingiva because of the deep curve of Spee.
to SN angle of 126.5u reflected proclination of the was diagnosed as skeletal Class I deepbite with upper
maxillary incisors and resulted in an acute nasolabial anterior protrusion and congenitally missing two man-
angle of 88.5u. The upper lip was 6.5 mm and the lower dibular lateral incisors.
lip was 3.8 mm in front of the Sn-Pog9 plane. Vertical
exposure of the maxillary incisors at rest was 4 mm Treatment Plans
(Table 1; Figure 4). Based on these findings, the patient The treatment objectives for this patient were to
achieve a normal soft tissue profile and to obtain ideal
overjet and overbite. By correcting the protrusion of the
maxillary anterior teeth, the lip incompetence would be
eliminated and the acute nasolabial angle would be
improved. This would lead to better facial esthetics and
normal incisal and canine guidance.
The treatment plan to achieve these treatment
objectives was established as follows: the maxillary
anterior teeth would be retracted after extraction of the
maxillary first premolars. The mandibular lateral
incisors and canines would be replaced with the
canines and premolars, respectively.
A maximum anchorage was essential to maintain a
Class I molar relationship throughout the entire period
Figure 3. Pretreatment panoramic radiograph. of treatment. To replace lateral incisors and canines

Table 1. Cephalometric Measurements

Retention
Measurements Pretreatment Posttreatment (10 y)
SNA, u 83.5 82.0 82.5
SNB, u 81.5 81.0 81.0
ANB, u 2.0 1.0 1.5
Wits appraisal, mm 3.5 2.5 2.5
SN-GoGn, u 27.5 27.5 27.5
Anterior facial height, Figure 5. The 0.017 3 0.025–inch titanium molybdenum alloy (TMA)
mm 117.9 118.7 120.9 T-loop spring19 used in this case.
Posterior facial height,
mm 84.9 84.7 86.6
with canines and premolars, reshaping and reduction
U1-SN, u 126.5 101.5 102.0
IMPA, u 94.5 91.5 91.5 of tooth size would need to be performed.
U1 to Facial plane, mm 15.3 3.9 4.1
L1 to Facial plane, mm 4.2 2.2 1.9 Treatment Progress
Upper lip to Sn-Pog9
line, mm 6.5 2.4 2.7 The maxillary first premolars were extracted. All
Lower lip to Sn-Pog9 teeth were sequentially bonded or banded with 0.018
line, mm 3.8 1.6 1.5 3 0.025–inch preadjusted edgewise brackets (Roth-
Nasolabial angle, u 88.5 102.0 99.5
type prescription). To gain space for anterior align-
Figure 4. Pretreatment cephalometric radiograph (A) and tracing (B).
Figure 6. Progress intraoral photographs: (A) Four months later, the maxillary anterior teeth were retracted with a T-loop and the mandibular
anterior teeth were intruded using one-piece intrusion archwire. (B) Forces (vertical arrows) and moment (round arrow) produced by a one-piece
intrusion arch in the mandible. (C) Nine months later, the mandibular anterior segment was leveled, but labioversion was observed. (D) Fourteen
months after the beginning of treatment, the extraction space was almost closed and an L-loop was inserted in the mandibular arch for
torque control.


Table 2. Forces and Moments According to Activation of 0.17 3 0.025 Inch Titanium Molybdenum Alloy Standard T-Loop Spring (B/L 5 0.63)19
D, Ma, Mb, Fh, Fv, Ma/Fh, Mb/Fh, F/D,
mm g-mm g-mm g g mm mm (Ma2Mb)/Fh, g/mm
0.0 1283.5 1288.7 17.1 24.7
0.5 1329.5 1448.0 49.5 211.9 26.8 29.2 22.4 64.8
1.0 1377.2 1561.3 77.6 216.4 17.7 20.1 22.4 56.2
1.5 1420.9 1672.5 105.4 220.6 13.5 15.9 22.4 55.6
2.0 1464.8 1780.9 133.4 225.2 11.0 13.4 22.4 55.9
2.5 1509.2 1884.2 160.9 228.7 9.4 11.7 22.3 55.0
3.0 1563.2 1983.9 188.9 232.2 8.3 10.5 22.2 56.0
3.5 1611.4 2076.1 216.4 235.2 7.4 9.6 22.1 55.0
4.0 1669.1 2170.4 244.1 238.5 6.8 8.9 22.1 55.4
4.5 1716.8 2256.6 272.4 241.6 6.3 8.3 22.0 56.6
5.0 1766.4 2343.2 300.7 244.4 5.9 7.8 21.9 56.6
5.5 1791.7 2425.4 329.6 246.7 5.4 7.4 21.9 57.8
6.0 1810.6 2505.8 358.2 248.1 5.1 7.0 21.9 57.2
ment, the maxillary canines were retracted partially tient was instructed to wear a short highpull headgear
with a transpalatal arch and tipback spring to reinforce at night to counteract the excessive moment of the
the anchorage. The mandibular teeth were aligned posterior segment. Since the cross-tube detached
sectionally for segmental arch leveling. from the stabilizing wire of the anterior segment about
Four months later, the maxillary anterior teeth were 1 month later, the canine brackets were exchanged for
retracted with an A-type segmented T-loop (0.017 3 Burstone’s canine brackets, and new universal T-loop
0.025–inch titanium molybdenum alloy), which was springs were fabricated.
Burstone’s universal T-loop (Figure 5). The loop was The mandibular anterior and posterior segments
located approximately 3 mm posteriorly in relation to were stabilized separately, and a one-piece intrusion
the interbracket distance, keeping a B/L ratio of 0.63 to archwire was attached to the mandibular first molars to
obtain a moment differential.18,19 The anterior and intrude the anterior teeth (Figure 6B). Nine months
posterior segments were stabilized with 0.017 3 later, the anterior segment was intruded and posi-
0.025–inch stainless-steel wire (Figure 6A). The pa- tioned lower than the posterior segment, but labiover-
Figure 7. Posttreatment dental casts.

Figure 8. Posttreatment facial (A) and intraoral (B) photographs.
sion of the anterior segment was observed (Fig-

ure 6C). To correct the axis of the anterior segment,
an 0.016 3 0.022–inch stainless-steel wire with a 7u
lingual crown torque in the anterior region and L-loops
between canine, first premolar, and second premolar
was inserted to the lower arch (Figure 6D). The
mandibular canines were interproximally recontoured
several times throughout the treatment.
The T-loop was activated 3 mm initially (M/F alpha 5
8.3, M/F beta 5 10.5) and reactivated when a space of
1.5 mm was closed (M/F alpha 5 13.5, M/F beta 5
15.9), as seen in Table 2,19 and this procedure was
repeated a couple of times until the extraction space was
closed (Figure 6D). After 10 months of retraction of the
maxillary anterior segment, the T-loop was replaced with Figure 9. Posttreatment panoramic radiograph.

Figure 10. Posttreatment cephalometric radiograph (A) and superimposed tracings (B).
a root spring of 0.016 3 0.022–inch stainless-steel wire ular incisors were intruded 3.0 mm and 1.5 mm,
for the control of angulation of root axis. The canine and respectively, and retracted 11.4 mm and 2.3 mm,
second premolar were tied tightly to prevent opening of respectively. Consequently, the overjet was reduced
the extraction space during the root movement. As from 10.0 mm to 1.5 mm, and the overbite was
canting of the maxillary anterior teeth was noticed, a reduced from 6.0 mm to 2.0 mm. The U1 to SN plane
three-piece intrusion wire was constructed, and the was decreased from 126.5u to 101.5u (Table 1). The
intrusion force applied only to the right side. movement of the maxillary incisors contributed to
After 26 months of treatment, proper overjet and correction of the soft tissue profile and mentalis strain.
overbite were achieved, and the fixed appliance was Even though the anterior teeth were fully retracted in
removed. For retention, fixed lingual retainers were the maxilla, there was little movement of the molars
bonded from second premolar to second premolar in anteriorly (Figure 10). After 10 years of retention,
the maxilla and from first premolar to first premolar in posttreatment stability of the occlusion was observed
the mandible (Figure 7). The patient was requested to (Figure 11), and the lateral cephalometric radiograph
wear removable circumferential retainers 24 hours a and superimposition showed no marked skeletal or
day for the first 6 months and thereafter for 18 months dental changes (Figures 10 and 12B). A minor space
at night only. between the upper central incisors was observed even
though a lingual fixed retainer had been bonded
Treatment Results throughout the retention period (Figure 13).
The posttreatment facial photographs showed
DISCUSSION
marked improvement of the facial profile, and the
patient’s smile improved. Protrusion of the maxillary The universal T-loop has been recognized as an
anterior teeth was corrected, and a Class I molar effective means to achieve desired tooth movement by
relationship was achieved with proper overjet and differential moments between the anterior and poste-
overbite. As a result of the reshaping of the mandibular rior segments.13–20 Although TADs have been widely
canines to incisors, normal incisal and canine guid- used for anchorage reinforcement, there are unpre-
ance could be established (Figure 8). dictable factors such as anatomical limitations and the
The posttreatment panoramic radiograph confirmed possibility of failure. However, precise control of tooth
root paralleling (Figure 9). The maxillary and mandib- movement is possible in a predictable manner with the

Figure 11. Facial (A) and intraoral (B) photographs 10 years after treatment.
T-loop spring by changing the dimensions,14,15,17 Thus, we used a universal T-loop spring (Figure 5)
shape,13 or position of the T-loop.18,19 According to and positioned it approximately 3 mm posteriorly (keep-
Kuhlberg and Burstone,18 eccentric positioning of a T- ing a B/L ratio of 0.63) to retract and intrude the upper
loop with a symmetric shape could be used to achieve anterior teeth and to prevent posterior anchorage loss. A
a moment differential and maintenance of the moment posteriorly positioned T-loop spring produces a greater
differential as the spaces close, improving anchorage moment to posterior teeth than to anterior teeth, and it
control and force system predictability. As it was also causes extrusive force on the posterior teeth and
modified to increase the preactivation moment accord- intrusive force on the anterior teeth.18,19 The T-loop
ing to the increase of intertube distance by adding spring in this case was assumed to produce a M/F alpha
continuous curvature to the horizontal leg, all kinds of of 8.3 and a M/F beta of 10.5 when the spring was
space closure are possible with only one spring, activated 3 mm, and the M/F alpha and beta would
regardless of intertube distance.16,19 increase with deactivation (Table 2).19 During deactiva-

To correct a deep overbite in this case, intrusion of the

incisors was indicated because of the significant
exposure of maxillary incisors, deep COS, and no
remaining growth. Intrusion of the incisors can be
performed by the segmented arch or the continuous
archwire technique.2,21–25 One of the differences between
the two methods seems to be whether the extrusion of
posterior teeth is allowed or not. In terms of the long-term
stability of deepbite treatment, extrusion of the premolar
teeth by a continuous archwire would increase a patient’s
lower facial height, and this change would tend to relapse
following treatment unless suitable growth occurred.3–7 In
contrast, other reports26,27 have shown that vertical
extrusion of premolars and molars gives rise to a stable
change and that lower incisor intrusion frequently
relapses to produce an increase in overbite. Considering
these conflicting opinions, incisor intrusion seems to be a
more stable approach if intruded incisors can be
maintained. Therefore, both maxillary and mandibular
incisors were intruded without extrusion of posterior teeth
by the segmented arch technique, and fixed lingual
retainers were bonded, including both sides of the
premolars, to prevent relapse of the intruded incisors.
In many long-term studies of deepbite correction,
several factors contributing to stability have been
considered. The long retention period and the retainer
design are of importance, while the relative stability of
overjet plays a role in overbite stability as well.22
Protrusion of the mandibular incisors during orthodontic
correction of overbite28 and increased curve relapse21
can increase the possibility of overbite relapse. In terms
of COS relapse, patients who were completely leveled
Figure 12. Panoramic (A) and lateral cephalometric (B) radiographs
posttreatment showed a significantly lower incidence of
10 years after treatment.
COS relapse than did those who were not.24
In our case, the amount of overbite relapse was
tion of the spring, however, the whole system of force 0.5 mm, which is similar to values quoted in previous
can change by the movement of the teeth, requiring a reports.23,29,30 The long retention period by a fixed
self-corrective loop with proper compensation,17 or the lingual retainer may have played an important role in
spring must be readjusted every month.20 Because we the stability of the overbite. The mandibular incisors
used Burstone’s universal T-loop, the spring was were protruded right after intrusion, but this was
readjusted every visit and reactivated when the space corrected by stripping of the mandibular canines and
of 1.5 mm was closed. a L-loop wire with lingual crown torque. Therefore, the
The patient was instructed to wear a short highpull COS could be flattened completely without protrusion
headgear at night to cancel out the excessive moment of the mandibular incisors, which may also have
exerted on the posterior teeth by production of distal contributed to the long-term stability. In addition, the
root tipping moment (Figure 14) and consequently size of the mandibular canines was decreased 1.4 mm
preventing mesial root movement of the posterior by stripping, and the sum of the incisor ratio was
teeth. The vertical (extrusive) force on the beta changed from 4:3.26 to 4:2.92. Consequently, proper
position seemed to be balanced by bite force as it is overjet and overbite and a Class I molar relationship
known,19 and the vertical (intrusion) force on alpha could be achieved and maintained. The relationship of
intruded the anterior teeth. From the superimposition appropriate overjet and overbite may also have had a
of lateral cephalograms (Figure 10B), it is confirmed causal effect on posttreatment stability.
that the position of the maxillary first molar was Posttreatment results showed good stability after 10
maintained while the anterior teeth were retracted years; prevention of extrusion of the posterior teeth by
11.4 mm and intruded 3.0 mm. the segmented arch technique and the long retention

Figure 13. Dental casts 10 years after treatment.
Figure 14. Forces and moments produced by a T-loop spring and a short highpull headgear. A posteriorly positioned T-loop spring produces a
greater moment (round arrows) to posterior teeth than to anterior teeth, and it also causes extrusive force (vertical arrows) on posterior teeth and
intrusive force on anterior teeth. A short highpull headgear was intended to counteract the excessive moment exerted on the posterior teeth by
delivering a distal force (dashed arrow) above the center of resistance.
period with fixed lingual retainers seemed to contribute ACKNOWLEDGMENT

to this stability. We thank Dr K. Y. Jung for his help with the successful
treatment.
CONCLUSION
REFERENCES
N A patient with deepbite and protrusion of the
maxillary anterior teeth was treated successfully by 1. Horiuchi Y, Horiuchi M, Soma K. Treatment of severe Class II
Division 1 deep overbite malocclusion without extractions in an
maximum retraction and intrusion using a segmented adult. Am J Orthod Dentofacial Orthop. 2008;133:S121–S129.
arch technique with universal T-loop spring, and the 2. Weiland FJ, Bantleon HP, Droschl H. Evaluation of
treatment result was stable after 10 years. continuous arch and segmented arch leveling techniques

in adult patients—a clinical study. Am J Orthod Dentofacial 17. Viecilli RF. Self-corrective T-loop design for differential
Orthop. 1996;110:647–652. space closure. Am J Orthod Dentofacial Orthop. 2006;129:
3. Bell WH, Jacobs JD, Legan HL. Treatment of Class II deep 48–53.
bite by orthodontic and surgical means. Am J Orthod. 1984; 18. Kuhlberg AJ, Burstone CJ. T-loop position and anchorage
85:1–20. control. Am J Orthod Dentofacial Orthop. 1997;112:12–18.
4. Wylie WL. Overbite and vertical facial dimensions in terms of 19. Burstone CJ, van Steenbergen E, Hanley KJ. Modern
muscle balance. Angle Orthod. 1944;14:13–17. Edgewise Mechanics and the Segmented Arch Technique.
5. Schudy FF. The association of anatomical entities as Glendora, Calif: Ormco; 1995.
applied to clinical orthodontics. Angle Orthod. 1966;36: 20. Martins RP, Buschang PH, Gandini LG Jr. Group A T-loop
190–203. for differential moment mechanics: an implant study.
6. McDowell EH, Baker IM. The skeletodental adaptations in Am J Orthod Dentofacial Orthop. 2009;135:182–189.
deep bite correction. Am J Orthod Dentofacial Orthop. 1991; 21. Shannon KR, Nanda RS. Changes in the curve of Spee with
100:370–375. treatment and at 2 years posttreatment. Am J Orthod
7. Bench RW, Gugino CF, Hilgers JJ. Bio-progressive therapy. Dentofacial Orthop. 2004;125:589–596.
Part 2. J Clin Orthod. 1977;11:661–682. 22. Al-Buraiki H, Sadowsky C, Schneider B. The effectiveness
8. Burstone CJ. Deep overbite correction by intrusion. and long-term stability of overbite correction with incisor
Am J Orthod. 1977;72:1–22. intrusion mechanics. Am J Orthod Dentofacial Orthop. 2005;
9. Kim SH, Lee KB, Chung KR, Nelson G, Kim TW. Severe 127:47–55.
bimaxillary protrusion with adult periodontitis treated by 23. Fidler BC, Artun J, Joondeph DR, Little RM. Long-term
corticotomy and compression osteogenesis. Korean J stability of Angle Class II, Division 1 malocclusions with
Orthod. 2009;39:54–65. successful occlusal results at end of active treatment.
10. Sugawara J. Orthodontic reduction of lower facial height in Am J Orthod Dentofacial Orthop. 1995;107:276–285.
open bite patients with skeletal anchorage system: beyond 24. Preston CB, Maggard MB, Lampasso J, Chalabi O. Long-
traditional orthodontics. World J Orthod. 2005;6(suppl): term effectiveness of the continuous and the sectional
24–26. archwire techniques in leveling the curve of Spee. Am J
11. Park YC, Choi YJ, Choi NC, Lee JS. Esthetic segmental Orthod Dentofacial Orthop. 2008;133:550–555.
retraction of maxillary anterior teeth with a palatal appliance 25. Dake ML, Sinclair PM. A comparison of the Ricketts and
and orthodontic mini-implants. Am J Orthod Dentofacial Tweed-type arch leveling techniques. Am J Orthod Dento-
Orthop. 2007;131:537–544. facial Orthop. 1989;95:72–78.
12. Lee KJ, Joo E, Kim KD, Lee JS, Park YC, Yu HS. Computed 26. Schudy FF. Cant of the occlusal plane and axial inclinations
tomographic analysis of tooth-bearing alveolar bone for of teeth. Angle Orthod. 1963;33:69–82.
orthodontic miniscrew placement. Am J Orthod Dentofacial 27. Schudy FF. Vertical growth versus anteroposterior growth
Orthop. 2009;135:486–494. as related to function and treatment. Angle Orthod. 1964;34:
13. Burstone CJ. The segmented arch approach to space 75–93.
closure. Am J Orthod. 1982;82:361–378. 28. Simons ME, Joondeph DR. Change in overbite: a ten-year
14. Burstone CJ, Koenig HA. Optimizing anterior and canine postretention study. Am J Orthod. 1973;64:349–367.
retraction. Am J Orthod. 1976;70:1–19. 29. Carcara S, Preston B, Jureyda O. The relationship between
15. Hoenigl KD, Freudenthaler J, Marcotte MR, Bantleon HP. the curve of Spee, relapse and the Alexander Discipline.
The centered T-loop—a new way of preactivation. Am J Semin Orthod. 2001;7:90–99.
Orthod Dentofacial Orthop. 1995;108:149–153. 30. Sadowsky C, Schneider BJ, BeGole EA, Tahir E. Long-term
16. Choy KC, Kim KH, Park YC. Factors affecting force system stability after orthodontic treatment: nonextraction with
of orthodontic loop spring. Korean J Orthod. 1999;29:511– prolonged retention. Am J Orthod Dentofacial Orthop.
519. 1994;106:243–249.

Original Article
Canine retraction rate with self-ligating brackets vs

conventional edgewise brackets
S. Jack Burrowa
ABSTRACT
Objective: To compare the rates of retraction down an archwire of maxillary canine teeth when
bracketed with a self-ligating bracket was used on one side and a conventional bracket on the
other.
Materials and Methods: In 43 patients requiring maxillary premolar extraction, a self-ligating
bracket (Damon3, SmartClip) was used on the maxillary canine on one side and a conventional
bracket (Victory Series) on the other. The teeth were retracted down a 0.018-inch stainless steel
archwire, using a medium Sentalloy retraction spring (150 g). The rates of retraction were analyzed
using a paired t-test.
Results: The mean movement per 28 days for the conventional bracket was 1.17 mm. For the
Damon bracket it was 0.9 mm and for the SmartClip bracket it was 1.10 mm. The differences
between the conventional and self-ligating brackets were statistically significant: paired t-test,
SmartClip, P , .0043; Damon3, P , .0001).
Conclusion: The retraction rate is faster with the conventional bracket, probably because of the
narrower bracket width of the self-ligating brackets. (Angle Orthod. 2010;80:626–633.)
KEY WORDS: Canine retraction; Self-ligating; Edgewise brackets
INTRODUCTION Friction is the result of chemical bonding between

surfaces. Surface area does not affect friction because
Orthodontists, when moving a tooth down a wire,
as surface area increases, the force per unit area
have to contend with biologic barriers and mechanical
decreases (not to be confused with traction where
phenomena. Three of the mechanical phenomena are surface area is important). On a microscopic scale,
currently of interest because of the marketing of self- even highly polished surfaces of orthodontic brackets
ligating brackets that are said to reduce resistance to and wires are irregular, and the true area of physical
sliding: friction, binding, and notching. These collec- contact is determined by asperities (Figure 1).2
tively determine the resistance to sliding a wire through In clinical orthodontics, we are dealing with a
a bracket or a bracket along a wire. quasistatic thermodynamic process, which means the
Friction is the resistance force between objects that movements are very slow and fairly close to static
oppose movement.1 It is always exerted in a direction equilibrium. Kusy and Whitley3 defined resistance to
opposing movement between the two surfaces. Fric- sliding (RS) as a combination of three components:
tion is not a fundamental force, it cannot be calculated friction (FR), binding (BI), and notching (NO):
from fundamental principles; it must be calculated
empirically. In contrast, moments or binding can be N FR, static or kinetic, is due to the contact forces
calculated using fundamental principles, eg, a binding between the wire, brackets, and ligatures;
force that equals the moment of a couple (MC) can be N BI is created as soon as the tooth begins to move
calculated using the equation MC 5 F/D. and the wire contacts the edge of the bracket; and
N NO occurs when permanent deformation of the wire
a
Adjunct Professor, Department of Orthodontics, University of occurs at the wire-bracket corner interface (Figure 2).
North Carolina School of Dentistry, Chapel Hill, N.C. Tooth movement stops as a notch in the wire catches
Corresponding author: Dr S. Jack Burrow, 2711 Randolph on the bracket or when the binding angle increases to
Road, Suite 600, Charlotte, NC 28207 a point (hz) that plastic deformation occurs. Movement
(e-mail: hbcteam@aol.com)
resumes only after the notch is released.
Accepted: November 2009. Submitted: June 2009.
G 2010 by The EH Angle Education and Research Foundation, Consequently, RS 5 FR + BI, or RS 5 NO because
Inc. sliding stops when NO begins. This equation can be

SELF-LIGATING VS CONVENTIONAL EDGEWISE BRACKETS 627
Figure 1. Wire-bracket interface. The surface is not smooth, but

irregular. The areas of contact between the wire surface and bracket
surface is determined by asperities.
applied to the passive and active stages of tooth

movement in a laboratory setting. The passive stage of
movement is when the contact angle (h) between the
archwire and bracket slot is less than the critical angle (hc)
(Figure 3A) before the wire touches the corner of the
bracket, and resistance to sliding is due only to friction.4
The passive stage really exists only in laboratory settings
because as soon as a tooth moves in response to a force
against its crown, it tips until the wire touches the corner of
a bracket. In a laboratory setting, the investigator can
stabilize the wire-bracket system in a fixed position, so the
bracket does not tip and the wire does not flex (Figure 3B);
ie, the wire-bracket system can be positioned so it does
not have any forces whatsoever on it as the wire passes
through the bracket. This never happens in clinical
orthodontics because the wire actually touches the
bracket as soon as tooth movement begins. The active
stage is defined as any angle above hc (Figure 3C).
Kusy reported the effect of binding and notching in
the late 1990s. Articolo and Kusy5 concluded that the
binding influence increased as the second-order
angulation increased, which is in agreement with Figure 2. Notching occurring when the wire catches due to
Nicolls and others.6–11 Articolo and Kusy further deformation of the wire, ie, contact of a notch in the wire against
reported that ‘‘RS becomes dependent on BI very the edge of the bracket: or, when binding reaches an angle in which
quickly, after the initiation of the active configuration.’’ plastic deformation occurs (h . hz). When binding becomes sufficient
to cause notching, sliding mechanics cease.
‘‘In fact BI was calculated to be at least 80% of RS at h
5 7u for all couples, and as much as 99% at h 5 13u for
one couple (SS-SC).’’ Friction was hardly an influence.
In 2002, Thorstenson and Kusy12,13 reported two
These laboratory reports do not support advertising
studies that investigated the effect of second-order
claims of faster tooth movement with self-ligating
angulation (binding) on self-ligating brackets. Both
brackets. The goal of this clinical study was to
studies basically had the same conclusion. ‘‘RS
compare the velocity of canine retraction with self-
increased proportionally with the second-order angula-
ligating brackets to conventional brackets.
tion’’12 and the binding was ‘‘independent of bracket
design.’’13 Figure 4 shows the results when only friction
is involved. Figure 5 shows the data for binding.
Thorstenson14 concluded that ‘‘binding does not appear A sample size of 43 patients was used in this
to be affected by ligation method,’’ ie, binding was the investigation (21 Damon3, 22 SmartClip, 43 conven-
same with conventional and self-ligating brackets. tional Victory Series).

628 BURROW
Figure 3. (A) Passive stage of orthodontic tooth movement when the contact angle, h, is less than the critical angle, hc. In the passive stage, the
wire does not touch the edges. (B) In an experimental design, the wire can be stabilized so it does not move, and the wire can be stabilized so it
cannot flex. The wire can actually be positioned so there are no forces whatsoever on the bracket-wire interface. This, however, never happens in
clinical orthodontics. (C) This represents the active stage of orthodontic tooth movement when h is more than hc, and the wires start to contact the
corners of the bracket and binding starts.
The inclusion criteria for the study were as follows: The canines were retracted using a GAC Sentalloy
retraction spring (150 g). The conventional canine
N Class II malocclusion with proclined maxillary inci-
bracket was tied in with a stainless steel ligature.
sors or crowding, or Class I malocclusion with
The retraction springs were attached to the bracket
proclined maxillary and mandibular teeth;
attachments. The SmartClip and Victory Series brack-
N treatment plan for extraction of maxillary first
ets have the attachments manufactured to the bracket,
premolars and retraction of maxillary canines; and
and the Damon3 attachment is manually placed and
N excellent periodontal health.
crimped to the bracket. Intra-arch mechanics were
All patients who met the inclusion criteria were used and the canines were retracted down a 0.018-
entered into the study. Their demographic character- inch stainless steel wire. Patients were seen every 4
istics are shown in Table 1. weeks (28 days).
Each patient had a 0.022-inch slot conventional All changes in the amount of retraction were
bracket placed on one canine and a 0.022-inch slot measured intra-orally by the investigator using a
Damon3 or SmartClip bracket placed on the other, with flexible millimeter ruler. The amount of retraction was
the left or right side for the self-ligating bracket chosen measured from the maxillary dental midline to the
using a randomization sequence. The molars were mesial of the canine. Each measurement was made
banded with Victory Series bands, and a transpalatal four times and the results averaged. The measure-
arch was placed. All remaining teeth were bracketed ments were made to the one-half millimeter.
with Victory Series brackets. The arches were leveled The rate of retraction of the canines was defined as
and aligned before commencing canine retraction. the distance traveled divided by the time interval

Figure 4. Classical friction results when only friction is involved.
needed for closure of the extraction space. The A paired t-test was used to compare the retraction
amount of retraction was measured at each appoint- rates of the conventional bracket to those of the
ment during space closure, but the only measurement Damon3 and SmartClip bracket.
used in this study was the distance traveled divided by
the number of 28-day intervals until one of the canines
RESULTS
was in the proper position; ie, the finish date for the
retraction was when the extraction space was closed The amount of movement between appointments
on one side (Figure 6). (28 days) for the three bracket types is shown in
Figure 5. Coefficient of binding data.

630 BURROW
Table 1. Demographic Characteristics of Sample Table 2. Rate of Movement by Bracket Type

n 5 43 Bracket Type Average Movement, mm Range, mm
Mean age at first appointment, years 14.8 6 6.24 Conventional 1.17 6 0.28 0.60 to 2.00
Range, years 11.3 to 27.6 Self-ligating 1.00 6 0.28 0.50 to 1.70
Damon3 0.90 6 0.24 0.50 to 1.30
Gender 44% Female SmartClip 1.10 6 0.28 0.70 to 1.70
56% Male
Table 2. The rate of movement for the conventional

bracket side was greater than that for either of the self- increased efficiency and possibly shorter treatment
ligating brackets, with the SmartClip bracket faster times. The force levels, size of the wire, and geometry
than the Damon3 bracket. Although the mean differ- of the bracket could have an impact on the efficiency of
ences at successive appointments were small, the tooth movement.
difference between the conventional bracket and both Many studies of the relationship of force to tooth
the self-ligating brackets was statistically significant on movement used closing loops. Although this does not
a paired t-test: SmartClip, P , .0043; Damon3, P , directly compare to sliding mechanics, it provides
.0001. some insight into the effect of force magnitude on the
The average movement per 28 days was 0.27 mm rate of tooth movement and the pattern of typical tooth
faster with the conventional brackets than with the movement. Boester and Johnston15 used sectional
Damon bracket; it is statistically significant (P , closing loops to retract canines in extraction situations
.0001). The movement per 28 days was 0.07 mm using force levels of 60, 150, 240, and 330 g. Their
faster with the conventional bracket than with the objective was to study the rate of tooth movement at
SmartClip bracket; this was also statistically significant various force levels. Maxillary canine retraction was
(P , .0043). When the self-ligating brackets were 0.8 mm/month for 60 g, 1.3 mm/month for 150 g,
combined and compared with the conventional bracket, 0.8 mm/month for 240 g, and 1 mm/month for 330 g
the average movement per 28 days was 0.17 mm faster of force. They concluded that ‘‘space closure proceeds
with the conventional; this was statistically significant equally rapidly at forces ranging from five (possibly
(P , .0001). Although these average differences were less) to eleven ounces. In this range, bone resorption,
small, they were statistically significant. per se, appears to be occurring at a minimal rate and
accordingly, may constitute the rate limiting factor.’’
DISCUSSION Iwasaki et al.16 used sectional closing loops to retract
canines and studied the velocity of tooth movement
In comparing these results with those of other
using force levels of 18 g and 60 g. When he used
studies of the rate of tooth movement, several points
60 g of force to retract canines, the velocity averaged
described below should be considered.
1.27 mm/month. With 18 g of force, the velocity was
0.87 mm/month, although there was quite a bit of
Rate of Movement in This Study of Sliding
individual variation. They concluded that ‘‘cell biology
Compared with Other Types of Tooth Movement
and metabolic factors must account for the variability in
The properties that influence resistance to sliding tooth movement.’’ From all three of these reports, it
are of great interest in the orthodontic community seems clear that once a level of force sufficient to
because lower resistance to sliding could lead to produce a biologic response is present, the rate of
Figure 6. (A) Maxillary right canine retracted using GAC retraction spring (150 g). Canine in final position (SmartClip). (B) Maxillary right canine in
final position (Damon). (C) Maxillary canine is retracted with conventional bracket (Victory Series).

tooth movement is a function of the biology, not the comparing clinical effectiveness on mandibular tooth
force magnitude. alignment comparing conventional brackets with Da-
The next studies investigated sliding mechanics. mon brackets and concluded that there was no
Paulson et al.17 measured canine retraction using difference in reducing crowding.
sliding mechanics down a 0.016-inch wire. The Pandis et al.23 compared the magnitude of moments
retraction force was 50 g to 100 g. His sample generated during a laboratory-simulated rotational
averaged 1.08 mm/month, but individual velocities correction of teeth using three different brackets:
ranged from 0.7 mm/month to 2.4 mm/month. Huff- Orthos 2, Damon2, and In-Ovation R. They found the
man and Way18 did an vivo study to determine the highest moments were generated using the Damon
amount of movement, rate of movement, and amount bracket, and the conventional brackets generated the
of tipping when retracting a canine down a 0.016-inch lowest moments. Thorstenson and Kusy12,14 also found
wire and a 0.020-inch wire using sliding mechanics. Damon brackets to have higher binding forces
They used 200 g of force to retract the canine. The (second-order angulation) compared with conventional
velocities were 1.37 mm/month when retracting down brackets.
the 0.016-inch wire and 1.20 mm/month with the These studies comparing self-ligating brackets with
0.020-inch wire. The difference was not significant. conventional brackets all point to the same conclusion
Sonis et al.19 used energy chains and latex thread to as this study, ie, ligation type makes little difference in
retract canines down a 0.016 3 0.022 inch archwire. initial alignment or space closure.
The initial forces were 250–400 g. The mean velocity
of tooth movement calculated over a 3-week period Bracket Geometry as a Factor in Resistance
was 1.28 mm for elastic threads and 1.51 mm for the to Sliding
elastic chains. They concluded that ‘‘all the materials
tested produced approximately equal amounts of tooth Hamdan and Rock,24 who investigated different
movement.’’ combinations of torque and tip, reported that every 4-
Based on the above studies, it seems reasonable to degree increase in bracket tip produced a significant
have used 150 g of force when retracting canines increase in resistance to sliding. ‘‘RS was significantly
along an undersized wire in this study. increased by tip and torque separately and in
combination, although tip was the more powerful
Other Studies of Self-ligating vs influence.’’
Conventional Brackets The amount of tip at the point of wire contact with the
corner of a bracket is a function of bracket width. The
Advocates of self-ligating brackets recommend their Damon bracket is 2.67 mm wide, the SmartClip is
bracket based on light forces, less friction, low 2.79 mm wide, and the Victory Series bracket is
moments, and consequentially more efficient (faster) 3.81 mm wide. When a force is placed on the canine
tooth movement. Is there any evidence that these (or any other) bracket to move it down the archwire, a
purported advantages are real? moment of force is created (MF). To counter the MF, a
Miles et al.20 compared the effectiveness of Damon moment of a couple is created (MC). The MC is equal to
and conventional twin brackets during initial alignment. the forces at the edges of the brackets times the width
This was a clinical trial using 60 consecutive patients in of the bracket. The maximum bending moment (MBM)
a split-mouth design. The irregularity index was when a tooth is moved down an archwire is MBM 5
measured at three times; original baseline, 10 weeks, (FWX)/L (Figure 7).
and 20 weeks. The conventional bracket achieved a To evaluate the effect the bracket width has on tooth
better irregularity index score at both 10 and 20 weeks. movement down an archwire, we need to algebraically
The authors reported that the difference in alignment rearrange MC 5 FW to F 5 MC/W. Substituting the
could be due to the fact that the Damon bracket does equation into the maximum binding moment gives
not engage the initial wires and allows 8.5u of rotational us:
play when compared with the conventional bracket.
The second wire, a 0.016 3 0.025 inch archwire, was MC
FWX WX
not fully engaged either. They surmised that the M~ ~W :
degree of rotational play involved in the Damon L L
bracket attributed to the better alignment of the Substituting (L2W ) for X gives us:
conventional bracket. Miles21 also compared SmartClip
self-ligating brackets to conventional brackets and MC
W (L{W ) M (L{W )
concluded that there was no difference in reducing M~ W ~
C
,
crowding. Scott et al.22 did a randomized clinical trial L L

632 BURROW
2. Jastrebski ZB. The Nature and Properties of Engineering

Materials. 3rd ed. New York, NY: Wiley; 1987.
3. Kusy RP, Whitley JQ. Influence of archwire and bracket
dimensions on sliding mechanics: derivations and determi-
nations of the critical contact angles for binding. Eur J Orthod.
1999;21:199–208.
4. Kusy RP. Ongoing innovations in biomechanics and
materials for the new millennium. Angle Orthod. 2000;70:
366–376.
5. Articolo LC, Kusy RP. Influence of angulation on the
resistance to sliding in fixed appliances. Am J Orthod
Dentofacial Orthop. 1999;115:39–51.
6. Nicolls J. Frictional forces in fixed orthodontic appliances.
Dent Pract. 1968;18:362–366.
7. Peterson L, Spencer, Andeasen GF. Comparison of
frictional resistance of Nitinol and stainless steel wires in
edgewise brackets. Quint Inter Digest. 1982;13:563–571.
Figure 7. The beam illustration demonstrates the position of the 8. Frank CA, Nikolai RJ. A comparative study of frictional
maximum bending moment (MBM). The maximum bending moment resistance between orthodontic bracket and archwire.
is MBM 5 (FWX)/L. Am J Orthod. 1980;78:593–609.
9. Andeasen GF, Quevedo FR. Evaluation of frictional forces in
which shows us that as the W (width) of the bracket the 0.022 3 0.028 edgewise bracket in vitro. J Biomech.
increases, the maximum bending moment decreases. 1970;3:151–160.
10. Sims AP, Waters NE, Birnie DJ. A comparison of the forces
Consequently, a narrower bracket generates higher required to produce tooth movement ex vivo through three
moments (higher forces at the edges) than a wider types of pre-adjusted brackets when subjected to deter-
bracket. These increased forces could increase mined tip or torque values. Br J Orthod. 1994;21:367–373.
resistance to sliding. This concept already is incorpo- 11. Ogata RH, Nanda RS, Duncanson MG Jr, Sinha PK, Currier
rated into standard texts of orthodontics25 that advo- GF. Frictional resistances in stainless steel bracket-wire
combinations with effects of vertical deflections. Am J Orthod
cate wider brackets for better sliding mechanics due to Dentofacial Orthop. 1996;109:535–542.
the decreased contact angle. 12. Thorstenson BS, Kusy RP. Effect of archwire size and
In this study, the average movement per 28 days material on the resistance to sliding of self-ligating brackets
was faster on the side being treated with the with second-order angulation in the dry state. Am J Orthod
conventional bracket than on the side being treated Dentofacial Orthop. 2002;122:295–305.
13. Thorstenson BS, Kusy RP. Comparison to resistance to
with the self-ligating bracket. Although this is a
sliding between different self-ligating brackets with second-
statistically significant difference, clinically it probably order angulation in the dry and saliva states. Am J Orthod
would not matter. Even though bracket geometry, Dentofacial Orthop. 2002;121:472–482.
particularly bracket width, influences resistance to 14. Thorstenson GA. SmartClip self-ligating brackets frictional
sliding, it still is important to keep in mind the limiting study. In: Orthodontic Perspectives. Vol XII, No 1. Monrovia,
CA: 3M Unitek; 2005:8–11.
factor in the rate of tooth movement appears to be
15. Boester CH, Johnston LE. A clinical investigation of the
more the biologic response to the force than mechan- concepts of differential and optimum force in canine
ical aspects of the orthodontic appliance. retraction. Angle Orthod. 1974;44:113–119.
16. Iwasaki LR, Haack JE, Nickel JC, Morton J. Human tooth
CONCLUSION movement in response to continuous stress of low magni-
tude. Am J Orthod Dentofacial Orthop. 2000;117:175–183.
N Canine retraction by sliding the tooth along an 17. Paulson RC, Speidel TM, Isaacson RJ. A laminagraphic
undersized archwire tends to be faster with conven- study of cuspid retraction versus molar anchorage loss.
Angle Orthod. 1970;40:20–27.
tional than self-ligating brackets, probably because
18. Huffman JD, Way DC. A clinical evaluation of tooth
the narrower self-ligating brackets lead to greater movement along arch wires of two different sizes.
elastic binding and resistance to sliding is much more Am J Orthod. 1983;83:453–459.
determined by this than by friction. 19. Sonis AL, Van der Plas E, Gianelly A. A comparison of
elastomeric auxiliaries versus elastic thread on premolar
extraction site closure: an in vivo study. Am J Orthod. 1986;
ACKNOWLEDGMENT
89:73–78.
I would like to thank Dr William R. Proffit for editorial 20. Miles PG, Weyant RJ, Rustveld L. A clinical trial of Damon2
assistance. vs conventional twin brackets during initial alignment. Angle
Orthod. 2006;76:480–485.
REFERENCES 21. Miles PG. SmartClip versus conventional brackets for initial
alignment: is there a difference? Aust Orthod J. 2005;21:
1. Blau JP. Tribology and Its Nomenclature in Friction and 123–127.
Wear Transitions of Materials. Park Ridge, NJ: Noyes 22. Scott P, DiBiase AST, Sherriff M, Coburne MT. Alignment
Publications; 1989. efficiency of Damon3 self-ligating and conventional ortho-

dontic bracket systems: a randomized clinical trial. 24. Hamdan A, Rock P. The effect of different combinations of
Am J Orthod Dentofacial Orthop. 2008;134:470–471. tip and torque on archwire/bracket friction. Eur J Orthod.
23. Pandis N, Eliades T, Partowi S, Bourauel C. Moments 2008;30:508–514.
generated during simulated rotational correction with self- 25. Proffit WR. Mechanical principles in orthodontic force control.
ligating and conventional brackets. Angle Orthod. 2008;78: In: Proffit WR, Fields HW, Sarver DH, eds. Contemporary
1030–1034. Orthodontics. 4th ed. St Louis, Mo: Elsevier; 207:376.

Original Article
Reliability of reference distances used in photogrammetry

Muge Aksua; Demet Kayab; Ilken Kocaderelic
ABSTRACT
Objective: To determine the reliability of the reference distances used for photogrammetric
assessment.
Materials and Methods: The sample consisted of 100 subjects with mean ages of 22.97 6 2.98
years. Five lateral and four frontal parameters were measured directly on the subjects’ faces. For
photogrammetric assessment, two reference distances for the profile view and three reference
distances for the frontal view were established. Standardized photographs were taken and all the
parameters that had been measured directly on the face were measured on the photographs. The
reliability of the reference distances was checked by comparing direct and indirect values of the
parameters obtained from the subjects’ faces and photographs. Repeated measure analysis of
variance (ANOVA) and Bland-Altman analyses were used for statistical assessment.
Results: For profile measurements, the indirect values measured were statistically different from
the direct values except for Sn-Sto in male subjects and Prn-Sn and Sn-Sto in female subjects. The
indirect values of Prn-Sn and Sn-Sto were reliable in both sexes. The poorest results were
obtained in the indirect values of the N-Sn parameter for female subjects and the Sn-Me parameter
for male subjects according to the Sa-Sba reference distance. For frontal measurements, the
indirect values were statistically different from the direct values in both sexes except for one in male
subjects. The indirect values measured were not statistically different from the direct values for Go-
Go. The indirect values of Ch-Ch were reliable in male subjects. The poorest results were obtained
according to the P-P reference distance.
Conclusions: For profile assessment, the T-Ex reference distance was reliable for Prn-Sn and Sn-
Sto in both sexes. For frontal assessment, Ex-Ex and En-En reference distances were reliable for
Ch-Ch in male subjects. (Angle Orthod. 2010;80:670–677.)
KEY WORDS: Photogrammetry; Reference distances
INTRODUCTION been performed by using lateral cephalometric radio-

graphs classically, and several cephalometric analy-
One of the most important factors in the planning of
ses have been developed and proposed.3–5
orthodontic treatment and the assessment of treat-
It is not possible to evaluate the soft tissues from the
ment changes is to evaluate the soft tissue.1,2 For this
frontal view by using cephalometric radiographs. In
purpose, quantitative assessments of soft tissue have
addition to cephalometric radiographs, soft tissue
evaluation has been carried out by means of different
a
Assistant Professor, Department of Orthodontics, Hacettepe methods such as anthropometry,6–10 two- or three-
University Faculty of Dentistry, Hacettepe University, Ankara, dimensional photogrammetry,11–23 and three-dimen-
Turkey. sional imaging techniques.24,25 Among these methods,
b
Research Assistant, Department of Orthodontics, Hacettepe
University, Ankara, Turkey.
two-dimensional photogrammetry has the advantage
c
Professor and Chair, Department of Orthodontics, Hacettepe of being a basic, noninvasive, cost-effective, and quick
University Faculty of Dentistry, Hacettepe University, Ankara, method that requires minimal time and equipment in
Turkey. the assessment of soft tissue.26–28
Corresponding author: Dr Muge Aksu, Department of Ortho- Most of the studies about soft tissue evaluation on
dontics, Faculty of Dentistry, Hacettepe University, Ankara -
06100, Turkey standardized two-dimensional life-sized photographs
(e-mail: mugepeh@hacettepe.edu.tr) reported the assessment or comparison of racial
Accepted: November 2009. Submitted: July 2009.
characteristics, differences between genders, and
G 2010 by The EH Angle Education and Research Foundation, treatment changes.13,16–18,20–22,29 Only one study con-
Inc. cluded the reliability of reference distances for facial

REFERENCE DISTANCES IN PHOTOGRAMMETRY 671
asymmetry assessment.19 Since then, researchers

have never attempted to study the reliability of
reference distances that can be used for photogram-
metric assessment. Nonetheless, such information is
important for clinicians because the reliability of the
measurements obtained from the photographs de-
pends on the reliability of the reference distances used
on photographs. Therefore, the aim of our study is to
classify the reliability of the five reference distances
used for photogrammetric assessment on subjects’
two-dimensional extraoral photographs obtained under
three postural conditions (centric relation, relaxed lip
posture, natural head orientation, and sitting position).
These reference distances are Sa-Sba (superior
auricula-subauricula), T-Ex (tragus-exocanthion) on
the profile view, and Ex-Ex (exocanthion-exocanthion),
En-En (endocanthion-endocanthion), and P-P (pupil
center-pupil center) on the frontal view.

The sample consisted of 100 healthy subjects (50
male and 50 female). The mean ages of the male and
female subjects were 23 6 3.39 and 22.94 6 2.41
years, respectively. All subjects had facial symmetry,
no history of trauma and no craniofacial anomaly. The Figure 1. Soft tissue parameters measured on the profile view: T-Ex
(tragus-exocanthion) tragus-exocanthion distance, Sa-Sba (super-
subjects were informed about the procedures; they aurale-subaurale) ear length, Tri-N (trichion-nasion) superior facial
accepted to participate in this study and signed third, N-Sn (nasion-subnasale) nose height or middle facial third, N-
informed consents. Prn (nasion-pronasale) nasal bridge length, Prn-Sn (pronasale-
subnasale) nasal tip protrusion, Sn-Sto (subnasale-stomion) upper
lip height, and Sn-Me (subnasale-menton) inferior facial third.
Direct Measurement
Direct measurements on each subject’s face were parameters used in the direct method were measured
done with a millimetric compass (Sylvac, Fowler, on photographic records using two reference distanc-
OPTO-RS232 Simplex/Duplex, Sweden) in centric es on the profile view (Sa-Sba, T-Ex) and three
relation, relaxed lip posture,30 natural head orienta- reference distances on the frontal view (Ex-Ex, En-
tion,31 and sitting position. Eight frontal and eight lateral En, P-P). Each measurement was repeated three
distances were measured directly. The parameters times by the same investigator following a 1-week
measured are shown in Figures 1 and 2. interval, and the mean values were used. Magnifica-
tion error was calculated from a basic proportion
Indirect Measurement using reference distances: X 5 A 3 B/C, where A is
Standardized lateral and facial photographs of each the selected reference plane distance measured on
subject were taken for the indirect measurements. All the subject’s face, B is any parameter measured on
subjects were positioned on a line marked on the the subject’s extraoral photograph, and C is the same
floor during the recording. The photographic set-up reference plane distance measured on the subject’s
consisted of a tripod supporting a digital camera extraoral photograph. Two different values on the
(Nicon Coolpix L1, 6.2 Megapixels, 53 zoom) 60 cm profile view and three different values on the frontal
away from the subject. No optical focus was used view were obtained for each parameter according to
during the taking of the photograph. Photographs the reference distances using this proportion. These
were taken with each subject in natural head values were compared with the direct values obtained
orientation,31 centric relation, and relaxed lip pos- from the patients. When the differences between the
ture30 as in the direct method. The photographic indirect values measured according to reference
records were transferred to the computer and distances and the direct values measured on sub-
analyzed with the software for Windows, Image Tool jects’ faces were no greater than 1 mm, the reference
version 3.0 (UTHSCSA, San Antonio, Tex). The distance was considered reliable.

672 AKSU, KAYA, KOCADERELI
Table 1. Comparison of Linear Values (in mm) of Reference Lines

Between Male and Female Subjects
Male (n 5 50) Female (n 5 50)
Reference Line Mean SD Mean SD P
Sa-Sba 61.42 3.85 56.87 3.44 .00
T-Ex 85.92 3.42 80.89 2.43 .00
Ex-Ex 94.03 3.62 91.09 3.60 .00
En-En 34.11 2.33 33.03 2.60 .03
P-P 48.31 4.44 51.35 3.38 .00
SD indicates standard deviation. P , .05.
For profile measurements, the indirect values

measured according to the Sa-Sba reference distance
were statistically different from the direct values
measured on faces of male and female subjects (P
, .05). In male subjects, the indirect values measured
according to the T-Ex distance were statistically
different from the direct values measured on subjects’
faces (P , .05), except for one parameter (Sn-Sto, P
5 .91). In female subjects, the indirect values
measured according to the T-Ex distance were also
statistically different from the direct values measured
Figure 2. Soft tissue parameters measured on the frontal view: Ex- on subjects’ faces (P , .05), except for two parame-
Ex (right exocanthion-left exocanthion) biocular width, En-En (right ters (Prn-Sn, P 5 .57; Sn-Sto, P 5 .55) (Table 2).
endocanthion-left endocanthion) intercanthal width, P-P (midpoint of The indirect values of two parameters (Prn-Sn and
right pupil-midpoint of left pupil) interpupillary width, Al-Al (right alare- Sn-Sto) according to the T-Ex reference distance were
left alare) alar width, Ch-Ch (right cheilion-left cheilion) mouth width,
reliable in both sexes. In male and female subjects, the
Go-Go (right gonion-left gonion) gonial width, and Sn-Sto (subna-
sale-stomion) upper lip height. indirect values measured according to the T-Ex
reference distance were shorter than the direct values
measured on subjects’ faces and closer to the direct
Statistical Analysis values than the indirect values measured according to
To determine the differences of reference distances the Sa-Sba reference distance in both sexes. The
between the female and male subjects, independent poorest results were obtained in the indirect values of
sample t-test was used. For the profile and frontal N-Sn parameter for female subjects and Sn-Me
measurements in both sexes, differences between the parameter for male subjects according to the Sa-Sba
values measured on the subject’s face and photograph reference distance. Among the indirect values mea-
were calculated with repeated measure analysis of sured according to the T-Ex reference distance, the
variance (ANOVA). Pair-wise comparisons were made greatest difference was in Sn-Me for male subjects
by using Bonferroni correction when the differences and N-Prn for female subjects (Tables 3 and 4).
between the direct and indirect measurements were For frontal measurements, the indirect values
statistically significant. To find the reliability of refer- measured according to Ex-Ex, En-En, and P-P
ence distances, Bland- Altman analysis was used. distances were statistically different from the direct
Intra-investigator reliability was assessed with intra- values measured on subjects’ photographs in both
class correlation coefficient for repeated measure- sexes (P , .05), except for one in male subjects. The
ments. indirect values measured according to Ex-Ex and En-
En distances were not statistically different from the
RESULTS direct values measured on subjects’ faces for the Go-
Go parameter (P 5 .06 for Ex-Ex line; P 5 .09 for En-
The reliability of the investigator was 0.97–0.99 for En line) in male subjects (Table 5).
repeated measurements, indicating excellent reliability. Of the four parameters, the indirect values of Ch-Ch
The sample was divided into two groups as female according to the Ex-Ex and En-En reference distances
and male because Sa-Sba, T-Ex, Ex-Ex, En-En, P-P were reliable in male subjects. The indirect values
distances were statistically different between the sexes measured according to the En-En reference distance
according to the independent samples t-test (P , .05; were closer to the direct values measured on subjects’
Table 1). faces in both sexes except for two parameters (Al-Al

Table 2. Soft Tissue Parameters Measured on Subjects’ Faces and Photographs From Profile View (in mm)a
Indirect Values According Indirect Values According
Direct Values (A) to Sa-Sba (B) to T-Ex (C) P
Parameter Sex Mean 6 SD Mean 6 SD Mean 6 SD ABC AB AC
N-Sn Male 56.44 6 4.09 48.88 6 4.35 53.17 6 3.78 .00* .00* .00*
Female 53.29 6 4.10 46.83 6 4.05 50.29 6 3.94 .00* .00* .00*
N-Prn Male 49.99 6 3.89 42.95 6 4.21 46.65 6 3.91 .00* .00* .00*
Female 45.35 6 4.25 39.34 6 3.21 42.32 6 3.24 .00* .00* .00*
Prn-Sn Male 21.33 6 1.80 18.79 6 1.55 20.44 6 1.54 .00* .00* .00*
Female 19.70 6 1.39 18.08 6 1.55 19.45 6 1.58 .00* .00* .57
Sn-Sto Male 22.58 6 2.47 20.64 6 2.35 22.43 6 2.42 .00* .00* .91
Female 20.46 6 1.78 18.98 6 1.85 20.37 6 1.90 .00* .00* .55
Sn-Me Male 74.34 6 6.06 65.02 6 5.92 70.62 6 5.53 .00* .00* .00*
Female 66.07 6 4.26 60.11 6 3.67 64.68 6 4.20 .00* .00* .00*
a
* Indicates that the direct values measured on subjects’ faces are statistically different from the indirect values measured on subjects’
photographs according to Sa-Sba and T-Ex lines.
and Sn-Sto) in male and one parameter (Go-Go) in conditions (relaxed lip posture, natural head orienta-
female subjects. The poorest results were obtained tion, and sitting position). Until now, there has been
according to the P-P reference distance (Tables 6 and 7). evidence about the usefulness of photographic as-
sessment.11,28,33–36
DISCUSSION The usefulness of the patients’ photographs is
Two-dimensional photogrammetry has been used limited unless the prints are of standardized view
for evaluating the soft tissues in orthodontic treatment. and size. Farkas et al.37 compared a large number
The method was shown to be sufficiently reproducible of facial measurements taken from standardized
since it was simple to achieve in a conventional setting, photographs. Determination of absolute sizes on
without the need for special equipment.27,28 Several photographs necessitates the calibration of the
authors have published the profile and frontal charac- image such as marking on the subject’s face and
teristics of the face by collecting the data via using a millimeter ruler unless life-size photographs
anthropometric measurements6–10 or by using three- are used. In our study, a basic proportion using
dimensional imaging techniques.15,19,23,25,32 The deter- reference distances served as a standard of image
mination of the reliability of 2-dimensional photogram- calibration. To reduce method error, all measurements
metry for soft tissue evaluation might provide clinicians were made with each subject in natural head orienta-
the ability to assess soft tissue from both profile and tion,31 centric relation, sitting position, and relaxed lip
frontal views after orthodontic treatment. posture.30
This study was designed to classify the reliability of The reliability of the investigator was excellent,
the five reference line distances used for photogram- indicating that soft tissue landmarks can be located
metric assessment on subjects’ two-dimensional ex- consistently. The arguments for using the ear and eye
traoral photographs obtained under three postural are that the main development of these parts of the
Table 3. For Profile Measurements, Differences Between the Direct and Indirect Measurements According to T-Ex and Sa-Sba Reference
2a
Distances In Male Subjectsd

Confidence Interval for Agreement
Measurements d̄ SD d̄ 2 1.96 * SD d̄ + 1.96 * SD
N-Sn(direct) 2 N-Sn(T-Ex) 3.271 1.360 0.605 5.937
N-Sn(direct) 2 N-Sn(Sa-Sba) 7.563 2.030 3.585 11.541
N-Prn(direct) 2 N-Prn(T-Ex) 3.338 1.495 0.408 6.267
N-Prn(direct) 2 N-Prn(Sa-Sba) 7.034 1.993 3.127 10.942
Prn-Sn(direct) 2 Prn-Sn(T-Ex) 0.891 1.167 21.396 3.179
Prn-Sn(direct) 2 Prn-Sn(Sa-Sba) 2.547 1.037 0.514 4.580
Sn-Sto(direct) 2 Sn-Sto(T-Ex) 0.155 1.051 21.906 2.215
Sn-Sto(direct) 2 Sn-Sto(Sa-Sba) 1.942 1.313 20.632 4.516
Sn-Me(direct) 2 Sn-Me(T-Ext) 3.723 2.579 21.332 8.779
Sn-Me(direct) 2 Sn-Me(Sa-Sba) 9.325 2.959 3.525 15.124
2a
d
d̄ indicates mean values of the differences between the direct and indirect measurements; SD, standard deviation.

Table 4. For Profile Measurements, Differences Between the Direct and Indirect Measurements According to T-Ex and Sa-Sba Reference
2a
Distances in Female Subjectsd

N-Sn(direct) 2 N-Sn(T-Ex) 3.004 2.255 21.415 7.423
N-Sn(direct) 2 N-Sn(Sa-Sba) 6.454 3.087 0.403 12.505
N-Prn(direct) 2 N-Prn(T-Ex) 3.029 3.085 23.018 9.075
N-Prn(direct) 2 N-Prn(Sa-Sba) 6.008 3.394 20.644 12.660
Prn-Sn(direct) 2 Prn-Sn(T-Ex) 0.242 0.879 21.481 1.966
Prn-Sn(direct) 2 Prn-Sn(Sa-Sba) 1.618 1.068 20.475 3.712
Sn-Sto(direct) 2 Sn-Sto(T-Ex) 0.090 1.058 21.984 2.164
Sn-Sto(direct) 2 Sn-Sto(Sa-Sba) 1.488 1.145 20.757 3.733
Sn-Me(direct) 2 Sn-Me(T-Ext) 1.388 2.972 24.438 7.214
Sn-Me(direct) 2 Sn-Me(Sa-Sba) 5.954 3.148 20.217 12.124
2a
face occurs in the early ages and are stable during elasticity of the ear might account for some error
growing.38–41 during the assessment.
In our study, Sa-Sba, T-Ex, Ex-Ex, En-En, and P-P Of the four parameters obtained from the frontal
distances were different between male and female views, the difference between direct and indirect
subjects. Therefore, we assessed the subjects sepa- measurements of Ch-Ch was less than 1 mm in male
rately as male and female. subjects according to the Ex-Ex and En-En reference
Of the five parameters obtained from the profile distances. In contrast to our results, Farkas et al.37 and
views, the difference between direct and indirect Tanner and Weiner42 showed that the difference
measurements of Prn-Sn and Sn-Sto were less than between the indirect and direct measurements for
1 mm in both sexes according to the T-Ex line. The Ch-Ch parameter was more than 1 mm. The difference
highest difference was seen in Sn-Me (3.72 mm) for in the other remaining parameters was less than 2 mm
the male group and in N-Prn (3.03 mm) for the female except for the parameter Go-Go in female subjects.
group when the T-Ex reference distance was used. The differences between direct and indirect measure-
However, the differences between direct and indirect ments were dramatically higher when P-P reference
measurements were higher (1.48–9.32 mm) when the distance was used. It must be kept in mind that all
Sa-Sba reference distance was used. The poorest subjects were asked to look straight ahead to a distant
results were obtained with the Sa-Sba reference point at eye level during the assessment. The use of a
distance. The results of this study for the profile stable point might eliminate possible errors resulting
measurements showed that the indirect measure- from pupils and might give different results. Ras et al.19
ments according to T-Ex distance were closer to direct concluded that the best reference line among four
measurements than the indirect measurements ac- reference distances (exocanthion-exocanthion, endo-
cording to the Sa-Sba distance for both sexes. The canthion-endocanthion, superalare-superalare, and
Table 5. Soft Tissue Parameters Measured on Subjects’ Faces and Photographs From Frontal View (in mm)
Indirect Values Indirect Values Indirect Values
According to According to According to
Direct Values (A) Ex-Ex (B) En-En (C) P-P (D) P
Parameter Sex Mean SD Mean 6 SD Mean 6 SD Mean 6 SD ABCD AB AC AD
Al-Al Male 37.93 6 2.60 39.48 6 2.70 39.60 6 2.74 29.42 6 3.19 .00* .00* .00* .00*
Female 33.94 6 1.99 35.77 6 2.32 35.74 6 2.31 28.87 6 2.21 .00* .00* .00* .00*
Ch-Ch Male 52.11 6 3.07 52.96 6 3.26 53.03 6 3.26 39.40 6 4.01 .00* .02* .00* .00*
Female 47.99 6 3.46 49.72 6 3.71 49.63 6 3.91 40.24 6 3.62 .00* .00* .00* .00*
Go-Go Male 124.37 6 5.18 122.59 6 5.75 122.71 6 5.83 89.87 6 8.85 .00* .06 .09 .00*
Female 118.46 6 5.72 112.11 6 6.34 111.96 6 6.42 90.69 6 6.50 .00* .00* .00* .00*
Sn-Sto Male 22.08 6 2.58 23.63 6 2.64 23.66 6 2.66 17.58 6 2.29 .00* .00* .00* .00*
Female 20.01 6 1.72 21.24 6 1.89 21.19 6 1.87 17.17 6 1.84 .00* .00* .00* .00*
* Indicates that the direct values measured on subjects’ faces are statistically different from the indirect values measured on subjects’
photographs according to Ex-Ex, En-En, and P-P lines.

Table 6. For Frontal Measurements, Differences Between the Direct and Indirect Measurements According to Ex-Ex, En-En, and P-P
Reference Distances in Male Subjectsa
Al-Al(direct) 2 A-Al(ex-ex) 21.552 1.010 23.531 0.428
Al-Al(direct) 2 Al-Al(en-en) 21.671 0.988 23.608 0.266
Al-Al(direct) 2 Al-Al(p-p) 8.504 2.790 3.035 13.973
Ch-Ch(direct) 2 Ch-Ch(ex-ex) 20.855 2.011 24.797 3.086
Ch-Ch(direct) 2 Ch-Ch(en-en) 20.920 1.901 24.648 2.803
Ch-Ch(direct) 2 Ch-Ch(p-p) 12.707 3.965 4.936 20.478
Go-Go(direct) 2 Go-Go(ex-ex) 1.536 4.261 26.815 9.887
Go-Go(direct) 2 Go-Go(en-en) 1.429 4.269 26.939 9.797
Go-Go(direct) 2 Go-Go(p-p) 33.036 8.951 15.492 50.580
Sn-Sto(direct) 2 Sn-Sto(ex-ex) 21.555 1.369 24.237 1.128
Sn-Sto(direct) 2 Sn-Sto(en-en) 21.577 1.339 24.201 1.046
Sn-Sto(direct) 2 Sn-Sto(p-p) 4.495 1.848 0.874 8.116
a
cheilion-cheilion) was formed by the one which is the measurement precision was 1 mm. The reliability
perpendicular and bisects the line that connects the of our results depends on the clinic sense of the
landmarks exocanthion. However, Farkas et al.37 found orthodontist. Also, the results could change if life-sized
that Ex-Ex was not reliable while En-En was reliable. In photographs are used.
our study, the distortion caused by photographing,
measuring without previously indicated landmarks on
CONCLUSIONS
the face, might have accounted for the unreliability of
reference distances. Farkas et al.37 stated that the N For profile measurements, the T-Ex reference
magnitude of the error depends on the thickness of the distance is reliable for the indirect values of Prn-Sn
soft tissue covering the bony landmark, and measure- and Sn-Sto in both sexes. The poorest results were
ments of some landmarks (eg, Al, Sa, Sba) may not be obtained for the Sa-Sba reference distance and the
precise if photographs are not sharp enough to allow indirect values of N-Sn parameter in female and Sn-
accurate identification of these landmarks. Me parameter in male subjects.
Our results showed that some of the measurements N For frontal measurements, Ex-Ex and En-En refer-
according to reference distances are less reliable ence distances are reliable only for the indirect
when compared with direct measurements, while few values of one parameter (Ch-Ch) in male subjects.
but reliable indirect measurements exist. The mea- The poorest results were obtained for the indirect
surement precision is important for evaluating the values measured according to P-P reference dis-
reliability of direct and indirect methods. In this study, tance and for Go-Go parameter in both sexes.
Table 7. For Frontal Measurements, Differences Between the Direct and Indirect Measurements According to Ex-Ex, En-En, and P-P
Reference Distances in Female Subjectsa
Al-Al(direct) 2 Al-Al(ex-ex) 21.831 1.331 24.439 0.778
Al-Al(direct) 2 Al-Al(en-en) 21.799 1.517 24.772 1.174
Al-Al(direct) 2 Al-Al(p-p) 5.072 1.663 1.813 8.330
Ch-Ch(direct) 2 Ch-Ch(ex-ex) 21.729 2.052 25.751 2.292
Ch-Ch(direct) 2 Ch-Ch(en-en) 21.635 2.418 26.375 3.105
Ch-Ch(direct) 2 Ch-Ch(p-p) 7.753 2.475 2.903 12.604
Go-Go(direct) 2 Go-Go(ex-ex) 6.343 4.039 21.574 14.260
Go-Go(direct) 2 Go-Go(en-en) 6.497 4.477 22.277 15.272
Go-Go(direct) 2 Go-Go(p-p) 27.762 5.953 16.094 39.429
Sn-Sto(direct) 2 Sn-Sto(ex-ex) 21.225 1.528 24.219 1.770
Sn-Sto(direct) 2 Sn-Sto(en-en) 21.182 1.484 24.0192 1.727
Sn-Sto(direct) 2 Sn-Sto(p-p) 2.842 1.831 20.748 6.431
a

REFERENCES 20. Fernández-Riveiro P, Suárez-Quintanilla D, Smyth-Cha-

mosa E, Suárez-Cunqueiro M. Linear photogrammetric
1. Fränkel R, Fränkel C. Orthodontics in orofacial region with analysis of the soft tissue facial profile. Am J Orthod
help of function regulators. Inf Orthod Kieferorthop. 1988;20: Dentofacial Orthop. 2002;122:59–66.
277–309. 21. Scavone H Jr, Trevisan H Jr, Garib DG, Ferreira FV. Facial
2. McNamara JA Jr, Brust EW, Riolo ML. Soft tissue evaluation profile evaluation in Japanese-Brazilian adults with normal
of individuals with an ideal occlusion and a well-balanced occlusions and well-balanced faces. Am J Orthod Dentofa-
face. In: McNamara JA Jr, ed. Esthetics and the Treatment cial Orthop. 2006;129:721.e1–5.
of Facial Form. Monograph 28, Craniofacial Growth Series. 22. Choe KS, Yalamanchili HR, Litner JA, Sclafani AP, Quatela
Ann Arbor, Mich: Center for Human Growth and Develop- VC. The Korean American woman’s nose: an in-depth nasal
ment, The University of Michigan; 1993:115–146. photogrammatic analysis. Arch Facial Plast Surg. 2006;8:
3. Broadbent BH Sr, Broadbent BH Jr, Golden WH. Bolton 319–323.
Standards of Dentofacial Developmental Growth. St Louis, 23. Honrado CP, Lee S, Bloomquist DS, Larrabee WF Jr.
Mo: CV Mosby; 1975. Quantitative assessment of nasal changes after maxillo-
4. Garner LD. Soft-tissue changes concurrent with orthodontic mandibular surgery using a 3-dimensional digital imaging
tooth movement. Am J Orthod. 1974;66:367–377. system. Arch Facial Plast Surg. 2006;8:26–35.
5. Roos N. Soft-tissue profile changes in Class II treatment. 24. Guess MB, Solzer WV. Computer treatment estimates in
Am J Orthod. 1977;72:165–175. orthodontics and orthognathic surgery. J Clin Orthod. 1989;
6. Farkas LG. Anthropometry of the Head and Face in 23:262–268.
Medicine. New York, NY: Elsevier North Holland Inc; 25. Soncul M, Bamber MA. Evaluation of facial soft tissue
1981:285. changes with optical surface scan after surgical correction of
7. Uzun A, Akbas H, Bilgic S, Emirzeoglu M, Bostanci O, Sahin Class III deformities. J Oral Maxillofac Surg. 2004;62:
B, Bek Y. The average values of the nasal anthropometric 1331–1340.
measurements in 108 young Turkish males. Auris Nasus 26. Nechala P, Mahoney J, Farkas LG. Digital two-dimensional
Larynx. 2006;33:31–35. photogrammetry: a comparison of three techniques of
8. Chung C, Lee Y, Park KH, Park SH, Park YC, Kim KH. obtaining digital photographs. Plast Reconstr Surg. 1999;
Nasal changes after surgical correction of skeletal Class III 103:1819–1825.
malocclusion in Koreans. Angle Orthod. 2008;78:427–432. 27. Good S, Edler R, Wertheim D, Greenhill D. A computerized
9. Arslan SG, Genç C, Odabaş B, Kama JD. Comparison of photographic assessment of the relationship between
facial proportions and anthropometric norms among Turkish skeletal discrepancy and mandibular outline asymmetry.
young adults with different face types. Aesthetic Plast Surg. Eur J Orthod. 2006;28:97–102.
2008;32:234–242. 28. Edler R, Wertheim D, Greenhill D. Comparison of radio-
10. Bozkir MG, Karakas P, Oguz O. Measurements of soft graphic and photographic measurement of mandibular
asymmetry. Am J Orthod Dentofacial Orthop. 2003;123:
tissue orbits in Turkish young adults. Surg Radiol Anat.
167–174.
2003;25:54–57.
29. Fernández-Riveiro P, Smyth-Chamosa E, Suárez-Quinta-
11. Zhang X, Hans MG, Graham G, Kirchner HL, Redline S.
nilla D, Suárez-Cunqueiro M. Angular photogrammetric
Correlations between cephalometric and facial photographic
analysis of the soft tissue facial profile. Eur J Orthod.
measurements of craniofacial form. Am J Orthod Dentofa-
2003;25:393–399.
cial Orthop. 2007;131:67–71.
30. Arnett GW, Bergman RT. Facial keys to orthodontic
12. Gavan JA, Wahburn SL, Lewis PH. Photography: an
diagnosis and treatment planning. Part I. Am J Orthod
anthropometric tool. Am J Phys Anthropol. 1952;10:331–
353.
31. Lundström A, Lundström F, Lebret LM, Moorrees CF.
13. Milosević SA, Varga ML, Slaj M. Analysis of the soft tissue Natural head position and natural head orientation: basic
facial profile of Croatians using of linear measurements. considerations in cephalometric analysis and research.
J Craniofac Surg. 2008;19:251–258. Eur J Orthod. 1995;17:111–120.
14. Stoner MM. A photometric analysis of the facial profile. 32. Terajima M, Yanagita N, Ozeki K, et al. Three-dimensional
Am J Orthod. 1955;41:453–469. analysis system for orthognathic surgery patients with jaw
15. Hajeer MY, Ayoub AF, Millett DT. Three-dimensional deformities. Am J Orthod Dentofacial Orthop. 2008;134:
assessment of facial soft-tissue asymmetry before and after 100–111.
orthognathic surgery. Br J Oral Maxillofac Surg. 2004;42: 33. Proffit W. Contemporary Orthodontics. St Louis, Mo: CV
396–404. Mosby; 1986:127–134.
16. Leong SC, White PS. A comparison of aesthetic proportions 34. Graber TM. Orthodontics; Principles and Practice. 3rd ed.
between the Oriental and Caucasian nose. Clin Otolaryngol Philadelphia, Pa: WB Saunders; 1972:275–288.
Allied Sci. 2004;29:672–676. 35. Robertson NR, Volp CR. Telecentric photogrammetry: its
17. Dimaggio FR, Ciusa V, Sforza C, Ferrario VF. Photographic development, testing, and application. Am J Orthod. 1981;
soft-tissue profile analysis in children at 6 years of age. 80:623–637.
Am J Orthod Dentofacial Orthop. 2007;132:475–480. 36. Lauweryns I, van Cauwenberghe N, Carels C. Interobserver
18. Scavone H, Zahn-Silva W, do Valle-Corotti KM, Nahás AC. and intraobserver agreement of clinical orthodontic judg-
Soft tissue profile in white Brazilian adults with normal ments based on intraoral and extraoral photographs. Angle
occlusions and well-balanced faces. Angle Orthod. 2008;78: Orthod. 1994;64:23–30.
58–63. 37. Farkas GL, Bryson W, Klotz J. Is photogrammetry of the
19. Ras F, Habets LL, van Ginkel FC, Prahl-Andersen B. face reliable? Plast Reconstr Surg. 1980;66:346–355.
Method for quantifying facial asymmetry in three dimensions 38. Farkas LG, Posnick JC, Hreczko TM. Anthropometric
using stereophotogrammetry. Angle Orthod. 1995;65:233– growth study of the ear. Cleft Palate Craniofac J. 1992;29:
239. 324–329.

39. Ferrario VF, Sforza C, Ciusa V, Serrao G, Tartaglia GM. 41. Peck S, Peck L, Kataja M. Skeletal asymmetry in
Morphometry of the normal human ear: a cross-sectional esthetically pleasing faces. Angle Orthod. 1991;61:43–48.
study from adolescence to mid-adulthood. J Craniofac 42. Tanner JM, Weiner JS. The reliability of the photogrammet-
Genet Dev Biol. 1999;19:226–233. ric method of anthropometry, with a description of a
40. Vig PS, Hewitt AB. Asymmetry of the human facial skeleton. miniature camera technique. Am J Phys Anthropol. 1949;
Angle Orthod. 1975;45:125–129. 7:145–186.

Original Article
Temporary anchorage device insertion variables: effects on retention

Joseph S. Petreya; Marnie M. Saundersb; G. Thomas Kluemperc;
Larry L. Cunninghamd; Cynthia S. Beemane
ABSTRACT
Objective: To quantify the influence of temporary anchorage device (TAD) insertion variables on
implant retention.
Materials and Methods: Three hundred thirty TADs from three companies were placed in
synthetic bone replicas at variable depths and angulations and compared. Clinically relevant forces
were applied to the TADs until failure of retention occurred.
Results: In all three implants, increased insertion depth increased implant retention. As the
distance from the abutment head to the cortical plate increased, the retention of all three implants
decreased. A significantly greater force to fail was required for a 90u insertion angle than for 45u or
135u insertion angles. No significant difference was found between the 45u and 135u insertion
angles. A significant reduction in force to fail occurred when comparing 90u and 45u oblique
insertion angles.
Conclusions: Increasing penetration depth of TADs results in greater retention. Increased
abutment head distance from cortical plate leads to decreased retention. Placement of TADs at 90u
to the cortical plate is the most retentive insertion angle. Insertion at an oblique angle from the line
of force reduces retention of TADs. (Angle Orthod. 2010;80:634–641.)
KEY WORDS: Temporary anchorage device; Mini-implant; Implant failure; Implant retention;
Bone; Force to fail
INTRODUCTION spurred interest in skeletal anchorage systems, which

appeal to practitioners because they have the potential
For more than 100 years, orthodontists have
to provide absolute anchorage and do not depend on
searched for ideal anchorage that fits two criteria:
patient compliance.1 Skeletal anchorage has been the
absolute resistance to unwanted tooth movement and
subject of study for more than 60 years in orthodon-
independence from patient compliance. Conventional
tics.2,3 The successful use of osseointegrated retro-
intra- and extraoral anchorage systems often fall short
molar, palatal, and restorable implants has been
of providing absolute anchorage. This deficiency has
demonstrated in the literature.4–6 Unfortunately, these
implant systems require osseointegration before or-
a
Private practice, Somerset, Ky. thodontic force can be applied; in addition, they may
b
Assistant Professor, Center for Biomedical Engineering, increase treatment time, they are expensive, their
University of Kentucky, Lexington, Ky.
size limits placement location, and their removal is
c
Associate Professor and Chief, Division of Orthodontics,
University of Kentucky College of Dentistry, Lexington, Ky. difficult.
d
Associate Professor and Chief, Division of Oral and Unlike osseointegrated implants, temporary anchor-
Maxillofacial Surgery, University of Kentucky College of Dentist- age devices (TADs) are easily inserted and removed,
ry, Lexington, Ky. and can be immediately loaded.1,7 They are relatively
e
Associate Professor and Program Director, Division of
Orthodontics, University of Kentucky College of Dentistry, inexpensive and can be placed in a variety of locations,
Lexington, Ky. vastly increasing their versatility.1 Consequently, TADs
Corresponding author: Dr Cynthia S. Beeman, Associate are quickly becoming the preferred method of skeletal
Professor and Program Director, Division of Orthodontics, Room anchorage. The risk and utility of TADs are questions
D442 Chandler Medical Center, University of Kentucky College
of Dentistry, 800 Rose Street, Lexington, KY 40536-0297
that still need to be answered. Since TADs are not
(e-mail: csbeeman@gmail.com) osseointegrated, the magnitude of orthodontic forces
they can withstand, and how various insertion param-
Accepted: November 2009. Submitted: July 2009.
G 2010 by The EH Angle Education and Research Foundation, eters affect their ability to withstand these forces is not
Inc. well understood.

TEMPORARY ANCHORAGE DEVICE INSERTION VARIABLES 635
Figure 1. Soft tissue difference effects on implant insertion.

Figure 2. Angular insertion of 45u, 90u, and 135u to the cortical plate.
Reported success rate of TADs ranges from 80.5%

to 95.2%.1,7–9 Implant diameter, length, time of loading,
and presence of inflammation have been reported as tion, and attachment head distance from the cortical
factors affecting success of TADs.1,7–9 Implant reten- plate.
tion is also affected by loading parameters. Previous
studies have employed direct pull-out tests with forces MATERIALS AND METHODS
running parallel to the long axis of the TAD.10 Because Three hundred thirty tests were completed on TADs
TADs are not loaded vertically in clinical situations, a of three companies: Ormco, Orange, CA; Synthes
more appropriate model is to load TADs perpendicular North America, West Chester, PA; and Dentaurum,
to their long axes, placing horizontal forces on the Newtown, PA. Implants were placed in synthetic bone
implants. replicas (Sawbones, Pacific Research Laboratories)
Previous reports suggest greater moment arms and matched for cortical plate and cancellous bone
created by increased abutment head distance from thickness and density of the maxillary premolar region.
the cortical plate reduce implant stability.11 Investiga- Self-drilling TADs, which included Ormco VectorTAS,
tions have been limited to minimal variations of 6 mm and 8 mm; Synthes OBA (Orthodontic Bone
abutment head distance from the cortical plate and Anchor), 6 mm, 8 mm, and 10 mm; and Dentaurum
are not well correlated to the clinical environment.12 As tomas (temporary orthodontic micro anchorage sys-
soft tissue thickness for TAD placement may be as tem), 6 mm, 8 mm, and 10 mm, were tested. TADs
high as 4.5 mm, implants seated to the soft tissue from three different companies were used, not to test
collar potentially create increased lever arms, which differences between products, but to study how TADs
affect stability9,13–15 (Figure 1). These features and their respond overall to specific placement protocols.
potential impact on TAD placement and stability Synthetic bone replicas were matched for the
require further investigation. maxillary premolar region in cortical plate thickness,
Previous reports on the potential impact of insertion as well as plate and cancellous bone densities.17–19
angles on retention are conflicting. Published reports The maxillary premolar region was selected due to its
have recommended insertion angles of 45u, while combination of the thinnest cortical plate, the least
others advise that a 90u angle to the cortical plate is dense cancellous bone and because clinically, it
more retentive (Figure 2).1,16 To date, reviews on the represents a common location for TAD placement.17–19
impact of loading implants obliquely when inserted at Biomechanical test blocks were chosen over a
extreme angles to the occlusal plane have not been cadaver model because they offer uniform and
evaluated. With such wide variation in advice, it is consistent physical properties that eliminate the
difficult to determine best-evidence clinical guidelines. variability encountered when testing with human
Moreover, studies evaluating the potential impact of cadaver bone. Synthetic bone also does not have the
insertion angles define TAD failure differently. Some concerns of desiccation and quality change of cadaver
investigations have defined failure as complete remov- bone. Bone blocks consisted of a cancellous region of
al, while others are less stringent. 0.08 g/cc (5 pcf) density cellular rigid polyurethane
The purpose of this study was to investigate foam with a 1.5 mm 0.64 g/cc (40 pcf) laminated solid
insertion variables and their affect on TAD stability. rigid polyurethane foam cortical plate analog. Blocks
Variables included penetration depth, angle of inser- were milled to 1 inch2 by 1.8 inches tall.

636 PETREY, SAUNDERS, KLUEMPER, CUNNINGHAM, BEEMAN
Figure 5. Abutment distance from the cortical plate deep insertion.

Figure 3. Implants of 6 mm, 8 mm, and 10 mm inserted to base of
soft tissue collar.
8-mm implants inserted to depths of 6 mm, creating 4-
The ability of each TAD to resist increasing force and 6-mm abutment distances from the cortical plate,
levels was evaluated for three separate insertion respectively (Figure 5).
variables: insertion depth, abutment head from the Effects of insertion angle were also investigated,
cortical plate, and angle of insertion. For evaluation of inserting 8-mm implants at 45u, 90u, and 135u angles to
insertion depth, implants were inserted 90u to the the cortical plate. Each of the three angular place-
cortical plate to the depth of the soft tissue collar ments was made in the same plane as the line of force
creating 6-, 8-, and 10-mm insertion depths, respec- applied. A final angular test was completed placing the
tively (Figure 3). For evaluation of abutment distance implants perpendicular to the applied force, but at a
from the cortical plate, all implants were inserted to a 45u oblique angle (Figure 6). A custom jig was
uniform 6-mm depth. Implants of 6 mm were inserted to designed to ensure all implants were inserted at the
depth, leaving a 4-mm abutment head above the appropriate angle.
cortical plate. Implants of 8 mm and 10 mm were A small-scale loading machine (Figure 7) was used to
inserted to 6-mm depths, leaving their abutment heads apply force to the implants. Bone blocks were placed in
and additional threads above the cortical plate, creating the machine with the implants oriented tangent to the
increased abutment head distances from the cortical load cell secured with vise grips and a backing plate to
plate (Figure 4). An additional investigation of distance
from the cortical plate was undertaken, inserting 10-mm
implants to a depth of 8 mm, creating a 6-mm distance
from the cortical plate. This was compared to 6-mm and
Figure 4. Abutment distances from the cortical plate. Figure 6. Angular oblique insertion.

Figure 7. Small scale loading machine with a blowup of bone blocks placed in the machine with the implants oriented tangent to the load cell.
counteract block rotation. Implants were attached via a Insertion Depth

.30-mm monofilament nylon ligature looped around the
In all three implant systems, a significant increase in
abutment head and attached to a 10-lb load cell via a
force to fail occurred as insertion depth increased (P ,
compression grip. Force was applied to the implants at a
.001; Figure 9; Table 1). The force to fail required for
rate of 4 mm/sec until failure of retention occurred.
all 6-mm penetration depths was significantly less than
Failure point was determined at the inflection point, the
point beyond bending at which the implant initially began
to move in the bone block (Figure 8). Ten data points
were collected per second and the load displacement
data were recorded. Peak load force to fail was
obtained. Force to fail was analyzed by using unpaired
t-tests for two group comparisons and one-way analysis
of variance (ANOVA) for three or more comparison
groups. Post-hoc analyses used the Tukey-Kramer
method. Results were considered significant at P , .05.
RESULTS
The initial force to fail tests on 330 implants was
successfully recorded. Significant differences (P ,
.05) in force to fail were detected in all groups Figure 8. Test demonstrating failure point determined via inflection
examined. point indicating translation of implant in bone block.

Figure 9. Failure loads for insertion depths. Figure 10. Failure loads for abutment distances from the
cortical plate.
that for 8-mm depths. There were no significant
differences between the 8-mm and 10-mm insertion fail with a 24.2% decrease (P , .001) as distance of
depths. Reduction in force to fail was 25.0% (P , .05) the abutment head from the cortical plate increased
for Ormco VectorTAS 6-mm vs 8-mm insertion depths, from 4 mm to 8 mm. Results were not significant when
17.0% (P , .01) for Synthes OBA 6-mm vs 8-mm, comparing abutment distances of 6-mm and 8-mm
21.8% (P , .001) for Synthes OBA 6-mm vs 10-mm, from the cortical plate. When evaluating increased
34.1% (P , .01) for Dentaurum tomas 6-mm vs 8-mm, insertion depth combined with increased abutment
and 36.5% (P , .001) for Dentaurum tomas 6-mm vs distance from the cortical plate, no significant differ-
10-mm insertion depths. ences in implant retention were found in either the
Synthes OBA or the Dentaurum tomas implant
Abutment Distance From Cortical Plate systems (Figure 11; Table 3).
In all three implant systems, a reduction in force to Angular Measures

fail occurred as distance from the cortical plate
increased (Figure 10; Table 2). The results for the A significantly reduced failure force was required
VectorTAS implants demonstrated a reduction in force with implants placed at 45u or 135u insertion angles
to fail when comparing 6-mm and 4-mm distances compared with a 90u insertion angle (P , .001) in all
from the cortical plate, but were not significant due to three implant systems. No significant difference was
increased variability in the sample (P . .05). The found between the 45u and 135u angles in any implant
Synthes OBA demonstrated significant differences system (Figure 12; Table 4).
between abutment head distances from the cortical Ormco VectorTAS implants placed at 45u or 135u
plate when comparing 4-mm to 6-mm distances and 4- insertion angles to the cortical plate demonstrated
mm to 8-mm distances. No significant difference in 29.6% and 33.3% reduction in force to fail when
force to fail was found between abutment head compared to the 90u insertion (P , .001). No
distances of 6 mm and 8 mm. In the Synthes OBA, significant difference was found between the 45u and
the 6-mm abutment distance from the cortical plate 135u angles. The Synthes OBA also had a significantly
required 24.7% less force before initial failure than the reduced force to fail when comparing implants placed
4-mm abutment distance (P , .05). The 8-mm at 45u or 135u insertion angles with those placed at a
90u insertion angle (P , .001). These differences
abutment distance from the cortical plate required
represented a 31.5% and 31.9% reduction in force to
31.4% less force before initial failure than the 4-mm
fail when compared with the 90u insertion, respectively.
abutment distance (P , .001). The Dentaurum tomas
No significant difference was found between the 45u
implant demonstrated significant reduction in force to
Table 2. Failure Loads (in Newtons) for Abutment Distances From

Table 1. Failure Loads (in Newtons) for Insertion Depths for Ormco the Cortical Plate for Ormco Vector, Synthes OBA, and Dentaurum
Vector, Synthes OBA, and Dentaurum Tomas Systems Tomas Systems
6-mm Insert 8-mm Insert 10-mm Insert 4-mm Abutment 6-mm Abutment 8-mm Abutment
Ormco Vector 5.963 6 0.336* 8.805 6 0.278 – Ormco Vector 5.963 6 0.336 5.057 6 0.699 –
Synthes OBA 5.729 6 0.073** 6.927 6 0.136 7.327 6 0.097 Synthes OBA 5.729 6 0.073* 4.316 6 0.196 3.186 6 0.131
Dentaurum Dentaurum
tomas 5.179 6 0.208** 7.856 6 0.097 8.159 6 0.058 tomas 5.179 6 0.208 4.623 6 0.122 3.925 6 0.156
* P , .05; ** P , .001. * P , .01.

Figure 11. Failure loads for abutment distance from cortical plate Figure 12. Failure loads for angular insertions 45u, 90u, and 135u.
deep insertion.
that implant failure increases the further the abutment is
and 135u angles. The Dentaurum tomas implants from the cortical plate.
placed at 45u and 135u insertion angles had 43.1% and As has been shown in previous studies,1,8–10 our
41.3% reduction in force to fail when compared with study found TAD stability increases with implant
the 90u insertion, respectively (P , .001). Again, no length, as long as the increased length is inserted in
significant difference was found between the 45u and bone. While implants in this study required more force
135u angles with the Dentaurum tomas implant. to fail as the insertion depth increased, the lowest initial
failure with a 6-mm implant was 461 grams, well above
Angular Oblique most traditional orthodontic force levels.20 In applica-
In evaluating the 45u oblique insertion, a significantly tions of high forces (including orthopedic forces) and
reduced force to fail was found when comparing to 90u simultaneous placement of multiple loads on the same
in all three implant systems (P , .0001; Figure 13; implant, longer implants may be necessary for
Table 5). retention. However, in most orthodontic force systems,
There was a 45.1% reduction in force to fail from 45u 6-mm implants appear to be sufficient as long as they
to 90u (P , .0001) using the Ormco VectorTAS are inserted 90u to the cortical plate. Because shorter
system, while there was a 47.8% reduction using the implants run less risk of damaging roots and their
Synthes OBA system (P , .0001). The Dentaurum supporting tissues and since they are also likely to
withstand most clinical orthodontic force levels, the 6-
tomas implant system demonstrated a 45.6% reduc-
mm length is appropriate for most applications.
tion in force to fail (P , .0001) when comparing the 45u
oblique and 90u insertion. Regarding abutment distance to the cortical plate,
our study confirms previous reports on the impact of
DISCUSSION abutment distance from the cortical plate.13,14 Implant
abutments 6 mm from the cortical plate failed at loads
The results of our study both confirmed and varied as low as 375 grams and those inserted 8 mm from the
from previous reports. Our results confirmed broad plate failed at loads as low as 282 grams. With this
clinical interpretations, such as the impact of length on pattern clearly established in our study, we recom-
implant success. Our results were also consistent in all mend that implants are inserted so their abutment
three implant systems, demonstrating the results are heads are as close to the cortical plate as possible.
not TAD specific. Using more clinically relevant Unfortunately, in areas of thick tissue this is not
definitions of failure, our results refute previous reports possible and increased torquing arms will be created.
of the impact of insertion angle on implant success.16 Knowing this, we tested if increasing insertion depth
Our results also confirm the hypothesis of Kyung et al.11 would compensate on the occasions where abutment
Table 4. Failure Loads (in Newtons) for Angular Insertions 45u, 90u,
Table 3. Failure Loads (in Newtons) for Abutment Distance From and 135u
Cortical Plate with Deep Insertion
45u 90u 135u
4-mm Abutment, 6-mm Abutment, 6-mm Abutment,
Ormco Vector 5.503 6 0.441* 8.805 6 0.278 5.212 6 0.564*
6-mm Insertion 6-mm Insertion 8-mm Insertion
Synthes OBA 4.743 6 0.116* 6.927 6 0.136 4.72 6 0.173*
Synthes OBA 5.729 6 0.073 4.316 6 0.196 4.239 6 0.131 Dentaurum
Dentaurum tomas 4.477 6 0.157* 7.856 6 0.097 4.611 6 0.168*
tomas 5.179 6 0.208 4.623 6 0.122 4.426 6 0.156
* P , .001.

inserting implants at an extreme angle to the cortical

plate to reduce the risk of root contact on insertion, this
recommendation significantly reduces the amount of
force the implant will withstand.11 In situations where
root proximity is a concern, insertion site should be
reconsidered or preimplant orthodontics should be
conducted to facilitate root divergence. The results of
all angular tests indicate placement of TADs at 90u is
suggested for optimal retention.
While the results of this study are definitive, they are
not without limitations. This study was conducted with
bone analogs and simulated loading. While this design
Figure 13. Failure loads for angular insertions 45u oblique and 90u.
reduces confounding variables, clinical investigation of
the principles germane to this study would increase
head distance needed to be increased. However, our
understanding. Readers should also be mindful this
data do not support a protective component to the
study focused on a single insertion location, specifi-
increased insertion depth in these situations. There-
cally chosen for the thin cortical plate and least dense
fore, we cannot suggest that increasing depth of
cancellous bone of the maxillary premolar region, and
insertion will markedly increase retention in situations
did not evaluate the effects of static loading over long-
where longer torquing arms are required.
term force application. Future investigations in areas of
The results of our study on the angle of insertion
various hard and soft tissue anatomy are needed.
differ from previously published reports.16 Though
Additionally, evaluations with different implant designs
previous reports have recommended insertion angles
may prove to be beneficial, especially implants with
with implants angled toward the force acted upon
tighter thread patterns and different pitches.
them, our study refutes these findings. In our study,
failure of the extreme angles of 45u and 135u was
CONCLUSIONS
found to occur within orthodontic force levels.20 Our
study also showed that 90u insertion to the cortical N Increased insertion depth increases retention,
plate is the most retentive insertion angle, contrary to though shorter implants should be sufficient in most
previously published reports.16 While previous reports orthodontic force systems if placed at 90u to the
found 45u to the force to be the most retentive, failure cortical plate.
was determined as total implant pull-out, well beyond N Increased abutment distance from the cortical plate
the point at which the implant would be of clinical decreases retention.
benefit. By defining failure as the point at which the N Placement at 90u to the cortical plate is the most
anchor unit begins to move, it is clear a 90u insertion to retentive insertion angle.
the cortical plate is the most retentive. Whenever N Insertion at an oblique angle from the line of force
possible, implants should be placed at 90u to the reduces retention.
cortical plate for maximum retention.
Oblique angles of 45u to the cortical plate also failed
ACKNOWLEDGMENTS
within orthodontic force levels.20 Because this off-axis
insertion and load relationship led to significant loss of The authors wish to thank Synthes, Ormco, and Dentaurum
retention at less than 45%, the amount of force for the for donations of the implants used in this study. Partial support
provided by NIH Grant K25AG022464.
90u insertion angle, this insertion angle is contraindi-
cated when medium to heavy orthodontic force levels
REFERENCES
are used. Although previous studies recommend
1. Park HS, Jeong SH, Kwon OW. Factors affecting the clinical
success of screw implants used as orthodontic anchorage.
Table 5. Failure Loads for Angular Insertions 45u Oblique & 90u - Am J Orthod Dentofacial Orthop. 2006;130:18–25.
Ormco Vector, Synthes OBA, and Dentaurum Tomas with 2. Gainsforth BL, Higley LB. A study of orthodontic anchorage
Standard Errors. possibilities in basal bone. Am J Orthod Oral Surg. 1945;31:
406–417.
Angular Oblique (N) 3. Creekmore TD, Eklund MK. The possibility of skeletal
90 degree 45 oblique anchorage. J Clin Orthod. 1983;17:266–269.
4. Roberts WE, Marshal KJ, Mozsary PG. Rigid endos-
Vector 4.787 6 0.574 8.805 6 0.278
seous implant utilized as anchorage to protract molars and
OBA 3.614 6 0.074 6.927 6 0.136
close an atrophic extraction site. Angle Orthod. 1990;60:
Tomas 3.377 6 0.131 7.856 6 0.097
135–152.

5. Wehrbein H, Merz BR, Diedrich P, Glatzmaier J. The use of used for orthodontic anchorage. Clin Oral Implants Res.
palatal implants for orthodontic anchorage. Design and 2005;16:473–479.
clinical application of the orthosystem. Clin Oral Implants 13. Costa A, Pasta G, Bergamaschi G. Intraoral hard and soft
Res. 1996;7:410–416. tissue depths for temporary anchorage devices. Semin
6. Kokich VG. Managing complex orthodontic problems: the Orthod. 2005;11:10–15.
use of implants for anchorage. Semin Orthod. 1996;2: 14. Kim HJ, Yun HS, Park HD, Kim DH, Park YC. Soft-tissue
153–160. and cortical-bone thickness at orthodontic implant sites.
7. Tseng YC, Hsieh CH, Chen CH, Shen YS, Huang IY, Chen Am J Orthod Dentofacial Orthop. 2006;130:177–182.
CM. The application of mini-implants for orthodontic 15. Goaslind GD, Robertson PB, Mahan CJ, Morrison WW,
anchorage. Int J Oral Maxillofac Surg. 2006;35:704– Olson JV. Thickness of facial gingiva. J Periodontol. 1977;
707. 48:768–771.
8. Cheng SJ, Tseng IY, Lee JJ, Kok SH. A prospective study of 16. Pickard MB, Dechow P, Rossouw PE, Buschang PE. Effects
the risk factors associated with failure of mini-implants used of mini-screw orientation on implant stability and resistance
for orthodontic anchorage. Int J Oral Maxillofac Implants. to failure. Am J Orthod Dentofacial Orthop. 2010;137:91–99.
2004;19:100–106. 17. Misch CE, Qu Z, Bidez MW. Mechanical properties of
9. Miyawaki S, Koyama I, Inoue M, Mishima K, Sugahara T, trabecular bone in the human mandible: implications for
Takano-Yamamoto T. Factors associated with the stability dental implant treatment planning and surgical placement.
of titanium screws placed in the posterior region for J Oral Maxillofac Surg. 1999;57:700–708.
orthodontic anchorage. Am J Orthod Dentofacial Orthop. 18. Peterson J, Wang Q, Dechow PC. Material properties of the
2003;124:373–378. dentate maxilla. Anat Rec A Discov Mol Cell Evol Biol. 2006;
10. Huja SS, Litsky AS, Beck FM, Johnson KA, Larsen PE. Pull- 288:962–972.
out strength of monocortical screws placed in the maxillae 19. Deguchi T, Nasu M, Murakami K, Yabuuchi T, Kamioka H,
and mandibles of dogs. Am J Orthod Dentofacial Orthop. Takano-Yamamoto T. Quantitative evaluation of cortical
2005;127:307–313. bone thickness with computed tomographic scanning for
11. Kyung HM, Park HS, Bae SM, Sung JH, Kim IB. orthodontic implants. Am J Orthod Dentofacial Orthop.
Development of orthodontic micro-implants for intraoral 2006;129;6:721.e7–721.e12.
anchorage. J Clin Orthod. 2003;37:321–328. 20. Ren Y, Maltha JC, Kuijpers-Jagtman AM. Optimum force
12. Buchter A, Wiechmann D, Koerdt S, Wiesmann HP, Piffko J, magnitude for orthodontic tooth movement: a systematic
Meyer U. Load-related implant reaction of mini-implants literature review. Angle Orthod. 2003;73:86–92.

Original Article
A cephalometric study to investigate the skeletal relationships in patients

with increasing severity of hypodontia
Priti N. Acharyaa; Steven P. Jonesb; David Molesc; Daljit Gilld; Nigel P. Hunte
ABSTRACT
Objectives: To determine the skeletal relationships in patients with hypodontia and analyze the
effects of severity and pattern.
Materials and Methods: Pretreatment lateral cephalograms from 277 patients with hypodontia,
categorized by the number of missing teeth as mild (1–2), moderate (3–5), or severe ($6), were
digitized recording angular measurements and ratios and compared with published norms matched
for age and gender. Pattern was determined as mandibular, maxillary, bimaxillary, bilateral,
anterior, posterior, and anteroposterior. Linear regression models assessed relationships between
number of missing teeth and cephalometric parameters, controlling for the pattern of hypodontia.
Results: For every additional missing tooth, SNA, SNB, and ANB decreased 0.3u, 0.1u, and 0.2u,
respectively; this was clinically significant for .4, .10, and .5 missing teeth, respectively.
Mandibular to cranial base ratio decreased 0.3% for every additional missing tooth; this was
clinically significant for .10 missing teeth. The MMPA decreased 0.3u for every additional missing
tooth; this was clinically significant for .7 missing teeth. Percentage LAFH decreased 0.2% for
every additional missing tooth; this was significant for .7 missing teeth. Jarabak ratio increased
0.2% for each additional missing tooth; this was clinically significant for .10 missing teeth. Anterior
hypodontia significantly decreased most cephalometric parameters.
Conclusions: Patients with hypodontia demonstrated a tendency toward a Class III relationship,
caused by decreased maxillary and mandibular angular prognathism and MnCB ratio, though the
effect was greater on the maxilla than the mandible. Clinical significance was only associated with
severe hypodontia. Vertically, there was a tendency toward decreased MMPA and %LAFH; this
was clinically relevant only with severe hypodontia. Anterior hypodontia had a significant effect on
skeletal relationship. (Angle Orthod. 2010;80:699–706.)
KEY WORDS: Hypodontia; Skeletal; Cephalometric analysis; Pattern
INTRODUCTION
Hypodontia is the developmental absence of one or
a
Specialist Registrar in Orthodontics, Eastman Dental Hospi-
more teeth, excluding third molars,1 and it is classified
tal, UCLH, London, UK. according to the number of missing teeth: mild for 1 or
b
Consultant Orthodontist, Eastman Dental Hospital, UCLH, 2 missing teeth,2 moderate for 3 to 5 missing teeth,2
London, UK. and severe where 6 or more permanent teeth are
c
Professor of Oral Health Services Research, Peninsula missing.3,4 A meta-analysis of white population surveys
Dental School, Plymouth, UK.
showed a mean gender prevalence of 5.5% in Europe,
d
Consultant Orthodontist, Eastman Dental Hospital, UCLH,
London, UK. 6.6% in Australia, and 3.9% in North America; 82.9%
e
Chairman of the Division of Craniofacial Development of affected individuals have mild hypodontia.5 Females
Sciences, Head of Orthodontic Unit, UCL Eastman Dental are 1.37 times more likely to have dental agenesis than
Institute, London, UK. males.5
Corresponding author: Professor Nigel P. Hunt, Orthodontic
The management of hypodontia often involves
Department, Eastman Dental Hospital, 256 Grays Inn Rd,
London, WC1X 8LD, UK orthodontic repositioning of the remaining teeth to
(e-mail: N.hunt@eastman.ucl.ac.uk) allow for strategic prosthetic replacements, and it is
Accepted: November 2009. Submitted: July 2009. best undertaken within a multidisciplinary team.1,3
G 2010 by The EH Angle Education and Research Foundation, Orthodontic treatment must be carried out within the
Inc. anatomic constraints of an individual’s three-dimen-

700 ACHARYA, JONES, MOLES, GILL, HUNT
sional dentoalveolar structure, and the ultimate aim is Table 1. Demographic Overview of the Hypodontia Sample
to create a harmonious facial form through dental and Parameter Value Percentage
skeletal changes. Devising a suitable long-term treat- Total number of patients 277
ment plan requires knowledge of the likely pattern of
Age (years)
skeletal development in individuals with hypodontia.
Mean 13.34
The aims of this study were to investigate the Median 13.34
vertical and antero-posterior facial skeletal relation- Standard deviation 1.78
ships in patients with increasing severity of hypodontia Minimum 8.47
and to determine whether the pattern of hypodontia Maximum 19.95
affects any specific component of facial form. Gender
Male 124 44.8%
MATERIALS AND METHODS Female 153 55.2%
Pattern of hypodontia
Ethical approval was sought and granted by the
Bilateral 226 81.6%
University College London Hospitals Research and Maxillary 222 80.1%
Ethics Committee. The sample consisted of patients Mandibular 206 74.4%
with hypodontia of their permanent dentition who were Bimaxillary 151 54.5%
treated in the Orthodontic Department at the Eastman Anterior 184 66.4%
Dental Hospital. Subjects were classified into the Posterior 208 75.1%
Anteroposterior 116 41.9%
following categories: mild (1 or 2 teeth missing),
moderate (3 to 5 teeth missing), and severe hypodon- Severity of hypodontia
tia ($6 teeth missing). The severity and distribution of Mild (1 or 2 teeth) 110 39.7%
Moderate (3–5 teeth) 82 29.6%
hypodontia were identified from dental panoramic Severe ($6 teeth) 85 30.7%
tomograms and supplemented with clinical notes.
A sample-size calculation was performed using
Altman’s nomogram and the standard deviation of N Total number of missing teeth
angular measurements from a UK sample of patients The pattern of hypodontia was classified as follows:
with hypodontia.6 Clinically meaningful differences for
ANB and MMPA were calculated as 1u greater than N Bilateral
their standard deviations.7 For horizontal measure- N Mandibular
ments, 70 patients in each severity group provided N Maxillary
96% power (a 5 .05). For vertical measurements, 70 N Bimaxillary
patients in each group provided 97% power (a 5 .05). N Anterior (incisors only)
N Posterior (canines, premolars, molars)
The following inclusion criteria were used: N Anteroposterior
N Pretreatment dental panoramic tomogram and lateral A total of 277 patients were included in the study
cephalogram of good quality, taken as part of routine (Table 1). Pretreatment cephalograms were analyzed
orthodontic treatment by direct digitization using a digitizer (Numonics Model
N Age 8–20 years old at time of cephalogram being IPS/BLG, Numonics Corp, 101 Commerce Drive,
taken Montgomeryville, Pa) linked to a computer installed
N White with a customized geometric digitizing program (GELA
Version 1.5, British Orthodontic Society, London, UK).
The following exclusion criteria were used:
The program prompted sequential identification of
N Craniofacial syndromes, including ectodermal dys- landmarks, calculating linear and angular measure-
plasia ments. Each radiograph was secured to a flat light box
N History of trauma to anterior teeth with surrounding light blocked out using a black
N Digit-sucking habits cardboard frame and digitized in dark ambient condi-
N Previous orthodontic treatment tions. For bilateral landmarks, the midpoint was
determined by hand tracing.8 No more than 10
The following additional data were captured:
radiographs were digitized in any session, to reduce
N Date of birth operator fatigue.
N Age at time of cephalogram (rounded down to the To eliminate the effect of magnification from the
nearest whole year) cephalograms, ratios and angular cephalometric mea-
N Gender surements were used to determine the vertical and A-
N Charting of missing teeth P skeletal relationships. The results were compared to

THE EFFECT OF HYPODONTIA ON SKELETAL PATTERN 701
Figure 1. Histogram of the number of missing teeth.
UK published norms for equivalent control groups, had maxillary or bilateral hypodontia (Table 1). A
matched for age and gender.9 histogram demonstrated that most patients had 1 or
Intraoperator error and repeatability were mea- 2 teeth missing; the most severely affected patient in
sured by digitizing each cephalogram twice, at least the sample had 18 missing teeth (Figure 1).
1 week apart, to avoid landmark memorization.10 The Systematic error was very small (0.1–0.2u) for SNA,
mean of both measurements was used in the final SNB, SNPog, SNMxP, FMPA, S-Ar-Go, and Ar-Go-Me,
statistical analysis. Systematic and random errors whereas N-S-Ar and MMPA showed systematic errors
were assessed using STATA 10.0 (Stata Corp, of 0.4u and 0.3u, respectively. The percentage face
Intercooled STATA 10.0, 1984–2007 for Windows, height and skeletal base ratios showed systematic
College Station, Tex) to calculate the paired t-test and errors from 0.1%–1.2%, although the A-P ratios were
the repeatability coefficient, respectively, and the more substantially affected (Table 2).
Bland and Altman method was used to calculate the Linear regression testing revealed that SNA, SNB,
combined error. Linear regression was used to ANB, and MMPA were significantly affected by the
calculate the effect of increase in the number of number of missing teeth; the effect for SNA and ANB
missing teeth on each cephalometric parameter. Any was highly statistically significant (P 5 .0001). SNA
parameters showing statistical significance were sub- decreased by 0.3u, SNB decreased by 0.1u, ANB
sequently subjected to multiple linear regressions decreased by 0.2u, and MMPA decreased by 0.3u for
using the number and pattern of missing teeth as the every additional tooth missing. The results for the
independent variables. %LAFH, Jarabak ratio, and MnCB ratio were also
significant, and %LAFH was highly statistically signif-
icant (P 5 .0001). The %LAFH decreased by 0.2%,
RESULTS
Jarabak ratio increased by 0.2%, and MnCB ratio
The sample consisted of 124 male and 153 female decreased by 0.3% for every additional tooth missing.
subjects with an age range of 8.47–19.95 years (mean All other measurements were unaffected by the
age 13.34 years). Just over 41% of subjects had severity of hypodontia (Table 3).
anteroposterior hypodontia, more than 54% had Multiple linear regressions for seven parameters that
hypodontia affecting both jaws, 66% had anterior had a statistically significant association with the
hypodontia, approximately 75% had mandibular hypo- number of missing teeth were further examined with
dontia or posterior hypodontia, and more than 80% respect to the pattern of hypodontia (Table 4). A

Table 2. Bland and Altman Table of Repeatability Testing

Systematic Error Random Error
Mean P value from Standard Deviation Coefficient of Limits of
Parameter Difference Paired t-Test of Differences Repeatability Agreement
SNA (degrees) 0.1 .002** 0.7 1.3 21.2, 1.5
SNB (degrees) 0.1 .0001**** 0.5 1.0 20.8, 1.1
ANB (degrees) 20.1 NS 0.5 1.0 21.0, 0.9
SNPog (degrees) 0.1 .0001**** 0.5 1.0 20.8, 1.1
SNMxP (degrees) 20.2 .0001**** 0.8 1.6 21.8, 1.4
SNMnP (degrees) 0.0 NS 0.9 1.7 21.7, 1.7
MMPA (degrees) 0.3 .0001**** 1.0 2.0 21.8, 2.3
FMPA (degrees) 0.1 NS 1.0 2.0 21.9, 2.1
NSAr (degrees) 20.4 .0001**** 1.5 3.0 23.4, 2.6
SArGo (degrees) 0.2 .023* 1.8 3.6 23.3, 3.8
ArGoMe (degrees) 0.2 .002** 1.0 2.0 21.8, 2.2
Björk’s sum angle (degrees) 0.0 NS 0.9 1.7 21.7, 1.7
%LAFH (percentage) 0.1 .028* 0.7 1.4 21.2, 1.4
%LPFH (percentage) 20.2 .041* 1.5 2.9 23.1, 2.7
Jarabak ratio (percentage change) 0.0 NS 1.0 2.0 21.8, 1.9
MxCB ratio (percentage change) 21.2 .0001**** 2.7 5.3 26.4, 4.0
MnCB ratio (percentage change) 20.2 .021* 1.8 3.5 23.7, 3.2
MxMn ratio (percentage change) 21.0 .0001**** 2.9 5.7 26.7, 4.7
* P 5 .05; ** P 5 .01; *** P 5 .001; **** P 5 .0001; NS 5 nonsignificant.
pattern of anterior hypodontia was associated with a for SNA, 0.2u for ANB, and 0.2% for %LAFH,
statistically significant reduction in all cephalometric respectively (P 5 .0001).
parameters, excluding the Jarabak ratio, and there For the other patterns of hypodontia, only six
were highly statistically significant reductions of 0.3u parameters were affected by the pattern of hypodontia
after adjusting for the number of missing teeth. The
Table 3. Univariate Effect of the Number of Missing Teeth on Jarabak ratio decreased by 1.5% with mandibular
Cephalometric Parameters Measured (Compared with Age- and hypodontia; MMPA increased by 2.3u with bimaxillary
Gender-matched Controls)
hypodontia; the Jarabak ratio and MnCB ratio de-
Unstandardized creased by 1.6% and 2.6%, respectively, with bimax-
Coefficient for 95%
illary hypodontia; MMPA increased by 1.9u with
the Number of Confidence
Parameter Missing Teeth Intervals P value posterior hypodontia; and the Jarabak ratio decreased
by 1.7% with posterior hypodontia . Excluding anterior
SNA (degrees) 20.3 20.4, 20.2 .0001****
SNB (degrees) 20.1 20.2, 0.0 .032* hypodontia, there was no consistent underlying rela-
ANB (degrees) 20.2 20.3, 20.1 .0001**** tionship between skeletal morphology and the pattern
SNPog (degrees) 0.0 20.1, 0.1 NS of hypodontia.
SNMxP (degrees) 0.1 20.0, 0.2 NS Regression analyses provide an indication of the
SNMnP (degrees) 20.2 20.4, 0.0 NS
MMPA (degrees) 20.3 20.5, 20.1 .004**
unit effect of hypodontia severity on various clinical
FMPA (degrees) 20.2 20.4, 0.0 NS parameters in terms of missing teeth. Clinicians more
NSAr (degrees) 0.0 20.1, 0.2 NS usually consider severity of hypodontia as a series of
SArGo (degrees) 20.2 20.4, 0.0 NS categories (mild, moderate, and severe). To aid clinical
ArGoMe (degrees) 20.1 20.2, 0.1 NS
interpretation of the results, those cephalometric
Björk’s sum angle
(degrees) 20.2 20.4, 0.0 NS measurements that showed a statistically significant
%LAFH (percentage) 20.2 20.2, 20.1 .0001**** linear association with the total number of missing
%LPFH (percentage) 0.1 0.0, 0.3 NS teeth (Table 3) were recategorized into the clinical
Jarabak ratio (percentage grades of severity. The anticipated typical ranges of
change) 0.2 0.0, 0.3 .029*
MxCB ratio (percentage
average effects on the clinical parameters for each
change) 20.1 20.3, 0.1 NS severity category are displayed in Figures 2 and 3.
MnCB Ratio (percentage
change) 20.3 20.5, 20.1 .006 * DISCUSSION
MxMn Ratio (percentage
change) 0.1 0.0, 0.3 NS The effect of hypodontia on mandibular and maxil-
* P 5 .05; ** P 5 .01; *** P 5 .001; **** P 5 .0001; NS 5 lary growth has been subject to relatively few research
nonsignificant. publications, perhaps because of the difficulties of

Table 4. Bivariate Effect of Pattern of Hypodontia Adjusted for the Table 4. Continued
Number of Missing Teeth on Cephalometric Parameters Measured
(Compared with Age- and Gender-matched Controls) Unstandardized
Coefficient for
Unstandardized Total Number of
Coefficient for Missing Teeth 95%
Total Number of Parameter and and Pattern of Confidence
Missing Teeth 95% Pattern of Hypodontia Hypodontia Intervals P value
Parameter and and Pattern of Confidence
Pattern of Hypodontia Hypodontia Intervals P value SNB (degrees) 20.8 21.8, 0.3 NS
ANB (degrees) 0.4 20.4, 1.2 NS
Mandibular MMPA (degrees) 1.9 0.3, 3.5 .021*
SNA (degrees) 20.4 21.5, 0.6 NS %LAFH (percentage) 0.0 20.6, 0.7 NS
SNB (degrees) 20.9 21.9, 0.2 NS Jarabak ratio
ANB (degrees) 0.5 20.3, 1.3 NS (percentage change) 21.7 23.0, 20.3 .015*
MMPA (degrees) 1.3 20.3, 2.9 NS MnCB ratio
%LAFH (percentage) 20.3 20.9, 0.4 NS (percentage change) 21.7 23.5, 0.1 NS
Jarabak ratio
Anteroposterior
(percentage change) 21.5 22.9, 20.2 .024*
MnCB Ratio SNA (degrees) 0.4 20.6, 1.4 NS
(percentage change) 21.6 23.4, 0.2 NS SNB (degrees) 0.2 20.8, 1.2 NS
Maxillary MMPA (degrees) 0.6 20.9, 2.1 NS
SNA (degrees) 20.3 21.4, 0.8 NS %LAFH (percentage) 0.2 20.4, 0.9 NS
SNB (degrees) 0.1 21.0, 1.2 NS Jarabak ratio
ANB (degrees) 20.4 21.3, 0.4 NS (percentage change) 0.4 20.8, 1.7 NS
MMPA (degrees) 0.6 21.1, 2.4 NS MnCB ratio
%LAFH (percentage) 0.6 20.1, 1.3 NS (percentage change) 21.1 22.7, 0.6 NS
Jarabak ratio a
Only those parameters that were statistically significant are
(percentage change) 0.2 21.2, 1.7 NS
reported here.
MnCB ratio
* P 5 .05; ** P 5 .01; *** P 5 .001; **** P 5 .0001. NS 5 non-
significant.
Bilateral
SNA (degrees) 20.5 21.7, 0.7 NS
SNB (degrees) 0.2 21.0, 1.3 NS obtaining a sufficiently large sample size, variation in
ANB (degrees) 20.5 21.4, 0.4 NS age groups and a range of cephalometric analyses
MMPA (degrees) 0.2 21.6, 2.1 NS used. Differences in the classification of the severity of
%LAFH (percentage) 0.0 20.7, 0.8 NS
hypodontia make comparisons difficult.
Jarabak ratio
(percentage change) 20.7 22.3, 0.8 NS To recruit a sufficient sample size, hypodontia was
MnCB Ratio initially classified by clinical severity. No consistent
(percentage change) 0.2 21.9, 2.3 NS pattern of classification exists, except for severe
Bimaxillary hypodontia with $6 teeth missing.3,4 In this study,
SNA (degrees) 20.9 22.0, 0.3 NS moderate hypodontia was classified as 3–5 missing
SNB (degrees) 20.9 22.1, 0.2 NS teeth and mild hypodontia as 1–2 missing teeth.2 The
pattern of hypodontia in previous studies used the
MMPA (degrees) 2.3 0.5, 4.0 .011**
%LAFH (percentage) 0.3 20.4, 1.1 NS tooth type missing6 or the affected jaw.11 The simple
Jarabak Ratio classification of mandibular, maxillary, bilateral, bimax-
(percentage change) 21.6 23.0, 20.1 .033* illary, anterior, posterior, and anteroposterior was
MnCB Ratio considered appropriate for this study. We used
(percentage change) 22.6 24.6, 20.6 .011**
published UK norms to permit direct comparison of
Anterior the hypodontia sample with controls matched for age,
SNA (degrees) 20.3 20.4, 20.2 .0001**** ethnicity, and gender.9
SNB (degrees) 20.1 20.3, 20.03 .018*
ANB (degrees) 20.2 20.3, 20.1 .0001****
Systematic error was very small (0.1u and 0.1%–
MMPA (degrees) 20.3 20.4, 20.1 .007** 0.2%) for all significant measurements (SNA, SNB,
%LAFH (percentage) 20.2 20.2, 20.1 .0001**** ANB, %LAFH, MnCB ratio) except the angle MMPA
Jarabak ratio (0.3u). Although Björk’s sum angle was not prone to
systematic error, its three component angles (N-S-Ar,
MnCB ratio
(percentage change) 20.3 20.5, 20.1 .005** S-Ar-Go, and Ar-Go-Me) were.
The present study found a linear correlation be-
Posterior
tween a reduction in SNA, SNB, and ANB and the
SNA (degrees) 20.4 21.4, 0.7 NS
number of missing teeth. The SNA angle decreased

Figure 2. Plots for angular measurements showing a statistically significant linear association with the total number of missing teeth, categorized
as mild, moderate and severe hypodontia.
more than the SNB angle, resulting in a decreased mandibular alveolar prognathism. Overall, it seems
ANB angle. Previous studies have reported SNA angle that hypodontia affects relative growth of the maxilla
reduction12–15 and maxillary retrusion.16 Maxillary base more than the mandible,11,16 but the effects are only
shortening is possibly attributed to the lack of bony evident with severe hypodontia.6,13 MMPA decreased
apposition in the tuberosity and anterior alveolar as the severity of hypodontia increased, which concurs
process, caused by molar and incisor agenesis.17 with previous studies.6,18 Some authors relate these
The present study found no direct relationship between changes to dental and functional compensation due to
hypodontia and maxillary base length, in keeping with the lack of posterior dental support.13,23 Although the
previous studies,11,13 although one study found a present study found no statistically significant associ-
biassociation of Nasion and A point retrusion, resulting ation between hypodontia and the SNMnP angle,
in a normal SNA angle.17 others have.13,22,24,25 Similarly, the present study found
Severe hypodontia has been associated with a no association between FMPA and the number of
decreased ANB13–15 angle and Class III skeletal missing teeth, which is in keeping with one previous
relationship,6,14,18–20 although previous studies have study20 but not others.14,15,26
highlighted the problems associated with drawing No association was found between hypodontia and
conclusions from ANB changes.18,21 Published results changes in the saddle angle (N-S-Ar), joint angle (S-
on the mandibular effects of hypodontia vary. Some Ar-Go), gonial angle (Ar-Go-Me), or sum of the three
report an increased mandibular corpus length and (Björk’s sum angle). In contrast, a previous study
SNB angle,16,19 whereas others have found no changes found a reduction in Björk’s sum angle25 and lower
in SNB,13 even in patients with $10 missing teeth. gonial angle (nasion-gonion-gnathion),25 which was not
Interestingly, although SNB decreased in the present measured in this study.
study, SNPog did not. In direct contrast, one study22 The %LAFH decreased, which was in keeping with
reported increased mandibular skeletal prognathism several previous studies.13,25 The Jarabak ratio in-
as assessed by the SNPog angle but decreased creased to a clinically significant level when .10 teeth

Figure 3. Plots for ratio measurements showing a statistically significant linear association with the total number of missing teeth, categorized as
mild, moderate and severe hypodontia.
were missing, indicating that in the presence of severe effects of total hypodontia on the A-P and vertical
hypodontia the total posterior face height increases, skeletal relationship. However, it should be noted that
the total anterior face height decreases, or a combi- none of the parameters showed a significant correla-
nation of both occurs. This demonstrates a closing tion to anteroposterior hypodontia.
growth rotation in individuals with severe hypodontia, One other previous study12 categorized the pattern
as previously reported.13,24 of hypodontia in a similar fashion but limited it to
Given the magnitude of vertical and A-P skeletal hypodontia affecting the mandible only or maxilla only,
differences seen, the clinical relevance of the effect of and there was no gender differentiation. Individuals
hypodontia on each of the seven statistically significant with maxillary hypodontia had significantly smaller
cephalometric parameters is likely to manifest only in SNA and ANB angles and increased N-A-Pog angles.
patients with severe hypodontia. Underlying relation- Mandibular hypodontia was associated with significant
ships were found between the pattern of hypodontia reduction of ANB. The authors published linear
and each cephalometric parameter, and there was a measurements of vertical face height, which cannot
statistically significant association to the number of be directly related to the present findings. The authors
missing teeth. Of the 49 possible bivariate associations concluded that patients with hypodontia had a slightly
tested, 12 showed statistical significance. Interestingly, increased face height and that hypodontia affects A-P
half of these were associated with a pattern of anterior maxillary growth most, irrespective of gender or
hypodontia, and the results of the present study seem pattern. A later study reported no significant cephalo-
to suggest that the effects of anterior hypodontia are metric changes associated with hypodontia but found a
entirely responsible for the reduction in the ANB angle. significant decrease in maxillary jaw size associated
Also, three of four bivariable associations involved with maxillary hypodontia. No significant associations
bimaxillary hypodontia. This probably reflects the were found for mandibular tooth agenesis. Bolton
severity of hypodontia, as seen with the univariate growth study templates were used for comparison.11 A

more recently published study27 categorized pattern as 9. Bhatia SN, Leighton BC. A Manual of Facial Growth.
anterior, posterior, and anteroposterior and tested for A Computer Analysis of Longitudinal Cephalometric Growth
Data. Oxford, UK: Oxford University Press / Oxford Medical
their effect on three cephalometric parameters that Publications; 1993.
were also measured in the present study: SNA, SNB, 10. You QL, Hägg U. A comparison of three superimposition
and ANB. The authors of that study also concluded methods. Eur J Orthod. 1999;21:717–725.
that an anterior hypodontia pattern has a predominant 11. Tavajohi-Kermani H, Kapur R, Sciote JJ. Tooth agenesis
effect on the dentoskeletal pattern. and craniofacial morphology in an orthodontic population.
Am J Orthod Dentofacial Orthop. 2002;122:39–47.
12. Wisth PJ, Thunold K, Böe OE. The craniofacial morphology
CONCLUSIONS of individuals with hypodontia. Acta Odontol. Scand. 1974;
N With increasing severity of hypodontia, the A-P 32:281–290.
13. Øgaard B, Krogstad O. Craniofacial structure and soft tissue
skeletal relationship demonstrated a tendency to-
profile in patients with severe hypodontia. Am J Orthod
ward a Class III pattern. This was caused by a Dentofacial Orthop. 1995;108:472–477.
combination of a decrease in maxillary and mandib- 14. Chan DWS, Samman N, McMillan AS. Craniofacial profile in
ular angular prognathism and mandibular length in southern Chinese with hypodontia. Eur J Orthod. 2009;31:
ratio to the anterior cranial base length. 300–305.
N Hypodontia affected the A-P size of the maxilla more 15. Ben-Bassat Y, Brin I. Skeletal and dental patterns in patients
with severe congenital absence of teeth. Am J Orthod
than the mandible, although these effects were
clinically significant only in the presence of severe 16. Roald KL, Wisth PJ, Bøe OE. Changes in cranio-facial
hypodontia. morphology of individuals with hypodontia between the ages
N The vertical skeletal pattern with increasing severity of 9 and 16. Acta Odontol. Scand. 1982;40:65–74.
of hypodontia demonstrated a decreased MMPA and 17. Endo T, Yoshino S, Ozoe R, Kojima K, Shimooka S.
%LAFH. This was only clinically relevant in severe Association of advanced hypodontia and craniofacial mor-
phology in Japanese orthodontic patients. Odontology.
hypodontia and may be attributed to a lack of
2004;92:48–53.
posterior occlusal support. 18. Woodworth DA, Sinclair PM, Alexander RG. Bilateral
N A pattern of anterior hypodontia had a significant congenital absence of maxillary lateral incisors: a craniofa-
effect on the A-P and vertical skeletal relationships cial and dental cast analysis. Am J Orthod Dentofacial
with increasing severity of hypodontia. Orthop. 1985;87:280–293.
19. Bondarets N, Jones RM, McDonald F. Analysis of facial
growth in subjects with syndromic ectodermal dysplasia:
ACKNOWLEDGMENT a longitudinal analysis. Orthod Craniofac Res. 2002;5:
71–84.
The GELA program was kindly written by Mr Adrian Hart, 20. Chung CJ, Han JH, Kim KH. The pattern and prevalence of
Consultant Orthodontist, Raigmore Hospital, Inverness, UK. hypodontia in Koreans. Oral Dis. 2008;14:620–625.
21. Hussels W, Nanda RS. Analysis of factors affecting angle
REFERENCES ANB. Am J Orthod Dentofacial Orthop. 1984;85:411–
423.
1. Goodman JR, Jones SP, Hobkirk JA, King PA. Hypodontia:
22. Nodal M, Kjær I, Solow B. Craniofacial morphology in
1. Clinical features and the management of mild to moderate
patients with multiple congenitally missing permanent teeth.
hypodontia. Dent Update. 1994;21:381–384.
Eur J Orthod. 1994;16:104–109.
2. Dhanrajani PJ. Hypodontia: etiology, clinical features, and
management. Quintessence Int. 2002;33:294–302. 23. Dermaut LR, Goeffers KR, De Smit AA. Prevalence of tooth
3. Hobkirk JA, Brook AH. The management of patients with agenesis correlated with jaw relationship and dental
severe hypodontia. J Oral Rehabil. 1980;7:289–298. crowding. Am J Orthod Dentofacial Orthop. 1986;90:
4. Hobkirk JA, King PA, Goodman JR, Jones SP. Hypodontia: 204–210.
2. The management of severe hypodontia. Dent Update. 24. Sarnäs KV, Rune B. The facial profile in advanced
1995;22:8–11. hypodontia: a mixed longitudinal study of 141 children.
5. Polder BJ, Van’t Hof MA, Van der Linden FPGM, Kuijpers- Eur J Orthod. 1983;5:133–143.
Jagtman AM. A meta-analysis of the prevalence of dental 25. Bondarets N, McDonald F. Analysis of the vertical facial
agenesis of permanent teeth. Community Dent Oral form in patients with severe hypodontia. Am J Phys Anthro-
Epidemiol. 2004;32:217–226. pol. 2000;111:177–184.
6. Chung LKL, Hobson RS, Nunn JH, Gordon PH, Carter NE. 26. Endo T, Ozoe R, Yoshino S, Shimooka S. Hypodontia
An analysis of the skeletal relationships in a group of young patterns and variations in craniofacial morphology in
people with hypodontia. J Orthod. 2000;27:315–318. Japanese orthodontic patients. Angle Orthod. 2006;76:
7. Mills JRE. Principles and Practice of Orthodontics. 2nd ed. 996–1003.
Edinburgh, UK: Churchill Livingstone; 1987:80. 27. Ben-Bassat Y, Brin I. Skeletodental patterns in patients with
8. Sandler PJ. Reproducibility of cephalometric measure- multiple congenitally missing teeth. Am J Orthod Dentofacial
ments. Br J Orthod. 1988;15:105–110. Orthop. 2003;124:521–525.

Original Article
Digital models vs plaster models using

alginate and alginate substitute materials
Gilda Torassiana; Chung How Kaub; Jeryl D. Englishc; John Powersd; Harry I. Bussae;
Anna Marie Salas-Lopeze; John A. Corbettf
ABSTRACT
Objective: To compare the dimensional stability of four impression materials over time and to
compare OraMetrix digital models vs traditional plaster models.
Materials and Methods: Two traditional alginates (Identic and imprEssix) and two alginate
substitutes (Alginot FS and Position PentaQuick) were used to take multiple impressions of a
maxillary typodont. Fifteen impressions for each material were taken and poured with plaster at
three time points: 72 hours, 120 hours, and 1 week. Five impressions for each material were taken
and were sent to OrthoProof for digital model reproduction at 72 hours. Digital models were then
integrated with OraMetrix software. Plaster and digital models were measured in the anterior-
posterior, transverse, and vertical dimensions. The control typodont and plaster models were
measured using a digital caliper, and digital models were measured using OraMetrix software.
Results: Statistically significant changes were found for models replicated from Identic impression
material in all three dimensions by 72 hours. Statistically significant changes were seen in
imprEssix impressions in the vertical and intercanine dimensions. Digital models were significantly
smaller in all dimensions compared with plaster models and the control.
Conclusions: Identic impression material showed a statistically and clinically significant change in
all dimensions within 72 hours and therefore should not be used if impressions are not going to be
poured immediately. Alginate substitutes were dimensionally stable over an extended period.
Digital models produced by OraMetrix were not clinically acceptable compared with plaster models.
(Angle Orthod. 2010;80:662–669.)
KEY WORDS: Digital models; Impression materials
INTRODUCTION
a
Resident, Department of Orthodontics, University of Texas
Health Science Center at Houston, Houston, Tex. Diagnosis, the single most important phase of
b
Professor and Department Chair, Department of Orthodon- orthodontic treatment, is dependent on accurate and
tics, University of Alabama, Birmingham, Ala. reliable orthodontic records. The vital information
c
Professor and Chair, Department of Orthodontics, University
of Texas Health Science Center at Houston, Houston, Tex.
required to diagnose a malocclusion and develop an
d
Adjunct Professor in the Department of Orthodontics, orthodontic treatment plan consists of models, photo-
University of Texas Health Science Center at Houston, Houston, graphs, panoramic and lateral cephalometric radio-
Tex. graphs, and a clinical examination.1 Digital technology
e
Clinical Associate Professor, Department of Orthodontics, has made significant changes in orthodontics. Digital
University of Texas Health Science Center at Houston, Houston,
Tex.
photography and radiographs are rapidly replacing
f
Clinical Assistant Professor, Department of Orthodontics, traditional methods. The progression to a completely
University of Texas Health Science Center at Houston, Houston, ‘‘paperless office’’ has incorporated the use of digital
Tex. models, records, consents, and financial agreements.
Corresponding author: Dr Chung How Kau, Professor and
Digital models have eliminated the need for storage
Department Chair, University of Alabama. 1919 7th Avenue
South, SDB305, Birmingham, AL 35294 space and have made retrieval and transfer of models
(e-mail: chung.h.kau@inbox.com) easier. These three-dimensional models can be easily
Accepted: October 2009. Submitted: July 2009.
manipulated to gather measurements to facilitate
G 2010 by The EH Angle Education and Research Foundation, diagnosis and treatment planning. With the numerous
Inc. advantages of digital models and the progression to a

COMPARISON OF DIGITAL MODELS VS PLASTER MODELS 663
‘‘paperless office,’’ digital models will replace tradition- maxillary typodont, and measurements were made
al plaster models.2–5 using the digital caliper directly from the typodont.
Currently, three companies offer computer-based Anterior-posterior, transverse, and vertical measure-
three-dimensional models: SureSmile (OraMetrix Inc, ments were taken five times on the typodont and were
Dallas, Tex), OrthoCAD (Cadent Inc, Carlstadt, NJ), averaged.
and E-Models (GeoDigm Corporation Inc, Chanhas- Four different impression materials were used,
sen, Minn). Diagnostic impressions of the patient’s ranging from traditional alginates to alginate-alterna-
dentition are mailed to the company, and the impres- tive materials:
sions are scanned using various technologies unique
1. Material 1—Identic Alginate (Dux Dental, Oxnard,
to the company. These scanned images are uploaded
Calif), a traditional alginate
to the company software, allowing viewing of the three-
dimensional models. The clinician, using the com- 2. Material 2—imprEssix Color Change Alginate (Dents-
pany’s software program, can access these images. ply, York, Pa), a color change traditional alginate
The program also allows the orthodontist to perform 3. Material 3—Alginot FS (Kerr USA, Romulus, Mich),
measurements and manipulate the models to achieve an alginate-alternative material
proper diagnosis. 4. Material 4—Position PentaQuick (3M ESPE Dental
The accuracy of the digital models depends initially Products, Seefeld, Germany), an alginate-alterna-
on the accuracy of the impression. Typically, an tive material
irreversible hydrocolloid or alginate material is used
for orthodontic diagnostic models because of inexpen- Impressions were taken according to manufacturers’
siveness, ease of use, and relative accuracy. Howev- recommendations, including the recommended tray
er, alginate does have a short-term dimensional adhesives for each material. Identic and imprEssix
stability. Impression materials such as polyether and Color Change Alginates were mixed with a mechanical
polyvinyl siloxane (PVS) have been used for their mixer, the Alginator II (Dux Dental).
accuracy and extended dimensional stability, but these Three time intervals were used to evaluate the
materials are significantly more expensive. With plaster models, as these are reasonable time intervals
advances in dental materials, manufacturers have from the time impressions are taken until they can be
created alginate substitutes that incorporate the longer reproduced as plaster or digital models:
dimensional stability of polyether and PVS without a
significant price difference. 1. T1—72 hours
As digital models become more common, and as 2. T2—120 hours
advances in dental materials introduce new impression 3. T3—1 week
products, more research needs to be done. This study
will examine the three-dimensional accuracy of mea- Digital models were evaluated only at T1, as
surements made on digital models using OraMetrix material changes over time could be seen in plaster
and traditional plaster models using four different models.
impression materials. The materials will also be
evaluated for their dimensional stability over three
Technique for Model Representation
time periods.
Digital models (Twenty impressions were taken for
Clinical Application digital models.)
With the increased use of digital diagnostic models, N Five impressions per material
the purpose of this study was to compare digital and N Each impression was scanned three times; however,
plaster models using various alginate and alginate- evaluated only at T1
alternative impression materials. Clinicians can use
the findings to determine whether digital or plaster
Impressions were packaged according to OraMetrix
models are comparable in diagnostic accuracy and
(Richardson, Tex) guidelines and were shipped to
which impression material will provide a diagnostically
OrthoProof USA (Albuquerque, NM) via 2-day shipping.
accurate model.
Digital models were created using proprietary FlashCT
cone beam computed tomography (CBCT) technology
(Hytec Inc, Los Alamos, NM). This is a patented
Eighty single-arch impressions were taken of a technology that requires the impressions to be scanned
standard maxillary typodont (Kilgore International Inc, using CBCT technology based on the interaction of the
Coldwater, Mich). The control used was the standard radiation and the impression material. Digital files were

664 TORASSIAN, KAU, ENGLISH, POWERS, BUSSA, SALAS-LOPEZ, CORBETT
Figure 1. Digital measurements.
sent to OraMetrix, and digital models were viewed and the MF line angle of the respective central incisor
measured using SureSmile v5.6 software. Measure- (Figure 1a).
ments were made in all three dimensions: anterior-
posterior, transverse, and vertical. Transverse.
N The typodont, digital, and plaster models were
Plaster Models measured from the central pit of the first molar to
Sixty impressions were taken for plaster models. the central pit of the contralateral first molar. The
typodont and models were also measured from the
N 5 impressions for each material for each time cusp tip of the canine to the contralateral canine cusp
interval tip (Figure 1b).
N Impressions were poured at the appropriate time
intervals with 0–67 Snow White plaster (Heraeus Vertical.
Kulzer Inc, South Bend, Ind) using the manufactur-
er’s recommendations of 100 g plaster to 26 mL N The typodont, digital, and plaster models were
water, vacuum mixed for 15 seconds. measured from the incisal edge at the midline of
the maxillary right and left central incisors to the
gingival margin (Figure 1c).
Plaster models were measured using digital calipers
(accuracy 6 0.03 mm). Measurements were taken in
all three dimensions: anterior-posterior, transverse, Parameters Measured
and vertical.
The following parameters were evaluated:
Dimensions Evaluated N Quality of plaster models produced from different
impression materials compared with control mea-
Anterior-posterior.
surements
N The typodont, digital, and plaster models were N Effects of time on the dimensional stability of the
measured from the central pit of the first molar to impression material used

N Comparison of control typodont measured with over time. Material 3 (Alginot FS) and Material 4
digital calipers, plaster models measured with digital (PentaQuick) did not show any significant changes
calipers, and digital models measured with OraMetrix over time.
software At T1, Material 1 showed the greatest decrease in
measurements compared with the control in all
dimensions. Although some increases in measure-
All digital models were compared with the control
ments were seen at T2 and T3, overall the measure-
model, and all digital models were compared with the
ments were significantly smaller for Material 1 (Fig-
plaster models.
ure 2a through 2c).
Statistical Evaluations
The Effect of Time on the Dimensional Stability of
Differences in impression material over time were the Impression Material Used
evaluated for plaster models compared with the
The materials in each dimension that showed a
control using a multilevel mixed-effect linear regres-
significant change in Table 1a were compared in
sion. The vertical, transverse, and anterior-posterior
Table 1b at T1 and T3 using a paired t-test. This was
dimensions were evaluated for each material for
done to see whether there was a significant change in
each time point. Materials that showed a significant
measurements at the two extreme measurement
change (P , .05) in any dimension compared with
points. As can be seen in Table 1b, there was no
the control were evaluated at T1 and T3 using a
significant change for Material 1 in any of the
paired t-test. Technique differences were compared
dimensions between T1 and T3. Therefore it is evident
at T1 to determine whether significant differences
that the significant dimensional change for Material 1
existed between plaster and digital models. The
occurred from the control to T1. Material 2 showed a
control was compared vs digital models for each
similar result, except for the change in the left vertical
material and dimension at T1. The plaster and
measurement, which had a significant change from T1
digital models were compared for each material and
to T3.
dimension at T1.
Comparisons were done using Kruskal-Wallis Comparison of Control Typodont Measured with
equality of populations rank test and one-way Digital Calipers, Plaster Models Measured with
analysis of variance. All analyses were performed Digital Calipers, and Digital Models Measured with
with STATA 10.0 (StatCorp, College Station, Tex). OraMetrix Software
The following results were obtained in the digital
RESULTS
models compared with control and plaster vs digital
Error studies were carried out on the various models.
methods based on the repeated measures of one Digital models compared with control. Comparisons
observer. The error was less than 0.5 mm and was were done for each dimension at T1 for each
found to be not statistically significant (P , .05). material. Intercanine and intermolar dimensions
Measurements in the anterior-posterior, transverse, showed a statistically significant difference, with
and vertical dimensions were obtained and de- digital models having a smaller dimension. Anterior-
scribed in accordance with the parameters de- posterior dimension also showed a statistically
scribed. significant difference, with digital models having a
smaller dimension. However, the left anterior-poste-
rior dimension for Material 2 and the right anterior-
The Quality of Plaster Models Produced from
posterior measurement for Material 1 were not
Different Impression Materials Compared with
statistically significant. Right vertical measurements
Control Measurements
showed a statistically significant difference for all
The mean of the five plaster models for each materials, with digital models having a smaller
material for each time period was compared with the dimension compared with the control. Only Materials
control measurements using a multilevel mixed-effect 1 and 2 showed a statistically significant difference in
linear regression. Compared with the control measure- left vertical measurements, with digital models having
ment (Table 1a), Material 1 (Identic alginate) showed a a smaller dimension. Overall the control measure-
statistically significant (P , .05) difference in all ments in all dimensions were larger than the digital
dimensions. Material 2 (imprEssix) showed a statisti- measurements.
cally significant change in intercanine width and Comparison of plaster and digital models at T1.
vertical measurements as measurements decreased Using a one-way analysis of variance and a Kruskal-

Table 1a. Plaster Technique—Differences in Impression Materials Over Time

Control 72 Hours 120 Hours 1 Week
Mean (SD) Mean (SD) Mean (SD) Mean (SD) P Valuea
Intercanine
Control vs Material 1 36.93 (0.11) 36.21 (0.24) 36.44 (0.22) 36.40 (0.36) ,.001
Control vs Material 2 36.93 (0.11) 36.78 (0.25) 36.72 (0.15) 36.75 (0.18) .013
Intermolar
Left Anterior-Posterior Measurements
Right Anterior-Posterior Measurements
Left Central
Right Central
a
Multilevel mixed-effects linear regression.
Wallis equality of populations rank test, plaster and Table 1b. Plaster Technique—Mean Differences in Impression
digital techniques for the four materials were compared Materials Over Time
at T1. Significant findings were found for each 72 Hours 1 Week
dimension with various materials. Materials 2, 3, and Mean (SD) Mean (SD) P Valuea
4 were significantly different in terms of intercanine and Intercanine
intermolar measurements. Materials 3 and 4 were Material 1 36.21 (0.24) 36.40 (0.36) .1864
significantly different for right and left anterior-posterior Material 2 36.78 (0.25) 36.75 (0.18) .8321
measurements. Materials 1 and 2 were significantly Intermolar
different for the left central measurement, and all Material 1 47.80 (0.13) 47.76 (0.25) .7338
materials were significantly different for the right
Left Anterior-Posterior Measurements
central measurement. Plaster measurements were
Material 1 39.06 (0.31) 38.67 (0.16) .0927
always larger than digital measurements (Figure 3a
through 3c). Right Anterior-Posterior Measurements
Material 1 38.61 (0.03) 38.70 (0.08) .0562
Left Central
DISCUSSION
Material 1 10.67 (0.08) 10.66 (0.15) .8993
The results of this study show that material and Material 2 10.98 (0.08) 10.58 (0.13) .0050
time are crucial for the dimensional stability of Right Central
impression materials, and the technique used can Material 1 10.83 (0.13) 10.81 (0.12) .8807
introduce variability among measurements. Many Material 2 10.92 (0.10) 10.77 (0.15) .1292
impression materials are dimensionally stable only a
Paired t-test.

Figure 2. (a) Plaster technique: Differences in impression material over time in transverse dimension: Only Materials 1 and 2 showed a
significant change in intercanine dimension, showing a decrease in measurement. Only Material 1 showed a significant decrease in intermolar
measurements. (b) Plaster technique: Differences in impression material over time in anterior-posterior dimension: Only Material 1 showed a
significant change compared with the control. In both left and right anterior-posterior dimensions, Material 1 showed a decrease in dimension. (c)
Plaster technique: Differences in impression material over time in vertical dimension: Both Materials 1 and 2 showed a significant change in
vertical dimension compared with the control. Both Materials 1 and 2 showed a decrease in vertical dimension.
for immediate use; however, with digital model found that traditional alginates showed dimensional
services, turnaround time can be about 7 days. instability in the vertical dimension within 72 hours.
Therefore the dimensional stability of materials Similar results are seen in this study, as Identic
typically used for diagnostic models must be evaluat- showed as much as 0.72 mm of dimensional change
ed for a more extended period. within 72 hours—a statistically significant change.
Tennison et al.6 compared the dimensional stability As was discussed in the Results, statistical tests
of various alginates at 1 hour, 24 hours, 48 hours, 72 were found to be significant if P # .05. However, in a
hours, and 120 hours. It was found that Identic clinical setting, according to the American Board of
exhibited shrinkage at all time points, ranging from Orthodontics Objective Grading System (ABO OGS),
0.92% at 1 hour to 2.81% at 120 hours. In a study discrepancies in the vertical, transverse, and anterior-
that compared traditional alginate (Kromopan N100, posterior dimensions greater than 0.5 mm are consid-
Kromopan USA, Inc, Des Plaines, Ill) vs alginate ered to be significant.6 When the impression materials
substitutes and PVS impression materials,7 it was used for plaster models were compared with the

Figure 3. (a) Differences between plaster and digital techniques for all materials at T1 for transverse dimensions: Materials 2, 3, and 4 were
significantly different, with digital model measurements being smaller. (b) Differences between plaster and digital techniques for all materials at
T1 for anterior-posterior measurements: Only Materials 3 and 4 were significantly different for both right and left anterior-posterior dimensions
when plaster models were compared with digital models. (c) Differences between plaster and digital techniques for all materials at T1 for vertical
measurements: All materials were significantly different between plaster and digital models for the right vertical measurement. Only Materials 1
and 2 were significantly different for the left vertical measurement.
control measurements, only Material 1 (Identic) models produced by OrthoCad or E-models (Geo-
showed a statistically and clinically significant de- Digm). When mesiodistal tooth width and intercanine
crease in all dimensions measured. Material 1 showed and intermolar width were measured, digital models
a significant change at T1; therefore a significant were found to be clinically acceptable. Studies done to
dimensional change in Material 1 is seen within 72 compare plaster models graded by the ABO OGS vs
hours. Although Material 2 did show statistically OrthoCad digital models graded by the ABO OGS
significant differences in the transverse and vertical found significant differences in scores; therefore it was
dimensions, the differences were not clinically signif- concluded that the OrthoCAD software was not
icant. adequate for scoring all parameters required by the
Numerous studies have been done comparing ABO OGS. No studies could be found that used digital
plaster models measured by digital calipers vs digital models by OrthoProof and OraMetrix software. Ortho-

Proof does not release detailed technical information CONCLUSIONS

regarding scanning and digitizing methods used. One
N Material choice and model replication technique are
has to assume that OrthoProof scanned the impres-
very important because they can introduce clinically
sions as instructed based on the strict guidelines of the
relevant errors.
study.
N A long-term dimensionally stable alginate is neces-
When the plaster and digital models were compared,
sary if impressions are not going to be poured or
overall the digital model measurements were smaller
scanned in a timely fashion.
compared with the plaster model measurements.
N Digital models measured with OraMetrix software
Differences between the measurements were greater showed a clinically significant difference compared
than 0.5 mm; therefore a clinically significant differ- with traditional plaster models.
ence is seen between plaster and digital models.
Because the control typodont was measured using
digital calipers, similar to the plaster models, the REFERENCES
control measurements were closely correlated with 1. Han UK, Vig KW, Weintraub JA, Vig PS, Kowalski CJ.
the plaster measurements. One possible reason for Consistency of orthodontic treatment decisions relative to
this dimensional discrepancy may be the method of diagnostic records. Am J Orthod Dentofacial Orthop. 1991;
image capture. Radiation sources have produced a 100:212–219.
2. Joffe L. Current products and practices OrthoCADTM: digital
burnout effect on the periphery that may have models for a digital era. J Orthod. 2004;31:344–347.
contributed to a smaller reading. 3. Mayers M, Firestone AR, Rashid R, Vig KW. Comparison of
The control typodont was not scanned and digitized, peer assessment rating (PAR) index scores of plaster and
and control measurements were not taken with OraMe- computer-based digital models. Am J Orthod Dentofacial
Orthop. 2005;128:431–434.
trix software; this may have introduced the difference
4. Quimby ML, Vig KW, Rashid RG, Firestone AR. The
between digital and plaster models. The difference in accuracy and reliability of measurements made on comput-
plaster and digital measurements may be due to the er-based digital models. Angle Orthod. 2004;74:298–303.
ability to magnify the scanned image on the screen and 5. Bell A, Ayoub AF, Siebert P. Assessment of the accuracy of a
the ability to cross-section the model to locate points. three-dimensional imaging system for archiving dental study
models. J Orthod. 2003;30:219–223.
Also, digital models were scanned only at T1 and were 6. Tennison J, English JD, Bussa H, Powers J, Frey G, Duke J.
compared with control and plaster models produced at Dimensional Stability of Orthodontic Alginates [master’s
T1. This was done to prevent dimensional changes in thesis]. Houston, Tex: University of Texas Dental Branch;
the impression materials that may introduce variations 2007.
in measurements between digital and plaster models. 7. Shambarger JH, English JD, Darsey D, Powers JM, Frey GN,
Lee RP, Bussa HI. The accuracy of OrthoCAD digital models
Future trials can study digital models at various time compared to plaster models. Presented at: International
points to determine whether time may have affected Association for Dental Research Annual Meeting; April 3,
digital model measurements. 2008; Dallas, Tex. Poster #0382.

Original Article
Feeding and nonnutritive sucking habits and prevalence of

open bite and crossbite in children/adolescents with Down syndrome
Ana Cristina Oliveiraa; Isabela Almeida Pordeusb; Cintia Silva Torresc; Milene Torres Martinsc;
Saul Martins Paivab
ABSTRACT
Objective: To analyze the influence of breastfeeding, bottle feeding, and nonnutritive sucking
habits on the prevalence of open bite and anterior/posterior crossbite in children with Down
syndrome (DS).
Materials and Methods: A cross-sectional study was carried out in 112 pairs of mothers/children
with DS between 3 and 18 years of age at a maternal/children’s hospital in Rio de Janeiro, Brazil.
The children with DS were clinically examined for the presence of open bite as well as anterior and
posterior crossbite. Information on breastfeeding, bottle feeding, and nonnutritive sucking habits
was collected using a structured questionnaire. The control variables were age and mouth posture
of children/adolescents and mother’s schooling. Statistical analysis of the data was performed
using the chi-square test and multiple logistic regression.
Results: The prevalence of anterior open bite was 21%, anterior crossbite was 33%, and posterior
crossbite was 31%. The use of bottle feeding for more than 24 months (prevalence ratio [PR] 5
1.6) was associated with the occurrence of open bite. Having breastfed for less than 6 months (PR
5 1.4) and pacifier sucking for more than 24 months (PR 5 3.1) were associated with the
prevalence of anterior crossbite. Finger sucking (PR 5 2.9) and the use of bottle feeding for more
than 24 months (PR 5 2.6) were associated with posterior crossbite.
Conclusion: The prevalence of open bite and crossbite in children with DS was associated with
the use of bottle feeding and pacifier sucking for more than 24 months, breastfeeding for less than
6 months, and finger sucking. (Angle Orthod. 2010;80:748–753.)
KEY WORDS: Down syndrome; Malocclusion; Open bite; Crossbite; Breastfeeding; Sucking
behavior
INTRODUCTION mosomal abnormality recognized in humans.1,2 Chil-

dren with DS have insufficient bone development
Down syndrome (DS) is the best known of all
associated with muscle hypotonicity. This becomes
malformation syndromes, since it was the first chro-
more accentuated with age and leads to a greater
occurrence of malocclusions in comparison to the
a
Professor, Department of Social and Preventive Dentistry, general population.3–6
Faculty of Dentistry, Federal University of Minas Gerais, Minas These orofacial disorders have a negative impact on
Gerais, Brazil. the daily life of individuals with DS by stigmatizing them
b
Professor, Department of Orthodontics and Pediatric Den- because of their facial appearance. Consequently,
tistry, Faculty of Dentistry, Federal University of Minas Gerais,
social relationships and potential employment oppor-
Minas Gerais, Brazil.
c
PhD student, Department of Orthodontics and Pediatric tunities are hindered.2,6,7 Morphological deviations
Dentistry, Faculty of Dentistry, Federal University of Minas affect the dentition and oral cavity and cause vertical
Gerais, Minas Gerais, Brazil. and transverse alterations, such as anterior open bite
Corresponding author: Dr Ana Cristina Oliveira, Department and anterior/posterior crossbite.4,6,8,9 These dentofacial
of Social and Preventive Dentistry, Faculty of Dentistry, Federal
alterations lead to problems with sucking, swallowing,
University of Minas Gerais, Av. Pres. Antônio Carlos, 6627,
Pampulha, Belo Horizonte, Minas Gerais 31270-901 Brazil and salivation.1,6,9–11
(e-mail: anacboliveira@yahoo.com.br) Craniofacial growth and occlusal abnormalities may
Accepted: October 2009. Submitted: July 2009. be accentuated by the interaction between genetic and
G 2010 by The EH Angle Education and Research Foundation, environmental factors, such as breastfeeding and oral
Inc. habits (pacifier sucking, finger sucking, nail biting, etc),

OPEN BITE AND CROSSBITE ON DOWN SYNDROME 749
depending on the duration, intensity, and frequency of Periodontal Index probe (Golgran, São Paulo, SP,
the habit.2,9,12–18 A number of reports have suggested Brazil), also known as a ball-point probe.20
that nonnutritive sucking habits (usually pacifier or The study received approval from the Human
thumb sucking) may be responsible for some forms of Research Ethics Committee of the National School of
malocclusion in childhood, but the role of early feeding Public Health, Oswaldo Cruz Foundation, Rio de
in occlusion needs to be further evaluated.7,9,11,13–18 Janeiro, Brazil. Intraexaminer calibration, test/retest
Children with a history of pacifier sucking demonstrat- of the questionnaire, and a pilot study were carried out
ed a higher prevalence of posterior crossbite and prior to the main study.
anterior open bite.16 Although oral/motor dysfunction in To determine intraexaminer agreement, 25 children/
DS is well known, aspects related to these factors in adolescents with DS were examined and reexamined
children with Down syndrome have not been fully after a 10-day interval. These individuals were from a
clarified.2,6,11,19 nongovernmental organization that offers care to
The aim of the present study was to analyze the children with DS in the city of Rio de Janeiro and did
influence of breastfeeding, bottle feeding, and nonnu- not participate in the main study. Intraexaminer
tritive sucking habits on the prevalence of open bite diagnostic agreement was considered very good.21
and anterior/posterior crossbite in children/adolescents The following kappa values were achieved: 0.91 for
with DS. open bite, 0.89 for anterior crossbite, and 0.94
posterior crossbite.
MATERIALS AND METHODS To test the internal validity of the questionnaire,
testing and retesting of the measure were conducted
A cross-sectional study was carried out with 112
with the 25 mothers of the children/adolescents who
children/adolescents with DS, aged 3 to 18 years,
participated in the calibration test. The retest was
and their respective mothers. Data collection took
carried out after a 10-day interval. The results of the
place in a maternal/children’s hospital in Rio de
test/retest agreement revealed kappa values ranging
Janeiro, Brazil, which is a healthcare reference center
from 0.74 to 1.00, which are considered very good to
for individuals with DS. The data were obtained from
excellent.21
an oral exam of the patients and a structured
A pilot study was then carried out with 20 pairs of
questionnaire to which the mothers responded in an
mothers/children with DS treated at a maternal/
interview format.
children’s hospital in Rio de Janeiro. These individuals
While awaiting the medical appointment, each
also did not participate in the main study. The aim of
mother-child pair received information regarding the
this step was to test the method and data collection
study and was asked to participate. Pairs that agreed
instruments, confirming the validity of the methodology
to participate were directed to an examining room in
to be employed.
which the mother signed the informed consent,
Statistical analysis was performed employing the
responded to the questionnaire, and watched the
Software Package for the Social Sciences (SPSS for
clinical examination of her child. The questionnaire
Windows, version 15.0, SPSS Inc, Chicago, Ill).
contained items addressing breastfeeding, bottle
Univariate analysis was first performed. The chi-
feeding, and nonnutritive sucking habits of the chil-
square test (P , .10) was used to determine the
dren. The clinical examination recorded the presence
relationship between the dependent variables (preva-
of anterior open bite and anterior/posterior crossbite.
lence of open bite and anterior/posterior crossbite) and
The World Health Organization criteria were used to
the independent variables. Multiple logistic regression
assess malocclusions.20
was performed to identify the independent impact of
Confounders and other determent factors of interest
each variable studied. The independent variables were
were obtained, including age of the child/adolescent,
included in the decreasing logistic model in accor-
mouth posture (closed/open), and mother’s schooling.
dance with their statistical significance (P , .25;
Mouth posture was assessed during the clinical
backward stepwise procedure) or clinical-epidemiolog-
examination and interviews with the mothers, when
ical importance.
the child thought that he or she was not being
observed and revealed a habit of the mouth remaining
RESULTS
open or closed.
The clinical exam was performed by one of the The final sample included 112 children and adoles-
researchers with the help of a research assistant, who cents with DS between 3 and 18 years of age (mean
recorded the data. The exam was performed under age: 8.3 6 4.3 years; median: 7). Sixty-five partici-
artificial light with the aid of a disposable mouth mirror pants were 3 to 8 years old (58.0%) and 47 were 9 to
(Prisma, São Paulo, SP, Brazil) and a Community 18 years old (42.0%); 52 (46.4%) were girls and 60

750 OLIVEIRA, PORDEUS, TORRES, MARTINS, PAIVA
DISCUSSION
The results of the present study revealed a
significant prevalence of malocclusion in individuals
with DS, thereby confirming the findings of previous
studies, including those comparing individuals with DS
to children/adolescents with normal development or
some other type of disability.3,4,6,8,19 This is related to
more frequent occurrences of craniofacial deformities,
abnormal growth and development, and a higher
incidence of abnormal tongue posture and orofacial
muscle disorders.2,6,7,9 Some types of malocclusions
Figure 1. Prevalence of open bite, anterior crossbite, and posterior are caused and maintained by abnormal tongue
crossbite in children/adolescents with DS, Rio de Janeiro, Brazil (n function or posture.12
5 112).
Regardless of the presence of DS, studies stress
the multifactorial etiopathogenesis of malocclu-
(53.6%) were boys, and all were accompanied by their sion.6,8,11,13–18,22 There is interaction among congenital,
mothers (mean age 5 41 6 8.4 years). Only four pairs morphological, biomechanical, and environmental fac-
of subjects refused to participate because of a lack of tors during the growth and development of children,
time, disinterest, or the child’s refusal to be examined. including factors linked to breastfeeding and nonnutri-
Figure 1 displays the distribution of the participants tive oral habits.
based on the prevalence of open bite (21% [90% Because the present study analyzed the influence of
confidence interval [CI]: 15%–28%]), anterior crossbite breastfeeding, bottle feeding, and nonnutritive oral
(33% [90% CI: 26%–40%]) and posterior crossbite habits on the prevalence of malocclusion, the authors
(31% [90% CI: 24%–39%]). opted to adjust the logistic regression model using the
The bivariate analysis revealed that open bite was following confounding and risk factors: age, mouth
statistically associated with mouth posture (P 5 .03) posture, and mother’s schooling. Mouth posture was
(Table 1). The prevalence of anterior crossbite was measured because many individuals with DS have the
statistically significantly associated with the use of habit of maintaining a mouth open position. According
pacifier sucking (P 5 .01). Posterior crossbite was to a number of authors, underdevelopment of the
statistically associated with the use of bottle feeding (P maxilla, orbicular muscles of the lips, and facial
5 .07) and age of the child/adolescent (P , .01). muscles results in insufficient lip seal and can lead to
Table 2 displays the results of the multiple logistic mouth breathing.2,5,9,10
regression analysis. Regarding open bite, the use of Hypotonia of the facial and lip muscles, associated
bottle feeding remained in the final model, revealing a with a hypotonic, protruded tongue, causes many
1.6-fold increased prevalence of anterior open bite (P children/adolescents with DS to experience difficulties
. .10) among children/adolescents who bottle fed for in sucking and swallowing. According to Mizuno and
more than 24 months. There was an approximately Ueda,11 sucking difficulties in infants with DS may
threefold increased prevalence of anterior crossbite result from the low sucking pressure and short
among participants with prolonged use of pacifier duration, rather than a lack of will to suck. It is caused
sucking (more than 24 months); this result achieved not only by hypotonicity of the perioral muscles, lips,
statistical significance (P 5 .02). Although it was not and masticatory muscles, but also by a deficiency in
statistically significant, breastfeeding also remained in smooth peristaltic-like tongue movements in the early
the logistic model; there was a 1.4-fold increased neonatal period. Sucking pressure, which is negative
prevalence of anterior crossbite among those children/ intraoral pressure, is generated by the lowering of the
adolescents who were not breastfed or were breastfed jaw and posterior depression of the tongue, which
for fewer than 6 months. occurs during the tongue movement sequence.
Regarding posterior crossbite, finger sucking and However, these characteristics do not impede such
the use of bottle feeding remained in the logistic model, individuals from being able to breastfeed or having
with both variables achieving statistical significance. nonnutritive sucking habits.2,9,11,23,24 In a study carried
There was a nearly threefold increased prevalence of out in Japan involving children with DS and those with
posterior crossbite among children/adolescents with normal development, Mizuno and Ueda11 found signif-
finger sucking habits. There was a 2.6-fold increased icant development in the sucking behavior of the
prevalence of posterior crossbite among those who individuals with DS over time, but the sucking capacity
bottle fed for more than 24 months. of these infants was lower than that of normal infants.

Table 1. Distribution of the Sample (n 5 112) According to the Prevalence of Open Bite, Crossbite, and Associated Factors
Open Bite Anterior Crossbite Posterior Crossbite
Factor n n (%) P n (%) P n (%) P
Breastfeeding .26 .72 .22
,6 mo (or absent) 67 12 (17.9) 23 (34.3) 18 (26.9)
6 mo or more 45 12 (26.7) 14 (31.1) 17 (37.8)
Bottle feeding .22 .31 .07*
24 mo or more 53 14 (26.4) 20 (37.7) 21 (39.6)
,24 mo (or absent) 59 10 (16.9) 17 (28.8) 14 (23.7)
Pacifier sucking .86 .01* .27
24 mo or more 22 5 (22.7) 12 (54.5) 9 (40.9)
,24 mo (or absent) 90 19 (21.1) 25 (27.8) 26 (28.9)
Finger sucking .56 .69 .69
Yes 19 5 (26.3) 7 (36.8) 8 (42.1)
No 93 19 (20.4) 30 (32.3) 27 (29.0)
Mouth posture .03* .93 .77
Open 49 15 (30.6) 16 (32.7) 16 (32.7)
Closed 63 9 (14.3) 21 (33.3) 19 (30.2)
Age .61 .54 , .01*
3–8 y 65 15 (23.1) 20 (30.8) 9 (13.8)
9–18 y 47 9 (19.1) 17 (36.2) 26 (55.3)
Maternal schooling .18 .24 .47
,8 y 52 14 (26.9) 20 (38.5) 18 (34.6)
8 y or more 60 10 (16.7) 17 (28.3) 17 (28.3)
* Chi-square test (P , .10).
After the logistic analysis, the malocclusions inves- growth of the palate—conditions that lead to posterior
tigated were strongly associated with the duration of crossbite.8 The nonnutritive sucking habit also seems
breastfeeding and bottle feeding and with nonnutritive to be one of the most important factors influencing
sucking habits. The use of bottle feeding for more than malocclusion, regardless of whether a child has DS or
24 months denoted an increased prevalence of open not.8,11,25 A Brazilian study involving 330 four-year-old
bite or posterior crossbite, which confirms the results children observed anterior open bite and posterior
described in previous studies carried out on children crossbite in children with nonnutritive sucking habits,
with normal development.8,11,15,13,25 Several authors regardless of the duration of the habit.16
have observed an association between nonnutritive Children with pacifier sucking habits lasting for more
sucking habits and the prolonged use of bottle than 24 months and finger sucking habits had an
feeding.8,13,18 approximately threefold increased prevalence of ante-
In children with nonnutritive sucking habits and rior and posterior crossbite, which is in agreement with
those with prolonged bottle feeding, the different the findings of previous studies.8,13,14,16,25 Such results
involvement of orofacial muscles and different impact stress the influence of external factors on the genetic
on the palate are presumably responsible for the poor trait of DS in the etiology of these malocclusions. A
alignment of teeth and the anomalous transverse number of studies point out that individuals with DS are
Table 2. Multiple Logistic Regression Models Explaining the Prevalence of Open Bite and Crossbite in Children and Adolescents with Down
Syndrome in Rio de Janeiro, Brazil (n 5 112)
Dependent Variables Independent Variables PR (IC 90%) Crudea PR (IC 90%) Adjusted*
Open bite Bottle feeding (24 mo or more) 1.75 (0.70–4.38) 1.68 (0.75–3.72)
Anterior crossbite Pacifier sucking (24 mo or more) 3.12 (1.19–8.12)** 3.14 (1.39–6.96)**
Breastfeeding (,6 mo or absent) 1.15 (0.51–2.59) 1.41 (0.68–2.91)
Posterior crossbite Finger sucking 1.77 (0.64–4.90) 2.90 (1.06–7.96)**
Bottle feeding (24 mo or more) 2.10 (0.93–4.76) 2.67 (1.21–5.92)**
a
PR indicates prevalence ratios; CI 90%, 90% confidence intervals.
* Adjusted for control variables (age, mouth posture, and maternal schooling).
** Statistically significant.

752 OLIVEIRA, PORDEUS, TORRES, MARTINS, PAIVA
more prone to anterior/posterior crossbite owing to the malocclusion compromises the lives of children with
combination of clinical aspects of the syndrome (small DS.9,28 Mothers need to be persistent as well as duly
midface, protrusive tongue, and high palate).2,4,9,10,23 encouraged and oriented immediately postpartum with
Nonetheless, the influence of extrinsic factors, such as regard to the importance of breastfeeding their children
nonnutritive sucking habits, cannot be ignored. with DS and exercising discipline in the use of bottle
The duration of thumb sucking habits was not feeding and nonnutritive sucking habits.23,24
analyzed in the present study, as there was a report
of only one child with a thumb sucking habit for fewer CONCLUSIONS
than 24 months. Thus, only the presence or absence
N The children and adolescents with DS in the present
of this habit was considered. In a longitudinal study
sample had a high prevalence of open bite, anterior
carried out in the United States with 372 normal
crossbite, and posterior crossbite.
children with primary teeth, an increase in the
prevalence of anterior open bite and posterior cross- N The use of bottle feeding for more than 24 months
bite was found among those children with nonnutritive was associated with the occurrence of open bite and
sucking habits lasting until 24 or more months of age. posterior crossbite in children/adolescents with DS.
However, when these children continued the habit to N Having breastfed for fewer than 6 months was
48 months of age, the prevalence of these types of strongly associated with the presence of anterior
malocclusion was even higher.15 Anterior open bite is crossbite, as was the habit of pacifier sucking for
often spontaneously corrected in individuals who more than 24 months.
abandon a finger sucking habit before the pubertal N The habit of finger sucking was associated with
growth spurt.14 posterior crossbite in this portion of the population.
Having breastfed for fewer than 6 months and the
presence of a pacifier sucking for more than 24 months ACKNOWLEDGMENTS
denoted an increased prevalence of anterior crossbite.
This study was supported by the Brazilian fostering agencies
In an investigation of 300 preschool children in Brazil,
National Council of Scientific and Technological Development
Góis et al.18 found that children with a pacifier-sucking (CNPq) and Coordination of Higher Education (CAPES), Ministry
habit that continued beyond 2 years of age had a 13- of Education, Brazil.
fold increased prevalence of malocclusion in compar-
ison with children with no such habit. According to REFERENCES
Trawitzki et al.17 and Luz et al.,22 early weaning with the
1. Mitchell RB, Call E, Kelly J. Ear, nose and throat disorders in
introduction of nonhuman milk and other ‘‘substitutive’’ children with Down syndrome. Laryngoscope. 2003;113:
foods can favor the development of nonnutritive habits, 259–263.
such as pacifier and finger sucking. 2. Oliveira AC, Paiva SM, Campos MR, Czeresnia D. Factors
Although the present study design is strong, it has associated with malocclusions in children and adolescents
some limitations. A weakness of cross-sectional with Down syndrome. Am J Orthod Dentofacial Orthop.
2008;133:489.e1–489.e8.
studies resides in the difficulty in establishing causal 3. Quintanilha JS, Biedma BM, Rodrı́guez MQ, Mora MT,
relationships based on a cross-section in time.25 Cunqueiro MM, Pazos MA. Cephalometrics in children with
Moreover, the data on nutritive and nonnutritive oral Down’s syndrome. Pediatr Radiol. 2002;32:635–643.
habits were obtained from mothers’ reports and not 4. Bradley C, McAlister T. The oral health of children with
collected directly, which makes such data subject to Down syndrome in Ireland. Spec Care Dentist. 2004;24:
55–60.
information and memory bias. Thus, caution should be
5. Korbmacher H, Limbrock J, Kahl-Nieke B. Orofacial devel-
taken to avoid influencing the respondents and avoid opment in children with Down’s syndrome 12 years after
bias when interpreting the results.26 In Brazil, it is very early intervention with a stimulating plate. J Orofac Orthop.
difficult to obtain a representative sample of children/ 2004;65:60–73.
adolescents with DS, as there is no official database 6. Musich DR. Orthodontic intervention and patients with Down
that registers all individuals affected by this syndrome. syndrome. Angle Orthod. 2006;76:734–735.
7. Waldman HB, Perlman SP, Swerdloff M. Orthodontics and
Thus, the option was made to use a convenience
the population with special needs. Am J Orthod Dentofacial
sample in which important aspects related to the Orthop. 2000;118:14–17.
presence of malocclusion were identified in this portion 8. Viggiano D, Fasano D, Monaco G, Strohmenger L. Breast
of the population.27 feeding, bottle feeding, and non-nutritive sucking: effects on
Identifying factors associated with malocclusion in occlusion in deciduous dentition. Arch Dis Child. 2004;89:
1121–1123.
children with DS may contribute to interventions and
9. Bäckman B, Grevér-Sjölander AC, Bengtsson K, Persson J,
orientation regarding breastfeeding and prolonged Johansson I. Children with Down syndrome: oral develop-
nonnutritive sucking habits. Caregivers and healthcare ment and morphology after use of palatal plates between 6
professionals should be aware of the extent to which and 48 months of age. Int J Paediatr Dent. 2007;17:19–28.

10. Hennequin M, Faulks D, Veyrune JL, Bourdiol P. Signifi- 19. Carlstedt K, Henningsson G, Dahllöf G. A four-year
cance of oral health in persons with Down syndrome: a longitudinal study of palatal plate therapy in children with
literature review. Dev Med Child Neurol. 1999;41:275–283. Down syndrome: effects on oral motor function, articulation
11. Mizuno K, Ueda A. Development of sucking behavior in and communication preferences. Acta Odontol Scand.
infants with Down’s syndrome. Acta Paediatr. 2001;90: 2003;61:39–46.
1384–1388. 20. World Health Organization. Oral Health Surveys: Basic
12. Huang GJ, Justus R, Kennedy DB, Kokich VG. Stability of Methods. Geneva: World Health Organization; 1997.
anterior openbite treated with crib therapy. Angle Orthod. 21. Rigby AS. Statistical methods in epidemiology. Towards an
1990;60:17–24. understanding of the kappa coefficient. Disabil Rehabil.
13. Serra-Negra JM, Pordeus IA, Rocha Jr JF. Study of the 2000;22:339–344.
relationship between infant feeding methods, oral habits, and 22. Luz CL, Garib DG, Arouca R. Association between
malocclusion. Rev Odontol Univ São Paulo. 1997;11:79–86. breastfeeding duration and mandibular retrusion: a cross-
14. Villa NL, Cisneros GJ. Changes in the dentition secondary to sectional study of children in the mixed dentition.
palatal crib therapy in digit-suckers: a preliminary study. Am J Orthod Dentofacial Orthop. 2006;130:531–534.
Pediatr Dent. 1997;19:323–326. 23. Hennequin M, Allison PJ, Veyrune JL. Prevalence of oral
15. Warren J, Bishara S, Steinbock K, Yonezu T, Nowak A. health problems in a group of individuals with Down
Effects of oral habits’ duration on dental characteristics in syndrome in France. Dev Med Child Neurol. 2000;42:
691–698.
the primary dentition. J Am Dent Assoc. 2001;132:
24. Pisacane A, Toscano E, Pirri I, et al. Down syndrome and
1685–1693.
breastfeeding. Acta Paediatr. 2003;92:1479–1481.
16. Katz CR, Rosenblatt A, Gondim PP. Nonnutritive sucking
25. Cozza P, Baccetti T, Franchi L, Mucedero M, Polimeni A.
habits in Brazilian children: effects on deciduous dentition
Transverse features of subjects with sucking habits and
and relationship with facial morphology. Am J Orthod
facial hyperdivergency in the mixed dentition. Am J Orthod
17. Trawitzki LV, Anselmo-Lima WT, Melchior MO, Grechi TH, 26. Martins CC, Ramos-Jorge ML, Cury JA, Pordeus IA, Paiva
Valera FC. Breast-feeding and deleterious oral habits in SM. Agreement between data obtained from repeated
mouth and nose breathers. Braz J Otorhinolaryngol. 2005; interviews with a six-years interval. Rev Saúde Pública.
71:747–751. 2008;42:346–349.
18. Góis EG, Ribeiro-Júnior HC, Vale MP, Paiva SM, Serra- 27. Kirkwood BR, Stern J. Essentials of Medical Statistics.
Negra JM, Ramos-Jorge ML, Pordeus IA. Influence of London: Blackwell; 2003.
nonnutritive sucking habits, breathing pattern and adenoid 28. Trottman A, Elsbach HG. Comparison of malocclusion in
size on the development of malocclusion. Angle Orthod. preschool black and white children. Am J Orthod Dentofacial
2008;78:647–654. Orthop. 1996;110:69–72.

Case Report
New treatment modality for maxillary hypoplasia in cleft patients

Protraction facemask with miniplate anchorage
Seung-Hak Baeka; Keun-Woo Kimb; Jin-Young Choic
ABSTRACT
Objective: To present cleft patients treated with protraction facemask and miniplate anchorage
(FM/MP) in order to demonstrate the effects of FM/MP on maxillary hypoplasia.
Materials and Methods: The cases consisted of cleft palate only (12 year 1 month old girl,
treatment duration 5 16 months), unilateral cleft lip and alveolus (12 year 1 month old boy,
treatment duration 5 24 months), and unilateral cleft lip and palate (7 year 2 month old boy,
treatment duration 5 13 months). Curvilinear type surgical miniplates (Martin, Tuttlinger, Germany)
were placed into the zygomatic buttress areas of the maxilla. After 4 weeks, mobility of the
miniplates was checked, and the orthopedic force (500 g per side, 30u downward and forward from
the occlusal plane) was applied 12 to 14 hours per day.
Results: In all cases, there was significant forward displacement of the point A. Side effects such
as labial tipping of the upper incisors, extrusion of the upper molars, clockwise rotations of the
mandibular plane, and bite opening, were considered minimal relative to that usually observed with
conventional protraction facemask with tooth-borne anchorage.
Conclusions: FM/MP can be an effective alternative treatment modality for maxillary hypoplasia
with minimal unwanted side effects in cleft patients. (Angle Orthod. 2010;80:783–791.)
KEY WORDS: Maxillary protraction; Facemask; Miniplate
INTRODUCTION unwanted side effects such as labioversion of the

upper incisors, extrusion of the upper molars, coun-
In Class III malocclusion patients with mild to
terclockwise rotation of the upper occlusal plane, and
moderate maxillary hypoplasia, the protraction face-
eventual clockwise rotation of the mandible.3–6 There-
mask has been used to stimulate sutural growth at the
fore, labial inclined maxillary incisors and/or a vertical
circum-maxillary suture sites in growing patients.1–3 To
facial growth pattern would be contraindications for
transmit the orthopedic force from the protraction
facemask therapy with tooth-borne anchorage.
facemask to the maxilla, intraoral devices such as a
To allow the direct transmission of orthopedic force
labiolingual arch, quad helix, and rapid maxillary
to the circum-maxillary sutures, intentionally ankylosed
expansion (RME) have been used. However, the use
primary canines, osseointegrated implants, and ortho-
of the upper dentition as anchorage cannot avoid
dontic miniscrews have been used as skeletal anchor-
a
Associate Professor, Department of Orthodontics, School of age for protraction facemasks.7–11 Since surgical
Dentistry, Dental Research Institute, Seoul National University, miniplates are a reliable means for applying orthodon-
Seoul, South Korea. tic and orthopedic forces,12 Kircelli et al.,13 Cha et al.,14
b
Resident and Graduate Masters Student, Department of and Kircelli and Pektas15 introduced the protraction
Orthodontics, School of Dentistry, Seoul National University,
facemask with miniplate anchorage (FM/MP) therapy
Seoul, South Korea.
c
Associate Professor, Department of Oral and Maxillofacial to treat Class III malocclusion with maxillary hypopla-
Surgery, School of Dentistry, Dental Research Institute, Seoul sia and hypodontia (Figure 1).
National University, Seoul, South Korea. The protocol of FM/MP is as follows: (1) After an
Corresponding author: Dr Jin-Young Choi, Department of Oral approximate 1–2 cm horizontal vestibular incision is
and Maxillofacial Surgery, School of Dentistry, Dental Research
made just below the zygomatic buttress area under
Institute, Seoul National University, Yeonkun-dong #28, Jongro-
ku, Seoul 110-768, South Korea local anesthesia, the zygomatic buttress is exposed
(e-mail: jinychoi@snu.ac.kr) with a subperiosteal flap. (2) Curvilinear type surgical
Accepted: September 2009. Submitted: July 2009. miniplates (Martin, Tuttlinger, Germany) are bent
G 2010 by The EH Angle Education and Research Foundation, according to the anatomical shape of the zygomatic
Inc. buttress. (3) The distal end hole of the miniplate should

784 BAEK, KIM, CHOI
Figure 1. Comparison of pretreatment (left) and posttreatment (right) in patient with Class III malocclusion. (a) Facial photographs. (b) Intraoral
photographs. (c) Lateral cephalograms. (d) Superimposition (solid line: pretreatment; dotted line: posttreatment).
be cut to make a hook for elastics. (4) After the patients who were treated with FM/MP and to
miniplates are placed into the zygomatic buttress demonstrate the effect of FM/MP on maxillary hypo-
areas, three self-tapping type screws are used per plasia in cleft patients.
side to fix the miniplates (Figure 2a). (5) The distal end
of the miniplate should be exposed through the CASE REPORTS
attached gingiva between the upper canine and first CASE 1
premolar to control the vector of elastic traction
(Figure 2b). (6) Four weeks after placement of the Skeletal Class III malocclusion with cleft palate (CP)
miniplates, their mobility is checked and the orthopedic and anterior open bite (Figure 3, Table 1).
force (500 g per side, 30u downward and forward from
the occlusal plane) is applied for 12 to 14 hours per Diagnosis
day. (7) It is recommended to overcorrect the The patient was a 12 year 1 month old girl with CP
malocclusion into positive overjet and a slight Class only. She presented with concave facial profile,
II canine and molar relationship. anterior crossbite (29 mm overjet), and anterior open
Cleft patients often develop Class III malocclusion bite (22 mm overbite). Cephalometric analysis
with maxillary hypoplasia and vertical facial growth showed skeletal Class III malocclusion with maxillary
pattern due to the combined effects of the congenital hypoplasia (ANB, 25.4u; A to N perp, 23.4 mm), steep
deformity itself and the scar tissues after surgical mandibular plane angle (FMA, 32.7u), and a skeletal
repair.16 These are contraindications for conventional age after the pubertal growth spurt according to the
facemask therapy. However, little research has been cervical vertebrae maturation index (CVMI, stage 4).17
done on the use of FM/MP in cleft patients. Therefore, Her condition was one of the contraindications for
the purpose of this case report is to present three cleft conventional facemask therapy.

MAXILLARY PROTRACTION: FACEMASK AND MINIPLATE 785
anterior open bite was corrected by downward and

forward movement of the maxilla. Slight labial tipping of
the upper incisors (DU1 to SN, 2.0u) occurred after
correction of anterior crossbite and open bite.
CASE 2
Skeletal Class III malocclusion with unilateral cleft lip
and alveolus (UCLA) and vertical facial growth pattern
(Figure 4, Table 1).
Diagnosis
The patient was a 12 year 1 month old boy with
UCLA on the left side. Although he presented with a
straight facial profile, he had an anterior crossbite
(22.5 mm overjet), upper anterior crowding, and peg
laterals on the cleft side. Although the anteroposterior
skeletal relationship (ANB, 1.4u) was within normal
range and the upper and lower incisors were lingually
inclined (U1 to SN, 95.1u; IMPA, 86.9u), a vertical facial
growth pattern (FMA, 33.2u) existed. His skeletal age
was before his pubertal growth spurt according to the
CVMI (stage 3).17
Treatment Plan
Conventional facemask protraction with a tooth-
borne anchorage device was not appropriate because
the patient had a vertical facial growth pattern.
Figure 2. Schematic drawing of the surgical positioning (a) and
intraoral position of the miniplate (b).
Therefore, the FM/MP was used to avoid unwanted
side effects.
Treatment Plan Treatment Progress

Although growth observation and reassessment Initially, the fixed orthodontic appliance was placed
after 2 years were proposed, her parents wanted to to correct the anterior crowding in the upper arch. The
receive the FM/MP therapy. The possibility of orthog- FM/MP therapy was started 4 weeks after placement
nathic surgery after pubertal growth was explained. of the miniplates according to the protocol.
Treatment Progress Treatment Results

FM/MP therapy was started 4 weeks after place- After 24 months of FM/MP therapy, there was a 3.1-
ment of the miniplates according to the protocol. mm forward movement of point A (DA to N perp). ANB
During protraction, the fixed appliances were placed angle was changed from 1.4u to 3.5u, and a Class II
to align the dentition. canine and molar relationship was obtained. The
finding that there was a negligible counterclockwise
Treatment Results rotation of the mandibular plane (0.4u) and occlusal
plane angle (20.9u) indicated that there were almost
After 16 months of FM/MP therapy, there was
no side effects such as extrusion of the upper molars
significant forward movement of the point A (DA to N
and bite opening. Labial tipping of the upper incisors
perp, 5.6 mm). The ANB angle was changed from
(DU1 to SN, 4.9u) occurred due to alignment.
25.4u to 2.9u, and a Class II canine and molar
relationship, normal overbite, and overjet were ob-
CASE 3
tained. A slight counterclockwise rotation of the occlusal
plane angle (21.8u) was interpreted to mean that there Skeletal Class III malocclusion with unilateral cleft lip
was almost no side effect such as extrusion of the upper and palate (UCLP) and vertical facial growth pattern
molars. Although the FMA was increased 4.3u, the (Figure 5, Table 1).

786 BAEK, KIM, CHOI
Figure 3. Comparison of pretreatment (left) and posttreatment (right) in patient with cleft palate (case 1). (a) Facial photographs. (b) Intraoral
photographs. (c) Lateral cephalograms. (d) Superimposition (solid line: pretreatment; dotted line: posttreatment).
Diagnosis Treatment Results

The patient was a 7 year 2 month old boy with a Similar to Case 2, there was a 3.0-mm forward
UCLP on the right side. Although he presented with a movement of point A (DA to N perp) after 13 months of
straight facial profile, he had an anterior crossbite protraction facemask therapy. The ANB angle was
(22.7 mm overjet). Although the anteroposterior skel- changed from 1.6u to 3.1u, and Class II canine and
etal relationship (ANB, 1.6u) was within normal range molar relationships were obtained. Although there was
and the upper incisors were lingually inclined (U1 to a slight counterclockwise rotation of the mandibular
SN, 93.9u), a vertical facial growth pattern existed plane (20.9u) and occlusal plane angle (22.5u), there
(FMA, 34.6u). His skeletal age was before his pubertal was no bite opening in the anterior teeth. Some labial
growth spurt according to CVMI (stage 2).17 tipping of the upper incisors (DU1 to SN, 2.7u) occurred
after correction of the anterior crossbite.
Treatment Plan
DISCUSSION
FM/MP was planned to maximize protraction of the
maxilla and to avoid unwanted side effects. Site for Placement of Miniplates
The zygomatic buttress area was used as the site for
Treatment Progress
placement of the miniplates due to following reasons:
FM/MP therapy was started 4 weeks after place- (1) It has enough thickness and adequate bone
ment of the miniplates according to the protocol. quality.18 (2) It is near to the center of resistance of

Table 1. Comparison of the Skeletal, Dental, and Soft Tissue Variables Between Pretreatment (T0) and Posttreatment (T1)
Case 1 Case 2 Case 3
Variable T0 T1 T0 T1 T0 T1
Anteroposterior skeletal relationship
SNA (u) 76.0 80.6 74.9 77.1 78.5 80.4
SNB (u) 81.4 77.7 73.5 73.6 76.9 77.3
ANB (u) 25.4 2.9 1.4 3.5 1.6 3.1
A to N perp (mm) 23.4 2.2 25.2 22.1 23.3 20.3
Pog to N perp (mm) 4.7 20.6 212.2 210.2 28.0 28.0
Wits appraisal (mm) 215.5 23.6 25.2 21.9 25.4 22.5
Vertical skeletal relationship
Bjork sum (u) 403.7 408.2 403.2 403.6 402.3 401.5
Saddle angle (u) 128.4 125.4 120.7 122.1 116.9 117.2
Articular angle (u) 143.9 154.1 152.4 152.8 150.1 151.1
Gonial angle (u) 131.4 128.7 130.1 128.7 135.3 133.2
Facial height ratio (%) 55.7 52.7 58.6 58.1 60.4 61.6
Palatal plane angle (u) 22.3 21.5 2.7 2.9 2.2 1.8
FMA (u) 32.7 37.0 33.2 32.7 34.6 33.7
Mandibular plane to SN plane angle (u) 43.7 48.2 43.2 43.6 42.3 41.5
Occlusal plane to SN plane angle (u) 21.6 19.8 24.6 23.7 23.2 20.7
Dental relationship
U1 to SN (u) 106.4 108.4 95.1 100.0 93.9 96.6
IMPA (u) 88.2 76.3 86.9 87.5 86.8 82.2
Soft tissue
Nasolabial angle (u) 105.8 107.3 117.4 119.0 115.4 109.3
the nasomaxillary complex so that the force vector can movement in the maxilla when protraction was in
be placed close to the center of rotation of the conjunction with RME compared with protraction
nasomaxillary complex.12,19 without RME. Liou and Tsai22 presented the combined
use of repeated rapid maxillary expansion and
Vector Control constriction and intraoral springs for maxillary protrac-
tion and concluded that significant advancement of the
The direction of force vector was 30u downward and point A could be obtained.
forward from the occlusal plane. Tanne et al.20 and However, a recent prospective, randomized clinical
Miyasaka-Hiraga et al.21 reported that downward and trial23 showed that facemask therapy, with or without
forward force produced uniform stretch and translatory RME, produced equivalent changes in the dentofacial
repositioning of the nasomaxillary complex in an complex and insisted that RME might not be indis-
anterior direction. However, conventional dental an- pensable to maxillary protraction unless a transverse
chorage usually results in counterclockwise rotation of deficiency exists. In our cases, we did not use RME
the palatal plane, and clockwise rotation of the because cleft lip and palate patients do not have some
mandible, which would be unfavorable in a patient or whole parts of the midpalatal suture. Since this
with vertical growth pattern. could affect the amount of maxillary advancement in
The miniplate can transmit the orthopedic force cleft patients, further studies will be necessary.
directly to the maxilla and minimize rotational effect.
Although there was a slight increase of FMA (4.3u) in Timing of Treatment
Case 1, Cases 2 and 3 showed negligible changes of
the palatal plane angle, FMA, and mandibular plane to There are numerous articles that advocate the
SN angle (Table 1). protraction therapy at an early stage.5,24–28 Because
the palatomaxillary suture becomes highly interdigitat-
ed with increasing age, it becomes difficult to
Conjunction with RME
disarticulate the palatal bone from the pterygoid
Expansion of the maxilla before or during protraction process.29 After the pubertal growth peak, side effects
of the maxilla has been performed to facilitate such as tooth movement and/or mandibular rotation
protraction by disarticulating the circum-maxillary rather than maxillary protraction are likely to be the
sutures and initiating a cellular response in these major response to treatment.5,30 However, Baik4 and
sutures.1,3,5 Baik4 reported that there was more forward Sung and Baik31 insisted that there was no statistical

788 BAEK, KIM, CHOI
Figure 4. Comparison of pretreatment (left) and posttreatment (right) in patient with unilateral cleft lip and alveolus (case 2). (a) Facial
photographs. (b) Intraoral photographs. (c) Lateral cephalograms. (d) Superimposition (solid line: pretreatment; dotted line: posttreatment).
difference when changes due to treatment were 0.9 mm to 2.9 mm advancement of the point A. So25
compared according to ages. insisted that the effect of protraction facemask therapy
On the other hand, Kircelli and Pektas15 reported on the maxilla was two thirds skeletal and one third
that protraction using the FM/MP in relatively older dental changes.
patients (11 to 13 years) was successful with minimum In cases with facemask and skeletal anchorage,
dentoalveolar side effects. We also confirmed that amounts of the maxillary advancement have been
there was a significant maxillary protraction with fewer reported to be 4.0 to 4.8 mm.9,15,34 Therefore, maxillary
dentoalveolar side effects using the MP/FM in the advancement can be enhanced by skeletal anchorage
patient after the pubertal growth peak and menarche rather than conventional dental anchorage in growing
(Case 1, CVMI stage 4).17 patients.
In our cases, although a similar treatment protocol
was used, the amounts of maxillary advancement
varied according to cleft types (approximately 3.0 mm–
Comparison of the Amount of
5.6 mm). Since the duration of protraction (approxi-
Maxillary Advancement
mately 13–24 months) was relatively longer than in the
In cases of untreated Class III malocclusion with other studies,9,15,34 the scar tissues of cleft patients can
maxillary hypoplasia, Shanker et al.32 reported that be one of the reasons for variations in the amount of
point A came forward only 0.2 mm over a 6-month maxillary protraction. This result was in accordance
period. With conventional facemask therapy, Kim et with Buschang et al.35 concerning limited protraction
al.33 from meta-analysis, reported that it produced results in cleft patients.

Figure 5. Comparison of pretreatment (left) and posttreatment (right) in patient with unilateral cleft lip and palate (case 3). (a) Facial photographs.
(b) Intraoral photographs. (c) Lateral cephalograms. (d) Superimposition (solid line: pretreatment; dotted line: posttreatment).
Overcorrection are needed. Because the miniplates are independent

from dentition, simultaneous orthodontic treatment and
Ngan et al.36 and MacDonald et al.37 insisted that
maxillary protraction is an attractive advantage.
facemask therapy does not normalize the forward
The FM/MP can be used over a relatively longer
growth of the maxilla and that patients resume a Class
period than conventional facemask because it is
III growth pattern by deficient maxillary growth during the
independent of the upper dentition. According to
follow-up period. Therefore, overcorrection into Class II Ishikawa et al.,39 the effects of conventional facemask
canine and molar relationships is mandatory to com- therapy were significantly less in the second year, and
pensate for deficient posttreatment maxillary growth. no benefit from any treatment longer than 1 year was
Cleft patients seem to need more overcorrection established. However, we confirmed that FM/MP could
than ordinary Class III malocclusion cases with result in uniform advancement of the maxilla during the
maxillary hypoplasia because there are limitations in entire treatment period (Table 1).
the amounts of maxillary advancement and a high In addition, cleft patients have more vertical growth
relapse rate due to scar tissues.38 If the amount of pattern than noncleft patients,40 and excessive clock-
maxillary advancement is large and the patients and wise rotation of the mandible during facemask therapy
parents want a relatively short-term treatment, distrac- can worsen the facial profile. FM/MP can minimize
tion osteogenesis during adolescence or orthognathic clockwise rotation of the mandible and prevent
surgery in adulthood can be recommended. aggravation of the facial profile.
Advantages of FM/MP in Cleft Patients CONCLUSION

In adolescent cleft patients, multiple orthodontic N FM/MP can be an effective alternative treatment
treatment procedures such as alignment, leveling, modality for cleft patients with maxillary hypoplasia
arch expansion, and preparation for bone graft surgery with minimal unwanted side effects.

790 BAEK, KIM, CHOI
REFERENCES on the craniofacial complex: a study using the finite element

method. Am J Orthod Dentofacial Orthop. 1989;95:
1. McNamara JA. An orthopedic approach to the treatment of 200–207.
Class III malocclusion in young patients. J Clin Orthod. 21. Miyasaka-Hiraga J, Tanne K, Nakamura S. Finite element
1987;21:598–608. analysis for stresses in the craniofacial sutures produced by
2. Delaire J. Maxillary development revisited: relevance to the maxillary protraction forces applied at the upper canines.
orthopaedic treatment of Class III malocclusions. Br J Orthod. 1994;21:343–348.
Eur J Orthod. 1997;19:289–311. 22. Liou EJ, Tsai WC. A new protocol for maxillary protraction in
3. Nartallo-Turley PE, Turley PK. Cephalometric effects of cleft patients: repetitive weekly protocol of alternate rapid
combined palatal expansion and facemask therapy on Class maxillary expansions and constrictions. Cleft Palate
III malocclusion. Angle Orthod. 1998;68:217–224. Craniofac J. 2005;42:121–127.
4. Baik HS. Clinical results of the maxillary protraction in 23. Vaughn GA, Mason B, Moon HB, Turley PK. The effects of
Korean children. Am J Orthod Dentofacial Orthop. 1995;108: maxillary protraction therapy with or without rapid palatal
583–592. expansion: a prospective, randomized clinical trial.
5. Kapust AJ, Sinclair PM, Turley PK. Cephalometric effects of Am J Orthod Dentofacial Orthop. 2005;128:299–309.
face mask/expansion therapy in Class III children: a 24. Tindlund RS. Orthopaedic protraction of the midface in the
comparison of three age groups. Am J Orthod Dentofacial deciduous dentition. Results covering 3 years out of
Orthop. 1998;113:204–212. treatment. J Craniomaxillofac Surg. 1989;17(suppl 1):17–9.
6. Hickham JH. Maxillary protraction therapy: diagnosis and 25. So LL. Effects of reverse headgear treatment on sagittal
treatment. J Clin Orthod. 1991;25:102–113. correction in girls born with unilateral complete cleft lip and
7. Kokich VG, Shapiro PA, Oswald R, Koskinen-Moffett L, cleft palate—skeletal and dental changes. Am J Orthod
Clarren SK. Ankylosed teeth as abutments for maxillary Dentofacial Orthop. 1996;109:140–147.
protraction: a case report. Am J Orthod. 1985;88:303–307. 26. Saadia M, Torres E. Sagittal changes after maxillary
8. Silva Filho OG, Ozawa TO, Okada CH, Okada HY, Carvalho protraction with expansion in Class III patients in the
RM. Intentional ankylosis of deciduous canines to reinforce primary, mixed, and late mixed dentitions: a longitudinal
maxillary protraction. J Clin Orthod. 2003;37:315–320. retrospective study. Am J Orthod Dentofacial Orthop. 2000;
9. Enacar A, Giray B, Pehlivanoglu M, Iplikcioglu H. Facemask 117:669–680.
therapy with rigid anchorage in a patient with maxillary 27. Suda N, Suzukl MI, Hirose K, Hiyama S, Suzuki S, Kuroda
hypoplasia and severe oligodontia. Am J Orthod Dentofacial T. Effective treatment plan for maxillary protraction: is the
Orthop. 2003;123:571–577. bone age useful to determine the treatment plan?
10. Hong H, Nagn P, Li HG, Qi LG, Stephen HY. Use of Am J Orthod Dentofacial Orthop. 2000;118:56–62.
onplants as stable anchorage for facemask treatment. Angle 28. Franchi L, Baccetti T, McNamara JA Jr. Postpubertal
Orthod. 2005;75:453–460. assessment of treatment timing for maxillary expansion
11. Vachiramon A, Urata M, Kyung HM, Yamashita DD, Yen SL. and protraction therapy followed by fixed appliances.
Clinical applications of orthodontic microimplant anchorage Am J Orthod Dentofacial Orthop. 2004;126:555–568.
in craniofacial patients. Cleft Palate Craniofac J. 2009;46: 29. Melsen B, Melsen F. The postnatal development of the
136–146. palatomaxillary region studied on human autopsy material.
12. Umemori M, Sugawara J, Mitani H, Nagasaka H, Kamamura Am J Orthod. 1982;82:329–342.
H. Skeletal anchorage for open-bite correction. Am J Orthod 30. Cha KS. Skeletal changes of maxillary protraction in patients
Dentofacial Orthop. 1999;115:166–174. exhibiting skeletal class III malocclusion: a comparison of
13. Kircelli BH, Pektaş ZO, Uçkan S. Orthopedic protraction with three skeletal maturation groups. Angle Orthod. 2003;73:
skeletal anchorage in a patient with maxillary hypoplasia 26–35.
and hypodontia. Angle Orthod. 2006;76:156–163. 31. Sung SJ, Baik HS. Assessment of skeletal and dental
14. Cha BK, Lee NK, Choi DS. Maxillary protraction treatment of changes by maxillary protraction. Am J Orthod Dentofacial
skeletal Class III children using miniplates anchorage. Orthop. 1998;114:492–502.
Korean J Orthod. 2007;37:73–84. 32. Shanker S, Ngan P, Wade D, Beck M, Yiu C, Hägg U, Wei
15. Kircelli BH, Pektas ZO. Midfacial protraction with skeletally SH. Cephalometric A point changes during and after
anchored face mask therapy: a novel approach and maxillary protraction and expansion. Am J Orthod Dentofa-
preliminary results. Am J Orthod Dentofacial Orthop. 2008; cial Orthop. 1996;110:423–430.
133:440–449. 33. Kim JH, Viana MA, Graber TM, Omerza FF, BeGole EA.
16. Baek SH, Moon HS, Yang WS. Cleft type and Angle’s The effectiveness of protraction face mask therapy: a meta-
classification of malocclusion in Korean cleft patients. analysis. Am J Orthod Dentofacial Orthop. 1999;115:
Eur J Orthod. 2002;24:647–653. 675–685.
17. Hassel B, Farman AG. Skeletal maturation evaluation using 34. Singer SL, Henry PJ, Rosenberg I. Osseointegrated
cervical vertebrae. Am J Orthod Dentofacial Orthop. 1995; implants as an adjunct to facemask therapy: a case report.
107:58–66. Angle Orthod. 2000;70:253–262.
18. Sugawara J, Nishimura M, Nagasaka H, Kawamura H. 35. Buschang PH, Porter C, Genecov E, Genecov D, Sayler KE.
Temporary anchorage devices in orthodontics. Skeletal Facemask therapy of preadolescents with unilateral cleft lip
Anchorage System Using Orthodontic Miniplates. St Louis, and palate. Angle Orthod. 1994;64:145–150.
Mo: Mosby; 2009;15:320. 36. Ngan PW, Hagg U, Yiu C, Wei SH. Treatment response and
19. Tanne K, Matsubara S, Sakuda M. Location of the centre of long-term dentofacial adaptations to maxillary expansion
resistance for the nasomaxillary complex studied in a three and protraction. Semin Orthod. 1997;3:255–264.
dimensional finite element model. Br J Orthod. 1995;22: 37. MacDonald KE, Kapust AJ, Turley PK. Cephalometric
227–232. changes after the correction of Class III malocclusion with
20. Tanne K, Hiraga J, Kakiuchi K, Yamagata Y, Sakuda M. maxillary expansion/facemask therapy. Am J Orthod Den-
Biomechanical effect of anteriorly directed extraoral forces tofacial Orthop. 1999;116:13–24.

38. Baek SH, Lee JK, Lee JH, Kim MJ, Kim JR. Comparison of unilateral cleft lip and palate patients. Cleft Palate
treatment outcome and stability between distraction osteo- Craniofac J. 2000;37:92–97.
genesis and LeFort I osteotomy in cleft patients with 40. Van den Dungen GM, Ongkosuwito EM, Aartman IH,
maxillary hypoplasia. J Craniofac Surg. 2007;18: Prahl-Andersen B. Craniofacial morphology of Dutch pa-
1209–1215. tients with bilateral cleft lip and palate and noncleft controls
39. Ishikawa H, Kitazawa S, Iwasaki H, Nakamura S. Effects of at the age of 15 years. Cleft Palate Craniofac J. 2008;45:
maxillary protraction combined with chin-cap therapy in 661–666.

Original Article
Longitudinal growth changes of the cranial base from puberty to adulthood

A comparison of different superimposition methods
Zuleyha Mirzen Arata; Hakan Türkkahramanb; Jeryl D. Englishc; Ronald L. Galleranod; Jim C. Boleye
ABSTRACT
Objective: To investigate the stability of cranial reference landmarks from puberty through
adulthood and to compare the displacement of these landmarks among the superimposition
methods of Björk, Ricketts, Steiner, and the proposed tuberculum sella-wing (T-W) reference line.
Materials and Methods: The sample consisted of serial lateral cephalometric radiographs of 30
Class II division 1 patients taken at the pretreatment (T1; mean age, 11.98 years), posttreatment
(T2; mean age, 15.32 years) and postretention (T3; mean age, 32.12 years) periods. All
cephalometric radiographs were superimposed at the cranial base according to the overall
superimposition methods of Björk, Ricketts, Steiner, and the T-W method. The horizontal and
vertical displacements of cranial landmarks (nasion, wing, tuberculum sella, sella, basion, and
pterygomaxillare) were assessed by paired t-test according to Björk’s structural method. One-way
analysis of variance (ANOVA) was used for comparison of the displacement of cranial landmarks
among the superimposition methods.
Results: The tuberculum sella and wing were the most stable cranial landmarks of the cranial base.
The stability of sella and pterygomaxillare points were somewhat questionable. Nasion and basion were
highly variable. The displacements of all cranial landmarks were similar between the Björk and T-W
methods in all study periods. Most of the cranial landmarks displaced similarly in the horizontal direction
among the methods. Vertically, the behaviors of the cranial landmarks were frequently different.
Conclusions: T-W is the most similar superimposition method to Björk’s structural method; thus, it
is a reliable method for examining overall facial changes. (Angle Orthod. 2010;80:725–732.)
KEY WORDS: Cephalometrics; Superimposition methods; Cranial landmarks; Longitudinal growth
INTRODUCTION approach requires comprehensive examination of

current concepts. Two basic principles should be taken
The cranial base has long been suggested for
into account when the development of the cranial base
overall facial superimposition because of its stability
is evaluated: functional matrix theory8,9 and counter-
during the early ages of adolescence.1–7 This view had
part analysis.10 The cranial base is located between
widely been recognized in our field, but a scientific the brain and the face. Thus, it continues development
according to neural (neurocranial capsule) and skeletal
a
Professor, Department of Orthodontics, University of Ankara, (orofacial capsule) demands. Neural, skeletal, and
Turkey. muscular factors are closely interrelated. To avoid
b
Associate Professor, Department of Orthodontics, Süleyman errors, cranial base development should be addressed
Demirel University, Isparta, Turkey. by this rationale.
c
Professor, Department of Orthodontics, University of Texas
The cranial base consists of three segments (ie,
Health Science Center, Houston, Tex.
d
Associate Professor, Department of Orthodontics, University anterior cranial base, middle cranial base, and posterior
of Texas Health Science Center, Houston, Tex. cranial base). The anterior and posterior segments of
e
Professor, Department of Orthodontics, Baylor College of the cranial base grow at the same rate as craniofacial
Dentistry, Dallas, Tex. skeletal growth; thus, development of these segments
Corresponding author: Dr Zuleyha Mirzen Arat, Department of
continues for many years in line with the growth of the
Orthodontics, School of Dentistry, University of Ankara, Besev-
ler, Ankara, TR 06500 Turkey jaws. However, the middle cranial base completes its
(e-mail: mirzenarat@hotmail.com) development earlier due to the protection of the brain
Accepted: December 2009. Submitted: August 2009. and other vital organs.11–15 Thus, the stability of the
G 2010 by The EH Angle Education and Research Foundation, middle cranial base after age 8 makes it an excellent
Inc. baseline for the study of facial growth.1–3,11–13,15 The

726 ARAT, TÜRKKAHRAMAN, ENGLISH, GALLERANO, BOLEY
Table 1. Chronologic Age and the Distribution of Patients According to Cervical Vertebral Maturation Stagesa
Pubertal Growth Stage Postpubertal Growth Stage
Chronologic Age, Years CS1 CS2 CS3 CS4 CS5 CS6
T1 11.98 6 1.30 10 8 9 3 0 0
T2 15.32 6 1.12 0 0 2 15 12 1
T3 32.12 6 6.85 0 0 0 0 0 30
a
CS indicates cervical stage; T1, pretreatment; T2, posttreatment; T3, postretention.
stability of the anterior (N) and posterior (Ba) borders of division 1 patients (18 girls, 12 boys) treated by the
the cranial base1–5,11,13,16–18 and even of S point19–21 have same orthodontist. The mean ages of the subjects at
been discussed, and it was shown that these points are the start (T1) and at the end (T2) of treatment and long
highly variable during growth. follow-up period (T3) were 11.98 6 1.30, 15.32 6 1.12,
It has long been known that the tuberculum sella (T) and 32.12 6 6.85 years, respectively. The time interval
and wing (W) points, located at the middle cranial between T1–T3 periods was 20.15 6 6.73 years.
base, are highly stable.22,23 Because of this, the Skeletal maturity of the patients was assessed by
ethmoid triad system16 and the cranial base triangle20 using the cervical vertebral maturation (CVM) criteria.
superimposition methods were developed, but neither Accordingly, 27 of 30 subjects were included in CS1–
of them has become widely used. Information obtained CS3 maturation stage in T1 time and CS4–CS5
from implant studies as well as from studies on human maturation stage in T2 time. The maturation stage
autopsy material5,19,23 have shown that there are highly between CS1–CS3 coincides with the accelerative
stable regions in the cranial base. Thus, the superim- growth phase (pubertal growth stage), and the cervical
position of serial head films on relatively stable maturation stage between CS4–CS6 indicates the
anatomic structures is considered the most precise decelerative growth stage (postpubertal growth stage).
and the most reliable method for overall facial Consequently, the intervals between T1–T2 and T2–
superimpositions. T3 periods are considered as pubertal and postpuber-
Björk’s structural superimposition method has been tal growth stages, respectively. The distribution of the
the gold standard for both overall facial and regional patients according to chronologic age and cervical
superimpositions. It has high validity and moderate-to- maturation stages is indicated in Table 1.
high reproducibility.24 However, this method requires All cephalometric radiographs were traced and
high-quality radiographs and dedication of time and superimposed by one operator. To avoid errors in
effort. Although this limits the application of Björk’s landmark identification, a template with all six land-
method, the more easily applied methods (ie, Steiner, marks (nasion, wing, tuberculum sella, sella, basion,
Ricketts) have lower validity. and pterygomaxillare) was prepared for T1, T2, and T3
Today, we are using multiple superimposition radiographs. This template was used for landmark
techniques including Steiner, Ricketts, the ethmoid identification in all superimpositions. Therefore, errors
triad system, best fit, and the Björk method. This leads that could be due to landmark identification were
to chaos in the evaluation of treatment results.25–28 To eliminated. Then, all radiographs were superimposed
overcome this chaos, a superimposition method that is at the cranial base according to the most commonly
as reliable as the Björk method and as easily used superimposition methods and the newly suggest-
reproducible as the Steiner and Ricketts methods is ed T-W method.
required. Therefore, a two-step project has been The stability of cranial landmarks was evaluated by
planned. The first step is to investigate the stability of using Björk’s structural method. Besides, the degree of
cranial reference landmarks from puberty to adulthood. stability of the cranial landmarks was scored from 0 to
6 in consideration of the directions (horizontal and
The second step is to compare the changes of cranial
vertical) and the study periods (T1–T2, T2–T3, T1–
reference landmarks noted by the Björk, Ricketts, and
T3). The stability score and percentage of points found
Steiner methods with the proposed tuberculum sella-
stable at both directions and in all study periods were
wing (T-W) reference line.
reported as 6 and 100%, respectively. However, for
the points that were found stable neither at the sagittal
nor at the vertical direction along the study, the stability
The material consisted of pretreatment (T1), post- score and rate were indicated as 0 and 0%.
treatment (T2), and postretention (T3) standardized A brief explanation of the superimposition methods
lateral cephalometric radiographs of 30 Class II follows.

A COMPARISON OF DIFFERENT SUPERIMPOSITION METHODS 727
Björk’s Structural Method

The two radiographs were superimposed on the
reference bony structures in the anterior cranial base
as described by Björk and Skieller.7 These anatomical
reference structures are (1) the contour of the anterior
wall of sella turcica, (2) the anterior contour of the median
cranial fossa, (3) the mean intersection point of the lower
contours of the anterior clinoid processes and the
contour of the anterior wall of the sella, (4) the inner
surface of the frontal bone, (5) the contour of the
cribriform plate, (6) the contours of the bilateral
frontoethmoidal crests, and (7) the contour of the median
border of the cerebral surfaces of the orbital roofs.
Steiner (S-N Line)

The two tracings were superimposed on the S-N line
with registration at S point.
Figure 1. Cranial reference landmarks used in this study. N indicates

Ricketts (N-Ba Line) nasion; W, wing point (the intersection of the contour of the ala major
with the jugum sphenoidale); T, tuberculum sella (the intersection
The two tracings were superimposed on the Ba-N point of the lower contours of the anterior clinoid processes and the
line with registration at CC point (the point where the contour of the anterior wall of the sella); S, sella; Ba, basion; and
Ba-N plane and the Ptm-gnathion line intersect). Ptm, pterygomaxillare.
T-W Method Method Error

The two tracings were superimposed on the T-W All procedures were repeated for 10 patients by the
reference line with registration at T point. same orthodontist 1 month later. The reliability of
During the superimposition of the craniofacial measurements was calculated by the Cronbach alpha
structures, the N and S points were transferred from reliability test. Reliability coefficient (0.942–0.999) was
the first film (T1) to the second (T2) and third film (T3) found to yield sufficient reliability.
to serve as fiducial reference points, and the horizontal
(x) and vertical reference planes (y) were constructed RESULTS
using these fixed registration points. The projected
distances between the landmarks and reference The Stability of Cranial Landmarks
planes (x, y) were manually measured using a digital The results of paired t-test revealed that forward
caliper. The differences between the first and second movement of N and backward movement of Ba were
measurements (T1–T2), the second and third mea- statistically significant in all study periods. N point also
surements (T2–T3), and the first and third measure- showed a downward displacement (P , .01) during
ments (T1–T3) were recorded as the amount of puberty and the long follow-up period. Downward
displacement of the landmarks (Figure 1). displacement of Ba was statistically significant in all
study periods. T point remained stable along the study
periods in both vertical and horizontal directions. W
Statistical Method
point was found stable horizontally throughout all study
Paired t-test was used to assess the amount of periods and vertically in the pubertal and long follow-up
displacement of cranial landmarks according to the periods. In the postpubertal period, however, the
Björk method during pubertal (T1–T2), postpubertal downward displacement (0.29 mm) of this point was
(T2–T3), and over all (T1–T3) study periods. One-way significant (P , .05). S point was stable in the
analysis of variance (ANOVA) and Bonferroni test horizontal direction through pubertal and postpubertal
were used for comparison of the displacement of periods; however, in the long follow-up period,
cranial landmarks according to Björk, T-W, Ricketts, backward displacement of this point was significant
and Steiner methods in all study periods. (P , .01). In the vertical direction, S point displaced

Table 2a. Horizontal and Vertical Distances (in mm) of the Cranial Landmarks to the Reference Lines (X, Y) According to the Björk Method in
the Pretreatment (T1), Posttreatment (T2), and Postretention (T3) Periods
T1 T2 T3 Paired t-Test
Mean 6 SD Mean 6 SD Mean 6 SD T1–T2 T2–T3 T1–T3
Nasion Horizontal 102.75 6 6.05 104.59 6 5.84 106.41 6 5.97 *** *** ***
Vertical 24.97 6 0.13 26.18 6 2.14 26.37 6 2.39 ** NS **
Wings Horizontal 56.32 6 3.94 56.46 6 4.09 56.33 6 4.08 NS NS NS
Vertical 25.77 6 1.86 25.95 6 2.08 25.65 6 2.01 NS * NS
Tuberculum sella Horizontal 35.88 6 3.39 35.82 6 3.46 36.01 6 3.64 NS NS NS
Vertical 23.81 6 0.99 23.94 6 1.17 23.70 6 1.01 NS NS NS
Sella Horizontal 31.21 6 3.18 30.99 6 3.39 30.70 6 3.36 NS NS **
Vertical 25.01 6 0.09 25.32 6 0.65 25.37 6 0.67 * NS **
Pterygomaxillare Horizontal 46.36 6 3.62 46.62 6 3.83 46.40 6 3.74 NS NS NS
Vertical 39.79 6 3.31 40.41 6 2.97 41.39 6 2.94 NS * **
Basion Horizontal 5.99 6 2.28 4.63 6 1.95 3.58 6 1.68 *** ** ***
Vertical 58.75 6 3.56 59.70 6 3.39 61.42 6 4.43 ** *** ***
* P , .05; ** P , .01; *** P , .001; NS indicates nonsignificant.
downward both in the pubertal (P , .05) and long DISCUSSION

follow-up periods (P , .01) (Table 2a,b).
There are various error sources in cephalometric
superimposition.29–33 Some of them are inherent and
Comparison of the Methods
unavoidable, but there are also error sources that can
The results of variance analysis and Bonferroni tests be overcome. These include the selection of stable
are shown in Figure 2a through f and Tables 3 through reference landmarks to be used and standardization of
5. Table cells with bold font indicate the values most calibration.
similar to the Björk method. Accordingly, the T-W
method was the most similar to Björk’s superimposition
The Stability of Cranial Base Landmarks
method both in horizontal and vertical directions during
puberty (T1–T2) and in the long follow-up period (T1– The results of the study indicated that T point is the
T3) (Tables 3 and 5). In the pubertal growth stage (T1– most stable (100%) landmark in both directions and
T2), the horizontal displacement of N and Ba points through all stages. This is followed by W point with
was similar among the methods. In this stage of 83% stability. W and Ptm points located in the middle
growth, vertical displacement of all cranial landmarks cranial base remained constant in the horizontal
(N, W, T, and S) except Ptm and Ba showed direction in all study periods. However, W point in the
statistically significant differences between the Björk- postpubertal period (0.29 mm, P , .05) and Ptm both
Ricketts and T-W-Ricketts methods (Table 3). In the in the postpubertal (0.98 mm, P , .05) and long follow-
vertical direction, however, the displacement of all up (1.60 mm, P , .01) periods were displaced
cranial landmarks except Ptm and Ba was different downwards. This vertical displacement decreased
between the Björk-Ricketts (P , .001), T-W-Ricketts stability of Ptm (66%). Displacements shown during
(P , .05), and Steiner-Ricketts (P , .001) methods pubertal and long follow-up periods decreased stability
(Table 4). During the long follow-up period (T1–T3), of S point by 50% (Table 2b).
vertical displacement of all landmarks except Ba W point represents the anterior outline of the middle
showed statistically significant differences between cranial fossa and is based on the morphology of neural
Björk-Ricketts and T-W-Ricketts methods (Table 5). tissues.13 The stability of this landmark in the horizontal
direction has been indicated by longitudinal cephalo-
metric studies13,15 and in dry skulls.13,15 Knott12 indicat-
Table 2b. The Stability Scores of the Cranial Landmarks According
to the Björk Method in Both Directions and Along the Study Periods ed that the distance between point W and the pituitary
(T1–T2, T2–T3, T1–T3) point increased 0.1 mm between the ages of 6 and 9
Landmark Stability Score (0–6) Percent
years and remains completely stable from 9 to 15
years of age. Similarly, Arat et al.18 recently showed
Tuberculum sella 6 100
Wings 5 83
the T-W distance remained stable in all stages of
Pterygomaxillare 4 66 puberty. As mentioned above, in the current study
Sella 3 50 point W remained constant in the horizontal direction in
Nasion 1 16 all of the study periods; however, it moved slightly
Basion 0 0
(0.29 mm, P , .05) downward only in one stage (T2–

Figure 2. The graphical presentation of the displacements (in mm) of cranial reference landmarks according to the superimposition methods
along the study periods: (a) nasion, (b) wings, (c) tuberculum sella, (d) sella, (e) pterygomaxillare, and (f) basion.
T3) of the study. Here, the relationship existing it was reported that reliability of W point is high, and
between the development of the brain and the cranial reproducibility is moderate to high.13,24 In this case, it
base must again be remembered. As reported by was concluded that T and W points can be used in
Enlow,10 development of the cranial base is not solely overall superimpositions representing the middle cra-
based on sutural and synchondrosis activity. To nial base.
accommodate brain expansion, growth of the endo- The results also indicated that Ba is the most
cranial fossa was accomplished by direct cortical drift, variable landmark of the cranial base (0%). This was
involving deposition on the outside, with resorption followed by N point with 16% stability (Table 2b). It has
from the inside. This view explains the reason for been known that development of Ba and N continues
vertical displacement of the cranial base landmarks. for many years due to spheno-occipital synchondrosis
However, this does not reduce stability of W point. on one side and development of the frontal sinus on
The results of the present study indicated that W the other side. This study also showed this finding.
point is sufficiently stable (83%) (Table 2b). Moreover, Therefore, the contradiction arising from superimposi-

Table 3. The Comparison of Displacement (in mm) of the Cranial Landmarks Among Björk, T-W, Ricketts, and Steiner Methods in the Pubertal
Period (T1–T2)a,b
Björk (I) T-W (II) Ricketts (III) Steiner (IV) ANOVA Bonferroni Test
Mean 6 SD Mean 6 SD Mean 6 SD Mean 6 SD Sig I-II II-III III-IV I-III I-IV II-IV
Nasion Horizontal 1.84 6 1.51 1.79 6 1.71 1.29 6 1.82 2.01 6 1.82 NS NS NS NS NS NS NS
Vertical 1.21 6 2.15 1.06 6 2.29 20.52 6 0.62 0.10 6 0.18 *** NS ** NS ** ** NS
Wings Horizontal 0.14 6 0.89 0.23 6 0.91 20.35 6 1.34 0.41 6 0.87 * NS NS * NS NS NS
Vertical 0.18 6 0.80 0.02 6 0.17 20.94 6 1.10 20.42 6 0.66 *** NS ** NS ** ** NS
Tuberculum Horizontal 20.06 6 0.69 -0.01 6 0.06 20.57 6 1.14 0.15 6 0.52 ** NS ** ** ** NS NS
sella Vertical 0.12 6 0.78 0.01 6 0.04 20.76 6 1.32 20.12 6 0.65 *** NS ** ** ** NS NS
Sella Horizontal 20.22 6 0.72 20.17 6 0.53 20.58 6 1.14 0.01 6 0.03 * NS NS * NS NS NS
Vertical 0.31 6 0.64 0.26 6 0.76 20.50 6 1.41 0.01 6 0.06 ** NS ** NS ** NS NS
Pterygo- Horizontal 0.26 6 1.02 0.26 6 1.08 20.12 6 0.49 0.73 6 0.97 ** NS NS ** NS NS NS
maxillare Vertical 0.62 6 1.78 0.56 6 1.67 20.31 6 1.47 0.22 6 1.66 NS NS NS NS NS NS NS
Basion Horizontal 21.37 6 1.63 21.31 6 1.82 21.48 6 1.76 20.84 6 1.38 NS NS NS NS NS NS NS
Vertical 0.95 6 1.78 0.87 6 2.64 0.66 6 0.65 1.24 6 2.19 NS NS NS NS NS NS NS
a
T-W indicates tuberculum sella-wing; ANOVA, analysis of variance; Sig, significance.
b
Bold data indicate the most similar value to the Björk method.
* P , .05; ** P , .01; *** P , .001. NS indicates nonsignificant.
tion methods based on N and Ba points has become displacement of N both in the pubertal and long
definite. follow-up periods differs (P , .01) between the Björk
and Steiner methods. This difference is due to the fact
that downward displacement of N point is masked in
Comparison of the Methods
the Steiner method. Regardless of the reason, the
According to the results of variance analysis, difference between the Björk and Steiner methods may
displacements of all examined cranial landmarks were cause variations when vertical facial changes are
similar between the Björk and T-W methods in both measured. That is, in the Steiner superimposition
directions and at all study periods (Figure 2a through f; method, facial landmarks were displaced upward more
Tables 3 through 5). In fact, the T-W method had than they would in the Björk method. This creates
reduced the anatomy of the cranial base to a reference confusion in the interpretation, particularly of respons-
line passing through the most stable two points of the es to functional/orthopedic interventions.27,28
middle cranial base. Thus, it can be suggested that the The horizontal displacements of all cranial land-
T-W method can replace the Björk superimposition marks, except T point in the pubertal stage, were similar
method. between Björk-Ricketts and T-W-Ricketts methods in all
The horizontal displacements of all cranial land- study periods. Vertically, however, most of the cranial
marks were also similar between the Björk and Steiner landmarks displaced differently in the Ricketts method
methods. In the vertical direction, however, the compared to both the Björk and the T-W methods
Table 4. The Comparison of Displacement (in mm) of the Cranial Landmarks Among the Björk, T-W, Ricketts, and Steiner Methods in the
Postpubertal Period (T2–T3)a,b
Nasion Horizontal 1.82 6 1.66 1.90 6 1.82 1.79 6 2.00 1.98 6 2.09 NS NS NS NS NS NS NS
Vertical 0.19 6 2.18 0.56 6 1.92 20.57 6 0.73 20.01 6 0.16 * NS * NS NS NS NS
Wings Horizontal 20.13 6 1.06 20.28 6 0.73 20.08 6 1.32 0.17 6 0.90 NS NS NS NS NS NS NS
Vertical 20.29 6 0.72 20.01 6 0.18 21.24 6 1.26 20.30 6 0.55 *** NS * * * NS NS
Tuberculum sella Horizontal 0.19 6 0.99 0.01 6 0.06 0.13 6 1.16 0.44 6 0.40 NS NS NS NS NS NS NS
Vertical 20.24 6 0.82 0.00 6 0.02 21.17 6 1.61 20.33 6 0.77 *** NS * * * NS NS
Sella Horizontal 20.29 6 0.94 20.41 6 0.48 20.27 6 1.19 0.01 6 0.10 NS NS NS NS NS NS NS
Vertical 0.05 6 0.84 0.18 6 0.80 20.99 6 1.35 0.01 6 0.11 *** NS * * NS NS NS
Pterygomaxillare Horizontal 20.22 6 1.23 20.31 6 1.11 20.11 6 0.59 0.13 6 1.03 NS NS NS NS NS NS NS
Basion Horizontal 21.05 6 1.78 21.10 6 2.03 21.22 6 1.98 20.50 6 1.81 NS NS NS NS NS NS NS
a
b
* P , .05; *** P , .001. NS indicates nonsignificant.

Table 5. The Comparison of Displacement (in mm) of the Cranial Landmarks Among the Björk, T-W, Ricketts, and Steiner Methods in All
Periods (T1–T3)a,b
Nasion Horizontal 3.66 6 2.53 3.69 6 3.02 3.07 6 2.69 3.99 6 2.97 ns NS NS NS NS NS NS
Vertical 1.40 6 2.41 1.62 6 2.80 21.09 6 0.83 0.09 6 0.15 *** NS * NS * * *
Wings Horizontal 0.01 6 1.18 20.05 6 1.12 20.43 6 1.57 0.58 6 0.97 * NS NS * NS NS NS
Vertical 20.12 6 0.75 0.01 6 0.17 22.18 6 1.55 20.72 6 0.87 *** NS * * * NS *
Tuberculum Horizontal 0.13 6 0.83 0.00 6 0.00 20.44 6 1.47 0.59 6 0.49 *** NS NS * NS NS NS
sella Vertical 20.12 6 0.76 0.01 6 0.04 21.93 6 1.94 20.44 6 0.68 *** NS * * * NS NS
Sella Horizontal 20.51 6 0.77 20.58 6 0.54 20.84 6 1.53 0.02 6 0.09 ** NS NS * NS NS NS
Vertical 0.35 6 0.66 0.45 6 0.72 21.49 6 2.00 0.02 6 0.09 *** NS * * * NS NS
Pterygo- Horizontal 0.03 6 1.35 20.05 6 1.46 20.23 6 0.67 0.86 6 1.40 ** NS NS * NS NS *
maxillare Vertical 1.60 6 2.70 1.77 6 2.52 20.39 6 2.05 1.17 6 2.51 ** NS * NS * NS NS
Basion Horizontal 22.41 6 2.41 22.41 6 2.53 22.70 6 2.60 21.33 6 2.05 ns NS NS NS NS NS NS
Vertical 2.66 6 2.67 2.45 6 3.62 1.01 6 0.93 2.78 6 3.07 ns NS NS NS NS NS NS
a
b
* P , .05; ** P , .01; *** P , .001. NS indicates nonsignificant.
through all study periods. The different behaviors of the this manuscript and Dr Ozgur Koskan from the Süleyman
cranial landmarks, particularly of N, would conceal the Demirel University Faculty of Agriculture, Department of Animal
Sciences for his help with the statistical analysis.
changes of the face in the vertical direction.17,33
The results of this study indicate that the T-W
REFERENCES
method is as reliable as Björk’s structural superimpo-
sition method and is easily applied. Therefore, this 1. Bergersen EO. A comparative study of cephalometric
study has encouraged us to propose the T-W superimposition. Angle Orthod. 1961;31:216–229.
2. Steuer I. The cranial base for superimposition of lateral
superimposition method for examining overall facial cephalometric radiographs. Am J Orthod. 1972;61:493–500.
changes both in the active growth period and in the 3. Moss ML, Greenberg SN. Postnatal growth of the human
long follow-up periods. skull base. Angle Orthod. 1955;25:77–84.
4. Björk A. Cranial base development. Am J Orthod. 1955;41:
CONCLUSIONS 198–255.
5. Melsen B. The cranial base. Acta Odontol Scand. 1974;
N Among the landmarks investigated in this study, T 32(suppl 62):1–126.
and W points were the most stable points of the 6. Ghafari J, Engel FE, Laster LL. Cephalometric superimposi-
tion on the cranial base: a review and a comparison of four
cranial base.
methods. Am J Orthod Dentofacial Orthop. 1987;91:403–413.
N Superimposition at T point along the T-W line was a reli- 7. Björk A, Skieller V. Normal and abnormal growth of the
able method when examining the overall facial changes mandible. A synthesis of longitudinal cephalometric implant
both in the active growth and long follow-up periods. studies over a period of 25 years. Eur J Orthod. 1983;5:
N N and Ba points were the most variable points both in 1–46.
the horizontal and vertical directions along all study 8. Moss ML, Salentijin L. The primary role of functional
matrices in facial growth. Am J Orthod. 1969;55:566–577.
periods. The stability of S and Ptm points was 9. Nepola SR. The intrinsic and extrinsic factors influencing the
somewhat questionable, especially in the long follow- growth and development of the jaws: heredity and functional
up period. This devaluates the reliability of the matrix. Am J Orthod. 1969;55:499–505.
superimposition methods depending on those points 10. Enlow DH. Handbook of Facial Growth. 2nd ed. Philadel-
in the studies of longitudinal basis. phia, Pa: WB Saunders; 1982.
11. Ford EHR. Growth of the human cranial base. Am J Orthod.
N The behavior of the cranial landmarks differed 1958;44:498–506.
among the superimposition methods. The differenc- 12. Knott VB. Ontogenetic change of four cranial base
es were more prominent in the vertical direction. segments in girls. Growth. 1969;33:123–142.
N T-W was the superimposition method most similar to 13. van der Linden F, Enlow DH. A study of the anterior cranial
Björk’s structural method in both directions along the base. Am J Orthod. 1971;41:119–124.
14. Ricketts RM. Provocations and Perceptions in Craniofacial
study.
Orthopedics. Vol I. San Diego, Calif: Rocky Mountain, Inc;
1989.
ACKNOWLEDGMENT 15. Hilloowala RA, Trent RB, Pifer RG. Interrelationships of
brain cranial base and mandible. Cranio. 1998;16:267–274.
We thank Dr Emel Arat from the University of Toronto, 16. Moore AW. Observation on facial growth and its clinical
Department of Orthodontics for her assistance in preparation of significance. Am J Orthod. 1959;45:399–423.

17. Arat ZM, Rubenduz M, Akgul AA. The displacement of 25. Coben SE. The sphenoccipital synchondrosis: the missing
craniofacial reference landmarks during puberty: a compar- link between the profession’s concept of craniofacial growth
ison of three superimposition methods. Angle Orthod. 2003; and orthodontic treatment. Am J Orthod Dentofacial Orthop.
73:374–380. 1998;114:709–712.
18. Arat M, Koklu A, Ozdiler E, Rubenduz M, Erdogan B. 26. Buschang PH, Santos-Pinto A. Condylar growth and glenoid
Craniofacial growth and skeletal maturation: a mixed fossa displacement during childhood and adolescence.
longitudinal study. Eur J Orthod. 2001;23:355–363. Am J Orthod Dentofacial Orthop. 1998;113:437–442.
19. Björk A. The use of metallic implants in the study of facial 27. Sellke TA, Cook AH. Reply, reader’s forum. Am J Orthod
growth in children: method and application. Am J Phys Dentofacial Orthop. 1995;107:18A.
Anthropol. 1968;29:243–254. 28. Swartz ML. Comment on superimposition techniques.
20. Viazis AD. The cranial base triangle. J Clin Orthod. 1991;25: Am J Orthod Dentofacial Orthop. 1995;107:17A–18A.
565–570. 29. Baumrind S, Frantz RC. The reliability of head film
21. Tollaro I, Bacetti T, Franchi L. Mandibular skeletal changes measurements. 1. Landmark identification. Am J Orthod.
induced by early functional treatment of Class III malocclu- 1971;60:111–127.
sion. A superimposition study. Am J Orthod Dentofacial 30. Houston WJB, Lee RT. Accuracy of different methods of
Orthop. 1995;108:525–532. radiographic superimposition on cranial base structures.
22. Nelson TO. Analysis of facial growth utilizing elements of Eur J Orthod. 1985;7:127–135.
the cranial base as registrations. Am J Orthod. 1960;46: 31. Houston WJB, Maher RE, McElroy D, Sherriff M. Sources of
379. error in measurements from cephalometric radiographs.
23. Melsen B, Melsen F. The postnatal development of the Eur J Orthod. 1986;8:149–151.
palatomaxillary region studied on human autopsy material. 32. You QL, Hagg U. A comparison of three superimposition
Am J Orthod. 1982;82:329–342. methods. Eur J Orthod. 1999;21:717–725.
24. Athanasiou AE. Orthodontic Cephalometry. London, UK: 33. Houston WJB. The analysis of errors in orthodontic
Mosby-Wolfe; 1995. measurements. Am J Orthod. 1983;83:382–390.

Original Article
3-D analysis of facial asymmetry in children with hip dysplasia

Shane Rex Tollesona; Chung How Kaub; Robert P. Leec; Jeryl D. Englishd; Virpi Harilae;
Pertti Pirttiniemif; Marita Valkamag
ABSTRACT
Objective: To determine whether facial asymmetry existed in patients with developmental
dysplasia of the hip (DDH).
Materials and Methods: Subjects consisted of children between ages 5 and 10 years having
DDH, and treated by the Von Rosen splint method. Three-dimensional (3-D) facial photographs
were taken on all subjects using the 3dMDface system. Using RF6 PP2 software, anthropometric
landmarks were plotted and used to calculate asymmetry based on 3-D coordinates in a reference
framework.
Results: Of a total of 60 subjects with a mean age of 8 years (SD, 1.4 years), 30 had dysplasia of
the left hip; 13, of the right; and 17 were bilateral. Twenty-seven subjects had upper face (UF)
dominance values of 2 mm or more; of those, 26 were right-side dominant. Twenty-four subjects
(40%) had a chin-point (CP) deviation of 2 mm or more; of those, 21 had right-side deviations.
Statistically, UF and CP deviations were not significantly independent of each other (P . .05).
Thirty percent of subjects had a posterior dental crossbite.
Conclusions: The results indicate that facial asymmetry exists in patients with DDH. (Angle
Orthod. 2010;80:707–712.)
KEY WORDS: 3-D analysis; Facial asymmetry
INTRODUCTION tries in DDH children. It might seem reasonable to

expect that craniofacial asymmetries would translate
Developmental dysplasia of the hip (DDH), the
into concomitant dental asymmetries, and studies do
etiology of which is multifactorial, is medically well
exist showing statistically significant associations
known and usually diagnosed in infancy.1 In the
between facial and dental asymmetry,2 suggesting an
literature, there is very little regarding facial asymme-
increased need for orthodontic therapy. Because some
of these problems may require orthopedic correction
a
Resident, Department of Orthodontics, University of Texas during the growth period or surgical management
Health Science Center at Houston, Houston, Texas. later,3 additional information on the development and
b
Professor and Department Chair, Department of Orthodon- detection of facial asymmetry is important in ortho-
tics, University of Alabama, Birmingham, Ala. dontic treatment planning.
c
Clinical Assistant Professor, Department of Orthodontics,
University of Texas Health Science Center at Houston, Houston, Measurement of facial asymmetry using three-
Tex. dimensional (3-D) photography is a relatively new
d
Professor and Department Chair, Department of Orthodon- concept, with just a few studies exploring its capa-
tics, University of Texas Health Science Center at Houston, bilities4–6 and no standard technique having yet
Houston, Tex.
been accepted. Past methods of study have
e
Associate Professor, Department of Orthodontics, University
of Oulu, Oulu, Finland. included various 2-D photographs and radio-
f
Professor and Department Chair, Department of Orthodon- graphs. These records have proven useful but they
tics, University of Oulu, Oulu, Finland. are limited because of their 2-D representation of
g
Pediatrician, University of Oulu, Oulu, Finland. a 3-D structure. Three-dimensional photography
Corresponding author: Dr Chung How Kau, Professor
seems to solve many of these dilemmas, as it is
and Department Chair, Department of Orthodontics, University
of Alabama, 1919 7th Avenue South SDB305, Birmingham, AL noninvasive and does not expose subjects to radiation.
(e-mail: chung.h.kau@inbox.com) This technology allows accurate representation of
Accepted: November 2009. Submitted: August 2009. facial soft tissue and morphologies,7–9 and it can
G 2010 by The EH Angle Education and Research Foundation, be used to compare10,11 and predict orthodontic
Inc. outcomes.12–14

708 TOLLESON, KAU, LEE, ENGLISH, HARILA, PIRTTINIEMI, VALKAMA
This study was designed to determine whether facial

asymmetry exists in patients with DDH and to evaluate
a method for the study of facial asymmetries and
malocclusions in children born with DDH and treated
with splint therapy.

Subjects
The subjects recruited for this study had to meet the
following inclusion criteria:
N Subjects were born during the years 1997–2001 in
Northern Ostrobothnia.
N Hospital District of Oulu, Finland.
N Subjects had DDH.
N Subjects were treated by the Von Rosen method.
All children with previously diagnosed plagiocephaly
or craniosynostosis were excluded from the study.
Imaging System
The imaging system used in this study was the
portable 3dMDface System (3dMD, Atlanta, GA), an
imaging system that combines stereophotogrammetry Figure 1. A sample illustrating the anthropometric landmarks used.
and structured light techniques.12 This system uses a
multicamera configuration, with three cameras on each
positions along an x-, y-, and z-coordinate system. The
side (one color and two infrared), that records high-
3-D evaluation methods described below were pat-
quality, photo-realistic pictures. It is able to capture full
terned after a recent study using the same software.4
facial images from ear to ear and under the chin in 1.5
milliseconds at the highest resolution. Manufacturer’s
3-D Evaluation Methods
stated accuracy is less than 0.5 mm, and the quoted
clinical accuracy is 1.5% of total observed variance.15 Each craniofacial image was oriented in the com-
Three-dimensional surface images captured by surface puter virtual space to have a natural head position
acquisition systems are highly repeatable and pre- before analysis. This was done by confirming the
cise.16,17 orientation of the interpupillary line to be parallel to the
Images taken with the 3dMDface System were horizontal x-axis (from both a frontal and coronal view)
analyzed and viewed on a computer using the and the orientation of the patient’s line of sight parallel
3dMDpatient Software Platform. to horizontal (z-axis). Various anthropometric land-
marks were chosen for analysis and were identified by
Image Acquisition marking them on the surface of the facial contour using
the cursor. Anthropometric landmarks included five
Images were acquired with the subjects in their
midline points (n, prn, sn, ls, pg) and four bilateral
natural head position, which has proven to be clinically
points (ex, en, ac, cph) (Figure 1 and Table 1).
reliable.18 The subjects sat on an adjustable chair with
Landmarks were chosen carefully so as to be
their face centered on a computer screen and were
easily identifiable and repeatable, and they appear
asked to keep the facial musculature as relaxed as
as color points with reference coordinates. The
possible.
surface shell was translated in the 3-D space so as
to center soft tissue nasion (n) as point (0, 0, 0) in the
Image Analysis
x-, y-, z-coordinate system, Figure 2. The values of
All images acquired were transferred to a reverse other points’ coordinates therefore represent distances
modeling software package, Rapidform 2006 Plus Pack from n on the chosen axis in millimeters; their
2 (RF6 PP2) (INUS Technology, Seoul, Korea) for corresponding positive or negative value indicates
analysis.19 The software allows the surface data to be directions (ie, positive x 5 left, positive y 5 up,
assessed as a collection of points interrelated by their positive z 5 anterior).

3-D ANALYSIS OF FACIAL ASYMMETRY 709
Table 1. Definitions of Anthropometric Landmarks Used in

This Study
Abbreviation Landmark
N nasion
En endocanthion
Ex exocanthion
T tragus
Prn pronasale
Ac alar curvature
Sn subnasale
Cph crista philtri
Ls labiale superius
Ch cheilion
Pg pogonion
N Finally, UF dominance was computed using these

raw x-values for midline landmarks and differences in
x-values for bilateral landmarks with the appropriate
negative or positive sign assigned to correlate with
the side of dominance. For each subject, x-values
were summated and averaged to formulate a
negative or positive value to represent a right or left
dominance in the UF.
Figure 2. Illustration of the x-, y-, and z-coordinate system centered
at landmark nasion. (z-axis is oriented perpendicular to the page). CP Deviation analysis. Chin deviation can be one of
the most notable indicators of facial asymmetry, as
laterality is most common on the lower one-third of the
Parameters Measured face.20,21 CP deviation was measured by the x-value of
landmark pg, which represents the deviation from the
The following three parameters were studied and facial midline in millimeters. This was recorded for
evaluated: each subject, and a positive value represents a left
UF dominance analysis deviation while a negative value represents a right
N All anthropometric landmarks except pg were con- deviation. Based on previous literature, a distance of
sidered in the upper two-thirds of the face (above the less than 2 mm was considered to be within normal
mandibular region) and were used to calculate the facial proportions of symmetry and therefore not
side of dominance of the upper face (UF). For these significant.21–23
points, the more lateral (x-axis) the landmark, the Dental occlusion analysis. Subjects underwent a
more dominant is the landmark to that side of the clinical dental exam, which recorded the presence of
face. posterior cross bites (left, right, or bilateral).
N For midline landmarks (prn, sn, ls), x-values were
recorded, representing their deviation from the facial
midline. By conventions set out in this study, a STATISTICAL ANALYSIS
negative value for x represents a deviation toward
the right side of the face. Theoretically, a perfectly Paired t-tests (SPSS 17.0, Chicago, Ill) were used to
symmetrical face would have an x-value of 0 for determine whether significant differences existed
every midline point. between UF and CP deviations in the x-direction of
N For bilateral landmarks (ex, en, ac, cph), differences space by comparing UF and CP values in all subjects
in the left and right x-values represent their degree of with
asymmetry in that plane of space. Theoretically, Pair 1: significant UF dominance values.
bilateral landmarks on a perfectly symmetrical face Pair 2: significant CP deviation values.
should have equal but opposite values for the x-
coordinate (eg, right 5 25 mm; left 5 +5 mm).
Differences in corresponding x-values for bilateral RESULTS
landmarks were recorded. Negative values were
Subjects
assigned to all right-side dominant measures,
and positive values for all left-side dominant mea- A total of 60 out of 130 subjects voluntarily
sures. participated in the study. All were children between

Table 2. Results of the Side of DDH of Subjects Participating in Table 3. Results of the Three Parameters Measured
the Study
Results of Parameters Measured
Subjects
Significant (2 mm or more)
Side of DDHa Male Female Total
Upper Face Chin-Point Dental Cross
Left 11 19 30 Side Dominance Deviation Bite
Right 2 11 13
Left 1 3 5
Bilateral 7 10 17
Right 26 21 7
Total 20 40 60
Bilateral n/a n/a 6
a
DDH indicates developmental developmental dysplasia of the Total 27 24 18
hip.
DISCUSSION
ages 5 and 10 years, and the average age of all Until now, few studies have focused on analyzing
subjects was 8 years (see Table 2 for breakdown). facial asymmetries and malocclusions specifically
Ten measurements were randomly made on five within a population of DDH children. Vlimmeren,24 in
subjects by two investigators to test inter- and his review of diagnostic strategies for the evaluation of
intraobserver reliability. A t-test showed no statistically asymmetry in infants, stated that children with defor-
significant differences between operators. mational plagiocephaly have an elevated risk of
mandibular asymmetry, and cranial asymmetry was
Parameters Measured found in about 30% of infants with muscular torticollis.
Studies exist showing statistically significant associa-
UF dominance analysis. Using the average value tions between facial and dental asymmetry.2 Pirttiniemi
from all UF landmarks, 27 subjects (45%) had a et al.22 found significant asymmetries of the facial
significant (2 mm or more) UF asymmetry in the x- skeleton and dental arches with muscular torticollis.
dimension. Of these, 26 were right-side dominant and The patients had a high prevalence of treated or
only 1 was left-side dominant (Table 3). Looking at the diagnosed lateral malocclusions (50%) compared with
sample as a whole, the average UF dominance value the control group (12.5%), and they also had more
was 21.5 mm, with a range from 24.6 mm to 9.7 mm. dental arch asymmetry and midline deviation in the
Without regard to right or left, by taking the absolute maxillary than in the mandibular arch.
value of each number, the average asymmetry value
In addition, a few reports in the literature link DDH to
for the UF was 1.8 mm.
these asymmetric head and neck disorders. Watson
CP deviation analysis. Regarding lower face asym- studied the relation between the side of plagiocephaly,
metry, 24 subjects (40%) had a CP deviation of 2 mm dislocation of hip, scoliosis, bat ears, and sternomas-
or more. Within this group, 21 had right-side chin
toid tumor.25 Iwahara and Ikeda26 reported that 14.8%
deviations (87.5%) and 3 had left-side chin deviations
of their patients with congenital muscular torticollis also
(12.5%; Table 3). In all 60 subjects, the average chin
had dysplasia of the hip, whereas Hummer and
deviation was 21.1 mm, ranging from 25.1 mm to
MacEwen observed 20%.27 Cady1 reiterated that the
7.6 mm. Without regard to right or left, by taking the
risk factors generally used for DDH are questionable
absolute value of each number, the average deviation
physical exam, female sex, breech presentation, and
of pg from the facial midline was 1.9 mm.
positive family history, and added that there is some
Dental occlusion analysis. Posterior dental cross
evidence that torticollis may be a risk factor as well.
bites were present in 18 of the 60 subjects (30%), with
Some studies even relate the side of the disorder (left
7 being right, 5 left, and 6 bilateral (Table 3).
or right) to the side of craniofacial asymmetry. For
example, infants with posterior deformational plagio-
cephaly are characterized by having the ipsilateral ear
Statistical analysis to determine whether signif- and cheek anteriorly displaced, and mandibular asym-
icant differences existed between UF and CP devia- metry with deviation toward the unaffected side.28,29
tions in the x-direction of space indicated that there Kane et al.,30 in a computer tomography study, reported
were no statistically significant differences for UF vs a 3.8% larger hemimandibular volume, a 3.5% shorter
CP (pair 1) and CP vs UF (pair 2) (P . .05). The ramal height, and a 3% longer mandibular body length
results suggest that subjects in this study with a on the ipsilateral side of the occipital flattening.
presenting asymmetry in either the UF or CP tended to Watson25 related this to DDH, reporting that the flat
be asymmetric on the same side in the corresponding temple in plagiocephaly and a unilateral congenitally
parameter (Table 4). dislocated hip tend to be on the same side.

3-D ANALYSIS OF FACIAL ASYMMETRY 711
Table 4. Results of Paired t-Test for Different Groupingsa

Paired Sample Test
95% Confidence Interval of the Difference
Mean Std. Deviation Lower Upper Sig. (2-tailed)
Pair 1 CPb1 - UFc1 0.08 1.18 20.39 0.54 0.73
Pair 2 CP2 - UF2 20.38 1.28 20.92 0.17 0.17
a
Pair 1 included the CP and UF values for all subjects with UF dominance values .2 mm (CP12UF1). Pair 2 included the CP and UF values
for all subjects with CP deviation values .2 mm (CP22UF2).
b
CP indicates chin point.
c
UF indicates upper face.
The image analysis and 3-D evaluation methods and 200432,33 reported the incidence of cross bites in
used in this study mimicked techniques from a recently Finnish children to be 13% and 4%–10%, respectively.
published report studying asymmetry using the RF6 This information suggests a much higher prevalence of
software.4 In addition, the use of anthropometric this malocclusion than that in the general population.
landmarks with the 3dMDface System have been
proven valid and reliable.31 CONCLUSIONS
In our study, a careful inspection of the data found
N The results indicate that facial asymmetry exists in
no strong relationship between the side of DDH
patients with DDH.
compared with the side of UF dominance, the side of
CP deviation, or the side of dental cross bite.
REFERENCES
Surprisingly, there was also no strong relationship
between CP deviation and the presence of a posterior 1. Cady RB. Developmental dysplasia of the hip: definition,
cross bite. Only 5 of the 24 significant CP deviations recognition, and prevention of late sequelae. Pediatr Ann.
2006;35:92–101.
had concomitant posterior cross bites, which might
2. Sheats RD, McGorray SP, Musmar Q, Wheeler TT, King GJ.
influence the asymmetric mandibular position. Prevalence of orthodontic asymmetries. Semin Orthod.
In this study, 45% of the study sample had 1998;4:138–145.
significant (2+ mm) UF asymmetry; of those, 26 of 27 3. Bishara SE, Burkey PS, Kharouf JG. Dental and facial
were right-side dominant. Additionally, 40% of the asymmetries: a review. Angle Orthod. 1994;64:89–98.
4. Bozic MKC, Richmond S, Maja O, Hren NI. A novel method
study sample had significant (2+ mm) CP deviations;
of analysis of Class III patients using 3-dimensional soft
of those, 21 of 24 were right-side deviations. Looking tissue analysis: Elsevier Editorial System(tm) for Am J
at the sample as a whole, both the average UF Orthod Dentofacial Orthop. In press. 2008.).
dominance (21.5 mm) and the average CP deviation 5. Ferrario VF, Sforza C, Miani A Jr, Serrao G. A three-
(21.1 mm) were negative numbers. This reflects a dimensional evaluation of human facial asymmetry. J Anat.
majority of right-side facial tendencies in the study 1995;186(pt 1):103–110.
6. Hartmann J, Meyer-Marcotty P, Benz M, Hausler G, Stellzig-
sample. The results also suggest that subjects in this Eisenhauer A. Reliability of a method for computing facial
study did not have upper or lower facial asymmetry symmetry plane and degree of asymmetry based on 3D–
independent from the other, as a presenting asymme- data. J Orofac Orthop. 2007;68:477–490.
try in either the UF or CP tended to be asymmetric on 7. Kau CH, Cronin AJ, Richmond S. A three–dimensional
the same side in the corresponding parameter. evaluation of postoperative swelling following orthognathic
surgery at 6 months. Plast Reconstr Surg. 2007;119:
According to a criterion used in previous studies, 2192–2199.
lateral deviation of 2 mm or more was employed as a 8. McCance AM, Moss JP, Wright WR, Linney AD, James DR.
critical value to separate asymmetry from symmetry.31 A three-dimensional soft tissue analysis of 16 skeletal Class
Severt and Proffit21 reported that, in patients showing III patients following bimaxillary surgery. Br J Oral Maxillofac
dentofacial deformity including jaw deviation, laterality Surg. 1992;30:221–232.
9. Moss JP, Ismail SF, Hennessy RJ. Three-dimensional
toward the left side was present in more than 85% of
assessment of treatment outcomes on the face. Orthod
their sample. Our study of DDH children, however, Craniofac Res. 2003;6(suppl 1):126–131; discussion 79–82.
showed quite the opposite, with 87.5% of the 10. Hajeer MY, Ayoub AF, Millett DT. Three-dimensional
significant CP deviations being toward the right side. assessment of facial soft-tissue asymmetry before and after
Perhaps this suggests something different occurring orthognathic surgery. Br J Oral Maxillofac Surg. 2004;42:
with subjects having DDH as compared with the 396–404.
11. Palomo JM, Hunt DW Jr, Hans MG, Broadbent BH Jr. A
general population. longitudinal 3-dimensional size and shape comparison of
Another interesting finding was the number of untreated Class I and Class II subjects. Am J Orthod
subjects with cross bites—30%. Two studies in 1990 Dentofacial Orthop. 2005;127:584–591.

12. Kau CH, Richmond S, Incrapera A, English J, Xia JJ. Three- 23. Haraguchi S, Takada K, Yasuda Y. Facial asymmetry in
dimensional surface acquisition systems for the study of subjects with skeletal Class III deformity. Angle Orthod.
facial morphology and their application to maxillofacial 2002;72:28–35.
surgery. Int J Med Robot. 2007;3:97–110. 24. van Vlimmeren LA, Helders PJ, van Adrichem LN,
13. Mah J, Ritto AK. Imaging in orthodontics: present and future. Engelbert RH. Diagnostic strategies for the evaluation of
J Clin Orthod. 2002 Nov;36:619–625. asymmetry in infancy—a review. Eur J Pediatr. 2004;163:
14. Marmulla R, Hassfeld S, Luth T, Muhling J. Laser- 185–191.
scan-based navigation in cranio-maxillofacial surgery. 25. Watson GH. Relation between side of plagiocephaly,
J Craniomaxillofac Surg. 2003;31:267–277. dislocation of hip, scoliosis, bat ears, and sternomastoid
15. Aldridge K, Boyadjiev SA, Capone GT, DeLeon VB, tumours. Arch Dis Child. 1971;46:203–210.
Richtsmeier JT. Precision and error of three– 26. Iwahara T, Ikeda S. On the ipsilateral involvement of
dimensional phenotypic measures acquired from 3dMD congenital muscular torticollis and congenital dislocation of
photogrammetric images. Am J Med Genet A. 2005; the hip. J Jpn Orthop Assn. 1962;35:1221.
15;138A:247–253. 27. Hummer CD, MacEwen GD. The coexistence of torticollis
16. Kau CH, Zhurov A, Richmond S, Cronin A, Savio C, Mallorie and congenital dysplasia of the hip. J Bone Joint Surg Am.
C. Facial templates: a new perspective in three dimensions. 1972;54:1255–1256.
Orthod Craniofac Res. 2006;9:10–17. 28. Huang MH, Gruss JS, Clarren SK, Mouradian WE, Cunning-
ham ML, Roberts TS, et al. The differential diagnosis of
17. Kau CH, Hunter LM, Hingston EJ. A different look:
posterior plagiocephaly: true lambdoid synostosis versus
3-dimensional facial imaging of a child with Binder syn-
positional molding. Plast Reconstr Surg. 1996;98:765–774;
drome. Am J Orthod Dentofacial Orthop. 2007;132:
discussion 75–76.
704–709.
29. Mulliken JB, Vander Woude DL, Hansen M, LaBrie RA,
18. Chiu CS, Clark RK. Reproducibility of natural head position.
Scott RM. Analysis of posterior plagiocephaly: deformational
J Dent. 1991;19:130–131.
versus synostotic. Plast Reconstr Surg. 1999;103:371–380.
19. Zhurov AI K, C.H, Richmond S. Computer Methods for 30. Kane AA, Mitchell LE, Craven KP, Marsh JL. Observations
Measuring 3D Facial Morphology. In: J Middleton on a recent increase in plagiocephaly without synostosis.
MGSaMLJ, ed, Computer Methods in Biomechanics & Pediatrics. 1996;97(6; Pt 1):877–885.
Biomedical Engineering–5. FIRST Numerics Ltd, Cardiff, 31. Wong JY, Oh AK, Ohta E, Hunt AT, Rogers GF, Mulliken JB,
UK. et al. Validity and reliability of craniofacial anthropometric
20. Samman N, Tong AC, Cheung DL, Tideman H. Analysis of measurement of 3D digital photogrammetric images. Cleft
300 dentofacial deformities in Hong Kong. Int J Adult Orthod Palate Craniofac J. 2008:232–299.
Orthog Surg. 1992;7:181–185. 32. Mugonzibwa EA, Eskeli R, Kuijpers-Jagtman AM, Laine-
21. Severt TR, Proffit WR. The prevalence of facial asymmetry Alava MT, van’t Hof MA. Occlusal characteristics during
in the dentofacial deformities population at the University of different emergence stages of the permanent dentition in
North Carolina. Int J Adult Orthod Orthog Surg. 1997;12: Tanzanian Bantu and Finnish children. Eur J Orthod.
171–176. 2004:251–260.
22. Pirttiniemi P. Normal and increased functional asymmetries 33. Kerosuo H. Occlusion in the primary and early mixed
in the craniofacial area. Acta Odontol Scand. 1998;56: dentitions in a group of Tanzanian and Finnish children.
342–345. ASDC J Dent Child. 1990;57:293–298.

Original Article
Influence of ceramic (feldspathic) surface treatments on the

micro-shear bond strength of composite resin
Sumit Yadava; Madhur Upadhyayb; Gilberto Antonio Borgesc; W. Eugene Robertsd
ABSTRACT
Objective: To test the null hypothesis that surface treatment has no influence on the micro-shear
bond strength between orthodontic composite resin cement and ceramics (feldspathic porcelain).
Materials and Methods: Circular specimens of feldspathic porcelain were fabricated and
randomly divided into six groups: (1) no treatment; (2) treatment with a mixture of acidic primer and
silane agent for 20 seconds; (3) etching with 9.5% hydrofluoric acid; (4) etching with 9.5%
hydrofluoric acid and coating with a mixture of acidic primer and silane agent for 20 seconds; (5)
airborne-particle abrasion with 50-mm aluminum oxide; and (6) airborne-particle abrasion and
coating with a mixture of acidic primer and silane agent for 20 seconds. The porcelain disks were
then bonded to resin cylinders with composite resin cement. A micro-shear bond test was carried
out to measure the bond strength. Moreover, each ceramic surface was observed morphologically
by scanning electron microscopy. One-way analysis of covariance was used to compare the
groups for differences in micro-shear bond strength.
Results: The mean micro-shear bond strength varied as a function of surface treatment. It ranged
from 3.7 to 20.8 MPa. The highest values for micro-shear bond strength were found when the
surface was acid-etched with hydrofluoric acid and coated with silane. On the other hand, the
control group (no treatment) had significantly lower micro-shear bond strength than all the other
groups.
Conclusion: The null hypothesis that the surface treatment has no influence on the micro-shear
bond strength of orthodontic composite resin was rejected. The bond strength between ceramics
and orthodontic resin cement is affected by the ceramic surface treatment. The bond failure was of
the adhesive type, except with the hydrofluoric acid + silane group, where it was a cohesive bond
failure. (Angle Orthod. 2010;80:765–770.)
KEY WORDS: Orthodontic resin; Cementation; Surface treatment; Shear bond strength; Acid
etching; Abrasion
INTRODUCTION
PhD Student, Mineralized Tissue Histology and Research
a
Dental ceramics have increasingly become the
Laboratory, Department of Orthodontics, Indiana School of
Dentistry, Indianapolis, Indiana; Graduate Researcher, Dental
material of choice for achieving natural-looking resto-
Biomechanics Laboratory, Department of Mechanical Engineer- rations that can substitute for destroyed or missing
ing, Indiana University–Purdue University, Indianapolis, Ind. anterior teeth.1,2 These materials have desirable
b
Division of Orthodontics, Department of Craniofacial Scienc- characteristics, such as chemical stability, biocompat-
es, School of Dentistry, University of Connecticut, Farmington, ibility, high compressive strength, and a coefficient of
Conn.
thermal expansion that is similar to that of natural tooth
c
Assistant Professor, Biomaterials Laboratory, University of
Uberaba, Minas Gerais, Brazil. structure, thereby producing more natural restorations.
d
Professor, Department of Orthodontics, Indiana University According to the NHANES III data, the frequency of
School of Dentistry, Indianapolis, Ind. malocclusion in adults in the United States is around
Corresponding author: Dr Sumit Yadav, Mineralized Tissue 50%.3 The prevalence of malocclusion in Western
Histology and Research Laboratory, Section of Orthodontics,
European adults is between 40% and 76%.4 Data from
Indiana School of Dentistry, DS 238, 1121 W. Michigan Street,
Indianapolis, IN 46202 previous research indicate that the prevalence of
(e-mail: syadav@iupui.edu) malocclusion in adults is equal to or greater than that
Accepted: December 2009. Submitted: August 2009. of children and adolescents.5 With the increase in adult
G 2010 by The EH Angle Education and Research Foundation, orthodontic treatment comes the need to find a reliable
Inc. method for bonding orthodontic brackets onto metal or

766 YADAV, UPADHYAY, BORGES, ROBERTS
Table 1. Surface Treatments Applied to Each Group of Disks

Group
Surface Treatment 1 2 3 4 5 6
Acidic primer + silane agent (20 s) X X X
9.5% hydrofluoric acid (60 s) X X
Airborne-particle abrasion X X
No surface treatment X Figure 1. Formation of resin cylinders on ceramic surfaces.
ceramic restorations. Over the past few years, a (according to the manufacturer’s instructions). Sub-
number of mechanical and chemical retention systems sequently, the specimens were washed with tap
have been developed to achieve satisfactory bond water for 1 minute, ultrasonically cleaned in a water
strength between adhesives and ceramics. Organosi- bath for 10 minutes, and air dried.
lane coupling agents have been reported to increase N Group 4: Similar to group 3, the specimens were
the bond strength of composite resin to porcelain.6 It etched with 9.5% HF (Ultradent). After etching, the
has even been reported that airborne-particle abrasion specimens were washed under tap water for
(APA) and acid etching with hydrofluoric acid (HF) 1 minute, ultrasonically cleaned in a water bath for
increase the surface area of ceramic surfaces and 10 minutes, and air dried. After air drying, a mixture
create an irregular topography that enhances the of acidic primer and silane agent (Kuraray Co) was
potential for micromechanical retention of the adhesive applied to the ceramic surface for 20 seconds.
cement. However, HF has been found to be harmful N Group 5: Specimens were treated with APA with 50-
and can irritate the soft tissues; therefore, clinicians mm aluminum oxide for 5 seconds at 4 bar pressure.
should be cautious when using it in the oral cavity. Because the distance of the abrasion tip from the
Few studies have reported on the bond strength of ceramic surface can affect surface topography, it
orthodontic composite resin to feldspathic porcelain. was maintained at a fixed distance of 4 mm for all
Moreover, the micro-shear bond strength (MSBS) of specimens. Following ABA, the specimens were
orthodontic composite resins to most porcelain surfac- washed with tap water for 1 minute, ultrasonically
es has been reported to be inadequate. Therefore, the cleaned in a water bath for 10 minutes, and air dried.
purpose of this in vitro study was to compare the N Group 6: Specimens were treated with APA with 50-
MSBS of orthodontic composite resin bonded to a mm aluminum oxide for 5 seconds at 4-bar pressure,
feldspathic porcelain disk prepared by six different with the distance of the tip from the ceramic surface
surface treatments and to ascertain whether surface kept at 4 mm. The specimens were then washed with
treatment has any significant effect on the overall tap water for 1 minute, ultrasonically cleaned in a
shear bond strength. Additionally, the conditioned water bath for 10 minutes, and air dried. After air
surfaces were evaluated by scanning electron micros- drying, a mixture of acidic primer and silane agent
copy (SEM). The null hypothesis was that different (Kuraray Co) was applied to the surfaces for
surface treatment methods have no significant influ- 20 seconds.
ence on the MSBS of the orthodontic composite resin.
To prepare the orthodontic adhesive (Transbond
MATERIALS AND METHODS Light Cure Adhesives, 3M Unitek, Monrovia, Calif) for
cementation, adhesive was directly used to fill an iris
Bonding Procedure that was cut from microbore Tygon tubing (TYG-030,
Small Parts Inc, Miami Lakes, Fla) with internal
Thirty-seven circular specimens (6 mm in diameter
diameter and height of approximately 1 mm and
and 4 mm in thickness) of feldspathic porcelain were
0.5 mm, respectively (Figures 1 and 2). The Tygon
divided into six different groups according to the
tubing containing the composite resin was put on the
surface treatment. Group 1 included seven circular
ceramic surface and light-cured for 40 seconds. In this
specimens, whereas the other groups were made up of
manner, each ceramic surface was bonded at four
six circular specimens (Table 1).
different locations with the resin cylinders. The
N Group 1: Specimens were not given any additional assembly of porcelain plus composite resin was stored
surface treatment (control). at room temperature (23uC 6 2uC) for 1 hour prior to
N Group 2: Specimens were treated with a mixture of removal of the Tygon tubing. Subsequently, the
acidic primer and silane coupling agent (Kuraray Co, specimens were immersed in water at 37uC for
Osaka, Japan) for 20 seconds. 24 hours before testing for MSBS. The bonding
N Group 3: Specimens were treated with 9.5% HF procedures were carried out by the same operator in
(Ultradent, South Jordan, Utah) for 60 seconds accordance with the manufacturers’ instructions.

MICRO-SHEAR BOND STRENGTH OF COMPOSITE RESIN 767
Aktiengeselischaft Fuürstentun, Liechtenstein) for

180 seconds at 40 mA and they were examined using
SEM (LEO 435 VP, Cambridge, England) at 20 Kv by
the same operator (Figures 3 and 4).
Statistical Methods
One-way analysis of variance was used to compare
the groups for differences in failure strength. Because
the failure strength measurements were not normally
distributed, a transformation of the data (natural
logarithm) was used for the analyses to satisfy the
Figure 2. Sketch depicting the bond strength testing procedure. analysis of variance assumptions.
RESULTS
Micro-Shear Bond Testing
The results are shown in Tables 2 and 3 and
Before the test, all porcelain/resin cylinder interfaces Figure 5. The MSBS values ranged from a minimum
were checked under an optical microscope at 203 of 0.7 MPa for the control group to a maximum of
(Olympus, Tokyo, Japan) for bonding defects. Any 35.5 MPa for the HF/silane-treated group. The mean
cylinders that showed apparent interfacial gaps, MSBS was 20.8 MPa for the HF/silane-treated group.
bubbles, or any other defects were excluded and The control had significantly lower mean MSBS (P ,
replaced by other cylinders. The assembly of the round .01) than all other groups. Silane had a significantly
porcelain disk and the composite resin was attached to different mean MSBS than HF (P 5 .0187), APA (P ,
the testing device using cyanoacrylate adhesive .0001), APA/silane-treated (P , .0001), and HF/silane-
(Superbond, Loctite, Hunt Valley, Md), which in turn treated (P , .0001). The HF-treated specimens had
was placed in a universal testing machine (MTS Sine significantly lower mean MSBS than APA/silane-
Tech Re New 1123, MTS Systems Corp, Eden Prairie, treated (P , .0191) and HF/silane-treated (P ,
Minn) for shear bond testing (Figure 2). An edge of .0001) disks and a marginally lower mean MSBS than
stainless steel, 0.5 mm in thickness, was fixed on the APA specimens. APA and APA/silane-treated disks
superior part of the universal testing machine and was had significantly lower failure stress than did HF/
gently adapted against the ceramic/resin cement silane-treated disks (P 5 .0006 and P 5 .006,
interface. A shear force was applied to each specimen respectively).
at a crosshead speed of 0.5 mm/min until failure APA and APA/silane-treated disks (P 5 .469) were
occurred. not significantly different from each other. It was also
Two specimens from each group were gold coated noted the bond failures for all the specimens, except
with a sputter coater (Balzers-SCD 050, Balzers Union for the HF/silane group, were of the adhesive type,
whereas the HF/silane specimens revealed a cohesive
type of bond failure (Figures 6 and 7).
Figure 3. Comparison of HF-treated ceramic with a glazed (control) Figure 4. Comparison of APA ceramic with a glazed ceramic
ceramic under SEM. under SEM.

Table 2. Micro-Shear Bond Strength of Different Groups Table 3. Differences in Micro-Shear Bond Strength Between
Groups
Groupa N Mean SD SE Min Max
Comparisona P
Control 28 3.7 1.9 0.4 0.7 7.5
HF 20 9.9 2.7 0.6 6.7 18.3 Control vs HF , .0001
HF/silane 20 20.8 6.4 1.4 11.5 35.5 Control vs HF/silane , .0001
APA 21 13.5 6.1 1.3 4.2 27.7 Control vs APA , .0001
APA/silane 20 14.3 5.3 1.2 6.8 24.3 Control vs APA/silane , .0001
Silane 20 7.7 3.7 0.8 2.7 13.5 Control vs silane , .0001
a HF vs HF/silane , .0001
APA indicates airborne-particle abrasion; SD, standard deviation;
HF vs APA .0960
and SE, standard error.
HF vs APA/silane .0191
HF vs silane .0187
DISCUSSION HF/silane vs APA .0006
HF/silane vs APA/silane .0067
The null hypothesis of this study—that surface HF/silane vs silane , .0001
treatment of feldspathic porcelain has no influence APA vs APA/silane .4696
on the MSBS of the orthodontic composite resin—was APA vs silane , .0001
rejected. The results showed that the MSBS varied as APA/silane vs silane , .0001
a
a function of surface treatment. Adequate bond HF indicates hydrofluoric acid; APA, airborne-particle abrasion.
strength between a metal bracket and enamel is in
the range of 6 to 8 MPa.7 Except for the control group, significantly greater MSBS than treatment with HF and
all the mean values of MSBS between composite resin APA alone, respectively.
and feldspathic ceramic were either within or above the In all the groups except HF/silane, adhesive failures
optimal range. between the porcelain and composite resin were seen.
The previous literature has suggested the effective- Adhesive failures at the porcelain/composite interface
ness of silane coupling agents in establishing a bond are preferred to avoid porcelain fractures during
between resin and a ceramic surface.6 Acid catalysis debonding. It has been reported that if the compos-
increases the bond strength of adhesive cement to ite/porcelain bond strength is above 13 MPa, cohesive
ceramics treated with silane because of the initiation failure is seen in the porcelain; however, in the present
and formation of a siloxane bond between the silane study, although the APA group (13.5 MPa) and the
coupling agent and the ceramic surface. The current APA/silane group (14.3 MPa) had a mean MSBS
research confirms this finding, where ceramic treated slightly higher than 13 MPa, adhesive failure was
with a mixture of acidic primer and silane showed a observed between the porcelain and the composite
significantly greater MSBS than control ceramic resin.8 The HF/silane group had a mean MSBS
samples that were left untreated; similarly, HF/silane (20.8 MPa) that was significantly higher than all the
and APA/silane treatment of ceramic resulted in a other groups and showed a cohesive failure in the
Figure 5. Graph depicting MSBS values of different groups.

MICRO-SHEAR BOND STRENGTH OF COMPOSITE RESIN 769
Figure 6. HF/silane-treated specimen under SEM.
porcelain. This finding indicates that clinicians should porcelain. This methodology involved small bonding
be extremely cautious when using the HF/silane surface areas and a uniform distribution of stress;
method to prepare porcelain, because debonding when compared to the micro-tensile bond test,
may result in a fracture or a crack in the porcelain trimming of the sample after the bonding procedure
surface. is not necessary. Aside from this, preparing specimens
In the present study tests of MSBS were performed for this test is easy and multiple samples can easily be
between an orthodontic adhesive and feldspathic made, even using brittle materials. Although Shimada
Figure 7. APA/silane-treated specimen under SEM.

et al.9 used a thin wire (diameter of 0.20 mm) looped N A silane coupling agent effectively increased the
around the resin cylinder to carry out MSBS testing, a mean MSBS between orthodontic adhesive and
pilot experiment indicated no difference in using a wire feldspathic porcelain.
or a stainless steel edge.
Because this is an in vitro study, extrapolation of the
REFERENCES
current findings to the clinical situation must be done
with caution. Furthermore, in clinical situations, the 1. Heydecke G, Thomason JM, Lund JP, Feine JS. The impact
bonds between the composite resin and porcelain are of conventional and implant supported prostheses on social
and sexual activities in edentulous adults. Results from a
influenced by additional oral environmental factors, eg, randomized trial 2 months after treatment. J Dent. 2005;33:
saliva, forces of mastication, type of stress being 649–657.
applied, etc. Nevertheless, these in vitro studies can 2. Raptis NV, Michalakis KX, Hirayama H. Optical behavior of
be used to compare mean MSBS values of different current ceramic systems. Int J Periodontics Restorative Dent.
composite resins and thus can suggest clinical usage. 2006;26:31–41.
3. Proffit WR, Fields HW Jr, Moray LJ. Prevalence of
Further research with other composite resins and malocclusion and orthodontic treatment need in the United
ceramics is indicated to effectively compare the mean States: estimates from the NHANES III survey. Int J Adult
MSBS values of different composite resins. Orthodon Orthognath Surg. 1998;13:97–106.
4. Salonen L, Mohlin B, Gotzlinger B, Hellden L. Need and
demand for orthodontic treatment in an adult Swedish
population. Eur J Orthod. 1992;14:359–368.
CONCLUSIONS 5. McLain JB, Proffitt WR. Oral health status in the United
States: prevalence of malocclusion. J Dent Educ. 1985;49:
N The mean MSBS varied as the function of surface 386–397.
treatments. The mean MSBS values were between 6. Lacy AM, LaLuz J, Watanabe LG, Dellinges M. Effect of
3.7 and 20.8 MPa and were adequate for all groups porcelain surface treatment on the bond to composite.
except for the control group. J Prosthet Dent. 1988;60:288–291.
7. Oliver BM, Dujovne AR. Shear strength of APC brackets
N The HF/silane-treated group had a significantly bonded to extracted teeth. A pilot study. Orthod Cyber J.
greater MSBS than all the other groups. The MSBS http://www.oc-j.com/issue2/liver2a.htm. Accessed on June
values were greatest for HF/silane-treated, followed 19, 2009.
by APA/silane-treated, APA, HF-treated, silane- 8. Piascik JR, Swift EJ, Thompson JY, Grego S, Stoner BR.
treated, and the control group. Surface modification for enhanced silanation of zirconia
ceramics. Dent Mater. 2009;25:1116–1121.
N The MSBS values of the APA specimens and the 9. Shimada Y, Yamaguchi S, Tagami J. Micro-shear bond
APA/silane-treated specimens were not significantly strength of dual-cured resin cement to glass ceramics. Dent
different from each other. Mater. 2002;18:380–388.

Original Article
Quantification of three-dimensional orthodontic force systems of

T-loop archwires
Jie Chena; Serkis C. Isikbayb; Edward J. Brizendinec
ABSTRACT
Objective: To demonstrate the three-dimensional (3D) orthodontic force systems of three
commercial closing T-loop archwires using a new method and to quantify the force systems of the
T-loop archwires.
Materials and Methods: An orthodontic force tester (OFT) and a custom-made dentoform were
developed to measure force systems. The system simulated the clinical environment for an
orthodontic patient requiring space closure, which included measurement of three force
components along, and three moment components about, three clinically defined axes on two
target teeth. The archwires were attached to the dentoform and were activated following a standard
clinical procedure. The resulting force system was measured using the OFT.
Results: The force systems of the T-loops on the teeth were 3D. Activation in one direction
resulted in force and moment components in other directions (side effects). The six force and
moment components as well as the moment-to-force ratios in the clinically defined coordinate
system were quantified.
Conclusions: The commercial archwires do not provide force systems for pure translation.
Quantification of the force system is critical for the selection and design of optimal orthodontic
appliances. (Angle Orthod. 2010;80:754–758.)
KEY WORDS: Orthodontic force system; Biomechanics
INTRODUCTION However, quantification of the force system delivered

by a continuous closing loop archwire has only been
The control of orthodontic movement relies on the
studied on a limited basis. Various experimental or
ability to quantify and manipulate the force system,
computational studies3–8 have been performed on the
specifically the moment-to-force ratio (M/F). Small
force systems. These studies4–6 have concluded that the
differences in M/F can produce different clinical
force system is multifactorial and can be adjusted, and a
results.1,2 The inability to control the orthodontic force
3D analysis is needed to study the force system of an
system can result in undesirable tooth movement,
archwire.8 However, they have not contributed much to
which reduces overall treatment efficiency.
the understanding of the continuous closing archwires,
because either the data were obtained with segmented
a
Professor, Departments of Engineering/Orthodontics and
Oral Facial Genetics, School of Dentistry, Indiana University appliances in a two-dimensional space or the archwires
Purdue University Indianapolis, Indianapolis, Ind. were tested under nonclinical conditions.4–6,9–11
b
Associate Professor, Department of Orthodontics and Oral A segmental wire provides force systems on two
Facial Genetics, School of Dentistry, Indiana University Purdue teeth that are equal in magnitude but opposite in
University Indianapolis, Indianapolis, Ind.
c
Staff Scientist, Department of Biostatistics, Indiana Univer-
direction. However, an archwire is commonly attached
sity School of Medicine, Indiana University Purdue University to more than two teeth. The force systems on the two
Indianapolis, Indianapolis, Ind. teeth adjacent to the space are affected by constraints
Corresponding author: Dr Jie Chen, Mechanical Engineering/ from their neighboring teeth; thus, the force systems
Orthodontics and Oral Facial Genetics, Indiana University
from these two kinds of appliances are different.
Purdue University Indianapolis, 723 West Michigan St, Indiana-
polis, IN 46202 Furthermore, most previous studies5–7 fixed the appli-
(e-mail: jchen3@iupui.edu). ance to the brackets rigidly, whereas ligature ties are
Accepted: October 2009. Submitted: August 2009.
currently used in clinical settings. The archwires are
G 2010 by The EH Angle Education and Research Foundation, secured firmly—not rigidly—with stainless-steel or
Inc. elastomeric ligature ties. Recent data10 clearly demon-

3D ORTHODONTIC FORCE SYSTEMS 755
cells were able to measure three force and three

moment components simultaneously.
The dentoform was fixed to a platform (Figure 1).
The target teeth (left canine and left lateral incisor)
were attached to the load cells and were then
separated from the dentoform. In this way, their
original positions and orientations with respect to the
remainder of the dentoform were maintained. The
design ensured that the boundary condition of the lab
model was the same as that in the clinic.
The load cells were aligned with the clinically used
coordinate system, the origin of which was at the
center of the bracket. The x-axis was directed buccally
Figure 1. The orthodontic force tester used to measure three- (normal to the crown and pointing toward the cheek),
dimensional (3D) load systems on the canine and incisor by a T-loop
the y-axis distally (tangent to the crown and pointing
archwire. (A) Orthodontic force tester (OFT). (B) The T-loop archwire
ligated to the brackets. The canine and lateral incisor were, toward the molar), and the z-axis apically (perpendic-
respectively, separated from the rest of the typodont and attached ular to the x–y plane and pointing toward the root). The
to the load cells. The experimental setup allows simulation of the definition was good for only the left quadrant of the
space closure cases. maxilla teeth. The dentoform base was parallel to the
bracket slots, which allowed the occlusogingival (OG)
strate that the force system of an appliance depends axes of the target teeth to be parallel to the z axis. The
heavily on the ligation method. Therefore, the clinical transducers were further positioned so that buccolin-
ligation method needs to be simulated in order to gual (BL) and mesiodistal (MD) axes of each target
accurately measure the force system. tooth corresponded to the x and y axes, respectively.
Currently, archwire selection has not been based on The directions (OG, BL, and MD) are clinical terms to
the force system that the springs can deliver but rather describe tooth displacements.
on personal trial-and-error experience and preference. The T-loop closing archwires (0.016 3 0.022-inch
Quantification of the 3D force system with a continu- Stainless Steel Natural Form Arch, Oscar Inc, Fishers,
ous closing loop archwire will allow the clinician to Ind) in interloop distances of 38 mm, 42 mm, and
make an informed decision when selecting the arch- 46 mm were tested. The 42-mm T-loop closing
wires. The objectives of this study were to describe the archwire (TL42) positioned the T-loop in the middle of
technology and to evaluate the force systems of the the interbracket distance of the lateral incisor and
commercial T-loop archwires. canine brackets. The 38-mm (TL38) and the 46-mm
(TL46) archwires positioned the T-loop 2 mm anterior
MATERIALS AND METHODS and 2 mm posterior to the middle of the interbracket
distance, respectively.
An orthodontic force tester was developed.12 To- Three measurements were made for each archwire.
gether with a custom-made dentoform made of The OFT was zeroed before placing the closing loops.
autopolymerizing acrylic (Duralay, Reliance Dental An archwire was placed on the dentoform, with the
Mfg Co, Worth, Ill), treatment of a space closure after vertical legs separated by a 0.05-mm metal shim to
first premolar extractions and canine retractions was maintain a consistent baseline. The closing loop
simulated (Figure 1). For this purpose, the system archwire was secured to each bracket with 0.010-inch
allowed attachment of a clinically used orthodontic stainless-steel ligature wires (Figure 1). Crimpable
appliance to the full dentoform, activation of the stops were secured distal to the second molar tubes
appliance following established procedures, and mea- bilaterally. The resulting orthodontic force system (Fx,
surement of the force systems on target teeth to be Fy, Fz, Mx, My, and Mz) on the tooth was measured as
moved or to serve as an anchor. the baseline. The closing loops were activated using 1-
The orthodontic force tester consisted of two load mm and 2-mm shims between the crimpable stop and
cells (Multi-axis force/torque Nano17, ATI Industrial the molar tubes bilaterally. Measurements were made
Automation, Apex, NC). The force range of each load after all activations.
cell was 0–20 N, with a 60.025 N resolution, and the To assess variation due to operating errors, such as
moment range was 0–100 N-mm, with a 60.003 variation of ligature tie tightness, activation level, and
N-mm resolution. These ranges are adequate because wire shape, etc, experiments on each type of archwire
commonly used clinical force and moment fall into the were repeated on five wires with the same specifica-
load cell’s range with proper resolutions.13,14 The load tions. The average and standard deviation were

756 CHEN, ISIKBAY, BRIZENDINE
Figure 2. Three force components of the three T-loops on the canine Figure 3. Three moment components of the three T-loops on the
and lateral incisor corresponding to 1 mm and 2 mm of activations. canine and lateral incisor corresponding to 1 mm and 2 mm
of activations.
computed to statistically determine the load compo-
nents and their variations. A two-way full-factorial
analysis of variance model was used to assess the force increases linearly with the activation (Figure 2).
effect of activation amount (1 mm or 2 mm) and The force was higher on the lateral incisor. Fy was low
location of the T-loop on the resulting forces and (less than 0.6 N), primarily on the incisor side. The force
moments. A separate analysis was performed for each component on the canine side was negative (directed
tooth sensor. A significance level of .05 was used to mesially) at 1-mm and 2-mm activations. The magnitude
test all hypotheses. increased linearly with the activation for the TL38 and
TL42 cases (Figure 2). In the z direction, the sign of Fz
RESULTS was controlled by the loop location. For all the wires, the
force components on the two target teeth had different
The six load components were measured for both
signs. Mx was positive on the canine and negative on the
the left maxillary lateral incisor and the canine. T-loop
incisor (Figure 3). However, the sign of My depends on
archwires corresponding to three loop locations (TL38,
the loop location. For example, My on the canine was
TL42, and TL46) were evaluated. The force and
positive for the TL38 and TL42 wires and negative for
moment components corresponding to 1-mm and 2-
TL48. Finally, Mz was negative on the canine and positive
mm activations and their standard deviations are
on the incisor (Figure 3).
shown in Figures 2 and 3, respectively. The positive
The location of the T-loop had a significant effect on all
Fx was directed buccally. That component was
of the forces and moments except for the Fy force on the
responsible for a buccal (if it was positive) or lingual
canine tooth (Figures 2 and 3). The amount of activation
(if it was negative) crown tipping (first-order tipping).
also had a significant effect on all of the outcomes
Similarly, the positive Fy was directed distally. That
except, again, for the Fy force on the canine tooth. The
component caused a mesial (if it was negative) or
two-way analysis of variance additionally assessed if
distal (if it was positive) crown tipping (second-order
there was a multiplicative effect between the T-loop
tipping). The positive Fz was directed apically (toward
location and the amount of activation. In the case of the
the root). This component resulted in an intrusion (if it
Mx moment on the incisor and the My moment on both
was positive) or extrusion (if it was negative) (Fig-
the canine and incisor teeth, there was a significant
ure 2). The moment components were responsible for
interaction between location and activation.
tooth rotations about the corresponding coordinate
axes. The moment about the x-axis, Mx, created a
DISCUSSION
second-order rotation. The positive Mx rotated the
crown distally. The moments about the y-axis, My, As a result of ligation there were initial values for all
caused a first-order rotation. The positive My rotated the load components corresponding to the zero
the crown lingually. Similarly, the moments about the activation. These values were small compared to the
z-axis, Mz, created a third-order rotation. The positive corresponding components at 1-mm and 2-mm acti-
Mz caused distal-crown-in rotation (Figure 3). vations. Since zero activation is not used clinically, the
The activation of the spring in the distal direction associated force values are not reported here.
resulted in force and moment components in all three There were variations in the measured force and
directions. For all the loops, Fx was negative (against the moment components (Figures 2 and 3). The variations
crown) for all three archwires (TL38, TL42, and TL46). The were primarily due to the following inevitable factors:

3D ORTHODONTIC FORCE SYSTEMS 757
Table 1. Moment-to-Force (M/F) Ratios of the Three Types (TL38, TL42, TL46) of T-Loop Archwires. Values Presented Are the Average M/F
Ratios. N 5 5 in All Cases
Activation, TL38 TL42 TL46
Tooth mm Mx/Fy My/Fx Mz/Fy Mx/Fy My/Fx Mz/Fy Mx/Fy My/Fx Mz/Fy
Canine 1 28.4 20.3 12.7 213.4 20.4 16.7 29.6 1.6 12.7
2 26.1 20.3 6.4 210.9 0.0 8.8 37.3 1.9 225.1
Incisor 1 297.6 4.3 123.0 3.7 0.3 231.3 48.0 0.7 2126.3
2 190.2 4.5 2159.6 8.8 0.7 223.6 5.0 0.2 222.1
Fx, Fy, Fz are force components in the x, y, z directions. Mx, My, Mz are moment components about the x, y, z axes.
forces for tightening the stainless-steel ligature wire, and intruded the canine, while the TL42 wire showed
archwire shape, and activations. These factors were the opposite, meaning selection of loop location could
accounted for in the statistical evaluations. The varia- adjust the force component (Figure 2). The dominant
tions were, in general, small with respect to the means. moment component was Mx. Positive Mx on the canine
Large variations on Fy and Mx occurred for the TL46 rotates the crown distally, which counters the tipping
wires, especially corresponding to the 2-mm activation. created by a mesially directed Fy. Negative Mx on the
This was because the loop was too close to the canine incisor balances the moment created by distally
bracket at the level of activation, such that the loop directed Fy. The TL38 wire created the largest Mx on
interfered with the bracket and resulted in an unstable the lateral incisor, while the TL46 wire generated the
force system. This loop location and activation must be largest Mx on the canine, meaning the component is
discouraged clinically, because an unstable force system sensitive to the loop location. The Mx/Fy controlled the
may result in an unpredictable tooth displacement. translation or tipping in the MD direction. The low Fy on
Clinical control of tooth displacement is determined by the incisor contributed to a large Mx/Fy (see Table 1).
the design or selection of appropriate appliances, which In the BL direction, a negative My on the incisor
requires knowledge of M/F ratios in all three directions.15 counters the moment created by the negative Fx,
Furthermore, the magnitude of the force needs to be meaning that only a positive My/Fx may reduce the
restricted within a certain range for the best clinical tipping caused by Fx, and a negative ratio will enhance
effects.13 For the first time, all six components of the the lingual crown tipping. Although most My/Fx were
force systems of the three archwires were quantified positive, their values were low, which means they were
experimentally. The force increased linearly with the not sufficient to translate the teeth (Table 1). Finally,
activation. Each type of archwire had a distinct force Mz from the three archwires was high, which created a
system. Activation in the distal direction results in force high Mz/Fy as a result of low Fy. This would produce a
and moment components in both BL and OG directions, distal-crown-out rotation for the canine and a mesial-
which may cause undesired tooth displacement com- crown-out rotation for the lateral incisor (Figure 3).
ponents. These unwanted components are side effects This knowledge helps one to select appliances for a
that need to be eliminated. clinical need, which will not be possible without the
Understanding of the 3D force system influences the quantification of the 3D force system.
clinical decision of choosing the appliances. For all of The measured 3D force system confirmed some of
the loops, the dominant force component was Fx. the conclusions from previous two-dimensional studies
Compressive Fx against the crown was observed for all and revealed new information on coupling effects. When
three archwires (TL38, TL42, and TL46). The force was a T-loop is placed eccentrically in the space to be closed,
higher on the lateral incisor. It reached 5 N with 2-mm the moment, Mx, is higher on the tooth closer to the loop
activation. Consequently, a large moment My would be (Figure 3), confirming the conclusion made by Burstone
needed if translation was required. Distal force, Fy, and Koenig.16 In this study, coupling effects were
was low primarily on the incisor side, which would not quantified when the loop was activated distal to the
be sufficient for its distal movement. The mesial force molars. A large force, Fx, in BL direction was generated
was on the canine side. The force component reached against the crown, causing the teeth to tip crown-in. If
1.6 N at 2-mm activation for TL38, meaning the the desired tooth displacement was translation, there
posterior segment would be protracted. In the OG was not enough correction moment, My (low My/Fx),
direction, the forces on the two target teeth had causing tipping (Table 1). The intrusion/extrusion force,
different signs when an archwire was activated, Fz, would also occur depending on the loop locations.
meaning if one tooth was extruded, then the other Placing the loop mesially (TL38) would extrude the
one was intruded. The TL38 wire extruded the incisor incisor and intrude the canine, while placing the loop in

758 CHEN, ISIKBAY, BRIZENDINE
the middle (TL42) or distally (TL46) showed an opposite ACKNOWLEDGMENT

effect (Figure 2). Thus, adjusting the loop location could The study was supported by grant NIH-NIDCR R41-
be used as a method to control intrusion and extrusion. DE017025.
The force in the MD direction, Fy, was generally small on
the incisor side. Large Fy would be needed if the incisor REFERENCES
was prescribed to be translated distally. An understand-
1. Smith RJ, Burstone CJ. Mechanics of tooth movement.
ing of coupling effects is important for estimating Am J Orthod. 1984;85:294–307.
potentially unwanted displacement, which cannot be 2. Gjessing P. Controlled retraction of maxillary incisors.
quantified in two-dimensional models. The results Am J Orthod Dentofacial Orthop. 1992;101:120–131.
demonstrated that the 3D force systems would need to 3. Braun S, Garcia JL. The gable bend revisited. Am J Orthod
be quantified so that they could be manipulated Dentofacial Orthop. 2002;122:523–527.
4. Chen J, Bulucea I, Katona TR, Ofner S. Complete
quantitatively in order to control tooth displacement orthodontic load systems on teeth in a continuous full
and to reduce undesirable side effects. archwire: the role of triangular loop position. Am J Orthod
It is beneficial for clinicians to use an appliance with Dentofacial Orthop. 2007;132:e141–e148.
predictable clinical outcomes. This kind of appliance will 5. Chen J, Markham DL, Katona TR. Effects of T-loop
geometry on its forces and moments. Angle Orthod. 2000;
result in fewer side effects and fewer office visits, which
70:48–51.
greatly benefits both the orthodontist and the patient. For 6. Katona TR, Le YP, Chen J. The effects of first- and second-
example, the TL42 at a 2-mm activation produced a ratio order gable bends on forces and moments generated by
of 210.9 mm for Mx/Fy and a ratio of 0 for My/Fx (see triangular loops. Am J Orthod Dentofacial Orthop. 2006;129:
Table 1). If a desired M/F ratio for translating the tooth is 54–59.
7. Kuhlberg AJ, Burstone CJ. T-loop position and anchorage
8 for Mx/Fy and My/Fx, then the force system would
control. Am J Orthod Dentofacial Orthop. 1997;112:12–18.
overcorrect the tipping in the MD direction and tip the 8. Isaacson RJ, Lindauer SJ, Conley P. Responses of 3-
canine crown-in in the BL direction. If the desired tooth dimensional arch wires to vertical v-bends: comparisons
movement is translation, which is a common require- with existing 2-dimensional data in the lateral view. Semin
ment for space closure, the archwire does not meet that Orthod. 1995;1:57–63.
9. Lisniewska-Machorowska B, Cannon J, Williams S, Bant-
requirement. However, the M/F can be adjusted to the
leon HP. Evaluation of force systems from a ‘‘free-end’’ force
target values by introducing gable angles,6 different loop system. Am J Orthod Dentofacial Orthop. 2008;133(6):791.
locations, and activations.4 Future studies should be e1–10.
conducted to investigate the effects of the gable bends 10. Gregg J, Chen J. The Effect of Wire Fixation Methods on the
on the force system. Measured Force Systems of a T-Loop Orthodontics Spring.
AAO Meeting. Philadelphia, Penn: American Association of
Orthodontists. 1997.
CONCLUSIONS 11. Raboud D, Faulkner G, Lipsett B, Haberstock D. Three-
dimensional force systems from vertically activated ortho-
N The proposed method can quantify 3D force systems
dontic loops [erratum appears in Am J Orthod Dentofacial
of an archwire under a simulated clinical condition. Orthop 2001;120(1):80]. Am J Orthod Dentofacial Orthop.
N All three commonly used T-loop archwires produce 2001;119:21–29.
3D force systems. The force systems are tooth 12. Chen J. Apparatus and method for measuring orthodontic
specific and dependent on loop location. force applied by an orthodontic appliance (US Patient No.
6,120,287). USA, 2000.
N Significant coupling effects occur when a distal 13. Proffit W. Contemporary Orthodontics. St Louis, Mo: Mosby,
activation is applied; these effects need to be Inc; 2000.
quantified in order to control the unwanted side effects. 14. Ren Y, Maltha JC, Kuijpers-Jagtman AM. Optimum force
N The commercial archwires do not provide force magnitude for orthodontic tooth movement: a systematic
systems for pure translation. literature review. Angle Orthod. 2003;73:86–92.
15. Nanda RS. Biomechanics in Clinical Orthodontics. Philadel-
N Quantification of the force system is critical for the phia, Penn: WB Saunders Company; 1996.
selection and design of optimal orthodontic applianc- 16. Burstone CJ, Koenig HA. Optimizing anterior and canine
es. retraction. Am J Orthod. 1976;70:1–19.

Original Article
Pattern and amount of change after orthodontic correction of

upper front teeth 7 years postretention
Anders Andréna; Sasan Naraghib; Bengt Olof Mohlinc; Heidrun Kjellbergd
ABSTRACT
Objective: To investigate the amount and pattern of changes of maxillary front teeth 7 years
postretention, which previously were retained with a bonded retainer.
Materials and Methods: The study group consisted of 27 patients. Study models before treatment
(T1), at debonding (T2), 1 year after removal of the upper bonded retainer (T3), and 7 years
postretention (T4) were present. The irregularity index (sum of contact point displacements) and
the rotations of front teeth toward the raphe line were calculated.
Results: The irregularity index of the maxillary front teeth changes very little or not at all during the
first year postretention. Further change long term resulted in an irregularity index of mean 2.0
(range 0.0–5.8). The contact relationship between the laterals and centrals seems to be the most
critical. Forty rotated teeth in 21 patients were corrected more than 20u. Mean relapse during the
first year postretention was 6.7u (range 0.0u–14.7u). Mean changes during 7 years was 8.2u (range
0.0u–19.3u).
Conclusions: Relapse of upper front teeth retained with a bonded retainer is minor in both the
short and long term. If permanent retention is required after 3 years of retention, it is enough to
retain the incisors. (Angle Orthod. 2010;80:620–625.)
KEY WORDS: Retention; Rotation; Relapse; Irregularity; Long-term
INTRODUCTION The present study is a follow-up of a previous article

where the relapse tendency of the maxillary front, 1
From an esthetic standpoint, alignment of the
year after removal of upper bonded retainers was
anterior teeth is of considerable importance as most
reported.10 Minor or no relapse was found. What we do
patients focus on the alignment of incisors and
not know is if the short-term stable cases change in the
canines.1,2 The problems of lower front stability after
long term, or if slightly displaced contact points or
treatment has been discussed in several studies3–6 and
partly relapsed rotations 1 year postretention deterio-
some information can also be found regarding the
rate with time. The long-term pattern of relapse
severity of maxillary irregularity long term postreten-
concerning rotations and contact point displacements
tion.6–9 In most of these studies a maxillary Hawley
(CPDs) for canines, laterals, and centrals is unknown.
retainer has been used for upper retention, but there
The objectives of this study therefore are to:
are no long-term postretention studies after use of
bonded upper retainers. N Study the amount of changes in alignment of the
upper front teeth long-term after retention with a
a
Consultant Orthodontist, The County Orthodontic Clinic in bonded retainer.
Mariestad, Västra Götaland, Sweden. N Study the relapse pattern of corrected rotations.
b
Consultant Orthodontist, The County Orthodontic Clinic in
N Investigate the pattern of change of contact point
Växjö, Kronoberg, Sweden.
c
Professor and Head, Orthodontic Department, Institution of displacements due to rotations and labiolingual
Odontology, the Sahlgrenska Academy at Göteborg University, movements.
Göteborg, Sweden. N Examine the effect of overcorrected contact points.
d
Associate Professor, Institution of Odontology, The Sahl-
grenska Academy at Göteborg University, Göteborg, Sweden.
Corresponding author: Dr Anders Andrén, Tandregleringsklin, MATERIALS AND METHODS
Bråtenvägen 4, S-542 42 Mariestad, Sweden
(e-mail: anders.andren@vgregion.se) The 45 patients described in a previous study10 were
Accepted: November 2009. Submitted: September 2009. invited to participate in the follow-up study. In spite of
G 2010 by The EH Angle Education and Research Foundation, several attempts to contact the patients, only 27
Inc. agreed to participate. The reason for this was mainly

PATTERN AND AMOUNT OF CHANGE 621
Figure 2. Contact point displacement (CPD) before treatment (T1),

after treatment (T2), 1 year postretention (T3), and 7 years
Figure 1. Teeth angles on the right side to the raphe line and
postretention (T4).
intercanine distance.
Measurement Error
The measurement error was calculated from double
that they lived far away from the clinic, and in some measurements of 27 models (T4), using Dahlberg’s
cases there was a lack of interest to participate in the formula.13 The error for CPD measurements was
follow-up. 0.2 mm.
The present group of 27 patients is, considering The measurement errors for rotation were 3.1u for
treatment and duration of retention, similar to the canines, 2.8u for laterals, and 2.4u for centrals.
former group of 45. Mean irregularity index at T1
(pretreatment), T2 (posttreatment), and T3 (1 year Statistical Analysis
postretention) was also of the same magnitude.
The mean age of patients was 25.3 years (range Pearson’s product-moment correlation coefficients
21.7–30.4 years); there were 10 male and 17 female were calculated to test for associations between
patients. irregularity index at T1/T3, T1/T4, and T3/T4. The
The patients in the present study, including 19 same analysis was also used to test for correlations
treated with extraction and eight with nonextraction, between correction of rotations/relapse of rotations
who had initial upper front irregularity had been treated and the change in mean CPD T2/T4 for the canine/
with fixed edgewise appliances at the County Ortho- lateral contact, the lateral/central contact, and the
dontic Clinic in Mariestad, Sweden. Their maxillary central/central contact.
arches were retained with bonded retainers only. The
mean duration of the retention period was 34.3 months RESULTS
(range 25–48 months). Contact Point Discrepancies
Study models were collected at mean 7.6 years
(range 6.7–10.9 years) out of retention (T4). Models CPDs (mean) for each of the five contact points in
from (T1), (T2), and (T3) were available. The method the maxillary front are presented in Figure 2. There
of retention with a bonded wire has been previously was no statistically significant difference between the
described.11 change in mean CPD for the contacts canines/laterals,
Labiolingual CPDs of the five contacts between laterals/centrals, or centrals/centrals. The irregularity
the mesial of the upper right canine and the mesial index of the upper front teeth is seen in Table 1. No
of the upper left canine were measured with a correlations were found between the pretreatment and
digital caliper with 0.1 mm accuracy. CPDs less postretention irregularity T1/T3 and T1/T4. There was
than 0.5 mm were noted as 0 mm. The sum of the
five CPDs on each model (irregularity index) was Table 1. Mean Irregularity Index T1, T2, T3, T4 (n 5 27)a
calculated according to Little.12 All models were Irregularity Index SD Range
scanned and measured as described in the 1-year T1 10.3 5.11 3.7–29.9
follow-up.10 Intercanine distance and rotations of the T2 0.9 0.74 0.0–2.1
upper front teeth in relation to the raphe line were T3 1.3 1.05 0.0–3.5
measured with a modified computer program (Scion T4 2.0 1.90 0.0–5.8
Image) (Figure 1). a
SD indicates standard deviation.

622 ANDRÉN, NARAGHI, MOHLIN, KJELLBERG
6.4 mm). Five of these showed relapse of 1 mm or

more at T4 (range 1.1–2.3). Among the patients,
without increased intercanine distance during treat-
ment, three showed a decreased intercanine distance
of 1.5–2.3 mm. In these cases, there was a tendency
for small changes during both the first year postreten-
tion and long term, but no obvious influence on the
irregularity.
DISCUSSION
This study shows that the irregularity index of the
maxillary front teeth changes very little or not at all
Figure 3. Twenty overcorrections at T2, T3, and T4. during the first year postretention. Further change long
term resulted in an irregularity index of less than 3 for
70% of the patients. A weakness in our material is the
a significant association between the irregularity index relatively small number of patients (n 5 27) with
at T3 and T4 (R 5 0.938, P , .0001). records 1 year postretention and long-term postreten-
Most of the posttreatment irregularity of 0.9 (T2) was tion. The 27 patients we could examine long term were
because of overcorrections. When overcorrections in all aspects similar to the larger group10 (initial
were excluded, the mean irregularity index was 0.4. irregularity, treatment, duration of retention). A
Twenty overcorrected contacts were noted. Of the 20 strength with this study is that the original 45 patients
overcorrected CPDs, 10 showed perfect contacts at T4 reported in the 1-year study10 were selected at the
and four were to some degree still overcorrected. Six appointment when the retainer was removed. Other
contacts had relapsed 0.8–2.1 mm (Figure 3). studies7–9,14–17 are based on retrospective materials
selected from larger collections. All our patients had
Rotations the same method of retention, ie, upper bonded
A total of 40 rotated teeth in 21 patients were retainer, and we can specify the length of the retention
corrected more than 20u during treatment (range period and the postretention period. These variables
20.3u–51.9u). Mean relapse during the first year have a wider range in many studies or are not reported
postretention (T2–T3) was 6.7u (range 0.0u–14.7u). at all.6–8,14–17
Seven years postretention (T2–T4) the mean relapse Difficulty in locating the raphe line on all four casts,
was 8.2u (range 0.0u–19.3u). Mean for the different variation of quality of plaster casts, and changes in
tooth groups are seen in Table 2. Most of the changes arch form causing relatively large measurement errors
were seen at 1 year postretention (T3). Positive have been discussed in the previous paper.10 Although
correlation between rotational correction in treatment using implants as fixed reference points is the most
and long-term relapse was significant for centrals (P 5 stable way to measure rotations, the raphe line is a
.0004), laterals (P 5 .0007), and canines (P 5 .0056). useful and relatively easy tool to find reference to
measure rotations of upper front teeth.
Circumferential supracrestal fiberotomy was per-
formed on seven incisors in four patients (initially Most of the patients who showed minor irregularity 1
year postretention were more irregular at long-term
corrected mean 34u). Mean rotational relapse (T2–T4)
follow-up resulting in that 14% of the contacts were
in this group was 9.1u (range 1.2u–17.6u).
displaced more than 1 mm, maximum 2.2 mm. There
was a strong correlation between irregularity 1 year
Intercanine Distance
postretention and long-term, but we could not confirm
During treatment the upper intercanine distance the finding of Surbeck et al.14 that pretreatment
increased 1.5 mm or more in nine patients (range 1.5– irregularity is a significant risk factor for postretention
Table 2. Teeth Derotated More Than 20.0u (n 5 40)

Mean Correction T2, Mean Relapse T3, Mean Relapse T4, Mean Relapse T2–T4,
Number Degrees Degrees Degrees % Pearson r
Cuspids 8 28.5 5.1 5.7 20 0.518**
Laterals 17 31.5 7.8 10.2 32 0.614***
Centrals 15 27.4 6.2 7.2 26 0.632***
** P , .01; *** P # .001.

Figure 4. Patient HE at T1 (a), T2 (b), T3 (c), T4 (d,e), and diagram of rotations (f) (angle to raphe line in degrees); 22 angle is almost the same
as T2–T4, but there is a CPD between 22 and 21.
relapse. However, half of the group of 27 patients did From an esthetic point of view, a slightly rotated
not change at all and they were stable during the whole upper canine is seldom disturbing due to the curved
postretention period. buccal surface, especially if the distal aspect of the
Concerning corrected rotations, almost all relapse lateral is located buccal to the mesial aspect of the
was seen 1 year postretention with very small further canine. A rotation of a lateral or central that causes a
changes long term. The laterals showed more rota- broken contact is more displeasing (Figure 7). Our
tional mean relapse than centrals and canines, and of clinical impression is that the contact between lateral
the 12 rotations that relapsed more than 10u, eight and central is the most critical concerning correction
were laterals. Some of the severely corrected rotations and stability. If after a 3-year period of retention a
were perfectly stable at T4 as can be seen from the decision is made to use permanent retention of the
range of relapse, 0.0u–19.3u. Our data confirm the maxillary front teeth, a retainer bonded to only the
findings of Surbeck et al.14 that the majority of rotational incisors seems to be a relevant choice.
relapse of the maxillary incisors is approximately 10u.
Half of the overcorrected contacts were nicely aligned at
T4. The overcorrections that started to relapse 1 year CONCLUSIONS
postretention, deteriorated in the long term.
The irregularity index is not always reflecting the N The irregularity index of the maxillary front was ,3 in
esthetic impression of the teeth; evenly distributed 70% of the patients long term with 14% of the contact
small CPDs are obviously better than one or two major relationships displaced .1 mm.
displaced contacts with the lateral/central contact often N Pretreatment irregularity was not correlated to long-
being the most critical (Figures 4 and 5). Our term outcome.
experience is that rotations of about 10u are not N There was a strong correlation between irregularity 1
visible. A relapse in the range of 15u to 20u can be and 7 years postretention.
detected at close examination. Of the 40 severe N Most of the rotational relapse was seen 1 year
rotations in this study, 15% relapsed within that range postretention with small changes long-term, 15% of
(15.6u–19.3u; Figure 6). corrections more than 20u relapsed within a range of

624 ANDRÉN, NARAGHI, MOHLIN, KJELLBERG
Figure 5. Patient ED at T1 (a), T2 (b), T3 (c), T4 (d,e), and diagram of rotations (f) (angle to raphe line in degrees); 21 has relapsed between T2
and T4 about 13u, but there is no CPD between 21 and 11. Irregularity index 5.1 (T4).
Figure 6. Patient EI at T1 (a), T2 (b), T3 (c), T4 (d,e), and diagram of rotations (f) (angle to raphe line in degrees). It shows some rotational
relapse on 12 and 22 but almost no effect on CPD of those teeth.

5. Boese LR. Fiberotomy and reproximation without lower

retention 9 years in retrospect: part II. Angle Orthod. 1980;
50:169–178.
6. Boley JC, Mark JA, Sachdeva RC, Buschang PH. Long-term
stability of Class I premolar extraction treatment. Am J Orthod
7. Sadowsky C, Schneider BJ, BeGole EA, Tahir E. Long-term
stability after orthodontic treatment: nonextraction with
prolonged retention. Am J Orthod Dentofacial Orthop.
1994;106:243–249.
8. Vaden JL, Harris EF, Gardner RL. Relapse revisited.
9. Edwards JG. A long-term prospective evaluation of the
circumferential supracrestal fiberotomy in alleviation ortho-
Figure 7. Patient K at T1 (a), T2 (b), T3 (c), and T4 (d) shows dontic relapse. Am J Orthod Dentofacial Orthop. 1988;93:
overcorrection of tooth 21 between T1 and T2. Relapse continues T3 380–387.
to T4. 10. Naraghi S, Andrén A, Kjellberg H, Mohlin BO. Relapse
tendency after orthodontic correction of upper front teeth
retained with a bonded retainer. Angle Orthod. 2006;76:
15u–20u. This amount of rotation is normally not 570–576.
visible for the patient. 11. Zachrisson BU. Clinical experience with direct-bonded
N Of the overcorrected contacts, 50% returned to orthodontic retainers. Am J Orthod Dentofacial Orthop.
perfect alignment. 1977;71:440–448.
12. Little RM. The irregularity index: a quantitative score of
mandibular anterior alignment. Am J Orthod Dentofacial
REFERENCES Orthop. 1975;68:554–563.
13. Dahlberg G. Statistical Methods for Medical and Biological
1. Espeland LV, Stenvik A. Perception of personal dental Students. London, UK: George Allen and Unwin Ltd; 1940.
appearance in young adults: relationship between occlusion, 14. Surbeck BT, Årtun J, Hawkins NR, Leroux B. Associations
awareness, and satisfaction. Am J Orthod Dentofacial between initial, posttreatment, and postretention alignment
Orthop. 1991;100:234–241. of maxillary anterior teeth. Am J Orthod Dentofacial Orthop.
2. Shaw WC. Factors influencing the desire for orthodontic 1998;113:186–195.
treatment. Eur J Orthod. 1981;3:151–162. 15. Jones ML. The Barry project-a further assessment of
3. Heiser W, Niederwanger A, Bancher B, Bittermann G, occlusal treatment change in a consecutive sample:
Neunteufel N, Kulmer S. Three-dimensional dental arch and crowding and arch dimensions. Br J Orthod. 1990;17:
palatal form changes after extraction and nonextraction 269–285.
treatment after extraction and nonextraction treatment. Part 16. Sadowsky C, Sakols EI. Long-term assessment of ortho-
1. Arch length and area. Am J Orthod Dentofacial Orthop. dontic relapse. Am J Orthod Dentofacial Orthop. 1982;82:
2004;126:71–81. 456–463.
4. Little RM, Riedel RA, Årtun J. An evaluation of changes in 17. Ormiston JP, Huang GJ, Little RM, Decker JD, Seuk GD.
mandibular anterior alignment from 10 to 20 years post- Retrospective analysis of long-term stable and unstable
retention. Am J Orthod Dentofacial Orthop. 1988;93: orthodontic treatment outcomes. Am J Orthod Dentofacial
423–428. Orthop. 2005;128:568–574.

Original Article
Quantity and quality assessment of randomized controlled trials on

orthodontic practice in PubMed
Tatsuo Shimadaa; Hisako Takayamab; Yoshiki Nakamurac
ABSTRACT
Objective: To find current high-quality evidence for orthodontic practice within a reasonable time,
we tested the performance of a PubMed search.
Materials and Methods: PubMed was searched using publication type randomized controlled trial
and medical subject heading term ‘‘orthodontics’’ for articles published between 2003 and 2007. The
PubMed search results were compared with those from a hand search of four orthodontic journals to
determine the sensitivity of PubMed search. We evaluated the precision of the PubMed search result
and assessed the quality of individual randomized controlled trials using the Jadad scale.
Results: Sensitivity and precision were 97.46% and 58.12%, respectively. In PubMed, of the 277
articles retrieved, 161 (58.12%) were randomized controlled trials on orthodontic practice, and 115
of the 161 articles (71.42%) were published in four orthodontic journals: American Journal of
Orthodontics and Dentofacial Orthopedics, The Angle Orthodontist, the European Journal of
Orthodontics, and the Journal of Orthodontics. Assessment by the Jadad scale revealed 60 high-
quality randomized controlled trials on orthodontic practice, of which 45 (75%) were published in
these four journals.
Conclusion: PubMed is a highly desirable search engine for evidence-based orthodontic practice.
To stay current and get high-quality evidence, it is reasonable to look through four orthodontic
journals: American Journal of Orthodontics and Dentofacial Orthopedics, The Angle Orthodontist,
the European Journal of Orthodontics, and the Journal of Orthodontics. (Angle Orthod.
2010;80:713–718.)
KEY WORDS: Hand search; Jadad scale; Orthodontics; PubMed; Randomized controlled trial
INTRODUCTION practice has recently risen. Peck6 stated ‘‘We orthodon-

tists have an obligation to be clinical scientists providing
Evidence-based medicine is defined as the conscien-
the best evidence-based service.’’ Ackerman7 also
tious, explicit, and judicious use of the best current
stated, ‘‘The challenge facing orthodontists in the 21st
evidence in making decisions about the care of individual
century is the need to integrate the accrued scientific
patients.1 Since the phrase ‘‘evidence-based medicine’’ evidence into clinical orthodontic practice.’’
appeared in the ACP Journal Club for the first time in Results of high-quality randomized controlled trials
1991,2 evidence-based medicine has been widely (RCTs) are considered to be strong evidence in
incorporated in clinical practice all over the world.3–5 In evidence-based orthodontic practice8–11 and are recog-
this context, the demand for evidence of orthodontic nized as the gold standard for providing clinical
research evidence.8 To retrieve medical and dental
a
Assistant Professor, Department of Orthodontics, School of
articles, the PubMed version of the Medline database is
Dentistry, Tsurumi University, Yokohama, Japan.
b
Research Fellow, Department of Orthodontics, School of widely used, because it is the largest single database
Dentistry, Tsurumi University, Yokohama, Japan. for biomedical references, indexing abstracts from
c
Professor and Department Chair, Department of Orthodon- about 4,000 journals worldwide free of charge.12–14
tics, School of Dentistry, Tsurumi University, Yokohama, Japan. Therefore, it is convenient to utilize PubMed to search
Corresponding author: Dr Tatsuo Shimada, Department of
for RCTs on orthodontic practice. Sjögren and Halling15
Orthodontics, School of Dentistry, Tsurumi University, 2-1-3
Tsurumi, Tsurumi-ku, Yokohama, 2300063 Japan stated that the medical subject headings (MeSH terms)
(e-mail: shimada-t@tsurumi-u.ac.jp) and publication type RCT were valid in PubMed
Accepted: December 2009. Submitted: September 2009. searches for RCTs in dental research. However, some
G 2010 by The EH Angle Education and Research Foundation, researchers pointed out problems with the Medline
Inc. search, for example, inherent software and operator

714 SHIMADA, TAKAYAMA, NAKAMURA
limitations,12 difficulty in indexing articles and misclas- Orthodontic article

sification of articles,13 and change in database content
N Including a related word; for example, orthodontics,
and indexing practices over time.16 At the same time, to
orthopedics, fixed appliance, bracket, or orthognathic
offer the best evidence-based service, it is important not
surgery.
only to find the articles but also to assess the quality of
N Judged as an orthodontic clinical article in terms of
the articles. However, most of the previous studies
article content.
searching for dental RCTs on Medline focused on the
quantity of articles.17–19 Nonorthodontic article
We consider that the most important goal for
performing evidence-based orthodontic practice is to N Unable to judge as an orthodontic clinical article in
find current high-quality evidence within a short time. terms of article content.
We investigated whether highly sensitive PubMed N Relating only to temporomandibular disorder, sleep
search strategies15,16,20 were appropriate for a search apnea syndrome, and bruxism and not to malocclu-
of RCTs on orthodontic practice. We also assessed all sion.
retrieved articles using the Jadad scale.21 N Not using an orthodontic appliance.
MATERIALS AND METHODS Quality Assessment of RCTs Using the

Jadad Scale
Medline Searches
The Jadad scale includes three items related to the
To identify RCTs on orthodontic practice with high validity of RCTs. Perfect scores for each item are 2, 2,
probability, we searched PubMed based on approach- and 1 for random allocation, double-blinding, and
es used in earlier studies.15,16,20 The Medline database withdrawals or dropouts, respectively. The total score
(Entrez PubMed, www.ncbi.nlm.nih.gov) was used to is 0 to 5 points. A score of less than or equal to 2 points
search for RCTs on orthodontic practice (April 2008) is evaluated as poor quality and a score of 3 to 5 points
using the MeSH term ‘‘orthodontics’’ and limiting is evaluated as high quality.22 When there was
publication type to ‘‘randomized controlled trial,’’ disagreement regarding the scores of the three items,
publication date to ‘‘from 2003/1/1 to 2007/12/31,’’ we referred to the results of earlier studies23,24 to make
language ‘‘English,’’ and subjects to ‘‘humans.’’ a final decision.
Comparison of the Results from PubMed Search Sensitivity and Precision of PubMed Search
with Those from a Hand Search of Four
Orthodontic Journals Sensitivity was evaluated in comparison with the
results of a hand search (gold standard) of four
Because 70% of the RCTs on orthodontic practice in orthodontic journals. Sensitivity was defined as the
the PubMed search were found in only four orthodontic number of RCTs on orthodontic practice in the four
journals, we defined the result of the hand search for orthodontic journals on PubMed divided by the number
RCTs on orthodontic practice in these four journals as of RCTs on orthodontic practice in the four orthodontic
the gold standard. This result was used for calculating journals by hand search. Precision was defined as the
sensitivity. number of true RCTs on orthodontic practice divided
by the number of records retrieved.16
Discrimination Between RCT or non-RCT
We chose articles that had the words, ‘‘randomly,’’ RESULTS
‘‘random.’’ or ‘‘randomization’’ in the title and ab-
PubMed Searches
stract.21 Initial screening of articles was performed
independently by two of the authors. When disagree- As shown in Table 1, 277 hits for RCTs on
ment in the classification of articles about RCTs orthodontic practice were obtained in the PubMed
occurred, the article was reread and discussed until search. There were 230 (83.03%) orthodontic articles,
a consensus was obtained. and of these, 201 (72.56%) were RCTs. Upon further
examination, 161 (58.12%) of these were true RCTs on
Discrimination of Orthodontic Articles orthodontic practice. Thus, the precision was 58.12%.
The retrieved articles were scrutinized for their
Number of RCTs on Orthodontic Practice
relevance to orthodontic practice. Articles covering
According to the Journal
other dental areas were excluded. We appraised all
the retrieved articles and discriminated the orthodontic As shown in Table 2, the 161 RCTs on orthodontic
articles as follows. practice were listed in 30 journals. Of these, 115

MOST RCTS ARE PUBLISHED IN FOUR JOURNALS 715
Table 1. Results of PubMed Search

Randomized Controlled Trials, (%) Nonrandomized Controlled Trials, (%) Total, (%)
Orthodontic articles 161 (58.12) 69 (24.91) 230 (83.03)
Nonorthodontic articles 40 (14.44) 7 (2.53) 47 (16.97)
Total 201 (72.56) 76 (27.44) 277 (100.00)
articles (71.43%) were listed in four orthodontic blinding but no description of the methods, and 147
journals. About half (n 5 57) of these articles articles (91.30%) contained no description of the
(49.57%) were found in the American Journal of blinding technique.
Orthodontics and Dentofacial Orthopedics (AJO-DO), The mean score of withdrawals was 0.60. The
23 (20.00%) in The Angle Orthodontist (AO), 22 adequately reported number of and reason for
(19.13%) in the European Journal of Orthodontics withdrawals were shown in 97 (60.25%) articles. The
(EJO), and 13 (11.30%) in the Journal of Orthodontics other 64 (39.75%) articles reported only the number of
(JO). The remaining 46 articles (28.57%) were withdrawals and contained no description of number or
published in 26 other journals. reason.
The numbers of publications of high- and poor-
Hand Search of the Top Four Orthodontic Journals quality RCTs according to journal are indicated in
Tables 4 and 5. The high-quality RCTs were not
We hand searched all of the articles in the four
limited to any specific journal. The total number of
journals, each of which had more than 10 articles.
high-quality RCTs was 60; 45 (75%) of these were
Besides the articles searched in PubMed, only three
published in the top four journals. There were 101
additional RCTs were found in the hand search; two
poor-quality RCTs; 70 (69.31%) of these were pub-
were from AJO-DO and one was from JO.
lished in the top four journals.
We set the results from the hand search of the top
four orthodontic journals as the gold standard. The
DISCUSSION
sensitivity for PubMed search was 97.46%.
For evidence-based orthodontic practice, the latest
Jadad Scale of RCTs of Orthodontic Practice high-quality evidence is essential. To find a simple and
As shown in Table 3, the mean Jadad scale score
was 2.17, which was between poor quality (0–2 points) Table 3. Number and Percentage (within parentheses) of
Randomized Controlled Trials (n 5 161) on Orthodontic Practice
and high quality (3–5 points). The mean score of Searched by PubMed that Reported Randomization, Double-
randomization was 1.41. Randomization with ade- blinding, Withdrawals or Dropouts and Jadad Scale Score
quate methods was described in 72 (44.72%) articles.
Variable No. (%)
Eighty-three (51.55%) articles did not mention any
Randomization
randomization methods. In six (3.73%) articles, the
Adequate method: 2 points 72 (44.72)
methods described were inadequate.
Method not reported: 1 point 83 (51.55)
The mean score of double-blinding was 0.16. Only Inadequate method: 0 points 6 (3.73)
12 (7.45%) articles showed adequate methods. Two Mean score 1.41
articles (1.24%) contained description of the double- Double-blinding
Adequate method: 2 points 12 (7.45)
Table 2. Number and Percentage of Randomized Controlled Trials Method not reported: 1 point 2 (1.24)
on Orthodontic Practice According to Journal Inadequate method: 0 points 147 (91.30)
Rank Journal No. (%) Mean score 0.16
1 American Journal of Orthodontics and Withdrawals/dropouts

Dentofacial Orthopedics 57 (35.40) Adequate method: 1 point 97 (60.25)
2 The Angle Orthodontist 23 (14.29) Inadequate method: 0 points 64 (39.75)
3 European Journal of Orthodontics 22 (13.66) Mean score 0.60
4 Journal of Orthodontics 13 (8.07) Jadad scale score
5 Cleft Palate-Craniofacial Journal 5 (3.11)
5 7 (4.35)
5 Journal of Clinical Orthodontics 5 (3.11)
4 3 (1.86)
7 Journal of Clinical Periodontology 3 (1.86)
3 50 (31.06)
7 European Journal of Oral Sciences 3 (1.86)
2 55 (34.16)
9 8 journals 2 each (1.24)
1 43 (26.71)
17 14 journals 1 each (0.62)
0 3 (1.86)
Total 161 (100.00) Mean score 2.17

Table 4. Number of High-Quality Randomized Controlled Trials on Table 5. Number of Poor-Quality Randomized Controlled Trials on
Orthodontic Practice According to Journal Orthodontic Practice According to Journal
5 4 3 2 1 0
Journal points points points Total Journal points point points Total
American Journal of American Journal of
Orthodontics and Orthodontics and
Dentofacial Orthopedics 2 1 18 21 Dentofacial Orthopedics 18 16 2 36
The Angle Orthodontist 1 0 5 6 The Angle Orthodontist 7 10 0 17
European Journal of European Journal of
Orthodontics 1 0 9 10 Orthodontics 9 3 0 12
Journal of Orthodontics 1 0 7 8 Journal of Orthodontics 2 3 0 5
Subtotal 5 1 30 45 (75%) 70
Journal of Clinical Subtotal 36 32 2 (69.31%)
Periodontology 1 0 1 2 Journal of Clinical Orthodontics 3 2 0 5
Caries Research 1 0 0 1 Australian Orthodontic Journal 2 0 0 2
Orthodontics and Craniofacial Cleft Palate-Craniofacial
Research 0 1 0 1 Journal 2 0 0 2
Acta Odontologica Progress in Orthodontics 2 0 0 2
Scandinavica 0 1 0 1 Texas Dental Journal 2 0 0 2
Cleft Palate-Craniofacial European Journal of Oral
Journal 0 0 5 5 Science 1 1 0 2
British Dental Journal 0 0 1 1 Orthodontics and Craniofacial Re-
European Journal of Oral search 0 2 0 2
Science 0 0 1 1 British Dental Journal 1 0 0 1
Oral Health & Preventive International Journal of
Dentistry 0 0 1 1 Dental Hygiene 1 0 0 1
The Journal of Pediatrics 0 0 1 1 Journal of Clinical
World Journal of Orthodontics 0 0 1 1 Periodontology 1 0 0 1
Total 7 3 50 60 (100%) Oral Health & Preventive
Dentistry 1 0 0 1
Rhinology 1 0 0 1
The Journal of Craniofacial Surgery 1 0 0 1
convenient method to retrieve RCTs for orthodontic World Journal of Orthodontics 1 0 0 1
British Journal of Oral and Maxillo-
practice, we tested the performance of PubMed search
facial Surgery 0 1 0 1
with appropriate strategies. We performed the PubMed European Journal of Dental Edu-
search with the MeSH term ‘‘orthodontics,’’ using the cation 0 1 0 1
limit criterion to confine the results to RCTs. According Journal of Orofacial
to a past report,17 we also used the terms ‘‘humans’’ Orthopedics 0 1 0 1
journal of the South African Dental
and ‘‘English’’ in the PubMed search to limit the results. Association 0 1 0 1
Sensitivity and precision were assessed as an index of Oral Surgery, Oral Medicine, Oral
validity. For evidence-based decision making, two Pathology, Oral
points are important. The first is to get the latest Radiology, and
Endodontology 0 1 0 1
high-quality evidence, and the second is to get such
The International Journal of Peri-
evidence in a matter of minutes. For the former, high odontics & Restorative Dentistry 0 1 0 1
sensitivity is needed. If the sensitivity is low, it will be Plastic and Reconstructive Surgery 0 0 1 1
difficult to identify evidence that was not retrieved. This Total 55 43 3 101 (100%)
is a great concern when making evidence-based
decisions in orthodontic practice. Because the sensi-
tivity was 97.46%, extremely few RCTs were excluded.
For the latter, high precision is needed. If the precision Past investigations of this strategy showed a high
is low, one more step will be needed to appraise the probability in medical and orthodontic research. In the
retrieved information manually, therefore prolonging medical field, Glanville et al.16 reported a high level of
the time needed to get appropriate information. sensitivity (82.78%) and precision (100%) for the
Because the precision was 58.12%, this search Medline search using the MeSH term and ‘‘randomized
technique might retrieve many unrelated articles. From controlled trial.’’ Our result showed that it was highly
the aforementioned validity, this PubMed search sensitive (97.46%), while retaining reasonable levels
strategy is valid for retrieving RCTs on orthodontic of precision (58.12%). Sjögren and Halling15 showed
practice if one is careful about definitely appraising the that PubMed search had high validity in endodontics
retrieved information. and orthodontics and low validity in pediatric dentistry

MOST RCTS ARE PUBLISHED IN FOUR JOURNALS 717
and oral surgery. The reason for difference between CONCLUSIONS

our results and a past investigation16 may be the
N The PubMed search strategy showed nearly 100%
variation of validity for MeSH terms in different
sensitivity and 60% precision. This would be useful
disciplines.
for evidence-based orthodontic practice.
In addition to sensitivity, evaluation of quality is also
N We showed that more than 70% of the RCTs
important. We evaluated the retrieved articles using
retrieved by PubMed were localized in four journals,
the Jadad scale. The results indicated that 75% of the
and 75% of RCTs were also localized even when
high-quality articles were published in four journals:
restricted to high-quality RCTs. Therefore, we
AJO-DO, AO, EJO, and JO. We identified these
defined AJO-DO, AO, EJO, and JO as the four
journals as the four major orthodontic journals for
major journals for evidence-based orthodontic prac-
evidence-based orthodontic practice.
tice. If one looks through these four journals on a
Our results showed a low percentage (7.45%) for
daily basis, one can gather plenty of high-quality
double-blind description compared with withdrawals or
orthodontic information for evidence-based decision
dropouts (60.25%). All articles were described as
making.
being randomized in our investigation, because we
N Only eight articles showed a perfect score for
investigated only RCTs. In the past, almost the same
double-blinding among the 115 RCTs on orthodontic
result was reported in a study that evaluated RCTs
practice published in the top four journals. Double-
and controlled clinical trials of orthodontic practice
blinding is impossible when evaluating the effect of
using the Jadad scale.25,26 The low percentage of
an orthodontic appliance in principle. Quality evalu-
double-blind description may indicate inherent char-
ation of an orthodontic clinical trial, especially
acteristics of the study for comparison of different
double-blinding, will be an issue in the future.
orthodontic treatments. The Jadad scale is the only
scale for evaluation of RCTs, and its validity has been
proved. Jadad and colleagues developed this scale for REFERENCES
the evaluation of pain treatment and indicated its
1. Sackett DL, Rosenberg WMC, Gray JAM, Haynes RB,
capability of universal application for any medical Richardson WS. Evidence based medicine: what it is and
area.21 what it isn’t. Br Med J. 1996;312:71–72.
Because removing systematic error (bias) is impor- 2. Guyatt GH. Evidence-based medicine. Ann Intern Med.
tant in clinical trials, a strong study design is desired. 1991;114(ACP J Club. suppl 2):A–16.
Therefore, randomization, double-blinding, and with- 3. Evidence-Based Medicine Working Group. Evidence-Based
medicine: a new approach to teaching the practice of
drawals or dropouts are emphasized in general. In this medicine. JAMA. 1992;268:2420–2425.
investigation only eight articles were double-blinded 4. Haynes RB, Devereaux PJ, Guyatt GH. Clinical expertise in
with two points of perfect scores in the top four the era of evidence-based medicine and patient choice.
journals.27–34 All studies evaluated the effect of drugs, ACP J Club. 2002;136:A-11–A-14.
and no study was concerned about the effect of 5. Eddy DM. Evidence-based medicine: a unified approach.
Health Aff. 2005;24:9–17.
orthodontic appliances. This fact may indicate that
6. Peck S. Philosophy and evidence-based orthodontics. Angle
RCT is a supreme study design for evaluating the Orthod. 1997;67:403.
effect of drugs that can be performed double-blinded 7. Ackerman M. Evidence-based orthodontics in the 21st
by active drug and placebo. However, it may be century. J Am Dent Assoc. 2004;135:162–167.
inappropriate to use RCTs for estimating the ortho- 8. Rinchuse DJ, Sweitzer EM, Rinchuse DJ, Rinchuse DL.
dontic treatment outcome because it is impossible to Understanding science and evidence-based decision mak-
ing in orthodontics. Am J Orthod Dentofacial Orthop. 2005;
perform double-blinding in such a clinical trial. Rinch- 127:618–624.
use et al.8 stated that results of research studies 9. O’Brien KD, Wright JL, Mandall NA. How to … do a
without placebo or sham procedures might be ques- randomized controlled trial. J Orthod. 2003;30:337–341.
tionable because of the influence of various psycho- 10. Phillips C, Tulloch JFC. The randomized clinical trial as a
logical factors. However, he did not suggest any powerful means for understanding treatment efficacy. Semin
Orthod. 1995;1:128–138.
alternative methods if blinding was impossible. Dou-
11. Newcombe RG. Research in orthodontics—A statistical
ble-blinded orthodontic clinical trials are yet to be perspective. Br J Orthod. 1995;1:299–302.
performed. 12. Bigby M. Evidence-based medicine in a nutshell. Arch
To conclude, RCTs on orthodontic practice are Dermatol. 1998;134:1609–1618.
easily and reliably accessible in a few minutes free of 13. Greenhalgh T. How to read a paper. The Medline database.
Br Med J. 1997;315:180–183.
charge. To gather orthodontics-related, high-quality,
14. Falagas ME, Pitsouni EI, Malietzis GA, Pappas G. Com-
evidence-based information, we should hand search parison of PubMed, Scopus, Web of Science, and Google
the four orthodontic journals AJO-DO, AO, EJO, and Scholar: strengths and weaknesses. FASEB J. 2008;22:
JO, on a daily basis. 338–342.

15. Sjögren P, Halling A. Medline search validity for randomized 25. Harrison JE. Clinical trials in orthodontics I: demographic
controlled trials in different areas of dental research. Br details of clinical trials published in three orthodontic
Dent J. 2002;192:97–99. journals between 1989 and 1998. J Orthod. 2003;30:25–30.
16. Glanville JM, Lefebvre C, Miles JNV, Camosso-Stefinovic J. 26. Harrison JE. Clinical trials in orthodontics II: assessment of
How to identify randomized controlled trials in MEDLINE: ten the quality of reporting of clinical trials published in three
years on. J Med Libr Assoc. 2006;94:131–136. orthodontic journals between 1989 and 1998. J Orthod.
17. Nishimura K, Rasool F, Ferguson MB, Sobel M, Niederman 2003;30:309–315.
R. Benchmarking the clinical prosthetic dental literature on 27. Willmot DR. White lesions after orthodontic treatment: does
MEDLINE. J Prosthet Dent. 2002;88:533–541. low fluoride make a difference? J Orthod. 2004;31:235–242.
18. Kim MY, Lin J, White R, Niederman R. Benchmarking the 28. Ogaard B, Alm AA, Larsson E, Adolfsson U. A prospective,
endodontic literature on MEDLINE. J Endod. 2001;27: randomized clinical study on the effects of an amine fluoride/
470–473. stannous fluoride toothpaste/mouth rinse on plaque, gingi-
19. Russo SP, Fiorellini JP, Weber HP, Niederman R. Bench- vitis and initial caries lesion development in orthodontic
marking the dental implant evidence on MEDLINE. Int J Oral patients. Eur J Orthod. 2006;28:8–12.
Maxillofac Implants. 2000;15:792–800. 29. Polat O, Karaman AI, Durmus E. Effects of preoperative
20. Royle PL, Waugh NR. Making literature searches easier: a ibuprofen and naproxen sodium on orthodontic pain. Angle
rapid and sensitive search filter for retrieving randomized Orthod. 2005;75:791–796.
controlled trials from PubMed. Diabetes Med. 2007;24: 30. Polat O, Karaman AI. Pain control during fixed orthodontic
308–311. appliance therapy. Angle Orthod. 2005;75:214–219.
21. Jadad AR, Moore A, Carrol D, Jenkinson C, Reynolds DJM, 31. Young AN, Taylor RW, Taylor SE, Linnebur SA, Buschang
Gavaghan DJ, McQuay HJ. Assessing the quality of reports PH. Evaluation of preemptive valdecoxib therapy on initial
of randomized clinical trials: is blinding necessary? Control archwire placement discomfort in adults. Angle Orthod.
Clin Trials. 1996;17:1–12. 2006;76:251–259.
22. Sjögren P, Halling A. Quality of reporting randomized clinical 32. Gorton J, Featherstone DB. In vivo inhibition of demineral-
trials in dental and medical research. Br Dent J. 2002;192: ization around orthodontic brackets. Am J Orthod Dentofa-
100–103. cial Orthop. 2003;123:10–14.
23. Schulz KF, Chalmers I, Hayes RJ, Altman DG. Empirical 33. Bird SE, Williams K, Kula K. Preoperative acetaminophen vs
evidence of bias. Dimension of methodological quality ibuprofen for control of pain after orthodontic separator
associated with estimates of treatment effects in controlled placement. Am J Orthod Dentofacial Orthop. 2007;132:
trials. JAMA. 1995;273:408–412. 504–510.
24. Dickinson K, Bunn F, Wentz R, Edwards P, Roberts I. Size 34. Bradley RL, Ellis PE, Thomas P, Bellis H, Ireland AJ, Sandy
and quality of randomized controlled trials in head injury: JR. A randomized clinical trial comparing the efficacy of
review of published studies. Br Med J. 2000;320: ibuprofen and paracetamol in the control of orthodontic pain.
1308–1311. Am J Orthod Dentofacial Orthop. 2007;132:511–517.

Original Article
The role of heme oxygenase-1 in mechanical stress- and

lipopolysaccharide-induced osteogenic differentiation in human
periodontal ligament cells
Jin-Hyoung Choa; Sun-Kyung Leeb; Jin-Woo Leec; Eun-Cheol Kimd
ABSTRACT
Objective: To investigate the mechanisms through which mechanical stress and lipopolysaccha-
ride treatment modulate osteoblastic differentiation in periodontal ligament cells.
Materials and Methods: Cells were treated with lipopolysaccharide and/or mechanical strain
applied with a Flexercell Strain Unit. Protein expression and mRNA were analyzed by Western
blotting and reverse transcription–polymerase chain reaction, respectively.
Results: When lipopolysaccharide was co-applied with mechanical strain, the increase in the
expression of bone morphogenetic protein-2, bone morphogenetic protein-7, and Runx2 mRNA
seen with mechanical strain alone was restricted, but heme oxygenase-1 expression was further
enhanced. Furthermore, pretreatment with an inhibitor of heme oxygenase-1 or inhibitors of p38,
mitogen-activated protein kinase, JNK, phosphoinositide 3-kinases, protein kinase G, and nuclear
factor kB restricted osteogenic differentiation induced by the application of lipopolysaccharide and
mechanical strain.
Conclusions: These results suggest that orthodontic force-induced osteogenesis in alveolar bone
is inhibited by the accompanying periodontal inflammation via the upregulation of heme
oxygenase-1 expression. Thus, the heme oxygenase-1 pathway could provide a possible
therapeutic strategy to improve bone formation in orthodontic treatment. (Angle Orthod.
2010;80:740–747.)
KEY WORDS: Mechanical stress; LPS; Osteogenic differentiation; HO-1; PDLCs
INTRODUCTION affinity for orthodontic brackets.1 This affinity can affect

the concentration of LPS in the gingival sulcus, thereby
Lipopolysaccharide (LPS), a main component of the
contributing to inflammation in periodontal tissues
cell surface of Gram-negative bacteria, is thought to be
adjacent to the brackets.2 In a dog model, orthodon-
a major cause of inflammation by bacterial infections.
tically treated teeth showed additional loss of connec-
A previous study demonstrated that Porphyromonas
tive tissue attachment in sites with infrabony pockets
gingivalis and Escherichia coli LPS exhibit a high
and plaque-induced inflammation.3 However, the
effects of the combination of orthodontic force and
a
Assistant Professor, Department of Orthodontics, College of
Dentistry, Wonkwang University, Iksan, South Korea. oral bacterial infection on bone remodeling are not
b
Graduate PhD student, Department of Oral and Maxillofacial completely understood.
Pathology, College of Dentistry, Wonkwang University, Iksan, The transmission of orthodontic force to the alveolar
South Korea. bone is mediated by the response of the periodontal
c
Professor, Department of Orthodontics, College of Dentistry,
Dankook University, Cheonan, South Korea.
ligament cells (PDLCs).4 The cells in the periodontal
d
Professor, Department of Oral & Maxillofacial Pathology, ligament are capable of differentiating into osteoblasts
College of Dentistry, Wonkwang University, Iksan, South Korea. or cementoblasts in response to a mechanical stimu-
Corresponding author: Eun-Cheol Kim, DDS, PhD, Professor, lus.5,6 The osteogenic differentiation of PDLCs is
Department of Oral and Maxillofacial Pathology, Dental College, known to play a pivotal role in alveolar bone
Wonkwang University, 344-2 Shinyoungdong, Iksan City, Jeon-
buk, 570-749, South Korea remodeling during orthodontic tooth movement.4
(e-mail: eckwkop@wonkwang.ac.kr) In the bone microenvironment, bone morphogenetic
Accepted: November 2009. Submitted: September 2009.
proteins (BMP) play a critical role in the regulation of
G 2010 by The EH Angle Education and Research Foundation, osteoblast differentiation and function. BMP-2, BMP-4,
Inc. BMP-6 and BMP-7 are growth factors in promoting the

LPS AND STRESS IN HUMAN PDL CELLS 741
differentiation of mesenchymal stem cells (MSCs) approved by the institutional review board and ethical
into osteoblasts or chondroblasts.7 Moreover, BMP committee at Wonkwang University.
activity is regulated by antagonists such as noggin.8 In
human osteoblastic cells, mechanical stress (MS) Application of MS
stimulates mineralization by increasing the production
of BMPs and decreasing the production of BMP Human PDLCs (3 3 105/well) were subcultured into
antagonists.9 six-well, 35-mm flexible-bottomed Uniflex culture
Mechanical signals may promote cell differentiation plates (Flexcell Corp, Hillsborough, NC) with a
into a particular phenotype.10 For example, MS has centrally located rectangular portion (15.25 mm 3
been shown differentiate MSCs into the ligament 24.18 mm) coated with type I collagen designed to
lineage but not into bone or cartilage,11 PDLCs into provide a uniform uniaxial strain. They were then
osteoclasts,12 MSCs into chondrocytes,13 MSCs into subjected to an intermittent deformation of 3%, 6%,
bone,14 and dental pulp stem cells into dentin.15 12% or 15% of maximum stretch for 2.5 seconds
However, MS such as cyclic tension has been reported followed by 2.5 seconds of relaxation (12 cycle/min for
to both inhibit16 and stimulate17 the synthesis of alkaline 24 h) with a Flexercell FX-4000 Strain Unit (Flexcell
phosphatase. In mice, orthodontic force-induced os- Corporation, Hillsborough, NC) according to the
teoclastogenesis in alveolar bone was inhibited by the manufacturer’s instructions.
accompanying periodontal inflammation and resulted
in reduced orthodontic tooth movement.18 But there is RNA Isolation and Reverse Transcription–
no evidence to show that MS and LPS affect the Polymerase Chain Reaction
induction of growth factor genes, specifically BMPs, in
PDLCs. After applying the MS, total RNA was isolated from the
Heme oxygenase (HO) is the rate-limiting enzyme in cells using the Trizol reagent (Invitrogen Life Technolo-
the catabolism of heme-generating biliverdin. Previ- gies, Carlsbad, Calif) according to the manufacturer’s
ously, we reported that the HO-1 pathway is a key instructions. Briefly, 1 mg of RNA isolated from each
mechanism for the adaptation to stressful conditions cultures was reverse transcribed using oligo (dT)15
and the recovery from injurious events by dental primers (Roche Diagnostics, Mannheim, Germany) and
cells.19–24 Moreover, the expression of HO-1 is related AccuPower RT PreMix (Bioneer, Daejon, Korea) accord-
to adipogenesis by human MSCs,25 osteoblastic ing to the manufacturer’s protocols. An amount of cDNA
differentiation by PDLCs23 and neuronal differentiation equivalent to 25 ng of total RNA was then subjected to
by MSCs.26 polymerase chain reaction (PCR). The primers used for
Because HO-1 is associated with differentiation, it is cDNA amplification are listed in Table 1. PCR products
plausible to postulate that HO-1 is involved in the were subjected to electrophoresis on 1.2% agarose gel
response of PDLCs to MS and bacterial infection as and were stained with ethidium bromide.
well as in the differentiation of PDLCs to osteoblast-like
cells. The objective of this study was to examine the Western Blot Analysis
effects of MS and LPS on the osteodifferentiation of
PDLCs. We also examined the underlying signaling An equal volume of 2 3 sodium dodecyl sulfate
pathways, by measuring changes in the expression of (SDS) sample buffer was added and the samples were
HO-1, BMPs, BMP antagonists, and transcription then boiled for 5 minutes. A sample (40 mg) was
factor involved in osteoblastic differentiation. subjected to electrophoresis on 12% SDS-polyacryl-
amide gels for 2 hours at 20 mA and then transferred
MATERIALS AND METHODS onto nitrocellulose. The membrane was incubated for
1 hour in 5% (wt/vol) dried milk protein in phosphate
Cell Culture
buffer solution (PBS) containing 0.05% (vol/vol)
Human immortalized PDLCs were established from Tween-20 (PBS-T), washed in PBS-T and then
normal PDLCs through HPV16 E6 and E7 genes incubated for 1 hour in the presence of primary
transfection.27 The PDLCs were cultured in Dulbecco’s antibody (1:1,000). The membrane was washed
modified Eagle’s medium (DMEM; Gibco BRL, Gai- extensively with PBS-T and then incubated with anti-
thersburg, Md) supplemented with 10% fetal bovine mouse IgG antibody conjugated to horseradish perox-
serum (FBS) in a humidified atmosphere of 5% CO2 at idase (1:3,000) for 1 hour. After extensive washes,
37uC. For differentiation, PDLCs were cultured with immunoreactive bands on the membrane were visual-
differentiation medium (10% FBS/DMEM, including ized using chemiluminescent reagents according to
50 mg/mL ascorbic acid and 10 mM b-glycerophos- the manufacturer’s protocol (Amersham-Pharmacia,
phate), as described previously.28 This study was Piscataway, NJ).

742 CHO, LEE, LEE, KIM
Table 1. Reverse Transcription–Polymerase Chain Reaction Primers

Gene Sequence (59 -39) Size (bp)
Heme oxygenase-1 Forward: AAGATTGCCCAGAAAGCCCTGGAC 405
Reverse: AACTGTGCCACCAGAAAGCTGAG
Bone morphogenetic protein-2 Forward: CCAACCATGGATTCGTGGTG 456
Reverse: GGTACAGCATCGAGATAGCA
Bone morphogenetic protein-7 Forward: CAGCCTGCAAGATAGCCATT 276
Reverse: AATCGGATCTCTTCCTGCTC
Runx2 Forward: AACCCACGAATGCACTATCCA 75
Reverse: CGGACATACCGAGGGACCTG
Noggin Forward: GCACCCAGCGACAACCTGCCC 399
Reverse: GCTGCCCACCTTCACGTAGCG
Glyceraldehyde 3-phosphate dehydrogenase Forward: ACCACAGTCCATGCCATCAC 452
Reverse: TCCACCACCCTGTT GCTGTA
Statistical Analysis differentiation in PDLCs. However, 0.1 mg/mL and 1 mg/

mL LPS enhanced MS-induced HO-1 upregulation and
The statistical analyses of the data were performed
noggin downregulation (Figure 2).
by one-way analysis of variance followed by a multiple-
comparison Turkey’s test with the use of the SPSS
Effects of Inducing or Inhibiting HO-1 Expression
program (SPSS 12.0, SPSS GmbH, Munich, Ger-
on MS-induced and LPS-induced
many). Statistical significance was determined at P ,
Osteogenic Differentiation
.05.
To verify the role of HO-1 in counteracting the
RESULTS osteoblastic differentiation caused by LPS and MS in
PDLCs, we investigated the effect of pretreating cells
The Effects of MS on Osteodifferentiation in PDLCs with protoporphyrin IX chloride (CoPP, a potent HO-1
In PDLCs, MS increased the levels of BMP-2, BMP- inducer) or tin-protoporphyrin IX (SnPP, a potent HO-1
7, and Runx2 mRNAs in a force-dependent and time- inhibitor) for 16 hours. CoPP pretreatment increased
dependent manner (Figure 1A,B). In contrast, MS while SnPP blocked the induction of HO-1 mRNA
resulted in a force-dependent and time-dependent expression by LPS treatment and MS. The LPS-
decrease in noggin mRNA. The maximal expression of induced and MS-induced BMP-2, BMP-7, and Runx-2
BMP-2, BMP-7, and Runx2 mRNAs was observed in mRNA upregulation was enhanced by CoPP pretreat-
cells subjected to MS applied at a force causing a 12% ment (Figure 3). Also, pretreatment with CoPP and the
increase in cell length for 48 hours. corresponding rise in HO-1 prevented the downregu-
lation of noggin mRNA seen with LPS and MS. In
The Effects of MS on HO-1 Expression in PDLCs contrast, SnPP pretreatment prevented the LPS-
induced and MS-induced osteogenic differentiation.
As shown in Figure 1A,B, HO-1 mRNA expression SnPP also inhibited HO-1 expression, and this
gradually increased in PDLCs treated with MS. This inhibition enhanced the decrease in noggin mRNA
time-dependent and force-dependent upregulation expression by exposure to LPS and MS.
peaked at 12% of MS and decreased with greater
force. Along with this upregulation in HO-1 mRNA Effects of Signal Transduction Modulators on MS-
expression, we also found a corresponding increase in induced and LPS-induced
HO-1 protein expression (Figure 1C,D). Osteogenic Differentiation
Effects of MS and LPS on Osteogenic To examine the signaling pathways involved in MS-
Differentiation and HO-1 Expression in induced and LPS-induced osteogenic differentiation
Human PDLCs and HO-1 expression, PDLCs were pretreated with
various inhibitors of key signaling molecules. MS-
To examine the combined effect of LPS and MS on induced and LPS-induced osteogenic differentiation
osteogenic differentiation in human PDLCs, we applied and HO-1 expression were inhibited in PDLCs by the
a force resulting in 12% cellular elongation with LPS selective p38 mitogen-activated protein kinase
derived from P gingivalis for 48 hours, and then (MAPK) inhibitor SB203580, the JNK inhibitor
reverse transcription–PCR was performed. LPS de- SP600125, the specific membrane-permeable protea-
creased MS-induction of BMP-2, BMP-7, and Runx-2, some inhibitor MG132, the phosphoinositide 3-kinases
a key transcription factor associated with osteoblast (PI3K) inhibitor LY294002, the protein kinase G (PKG)

Figure 1. Effects of mechanical stress on expression of heme oxygenase-1 and osteogenic differentiation markers in periodontal ligament cells.
Cells were cultured with or without mechanical stress (3%–15%) for up to 48 hours. mRNA and protein were assayed by semiquantitative reverse
transcription–polymerase chain reaction (A and C) and Western blotting (E and G). Quantitative data on the relative amounts of mRNA or protein
of genes to glyceraldehyde 3-phosphate dehydrogenase or b-actin are provided (B, D, E, and F). Experiments were performed in triplicate for
each data point, and the standard errors are shown as error bars. *Statistically significant difference compared with control, P , .05.

Figure 2. Effects of mechanical stress (MS) and lipopolysaccharide from Porphyromonas gingivalis on osteogenic differentiation and heme
oxygenase-1 expression. Cells cultured with or without indicated concentrations of lipopolysaccharide and MS (12%) for 48 hours were assayed
by reverse transcription–polymerase chain reaction (A). Quantitative data on the relative amounts of mRNA genes to glyceraldehyde 3-
phosphate dehydrogenase are provided on the right (B). These data are representative of three independent experiments. *Statistically
significant difference compared with control, P , .05; #statistically significant difference compared with MS-treated group, P , .05.
inhibitor KT5823, and the nuclear factor kB (NF-kB) expression of inflammatory cytokines,29 cytoskeletal
inhibitor PDTC. Osteogenic differentiation and HO-1 components,16 and osteogenic genes.17,30,31 In this
expression were not inhibited by the ERK1/2 pathway study, we provide evidence for a role of the HO-1
inhibitor PD98059 (Figure 4). pathway and relevant signaling molecules in the
regulation of LPS-induced and strain-induced osteo-
blastic differentiation.
DISCUSSION
BMPs are known to play a critical role in prolifera-
Human PDLCs have been widely used to study the tion, differentiation, and matrix secretion of bone
effects of tensile MS on the protein and mRNA cells.31 In the present study, we found that MS on
Figure 3. Effects of heme oxygenase-1 (HO-1) inducer and inhibitor on osteogenic differentiation induced by mechanical stress and
lipopolysaccharide (LPS). Cells were pretreated with different concentrations of tin-protoporphyrin IX (an HO-1 inhibitor) or protoporphyrin IX
chloride (an HO-1 inducer) for 16 hours and then stimulated with or without strain and LPS for an additional 48 hours (A). Quantitative data on the
relative amounts of mRNA genes to glyceraldehyde 3-phosphate dehydrogenase are provided on the right (B). *Statistically significant difference
compared with control, P , .05; #statistically significant difference compared with group treated with mechanical stress and LPS, P , .05. The
experiments were performed three times, and representative data are shown.

Figure 4. Effects of signal transduction modulators on osteogenic differentiation induced by mechanical stress and lipopolysaccharide (LPS).
Cells were pretreated with 20 mM SB203580, 20 mM PD098059, 20 mM SP600125, 10 mM PDTC, 20 mM KT5823, and 20 mM LY694002 for 1 hour
and then stimulated with or without strain and LPS for an additional 48 hours. Quantitative data on the relative amounts of mRNA genes to
glyceraldehyde 3-phosphate dehydrogenase are provided (B). The experiments were performed three times, and representative data are shown.
*Statistically significant difference compared with control, P , .05; #statistically significant difference compared with group treated with
mechanical stress and LPS, P , .05.
PDLCs induced the transcriptional expression of BMP- It has been well documented that BMPs upregulate
2 and BMP-7. These data, therefore, are in agreement various transcription factors involved in osteoblastic
with previous studies showing that MS upregulates differentiation.14,33 Thus, we determined the effect of
BMP-2, BMP-4, BMP-6, and BMP-7 in osteoblasts14; MS on the expression of Runx2 transcription factor.
BMP-2, BMP-6, and BMP-7 in osteoblasts32; and BMP- Our results showed that the expression of Runx2
2 and BMP-6 in PDLCs.30 mRNA was significantly increased by the application of
There is evidence for the autoregulation of BMP MS, which suggests that MS may promote osteoblastic
expression in osteoblasts in the form of a negative differentiation through the regulation of BMP-respon-
feedback loop that decreases cellular exposure to sive transcription factors in PDLCs.
BMPs. Thus, in this study, we focused on extracellular HO-1 is emerging as the prototypic endogenous
antagonists of BMP, such as noggin. We found that the cytoprotective enzyme essential for cells to adapt to
expression of BMP antagonists decreased with the stressful conditions and to recover from injurious
application of MS in a force-dependent manner. From events.19–24 In the present study, we found that MS
these findings, mechanical loading may stimulate increased HO-1 mRNA and protein expression in
osteoblastic differentiation by regulating the expres- PDLCs. These findings are consistent with our
sion of BMPs and BMP antagonists. previous study in which osteogenic differentiation

CONCLUSIONS
N To our knowledge this study is the first to demon-
strate that MS-induced osteogenic differentiation of
PDLCs is inhibited by the accompanying periodontal
inflammation.
N HO-1 may influence changes on the differentiation in
PDLCs.
ACKNOWLEDGMENT
This article was supported by Wonkwang University in 2008.
REFERENCES
1. Alexander SA. Effects of orthodontic attachments on the
Figure 5. Schematic diagram illustrating the signaling pathways
gingival health of permanent second molars. Am J Orthod
triggered by exposure to mechanical stress and lipopolysaccharide in
periodontal ligament cells.
2. Paschos E, Limbach M, Teichmann M, Huth KC, Folwaczny
M, Hickel R, Rudzki-Janson I. Orthodontic attachments and
was found to be associated with HO-1 expression in chlorhexidine-containing varnish effects on gingival health.
PDLCs.24 Thus, upregulation of HO-1 may provide a Angle Orthod. 2008;78:908–916.
major contribution to cellular adaptation and resistance 3. Wennstrom JL, Stokland BL, Nyman S, Thilander B.
Periodontal tissue response to orthodontic movement of
to MS. teeth with infrabony pockets. Am J Orthod Dentofacial
It is not known whether LPS and MS affect Orthop. 1993;103:313–319.
osteoblastic differentiation, such as BMP expression 4. Kawarizadeh A, Bourauel C, Götz W, Jäger A. Early
in PDLCs. In this study we found that LPS potently responses of periodontal ligament cells to mechanical
inhibited MS-induced BMP-2, BMP-7, and Runx2 stimulus in vivo. J Dent Res. 2005;84:902–906.
5. Wada N, Maeda H, Tanabe K. Periodontal ligament cells
expression, whereas LPS recovered noggin expres- secrete the factor that inhibits osteoclastic differentiation and
sion. These results imply that LPS acts as a negative function: the factor is osteoprotegerin/osteoclastogenesis
modulator of MS-induced osteoblastic differentiation. inhibitory factor. J Periodontal Res. 2001;36:56–63.
Antioxidant and cytoprotective as well as anti- 6. Kobayashi M, Takiguchi T, Suzuki R, et al. Recombinant
inflammatory and immunosuppressive effects have human bone morphogenetic protein-2 stimulates osteoblast
differentiation in cells isolated from human periodontal
been described following HO-1 induction by CoPP.34 ligament. J Dent Res. 1999;78:1624–1633.
Thus, the effect of pharmacologic HO-1 induction was 7. Wright E, Hargrave MR, Christiansen J, et al. The Sry-
tested using CoPP. When applied in our PDLCs, the related gene Sox9 is expressed during chondrogenesis in
HO-1 inducer CoPP enhanced the LPS-induced and mouse embryos. Nature Genetics. 1995;69:15–20.
MS-induced osteoblastic differentiation. This CoPP- 8. Canalis E, Economides AN, Gazzerro E. Bone morphoge-
netic proteins, their antagonists, and the skeleton. Endocr
enhanced differentiation was abrogated by SnPP Rev. 2003;24:218–235.
treatment, proving the specificity of this HO-1 effect. 9. Mitsui N, Suzuki N, Maeno M, et al. Optimal compressive
Thus, for the first time, we have demonstrated the force induces bone formation via increasing bone morpho-
effects of HO-1 induction on MS-induced and LPS- genetic proteins production and decreasing their antagonists
induced osteogenic differentiation in PDLCs. production by Saos-2 cells. Life Sci. 2006;78:2697–2706.
10. Estes BT, Gimble JM, Guilak F. Mechanical signals as
HO-1 gene expression can be induced via signaling
regulators of stem cell fate. Curr Top Dev Biol. 2004;60:
pathways involving MAPKs (ERK, JNK, p38), PI3K, 91–126.
and NF-kB in PDLCs.21,22,24 Increases in the expres- 11. Altman GH, Horan RL, Martin I, et al. Cell differentiation by
sion of HO-1, BMP-2, BMP-7, and Runx-2 resulting mechanical stress. FASEB J. 2002;16:270–272.
from treatment with LPS and MS were attenuated by 12. Kanzaki H, Chiba M, Shimizu Y, Mitani H. Periodontal
ligament cells under mechanical stress induce osteoclasto-
preincubation with selective inhibitors of p38, ERK,
genesis by receptor activator of nuclear factor kappaB
PI3K, PKG, and NF-kB (SB203580, U0126, ligand up-regulation via prostaglandin E2 synthesis. J Bone
LY294002, KT5823, and PDTC, respectively). These Miner Res. 2002;17:210–220.
results suggest that LPS-induced and MS-induced 13. Huang CYC, Hagar KL, Frost LE, Sun Y, Cheung HS.
HO-1 and BMP expression occurs via the activation of Effects of cyclic compressive loading on chondrogenesis of
rabbit bone-marrow derived mesenchymal stem cells. Stem
p38, ERK MAPK, PI3K, PKG, and NF-kB. Figure 5 is a
Cells. 2004;22:313–323.
schematic representation of the signaling pathway 14. Sumanasinghe RD, Bernacki SH, Loboa EG. Osteogenic
involved in HO-1 and BMP activation in response to differentiation of human mesenchymal stem cells in collagen
LPS and MS in PDLCs. matrices: effect of uniaxial cyclic tensile strain on bone

morphogenetic protein (BMP-2) mRNA expression. Tissue and inflammatory cytokines tumor necrosis factor-alpha and
Eng. 2006;12:3459–3465. interleukin-6 in Zucker rats and reduce adipogenesis in
15. Yu V, Damek-Poprawa M, Nicoll SB, Akintoye SO. Dynamic human mesenchymal stem cells. J Pharmacol Exp Ther.
hydrostatic pressure promotes differentiation of human 2008;325:833–840.
dental pulp stem cells. Biochem Biophys Res Commun. 26. Barbagallo I, Tibullo D, Di Rosa M, et al. A cytoprotective
2009;386:661–665. role for the heme oxygenase-1/CO pathway during neural
16. Chiba M, Mitani H. Cytoskeletal changes and the system of differentiation of human mesenchymal stem cells. J Neurosci
regulation of alkaline phosphatase activity in human Res. 2008;86:1927–1935.
periodontal ligament cells induced by mechanical stress. 27. Pi SH, Lee SK, Hwang YS, Choi MG, Lee SK, Kim EC.
Cell Biochem Funct. 2004;22:249–256. Differential expression of periodontal ligament-specific
17. Yang YQ, Li XT, Rabie AB, Fu MK, Zhang D. Human markers and osteogenic differentiation in human papilloma
periodontal ligament cells express osteoblastic phenotypes virus 16-immortalized human gingival fibroblasts and peri-
under intermittent force loading in vitro. Front Biosci. 2006; odontal ligament cells. J Periodontal Res. 2007;42:104–
11:776–781. 113.
18. Okamoto A, Ohnishi T, Bandow K, et al. Reduction of 28. Fujii S, Maeda H, Wada N, Kano Y, Akamine A. Establishing
orthodontic tooth movement by experimentally induced and characterizing human periodontal ligament fibroblasts
periodontal inflammation in mice. Eur J Oral Sci. 2009; immortalized by SV40T-antigen and hTERT gene transfer.
117:238–247. Cell Tissue Res. 2002;324:117–125.
19. Min KS, Hwang YH, Ju HJ, et al. Heme oxygenase-1
29. Long P, Hu J, Piesco N, Buckley M, Agarwal S. Low
mediates cytoprotection against nitric oxide-induced cyto-
magnitude of tensile strain inhibits IL-1-dependent induction
toxicity via the cGMP pathway in human pulp cells. Oral
of proinflammatory cytokines and induces synthesis of IL-10
Surg Oral Med Oral Pathol Oral Radiol Endod. 2006;102:
in human periodontal ligament cells in vitro. J Dent Res.
803–808.
2001;80:1416–1420.
20. Min KS, Kwon YY, Lee HJ, et al. Effects of proinflammatory
30. Wescott DC, Pinkerton MN, Gaffey BJ, Beggs KT, Milne TJ,
cytokines on the expression of mineralization markers and
Meikle MC. Osteogenic gene expression by human peri-
heme oxygenase-1 in human pulp cells. J Endod. 2006;32:
39–43. odontal ligament cells under cyclic tension. Dent Res. 2007;
21. Pi SH, Kim SC, Kim HT, Lee HJ, Lee SK, Kim EC. Defense 86:1212–1216.
mechanism of heme oxygenase-1 against cytotoxic and 31. Wongkhantee S, Yongchaitrakul T, Pavasant P. Mechanical
receptor activator of NF kappa-B ligand inducing effects of stress induces osteopontin expression in human periodontal
hydrogen peroxide in human periodontal ligament cells. ligament cells through rho kinase. J Periodontol. 2007;78:
J Periodontal Res. 2007;42:331–339. 1113–1119.
22. Lee SK, Pi SH, Kim SH, et al. Substance P regulates 32. Siddhivarn C, Banes A, Champagne C, Riché EL, Weer-
macrophage inflammatory protein 3a/CCL20 with heme apradist W, Offenbacher S. Mechanical loading and
oxygenase-1 in human periodontal ligament cells. Clin Exp D12prostaglandin J2 induce bone morphogenetic protein-2,
Immunol. 2007;150:567–575. peroxisome proliferator-activated receptor c-1, and bone
23. Lee SK, Min KS, Kim Y, et al. Mechanical stress activates nodule formation in an osteoblastic cell line. J Periodont
proinflammatory cytokines and antioxidant defense en- Res. 2007;42:383–392.
zymes in human dental pulp cells. J Endod. 2008;34: 33. Ducy P, Zhang R, Geoffroy V, Ridall AL, Karsenty G. Osf2/
1364–9136. Cbfa1: a transcriptional activator of osteoblast differentia-
24. Kook YA, Lee SK, Son DH, et al. Effects of substance P on tion. Cell. 1997;89:747–754.
osteoblastic differentiation and heme oxygenase-1 in human 34. Akamatsu Y, Haga M, Tyagi S, et al. Heme oxygenase-1-
periodontal ligament cells. Cell Biol Int. 2009;33:424–428. derived carbon monoxide protects hearts from transplant
25. Kim DH, Burgess AP, Li M, et al. Heme oxygenase- associated ischemia reperfusion injury. FASEB J. 2004;18:
mediated increases in adiponectin decrease fat content 771–772.

Original Article
Skeletal changes of maxillary protraction without rapid maxillary expansion

A comparison of the primary and mixed dentition
Dong-Yul Leea; Eun-Soo Kimb; Yong-Kyu Lima; Sug-Joon Ahnc
ABSTRACT
Objective: To determine potential differences in treatment efficiencies of face mask therapy
without rapid maxillary expansion (RME) at different early dental stages.
Materials and Methods: Forty-nine Class III children who were treated with a face mask without
RME were divided into two groups according to their pretreatment dental stage. The primary
dentition treatment group consisted of 26 subjects and the mixed dentition treatment group
consisted of 23 subjects. Lateral cephalograms before treatment (T0), at the end of treatment (T1),
and at least 1 year after the end of treatment (T2) were calculated and analyzed. Fourteen
cephalometric variables were evaluated by t-test to identify any significant differences in skeletal
changes between the two groups during T1-T0, T2-T1, and T2-T0.
Results: The primary dentition group showed not only a greater response to maxillary protraction
without RME than did the mixed dentition group during T1-T0, but also a greater relapse tendency
during T2-T1. As a result, no significant differences were noted between the two groups in the
treatment effects of face masks without RME over the time period T2-T0.
Conclusion: This study suggests that face mask therapy without RME may be postponed to the
early to mid mixed dentition period because the therapy induces similar skeletal changes when
initiated at primary or mixed dentition. (Angle Orthod. 2010;80:692–698.)
KEY WORDS: Skeletal changes; Maxillary protraction; Face mask; Dental age
INTRODUCTION improving occlusion in skeletal Class III malocclusion

subjects.3 However, several long-term studies have
Class III malocclusions with anterior crossbites are
shown that the orthopedic effects of chin cup therapy
common clinical problems among East Asians, includ-
are questionable.4,5 In addition, it has been reported
ing Koreans.1 More than half of the preadolescent
that the orthopedic advantages of chin cup therapy
Korean patients visiting an orthodontist have a Class seem to outweigh the disadvantage of occasionally
III malocclusion.2 inducing a temporomandibular joint disorder.6 The
Class III malocclusions can be treated in several recognition that maxillary deficiency is another cause
ways in growing patients. The chin cup has been used of Class III malocclusion in children has led to an
to control mandibular overgrowth, which is one of the increase in face mask treatment, a therapy used to
main reasons for Class III malocclusions. Many studies enhance maxillary forward growth.1,7 Approximately half
have demonstrated that chin cup therapy is effective in of Korean children with a Class III malocclusion have a
maxillary deficiency.7
a
Professor, Department of Orthodontics, Graduate School of One important question is whether there exists an
Clinical Dentistry, Korea University, Seoul, South Korea.
optimal time to start treatment in growing children. A
b
Private Practice, Seoul, South Korea.
c
Associate Professor, Department of Orthodontics, School of number of authors believe that early face treatment
Dentistry and Dental Research Institute, Seoul National Univer- may be more effective in improving the skeletal
sity, Seoul, Korea. relationship in Class III malocclusion children,8–12 but
Corresponding author: Dr Sug-Joon Ahn, Department of others have found no differences in the treatment of
Orthodontics and Dental Research Institute, School of Dentistry,
different age groups.13–15 However, most studies have
Seoul National University, 28-22 Yeungeon-dong, Jongro-gu,
Seoul, Korea 110-768 South Korea analyzed the treatment efficiency of maxillary protrac-
(e-mail: titoo@snu.ac.kr) tion using chronological age rather than developmental
Accepted: November 2009. Submitted: September 2009. stage. This results in a high level of variation, making it
G 2010 by The EH Angle Education and Research Foundation, difficult to determine treatment efficiency. Evaluating
Inc. treatment efficiency using developmental stages such

SKELETAL CHANGES BY MAXILLARY PROTRACTION 693
Table 1. Ages at the Start of Orthodontic Treatment (T0), at the End of Treatment (T1), and after Retention (T2)
Primary Dentition Group (n 5 26) Mixed Dentition (n 5 23)
Mean (SD) Range Mean (SD) Range
T0 (y) 6.1 (0.5) 5.1–6.9 8.4 (0.8) 6.4–9.7
T1 (y) 7.2 (0.5) 6.2–8.3 9.7 (0.9) 7.8–11.0
T2 (y) 8.6 (0.6) 7.6–10.1 11.1 (1.0) 9.3–12.9
as dental age is better because it is simple and easy to consisted of 23 subjects (14 girls and 9 boys) in the
perform in a clinical setting. early to mid mixed dentition stage (stage from
Rapid maxillary expansion (RME) has been used complete eruption of permanent first molars and
widely with various maxillary protraction devices to permanent incisors to shedding of primary canines
expand constricted maxillary arches, as posterior and molars and the eruption of successors; Hellman’s
crossbites are relatively common in maxillary deficien- developmental stages IIIA-IIIB15). The mean ages of
cy cases.11,13,14 In addition, RME can produce suture both groups at T0, T1, and T2 are shown in Table 1.
opening within all maxillary bone sutures, which helps Differences between groups in treatment time (T1-T0),
maxillary protraction by face mask.10–13 However, many the observation period (T2-T1), and total observation
patients are too young or have a transverse dimension time (T2-T0) are described in Table 2.
of the maxilla that is too to be fitted with an RME, Treatment was carried out using a Delaire-type face
particularly in the primary dentition. mask16 (KJ Meditech, KwangJu, Korea) with a remov-
The purposes of this retrospective investigation able intraoral appliance and heavy elastics (Figures 1A
were to analyze differences in treatment efficiencies and 1B). Adams’ clasps and the covered occlusal
of maxillary protraction without RME between earlier surface in the intraoral appliance provided retention for
treatment during the primary dentition and later the elastics (Figures 1C and 1D). Adams’ clasps were
treatment during the early to mid mixed dentition, placed on the primary molars before eruption of the first
and to compare skeletal changes during treatment and molar, and on the permanent first molar in the case of
observation periods. The research protocol was complete eruption of the maxillary first molar. Hooks for
reviewed and approved by the institutional review the elastics were placed between the primary canine and
board of the University Hospital (IRB No: GT0914). the primary first molar. Elastics (3M/Unitek, Monrovia,
Calif) were attached from the hooks on the appliance to
MATERIALS AND METHODS the support bar of the face mask in a downward and
forward direction (30 degrees from the occlusal plane),
Forty-nine Class III children (32 girls and 17 boys) producing an orthopedic force of 350 g per side. To
who satisfied the following criteria were selected from prevent dislodging, the appliances were readjusted by
176 patients who visited the Department of Orthodon- putting acrylic in the area of both upper buccal segments
tics, Korea University Guro Hospital, from 1999 to when required. Patients were instructed to wear the face
2001, and were treated with a face mask. These mask for at least 12 hours per day. All patients were
patients showed (1) primary or early to mid mixed treated to a positive dental overjet before discontinuing
dentition; (2) Class III skeletal pattern with anteropos- treatment. Most patients were overcorrected toward a
terior maxillary deficiency, mesial step, and anterior distal step of the primary molars.
crossbite; (3) available cephalograms from before the Lateral cephalograms of each patient were taken at
initiation of treatment (T0), at the end of treatment (T1), T0, T1, and T2 and were traced and analyzed by a
and at least 1 year after treatment (T2); (4) no retention single investigator. Thirteen landmarks were digitized
device, such as a fixed appliance or functional on each radiograph (Figure 2), from which 14 variables
appliance during T2-T1; (5) no RME therapy; (6) no
other craniofacial anomalies or skeletal asymmetry; (7)
Table 2. Comparison of Treatment Time (T1-T0), Retention Time
no skeletal transverse problems; and (8) no previous (T2-T1), and Total Observation Time (T2-T0) Between the Primary
orthodontic treatment. Dentition Group (PG) and the Mixed Dentition Group (MG)
The sample was divided into two groups according PG (n 5 26) MG (n 5 23) Significancea
to dental development stage at T0. The primary
T1-T0 (mo) 13.0 (3.7) 15.1 (5.2) NS
dentition group (PG) consisted of 26 subjects (18 girls T2-T1 (mo) 17.2 (6.7) 16.5 (6.4) NS
and 8 boys) in the primary dentition stage (from T2-T0 (mo) 30.3 (7.8) 31.6 (7.0) NS
completion of primary dentition to beginning of eruption NS indicates not significant.
of permanent first molars; Hellman’s developmental a
t-test was performed with a significance level of a 5 .05 to
stages IIA-IIC15). The mixed dentition group (MG) compare differences between groups.

694 LEE, KIM, LIM, AHN
Figure 1. (A) Delaire-type face mask with elastics that delivered 350 g of maxillary protraction force on each side, 30 degrees downward from the
occlusal plane. (B) An intraoral removable appliance with hooks at the canine-premolar area.
were selected for evaluation of skeletal changes repeated cephalometric measurements at T0, T1, and
induced by the face mask treatment (Table 3). Only T2 were correlated or clustered within each subject.
angular (Figures 3 and 4) and proportional measure- Therefore, a multivariate statistical approach was
ments were used to minimize the differences in required. After the three basic assumptions were
magnitude between PG and MG according to the confirmed, the normality of the distribution, the equality
patient’s developmental stage. of the variance, and the spherical assumption,
Skeletal changes during treatment, during the repeated measure analysis of variance (ANOVA),
observation period, and over the total observation time were performed. In the analysis model, we incorporat-
were found for each variable by subtracting T0 from ed gender variable as well as between-group (MG vs
T1, subtracting posttreatment T1 from T2, and PG) variables. All values were considered significant at
subtracting T0 from T2, respectively (Table 4). Multiple P , .05.

RESULTS
Table 2 shows the differences in treatment time (T1-
T0), retention time (T2-T1), and total time (T2-T0)
between PG and MG. No significant time difference
was noted between the two groups during any time
period, indicating that we could eliminate the effects of
time on changes in skeletal parameters to determine
the treatment efficiency of face mask therapy during
any time period.
Table 3 shows the differences in skeletal morphol-
ogy at T0, T1, and T2. No significant difference in
skeletal morphologies was noted between PG and
MG, except the mandibular body-to-anterior cranial
base ratio at T0. This indicates that the patients in MG
had relatively larger mandibles at T0 than did those in
PG. Anteroposterior skeletal discrepancies at T1 were
greatly improved, although the difference in the
mandibular body-to-anterior cranial base ratio still
persisted after face mask therapy (Table 3). However,
a significant difference in the ANB angle was noted
between the two groups at T1. The ANB angle was
significantly larger in PG than in MG at the end of
treatment. All skeletal differences, including the ANB
and the mandibular body-to-anterior cranial base ratio,
showed no significant difference between the two
groups at T2.
Figure 2. The cephalometric landmarks used in this study. (1) Table 4 shows the changes in skeletal variables of
Nasion, (2) orbitale, (3) anterior nasal spine, (4) point A, (5) point B, the two groups over T1-T0, T2-T1, and T2-T0,
(6) pogonion, (7) gnathion, (8) menton, (9) gonion, (10) posterior respectively. SNA, ANB, SN-GoGn, and mandibular
nasal spine, (11) articulare, (12) porion, and (13) sella.
plane angle increased, while SNB and facial plane
angle decreased during treatment, which means that
skeletal Class III malocclusions were corrected by
forward movement of the maxilla and downward-
Table 3. Cephalometric Comparisons between the Primary Dentition Group (PG) and the Early Mixed Dentition Group (MG) at the Beginning of
Treatment (T0), after Treatment (T1), and at Least 1 Year after Treatment (T2)
T0 T1 T2
PG (n 5 26) MG (n 5 23) PG (n 5 26) MG (n 5 23) PG (n 5 26) MG (n 5 23)
Variables Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD)
SNA 78.95 (2.46) 78.13 (2.74) 81.20 (2.30) 80.19 (2.79) 81.03 (2.61) 80.88 (3.04)
SNB 79.42 (2.14) 79.17 (2.90) 77.67 (1.95) 78.39 (2.98) 78.60 (2.23) 79.11 (3.13)
ANB 20.47 (1.58) 21.04 (1.78) 3.53 (1.91)** 1.81 (1.95)** 2.43 (1.47) 1.78 (1.19)
Saddle angle 122.78 (3.72) 123.56 (4.65) 123.45 (4.16) 123.55 (4.29) 123.70 (4.03) 123.27 (5.12)
Articular angle 147.82 (4.96) 147.45 (5.54) 149.86 (4.97) 149.09 (5.65) 149.21 (4.78) 148.43 (5.46)
Gonial angle 127.74 (4.25) 129.10 (4.89) 126.85 (4.00) 128.67 (4.55) 126.22 (3.58) 128.24 (5.02)
Sum 398.35 (3.63) 400.12 (4.00) 400.16 (3.46) 401.31 (4.62) 399.13 (3.26) 400.94 (4.27)
SN-GoGn 38.32 (3.63) 40.08 (4.00) 40.13 (3.46) 41.30 (4.63) 39.10 (3.25) 40.92 (4.27)
Facial plane angle 78.77 (1.99) 78.86 (3.23) 77.22 (1.83) 77.91 (3.28) 78.22 (2.16) 78.71 (3.41)
Palatal plane angle 1.77 (2.61) 1.27 (1.89) 1.06 (2.68) 0.91 (2.42) 1.87 (2.34) 1.61 (2.73)
Mandibular plane angle 30.36 (3.37) 31.63 (3.65) 31.95 (3.31) 33.23 (4.06) 31.49 (2.92) 32.49 (4.12)
Mandibular body length-to-anterior
cranial base ratio (%) 99.96 (4.02)* 103.30 (6.90)* 101.53 (4.80)* 105.30 (6.22)* 103.44 (5.20) 106.27 (8.13)
Facial height ratio (%) 60.44 (2.84) 59.50 (3.03) 59.65 (2.45) 59.06 (3.57) 60.56 (2.59) 59.77 (3.28)
ANS-Me/Nasion-Me (%) 55.81 (1.40) 55.19 (1.64) 56.59 (1.44) 56.08 (1.69) 56.04 (1.57) 55.33 (1.74)
* P , .05; ** P , .01.

Figure 4. Cephalometric variables used in this study (cont’d). (7)

Facial plane angle, (8) palatal plane angle, (9) mandibular plane
angle, and (10) SN-GoGn.
be closely related to skeletal age,18,19 it can be a useful

alternative for determining the most effective timing of
Figure 3. Cephalometric variables used in this study. (1) SNA, (2) maxillary protraction.
SNB, (3) ANB, (4) saddle angle, (5) articular angle, and (6) gonial Many studies have described the general treatment
angle. Sum is the total sum of the saddle angle, articular angle, and efficiency of face mask with RME therapy as a
gonial angle. combination of effective skeletal and dental modifica-
tions.9–13 Although RME can facilitate effective maxil-
backward rotation of the mandible in both groups. lary forward movement by disrupting the circum-
However, changes in SNB and ANB were significantly maxillary sutural system,10,11 patients who have a
different between the two groups during T1-T0 sufficient transverse dimension of the maxilla need
(Table 4). Changes in the ANB showed a statistically only maxillary protraction. In addition, spontaneous
significant difference between the two groups during improvement of posterior crossbites is observed after
T2-T1, but no significant difference in skeletal changes maxillary protraction because the protracted maxillary
was noted during T2-T0 (Table 4). arch fits well with its smaller counterpart of the
mandible.
DISCUSSION Furthermore, most studies have not investigated the
The optimal time to start early face mask therapy in treatment efficiency of face mask treatment in very
patients with maxillary deficiencies is still controversial. young patients with primary dentition. Although a
This can be explained in part by the fact that most previous study compared the treatment effects of
studies investigating the optimal treatment time for maxillary protraction between a deciduous dentition
face mask therapy have used chronological age as a group and an early mixed dentition group, maxillary
classification criterion.8,10,17 The age ranges of patients protraction was performed with RME and data on the
in previous studies were fairly wide, meaning that the observation period were not included.12 The purpose of
skeletal effects induced by maxillary protraction might this study was to determine the most effective timing
vary with age. Although skeletal age is a more reliable for face mask therapy without RME in young growing
clinical indicator than is chronological age for deter- patients with PG or MG by comparing skeletal changes
mining the treatment efficiency of face mask therapy, during T1-T0 and T2-T1 on the basis of dental age.
an additional record, such as a hand-and-wrist No significant differences were noted in pretreat-
radiograph, should be used to measure skeletal age. ment skeletal structures between PG and MG for most
Therefore, because dental age has been reported to variables, except mandibular body-to-anterior cranial

Table 4. Cephalometric Comparisons of Skeletal Changes between the Primary Dentition Group (PG) and the Mixed Dentition Group (MG)
during Treatment (T1-T0), the Observation Period (T2-T1), and the Total Period (T2-T0)
T1-T0 T1 T2
PG (n 5 26) MG (n 5 23) PG (n 5 26) MG (n 5 23) PG (n 5 26) MG (n 5 23)
Variables Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD)
SNA 2.24 (1.34) 2.07 (1.77) 20.17 (1.29) 0.69 (1.86) 2.07 (1.38) 2.75 (1.38)
SNB 21.75 (1.33)* 20.78 (1.59)* 0.92 (1.28) 0.72 (1.98) 20.83 (1.68) 20.06 (2.32)
ANB 4.00 (1.80)* 2.85 (1.38)* 21.10 (1.39)* 20.03 (1.81)* 2.90 (1.58) 2.82 (1.58)
Saddle angle 0.66 (1.49) 20.02 (2.54) 0.25 (2.19) 20.28 (2.46) 0.91 (2.09) 20.29 (2.33)
Articular angle 2.03 (3.29) 1.64 (3.46) 20.64 (4.57) 20.66 (3.26) 1.39 (3.37) 0.98 (2.81)
Gonial angle 20.89 (2.09) 20.43 (1.65) 20.64 (2.86) 0.58 (2.07) 21.53 (2.80) 0.14 (2.54)
Sum 1.81 (1.62) 1.19 (1.61) 21.03 (1.38) 20.36 (1.80) 0.77 (1.79) 0.83 (1.73)
SN-GoGn 1.81 (1.63) 1.21 (1.62) 21.03 (1.37) 20.38 (1.79) 0.78 (1.79) 0.83 (1.73)
Facial plane angle 21.54 (1.29) 20.94 (1.60) 1.00 (1.27) 0.79 (1.99) 20.55 (1.57) 20.15 (2.22)
Palatal plane angle 20.71 (2.00) 20.36 (1.82) 0.81 (1.97) 0.70 (1.60) 0.09 (2.47) 0.33 (1.97)
Mandibular plane angle 1.58 (1.69) 1.59 (1.58) 20.45 (1.79) 20.73 (1.57) 1.12 (2.07) 0.85 (2.15)
Mandibular body length-to-anterior cranial base
ratio (%) 1.58 (2.58) 2.02 (4.20) 0.93 (1.61) 0.96 (5.31) 3.49 (3.66) 2.98 (3.49)
Facial height ratio (%) 20.80 (1.47) 20.44 (1.09) 0.91 (1.40) 0.72 (1.59) 0.11 (1.62) 0.27 (1.28)
ANS-Me/Nasion-Me (%) 0.78 (1.16) 0.88 (1.10) 20.55 (0.86) 20.74 (0.94) 0.22 (1.31) 0.13 (1.22)
* P , .05; ** P , .01.
base ratio (Table 3). This indicates that at T0, the between the two groups. The change in ANB during
patients in MG showed a relatively larger mandibular the observation period was greater in the PG than in
body length than did those in PG. The difference in the MG (Table 4). The ANB difference between the
mandibular body length may be due to the fact that two groups came mostly from the change in SNA
before puberty, the mandible grows at a steadier rate during the observation period, which indicates an
than does the anterior cranial base.20 increased maxillary protraction relapse tendency in
The skeletal Class III malocclusions in the two PG compared with that in MG during this period. It
groups were improved by advancement of the maxilla seems that maxillary protraction induced by face
and backward movement of the mandible (Table 3). masks without RME may be less stable in very young
Statistical analyses showed the differences in treat- patients.
ment effects between different stages of dental Table 4 shows that no significant difference in
development (Table 4). The magnitudes of maxillary skeletal changes was seen between the two groups
protraction induced by face mask therapy were similar over the total time period. The insignificant differences
between the two groups, but changes in mandibular in skeletal changes may contribute to the similar
position were greater in PG than they were in MG. The skeletal patterns noted between the two groups at T2
different treatment response between PG and MG is (Table 3). This can be explained by the fact that the
consistent with significant differences in ANB at T1 relapse tendency during the observation period was
(Table 3). These results are in part consistent with greater in PG than in MG, although face masks
previous studies, which showed that the magnitudes of induced more favorable corrections of the skeletal
skeletal changes in the deciduous dentition group were Class III pattern in PG. This means that treatment
greater than in the early mixed dentition group.8,12 efficiency could not be guaranteed, although the
However, previous studies also showed that both effectiveness of face mask treatment without RME
maxillary advancement and mandibular relocation during the primary dentition could be accepted.
were greater in early treatment groups than in late When we reclassified patients using the stages in
treatment groups. This may be due to the effects of cervical vertebral maturation (CVMS),23 all patients in
RME on maxillary protraction, as RME can promote PG showed CVMS1. In cases of MG, about half of the
the effects of maxillary protraction by increasing patients showed CVMS1 and the remaining showed
sutural activity around the maxilla.21 In addition, CVMS2. When we reanalyzed skeletal patterns of the
Delinger21 has reported that maxillary expansion alone early treatment group (face mask therapy was started
can produce forward movement of the maxilla. in CVMS1) and the late treatment group (treatment
After correction of the Class III malocclusion by face was started in CVMS2), the differences in skeletal
mask treatment, a relapse tendency was noted during patterns between early and late treatment groups were
the observation period (Tables 3 and 4), with a not significantly different from those observed between
significant difference in relapse patterns observed PG and MG at T0, T1, and T2 (data not shown). In

addition, results of treatment efficiencies were similar 4. Sugawara J, Asano T, Endo N, Mitani H. Long-term effects of
between dental and skeletal groupings. Generally, the chincup therapy on skeletal profile in mandibular prognathism.
early treatment group (CVMS1) showed a more 5. Sugawara J, Mitani H. Facial growth of skeletal Class III
effective orthopedic response in the mandible during malocclusion and the effects, limitations, and long-term
T1-T0, and the late treatment group (CVMS2) main- dentofacial adaptations to chin cup therapy. Semin Orthod.
tained a more stable result of maxillary protraction 1997;3:244–254.
during T2-T1. 6. Deguchi T, Kitsugi A. Stability of changes associated with
chin cup treatment. Angle Orthod. 1996;66:139–146.
In this study, the observation period was about 1.5
7. Kim SC, Lee KS. The cephalometric study of facial types in
years. Fixed appliances were usually needed to move Class III malocclusion. Korean J Orthod. 1990;20:519–539.
each tooth more precisely after maxillary protraction. 8. Takada K, Petdachai S, Sakuda M. Changes in dentofacial
Individual tooth movement with Class III elastics during morphology in skeletal Class III children treated by a
fixed appliance therapy can influence skeletal mor- modified maxillary protraction headgear and chin cup: a
phology during the retention period. In addition, some longitudinal cephalometric appraisal. Eur J Orthod. 1993;15:
211–221.
patients should wear a functional appliance, a sagittal 9. Baccetti T, McGil JS, Franchi L, McNamara JA Jr. Skeletal
appliance, or a chin cup to maintain treatment results. effects of early treatment of Class III malocclusion with
Therefore, it was difficult to collect matched samples maxillary expansion and face mask therapy. Am J Orthod
with sufficient observation time. However, most pa- Dentofacial Orthop. 1998;113:333–343.
tients in MG passed through the peak in skeletal 10. Kapust AJ, Sinclair PM, Turley PK. Cephalometric effects of
face mask/expansion therapy in Class III children: a
growth at T2 based on CVMS, and the patients in PG
comparison of three age groups. Am J Orthod Dentofacial
showed stable results until the postpubertal period. Orthop. 1998;113:204–212.
Additional studies using long-term observation data 11. Franchi L, Baccetti T, McNamara JA Jr. Postpubertal
and a control group are needed to better understand assessment of treatment timing for maxillary expansion
the clinical implications of the timing of face mask and protraction therapy followed by fixed appliances.
therapy in growing Class III patients. Am J Orthod Dentofacial Orthop. 2004;126:555–568.
12. Kajiyama K, Murakami T, Suzuki A. Comparison of
orthodontic and orthopedic effects of a modified maxillary
CONCLUSIONS protractor between deciduous and early mixed dentition.
N The primary dentition group showed not only a more 13. Baik HS. Clinical results of the maxillary protraction in
effective response to orthopedic correction during Korean children. Am J Orthod Dentofacial Orthop. 1995;
the treatment period but also a higher relapse 108:583–592.
tendency than did the early to mid mixed dentition 14. Saadia M, Torres E. Sagittal changes after maxillary
group after active treatment. protraction with expansion in Class III patients in the
primary, mixed, and late mixed dentition: a longitudinal
N Similar skeletal effects can be obtained when maxillary
retrospective study. Am J Orthod Dentofacial Orthop. 2000;
protraction is initiated before eruption of the first 117:669–680.
permanent molar (Hellman’s developmental stages 15. Hellman M. Development of face and dentition in its
IIA-IIC) or after complete eruption of the first perma- application to orthodontic treatment. Am J Orthod. 1940;
nent molar (Hellman’s developmental stages IIIA-IIIB). 26:424–447.
N Face mask therapy without RME can be postponed 16. Verdon P. Professor Delaire Facial Orthopedic Mask.
Denver, Colo: Rocky Mountain Orthodontic Products; 1992.
to the early mixed dentition period on the basis of 17. Deguchi T, Kanomi R, Ashizawa Y, Rosenstein SW. Very
results of this study. early face mask therapy in Class III children. Angle Orthod.
1999;69:349–355.
18. Marshall D. Radiographic correlation of hand, wrist, and
ACKNOWLEDGMENT tooth development. Dent Radiogr Photogr. 1976;49:51–72.
This study was supported by the Korea University Fund. 19. Krailassirl S, Anuwongnukroh N, Dechkunakorn S. Rela-
tionships between dental calcification stages and skeletal
maturity indicators in Thai individuals. Angle Orthod. 2002;
REFERENCES 72:155–166.
1. Proffit WR, Fields HW Jr. Contemporary Orthodontics. St 20. Haas AJ. Palatal expansion: just the beginning of dentofa-
Louis, Mo: Mosby; 2000:270–518. cial orthopedics. Am J Orthod. 1970;57:219–225.
2. Yang WS. The study on the orthodontic patients who visited 21. Delinger EL. A preliminary study of anterior maxillary
department of orthodontics, Seoul National University displacement. Am J Orthod. 1973;63:509–516.
Hospital during last 10 years (1985–1994). Korean J Orthod. 22. Riolo ML, Moyers RE, McNamara JA Jr, Hunter WS. An
1995;25:497–509. Atlas of Craniofacial Growth. Monograph No. 2, Craniofacial
3. Deguchi T, Kuroda T, Minoshima Y, Graber TM. Craniofacial Growth Series, Center for Human Growth and Development.
features of patients with Class III abnormalities: growth- Ann Arbor, MI: University of Michigan; 1974:121–171.
related changes and effects of short-term and long-term chin 23. O’Reilly M, Yanniello GJ. Mandibular growth changes and
cup therapy. Am J Orthod Dentofacial Orthop. 2002;121: maturation of cervical vertebrae—a longitudinal cephalo-
84–92. metric study. Angle Orthod. 1988;58:179–184.

Original Article
Stability of surgically assisted rapid maxillary expansion and

orthopedic maxillary expansion after 3 years’ follow-up
Gökmen Kurta; Ayşe Tuba Altug-Ataçb; Mustafa Sancar Ataçc; Hakan Alpay Karasud
ABSTRACT
Objective: To evaluate the stability of surgically assisted rapid palatal expansion (SARME) and
orthopedic maxillary expansion (OME) after 3 years of follow-up, and compare these changes with
a control group.
Materials and Methods: The subjects of the study were divided into three groups. Group 1 was
composed of 10 patients (6 males, 4 females) with a mean age of 15.51 years (range: 13.33–17.58
years) and treated with OME, Group 2 comprised 10 patients (7 males, 3 females) with a mean age
of 19.01 years (range: 16.25–25.58 years) and treated with SARME. Group 3 was the control
group, consisting of 10 untreated, skeletal Class 1 subjects (6 males, 4 females) with a mean age
of 15.27 years (range: 13.42–17.00 years) and matched to the OME group for sex and age. Lateral
cephalometric and posteroantererior films were taken before expansion (T1), postexpansion (T2),
and 3 years after the retention period (T3).
Results: After OME and SARME, significant increases were observed for both dental and skeletal
transverse widths (P , .01). After 3 years of follow-up, maxillary basal width decreased 1.35 6
0.44 mm in the SARME group and 1.19 6 0.41 mm in the OME group, while upper molar width
decreased 2.23 6 1.24 mm in the SARME group and 2.79 6 1.01 mm in the OME group.
Conclusions: Both the OME and SARME procedures remained stable after 3 years of follow-up
with some amount of postretention relapse, compared with the control group. (Angle Orthod.
2010;80:613–619.)
KEY WORDS: Expansion; OME; SARME; Stability
INTRODUCTION and without maxillary constriction, skeletal or pseudo

Class III patients, cleft lip and palate patients, and
The origin of a constricted maxilla can be skeletal,
patients exhibiting moderate maxillary crowding to gain
dental, or combination of both structures. Angell1 first
arch length.2
described maxillary expansion in 1860, and since then
Krebs6 showed that, with advancing age, more force
it has been used for the treatment of both skeletal and is required, more dental tipping occurs, and less
dental posterior cross bites. Rapid maxillary expansion skeletal expansion can be achieved. Bishara and
has been used for unilateral and bilateral posterior Staley2 stated that the optimal age for expansion is
crossbites,2–5 skeletal Class II division 1, patients with before 13 to 15 years of age and that expansion in
older patients can yield unpredictable and unstable
a
Assistant Professor, Department of Orthodontics, School of results. Vanarsdall7 suggested that, as sutural closure
Dentistry, Erciyes University, Kayseri, Turkey. ends, maxillary expansion is generally unsuccessful
b
Associate Professor, Department of Orthodontics, School of
because of alveolar or dental tipping with little or no
Dentistry, Ankara, Turkey.
c
Assistant Professor, Department of Oral and Maxillofacial basal skeletal movement. Complications such as
Surgery, Gazi University, Ankara University, Ankara, Turkey. severe pain, pressure necrosis of soft tissue, tipping
d
Associate Professor, Department of Oral and Maxillofacial and extrusion of maxillary teeth, bending of alveolar
Surgery, Ankara University, Ankara, Turkey. bone, uncontrolled relapse, and periodontal complica-
Corresponding author: Dr Gökmen Kurt, Erciyes University,
tions can also be observed in mature adolescent and
School of Dentistry, Department of Orthodontics, Kayseri, 38039
Turkey adult patients.7,8
(e-mail: gokmenkurt@hotmail.com) Many reasons have been suggested for the limita-
Accepted: October 2009. Submitted: September 2009. tion of orthopedic maxillary expansion (OME) in
G 2010 by The EH Angle Education and Research Foundation, skeletally mature patients. However, few reports have
Inc. shown successful findings after nonsurgical maxillary

614 KURT, ALTUG-ATAÇ, ATAÇ, KARASU
Table 1. Mean (D), Standard Deviations (Sd), and Minimum (Min) completed 99.2% (minimum: 98.61%, maximum:
and Maximum (Max) Values of Subjects’ Ages in the SARMEa, 99.8%) of their growth potential. The pretreatment
OMEb, and Control Groups ANB was –.98u 6 1.30u, indicating a skeletal Class III
Chronological Age (year) pattern (Tables 1 and 2).
Group N D Sd MIN MAX
SARME Group
SARME 10 R: 7 =: 3 19.01 1.22 16.25 25.58
OME 10 R: 6 =: 4 15.51 1.09 13.33 17.58 The SARME group was composed of 10 patients (7
Control 10 R: 6 =: 4 15.27 1.43 13.42 17.00
males, 3 females) with a mean age of 19.01 years
a
SARME indicates surgically assisted rapid palatal expansion. (range: 16.25–25.58 years) and a minimum skeletal age
b
OME indicates orthopedic maxillary expansion.
of 17 years, according to the Greulich and Pyle hand-
wrist analysis.22 Of these 10 patients, 6 patients were
expansion in adults, as shown in children.9–11 Surgical older than 17 years and 4 patients were 17 or younger.
procedures such as LeFort osteotomy for segmenting These four patients were initially treated by OME, but
the maxilla to widen it in the transverse dimension and their expansion procedure was continued with surgical
surgically assisted rapid maxillary expansion (SARME) assistance owing to discomfort, pain, or resistance to
have been suggested to eliminate possible complica- expansion. The patients in the SARME group had
tions after OME in adults.9 SARME has similar completed 99.51% (minimum: 99.1%, maximum:
indications for use with conventional expansion, such 100%) of their growth potential before treatment.22 Their
as maxillomandibular deficiency of more than 5 mm or pretreatment ANB was 0.08u 6 0.67u, indicating a
mild crowding, or to facilitate later treatment of skeletal Class III tendency (Table 2).
anteroposterior discrepancies.2,5,12–14
A number of studies have evaluated the long-term Control Group
stability of SARME using dental cast measurements for
the most part12,15–18 which, therefore, show only dental The control group consisted of 10 untreated, skeletal
changes. Berger et al.19 and Byloff and Mossaz20 used Class 1 subjects (6 males, 4 females) matched to the
PA films to measure transverse skeletal changes and OME group for sex and age (Table 1) in order to
reported on a 1-year follow-up after SARME. Byloff and assess the effects of normal skeletal growth changes
Mossaz20 evaluated the skeletal changes only in a over a 3-year follow-up period.
SARME group, and Berger et al.19 compared the SARME
group with an OME group, but did not use a control Surgical Procedure
group. Therefore, the aim of our study was to evaluate
Surgery was performed under local anesthesia.
the stability of SARME and OME after 3 years of follow-
Bilateral incisions were made at the depth of the
up, and compare these changes with a control group.
vestibule from the first molar area to the distal aspect
of the lateral incisor. The mucoperiosteum was
elevated, and the maxillary bone exposed from the
The study groups consisted of 20 patients. Ten piriform aperture to the pterygomaxillary fissure. After
additional subjects were randomly selected from the identifying the infraorbital nerve, an osteotomy was
longitudinal archive of Ankara University, Faculty of performed horizontally from the piriform aperture to the
Dentistry and the Department of Orthodontics to serve pterygomaxillary fissure well above the tooth apices.
as the control group. The distribution of the sample is The pterygoid plates were not separated from the
presented in Table 1. Inclusion criteria for the maxillary maxilla. An additional vertical incision was made
expansion groups were parallel to the labial frenulum, and the maxilla was
separated by malleting a thin osteotome between the
—the presence of at least 5 mm of bilateral posterior
central incisors at a level below the anterior nasal
skeletal cross bite;
spine. The surgical sites were irrigated and sutured. An
—no future orthognathic surgery required;
anterior nasal pack and pressure bandage were
—no pre-expansion orthodontics performed.21
applied for 24 hours; and antibiotics, analgesics, and
an oroantral regime were prescribed.21
OME Group
Treatment Protocol
The OME group was composed of 10 patients (6
males, 4 females) with a mean age of 15.51 years All patients were treated with occlusal-coverage,
(range: 13.33–17.58 years) and a minimum skeletal Hyrax-type expanders (Figure 1). In the OME group,
age of 15 years, according to the Greulich and Pyle the screws were activated immediately after bonding
hand-wrist analysis.22 Patients in the OME group had and in the SARME group after surgery. The activation

STABILITY OF SARME AND OME 615
Table 2. Mean (D) and Standard Deviations (Sd) of the Cephalometric Values of Subjects Before Treatment in the SARME, OME, and
Control Groups
SARME OME Control
Parameter D 6Sd D 6Sd D 6Sd
Lateral Cephalometric Measurements
Maxilla
1. SNA deg 75.28 0.94 74.57 0.96 8.020 3.21
2. N-ANS mm 57.50 0.97 60.08 1.94 58.63 1.01
3. SN/Palatal plane deg 10.99 0.94 11.83 1.11 11.04 1.23
Mandible
4. SNB deg 75.20 1.32 77.56 1.07 77.55 3.31
Maxillomandibular relations
5. ANB deg 0.08 0.67 22.98 1.30 2.65 0.75
6. SN/GoMe deg 44.09 1.91 43.77 1.14 36.80 2.62
7. Overjet mm 1.92 0.82 21.82 1.67 2.05 0.86
8. Overbite mm 21.61 1.48 1.15 0.52 2.10 1.39
Posteroanterior Cephalometric Measurements
9. MXR-MXL deg 66.79 2.16 62.80 0.56 70.83 1.58
10. UMOLR-UMOLL deg 54.88 2.38 55.04 0.93 61.23 1.74
11. LMOLR-LMOLL mm 61.60 1.46 58.62 1.46 60.88 1.24
12. MxR/cg/mXl mm 50.61 1.24 48.93 1.41 52.71 2.01
protocol for both expansion groups was two turns a were also performed to analyze changes within the
day (0.25 mm per turn). In all 20 patients, the opening observation periods.
of the midpalatal suture was followed using occlusal
radiographs. After the desired expansion was Error Study
achieved, the expander was kept on the teeth as a
passive retainer for an average of about 90 days. All cephalograms were retraced 1 month later by the
Fixed appliance treatment was not initiated until after same examiner. No significant differences were found,
the postexpansion radiographs were taken so as not to and the reliability coefficients (r) ranged between 0.91
affect the dentoalveolar measurements. Immediately and 0.96.
after the expander was removed, fixed straightwire
appliances were placed, and transpalatal arches were RESULTS
used for the remainder of the conventional orthodontic OME-SARME-Control Group Comparison at T1–T2
treatment period. At the end of fixed orthodontic
treatment, the transpalatal arches were removed and Sagittal measurements. The maxilla showed poste-
a Hawley plate was used full time for 6 months and rior rotation in the SARME group (SN/palatal plane),
only at night for 6 months for a total of 12 months. The and anterior rotation in the OME group, and the
treatment then was finished. difference was significant (P , .01). The difference
between the OME and the control group was also
Cephalometric Measurements significant (P , .01) (Table 3). The most anterior
displacement of the maxilla (SNA) was measured in
Lateral and posteroanterior cephalograms and the OME group (2.18u 6 1.60u) (P , .05); this
hand-wrist films were obtained for each patient at displacement was significantly different from the
preexpansion (T1), postexpansion (T2) and after a SARME (P , .05) and control groups (P , .05).
time period including fixed appliance treatment, a The SARME group exhibited significant vertical
retention period of 12 months, and 3 years without downward maxillary displacement (N-ANS) (P , .01),
retention (T3). Eight lateral cephalometric (Figure 2) and the OME group showed vertical upward move-
and four posteroanterior (Figure 3) measurements ment, which may explain the maxillary rotations in both
were performed at T1, T2, and T3.
groups. The differences among the three groups were
significant (Table 3).
Statistics
Bite opening (SN/GoMe) was observed both in the
Analysis of variance and Duncan’s tests were used, SARME (P , .05) and the OME (P . .05) groups, and
respectively, to compare the cephalometric measure- the difference was significant between the SARME and
ments of patients at T1, T2, and T3. Paired t-tests the control groups (P , .01).

Figure 1. Intraoral photograph of the occlusal-coverage, Hyrax-type expander.
Significant decrease in SNB (P , .05) in the SARME differences in the SARME and control groups in ANB
group and insignificant increases were measured in were insignificant. The significant increase in the ANB
both the OME and control groups. The ANB angle angle caused a significant increase in the overjet
showed a significant increase in the OME group (P , measurement in the OME group (P , .05).
.05) because of an increase in the SNB angle, and the Transverse measurements. The maxillary width
(MxR-MxL) increased significantly in both the SARME
and OME groups (P , .01), and both expansion
groups showed significant differences compared with
the control group (P , .01). The maxillary posterior
segment exhibited significant transverse tipping (MxR/
cg/MxL) in both the SARME and OME groups
compared with the control group (P , .01). The most
tipping (P , .01) was observed in the SARME group
(3.28u 6 0.75u) (Table 3).
The upper first molar width (UmolR-UmolL) showed
significant increases in both treatment groups (P ,
.01), and the differences between the control group are
significant (P , .001). Lower molar width (LmolR-
LmolL) increased significantly in the OME group (P ,
.05).
OME-SARME-Control Group Comparison at T2–T3

Sagittal measurements. After 3 years of follow-up,
most of the lateral cephalometric measurements
showed similar changes among the three groups.
The SN/palatal plane angle showed a significant
increase in the OME group (2.12u 6 1.21u) (P ,
.01), and a comparison of the SARME and OME
groups with the control group displayed significant
differences (P , .05) (Table 4).
Opening of the bite continued in both expansion
groups, and the differences between the SARME and
OME groups and the control group were significant (P
, .05). Overbite changes in the expansion groups
were significant compared with the control group after
Figure 2. Lateral cephalometric measurements: SNA, SNB, ANB, N-
3 years, which might be due to compensation for the
ANS, SN/palatal plane, SN/mandibular plane, in degrees; overjet bite opening during fixed orthodontic treatment after
and overbite, in mm. rapid maxillary expansion.

The SARME and OME groups exhibited decreases

in upper molar widths, and the changes in the OME
group were significant (P , .05). The differences
between the OME group and the SARME group were
significant (P , .05). Also, the expansion groups
showed significant changes when compared with the
control group (P , .001).
Lower molar width (LmolR-LmolL) also relapsed
significantly in the OME group (P , .05), and
insignificantly in the SARME group; the differences
among the three groups were statistically significant (P
, .05) (Table 4).
DISCUSSION
PA films were used to evaluate the transverse
Figure 3. Posteroanterior cephalometric measurements: basal skeletal changes pre-expansion, postexpansion, and
maxillary width (MxR-MxL), maxillary dentoalveolar width (UmolR-
after 3 years of follow-up in our study. Betts et al.14
UmolL), in millimeters; mandibular dentoalveolar width (LmolR-
LmolL), angles between crista galli and maxillary base points (MxR/ suggested that posteroanterior cephalograms are
cg/MxL). available and reliable tools to identify and evaluate
transverse skeletal discrepancies between the maxilla
and the mandible. Lateral cephalometric films were
Transverse measurements. The achieved increase also used to measure sagittal changes during the
in maxillary bony width (MxR-MxL) was reduced follow-up periods. Studies assessing the stability of the
insignificantly within both expansion groups, and the SARME procedure mostly used dental measurements
amount of relapse was also not statistically significant to evaluate the long-term changes.12,15–18 Few studies
(Table 4), but the differences between the expansion used PA films for evaluating long-term skeletal
groups and the control group were significant (P , .05). changes,19,20 so the number of studies showing
Relapse of the transverse tipping of maxillary halves skeletal stability of SARME is limited.
was insignificant and similar in both expansion groups The mean ANB in the OME group was 22.98u 6
(MxR/cg/MxL), and the control group showed significant 1.30u, indicating a skeletal Class III pattern, and in the
changes compared with the expansion groups (P , .01). SARME group, 0.08u 6 0.67u, indicating a skeletal
Table 3. Comparison of SARME, OME, and Control Groups Between T2 and T1 by Analysis of Variance (ANOVA) and Duncan Tests
SARME OME Control
SARME vs OME vs SARME vs
Parameters D 6Sd D 6Sd D 6Sd Test OME Control Control
Maxilla
1. SNA deg 0.18 0.36 2.18* 1.60 0.56 0.67 ** * *
2. N-ANS mm 1.21** 0.28 20.22 0.31 0.84 1.77 ** ** * **
3. SN/Palatal Plane deg 0.65 0.33 21.09* 0.44 1.01 0.42 ** ** **
Mandible
4. SNB deg 20.93* 0.34 0.52 0.34 0.67 1.38 * * *
Maxillomandibular
relations
5. ANB deg 1.11 0.59 1.66* 0.50 20.11 0.77 ** * ** **
6. SN/GoMe deg 0.90* 0.34 0.15 0.60 20.79 1.84 ** **
7. Overjet mm 0.26 0.42 1.64* 0.56 20.38 0.72 ** **
8. Overbite mm 1.51 1.07 0.42 0.60 20.14 1.12
9. MxR-MxL mm 2.45** 0.52 2.22** 0.51 0.36 0.16 *** *** ***
10. UmolR-UmolL mm 7.81** 1.01 7.38** 1.01 1.32 0.44 *** *** ***
11. LmolR-LmolL mm 0.27 0.43 1.61** 0.37 0.75 0.33 * * * *
12. MxR/cg/MxL deg 3.28** 0.75 1.08** 0.20 0.69 0.26 *** * *** ***
* P , .05; ** P , .01; *** P , .001; D, mean of the differences; Sd, standard deviation of mean of the differences.

Table 4. Comparison of SARME, OME, and Control Groups Between T3 and T2, by Analysis of Variance (ANOVA) and Duncan Tests
SARME OME Control
SARME OME vs SARME vs
Parameter D 6Sd D 6Sd D 6Sd Test vs OME Control Control
Maxilla
1. SNA deg 0.13 0.23 0.20 0.10 20.12 0.11
2. N-ANS mm 0.17 0.13 0.11 0.14 0.13 0.07
3. SN/Palatal deg 1.15 1.18 2.12* 1.21 20.81 0.32 * * * *
plane
Mandible
4. SNB deg 0.29 0.21 0.78 0.22 0.17 0.11
Maxillomandibular
relations
5. ANB deg 20.11 0.17 20.24 0.09 20.21 0.35
6. SN/GoMe deg 0.63 0.28 0.35 0.44 20.39 0.31 * * *
7. Overjet mm 0.15 0.13 0.50 0.35 0.28 0.32
8. Overbite mm 0.29 0.27 0.55 0.21 20.23 0.20 * * *
9. MxR-MxL mm 21.35 0.44 21.19 0.41 0.15 0.07 ** ** **
10. UmolR-UmolL mm 22.23 1.24 22.79* 1.01 1.08 0.24 *** * *** ***
11. LmolR-LmolL mm 20.07 0.03 20.82* 0.32 0.23 0.11 * * * *
12. MxR/cg/MxL deg 22.17 0.34 22.08 0.30 0.48 0.36 ** ** **
* P , .05; ** P , .01; *** P , .001; D, Mean of the differences; Sd, standard deviation of mean of the differences.
Class III tendency. The control group comprised expansion. Insignificant sagittal maxillary displacement
skeletal Class I cases and did not match with the in the SARME group could be due to minimal or no
expansion groups by means of skeletal features, for growth potential of the patients. Byloff and Mossaz20
ethical reasons. concluded that maxillary skeletal expansion with the
Many authors accept patient age in choosing SARME is mainly a lateral rotation of the two maxillary
between the OME and SARME.9 However, conflicting halves with minimum horizontal translation, which
suggestions are found in the literature regarding when supports our findings. All three groups showed insignif-
the OME or SARME should be used for achieving icant maxillary movements in the long term.
successful skeletal expansion. Timms and Vero23 Similar transverse maxillary skeletal increase was
accepted 25 years as an upper limit for applying found in the SARME and OME groups after expansion
OME. Epker and Wolford24 recommended SARME in (2.45 6 0.52 mm and 2.22 6 0.51 mm, respectively).
patients over 16 years of age. Mommaerts25 stated that In the long-term follow-up, both expansion groups
OME is indicated for patients younger than 12 years exhibited 50% of skeletal maxillary relapse, and these
and, for those over 14 years, corticotomy-assisted transverse decreases were significant compared with
expansion is needed. Some of the patients in the OME the control group. The main maxillary skeletal trans-
group were adults or young adults, but they still had verse difference between expansion groups was
growth potential, so successful expansion was maxillary transverse rotation. The SARME group
achieved in these patients. Results in the OME group showed significant lateral rotation of the maxillary
support Suri and Taneja,9 who stated that successful halves compared with the OME group (3.28u 6 0.75u
OME can be achieved in chronologically advanced— and 1.08u 6 0.20u, respectively), and this finding
but skeletally immature—patients. On the other hand, supports the idea that skeletal expansion with SARME
OME might be unsuccessful in chronologically younger is mainly a lateral rotation of the two maxillary halves.20
patients with advanced skeletal maturity. The decrease in the transverse angular measurement
Statistically significant forward displacement of was insignificant after 3 years in both expansion
the maxilla was observed only in the OME group. groups. Although some relapse was observed in the
Altug-Ataç et al.21 stated that this forward displacement expansion groups, the total amount of skeletal
could be explained by the occlusal coverage of the transverse changes in both expansion groups exceed-
expanders, which helped unlock the occlusion and set ed that in the control group in the long term.
the maxilla free in the OME patients who had a negative Different relapse rates have been reported for
ANB value initially. Similar maxillary movement was SARME in the literature, from 5% to about
observed both in the SARME and control groups after 25%,9,18,19,25; reported relapse rates for OME is much

higher, and can be as high as 63%.2,9,26,27 Both 5. Wertz RA. Skeletal and dental changes accompanying rapid
groups exhibited 50% decrease of skeletal maxillary midpalatal suture opening. Am J Orthod. 1970;58:41–66.
6. Krebs A. Mid–palatal expansion studied by the implant
transverse dimension after 3 years. Also in our study, method over a seven year period. Trans Eur Orthod Soc.
the pterygoid plates were not separated from the 1964:131–142.
maxilla and no midpalatal osteotomy was performed. 7. Vanarsdall RL. Transverse dimension and long-term stabil-
Bays and Greco18 and Northway and Meade17 sug- ity. Semin Orthod. 1999;5:171–180.
gested not separating the maxilla from the pterygoid 8. Graber TM, Vanarsdall RL, eds. Orthodontics: Current
Principles and Techniques. 2nd ed. St Louis, Mo: Mosby;
plates to avoid invading the pterygomaxillary junction. 1994:715–721.
The authors stated that such a separation requires 9. Suri L, Taneja P. Surgically assisted rapid palatal expansion:
extreme force and usually causes the plates to a literature review. Am J Orthod Dentofacial Orthop. 2008;
fracture.17 133:290–302.
The SARME and OME groups showed similar 10. Handelman CS. Nonsurgical rapid maxillary alveolar expan-
sion in adults: a clinical evaluation. Angle Orthod. 1997;67:
maxillary molar width increases after expansion (7.81 291–305.
6 1.01 mm and 7.38 6 1.01 mm, respectively). Both 11. Handelman CS, Wang L, BeGole EA, Haas AJ. Nonsurgical
expansion groups showed 30% decreases in trans- rapid maxillary expansion in adults: report on 47 cases using
verse molar width after a follow-up period; changes in the Haas expander. Angle Orthod. 2000;70:129–144.
the OME group were significant. Magnusson et al.12 12. Magnusson A, Bjerklin K, Nilsson P, Marcusson A.
Surgically assisted rapid maxillary expansion: long-term
concluded that relapse is most pronounced during the stability. Eur J Orthod. 2009;31:142–149.
first 3 years after treatment and suggested retention 13. Haas AJ. Long-term posttreatment evaluation of rapid
during this period. Byloff and Mossaz20 found a one- palatal expansion. Angle Orthod. 1980;50:189–217.
third decrease in lateral tipping of the molars during the 14. Betts NJ, Vanarsdall RL, Barber HD, Higgins-Barber K,
retention period. Magnusson et al.12 concluded that Fonseca RJ. Diagnosis and treatment of transverse
maxillary deficiency. Int J Adult Orthod Orthognath Surg.
pterygoid detachment could not fully eliminate the 1995;10:75–96.
posterior resistance and that buccal tipping of the 15. Anttila A, Finne K, Keski–Nisula K, Somppi M, Panula K,
molars can still be observed. Wertz5 stated that flaring Peltomaki T. Feasibility and long–term stability of surgically
or tipping of the maxillary molars was a demonstrable assisted rapid maxillary expansion with lateral osteotomy.
and expected response to expansion. Although dental Eur J Orthod. 2004;26:391–395.
16. Sokucu O, Kosger HH, Bicakci AA, Babacan H. Stability in
relapse in the OME group was statistically significant,
dental changes in RME and SARME: a 2-year follow-up.
both expansion groups maintained most of their Angle Orthod. 2009;79:207–213.
transverse dental width stability after 3 years com- 17. Northway WM, Meade JB. Surgically assisted rapid maxil-
pared with the control group. lary expansion: a comparison of technique, response and
stability. Angle Orthod. 1997;67:309–320.
18. Bays RA, Greco JM. Surgically assisted rapid palatal
CONCLUSIONS
expansion: an outpatient technique with long-term stability.
N Although invasive surgical protocols such as ptery- J Oral Maxillofac Surg. 1992;50:110–113.
19. Berger JF, Kulbersh VA, Borgula T, Kaczynski R. Stability of
goid detachment and palatal separation were not
orthopedic and surgically assisted rapid palatal expansion over
used, both skeletal and dental widths were stable time. Am J Orthod Dentofacial Orthop. 1998;114:638–645.
after 3 years’ follow-up, with some relapse. 20. Byloff FK, Mossaz CF. Skeletal and dental changes
N Overexpansion is suggested for both SARME and following surgically assisted rapid palatal expansion.
OME for more stable results. Eur J Orthod. 2004;26:403–409.
21. Atac ATA, Karasu HA, Atac D. Surgically assisted rapid
N Results of this study confirm the idea that expansion
maxillary expansion compared with orthopedic rapid maxil-
with SARME is mainly a lateral rotation of the two lary expansion. Angle Orthod. 2006;76:353–359.
maxillary halves. 22. Greulich WW, Pyle SI. Radiographic Atlas of Skeletal
N Increased follow-up periods may be more effective Development of Hand and Wrist. 2nd ed. Stanford, Calif:
for assessing long-term changes after expansion Stanford University Press; 1959.
23. Timms DJ, Vero D. The relationship of rapid maxillary
procedures.
expansion to surgery with special reference to midpalatal
synostosis. Br J Oral Surg. 1981;19:180–196.
24. Epker BN, Wolford LM. Transverse Maxillary Deficiency
REFERENCES
Dentofacial Deformities: Integrated Orthodontic and Surgi-
1. Angell EC. Treatment of irregularities of the permanent adult cal Correction. St Louis, Mo: Mosby; 1980.
teeth. Dent Cosmos. 1860;1:540–545. 25. Mommaerts MY. Transpalatal distraction as a method of maxil-
2. Bishara SE, Staley RN. Maxillary expansion: clinical implica- lary expansion. Br J Oral Maxillofac Surg. 1999;37:268–272.
tions. Am J Orthod Dentofacial Orthop. 1987;91:3–14. 26. Mew J. Long-term effect of rapid maxillary expansion.
3. Haas AJ. The treatment of maxillary deficiency by opening Eur J Orthod. 1993;15:543.
the mid-palatal suture. Angle Orthod. 1965;35:200–217. 27. Velazquez P, Benito E, Bravo LA. Rapid maxillary expan-
4. Haas AJ. Just the beginning of dentofacial orthopedics. sion: a study of the long-term effects. Am J Orthod
Am J Orthod. 1970;57:210–254. Dentofacial Orthop. 1996;109:361–367.

Original Article
Cytotoxic effects of orthodontic composites

Siddik Malkoca; Bayram Corekcib; Hayriye Esra Ulkerc; Muhammet Yalçınd; Abdülkadir Şengüne
ABSTRACT
Objectives: To evaluate the cytotoxic effects of five different light-cured orthodontic bonding
composites.
Materials and Methods: The orthodontic composites Heliosit Orthodontic (Ivoclar), Transbond XT
(3M Unitek), Bisco ORTHO (Bisco), Light Bond (Reliance), and Quick Cure (Reliance) were
prepared, and the samples were extracted in 3 mL of BME (Basal Medium Eagle) with 10%
newborn calf serum for 24 hours. The L929 cells were plated (25,000 cells/mL) in a 96-well dish
and maintained in a humidified incubator for 24 hours at 37uC, 5% CO2, and 95% air. After
24 hours of incubation of the cells, the incubation medium was replaced by the immersed medium
in which the samples were stored. Then, L929 cells were incubated in contact with eluates for
24 hours. The cell mitochondrial activity was evaluated by the methyl tetrazolium (MTT) test.
Twelve wells were used for each specimen, and the MTT tests were applied two times. The data
were statistically analyzed by one-way analysis of variance (ANOVA) and Tukey HSD tests.
Results: Results with L929 fibroblasts demonstrated that except for Transbond XT, freshly
prepared composite materials did not reduce vital cell numbers (P . .05) compared with the control
group. Our data demonstrate that Transbond XT showed significant cytotoxicity compared with the
control group.
Conclusion: Results indicate that tested orthodontic bonding composites are suitable for clinical
application, but that further studies using different test methods are needed for Transbond XT.
(Angle Orthod. 2010;80:759–764.)
KEY WORDS: Biocompatibility; Light-cured orthodontic composites; Cytotoxicity
INTRODUCTION cure on demand has caused an increasing number of

orthodontic practices to use light-cure adhesives
The introduction of the acid etch bonding technique
instead of the more traditional paste-paste adhesives
has led to dramatic changes in clinical practice, and
requiring in-office mixing.1 The light initiated resins
orthodontic bonding has been a critical issue. Today,
have become the most popular adhesives for a
orthodontists are approaching 40 years of successful,
majority of orthodontists. Photo-activated resin com-
reliable orthodontic bonding in clinics. The desire to
posites are the choice of adhesive for orthodontic
bonding because of their ease of use and the extended
a
Associate Professor and Department Chair, Department of
time they allow for bracket placement.2 A number of
Orthodontics, Faculty of Dentistry, Inönü University, Malatya,
Turkey. studies published during the 1990s have investigated
b
Research Assistant, Department of Orthodontics, Faculty of the bond strength of dual-cure and light-cure adhe-
Dentistry, Selçuk University, Konya, Turkey. sives of transparent ceramic brackets and metallic
c
Private Practice, Konya, Turkey. brackets. The majority of these studies concluded that
d
Assistant Professor, Department of Conservative Dentistry,
Faculty of Dentistry, Inönü University, Malatya, Turkey.
the light-cured material was not different from that for
e
Professor and Department Chair, Department of Conserva- the chemically cured adhesives, and even these
tive Dentistry, Faculty of Dentistry, Selçuk University, Konya, materials are found to have higher bond strength.3
Turkey. Although the developments and improvements of
Corresponding author: Dr Siddik Malkoc, Department of the orthodontic light cure adhesive materials are very
Orthodontics, Faculty of Dentistry, Inönü University, Malatya,
Malatya 44280, Turkey satisfying and amazing, biocompatibility of the ortho-
(e-mail: siddikmalkoc@yahoo.com) dontic composites is still a problem for orthodontists
Accepted: November 2009. Submitted: September 2009.
today. In addition, very few data are available in the
G 2010 by The EH Angle Education and Research Foundation, literature regarding biocompatibility of the commercial-
Inc. ly available orthodontic composites and the length of

760 MALKOC, COREKCI, ULKER, YALÇIN, ŞENGÜN
Table 1. The Adhesive Resins Included in the Investigation

Product Material Typea Manufacturer and Lot Number
Bisco ORTHO Bis-GMA (15%–40%), TEGDMA (5%–15%), fused silica (30%–60%) Bisco Inc, lot no: 0600004598
Heliosit Orthodontic UDMA, bis-GMA and decamethylen-DMA (85%), dispersed silicon dioxide Ivoclar Vivadent AG, lot no: G21064
(14%), catalysts and stabilizers (1%)
Light Bond UDMA, TEGDMA (22%), fused silica, sodium fluoride (78%) Reliance Orthodontic Products, lot no:
0609234
Quick Cure Silica-crystalline, fused silica (50%–90%), bis-GMA (1%–10%), Reliance Orthodontic Products, lot no:
TEGDMA (5%–10%), sodium fluoride (0.1%–2.0%) 0604936
Transbond XT Bis-GMA (5%–10%), bis-EMA (10%–20%), TEGDMA (5%–10%), 3M Unitek, lot no: 6CX6WM0088
silane-treated quartz (70%–80%), silane-treated silica (2%)
a
Bis-GMA indicates bisphenol A diglycidyl methacrylate; UDMA, urethane dimethacrylate; bis-EMA, bisphenol A bis (2-hydroxethyl ether)
dimethacrylate; TEGDMA, triethylene glycol dimethacrylate; decamethylen-DMA, decamethylen-dimethacrylate.
time after light polymerization.4 Orthodontists are using Cytotoxicity Testing

a large variety of bonding agents, and newer ortho-
L929 cells (ATCC CCL 1, Şap Enstitüsü, Ankara,
dontic composite materials present new challenges
Turkey) were cultured in BME (Basal Medium Eagle)
because of the potential for interaction.
containing 10% newborn calf serum and 100 mg/mL
No comprehensive data are available in the ortho-
penicillin/streptomycin at 37uC in a humidified atmo-
dontic literature regarding the toxicity of light-cured
sphere of 95% air, 5% CO2. Cell cultures between the
orthodontic resin composites. It should be critically
12 and 15 passages were used in this study. Confluent
emphasized at this point that the manufacturers of
cells were detached with 0.25% trypsin and seeded at
these materials possess comprehensive test data.
a density of 5 3 103 into each well of a 96-well plate for
Therefore, the aim of the present study was to
24 hours at 37uC and 5% CO2. After 24 hours of
evaluate the cytotoxic effects of five different light-
incubation, the culture medium was replaced with
cured orthodontic bonding composites.
200 mL of culture medium containing material extracts
of orthodontic composites. The original culture medium
served as control in this study.
Five different orthodontic composites were tested in Cultures were incubated for 24 hours at 37uC
the experiments: Heliosit Orthodontic (Ivoclar Vivadent and 5% CO2. The viability of cells exposed to
AG, Schaan, Liechtenstein), Transbond XT (3M Unitek material extracts was assessed using succinic dehy-
Ortho Prod, Monrovia, Calif), Bisco ORTHO (Bisco Inc, drogenase activity. The succinic dehydrogenase ac-
Schaumburg, Ill), Light Bond (Reliance Orthodontic tivity has been shown to be reasonably representative
Products Inc), and Quick Cure (Reliance, Itasca, III). of mitochondrial activity in the cells and reflects both
Their components and details are listed in Table 1. cell number and activity.5 The old medium was
Test specimens were prepared according to the removed and cell cultures were rinsed with PBS, and
manufacturers’ instructions in standard Teflon molds of 0.5-mL aliquots of freshly prepared MTT (3-[4,5-
5 mm in diameter and 2 mm in depth. All specimens dimethyl-thiazol-2-yl]-2,5-diphenyl-tetrazolium bro-
were prepared and handled under aseptic conditions mide) solution (0.5 mg/mL in BME) were added to
to limit the influence of biologic contamination on the each well.
cell culture tests. Specimens were prepared between After a 2-hour incubation period (37uC, 5% CO2), the
Mylor and glass slabs to minimize the oxygen inhibition supernatant was removed and the intracellularly stored
and maximize the surface smoothness. Specimens MTT formazan was solubilized in 200 mL dimethyl
that required light curing were cured using a standard sulfoxide for 30 minutes at room temperature. The
light curing unit (LED, Elipar FreeLight 2, 3M ESPE absorbance at 540 nm was spectrophotometrically
Dental Products, St Paul, Minn). measured.
Four samples were prepared for each group for The worksheets were incorporated into the software,
cytotoxicity testing. The samples were immersed in Excel version XP, and then recalculated as follows: cell
7 mL of culture medium for 24 hours at 37uC to extract viability percentage 5 (a{b)
(c{b) |100 where a is the OD
residual monomer or cytotoxic substances. The culture value at 540 nm derived from a well added with a test
medium containing material extracts was sterile filtered chemical, b is the mean OD value at 540 nm derived
for use on the cell cultures. from blank wells, and c is the mean OD value at

CYTOTOXIC EFFECTS OF ORTHODONTIC COMPOSITES 761
Table 2. The Cell Viability Percentages by Methyl Tetrazolium

(MTT) Assaya
Groups n Mean SD Tukey HSDb
Control 24 100.08 12.88 A
Bisco ORTHO 24 90.36 14.46 AB
Heliosit Orthodontic 24 91.94 11.58 AB
Light Bond 24 91.69 14.95 AB
Quick Cure 24 91.26 14.16 AB
Transbond XT 24 87.03 10.66 B
a
SD indicates standard deviation. Figure 2. Cultured L929 cells for Heliosit Orthodontic.
b
Means of the same letter are not significantly different at a 5 .05.
significantly decreased L929 cell survival rates com-
540 nm derived from control wells (ie, added culture pared to the control group (P , .05; Table 2).
medium as a test chemical).
Twelve replicate cell cultures were exposed to each Morphologic Assessment
concentration of a single material in at least two
independent experiments. Cell survival in treated L929 cells were elongated and spindle-shaped in
groups was compared with that in the untreated appearance. While Heliosit Orthodontic led to enlarge-
controls. Differences between median values were ment of the intercellular space, the cells kept their
statistically analyzed using the one-way analysis of shape as spindle in appearance. However, the cell
variance (ANOVA) and Tukey HSD tests. densities were decreased when compared with the
Cell survival of L929 cells was evaluated in a methyl control. For Bisco ORTHO, Quick Cure, and Light
tetrazolium test after exposure to orthodontic bracket Bond, the cells were retracted, rounded in appearance,
adhesive materials. Data are expressed as percentage and also led to enlargement of the intercellular space,
of the control cultures. Cell survival rates were but the cells did not show any differences with the
calculated from independent experimental cultures. control (Figures 1 through 6). However, in the Trans-
bond XT group the cells were significantly retracted,
Cell Morphology Evaluation rounded, and also increased in the intercellular space,
while there was no significantly different appearance in
Morphologic alteration of L929 cells was observed the Transbond XT group cell population when com-
directly using an inverted microscope (TS100 Nikon pared with the control.
Eclipse, Japan) (103) and photographed by a camera
(Nikon Eclipse, Tokyo, Japan). DISCUSSION
RESULTS It is very important to evaluate the biocompatibility of

the orthodontic bonding adhesives, because these
Means and standard deviations of cell survival rates materials are located proximate to the periodontal
for each group are given in Table 2. There were tissue and alveolar bone. Substances released from
significant differences among the orthodontic compos- orthodontic composites may cause a reaction (inflam-
ite resins in the cell survival percentage (P , .05). mation or necrosis) in adjacent tissues, such as the
Results demonstrated that all of the orthodontic oral mucosa and gingiva, or alveolar bone. There are
bracket adhesive materials except Transbond XT did several ways that materials may influence the health of
not reduce L929 cell survival (P . .05) when compared soft tissues—by delivering water-soluble components
with the control group. However, Transbond XT into the saliva and the oral cavity as well as by
interacting directly with adjacent tissues.6 The extrac-
Figure 1. Cultured L929 cells for control group. Figure 3. Cultured L929 cells for Transbond XT.

Figure 4. Cultured L929 cells for Bisco ORTHO.
tion assay described above is one of the most

frequently used methods to assay the mechanism of
intraoral cytotoxicity in the study of orthodontic bonding
composites.
Another important topic, the area of biocompatibility Figure 6. Cultured L929 cells for Quick Cure.
of materials, is also relevant to the practitioner from the
standpoint of the health of the dental team. In many
possible reactions. Moreover, individual test methods
cases, the risk of the adverse effects of biomaterials is
are usually adequate only to describe or document a
much higher for the dental team than for the patients
single aspect of a certain type of unwanted reaction.
because of chronic exposure of the dental team and
Cell culture tests will detect only the influence of a
manipulation of the materials when they are being
material on isolated cells.7 Isolated cells derived from
placed, set or removed.5 Furthermore, there is the
necessity of informing the affected personnel that the animal or human tissues are grown in culture plates
dental materials used by orthodontists can pose some and then are used for these tests.7 Today, primarily
risk to the patient and the dental team. It is the permanently growing cells are used for this purpose
orthodontic clinician’s problem to conclude whether because these cells can be easily amplified and their
this evidence is deserved and estimate the risk of behavior is well known, relatively consistent, and
these topics in orthodontic practice. constant.7 Frequently, permanent mouse fibroblast
A great variety of different test methods are used to (L929) cells are used. These cell cultures are
determine the risk of such damage to ensure material ‘‘incubated’’ with the materials or their extracts.
compatibility. However, the results of such evaluations Subsequently, a series of various parameters will be
are dependent not only on the tested material, but also measured; for example, the number of ‘‘surviving’’
on the test method used. Evaluation of the biocom- cells, protein synthesis, enzyme activity, or synthesis
patibility of dental materials is complex and compre- of inflammatory mediators.7 L929 fibroblasts and
hensive because unwanted tissue reactions may occur gingival fibroblasts have previously been shown to
in a great variety of types.7 have similar cytotoxicity levels. Consequently, L929
Previously, the tissue compatibility of orthodontic fibroblasts make a useful screening model for in vitro
bonding agents was studied in animal experiments.8 toxicity testing of dental materials. Because of its
Ethical considerations, poor reproducibility, and small excellent reproducibility, the L929 cell line was
sample sizes resulted in the development of in vitro preferred to primary gingival fibroblasts.9
cytotoxicity tests and their standardization.9 Any single The advantages of the MTT procedure are simplic-
test method is applicable only for investigating one ity, accuracy, reliability, and the saving of time. The
type of unwanted reaction out of a great variety of MTT method proved to be useful to estimate cell
densities in small culture volumes. The cultivation in
small culture volumes and the sensitive evaluation with
the MTT assay allow the screening and testing of
many different substances and fractions for the
determination of cytotoxicity. For these reasons, we
also used an MTT assay procedure.
Resin matrixes of the orthodontic composites
consist of mainly two monomers: bisphenol A diglycidyl
dimethacrylate (bis-GMA) and triethylene glycol di-
methacrylate (TEGDMA). In addition, the matrix resin
Figure 5. Cultured L929 cells for Light Bond. consists of a mixture of various monomers, for

CYTOTOXIC EFFECTS OF ORTHODONTIC COMPOSITES 763
example, bis-GMA and/or UDMA, as well as various was recently studied.17 Hence, cytotoxicity of Trans-
modifications of these molecules. Other ingredients of bond XT could be explained by the presence of bis-
the composite matrix are co-monomers (EGDMA, EMA in its matrix.
DEGDMA, TEGDMA) and various additives such as The result of this study showed that light-cured
photo-initiators (eg, camphorquinone), co-initiators orthodontic bonding adhesives have acceptable high
(eg, DMABEE, DEAEMA), inhibitors (eg, BHT), ultra- biocompatibility when compared with other dental
violet absorbers, photo-stabilizers, and pigments.10,11 adhesives.18,19 This issue may be explained by mixture
Polymerization in products used today is mainly of orthodontic adhesive monomers and a high degree
initiated by light; the light-sensitive initiator camphor- of cure.20
quinone acts together with an aliphatic amine-type However, the results of the present in vitro study
catalyst. TEGDMA has an important function because remain unclear, and further studies using different test
it decreases the viscosity of the matrix, thus allowing methods are needed for Transbond XT. Research
increased filler content. Resin-based composites efforts should focus on assessing long-term biologic
(ormocers), which are presented recently are based effects of orthodontic composites.
on a Si-O scaffold with methacrylic side chains, which
are necessary for polymerization.7 Orthodontic com- CONCLUSION
posite resins may be released bis-phenol A, a bis-GMA
precursor that exhibits cytotoxic effects concluded N The tested orthodontic bonding composites are
potential biologic adverse reactions. Furthermore, suitable for clinical application. However, Transbond
various ions are leached out at different times and in XT was cytotoxic (87%), while the other orthodontic
different conditions. Recently, data from animal studies composites caused no or only slight cellular alter-
have been presented concerning biodegradation of ations (90%, 91%).
HEMA/TEGDMA.12,13 Both ‘‘water-soluble’’ substances
are used in a variety of resin-based composites REFERENCES
(TEGDMA) and adhesives (HEMA/TEGDMA), and 1. Zachrisson BU, Buyukyilmaz T. Bonding in orthodontics. In:
thus are released from materials. Swallowed HEMA/ Graber TM, Vanarsdall RL, Vig K, eds. Orthodontics:
TEGDMA were almost completely absorbed by the Current Principles and Techniques. Philadelphia, Pa: Else-
organism. These ions are released from orthodontic vier Inc; 2005:579–659.
2. Malkoc S, Uysal T, Usumez S, Isman E, Baysal A. In vitro
resin-composite, diffuse through oral tissues, and are
assessment of temperature rise in the pulp during ortho-
cytotoxic. dontic bonding. Am J Orthod Dentofacial Orthop. In press.
In addition, Hansel et al.14 investigated the influence 3. Eliades T, Eliades G. Orthodontic adhesive resins. In:
of base monomers (bis-GMA, UDMA) and co-mono- Brantley WA, Eliades T, eds. Orthodontic Materials:
mers (TEGDMA, EGDMA) on the in vitro proliferation Scientific and Clinical Aspects. Stuttgart, Germany: Thieme;
of caries-relevant bacteria. They found that the base 2001:201–220.
4. Gioka C, Bourauel C, Hiskia A, Kletsas D, Eliades T, Eliades
monomers had no influence or only a slightly growth-
G. Light-cured or chemically cured orthodontic adhesive
inhibiting effect on these cultures, but that both of the resins? A selection based on the degree of cure, monomer
co-monomers tested (TEGDMA, EGDMA) promoted leaching, and cytotoxicity. Am J Orthod Dentofacial Orthop.
bacterial proliferation. Because these substances 2005;127:413–419; quiz 516.
usually leach from resin-based composites at higher 5. Wahata JC. Principles of biocompability. In: Brantley WA,
concentrations than base monomers do, an overall Eliades T, eds. Orthodontic Materials: Scientific and Clinical
Aspects. Stuttgart, Germany: Thieme; 2001:271–286.
increased bacterial growth may be the consequence 6. Huang TH, Tsai CY, Chen SL, Kao CT. An evaluation of the
in the presence of resin-based composites. Söhoel cytotoxic effects of orthodontic bonding adhesives upon a
et al.15 tested two different bis-GMA/TEGDMA con- primary human oral gingival fibroblast culture and a
taining resins that were used for the bonding of permanent, human oral cancer-cell line. J Biomed Mater
brackets. Both substances generated a sensitization Res. 2002;63:814–821.
in 50% of the experimental animals, with a subsequent 7. Schmalz G. Resin-based composites. In: Schmalz G,
Arenholt-Bindlev D, eds. Biocompatibility of Dental Materi-
allergic reaction.15 als. Berlin, Heidelberg, Germany: Springer-Verlag;
Usually, mixtures of these monomers are used in 2009:99–138.
orthodontic composites. When evaluating the ingredi- 8. Davidson WM, Sheinis EM, Shepherd SR. Tissue reaction
ent of tested materials in the present study, there were to orthodontic adhesives. Am J Orthod. 1982;82:502–507.
significant similarities in resin matrixes. However, 9. Jonke E, Franz A, Freudenthaler J, Konig F, Bantleon HP,
Schedle A. Cytotoxicity and shear bond strength of four
Transbond XT also contains bis-EMA. In addition, a
orthodontic adhesive systems. Eur J Orthod. 2008;30:
bis-EMA monomer showed a cytotoxic effect analo- 495–502.
gous to that of TEGDMA.16 The mechanism of 10. Ferracane JL. Current trends in dental composites. Crit Rev
cytotoxicity induced by TEGDMA in human fibroblasts Oral Biol Med. 1995;6:302–318.

11. Spahl W, Budzikiewicz H, Geurtsen W. Determination of permanent 3T3 and three human primary fibroblast cultures.
leachable components from four commercial dental com- J Biomed Mater Res. 1998;41:474–480.
posites by gas and liquid chromatography/mass spectrom- 17. Stanislawski L, Lefeuvre M, Bourd K, Soheili-Majd E,
etry. J Dent. 1998;26:137–145. Goldberg M, Perianin A. TEGDMA-induced toxicity in
12. Reichl FX, Durner J, Hickel R, et al. Uptake, clearance and human fibroblasts is associated with early and drastic
metabolism of TEGDMA in guinea pigs. Dent Mater. 2002; glutathione depletion with subsequent production of
18:581–589. oxygen reactive species. J Biomed Mater Res A. 2003;66:
13. Reichl FX, Durner J, Manhart J, et al. Biological clearance of 476–482.
HEMA in guinea pigs. Biomaterials. 2002;23:2135–2141. 18. Koulaouzidou EA, Helvatjoglu-Antoniades M, Palaghias G,
14. Hansel C, Leyhausen G, Mai UE, Geurtsen W. Effects of Karanika-Kouma A, Antoniades D. Cytotoxicity of dental
various resin composite (co)monomers and extracts on two adhesives in vitro. Eur J Dent. 2009;3:3–9.
caries-associated micro-organisms in vitro. J Dent Res. 19. Ulker HE, Sengun A. Cytotoxicity evaluation of self adhesive
1998;77:60–67. composite resin cements by dentin barrier test on 3D pulp
15. Söhoel H, Gjerdet NR, Hensten-Pettersen A, Ruyter IE. cells. Eur J Dent. 2009;3:120–126.
Allergenic potential of two orthodontic bonding materials. 20. Usumez S, Buyukyilmaz T, Karaman AI, Gunduz B.
Scand J Dent Res. 1994;102:126–129. Degree of conversion of two lingual retainer adhesives
16. Geurtsen W, Lehmann F, Spahl W, Leyhausen G. Cytotox- cured with different light sources. Eur J Orthod. 2005;27:
icity of 35 dental resin composite monomers/additives in 173–179.

Original Article
Early headgear effect on the eruption pattern of maxillary second molars

Yossi Abeda; Ilana Brinb
ABSTRACT
Objective: To test the hypothesis that the use of a combination headgear (HG) during the first
phase of orthodontic treatment has no effect on the eruption pattern of the maxillary second
permanent molars.
Materials and Methods: The records of the patients in a two-phase randomized clinical trial of
early Class II treatment were utilized. The HG group comprised 47 patients, and the control (CON)
group comprised 52 patients. The mean age of both groups was 9.4 years at the beginning of the
clinical trial (T1). Cephalograms and panoramic views obtained at T1 and at the end of 15 months
of phase I treatment or observation (T2) were utilized. The vertical and horizontal movements of
the first and second upper molars (U6 and U7, respectively) were measured. The beginning of
phase II (T3) and the end of phase II (T4) records were visually reviewed for follow-up of the
eruption of the U7.
Results: The pattern of movement for the distal and vertical displacement of the U6 and U7 was
significantly different in the HG and CON groups (P , .001). At the end of phase I, none of the U7
in either group were diagnosed as malposed or suspected for impaction. At the end of phase II, all
but one U7 with a possible cystic lesion had erupted.
Conclusions: The hypothesis is rejected. Forces exerted by combination headgear to the U6 in
phase I have a distalizing effect and a transitory slowing down effect on the eruption of the U7
buds. These latter teeth always erupted except when pathology occurred. (Angle Orthod.
2010;80:642–648.)
KEY WORDS: Second molar eruption; Combined headgear; Class II
INTRODUCTION panoramic radiographs, it is time to start extraoral

forces attached to the U6 to avoid impaction. None of
The eruption pattern of the maxillary second molar
these studies related to the use of a specific type of
(U7) may be affected by early application of headgear
HG.
(HG) prior to its intraoral eruption. Nanda and
A few studies investigated the effect of the unerupt-
Dandajena1 observed that use of HG for an extended
ed U7 on the efficiency of noncompliance appliances
period of time may result in delayed eruption of the
for maxillary molar distalization.3,4 Kinzinger et al.4
second molars. Bishara2 suggested the use of a
noted that the U7 tooth bud acts like a fulcrum on its
clinical landmark to determine the optimal timing for
mesial neighbor and that the degree of tipping of the
HG treatment based on the relationship of the U7 to
U6 was much greater in patients whose U7 was still in
the maxillary first molar (U6). He suggested that when
the budding stage. Bussick and McNamara3 observed
the crowns of the U7 have erupted past the apical third
no difference in the amount of molar correction with the
of the roots of the U6, as determined from periapical or
pendulum appliance between patients who had the U7
erupted and those who did not. Taner et al.5 used a
a
Lecturer, Department of Orthodontics, Hadassah Faculty of
Dental Medicine, Hebrew University Jerusalem, Israel. cervical headgear and Tortop and Yuksel6 a combina-
b
Professor, Department of Orthodontics, Hadassah Faculty of tion headgear and both reported a significant effect on
Dental Medicine, Hebrew University Jerusalem, Israel. distal tipping and eruption of the U7 when a distal force
Corresponding author: Yossi Abed, Department of Orthodon- was applied to the U6.
tics, Hadassah School of Dental Medicine, Hebrew University,
PO Box 12272, Jerusalem, Israel 91120
However, no quantitative study of the effect of
(e-mail: abedy@cc.huji.ac.il) combination HG treatment on the eruption pattern of
the U7 in a randomized clinical trial design was found
Accepted: November 2009. Submitted: October 2009.
G 2010 by The EH Angle Education and Research Foundation, in the literature. It was therefore the purpose of this
Inc. project to investigate the possible effect of combination

EARLY HEADGEAR EFFECT ON MAXILLARY SECOND MOLARS 643
HG treatment in the first phase of orthodontic Measurements on Cephalograms

treatment on the eruption pattern of the U7 in a
For the evaluation of the displacement of the U6 and
random population of Class II patients.
U7, the following measurements were performed
(Figure 1):
N Vertical displacement was measured on a perpen-
The records of the patients in the University of North
dicular line from the mid-crown point to the SN line.
Carolina (UNC) two-phase randomized clinical trial of
N Horizontal displacement was measured from the
early Class II treatment7 were utilized. Fourteen
mid-crown point to the SN-perp.
patients were removed from the original UNC material
N Axial inclination of the U7 was measured as the
because of the quality of the scanned cephalograms:
angle between the long axis of the molar and the SN
five from the headgear (HG) group and nine from the
line. The long axis was defined as a line connecting
control (CON) group. Finally, the HG group comprised
the MC and RF points.
47 patients treated with a combination HG with a short
outer bow (ending approximately at the mesial of the The measurements were performed by computing
molar tubes) adjusted to deliver between 8 and 10 the positions of the mid-crown points for each side
ounces to the headcap and with the neck strap force separately, and the bilateral mean was applied for the
just sufficient to prevent buccal flaring of the upper statistical evaluation. When T1–T2 calculations were
molars. The CON group comprised 52 patients who used, negative signs denoted eruption in the vertical
did not receive any phase I treatment. The mean age dimension or mesial movement in the horizontal
of both groups was 9.4 years (HG group SD 5 1.0 dimension.
years; CON group SD 5 1.2 years) at the beginning of
the clinical trial. About 60% of the participants were Statistical Analyses
male (60% in the HG group; 57% in the CON group),
Ten random cases were remeasured, and the
and about 90% of the patients exhibited a bilateral
standard error was calculated using Dahlberg’s8
Class II molar (90% in the HG group; 93% in the CON
formulas for each variable. The method error was
group).
found to be low (Table 1) except for the furcation
Cephalograms obtained at the beginning (T1) and at
measurements and these were omitted from further
the end of 15 months (T2) of phase I treatment (for
analysis. Thus, the mean of the right and left linear
the HG group) or observation (for the CON group)
measurements and their SD values were calculated.
were measured in this study. In addition, the records
The paired t-test was applied to test the significance
of the beginning of phase II and the end of treatment
of the change between the two time points within each
(T3 and T4, respectively) were available and were
group and between the groups. A P value of 5% or less
visually evaluated for confirmation of the U7 full
was considered as statistically significant.
eruption at these stages. However, no measurements
were performed on these final sets because the
RESULTS
influence of the fixed appliances and/or extractions on
the eruption of U7 was beyond the scope of this study. For the linear measurements, the range of the
Because the molar bands were not removed at the standard error was 0.4 mm to 0.7 mm with one
end of phase I in some of the patients, the technician exception of 2.3 mm for the horizontal displacement of
was not blinded as to the treatment group of these U6 at T2 (Table 1). The error for the angular
patients. measurements was high, ranging from 4.7u to 8.3u,
The cephalograms were scanned. For digitization and consequently the angular measurements were
and measurements, the Viewbox 3.1.1.12 version of deleted from further consideration.
cephalometric software (dHAL SOFTWARE, Kifissia, At the beginning of phase I, the horizontal and
Greece) was used. The following landmarks were vertical positions of the U6 and U7, respectively, were
identified (Figure 1): sella turcica (S); nasion (N); the similar in both groups (Table 2). The change in the
most mesial (MM) and the most distal (MD) point on horizontal and vertical positions of these teeth in the
the crown contour of the U6 and U7; and root furcation HG and CON groups with time (T1–T2) are presented
(RF) point of the U6 and U7. The midpoint of the molar in Tables 3 and 4, respectively. These changes within
crown (MC) width was calculated, and the position of each group were statistically significant except for the
these teeth was measured from the midpont land- maxillary U7 in the horizontal dimension.
marks, utilizing the Viewbox software. In the process of Comparison between the HG and CON groups by
landmark identification, the enhancement function was the molar response is presented in Table 5, and
utilized. demonstrates significant differences for both molars

644 ABED, BRIN
Figure 1. Cephalometric landmarks utilized in this study: S indicates sella turcica; N, nasion; MM, the most mesial point on the crown contour of
the molar; MD, the most distal point on the crown contour of the molar; MC, mid-distance between MM and MD; RF, root furcation point of the
molar. Horizontal measurements were performed to SN-perp. line and the vertical measurements were performed to the SN line. All
measurements were performed for the right and left molars, and the average value was used for statistical evaluation.
in both dimensions. When the changes between the

U6 and U7 were studied closely within each group, a
significant change in both dimensions was found for
the U6 and U7 in the CON group (Table 6). This was
true for the HG group as well for the vertical dimension Table 1. Measurement Error of the Various Variablesa
only. However, the U6 and U7 moved the same Measurement Measurement
distance distally in this group. Variable Error at T1 Error at T2
In the HG group, the U6 demonstrated an average U6 molar vertical
distal displacement of 2.6 mm and a moderate (1.5- displacement 0.6 mm 0.6 mm
U6 molar horizontal
mm) vertical eruption, while in the CON group it moved displacement 0.7 mm 2.3 mm
to the mesial 1.2 mm, on average, with a 2.5-mm U6 molar furcation 5.9u 4.7u
eruption (Table 5). The U7 erupted toward the occlusal U7 molar vertical
plane in both groups, although the eruption in the HG displacement 0.5 mm 0.7 mm
U7 molar horizontal
group, on average, was significantly slowed down
displacement 0.4 mm 0.5 mm
when compared to that of the CON group as can be U7 molar furcation 8.3u 6.6u
seen in Table 5 (CON mean: 4.8 mm; HG mean: a
U6 indicates upper maxillary first molar; U7, upper maxillary
2.6 mm). In the HG group, the average distal second molar; T1, at the beginning of the clinical trial; T2, at the end
displacement of the U7 was 2.8 mm, while there was of 15 months of phase I treatment or observation.

Table 2. Comparison of the Position (6 SD) in mm of the Maxillary Table 4. Comparison of the Horizontal and Vertical Change in
Molars in the Headgear (HG) and Control (CON) Groups at the Molar Position in mm (6 SD) in the CON group (n 5 52) (Paired
Beginning of Phase I (Independent Samples t-Test)a Samples t-Test)a
Position HG (n 5 47) CON (n 5 52) P Position T1 T2 T1–T2 P
b
U7 molar vertical position 49.5 6 3.6 48.7 6 4.4 .348 U7 molar vertical position 48.7 6 4.4 53.5 6 5.5 24.8 6 1.9 .000
U7 molar horizontal position 23.8 6 3.2 24.6 6 3.1 .194 U7 molar horizontal position 24.6 6 3.1 24.5 6 3.5 0.1 6 1.5 .779
U6 molar vertical position 61.0 6 3.1 60.2 6 3.5 .244 U6 molar vertical position 60.2 6 3.5 62.7 6 3.7 22.5 6 1.1b .000
U6 molar horizontal position 29.4 6 3.8 30.2 6 4.3 .360 U6 molar horizontal position 30.2 6 4.3 31.4 6 4.5 21.2 6 1.5c .000
a a
U7 indicates upper maxillary second molar; U6, upper maxillary U7 indicates upper maxillary second molar; U6, upper maxillary
first molar. first molar; T1, at the beginning of the clinical trial; T2, at the end of
15 months of phase I treatment or observation.
b
Negative value denotes vertical eruption.
virtually no displacement of this tooth, on average, in c
Negative value denotes distal movement.
the CON group (mean: 0.1 mm) (Table 5).
Visual evaluation of the records at the end of phase I
very compliant patients, who used their headgears
indicated a normal eruption path of all U7 in both
more than the recommended 14 hours per day. This
groups. At the end of phase II, visual review of the
phenomenon prompted us to analyze the possible
records suggested that all but one U7 were erupted.
changes in the eruptive pattern of U7 adjacent to U6 to
The unerupted U7 was possibly impeded by a cystic
which HG was applied.
lesion.
Our results are limited to linear measurements since
angular measurements of molar displacement were
DISCUSSION
omitted due to a large method error (Table 1). This
Study of the U7 eruption pattern in untreated error can be explained by the relatively early develop-
patients with skeletal Class II suggested that in these mental stage of the U7 of the young patients in this
individuals, the U7 may erupt earlier than in individuals study, rendering the root-furcation difficult to locate
with skeletal Class I malocclusion.9 In addition, it was precisely on the cephalograms. This premise is
found that the U7 erupt as soon as there is enough supported by the larger method error found in the U7
space in the upper jaw (Tschechne et al.).10 However, angular measurement compared with the U6 (Ta-
we do not know if this is also true for patients ble 1), which was high as well. Most other measure-
undergoing early HG treatment for Class II malocclu- ment errors were comparable to errors in studies with
sion. A common treatment sequence in these patients similar methodologies,5,6 except for the U6 horizontal
consists of distal movement and/or growth modification displacement at T2.
by HG to the first molars in the first phase of treatment, In this study the SN reference plane was chosen
and application of fixed appliances with or without HG because of the high reproducibility of the relevant
in the second phase. The clinical experience indicates cephalometric landmarks. We are aware of the fact
that in most cases at the end of treatment, the U7 erupt that this measurement method will not differentiate
into the oral cavity uneventfully following this treatment between growth effects and dental movement effect.
protocol. However, there are cases in which one or However, because of the randomization effect, the
both of the U7 are severely delayed or even impacted position of the molars was similar in both groups
following the HG treatment. It should be kept in mind (Table 2), and the growth pattern could be assumed to
that the incidence of impacted U7 in a random be similar as well. Thus, any difference between the
population is 0.08% (Andreasen et al.).11 The delay groups regarding the pattern of eruption of U7 is
was observed in our clinical experience especially in supposed to be due to the treatment only.
The eruption pattern of U7 under the influence of
Table 3. Comparison of the Horizontal and Vertical Molar Position combination HG to the U6 in Class II patients was
in mm (6 SD) in the Headgear (HG) Group (n 5 47) (Paired found to be significantly altered during phase I when
Samples t-Test)a compared with the CON group (Table 5).
Position T1 T2 T1–T2 P
U7 molar vertical position 49.5 6 3.6 52.1 6 5.3 22.6 6 3.0b .000
In the Horizontal Dimension
U7 molar horizontal position 23.8 6 3.2 21.0 6 3.5 2.8 6 1.6 .000
In the CON group, the U7 did not change position
U6 molar vertical position 61.0 6 3.1 62.5 6 3.5 21.5 6 1.7b .000
U6 molar horizontal position 29.4 6 3.8 26.8 6 4.0 2.6 6 2.2 .000 horizontally at this stage of development. However, in
a
the HG group U7 and U6 were displaced distally in the
U7 indicates upper maxillary second molar; U6, upper maxillary
first molar; T1, at the beginning of the clinical trial; T2, at the end of
horizontal dimension to a similar extent (Table 6). The
15 months of phase I treatment or observation. amount of distal movement of the U7 in this study was
b
Negative value denotes vertical eruption. somewhat larger than that found by Tortop and

646 ABED, BRIN
Table 5. Comparison Between the Changes in Molar Position in mm (6 SD) With Time in the Headgear (HG) and Control (CON) Groups
(Independent Samples t-Test)a
Difference HG (n 5 47) CON (n 5 52) P
U7 molar vertical difference T1–T2b 22.6 6 3.0 24.8 6 1.9 .000
U7 molar horizontal difference T1–T2 2.8 6 1.6 0.1 6 1.5 .000
U6 molar vertical difference T1–T2b 21.5 6 1.7 22.5 6 1.1 .001
U6 molar horizontal difference T1–T2 2.6 6 2.2 21.2 6 1.5c .000
a
U7 indicates upper maxillary second molar; U6, upper maxillary first molar; T1, at the beginning of the clinical trial; T2, at the end of 15 months
of phase I treatment or observation.
b
c
Negative value denotes mesial movement.
Yuksel.6 They used a similar appliance for approxi- the controls. The amount of vertical displacement of
mately 5 months and an observation period of 2 years the U7 in this study was larger than that found by
and found 2.3 mm of distal movement. Taner et al.5 Taner et al.5 who used a cervical HG, but for a shorter
who used a cervical HG found 2.3 mm of distal period of time. On the other hand, Tortop and Yuksel6
movement of the U7 over 11.3 months. Both studies suggested a vertical displacement of the U7 of 0.5 mm
however, measured the movement to a different only during treatment with combined HG for about 5
reference plane. months and a 2-year observation period. The large
It is also interesting to note the difference between variability in the eruption response of the U7 to HG
the U6 displacements in the two groups. In the horizontal therapy as indicated by the various authors may also
dimension, the U6 in the CON group was displaced to the be caused by differences in compliance with HG use.
mesial 1.2 6 1.5 mm (Table 4). This can be explained by Thus, overall it can be concluded that a 15-month
the normal growth process as well as by mesial drift duration of use of a combination headgear may result
following the loss of the second deciduous molar, which in delayed eruption of the U7 but not in its impaction.
could occur concomitantly. In the HG group, the U6 were This is in agreement with Nanda and Danjema1 who
displaced distally 2.6 6 2.2 mm. Thus, when the mesial observed that prolonged use of HG results in delayed
migration that normally occurs (CON group) and the eruption of U7. It can also be hypothesized that the
distal displacement due to the HG were combined, the more distal eruption of U7 points to a longer eruption
change in position of the U6 in the two groups amounted path, which may last for a longer period of time.
to a mean of 3.8 mm (1.2 + 2.6 5 3.8). The vertical delay could be explained by the fact that
When the negligible migration which normally occurs the second molars might have served as a fulcrum for
(CON group) and the distal displacement due to the the distal tipping of the U6 (Figure 2), similar to the
HG were combined, the change in the U7 position in situation described in the case of the pendulum
the HG groups amounted to a mean of 2.7 mm (2.8 2 appliance.4 Thus, the temporary distal tipping of the
0.1 5 2.7). It can be suggested then, that the net distal U6 blocked the eruption path of the U7 until uprighting
displacement of the U7 in the HG group was about the of the U6 occurred and the U7 completed its eruption.
same as that of the U6 in the HG group (2.8 and In addition, it is known that the eruption of the U7 is
2.6 mm, respectively) (Table 6). However, the change limited by the space distal to the U6 in the dental
in position of the U7 in the treatment group was about arch.12,13 The distal movement of the U6 in the HG
71% (2.7/3.8) of the change in position of the U6. group possibly also encroached on the U7 space, thus
causing a delay in its eruption. Thus, the tendency
In the Vertical Dimension
In the HG group, the vertical eruption of the U6 was Table 6. Comparison of the Change in Position in mm (6 SD)
Between the U6 and U7 in the Headgear (HG; n 5 47) and Control
impeded by 1 mm. The U6 erupted 21.5 6 1.7 mm for
(CON; n 5 52) Groups (Paired Samples t-Test)a
the HG group and 22.5 6 1.1 mm for the CON group
Change U7 U6 P
(Table 6), probably because of the combination HG
configuration exerting an intrusive component of force. Vertical change in HG groupb 22.6 6 3.0 21.5 6 1.7 .000
Horizontal change in HG group 2.8 6 1.6 2.6 6 2.2 .734
Although the HG treatment allowed for the U7
Vertical change in CON groupb 24.8 6 1.9 22.5 6 1.1 .000
vertical intra-alveolar eruption, this process was Horizontal change in CON group 0.1 6 1.5 21.2 6 1.5c .000
delayed and more variable compared with the controls a
U7 indicates upper maxillary second molar; U6, upper maxillary
(2.6 6 3.0 mm vs 4.8 6 1.9 mm, respectively). This first molar.
mean delay of 2.2 mm (4.8 2 2.6 5 2.2) amounted to b
about 46% (2.2/4.8) of the normal vertical eruption in c
Negative value denotes mesial movement.

Figure 2. Patient 4801. (A) Panoramic view before treatment (T1). (B) End of phase I—note the temporary distal tipping of the U6, which blocks
the eruption path of the U7. (C) Uprighting of the U6 occurred during phase II, and the U7 completed its eruption.

648 ABED, BRIN
towards early eruption of maxillary U7 in Class II N Most of the horizontal movement of the U6 was
patients9 is reversed by early orthodontic treatment expressed in the U7 movement.
with combination headgear (phase I). It should be
noted that this delay in eruption of the U7 may lead to a
ACKNOWLEDGMENTS
later start of the second phase of orthodontic treatment
while the orthodontist awaits the eruption of these The authors wish to thank Dr Joan Tulloch for the possibility to
teeth. The occurrence of impaction of U7 is probably study the clinical material obtained with the help of the NIH grant
R01-DE 08708. Also, our thanks go to Dr Ceib Phillips for her
very rare and follows only extreme HG use, beyond the
constructive remarks.
recommended number of hours, which is also rarely
encountered. Visual evaluation of the roentgenograms
REFERENCES
of the patients observed in this study, obtained at the
end of phase II (T4), did not reveal any impacted U7 1. Nanda R, Dandajena T. The role of the headgear in growth
except for one U7. Close inspection of the posttreat- modification. Semin Orthod. 2006;12:25–33.
2. Bishara S. Class II malocclusion: diagnostic and clinical
ment panoramic radiograph reveals a possible cystic considerations with and without treatment. Semin Orthod.
lesion. This occurrence may be unrelated to the HG 2006;12:11–12.
treatment, but clear etiology cannot be determined. 3. Bussick TJ, McNamara JA Jr. Dentoalveolar and skeletal
The contention of Bishara2 that in order to avoid changes associated with the pendulum appliance. Am J
impaction of U7 following the application of HG forces Orthod Dentofacial Orthop. 2000;117:333–343.
to the U6, treatment should be started only when the 4. Kinzinger GS, Fritz UB, Sander FG, Diedrich PR. Efficiency
of a pendulum appliance for molar distalization related to
U7 buds are in the height of the apical third of the second and third molar eruption stage. Am J Orthod
adjacent U6 was not examined in this study. However, Dentofacial Orthop. 2004;125:8–23.
visual evaluation of the T1 panoramic views in all 5. Taner TU, Yukay F, Pehlivanoglu M, Cakirer B. A
cases in this study revealed the presence of the comparative analysis of maxillary tooth movement produced
above-mentioned relationships between the U6 and by cervical headgear and pend-x appliance. Angle Orthod.
2003;73:686–691.
U7. Thus, the HG treatment was started at the optimal
6. Tortop T, Yuksel S. Treatment and posttreatment changes
timing according to Bishara,2 which could also explain with combined headgear therapy. Angle Orthod. 2007;77:
the lack of impactions (except for one case with a 857–863.
possible cystic lesion). 7. Tulloch JF, Phillips C, Koch G, Proffit WR. The effect of early
There are some limitations to this study, one of them intervention on skeletal pattern in Class II malocclusion: a
being the conservative two-dimensional presentation of randomized clinical trial. Am J Orthod Dentofacial Orthop.
1997;111:391–400.
the region of interest. It would be significantly more 8. Dahlberg E. Statistical Methods for Medical and Biological
informative to follow the eruptive process of the U7 in Students. New York, NY: Interscience Publications; 1940.
three dimensions, but high cost and high radiation 9. Brin I, Camasuvi S, Dali N, Aizenbud D. Comparison of
exposure to the patient limit 3D clinical use. In addition, second molar eruption patterns in patients with skeletal
the relatively small sample size did not allow subdivision Class II and skeletal Class I malocclusions. Am J Orthod
into maxillary or mandibular Class II or subdivision by
10. Tschechne S, Muller B, Dibbets J. Sagittal space relations in
gender. These approaches could give additional insight the maxilla during molar eruption. J Orofac Orthop. 2008;69:
into the question of altered eruption pattern of the U7. 94–98.
11. Andreasen JO, Petersen JK, Laskin DM. Textbook and
CONCLUSIONS Color Atlas of Tooth Impactions: Diagnosis, Treatment,
Prevention. Copenhagen, Denmark: Munksgaard; 1997.
N The forces exerted by combination headgear to the 12. Mitani H. Behavior of the maxillary first molar in three planes
U6 at phase I orthodontic treatment have a slowing with emphasis on its role of providing room for the second
down effect by 46% on the vertical eruption of U7 and third molars during growth. Angle Orthod. 1975;45:
159–168.
compared with untreated controls.
13. Godt A, Berneburg M, Kalwitzki M, Goz G. Cephalometric
N This effect is transitory because all U7 without analysis of molar and anterior tooth movement during
pathologic signs were in occlusion at the end of cervical headgear treatment in relation to growth patterns.
phase II. J Orofac Orthop. 2008;69:189–200.

Original Article
Cervical vertebra morphology in different skeletal classes

A three-dimensional computed tomography evaluation
Miyuki Watanabea; Tetsutaro Yamaguchib; Koutaro Makic
ABSTRACT
Objective: To describe cervical vertebra morphology in subjects with different anteroposterior jaw
relationships.
Materials and Methods: Cone-beam computed tomography images of 31 female subjects aged
19 to 41 years were evaluated. Subjects constituted two groups according to the ANB angle: group
1, skeletal Class II (ANB angle .5); and group 2, skeletal Class III (ANB angle ,1). Nine linear
measurements and one angular measurement were used to assess the vertebral morphology. The
Mann-Whitney U-test was used for statistical analysis.
Results: The mean atlas dorsal arch height was significantly shorter in Class II subjects compared
with those in Class III (P , .05). The cervical vertebra morphological analysis by cone-beam
computed tomography was of comparable precision to three-dimensional computed tomography
evaluations. This study confirmed previous findings that Class II subjects have significantly lower
atlas dorsal arch heights.
Conclusion: The height of the atlas dorsal arch of cervical vertebrae is affected by the
anteroposterior skeletal pattern. (Angle Orthod. 2010;80:719–724.)
KEY WORDS: Cervical vertebrae; Skeletal Class II; Skeletal Class III; Cone-beam computed
tomography (CBCT)
INTRODUCTION space,5 occlusion,2–7 and temporomandibular disor-

ders.8,9 Furthermore, head posture is linked to the
The cervical vertebra column supporting the head
development and function of dentofacial structures.9
comprises seven vertebrae. The first vertebra (C1) or
The relationship between C1 dimensions and cranio-
atlas and the second vertebra or axis together form the
cervical posture has also been studied.2 Cervical
superior or suboccipital segment connecting the spine posture was linked to mandibular length, with longer
to the occiput and involving a complex chain of joints. mandibles associated with cervical columns more
Suboccipital muscles attached to this region determine inclined to the true horizontal.10 Mandibular length was
head posture, controlling fine through complicated also directly correlated with straightness of the cervical
movements for compound flexion and extension, as column (ie, a lower cervical lordosis angle).11 However,
well as lateral flexion with rotation.1 no previous studies have described the relationship
Dimensions of C1 as well as head and neck posture between cervical vertebra morphology using three-
are associated with factors such as craniofacial dimensional imaging and maxillofacial morphology.
morphology, including the cranial base,2–4 upper airway Computed tomography (CT) was first considered for
such a study because of the accurate three-dimen-
a
Research Associate, Department of Orthodontics, School of
sional imaging that is possible with CT. However,
Dentistry, Showa University, Tokyo, Japan.
b
Assistant Professor, Department of Orthodontics, School of limitations such as radiation, machine size, and cost
Dentistry, Showa University, Tokyo, Japan. made this approach impractical. The more recently
c
Professor and Chairman, Department of Orthodontics, established cone-beam computed tomography
School of Dentistry, Showa University, Tokyo, Japan. (CBCT) presents a more feasible alternative, as these
Corresponding author: Dr Tetsutaro Yamaguchi, Department
lower-cost and smaller machines still produce high-
of Orthodontics, School of Dentistry, Showa University, 2-1-1,
Kitasenzoku, Outa-ku, Tokyo 145-8515, Japan quality data. With several CBCT scanners now
(e-mail: tyamaguchi@dent.showa-u.ac.jp) available, involving lower radiation dosages12–15 and
Accepted: December 2009. Submitted: October 2009. lower costs,16 three-dimensional (3D) radiography is
G 2010 by The EH Angle Education and Research Foundation, becoming more commonplace in the dental profession
Inc. as a valuable diagnostic tool.

720 WATANABE, YAMAGUCHI, MAKI
This study investigated the detailed morphology of Table 1. The Maxillofacial Characteristics of Class II and Class
cervical vertebrae in three dimensions in subjects with III Groups
different anteroposterior skeletal patterning. Class II Class III
Group Group
MATERIALS AND METHODS Ave SD Ave SD P Value
Subjects SNA, degrees 80.0 5.0 80.8 3.9 NS
SNB, degrees 74.7 3.8 81.7 5.2 .00**
CBCT scan radiographs were obtained from the ANB, degrees 5.4 2.3 -0.9 2.5 .00**
Department of Orthodontics files at Showa University Mandibular plane, degrees 32.2 6.2 30.8 5.2 NS
Dental Hospital for orthognathic surgery. Images were Ave indicates average; SD, standard deviation; and NS, not
derived from pretreatment CBCT scans of 31 female significant.
subjects aged 19 to 41 years, for the following ** Significant difference, P , .01
conditions: maxillofacial injury without fracture, diffuse
inflammation, soft tissue tumor, neuralgia, and un- strates both the angular and linear measurements
known lesion. Subjects were excluded from the study if obtained by CBCT. Of those obtained, eight linear
they had congenital disorders such as cleft palate, measurements and one angular measurement were
general physical problems, or disorders of the pharyn- used to assess cervical vertebra morphology with
geal soft tissue, including adenoiditis or enlarged different sagittal skeletal patterns.
tonsils, or if they were not of Japanese racial heritage. 1. Horizontal outer anteroposterior (AP) diameter of
Final participants met the following requirements: all the first cervical vertebra (C1)17 (mm) (HOAPC1;
permanent teeth erupted, except third molars; no Figure 1)
functional mandibular deviations; no evident facial 2. Horizontal inner AP diameter of C1 17 (mm)
asymmetry; no history of orthodontic treatment during (HIAPC1; Figure 1)
childhood; and no neck pain or medical history of 3. Horizontal outer transverse diameter of C117 (mm)
cervical disorders. Only females were included in this (HOTDC1; Figure 1)
study because the number of males with skeletal Class 4. Distance between outer margin of transverse
II (ANB angle .5) and/or skeletal Class III (ANB angle foramen and outer margin of lateral mass17 (mm)
,1) established with CBCT images was insufficient for (outer margin; Figure 1)
meaningful analysis. This study was approved by the 5. AP diameter of superior surface of C1 anterior
Showa University Dental Hospital Ethics Committee. arch17 (mm) (superior surface; Figure 1)
6. Lateral outer AP diameter of C118 (mm) (LOAPC1;
Obtaining CBCT Images Figure 2)
Cervical vertebrae were scanned and analyzed 7. Height of the atlas dorsal arch18 (mm) (dorsal arch;
using the CB MercuRay (Hitachi Medico Technology, Figure 2)
Tokyo, Japan). The scanning conditions were 100 kVp, 8. Frontal outer transverse diameter of C1 (mm)
10 mA, F-mode 512 slices/scan (slice width of 377 mm), (FOTDC117; Figure 3)
and 9.6 seconds. Data obtained were reconstructed 9. Angle along axis line of the dens to occlusal plane
using the CBworks 2.0 three-dimensional reconstruc- (dens angle19 [degrees]; Figure 4)
tion software (Hitachi Medico Technology).
For CT imaging, patients were positioned in centric
Reliability
occlusion (maximum dental intercuspation), and their
heads were positioned such that the Frankfort and Measurement error was determined by the random
midsagittal planes were perpendicular to the floor. selection of four CBCT images on two separate
Subjects were seated in the CB MercuRay system, occasions. One-way analysis of variance, used to test
with their facial median line vertical to the floor and the quality of means for the three measurements,
Frankfort plane parallel to the floor. This position was suggested that this sampling was consistent. Mean
controlled by a guideline directed from the front and scores for the three measurements did not differ
sides. significantly; the measurement error was thus consid-
ered to be negligible.
Measurements
CBCT images were used to assess the maxillofacial
characteristics of Class II and Class III subjects, SNA, The statistical significance of differences between
SNB, ANB, and mandibular plane (linear measure- the Class II and Class III groups was determined using
ments detailed in Tables 1 and 2). Figure 4 demon- the Mann-Whitney U-test. Analyses were performed

THE MORPHOLOGY OF CERVICAL VERTEBRAE 721
Table 2. Mean Value of Cervical Vertebra Measurements ular plane angle was noted between Class II and Class
Class II Group Class III Group
III subjects.
Table 2 details the mean value of cervical vertebra
Ave SD Ave SD P Value
measurements in Class II and Class III groups. Only
(1) HOAPC1, mm 72.1 3.2 71.5 3.1 .62 the height of the atlas dorsal arch showed a significant
(2) HIAPC1, mm 42.1 3.0 42.8 2.2 .47
(3) HOTDC1, mm 72.1 3.2 71.5 3.1 .62
difference between Class II and Class III (P 5 .029),
(4) Outer margin, mm 8.2 1.1 8.7 1.4 .26 with the other measurements not significantly different
(5) Superior surface, mm 6.5 1.1 6.5 1.6 .97 among subjects regardless of skeletal class.
(6) LOAPC1, mm 43.0 1.8 42.8 2.5 .81
(7) Dorsal arch, mm 8.3 1.1 9.9 2.6 .03*
DISCUSSION
(8) FOTDC1, mm 72.3 3.1 71.4 3.1 .43
(9) Dens angle, degrees 79.7 8.2 76.7 8.6 .33 Skeletal class and cervico-vertebral anatomy both
Ave indicates average; SD, standard deviation. are associated with craniofacial structure. CT imaging
* Significant difference, P , .05 has proved remarkably accurate in these association
studies for linear,20–23 geometric,24 and volumetric25,26
measurements within the maxillofacial complex. Fur-
using Statcel 2 (four-step Excel statistics, version 2,
ther to this, we now reveal differences in cervical
OMS Publishing, Saitama, Japan). P values less than
vertebra morphology in subjects with different antero-
.05 were considered statistically significant.
posterior jaw relationships.
This study demonstrated significant differences
RESULTS
between Class II and Class III subjects in height of
Table 1 details the mean value of maxillofacial the atlas dorsal arch. A pattern of positive correlation
dimensions in the Class II and Class III groups. No between mandibular length (Ar-Me) and the antero-
statistically significant difference in SNA and mandib- posterior length of the atlas was previously shown to
Figure 1. a Horizontal outer anteroposterior (AP) diameter of the first cervical vertebra (C1) (mm); HOAPC1. b Horizontal inner AP diameter of
C1 (mm); HIAPC1. c Horizontal outer transverse diameter of C1 (mm); HOTDC1. d Distance between outer margin of transverse foramen and
outer margin of lateral mass (mm); outer margin. e AP diameter of the superior surface of C1 anterior arch (mm); superior surface.

Figure 2. f Lateral outer AP diameter of C1 (mm); LOAPC1. g Height of the atlas dorsal arch (mm); dorsal arch.
increase in growing subjects.18 In addition, morpholog- the atlas dorsal arch seems independent of craniofa-
ical analysis of the atlas dorsal arch revealed a clear cial vertical differences. Several studies have suggest-
association with the growth direction of the mandible, ed that craniofacial vertical differences are related to
whereby a lower atlas dorsal arch indicated less differences in cervical morphology.6,7 The present
horizontal growth of the mandible.27 This is probably findings showed no difference in outer margin values
due to the fact that subjects with a low dorsal arch had between groups, although the vertical differences in
a relatively elevated head position and thus altered craniofacial morphology were related to the outer
suprahyoidal muscular activity, which would perma- margin measurements.28
nently affect the position of the mandible.27 Therefore, The anatomy and position of the cervical curvature
it was proposed that understanding the form of the in space depend on various body factors, including
cervical vertebra and the orthodontic treatment would ethnicity,29–31 gender,29,30,32–34 age,3,32,33 stature,35 and
enhance the treatment plan for subsequent growth.2 craniofacial morphology.4,10,11,36–39 Differences in these
This study found no difference in mandibular plane correlations might be attributed to factors such as
angle between Class II and Class III groups (Table 1). racial difference,39 although the association between
Therefore, the variation demonstrated here in height of craniofacial measurements and atlas morphology
Figure 3. h Frontal outer transverse diameter of C1 (mm); FOTDC1.

THE MORPHOLOGY OF CERVICAL VERTEBRAE 723
Figure 4. i Angle along axis line of the dens to occlusal plane; dens angle.
varied with sex and age. The growth directions in REFERENCES

cervical vertebrae and mandible growth are certainly 1. Kapandji IA. The Physiology of the Joints. Annotated
associated, although influences here of gender and Diagrams of the Mechanics of the Human Joints. Vol 3.
racial differences are possible.18 Moreover, this study The Trunk and the Vertebral Column. Edinburgh and
suggested that Japanese female Class II subjects had London: Churchill Livingstone; 1970–1974:170–251.
lower dorsal arch heights than Class III subjects. 2. Huggare J. Association between morphology of the first
cervical vertebra, head posture, and craniofacial structures.
Eur J Orthod. 1991;13:435–440.
CONCLUSIONS 3. Hellsing E, Reigo T, McWilliam J, Spangfort E. Cervical and
N This study confirmed past findings that Class II lumbar lordosis and thoracic kyphosis in 8, 11 and 15-year-
old children. Eur J Orthod. 1987;9:129–138.
subjects are significantly shorter in terms of atlas
4. Solow B, Tallgren A. Head posture and craniofacial
dorsal arch height. Our findings on the relationship of morphology. Am J Phys Anthropol. 1976;44:417–435.
anteroposterior skeletal pattern to the transverse 5. Solow B, Siersbaek-Nielsen S. Airway adequacy, head
diameter of C1 parallel the results of previous posture, and craniofacial morphology. Am J Phys Anthropol.
studies.17,40–43 1984;86:214–223.
N The atlas dorsal arch height of cervical vertebrae is 6. Sonnesen L, Kjaer I. Cervical column morphology in
patients with skeletal class III malocclusion and mandibular
affected by the anteroposterior skeletal pattern. overjet. Am J Orthod Dentofacial Orthop. 2007;132:427e7–
427e12.
7. Sonnesen L, Kjaer I. Anomalies of the cervical vertebrae in
ACKNOWLEDGMENT
patients with skeletal class II malocclusion and horizontal
This study was supported by a Showa University Research maxillary overjet. Am J Orthod Dentofacial Orthop. 2008;
Grant for Young Researchers. 133:188e15–188e20.

8. Hackney J, Bade D, Clawson A. Relationship between 24. Marmulla R, Wortche R, Muhling J, Hassfeld S. Geometric
forward head posture and diagnosed internal derangement accuracy of the NewTom 9000 Cone Beam CT. Dentomax-
of the temporomandibular joint. J Orofac Pain. 1993;7: illofac Radiol. 2005;34:28–31.
386–390. 25. Stratemann SA, Huang JC, Maki K, Miller AJ, Hatcher DC.
9. Sonnesen L, Bakke M, Solow B. Temporomandibular Comparison of cone beam computed tomography imaging
disorders in relation to craniofacial dimensions, head with physical measures. Dentomaxillofac Radiol. 2008;37:
posture and bite force in children selected for orthodontic 80–93.
treatment. Eur J Orthod. 2001;23:179–192. 26. Breiman RS, Beck JW, Korobkin M, Glenny R, Akwari OE,
10. Özbek MM, Köklü A. Natural cervical inclination and Heaston DK, Moore AV, Ram PC. Volume determinations
craniofacial structure. Am J Orthod Dentofacial Orthop. using computed tomography. AJR Am J Roentgenol. 1982;
1993;104:584–591. 138:329–333.
11. Festa F, Dolci M, Filippi M. Relationship between cervical 27. Huggare J. The first cervical vertebra as an indicator of
lordosis and facial morphology in Caucasian women with mandibular growth. Eur J Orthod. 1989;11:10–16.
skeletal Class II malocclusion: a cross-sectional study. 28. Kato C, Yamaguchi T, Watanabe M, Tomoyasu Y, Asama Y,
Cranio. 2003;21:121–129. Maki K. Cone-beam computed tomography evaluation of
12. Kau CH, Richmond S, Palomo JM, Hans MG. Three- cervical vertebra morphology in female patients with skeletal
dimensional cone beam computerized tomography in open bite. Hosp Dent (Tokyo). 2009;21:7–11.
orthodontics. J Orthod. 2005;32:282–293. 29. Solow B, Barrett MJ, Brown T. Craniocervical morphology
13. Hatcher DC, Aboudara CL. Diagnosis goes digital. and posture in Australian Aboriginals. Am J Phys Anthropol.
Am J Orthod Dentofacial Orthop. 2004;125:512–515. 1982;59:33–45.
14. Ludlow JB, Davies-Ludlow LE, Brooks SL, Howerton WB. 30. Cooke MS, Wei SH. Intersex differences in craniocervical
Dosimetry of 3 CBCT devices for oral and maxillofacial morphology and posture in southern Chinese and British
radiology: CB Mercuray, NewTom 3G and i-CAT. Dento- Caucasians. Am J Phys Anthropol. 1988;77:43–51.
maxillofac Radiol. 2006;35:219–226. 31. Grave B, Brown T, Townsend G. Comparison of cervicov-
15. Walker L, Enciso R, Mah J. Three-dimensional localization ertebral dimensions in Australian Aborigines and Cauca-
of maxillary canines with cone-beam computed tomography. sians. Eur J Orthod. 1999;21:127–135.
Am J Orthod Dentofacial Orthop. 2005;128:418–423. 32. Borden AG, Rechtman AM, Gershon-Cohen JG. The normal
16. Sukovic P. Cone beam computed tomography in craniofa- cervical lordosis. Radiology. 1960;74:806–809.
cial imaging. Orthod Craniofac Res. 2003;6(suppl 1):31–36; 33. Rechtman AM, Boreadis Borden AG, Gershon-Cohen J.
discussion 179–182. The lordotic curve is the cervical spine. Clin Orthop. 1961;
17. Naderi S, Cakmakçi H, Acar F, Arman C, Mertol T, Arda MN. 20:208–216.
Anatomical and computed tomographic analysis of C1 34. Hardacker JW, Shuford RF, Capicotto PN, Pryar PW.
vertebra. Clin Neurol Neurosurg. 2003;4:245–248. Radiographic standing cervical segmental alignment in adult
18. Ronaldo GV, Mohri T, Morita S, Hanada K. Cephalometric volunteers without neck symptoms. Spine. 1997;22:1472–
study on the morphology of the first cervical vertebra to 1480.
craniofacial structure. Part 1: Is there any relationship? 35. Huggare JÅV, Laine-Alava MT. Nasorespiratory function
Niigata Dent J. 1996;26:13–24. and head posture. Am J Orthod Dentofacial Orthop. 1997;
19. Akimoto Y, Noma H, Takaki T. A study of the relationship 112:507–511.
between the occlusal plane and the dens of the axis by 36. Gresham H, Smithells PA. Cervical and mandibular posture.
lateral cephalography, regarding individual-normal occlusion Dental Record. 1954;74:261–264.
persons and patients with jaw deformity. Jpn J Jaw Deform. 37. Bench RW. Growth of the cervical vertebrae as related to
2000;10:89–98. tongue, face, and denture behavior. Am J Orthod. 1963;49:
20. Hilgers ML, Scarfe WC, Scheetz JP, Farman AG. Accuracy 183–214.
of linear temporomandibular joint measurements with cone 38. Marcotte MR. Head posture and dentofacial proportions.
beam computed tomography and digital cephalometric Angle Orthod. 1981;51:208–213.
radiography. Am J Orthod Dentofacial Orthop. 2005;128: 39. Nezu H, Nagata K, Yoshida Y, Kosaka H, Kikuchi M.
803–811. Cephalometric comparison of clinical norms between the
21. Lascala CA, Panella J, Marques MM. Analysis of the Japanese and Caucasians. J Jpn Orthod Soc. 1982;41:
accuracy of linear measurements obtained by cone beam 450–465.
computed tomography (CBCT-NewTom). Dentomaxillofac 40. Doherty BJ, Heggeness MHO. The quantitative anatomy of
Radiol. 2004;33:291–294. the atlas. Spine. 1994;19:2497–2500.
22. Kobayashi K, Shimoda S, Nakagawa Y, Yamamoto A. 41. Francis CC. Variations in the articular facets of the cervical
Accuracy in measurement of distance using limited cone- vertebrae. Anat Res. 1995;122:589–602.
beam computerized tomography. Int J Oral Maxillofac 42. Francis CC. Dimensions of the cervical vertebrae. Anat Res.
Implants. 2004;19:228–231. 1995;122:603–609.
23. Misch KA, Yi ES, Sarment DP. Accuracy of cone beam 43. Williams PL, Warwick R, Dyson A, Bannister LH. In: Gray’s
computed tomography for periodontal defect measure- Anatomy. Norwick, UK: Churchill Livingstone; 1998:317–
ments. J Periodontol. 2006;77:1261–1266. 318.

Original Article
Accuracy and reliability of palatal superimposition of three-dimensional

digital models
Dong-Soon Choia; Young-Mok Jeongb; Insan Jangc; Paul George Jost-Brinkmannd;
Bong-Kuen Chae
ABSTRACT
Objective: To evaluate the accuracy of the superimposition of three-dimensional (3D) digital
models using the palatal surface as a reference for measuring tooth movements.
Materials and Methods: Maxillary plaster models were selected from 20 patients. The right and
left canines, premolars, and molars were individually cut underneath the gingival margins and set
up in wax (plaster model 1 5 PM1). The PM1s were scanned to create 3D digital models (digital
model 1 5 DM1). Teeth on the PM1s were randomly moved (plaster model 2 5 PM2) and
subsequently scanned to produce another set of 3D digital models (digital model 2 5 DM2). DM1s
and DM2s were superimposed using the palatal area as reference via surface-to-surface matching
software, and the changes in tooth movement were calculated. In the plaster models, the tooth
movements were directly measured using the Reference Measurement Instrument. A paired t-test
and a correlation analysis were performed to determine whether the two measurement methods
differed significantly.
Results: The means of the anteroposterior (x-axis), transverse (y-axis), and vertical (z-axis) tooth
movements of the plaster models and the digital models did not differ significantly, and very high
correlations were found between the plaster models and the digital models.
Conclusion: From a technical point of view, the superimposition of 3D digital models using the
palatal surface provides accurate and reliable measurements, but it remains to be investigated how
stable the palatal surface is longitudinally after growth and/or orthopedic treatment take place.
(Angle Orthod. 2010;80:685–691.)
KEY WORDS: Three-dimensional digital model; Superimposition
INTRODUCTION
Assistant Professor, Department of Orthodontics, College of
a
Changes in tooth positions following orthodontic
Dentistry, Gangneung-Wonju National University, Gangneung,
South Korea. treatment have traditionally been evaluated by super-
b
PhD Graduate Student, Department of Orthodontics, College imposition of serial cephalometric radiographs. How-
of Dentistry, Gangneung-Wonju National University, Gang- ever, this method has a number of drawbacks:
neung, South Korea. difficulties in evaluating three-dimensional (3D) tooth
c
Full-Time Lecturer, Department of Orthodontics, College of
movements, problems with identifying inherent land-
Dentistry, Gangneung-Wonju National University, Gangneung,
South Korea. marks,1 tracing errors, and frequent radiation expo-
d
Professor, Department of Orthodontics, Dentofacial Ortho- sure.2
pedics and Pedodontics, Center for Dental and Craniofacial The plaster model is the traditional 3D patient record
Sciences, Charité-Universitätsmedizin Berlin, Berlin, Germany. for measuring linear changes in the dental arch. To
e
Professor, Department of Orthodontics, College of Dentistry,
Gangneung-Wonju National University, Gangneung, South analyze tooth movements, accurate superimposition of
Korea. serial models on a stable and identifiable structure is
Corresponding author: Prof Bong-Kuen Cha, Department of necessary. Many studies have reported on the stability
Orthodontics, College of Dentistry, Gangneung-Wonju National of the palatal rugae as reference points for the
University, 120 Gangneung Daehangno, Gangneung City,
comparison of the pretreatment and posttreatment
Gangwon Province, 210-702, South Korea
(e-mail: korth@gwnu.ac.kr) conditions on plaster models.3–7 Unfortunately, these
studies did not provide important information about the
Accepted: December 2009. Submitted: October 2009.
G 2010 by The EH Angle Education and Research Foundation, structural and volumetric changes in the palate or
Inc. regarding 3D orthodontic tooth movements.

686 CHOI, JEONG, JANG, JOST-BRINKMANN, CHA
Digital 3D models have become standard technolo-

gy.8,9 Digital models have a number of advantages in
terms of storage, retrieval, diagnostic versatility,
transferability, and durability.10,11 Numerous studies
have shown that 3D digital models can be used for
model analysis and diagnosis,10,12–16 treatment plan-
ning,8,17 design and manufacture of orthodontic appli-
ances,18–20 and evaluation of tooth movement.21–24 To
date, however, studies of the reliability of computed
superimposition of 3D digital models to assess the
outcomes of orthodontic treatments have been limit-
ed.22,23
Our previous publication suggests that 3D digital
model superimposition using the palatal surface as a
reference is as reliable as cephalometric superimpo-
sition for assessing orthodontic tooth movement in Figure 1. Measurement points (blue dots) and established coordi-
nate system on a plaster model.
maxillary premolar extraction cases.22 Nevertheless, it
remains to be investigated how accurate digital
superimposition techniques are for the quantification (RMI) (RMI 550, SAM Präzisionstechnik, Munich,
of tooth movements. Consequently, it was the purpose Germany), which has three digital calipers with a
of this study to evaluate the accuracy of mathematical resolution of 0.01 mm in the x-, y-, and z-axes and a
superimposition using the constant palatal surfaces as cone-shaped measuring tip with a diameter of 0.5 mm
a reference for measuring the changes in tooth (Figure 2). The plaster models were mounted on the
movement on 3D digital models in comparison with RMI, while the occlusal plane was kept parallel to the
the actual tooth movement in setup models. floor of the RMI, and the midpalatal suture was fit into
the x-axis of the RMI (Figure 2).
MATERIALS AND METHODS On each cast (PM1), a coordinate system was
The protocols of this study were reviewed and constructed according to Ashmore et al.,21 with the
approved by the ethics committee of Gangneung- junction of the incisive papilla and the palatine raphe
Wonju National University Dental Hospital (IRB 2009- as the origin (0, 0, 0), which resulted in the x-, y-, and
11). Posttreatment maxillary plaster models of 20 z-axes (Figure 1). A movement in the positive direction
patients were randomly selected from the archive of along the x-axis indicated mesial movement. Positive
the Department of Orthodontics of Gangneung-Wonju values in the y- and z-axes indicated right and
National University Dental Hospital. The inclusion extrusive tooth movements, respectively (Figures 1
criterion were that the plaster models have (1) and 2). The cusp tip of the canine, the first premolar,
permanent dentition, (2) complete dentition from the and the first molar of the PM1 and the origin (0, 0, 0)
central incisors to the second molars, and (3) no were marked with black pencil. After the digital calipers
porosities on the teeth and the palatal surfaces. There were set in the x-, y, and z-axes at zero at the origin of
was no consideration of age and gender. The right and PM1, the distances from the cusp tip of the canine, the
left canines, premolars, and molars were individually first premolar, and the first molar to the origin were
cut underneath the gingival margins and set up in wax measured. PM1 was removed from the RMI, and PM2
(plaster model 1 5 PM1) (Figure 1). PM1s were was placed into the same position on the RMI. The
scanned to reconstruct 3D digital models (digital model distance from the cusp tips to the origin was measured
1 5 DM1). The teeth on the PM1s were randomly again on PM2. The differences between PM1 and PM2
moved (plaster model 2 5 PM2) and scanned again to were calculated.
produce another set of 3D digital models (digital model
2 5 DM2). Measurements were performed on the cusp Measurements on the 3D Digital Models
tip of the canines, the buccal cusp tip of the first
Three-dimensional scanning of the plaster models
premolars, and the mesiobuccal and mesiopalatal
was performed using the Orapix 3D scanner (laser slit–
cusp tips of the first molars (Figure 1).
type noncontact 3D scanner, Orapix Co Ltd, Seoul,
South Korea; accuracy 6 20 mm) and a 3D reverse
Measurements on the Plaster Models
modeling software program (Rapidform 2002, INUS
Measurements on the plaster models were per- Technology Inc, Seoul, South Korea) (Figure 3). The
formed with the Reference Measurement Instrument same coordinate system that had been used for PM1

RELIABILITY OF SUPERIMPOSITION OF DIGITAL MODELS 687
Figure 2. Measurement of the plaster model with the Reference Measurement Instrument (RMI). (A) Digital calipers in the x-, y-, and z-axes. (B)
Plaster model, which could be placed repeatedly in the same position on the RMI. (C) Measuring tip on the canine.
was established on DM1 (Figure 3). To measure tooth least 5 mm from the gingival margins of the posterior
movement, DM1 and DM2 were superimposed on the teeth bilaterally. The distal margin did not extend
surface across the palate. The area of superimposition distally beyond the line in contact with the distal
is presented in Figure 4. It included the first, second, surfaces of the maxillary second molars bilaterally.
and third palatal rugae, but the nasopalatine papilla This procedure, designated as 3D surface-to-surface
was excluded. The lateral margins were located at matching (best-fit method), employed a least-mean-
square technique using a function of Rapidform
2002.9,22,25,26 Analogous to what is shown in Figure 5,
the distances of all eight measuring points described
previously between DM1 and DM2 were calculated
along the x-, y-, and z-axes.
Figure 3. The three-dimensionally scanned digital model. The digital

model and the plaster model employed the same coordinate system. Figure 4. The reference region for the palatal superimposition.22

on the plaster models were compared with those

measured on the superimposed 3D digital models. A
paired t-test and a correlation analysis were performed
to determine whether the two measurement methods
differed significantly.
To determine identification errors of the same points
on the plaster model and the digital model, one
examiner measured the 3D distance from the origin
to the cusp tip of the right and left canines, first and
second premolars, and first and second molars (total
of 10 points) and repeated the measurements 2 weeks
later. The mean differences on the plaster model were
0.04 mm, 0.07 mm, and 0.08 mm along the x-, y-, and
z-axes, respectively. On the digital model, the mean
difference was 0.01 mm along all three axes.
Figure 5. Measurement of the superimposed 3D digital models. RESULTS

Point 1 indicates the mesiobuccal cusp tip of the left first molar of first
digital model (DM1); point 2 is that of the second digital model (DM2). Table 1 shows the mean differences between the
measurements of the plaster models and the digital
models. The P values from the paired t-test assessed
Statistical Analysis and Error Test
whether or not the plaster and digital models yielded
The mean anteroposterior (x-axis), transverse (y- equivalent mean values of the tooth movements. The
axis), and vertical (z-axis) tooth movements measured P values indicated that the means of the anteropos-
Table 1. Differences in the Tooth Movements Evaluated on Superimposed 3D Digital Models and Plaster Models, Paired t-Tests, and Pearson
Correlation Coefficients
Difference (plaster model – digital model) Paired t-Test Pearson Correlation
a
Location Mean (mm) SD (mm) t P r P
Anteroposterior (x-axis)
Right canine 20.02 0.15 20.695 .495 0.991 .000
Right first premolar 20.04 0.17 21.139 .269 0.985 .000
Right first molar (MB) 20.03 0.17 20.751 .462 0.984 .000
Right first molar (MP) 0.05 0.19 1.258 .224 0.979 .000
Left canine 0.03 0.12 1.097 .286 0.995 .000
Left first premolar 0.04 0.20 0.941 .359 0.969 .000
Left first molar (MB) 20.04 0.12 21.504 .149 0.994 .000
Left first molar (MP) 20.04 0.14 21.149 .265 0.986 .000
Transverse (y-axis)
Right canine 20.03 0.00 20.800 .434 0.994 .000
Left canine 0.04 0.20 0.940 .359 0.994 .000
Vertical (z-axis)
Right canine 0.00 0.17 20.104 .918 0.988 .000
Left canine 20.01 0.13 20.193 .849 0.996 .000
a
MB indicates mesiobuccal cusp; MP, mesiopalatal cusp.

terior (x-axis), transverse (y-axis), and vertical (z-axis) complicated measurement process, and two-dimen-
tooth movements of the plaster and digital models did sional measurement of the 3D curvature of the palatal
not differ significantly. vault.22,25 Ashmore et al.21 employed a mechanical 3D
The Pearson correlation coefficients of the plaster digitizer for a 3D analysis of molar movements during
and digital models are shown in Table 1. The headgear treatment. Miller et al.24 superimposed 3D
correlation analysis revealed that the r values of all digital models to evaluate orthodontic treatment
the variables were very high (highest [0.998] for the y- outcomes in three dimensions, again using the palatal
axis movement of the left first molar and lowest [0.969] rugae as a reference structure.
for the x-axis movement of the left first premolar). There seems to be no consensus on the stability of
Figure 6 shows a scatter plot and regression lines for the palatal rugae as to the effect of growth or
the tooth movements along each axis, as determined treatment. Friel28 demonstrated in a study that the
on the plaster and digital models. Good correlations teeth move forward in relation to the palatal rugae in
were revealed again for all the tooth movements. This conjunction with growth of the jaws. Simmons et al.,29
means that the measurements of the tooth movements in a longitudinal study (from primary dentition to young
were the same whether they were measured directly adult) of the anteroposterior stability of the medial
from the plaster models or by palatal superimposition rugae region, concluded that the medial rugae
on the 3D digital models. landmarks did not appear to be stable reference points
for investigation of tooth migration. Future research
DISCUSSION should evaluate the 3D positional stability of the palatal
rugae using another stable reference plane. The
Traditional two-dimensional cephalometric radio- orthodontic miniscrew may serve as an alternative
graphs have played an important role in evaluating reference landmark, but only in a limited number of
the results of orthodontic treatment. However, cepha- cases.22,23,25 Jang et al.23 evaluated the stability of
lometric evaluation involves difficulties in measuring palatal rugae using digital models superimposed on
tooth movements and identifying inherent landmarks in three miniscrews as registration landmarks and con-
all three dimensions.1,2 cluded that the medial points of the third palatal rugae
Plaster models have been an essential component and the shape of the palatal vault were stable
of 3D diagnostic records in the orthodontic treatment throughout orthodontic treatment with premolar ex-
procedure. The palatal rugae form their pattern by the traction.
12th to 14th week of prenatal life and are reasonably In the present study, only a few rugae points21,24
stable during a person’s growth.27 Thus they may serve were not used, but the entire palatal vault including the
as a suitable reference structure when studying serial rugae22,23,25 was used as a reference landmark to
models. Many authors have investigated the use of the support the hypothesis that the so-called best-fit
palatal rugae as reference points for measuring tooth method using the palatal surface could be used for
movements on serial dental casts3–7 and on 3D digital accurate superimposition of serial 3D digital models.
models.21,23,24 In a study of changes occurring in 15 Advanced technologies such as 3D scanning, 3D
patients who underwent extraction of four premolars, reverse technology for the construction of the digital
Peavy and Kendrick6 reported that the lateral ends of model, and surface-to-surface matching technology
the rugae close to the teeth followed the teeth in the were applied in this study.
sagittal plane, while the so-called O point on the In the present study, the mean anteroposterior,
midsagittal plane was least affected. Van der Linden7 horizontal, and vertical tooth movements measured by
evaluated changes in rugae and interrugal dimensions the palatal superimposition in the 3D digital models did
in 65 normally growing children (aged 6 to 16 years) not statistically differ from those directly obtained from
and in six orthodontically treated patients. The author the plaster models (Table 1). Moreover, there was a
noted little or no change in the length of the individual high correlation between the two methods (Table 1
rugae and interrugal distances. Almeida et al. 3 and Figure 6). These results suggest that the super-
suggested that the transverse offsets and distances imposition of 3D digital models using surface-to-
between the medial rugae points are generally stable, surface matching technology in the palatal area can
particularly for the first rugae. Hoggan and Sadowsky5 result in accurate and reliable measurements for the
reported that the medial and lateral ends of the third assessment of orthodontic tooth movements. The
palatal rugae could be used as reliably as cephalo- present study investigated the accuracy of the best-fit
metric superimposition to assess anteroposterior tooth method when identical palatal surfaces were scanned
movements. However, the evaluation of tooth move- twice and superimposed. Whether similar accuracy
ments on plaster models has many clinical drawbacks, can be achieved when repeated impressions are made
such as difficulties in establishing reference points, the in growing patients remains to be determined.

Figure 6. Scattergrams and regression lines for the tooth movements measured on the plaster models and the 3D digital models.
Some doubt remains about the validity of direct 2. Ghafari J, Baumrind S, Efstratiadis SS. Misinterpreting
measurement by means of the RMI. Further cephalo- growth and treatment outcome from serial cephalographs.
Clin Orthod Res. 1998;1:102–106.
metric studies may be required to assess actual tooth 3. Almeida MA, Phillips C, Kula K, Tulloch C. Stability of the
movement in animals after orthodontic tooth move- palatal rugae as landmarks for analysis of dental casts.
ment and directly by instruments such as the RMI Angle Orthod. 1995;65:43–48.
device. In addition, the growth-dependent stability of 4. Bailey LT, Esmailnejad A, Almeida MA. Stability of the
the palatal surface in growing patients, as well as palatal rugae as landmarks for analysis of dental casts in
extraction and nonextraction cases. Angle Orthod. 1996;66:
evidence of the stability of the area in subjects treated
73–78.
with expansion mechanics or in the mandible, have not 5. Hoggan BR, Sadowsky C. The use of palatal rugae for the
yet been fully explained. We are studying a possible assessment of anteroposterior tooth movements. Am J
landmark for the superimposition of mandibular digital Orthod Dentofacial Orthop. 2001;119:482–488.
models. 6. Peavy DC Jr, Kendrick GS. The effects of tooth movement
Virtual study models can replace conventional study on the palatine rugae. J Prosthet Dent. 1967;18:536–542.
7. van der Linden FP. Changes in the position of posterior
casts for many purposes, such as model analysis, teeth in relation to ruga points. Am J Orthod. 1978;74:
diagnosis, diagnostic setup, and treatment planning. 142–161.
Moreover, with the superimposition method used in the 8. Motohashi N, Kuroda T. A 3D computer-aided design
present study, it seems promising that, in the future, a system applied to diagnosis and treatment planning in
simple mouse click will enable computer-assisted orthodontics and orthognathic surgery. Eur J Orthod. 1999;
21:263–274.
evaluation of 3D tooth movements. This knowledge
9. Commer P, Bourauel C, Maier K, Jager A. Construction and
will form the basis for future studies of the effects of testing of a computer-based intraoral laser scanner for
multiple impressions or intraoral optical scans and determining tooth positions. Med Eng Phys. 2000;22:
growth of the palatal vault on the appropriateness of 625–635.
using the palatal vault for best-fit superimposition. 10. Asquith J, Gillgrass T, Mossey P. Three-dimensional
imaging of orthodontic models: a pilot study. Eur J Orthod.
2007;29:517–522.
CONCLUSIONS
11. Joffe L. OrthoCAD: digital models for a digital era. J Orthod.
The best-fit mathematical superimposition method of 2004;31:344–347.
12. Cha BK, Choi JI, Jost-Brinkmann PG, Jeong YM. Applica-
maxillary casts on the identical palatal vault is very
tions of three-dimensionally scanned models in orthodon-
accurate and allows for 3D evaluation of tooth tics. Int J Comput Dent. 2007;10:41–52.
movement. It remains to be investigated how stable 13. Costalos PA, Sarraf K, Cangialosi TJ, Efstratiadis S.
the palatal surface is longitudinally after growth and/or Evaluation of the accuracy of digital model analysis for the
orthopedic treatment take place. American Board of Orthodontics objective grading system
for dental casts. Am J Orthod Dentofacial Orthop. 2005;128:
624–629.
ACKNOWLEDGMENT 14. Gracco A, Buranello M, Cozzani M, Siciliani G. Digital and
This study was supported by the 2007 research fund of plaster models: a comparison of measurements and times.
Gangneung-Wonju National University (2007-0102). Prog Orthod. 2007;8:252–259.
15. Mullen SR, Martin CA, Ngan P, Gladwin M. Accuracy of
space analysis with emodels and plaster models.
REFERENCES
1. Trpkova B, Major P, Prasad N, Nebbe B. Cephalometric 16. Quimby ML, Vig KW, Rashid RG, Firestone AR. The
landmarks identification and reproducibility: a meta-analysis. accuracy and reliability of measurements made on comput-
Am J Orthod Dentofacial Orthop. 1997;112:165–170. er-based digital models. Angle Orthod. 2004;74:298–303.

17. Macchi A, Carrafiello G, Cacciafesta V, Norcini A. Three- dimensional tooth movement during orthodontic treatment.
dimensional digital modeling and setup. Am J Orthod Angle Orthod. 2009;79:447–453.
Dentofacial Orthop. 2006;129:605–610. 24. Miller RJ, Kuo E, Choi W. Validation of Align Technology’s
18. Fischer-Brandies H, Orthuber W, Laibe J, Menzel E. Treat III digital model superimposition tool and its case
Continuous arch wire technique using the bending art application. Orthod Craniofac Res. 2003;6(suppl 1):143–
system. J Orofac Orthop. 1997;58:198–205. 149.
19. Owen AH III. Accelerated Invisalign treatment. J Clin 25. Cha BK, Lee JY, Bae SH, Park DI. Preliminary study of
Orthod. 2001;35:381–385. future orthodontic model analysis: the orthodontic applica-
20. Sachdeva RC. SureSmile technology in a patient- tion of 3D reverse engineering technologies. Taehan
centered orthodontic practice. J Clin Orthod. 2001;35: Chikkwa Uisa Hyophoe Chi. 2002;40:107–117.
245–253. 26. McDonagh S, Moss JP, Goodwin P, Lee RT. A prospective
21. Ashmore JL, Kurland BF, King GJ, Wheeler TT, Ghafari J, optical surface scanning and cephalometric assessment of
Ramsay DS. A 3-dimensional analysis of molar movement the effect of functional appliances on the soft tissues.
during headgear treatment. Am J Orthod Dentofacial Eur J Orthod. 2001;23:115–126.
Orthop. 2002;121:18–29, discussion 29–30. 27. Lebret L. Growth changes of the palate. J Dent Res. 1962;
22. Cha BK, Lee JY, Jost-Brinkmann PG, Yoshida N. Analysis 41:1391–1404.
of tooth movement in extraction cases using three-dimen- 28. Friel S. Migration of teeth. Dent Rec (London). 1949;69:
sional reverse engineering technology. Eur J Orthod. 2007; 74–84.
29:325–331. 29. Simmons JD, Moore RN, Erickson LC. A longitudinal study
23. Jang I, Tanaka M, Koga Y, Iijima S, Yozgatian JH, Cha BK, of anteroposterior growth changes in the palatine rugae.
Yoshida N. A novel method for the assessment of three- J Dent Res. 1987;66:1512–1515.

Original Article
Craniofacial growth in ectodermal dysplasia

An 8 year longitudinal evaluation of Italian subjects
Claudia Dellaviaa; Francesca Cattib; Chiarella Sforzac; Davide G. Tommasid;

Virgilio Ferruccio Ferrarioe
ABSTRACT
Objective: To identify the main directions of growth of facial structures in subjects with
hypohidrotic ectodermal dysplasia (HED).
Materials and Methods: The 3D noninvasive facial measurements were collected in 12 subjects
(6 boys, 6 girls) with HED during four assessments (at 8, 11, 12, and 15 years) using an
electromagnetic digitizer. The modifications of linear distances in the upper, middle, and lower third
of the face were analyzed and compared with cross-sectional data obtained in normal healthy
coetaneous. For each distance, differential values between the last and the initial data were
calculated individually, separately for a first (8–11 years) and a second growth period (12–15
years).
Results: In the first time span, the growth of all facial measurements was reduced in HED subjects
compared with control subjects. During this interval, most of the HED children underwent a
functional and/or prosthetic treatment. During adolescence, the width and height of the lower and
upper facial thirds showed a larger growth in HED subjects than in control subjects, while all facial
depths and all distances in the middle facial third maintained a reduced growth.
Conclusions: The deviation from normal facial growth of HED subjects tends to lessen with age.
Functional and prosthetic appliances may have enhanced facial growth. (Angle Orthod.
2010;80:733–739.)
KEY WORDS: Growth; Ectodermal dysplasia; Longitudinal; Face; 3D
INTRODUCTION
Craniofacial growth is the result of multiple interac-
tions between genetic and epigenetic elements,
involving both soft and hard tissue structures. During
a
Assistant Professor, Department of Human Morphology and
Biomedical Sciences ‘‘Città Studi,’’ School of Dentistry, Uni- growth, the epithelium and the mesenchyme undergo
versità degli Studi di Milano, Milano, Italy. a continuous development with a cascade of reciprocal
b
Research Assistant, Department of Human Morphology and inductions to finally construct an overall harmonic
Biomedical Sciences ‘‘Città Studi,’’ School of Dentistry, Uni- complex.1–3 Genetic modifications may cause abnor-
versità degli Studi di Milano, Milano, Italy.
malities in any phase of this morphogenetic process,
c
Professor and Department Chair, Department of Human
Morphology and Biomedical Sciences ‘‘Città Studi,’’ School of thus resulting in a nonharmonious facial morphology
Dentistry, Università degli Studi di Milano, Milano, Italy. with associated functional and esthetic impairments.4,5
d
Post Doctoral Fellow, Department of Human Morphology The craniofacial structures derived from ectoderm
and Biomedical Sciences ‘‘Città Studi,’’ Università degli Studi di and neural crests—more rarely also from mesodermal
Milano, Milano, Italy.
e
Professor, Department of Human Morphology ‘‘Città Studi,’’ and endodermal layers—are altered in subjects
School of Dentistry, Università degli Studi di Milano, Milano, affected by ectodermal dysplasia (ED).
Italy. ED is a rare group of genetic syndromes inherited by
Corresponding author: Dr Claudia Dellavia, Università degli autosomal recessive, autosomal dominant, or x-linked
Studi di Milano, Human Morphology and Biomedical Sciences
‘‘Città Studi,’’ Via Mangiagalli 31 Milano, Italy 20133
recessive transmission.3,6,7 More recently, molecular
(e-mail: claudia.dellavia@unimi.it) analyses have identified the mutations of genes
responsible for about 50 types of ED that are involved
Accepted: December 2009. Submitted: October 2009.
G 2010 by The EH Angle Education and Research Foundation, in (1) cell adhesion, (2) transcription regulation, (3)
Inc. cell-cell signaling, (4) development, and (5) other

734 DELLAVIA, CATTI, SFORZA, TOMMASI, FERRARIO
functions (eg, structural proteins, placode forma- The current study aimed to identify the actual
tion).3,6,7 Future advances in cell biology and embryo- directions of growth of the facial structures in young
genic pathways could allow a reclassification of subjects with HED by analyzing the modifications of
ectodermal dysplasia according to the functions of linear measurements in the upper, middle, and lower
mutated genes.6,7 third of the face. The morphometric evaluation was
The most frequent form of ED is hypohidrotic performed noninvasively using a 3D computerized
ectodermal dysplasia (HED; OMIM 305100) charac- digitizer on the facial soft tissues of HED and reference
terized by hypotrichosis, hypodontia, and hypohidro- subjects during an 8-year period of growth. The
sis with major manifestations in the male sex. The method has already been proved to be reliable in the
involved genes encode a collagenous transmem- quantitative assessment of craniofacial variations in
brane protein—ectodysplasin—and its two receptors, both normal and syndromic patients.4,5,10,11,18
regulating the epithelial-mesenchymal interactions
and the hair follicle morphogenesis.3,7 Affected MATERIALS AND METHODS
patients present with a typical ‘‘aged-face’’ associat-
Twelve white Italian subjects diagnosed as having
ed with prominent forehead and chin, saddle nose,
HED (six boys and six girls) were analyzed. Subjects
everted lips, sunken cheeks, periorbital wrinkles,
were referred for examination by the Italian National
high-set orbits, large and low set ears, small hard
Ectodermal Dysplasia Association (ANDE). No subject
tissue palate, hypoplasia of the alveolar process, and
had undergone any previous craniofacial surgical
multiple agenesis of both primary and permanent
procedure. In all HED subjects, 3D noninvasive facial
teeth.4,5,8–15
measurements and a dental formula (only erupted
Clinical management of such patients with craniofa-
teeth at clinical examination excluding third molars)
cial deformities and functional alterations should
were collected by the same expert operator during four
consider the quantitative assessment of the dimen-
assessments (at 8, 11, 12, and 15 years). For the
sions, reciprocal spatial positions, and relative propor-
longitudinal evaluation, data were computed separate-
tions of the facial structures during growth to intercept
ly for a first (8–11 years) and a second (12–15 years)
deviations from the norm and possibly correct them at
growth period.
the appropriate time.10,15,16 Detailed knowledge of the
Reference cross-sectional data were recorded in
typical growth pattern of HED subjects could provide
previous investigations performed by the staff of the
the clinicians useful information to plan a multidisci-
Functional Anatomy Research Center (FARC) at the
plinary specific treatment.
University of Milan on 160 healthy subjects of the
Most previous reports investigated the facial fea-
same ethnic group (40 subjects for each age and sex
tures of HED by cross-sectional analyses,4,5,8,9,11–13,15
subgroup). Control subjects did not have a previous
and few longitudinal studies are currently available.10,17
history of craniofacial trauma or surgery and congenital
In their cephalometric evaluation of ED young patients,
anomalies. Part of their data had already been
Bondarets et al.17 observed a peculiar trend of
published.18,19
craniofacial growth towards a retrognathic maxilla
The parents or legal guardians of all the analyzed
and Class III sagittal relationships of the jaws. A more
individuals gave their informed consent to participate in
recent longitudinal study, performed by our research
the analysis. All procedures were noninvasive, did not
group, compared the growth of HED young patients
provoke damages, risks, or discomfort to the subjects,
with that of healthy reference peers and found a global
and were preventively approved by the local ethics
reduction of all facial volumes in the syndromic
committee in accordance with the ethical principles of
subjects during childhood.10 Nevertheless, facial vol-
the World Medical Association Declaration of Helsinki
umes increased their growth by time and, at the end of
(version, 2002).
adolescence, the analyzed HED subjects had similar
growth patterns of facial volumes compared to their
Data Collection
reference peers and nearly double that of nonrehabili-
tated HED subjects from the study of Bondarets et al.17 The data collection procedure and subsequent off-
It has been hypothesized that early orthodontic and line calculations were previously published in detail.19
prosthetic devices worn by the analyzed HED subjects In summary, a single experienced operator located
could have improved masticatory function and promot- and marked 50 landmarks on the cutaneous facial
ed the growth of the middle third and lower third facial surface of each subject. During landmark marking, the
structures to levels above those found in untreated children sat relaxed with a natural head position and
subjects with HED.10 Besides, facial volumes give an the teeth in an intercuspal position. For each child, this
estimate of the overall structures, but they do not phase lasted about 5 minutes. Then, all subjects
provide the vectorial directions of growth.11 wearing removable prosthetic devices were asked to

FACIAL GROWTH AND ECTODERMAL DYSPLASIA 735
Figure 1. Facial distances in a 12-year-old HED boy. (a) Facial depths. (b) Facial widths. (c) Facial heights. The landmarks used are labeled.
remove the appliance for data digitization lasting about collection were calculated individually, separately for
60 seconds. The x, y, and z coordinates of the facial the first (D1, 8–11 years) and the second growth period
landmarks were recorded with a computerized elec- (D2, 12–15 years). For further comparisons, differential
tromagnetic digitizer (3Draw, Polhemus Inc, Colches- data of both growth intervals were then normalized as
ter, Vt) and analyzed using customized computer a percentage of the initial values.
algorithms written by one of the authors. The repro- Mean differential normalized values were computed
ducibility of landmark identification and digitization separately for boys and girls in each growth period.
were previously reported and found to be reliable.10 Differential normalized data were also computed for
The following soft tissue landmarks were used: the reference subjects, using measurements obtained
in boys and girls of comparable ages.
(1) Midline landmarks: tri, trichion; n, nasion; sn,
For each facial distance, the differential normalized
subnasale; pg, pogonion.
growths (mean normalized D1 and D2) were visualized
(2) Paired landmarks (right and left side, noted r and
through polar or nysquit diagrams, which summarize
l): exr, exl, exocanthion; tr, tl, tragion; gor, gol,
the quantitative variation of the considered parameters
gonion.
of HED compared with average data obtained in
reference subjects. The polar diagram provides a
Data Analysis quantitative overview of several measuring points in
one circular normalized form as all the segments
The 3D coordinates of the landmarks obtained on
approximate a circle. Therefore, diagrams were
each subject were used to estimate linear distances
composed of two circles: the inner circle (circle 1)
(unit: mm) of depth, width, and height of the three facial
provided the global mean growth of the reference
thirds (Figure 1) as reported in Table 1.
group, and the outer circle (circle 2) represented a
For each facial measurement, differential values
200% increment of the reference global mean growth.
between data from the last and the initial data
The origin of the axes (marked as 0) represented a null
reference global mean growth.
Table 1. Measurements Calculated From the Digitized Landmarks The global differential normalized growths of HED
Distances Landmarks subjects in the analyzed time span were displayed on
Upper facial depth (Du) n2(tr2tl) polar diagrams separately for boys and girls. The
Upper facial width (Wu) exr2exl segments included between 0 and circle 1 indicated a
Upper facial height (Hu) tri2n reduced growth of HED subjects compared with
Middle facial depth (Dm) sn2(tr2tl)
Middle facial width (Wm) tr2tl
reference subjects; by contrast, segments included
Middle facial height (Hm) n2sn between circle 1 and circle 2 indicated an increased
Lower facial depth (Dl) pg2(tr2tl) growth of HED subjects. No inferential statistics were
Lower facial width (Wl) gor2gol performed because of the limited number of analyzed
Lower facial height (Hl) sn2pg
HED subjects.

Figure 3. Intraoral photograph in an 8-year-old girl with HED

showing a complete removable mandibular prosthesis and a partial,
removable maxillary denture.
the data of reference peers, differential values from the

initial and last examinations were obtained and
normalized.
Figure 2. Panoramic radiograph of one HED boy at (a) 8 and (b) In the first time interval (8–11 years), the growth of
11 years.
all facial measurements was reduced in HED subjects
in comparison with control subjects of the same age
RESULTS and sex. From 12 to 15 years, upper facial third width
and height increased in HED subjects compared with
At oral examination, hypodontia was found in all
reference peers of both sexes, while upper facial third
HED subjects with a large variability in the number and
depth was reduced. The growth of the middle facial
distribution of the erupted permanent teeth. Globally,
third was reduced in all spatial dimensions in the HED
the upper central incisors were conserved in 62% of
subjects. In the lower third, the growth of facial depth
the analyzed children and had a conical shape in
was reduced, while that of facial height was increased
seven patients. Half of the HED individuals presented
in HED subjects of both genders; by contrast, facial
almost three first molars. About 25% of the HED
width growth was strictly increased in girls, but reduced
subjects had pegged maxillary canines. Radiographic
in HED boys. The mean differential normalized growth
images were available only in one syndromic boy and
of facial distances in HED subjects compared to
revealed agenesis of multiple primary and permanent
reference subjects is visualized in Figure 4.
teeth (Figure 2).
In the first growth period, the number of teeth ranged
DISCUSSION
from 9 to 13 in girls and from 4 to 10 in boys (at 8
years, mean 6 SD: 9.86 6 3.02; at 11 years, mean Previous anthropometric and cephalometric cross-
value 6 SD: 12.28 6 5.85). In the second growth sectional assessments revealed individuals with HED
period, the number ranged from 12 to 24 in girls and having distinct ‘‘facies’’ with marked deviations from
from 8 to 11 in boys (at 12 years, mean 6 SD: 12.83 6 the norm.4,5,8,11–13,15,16,20,21 A recent longitudinal investi-
5.41; at 15 years, mean 6 SD: 14.50 6 5.16). Globally, gation found a global reduction of all facial volumes in
girls with HED had more teeth than boys. the HED subjects compared with reference individuals
During the analyzed time span, all the syndromic during childhood, but some improvements in the
subjects, except for three girls, underwent orthodontic maxillary and mandibular size were reported by time.10
treatment with both orthopedic and functional supports In the current study, the actual directions of growth
and/or prosthetic rehabilitation with partial removable of facial thirds were calculated in HED subjects by a
dentures (Figure 3). Modification or replacement of the longitudinal 8-year quantitative evaluation using the
dental prostheses was performed annually to follow same equipment. During childhood, the current HED
facial growth changes. boys and girls grew less than the reference peers in all
To evaluate the global growth variations of facial analyzed facial distances, as already found in the
distances of HED children and to compare them with study of facial volumes by Dellavia et al.10

At the beginning of adolescence, some modifications

in the craniofacial pattern of growth occurred in the HED
individuals. Forehead width and height showed a larger
growth in the syndromic than in their healthy peers of
both genders, in accordance with the previously
reported marked growth of upper facial third volume.10
However, the measuring of forehead height in the HED
subjects could be influenced by some imprecision in
trichion landmark identification because of their rare,
sparse, and fair hair. Besides, some reference points
and distances (namely trichion and upper facial height)
are difficult to define also in nonsyndromic individuals.2
During the entire growth interval, anteroposterior
facial dimensions remained reduced in comparison
with reference values as already reported in previous
anthropometric cross-sectional evaluations.4,15 These
findings suggest that diminished facial depths, with a
consequently flat ‘‘aged’’ profile, are characteristic
features of the present syndromic patients that do
not modify during growth.13 Besides, the retruded
position of landmark nasion determines the developing
of well-described frontal bossing.10,15,16,20
The current ED-afflicted individuals had midface
hypoplasia with a reduced growth of all middle facial
third dimensions in both time span intervals. This result
is in accordance with previous cephalometric and
anthropometric data on maxillary and palatal
size.8,9,13,15,17,20 In contrast, the longitudinal assessment
of facial volumes growth performed by Dellavia et al.10
reported an increased growth of maxillary HED volume
during adolescence. Nevertheless, there is only an
apparent discordance because the maxillary alveolar
process was comprised in the volume assessment of
the maxilla but not in the present measurement of
middle facial height. The distance between landmark
subnasale and pogonion (lower facial height) involves
both the maxillary and mandibular alveolar process.
Therefore, the reduced growth of the middle facial
third could be attributed to diminished transversal and
anteroposterior distances. In the study by Ward and
Bixler,15 all facial widths were more similar to reference
Figure 4. Polar diagram illustrating the global differential normalized values, while Sforza et al.4 observed a large variability
mean growth of the analyzed facial distances in HED and reference of facial distances with an overall disharmonious
subjects of both genders (boys, solid line; girls, broken line) (a)
modification of ED faces compared with normal
during the first time interval (D1, 8–11 years) and (b) during the
second time interval (D2, 12–15 years). Du indicates upper facial nonsyndromic faces.
depth; Wu, upper facial width; Hu, anterior upper facial height; Dm, In the lower facial third, the height increased its
middle facial depth; Wm, middle facial width; Hm, anterior middle growth in the second analyzed time span interval with
facial height; Dl, lower facial depth; Wl, lower facial width; and Hl, possible enhancement by dental eruption and use of
anterior lower facial height. Circle 1 provides the global mean growth
functional appliances and prosthetic devices. In fact,
of the reference subjects; circle 2 represents a 200% increment of
the reference global mean growth. Origin of the axes (marked as 0) prosthodontic therapy seemed to emphasize growth
represents a null reference global mean growth. normalization as observed in implant-treated ED
subjects.12,14 In the current study, the maxillary central
incisors and the maxillary and mandibular first molars
were the most conserved permanent teeth, even if
they showed shape modifications in most HED

children as found in previous investigations.12,14,21,22 In REFERENCES

the 8-year interval, the average number of teeth of the 1. Enlow DH. Facial Growth. Philadelphia, Pa: WB Saunders;
current HED young individuals increased from 9 to 14. 1990.
This factor may have induced a significant vertical 2. Ferring V, Pancherz H. Divine proportions in the growing
growth of the alveolar bone as suggested by Johnson face. Am J Orthod Dentofacial Orthop. 2008;134:472–
et al.12 who found a positive correlation between the 479.
3. Thesleff I. The genetic basis of tooth development and
number of maxillary missing permanent teeth and
dental defects. Am J Med Genet. 2006;140A:2530–
craniofacial dysmorphology in ED children. Male 2535.
subjects with severe hypodontia had marked midface 4. Sforza C, Dellavia C, Vizzotto L, Ferrario VF. Variations
hypoplasia with reduced lower and total facial height.12 in the facial soft-tissues of Italian individuals with ecto-
Similarly, Dellavia et al.9 reported larger palatal height dermal dysplasia. Cleft Palate Craniofac J. 2004;41:
in partially dentate than in completely edentulous 6- 262–267.
5. Sforza C, Dellavia C, Goffredi M, Ferrario VF. Soft tissue
year-old HED boys.
facial angles in individuals with ectodermal dysplasia: a
The growth of lower facial width was larger in the three-dimensional noninvasive study. Cleft Palate
HED than in control girls, while the growth pattern was Craniofac J. 2006;43:339–349.
similar in HED and in reference boys during adoles- 6. Itin PH, Fistarol SK. Ectodermal dysplasias. Am J Med
cence. Besides, the HED-syndromic features appear Genet. 2004;131C:45–51.
more evident in the male sex.14 Also, it has been 7. Lamartine J. Towards a new classification of ectodermal
dysplasias. Clin Exp Dermatol. 2003;28:351–355.
reported that the mandible size tends to normalize 8. Bondarets N, McDonald F. Analysis of the vertical facial
during growth in both sexes.10,13,17 The different form in patients with severe hypodontia. Am J Phys Anthro-
dimensional variations between maxilla and mandible pol. 2000;111:177–184.
may be explained by different growth mechanisms in 9. Dellavia C, Sforza C, Malerba A, Strohmenger L, Ferrario
the two jaws: the maxilla undergoes a sutural growth, VF. Palatal size and shape in 6-year-old patients affected by
while the mandible is primarily characterized by hypohidrotic ectodermal dysplasia. Angle Orthod. 2006;76:
978–983.
endochondral growth at the condyles.1,13 In the ED 10. Dellavia C, Catti F, Sforza C, Grandi G, Ferrario VF. Non-
patients, the mandibular ramus height increases over invasive longitudinal assessment of facial growth in children
time, but the alveolar bone remains atrophic with and adolescents with hypohidrotic ectodermal dysplasia.
consequent low angle vertical growth pattern.13,20 Eur J Oral Sci. 2008;116:305–311.
The present preliminary data confirm that early 11. Ferrario VF, Dellavia C, Serrao G, Sforza C. Soft tissue
facial areas and volumes in individuals with ectodermal
dental rehabilitation is paramount to enhance the
dysplasia: a three-dimensional non invasive assessment.
growth potential of facial hard and soft tissues, thus Am J Med Genet. 2004;126A:253–260.
permitting the attainment of more normal (and possibly 12. Johnson ER, Roberts MW, Guckes AD, Bailey LJ, Phillips
more pleasant) facial features. Although facial depths CL, Wright JT. Analysis of craniofacial development in
and maxillary width still show a reduced growth during children with hypohidrotic ectodermal dysplasia. Am J Med
adolescence, a positive increase in vertical facial Genet. 2002;112:327–334.
13. Lexner MO, Bardow A, Bjorn-Jorgensen J, Hertz JM, Almer
growth can be achieved together with improvements L, Kreiborg S. Anthropometric and cephalometric measure-
in speech, deglutition, and mastication. Hence, the ments in X-linked hypohidrotic ectodermal dysplasia. Orthod
orthodontist has to monitor functional appliances and Craniofac Res. 2007;10:203–215.
removable dentures frequently, with continuous adap- 14. Suri S, Carmichael RP, Tompson BD. Simultaneous
tation and replacement during growth.14,21 functional and fixed appliance therapy for growth modifica-
tion and dental alignment prior to prosthetic habilitation
in hypohidrotic ectodermal dysplasia: a clinical report. J
CONCLUSIONS Prosthet Dent. 2004;92:428–433.
N Facial growth of HED patients can be evaluated by a 15. Ward RE, Bixler D. Anthropometric analysis of the face in
hypohidrotic ectodermal dysplasia: a family study. Am J Phys
simple, low-cost, fast, and noninvasive examination. Anthropol. 1987;74:453–458.
N The main directions of growth of facial structures 16. Saksena SS, Bixler D. Facial morphometrics in the
were identified. identification of gene carriers of X-linked hypohidrotic
N Future longitudinal analyses may quantify the effect ectodermal dysplasia. Am J Med Genet. 1990;35:105–
of specific oral treatments on individual facial growth 114.
17. Bondarets N, Jones RM, McDonald F. Analysis of facial
patterns.
growth in subjects with syndromic ectodermal dysplasia: a
longitudinal analysis. Orthod Craniofac Res. 2002;5:71–
84.
ACKNOWLEDGMENTS
18. Sforza C, Grandi G, Binelli M, Tommasi DG, Rosati R,
The precious and indispensable collaboration of all patients, Ferrario VF. Age- and sex-related changes in the
their families, and of the Associazione Nazionale Displasie normal human ear. Forensic Sci Int. 2009;187:
Ectodermiche (ANDE, Italy) is gratefully acknowledged. 110.e1–7.

19. Ferrario VF, Sforza C, Poggio CE, Cova M, Tartaglia G. 21. Worsaae N, Jensen BN, Holm B, Holsko J. Treatment of
Preliminary evaluation of an electromagnetic three-dimen- severe hypodontia-oligodontia—an interdisciplinary con-
sional digitizer in facial anthropometry. Cleft Palate cept. Int J Oral Maxillofac Surg. 2007;36:473–480.
Craniofac J. 1998;35:9–15. 22. Guckes AD, Scurria MS, King TS, McCarthy GR, Brahim JS.
20. Alcan T, Basa S, Kargül B. Growth analysis of a patient with Pattern of permanent teeth present in individuals with
ectodermal dysplasia treated with endosseous implants: 6- ectodermal dysplasia and severe hypodontia suggests treat-
year follow-up. J Oral Rehabil. 2006;33:175–182. ment with dental implants. Pediatr Dent. 1998;20:278–280.

Original Article
Patient attitudes toward retention and perceptions of treatment success

Nikolay D. Mollova; Steven J. Lindauerb; Al M. Bestc; Bhavna Shroffd; Eser Tufekcie
ABSTRACT
Objective: To discern patients’ opinions regarding responsibility for orthodontic retention and to
determine whether patient attitudes toward retention are related to perceptions of treatment
success.
Materials and Methods: Questionnaires regarding orthodontic retention were distributed to first-
year undergraduate college students (n 5 158), first-year dental students (n 5 183), and retention
patients at orthodontic offices (n 5 214). Items included treatment satisfaction, perceived
responsibility for retention, type of retainer prescribed, reasons for discontinuing use of retainers,
and relapse experienced.
Results: Four hundred twenty-eight of 555 participants indicated that they had received
orthodontic treatment. Most indicated they were either ‘‘satisfied’’ or ‘‘very satisfied’’ with their
teeth, both at the end of treatment (96%) and currently (84%). There was a strong relationship
between the perception of stability of tooth position and current satisfaction level (P , .0001). Most
individuals (88%) indicated that they themselves were responsible for maintaining the alignment
and fit of their teeth. Those who indicated that someone else was responsible were nearly twice as
likely to be dissatisfied with their teeth (P 5 .0496). Patients who had been prescribed clear,
invisible retainers were significantly more likely to be ‘‘very satisfied’’ currently (50%) compared to
those with Hawley (35%) or permanently bonded (36%) retainers (P 5 .0002). Patients with
Hawley retainers were significantly less likely to be wearing them currently as prescribed (45%)
than those with invisible (65%) or bonded (68%) retainers (P , .0001).
Conclusions: Satisfaction with orthodontic results after treatment is related to patient perceptions
of responsibility for retention and perceived stability of tooth position. Patients should play a
contributory role in formulating orthodontic retention plans. (Angle Orthod. 2010;80:656–661.)
KEY WORDS: Retention; Satisfaction; Responsibility; Compliance; Retainer
INTRODUCTION
Undesirable changes in the alignment of teeth
following orthodontic treatment commonly occur un-
less some form of retention is employed.1–4 Kaplan5
a
Student, Department of Orthodontics, School of Dentistry, suggested that patients should be informed of the high
Virginia Commonwealth University, Richmond, Virginia.
b
Professor and Chair, Department of Orthodontics, School probability that some relapse will occur after applianc-
of Dentistry, Virginia Commonwealth University, Richmond, es are removed and of the natural adaptations that
Va. take place over time. In this way, patients become an
c
Associate Professor, Department of Biostatistics, Virginia integral part of the decision-making process, along
Commonwealth University, Richmond, Va. with the orthodontist, regarding the appropriate dura-
d
Professor, Department of Orthodontics, School of Dentistry,
Virginia Commonwealth University, Richmond, Va. tion of retention procedures. Undoubtedly, successful
e
Associate Professor, Department of Orthodontics, School preservation of orthodontic results is achieved most
of Dentistry, Virginia Commonwealth University, Richmond, effectively if patients accept responsibility for wearing
Va. and maintaining appropriate retention appliances.
Corresponding author: Dr Steven J. Lindauer, Department of
Blake and Bibby6 presented six basic treatment
Orthodontics, School of Dentistry, Virginia Commonwealth
University, 520 North 12th Street, Richmond, VA principles to which orthodontists should adhere to
(e-mail: slindauer@vcu.edu) improve posttreatment stability of orthodontic align-
Accepted: December 2009. Submitted: October 2009. ment. Others7–10 advocated long-term retention for
G 2010 by The EH Angle Education and Research Foundation, most patients because of the uncertainty involved in
Inc. determining which patients will remain stable. Ideally,

RETENTION AND PERCEPTIONS OF TREATMENT SUCCESS 657
Table 1. Demographic Characteristics of the Groups Surveyed

College Students (n 5 158) Dental Students (n 5 183) Patients (n 5 214)
Characteristic N % N % N %
Sex
Female 122 77 71 39 127 59
Male 36 23 112 61 87 41
Ethnicity
Asian 20 13 26 14 11 5
Black 34 22 4 2 34 16
Hispanic 6 4 5 3 6 3
White 93 59 134 74 154 73
Other 4 3 13 7 5 2
Treatment history
No 61 39 62 34 3 1
Yes 97 61 120 66 211 99
Age (y)
Mean 20.3 24.5 21.9
Standard deviation 2.0 3.2 12.2
Range 18–29 21–43 8–67
the choice of the type of retainer appropriate for each for orthodontic retention and to determine associations
individual patient, whether fixed or removable, should between patient attitudes toward retention and per-
be made by considering the unique circumstances ceptions of treatment success.
involved.7,8 These include both the potential and
expected changes that may occur over time and the MATERIALS AND METHODS
patient’s ability and willingness to comply with the
A survey was constructed for distribution to former
retention plan. However, results from one survey11
orthodontic patients with questions regarding their
suggested that orthodontists are more likely to choose
retention experiences. Items included demographic
retention procedures according to their own personal
information and questions pertaining to treatment
preferences.
satisfaction, perceived responsibility for retention, type
The 2008 JCO Study of Orthodontic Diagnosis and
of retainer prescribed, reasons for discontinuing use of
Treatment Procedures12 demonstrated a continuing
retainers, and relapse experienced. Institutional review
increase in the routine use of permanent bonded
board approval was obtained and the survey was
retainers by orthodontists. Studies7,8 have shown fixed
distributed to 555 individuals, including first-year
retainers to be safe and effective for most patients over
undergraduate college students (n 5 158), first-year
the long term. Others,13,14 however, have found them to
dental students (n 5 183), and retention patients at
be associated with an increased accumulation of
several local orthodontic offices (n 5 214). Participa-
calculus and point to them as a cause of gingival
tion was voluntary. College students and dental
recession. Orthodontists acknowledge that permanent
students were surveyed during class time and were
retention may be inappropriate for certain patients.8,9
requested to return the survey whether or not they had
Routine removable retainer use is still reported by
ever undergone orthodontic treatment. Differences
more than 50% of orthodontists.12 Compliance with
among groups, and differences and associations
removable retainer wear is out of the control of the
among responses, were determined using Chi-square
orthodontist and may lead to frustration for both
analysis or the Wilcoxon rank-sum test with a
practitioners and patients.9,15 Wong and Freer16 report-
significance threshold of P , .05.
ed that more than 50% of patients admitted that they
did not wear retainers as instructed, with the most
RESULTS
common reasons being discomfort and forgetfulness.
Whereas several studies have investigated the Demographic characteristics of the groups surveyed
attitudes and preferences of orthodontists toward are shown in Table 1. Of the 555 surveys distributed,
various retention protocols, few have reported on the 428 individuals reported a history of orthodontic
perceptions of patients regarding orthodontic retainers treatment, and their characteristics are shown in
and stability.4,16,17 The purpose of the current study was Table 2. While nearly all of the patients surveyed in
to discern patients’ opinions regarding responsibility orthodontic offices had a history of wearing braces,

658 MOLLOV, LINDAUER, BEST, SHROFF, TUFEKCI
Table 2. Demographic Characteristics of the Participants satisfaction currently. The remaining 58% (n 5 244)
Reporting a History of Wearing Braces indicated the same level of satisfaction at both time
Characteristic N % points. There was a strong relationship between the
Sex perception of stability of tooth position since treatment
Female 248 58 concluded and the current level of satisfaction (P ,
Male 180 42 .0001). Of those who reported that their teeth had not
Ethnicity moved, 82% said they were very satisfied with the
Asian 32 8 current straightness and fit of their teeth, compared to
Black 49 12 39% overall. Only 2% of those who perceived no
Hispanic 10 2 posttreatment tooth movement said they were not
White 319 75
Other 13 3
satisfied currently, compared to 18% of those who
perceived a little movement and 55% of those who
Current age (y)
perceived a lot of movement.
Mean 22.1
Standard deviation 8.8
Regarding responsibility for orthodontic retention,
Range 8–67 most (88%) individuals perceived that they themselves
Time since treatment completed (y) were responsible for maintaining the alignment of their
Mean 5.3 teeth after treatment (Table 4). However, some indi-
Standard deviation 5.4 cated that the orthodontist (n 5 45), the general dentist
Range 0–41 (n 5 7), parents (n 5 6), or others (n 5 4) were
responsible. Those who reported that anyone other
there was no significant difference in the proportion of than themselves was responsible for orthodontic
dental (66%) and college (61%) students who reported retention were nearly twice as likely to report dissat-
previous treatment (P . .05). Among college students, isfaction with the current straightness and fit of their
Asians (45%), Blacks (41%), and Hispanics (50%) teeth, versus those who claimed responsibility them-
were significantly less likely to have undergone selves (29% vs 15%; P 5 .0496).
orthodontic treatment than Whites (72%) (P , Most of the patients said that they received a
.0001). Participants without a history of wearing braces removable mandibular retainer after treatment, either
were asked to return the survey without answering with (48%) or without (32%) wires (Table 5). Only 17%
the remaining questions. Not all respondents an- said they had a bonded or banded mandibular retainer,
swered all questions, and some indicated multiple and the remainder (4%) received no mandibular
answers to questions that sought only one answer. retainer. Of those receiving retainers, 26% reported
Answers given to the various questions by those with a that their retainer had broken or needed repair or
history of orthodontic treatment are reported in replacement at least once. There were no statistically
Tables 3 through 5. significant differences among the patients with Hawley
Most respondents indicated that they were either (27%), invisible (clear) (21%), or bonded (35%)
‘‘satisfied’’ or ‘‘very satisfied’’ with the alignment and fit retainers with past breakage, repair, or replacement
of their teeth, both at the end of treatment (96%) and reported (P . .05). However, there was a relationship
currently (84%) (Table 3). Of the 422 subjects who between the type of retainer prescribed and current
indicated their level of satisfaction at both times, 40% satisfaction with treatment. Patients with invisible
(n 5 170) reported a decrease in satisfaction since the retainers were significantly more likely to report that
end of treatment, and only 2% (n 5 8) reported greater they were ‘‘very satisfied’’ currently (50%) compared to
Table 3. Answers to Questions Regarding Treatment Satisfaction and Stability

Question Response N %
How satisfied were you with the straightness and fit of your Very satisfied 292 69
teeth when your braces were removed? (n 5 422) Satisfied 113 27
Not satisfied 17 4
How satisfied are you now with the straightness and fit of Very satisfied 164 39
your teeth? (n 5 422) Satisfied 188 45
Not satisfied 70 17
Did your teeth move or get crooked since you got your braces Yes, they moved a lot and it bothers me 20 5
off? (n 5 420) Yes, they moved a lot, but it does not bother me 11 3
Yes, they moved a little and it bothers me 128 30
Yes, they moved a little, but it does not bother me 138 33
No, they did not move 123 29

Table 4. Answers to Questions Regarding Responsibility for Retention

Who do you think is responsible for keeping your teeth My orthodontist 45 11
straight after your braces come off? (n 5 434) My general dentist 7 2
My parents 6 1
I am 372 88
Other answer 4 1
If your teeth moved or got crooked, what do you think The orthodontist didn’t do a good job in the first place 12 4
is the main reason that your teeth moved since you The orthodontist didn’t follow up long enough after my 9 3
got your braces off? (n 5 324) braces were done
I didn’t follow up with my visits to the orthodontist after my 12 4
braces were done
I didn’t wear my retainer as much every day as I was 133 41
supposed to
I stopped wearing my retainer too soon 70 22
My wisdom teeth came in and caused my teeth to move 19 6
My teeth just moved naturally as I got older 45 14
Other answer 24 7
those with Hawley (35%) or permanently bonded When only those patients out of treatment for more
(36%) retainers (P 5 .0002). There were no differenc- than 1 year were considered (n 5 354), 66% of those
es in the number of individuals who reported that they with bonded retainers said they were still in place,
were not satisfied with the alignment and fit of their compared to 54% wearing invisible retainers and 39%
teeth currently among the various retainer types (P . wearing Hawley retainers (P , .0001). Reasons given
.05). Patients given no retainer were significantly more for discontinuing retainer use were significantly differ-
likely to report that they were currently ‘‘not satisfied’’ ent depending on the type of retainer prescribed (P 5
(53%; P 5 .0002). .0002). For removable retainers (Hawley or invisible),
Of the patients who reported that they received the most common reason for stopping wear was ‘‘I just
mandibular retainers, 45% said that they had stopped stopped wearing it eventually’’ (n 5 52; 33%). Those
wearing them regularly. Discontinuation of retainer use with bonded retainers most often said the retainer was
was significantly related to the type of retainer removed by an orthodontist or dentist (n 5 8; 42%) or
prescribed (P , .0001), with only 45% of those with that it was lost or broken (n 5 5; 26%).
Hawley retainers claiming to still wear them compared
to 65% of those with invisible retainers and 68% of DISCUSSION
those with bonded retainers. However, bonded retain- This study surveyed the opinions of 428 patients
ers had been in place, on average, for a significantly who had completed fixed orthodontic treatment an
longer period of time (7.8 years) than Hawley (3.5 average of 5.3 years previously. No attempt was made
years) or invisible (1.8 years) retainers (P , .0001). to determine characteristics of the original malocclu-
Table 5. Answers to Questions Regarding Retainer Type and Duration of Retention

What kind of retainer did you get for your lower Permanent bonded or banded retainer (glued in place) 70 17
teeth after your braces were removed? Removable clear retainer without wires (invisible retainer) 132 32
(n 5 419) Removable plastic retainer with wires 202 48
No retainer 15 4
Other answer 0 0
Did your lower retainer ever break or need to be No 299 74
repaired or replaced? (n 5 406) Yes 107 26
Have you stopped wearing your lower retainer on No, I still wear it 222 55
a regular basis? (n 5 407) Yes 185 45
Why did you stop wearing your retainer? (n 5 175) The original orthodontist told me to stop wearing it 11 6
A new/different dentist told me to stop wearing it 5 3
I figured I had been wearing it long enough 21 12
I lost it and I didn’t want to replace it 14 8
It broke and I didn’t want to fix it 8 5
It did not fit anymore 34 19
I just stopped wearing it eventually 55 31
I never really wore it like I was supposed to 27 15

660 MOLLOV, LINDAUER, BEST, SHROFF, TUFEKCI
sions or any other treatment factors involved. Addi- wearing their retainer enough (41%), stopping retainer
tionally, there was a large amount of variation among wear too soon (22%), or not following up with
patients in the time since appliances had been appointments (4%). Others considered posttreatment
removed, with 84% reporting completion of active movement to be a natural phenomenon (14%) or a
treatment at least 1 year previously. However, the result of eruption of third molars (6%). Only 7% blamed
primary purpose of the study was to determine the orthodontist for not following up long enough or for
whether there were associations between patients’ performing inadequate treatment.
attitudes toward orthodontic retention and perceptions Perception of stability of tooth alignment and fit was
of treatment success. strongly related to current satisfaction with the ortho-
The overwhelming majority of those surveyed dontic treatment outcome. The literature suggests that
perceived that orthodontic retention was their own some form of retention is required to maintain tooth
responsibility (88%), with only 11% reporting that the alignment in the long term. Invisible retainers have
orthodontist was primarily responsible. Those who been shown to maintain individual tooth positions as
considered anyone other than themselves to have well as or better than traditional Hawley retainers.22,23
primary responsibility for the stability of their treatment However, Hawley retainers allow more beneficial
results were about twice as likely to report that they settling of the occlusion after treatment than clear,
were not satisfied with the current alignment and fit of full-coverage retainers.24 A recent study25 found that
their teeth. This is consistent with other fields of the greatest increase in posterior tooth contacts
medicine, in which shared responsibility for treatment occurred when bonded retainers were used for
has also been found to be an important component of retention. Although bonding permanent retainers
treatment success.18–20 It also suggests that getting seems to be the most predictable way to prevent
patients involved in treatment decisions and making unwanted tooth movements and encourage posterior
them aware of their own responsibility for retention settling, undesirable changes in tooth position have
regimens is important for maintaining high levels of been reported even with bonded retainers in place.26
satisfaction. This may also help to improve patient Psychological traits of individual patients may affect
cooperation. their satisfaction with orthodontic treatment out-
Although recent surveys8,12,21 suggest that the routine comes.27 Anderson et al.28 found that posttreatment
placement of bonded permanent retainers is becoming satisfaction was positively correlated with pretreatment
more common among orthodontists, only 17% of the motivation levels. In the present study, satisfaction
patients in this study reported receiving one. Of the was greater for patients who claimed responsibility for
patients with bonded retainers, 35% reported that they maintaining posttreatment outcomes. However, the
had previously broken or needed repair or replacement, perception that alignment and fit of the teeth remained
which was not significantly different from the failure rate stable over time was also an important factor. Current
reported for other types of retainers. This failure rate satisfaction was highest for the patients who received
was consistent with that found by Booth et al.10 but was clear, invisible retainers. It is apparent that no single
substantially higher than the approximately 5% rate approach is appropriate for all patients and that
reported by Zachrisson.7 orthodontic retention decisions should be made with
Patients with bonded and invisible retainers were consideration of differences among individuals. Most
more likely to be wearing them regularly at the time of likely, the best way to achieve long-term patient
the survey than those who were prescribed Hawley satisfaction is to devise an appropriate retention plan
retainers. For those patients who had completed with input from both the orthodontist and the patient so
treatment at least 1 year prior to the present study, that the patient is informed of the options available and
compliance with removable retainer wear, for invisible is motivated to share responsibility for maintaining the
and especially Hawley types, was significantly lower outcome.
than for bonded retainers. About two-thirds of bonded
retainers were still in place after 1 year, whereas fewer CONCLUSIONS
than half of the patients with Hawley retainers reported
that they were still wearing them as prescribed. Wong N Most respondents indicated that they were either
and Freer16 also found that fewer than 50% of patients ‘‘satisfied’’ or ‘‘very satisfied’’ with the alignment and
wore removable mandibular retainers as instructed, fit of their teeth, both at the end of treatment (96%)
although they did not differentiate between Hawley and and currently (84%), at an average of 5.3 years
invisible retainer types. posttreatment.
Most of the patients in this study who perceived N There was a strong relationship between the
movement of their teeth following treatment attributed perception of stability of tooth position after treatment
those changes to factors under their own control: not and current satisfaction (P , .0001).

N Most individuals (88%) perceived that they them- 12. Keim RG, Gottlieb EL, Nelson AH, Vogels DS III. 2008 JCO
selves were primarily responsible for maintaining the study of orthodontic diagnosis and treatment procedures,
part 1: results and trends. J Clin Orthod. 2008;42:625–640.
alignment of their teeth after treatment. Those who 13. Levin L, Samorodnitzky-Naveh GR, Machtei EE. The
believed that anyone else was responsible for association of orthodontic treatment and fixed retainers with
retention were about twice as likely to feel dissatis- gingival health. J Periodontol. 2008;79:2087–2092.
fied with their teeth currently (P 5 .0496). 14. Pandis N, Vlahopoulos K, Madianos P, Eliades T. Long-term
periodontal status of patients with mandibular lingual fixed
retention. Eur J Orthod. 2007;29:471–476.
ACKNOWLEDGMENTS 15. Gross AM, Schwartz CL, Kellum GD, Bishop FW. The effect
of a contingency contracting procedure on patient compli-
This study was supported in part by an A. D. Williams Summer ance with removable retention. J Clin Orthod. 1991;25:
Research Fellowship from Virginia Commonwealth University 307–310.
and by the Medical College of Virginia Orthodontic Education 16. Wong P, Freer TJ. Patients’ attitudes towards compliance
and Research Foundation. with retainer wear. Aust Orthod J. 2005;21:45–53.
17. Hichens L, Rowland H, Williams A, Hollinghurst S, Ewings
REFERENCES P, Clark S, Ireland A, Sandy J. Cost-effectiveness and
patient satisfaction: Hawley and vaccum-formed retainers.
1. Rinchuse DJ, Miles PG, Sheridan JJ. Orthodontic retention
Eur J Orthod. 2007;29:372–378.
and stability: a clinical perspective. J Clin Orthod. 2007;41:
18. Gurmankin AD, Baron J, Hershey JC, Ubel PA. The role of
125–132.
physicians’ recommendations in medical treatment deci-
2. Little RM. Stability and relapse of mandibular anterior
sions. Med Decis Making. 2002;22:262–271.
alignment: University of Washington studies. Semin Orthod.
19. Lorenz AD, Mauksch LB, Gawinski BA. Models of collabo-
1999;5:191–204.
ration. Prim Care. 1999;26:401–410.
3. Eslambolchi S, Woodside DG, Rossouw PE. A descriptive
study of mandibular incisor alignment in untreated subjects. 20. Tang PC, Newcomb C, Gorden S, Kreider N. Meeting the
Am J Orthod Dentofacial Orthop. 2008;133:343–353. information needs of patients: results from a patient focus
4. Bondemark L, Holm AK, Hansen K, Axelsson S, Mohlin B, group. Proc AMIA Annu Fall Symp. 1997:672–676.
Brattstrom V, Paulin G, Pietila T. Long-term stability of 21. Singh P, Grammati S, Kirschen R. Orthodontic retention
orthodontic treatment and patient satisfaction. Angle Orthod. patterns in the United Kingdom. J Orthod. 2009;36:115–121.
2007;77:181–191. 22. Rowland H, Hichens L, Williams A, et al. The effectiveness
5. Kaplan H. The logic of modern retention procedures. of Hawley and vacuum-formed retainers: a single-center
Am J Orthod Dentofacial Orthop. 1988;93:325–340. randomized controlled trial. Am J Orthod Dentofacial
6. Blake M, Bibby K. Retention and stability: a review of the Orthop. 2007;132:730–737.
literature. Am J Orthod Dentofacial Orthop. 1998;114: 23. Lindauer SJ, Shoff RC. Comparison of Essix and Hawley
299–306. retainers. J Clin Orthod. 1998;32:95–97.
7. Zachrisson BU. Long-term experience with direct-bonded 24. Sauget E, Covell DA, Boero RP, Lieber WS. Comparison of
retainers: update and clinical advice. J Clin Orthod. 2007;41: occlusal contacts with use of Hawley and clear overlay
728–737. retainers. Angle Orthod. 1997;67:223–230.
8. Renkema AM, Sips ET, Bronkhorst E, Kuijpers-Jagtman 25. Sari Z, Uysal T, Basciftci FA, Inan O. Occlusal contact
AM. A survey on orthodontic retention procedures in The changes with removable and bonded retainers in a 1-year
Netherlands. Eur J Orthod. 2009;31:432–437. retention period. Angle Orthod. 2009;79:867–872.
9. Cerny R, Lloyd D. Dentists’ opinions on orthodontic retention 26. Katsaros C, Livas C, Renkema AM. Unexpected complica-
appliances. J Clin Orthod. 2008;42:415–419. tions of bonded mandibular lingual retainers. Am J Orthod
10. Booth FA, Edelman JM, Proffit WR. Twenty-year follow-up Dentofacial Orthop. 2007;132:838–841.
of patients with permanently bonded mandibular canine-to- 27. Al-Omiri MK, Abu Alhaija ES. Factors affecting patient
canine retainers. Am J Orthod Dentofacial Orthop. 2008; satisfaction after orthodontic treatment. Angle Orthod. 2006;
133:70–76. 76:422–431.
11. Wong PM, Freer TJ. A comprehensive survey of retention 28. Anderson LE, Arruda A, Inglehart MR. Adolescent patients’
procedures in Australia and New Zealand. Aust Orthod J. treatment motivation and satisfaction with orthodontic
2004;20:99–106. treatment. Angle Orthod. 2009;79:821–827.

Original Article
Suitability of orthodontic brackets for rebonding and reworking following

removal by air pressure pulses and conventional debracketing techniques
Michael Knösela; Simone Mattysekb; Klaus Jungc; Dietmar Kubein-Meesenburgd;
Reza Sadat-Khonsarie; Dirk Ziebolzf
ABSTRACT
Aim: To test the null hypothesis that there are no significant differences in the reusability of
debonded brackets with regard to debonding technique and adhesive used.
Method: Ninety-six osteotomed third molars were randomly assigned to two study groups (n 5 48)
for bonding of a 0.018-inch bracket (Ormesh, Ormco) with either a composite adhesive (Mono-
Lok2; RMO) or a glass ionomer cement (GIC; Fuji Ortho LC;GC). Each of these two groups were
then randomly divided into four subgroups (n 5 12) according to the method of debonding using (1)
bracket removal pliers (BRP; Dentaurum), (2) a side cutter (SC; Dentaurum), (3) a lift-off
debracketing instrument (LODI; 3M-Unitek), or (4) an air pressure pulse device (CoronaFlex;
KaVo). The brackets were subsequently assessed visually for reusability and reworkability with 23
magnification and by pull testing with a 0.017- 3 0.025-inch steel archwire. The proportions of
reusable brackets were individually compared in terms of mode of removal and with regard to
adhesives using the Fisher exact test (a 5 5%).
Results: The null hypothesis was rejected. Not taking into account the debonding method,
brackets bonded with GIC were judged to a significant extent (81%; n 5 39; P , .01) to be
reworkable compared with those bonded with composite (56%; n 5 27). All brackets in both
adhesive groups removed with either the LODI or the CoronaFlex were found to be reusable,
whereas 79% (46%) of the brackets removed with the BRP (SC) were not. The proportion of
reusable brackets differed significantly between modes of removal (P , .01).
Conclusion: With regard to bracket reusability, the SC and the BRP cannot be recommended for
debonding brackets, especially in combination with a composite adhesive. (Angle Orthod.
2010;80:649–655.)
KEY WORDS: Debonding; Rebonding; Bracket reusability; CoronaFlex
INTRODUCTION
a
Associate Professor, Department of Orthodontics and
Dentofacial Orthopedics, Center of Dentistry, Georg-August- The bracket adhesive technique is an important and
University, Göttingen, Germany.
routine part of contemporary orthodontics because it
b
Research Scientist, Department of Orthodontics and
Dentofacial Orthopedics, Center of Dentistry, Georg-August- provides the basis for ensuring a controlled force and
University, Göttingen, Germany. torque transmission from the archwire on the teeth.
c
Assistant Professor, Department of Medical Statistics, The enamel-adhesive-bracket interface has to provide
Georg-August-University, Göttingen, Germany. both aspects of reliable attachment of the bracket
d
Professor and Department Chair, Department of Orthodon-
tics and Dentofacial Orthopedics, Center of Dentistry, Georg- during treatment and also easy and quick removal
August-University, Göttingen, Germany. following treatment with the least possible amount of
e
Associate Professor, Department of Orthodontics and damage to the enamel surface.
Dentofacial Orthopedics, Center of Dentistry, Georg-August- Another important aspect is that of economics. The
University, Göttingen, Germany.
cost factor can be significantly reduced if the bracket
f
Assistant Professor, Department of Preventive Dentistry,
Periodontology and Cariology, Center of Dentistry, Georg- debonding is carried out in such a way that it does not
August-University, Göttingen, Germany. damage the brackets during removal; that is, the
Corresponding author: Priv-Doz Dr Michael Knösel, Depart-
ment of Orthodontics and Dentofacial Orthopedics, Georg-
August-University, Center of Dentistry, Robert-Koch-Str. 40 Accepted: December 2009. Submitted: October 2009.
37099 Göttingen, Germany 37099 G 2010 by The EH Angle Education and Research Foundation,
(e-mail: mknoesel@yahoo.de) Inc.

650 KNÖSEL, MATTYSEK, JUNG, KUBEIN-MEESENBURG, SADAT-KHONSARI, ZIEBOLZ
brackets can be reused following different forms of as an alternative to conventional orthodontic compos-
processing, for either rebonding incorrectly positioned ites. However, later studies confirmed the assumption
brackets or recycling, requiring only a fraction of the that composites produce significantly higher bond
price for a new bracket. strengths during bracket fixation than light-cured GIC.12
Matasa1 investigated differences in the damage to The aim of this in vitro study was therefore to
brackets as a result of different debonding techniques. optimize bracket removal to enable their reusability
Damage can be done to the wings of the bracket, in with rebonding after incorrect bonding and for rework-
addition to deformation of the bracket base or slot. If ing, by testing four different debonding instruments
these occur, it will no longer be possible to reuse or (side cutter, bracket removal pliers, lift-off debracketing
rework the bracket. instrument, and an air pressure crown remover;
Orthodontic bracket recycling, which is perhaps CoronaFlex, Kavo, Biberach/Riss, Germany) used in
more appropriately referred to as bracket processing prosthodontics for the removal of crowns and bridges)
or bracket reworking, as brackets are not reshaped but and two different types of orthodontic adhesives (a
only separated from stains and adhesive remnants composite adhesive and a GIC).
using heat and chemical agents, followed by cleaning The null hypothesis was that there are no significant
and polishing results in brackets that reach standards differences in terms of the reusability of debonded
of quality comparable to those shown by unused brackets in relation to the debonding technique and
brackets2 and are able to withstand the same draw-off adhesive used.
strengths.3
Because of the increasing importance of the eco- MATERIALS AND METHODS
nomic aspect of orthodontics, the use of reworked
Teeth
brackets has been adopted by a growing number of
orthodontists. In a questionnaire distributed among 300 Ninety-six third molars that were freshly osteotomed
members of the British Orthodontic Society, 48% said due to a lack of space were included in the study. The
they used processed brackets for economic reasons.4 exclusion criteria of enamel damage (fractures, de-
Another rationale for using a bracket-preserving mineralization, or decay) were assessed by visual
method of debonding is the intentional removal and inspection using a 23 magnifier. The patients or their
rebonding of incorrectly positioned brackets,5 a chal- guardians gave informed consent for donation of the
lenge orthodontists regularly face. Even the slightest extracted third molars for study purposes. Both for
deformations of the bracket slot can result in lowered better handling and a simulation of the viscoelastic
fitting accuracy of the inserted archwire, producing cushioning and elastic dental fiber suspension, the
unwanted friction.6 cleaned teeth were each embedded in small blocks of
Therefore, in addition to the enamel-preserving silicon (Silaplast; Detax, Ettlingen, Germany; Figure 1)
aspect of debonding, it is important for many ortho- and stored in physiological saline at 20uC, which was
dontists to choose a method of bracket removal that renewed every second day.
does not deform or damage the bracket and allows for The teeth were randomly assigned to one of two
instant rebonding. study groups (n 5 48) for either bonding with composite
To ensure the reusability of a bracket, it is important adhesive or with GIC. Then the two groups were each
to choose both an appropriate adhesive and the right once more randomly divided into four subgroups (n 5
instrument for debonding. Current bracket adhesives 12), and these were allocated one of the four bracket-
are mostly based on diacrylates. Their bonding debonding methods described in the following sections.
strength has been tested in numerous studies, and
they have produced good results,7,8 which to a Brackets
substantial extent is due to the conditioning of the
Maxillary premolar metal brackets (0.018-inch slot
enamel surface by etching and subsequent mechan-
system; Ormesh; Ormco, Orange, CA), which were
ical retention to microporosity.9
identical on both the left and right sides, proved to fit
Apart from composites, glass ionomeric cements
neatly to the buccal surface of the third molars. The
(GIC) are in common use in orthodontics. In contrast to
brackets had a common meshlike base to provide
composite adhesives, they do not work by using
better adhesive retention.13
mechanical retention but instead by acid-base interac-
tion between GIC and enamel.10 Compton et al.11
Adhesives
reported that light-cured GIC had a higher bonding
strength compared with chemically hardened GIC and We used two types of adhesive for bracket bonding
suggested the use of GIC with prior conditioning of the following the manufacturers’ instructions for applica-
enamel surface with a weak acid to enhance cohesion tion. Representative for the composites, Mono-Lok2

BRACKET DEFORMATION DURING BONDING 651
Figure 1. Instruments used for debonding with the corresponding force systems generated (red), moments (blue), and effective force (dotted) on
the bracket base. Please see the online version of this article for the color version of figure 1.
(Rocky Mountain Orthodontics, Denver, Colo) was LC (GC Co-operation, Tokyo, Japan), a light-cured,
used, a one-paste hybrid composite of methacrylmono- resin-modified GIC that is characterized by enhanced
mers and polymers, with prior enamel etching using bond strength compared with conventional GIC.11,14
37% phosphoric acid. For the GIC, we used Fuji Ortho Prior enamel conditioning was achieved by etching with

Table 1. Application Steps for Both the Composite Adhesive Mono-Lok2 and the Glass Ionomere Cement Fuji Ortho LC
Adhesive Group Size (n) Application Steps
Mono-Lok2 (Rocky Mountain 48 1. Cleaning of enamel surface for 30 s using fluoride-free polishing (Zirkate; L.D.
Orthodontics, Denver, Colo) Caulk Co, Milford, Del)
2. Enamel conditioning with 37% phosphoric acid for 30 s with subsequent rinsing
with water for 1 min and air drying
3. Primer application with single-use brush on enamel and bracket base
4. Adhesive application to bracket base
5. Immediate bracket positioning on enamel, by slight exertion of pressure for 45 s
6. Removal of excessive adhesive with a scaler
Fuji Ortho LC (GC Cooperation, 48 1. Cleaning of the enamel surface for 30 s using fluoride-free polishing (Zirkate;
Tokyo, Japan) L.D. Caulk Co, Milford, Del)
2. Enamel conditioning with 10% polyacrylic acid for 10 s
3. Mixing of fluid and powder according to the manufacturer’s instructions (ratio:
1:3) for 20 s
4. Application of adhesive to bracket base
5. Immediate positioning of the bracket on the enamel
6. Ultraviolet light curing for 40 s (wavelength: 470 nm)
7. Removal of excessive adhesive with a scaler
10% polyacrylic acid. Application steps for both toggle, releasing a short impact pulse of 3000 N for 10
adhesives are shown in Table 1. milliseconds, which removed the adhesion.
Debonding Criteria and Assessment of Reusability of

Debonded Brackets
Removal of the brackets was carried out 24 hours
after bonding. Four different debonding instruments At first, the brackets were visually inspected using a
were employed: (1) a medium-sized side cutter (SC; 23 magnification for deformations to the bracket base,
Dentaurum, Ispringen, Germany), (2) bracket removal wings, or slot, compared directly with an unused (new)
pliers (BRP; Dentaurum; Ispringen, Germany), (3) a bracket of the same type (Figure 2a), as recovery of
lift-off debracketing instrument (LODI; 3M Unitek, brackets with visual deformations cannot be reworked at
Monrovia, Calif), and (4) an air pressure pulse device a reasonable cost.6 Only slight deformations of the slot
(CoronaFlex; Kavo, Biberach, Germany), which is inhibit the wire insertion, or at least the gliding of the
used in prosthodontics for the removal of crowns and bracket along the archwire (Figures 2b,c). Therefore,
bridges (Figure 1a-d). brackets were subsequently tested with a straight 0.017-
Although variations in the use of the debonding 3 0.025-inch archwire (Ormco) for deadlock or crepita-
instruments are possible, debracketing was performed tion, which are signs of slot deformation. The assess-
in the typical, standardized manner. The SC was ment criterion for being not reworkable and nonreusable
orientated in such a way that it was inserted diagonally was whether the archwire resulted in any noticeable
at the bracket base at the cervical and incisor portion friction or jamming, either during complete manual
of the bracket wing. The bracket was removed by pulling through the slot (Figure 2c) or by regular insertion
gentle squeezing of the pliers and an additional into the slot. Both tests were carried out twice by two
clockwise rotational movement. orthodontists in the Dentistry Centre at the University of
The BRP was used by gripping below the bracket Göttingen, Germany. A bracket was judged to be
wings. By closing and downward tipping of the pliers, a reusable if it passed both tests. Testing was implement-
rotational axis was created at the apical bracket ed prior to and after debonding: all 96 brackets fresh
margin, thereby releasing the bracket. from the factory were therefore intact at baseline.
The LODI was positioned by linking its hanger to the
upper left bracket wing and simultaneously resting the Statistical Analyses
instrument on the tooth. Compression of the pliers
caused the bracket to lift off on application of a pulling The proportion of reusable brackets was compared
force. between instruments and adhesives, respectively,
The CoronaFlex was used by positioning its toggle using the Fisher exact test. The significance level for
parallel to the adhesive-enamel interface. By pulling the tests was chosen to be a 5 5%. All analyses were
the trigger, a piston with a weight of 2.5 g was impelled performed using the software R (version 2.8, www.
by an air pressure of 2.2 bar along the shaft on the r-project.org).

Figure 2. Examples of (a, left) deformed and (a, right) intact brackets. (b) Misfit of archwire due to slot deformation. (c) Pull testing.
RESULTS rejected. All brackets of both adhesive groups

removed with the LODI and also those removed with
Adhesive Factor
the impulse device were reusable; that is, they were
Leaving the method used for the bracket removal intact according to the standard used in the evaluation.
out of the analysis, of those brackets bonded with Fuji The results of the assessment for bracket reusability
Ortho LC, 81% (n 5 39) were judged to be reworkable. for the BRP and the SC were worse (Table 2) and
Of those bonded with Mono-Lok2, 56% (n 5 27) were differed significantly when compared with one another
judged to be reworkable (Table 2). Statistically, the (P 5 .04). Particularly in combination with the
proportions of reusable brackets was therefore signif- MonoLok bonding, the BRP and SC results were
icant (P , .01), and accordingly, the null hypothesis discouraging (Table 2).
was rejected. In general, brackets bonded with Fuji
Ortho LC were subjectively easier to remove com- DISCUSSION
pared with those bonded with MonoLok.
The proportion of reusable debonded brackets may
be influenced by three factors: the mode of enamel
Instrument Factor
conditioning, the adhesive used, and the mode of
The proportions of reusable brackets also differed debonding.
significantly between instruments (P , .01), so that the The composite adhesive MonoLok2 was used in
second element of the null hypothesis was also combination with prior enamel conditioning by etching
with 37% phosphoric acid, to enhance the interface
Table 2. Comparison of Bracket Reusability Between Instruments area between the adhesive and the enamel by
and Adhesives dissolving enamel prisms away to a depth of 22 mm,
Bracket Reusable thereby providing a basis for mechanical adhesive
P (Fisher
Factor No Yes Exact Test) retention.15
Fuji Ortho LC requires enamel conditioning with 10%
Instrument ,.01
Bracket removal pliers 19 (79%) 5 (21%)
polyacrylic acid. Its penetration depth can be judged as
CoronaFlex 0 (0%) 24 (100%) moderate compared with the use of 37% phosphoric
Lift-off debracketing 0 (0%) 24 (100%) acid.16 Its purpose is not to enhance mechanical
instrument retention but to reduce interface (surface) tension, to
Side cutter 11 (46%) 13 (54%) provide the basis for chemical bonding between the
Adhesive ,.01 enamel and GIC.17
Fuji Ortho LC 9 (19%) 39 (81%) Contemporary light-cured GIC produces clinically
Mono-Lok2 21 (44%) 27 (56%)
acceptable results for bond strength,14,18 although they

are below those for conventional composites.19,20 The pulling force exerted by the SC is generated by
Moreover, they are easier to handle than are compos- the wedge effect of the pliers’ cutting end (Figure 1b).
ites such as MonoLok2 as they do not require a Compression of the pliers creates a deformation of the
completely dry operation area,18 which is of advantage bracket base. As these forces are identical in
in, for example, the unforeseen necessity of instant magnitude but opposite in direction, they cancel each
rebonding of lost or incorrectly positioned brackets other out and do not contribute to the release of the
during practice hours or bonding lingual mandibular bracket. The insertion at two points of the bracket
teeth. generates a force couple (ie, a moment with an initial
On the subject of enamel damage following debond- axis of rotation at the center of the bracket and a force
ing, Kusy21 raised the question of whether and when that is perpendicular to the enamel surface). There-
higher bond strengths are necessary and recommend- fore, regardless of the adhesive used, the SC resulted
ed the use of GIC for orthodontic purposes. Our in a majority of the brackets’ having deformations at
findings indicate that a side effect of these lower bond the base or at the slot. We can therefore conclude that
strengths of GIC appears to be a significantly reduced the compression of brackets by an insertion of
proportion of brackets that are not reusable after opposite forces, as described for the BRP and the
debonding (Table 2). Putting the method of bracket SC method, may be decisive in excluding brackets as
removal to one side and focusing on the adhesive being suitable for rebonding.
factor, almost half of the brackets bonded with The equivalent force system generated by the LODI
MonoLok2 were not in such a condition as to allow is a pull-off force perpendicular to the bracket base,
reuse or reworking, whereas in the GIC group, less which creates a moment that is parallel to the enamel
than 20% of the brackets had corresponding deforma- surface (Figure 1c). There was no compression of the
tions. This result is not only in agreement with previous bracket, and although there was a pulling force
findings of higher bond strengths for composites than insertion at one bracket wing, all brackets debonded
for GIC8,14,18 but also seems to confirm our subjective with the LODI remained intact afterward. This result is
impression that the brackets used in our study that in accordance with findings from previous studies.23,24
were bonded with Fuji Ortho LC were easier to flip off The equivalent force system created by the Corona-
compared with those bonded with MonoLok2. Flex is a force and a moment that are both parallel to
Beyond a diversity of debonding methods and their the enamel surface and to the bracket base (Fig-
modifications, variations in the application of the tested ure 1d). In combination with the immediate elimination
devices themselves are possible, with the result of of the adhesive layer, the absence of compressive
different force influence lines. We therefore propose to forces and the moment parallel to the enamel provide
base discussions of the effects of debonding methods an explanation for the preservation of all brackets
not only on the type of the instrument but also on a clear during debonding.
description of the way it was applied, as variations may
have an influence on bracket deformation. Clinical Implications
The equivalent force systems created by the
Irrespective of the adhesive used, the LODI and the
different instruments during removal of the brackets
CoronaFlex provided the best results in relation to the
may provide an explanation for the proportions of
reusability of debonded brackets. If conventional
nonreusable brackets. The use of the BRP produces
debonding techniques (BRP, SC) are used, the use
symmetrical force insertion on all four bracket wings
of a GIC is favored, since its lower bond strength
(Figure 1a). These forces are transmitted to bracket
compared with composite adhesives18 seems to result
base and slot but do not contribute to bracket removal,
in a significantly higher portion of reusable brackets. In
as they cancel each other, but they do contribute to a
addition, the insertion of a gauge or archwire matching
deformation of bracket base and especially bracket
the slot width during debonding may be considered a
slot. The equivalent force system created by the BRP
useful step to prevent the slot from collapsing.
is a force perpendicular to enamel surface and bracket
base, which creates a moment parallel to the bracket
CONCLUSIONS
base (Figure 1a). The result is wing and slot deforma-
tion, which absorbs forces that would otherwise be N The proportions of reusable or reworkable brackets
transmitted on the enamel. In this context, Benett et differed significantly depending on the mode of
al.22 described the BRP method as more enamel removal and the adhesive used (P , .01).
preserving than those methods that produce forces on N In the case of those bonded with the GIC Fuji Ortho
the bracket base, as does the SC, but they also LC, a significantly higher number of brackets could
reported that brackets removed with the removal pliers be recovered or reused compared with those bonded
were not suitable for rebonding. with the composite MonoLok2 (P , .01).

N All brackets removed with the LODI and with the orthodontic bonding agent. Am J Orthod Dentofacial Orthop.
impulse device were reusable, that is, intact accord- 1992;101:138–144.
12. Cook PA, Youngson CC. An in vitro study of the bond
ing to the evaluation criterion, regardless of the strength of a glass ionomer cement in the direct bonding of
adhesive used, whereas 79% (46%) of the brackets orthodontic brackets. Br J Orthod. 1988;15:247–253.
removed with the removal pliers (SC) were not. 13. Maijer R, Smith DC. Variables influencing the bond strength
N When assessed in terms of bracket reusability, of metal orthodontic bracket bases. Am J Orthod. 1981;79:
especially in combination with the MonoLok2 com- 20–33.
14. Cacciafesta V, Jost-Brinkmann PG, Süßenberger U,
posite adhesive, the SC and the BRP cannot be
Miethke RR. Effects of saliva and water contamination on
recommended for debonding brackets. the enamel shear bond strength of light-cured glass ionomer
cement. Am J Orthod Dentofacial Orthop. 1998;113:
REFERENCES 402–407.
1. Matasa GC. Pros and cons of the reuse of direct-bonded 15. Shinchi MJ, Soma K, Nakabayashi N. The effect of
appliances. Am J Orthod Dentofacial Orthop. 1989;96: phosphoric acid concentration on resin tag length and bond
72–76. strength of a photo-cured resin to acid-etched enamel. Dent
2. Eliades T, Zinelis S, Eliades G, Athanasiou AE. Character- Mater. 2000;16:324–329.
ization of as-received, retrieved, and recycled stainless steel 16. Farquhar RB. Direct bonding comparing a polyacrylic acid
brackets. J Orofac Orthop. 2003;64:80–87. and a phosphoric acid technique. Am J Orthod. 1986;90:
3. Postlethwaite KM. Recycling bands and brackets. 187–194.
Br J Orthod. 1992;19:157–164. 17. Attin T, Buchalla W, Hellwig E. Influence of enamel
4. Coley-Smith A, Rock WP. Bracket recycling—who does conditioning on bond strength of resin-modified glass-
what? Br J Orthod. 1997;24:172–174. ionomer restorative materials and polyacid-modified com-
5. Mui B, Rossouw PE, Kulkarni GV. Optimization of a posites. J Pros Dent. 1996;76:29–33.
procedure for rebonding dislodged orthodontic brackets. 18. Fricker JP. A new self-curing resin-modified glass ionomer
Angle Orthod. 1999;69:276–281. cement for direct bonding of orthodontic brackets in vivo.
6. Machen DE. Orthodontic bracket recycling. Am J Orthod Am J Orthod Dentofacial Orthop. 1998;113:384–386.
Dentofacial Orthop. 1993;104:618–619. 19. Bishara SE, Von Wald L, Laffon JF, Jakobsen JR. Effect of
7. Miguel JA, Almeida MA, Chevitarese O. Clinical comparison altering the type of enamel conditioner on the shear bond
between glass ionomer cement and a composite for direct strength of a resin-reinforced glass ionomer adhesive.
bonding of orthodontic brackets. Am J Orthod Dentofacial Am J Orthod Dentofacial Orthop. 2000;118:288–294.
Orthop. 1995;107:484–487. 20. Toledano M, Osorio R, Osorio E, Romeo A, de la Higuera B,
8. Valente RM, de Rijk WG, Drummond JL, Evans CA. Etching Garcia-Godoy F. Bond strength of orthodontic brackets
conditions for resin-modified glass ionomer cement for using different light and self-curing cements. Angle Orthod.
orthodontic brackets. Am J Orthod Dentofacial Orthop. 2003;73:56–63.
2002;121:516–520. 21. Kusy R. Letter to the editor. Am J Orthod Dentofacial
9. Bishara SE, Gordon VV, Von Wald L, Jakobsen JR. Shear Orthop. 1994;106:17A.
bond strength of composite, glass ionomer and acid primer 22. Bennett CG, Shen C, Waldron JH. The effects of debonding
adhesive systems. Am J Orthod Dentofacial Orthop. 1999; on the enamel surface. J Clin Orthod. 1984;18:330–334.
115:24–28. 23. Oliver RG. The effect of different methods of bracket
10. Wilson AD, Kent BE. A new translucent cement for dentistry. removal on the amount of residual adhesive. Am J Orthod
Br Dent J. 1972;132:133–135. Dentofacial Orthop. 1988;93:196–200.
11. Compton AM, Meyers CE, Hondrum SO, Lorton L. 24. Coley-Smith A, Rock WP. Distortion of metallic orthodontic
Comparison of the shear bond strength of a light-cured brackets after clinical use and debond by two methods.
ionomer and a chemically cured glass ionomer for use as an Br J Orthod. 1999;26:135–139.

Angle

Enviado por

Dados do documento

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Angle

Enviado por

Direitos autorais:

Formatos disponíveis

Volume 80, Issue 4 (July 2010)

What's New in Dentistry

611 Latest research reports from the dental literature

613 Stability of surgically assisted rapid maxillary expansion and orthopedic

626 Canine retraction rate with self-ligating brackets vs conventional

634 Temporary anchorage device insertion variables: effects on retention

Joseph S. Petrey, Marnie M. Saunders, G. Thomas Kluemper, Larry L. Cunningham, and

Yossi Abed and Ilana Brin

649 Suitability of orthodontic brackets for rebonding and reworking following

670 Reliability of reference distances used in photogrammetry

Muge Aksu, Demet Kaya, and Ilken Kocadereli

Brian J. Schabel, Tiziano Baccetti, Lorenzo Franchi, and James A. McNamara Jr

685 Accuracy and reliability of palatal superimposition of three-dimensional

692 Skeletal changes of maxillary protraction without rapid maxillary

Dong-Yul Lee, Eun-Soo Kim, Yong-Kyu Lim, and Sug-Joon Ahn

699 A cephalometric study to investigate the skeletal relationships in patients

707 3-D analysis of facial asymmetry in children with hip dysplasia

713 Quantity and quality assessment of randomized controlled trials on

Tatsuo Shimada, Hisako Takayama, and Yoshiki Nakamura

719 Cervical vertebra morphology in different skeletal classes

Miyuki Watanabe, Tetsutaro Yamaguchi, and Koutaro Maki

725 Longitudinal growth changes of the cranial base from puberty to

733 Craniofacial growth in ectodermal dysplasia

740 The role of heme oxygenase-1 in mechanical stress- and

Jin-Hyoung Cho, Sun-Kyung Lee, Jin-Woo Lee, and Eun-Cheol Kim

754 Quantification of three-dimensional orthodontic force systems of T-loop

Jie Chen, Serkis C. Isikbay, and Edward J. Brizendine

759 Cytotoxic effects of orthodontic composites

765 Influence of ceramic (feldspathic) surface treatments on the micro-shear

783 New treatment modality for maxillary hypoplasia in cleft patients

Seung-Hak Baek, Keun-Woo Kim, and Jin-Young Choi

792 Why read this journal?

611 Angle Orthodontist, Vol 80, No 4, 2010

Angle Orthodontist, Vol 80, No 4, 2010

Clinical photography vs digital video clips for the

INTRODUCTION been prolific in recent years with articles that have

Angle Orthodontist, Vol 80, No 4, 2010 678 DOI: 10.2319/052207-243.1

Angle Orthodontist, Vol 80, No 4, 2010

were used. A CanonH EF 35 mm SLR camera

Angle Orthodontist, Vol 80, No 4, 2010

Angle Orthodontist, Vol 80, No 4, 2010

Angle Orthodontist, Vol 80, No 4, 2010

Angle Orthodontist, Vol 80, No 4, 2010

Angle Orthodontist, Vol 80, No 4, 2010

Absolute anchorage with universal T-loop mechanics for severe deepbite

INTRODUCTION convex profile and also increase the relapse potential

DOI: 10.2319/052809-304.1 771 Angle Orthodontist, Vol 80, No 4, 2010

Figure 1. Pretreatment facial (A) and intraoral (B) photographs.

Angle Orthodontist, Vol 80, No 4, 2010

Angle Orthodontist, Vol 80, No 4, 2010

Table 1. Cephalometric Measurements

Figure 4. Pretreatment cephalometric radiograph (A) and tracing (B).

Angle Orthodontist, Vol 80, No 4, 2010

Angle Orthodontist, Vol 80, No 4, 2010

Figure 7. Posttreatment dental casts.

Angle Orthodontist, Vol 80, No 4, 2010

Figure 8. Posttreatment facial (A) and intraoral (B) photographs.

sion of the anterior segment was observed (Fig-

Angle Orthodontist, Vol 80, No 4, 2010

Angle Orthodontist, Vol 80, No 4, 2010

Angle Orthodontist, Vol 80, No 4, 2010

To correct a deep overbite in this case, intrusion of the

Angle Orthodontist, Vol 80, No 4, 2010

Figure 13. Dental casts 10 years after treatment.